United States v. Fields

516 F.3d 923 (2008)

UNITED STATES of America, Plaintiff-Appellee,
v.
Edward Leon FIELDS, Jr., Defendant-Appellant.
State of Oklahoma, Amicus Curiae.

No. 05-7128.

United States Court of Appeals, Tenth Circuit.

February 25, 2008.

Vicki Mandell — King, Assistant Federal Public Defender (Raymond P. Moore, Federal Public Defender; James L. Hankins, Ogle & Welch, P.C., Oklahoma City, Oklahoma, with her on the briefs), Denver, Colorado, appearing for Defendant — Appellant.

Libra Lange, United States Department of Justice, Capital Case Unit (Jeffrey B. Kahan, United States Department of Justice, Capital Case Unit; Sheldon J. Sperling, United States Attorney Eastern District of Oklahoma, Muskogee, Oklahoma, with her on the briefs), Washington, D.C., appearing for Plaintiff — Appellee.

W.A. Drew Edmondson, Attorney General of Oklahoma, Sandra D. Rinehart, Senior Assistant Attorney General, Oklahoma City, Oklahoma, filed an amicus curiae brief for the State of Oklahoma.

Before HENRY, Chief Circuit Judge, TACHA, and KELLY, Circuit Judges.

TACHA, Circuit Judge.

Edward Leon Fields killed Charles and Shirley Chick at the Winding Stair Campground in the Ouachita National Forest on July 10, 2003. He had seen the couple there days earlier and drove there the evening of July 10 with a homemade ghillie suit (a covering for head and body made to resemble underbrush that Fields referred to as his sniper suit) and a camouflaged and powerfully scoped rifle in his truck. He found the Chicks on a vista some distance from their campsite. He retrieved the rifle, put on the ghillie suit, and hid near their campsite as it grew dark. In time, the Chicks came back to the campsite and sat at a table. Fields waited and watched them for about twenty minutes. When Charles told Shirley he was going to the tent, Fields shot him in the face. As Charles slumped to the table, Shirley got up and began running toward the couple's van. Fields shot at her and a bullet tore through her foot. She reached the passenger door of the van, but was shot again, on the side of her head. Fields caught up and shot her once more, in the back of the head, in the doorway of the van. Shirley died as a result of both head wounds. Fields returned to the table and shot Charles a second time in the head. Charles also died as a result of both of his wounds.

Physical evidence indicated that Fields then left the campsite and only returned hours later, when he broke the driver's window of the van and stole some items. He rummaged through only the driver's area of the van; the rest of the van and the Chicks' tent were untouched. A tip eventually led police to Fields' truck, where they found the rifle, the ghillie suit, and some of the items stolen from the Chicks' van. In the meantime, Fields had been taken in for questioning. He initially denied any connection to the crime, but confessed when confronted with the evidence taken from his truck.

An indictment was obtained in the Eastern District of Oklahoma against Fields charging, among other crimes, two first-degree murder counts under 18 U.S.C. § 1111. Fields pled guilty and a capital sentencing proceeding was conducted pursuant to the Federal Death Penalty Act of 1994 (FDPA),^[1] in particular 18 U.S.C. §§ 3591-94. At the conclusion of the proceeding, the jury determined that Fields was eligible for a death sentence under §§ 3591(a)(2) and 3593(e)(2) by finding, unanimously and beyond a reasonable doubt, (1) that he possessed the requisite homicidal intent, and (2) the presence of one (here, two) statutorily defined aggravating factors ("statutory aggravators"): substantial planning and premeditation to cause death (§ 3592(c)(9),), and multiple intentional killings committed in a single episode (§ 3592(c)(16)).

The jury then turned to the ad hoc nonstatutory aggravators framed and formally noticed by the government under § 3593(a). Such aggravators do not affect a defendant's eligibility for a death sentence, but contribute to the jury's assessment of whether a death sentence, if available, should be imposed. The jury found, again unanimously and beyond a reasonable doubt, that Fields (1) posed a future danger to the lives and safety of other persons; (2) caused permanent loss to Charles Chick's family, friends, and community; (3) caused permanent loss to Shirley Chick's family, friends, and community; and (4) inflicted mental anguish on Shirley Chick before her death.

Next, the jury considered a host of mitigating factors offered by the defense. At least one juror found, by the required preponderance of the evidence, that (1) Fields did not have a significant prior criminal history; (2) Fields served in and was honorably discharged from the Navy; (3) Fields had worked as a state prison guard; (4) Fields has special talents in cooking, art, and computers; (5) Fields is a loved father; (6) Fields is a loved brother; (7) Fields is a loved son; (8) Fields is a valued friend; (9) Fields' father died months before the offenses; (10) Fields' mother moved away weeks before the offenses; (11) Fields' ex-wife and their children moved away months before the offenses; (12) Fields' ex-wife recently had cancer that may or may not be in remission; (13) Fields' death will impact his children, family, and friends; (14) Fields cooperated with authorities after his arrest; (15) Fields confessed to the crimes; (16) Fields pled guilty to the crimes; and (17) Fields sought treatment for mental illness. All jurors, however, rejected several mitigators, including that (1) Fields' capacity to appreciate the wrongfulness of his conduct and conform his conduct to the law was significantly impaired; (2) Fields committed the offenses under severe mental or emotional disturbance; (3) Fields expressed remorse for the crimes; and (4) Fields will not present a future danger to society by being imprisoned for life without possibility of release.

Finally, pursuant to § 3593(e), the jury weighed all of the aggravating and mitigating factors to determine whether the aggravators sufficiently outweighed the mitigators to justify a sentence of death. The jury concluded, unanimously, that they did. Thereafter, the district court imposed death sentences on both murder counts.

Fields now appeals, challenging the jurisdictional basis for his federal conviction, which is a matter not waived by his guilty plea, see, e.g., United States v. Kunzman, 125 F.3d 1363, 1365 (10th Cir.1997), abrogated on other grounds by Slack v. McDaniel, 529 U.S. 473, 478, 120 S.Ct. 1595, 146 L.Ed.2d 542 (2000), and raising many other objections with respect to the sentencing proceeding. Addressing each of these matters in turn, we conclude that federal jurisdiction was properly exercised and that no reversible error occurred in the proceedings.

I. FEDERAL TERRITORIAL JURISDICTION

Fields was convicted of murder "[w]ithin the . . . territorial jurisdiction of the United States." 18 U.S.C. § 1111(b). Under 18 U.S.C. § 7(3) that territorial jurisdiction includes "lands reserved or acquired for the use of the United States, and under the exclusive or concurrent jurisdiction thereof." Thus, satisfaction of the latter conditions is a jurisdictional element of the offense of conviction. See United States v. Young, 248 F.3d 260, 275 (4th Cir.2001).

Fields challenges federal territorial jurisdiction here on three alternative grounds: (1) Oklahoma either reserved exclusive jurisdiction over the national forest land where the murders took place, or ceded only a limited proprietary jurisdiction to the United States, in neither case allowing for the exercise of federal territorial jurisdiction; (2) if Oklahoma purported to cede true concurrent jurisdiction, federal statutes controlling at the time the land was acquired did not authorize the United States to exercise concurrent territorial jurisdiction over the national forest land; and/or (3) the United States in any event did not accept such jurisdiction. We reject all of these contentions and hold that the United States properly exercised its concurrent territorial jurisdiction to prosecute Fields for murdering the Chicks on national forest land. Our conclusion is consistent with the only cases closely on point See United States v. Raffleld, 82 F.3d 611, 612-13 (4th Cir.1996); United States v. Gabrion, No. 1:99-CR-76, 2006 WL 2473978, at *2-*10 (W.D.Mich. Aug.25, 2006) (unpublished decision).

A. Oklahoma's Cession of General Concurrent Jurisdiction

1. Language of the Cession Act

Oklahoma consented to federal acquisition of its land for the creation and expansion of national forests in 1925, through legislation that also specified both the territorial jurisdiction retained by the State and the territorial jurisdiction ceded to the federal government:

That the consent of the State of Oklahoma be and is hereby given to the acquisition by the United States . . . of such lands in Oklahoma, as in the opinion of the Federal Government may be needed for the establishment, consolidation and extension of National Forests in the State; provided, that the State of Oklahoma shall retain a concurrent jurisdiction with the United States in and over the lands so acquired, so far that civil process in all cases, and such criminal process as may issue under the authority of the State of Oklahoma against any person charged with the commission of any crime without or within said jurisdiction, may be executed thereon in like manner as if this Act had not been passed.

Power is hereby conferred upon Congress of the United States to pass such laws and to make or provide for the making of such rules and regulations of both a civil and criminal nature, and provide punishment therefor, as in its judgment may be necessary for the administration, control and protection of such lands as may be from time to time acquired by the United States under the provisions of this act.

Act effective April 8, 1925 (Cession Act), ch. 42, §§ 1-2, 1925 Okla. Sess. Laws 61-62 (current version at Okla. State. tit. 80, §§ 6, 7) (emphasis added).

The first section of the Cession Act indicates that the United States is being ceded full civil and criminal jurisdiction, with a concurrent jurisdiction reserved to the State. Fields argues, however, that the specific reference in the second section to the conferral of power on Congtess "necessary for the administration, control, and protection of [the] lands," suggests that the ceded federal jurisdiction has a narrower range, relating only to the government's proprietary interests in the land, and thus is insufficient to support the imposition and enforcement of the criminal law under which he was prosecuted. As explained below, the only pertinent Oklahoma case construing the Cession Act supports the government's position that the Act cedes to the United States traditional concurrent jurisdiction. Fields' reading of the second section of the Act as imposing an enforceable substantive restriction on Congress's power, such that any legislation regarding national forest lands may be challenged for alleged deviation from the purposes of "administration, control, and protection," ignores the modifying phrase "as in [Congress's] judgment may be necessary for [such purposes]."

2. Cline and the interpretation of the Cession Act

In State v. Cline, 322 P.2d 208 (Okla. Crim.App.1958), the Oklahoma Court of Criminal appeals considered state jurisdiction over crimes in a federal game reserve and in that regard addressed the meaning of the Cession Act.^[2] The broadest holding in Cline, that "the State of Oklahoma did not cede exclusive jurisdiction to the United States," id. at 216, is not a point of dispute here. What is contested is the nature of the non-exclusive jurisdiction ceded by the State and whether it encompasses enforcement of federal criminal law, particularly the murder offense specified in § 1111. In that regard, Fields makes much of Cline's statement that Oklahoma could continue to exercise jurisdiction "not in conflict with the authority vested in the Secretary of Agriculture to punish violations of the wildlife regulations prescribed by him," 322 P.2d at 216. He contends this statement implies that the only jurisdiction ceded to the United States was for enforcement of federal regulations relating to proprietary interests inherent in the purposes for which state land was acquired — in Cline for the protection of wildlife, and here for the protection of forest land.

We disagree. First, regardless whether a state has fully ceded or withheld territorial jurisdiction over land acquired by the United States, a state cannot enforce laws that are inconsistent with the federal uses for which the land was acquired. See James v. Dravo Contracting Co., 302 U.S. 134, 141-42, 147, 58 S.Ct. 208, 82 L.Ed. 155 (1937); Surplus Trading Co. v. Cook, 281 U.S. 647, 650, 50 S.Ct. 455, 74 L.Ed. 1091 (1930); United States v. Lewisburg Area Sch. Dist., 539 F.2d 301, 307 (3d Cir.1976). Hence, the quote from Cline does not necessarily involve a judgment one way or the other as to the nature of the territorial jurisdiction ceded to the United States; rather, it merely observes the overriding priority accorded federal property use. See also Kleppe v. New Mexico, 426 U.S. 529, 543," 96 S.Ct. 2285, 49 L.Ed.2d 34 (1976). Second, the context of the quote suggests that, far from denying any authority beyond the proprietary, Cline was acknowledging the presence of an unexercised reservoir of such authority. Before referring to the wildlife regulations, Cline noted that the "United States does not now assert by statute anything more than proprietary control through [regulations issued by] the Secretary of Agriculture." 322 P.2d at 215 (emphasis added). Thus, the court's subsequent reference to the wildlife regulations reflected a recognition of the limited authority Congress had thus far chosen to exercise, implying that the potential authority ceded to it by the Cession Act reached beyond such proprietary interests.

There is another statement in Cline that at first blush may appear to support Fields' position but which, on closer examination, supports federal jurisdiction. Cline expressed concern that if the State did not exercise jurisdiction and enforce its criminal law in the reserve, there was nothing to fill the gap: "The area is without protection as to breaches of the peace . . ., unless Oklahoma is permitted to exercise the residuum of jurisdiction it retains over the area[.]" Id. at 216. This statement appears to imply that federal territorial jurisdiction is indeed absent, as otherwise there would be no gap—the Assimilative Crimes Act (ACA), 18 U.S.C. § 13, would have transformed all state crimes into federal crimes enforceable in federal courts, see id. § 13(a).

But that implication is undercut by the very next passage in Cline, which points out that "[i]n James Stewart & Co. v. Sadrakula, [309 U.S. 94, 60 S.Ct. 431, 84 L.Ed. 596 (1940)], reference is made to certain Congressional acts adopting state statutes as a means of promoting uniformity in law enforcement affecting such areas as the refuge herein, but these acts were repealed [in 1948]." Cline, 322 P.2d at 216. The cited case discussed an earlier version of the ACA repealed in 1948 (though replaced by 18 U.S.C. § 13(a), evidently unnoticed by the Cline court). Sadrakula, 309 U.S. at 100-01, 60 S.Ct. 431. In short, Cline was concerned about a gap in criminal-law protections created by the repeal of the ACA, not by the absence of jurisdiction sufficient to impose and enforce criminal law through such federal legislation. This whole discussion suggests that the exercise of federal authority reflected in the ACA (based explicitly on territorial jurisdiction) had previously filled the criminal-law gap that the Cline court was concerned state law now had to fill. Once again, Cline indicates that federal criminal authority was absent in the game reserve because it had not been exercised, not because the federal government lacked jurisdiction to do so.

In sum, Cline is fully consistent with, indeed lends substantial support to, our conclusion that the Cession Act cedes concurrent territorial jurisdiction to the United States. As discussed below, even if that jurisdiction is qualified by the Act's reference to the administration, control, and protection of forest lands, that qualification involves a matter committed solely to Congress's judgment and does not afford Fields any basis on which to challenge the jurisdictional validity of the murder statute under which he was prosecuted.

3. Necessity Clause of Cession Act and Congressional Judgment

Fields' argument under the Cession Act ultimately rests on the premise that if federal jurisdiction were restricted to matters "necessary for the administration, control, and protection of [national forest] lands," Cession Act, ch. 42, § 2, it would not encompass the murder offenses for which he has been prosecuted. He insists that this "necessity" clause does not authorize the general application of federal criminal law, and particularly federal offenses against persons rather than property, because the administration, control, and protection of national forests does not extend to anything beyond the land itself. While we do not necessarily share this cramped view of what is entailed in the proper administration, control, and protection of the national forests, we reject Fields' argument on the more fundamental basis that this condition on the exercise of ceded federal authority is, as the Cession Act provides, to be applied as appropriate "in [Congress's] judgment," id. (emphasis added), not in the judgment of the parties or the court.

Nearly identical "necessity" clauses were used in the state cession Acts at issue in the Raffield and Gabrion cases relied on by the government for its general position on concurrent jurisdiction. We find Gabrion's treatment of the clause especially instructive. It did not analyze the meaning of the terms used in the clause to resolve whether, in its view, the function of the federal criminal law at issue fell within their scope. Rather, it held that the clause as a whole imposed only such constraint as Congress recognized, because the state did not (as Oklahoma here did not) reserve "the authority to second guess the United States as to what prosecutions it might consider necessary for the administration, control and protection of the national forest lands." Gabrion, 2006 WL 2473978, at *4. Gabrion thus concludes, as we do, that the necessity clause does not require an independently confirmed administrative need before federal authority may be exercised, but only Congress's judgment that such a need warrants legislative action.

We are aware of no contrary authority, nor has Fields suggested any, to undercut this plain reading of the language in the Cession Act. Consequently, we hold that Congress's extension of federal criminal authority to all lands over which the United States has concurrent jurisdiction in 18 U.S.C. § 7(3), and its explicit application of the murder statute to this same range of lands in 18 U.S.C. § 1111(b), did not exceed the territorial jurisdiction ceded by Oklahoma to the federal government over property acquired as national forest land.

B. Territorial Jurisdiction Permitted by 16 U.S.C. § 480 (Weeks Act)

Turning to the relevant federal statutes, Fields argues that the Weeks Act so restricts federal authority' in national forests as to peremptorily preclude territorial jurisdiction altogether, even if the State has ceded it. This position is not only contrary to case law specifically construing the Weeks Act, but at odds with broader federal criminal jurisprudence.

1. Statutory language

The Weeks Act states:

The jurisdiction, both civil and criminal, over persons within national forests shall not be affected or changed by reason of their existence, except so far as the punishment of offenses against the United States therein is concerned; the intent and meaning of this provision being that the State wherein any such national forest is situated shall not, by reason of the establishment thereof lose its jurisdiction, nor the inhabitants thereof their rights and privileges as citizens, or be absolved from their duties as citizens of the State.

16 U.S.C. § 480 (emphasis added). Fields insists that this allocates exclusive territorial jurisdiction to the states, excepting only the federal government's authority to punish offenses relating solely to its proprietary interest in the land.

First of all, we emphasize that "offenses against the United States" does not refer to a subset of offenses relating only to federal property, which seems to be a tacit premise of Fields' argument. The quoted phrase, familiar from countless indictments, simply "means any federal offense," Iysheh v. Gonzales, 437 F.3d 613, 614 (7th Cir.2006)—an understanding reflected in the generic use of the phrase throughout the federal criminal code (for example, to define such general federal criminal concepts as principals and accessories, 18 U.S.C. §§ 2, 3, and criminal conspiracies, 18 U.S.C. § 371). Use of the phrase clearly does not signal a restriction to crimes against federal property.

2. Case law on § 480 and criminal practice in national forests

More generally, Fields does not cite a single case confirming his view that § 480 precludes federal territorial jurisdiction. The cases recognize, rather, that the effect of § 480 was to enable states to "maintain concurrent criminal and civil jurisdiction over national forests."^[3]United States v. California, 655 F.2d 914, 919 (9th Cir. 1980) (emphasis added); see also Raffield, 82 F.3d at 613 (explaining that § 480 "means only that the mere establishment of the forest does not alter the jurisdictional status of the land" and § 480 "does not in any way preclude state and federal governments from entering into a relationship of concurrent jurisdiction"); Gabrion, 2006 WL 2473978, at *2-*3 (same). This is not an aberrational holding of a few isolated cases, as Fields suggests. If § 480 precluded federal territorial jurisdiction as Fields argues, offenders could not be prosecuted for federal crimes that require territorial jurisdiction (such as § 1111 murder, or state-defined offenses adopted as federal crimes under the ACA), if such crimes occurred in national forests. The federal prosecutions in Raffield (ACA) and Gabrion (murder) are not unique in belying this implication of Fields' position. See also United States v. Avants, 367 F.3d 433, 438 (5th Cir.2004) (murder); United States v. Terry, 86 F.3d 353, 354 (4th Cir.1996) (ACA); United States v. Couch, 65 F.3d 542, 543 (6th Cir.1995) (ACA).

Fields tries to support his position that § 480 limits federal jurisdiction to the protection of proprietary interests by noting that another statute, originally enacted in 1897, includes a provision that authorizes the Secretary of Agriculture to

make provisions for the protection against destruction by fire and depredations upon the public forests and national forests which may have been set aside or which may be hereafter set aside . . . and he may make such rules and regulations to regulate their occupancy and use and to preserve the forests thereon from destruction . . .

16 U.S.C. § 551. Fields contends this delegation of regulatory authority to protect forest property indicates that such protection reflects the limits of legislative jurisdiction over forest land. This contention is meritless. Fields' underlying assumption, that the extent of regulatory authority Congress elects to delegate equals or delimits the reach of its own legislative jurisdiction, is obviously faulty as a general matter, and he cites no authority to support equating the two distinct concepts in this particular instance.

Fields also notes that before passage of the current version of 18 U.S.C. § 7(3) in 1940, federal criminal jurisdiction extended only to land within the. United States' exclusive territorial jurisdiction. He concedes that this has no effect on the federal government's present authority to prosecute crimes on land now within its concurrent jurisdiction, but contends that the limited reach of federal criminal jurisdiction at the time of the Weeks Act is an "indication that [it] did not authorize the. United States to acquire concurrent legislative jurisdiction over national forest lands." Aplt. Reply Br. at 7. He reasons "there would be little point in acquiring such jurisdiction when offenses committed on lands over which the United States had only concurrent jurisdiction did not constitute offenses against the laws of the United States." Id. at 8. This argument rests on a tacit premise that Fields makes no attempt to substantiate, i.e., that the sole purpose of acquiring federal territorial jurisdiction is to effectuate the application of federal criminal law. In any event, as Congress always could—and ultimately did—extend federal criminal jurisdiction to include lands within its concurrent territorial jurisdiction, the fact that such an extension had not yet been made when it passed the Weeks Act would not have made its acceptance of concurrent jurisdiction over forest lands pointless. The point would have been to preserve its power to extend federal criminal jurisdiction over forest lands if and when it wished to do so in the future.

3. Solicitor opinion letter and administrative deference

Fields relies on a 1942 opinion letter from the Solicitor of the Department of Agriculture concluding inter alia that § 480 "precluded the exercise of concurrent jurisdiction as well as exclusive jurisdiction by the United States over national forest lands." Aplt. Reply Br. at 4 & Attach. 2 at 3; see also Aplt. Suppl. Reply Br. at 5-7. Fields insists we owe this opinion "Chevron deference."^[4] We disagree.

Such deference applies "when it appears that Congress delegated authority to the agency generally to make rules carrying the force of law, and that the agency interpretation claiming deference was promulgated in the exercise of that authority." United States v. Mead Corp., 533 U.S. 218, 226-27, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001). These prerequisites are not satisfied here. Agency views expressed in opinion letters lack the force of law and thus do not warrant Chevron deference. Christensen v. Harris County, 529 U.S. 576, 587, 120 S.Ct. 1655, 146 L.Ed.2d 621 (2000). More fundamentally, the proper reach of the federal legislative power, and the jurisdiction it vests in the federal courts, is categorically not a matter of agency judgment: "Determining federal court jurisdiction is exclusively the province of the federal courts regardless of what an agency may say." Lindstrom v. United. States, 510 F.3d 1191, 1195 n. 3 (10th Cir.2007) (internal quotation marks omitted).

In sum, Fields' reliance on § 480 as a disavowal of federal territorial jurisdiction, permitting only a proprietary interest in national forest lands, is contrary to interpretations of the statute by other circuits and inconsistent with federal criminal jurisprudence as a general matter.

C. Federal Acceptance of Less than Exclusive Jurisdiction Prior to 1940

Alternatively, Fields argues that even if § 480 did not bar federal territorial jurisdiction explicitly, it did so implicitly by requiring the retention of concurrent jurisdiction by the ceding state, which in turn precluded the only form of jurisdiction (exclusive) the United States was authorized to accept at the time it acquired the land at issue here. This argument rests on the faulty premise that the United States could not accept a cession of less than exclusive jurisdiction.^[5]

When the United States acquired the land at issue here in 1931, federal acceptance of ceded jurisdiction was presumed absent an express refusal. See Fort Leavenworth R. Co. v. Lowe, 114 U.S. 525, 528, 5 S.Ct. 995, 29 L.Ed. 264 (1885); United States v. Lewisburg Area Sch. Dist., 539 F.2d 301, 306 n. 12 (3d Cir.1976); United States v. Johnson, 426 F.2d 1112, 1114 (7th Cir.1970). In 1940, Congress passed 40 U.S.C. § 255 (now § 3112), abrogating the rule of presumptive acceptance and substituting a notice procedure by which authorized federal officials expressly accept ceded jurisdiction.^[6] The statute also specifically permits the United States to accept less than exclusive territorial jurisdiction. 40 U.S.C. § 3112(a). Seizing on this latter point, Fields draws the negative implication that, prior to 1940, the. United States was unable to accept less than exclusive jurisdiction. But the recognition of non-exclusive territorial jurisdiction in § 3112(a) merely affirmed a point already well established by Supreme Court authority. See James v. Dravo Contracting Co., 302 U.S. 134, 147-48, 148-49, 58 S.Ct. 208, 82 L.Ed. 155 (1937) (citing cases, including Fort Leavenworth, for rule that state may qualify cession of jurisdiction, thereby creating concurrent state and federal jurisdiction, and extending rule to Enclave Clause property); see also United States v. Burton, 888 F.2d 682, 684-85 & n. 3 (10th Cir.1989) (citing James and Fort Leavenworth as establishing that "[t]he federal government can obtain concurrent or exclusive jurisdiction over specific property either by complying with the provisions of [the Enclave Clause], or by receiving a cession of legislative jurisdiction from the state in which the property is located." (footnote omitted)). As there were no restrictions on the cession and acceptance of concurrent territorial jurisdiction at the relevant times, Oklahoma effectively ceded and the United States could properly accept concurrent jurisdiction over the national forest land at issue here.

D. Presumptive Acceptance of Cession of Concurrent Jurisdiction

Fields further contends that, even if the United States was not categorically limited to exclusive jurisdiction, that was the only form of jurisdiction to which the presumption of acceptance applied. This argument is meritless. When the United States acquired the land in question, the case law (beginning with Fort Leavenworth in 1885) had established that (1) acceptance of ceded jurisdiction was presumed absent refusal and (2) jurisdiction obtained through cession could be less than exclusive. The natural, syllogistic conclusion is that acceptance of concurrent jurisdiction was presumed absent refusal, and Fields has not cited a single case holding otherwise. Fort Leavenworth itself applied the presumption to a cession of jurisdiction so qualified by the state that it prompted the Court to simultaneously recognize that states may cede less than exclusive jurisdiction. 114 U.S. at 528, 541, 5 S.Ct. 995.

E. Rebutting the presumption of acceptance and request for remand

Fields insists he can rebut the presumption of acceptance with evidence of affirmative disavowals of territorial jurisdiction. He cites internal agency reports and letters written over the years taking the general position that federal territorial jurisdiction does not, or should not, extend to national forests. We agree with the government that such "[e]vidence that Executive Branch employees expressed a preference for proprietary jurisdiction several decades ago [but after acquisition of the land here] is irrelevant to the analysis of the government's 1931 acceptance of concurrent legislative jurisdiction over the Winding Stair Campground." Aplee. Suppl. Br. at 19. The issue we must decide is whether, in 1931 when the land here was acquired, the United States affirmatively refused the cession of concurrent jurisdiction Oklahoma made in 1925 with respect to national forest acquisitions. None of the later materials Fields cites demonstrates such a refusal.

Anticipating this point, Fields argues that "it was not until 1940 (when Congress enacted 40 U.S.C § 255, requiring formal notice of acceptance of ceded jurisdiction), that there was any occasion to articulate the jurisdictional manner in which national forest lands were held by the United States." Aplt. Reply Br. at 14. We disagree. Given the presumption of acceptance, a pre-1940 acquisition of national forest land provided an essential occasion on which to articulate the mode of jurisdiction involved, i.e., to explicitly refuse the jurisdiction ceded by the state or accept it through silence.

As a position of last resort, Fields requests a remand to pursue discovery if we "remain[] unconvinced that the presumption of acceptance does not apply or has been rebutted." Id. at 17. This request misapprehends the nature of the matter. It is not a question of excavating yet more internal documents reflecting after-the-fact (and changeable) legal opinions or practical preferences of agency officials. The requisite explicit refusal of territorial jurisdiction over the lands acquired from Oklahoma in 1931 should be evident from the contemporaneous public record. We know of no such refusal, nor does Fields offer any indication that one exists. This is precisely the situation the presumption of acceptance was designed to address. That is, we need not be concerned that jurisdiction is fatally indeterminate on our present record (which would require a remand); given the presumption, territorial jurisdiction is established because there is no evidence affirmatively negating its acceptance. Under the circumstances, Fields is not entitled to a remand and we see no reason to grant one in our discretion.

II. EXCUSAL OF POTENTIAL JUROR FOR CAUSE UNDER WITHERSPOON-WITT DEATH-QUALIFICATION STANDARD^[7]

Fields contends that the district court erred in sustaining the government's challenge to a potential juror for cause based on his inability to properly consider imposition of the death penalty under the governing law. As explained below, given the extensive questioning of the potential juror and the district court's thorough and particularized consideration of the matter, including its assessment of the potential juror's credibility, we are convinced the court properly exercised its discretion in excusing the juror.

The Supreme Court recently summarized the standards for assessing Witherspoon-Witt challenges in Uttecht v. Brown, ___ U.S. ___, 127 S.Ct. 2218, 167 L.Ed.2d 1014 (2007). The dispositive question is whether a potential juror is "substantially impaired in his or her ability to impose the death penalty under the [governing legal] framework." Id. at 2224. The district court's ruling, "based upon determinations of demeanor and credibility that are peculiarly within a trial judge's province," warrants "deference even on direct review." Id. at 2223 (internal quotation marks omitted). And the district court is accorded broad discretion when "there is lengthy questioning of a prospective juror and the trial court has supervised a diligent and thoughtful voir dire." Id. at 2230. The excusal of a prospective juror must be affirmed unless "the record discloses no basis for a finding of substantial impairment." Id. And "when there is ambiguity in the prospective juror's statements, the trial court, aided as it undoubtedly [is] by its assessment of [the juror's] demeanor, [is] entitled to resolve it in favor of the State." Id. at 2223 (internal quotation marks omitted). Thus, the excusal of a potential juror "may be upheld even in the absence of clear statements from the juror that he or she is impaired." Id. Indeed, even "assurances that [the juror] would consider imposing the death penalty and would follow the law . . . d[o] not require the trial court to deny the State's motion to excuse," id. at 2229, if "these responses were interspersed with more equivocal statements," id. at 2227.

The potential juror here said he could not support the death penalty "except in very extreme cases," a category of death-eligibility far more limited than that countenanced by the FDPA, as reflected in the instances the juror cited: genocide; torture; and willful killing of children.^[8] R. Vol. 3 at 119, 122. Other than those cases, he stated that he would impose the death penalty only if specifically directed by law to do so. Id. at 119-20. Further questioning, by the government to bring out the potential juror's reservations about the death penalty and by the defense to defuse them, generated inconclusive, inconsistent, and/or ambiguous answers. Finally, the court explicitly asked him whether he could follow the court's instructions and approach the case on a "clean slate" separate from his personal moral criteria for the death penalty, or had a "definite impression" that his views "would substantially impair his ability to follow the law." Id. at 127. He admitted he would be substantially impaired.^[9]Id.

Various interpretations of all this may be possible, but a reasonable one is that, aside from the exception drawn for genocide, torture, and willful killing of children, the potential juror would balk at a death sentence unless the law left him no discretion but mandated that it be imposed. That would certainly qualify him as substantially impaired in his ability to comply with a juror's responsibility under the FDPA, which (as the instructions in this case explicitly affirmed) never mandates the death penalty but, rather, requires jurors to exercise their discretion and decide for themselves whether the death penalty is warranted under the court's instructions. Finally, we note that the district court's explanation of its assessment of the matter, both when it excused the juror and when it denied a request to reconsider the matter and recall him, reflected its reliance on classic credibility observations. See id. at 129; R. Vol. 11 at 1690-92.

Under the guiding principles from Uttecht noted at the outset, we affirm the district court's decision. The district court has discretion in such matters generally, and our affirmance of the particular ruling here is reinforced by the extensive individualized inquiry it conducted, the presence of ambiguities within its province to resolve, and its express reliance on its firsthand assessment of credibility. We see no basis for disturbing the district court's ruling.

III. GENERAL VERDICT OF DEATH

Although Fields was indicted, convicted, and sentenced to death on two murder counts, the verdict form and instructions did not direct the jury to weigh the aggravating and mitigating factors relevant to each count and determine whether a death sentence was warranted for either murder. Rather, the jury was directed to weigh all of the aggravators and mitigators in the case at once and reach a single sentencing verdict.

Specifically, the verdict form and associated instructions properly directed the jury to find those aggravators applicable to the murder of Charles Chick (substantial planning and premeditation; intentional killing of more than one person; future dangerousness; and victim impact relating to Charles) and those applicable to the murder of Shirley Chick (substantial planning and premeditation; intentional killing of more than one person; future dangerousness; victim impact relating to Shirley; and mental anguish of victim). R. Vol. 1, Doc. 228 at 34-37. The mitigators, which related to the defendant and thus did not vary with the victims, were properly found generally, without reference to each murder. Id. at 37-42.

At that point, instead of being told to weigh the aggravators and mitigators on each count and record the resultant sentences on separate forms, the jury was given a general form with these directions regarding the "Weighing Process" and "Imposition of Sentence," respectively:

The question you must answer at this stage of your deliberations is whether the proven aggravating factor(s) sufficiently outweigh the proven mitigating factors and information to justify a sentence of death . . . If you unanimously find that the weight of the aggravating factor(s) is sufficient to justify a sentence of death, answer "yes" below [and] record your verdict on [the general verdict form for death] . . . If you do not unanimously find that a death sentence is justified, answer "no" below, stop your deliberations [and] sign [the general verdict form for life imprisonment].

* * * *

This is the last step in your deliberations. If you have made all of the findings necessary to make the defendant eligible for a death sentence and have unanimously concluded that such a sentence is justified and that a sentence of death is therefore appropriate in this case, record your decision by collectively signing, the verdict [form for death] . . . and notify the court that you have reached a decision. If you do not unanimously conclude that sentence of death is justified and therefore must be imposed, sign the verdict for life imprisonment . . . and notify the court that you have reached a decision.

Id. at 42-43. The form specifying "that a sentence of death shall be imposed on the defendant"—without reference to any particular count—was signed by all of the jurors. Id. at 44. The court's subsequent poll of the jurors, all of whom affirmed the verdict, was likewise unitary. R. Vol. 22 at 3485-86. Pointing up the potential problem here, however, the Judgment and Commitment Order entered by the court reflected imposition of two separate sentences: "The defendant is hereby sentenced to Death on each of Counts One and Three of the Indictment." R. Vol. 1, Doc. 237 at 2.

Fields argues that it was error to allow the jury to consider a general death sentence in a multi-count case, based on a weighing process incorporating a set of aggravators that combined, and consequently exceeded, those it found applicable to the individual counts for which the death penalty was sought. Fields, however, did not object to the collective weighing process and general verdict below. Indeed, as the government emphasizes, the instructions and verdict form tendered by the defense were functionally the same as those ultimately used by the court. Therefore, because Fields invited any error of which he now complains, we decline to review it. See United States v. Deberry, 430 F.3d 1294, 1302 (10th Cir.2005) ("the invited-error doctrine precludes a party from arguing that the district court erred in adopting a proposition that the party had urged the district court to adopt"); United States v. Visinaiz, 428 F.3d 1300, 1310-11 (10th Cir.2005) (stating that because defense counsel had proffered the allegedly erroneous instruction, "a challenge thereto is precluded as invited error.")

In his reply brief, Fields casts his objection to the general verdict in terms of jury unanimity in order to sidestep application of the invited-error doctrine. According to Fields, his objection is not barred by the invited-error principle because, under United States v. Teague, 443 F.3d 1310 (10th Cir.), cert. denied, ___ U.S. ___, 127 S.Ct. 247, 166 L.Ed.2d 194 (2006), invited error is a waiver principle, id. at 1314-15, and jury unanimity cannot be waived, id. at 1317. This argument to avoid invited error fails because the purported verdict error here does not implicate jury unanimity.

Typically, the problem raised by the use of a general verdict does involve jury unanimity: presented with alternative bases for a verdict and simply asked whether it has unanimously reached that verdict, the jury's affirmative answer does not guarantee that all jurors agreed on the particular basis for it. In the classic example of Richardson v. United States, 526 U.S. 813, 816, 119 S.Ct. 1707, 143 L.Ed.2d 985 (1999), the jury found a continuing criminal enterprise based on the commission of three predicate crimes, but since more than three such crimes were alleged and the jury was not directed to specify which predicate crimes were the operative ones, there was no way of knowing whether it unanimously agreed on the same three. But that is not the type of purported error that occurred here. The jury was not presented with alternative routes to the death penalty; it was given a single basis for imposing a death sentence—the jury had to unanimously find that the aggravators (themselves unanimously found) collectively outweighed the mitigators. Thus, we know what the basis for the jury's verdict was and that it was unanimous.^[10]

In sum, because defense counsel induced any error by seeking to structure the verdict in precisely this improper way, we decline to review it.

IV. DOUBLE-COUNTING OF AGGRAVATOR FOR SUBSTANTIAL PLANNING AND PREMEDITATION

A. Improper Duplication

United States v. McCullah, 76 F.3d 1087, 1112 (10th Cir.1996), held that "the use of duplicative aggravating factors creates an unconstitutional skewing of the weighing process." Fields argues that the aggravator for substantial planning and premeditation (SPP) was improperly double-counted in the weighing process here, because the victims were killed in a single episode that inherently involved a unitary plan but the SPP aggravator was applied to each murder victim and then both iterations were redundantly counted in the weighing process. In reply, the government argues that the SPP aggravators were properly applied to each victim, and, thus, were not duplicative or impermissibly double-counted.

We agree with the government that both murder counts implicated the SPP aggravator, which was therefore properly applied to each. That point does not, however, respond to Fields' objection that the two instances of the aggravator were improperly aggregated when, in the unitary weighing process employed in this case, both murder counts were considered in combination. But, even on this favorable understanding of Fields' argument, relief on appeal is barred by the principle of invited error. Any double-counting potentially involved here is not a free-standing error, but just one aspect of the prejudice arising from the improper weighing process. As discussed above in connection with Fields' challenge to the general verdict, that weighing process was invited by the defense and for that reason is now beyond attack. Fields' objection to the double-counting of the SPP aggravator for both murders, caused by the same weighing process, is likewise not reviewable under the doctrine of invited error.

V. PROOF OF SUBSTANTIAL PLANNING AND PREMEDITATION

Fields also contends that there was insufficient evidence to prove he actually engaged in substantial planning and premeditation with respect to the murders. He argues that the evidence showed he had planned only to rob the victims. Given the established facts and the nature of evidentiary-sufficiency review, this argument is without merit.

A defendant challenging the sufficiency of the evidence faces a "high hurdle," as we "ask only whether, taking the evidence—both direct and circumstantial, together with the reasonable inferences therefrom—in the light most favorable to the government, a reasonable jury could [make the challenged finding] beyond a reasonable doubt." United States v. Jenkins, 175 F.3d 1208, 1215 (10th Cir.1999). The government aptly summarizes facts from which a reasonable jury could (indeed, very likely would) find the SPP aggravator:

[Fields] camouflaged his rifle, carefully constructed a ghillie suit, and practiced stalking people. He potentially began planning the Chicks' murder two days before the offense, when he first saw them. On the night of the murders, he drove to a secluded area and surveilled his victims while they sat on a vista. Instead of burgling their van and fleeing in the Chicks' absence, he methodically donned his ghillie suit, retrieved his rifle and waited for the victims to come within easy range. Even then, [he] watched the Chicks for 15 to 20 minutes before firing, then shot each victim repeatedly to ensure death. [He] left the area for at least an hour before returning to commit burglary, indicating his theft crimes were largely incidental to the homicides.

Br. of Aplee. at 49 (record citations omitted). It is quite possible that pecuniary gain was a motive for the killings, as Fields insists. But this would just reinforce, not negate, the fact that the killings were planned—as the means to achieve the admitted pecuniary end. There was no error in connection with the jury's finding of the SPP aggravator.

VI. FUTURE DANGEROUSNESS— CHALLENGES FOR VAGUENESS, OMISSION OF PRISON SETTING, AND INSUFFICIENT PROOF

The aggravator for future dangerousness was framed as follows:

1. The defendant poses a future danger to the lives and safety of other persons, as evidenced by one or more of the following:

(a) the offenses were a part of a continuing pattern of threatened or impending violence;

(b) the defendant's lack of remorse.

R. Vol. 1 doc. 228 at 36. Fields contends that this non-statutory aggravator was unconstitutionally vague, should have been limited to his future dangerousness in a prison setting, and was not proven beyond a reasonable doubt.

A. Vagueness

Vagueness review is very deferential. Tuilaepa v. California, 512 U.S. 967, 973, 114 S.Ct. 2630, 129 L.Ed.2d 750 (1994). We will not conclude that a non-statutory aggravating factor is unconstitutionally vague so long as "it has some common-sense core of meaning that criminal juries should be capable of understanding." Id. (internal quotation marks omitted). Fields insists the future-dangerousness aggravator was unconstitutionally vague because it failed to provide sufficient guidance on "specific criteria such as [1] criminal acts of violence, [2] the fact that the threat would be continuing, [3] a standard that such acts would be ``probable,' and [4] limiting consideration of this aggravating factor to the prison setting." Aplt. Opening Br. at 77.

1 Lack of reference to criminal violence

The lack of an explicit reference to the criminal nature of the danger posed by the defendant is immaterial. As noted, vagueness review is concerned with the jury's commonsense understanding. It is simply implausible that a jury would mistakenly believe it could impose the death penalty based on non-criminal danger the defendant might pose—as if, for example, being accident prone were relevant to whether someone should be sentenced to life or death.

2. Lack of "continuing" threat

The lack of an explicit reference to the "continuing" nature of the threat posed by the defendant is likewise innocuous. The critical idea of a continuing danger is clearly captured in the aggravator's reference to "future danger to the lives and safety of other persons." R. Vol. 1 doe. 228 at 36. As with the other "specific criteria" Fields summarily alludes to in connection with his vagueness challenge here, he does not present a developed argument as to why specific use of the term "continuing" is necessary. Rather, he simply cites to cases in which courts approved aggravator definitions including the term but nowhere held that it was required.

3. Lack of specific qualification regarding likelihood of future danger

We likewise do not find the aggravator unconstitutionally vague for lack of a qualifying reference regarding the likelihood of the future danger posed by the defendant. Again, Fields does not cite any authority holding that the Constitution requires the court to attach a precise level of probability to the danger involved. We think a commonsense understanding of what is at stake in connection with this intuitive aggravating circumstance is readily grasped without the articulation of particular probabilistic formulations. Given the deferential nature of our review under Tuilaepa, we discern no error here.

4. Failure to explicitly limit future danger to prison setting

When a defendant's future dangerousness is at issue in the weighing stage, "due process requires that the jury be informed that a life sentence carries no possibility for parole." Shafer v. South Carolina, 532 U.S. 36, 51, 121 S.Ct. 1263, 149 L.Ed2d 178 (2001). This requirement is not rigid as to form or mode, however, and may be satisfied "either by a jury instruction or in arguments by counsel." Id. at 39, 121 S.Ct. 1263 (internal quotation marks omitted). The aim is just to ensure that the jury is not misled about a defendant's future incarceration" status when assessing evidence of his future dangerousness. Id. at 49-50, 121 S.Ct. 1263 (explaining thrust of seminal plurality decision in Simmons v. South Carolina, 512 U.S. 154, 114 S.Ct. 2187, 129 L.Ed.2d 133 (1994)). Here, the jury clearly knew about Fields' ineligibility for parole, which was explained in the instructions and reflected in the verdict form itself. And the government effectively tied Fields' incarceration status to his future dangerousness in its only specific reference to the aggravator in closing argument: "Future dangerous[ness]? Which one of us would want to be in a cell next to someone who is capable of coldly attacking two innocent people who had done nothing to him?" R. Vol. 22 at 3465.

Fields nevertheless objects that the aggravator did not explicitly limit the jury's consideration to the prison setting. This is not required by the Simmons line of cases, which concern only the right to inform the jury of a defendant's ineligibility for parole. Noting this point, the Eighth Circuit rejected a similar argument in United States v. Allen, 247 F.3d 741, 788 (8th Cir.2001) (en banc), vacated and remanded on other grounds, 536 U.S. 953, 122 S.Ct. 2653, 153 L.Ed.2d 830 (2002), holding that "[b]ecause the jury was informed of [defendant's] ineligibility for parole," there were "no statutory or constitutional problems with arguing . . . [an] aggravating factor of future dangerousness [not limited to the prison setting] in support of a death sentence," id. at 788-89; see also United States v. Bernard, 299 F.3d 467, 482 (5th Cir.2002) (agreeing with Allen in dicta).

We agree with Allen that an explicit prison-setting limitation need not be included in a future-dangerousness aggravator where, as here, the jury is clearly informed that the defendant would not be eligible for parole if a life sentence were imposed. We emphasize that the government did not affirmatively mislead or confuse the jury on this point in its closing argument, but rather addressed future dangerousness specifically in terms of the prison setting.

B. Sufficiency of Proof of Aggravator

Other than the circumstances of the murders themselves, the government's case on future dangerousness was not particularly strong, but those circumstances were significant. This was not a crime provoked by victim-specific or unique situational factors, such as a sudden confrontation, a score, to settle, or fallout from a personal relationship, that might indicate such violence would not likely recur. On the contrary, this was a coldly calculated murder of strangers who could have been anyone, carried out for money or no particular reason at all, with evidence that the killer had no remorse—none of which suggests an inherently one-time occurrence. And evidence that the modus operandi had been developed over a period of time also does not suggest an abrupt isolated course of conduct to which the defendant would not return. While the particulars of Fields' stalk-and-snipe plan would obviously not directly translate to a prison setting, neither would such a setting deny future occasions for acting out the dangerous personal characteristics that the plan and its execution reflected.

Fields argues for a contrary interpretation of his conduct and character and, hence, the threat he posed to others. But advancing a different assessment of the evidence or urging conflicting inferences therefrom does not demonstrate a legal inadequacy in the government's proof. Rather, when reviewing a jury finding, we are obliged to consider the evidence and available inferences in the light most favorable to the government, Jenkins, 175 F.3d at 1215. Fields' challenge to the future dangerousness finding is meritless.

VII. JURY UNANIMITY ON NONSTATUTORY AGGRAVATOR FOR FUTURE DANGEROUSNESS

The jury was instructed, consistent with 18 U.S.C. § 3593(d), that it had to find all aggravators unanimously. But the future-dangerousness aggravator was framed in terms of two subsidiary grounds (continuing pattern of violence and lack of remorse) and jurors were not told they all had to agree on which ground(s) applied. Fields claims this violated his Sixth Amendment right to, a unanimous jury under Richardson v. United States, 526 U.S. 813, 119 S.Ct. 1707, 143 L.Ed.2d 985, and requires reversal on plain-error review (no objection was raised at trial). We hold that this claim does not satisfy plain-error standards for reversal.

Fields must show "(1) there was error, (2) that is plain, (3) that affects substantial rights, and (4) that seriously affects the fairness, integrity or public reputation of judicial proceedings." United States v. Robertson, 473 F.3d 1289, 1291 (10th Cir.2007) (summarizing test from United States v. Olano, 507 U.S. 725, 732, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993)). It is far from clear that the phrasing of the future-dangerousness aggravator was erroneous; indeed, Fields cites and we have located no authority directly on point. See United States v. Torres-Duenas, 461 F.3d 1178, 1182 (10th Cir.2006) (holding uncertainty in law precluded plain error), cert. denied, ___ U.S. ___, 127 S.Ct. 3054, 168 L.Ed.2d 766 (2007). Moreover, to show the requisite effect on substantial rights, Fields must demonstrate "a reasonable probability that, but for the error claimed, the result of the proceeding would have been different." United States v. Gonzalez-Huerta, 403 F.3d 727, 733 (10th Cir. 2005) (en banc) (internal quotation marks omitted) He does not begin to explain why it is reasonably likely that the jury would not have found the aggravator if instructed differently or that without the aggravator the weighing process would have led to a sentence other than death. His claim therefore fails under the third prong of the plain-error test.

VIII. NON-STATUTORY AGGRAVATOR FOR CAUSING MENTAL ANGUISH

A. Lack of Scienter and Vagueness/Overbreadth

Fields contends that aggravators relating to the infliction of anguish or other special suffering on the part of a victim are unconstitutionally vague and overbroad unless narrowed by a scienter tied to the suffering caused. Fields's reference to vagueness here involves a confusion of constitutional concepts. As we have seen, the vagueness principle has to do with whether an aggravator "``has some commonsense core of meaning . . . that criminal juries should be capable of understanding.'" Tuilaepa, 512 U.S. at 973, 114 S.Ct. 2630 (quotation marks omitted). This core of meaning is lacking if an aggravator "leave[s] the sentencer without sufficient guidance for determining [its] presence or absence." McCullah, 76 F.3d at 1110 (internal quotation marks omitted). Fields is not claiming that "mental anguish" lacks a commonsense core of meaning, leaving it too vague to apply. Rather, his objection is that this concept has too large a range of application, i.e., that it is overbroad and must be narrowed by the addition of a scienter requirement. We therefore construe Fields' objection as one of overbreadth only.

Fields cites Walton v. Arizona, 497 U.S. 639, 110 S.Ct. 3047, 111 L.Ed.2d 511 (1990), and Wade v. Calderon, 29 F.3d 1312 (9th Cir.1994), to support his objection. These cases are inapposite. Both involved overbreadth challenges to aggravators that under the operative capital schemes were prerequisites in the eligibility determination.^[11] At the eligibility stage, aggravators have a distinct constitutional function: narrowing the range of conduct for which the death penalty may even be considered. See Buchanan v. Angelone, 522 U.S. 269, 275-76, 118 S.Ct. 757, 139 L.Ed.2d 702 (1998) (contrasting eligibility and selection stages of sentencing); Tuilaepa, 512 U.S. at 971-75, 114 S.Ct. 2630 (same). Obviously, this narrowing function is inherently related to overbreadth concerns, as Walton and Wade reflect. In contrast, non-statutory aggravators play no role in the eligibility determination under the FDPA, but are relevant only in the weighing process at the ensuing sentence-selection stage. There, the critical constitutional imperative is no longer narrowing the range of potential defendants eligible for the death penalty but, rather, ensuring that the jury focuses on the particular case before it and makes "an individualized determination" whether the death-eligible defendant "should in fact receive that sentence . . . on the basis of [his] character . . . and the circumstances of the crime." Tuilaepa, 512 U.S. at 972, 114 S.Ct. 2630 (internal quotation marks omitted).

We do have relevant guidance from the Supreme Court, however. In Jones v. United. States, 527 U.S. 373, 119 S.Ct. 2090, 144 L.Ed.2d 370 (1999), the Court expressly considered the overbreadth of non-statutory aggravators under the FDPA—and aggravators (vulnerable victim and victim impact) that focused on the victim rather than the defendant and thus had no scienter requirement. Indeed, the Court remarked that this was the first time it had "specifically considered what it means for a factor to be overbroad when it is important only for selection purposes and especially when it sets forth victim vulnerability or victim impact evidence." Id. at 401, 119 S.Ct. 2090 (citation omitted) (internal quotation marks omitted).

While the selection stage does' not involve the narrowing function that is of primary importance at the eligibility stage, the Court identified another function relevant to both: protection "against bias or caprice in the sentencing decision" through the use of factors that "direct the jury to the individualized circumstances of the case." Id. at 402, 119 S.Ct. 2090 (internal quotation marks omitted). This is, of course, consistent with Tuilaepa's characterization of the selection stage, noted above, which Jones quoted. Id. Based on this functional understanding of non-statutory aggravators, the Court upheld the victim-centered (and scienter-less) aggravators against the overbreadth challenge— not because they narrowed the category of cases in which the death penalty applied, but "[b]ecause [they] directed the jury to the evidence specific to this case" and for that reason were not "overbroad in a way that offended the Constitution." Id. at 402 (emphasis added).

Under Jones, the mental-anguish aggravator in the present case was not overbroad. Regardless of its lack of a scienter requirement, it clearly directed the jury to focus on evidence specific to the particular victim and crime in this case, thereby satisfying the overarching function of a nonstatutory aggravator.

B. Congressional Preemption

Fields contends that Congress has "preempted" the use of the non-statutory aggravator for mental anguish through promulgation of the statutory aggravator for murder committed in a "heinous, cruel, or depraved manner." 18 U.S.C § 3592(c)(6). According to Fields, the mental-anguish aggravator is subsumed within the statutory aggravator, which demonstrates that Congress did not intend the jury to account for mental anguish during the selection stage.

We disagree. Even if we assume Fields is correct that the non-statutory mental anguish aggravator is subsumed within the statutory "heinous, cruel, or depraved manner" aggravator—an argument Fields has not supported with any relevant authority—it does not follow that the former reflects a deviation from the Congressional intent of the latter. Non-statutory aggravators, relevant only at the selection stage, may be similar or related to statutory aggravators, operative at the eligibility stage, without implicating preemption concerns, as each factor serves an independent purpose tied to the stage at which it functions:

[This] argument mistakenly assumes that non-statutory aggravating factors serve the same function as aggravating factors. In reality, statutory factors narrow the class of defendants eligible for the death penalty, whereas non-statutory factors serve the separate "individualizing" function that ensures the "jury [has] before it all possible relevant information about the individual defendant whose fate it must determine." [United States v.] Walker, 910 F.Supp. [837] at 855 [(N.D.N.Y.1995)] (quoting Jurek v. Texas, 428 U.S. 262, 276, 96 S.Ct. 2950, 49 L.Ed.2d 929 (1976)). There is no reason to believe that by choosing one factor for one purpose Congress excluded the use of related (and even broader) factors for a completely separate purpose.

United States v. Johnson, 1997 WL 534163, at *6 (N.D.Ill. Aug.20, 1997).

The mental-anguish aggravator properly focused the jury on the facts of this particular case—i.e., the anguish Shirley Chick felt seeing her husband Charles shot in the face right in front of her, and then being shot, chased down, and killed herself by an armed stranger cloaked in the freakish, de-humanizing ghillie suit displayed at trial. As discussed above in connection with Fields' overbreadth argument, this individualized assessment was sufficient under Jones to satisfy the purpose of a non-statutory aggravator at the selection stage of sentencing.

IX. NON-FAMILY VICTIM IMPACT

Fields contends that the district court permitted victim-impact evidence that went beyond the scope of permissible inquiry. Specifically, he objects to the introduction of matters relating to the impact of the victim's death not just on family but on "friends, co-workers and community," as the verdict form framed this non-statutory aggravator. In United States v. Barrett, 496 F.3d 1079 (10th Cir. 2007), we rejected such an objection regarding impact on a victim's friends, but we did not consider an extension of the inquiry to co-workers and community. Barrett took the matter as far as we are willing to go, at least in the absence of more specific guidance from the Supreme Court. While the court formalistically exceeded that limit here, in substance any error in this respect was harmless.^[12]

A. Impact on Friends

Barrett involved a challenge to victim-impact testimony from two friends of the victim regarding the effect his life and death had on them. The defendant objected to this testimony, arguing "that Congress has expressly limited impact evidence in federal death penalty cases to evidence concerning the effect of the offense on the victim and the victim's family." 496 F.3d at 1098. We recognized the FDPA's specific reference to "the effect of the offense on" and "loss suffered by" "the victim's family." Id. at 1098-99 (quoting 18 U.S.C. § 3593(a)). But, citing (1) the statute's additional inclusive reference to "any other relevant information," id. at 1099 (same), and (2) case law permitting evidence "giving the jury a glimpse of the victim's personality and the life he led," id., we rejected the defendant's rigid view of victim-impact evidence. We approved the challenged evidence, including testimony by "a longtime friend of [the victim], describing . . . the impact [the victim] had on his life," and by "a fellow Oklahoma Highway Patrol trooper, . . . describing the impact [the victim's] death had on him." Id. Similar evidence from friends in this case was likewise admissible.

B. Impact on Co-Workers and the Community

Barrett did not involve an inquiry into the impact of the victim's death on the community at large. Indeed, we are not aware of any precedent approving such an inquiry. The same is true regarding impact on the victim's co-workers. While one of the friends who testified in Barrett had worked with the victim, the defendant's challenge was framed solely in terms of "testimony from friends of the victim," id. at 1098, impact on "co-workers" was not cited in the aggravator presented to the jury, id. at 1087, and we did not invoke the friend's status as a co-worker in our analysis approving admission of the testimony, id. at 1099.

1. Impact on co-workers

Barrett supports the validity of victim-impact evidence regarding friends who (also) worked with the victim. But as to co-workers per se, a victim-impact inquiry would have a qualitatively different character. The loss of a co-worker in this sense (for example, the loss of her contribution to an office, unit, or team; the kind of loss that businesses insure with "key man" policies) is very far afield from the personal loss discussed in cases following the Supreme Court's initial approval of victim-impact evidence from family members in Payne v. Tennessee, 501 U.S. 808, 111 S.Ct. 2597, 115 L.Ed.2d 720 (1991). Without additional guidance from the Court encouraging further expansion of the victim-impact inquiry, we are not willing to extend it to the impersonal utilitarian considerations included within the idea of a loss to co-workers.

But any error in this regard was purely formalistic. The sole co-worker to testify, Charles Miliara, was also Charles Chick's close friend. He only briefly alluded to their work together, once to note their first meeting (a story illustrating Chick's easy-going manner in the face of an apprehensive and testy Miliara), R. Vol. 15 at 2607-08, and once to point up an irony in another personal story (involving the two men, whose jobs entailed creating instructions for assembling fighter jets, trying to assemble a train set for Miliara's son on Christmas without instructions), id. at 2620-21. In substance, Miliara's testimony concerned the loss of a friend and was not categorically different from that allowed in Barrett. Nor did the government argue' the matter differently to the jury. While the concept of victim impact should not have extended to the category of co-workers per se, that category was in fact empty and the error was harmless.

2. Impact on the community

Including the community in the victim-impact inquiry is fraught with complication. It would involve not just the incremental extension from family to friends (and even to co-workers), but a radical change in perspective: replacing a close-in focus on persons closely or immediately connected to the victim with a wide view encompassing generalized notions of social value and loss. Even if justified in principle, such an approach would be difficult to delimit and police to ensure it stayed within proper bounds. And there is no guidance to be gleaned from the case law, which is still on the first step of extending the victim-impact inquiry from family to friends. Lacking clear direction from the Supreme Court, we do not approve further expansion of the inquiry to the community.

But, once again, any error here is essentially academic in light of the evidence actually adduced in this case. There is nothing in the victim-impact testimony addressing community impact directly, nor did the government press the point in closing argument to the jury. As far as one can tell from the testimony of the Chicks' family and friends, everything this loving and generous—but decidedly unconventional and individualistic—couple, did for, and meant to, others was direct and personal. And in that respect the aggravator properly presented a powerful consideration in the weighing process, supported by moving testimony demonstrating the unique value of these particular people. In short, any error involved in formally framing the victim-impact aggravator to include community-level consequences was harmless given the lack of evidence directly relating to that improper consideration and the otherwise weighty considerations the aggravator properly put before the jury.

X. FAILURE OF ANY JUROR TO FIND SEVERE-DISTURBANCE MITIGATOR

Fields complains that no juror found that he committed the murders under severe mental or emotional disturbance. The government insists he does not raise a cognizable issue for our review. We agree.

A. Non-reviewability of Mitigator Findings under FDPA

The now-repealed capital scheme for continuing criminal enterprise (CCE) offenses expressly sanctioned review of mitigator findings by directing the court of appeals to remand if the record did not support either a jury's finding of an aggravator or its "failure to find[] any mitigating factors." 21 U.S.C. § 848(q)(3). In contrast, the FDPA directs the court to remand if the record fails to support "the existence of the required aggravator," i.e., the statutory aggravator(s) making the defendant eligible for the death penalty, and does not mention mitigating factors. 18 U.S.C. § 3595(c)(2). This clearly indicates that the FDPA does not contemplate evidentiary review of mitigator findings. Moreover, the government cites decisions from two circuits holding that, so long as the jury is not precluded from considering evidence relating to a mitigator, its determination of the matter is not subject to challenge on evidentiary grounds. See United States v. Higgs, 353 F.3d 281, 327 (4th Cir.2003); United States v. Paul, 217 F.3d 989, 999-1000 (8th Cir.2000). Fields has not cited any contrary authority.^[13]

He does point out that Higgs and Paul alternatively reached the merits to affirm the jury's rejection of the mitigatory in question, Higgs, 353 F.3d at 327; Paul, 217 F.3d at 1000, and notes that the Fifth Circuit (while expressing doubts on the matter) "[a]ssum[ed], arguendo, that [it] possessed] the authority to review the juror's special findings regarding mitigating factors" in United States v. Hall, 152 F.3d 381, 413 (5th Cir.1998), abrogated on other grounds by United States v. Martinez-Salazar 528 U.S. 304, 120 S.Ct. 774, 145 L.Ed.2d 792 (2000). These illustrations of cautious judicial practice do not substantively undermine the operative holdings in Higgs and Paul, or negate the FDPA's telling silence (particularly compared to the CCE scheme) as to review of mitigator findings. We agree with Higgs and Paul that a jury's failure to find a particular mitigator is not subject to challenge on evidentiary grounds in cases governed by the FDPA.

Indeed, the peculiar challenge advanced by Fields would not be cognizable even if we held that mitigator findings were subject to review for evidentiary sufficiency' in the traditional sense. Fields contends that at least one juror should have found he committed the offenses while under a severe mental or emotional disturbance. Aplt. Opening Br. at 100-02. He does not argue (with one possible exception in his reply brief) that the evidence was such that no reasonable juror could have rejected the mitigator; if that were his argument, his position would necessarily be that he was entitled to a unanimous finding on the mitigator. He makes the very different claim that his evidence was merely strong enough that some juror(s) should have found in his favor.

None of the cases we have discussed— which ultimately did review the merits of the appellant's challenge to the mitigator findings, under the CCE (McCullah), on an "arguendo" basis (Hall), or as an alternative rationale for disposition (Higgs and Paul)—even potentially recognize this kind of objection. Rather, they review the findings in the objective way challenges for evidentiary sufficiency are traditionally considered in the criminal law: was the evidence such that a reasonable trier of fact (and hence each and every juror) would be compelled to find the matter in the defendant's favor. See McCullah, 76 F.3d at 1112-13; Higgs, 353 F.3d at 327; Paul, 217 F.3d at 999; Hall, 152 F.3d at 413. Fields' attempt to frame an argument for reversal based on a lesser evidentiary claim is not authorized by any authority he has cited or that we have found. It is akin to seeking reversal on the legally fanciful basis that a defense too weak to render the prosecution's case insufficient should have at least convinced one juror to hang the jury.^[14]

At one point in his reply brief, Fields does characterize his argument as "the converse of an argument of insufficient evidence to support a conviction" and compares his situation to one where a properly instructed jury fails to convict "even when it can be said that no rational trier of fact could find guilt beyond a reasonable doubt." Aplt. Reply Br. at 54 (internal quotation marks omitted). This might suggest a recasting of his position to bring it in line with traditional principles. Yet the peculiar legal hedge clearly remains as he continues to assert only that some juror(s) should have ruled in his favor, see id. at 52, 56. And, of course, even if he did recast his objection in cognizable form, we would not be obliged to consider his belatedly adopted position. See United States v. Hall, 473 F.3d 1295, 1301 n. 1 (10th Cir.2007) (declining to address new argument on issue in reply brief).

XI. REASONABLE-DOUBT STANDARD IN WEIGHING PROCESS

Consistent with 18 U.S.C § 3593(e), the jury was instructed that it could impose the death penalty if it determined "that the aggravating factor or factors sufficiently outweigh the mitigating factor or factors to justify a sentence of death." R. Vol. 22 at 3404. Fields insists the jury should have been instructed that it was necessary to make this determination beyond reasonable doubt. This issue is controlled by our decision in Barrett, which followed the Fifth Circuit in holding that a reasonable-doubt standard is not required in the weighing process. Barrett, 496 F.3d at 1107-08 (following United States v. Fields, 483 F.3d 313, 346 (5th Cir.2007)). While Barrett involved the death penalty scheme for continuing criminal enterprises and drug activities in 21 U.S.C. § 848, that scheme was identical to the FDPA in this and most respects, see id. § 848(k). The objection raised here and in Barrett is constitutionally based, and its rejection in Barrett obviously applies with equal force in connection with the FDPA-indeed, the Fifth Circuit precedent Barrett explicitly followed was an FDPA case.

XII. GHILLIE SUIT IN THE JURY ROOM

The ghillie suit Fields used to conceal himself when stalking his victims was admitted into evidence and displayed briefly during trial. Over defense objection, the district court also allowed the suit in the jury room during deliberations. Fields does not challenge the admission of the suit, but does argue that it was improper to send it into the jury room. We discern no error.

While neither party has referred us to any cases similar in specifics, the broad parameters of our review are well-settled. "[I]t is within the discretion of the trial judge to decide what exhibits are permitted in the jury room" and "[t]his court will not overturn the trial court's exercise of discretion absent a clear showing of abuse and resulting prejudice." United States v. Hines, 696 F.2d 722, 734 (10th Cir.1982) (internal quotation marks omitted). Indeed, "the sending of tangible exhibits to the jury room is today so well established as to be practically irreversible." 2 McCormick on Evidence § 220 (6th ed.2006).

Fields cites the limited display of the ghillie suit permitted during trial and the district court's initial reluctance to allow it in the jury room as indications of the prejudice involved, and he insists that the disturbing nature of the suit likely prompted an adverse irrational emotional response from the jury. In response, the government argues that the appearance of the suit is probative in the context of this case. We agree with the government. The suit's camouflaging effect supported the substantial-planning aggravator, while its disturbing appearance supported the mental-anguish aggravator with respect to the murder of Shirley Chick, who was chased down by the ghillie-suited Fields.

Fields also argues that it was unnecessary to allow the suit in the jury room when a photograph would have sufficed. Whether a photograph would have been an adequate substitute is doubtful, especially in relation to the mental-anguish issue. In any event, as the government notes, the defense never asked the district court to consider this compromise, so it is a particularly inappropriate basis for saying the court abused its discretion in its handling of the matter.

The district court thoughtfully exercised its informed judgment in allowing the jury to examine firsthand an admitted exhibit of substantial significance to at least two disputed aggravating circumstances during its deliberations. This was not an abuse of discretion.

XIII. CUMULATIVE ERROR

Fields contends that even if the errors he asserts do not individually require reversal, their aggregate effect nevertheless mandates relief under the principle of cumulative error. "A cumulative-error analysis aggregates all errors found to be harmless and analyzes whether their cumulative effect on the outcome of the trial is such that collectively they can no longer be determined to be harmless." United States v. Toles, 297 F.3d 959, 972 (10th Cir.2002) (internal quotation marks omitted).

In this case, we have determined that the District Court erred in including co-workers and the community in the victim-impact inquiry, but we denied relief on the basis that the error was harmless. We also concluded that Fields had not demonstrated he was entitled to relief under plain-error review for any purported error in the compound phrasing of the future-dangerousness aggravator. It is unclear how to undertake a cumulative-error analysis where, as here, at least one purported "error" was rejected under plain-error review. See Barrett, 496 F.3d at 1121 n. 20. Indeed, there are inherent problems in cumulating issues rejected on plain-error grounds with matters rejected on harmless-error grounds.^[15] Nonetheless, even assuming we may cumulate the two issues, reversal is not warranted. The reference to co-workers and the community, unsupported by actual testimony and not referred to by the prosecution, could have played only a de minimis role in the jury's weighing of the aggravating and mitigating evidence. And to the extent the jury could not have considered Fields' future dangerousness (as it may not have unanimously determined the circumstances supporting such a finding), we do not think that "error" in conjunction with the victim-impact error tipped the scales against Fields. Upon review of the entire record, we conclude the collective impact was harmless.

The judgment of the district court is AFFIRMED.

NOTES

[1] Pub.L. No. 103-322, §§ 600001-600026, 108 Stat. 1796, 1959 (codified as amended in scattered Sections of 18 U.S.C.).

[2] Decisions of the Oklahoma Court of Criminal Appeals are binding as the highest authority on matters of state criminal law, Cummings v. Evans, 161 F.3d 610, 615 (10th Cir. 1998); United States v. Hines, 696 F.2d 722, 725 & n. 1 (10th Cir.1982), though two points diminish the authority of Cline for us here. First, its actual holding involved only the existence of state jurisdiction (to prosecute a breach of the peace), not the existence or extent of concurrent federal jurisdiction, Cline, 322 P.2d at 211, 216. Its statements in the latter regard are, as dicta, not binding on us. See DP-Tek, Inc. v. AT & T Global Info. Solutions Co., 100 F.3d 828, 831 (10th Cir. 1996) (placing dicta of state's highest court on same level as decisions of state's lower courts). Second, its comments on the Cession Act are at times entangled with its views regarding matters of federal law on which a state court decision cannot be authoritative. Even with these caveats, Cline's analysis of the Cession Act carries much weight here.

[3] The Supreme Court has not spoken to this precise issue. It has stated that through § 480 "Congress in effect has declined to accept exclusive legislative jurisdiction over forest reserve lands," Wilson v. Cook, 327 U.S. 474, 487, 66 S.Ct. 663, 90 L.Ed. 793 (1946), and hence "the States retain civil and criminal jurisdiction over the national forests," United States v. County of Fresno, 429 U.S. 452, 455, 97 S.Ct. 699, 50 L.Ed.2d 683 (1977). Neither of those statements is contrary to the concurrent jurisdiction recognized in the cases discussed above.

[4] See Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843-44, 104 S.Ct, 2778, 81 L.Ed.2d 694 (1984).

[5] We do not mean to imply agreement with Fields' initial premise that, in light of § 480, a state must retain concurrent jurisdiction over national forest land. Our analysis does not turn on this point because, as we have already held, that is what Oklahoma did here. We note that § 480 states only that the creation of a national forest does not in itself extinguish state jurisdiction, not that a state is precluded from voluntarily relinquishing its jurisdiction by ceding exclusive jurisdiction to the United States.

[6] These legislative changes to the existing common law were prospective only. See, e.g., United States v. Cassidy, 571 F.2d 534, 536-37 (10th Cir.1978); United States v. Redstone, 488 F.2d 300, 302 (8th Cir.1973) (citing cases).

[7] The seminal cases with respect to qualifying potential jurors for sentencing in death penalty cases are Witherspoon v. Illinois, 391 U.S. 510, 88 S.Ct. 1770, 20 L.Ed.2d 776 (1968), and Wainwright v. Witt, 469 U.S. 412, 105 S.Ct. 844, 83 L.Ed.2d 841 (1985).

[8] Fields argues that the mere existence of this exception to the juror's death penalty reservations is sufficient to preclude excusal, quoting Coleman v. Brown, 802 F.2d 1227 1232 n. 3 (10th Cir.1986), which approved the excusal of a juror unconditionally opposed to the death penalty but said: "Had the juror indicated that he could impose the death penalty in an exceptional case, the trial judge could have found the juror qualified under the Witt standard." Fields' reliance on Brown is unavailing: The statement he quotes from Brown is dicta and, at least as he reads it, is at odds with Uttecht, which held that jurors must be able to impose the death penalty under the scheme set by law. 127 S.Ct. at 2224, Uttecht in fact approved the excusal of a juror who admitted a narrow circumstantial acceptance of the death penalty. Id. at 2226.

[9] Fields objects that this direct inquiry framed in the terms of the governing standard constituted an improper delegation of the court's responsibility to decide the issue. We disagree. There is nothing to support Fields' speculation that the court let the juror decide for himself whether he was qualified to sit on the case; on the contrary, it is clear that the court made its own thorough determination, based on the whole record.

[10] We note that Fields' briefing on invited error completely glosses over this critical unanimity question. While he invokes unanimity in general fashion to avoid the invited-error bar, nowhere does he attempt to explain how the error here actually implicates unanimity. He simply states his objection that the jury imposed a unitary sentence rather than a sentence for each count—i.e., that he got the wrong unanimous sentence—and then proceeds to discuss how unanimity problems cannot be invited error under Teague. See Aplt. Reply Br. at 23-24.

[11] For Walton, see Ariz.Rev.Stat. §§ 13-703, 13-703.1, and for Wade, see Cal.Penal Code §§ 190.2, 190.4.

[12] The parties dispute whether a pretrial defense motion in limine objecting to non-family evidence preserved the matter for review (if not, plain-error rather than harmless-error standards apply). The dispute concerns whether the district judge definitively denied the motion or left open the possibility of changing his mind—if the former, further objection was unnecessary; if the latter, renewal of the objection when the evidence was offered at trial would have been required. See Fed.R.Evid. 103(a); United States v. Nichols, 169 F.3d 1255, 1264 (10th Cir.1999). Given that, as we explain above, there was no potential for improper prejudice arising from the evidence in question, relief is not available to Fields even if we apply the more favorable harmless-error standard.

[13] Fields relies on our holding in McCullah that "if the evidence does not support a juror's failure to find a mitigating factor, the resulting death sentence is unsound." 76 F.3d at 1112. But McCullah was decided under § 848(q)(3) and hence is irrelevant here.

[14] This characterization actually suggests a different objection that is cognizable in criminal proceedings, but one that has never been made here. A defendant may move under Fed.R.Crim.P. 33(a) for a new trial under the catch-all "interest of justice" rubric on the ground that a conviction, though supported with legally sufficient evidence, is nevertheless against the weight of the evidence. That, however, is a matter that involves assessing credibility and weighing evidence and is, accordingly, committed to the trial court's discretion. See United States v. Gabaldon, 91 F.3d 91, 93-94 (10th Cir.1996). In light of that, an appellate court "will not review a judgment [for weight of the evidence] where the trial court had no chance to make such a determination" due to the defendant's failure to file a Rule 33 motion. United States v. Hernandez, 227 F.3d 686, 695 (6th Cir.2000); see, e.g., United States v. Lopez, 576 F.2d 840, 845 (10th Cir.1978). Fields did not file such a motion.

[15] There is a fundamental distinction between plain-error and harmless-error review. If a preserved error is sufficiently prejudicial, reversal on harmless-error review is mandatory. And that is consonant with cumulative-error review, which similarly mandates relief if sufficient cumulative prejudice is evident. United States v. Harlow, 444 F.3d 1255, 1269 (10th Cir.2006) ("Unless an aggregate harmlessness determination can be made, collective error will mandate reversal, just as surely as will individual error that cannot be considered harmless." (internal quotation marks omitted)). But reversal for unpreserved error on plain-error review "is discretionary, not mandatory." United States v. Hernandez-Guevara, 162 F.3d 863, 870 (5th Cir.1998); United States v. Li, 115 F.3d 125, 129 (2d Cir.1997); see United States v. Gordon, 480 F.3d 1205, 1212 (10th Cir.2007). Injecting issues rejected on plain-error grounds into cumulative harmless-error review would allow a party to argue for mandatory reversal based on matters that, having been forfeited at trial, should be left to the discretion of the appellate court guided by the four-part Olano test. Additionally, the party who carries the burden also differs according to the review invoked. Compare United States v. Calzada-Maravillas, 443 F.3d 1301, 1306 (10th Cir. 2006) (the government bears the burden of showing that preserved errors are harmless) with United States Moreno-Trevino, 432 F.3d 1181, 1188 (10th Cir.2005).