United States v. Fell

571 F.3d 264 | 2d Cir. | 2009

571 F.3d 264 (2009)

UNITED STATES of America, Appellee,
v.
Donald FELL, Defendant-Appellant.

No. 06-2882-cr.

United States Court of Appeals, Second Circuit.

June 17, 2009.

William B. Darrow, Assistant United States Attorney, Burlington VT, for Appellee.

John Blume, Cornell Law School (Christopher Seeds, Sheri Lynn Johnson, on the brief), Ithaca, NY; Alexander Bunin, Federal Public Defender, Albany, NY, for Appellant.

ORDER

Defendant-Appellant Donald Fell, having filed a petition for panel rehearing or, in the alternative, for rehearing en banc, and the panel that determined the appeal having considered the request for panel rehearing, and the active members^[1] of the Court having considered the request for rehearing en banc, IT IS HEREBY ORDERED that the petition is DENIED. See Fed. R.App. P. 35(a).

Pursuant to Second Circuit Local Rule 0.28(7)(d), an automatic stay of execution of the sentence of death has been in place as of the date of the filing of the notice of appeal from the judgment of conviction, and remains in effect (unless vacated or modified) until the expiration of all proceedings available to the Defendant-Appellant (including review by the United States Supreme Court) as part of the direct review of the judgment of conviction.

Accordingly, the issuance of the mandate is held until the expiration of all proceedings available to the Defendant-Appellant (including review by the United States Supreme Court) as part of the direct review of the judgment of conviction.

With this Order, Judge RAGGI is filing a concurring opinion, in which Chief Judge JACOBS and Judges CABRANES, B.D. PARKER, WESLEY, and LIVINGSTON join; Judge CALABRESI is filing a dissenting opinion; Judge POOLER is filing a dissenting opinion; and Judge SACK is filing a dissenting opinion.

REENA RAGGI, Circuit Judge, with whom Chief Judge JACOBS, Judge CABRANES, Judge PARKER, Judge WESLEY, and Judge LIVINGSTON join, concurring:^[1]

On November 26, 2000, in Rutland, Vermont, Donald Fell and Robert Lee viciously stabbed to death Fell's mother Debra and her companion Charles Conway. Early the next morning, the killers went to a local mall where they kidnapped Teresca King, a 53-year old convenience store clerk arriving for work, and stole her car to make their escape from the state. After driving several hours and crossing into New York State, Fell and Lee stopped in a wooded area where they forced King out of the car and brutally beat her to death.

Of the three murders committed by Fell on November 26-27, 2000—two in Vermont and one in New York—only the New York murder qualified as a capital crime under federal law because it originated with an interstate kidnapping and carjacking. See 18 U.S.C. §§ 1201(a)(1) (capital kidnapping), 2119(3) (capital carjacking). In short, Teresca King's murder was no local crime. It implicated real and significant federal interests because it was achieved by transporting the victim across state lines.^[2] Accordingly, Fell was indicted by a federal grand jury sitting in the District of Vermont for the capital crimes of kidnapping and carjacking resulting in Teresca King's death in New York. In June 2005, a federal petit jury in Vermont found Fell guilty of these crimes and, in July, the same jury unanimously voted that he should be sentenced to death.

In a detailed opinion, a panel of this court rejected Fell's challenge to that sentence and affirmed the judgment of conviction. See United States v. Fell, 531 F.3d 197 (2d Cir.2008). Today, the court denies en banc review of this appeal. I join in that decision, and I write now only to respond to certain points raised by my colleague Judge Calabresi in his dissent from the denial of rehearing en banc.

At the outset, I note my agreement with the dissent's characterization of the trial court's conduct in this difficult case as "nothing short of exemplary" and of the panel opinion as "exceedingly careful" in its discussion of the various sentencing challenges raised by Fell. Post at 283. Similarly, I agree that Fell's specific claims of error in the district court's (1) removal for cause of Juror 64, and (2) refusal to admit a draft plea agreement at the penalty phase of the trial, fail under the "traditional rules" established by Supreme Court precedent. Post at 283, 286. Where I cannot agree with the dissent is in its suggestion that en banc review is needed to consider the possibility that something more than these traditional rules is necessary to address "federalism" concerns not raised by Fell either in the district court or on direct appeal: specifically, (1) whether a district court selecting a federal capital jury in a state—such as Vermont—that does not itself provide for the death penalty, must somehow take that fact into account in deciding whether to excuse jurors who express opposition to the death penalty; and (2) whether, in mitigation of sentence, a jury must be allowed to hear that the United States Attorney in the venue state had, at one point, been willing to enter into a plea agreement that provided for a non-capital disposition of the case. I respectfully submit that these "federalism" concerns are more imaginary than real and do not warrant our en banc consideration.

I. The Dissent's Jury Selection Concern

The Sixth Amendment provides that "[i]n all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury of the State and district wherein the crime shall have been committed." U.S. Const, amend VI. In Fell's case, there is no question that each and every juror who voted to impose the death penalty was a resident of the State and District of Vermont. Nor is there any question as to the impartiality of these jurors.

The dissent nevertheless urges us to consider en banc whether federalism principles require us to construe the Sixth Amendment vicinage requirement to demand not only that a federal criminal jury be drawn from the relevant state and district, but also that its members somehow represent the local "values" of that vicinage. Specifically, the dissent interprets Vermont's lack of death penalty legislation as evidence that "presumably a large portion of the population . . . is opposed to the death penalty." Post at 283. It submits that a proper respect for federalism might require a federal judge, in selecting a capital jury in such a state, to be "attuned to whether the jury members (and not just the jury pool . . .)—though willing to follow the law—are also representative of a state's overall opposition to the death penalty." Post at 284. The dissent suggests that we consider en banc whether to remand this case "to ask the able District Judge whether, in striking Juror 64, he fully considered the constitutional relevance of the values of Vermonters, the values of the jurisdiction in which he sat." Post at 285. Respectfully, I think no such en banc review, much less such a remand, is warranted in this case.

A. Juror 64's Voir Dire Responses Made It Appropriate To Remove Her Regardless of Vicinage

The above-quoted issue that the dissent proposes for en banc review relies on a critical assumption: that the juror under consideration, though opposed to the death penalty, was nevertheless "willing to follow the law." Post at 284. If that were true, we would hardly need to convene en banc to address the dissent's federalism theory because the Supreme Court has already made clear that the removal of such a juror constitutes reversible error. See Gray v. Mississippi, 481 U.S. 648, 668, 107 S. Ct. 2045, 95 L. Ed. 2d 622 (1987). In Witherspoon v. Illinois, 391 U.S. 510, 88 S. Ct. 1770, 20 L. Ed. 2d 776 (1968), the Court ruled that opposition to the death penalty is not enough, by itself, to support a prospective juror's removal for cause. See id. at 522, 88 S. Ct. 1770. Thereafter, in Wainwright v. Witt, 469 U.S. 412, 105 S. Ct. 844, 83 L. Ed. 2d 841 (1985), the Court held that, in a capital case, removal for cause based on a prospective juror's views about the death penalty is warranted only where the court forms "a definite impression" that "the juror's views would prevent or substantially impair" the juror's "performance of his duties . . . in accordance with his instructions and his oath." Id at 424-26, 105 S. Ct. 844 (internal quotation marks omitted); see United States v. Fell, 531 F.3d at 210 (discussing Supreme Court precedent regarding capital jury selection).

This case, however, does not fall within the dissent's paradigm. The juror referred to by the dissent—Juror 64—failed to demonstrate under Witt-Witherspoon and their progeny that she was willing to follow the law despite her personal objection to the death penalty. As the Supreme Court explained in Lockhart v. McCree, 476 U.S. 162, 106 S. Ct. 1758, 90 L. Ed. 2d 137 (1986), "those who firmly believe that the death penalty is unjust may nevertheless serve as jurors in capital cases so long as they state clearly that they are willing to temporarily set aside their own beliefs in deference to the rule of law." Id. at 176, 106 S. Ct. 1758 (emphasis added). The panel opinion demonstrates that Juror 64 made no such clear statement: "[t]hroughout the district court's painstaking and thoughtful voir dire, Juror 64 walked a fine line between her opposition to the death penalty and her willingness to follow the district court's instructions." United States v. Fell, 531 F.3d at 213. This is not to fault the juror. A juror's experiences, beliefs, and values may sometimes make it difficult for the juror to know, much less to state clearly or confidently, whether she will be able to set aside her own beliefs in deference to the rule of law. In such circumstances, a district court appropriately relies on its voir dire experience and its unique ability to observe the prospective juror during questioning to assess whether the juror will be able faithfully and impartially to apply the law. See Wainwright v. Witt, 469 U.S. at 426, 105 S. Ct. 844; see also United States v. Quinones, 511 F.3d 289, 301-02 (2d Cir.2007) (discussing Supreme Court cases according deference to district court's assessment of partiality where prospective juror's voir dire responses are ambiguous: "the more ambiguous a prospective juror's responses, the more useful demeanor, and thus oral inquiry, become in allowing a trial judge to identify partiality warranting removal for cause").^[3]

The able district judge, after extensive questioning, formed a definite impression that Juror 64 could not satisfy the Witt-Witherspoon standard for impartiality. Moreover, the appellate panel, although limited in its review to the cold record, reached the same conclusion. See United States v. Fell, 531 F.3d at 213. The Supreme Court's Sixth Amendment jurisprudence would neither have permitted the prospective juror to be seated upon such a finding nor tolerated the juror's removal on any lesser finding.^[4]

Consequently, it makes no difference that, in this case, the juror's partiality manifested itself in a non-death penalty state such as Vermont rather than a state that authorizes the death penalty. The juror was properly removed regardless of the vicinage. Accordingly, the particular challenge at issue on this appeal warrants no en banc review.

B. The Ability of a Juror Opposed to the Death Penalty To Serve on a Capital Jury Depends on a Constitutional Rule that Does Not Vary with the Vicinage

1. The Selection of Federal Juries To Hear Cases Arising Under Federal Law Does Not Implicate Federalism

I am, in any event, skeptical of the dissent's suggestion that federalism requires each state's adoption or rejection of the death penalty somehow to be factored into the selection of federal capital juries serving therein. Federalism is a principle concerned with "the constitutional distribution of power as between the Nation and the States." Staub v. City of Baxley, 355 U.S. 313, 325-26, 78 S. Ct. 277, 2 L. Ed. 2d 302 (1958) (emphasis added); accord Printz v. United States, 521 U.S. 898, 918-21, 117 S. Ct. 2365, 138 L. Ed. 2d 914 (1997) (discussing federalism as concern for distribution of authority between the state and federal governments); see also, e.g., Edward Rubin, Judicial Review and the Right To Resist, 97 Geo. L.J. 61, 118 n. 154 (2008) (noting that federalism is the "division of political authority between the state and federal governments"); Michael Stokes Paulsen, A Government of Adequate Powers, 31 Harv. J.L. & Pub. Pol'y 991, 992 (2008) ("Federalism, properly understood, is a descriptive term attached to the Constitution's allocation of powers."); Bradford R. Clark, Translating Federalism: A Structural Approach, 66 Geo. Wash. L.Rev. 1161, 1161 (1998) ("Federalism. . . refers to the Constitution's division of powers between the federal government and the states."); Larry Kramer, Understanding Federalism, 47 Vand. L.Rev. 1485, 1561 n. 5 (1994) (noting that federalism concerns "how authority is distributed between the political institutions of state and federal governments"). Federalism ensures "a proper respect for state functions" by limiting the exercise of federal authority when it "unduly interfere[s] with the legitimate activities of the States." Younger v. Harris, 401 U.S. 37, 44, 91 S. Ct. 746, 27 L. Ed. 2d 669 (1971) (discussing "[o]ur federalism" in context of abstention); accord Madeira v. Affordable Housing Found., Inc., 469 F.3d 219, 237 (2d Cir.2006) (discussing federalism in context of preemption).^[5]

The selection of a federal jury to hear a case arising under federal law involves the exercise of exclusive federal power. It does not intrude on any state function; much less does it trench on the exercise of any state power. It poses no interference with legitimate state activities. Thus, even though the states provide one geographic boundary for the vicinage requirement of the Sixth Amendment—with congressionally defined judicial districts providing the other boundary—no federalism concern warrants a construction of that requirement that reaches beyond geography. Certainly, nothing in the plain language of the Sixth Amendment indicates that its vicinage requirement reaches beyond simple geography to local ideology. Nor does the Amendment's history support such a theory. The demand for a constitutional amendment specifying a local vicinage for federal trials originated in recollections of Parliament's acts permitting American colonists to be brought to England for trial. See Drew L. Kershen, Vicinage, 29 Okla. L.Rev. 801, 807 (1976) (quoting grievance in Declaration of Independence faulting King "For transporting us beyond Seas to be tried for pretend offenses"); 3 J. STORY, COMMENTARIES ON THE CONSTITUTION OF THE UNITED STATES § 1775, p. 654 (photo, reprint 1999) (Boston, Hilliard, Gray & Co. 1833) (observing that object of Sixth Amendment's vicinage clause is to prevent "accused from being dragged to a trial in some distant state"); see also Williams v. Florida, 399 U.S. 78, 92-97, 90 S. Ct. 1893, 26 L. Ed. 2d 446 (1970) (discussing history of vicinage clause and particularly compromise that substituted congressionally determined judicial districts for counties, as had been characteristic of common law).

To be sure, the vicinage requirement, by defining the community from which a federal jury must be drawn, permits the jury to operate as the conscience of that community in judging criminal cases. See Kershen, Vicinage, 29 Okla. L.Rev. at 842-43; see generally A. DE TOCQUEVILLE, DEMOCRACY IN AMERICA 260 (H. Mansfield & D. Winthrop transls. and eds.2000) (observing that "the man who judges the criminal is really the master of society" (emphasis in original)). But this function of the vicinage requirement is satisfied by drawing a jury pool from a fair cross-section of the residents of the particular state and district. See generally Lockhart v. McCree, 476 U.S. at 173-74, 106 S. Ct. 1758 (holding that fair cross-section requirement applies to jury pool not petit jury). Once such a pool has been drawn, the Amendment's singular concern is with ensuring the petit jury's impartiality.

The dissent's suggestion that the Sixth Amendment, viewed through the prism of federalism, requires something more, i.e., a special solicitude for local values in the selection of a federal petit jury, would, I expect, appropriately be rejected out of hand if the local "value" revealed at jury selection were opposition to the sorts of civil rights, environmental, or gun trafficking requirements that are enforced through federal criminal law in ways not always mirrored in state legislation. The dissent disagrees, submitting that while juries must follow federal law, when "issues of judgment" arise jurors may exercise that judgment in light of "local values," even when hostile to the federal law at issue. See post at 291-92 n. 12. Respectfully, I think the dissent may be thinking of the process of jury deliberation, not jury selection.

In the former context, values, instincts, and predilections may inform jury judgment. The law does not necessarily approve or encourage such influences. Nevertheless, a proper respect for finality— not federalism—dictates that "[t]he mental processes of a deliberating juror with respect to the merits of the case at hand must remain largely beyond examination and second-guessing, shielded from scrutiny by the court as much as from the eyes and ears of the parties and the public." United States v. Thomas, 116 F.3d 606, 620 (2d Cir.1997); see Fed.R.Evid. 606(b) (precluding inquiry into the "mind or emotions" of deliberating juror).

This rule, however, does not reach beyond the deliberation room itself. Most important, it does not alter a trial court's Sixth Amendment obligation to select a jury that can fairly and impartially apply the relevant federal law. In other words, once the nation has duly enacted federal criminal law, neither the Sixth Amendment nor federalism confers an additional right on the citizens of states that choose not to enact similar local laws to have their opposition to the federal law given special consideration in the selection of a federal jury.

This is not to deny federalism a role in federal criminal law. See post at 292. Certainly, federalism informed the early republic's critical decision to reject a federal common law of crimes and to require the codification of federal criminal law. See United States v. Hudson & Goodwin, 11 U.S. (Cranch) 32 (1812); see generally Kathryn Preyer, Jurisdiction to Punish: Federal Authority, Federalism and the Common Law of Crimes in the Early Republic, 4 Law & Hist. Rev. 223 (1986). Even within such a statutory scheme, federalism is appropriately considered by Congress whenever it "criminalizes conduct already denounced as criminal by the States." United States v. Lopez, 514 U.S. 549, 561 n. 3, 115 S. Ct. 1624, 131 L. Ed. 2d 626 (1995) (observing that such legislation "effects a change in the sensitive relation between federal and state criminal jurisdiction" (internal quotation marks omitted)). To the extent federalism underlies the rule that Congress may "creat[e] offenses against the United States" only pursuant to its "delegated powers," id., federalism is also reasonably understood to play a part in a court's consideration of whether Congress has exceeded those powers, see, e.g., id. at 557-68, 115 S. Ct. 1624 (holding Gun-Free School Zones Act to exceed Congress's authority under the Commerce Clause).

But to play a part is not to appear in— much less to steal—every scene. Where, as here, there is no question as to Congress's constitutional authority (1) to proscribe interstate kidnappings and carjackings, see Perez v. United States, 402 U.S. 146, 150, 91 S. Ct. 1357, 28 L. Ed. 2d 686 (1971) (observing that 18 U.S.C. § 1201 was enacted pursuant to Congress's power to regulate "use of channels of interstate or foreign commerce which Congress deems are being misused"); United States v. Trupin, 117 F.3d 678, 685 n. 3 (2d Cir.1997) (collecting cases upholding § 2119 after Lopez), or (2) to prescribe a capital penalty when death results from such crimes, see generally United States v. Quinones, 313 F.3d 49, 69-70 (2d Cir.2002) (rejecting constitutional challenges to federal death penalty), no recognized theory of federalism supports the dissent's assertion that federal petit juries must be selected in light of the particular vicinage's support for or opposition to the federal law.

The conclusion that federalism does not accord local opposition to federal law any claim to special consideration in the selection of a capital jury finds support in Sparf v. United States, 156 U.S. 51, 15 S. Ct. 273, 39 L. Ed. 343 (1895), wherein the Supreme Court ruled that, while juries must determine the "facts as they find them to be from the evidence," it is their duty "to take the law from the court." Id. at 102, 15 S. Ct. 273. This effectively ratified a view that had gained momentum in the latter half of the Nineteenth Century, i.e., that juries should not themselves act as judges of the law. See generally United States v. Thomas, 116 F.3d at 614 (collecting authorities). After Sparf, the law continued to acknowledge that, in the deliberative process, a jury had the "power"—if not the "right"—to nullify law by returning verdicts seemingly contrary to a court's instructions. See id. at 614-15. But, as this court observed in Thomas, because "the jury's prerogative of lenity, introduces a slack into the enforcement of law, tempering its rigor by the mollifying influence of current ethical conventions, . . . the system of checks and balances embedded in the very structure of the American criminal trial . . . [gives rise to] a countervailing duty and authority of the judge to assure that jurors follow the law." Id. at 616 (internal citations and quotation marks omitted). In the selection of jurors in federal capital cases, Witt-Witherspoon and its progeny have established the appropriate constitutional balance between these two interests.

2. Constitutional Rules Concerning Capital Jury Selection Must Apply Equally Throughout the Nation

In selecting a federal jury in a capital case—and in reviewing that jury selection on appeal—courts must be ever mindful that what Witt-Witherspoon articulates is a constitutional rule for determining when jurors who oppose the death penalty may or may not be excused for cause consistent with the Sixth Amendment's guarantee of an impartial jury. Plainly, the Constitution must apply equally throughout the states. As Justice Story famously observed, "[t]he constitution of the United States was designed for the common and equal benefit of all the people of the United States." Martin v. Hunter's Lessee, 14 U.S. (1 Wheat.) 304, 348, 4 L. Ed. 97 (1816) (emphasis added). Thus, if a district court's decision to excuse a juror violates Witt-Witherspoon, it does so regardless of whether the voir dire occurred in a nondeath penalty state such as Vermont, a death penalty state such as Texas, or any of the other 48 states in the Union. Indeed, I am hard pressed to knew how we might explain to a capital defendant in Texas that he is entitled to any less rigorous voir dire of a potential juror who expresses opposition to the death penalty (because Texas law authorizes capital punishment) than we would insist on for a capital defendant in Vermont (because Vermont law proscribes capital punishment).

The dissent agrees that a capital jury voir dire cannot be less rigorous in one state than in another. See post at 291 n. 11. In the next sentence, it even suggests that its federalism argument is not to the contrary because "[r]ecognizing the effect of local values on what a potential juror's answer means is just that, no more and no less." Post at 291 n. 11. The dissent is careful never to specify just how a district court would go about "recognizing the effect of local values" in evaluating a juror's responses. Nevertheless, I suspect that what the dissent urges is a good deal "more" than the Witt-Witherspoon rule, the proper application of which is not challenged in this case. Otherwise there would be no need to call for a remand. Under the dissent's view of "federalism," in a non-death penalty state such as Vermont, it would not be enough to conclude, as the district court did in this case, that a person opposed to the death penalty could not serve because he failed to provide the assurance of impartiality discussed in Lockhart v. McCree, 476 U.S. at 176, 106 S. Ct. 1758. Why? Because the juror's opposition to the death penalty "was simply representative of the views of Vermont." Post at 291. This implies that federalism requires federal courts to show more tolerance for opposition to the death penalty when voiced by potential jurors in non-death penalty states such as Vermont. I cannot agree.

Capital defendants, regardless of the state in which they are tried, have an equal constitutional right not to have an opponent of the death penalty removed for cause from a capital jury absent a proper judicial determination that the juror could not faithfully and impartially apply the relevant law. But no defendant has a right—regardless of the "characteristics of the community where the trial is held," post at 285 — to have an opponent of the death penalty serve as a juror if he cannot clearly state a willingness "`to temporarily set aside [his] own beliefs in deference to the rule of law,'" United States v. Fell, 531 F.3d at 210 (quoting Lockhart v. McCree, 476 U.S. at 176, 106 S. Ct. 1758).

To the extent the dissent's invocation of federalism might be understood to suggest that the Witt-Witherspoon standard is insufficient, by itself, to satisfy the constitutional requirement that juries be "local," the point warrants little discussion. As already noted, see supra at 266, the Constitution's key locality requirement— that jurors be drawn from the same state and district wherein the charged crime was committed—was plainly met in this case. The dissent's federalism concern might therefore be better understood as a fair cross-section challenge, i.e., that the removal of certain jurors deprived Fell of a fair cross-section of Vermonters opposed to the death penalty. See Taylor v. Louisiana, 419 U.S. 522, 530, 95 S. Ct. 692, 42 L. Ed. 2d 690 (1975) ("[T]he fair-cross-section requirement [i]s fundamental to the jury trial guaranteed by the Sixth Amendment."). As noted supra at 269-70, however, the law is clear that the "fair cross-section" requirement applies to the jury pool, not the petit jury. Lockhart v. McCree, 476 U.S. at 174, 106 S. Ct. 1758. The dissent does not—and cannot—argue that the jury pool in this case did not represent a fair cross-section of Vermonters. With that vicinage requirement satisfied, it is the Sixth Amendment's guarantee of impartiality that gives every defendant—the Texan as well as the Vermonter—a right not to have jurors opposed to capital punishment removed for cause as long as they satisfy the Witt-Witherspoon standard.

The dissent nevertheless appears concerned that nationwide application of the Witt-Witherspoon standard is somehow insufficient to allow Vermont residents to exercise fully their capital sentencing discretion. I am not convinced that such a problem exists. If the dissent is correct that the lack of a state death penalty indicates that a majority of Vermonters do, in fact, oppose the death penalty, those opponents will presumably represent a larger percentage of the federal venire in that state than will their counterparts in states that have enacted death penalty legislation. Precisely because, under Witt-Witherspoon, a judge cannot remove a prospective juror for cause based solely on the juror's opposition to the death penalty, it logically follows that more opponents of the death penalty will likely serve on a federal capital jury in Vermont than in states with death penalties. Indeed, that conclusion appears to find support in the second part of the dissent's opinion, which asserts that juries in states that do not have local death penalties vote for capital punishment less frequently than juries in states that do provide for such punishment. See post at 289-90. Whether or not this is the case, Witt-Witherspoon assures a capital defendant in any state the equal right to have death-penalty opponents serve on the jury "so long as they state clearly that they are willing to temporarily set aside their own beliefs in deference to the rule of law." Lockhart v. McCree, 476 U.S. at 176, 106 S. Ct. 1758.^[6]

C. The Eighth Amendment Does Not Support the Dissent's Vicinage Argument

There appears to be a certain inconsistency in the dissent insofar as it argues both (1) that different voir dire rules should apply to jury selection in a non-death penalty state such as Vermont because, otherwise, federal capital juries will be ineffective conduits of local values, see post at 283-86; and (2) that juries are so effective at transmitting local values that they have made application of the federal death penalty unconstitutional because of its rarity, see post at 289-90. With respect to the second argument, the dissent asserts that "[b]ehind all this lies a still deeper constitutional question": whether this death penalty verdict is unconstitutionally "unusual" because it was imposed in Vermont, a non-death penalty state. Post at 289. Indeed, the dissent suggests that an affirmative answer to this question may be "necessary for . . . federalism to survive." Post at 291. This argument is unconvincing for several reasons.

1. Federalism Does Not Warrant Construing the Eighth Amendment Prohibition on Cruel and Unusual Punishments Differently Based on the Vicinage

The dissent's effort to import a federalism concern based on vicinage into the Eighth Amendment analysis of federal sentences is misguided. Although the Supreme Court has construed the Eighth Amendment to prohibit cruel and unusual punishment in accordance with "evolving standards of decency," Kennedy v. Louisiana, ___ U.S. ___, 128 S. Ct. 2641, 2649, 171 L. Ed. 2d 525 (2008), and considered "state practices" in identifying these standards, id. at 2650 (quoting Roper v. Simmons, 543 U.S. 551, 563, 125 S. Ct. 1183, 161 L. Ed. 2d 1 (2005)), it has done so only because in the aggregate such practices serve as a proxy for the "national consensus," id. at 2653 (noting that 45 states prohibited death penalty for child rape); see also Roper v. Simmons, 543 U.S. at 564-65, 125 S. Ct. 1183 (noting that 30 states prohibited death penalty for juveniles); Atkins v. Virginia, 536 U.S. 304, 314-15, 122 S. Ct. 2242, 153 L. Ed. 2d 335 (2002) (noting that 18 states exempted mentally retarded persons from death penalty, in addition to 12 that prohibited death penalty altogether). The Eighth Amendment, no less than other provisions of the Constitution, must apply equally throughout the states. Nothing in the Court's jurisprudence has ever suggested that federalism warrants re-tailoring the Eighth Amendment in each state—or each vicinage—to test federal death sentences by reference to local practices. In the absence of any supporting authority, it is no mere "formalism[ ]" for this court to decline to convene en banc to explore the dissent's novel theory of federalism. Post at 292-93; see post at 291-92.

2. The New York Capital Experience Does Not Raise Constitutional Concerns About the Death Sentence Returned Against Fell in Vermont

Even assuming that the relative rarity of federal death sentences in a single state or vicinage were cognizable under the Eighth Amendment, the dissent's argument rests almost entirely on federal capital experience in New York, not in Vermont. See post at 289-90 & n. 9 (noting that federal capital juries in New York have voted for the death penalty only once in 19 cases). The New York experience does not, however, support the dissent's "federalism" twist on the Eighth Amendment.^[7]

a. The Difficulty in Discerning Popular Values on Capital Punishment in New York

Like Vermont, New York does not currently use the death penalty to punish state crimes. This is the result of a 2004 decision by the New York Court of Appeals invalidating the state's death penalty statute in light of identified procedural deficiencies. See People v. LaValle, 3 N.Y.3d 88, 131, 783 N.Y.S.2d 485, 511, 817 N.E.2d 341 (2004). As even the dissent appears to recognize, we cannot assume that such a judicial invalidation of death penalty legislation reflects popular opposition to capital punishment. See post at 283 & n. 4.

The dissent nevertheless submits that the New York legislature's failure to reenact the death penalty since the LaValle decision "must be taken . . . as some evidence of the popular will, at least today." Post at 283 n. 4. This assertion, in fact, only highlights the difficulty in requiring federal judges to construe constitutional provisions by reference to sometimes shifting "local values." Since 1992, when this judge selected the first post-Furman federal capital jury in New York, see United States v. Pitera, 90-cr-0424 (RR) (E.D.N.Y.), the death penalty has been at the center of so many state political battles that its legal status in New York is necessarily an imperfect proxy for the "local values" of the state's citizenry. See, e.g., Joel Stashenko, Death Penalty Critics Doubt State Is Ready, Buff. News, Aug. 27, 1995, at A8 (noting that Governors Carey and Cuomo together vetoed 18 consecutive death penalty bills passed by the New York legislature from 1977-1994); James Dao, Death Penalty in New York Reinstated After 18 Years; Pataki Sees Justice Served, N.Y. Times, March 8, 1995, at 1 (discussing Governor Pataki's 1995 signature of legislation reenacting the state death penalty); see also People v. LaValle, 3 N.Y.3d at 131, 783 N.Y.S.2d 485, 817 N.E.2d 341 (invalidating 1995 death penalty law in 2004); Patrick D. Healy, Death Penalty Is Blocked by Democrats, N.Y. Times, April 13, 2005, at B1 (discussing legislative inaction following LaValle). Surely, a federal trial judge's Sixth Amendment obligations in selecting a capital jury in New York could not vary depending on how one interpreted state values on the death penalty in 1994, 1995, 2004, or 2005.

In any event, the dissent can only speculate that "local values" opposing the death penalty may have informed the decision of New York federal juries to vote for the death penalty in only a single case, United States v. Wilson, 493 F. Supp. 2d 537, 539 (E.D.N.Y.2007) (involving execution murders of two New York City police officers). Citing a press report that three jurors in another New York federal capital case voted for life without parole because the death penalty would not have been an option if the case had been tried in New York state court, the dissent suggests that "[p]erhaps that is the reason why New York juries have returned only one death sentence from all the federal capital prosecutions sought in New York." See post at 290 n. 9.

Such speculation is suspect for at least two reasons. First, as just noted, the state death penalty was available in New York from 1995 to 2004, i.e., for nine of the 17 years during which federal prosecutors sought the death penalty for federal crimes committed in the state. Thus, the dissent's speculation is inapplicable to federal juries sitting during that period. Second, during those nine years, New York state prosecutors sought the death penalty 18 times, and New York juries voted to impose it seven times. See Joseph Lentol et al., The Death Penalty in New York: To Examine the Future of Capital Punishment in New York State Before the Assembly Standing Comms. on Codes, Judiciary, & Correction 14, 32 (Dec. 15, 2004-Feb.11, 2005), available at http://assembly. state.ny.us/comm/Codes/20050403/ deathpenalty.pdf. This hardly indicates the sort of strong local resistance to capital punishment posited by the dissent.

b. The Dissent's Unsupported Assertion that New Yorkers' Sensitivity to Federalism Explains the Infrequency of Federal Capital Verdicts in that State

Confronted with the undeniable fact that New York state juries did impose the death penalty a number of times in the years before LaValle, the dissent attempts to recast its federalism argument. Rather than maintain that the New York federal capital experience demonstrates a local value of opposition to the death penalty (which federalism must respect), the dissent asserts that the willingness of New York juries "to impose death under their own law but not under federal law helps to underscore federalism as a likely independent variable in their deliberations." Post at 294. In short, the dissent submits that it is New Yorkers' desire to vindicate the "value" of "federalism . . . itself" that explains their willingness to impose the death penalty in state cases and their reluctance to do so in federal cases. Post at 294 (emphasis in original).

Even if the dissent could support this extraordinary hypothesis, it nowhere indicates how a state's willingness to impose the death penalty under its own laws demonstrates that doing so under federal law would be constitutionally cruel and unusual. No matter. The dissent's hypothesis is, in fact, not grounded in anything but imagination. No one who has ever participated in a criminal trial could possibly think that state and federal juries in New York, when contemplating imposition of a death sentence, have as their "likely" focus "federalism as [a] value in itself." Post at 294 (emphasis in original). Nor can the dissent support its theory by reiterating that three jurors in one New York federal capital case explained that they did not vote for the death penalty because they knew it was then unavailable under state law. See post at 290 n. 9, 14. The jurors' explanation for what informed their actions when New York did not have the death penalty hardly supports a hypothesis that, during the years when state law did provide for capital punishment, it was New Yorkers' sensitivity to federalism that "likely" prompted them to vote for the death penalty more often in state cases than in federal ones.

c. The Inadvisability of Speculation as to the Reasons for Capital Verdicts

The dissent's efforts to invent variations on its original federalism theme prompt me to sound a note of caution about making general assumptions—unsupported by record evidence or experience—as to why federal juries in New York have imposed the death penalty only once since Furman. Capital juries reject the death penalty for various and often complex reasons. One factor more likely than federalism to explain the New York capital experience is the character of victims. As capital defense counsel Kevin McNally, Director of the Federal Death Penalty Resource Counsel Project, observed in commenting on the federal capital verdict in United States v. Wilson, 493 F.Supp.2d at 539, "`[n]ot many New York [federal capital] cases have [had] innocent victims.'" Alan Feuer, An Aversion to the Death Penalty, but No Shortage of Cases, N.Y. Times, Mar. 10, 2008, at B1 (reporting that "many victims in New York [federal] capital cases are unsavory characters: drug dealers, mobsters, or members of street gangs— not the sort of people whose killers are likely to be punished with death"). Although the dissent cites this same article (presumably to support its own speculation as to why New York jurors have rejected the death penalty in every case except United States v. Wilson, see post at 290 n. 9), it dismisses the assessment quoted in the parenthetical as "remarkable and troubling speculation," post at 292. It does not, however, attempt to show that the New York Times' account inaccurately draws on the facts of New York's federal capital cases.

In fact, the dissent could not do so in light of the case histories of many New York federal capital cases revealing victims involved in criminal activity. See, e.g., United States v. Pitera, 90-cr-0424 (E.D.N.Y.) (declining to impose death sentence on defendant whose homicide victims were criminal associates in or competitors of defendant's continuing drug enterprise); United States v. Diaz, 94-cr-0328 (N.D.N.Y.) (declining to impose death sentence on defendants who murdered rival drug dealer); United States v. Matthews, 00-cr-0269 (N.D.N.Y.) (declining to impose death sentence on defendants who murdered rival drug dealer); United States v. Quinones, 00-cr-0761 (S.D.N.Y.) (declining to impose death sentence on defendants who murdered drug customer working as police informant); United States v. Williams, 00-cr-1008 (S.D.N.Y.) (declining to impose death sentence on defendants who murdered drug purchasers); United States v. Dixon, 01-cr-0389 (E.D.N.Y.) (declining to impose death sentence on defendant who murdered rival drug dealer working as police informant); United States v. McGriff, 04-cr-0966 (E.D.N.Y.) (declining to impose death sentence on defendant who murdered two rival drug dealers).^[8] In United States v. Wilson, however, the defendant's own counsel, experienced in capital litigation, specifically identified the good character of the police officer victims as a crucial factor in the federal jury's decision to impose the death penalty. See Alan Feuer, An Aversion to the Death Penalty, but No Shortage of Cases, N.Y. Times, Mar. 10, 2008, at B1 ("The victims in the Wilson case, two undercover police detectives, were crucial to the jury's decision, both [Wilson's defense counsel Ephraim] Savitt and Mr. McNally said. Crucial and unusual, they added, in that the detectives were perceived as `innocent' and `good.'").

Significantly, in many of the New York state cases in which juries voted for capital punishment, the victims could also be perceived of as "innocent." See, e.g., People v. Taylor, 9 N.Y.3d 129, 137-40, 848 N.Y.S.2d 554, 878 N.E.2d 969 (2007) (describing defendant's murder of five employees of fast-food restaurant in course of robbery); People v. LaValle, 3 N.Y.3d at 99-101, 783 N.Y.S.2d 485, 817 N.E.2d 341 (describing defendant's rape and killing of jogger); People v. Mateo, 2 N.Y.3d 383, 394-97, 779 N.Y.S.2d 399, 811 N.E.2d 1053 (2004) (describing defendant's killing of four persons while attempting to locate ex-girlfriend); People v. Cahill, 2 N.Y.3d 14, 35-37, 777 N.Y.S.2d 332, 809 N.E.2d 561 (2003) (describing defendant's beating and poisoning of wife); People v. Harris, 98 N.Y.2d 452, 471-72, 749 N.Y.S.2d 766, 779 N.E.2d 705 (2002) (describing defendant's murder of three persons in course of robbery); see also John Rather, Suffolk Jury Returns a Death Verdict for Supermarket Murder, N.Y. Times, Aug. 17, 2000, at B5 (describing defendant's murder of co-worker).

This is not to suggest that all capital verdicts, or even these particular verdicts, can necessarily be explained by reference solely to the victims' characters.^[9] It is simply to note that the role of character and the other factors discussed in footnote 9 have support in capital case records and the experience of capital counsel. By contrast, nothing supports the dissent's implausible assertion that federalism "in itself is the "likely" explanation for New York juries voting for the death penalty more often in state than federal cases.

For all these reasons, I submit that we cannot responsibly draw from the New York experience any inference that infrequent jury application of the death penalty in federal capital prosecutions in that state is a consequence of either local values generally opposing the death penalty or such a refined jury sensitivity to federalism as to limit death penalty verdicts largely to state prosecutions. Much less can we infer that a Vermont jury's death penalty verdict in this case — which certainly involved an innocent victim—somehow reflects a failure of federalism raising Sixth or Eighth Amendment concerns.

In sum, there simply is no federalism concern here warranting our en banc consideration of any local variance from the uniform, nationwide application of the Witt-Witherspoon standard for determining when opponents of the death penalty can be removed for cause from capital juries.

II. The Dissent's Plea Agreement Concern

The same conclusion obtains with respect to the dissent's suggestion that we convene en banc to review the district court's evidentiary ruling on the inadmissibility at the sentencing proceeding of an unexecuted plea agreement between Fell and the Vermont U.S. Attorney's Office. The district court allowed the jury to hear that Fell had offered to plead guilty to the capital kidnapping charge in exchange for a sentence of life imprisonment, but excluded from evidence the plea agreement to that effect drafted (but not signed) by the U.S. Attorney's Office in Vermont and subsequently rejected by the Justice Department. Fell sought to offer the plea agreement as evidence (1) "that substantial mitigating factors existed," largely related to his mental history, difficult youth, remorse, and assistance to the authorities; and (2) that he had accepted responsibility for his criminal conduct. United States v. Fell, 531 F.3d at 217. The dissent identifies no error in the panel's conclusion that the district court acted well within its discretion in excluding the agreement for these purposes. Instead, the dissent urges us to consider en banc the agreement's admissibility for a quite different purpose never advanced by Fell, but purportedly compelled by principles of federalism, i.e., that a capital jury should "be allowed to know—not for purposes of nullification, but for purposes of judgment— what local law enforcement officials believed was adequate" punishment for the crime of conviction. Post at 287.

Implicit in the dissent's discussion of this point is an unwarranted factual assumption: that a local United States Attorney's view regarding the appropriate sentence for a crime mirrors the values of the community in which he serves. This is hardly obvious given that United States Attorneys are not elected to their positions but are appointed by the President of the United States, who himself may or may not have received a majority of the votes cast in the district at issue.

In any event, the dissent's argument misperceives the role of a prosecutor in our adversarial system. He is the advocate of a contested position; he is not an expert witness. The dissent asserts that allowing the jury to hear about plea offers "does not entail transforming prosecutors into expert witnesses, but simply allow[s] the defense the option of submitting prosecutors' views—along with those of so many other sources—to a capital sentencing jury." Post at 287-88. This explanation is unconvincing because "prosecutors' views" on appropriate punishment would have no relevance unless prosecutors are recognized as experts who, as the dissent assumes, have particular insight into and expertise with "the needs and values of the communities in which they work." Post at 287-88. In a capital case, however, it is the jury alone that is charged with the responsibility of speaking for the community in deciding whether a particular defendant deserves to live or to die.

As noted in Part LA of this opinion, the jurors who serve on a capital jury may support or oppose the death penalty, but in returning a verdict their oath obliges them to "temporarily set aside their own beliefs in deference to the rule of law." Lockhart v. McCree, 476 U.S. at 176, 106 S. Ct. 1758. Thus, they do not speak for the community in the sense of an electorate voting on a referendum as to whether to prescribe or proscribe the death penalty generally. Rather, they represent the community in assuming the specific responsibility to give careful and thorough consideration to the relevant evidence in the case on trial with a view toward making a unique, individualized judgment about the appropriateness of the death penalty as a punishment for the particular defendant on trial in light of his particular crime. For precisely this reason, no expert opinions on the issue of sentence are warranted, whether from local prosecutors (either when they have vigorously supported the death penalty from the moment of arrest, or when they may have favorably considered lesser alternatives), their Justice Department counterparts, capital defense bar representatives, newspaper editorial boards, victims' rights groups, or even the trial judge.

To be sure, the law may permit certain witnesses—e.g., the defendant, his family members, his victims—to make a plea for the jury to exercise its sentencing judgment mercifully or severely. See generally 18 U.S.C. § 3595(c) (stating that information may be admitted at capital sentencing hearing "regardless of its admissibility under the rules governing admission of evidence at criminal trials"). But the allowance of such appeals from persons who have been affected by the crime or who will be affected by the sentence is very different from allowing purported experts to opine as to the appropriate sentence in the case.^[10]

Further, although the law liberally permits a capital defendant to put mitigating evidence before a jury, see Tennard v. Dretke, 542 U.S. 274, 284, 124 S. Ct. 2562, 159 L. Ed. 2d 384 (2004) (recognizing mitigating evidence to include any information that "tends logically to prove or disprove some fact or circumstance which a fact-finder could reasonably deem to have mitigating value"), there are limits. Specifically, Congress has authorized district courts to exclude information from a capital sentencing hearing when its "probative value is outweighed by the danger of creating unfair prejudice, confusing the issues, or misleading the jury." 18 U.S.C. § 3593(c). The probative value of plea-offer information is minimal with respect to the jury's evaluation of the relevant aggravating and mitigating factors in a case. Such factors may not have been fully identified at the time of the plea offer. Even if they were, the totality of circumstances relevant to those factors could not have been developed as completely as is possible through the crucible of trial. Thus, any inference as to the appropriate sentence that might be drawn from an unexecuted plea agreement will necessarily be prejudicial and confusing insofar as it focuses the jury's attention, even if only briefly, on what a third-party may have thought based on a different record from that developed at the sentencing hearing. More important, evidence of an unexecuted plea agreement would distract the jury from its sworn duty carefully to consider for itself the hearing record, and to make a unique, individualized judgment about the appropriateness of the death penalty as a punishment for the particular defendant on trial.

A final word on the dissent's discussion of the role of prosecutors in capital cases. The dissent asserts that "[s]urely the concurrers know that prosecutors—even federal prosecutors—are significantly affected by local values and attitudes in their choice of what is an appropriate sentence to pursue." Post at 290 (emphasis added). Respectfully, I submit that this statement, unsupported by any authority, offers too simple a view of prosecutorial decision making.

First, the vast majority of sentences sought by federal prosecutors are determined not by juries but by judges. Until recently, judges had their discretion circumscribed by mandatory Guidelines structured to minimize local sentencing disparity. See United States v. Booker, 543 U.S. 220, 233-34, 125 S. Ct. 738, 160 L. Ed. 2d 621 (2005). Even after Booker, judges may consider the local impacts of crime in imposing sentence, but not local mores or feelings about crime. See United States v. Cavera, 550 F.3d 180, 195 (2d Cir.2008) (en banc) (dictum). Thus, to the extent "local values" are not a significant factor in a judicial determination of sentence, I do not think there is any basis for the dissent's pronouncement that prosecutorial assessments as to the appropriate sentences they might urge judges to impose are "significantly affected" by local values.^[11]

Even if the dissent's "significant effect" pronouncement refers only to capital cases, it fails to recognize the myriad factors that can inform a prosecutor's decision to seek the death penalty. More "significant" than local values supporting or opposing the death penalty, I suggest, is the strength of the government's case for a capital sentence. This factor can vary enormously depending not only on the circumstances of the capital crime, the character of slain victims, the complicity of others who may not face capital punishment, and a range of mitigating or aggravating factors pertaining to the defendant himself, but also on a strategic assessment of how much of this evidence a jury is likely to hear (in light of possible legal challenges)^[12] or to question (in light of efforts to impeach prosecution witnesses). Prosecutors carefully weigh these circumstances because they know that, even in a state where local values generally support the death penalty, the punishment can be imposed in a particular case only on unanimous consent of twelve jurors. In short, regardless of local values, the law effectively gives a single juror the ability to veto capital punishment.

Because the dissent oversimplifies prosecutorial decision making, I fear that, in construing federalism to mandate the admissibility of unexecuted plea agreements in capital cases, it has found a Pandora's Box that this court should not open. If a jury's exercise of its capital sentencing judgment is usefully informed by evidence that local federal prosecutors were, at some point, willing to dispose of Fell's case with a life sentence rather than death, would it then be relevant for a Vermont jury in some other case to know that local federal prosecutors were not willing to extend such an offer? Similarly, if a plea offer to a life sentence in a capital case were to be received as some evidence that local prosecutors thought that punishment was sufficient to do justice, would local prosecutors be allowed to explain if their reasons for extending the offer were not so based? For example, would the scuttled agreement be relevant where the offer was informed not by mitigating circumstances pertaining to the defendant but by a desire to spare a defendant's frightened or seriously injured surviving victim the further trauma of having to testify about the experience? Or where the offer was informed by the anticipated high cost of a capital prosecution? See Frederic Block, A Slow Death, N.Y. Times, Mar. 15, 2007, at A27 (estimating that "more than $17 million" has been spent on "17 federal death penalty trials in New York State"). Countless scenarios relating to plea negotiations can be envisioned. To my mind, any such evidence is invariably more prejudicial and confusing than probative and should not be presented to a capital jury. This case presents no exception.

III. Conclusion

To conclude, because I do not think the dissent's appeal to "federalism" presents us with any real constitutional concerns about the jury selection or evidentiary rulings in this case, I am not inclined to have the full court convened to address theories not raised by the parties either in the district court or on appeal. I concur in the denial of en banc review.

CALABRESI, Circuit Judge, dissenting from the denial of rehearing:^[1]

This is the first direct appeal from a death sentence that we have had in decades, and that in itself seems to me to justify a rehearing en banc. Moreover, and seemingly paradoxically, such cases will repeat, and it is important and useful that every member of the Court be given an opportunity-after argument and in the context of a particular case—to give guidance to future panels and district courts. We do not go en banc frequently, but there are situations where the knowledge and views of the whole Court serve us better than panel-by-panel development. I believe this case presents just such a situation.

In addition, this is an appeal of a federal death sentence from a state that does not have capital punishment. That is an unusual occurrence for any federal court,^[2] and it is particularly important that our Circuit address it, because two of the three states in our jurisdiction, in different ways, have effectively done away with capital punishment.^[3] The imposition of the death penalty in states that have rejected it raises issues that have not yet been addressed. I will try in this opinion to discuss a few of these and explain why I think that the District Court might have erred with regard to them. But quite apart from how one ultimately feels about those issues or whether one believes that the District Court erred, their very existence raises question "of exceptional importance," Fed. R.App. P. 35(a)(2). And that, in itself, demands that our Court focus its collective wisdom on these problems, problems with which we have had little experience but will likely face again.

I.

Aside from these more general grounds for going en banc, to which I shall return, there are also specific reasons that, I believe, justify collective review of some of the decisions made by the District Court. Although the District Judge's behavior was nothing short of exemplary, and the panel opinion exceedingly careful, the unusual nature of this case presents questions that I am not sure the district judge fully considered.

A. Jury Selection

A juror—Juror 64—expressed strong opposition to the death penalty but made clear that she was willing to follow instructions and consider imposing it. She rated herself a "one" on a scale of "one to ten," with "one" indicating strong opposition to the death penalty, and on more than one occasion noted that she would "lean" towards life imprisonment without parole rather than death. But she also repeatedly said that she "could make that decision" to vote in favor of the death penalty and "could follow that process." When asked if she could honestly consider imposing the death penalty, she responded "Yes." The District Court excused her for cause.

In its careful opinion, United States v. Fell, 531 F.3d 197 (2d Cir.2008), the panel describes Juror 64 as the closest call among the excused jurors, id. at 211, and notes that "[w]hile Juror 64 strongly opposed the death penalty and was unprepared to conclude that a defendant deserved death simply because a murder was premeditated, she simultaneously claimed that she could impose the death penalty as part of her responsibilities as a juror in spite of her expressed reluctance to do so." Id. at 213. Nonetheless, the panel, guided by the traditional rules dealing with the striking of jurors for cause, gave deference to the District Court's decision to strike Juror 64.

The Supreme Court has given district courts an exceptional amount of discretion to strike for cause jurors who may not be able to apply the law faithfully and impartially. See Wainwright v. Witt, 469 U.S. 412, 426, 105 S. Ct. 844, 83 L. Ed. 2d 841 (1985). But I am aware of no Supreme Court case that addresses the selection of jurors in a federal capital case arising in a state that does not itself have the death penalty. And that, I think, is a major and undiscussed issue that must be considered in reaching a proper resolution of this case. The question is particularly crucial when it arises in a state like Vermont that lacks the death penalty not because of court action but because of legislative decision.^[4] In such a state, there is presumably a large portion of the population that is opposed to the death penalty. This must be so if the legislature's judgment reflects the will of the people, an assumption that, rightly or wrongly, we routinely make. Notably, the ability of juries to represent the will of the people in capital sentencing is no less important than the power of legislatures to reject the death penalty all together. Cf. Blakely v. Washington, 542 U.S. 296, 306, 124 S. Ct. 2531, 159 L. Ed. 2d 403 (2004) ("Just as suffrage ensures the people's ultimate control in the legislative and executive branches, jury trial is meant to ensure their control in the judiciary.").

Nor does it matter, in this respect, that the case before us is a federal trial. The Framers found the local nature of a jury, and local values embodied in that jury, to be so important that they made it a constitutional requirement that juries in federal cases be not only "impartial" but "of the State and district wherein the crime shall have been committed." U.S. Const, amend VI. The imposition of the death penalty in a situation where a jury finds aggravating and mitigating circumstances, as this jury did, is necessarily one of discretion—of judgment. And the Constitution requires that that judgment be exercised by people from the area where the crime was committed—in this case, Vermont. United States v. Fell, No. 2:01-CR-12-01, 2005 WL 1026599 (2005) (finding venue was statutorily and constitutionally proper in Vermont). That is, the federal jury—in exercising its judgment, not, let me be clear, in order to nullify—is constitutionally required to reflect the values of the people of Vermont.

In other words, as the Supreme Court has often recognized in the context of capital sentencing, "one of the most important functions any jury can perform in making... a selection [between life imprisonment and death for a defendant convicted in a capital case] is to maintain a link between contemporary community values and the penal system." Gregg v. Georgia, 428 U.S. 153, 181, 96 S. Ct. 2909, 49 L. Ed. 2d 859 (1976) (citing Witherspoon v. Illinois, 391 U.S. 510, 519 n. 15, 88 S. Ct. 1770, 20 L. Ed. 2d 776 (1968)). It is for this reason that courts "have, in our determination of society's moral standards, consulted the practices of sentencing juries: Juries maintain a link between contemporary community values and the penal system that [the Supreme] Court cannot claim for itself." Roper v. Simmons, 543 U.S. 551, 616, 125 S. Ct. 1183, 161 L. Ed. 2d 1 (2005) (Scalia, J., dissenting) (internal quotation marks omitted).

The relevant community values in the instant case are constitutionally defined as those of Vermont. And I believe it is at least worth considering whether juries in non-death penalty states like Vermont can "maintain a link between contemporary community values and the penal system" when, in selecting juries, trial courts are not attuned to whether the jury members (and not just the jury pool, as I will argue shortly)—though willing to follow the law—are also representative of a state's overall opposition to the death penalty. For a federalism like ours—made up as it is of states whose populations hold widely different moral viewpoints—to work, perhaps even to survive, it is at least arguable that the values of the citizens of the state in question—not just a minority of them— be reflected in trial juries, even in federal cases. That is, I believe, why our federal constitution prescribes juries of the state and district in federal cases. And if those values—and the Vermonters who hold them—were improperly excluded in this case, then the sentence the jury selected must fall. See Gray v. Mississippi 481 U.S. 648, 668, 107 S. Ct. 2045, 95 L. Ed. 2d 622 (1987) (improper exclusion of prospective jurors based on their views of the death penalty is reversible error).

This does not mean that a district court may not exclude a juror if the court has reasons, based on specific record facts, indicating that the juror will not follow the law. District judges are entitled to deference in making that determination. Witt, 469 U.S. at 425, 105 S. Ct. 844. But the court's discretion must be exercised in the proper context. And if in exercising its discretion the District Court was not fully aware of the importance of representing the constitutionally mandated vicinage or did not think it was permitted to consider it, then it is hard to know how we can defer to the court's decision. Answering that question could be as easy as remanding this case to ask the able District Judge whether, in striking Juror 64, he fully considered the constitutional relevance of the values of Vermonters, the values of the jurisdiction in which he sat.

Of course, the Sixth Amendment is often satisfied simply by drawing a fair cross section of the community for the jury pool. Lockhart v. McCree, 476 U.S. 162, 174, 106 S. Ct. 1758, 90 L. Ed. 2d 137 (1986). And one might argue, therefore, that federalism concerns will be adequately met so long as the jury pool in a federal capital case—as opposed to the sentencing jury itself—is a fair cross-section of a capital punishment-opposing community such as Vermont. But, if the issue is constitutionally important enough, that does not always suffice. Constitutional values must then be vindicated not just at the venire stage, but in the actual selection of a trial (or sentencing) jury. Thus in Batson v. Kentucky, 476 U.S. 79, 106 S. Ct. 1712, 90 L. Ed. 2d 69 (1986), the Supreme Court held that prosecutors may not use race as a factor when making peremptory challenges. The value that is at stake in Batson—equal protection of the laws—is surely a fundamental constitutional one. But one could argue that the value at issue here—federalism and its recognition of the different viewpoints that make up our nation—is also a fundamental constitutional one. And that value entails the right to be tried by a set of people who truly represent the point of view of a state and district.

One may debate whether the recognition of local juries that federalism represents is as important as the principles of non-discrimination at work in Batson. But the recognition of federalism's requirements is not obviously of lesser value, particularly in light of the fact that the right to local juries was manifestly one of the reasons the Revolution was fought. Brian C. Kalt, Crossing Eight Mile: Juries of the Vicinage and County-Line Criminal Buffer Statutes, 80 Wash. L.Rev. 271, 296-305 (2005) (describing Founding-era concerns that "put the right to local jury trials at the forefront of the Revolution"). The practical significance of such recognition and jury selection in the case before us simply has not been discussed, let alone examined fully through argument and briefing, as would occur in an en banc. And it is not enough to say that the constitutional rules surrounding jury selection in federal death penalty cases must apply "equally" in both death penalty and non-death penalty states. That is true enough. But what that means with respect to the exclusion of particular jurors in view of the characteristics of the community where the trial is held, is anything but easy to decide.^[5]

Nor, of course, does giving respect to federalism contradict the proposition that "the Constitution must apply equally throughout the states." See supra at 271 (Raggi, J., concurring in the denial of rehearing en banc). The fact that a constitutional right applies nationwide does not answer the question of how the right is operationalized in different communities, nor does it mean that juries must—or even may—look the same in different states. Respect for federalism as a constitutional principle must apply equally throughout the nation; it is only the underlying local values that vary. My concern is with whether and how we are to take account of those state-based variations.

The panel did a very good job applying the traditional rules, but those rules were developed in traditional cases. The case before us is anything but traditional, and I believe that the peculiar federalism issue it raises (informed by the constitutional requirement of local juries) is something that was not focused on sufficiently and that we should consider collectively, especially because it might well recur given the states over which we have jurisdiction.

B. The Plea Agreement

A second question arises from the District Court's decision to exclude a draft plea agreement, and the panel's related determination that Fell has failed to show that the prosecution misrepresented his willingness to plead guilty when it said that "if [Fell] wanted to plead guilty he could have." Fell, 531 F.3d at 220. Here again, the panel applied traditional rules, and gave great weight to the District Court's determination that allowing plea deals to be introduced would distract the jury from its own assessment of the situation. In doing so, the panel noted that all twelve of the jurors were aware that Fell was willing to plead guilty, leading the panel to conclude that admission of the draft plea agreement was "cumulative." Id.

But the draft plea agreement was important not only, as the panel suggested, because it showed Fell's willingness to admit responsibility for the death of Teresca King—which six jurors found to be mitigating factor—but also because it showed that the local prosecutors were willing to accept a sentence of life imprisonment in exchange for that plea.^[6] (Of course, following Department of Justice rules, the prosecutors could not actually sign the agreement unless the Attorney General approved, which he did not.) In fact, the draft plea agreement reflected the judgments of Vermont-based law enforcement enforcers on what was appropriate locally, and its overruling by Main Justice reflected a centralist, and in this case decidedly anti-federalist, decision. See John Gleeson, Supervising Federal Capital Punishment: Why the Attorney General Should Defer When U.S. Attorneys Recommend Against the Death Penalty, 89 Va. L.Rev. 1697, 1716 (2003) ("In a federal system that rightly accords great deference to states' prerogatives, the federalization of the death penalty should be limited to cases in which there is a heightened and demonstrable federal interest, one that justifies the imposition of a capital prosecution on communities that refuse to permit them in their own courts.").

This does not mean, of course, that the Attorney General should not have the authority to make such decisions. But given the Constitution's powerfully federalist assertion that matters of judgment should be decided by juries of the state, and given that this is a state that opposes the death penalty, it is at least worth considering whether the jury should be allowed to know—not for purposes of nullification, but for purposes of judgment—what local law enforcement officials believed was adequate. While the ultimate power to prosecute is given nationally, the power to sentence is constitutionally given to the local jury. And, arguably, that jury should be fully informed as to whether the decision to seek the extreme penalty was local or national.

To be sure, there are reasons not to allow the admission into evidence of deals that have not been finalized. But without infringing on these reasons, the high federalist values at stake could be protected by permitting the defendant to present to the sentencing jury the fact that the local U.S. Attorneys had, in their judgment, deemed it sufficient to have a life sentence. This crucial datum was totally lost in the instruction to the jury that "[t]he government refused that [plea] offer." Fell, 531 F.3d at 218 (emphasis added). Whatever our concerns about the admission of incomplete plea agreements, it must be noted that in this case the agreements were introduced, but in a way that misled the jury and effectively prejudiced Fell. This is because the agreements were presented in a way that suggested, quite wrongly, that the local prosecutors had rejected Fell's offer to plead guilty in exchange for a sentence of life imprisonment. Allowing the jurors to know that this was not so, whether through access to the incomplete plea negotiations or in some other way that the distinguished district judge could have crafted, would have corrected this misconception and would have helped vindicate the values of federalism. And it would have given the jury a relevant datum that was as crucial to its decision as the rest of the wide range of evidence to which it is entitled in the capital context.

Capital sentencing juries may consider almost anything in making their decisions. See Tennard v. Dretke, 542 U.S. 274, 284-85, 124 S. Ct. 2562, 159 L. Ed. 2d 384 (2004). Indeed, defendants' right to present broad evidence during sentencing is such that the Supreme Court has repeatedly found capital counsel constitutionally inadequate for failing to investigate fully or to present such evidence, even where the defendant himself discouraged it. See, e.g., Rompilla v. Beard, 545 U.S. 374, 377, 125 S. Ct. 2456, 162 L. Ed. 2d 360 (2005) ("We hold that even when a capital defendant's family members and the defendant himself have suggested that no mitigating evidence is available, his lawyer is bound to make reasonable efforts to obtain and review material that counsel knows the prosecution will probably rely on as evidence of aggravation at the sentencing phase of trial."); Wiggins v. Smith, 539 U.S. 510, 123 S. Ct. 2527, 156 L. Ed. 2d 471 (2003) (finding assistance to be ineffective where defense counsel unreasonably limited investigation). If defense counsel can be found constitutionally ineffective for failing to present to a jury information about a defendant's difficult family history, it seems at least plausible that a jury should also be permitted to know that local prosecutors were willing to seek a penalty less than death. The latter seems to me to be no less relevant to a jury's decision of whether to impose death than whether, as in Wiggins, a defendant's mother was an alcoholic or a defendant had "frequent, lengthy absences from school." 539 U.S. at 525, 123 S. Ct. 2527.

This does not entail transforming prosecutors into expert witnesses, but simply allowing the defense the option of submitting prosecutors' views—along with those of so many other sources—to a capital sentencing jury. There can be no doubt that despite their titles, "United States" Attorneys undoubtedly make charging and other decisions based on what they see as the needs and values of the communities in which they work. It is therefore not surprising that state-by-state disparities in the administration of federal death sentences correlate with the frequency with which U.S. Attorneys request Main Justice's approval to seek death sentences. Rory K. Little, The Federal Death Penalty: History and Some Thoughts About the Department of Justice's Role, 26 Fordham Urb. L.J. 347, 453 (1999). And, arguably, this is not just acceptable, but rather is fundamental if we are to survive as a united federalism in a nation whose values differ profoundly from one part to another.

Although judges retain a gate-keeping function to "exclude, as irrelevant, evidence not bearing on the defendant's character, prior record, or the circumstances of his offense," Lockett v. Ohio, 438 U.S. 586, 604 n. 12, 98 S. Ct. 2954, 57 L. Ed. 2d 973 (1978), the bounds of relevancy in capital sentencing are exceptionally broad. The Federal Death Penalty Act itself provides that evidence may be admitted "regardless of its admissibility under the rules governing admission of evidence at criminal trials." 18 U.S.C. § 3593(c). Other Circuits with more experience in capital cases have recognized that "[t]he Federal Death Penalty Act ... erects very low barriers to the admission of evidence at capital sentencing hearings. Since the need to regulate the scope of testimony is less at the penalty phase than at the guilt phase of trial, parties may present evidence `as to any matter relevant to the sentence.'" United States v. Lee, 274 F.3d 485, 494 (8th Cir. 2001) (quoting 18 U.S.C. § 3593(c)); see also United States v. Fields, 483 F.3d 313, 343 (5th Cir.2007) ("[T]he sole statutory restriction is that evidence may be excluded if it is more prejudicial than probative.").

Without further briefing or consideration by this Court en banc, I cannot say with confidence whether the admission of the plea agreement would have changed the mind of a juror.^[7] But as the Supreme Court has recognized in the context of other capital sentencing matters, such uncertainty "compels a remand for resentencing so that we do not `risk that the death penalty will be imposed in spite of factors which may call for a less severe penalty.'" Penry, 492 U.S. at 328, 109 S. Ct. 2934 (quoting Lockett, 438 U.S. at 605). What I can confidently assert is that the able District Judge gave no indication that he was considering federalism values when making his decision to exclude the views of the local prosecutors.

I do not fault the District Court or the panel for doing what is standard in capital cases. My concern is simply that they have given no sign that they thought about the broader federalist and jurisprudential aspects of the issue. And the answers to those and to other analogous questions seem to me—in our first direct capital appeal in decades—to be worth our specific and careful attention en banc.^[8]

II

Behind all this lies a still deeper constitutional question about the pursuit of the death penalty in states whose popularly enacted laws do not permit it. That question has not been directly raised to date in this case. That would not, however, preclude us from considering it, especially since it is simply a legal argument as to which we could get full briefing in an en banc argument.

It is no accident that although the federal death penalty was effectively reinstated in 1988, this is the first direct appeal in a capital case that we have had before us. The very asking of the death penalty had long been unusual for federal prosecutors in much of our Circuit. And even when they have asked for it, juries—constitutionally mandated juries of the state and district—have refused to impose it. In other words, as a matter of judgment, applying the norms of their "state and district," local federal juries have repeatedly and overwhelmingly rejected Washington's invitation to execute criminals in their states. See Alan Feuer, An Aversion to the Death Penalty, but No Shortage of Cases, N:Y. Times, Mar. 10, 2008, at B1 (noting that federal prosecutors have sought death sentences in 19 New York cases since 1988 and obtained such a sentence only once, compared to a one-third success rate nationally).

What is going on here is that the existence of certain local values makes the imposition of the federal death penalty in states that do not have the death penalty truly uncommon. It is in that sense significantly more "unusual" than was the execution of 16- and 17-year-olds or the execution of the mentally retarded, before the Supreme Court found those practices to be "cruel and unusual" in Roper, 543 U.S. 551, 125 S. Ct. 1183, 161 L. Ed. 2d 1, and Atkins, 536 U.S. 304, 122 S. Ct. 2242, 153 L. Ed. 2d 335. In all of these categories, values that made a death sentence unusual were at work. In cases involving juvenile and mentally retarded defendants, the values were apparently related to the culpability of the offender. See, e.g., Atkins, 536 U.S. at 306, 122 S. Ct. 2242; Roper, 543 U.S. at 568-75, 125 S. Ct. 1183. In cases from states without the death penalty, the constitutionally salient values are not just the "local" values, like the existence of substantial generalized opposition to capital punishment, but much more fundamentally the value and endurance of federalism itself—the recognition that we are part of a country, of a polity, that has to live with both Texan values and Northeastern values.

Naturally, a case will only reach us when, in the particular instance and despite those local values, the death penalty has been imposed. But that is always so where the claim is one that a punishment is cruel and unusual. It was the case with minors and the mentally retarded as well. Constitutionally, one cannot look at the individual case, but must instead consider whether the presence of important generalized values has made the particular death sentence constitutionally unusual. It seems to me at least arguable that the application of the death penalty in situations that involve predominantly local crimes in non-death penalty states may be sufficiently rare as to be constitutionally prohibited. And this may be so even though the conduct at issue constitutes a federal crime. The actions of local prosecutors and juries, in response to federalist values,^[9] may well have made the death penalty in such situations constitutionally void, just as their actions, in response to other values, made application of capital punishment constitutionally prohibited as to minors and the mentally retarded.

III

I add a few comments in reaction to the concurring opinion.

I do not understand why the concurring opinion begins by describing the viciousness of the murder, when the only fact relevant to the current discussion is that the victim was transported across state lines and hence that federal law could properly be applied. No one doubts that this murder—like most murders—was awful. But that, along with the full interplay of aggravating and mitigating factors, is something that our system leaves up to the judgment, first, of prosecutors empowered to seek what they believe to be an appropriate sentence, and then to properly selected juries. That is what my arguments focus on—the role of the prosecutors, and whether the jury was properly selected. And that, rather than appeals to emotion, is all that matters in this case.

In this regard, I believe that the concurring opinion misses the main point. Surely the concurrers know that prosecutors— even federal prosecutors—are significantly affected by local values and attitudes in their choice of what is an appropriate sentence to pursue. A key question in this case remains, therefore, whether that judgment, and the fact that it was overruled by the Attorney General of the United States—reflecting other values, perhaps national ones or perhaps of states that mattered more to him—is something that in some way or other the local, federal jury, the institution that is constitutionally charged with making the ultimate punishment decision, should know.^[10]

Surely, also, the concurrers know that a district judge, in deciding whether a juror, informed by his or her values, is capable or not of following the law, is not making a decision certain. The judge is taking an educated guess and in doing so is inevitably influenced by how much he believes that he is permitted to take account of, and respond to, the values of the jurisdiction in which he sits. The import of these values should, of course, be considered alongside the various other pieces of evidence regarding a juror's fitness for service on a capital jury. Here, for example, the three jurors whose exclusion is most disputed— Jurors 64, 141, and 195—were asked to rate their opposition to the death penalty on a scale of 1 to 10 (with 1 being strongly opposed and 10 being strongly in favor), and responded with rankings of 1, 4, and 8 respectively. All three were subsequently excluded based on their explanations of their personal, legal, religious, moral, and other views regarding the death penalty. I believe it is worth considering en banc whether, as part of this matrix of factors, we should clarify that district judges facing unusual capital sentencing situations such as this one may consider—alongside such other factors as jurors' religious beliefs, personal histories, and moral commitments, all of which are thought to be constitutionally relevant in the selection of death penalty jurors—the constitutionally mandated values of federalism.^[11] In this regard, one might contrast Jurors 64 and 195. Juror 64 stated that she was opposed to the death penalty but, essentially, that she could apply it in a particular case; Juror 195 explained, "At a philosophical level, I am in favor of the death penalty, but don't know if I could vote in favor of it when the decision is in my hands." Eliminating jurors who, like Juror 195, whether or not they support the death penalty as voters but cannot themselves vote in favor of as jurors is something that capital sentencing judges must do at every voir dire. But the position of Juror 64 is more difficult. In many ways, Juror 64 was simply representative of the views of Vermont, a fact that a district judge must be permitted to take account of. And here, it is not at all clear that he felt free to do so.

It is common in capital cases for jurors to be conflicted about whether they could, if asked, sentence another person to death. Until called for jury duty and questioned in court about whether they could vote in favor of death, few people have given the question serious, focused thought. See, e.g., (statement of Juror 64) ("I have never imagined myself being in a position to have to decide someone else's fate as their life goes, and I guess in theory, I have always felt that I wouldn't believe in the death penalty."). Dealing with this kind of indecision is always difficult, which is why the Witt line of cases exists. But I believe that it is especially hard in a state like Vermont, whose citizens may well have believed that—through their political decision not to have capital punishment—they have absolved themselves, as jurors, of the need to consider the question at all. Accordingly, special care may be required during capital jury selection in such jurisdictions. For in such states, citizens are likely to be forced to confront for the first time, as jurors, these momentous issues of life and death which they thought had been taken off the table by the voters. The District Judge in this case did an exemplary job with a very difficult case. But I am not as sure as the concurrers seem to be that the same District Judge, if he was told that he could take into account the local values of Vermont during jury selection, would have come out the same way with respect to Juror 64. That is why I would ask him.

Most significantly, the concurrers respond to my dissent only with formal objections and never answer the fundamental question: What is necessary for a federalism to survive, when it is made up of people of deeply different values who reside in sovereign states that reflect those different values.^[12] Judge Raggi's discussion of the role of federalism is a good example of what is wrong with the affirming opinions as a whole. To describe federalism as having only the role she asserts for it is to misunderstand completely both the importance of federalism and the difficulties federalisms face, and her citations to the distinguished scholars who assert that the role she describes is a part of federalism do nothing to strengthen her argument.

We are a nation with profoundly different values, and have been so from the very beginning. Moreover, these value differences are and have always been geographically correlated. To remain united and yet responsive to significant differences is anything but easy. Centrifugal forces (that led to the Civil War and lead to recurring suggestions of the advantages of secession, see James C. McKinley, Jr., Texas Governor's Secession Talk Stirs Furor, N.Y. Times, April 18, 2009, at A15) are always present. And so are centripetal forces, that would make us all the same; in one national state. Capital punishment, assisted suicide, abortion, polygamy, same sex marriage, segregation, and affirmative action are just a few historic and more recent tension points subject to these centrifugal and centripetal forces.

One can always give "formal answers" lodged, for example, in jurisdictional language or in procedural rules as to why, in a particular area, and depending on which side one is on in that area, the centrifugal or the centripetal forces ought to win. But such a crabbed view misses the point that all branches of government, courts no less than legislatures and executives, must be constantly aware of what is needed to keep us together and yet allow us to be different. It is the failure to face the broader issues at all, or even to deem them worth discussing en banc, that makes the concurring opinion's rejection of en banc review unsatisfactory. And that is so both (a) in its treatment of jury selection and the role of prosecutors—whose alleged imperviousness to local values would be risible were it not serious in this case—and (b) in its reductionist, even dismissive, description of what constitutes constitutional "unusualness"^[13]—leaving aside the remarkable and troubling speculation as to whether New York juries sentence defendants to death based on "the character of victims." See supra at 276 (Raggi, J., concurring in the denial of rehearing en banc).

Apart from formalisms, the principal point on which Judge Raggi's opinion seems to rely is that federal crimes do not implicate federalism. I believe that this is simply incorrect. The proliferation of federal crimes has been such that a very large number of things can be tried either locally or federally. See Daniel Richman, Prosecutors and Their Agents, Agents and Their Prosecutors, 103 Colum. L.Rev. 749, 795 (2003) ("[T]he federal criminal `code' may well be even broader than that of the states in the range of conduct it ostensibly covers."). For many reasons, including a sense of the importance of not making everything subject to centralized federal law, most criminal prosecutions are nevertheless left up to the states. See Jamie S. Gorelick & Harry Litman, Prosecutorial Discretion and the Federalization Debate, 46 Hastings L.J. 967, 969 (1995) (noting that 95% of criminal prosecutions happen in state courts). The question of when something is tried nationally or locally is itself a serious issue of federalism. But even when crimes are tried federally, local disparities in sentencing—some justified and some unjustified—are unavoidable. See, e.g., Stephanos Bibas, Regulating Local Variations in Federal Sentencing, 58 Stan. L.Rev. 137, 138 (2005) (noting local disparities in federal sentencing and arguing that "[e]qual treatment of similar offenders in different places is one important value in sentencing, but not the only one"). The fact that this case involved a federal crime does not absolve us from considering the relevance of federalism.

Judge Raggi rightly points out that judges are limited in the degree to which they can consider local values when imposing a sentence. See supra at 280 (Raggi, J., concurring in the denial of rehearing en banc) (citing United States v. Cavera, 550 F.3d 180, 195 (2d Cir.2008) (en banc)). But such appropriate limitations on a judge's discretion to consider some factors increase the power and responsibility of juries and prosecutors to do so, and that is why these are the two institutions with which my opinion is in large part concerned. Restricting judicial discretion over sentencing—and judicial discretion in federal capital sentencing is totally restricted—has long been recognized to increase the importance of prosecutorial discretion. Cf. Kate Stith & José A. Cabranes, Judging Under the Federal Sentencing Guidelines, 91 Nw. U.L.Rev. 1247, 1256-57 (1997) ("[T]here is little doubt that the Guidelines have augmented the authority of the prosecutor relative to that of the judge."); Daniel J. Freed, Federal Sentencing in the Wake of the Guidelines: Unacceptable Limits on the Discretion of Sentencers, 101 Yale L.J. 1681, 1723-24 (1992). This is one of the many reasons why I think the jury in this case might have been entitled to know more about why (and indeed if) the local U.S. Attorneys in this case exercised their discretion in seeking a death sentence.

Of course, I agree completely with Judge Raggi that "myriad factors ... can inform a prosecutor's decision to seek the death penalty." See supra at 281 (Raggi, J., concurring in the denial of rehearing en banc). Judge Raggi rightly lists among these factors the circumstances of the crime, the character of slain victims, and a range of mitigating and aggravating factors pertaining to the defendant himself. Id. Significantly, however, all of these "myriad factors" are admissible at capital sentencing. See, e.g., Kansas v. Marsh, 548 U.S. 163, 174, 126 S. Ct. 2516, 165 L. Ed. 2d 429 (2006) (state capital sentencing regime must "permit a jury to render a reasoned, individualized sentencing determination based on a death-eligible defendant's record, personal characteristics, and the circumstances of his crime"); Payne v. Tennessee, 501 U.S. 808, 827, 111 S. Ct. 2597, 115 L. Ed. 2d 720 (1991) (victim impact evidence may be admitted); Lockett v. Ohio, 438 U.S. 586, 604, 98 S. Ct. 2954, 57 L. Ed. 2d 973 (1978) (plurality opinion) (capital sentencer must "not be precluded from considering, as a mitigating factor, any aspect of a defendant's character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death"). For this reason too, I therefore believe it is worth our en banc consideration whether, among these "myriad factors," juries should also be permitted to know that local federal prosecutors were forced by Main Justice to seek death.

Judge Raggi is also quite right that the sentence ultimately determined by a jury is a product of jury deliberation, into which courts do not intrude. See supra at 270 (Raggi, J., concurring in the denial of rehearing en banc). But jury deliberation is in turn and inevitably a product of who is on the jury and of what values they hold. And that is why I think it is so important for us to consider how this jury was selected.

Finally, Judge Raggi notes that when capital punishment was permitted under New York law, there were quite a few death sentences handed down under that state law. See supra at 275 (Raggi, J., concurring in the denial of rehearing en banc). She uses this, not improperly, to argue that the degree of feeling about capital punishment in New York is by no means overwhelming one way or the other. But this simply makes it all the more remarkable that "no federal juries have been more reluctant to sentence federal defendants to death than those in New York." Feuer, supra (emphasis added). Indeed, the fact that New York juries were willing to impose death under their own law but not under federal law helps to underscore federalism as a likely independent variable in their deliberations, and it also suggests that the jurors in another federal case who said they voted for life because death would not have been an option in state court, see Drug Lord Sentenced to Life Without Parole, supra, were not alone in responding to federalism as value in itself.^[14]

The concurrers make many other arguments as to (a) why it might not be wise in the end for us to permit district judges to consider federalism when selecting juries for federal capital cases being tried in states whose laws forbid the death penalty, and (b) why the position of local prosecutors on life or death should not be made available to jurors in such cases. I do not exclude that some or even all of these arguments might, in the end, be convincing. But that can only be wisely determined after a complete discussion, informed by briefs, arguments, and citations to authority by parties whose attention has been fully focused on the issues by an en banc court.

All in all, Judge Raggi's opinion demonstrates very well why I can say that the District Court's conduct was exemplary, and yet ask that that conduct be at least reconsidered en banc in the light of what is needed in an area like capital punishment where our federalism has seemingly agreed to allow profound disagreement and yet where centripetal forces (on both sides) constantly challenge that federalist agreement. For this reason, even were I to find Judge Raggi's technical arguments convincing—which I do not—her opinion would still seem to me to miss the main point at stake in this case.

* * *

It has been said that "death is different." Ford v. Wainwright, 477 U.S. 399, 411, 106 S. Ct. 2595, 91 L. Ed. 2d 335 (1986) (opinion of Marshall, Brennan, Blackmun, and Stevens, JJ.). Because as a Circuit we have almost no experience in capital cases, we have not been afforded the opportunity to determine whether, and how, this is true. The issue before us, let me emphasize yet again, is not how we come out, but whether this particular case is worth reviewing en banc. I am not a great believer in en bancs. It seems to me, however, that these are issues to which the putting together of all our thought would be helpful. A rehearing en banc would help us all think about the important issues, framed by specific objections, properly before us on direct appeal, that are raised by the potential, nationally ordered, execution of a man in a state whose laws forbid it. I therefore respectfully dissent from the denial of such a rehearing.

POOLER, Circuit Judge, dissenting from denial of rehearing in banc.

I see no error in the district court's nor the panel's determinations using the "traditional rules" that guide striking jurors for cause and excluding a draft plea agreement as evidence of mitigating factors. See United States v. Fell, 571 F.3d 264 (2d Cir.2009) (Calabresi, J., dissenting from the denial of rehearing in banc). However, I agree with Judge Calabresi that our Court would benefit from in banc review since the factual circumstances of this case raise the question whether traditional rules suffice if "the death penalty verdict here is constitutionally `unusual.'" Id. at 273. We must first ask whether application of a precedent to a particular case is appropriate before it is followed. The law's stability is necessarily balanced with the need for flexibility in light of unusual or changing circumstances. Cf. Roper v. Simmons, 543 U.S. 551, 587, 125 S. Ct. 1183, 161 L. Ed. 2d 1 (2005) (Stevens, J., concurring) ("[T]hat our understanding of the Constitution does change from time to time has been settled since John Marshall breathed life into its text."). There are concerns of constitutional dimension in this death penalty case that warrant further inquiry.

I believe the value of in banc review of this case outweighs the reasons for declining review. I therefore have voted in favor of in banc review.

SACK, Circuit Judge, dissenting from denial of rehearing en banc.

Members of the Supreme Court have commented with some frequency that "death is different." See, e.g., Ford v. Wainwright, 477 U.S. 399, 411, 106 S. Ct. 2595, 91 L. Ed. 2d 335 (1986) (opinion of Marshall, Brennan, Blackmun, Stevens). That is doubtless so.

Although I am hardly ready to dismiss out of hand the views Judge Calabresi sets forth in his dissent from denial of rehearing en banc. I suspect that were we to grant rehearing, I would ultimately come to agree with the views expressed in the panel opinion. And in general terms and at first blush, it seems to me that the behavior of the trial judge was indeed "nothing short of exemplary." United States v. Fell, 571 F.3d 264 (2d Cir.2009) (Calabresi, J., dissenting from the denial of rehearing en banc).

Nonetheless, because "death is different" and this is our first decision in many decades addressing the imposition of the death penalty on appeal,^[*] I think that this is the rare case in which it makes institutional sense for us to render it as "the Court" and not as a panel thereof. Especially in light of the likelihood that other such cases will come before us in the reasonably near future, I think that an exchange of views among the members of the Court on these issues in this discrete context—with the benefit of briefing, argument, and deliberation—would be of considerable value to the Court and, through it, to the public.

I therefore have voted in favor of en banc review.

NOTES

[1] Judge Hall is recused from consideration of the petition for rehearing en banc.

[1] Senior Circuit Judge John M. Walker, Jr., who was a member of the three-judge panel in this case, has authorized me to note his agreement with this opinion.

[2] Title 18 U.S.C. § 1201(a) originated as the Lindbergh Law, enacted to address public concern that local authorities could not adequately respond to kidnappings once victims were transported across state lines. See United States v. Jackson, 390 U.S. 570, 589 n. 34, 88 S. Ct. 1209, 20 L. Ed. 2d 138 (1968). To the extent congressional authority to legislate in this area derives from the Commerce Clause, it is worth noting that Fell's crime did not simply have an effect on interstate commerce; it was itself interstate activity made possible by a channel of interstate commerce. See generally United States v. Lopez, 514 U.S. 549, 115 S. Ct. 1624, 131 L. Ed. 2d 626 (1995) (distinguishing between (1) crimes involving actual interstate activity or channels of interstate commerce and (2) crimes involving only intrastate activity but having interstate effects, in discussing particular federalism concerns raised by latter).

[3] The dissent states: "Surely, . . . the concurrers know that a district judge, in deciding whether a juror, informed by his or her values, is capable or not of following the law, is not making a decision certain. The judge is taking an educated guess and in doing so is inevitably influenced by how much he believes that he is permitted to take account of, and respond to, the values of the jurisdiction in which he sits." Post at 290. I respectfully suggest that this offers too simple a view of jury selection.

First, it trivializes the task of jury selection to describe it as "educated guess[ing]." As those who have watched responsible voir dire conducted—particularly in federal capital cases — can attest, the process requires the application of mature judicial judgment, informed by relevant legal precedent and often by considerable trial experience, to the assessment of each prospective juror's demeanor as he answers various questions. To be sure, that is what is reflected in the record in this case.

Second, when the dissent states that voir dire decisions are "inevitably influenced" by the degree to which the trial judge thinks he is permitted to take account of local values, it offers no authority for this assertion. Because nothing appellate judges do is analogous to jury selection, I respectfully submit that the dissent cannot rely simply on its own judicial experience to support this pronouncement. In Part II.B of this opinion, I explain why federal trial judges in fact have little reason to consider "local values" in conducting a capital voir dire, except to ensure that such values will not prevent a juror from following the legal instructions of the court. Where, as in this case, a trial judge questions a juror opposed to the death penalty, the obligation under Witt-Witherspoon is to determine whether the juror can clearly set aside her personal views and follow the law as stated by the court. If the judge reaches a definite conclusion that the juror cannot do so, the judge properly removes that juror without regard to whether her opposition to the death penalty does or does not comport with local values. Similarly, absent such a negative conclusion, the district judge could not remove the juror, again, regardless of local values.

[4] Although the dissent, in support of its federalism argument, conclusorily asserts that Juror 64 "essentially" demonstrated a willingness to set aside her opposition to the death penalty and follow the law, post at 291, it does not contend that the district court's contrary finding of fact was clearly erroneous. The point is significant because, if the district court's finding were clearly erroneous, one would not have to invoke "federalism" to identify constitutional error. But if the district court's finding merits appellate deference, then no federalism value would warrant remand.

[5] The dissent asserts that this describes only "a part of federalism." Post at 292 (emphasis in original). It cites no authority, however, to support a broader construction of federalism requiring federal courts to apply federal law, and particularly the Constitution, by reference to the different "local values" of individual states. Indeed, I am not sure such a requirement finds support even in the broad theories of federalism reflected in the concepts of state nullification and concurrent majorities expressed in the Virginia and Kentucky Resolutions of 1798, reprinted in 4 J. ELLIOT, DEBATES IN THE SEVERAL STATE CONVENTIONS ON THE ADOPTION OF THE FEDERAL CONSTITUTION, 528, 546 (2d ed. Phila.1866), or in the writings of John C Calhoun, see generally JOHN C. CALHOUN, A Disquisition on Government and A Discourse on the Constitution and Government of the United States, in 1 THE WORKS OF JOHN C. CALHOUN 1, 109 (Charleston, S.C., Walker & James 1851). In any event, these theories did not survive the Civil War. In the absence of authority, the dissent can hardly attack this opinion for "misunderstand[ing] completely. . . the importance of federalism." Post at 292.

[6] This is, in any event, a curious case in which to invoke the Sixth Amendment's vicinage requirement to demand special solicitude for Vermont values as they pertain to capital punishment. Fell's federal crimes are properly deemed "committed" in Vermont, U.S. Const. amend. VI, because the charged interstate kidnapping and carjacking originated in that state, see 18 U.S.C. § 3237(a) (providing that crimes committed in more than one district may be prosecuted "in any district in which such offense was begun, continued, or completed"); United States v. Rommy, 506 F.3d 108, 119 (2d Cir.2007). But the element of the crime that subjected Fell to the death penalty — Teresca King's murder — occurred in New York. To the extent the vicinage requirement originates in the common law expectation that a jury will be drawn from the locale where the "blood had been spilled," AKHIL AMAR. AMERICA'S CONSTITUTION: A BIOGRAPHY 233 (2005), one might think that if, as the dissent urges, vicinage appropriately considers local values, New York's values have as much claim to attention as Vermont's in this case. While I do not share the dissent's view of vicinage, I note that, at the time of Fell's 2005 trial, New York had enacted death penalty legislation, which had only recently been struck down based on procedural defects in the state law not relevant to the application of federal capital law. See infra at 274-78 (discussing history of New York capital legislation and difficulty in drawing conclusions therefrom as to local values). But, certainly, Fell's federal constitutional rights with respect to voir dire cannot differ depending on whether Teresca King's blood was spilled in Vermont or New York, or even on whether his case was prosecuted in one or the other of these appropriate venues.

[7] In discussing why I question the inferences the dissent draws from the New York experience and attempts to transfer to Vermont, I do not explore the not-insignificant question of whether different local values are held among the four federal judicial districts representing distinct vicinages in New York.

[8] The fact that many victims of New York federal capital crimes are themselves implicated in criminal activity is not surprising given that a high number of such prosecutions have arisen under statutes making it a crime to kill a person in furtherance of a continuing drug trafficking enterprise, see 21 U.S.C. § 848(e)(1)(A), or in aid of racketeering, see 18 U.S.C. § 1959(a)(1), (5). As discussed in the next paragraph of text, New York capital prosecutions have arisen in a variety of circumstances often involving innocent victims.

[9] In Pitera, for example, jurors identified the equal culpability of confederates who did not face the death penalty as well as the certainty that defendant would be incarcerated for life without parole as factors militating against imposition of the death penalty. See United States v. Pitera, 90-cr-0424 (E.D.N.Y.) (Verdict Form at Penalty Proceeding).

In United States v. Al-`Owhali, 98-cr-1023 (S.D.N.Y.) (prosecuting defendants for the murder of more than 200 innocent persons as a result of bombing U.S. embassies in Africa), the jury identified as a factor militating against death a concern that execution "could make [the defendant] a martyr." Benjamin Weiser, Life for Terrorist in Embassy Attack, N.Y. Times, June 13, 2001, at A1.

In United States v. Dhinsa, 97-cr-672 (E.D.N.Y.), the trial judge and defense counsel cited a film portraying defendant's family as the critical factor in the jury's decision not to impose the death penalty. See "Why Is Death Different? Litigating Death Penalty Cases in Federal Court," Panel Discussion at Second Circuit Judicial Conference (June 18, 1999), Tr. at 147 (quoting Judge Edward R. Korman: "[O]nce I saw this particular video, I thought to myself that the jury in this case is never going to vote the death penalty."); id. at 139 (quoting Dhinsa's counsel, Gerald L. Shargel, regarding "tremendous impact" of family film).

Of course, in many capital cases, the most significant mitigating factor may be the sympathetic portrayal of the defendant's own life history. See id. at 124-35 (reporting capital defense counsel David Bruck's efforts to construct and present defendant's life story effectively to jury).

[10] The dissent's conclusory assertion that this concurring opinion's objections to allowing a capital jury to hear the sentencing views of local prosecutors "apply just as much to letting the jury know that Fell had admitted guilt in the course of negotiations with prosecutors" is unconvincing. Post at 286 n. 6. Fell's admission of guilt was offered as evidence of conduct specifically, acceptance of responsibility, which has traditionally been viewed as a factor mitigating sentence. See, e.g., U.S.S.G. § 3E1.1. By contrast, nothing supports the dissent's suggestion that a prosecutor's view about appropriate punishment in a capital case is a proper factor for jury consideration.

[11] Rather than defend its original assertion about the significant effect of local values on prosecutorial decision making generally, the dissent submits that the point I make in this paragraph of the opinion otherwise supports its conclusion that Fell's jury should have heard about the unexecuted plea agreement. The dissent observes that, because the mandatory Guidelines regime "augmented the authority of the prosecutor relative to that of the judge," Kate Stith & José A. Cabranes, Judging Under the Federal Sentencing Guidelines, 91 Nw. U.L.Rev. 1247, 1257 (1997), Fell's jury "might have been entitled to know more about why (and indeed if) the local U.S. Attorneys in this case exercised their discretion in seeking a death sentence." Post at 293 (emphasis in original). The dissent does not explain how the quoted observation by Professor Stith and my concurring colleague Judge Cabranes supports its conclusion. This is no casual omission. The Sentencing Guidelines—whether mandatory or advisory—do not apply to capital sentencing decisions. Thus, the observed change in the balance of discretion between prosecutors and judges in Guidelines cases is simply irrelevant to capital sentencing decisions. I will not here repeat what I have already stated supra at 278-81 as to why it would, in fact, be inappropriate to interject prosecutorial opinions as to sentence into the penalty phase of a capital trial.

[12] The fact that prosecutorial decisions may be informed by some of the same evidence heard by the jury does not, of course, make those decisions themselves admissible. See supra at 279 (explaining why there is no place in the penalty phase of a capital trial for the prosecutor-as-expert).

[1] Judge Chester J. Straub, who voted in the en banc poll but who thereafter took senior status, authorizes me to say that he agrees with the views of this opinion.

[2] According to data collected by the Death Penalty Information Center, there are only nine inmates under federal death sentence in states effectively without their own death penalty law—this includes one inmate in New York and Fell in Vermont. Death Penalty Info. Ctr., The Federal Death Penalty, http:// www.deathpenaltyinfo.org/federal-death-penalty (last visited Apr. 1, 2009).

[3] Vermont has no death penalty. In 2004, the New York Court of Appeals declared New York's death penalty statute invalid on procedural grounds, People v. LaValle, 3 N.Y.3d 88, 783 N.Y.S.2d 485, 817 N.E.2d 341 (2004), and the New York legislature has, to date, declined to reinstate it by curing those procedural problems.

[4] Judge Raggi states that in New York, it was the courts that "invalidat[ed] the state's death penalty statute in light of identified procedural deficiencies" and concludes that "we cannot assume that such a judicial invalidation of death penalty legislation reflects popular opposition to capital punishment." See supra at 274 (Raggi, J., concurring in the denial of rehearing en banc). The most important fact about the case in which New York struck down its death penalty is that it explicitly noted that the legislature was free to cure the statutory defect that made the state's death penalty jury deadlock instruction unacceptable. See People v. LaValle, 3 N.Y.3d 88, 99, 783 N.Y.S.2d 485, 817 N.E.2d 341 (2004). That the New York legislature has declined to pass a new death penalty law must be taken, like Vermont's failure to enact capital punishment, as some evidence of the popular will, at least today.

[5] Cf. Witherspoon, 391 U.S. at 519-20, 88 S. Ct. 1770 ("[I]n a [jurisdiction] less than half of whose people believe in the death penalty, a jury composed exclusively of such people cannot speak for the community."). Witherspoon spoke of the nation not jurisdiction, because the High Court, at the time, believed that what it said was true of the nation as a whole. Regardless of whether the Court was correct in that belief, the statement applies directly to a jurisdiction like Vermont.

[6] Many of the objections raised by the concurrers to making this information available to the jury apply just as much to letting the jury know that Fell had admitted guilt in the course of negotiations with the prosecutors. In any event, negating the impression that local prosecutors were unwilling to accept a sentence of life imprisonment is anything but "cumulative."

[7] In Penry v. Lynaugh, 492 U.S. 302, 109 S. Ct. 2934, 106 L. Ed. 2d 256 (1989), the Supreme Court recognized that mitigating evidence (in that case, mental retardation and a history of abuse) can be "a two-edged sword: it may diminish his blameworthiness for his crime even as it indicates that there is a probability that he will be dangerous in the future," id. at 324, 109 S. Ct. 2934, but nonetheless found that resentencing was necessary "in the absence of instructions informing the jury that it could consider and give effect to the mitigating evidence of Penry's mental retardation and abused background by declining to impose the death penalty." Id. at 328, 109 S. Ct. 2934; see also Williams v. Taylor, 529 U.S. 362, 396, 120 S. Ct. 1495, 146 L. Ed. 2d 389 (2000) (noting possible double-edged nature of mitigation, but nonetheless finding counsel ineffective for failing to investigate and present it). Later, of course, the Supreme Court effectively superseded this part of Penry by holding that the execution of the mentally retarded violated the Eighth Amendment. Atkins v. Virginia, 536 U.S. 304, 122 S. Ct. 2242, 153 L. Ed. 2d 335 (2002).

[8] There are, I think, other problematic aspects of the decision which would not, in themselves, warrant going en banc but which would have to be dealt with if we did rehear the case. These include whether the prosecutor impermissibly suggested that Fell's mitigating evidence needed to have some nexus to the murder, whether Juror 141 was properly excluded (see Petition for Rehearing at 4-7), and whether the cumulative effect of any errors undermines our confidence in the fairness of the proceeding.

[9] In one recent federal case, a jury in New York refused to sentence to death a drug lord convicted of multiple murders. "Three jurors said they voted for life without parole because had the trial been held in state court, the death penalty would not have been an option." Drug Lord Sentenced to Life Without Parole, N.Y.L.J., Feb. 23, 2007, at 1. Perhaps that is the reason why New York juries have returned only one death sentence from all the federal capital prosecutions sought in New York. See Frederic Block, A Slow Death, N.Y. Times, Mar. 15, 2007, at A27 (estimating forty-two authorizations by attorneys general); Feuer, supra (stating that federal prosecutors had actually sought death sentences in nineteen cases). The authors agree that just one capital sentence resulted.

[10] See supra, note 5 (explaining that, contrary to the District Court's holding, the evidence of local prosecutors' preferences was not, in this case, "cumulative").

[11] As explained above, see supra at 268, this does not threaten the principle that "the Constitution must apply equally throughout the states," nor does it in any way suggest that a voir dire should be less rigorous in Texas than in Vermont. See supra at 278-81 (Raggi, J., concurring in the denial of rehearing en banc). Recognizing the effect of local values on what a potential juror's answer means is just that, no more and no less.

[12] The concurrers suggest that federalism cannot affect the choice of jurors in civil rights, environmental, or gun trafficking cases. See supra at 270 (Raggi, J., concurring in the denial of rehearing en banc). But of course it does, always has, and always must in a federalism as diverse as ours. To assert otherwise is to ignore reality, a reality that recognizes the tension between the centralizing effect of, inter alia, the Fourteenth Amendment and the necessity of national rules on the one hand, and the profoundly ingrained value differences among sovereign states on the other. The Constitution gives local juries a primary role in resolving that tension. In the areas cited by the concurrers, no less than in that of capital punishment, jurors are required to follow the national law, but when, perhaps less often in these areas than in capital sentencing, issues of judgment arise they will properly do so in light of local values.

[13] The concurring opinion focuses on whether capital punishment is unusual in New York. That, however, is not the issue I am addressing. The question I am concerned with is whether the application of capital punishment for federal crimes committed in states that do not have the death penalty is sufficiently "rare"—considering the country as a whole—as to be constitutionally "unusual." That is the question that is properly analogous to the question of whether the execution of minors or the mentally retarded was sufficiently unusual as to be constitutionally barred even in states which might be inclined to impose it with some frequency.

[14] It also suggests that it was not the characteristics of the New York victims that led to life sentences in the federal cases. See supra at 275-76 (Raggi, J., concurring in the denial of rehearing en banc). And that speculation by the concurring opinion is not made less troubling by the fact that it is supported by a citation to a work which I also cite, but for very different factual assertions. One does not normally cite a work for propositions one does not think appropriate.

[*] But cf. Ross v. Rell, 398 F.3d 203, 205 (2d Cir.2005) (per curiam) (vacating district court's temporary restraining order on Connecticut's imposition of death penalty, but staying order for two days to allow for further review), temporary stay vacated by No. 04A663, 543 U.S. 1134 (Jan. 28, 2005); Ross ex rel. Smyth v. Lantz, 396 F.3d 512, 514 (2d Cir. Jan.25, 2005) (per curiam) (denying motion to vacate stay of execution with respect to Connecticut's imposition of death penalty and dismissing appeal for determination of whether person who brought the relevant habeas petition had standing to do so); but see Lantz v. Ross, No. 04A656, 543 U.S. 1134 (Jan. 27, 2005) (vacating stay of execution).