[*] Richard and Anita Calkins Distinguished Professor of Law, Drake Law School. B.A., Illinois Wesleyan University, 1975; J.D., Harvard University, 1978. The author is especially grateful for the careful review and thoughtful comments of Professor David C. Baldus (which is not to imply that Professor Baldus agrees with any of the author's particular substantive points). The author, with the guidance of Professor Baldus, has made every effort to avoid misinterpreting or misusing any of the statistical data mentioned herein. Any errors, of course, are solely attributable to the author. The author also thanks his secretary, Karla Westberg, and his research assistants, Julie Greteman and Kristin Mueller, for their diligent assistance. Return to text.

[1] Carol S. Steiker & Jordan M. Steiker, Sober Second Thoughts: Reflections on Two Decades of Constitutional Regulation of Capital Punishment, 109 HARV. L. REV. 355 (1995). The Steiker article has been cited in the popular press. See Ted Gest, A House Without a Blueprint: After 20 Years, the Death Penalty Is Still Being Meted Out Unevenly, U.S. NEWS & WORLD REP., July 8, 1996, at 41 (quoting Steiker and Steiker for the proposition that Supreme Court death penalty jurisprudence has "grown like a house without a blueprint—with a new room here, a staircase there, but without the guidance of a master builder"). See also RAYMOND PATERNOSTER, CAPITAL PUNISHMENT IN AMERICA at xv (1991) ("[A] substantial number of death sentences continue to be imposed in a fashion that can only be described as 'freakish.' "); Daniel Givelber, The New Law of Murder, 69 IND. L.J. 375, 378 (1994) ("The new law [of capital] murder is a failure. By most measures, it has marginally reduced but by no means eliminated arbitrariness from capital punishment."). Return to text.

[2] See Steiker & Steiker, supra note 1, at 364. Return to text.

[3] See id. Return to text.

[4] Id. at 426. Return to text.

[5] Id. at 364. Return to text.

[6] The article is particularly effective in showing how easily states can fulfill the requirements the Court has set, see id. at 402, and at illuminating roads not taken by the Court that might have had better results, see id. at 414-26. Return to text.

[7] Steiker and Steiker never suggest that any of these goals are viewed by the Court as more important than any other, and in fact may suggest at one point that minimizing underinclusion is in fact the Court's primary goal: "Each of the three concerns or commitments (desert, fairness, and individualization) reflects different facets of the basic norm of equal treatment, the idea that like cases should be treated alike." Id. at 369. Return to text.

[8] In this context, as in most others, it is dangerous to view "the Court" as an entity with a single mind, when in fact it is composed of nine members who often have diametrically opposed opinions. As to the death penalty, as Steiker and Steiker point out, "the basic configuration on Eighth Amendment issues remained constant for two decades after Furman: two unwavering poles competed for the center." Id. at 428 (citing Furman v. Georgia, 408 U.S. 238 (1972)). On one pole were former Justices Brennan and Marshall, who advocated complete abolition of the death penalty. See id. at 427. On the other pole were former Chief Justice Burger and current Chief Justice Rehnquist (joined in more recent days by Justices Scalia and Thomas), who oppose significant regulation of the death penalty through the Eighth Amendment. See id. It was thus the remaining "centrist" justices like Blackmun (until late in his career, when he became a virtual abolitionist), O'Connor, Powell, Stevens, and White, whose votes controlled the outcomes of cases. See id. at 428. Thus, when I refer to "the Court," I am referring to that centrist bloc as augmented by whatever votes they could pull from the two poles on particular issues. Despite the cobbled-together nature of this body of doctrine, Steiker and Steiker acknowledge that the Court has promulgated some consistent themes, see id. at 364, and I agree (although I do not agree with Steiker and Steiker concerning what all those themes are). Return to text.

[9] 408 U.S. 238 (1972) (per curiam) (abolishing then-existing state death penalty laws). Return to text.

[10] See infra Part III.B.1-2. Return to text.

[11] See infra Part III.C. Return to text.

[12] Steiker & Steiker, supra note 1, at 364 (citing Furman v. Georgia, 408 U.S. 238 (1972); Gregg v. Georgia, 428 U.S. 153 (1976); Proffitt v. Florida, 428 U.S. 242 (1976); Jurek v. Texas, 428 U.S. 262 (1976); Woodson v. North Carolina, 428 U.S. 280 (1976); Roberts v. Louisiana, 428 U.S. 325 (1976)). Return to text.

[13] Id. Return to text.

[14] Id. at 372. Return to text.

[15] See id. Return to text.

[16] Id. at 366. Return to text.

[17] Id. Return to text.

[18] See id. at 369. Return to text.

[19] Id. at 371 (emphasis added). Return to text.

[20] See id. at 373-75. Return to text.

[21] See id. at 375-78. Return to text.

[22] See id. at 375-76 (citing Coker v. Georgia, 433 U.S. 584 (1977)). Return to text.

[23] Id. at 378-82. Return to text.

[24] See id. at 381. Return to text.

[25] See id. at 380-82. Return to text.

[26] See id. at 389-96. Return to text.

[27] See id. at 390-92. Return to text.

[28] See id. at 393-95. Return to text.

[29] See id. at 397 (citing Turner v. Murray, 476 U.S. 28, 37 (1986) (permitting voir dire concerning racial prejudice in cases involving interracial murders); Gardner v. Florida, 430 U.S. 349, 357-62 (1977) (invalidating a death sentence based in part upon a presentence report not made available to defense counsel); Caldwell v. Mississippi, 472 U.S. 320, 328-30 (1985) (preventing prosecutors from arguing that the jury's decision is not the final one concerning the death sentence since the jury verdict is subject to appellate review); Beck v. Alabama, 447 U.S. 625, 637-38 (1980) (requiring the inclusion of lesser-included offense instructions that would support a guilty verdict for a noncapital offense); Simmons v. South Carolina, 512 U.S. 154, 163-64 (1994) (permitting defendant to inform jury that a "life" sentence means life without parole); Johnson v. Mississippi, 486 U.S. 578, 584-87 (1988) (overturning a sentence based upon a prior conviction later invalidated); Herrera v. Collins, 506 U.S. 390, 417 (1993) (post-trial judicial consideration of newly discovered evidence may be constitutionally required in a truly compelling case)). Return to text.

[30] Steiker & Steiker, supra note 1, at 397. Return to text.

[31] See id. ("It is difficult to imagine a body of doctrine that is much worse—either in its costs of implementation or in its negligible returns—than the one we have now."); id. at 429 (describing the doctrinal structure as "functionally and aesthetically unsatisfying"); id. at 437 (describing the Court's doctrine as "failure as regulation"); id. at 438 ("We are left with the worst of all possible worlds: the Supreme Court's detailed attention to death penalty law has generated negligible improvements over the pre-Furman era, but has helped people to accept without second thoughts—much less 'sober' ones—our profoundly failed system of capital punishment."). Return to text.

[32] See, e.g., David C. Baldus et al., Comparative Review of Death Sentences: An Empirical Study of the Georgia Experiment, 74 J. CRIM. L. & CRIMINOLOGY 661, 664 (1983) ("[I]ndividual death sentences that are excessively severe in comparison to the sentences imposed in factually indistinguishable cases—what we call 'comparatively excessive'—do violate the [E]ighth [A]mendment."); Arnold Barnett, Some Distribution Patterns for the Georgia Death Sentence, 18 U.C. DAVIS L. REV. 1327, 1328 (1985) ("The notion animating proportionality review—one that has been explicitly endorsed by the Supreme Court—is that death sentences cannot be imposed in an arbitrary manner. It is considered objectionable if a given defendant is put to death while, in adjacent counties (or adjacent courtrooms), defendants in virtually the same situation are given prison terms."); PATERNOSTER, supra note 1, at 162-64 (arguing that Furman and Gregg both show that the Court has been concerned with underinclusion). Return to text.

[33] DAVID C. BALDUS ET AL., EQUAL JUSTICE AND THE DEATH PENALTY 14 (1990). Return to text.

[34] Id. Return to text.

[35] The two written works in which BWP set forth the findings that are most pertinent for purposes of this Article are BALDUS ET AL., EQUAL JUSTICE, supra note 33, and Baldus et al., Comparative Review, supra note 32. Return to text.

[36] See infra Parts III.B.1.a, III.B.2. Return to text.

[37] See Barnett, supra note 32, at 1342. The BWP group used complicated regression analysis. As will shortly be apparent, Barnett's methodology involves assigning simple integer values. See infra notes 41-44 and accompanying text. Return to text.

[38] See Barnett, supra note 32, at 1329. In Georgia, there is only one degree of murder. See GA. CODE ANN. § 26-1101(a) (Supp. 1996). Return to text.

[39] See Barnett, supra note 32, at 1335. This is the same group of post-Furman cases analyzed by BWP. See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 94. Return to text.

[40] See Barnett, supra note 32, at 1329. Return to text.

[41] See id. at 1339-41, 1364-66. Barnett's method may seem overly simplistic in focusing on only three variables, but the accuracy of his selection of factors is not crucial because choosing other factors would result in the same sort of continuum with low death sentencing rates for the least culpable cases, midrange death sentencing rates for midrange cases, and high death sentencing rates for the most culpable offenders. See, e.g., infra notes 177-81 and accompanying text (explaining system developed by BWP). Return to text.

[42] See Barnett, supra note 32, at 1341. Barnett's protocols for each of the three variables are reproduced at the end of this Article. See infra Appendix A. Return to text.

[43] See Barnett, supra note 32, at 1342. Return to text.

[44] Barnett's chart is found at Barnett, supra note 32, at 1342. Actually, the chart I have reproduced here is a reformulation of Barnett's chart produced by BWP in BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 95. I chose to use the Baldus reformulation because it is slightly easier to understand. There are a couple of minor discrepancies between the two charts for which I cannot account, but they do not make any difference in the analysis. Return to text.

[45] Furman v. Georgia, 408 U.S. 238, 309 (1972) (Stewart, J., concurring). Return to text.

[46] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 60. Return to text.

[47] See id. at 94-95 (discussing Barnett's work and identifying what would be underinclusive under his system). Return to text.

[48] See Baldus et al., Comparative Review, supra note 32, at 696-97 (citing language of cases quoted infra text accompanying notes 49-55); see also Steiker & Steiker, supra note 1, at 367-68 & nn.44-55 (citing several of those same cases). Return to text.

[49] Furman, 408 U.S. at 256 (Douglas, J., concurring). Return to text.

[50] Id. at 309-10 (Stewart, J., concurring) (footnotes omitted). Return to text.

[51] Id. at 311-12 (White, J., concurring). Return to text.

[52] 428 U.S. 262 (1976). Return to text.

[53] Id. at 276. Return to text.

[54] 428 U.S. 242 (1976). Return to text.

[55] Id. at 252. Return to text.

[56] 428 U.S. 153 (1976). Return to text.

[57] See id. at 188. Return to text.

[58] 446 U.S. 420 (1980). Return to text.

[59] Id. at 432. Return to text.

[60] 430 U.S. 349 (1977). Return to text.

[61] Id. at 361. Return to text.

[62] 462 U.S. 862 (1983). Return to text.

[63] Id. at 879. Return to text.

[64] Steiker & Steiker, supra note 1, at 366. Return to text.

[65] Id. Return to text.

[66] There were 2,484 such convictions in the six-year period following Furman. See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 45. This means there likely were about 2,000 in the five-year period included in Barnett's study. Return to text.

[67] See Baldus et al., Comparative Review, supra note 32, at 680. Return to text.

[68] See Barnett, supra note 32, at 1338. Return to text.

[69] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 44. Return to text.

[70] See id. Return to text.

[71] See id. at 43-45. Return to text.

[72] See id. at 44 (the penalty was sought in about 200 instances, versus about 400 instances in which it was not). For the exact figures, see supra text accompanying notes 69-70. Return to text.

[73] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 88-89 (of the 606 murder convictions studied, 483 were death-eligible). Return to text.

[74] Death penalty cases, if properly defended, cost a great deal more than other murder cases at virtually every stage, from voir dire through sentencing. Since most death-eligible defendants are indigent, the court ends up footing litigation costs for both sides. For horror stories of capital defendants who got bargain-basement representation, see Stephen B. Bright, Counsel for the Poor: The Death Sentence Not for the Worst Crime but for the Worst Lawyer, 103 YALE L.J. 1835 (1994). Return to text.

[75] See GA. CODE ANN. § 17-10-30.1 (Supp. 1996) (permitting sentencing by a judge presumably if jury is waived, although the statute is not entirely clear on that point). Return to text.

[76] See 4 WILLIAM BLACKSTONE, COMMENTARIES *358. Return to text.

[77] See, e.g., In re Winship, 397 U.S. 358, 363 (1970) ("The accused during a criminal prosecution has at stake interest of immense importance, both because of the possibility that he may lose his liberty upon conviction and because of the certainty that he would be stigmatized by the conviction."). Return to text.

[78] See id. at 371-72 (Harlan, J., concurring).

In a civil suit between two private parties for money damages, for example, we view it as no more serious in general for there to be an erroneous verdict in the defendant's favor than for there to be an erroneous verdict in the plaintiff's favor . . . . In a criminal case, on the other hand, we do not view the social disutility of convicting an innocent man as equivalent to the disutility of acquitting someone who is guilty . . . . I view the requirement of proof beyond a reasonable doubt in a criminal case as bottomed on a fundamental value determination of our society that it is far worse to convict an innocent man than to let a guilty man go free.
Id. Return to text.

[79] The Court found the beyond-a-reasonable-doubt standard of proof to be constitutionally required in In re Winship, 397 U.S. 358, 364 (1970). As to the supermajority requirement, the Constitution requires unanimity for conviction by a six-person jury. See Burch v. Louisiana, 441 U.S. 130, 138 (1979). As to a twelve-person jury, the Court has upheld a nine-to-three verdict in favor of conviction. See Johnson v. Louisiana, 406 U.S. 356, 362-63 (1972). However, the Court probably would not approve any lesser supermajority. Johnson was a five-to-four decision, and one of the five members of the majority, Justice Blackmun, stated in his concurrence that any lesser number of votes for conviction would cause him "great difficulty." Id. at 366 (Blackmun, J., concurring). Return to text.

[80] See, e.g., 18 U.S.C.A. § 3553(b) (West Supp. 1996) (providing that a court may depart from the sentencing guidelines due to a "circumstance of a kind, or to a degree, not adequately taken into consideration by the Sentencing Commission in formulating the guidelines that should result in a sentence different from that described"). Return to text.

[81] For discussions of mercy as a desirable aspect of criminal sentencing, see JEFFRIE G. MURPHY & JEAN HAMPTON, FORGIVENESS AND MERCY 20 (1988) ("To be merciful is to treat a person less harshly than, given certain rules, one has a right to treat that person."); Eric L. Muller, The Virtue of Mercy in Criminal Sentencing, 24 SETON HALL L. REV. 288, 335 (1993):

Mercy . . . is an attitude that the sentencer adopts, during the process of selecting a sentence from within a range of authorized punishments, by imagining both the nature of the criminal episode and the impact of the possible sentences from the defendant's perspective. Mercy reminds the judge of how things look from the defense table. Mercy's role in the sentencing process is therefore a limiting or restraining one; it provides a check against the risk that the sentencer will get things wrong by describing the nature of the crime in an imbalanced way or underestimating the impact of the sentence on the defendant.
See also Paul Whitlock Cobb, Jr., Note, Reviving Mercy in the Structure of Capital Punishment, 99 YALE L.J. 389, 391 (1989) (arguing that mercy is a particular virtue in capital sentencing). But see California v. Brown, 479 U.S. 538, 542 (1987) (holding that an instruction requiring jurors to ignore "mere sentiment, conjecture, sympathy, passion, prejudice, public opinion or public feeling" was not unconstitutional). Return to text.

[82] Justice Brennan created the "death is different" argument in Furman v. Georgia, 408 U.S. 238, 282 (1972) (Brennan, J., concurring). Since then, the Court has adopted this precept in part, although by no means to the extent it could have. See Steiker & Steiker, supra note 1, at 397 (discussing the relatively limited scope to which the Court has put the "death is different" doctrine into practice). Return to text.

[83] See ARK. CODE ANN. § 5-4-603(c) (Michie 1993); CAL. PENAL CODE § 190.3(k) (West 1988); CONN. GEN. STAT. ANN. § 53a-46a(f) (West Supp. 1996); GA. CODE ANN. § 17-10-30(c) (Supp. 1996); 720 ILL. COMP. STAT. ANN. 5/9-1 (West Supp. 1996); KAN. STAT. ANN. § 21-4624(e) (1995); KY. REV. STAT. ANN. § 532.025(2) (Michie 1990); LA. CODE CRIM. PROC. ANN. art. 905.6 (West Supp. 1997); MD. CODE ANN., CRIM. LAW § 413 (g), (h) (Supp. 1996); MISS. CODE ANN. § 99-19-101(2) (1994); MO. ANN. STAT. § 565.030 4 (West Supp. 1997); N.H. REV. STAT. ANN. § 630:5 IV (1996); N.J. STAT. ANN. § 2C:11-3(b) (West 1995); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); N.Y. CRIM. PROC. LAW § 400.27 10 (McKinney Supp. 1997); N.C. GEN. STAT. § 15A-2000(b) (1996); OHIO REV. CODE ANN. § 2929.03(D)(2) (Anderson 1996); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150 (Supp. 1994); 42 PA. CONS. STAT. ANN. § 9711(c) (West 1982 & Supp. 1996); S.C. CODE ANN. § 16-3-20(C) (Law. Co-op. Supp. 1996); S.D. CODIFIED LAWS § 23A-27A-4 (Michie 1988); TENN. CODE ANN. § 39-13-204(f) (Supp. 1996); TEX. CRIM. P. CODE ANN. § 37.071 2(e) (West Supp. 1997); UTAH CODE ANN. § 76-3-207(4) (Supp. 1996); VA. CODE ANN. § 19.2-264.4 (Michie 1995); WASH. REV. CODE ANN. § 10.95.060 (West 1990); WYO. STAT. ANN. § 6-2-102(f) (Michie Supp. 1996). Return to text.

[84] See ARK. CODE ANN. § 5-4-603(a) (Michie 1993); CAL. PENAL CODE § 190.4(a) (West 1988); CONN. GEN. STAT. ANN. § 53a-46a (West Supp. 1996); GA. CODE ANN. § 17-10-30 (Supp. 1996); MISS. CODE ANN. § 99-19-103 (1994); N.H. REV. STAT. ANN. § 630:5 IV (1996); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); N.Y. CRIM. PROC. LAW § 400.27 7c (McKinney Supp. 1997); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150 (Supp. 1994); 42 PA. CONS. STAT. ANN. § 9711(c)(iv) (West 1982 & Supp. 1996); S.C. CODE ANN. § 16-3-20(C) (Law. Co-op. Supp. 1996); S.D. CODIFIED LAWS § 23A-27A-4 (Michie 1988); TENN. CODE ANN. § 39-13-204(f)(1), (2) (Supp. 1996); TEX. CRIM. P. CODE ANN. § 37.071 2(d)(2), 2(f)(2) (West Supp. 1997); WYO. STAT. ANN. § 6-2-102(e) (Michie Supp. 1996). Return to text.

[85] See ARK. CODE ANN. § 5-4-603(a) (Michie 1993); CAL. PENAL CODE § 190.4(a) (West 1988); CONN. GEN. STAT. ANN. § 53a-46a (West Supp. 1996); GA. CODE ANN. § 17-10-30 (Supp. 1996); 720 ILL. COMP. STAT. ANN. 5/9-1(g) (West Supp. 1996); KAN. STAT. ANN. § 21-4624(e) (1995); KY. REV. STAT. ANN. § 532.025(3) (Michie 1990); LA. CODE CRIM. PROC. ANN. art. 905.3 (West Supp. 1997); MD. CODE ANN., CRIM. LAW § 413(g)-(i) (Supp. 1996); MISS. CODE ANN. § 99-19-101(2), (3) (1994); MO. ANN. STAT. § 565.030 4 (West Supp. 1997); N.H. REV. STAT. ANN. § 630:5 IV (1996); N.J. STAT. ANN. § 2C:11-3(a) (West 1995); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); N.Y. CRIM. PROC. LAW § 400.27 11(a) (McKinney Supp. 1997); N.C. GEN. STAT. § 15A-2000(b) (1996); OHIO REV. CODE ANN. § 2929.03(D)(1), (2) (Anderson 1996); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150(1)(d) (Supp. 1994); 42 PA. CONS. STAT. ANN. § 9711(c)(1)(iv) (West 1982 & Supp. 1996); S.C. CODE ANN. § 16-3-20(C) (Law. Co-op. Supp. 1996); S.D. CODIFIED LAWS § 23A-27A-4 (Michie 1988); TENN. CODE ANN. § 39-13-204 (f)(1) (Supp. 1996); TEX. CRIM. P. CODE ANN. § 37.071 2(c) (West Supp. 1997); UTAH CODE ANN. § 76-3-207(4) (Supp. 1996); VA. CODE ANN. § 19.2-264.4D(1) (Michie 1995); WASH. REV. CODE ANN. § 10.95.060(4) (West 1990); WYO. STAT. ANN. § 6-2-102(d)(ii) (Michie Supp. 1996). Return to text.

[86] See CAL. PENAL CODE § 190.4(a) (West 1988); Skaggs v. Commonwealth, 694 S.W.2d 672, 681 (Ky. 1985) (discussing that the failure of a jury to reach a verdict will result in the retrial of the sentencing). Return to text.

[87] See ARK. CODE ANN. § 5-4-603 (Michie 1993); CONN. GEN. STAT. ANN. § 53a-46a(f) (West Supp. 1996); GA. CODE ANN. § 17-10-31.1(c) (Supp. 1996); 720 ILL. COMP. STAT ANN. 5/9-1 (West Supp. 1996); KAN. STAT. ANN. § 21-4624(e) (1995); LA. CODE CRIM. PROC. ANN. art. 905.6 (West Supp. 1997); MD. CODE ANN., CRIM. LAW § 413(k)(2) (Supp. 1996); MISS. CODE ANN. § 99-19-103 (1994); MO. ANN. STAT. § 565.030 4 (West Supp. 1997); N.H. REV. STAT. ANN. § 630:5 IV (1996); N.J. STAT. ANN. § 2C:11-3(f) (West 1995); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); N.Y. CRIM. PROC. LAW § 400.27 10 (McKinney Supp. 1997); N.C. GEN. STAT. § 15A-2000(b) (1996); OHIO REV. CODE ANN. § 2929.03(D)(2) (Anderson 1996); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150 (Supp. 1994); 42 PA. CONS. STAT. ANN. § 9711(c)(1)(iv) (West 1982 & Supp. 1996); S.C. CODE ANN. § 16-3-20(C) (Law. Co-op. Supp. 1994); S.D. CODIFIED LAWS § 23A-27A-4 (Michie 1988); TENN. CODE ANN. § 39-13-204 (Supp. 1996); TEX. CRIM. PROC. CODE ANN. § 37.071 (West Supp. 1997); UTAH CODE ANN. § 76-3-207(4) (Supp. 1996); VA. CODE ANN. § 19.2-264.4E (Michie 1995); WASH. REV. CODE ANN. § 10.95.060 (West 1990); WYO. STAT. ANN. § 6-2-102(d)(ii) (Michie Supp. 1996). Return to text.

[88] See ARK. CODE ANN. § 5-4-603(a)(1) (Michie 1993); CAL. PENAL CODE § 190.4(a) (West 1988); GA. CODE ANN. § 17-10-30(c) (Supp. 1996); KAN. STAT. ANN. § 21-4624(e) (1995); KY. REV. STAT. ANN. § 532.025(3) (Michie 1990); LA. CODE CRIM. PROC. ANN. art. 905.3 (West Supp. 1997); MD. CODE ANN., CRIM. LAW § 413(d) (Supp. 1996); MISS. CODE ANN. § 99-19-103 (1994); MO. ANN. STAT. § 565.030 1 (West Supp. 1997); N.H. REV. STAT. ANN. § 630:5 III (1996); N.J. STAT. ANN. § 2C:11-3c(2)(a) (West 1995); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); N.Y. CRIM. PROC. LAW § 400.27 11(a) (McKinney Supp. 1997); N.C. GEN. STAT. § 15A-2000(c)(1) (1996); OHIO REV. CODE ANN. § 2929.03(D)(1) (Anderson 1996); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150 (Supp. 1994); 42 PA. CONS. STAT. ANN. § 9711(c)(iii) (West 1982 & Supp. 1996); S.C. CODE ANN. § 16-3-20(c) (Law. Co-op. Supp. 1996); S.D. CODIFIED LAWS § 23A-27A-5 (Michie 1988); TENN. CODE ANN. § 39-13-204(f)(1) (Supp. 1996); TEX. CRIM. P. CODE ANN. § 37.071 (West Supp. 1997); VA. CODE ANN. § 19.2-264.4C (Michie 1995); WYO. STAT. ANN. § 6-2-102(d)(i)(A) (Michie Supp. 1996). Return to text.

[89] See ARK. CODE ANN. § 5-4-603(a)(2) (Michie 1993); N.J. STAT. ANN. § 2C:11-2c(3)(a) (West 1995); N.Y. CRIM. PROC. LAW § 400.27 11(a) (McKinney Supp. 1997); OHIO REV. CODE ANN. § 2929.03(D)(2) (Anderson 1996); TENN. CODE ANN. § 39-13-204(f)(2) (Supp. 1996); WASH. REV. CODE ANN. § 10.95.060 (West 1990). Return to text.

[90] See CAL. PENAL CODE § 190.3(e) (West 1988); N.H. REV. STAT. ANN. § 630.5 III (1996); 42 PA. CONS. STAT. ANN. § 9711(c)(iii) (West 1982 & Supp. 1996). Return to text.

[91] See ARK. CODE ANN. § 5-4-603(a)(1)-(3) (Michie 1993); N.Y. CRIM. PROC. LAW § 400.27 6 (McKinney Supp. 1997); TENN. CODE ANN. § 39-13-204(f) (Supp. 1996). Return to text.

[92] See GA. CODE ANN. § 17-10-30 (Supp. 1996); KAN. STAT. ANN. § 21-4624(e) (1995); MISS. CODE ANN. § 99-19-101(2) (1994); N.M. STAT. ANN. § 31-20A-3 (Michie 1994); OKLA. STAT. ANN. tit. 21, § 701.11 (West Supp. 1997); OR. REV. STAT. § 163.150 (Supp. 1994); S.C. CODE ANN. § 16-3-2 (Law. Co-op. Supp. 1996); S.D. CODIFIED LAWS § 23A-27A-5 (Michie 1988); TEX. CRIM. P. CODE ANN. § 37.071 (West Supp. 1997); VA. CODE ANN. § 19.2-264.4C (Michie 1995); WYO. STAT. ANN. § 6-2-102 (Michie Supp. 1996). Return to text.

[93] See, e.g., CONN. GEN. STAT. ANN. § 53a-46a (West Supp. 1996) (requiring unanimity as to aggravating circumstances, but the burden is only preponderance of evidence). Six states require a finding of an aggravating circumstance beyond a reasonable doubt, but do not specify that the finding must be unanimous. See KY. REV. STAT. ANN. § 532.025(3) (Michie 1990); LA. CODE CRIM. PROC. ANN. art. 905.3 (West Supp. 1997); MD. CODE ANN., CRIM. LAW § 413(d), (g), (h), (i) (Supp. 1996); MO. ANN. STAT. § 565.030 4 (West Supp. 1997); N.C. GEN. STAT. § 15A-2000(b), (c) (1996); VA. CODE ANN. § 19.2-264.4C (Michie 1995). Three states require that an aggravating circumstance be found unanimously beyond a reasonable doubt and that the aggravating circumstances be found to outweigh the mitigating circumstances. See CAL. PENAL CODE § 190.3(k) (West 1988); N.H. REV. STAT. ANN. § 630.5 IV (1995); 42 PA. CONS. STAT. ANN. § 9711(c)(iv) (West 1982 & Supp. 1996). Two states require a finding of an aggravating circumstance beyond a reasonable doubt but do not specify that the finding be unanimous and require that the aggravating circumstances be found to outweigh the mitigating circumstances beyond a reasonable doubt. See N.J. STAT. ANN. § 2C:11-3c(3) (West 1995); OHIO REV. CODE ANN. § 2929.03 (Anderson 1996). Illinois requires unanimity regarding the existence of an aggravating circumstance. See 720 ILL. COMP. STAT. ANN. 5/9-1 (West Supp. 1996). Washington requires a finding that the aggravating circumstances outweigh the mitigating circumstances beyond a reasonable doubt. See WASH. REV. CODE ANN. § 10.95.060(4) (West 1990). Only Utah employs as the sole underinclusion device that the jury's sentencing recommendations be unanimous (i.e., does not require a finding by the jury of an aggravating circumstance by any particular burden of proof or number of jurors, and does not require any particular weighing of the aggravating and mitigating circumstances). See UTAH CODE ANN. § 76-3-207 (Supp. 1996). Return to text.

[94] See NEV. REV. STAT. ANN. §§ 175.554-.556 (Michie Supp. 1995). Return to text.

[95] See id. § 175.556(2). Return to text.

[96] See COLO. REV. STAT. ANN. § 16-11-103 (West Supp. 1996). Return to text.

[97] In four of these states, the jury renders an advisory verdict, but the judge has the power to override the jury's recommendation. See ALA. CODE § 13A-5-46(e) (1994); DEL. CODE ANN. tit. 11, § 4209 (1995); FLA. STAT. § 921.141(2) (1995); IND. CODE ANN. § 35-50-2-9(e) (Michie Supp. 1996). In the other four, the judge determines the sentence without any jury involvement. See ARIZ. REV. STAT. ANN. § 13-703E (West Supp. 1996); IDAHO CODE § 19-2515(c) (Supp. 1996); MONT. CODE ANN. § 46-18-305 (1995); NEB. REV. STAT. § 29-2522 (1995). Return to text.

[98] See DEL. CODE ANN. tit. 11, § 4209(d)(1)a (1995); IND. CODE ANN. § 35-50-2-9(i)(1) (Michie Supp. 1996). Return to text.

[99] See DEL. CODE ANN. tit. 11, § 4209(d)(1)b (1995). Return to text.

[100] See ALA. CODE § 13A-5-47(e) (1994); IND. CODE ANN. § 35-50-2-9(k) (Michie Supp. 1995); NEB. REV. STAT. § 29-2521 (1995). Return to text.

[101] See ARIZ. REV. STAT. ANN. § 13-703E (West Supp 1996); FLA. STAT. § 921.141 (1995); IDAHO CODE § 19-2515(c) (Supp. 1996); MONT. CODE ANN. § 46-18-305 (1995). In all of these states, the mitigating circumstances must outweigh the aggravating circumstances for the defendant to avoid a death sentence. Return to text.

[102] See, e.g., James Luginbuhl & Julie Howe, Discretion in Capital Sentencing Instructions: Guided or Misguided?, 70 IND. L.J. 1161, 1165-76 (1995) (discussing evidence of juror misapplication of instructions concerning burden of proof regarding aggravating and mitigating circumstances and unanimity requirement regarding finding of aggravating and mitigating circumstances). However, jurors can hardly miss the requirement that their ultimate decision to impose a death sentence must be unanimous. Return to text.

[103] See, e.g., BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 235.

In comparing post- Furman death sentencing rates among the states, it is worth noting that the states whose statutes require death sentencing by the judge rather than by the jury tend to have the highest rates in the region (e.g., Indiana in the East North Central region, Florida in the South Atlantic region, Alabama in the East South Central region, and Arizona and Idaho in the Mountain region).
Id.; see also Harris v. Alabama, 115 S. Ct. 1031, 1036 (1995) (noting that according to statistics compiled by the Alabama Prison Project, there have been only five cases in Alabama in which the judge rejected an advisory verdict of death and instead imposed life imprisonment, compared to 47 instances where the judge imposed a death sentence over a jury recommendation of life imprisonment); Stephen B. Bright & Patrick J. Keenan, Judges and the Politics of Death: Deciding Between the Bill of Rights and the Next Election in Capital Cases, 75 B.U. L. REV. 759 (1995) (exploring and decrying the propensity of judges to resort to death sentences in view of increasing public and political pressure). Return to text.

[104] See GA. CODE ANN. §§ 17-10-30, -31.1 (Supp. 1996). Return to text.

[105] See supra note 44 and accompanying text. Return to text.

[106] See sources cited supra notes 83-92. Return to text.

[107] The Court upheld sentencing by a one-person sentencer when it refused to find sentencing by the trial judge unconstitutional in Proffitt v. Florida, 428 U.S. 242, 259-60 (1976). The extent to which the Court might permit a death sentence to be imposed by a nonunanimous jury has never been tested, because all states with jury sentencing require a unanimous verdict. The Court has not yet decided whether a unanimous verdict is required at the guilt phase of a case in which the death sentence may be sought upon conviction. See Schad v. Arizona, 501 U.S. 624, 630 (1991). The Court also has not required that the beyond-a-reasonable-doubt standard be employed during capital sentencing, as is evidenced by the Court's upholding of state schemes that do not embody that requirement. See supra notes 82, 85, 92-93 and accompanying text (states not employing the beyond-a-reasonable-doubt standard in capital sentencing). Return to text.

[108] Examples include witnesses changing stories, co-defendants getting deals, and prosecutors' offices being short of resources. Return to text.

[109] See Bordenkircher v. Hayes, 434 U.S. 357, 364 (1978).

[S]o long as the prosecutor has probable cause to believe that the accused committed an offense defined by statute, the decision whether or not to prosecute, and what charge to file or bring before a grand jury generally rests entirely in his discretion [but the decision to prosecute may not be] . . . "deliberately based upon an unjustifiable standard such as race, religion, or other arbitrary classification."

Id. (quoting Oyler v. Boles, 368 U.S. 448, 456 (1962)). Return to text.

[110] 447 U.S. 625 (1980). Return to text.

[111] Id. at 637. Return to text.

[112] 391 U.S. 510 (1968). Although Steiker and Steiker begin their analysis with the 1972 Furman case, see Steiker & Steiker, supra note 1, at 363, I believe the 1968 Witherspoon case is the appropriate starting point for considering the Court's regulation of states' administration of the death penalty. Return to text.

[113] 391 U.S. at 512. Return to text.

[114] Id. at 512-23. Return to text.

[115] Id. at 519-21. Return to text.

[116] 428 U.S. 280 (1976). Woodson held unconstitutional a statute that made the death penalty mandatory for every defendant convicted of first-degree murder or felony murder. See id. at 301. Return to text.

[117] 428 U.S. 325 (1976). Roberts held unconstitutional a Louisiana statute that imposed a mandatory death sentence on five narrowly defined categories of first-degree murder. See id. at 335-36. Return to text.

[118] The Court appeared to reached the logical end of this line in Sumner v. Shuman, 483 U.S. 66, 85 (1987) (holding mandatory death penalty for murder committed by a person serving a life sentence unconstitutional). Return to text.

[119] 438 U.S. 586 (1978). Lockett held unconstitutional an Ohio statute that limited the categories of mitigating evidence that could be considered by the sentencer. See id. at 608. Return to text.

[120] 455 U.S. 104 (1982). Eddings held that it was unconstitutional for a sentencer to refuse, as a matter of law, to consider all relevant mitigating evidence. See id. at 112. Return to text.

[121] 486 U.S. 367 (1988). Return to text.

[122] See id. at 384. Return to text.

[123] See id. Return to text.

[124] 472 U.S. 320 (1985). Return to text.

[125] See id. at 325. Return to text.

[126] Id. at 341. Return to text.

[127] Id. at 330. Return to text.

[128] Id. (quoting Woodson v. North Carolina, 428 U.S. 280, 304 (1976)). Return to text.

[129] 481 U.S. 279 (1987). Return to text.

[130] See Steiker & Steiker, supra note 1, at 400-01. Return to text.

[131] The most famous portion of the opinion deals with McCleskey's claim of invidious discrimination. For a discussion of this claim and the Court's treatment of it, see infra notes 138-141 and accompanying text. Return to text.

[132] See McCleskey, 481 U.S. at 306-07. Return to text.

[133] Id. Return to text.

[134] Id. at 307 (quoting Gregg v. Georgia, 428 U.S. 153, 199 (1976)). Return to text.

[135] Id. at 308 (citations omitted) (emphasis added). Return to text.

[136] See supra Part II.C. Return to text.

[137] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 140-49 (discussing this racial phenomenon). Return to text.

[138] David C. Baldus et al., Law and Statistics in Conflict: Reflections on McCleskey v. Kemp, in HANDBOOK OF PSYCHOLOGY AND LAW 251, 259 (D.K. Kagehiro & W.S. Laufer eds., 1992). Return to text.

[139] See McCleskey, 481 U.S. at 298 ("[M]cCleskey would have to prove that the Georgia legislature enacted or maintained the death penalty statute because of an anticipated racially discriminatory effect.") (emphasis added). Return to text.

[140] See id. Return to text.

[141] Id. at 311-12 (citations omitted). Return to text.

[142] The Court may have missed the golden opportunity in McCleskey to pay more than lip service to this goal. Return to text.

[143] See Lockett v. Ohio, 438 U.S. 586, 604 (1978).

[T]he Eighth and Fourteenth Amendments require that the sentencer, in all but the rarest kind of capital case, 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.
Id. (emphasis added). Return to text.

[144] See Steiker & Steiker, supra note 1, at 369-70. Return to text.

[145] See id. at 370-71. Return to text.

[146] See id. at 402.

[C]ontemporary death penalty law is remarkably undemanding. The narrowing, channeling, and individualization requirements can be simultaneously and completely satisfied by a statute that defines capital murder as any murder accompanied by some additional, objective factor or factors and that provides for a sentencing proceeding in which the sentencer is asked simply whether the defendant should live or die.
Id. Return to text.

[147] See Lowenfield v. Phelps, 484 U.S. 231, 246 (1988).

[T]he narrowing function required for a regime of capital punishment may be provided in either of these two ways: The legislature may itself narrow the definition of capital offenses, as Texas and Louisiana have done, so that the jury finding of guilty responds to this concern, or the legislature may more broadly define capital offenses and provide for narrowing by jury findings of aggravating circumstances at the penalty phase.
Id. Return to text.

[148] See Lockett, 438 U.S. at 608 (invalidating a statute that limited the kinds of mitigating evidence the sentencer could consider). Return to text.

[149] See Caldwell v. Mississippi, 472 U.S. 320, 329-30 (1985) (reversing a death sentence because the prosecutor and judge had indicated to the jury that their decision was not final due to appellate review, thereby possibly detracting from the jury's awareness that it was exercising a "truly awesome responsibility"). Return to text.

[150] See 438 U.S. at 605. Return to text.

[151] The only exception is that the Court outlawed the death penalty for rape in Coker v. Georgia, 433 U.S. 584, 592 (1977). Return to text.

[152] See Steiker & Steiker, supra note 1, at 402 ("[C]ontemporary death penalty law is remarkably undemanding.") Return to text.

[153] See id. at 392 (noting the "near completeness" of the return to pre-Furman discretion with respect to mitigating factors). Return to text.

[154] See, e.g., id. at 402 (bemoaning "the fact of minimal regulation, which invites if not guarantees the same kinds of inequality as the pre-Furman regime"). Return to text.

[155] See discussion infra Part III.A. Return to text.

[156] See supra note 154 and accompanying text. Return to text.

[157] Steiker & Steiker, supra note 1, at 426. Return to text.

[158] At many points, Steiker and Steiker assert that state systems have the potential to operate arbitrarily. See, e.g., id. at 375 ("[T]he continuing failure of states to narrow the class of death-eligible invites the possibility that some defendants will receive the death penalty in circumstances in which it is not deserved according to wider community standards (overinclusion)."); id. at 378 ("[T]he fear of overinclusive application of the death penalty that accounted in part for the Court's decision to enter the constitutional thicket remains quite justified."); id. at 381-82 ("Narrowing the class of the death-eligible in no way addresses the problem of [overinclusion], because open-ended discretion after death-eligibility permits, even invites, the jury to act according to its own unaccountable whims."); id. at 391-92 ("Although such discretion cannot be used to render a defendant death-eligible contrary to community standards, it can be used to exempt favored defendants from the death penalty or to withhold severe punishment for crimes against despised victims."); id. at 402 ("And the fact of minimal regulation, which invites if not guarantees the same kinds of inequality as the pre-Furman regime, is filtered through time-consuming, expensive proceedings that ultimately do little to satisfy the concerns that led the Court to take a sober second look at this country's death penalty practices in the first place."); id. at 417-18 ("[A]llowing states to seek the death penalty against all offenders in these categories presents a real and substantial danger that many offenders will be selected for execution who do not 'deserve' it (and who will therefore be treated more harshly than many offenders who do 'deserve' death).") (emphasis added). Return to text.

[159] See id. at 426 ("We have argued that the Supreme Court's chosen path of constitutional regulation of the death penalty has been a disaster, an enormous regulatory effort with almost no rationalizing effect."); see also id. at 403 ("In short, the last twenty years have produced a complicated regulatory apparatus that achieves extremely modest goals with a maximum amount of political and legal discomfort."); id. at 426 ("It is difficult to imagine a body of doctrine that is much worse—either in its costs of implementation or in its negligible returns—than the one we have now."); id. at 429 (arguing that the Court's death penalty doctrinal structure is "functionally and ethically unsatisfying."); id. at 437 ("We began our exploration of legitimation theory in an effort to support the idea that the Court's deeply flawed death penalty law persists because of its success as a 'facade' that creates an appearance of stringent regulation but hides the incoherence and ineffectiveness of the underlying structure."); id. at 438 ("We are left with the worst of all possible worlds: the Supreme Court's detailed attention to death penalty law has generated negligible improvements over the pre-Furman era, but has helped people to accept without second thoughts—much less 'sober' ones—our profoundly failed system of capital punishment."). Return to text.

[160] See discussion infra Part III.B. Return to text.

[161] See Steiker & Steiker, supra note 1, at 375 (stating that the Baldus study "found that approximately 86% of all persons convicted of murder in Georgia over a five-year period after the adoption of Georgia's new statute were death-eligible under that scheme."). Return to text.

[162] See id. at 375 (citing BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 268 n.31). Return to text.

[163] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 88-89. Return to text.

[164] The difference between this 20% and the 14% exclusion figure used by Steiker and Steiker, see supra note 161, is that the 14% figure is based on 1974-79 cases, see BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 268 n.31, while my 20% figure is based on 1973-78 cases, see id. at 88-89. Return to text.

[165] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 43 (post-Furman cases analyzed included those from March 28, 1973, through June 20, 1978). In some jurisdictions, legislatures have drawn narrow death-eligibility criteria, resulting in a small percentage of homicides being death eligible. See id. at 233-34 (about 20% of Colorado murder and nonnegligent manslaughter cases death-eligible); see also David Baldus & George Woodworth, Proportionality: The View of the Special Master, 6 CHANCE: NEW DIRECTIONS FOR STATISTICS AND COMPUTING 9, 11 (1993) (only 227 of over 2,000 New Jersey homicide cases death-eligible). Return to text.

[166] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 25. Return to text.

[167] See id. Return to text.

[168] See id. (explaining the institution of the bifurcated proceeding). Return to text.

[169] Id. Return to text.

[170] Id. Return to text.

[171] That is, of course, as long as the defendant's attorney took advantage of the opportunity to present such evidence. For horror stories concerning attorneys who completely wasted this opportunity (not necessarily in Georgia), see David J. Gross, Sixth Amendment—Defendant's Dual Burden in Claims of Ineffective Assistance of Counsel, 75 J. CRIM. L. 755, 757 (1984); Welsh S. White, Effective Assistance of Counsel in Capital Cases: The Evolving Standard of Care, 1993 U. ILL. L. REV. 323, 325. Return to text.

[172] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 182 ("The most striking post-Furman change has been the statewide decline in the race-of-defendant effect. Indeed, on average, black defendants appear to enjoy a slight overall advantage compared to white defendants, although the effect is not statistically significant.") Return to text.

[173] See id. at 183-84. Instead, BWP attribute the finding to a general trend toward equal treatment of African-American defendants that began in urban areas and spread to rural areas. See id. Thus, they state, "Consequently, it does not appear that the decline in discrimination based on the defendant's race is attributable to Georgia's 1973 statutory changes." Id. at 184. Return to text.

[174] See supra note 138 and accompanying text. Return to text.

[175] Steiker & Steiker, supra note 1, at 358 n.8. Return to text.

[176] Id. (quoting PATERNOSTER, supra note 1, at xv). Return to text.

[177] See supra note 35. Return to text.

[178] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 229-48 (discussing three studies of pre- Furman cases, and three studies of post-Furman cases). An interesting recent research effort is reported in Robert E. Weiss et al., Assessing the Capriciousness of Death Penalty Charging, 30 L. & SOC'Y REV. 607, 617-25 (1996) (analyzing death penalty charging in 363 homicides in San Francisco County, California, through regression analysis and concluding that under the most optimistic assessment, the charging system "wring[s] out about two-thirds of the potential capriciousness"). Return to text.

[179] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 80-81 ("The purpose of each analysis was to determine whether those defendants who received sentences of death can be meaningfully distinguished from the many other defendants who received only life sentences."). Return to text.

[180] See id. at 44 (over 150 aggravating and mitigating factors were used). Return to text.

[181] See id. at 85. Return to text.

[182] Furman, 408 U.S. at 309 (Stewart, J., concurring). Return to text.

[183] In light of this statistic, it is possible to argue that the Court overreacted in Furman. However, there are important aspects of the world as seen by the Court in 1972 that are not reflected by this statistic. First, this statistic does not include cases that came before the Court from Georgia in which the defendants had been sentenced to death for rape or armed robbery, two crimes as to which the Court later determined that the death penalty was overinclusive as a matter of law. See Coker v. Georgia, 433 U.S. 584, 597 (1977); Hooks v. Georgia, 433 U.S. 917, 917 (1977). Second, it took quite a lot of legwork for the Baldus group to dig out the information that enabled them to determine just how aggravated each case was. The lack of a meaningful sentencing proceeding in Georgia prior to 1972 means that the Court did not have access to this information—the Georgia system from the Court's perspective must have resembled a "black box" from which defendants' names were pulled seemingly at random. Third, this statistic does not reflect the seemingly disproportionate number of African-Americans sentenced to death, particularly for rapes of white women. See supra note 128 and accompanying text. Return to text.

[184] BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 88-89. Return to text.

[185] See id. at 92. Return to text.

[186] I realize, of course, that death penalty opponents often contend that the very fact that the system cannot be made perfect is a cogent argument for its abolition. Return to text.

[187] This has prompted me to make a modest effort to do so. See infra Part III.B.1.b. Return to text.

[188] These cases are summarized in Appendix B. I identified these cases by obtaining a printout of all of the inmates on Georgia's death row from the NAACP Legal Defense Fund Death Penalty Project. I arranged the cases in reverse chronological order from date of sentence, then I searched the defendants' names electronically, to find a direct appeal decision by the Georgia Supreme Court. I excluded two of these cases. First, I did not use Drane v. State, 455 S.E.2d 27 (Ga. 1995), because the Georgia Supreme Court found that the trial court may have improperly excluded the co-defendant's confession—in which the co-defendant took major responsibility for the killing—and reversed the conviction. See id. at 30. Although this evidence was admitted at the sentencing phase, the failure to admit it at the guilt phase could have undermined the validity of the jury's verdict at the guilt phase. The second case I excluded was Potts v. State, 410 S.E.2d 89 (Ga. 1991). This case involved a kidnapping and murder that occurred in 1981, but took almost a decade to reach the Georgia Supreme Court. I excluded this case because the crime was not committed during the same time frame as the other cases I examined. I replaced these two cases with the next two cases in reverse chronological order. Return to text.

[189] The NAACP Legal Defense and Educational Fund's Spring 1995 "Death Row U.S.A." Report shows 104 inmates under a sentence of death in Georgia. Return to text.

[190] See infra notes 237-39 and accompanying text. Return to text.

[191] See Ledford v. State, 439 S.E.2d 917 (Ga. 1994); Bennett v. State, 414 S.E.2d 218 (Ga. 1992); Gibson v. State, 404 S.E.2d 781 (Ga. 1991); Hall v. State, 383 S.E.2d 128 (Ga. 1989); see also Appendix B, nos. 6, 13, 17, 23. Return to text.

[192] See Davis v. State, 426 S.E.2d 844 (Ga. 1993); Tharpe v. State, 416 S.E.2d 78 (Ga. 1992); Lynd v. State, 414 S.E.2d 5 (Ga. 1992); see also Appendix B, nos. 10, 12, 14. Return to text.

[193] See Hill v. State, 427 S.E.2d 770 (Ga. 1993); Todd v. State, 410 S.E.2d 725 (Ga. 1991); see also Appendix B, nos. 9, 16. Return to text.

[194] See Wellons v. State, 463 S.E.2d 868 (Ga. 1995), cert. denied, 117 S. Ct. 97 (1996); see also Appendix B, no. 1. Return to text.

[195] See Crowe v. State, 458 S.E.2d 799 (Ga. 1995), cert. denied, 116 S. Ct. 1021 (1996); see also Appendix B, no. 3. Return to text.

[196] See Hittson v. State, 449 S.E.2d 586 (Ga. 1994); see also Appendix B, no. 5. Return to text.

[197] See Bennett v. State, 149 S.E.2d 218 (Ga. 1992); see also Appendix B, no. 13. Return to text.

[198] See Jarrells v. State, 375 S.E.2d 842 (Ga. 1989); see also Appendix B, no. 25. Return to text.

[199] See Wellons v. State, 463 S.E.2d 868 (Ga. 1995), cert. denied, 117 S. Ct. 97 (1996); Ward v. State, 417 S.E.2d 130 (Ga. 1992); Tharpe v. State, 416 S.E.2d 78 (Ga. 1992); Pitts v. State, 386 S.E.2d 351 (Ga. 1989); see also Appendix B, nos. 1, 11, 12, 21. Return to text.

[200] See McMichen v. State, 458 S.E.2d 833 (Ga. 1995), cert. denied, 116 S. Ct. 819 (1996); Osborne v. State, 430 S.E.2d 576 (Ga. 1993); Ferrell v. State, 401 S.E.2d 741 (Ga. 1991); Stripling v. State, 401 S.E.2d 500 (Ga. 1991); Hightower v. State, 386 S.E.2d 509 (Ga. 1989); Hatcher v. State, 379 S.E.2d 775 (Ga. 1989); see also Appendix B, nos. 2, 8, 19, 20, 22, 24. Return to text.

[201] See Burgess v. State, 450 S.E.2d 680 (Ga. 1994); see also Appendix B, no. 4. Return to text.

[202] See Hill v. State, 427 S.E.2d 770 (Ga. 1993); see also Appendix B, no. 9. Return to text.

[203] See Pitts v. State, 386 S.E.2d 351 (Ga. 1989); see also Appendix B, no. 21. Return to text.

[204] See Burgess v. State, 450 S.E.2d 680 (Ga. 1994); see also Appendix B, no. 4. Return to text.

[205] See Davis v. State, 426 S.E.2d 844 (Ga. 1993); see also Appendix B, no. 10. Return to text.

[206] See Ward v. State, 417 S.E.2d 130 (Ga. 1992); see also Appendix B, no. 11. Return to text.

[207] See Lynd v. State, 414 S.E.2d 5 (Ga. 1992); see also Appendix B, no. 14. Return to text.

[208] See Todd v. State, 410 S.E.2d 725 (Ga. 1991); see also Appendix B, no. 16. Return to text.

[209] See Stripling v. State, 401 S.E.2d 500 (Ga. 1991); see also Appendix B, no. 20. Return to text.

[210] See Jarrells v. State, 375 S.E.2d 842 (Ga. 1989); see also Appendix B, no. 25. Return to text.

[211] See Wellons v. State, 463 S.E.2d 868 (Ga. 1995); Hall v. State, 415 S.E.2d 158 (Ga. 1991); see also Appendix B, nos. 1, 15. Return to text.

[212] See Fugate v. State, 431 S.E.2d 104 (Ga. 1993); see also Appendix B, no. 7. Return to text.

[213] See infra p. 587. Return to text.

[214] See supra notes 41-44 and accompanying text; see also Appendix A. Return to text.

[215] See Baldus et al., Comparative Review, supra note 32, at 684-86. Return to text.

[216] See id. at 685. Return to text.

[217] See id. Return to text.

[218] See id. at 686 n.90.

The major aggravating factors were: Torture, excessive and unnecessary pain, victim bound and/or gagged, execution-style killing, sexual perversion other than rape, victim pled for life, defendant showed pleasure with killing, mutilation, slashed throat, defendant an escapee, victim was a police or fire person, multiple shots to head or multiple stab wounds, insurance motive, or victim was held hostage.
Id. Return to text.

[219] See id. at 686. Return to text.

[220] See id. at 686 n.91.

The mitigating factors were: Defendant showed remorse, gave self up within 24 hours, was drunk or had a history of drug or alcohol abuse, had no intent to kill, believed he or she had a moral justification, the victim was a fugitive, provoked or aroused defendant, was drinking, or using drugs or had bad blood with defendant.
Id. Return to text.

[221] See id. at 686 n.92 ("The minor aggravating factors were: A race-related motive, victim was drowned, defendant resisted arrest, defendant created a great risk in a public place, or the victim was a hostage or female."). Return to text.

[222] Specifically, I assumed that the factors included within the Barnett analysis and the BWP analysis continue to identify the basic aggravating and mitigating factors, and continue to directly reflect their relative strength. As BWP explain:

There are two basic approaches to classifying cases as similar or dissimilar—the a priori and the empirical. The a priori approach endeavors to classify cases as similar on the basis of criteria that, from a legal or moral perspective, one believes should govern the appropriate sentence.
. . . . The empirical approach also begins by presupposing that certain factual characteristics of the case being reviewed can serve to identify other cases of "similar" culpability. In contrast to the a priori approach—which primarily selects those factual characteristics on a normative basis—the empirical approach tries to employ those legitimate case characteristics that, statistically, best explain the observed sentencing results . . . . The difference between the two methods is that the a priorist selects the factors he or she believes should influence the sentencing decision, while the empiricist selects the factors that actually appear to do so. BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 47-48. Thus, I am adopting an a priori approach, with my a priori choices being informed by the results of empirical studies. Return to text.

[223] See Hall v. State, 415 S.E.2d 158, 159 (Ga. 1991). Return to text.

[224] See id. at 160. Return to text.

[225] Id. Return to text.

[226] See id. Return to text.

[227] Barnett reported no cases in this box, but this is clearly a box that should have a high ratio—the ratios in the other boxes where the three Barnett integers totaled "4" are .56 and .81. Return to text.

[228] In this case, the defendant pulled the victim's head back by the hair and shot her in the forehead. If this is not an "execution-style" killing (BWP do not define this), then there is no serious aggravating circumstance, and the case scores .00 on the BWP scale. If it is an execution-style killing, then it scores .42. Return to text.

[229] There were not enough cases in this box in the BWP study (only one, in which a death sentence was not imposed) to form a valid proportion. The boxes on either side had moderate ratios (.40 and .50), but were based upon a small number of cases. Return to text.

[230] See Bennett v. Georgia, 414 S.E.2d 218, 220 (Ga. 1992). Return to text.

[231] See id. Return to text.

[232] See David Cook et al., The Decisionmakers: What Moves Prosecutors, Judges, & Jurors? (Mar. 4, 1996) (finding that jurors are more willing to find mental problems mitigating in the abstract than when applied to a real case) (materials based upon the Capital Jury Research project presented at Life in the Balance VIII conference in St. Louis, Missouri) (on file with the author). Return to text.

[233] See Taylor v. Georgia, 404 S.E.2d 255, 257 (Ga. 1991). Return to text.

[234] See id. Return to text.

[235] See id. Return to text.

[236] See id. at 258. Return to text.

[237] See Fugate v. Georgia, 431 S.E.2d 104, 106-07 (Ga. 1993). Return to text.

[238] See id. at 107. Return to text.

[239] See id. Return to text.

[240] See supra notes 237-39 and accompanying text. Return to text.

[241] See supra pp. 581-82. Return to text.

[242] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 42. Return to text.

[243] Id. at 60. Return to text.

[244] See id. at 90-91 ("Specifically, our 'overall culpability' index ranks the cases according to the presence or absence of seventeen legitimate case characteristics and combinations thereof that share a statistically significant relationship with the sentences imposed."). Return to text.

[245] See id. at 56.

We developed a regression-based culpability index for the [study] with a logistic multiple-regression analysis designed to identify statistically the factors that best explain which defendants received death sentences . . . . This procedure required us, first, to collect information for every case concerning a large number of legitimate case characteristics, such as prior record or a contemporaneous felony, that might have influenced the sentencing decision. We then computed for each such case characteristic a regression co-efficient (or 'weight') that reflected its individual contribution to the overall culpability index. Next, we calculated the relative culpability or blameworthiness of each case by summing the 'weights' of all the legitimate explanatory variables present in that case. We then ranked all the cases according to their relative culpability scores, thereby constructing an overall culpability index along which the cases were distributed. Finally, we defined as 'similar' six groups of cases with comparable overall culpability scores.
Id. Return to text.

[246] Id. at 91-92. Return to text.

[247] See supra pp. 580-81. Return to text.

[248] Steiker & Steiker, supra note 1, at 358. Return to text.

[249] See BALDUS ET AL., EQUAL JUSTICE, supra note 33, at 98-105. Return to text.

[250] Id. at 100. Return to text.

[251] See id. at 98-100. Return to text.

[252] See id. at 105. Return to text.

[253] See id. Return to text.

[254] See id. Return to text.

[255] See id. Return to text.

[256] Id. Return to text.

[257] Steiker & Steiker, supra note 1, at 357. Return to text.

[258] To me, the most regrettable road not taken by the Court has been its failure to require a heightened level of attorney competence above the relatively minimal standard set for all cases, including death penalty cases, in Strickland v. Washington, 466 U.S. 668, 687 (1984) (requiring that counsel's performance be shown to be deficient and that the deficiency prejudiced the defense in order to reverse a conviction or death sentence). A requirement that states provide super-competent trial counsel would either force states out of the death penalty business due to excessive costs or force states to provide top-notch counsel. Either possibility would be more fair to defendants and more efficient in the long run. Return to text.

[259] Barnett, supra note 32, at 1364-66. Return to text.

[260] Baldus at al., Comparative Review, supra note 32, at 687. Return to text.