dissenting from the denial of rehearing and rehearing en banc:
“[T]he way in which we choose those who will die reveals the depth of moral commitment among the living.” McCleskey v. Kemp, 481 U.S. 279, 344, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987) (Brennan, J., dissenting). It reveals our commitment to the Constitution’s bar on cruel and unusual punishment and to the Founding principle that all people are endowed with certain rights that are entitled to dignity and respect; a commitment that endures even as we pursue our legitimate interest in exacting retribution for, and deterring future commission of, the most heinous crimes. Our commitment is undermined, however, when we distinguish those who live from those who die on the basis of arbitrary factors that bear no relation to the goals underlying capital punishment.
In his appeal, Carlos Caro challenges the constitutionality of eligibility factors that select those who die on the basis of prior convictions for relatively minor, nonviolent drug offenses. See 18 U.S.C. § 3592(c)(10) & (12). These provisions have never before been used as the sole means to establish a defendant’s death eligibility. And no state has ever attempted to apply analogous eligibility factors to any other, potential capital defendant.
As I explain at length in my dissent from the panel opinion, see United States v. Caro, 597 F.3d 608, 636 (4th Cir.2010) (Gregory, J., dissenting), these factors fail to distinguish Caro from other murderers in “an objective, even-handed, and substantively rational way,” Zant v. Stephens, 462 U.S. 862, 879, 103 S.Ct. 2733, 77 L.Ed.2d 235 (1983). Fundamentally, eligibility factors that select defendants on the basis of nonviolent conduct are so detached from the constitutional justifications for capital punishment that they cannot be reconciled with the Eighth Amendment. Caro, 597 F.3d at 638-39 (Gregory, J., dissenting) (citing and explaining cases). Likewise, where eligibility factors could apply to millions of offenders but are applied, in actuality, to only one, there is an intolerably high risk that the death penalty is being applied arbitrarily and capriciously. Id. at 642.
The majority’s dismissal of these serious constitutional concerns with little more than “drugs are bad,” id. at 624 (majority opinion) (“Moreover, the felony drug offenses described by § 3592(c)(10) and (12) are serious indeed, however common may be their commission.”), is remarkably tone deaf. The fact that Congress targeted minor, nonviolent drug offenders for death-eligibility, but not any other class of *103nonviolent offenders, does not reduce the constitutional problems in Caro’s death sentence. It amplifies them.
Only two states’ death-penalty statutes can arguably be read to enhance a sentence from life in prison to death on the basis of a defendant’s prior, nonviolent drug convictions.1 And outside the death-penalty context, there is mounting and sustained criticism of laws that “treat[] a defendant who has committed a series of relatively minor and nonviolent drug crimes more severely than a murderer, and that take[ ] no account of the seriousness of the predicate crimes.” United States v. Pruitt, 502 F.3d 1154, 1167 (10th Cir.2007) (McConnell, J., concurring). Indeed, there is a growing consensus in our society — as reflected in the actions of our elected representatives and governing institutions — that even exceedingly harsh prison sentences for nonviolent drug offenders go beyond what is necessary to achieve the goals of sentencing and, by extension, create arbitrary disparities between offender classes. Cf. Kimbrough v. United States, 552 U.S. 85, 95-100, 128 S.Ct. 558, 169 L.Ed.2d 481 (2007) (discussing criticism of crack-powder disparity in federal sentencing guidelines).
Acting in response to heavy criticism of sentencing policy by judges, academics, and other members of the public, the Attorney General recently established a commission to reevaluate the application of harsh, mandatory sentences for nonviolent drug offenders.2 Of particular concern to the Department of Justice is data demonstrating that increasing sentencing disparities are “correlated with the demographics of offenders,” and that those unwarranted disparities arise from the fact that so many drug offenders are subject to mandatory-minimum sentences while even the most serious white-collar offenders are not.3
For its part, the United States Sentencing Commission has taken several recent steps to address the acknowledged unfairness in the application of the sentencing guidelines to nonviolent drug offenders. The Commission reduced the guideline range for offenders convicted of crack cocaine offenses based on its belief “that the 100-to-l [disparity between crack and powder cocaine sentences] significantly undermines various congressional objectives set forth in the Sentencing Reform Act and elsewhere.” U.S. Sentencing Guidelines Manual SuppApp. C, Arndt. 706. It also recommended that Congress repeal the 100-to-l, crack-powder disparity. Id. Congress responded by significantly narrowing the disparity in a bill signed by President Obama on August 3, 2010. See Fair Sentencing Act of 2010, Pub.L. No. 111-220 (July 28, 2010). Finally, the Commission introduced several changes to the guidelines, scheduled to go into effect this November, which would provide district courts with greater leeway to fashion downward departures and substitute treatment programs for prison terms for many nonviolent drug offenders. See U.S. Sentencing Guidelines Manual proposed Arndt. 1.
This same trend towards abolishing disproportionately high prison sentences for nonviolent drug offenders is taking hold at the state level. New York, for instance, repealed the so-called “Rockefeller Drug *104Laws,” by eliminating mandatory-minimum sentences for most nonviolent users and distributors and increasing judges’ discretion to sentence these offenders to treatment rather than prison.4 Lawmakers explained that repeal was necessary because the harshest provisions were being applied to low-level offenders and not kingpins;5 because treatment is often a more effective and efficient way of combating the drug epidemic that continues to plague society;6 and because mandatory-minimum sentences disproportionately targeted blacks and Hispanics.7 New Jersey likewise repealed its law imposing mandatory-minimum sentences for nonviolent drug dealers who distribute drugs within “school zones.”8 Legislators noted that a law originally intended to target drug dealers who preyed on children was not doing so; instead, it was ensnaring low-level offenders, ninety-six percent of whom were black or Hispanic, who lived in the poorest areas of the state.9 And Michigan, which once had arguably the nation’s harshest drug laws, repealed its mandatory-minimum sentencing scheme several years ago, in large part out of concern that the laws were predominantly affecting low-level carriers and drug mules.10
Factors that our political institutions view as deeply troubling when used to dramatically enhance an individual’s prison term, are constitutionally intolerable when used to justify a death sentence. See California v. Ramos, 463 U.S. 992, 998-99, 103 S.Ct. 3446, 77 L.Ed.2d 1171 (1983) (explaining that Supreme Court “has recognized that the qualitative difference of death from all other punishments requires a correspondingly greater degree of scrutiny of the capital sentencing determination”); Lockett v. Ohio, 438 U.S. 586, 604-05, 98 S.Ct. 2954, 57 L.Ed.2d 973 (1978) (explaining that while the judiciary generally defers to legislative policy choices in non-capital sentencing, the death penalty’s qualitative difference requires that those choices be subjected to heightened scrutiny in the capital punishment context). The death penalty may only be applied to those defendants who can “with reliability be classified among the worst offenders.” Roper v. Simmons, 543 U.S. 551, 569, 125 S.Ct. 1183, 161 L.Ed.2d 1 (2005). It may only be imposed to advance the state’s interest in retribution and deterrence— interests that are not sufficiently furthered even when applied to the average murderer. Id. at 571, 125 S.Ct. 1183; Atkins v. Virginia, 536 U.S. 304, 319, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002). It may not be applied “wantonly” and “freakishly” to certain offender classes. Furman v. Georgia, 408 U.S. 238, 310, 92 S.Ct. 2726, 33 *105L.Ed.2d 346 (1972) (Stewart, J., concurring). And it may not be applied in such a way as to create a “ ‘constitutionally unacceptable’ ” risk that racial prejudice enters into the capital sentencing process. McCleskey, 481 U.S. at 309, 107 S.Ct. 1756 (majority opinion) (quoting Turner v. Murray, 476 U.S. 28, 36 n. 8, 106 S.Ct. 1683, 90 L.Ed.2d 27 (1986)); see Graham v. Collins, 506 U.S. 461, 479-86, 113 S.Ct. 892, 122 L.Ed.2d 260 (1993) (Thomas, J., concurring) (discussing the way in which the Supreme Court’s capital jurisprudence has been motivated, in large degree, by the need to eliminate significant racial disparities in the death penalty’s application); Furman, 408 U.S. at 249-50, 92 S.Ct. 2726 (Douglas, J., concurring) (finding the death penalty unconstitutional as applied to the defendants because “[t]he death sentence is disproportionately imposed and carried out on the poor, the Negro, and the members of unpopular groups”).
Carlos Caro was a drug addict and a drug mule. His father and uncles raised him to join the family business, smuggling marijuana and cocaine across the Mexican border. In many ways, Carlos Caro never had a chance. Before going to prison, however, there is no indication that he was ever violent. And though his murder, like all murders, was heinous and despicable, the only difference between his crime and “average” murders for which death may not be prescribed, Atkins, 536 U.S. at 319, 122 S.Ct. 2242, is his history with drugs.
How can a factor that society has begun to reject as a basis for extraordinarily long prison sentences now, for the first time, be the sole ground to enhance a prison term to a death sentence? How can a factor that we recognize substantially over-punishes low-level, nonviolent offenders be compatible with the requirement that the death penalty only be applied to the worst of the worst? And how can we accept a factor that we know risks distinguishing between offenders based on irrational and pernicious characteristics?
The majority’s refusal to grapple with, let alone answer, any of these questions is telling. I am disappointed that the Court allows this silence to control.
. N.H.Rev.Stat. Ann. § 630:1 (2010); La. Code Crim. Proc. Ann. art. 905.4(A)(11) (2010).
. See Sally Quillian Yates, United States Attorney for the Northern District of Georgia, Testimony of the United States Department of Justice, Mandatory Minimum Sentencing Statutes, Before the United States Sentencing Commission 1-3 (May 27, 2010).
. See id. at 7.
. Jeremy W. Peters, Albany Reaches Deal to Repeal '70s Drug Laws, N.Y. Times, March 26, 2009, at Al; Press Release, Office of the Governor of New York, Governor Paterson Signs Rockefeller Drug Reforms Into Law (Apr. 24, 2009), available at http://www.state.ny.us/ governor/ printable/press_042409 l_printable.html [hereinafter Press Release ].
. See Madison Gray, New York's Rockefeller Drug Laws, Time Magazine, Apr. 02, 2009 (quoting New York Governor David Paterson).
. See Press Release, supra note 3 (statement of Senator Ruth Hassell-Thompson).
. See id. (statement of Assemblyman Jeffrion L. Aubry).
. Lisa Fleisher, School-Zone Law Relaxed, Star-Ledger, Jan. 13, 2010, at 028.
. Chris Megerian, Bill Eases Some School Zone Drug Penalties; Action in Jersey Senate Praised as a Lifesaver and Blasted as Going Soft on Crime, Star-Ledger, Dec. 11, 2009, at 001.
. Associated Press, Michigan to Drop Minimum Sentence Rules for Drug Crimes, N.Y. Times, Dec. 26, 2002, at A26.