\\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 1 5-MAY-05 11:55 C. RYAN RUSSELL* Death Anyways: Federal Habeas Corpus Retroactivity Law and the Decision in Schriro v. Summerlin It is the raw material from which legal fiction is forged: a vi-cious murder, an anonymous psychic tip, a romantic encoun- ter that jeopardized a plea agreement, an allegedly incompetent defense, and a death sentence imposed by a purportedly drug- addled judge. But, as Mark Twain observed, ?truth is often stranger than fiction because fiction has to make sense.?1 The above commentary by the Ninth Circuit on Warren Wes- ley Summerlin?s trip through the criminal justice system illus- trates the bizarre circumstances of his case. In short, the criminal justice system failed him. After being charged with murder and having been appointed a lawyer who had a love affair with the prosecutor,2 Summerlin was sentenced to death by a judge who may have confused the facts of Summerlin?s case with those of another capital defendant sentenced to death by the same judge on the same day.3 Summerlin challenged the constitutionality of having a judge determine his sentence at every turn, but he was continually denied relief.4 Years after Summerlin exhausted his * J.D., University of Oregon, 2005; B.A., Western Washington University, 2002. I would like to thank the following individuals: Douglas Park, Adjunct Professor of Law and Assistant Attorney General with the Oregon Department of Justice, for his valuable contributions and for teaching me everything I know about federal habeas corpus law; editors Tim Hering and Jessica King for their thoughtful revisions; and my wife, Jessica, and my parents, Chris and Kathi Russell, for their love, support, and unearthly amounts of patience. 1 Summerlin v. Stewart, 341 F.3d 1082, 1084 (9th Cir. 2003), rev?d sub nom. Schriro v. Summerlin, 124 S.Ct. 2519 (2004). 2 Summerlin, 341 F.3d at 1085-87. 3 Id. at 1089-91. The judge was later disbarred for marijuana use (including usage of marijuana during the period when Summerlin was sentenced to death). Id. at 1091; see also In re Disbarment of Marquardt, 503 U.S. 902 (1992); In re Marquardt, 821 P.2d 161 (Ariz. 1991). 4 Summerlin, 341 F.3d at 1091-92. [1389] \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 2 5-MAY-05 11:55 1390 OREGON LAW REVIEW [Vol. 83, 2004] direct appeals, the U.S. Supreme Court in Ring v. Arizona exe- cuted an about-face, overturning itself and embracing the argu- ment Summerlin had pushed for so long: Arizona?s capital sentencing scheme violated the Sixth Amendment right to a jury trial.5 Despite the Supreme Court?s decision that Arizona?s capi- tal sentencing structure was constitutionally flawed at the time Summerlin was sentenced, the state of Arizona continued its plan to execute him,6 because new rules of criminal procedure do not usually apply retroactively.7 The Ninth Circuit heard Summerlin?s case en banc and despite its observation that new rules of criminal procedure usually do not apply retroactively, the court held Ring should apply retroac- tively to Summerlin?s case.8 The court held there were two pri- mary reasons for applying Ring retroactively: (1) the Supreme Court?s holding in Ring was not a new rule of criminal proce- dure, but a new substantive rule of criminal law; and (2) even if the Ring holding was perceived as procedural, it fit one of the exceptions to the general presumption that new rules of criminal procedure do not apply retroactively.9 This holding had the prac- tical effect of overturning the death sentences of over one hun- dred prisoners in five different states.10 Arguably, it also had the effect of lowering the bar for applying new constitutional rules of criminal procedure retroactively. Not surprisingly, the United States Supreme Court granted certiorari to review the Ninth Cir- cuit?s decision.11 In a five-to-four decision, the Supreme Court overturned the Ninth Circuit, holding that Ring does not apply retroactively to prisoners challenging their capital sentences in a collateral pro- ceeding.12 The Supreme Court first held that, contrary to the Court of Appeals? finding, Ring was not a new substantive rule, but a new constitutional rule of criminal procedure.13 The Court also held that Ring did not apply retroactively under the Teague ?watershed? exception because having a judge as the fact-finder 5 536 U.S. 584, 609 (2002). 6 Summerlin, 341 F.3d at 1091. 7 Id. at 1096. 8 Id. at 1084. 9 Id. at 1121. 10 See Adam Liptak, Judges? Rulings Imposing Death are Overturned, N.Y. TIMES, Sept. 3, 2003, at A1. 11 Schriro v. Summerlin, 540 U.S. 1045 (2003). 12 Schriro v. Summerlin, 124 S. Ct. 2519, 2526 (2004). 13 Id. at 2524. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 3 5-MAY-05 11:55 Death Anyways 1391 during the sentencing proceeding did not ?seriously diminish? the accuracy of the sentencing proceeding.14 The Supreme Court?s opinion clarified Teague retroactivity law, reemphasized the ?accuracy? element of the watershed exception to Teague?s retroactivity bar, and maintained the overall narrowness of the watershed exception. This Comment discusses both the Ninth Circuit?s Summerlin and Supreme Court?s Schriro opinions in an attempt to explain the two courts? reasoning and to illustrate the significant and con- tentious elements of the Schriro opinion. It is also the goal of this Comment to sketch the contours of retroactivity law on col- lateral review and to come to conclusions on the state of the law in such a way that it extends beyond the immediate case presented to the Supreme Court. This Comment will show that there are aspects of retroactivity law that could be substantially affected by a court following the reasoning employed by the Ninth Circuit in its Summerlin decision. In many ways, the Ninth Circuit?s opinion would have significantly lowered the bar for when a new constitutional criminal rule is applied retroactively on collateral review. Finally, this Comment discusses the prece- dential effects that the Supreme Court?s Schriro opinion will have on future retroactivity litigation by specifically focusing on Schriro?s impact on the possible retroactive application of Blakely v. Washington,15 a new rule of criminal procedure an- nounced by the Supreme Court that could affect thousands of cases. Because Ring (the rule examined in Schriro) and Blakely both derive from applications of the Court?s holding in Apprendi v. New Jersey,16 the Court?s Schriro analysis will likely be ex- tremely influential on the inevitable litigation over whether Blakely applies retroactively. Part I of this Comment reviews the background law regarding retroactive application of new constitutional rules dealing with criminal matters on collateral review. Part II discusses the back- ground law of Summerlin?s case, including an explanation of the Supreme Court?s Ring decision as well as a general discussion of retroactivity law. Part III explains the Ninth Circuit?s holding and reasoning in Summerlin?s case, and Part IV discusses these elements of the Supreme Court?s opinion. Finally, Part V dis- 14 Id. at 2525. 15 124 S. Ct. 2531 (2004). 16 530 U.S. 466 (2000). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 4 5-MAY-05 11:55 1392 OREGON LAW REVIEW [Vol. 83, 2004] cusses the implications of these decisions on retroactivity law. First, it begins by discussing the impact on retroactivity law of the Supreme Court?s clarified discussion of the distinction between substantive and procedural rules. Second, it discusses the effect of the Court?s Schriro opinion on the second Teague exception, explaining how the Court maintained an extremely high hurdle for petitioners to clear in order to have new constitutional rules of criminal procedure apply retroactively to them. To conclude, this Comment discusses Schriro?s impact on Blakely, in an effort to illustrate the likely effects Schriro will have on that important case as well as other cases in the future. I FACTUAL AND PROCEDURAL HISTORY On April 29, 1981, Warren Wesley Summerlin brutally mur- dered Brenna Bailey, a delinquent-account investigator who came to Summerlin?s house to speak with his wife about an over- due account.17 As a result of his actions, Summerlin was con- victed of both first-degree murder and sexual assault.18 Pursuant to Arizona state law at the time, Summerlin received a sentenc- ing hearing over which the trial judge presided to determine whether Summerlin would be sentenced to life imprisonment or death.19 The judge sentenced Summerlin to death.20 Summerlin appealed his conviction and sentence.21 One of the issues Summerlin raised on appeal was that Arizona?s law of a judge, not a jury, sentencing him to death was unconstitutional.22 However, the Arizona Supreme Court rejected that argument, along with the rest of his claims, and affirmed Summerlin?s con- viction and death sentence.23 17 Summerlin v. Stewart, 341 F.3d 1082, 1084-85 (9th Cir. 2003), rev?d sub nom. Schriro v. Summerlin, 124 S. Ct. 2519 (2004). 18 Summerlin, 341 F.3d at 1088. 19 Id. at 1088-89; see also ARIZ. REV. STAT. ANN. ? 13-703 (West 1983) (amended 2002). 20 Summerlin, 341 F.3d. at 1089. The sentencing hearing was extremely short. It began with the State submitting one exhibit and then asking the judge to consider the trial testimony. Id. Summerlin?s attorney then called a doctor who had con- ducted a psychological evaluation on Summerlin, id. at 1085-86, but Summerlin told his attorney that he did not want the doctor to testify. Id. at 1089. The State con- cluded the sentencing hearing by calling two rebuttal psychiatric witnesses. Id. 21 State v. Summerlin, 675 P.2d 686 (Ariz. 1983). 22 Id. at 695. 23 Id. at 696. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 5 5-MAY-05 11:55 Death Anyways 1393 Summerlin next attempted to have his conviction overturned four different times in state court, all to no avail.24 He then peti- tioned for a writ of habeas corpus in federal district court but was once again denied.25 He appealed this judgment and a divided three-judge panel of the Ninth Circuit Court of Appeals reversed his conviction in part and remanded to the federal district court for an evidentiary hearing to determine whether Summerlin?s sentencing judge was competent when he sentenced Summerlin to death.26 In the meantime, the United States Supreme Court granted certiorari in State v. Ring27 to reexamine whether Arizona?s capi- tal sentencing scheme, which allowed judges to find the facts nec- essary to sentence persons to death, was constitutional.28 This was an issue Summerlin had raised both at the state and federal levels, thus far with no success.29 Accordingly, the Ninth Circuit withdrew its decision to remand the case to the district court for an evidentiary hearing and deferred submission of the case until the United States Supreme Court resolved Ring.30 Undoubtedly to Summerlin?s satisfaction, in Ring v. Arizona31 the Supreme Court held that Arizona?s capital sentencing procedure of having judges determine the aggravating factors necessary to impose a sentence of death violated the Sixth Amendment right to a trial by jury.32 Following Ring, Summerlin requested that the Arizona Su- preme Court recall the mandate in his direct appeal to apply Ring to his case.33 The Arizona Supreme Court denied Summer- lin?s motion to recall the mandate34 and the Ninth Circuit voted to rehear Summerlin?s case en banc.35 24 Summerlin, 341 F.3d at 1091. 25 Id. 26 Summerlin v. Stewart, 267 F.3d 926, 957 (9th Cir. 2001). 27 200 Ariz. 267 (2001), rev?d, 536 U.S. 584 (2002). 28 Summerlin, 341 F.3d at 1091. 29 Id. 30 Summerlin v. Stewart, 281 F.3d 836, 837 (9th Cir. 2002). 31 536 U.S. 584 (2002). 32 Id. at 609. See also U.S. CONST. amend. VI. 33 Summerlin, 341 F.3d at 1091. 34 Id. 35 Summerlin v. Stewart, 310 F.3d 1221 (9th Cir. 2002). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 6 5-MAY-05 11:55 1394 OREGON LAW REVIEW [Vol. 83, 2004] II BACKGROUND LAW A. The Ring Ruling and Related Authority Because the primary issue the Supreme Court decided in Sum- merlin was whether its holding in Ring applied retroactively, it is worth first discussing the Ring decision. In that case, Ring was convicted of felony murder.36 As with the law under which Sum- merlin was sentenced, Ring was sentenced to death after a judge found aggravating factors that supported a death sentence.37 Ring appealed his sentence, arguing ?Arizona?s capital sentenc- ing scheme violate[d] the Sixth and Fourteenth Amendments to the U.S. Constitution because it entrust[ed] to a judge the finding of a fact raising the defendant?s maximum penalty.?38 The State of Arizona relied on Walton v. Arizona,39 a United States Supreme Court opinion that upheld Arizona?s capital sen- tencing scheme.40 In Walton, the Court held that ??the Sixth Amendment does not require that the specific findings authoriz- ing the imposition of the sentence of death be made by the jury.??41 The Court noted the aggravating factors necessary to be found by a judge before imposition of a death sentence were not ?elements of the offense,? but were more like ?sentencing considerations.?42 Despite this obviously persuasive precedent, Ring relied on two recent Supreme Court cases that cast doubt on Walton?s va- lidity.43 In the first, Jones v. United States,44 the Supreme Court held: [U]nder the Due Process Clause of the Fifth Amendment and the notice and jury trial guarantees of the Sixth Amendment, any fact (other than prior conviction) that increases the maxi- mum penalty for a crime must be charged in an indictment, submitted to a jury, and proven beyond a reasonable doubt.45 36 Ring v. Arizona, 536 U.S. 584, 591 (2002). 37 Id. at 592. 38 Id. at 595. 39 497 U.S. 639 (1990). 40 Ring, 536 U.S. at 595. 41 Walton v. Arizona, 497 U.S. 639, 648 (1990) (quoting Hildwin v. Florida, 490 U.S. 638, 640-41 (1989) (per curiam)). 42 Id. at 648. 43 Ring, 536 U.S. at 595. 44 526 U.S. 227 (1999). 45 Id. at 243 n.6. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 7 5-MAY-05 11:55 Death Anyways 1395 The second helpful opinion to Ring?s case was Apprendi v. New Jersey.46 In that case, Apprendi fired several shots into the home of an African-American family.47 He pleaded guilty to possession of a firearm, a conviction that, standing on its own, carried a maximum sentence of ten years.48 However, after the plea was entered the prosecutor filed a motion to enhance Ap- prendi?s sentence based on a state hate-crime statute.49 That statute allowed a sentencing judge to enhance a sentence beyond the statutory maximum when the judge found by a preponder- ance of evidence that the crime was committed with the purpose of intimidating a person or group because of race.50 The judge in Apprendi?s case found the crime was racially motivated and sen- tenced Apprendi to twelve years in prison, two years beyond the statutory maximum without the aggravating factor.51 The Apprendi Court held that this practice of allowing a sen- tencing judge to find the racial motivation aggravator, by only a preponderance of evidence, violated the Sixth Amendment right to a jury and the Fourteenth Amendment?s right to due process of law.52 The Court reasoned that the state ?threatened Ap- prendi with certain pains if he unlawfully possessed a weapon and with additional pains if he selected his victims with a purpose to intimidate them because of their race.?53 The Court stated that determining whether a factor for a greater sentence was a ?finding of fact,? therefore subject to the Sixth Amendment right to a jury and the state?s burden of proof beyond a reasonable doubt, or a mere ?sentencing factor,? properly determined by a sentencing judge, is a question ?not of form, but of effect.?54 Rather than overruling Walton, the Apprendi Court concluded that Apprendi?s case was distinguishable.55 The Court explained that when a capital sentencing judge was required to find aggra- 46 530 U.S. 466 (2000). 47 Id. at 469. 48 Id. at 470. 49 Id. 50 Id. at 468-69. 51 Id. at 471. 52 Id. at 477. The role of the Fourteenth Amendment in Apprendi was more than holding the Sixth Amendment against the State of Arizona. The Due Process Clause also ?protects the accused against conviction except upon proof beyond a reasonable doubt of every fact necessary to constitute the crime with which he is charged.? In re Winship, 397 U.S. 358, 364 (1970). 53 Apprendi, 530 U.S. at 476 (emphasis added). 54 Id. at 494, 482-83. 55 Id. at 496-97. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 8 5-MAY-05 11:55 1396 OREGON LAW REVIEW [Vol. 83, 2004] vating factors to impose a death sentence, the jury had already decided the defendant was guilty of capital murder with a maxi- mum penalty of death.56 In that situation, the judge merely de- termines whether that maximum penalty should be imposed.57 Justice O?Connor?s dissent called this distinction ?baffling.?58 Justice O?Connor quite persuasively noted ?[a] defendant con- victed of first-degree murder in Arizona cannot receive a death sentence unless a judge makes the factual determination that a statutory aggravating factor exists. Without that critical finding, the maximum sentence to which the defendant is exposed is life imprisonment.?59 In Ring, however, the Court held ?Walton and Apprendi [were] irreconcilable,? and overruled the former.60 It overruled Walton ?[b]ecause Arizona?s enumerated aggravating factors op- erate as ?the functional equivalent of an element of a greater of- fense,? [and] the Sixth Amendment requires that they be found by a jury.?61 In other words, the Court held that a capital sen- tencing scheme violates the Sixth Amendment ?to the extent that it allows a sentencing judge, sitting without a jury, to find an ag- gravating circumstance necessary for imposition of the death penalty.?62 The Court reasoned that ?in effect,? the finding of a death sentence under Arizona law required a finding of fact.63 It noted that Arizona?s labeling the required aggravating findings ?sentencing factors? would not mean the Sixth Amendment does not apply.64 Accepting such an argument would reduce Ap- prendi ?to a ?meaningless and formalistic? rule of statutory drafting.?65 B. Retroactivity and Teague v. Lane When the Supreme Court issues a new constitutional rule, one might intuitively think the new rule would apply to every citizen of the United States. In fact, new judicial rulings generally apply 56 Id. 57 Id. 58 Id. at 538 (O?Connor, J., dissenting). 59 Id. 60 Ring v. Arizona, 536 U.S. 584, 609 (2002). 61 Id. (quoting Apprendi, 530 U.S. at 494 n.19). 62 536 U.S. at 609. 63 Id. at 604. 64 Id. 65 Id. (quoting Apprendi, 530 U.S. at 541 (O?Connor, J., dissenting)). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 9 5-MAY-05 11:55 Death Anyways 1397 both prospectively and retroactively.66 However, there are some situations where the Supreme Court will not apply judicial rul- ings retroactively.67 In the landmark case Teague v. Lane,68 the Supreme Court held that new constitutional rules of criminal procedure is one of those areas, at least with respect to cases that are considered final before the new rule is announced.69 The Court held that for such rules there is a presumption against ap- plying the rule retroactively, stating, ?[u]nless they fall within an exception to the general rule, new constitutional rules of criminal procedure will not be applicable to those cases which have be- come final before the new rules are announced.?70 That is, a new rule of criminal procedure will not affect persons who have ex- hausted their direct appeals and are challenging their convictions or sentences on a collateral attack. In coming to its determination that constitutional rules of crim- inal procedure generally would not be applied retroactively, the Court noted in Teague that it ?never has defined the scope of the writ [of habeas corpus] simply by reference to a perceived need to assure that an individual accused of crime is afforded a trial free of constitutional error,?71 but rather has ?recognized that in- terests of comity and finality must . . . be considered in determin- ing the proper scope of habeas review.?72 Although many have noted that nonretroactivity leads to arbitrarily denying some de- fendants their constitutional rights because the judiciary-created errors were corrected too late,73 the Court highlighted the frus- trations of applying new constitutional rules of criminal proce- dure on collateral review: a lack of finality, which undermines the 66 See, e.g., 2 RANDY HERTZ & JAMES S. LIEBMAN, FEDERAL HABEAS CORPUS PRACTICE AND PROCEDURE, ? 25.2, at 1034 n.1 (4th ed. 2001), which cites Landgraf v. USI Film Prods., 511 U.S. 244, 264 (1994) (noting general ?rule that ?a court is to apply the law in effect at the time it renders its decision??) (quoting Bradley v. Rich- mond Sch. Bd., 416 U.S. 696, 711 (1994)). 67 See, e.g., Linkletter v. Walker, 381 U.S. 618, 627 (1965). 68 489 U.S. 288 (1989) (plurality). 69 Id. at 310. 70 Id. 71 Id. at 308 (quoting Kuhlmann v. Wilson, 477 U.S. 436, 447 (1986)) (plurality opinion). 72 489 U.S. at 308. 73 See, e.g., Summerlin v. Stewart, 341 F.3d 1082, 1122 (Reinhardt, J., concurring) (?[E]xecuting people because their cases came too early?because their appeals en- ded before the Supreme Court belatedly came to the realization that it had made a grievous constitutional error . . . is surely arbitrariness that surpasses all bounds.?). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 10 5-MAY-05 11:55 1398 OREGON LAW REVIEW [Vol. 83, 2004] entire criminal process.74 However, Teague stated two exceptions to the general pre- sumption against retroactive application of new rules of criminal procedure: (1) if the rule ?place[d] certain kinds of primary, pri- vate individual conduct beyond the criminal law-making author- ity to proscribe?;75 or (2) if the rule ?require[d] the observance of those procedures that . . . are implicit in the concept of ordered liberty.?76 There are two elements to the Court?s second excep- tion to the rule of nonretroactivity for constitutional rules of criminal procedure.77 First, the new rule must be a ?watershed rule of criminal procedure? that ?will properly alter our under- standing of the bedrock procedural elements that must be found to vitiate the fairness of a particular conviction.?78 Second, the new rule must ?significantly improve the pre-existing factfinding procedures,?79 in a way that ?the likelihood of an accurate con- viction [has been] seriously diminished.?80 The Court has noted this second exception is quite narrow and should be employed on an infrequent basis.81 In fact, in Beard v. Banks the Supreme Court recently noted that ?because any qualifying rule ?would be so central to an accurate determination of innocence or guilt . . .? 74 Teague, 489 U.S. at 309, 310 (?Without finality, the criminal law is deprived of much of its deterrent effect.?) (?[S]tate courts are understandably frustrated when they faithfully apply existing constitutional law only to have a federal court discover, during a [habeas] proceeding, new constitutional commands.) (quoting Engle v. Isaac, 456 U.S. 107, 108 n.33). 75 Teague, 489 U.S. at 311 (quoting Mackey v. United States, 401 U.S. 667, 692 (Harlan, J., concurring in part and dissenting in part)). 76 Teague, 489 U.S. at 311 (quoting Mackey, 401 U.S. at 693 (Harlan, J., concur- ring in part and dissenting in part)). Although Teague was a plurality opinion, the next year a majority of justices made clear that the opinion substantially limited the retroactive application of new constitutional rules of criminal procedure to these two narrow exceptions for cases that reached the habeas corpus stage before the new rule was announced. See Sawyer v. Smith, 497 U.S. 227, 241 (1990); Saffle v. Parks, 494 U.S. 484, 487-88 (1990); Butler v. McKellar, 494 U.S. 407, 409 (1990). 77 Teague, 489 U.S. at 312. 78 Id. at 311 (quoting Mackey, 401 U.S. at 693-94 (Harlan, J., concurring in part and dissenting in part)). 79 Teague, 489 U.S. at 312 (quoting Desist v. United States, 394 U.S. 244, 262 (1969) (Harlan, J., dissenting)). 80 Teague, 489 U.S. at 313. 81 See, e.g., Graham v. Collins, 506 U.S. 461, 478 (1993) (the second exception ?clearly meant to apply only to a small core of rules?); Sawyer, 497 U.S. at 243 (?[I]t is ?unlikely that many such components of basis due process have yet to emerge.??) (quoting Teague); Spaziano v. Singletary, 36 F.3d 1028, 1043 (11th Cir. 1994) (stating a new rule under the second exception ?must be so fundamentally important that its announcement is a ?groundbreaking occurrence??) (quoting Caspari v. Bohlen, 510 U.S. 383, 396 (1994)). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 11 5-MAY-05 11:55 Death Anyways 1399 it should come as no surprise that we have yet to find a new rule that falls under the second Teague exception.?82 Further, the Court noted that ?[i]n providing guidance as to what might fall within this exception, we have repeatedly referred to the rule of Gideon v. Wainwright . . . (right to counsel), and only to this rule,?83 which indicates a very narrow view of the second Teague exception.84 Indeed, the same year Teague was announced, in Penry v. Lynaugh, the Court held the Teague nonretroactivity rule applied in capital cases.85 Although Teague exceptions are supposed to be rare, this holding was surprising because it essen- tially approved death sentences to persons who were denied cer- tain procedural rights while there was still an opportunity to remedy the wrong. In Bousley v. United States, the Court clarified that this Teague analysis only applies to new procedural rules, not substantive ones.86 Accordingly, a new constitutional rule will not be barred from applying retroactively by Teague if it is a new substantive rule.87 Further, because new rules of substantive criminal law generally apply retroactively,88 determining whether a new rule is substantive or procedural is critical. Although determining whether a new rule is procedural or substantive can be difficult,89 in Bousley the Court provided some guidance. A decision of ?substantive criminal law,? the Court stated, is one that ad- dresses the scope and application of a substantive federal crimi- nal statute.90 The Court did not speak as to what is a rule of procedure. Of course, another issue that must be dealt with before a court 82 124 S. Ct. 2504, 2513-14 (2004) (quoting Graham v. Collins, 506 U.S. 461, 478 (1993) (quoting Teague, 489 U.S. at 313)). Interestingly, Beard was decided by the Supreme Court on the same day it decided Summerlin?s case. Compare Beard, 124 S. Ct. at 2504 with Schriro v. Summerlin, 124 S. Ct. 2519 (2004). 83 Beard, 124 S. Ct. at 2514 (citing Gideon v. Wainwright, 372 U.S. 335 (1963)). 84 Beard, 124 S. Ct. at 2514; Saffle v. Parks, 494 U.S. 484, 495 (1990); Teague v. Lane, 489 U.S. 288, 311-12 (1989) (plurality opinion); Solem v. Stumes, 465 U.S. 638, 653-54, 644 n.4 (1984) (Powell, J., concurring). 85 Penry v. Lynaugh, 492 U.S. 302, 314 (1989). 86 Bousley v. United States, 523 U.S. 614, 620 (1998). 87 Id. 88 See, e.g., Coleman v. United States, 329 F.3d 77, 83 (2d Cir. 2003); Santana- Madera v. United States, 260 F.3d 133, 138 (2d Cir. 2001); Palmer v. Clarke, 293 F. Supp. 2d 1011, 1053 (D. Neb. 2003). 89 See, e.g., Robinson v. Neil, 409 U.S. 505, 509 (1973) (?[W]e would not suggest that the distinction that we draw is an ironclad one that will invariably result in the easy classification of cases in one category or the other.?). 90 Bousley, 523 U.S. at 620. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 12 5-MAY-05 11:55 1400 OREGON LAW REVIEW [Vol. 83, 2004] evaluates whether a new rule of criminal procedure falls into one of the two Teague exceptions is whether the rule is in fact ?new.?91 Determining whether a rule is new, as the Court noted in Teague, can also be difficult.92 However, the Teague court stated that: In general . . . a case announces a new rule when it breaks new ground or imposes a new obligation on the States or the Fed- eral Government . . . . To put it differently, a case announces a new rule if the result was not dictated by precedent existing at the time the defendant?s conviction became final.93 Although the Court has found it difficult to determine whether a rule is new, in cases where a new case extends an old rule,94 the Court has been quite clear when a case overrules an older one: ?[T]here can be no dispute that a decision announces a new rule if it expressly overrules a prior decision.?95 In O?Dell v. Netherland,96 the Supreme Court pulled all these issues together, laying out three elements a court must address under Teague: (1) the court must determine the date on which the defendant?s conviction became final; (2) the court must de- termine whether ?a state court considering [the defendant?s] claim at the time his conviction became final would have felt compelled by existing precedent to conclude that the rule [he] seeks was required by the Constitution?;97 and if not, then (3) whether the new rule falls into one of the two Teague exceptions discussed above.98 By the time the Ninth Circuit issued its opinion in Summerlin, the Eleventh Circuit Court of Appeals had already decided the 91 Whether a rule is ?new,? for retroactivity purposes, is a matter of law and re- viewed by federal courts de novo. See, e.g., Am. Trucking Ass?n v. Smith, 496 U.S. 167, 177 (1990) (plurality opinion) (?The determination whether a constitutional de- cision of this Court is retroactive . . . is a matter of federal law.?). 92 Teague v. Lane, 489 U.S. 288, 301 (1989). 93 Id. (citations omitted). 94 Saffle v. Park, 494 U.S. 484, 488 (1990). For a thorough discussion of the Court?s difficulty in defining what constitutes a new rule, and its accordingly com- plex jurisprudence in this area, see 2 HERTZ & LIEBMAN, supra note 66, at 1064- 1106. 95 Graham v. Collins, 506 U.S. 461, 467 (1993). See also Saffle, 494 U.S. at 488 (1990) (?The explicit overruling of an earlier holding no doubt creates a new rule.?). 96 521 U.S. 151 (1997). 97 Id. at 156 (quoting Lambrix v. Singletary, 520 U.S. 518, 527 (1997) (quoting Saffle, 494 U.S. at 488 (alterations in Lambrix)). 98 521 U.S. at 156. This approach was recently reaffirmed by the Supreme Court. See Beard v. Banks, 124 S. Ct. 2504, 2510 (2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 13 5-MAY-05 11:55 Death Anyways 1401 exact issue of whether Ring applied retroactively under the Teague doctrine.99 In Turner, the court first held that Ring an- nounced a procedural rule, rather than a substantive one.100 It reasoned that Ring was a procedural rule because it only affected ?what fact-finding procedure must be employed in a capital sen- tencing hearing.?101 Further, Ring did not change the underlying conduct or the burden of proof necessary to impose a death sen- tence.102 The court also argued Ring?s holding was ?procedural? because it was a mere extension of Apprendi, a ruling that nu- merous courts had already held was procedural.103 In Turner, the Eleventh Circuit also held that the rule from Ring did not fit either of the two exceptions stated in Teague.104 The court noted that the first Teague exception clearly does not apply to the Supreme Court?s Ring holding because it did not ?decriminalize any class of conduct or prohibit a certain category of punishment.?105 Next, the court found that Ring was not a ?watershed? rule that fit the second Teague exception because its purpose was not to improve the accuracy of the sentencing pro- cess.106 It argued that the new rule merely shifted ?fact-finding duties? from an impartial judge to an impartial jury.107 The point of Ring was not a problem with judges? conducting sentencing, 99 Turner v. Crosby, 339 F.3d 1247 (11th Cir. 2003). The Ninth Circuit noted the Eleventh Circuit?s opinion. See Summerlin v. Stewart, 341 F.3d 1082, 1096 n.4 (9th Cir. 2003). Another case that spoke directly to this issue was the Arizona Supreme Court?s application of Teague to the Ring rule. State v. Towery, 64 P.3d 828 (Ariz. 2003). In that case, the Arizona Supreme Court held that Teague barred the United States Supreme Court?s Ring holding from applying retroactively to the State?s pris- oners with death sentences. Towery, 64 P.3d at 830. The Ninth Circuit took note of the Arizona Supreme Court?s decision, but heavily criticized it and eventually went the other way from the state court?s holding. Summerlin, 341 F.3d at 1106-07. The Tenth Circuit Court of Appeals also held that Ring was not a substantive rule when the Ninth Circuit decided Summerlin. Cannon v. Mullin, 297 F.3d 989, 994 (10th Cir. 2002). Finally, Nevada?s Supreme Court also held that Ring did not apply retro- actively under the Teague standard. Colwell v. State, 59 P.3d 463, 470-73 (Nev. 2002). 100 Turner, 339 F.3d at 1284. 101 Id. 102 Id. 103 Id. The Supreme Court, in Apprendi, stated that its holding was procedural and did not go to the substance of the offense. Apprendi v. New Jersey, 530 U.S. 466, 475 (2000). 104 Turner, 339 F.3d at 1285. 105 Id. (quoting McCoy v. United States, 266 F.3d 1245, 1256-57 (11th Cir. 2001) (internal quotations omitted)). 106 Turner, 339 F.3d at 1286. 107 Id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 14 5-MAY-05 11:55 1402 OREGON LAW REVIEW [Vol. 83, 2004] but was an enlightened reading of the Sixth Amendment.108 III THE NINTH CIRCUIT?S HOLDING AND RATIONALE IN SUMMERLIN V. STEWART The Ninth Circuit had two alternative holdings for applying Ring retroactively.109 First, it found that the Ring holding, al- though partially procedural for Teague purposes, was a ruling of substantive criminal law. Because Ring affected the substance of criminal law, the court held it was not subject to the Supreme Court?s ruling in Teague.110 Second, the Ninth Circuit held that even if Ring were seen as a procedural rule, it still fit the second exception stated in Teague: the holding was a watershed rule that improved the accuracy of the underlying proceeding.111 A. Ring Announced a New Substantive Rule of Criminal Law In Summerlin, the Ninth Circuit first noted that Ring created at least a partially procedural rule.112 The court observed that Ring really just changed who found the facts, from judge to jury.113 However, it found that Ring went beyond that simple facial change of ?who decides.? The court also found that the rule at issue in Ring, despite being partially procedural, was a new substantive rule for Teague purposes.114 The Ninth Circuit used an intriguing argument to find that the Supreme Court established a new substantive rule in Ring. It observed that prior to Ring, the State claimed a jury was decid- ing, at the original criminal trial, whether the defendant was guilty of capital murder.115 The State?s theory, then, was that the judge was imposing a sentence on the jury?s already-determined capital murder verdict. However, as Ring pointed out, this was not really the case. Rather, the jury decided whether the defen- dant was guilty of murder, as charged in the indictment, and then the judge determined whether aggravating factors, which were termed the ?functional equivalent? of criminal elements by the 108 Id. 109 Id. at 1108. 110 Id. at 1099-1108. 111 Id. at 1108-21. 112 Id. at 1101. 113 Id. 114 Id. 115 Id. at 1101-02. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 15 5-MAY-05 11:55 Death Anyways 1403 Ring Court,116 were present before the judge had statutory au- thority to impose a death sentence. The way the Ninth Circuit saw it, the Ring Court created an entirely new substantive struc- ture of murder law in Arizona?murder at the trial level and cap- ital murder at the sentencing level, for which one would receive the death penalty.117 The court called this reordering of murder law a change from one offense (capital murder), pre-Ring, to a two-offense structure, post-Ring.118 To better understand how the Ninth Circuit saw a substantive change in law after Ring, it is helpful to consider what the Su- preme Court said in Ring. There the Court reasoned that Ari- zona could not have judges finding the aggravating factors necessary to impose a death sentence, because those were factual findings subject to the Sixth Amendment and Fourteenth Amendment guarantee to a jury finding facts beyond a reasona- ble doubt.119 Accordingly, if a jury is necessary at the sentencing level, there is no difference between the trial for innocence and guilt (termed simple ?murder? by the Ninth Circuit) and the trial for death or life imprisonment (termed ?capital murder? by the Ninth Circuit). Elements must be found at each stage. If ele- ments must be found at the sentencing stage, and not mere sen- tencing factors, then the Supreme Court, theoretically speaking, created a new substantive law. Put into its shortest form, capital murder is the highest kind of ?murder? in Arizona. To support this perception of the holding in Ring, the Ninth Circuit examined the history of capital sentencing in Arizona.120 The court noted that from 1919 to 1972, Arizona left capital sen- tencing in the complete discretion of the jury.121 However, in the wake of Furman v. Georgia,122 finding complete discretion of judge or jury unconstitutional, Arizona?s substantive capital sen- tencing structure was eliminated.123 After 1973, Arizona judges determined whether a defendant would be sentenced to death,124 with various changes to its structure mandated by Supreme Court 116 Ring v. Arizona, 536 U.S. 584, 609 (2002) (quoting Apprendi v. New Jersey, 530 U.S. 466, 494 n.19 (2000)). 117 Summerlin v. Stewart, 341 F.3d 1082, 1104-05 (9th Cir. 2003). 118 Id. at 1105. 119 Ring, 536 U.S. at 609. 120 Summerlin, 341 F.3d at 1102-05. 121 Id. at 1102. 122 408 U.S. 238 (1972) (per curiam). 123 Summerlin, 341 F.3d at 1103 (citing In re Tarr, 109 Ariz. 264 (1973)). 124 Summerlin, 341 F.3d at 1102-03. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 16 5-MAY-05 11:55 1404 OREGON LAW REVIEW [Vol. 83, 2004] decisions.125 The court of appeals then discussed its opinion in Adamson v. Ricketts,126 where it held the Arizona capital sen- tencing structure?s aggravating factors were elements of the dis- tinct offense of capital murder.127 However, the Ninth Circuit?s opinion in Adamson was overruled by the Supreme Court?s deci- sion in Walton.128 According to the court?s reasoning, once Ring overruled Walton, the Ninth Circuit?s interpretation that the ag- gravating factors were elements of the distinct crime of capital murder was restored.129 In noting that Ring was both a procedural and substantive de- cision, the Ninth Circuit found that all states must comply with the minimum procedural requirements outlined in Ring.130 How- ever, the court went on to state, as described above, that Ring was a substantive decision that should be applied retroactively.131 Because Ring merely extended the reasoning from Apprendi, and because the Ninth Circuit, and all other circuits, had found Apprendi was a procedural rule not applying retroactively,132 the court distinguished Ring and Apprendi on the substantive/proce- dural distinction.133 It reasoned that in Apprendi the Supreme Court directly stated New Jersey?s substantive criminal law was not at issue,134 but in Ring the substantive law was at issue.135 By the Ninth Circuit?s reasoning, the Supreme Court rendered a ?wholesale invalidation of Arizona?s capital sentencing 125 See, e.g., Alford v. Eyman, 408 U.S. 939 (1972); Lockett v. Ohio, 438 U.S. 586 (1978); Bell v. Ohio, 438 U.S. 637 (1978). 126 865 F.2d 1011 (9th Cir. 1988) (en banc). 127 Summerlin, 341 F.3d at 1104. 128 Id. 129 Id. at 1104-05. 130 Id. at 1106. 131 Id. 132 See, e.g., United States v. Buckland, 289 F.3d 558, 564 (9th Cir. 2003) (en banc) (holding that Apprendi was not a substantive determination, but applying the rule retroactively nevertheless, because the care arose on direct review); Goode v. United States, 305 F.3d 378, 382-85 (6th Cir. 2002); United States v. Sanchez- Cervantes, 282 F.3d 664, 670 (9th Cir. 2002) (holding Apprendi did not apply retro- actively under a Teague analysis for a case on collateral review); McCoy v. United States, 266 F.3d 1245, 1258 (11th Cir. 2001); United States v. Moss, 252 F.3d 993, 999-1000 (8th Cir. 2001); United States v. Sanders, 247 F.3d 139, 149-51 (4th Cir. 2001). 133 Summerlin, 341 F.3d at 1101-02, 1102 n.9. 134 Id. at 1101 (citing Apprendi v. New Jersey, 530 U.S. 466, 475 (2000)). 135 Summerlin, 341 F.3d at 1101. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 17 5-MAY-05 11:55 Death Anyways 1405 scheme.?136 The court relied on its reasoning that in capital cases the circumstances are unique in that the two crimes of capital murder and plain murder were split apart from the old (and un- constitutional) structure of just one class of murder, and that in creating such a new dichotomized structure in Ring, the Supreme Court had struck down the entire substantive law of murder in Arizona.137 B. Teague Analysis Applying the three-step Teague analysis described above, the Ninth Circuit first observed that Summerlin?s conviction was fi- nal when the Arizona Supreme Court denied ?rehearing of its opinion affirming his conviction and death sentence in 1984.?138 The court next determined that Ring announced a new rule that was not available to the Arizona courts before Summerlin ex- hausted all of his direct appeals.139 The court went on to determine whether either of the Teague exceptions applied to the Ring holding.140 The Ninth Circuit first held that ?[b]ecause Ring did not ?decriminalize a class of con- duct nor prohibit the imposition of capital punishment on a par- ticular class of persons, the first [Teague] exception is inapplicable? to the instant ruling.?141 However, the court also held that the Ring holding, if seen as a new procedural rule, fit the second Teague exception, thereby providing an alternative reason for applying the new constitutional rule retroactively.142 The court had several grounds for finding that Ring enhanced the accuracy of the proceeding.143 First, the court found that the new rule, on its face, improved the accuracy of the sentencing proceeding.144 The court, citing Sawyer v. Smith,145 noted that all capital sentencing procedures are aimed at improving the relia- bility and accuracy of the proceeding at issue.146 136 Id. 137 Id. 138 Id. at 1108 (citing State v. Summerlin, 675 P.2d 686 (Ariz. 1984)). 139 Summerlin, 341 F.3d at 1109. 140 Id. 141 Id. (quoting Graham v. Collins, 506 U.S. 461, 477 (1993) (quoting Saffle v. Parks, 494 U.S. 484, 495 (1990)). 142 Summerlin, 341 F.3d at 1121. 143 Id. at 1110-16. 144 Id. at 1110. 145 497 U.S. 227, 243 (1990). 146 Summerlin, 341 F.3d at 1110. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 18 5-MAY-05 11:55 1406 OREGON LAW REVIEW [Vol. 83, 2004] Second, the court found that ?fact-finding by a jury, rather than by a judge, is more likely to heighten the accuracy of capital sentencing proceedings.?147 It observed that sentencing proceed- ings with juries tend to resemble trial-like settings, with ?orderly presentation of evidence and argument.?148 The court noted penalty-phases presented to judges resembled mere ordinary sentencing proceedings.149 It also argued that this resulted in a significant amount of inadmissible evidence being used to decide a defendant?s fate.150 The court believed that these problems with the sentencing proceedings presented to judges led to less accurate determinations than would an orderly presentation to a jury.151 Third, the court argued that a large part of capital sentencing is the moral decision of deciding when a person should be sen- tenced to death.152 It noted that a jury is better equipped to ex- press the moral decision of whether to sentence a capital defendant to death because jurors are people from the commu- nity of the defendant.153 Also, because judges regularly sentence criminal defendants and may even regularly confront death pen- alty cases, the judge may be ?hardened,? or at least far ?less likely to reflect the current conscience of the community.?154 Fi- nally, the court of appeals noted that judges are affected by polit- ical pressures from having to run for office, suggesting that judges facing political pressure are more likely to impose the death penalty.155 147 Id. 148 Id. 149 Id. at 1110-11. 150 Id. at 1111-12. 151 Id. at 1113. 152 Id. 153 Id. 154 Id. at 1114. 155 Id. (citing John Blume & Theodore Eisenberg, Judicial Politics, Death Penalty Appeals, and Case Selection: An Empirical Study, 72 S. CAL. L. REV. 465, 470-75 (1999)); Stephen B. Bright et al., Breaking the Most Vulnerable Branch: Do Rising Threats to Judicial Independence Preclude Due Process in Capital Cases, 31 COLUM. HUM. RTS. L. REV. 123 passim (1999); 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, 793-94 (1995); Sam Kamin, Harmless Error and the Rights/Remedies Split, 88 VA. L. REV. 1, 62 (2002); Politics and the Death Penalty: Can Rational Discourse and Due Process Survive the Perceived Politi- cal Pressure?, 21 FORDHAM URB. L.J. 239, 270-73 (1994); Fred B. Burnside, Com- ment, Dying to Get Elected: A Challenge to the Jury Override, 1999 WIS. L. REV. 1017, 1039-44 (1999). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 19 5-MAY-05 11:55 Death Anyways 1407 The Ninth Circuit also found that Ring announced a new wa- tershed rule.156 It first found the Ring ruling ?fundamentally al- tered the procedural structure of capital sentencing applicable to all states.?157 This restructuring of capital sentencing was so fun- damental, according to the court, that it was a structural change in capital sentencing procedure that led to a finding that Ring was a watershed rule.158 Providing support for the finding that Ring announced a ?structural? change in capital sentencing was the fact that Ring error is not capable of harmless-error re- view.159 The court pointed out that because ?the wrong entity found Summerlin to be guilty of a capital crime . . . there was no jury verdict within the meaning of the Sixth Amendment and no constitutionally cognizable finding to review.?160 Put another way, the court could only speculate as to what the effect of the error in Summerlin?s sentencing proceeding was because Sum- merlin was provided the wrong fact-finder and the court would have to guess what a ?hypothetical jury? would have done.161 The court concluded by stating that ?Ring error is one ?affecting the framework within which the trial proceeds, rather than sim- ply an error in the trial process itself.??162 To complete its watershed analysis, the court of appeals ex- amined whether Ring was ?truly watershed,? and so fundamental that it announced a right ?implicit in the concept of ordered lib- erty.?163 The Ninth Circuit found Ring was such a ruling.164 It first noted Ring?s impact was far greater than the impact of the ?Mills/McKoy? rule which was found to be a ?watershed rule? by other circuits.165 The Mills/McKoy rule holds that a state may not limit mitigating evidence from the jury?s consideration of a capital sentence, even if the jury is not uninanimous in finding 156 Summerlin, 341 F.3d at 1116-21. 157 Id. at 1116. 158 Id. 159 Id. at 1116-17. 160 Id. at 1117. 161 Id. (citing Sullivan v. Louisiana, 508 U.S. 275, 280 (1993)); Bollenbach v. United States, 326 U.S. 607, 614 (1946). 162 Summerlin, 341 F.3d at 1119 (quoting Neder v. United States, 527 U.S. 1, 8 (1999) (quoting Arizona v. Fulminante, 499 U.S. 279, 310 (1991))). 163 Summerlin, 341 F.3d at 1119 (citations omitted). 164 Id. at 1121. 165 Id. at 1120. Ironically, after the Ninth Circuit issued its Summerlin opinion, the Supreme Court held that the new Mills rule does not apply retroactively, specifi- cally noting that Mills/McKoy did not announce a new watershed rule. Beard v. Banks, 124 S. Ct. 2504, 2508 (2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 20 5-MAY-05 11:55 1408 OREGON LAW REVIEW [Vol. 83, 2004] the mitigating evidence.166 Comparing the Ring rule to the Mills/ McKoy rule, the Ninth Circuit observed that ?Ring does not merely announce a supplemental procedural safeguard.?167 The court also noted that Ring rendered the constitutionality of one- fourth of capital punishment states? sentencing procedures inva- lid, and affected every single capital sentencing scheme in the country.168 Accordingly, Ring ?altered the fundamental bedrock principles applicable to capital murder trials.?169 The Ninth Circuit also had to distinguish Ring from Apprendi because in United States v. Sanchez-Cervantes, the Ninth Circuit held Apprendi did not apply retroactively under a Teague analy- sis.170 Clearly, holding Apprendi not to apply retroactively was related to whether the court should hold that Ring applies retro- actively, because the Supreme Court had stated Ring was merely an extension of Apprendi. The court distinguished Apprendi from Ring, for retroactivity purposes, in at least five ways: (1) Apprendi was not a decision of substantive law, in that it did not declare the statute unconstitutional; (2) Apprendi errors are not structural and are subject to harmless-error review; (3) Apprendi did not improve the accuracy of the sentencing proceeding and it was not sweeping because it would apply only in a limited num- ber of cases; (4) capital cases are structurally different than non- capital cases because the sentencing proceeding of a capital case with a jury resembles a trial; and (5) the Eighth Amendment of the Constitution imposes a heightened analysis on capital trials that was not present in the analysis of Apprendi.171 IV THE SUPREME COURT?S HOLDING AND RATIONALE The Supreme Court overturned the Ninth Circuit?s decision that Ring applied retroactively.172 The Court addressed the two 166 McKoy v. North Carolina, 494 U.S. 433, 435 (1990); Mills v. Maryland, 486 U.S. 367, 384 (1988). 167 Summerlin, 341 F.3d at 1120. 168 Id. (citing Ring v. Arizona, 536 U.S. 584, 621 (2002) (O?Connor, J., dissent- ing)); Brief for Amici Curiae Alabama, Colorado, Delaware, Florida, Idaho, Indi- ana, Mississippi, Montana, Nebraska, Nevada, New York District Attorney?s Ass?n, Pennsylvania, South Carolina, Utah, Virginia In Support of Respondent to the Su- preme Court in Ring at 4 n.2 (No. 01-488), available at 2002 WL 481140. 169 Summerlin, 341 F.3d at 1120. 170 282 F.3d 664, 670 (9th Cir. 2002). 171 Summerlin, 341 F.3d at 1121. 172 Schriro v. Summerlin, 124 S. Ct. 2519, 2527 (2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 21 5-MAY-05 11:55 Death Anyways 1409 primary holdings found by the Ninth Circuit, disagreeing with both of them.173 First, the Court ruled that Ring did not an- nounce a new substantive rule, because it merely altered the pro- cedure for determining whether a person is subject to a death sentence.174 Second, the Court found that Ring did not fit the second Teague exception, because judicial fact-finding does not seriously diminish the accuracy of an underlying proceeding.175 Justice Breyer, joined by three other Justices,176 filed a dissenting opinion arguing that there were three reasons the accuracy in Summerlin?s sentencing proceeding was seriously diminished in light of the Court?s Ring holding. A. Ring Is Not a New Rule of Criminal Procedure The Court began by making a somewhat significant clarifica- tion to retroactivity law. It initially explained that new substan- tive rules apply retroactively, citing Bousley, ?as well as constitutional determinations that place particular conduct or persons covered by the statute beyond the State?s power to pun- ish,? which is the first Teague exception.177 The Court went on to note that the Bousley rule is not separate from this first Teague exception.178 Specifically, the Court stated, ?[w]e have some- times referred to rules of this latter type as falling under an ex- ception to Teague?s bar on retroactive application of procedural rules . . . they are more accurately characterized as substantive rules not subject to the bar.?179 Accordingly, the Court made anomalous the Ninth Circuit?s holding that Ring was a substan- tive rule under Bousley which did not fit the first Teague exception. The Court then analyzed whether Ring announced a new sub- stantive rule of criminal law.180 It explained that in making the substantive/procedural distinction, a ?rule is substantive rather than procedural if it alters the range of conduct or the class of persons that the law punishes,? but is procedural if it ?regulate[s] 173 Id. at 2526-27. 174 Id. at 2523. 175 Id. at 2526. 176 Justices Stevens, Souter, and Ginsburg. See id. at 2527. 177 Id. at 2522. 178 Id. at 2522-23, 2523 n.4. 179 Id. at 2522 n.4 (citation omitted). 180 Id. at 2523-24. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 22 5-MAY-05 11:55 1410 OREGON LAW REVIEW [Vol. 83, 2004] only the manner of determining the defendant?s culpability.?181 Applying this method of distinguishing the two, it found that Ring did not ?alter the range of conduct Arizona law subjected to the death penalty.?182 In fact, the Court noted that because Ring was decided entirely upon Sixth Amendment jury-trial grounds, something that seems procedural on its face, it could have ?nothing to do with the range of conduct a State may criminalize.?183 The opinion also noted that rules allocating deci- sion-making authority are ?prototypical procedural rules.?184 The Court also rejected Summerlin and the Ninth Circuit?s ar- gument that Ring restructured the elements of Arizona?s aggra- vated murder structure to create a new offense of ?capital murder.?185 It began by explaining that a decision which modi- fies the elements of an offense usually announces a substantive rule.186 However, the Court explained that modification of ele- ments refers to the altering of ?the range of conduct the statute [in question] punishes.?187 In the case of Ring, there was no al- tering of the range of conduct that was punishable after the opin- ion.188 The Court in Ring did not hold that certain elements were necessary for Arizona to sentence a person to death. Rather, Ring merely held that Arizona must have juries decide the elements that Arizona chose, because those elements were ?factual? elements.189 In other words, Ring found that certain sentencing elements, as labeled by the Arizona state legislature, were the functional equivalent of criminal elements. Ring merely relabeled Arizona?s sentencing aggravators by looking at substance over form. However, at the end of the day, this re- labeling did not affect the conduct which is captured as criminal. Rather, it was a relabeling that affected who decides, as pre- scribed by the Sixth Amendment.190 181 Id. at 2523. 182 Id. 183 Id. 184 Id. 185 Id. at 2524. 186 Id. 187 Id. 188 Id. 189 Id. 190 The Court also criticized the Ninth Circuit?s holding for federalism reasons. It noted, as discussed in note 99, supra, that the Arizona Supreme Court did not find that Ring reordered Arizona?s capital murder structure in State v. Towery, 64 P.3d 828, 832-33 (Ariz. 2003). Schriro, 124 S. Ct. at 2524. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 23 5-MAY-05 11:55 Death Anyways 1411 B. Ring Is Not a Watershed Rule of Criminal Procedure Because the Supreme Court rejected the Ninth Circuit?s find- ing that Ring announced a new substantive rule, it still had to determine whether Ring fell under the watershed exception to Teague.191 In making this determination, the Court did not speak to whether Ring was a fundamental decision, ?implicit in the concept of ordered liberty.?192 Instead, it focused entirely on whether a judicial determination of the aggravating factors under Arizona?s pre-Ring statute ?seriously diminished [the] accuracy? of Summerlin?s sentencing proceeding.193 Despite the Ninth Cir- cuit?s list of reasons why juries are more accurate factfinders than judges, the majority held that it did not.194 The Court stated that rather than looking to see whether juries are better factfinders than judges, the Court must determine ?whether judicial factfind- ing so ?seriously diminishe[s]? accuracy that there is an ?imper- missibly large risk? of punishing conduct the law does not reach.?195 Applying this standard, the Court found that evidence sup- porting the questionability of judicial factfinding was too equivo- cal to support a conclusion that the accuracy of Summerlin?s sentencing proceeding was seriously diminished.196 Essentially, it found that because there was no clear answer to whether judicial factfinding is more or less accurate than that of juries, it could not be said that the accuracy of Summerlin?s conviction was seri- ously diminished.197 The Court pointed out that ?for every argu- ment why juries are more accurate factfinders, there is another why they are less accurate.?198 Upon finding this issue to be in- conclusive and controversial, the Court stated, ?[w]hen so many presumably reasonable minds continue to disagree over whether juries are better factfinders at all, we cannot confidently say that 191 Schriro, 124 S. Ct. at 2524. 192 Id. at 2527 (Breyer, J., dissenting) (quoting Teague v. Lane, 489 U.S. 288, 311- 13 (1989)). 193 Schriro, 124 S. Ct. at 2525-26. 194 Id. at 2524-25. 195 Id. at 2525 (quoting Teague v. Lane, 489 U.S. 288, 312-13 (1989) (quoting De- sist v. United States, 394 U.S. 244, 262 (1969) (Harlan, J., dissenting))). 196 Schriro, 124 S. Ct. at 2525. 197 Id. 198 Id. The Court noted that although the Ninth Circuit majority listed reasons why juries are more accurate, the dissent in the Ninth Circuit?s Summerlin opinion had advanced several arguments why juries are less accurate than judges. Id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 24 5-MAY-05 11:55 1412 OREGON LAW REVIEW [Vol. 83, 2004] judicial factfinding seriously diminishes accuracy.?199 The Court found precedential support for its finding that judi- cial factfinding was not seriously inaccurate in the pre-Teague de- cision of DeStefano v. Woods.200 In DeStefano the Supreme Court was presented with the issue of whether Duncan v. Louisi- ana,201 where the Court held that ?the States cannot deny a re- quest for jury trial in serious criminal cases,?202 applied retroactively.203 The DeStefano court gave three different rea- sons for holding that Duncan did not apply retroactively,204 one of which was important to the Schriro court.205 In finding that the purpose of the Duncan rule would not be served by retroac- tive application, DeStefano stated, ??[w]e would not assert, how- ever, that every criminal trial . . . held before a judge alone is unfair or that a defendant may never be as fairly treated by a judge as he would be by a jury.??206 In Schriro, the Court be- lieved that this was a finding by the DeStefano Court that an entire criminal trial without a jury was not impermissibly inaccu- rate.207 Accordingly, a judge finding only sentencing aggravators could not be any more inaccurate. The Schriro majority also addressed two significant points from the dissent.208 First, in response to the dissent?s claim that juries are more accurate because they are more adept at deciding factors that call for the weighing of community standards, the majority strictly read Arizona?s sentencing aggravators, noting that nowhere did Arizona?s death sentence statute require them to be ?determined by community standards.?209 Second, in re- sponse to the dissent?s arguments on the theme that death sentences should be viewed differently when it comes to retroac- tivity, the Court used formalistic reasoning, stating that this was not an application of Teague, but of a different sort of balancing 199 Id. 200 Id. at 2525 (citing DeStefano v. Woods, 392 U.S. 631 (1968)(per curiam)). 201 391 U.S. 145 (1968). 202 DeStefano, 392 U.S. at 632. 203 Id. at 633. 204 Id. at 633-34. 205 Schriro v. Summerlin, 124 S. Ct. 2519, 2525-26 (2004). 206 DeStefano, 392 U.S. at 633-34 (quoting Duncan v. Louisiana, 391 U.S. 145, 158 (1968)). 207 Schriro, 124 S. Ct. at 2526. 208 Id. 209 Id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 25 5-MAY-05 11:55 Death Anyways 1413 test that is nowhere found in Court precedent.210 C. Justice Breyer?s Dissent Justice Breyer?s dissent, joined by three other justices, began with the assumption that Ring announced a rule that meets the ?implicit in the concept of ordered liberty? piece of the Teague exception, reasoning that the majority did not deny this piece of the test and citing a number of cases that indicated the jury trial right is a ?fundamental guarantee.?211 Accordingly, like the ma- jority, the dissent focused entirely on whether Ring significantly undermined the accuracy of the underlying proceeding.212 The dissent provided three primary reasons why having a judge as the factfinder in a death-sentence proceeding seriously undermines the accuracy of the proceeding.213 First, it noted that the necessary aggravators for a death sen- tence are full of highly subjective, value-laden terms, which clearly imply that they should be evaluated according to ?com- munity standards,? which can only be provided by a jury.214 Jus- tice Breyer specifically pointed out one of Arizona?s aggravators, which required a finding that the crime ?was committed in an ?especially heinous, cruel, or depraved manner.??215 Second, Justice Breyer examined the policies underlying the Teague rule of non-retroactivity and balanced these competing interests to find that Ring should be applied retroactively.216 He began by noting two objectives of habeas corpus law, specifically, protecting the innocent and uniformity among all persons.217 He argued that these considerations are more pointed in the death penalty context because Eighth Amendment jurisprudence re- quires greater scrutiny because of the severity and irrevocability of the punishment.218 On the uniformity issue, the dissent noted the arbitrariness of a person being put to death under a proce- dure acknowledged as unconstitutional and how, in the death penalty context, that is unique because none of the ?sentence? 210 Id. The Court noted that even if it were willing to reconsider Teague, it would not be willing to adopt such a balancing formulation. 211 Id. at 2527 (Breyer, J., dissenting). 212 Id. 213 Id. at 2528. 214 Id. 215 Id. (quoting ARIZ. REV. STAT. ANN. ? 13-703(F)(6) (West Supp. 2003)). 216 Schriro, 124 S. Ct. at 2528. 217 Id. at 2528. 218 Id. at 2528-29. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 26 5-MAY-05 11:55 1414 OREGON LAW REVIEW [Vol. 83, 2004] has been carried out by the time the Court catches its constitu- tional mistake.219 The dissent also discussed the policy interests usually cited for not applying new rules of criminal procedure retroactively and how these interests were not particularly weighty in regards to Ring.220 The dissent noted that one inter- est in non-retroactivity is conservation of state resources, but ar- gued this was not weighty in this case because Ring only affected approximately 110 individual cases.221 Further, finality, a policy often cited for non-retroactivity, is not as important in the capital context as in usual criminal cases because the death sentence process goes on for many years anyway, and finality should be discounted due to the nature of the sentence.222 Third, the dissent argued that the majority?s reliance on De- Stefano was not persuasive because DeStefano was a pre-Teague case and the majority singled out only one of the three factors that DeStefano relied on in coming to its conclusion.223 The dis- sent noted that the other two factors discussed in DeStefano for non-retroactivity of the Duncan rule were quite persuasive, while the one factor seized by the Schriro Court was not as persuasive standing on its own.224 Further, the dissent distinguished DeStef- ano by noting Ring is quite different than the Duncan rule: it applies to a small subclass of defendants; the relative harm was greater to defendants deprived of Ring; administration of justice is not as difficult with Ring; and there were hardly any reliance interests damaged by Ring.225 V IMPLICATIONS A. The Supreme Court?s Substantive/Procedural Distinction The Schriro Court?s clarification of the substantive/procedural distinction is a positive movement in the law of habeas retroac- tivity for a number of reasons: it provides much need clarity in the field; it is appropriate in light of Supreme Court precedent; it follows the policies underlying the Teague rule; and it will pro- 219 Id. 220 Id. at 2529-30. 221 Id. at 2530. 222 Id. 223 Id. 224 Id at 2530-31. 225 Id. at 2531. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 27 5-MAY-05 11:55 Death Anyways 1415 vide far greater certainty for future retroactivity cases. The Court clarified that there are not two separate procedural Teague exceptions and a separate substantive Bousley rule that can lead to retroactive application of new rules of criminal law.226 Rather, there is a rule of retroactivity for new substantive rules and a rule of non-retroactivity for new procedural rules of criminal law, un- less the new rule fits the one ?watershed? exception.227 This ex- planation clarified a developing problem. Namely, there appeared to be little difference between the first Teague excep- tion and what the Ninth Circuit termed the ?substantive? excep- tion to the non-retroactivity bar. Nevertheless, courts began analyzing Teague issues under such a framework.228 The result was ambiguity, uncertainty, and bizarre results, such as the Ninth Circuit?s highly tenuous finding that Ring announced a substan- tive rule.229 A rule like Ring, about ?who decides? in a sentenc- ing proceeding, appears to be the classic example of a procedural rule, as the Schriro majority pointed out.230 For the Ninth Cir- cuit to find to the contrary illustrates the pre-Schriro framework in this area of Teague jurisprudence was lacking. Of course, prior to Schriro there was relatively little written on the distinc- tion of substance and procedure in the habeas context,231 and as the Ninth Circuit illustrated, the door was wide open for any fed- eral court to overturn a state?s conviction if it so desired. The Supreme Court?s Schriro opinion began to fill that gap. The Supreme Court?s explanation of the difference between procedural and substantive rules is simple and clear. If a new rule of constitutional criminal law ?alters the range of conduct or the class of persons that the law punishes,? then it is substantive and applies retroactively.232 The focus is on whether the new rule affects the number of people or the amount of conduct sub- ject to the criminal statute under which the petitioner was con- 226 See infra Part V.A. 227 Schriro, 124 S. Ct. at 2522-23, 2523 n.4. 228 See, e.g., Summerlin v. Stewart, 341 F.3d 1082 (9th Cir. 2003); Coleman v. United States, 329 F.3d 77, 83 (2d Cir. 2003); Santana-Madera v. United States, 260 F.3d 133, 138 (2d Cir. 2001). 229 Judge Rawlinson?s dissenting opinion in Summerlin noted ?the majority opin- ion wanders afield? in its holding that Ring announced a substantive decision. Sum- merlin, 341 F.3d at 1125. 230 Schriro v. Summerlin, 124 S. Ct. 2519, 2523 (2004). 231 Ethan Isaac Jacobs, Note, Is Ring Retroactive?, 103 COLUM. L. REV. 1805, 1828 (2003). 232 Schriro, 124 S. Ct. at 2523. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 28 5-MAY-05 11:55 1416 OREGON LAW REVIEW [Vol. 83, 2004] victed. In contrast, a new rule is procedural if it regulates the ?manner of determining the [criminal] defendant?s culpabil- ity.?233 The critical piece for determining whether a rule is pro- cedural, rather than substantive, appears to be whether it addresses the ?manner of determining? culpability.234 This stan- dard does not leave room for courts to examine other factors, like the Ninth Circuit?s tracing Arizona?s death penalty statute throughout the twentieth century, to come to the conclusion that Ring, a rule about who is the fact-finder, is a substantive rule of criminal law. The Schriro Court?s relatively simple and dichoto- mous standard for determining whether a new rule is procedural or not should cut off confusion when the Supreme Court issues future rules of constitutional criminal law. Further, Schriro?s holding regarding the substantive/pro- cedural distinction appears to be in accord with the spirit of Su- preme Court precedent on this issue. The analysis begins by closely looking at Bousley v. United States,235 a case relied upon by the Ninth Circuit in Summerlin and the Supreme Court in Schriro.236 In Bousley, the Supreme Court held that ?because Teague by its terms applies only to procedural rules, we think it is inapplicable to the situation in which this Court decides the meaning of a criminal statute enacted by Congress.?237 The focus of Bousley was whether the Supreme Court?s decision in Bailey v. United States238 applied retroactively on collateral review.239 Bailey held that use of a firearm under a firearms statute240 re- quired active employment as opposed to mere possession of the firearm.241 In Bousley, the Court held the Bailey rule was not barred by Teague because the narrowing of the term ?use? cre- ated ?a significant risk that a defendant stands convicted of ?an act that the law does not make criminal.??242 The opinion ob- 233 Id. (emphasis omitted). 234 Id. This appeared to be an important piece to the Court, as the Court empha- sized this phrase in text of its opinion. 235 523 U.S. 614 (1998). 236 See Summerlin v. Stewart, 341 F.3d 1082, 1099 (9th Cir. 2003); Schriro, 124 S. Ct. at 2523. In fact, the Supreme Court cited Bousley when giving its definition of ?substantive? and ?procedural? rules. Schriro, 124 S. Ct. at 2523. 237 Bousley, 523 U.S. at 620. 238 516 U.S. 137 (1995). 239 Bousley, 523 U.S. at 618. 240 18 U.S.C. ? 924(c)(1) (1994). 241 Bailey, 516 U.S. at 144. 242 Bousley, 523 U.S. at 620 (quoting Davis v. United States, 417 U.S. 333, 346 (1974)). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 29 5-MAY-05 11:55 Death Anyways 1417 served it would be ?inconsistent with the doctrinal underpinnings of habeas review?243 to bar collateral relief based on new rules like Bailey, where the Court interpreted the relevant statute to not reach certain conduct as criminal.244 Bousley established that a substantive decision is one that pro- vides that Teague will not block the retroactive application of a new rule that alters the scope of a criminal statute. The concern in Bousley was that a new rule had been established significantly limiting the amount of conduct that was ?criminal? under a fed- eral statute. The Court?s Schriro definition of ?substantive? looks directly at that concern?whether the new rule limits the conduct punishable as criminal. Bousley was never intended to apply to a rule like Ring because the only shift that was made was ?who decides.?245 Accordingly, the Schriro Court?s formula- tion of the substantive/procedural distinction appears to be more in line with this precedent than the Ninth Circuit?s opinion. The Schriro Court?s substantive/procedural distinction is also in line with the underlying doctrinal principles of the Teague rule. Beginning in the mid-1960s, Justice Harlan began writing a number of dissenting and concurring opinions which later be- came the foundation for the Court?s non-retroactivity rule in Teague.246 In explaining the first exception, Justice Harlan stated that ?[t]here is little societal interest in permitting the criminal process to rest at a point where it ought properly never to re- pose.?247 Justice Harlan went on further to give as an example ?[n]ew ?substantive due process? rules? that ?free[ ] individuals from punishment for conduct that is constitutionally pro- tected.?248 This understanding of the first Teague exception illus- trates that the determination of whether the new constitutional 243 Bousley, 523 U.S. at 621. 244 Id. at 620. 245 In fact, after the Supreme Court issued its opinion in Ring, the Arizona legisla- ture amended the State?s capital sentencing statute. See Act of Apr. 27, 2001, ch. 260, ? 1, 2001 Ariz. Sess. Laws 1334, 1334 (codified as amended at ARIZ. REV. STAT. ? 13-703.01 (2003)). It did not change the terms of the aggravating factors necessary to impose a sentence of death; rather, it only changed the factfinder from judge to jury. Id. 246 See United States v. Johnson, 457 U.S. 537, 546-47 nn.9-10 (1982) (citing opin- ions); Mackey v. United States, 401 U.S. 667, 675 (1971) (Harlan, J., concurring in part and dissenting in part); Desist v. United States, 394 U.S. 244, 256-57 (1969) (Harlan, J., dissenting). For an account of the historical development of non-retro- activity, see 2 HERTZ & LIEBMAN, supra note 66, ? 25.2, at 1034-45. 247 Mackey, 401 U.S. at 693 (Harlan, J., concurring in part and dissenting in part). 248 Id. at 692-93. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 30 5-MAY-05 11:55 1418 OREGON LAW REVIEW [Vol. 83, 2004] rule is substantive or procedural should focus on whether, in the end, the rule limits conduct that may be punished as criminal. Hertz and Liebman agree with this reading, stating, ?[i]n general, the first exception may be interpreted as distinguishing new rules of substantive criminal law, which always apply retroactively, from new rules of criminal procedure, which generally do not ap- ply retroactively in cases that were final as of the time the new rule was announced.?249 In other words, merging the Bousley ruling with the first Teague exception fits the underlying policy and intent of the original Teague rule. Aside from the Court?s precedent, rejecting the Ninth Circuit?s reasoning that Ring was a substantive rule averted the possibility of not being able to find a logical end to what is not substantive. Particularly, the Ninth Circuit?s distinction of Ring from Ap- prendi were unavailing, despite the fact that the Ninth Circuit had held Apprendi does not apply retroactively.250 The Ninth Circuit?s explanation that Ring established ?capital murder? and plain ?murder? in Arizona could just as easily have applied to Apprendi. One could say Apprendi reorganized the elements of assault plus the aggravator into two distinct crimes: assault and hate-motivated assault. The former was punishable by a maxi- mum of ten years in New Jersey, the latter punishable by up to twenty. Accordingly, the reasoning used by the Ninth Circuit to hold that Ring announced a substantive rule could apply to countless situations and is the crux of the problem with the court?s reasoning?it gutted the Teague rule. It removed whatever cases a court wants to exempt from the bar of retroac- tivity. A conception of the substantive/procedural distinction without a clear ending point is problematic in this area because such distinctions have never been easy.251 The Ninth Circuit?s ?non-existent? distinctions from Apprendi, as well as its overall characterization of substantive versus procedural, provided fu- ture courts with little guidance or potential consistency. Not sur- prisingly, the Supreme Court immediately corrected these problems by its formulation of the substantive/procedural distinction. The Supreme Court?s policy interests in Teague?s retroactivity 249 2 HERTZ & LIEBMAN, supra note 66, ?25.7, at 1119. 250 See supra Parts III.B, IV.A. 251 D. Michael Risinger, ?Substance? and ?Procedure? Revisited with Some After- thoughts on the Constitutional Problems of ?Irrebuttable Presumptions,? 30 UCLA L. REV. 189, 189-90 (1982). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 31 5-MAY-05 11:55 Death Anyways 1419 holding are also advanced by the Schriro Court?s substantive/ procedural distinction. The Court has offered roughly three ex- planations for its rule of non-retroactivity of new rules of crimi- nal procedure: (1) applying new rules retroactively does not advance the deterrent purpose of habeas corpus law, specifically meaning habeas corpus usually is a way to get state courts to abide by the United States Constitution; (2) retroactivity of new constitutional rules frustrates judicial concern over comity and finality; and (3) in evaluating the relative costs and benefits of retroactivity, the costs imposed upon the states far outweigh the benefits.252 These policies are furthered by defining a substan- tive rule, for Teague purposes, as only one that readjusts the con- duct punishable as criminal. Such a rule limits the number of new rules being applied retroactively, advances comity and final- ity, and does not impose such a great cost on the states.253 It also creates far more clarity as to what constitutes a substantive ver- sus a procedural rule. For example, no court other than the Ninth Circuit seriously questioned whether Ring was procedural or substantive because on its face the decision was procedural. Allowing courts to read into a ruling, probing around for a com- prehensive history of the development of a criminal statute, will result in far less finality of criminal judgments. Courts could be- gin splitting over what is and is not a new substantive rule, as was the case under Ring after the Ninth Circuit?s Summerlin decision. In sum, the distinction between procedural and substantive rules under Teague greatly needed to be clarified, and was by Schriro. This was an area with little guidance and a significant probability for conflicting interpretations. The Court?s answer came in a form that fits its precedent as well as the overall poli- cies underlying the Teague rule. B. The Teague ?Watershed? Analysis As discussed above, the Supreme Court concluded the Ring 252 Eric J. Beane, Note, When It Comes to Capital Sentencing, You Be the Judge: Ring v. Arizona, 45 ARIZ. L. REV. 225, 233 (2003). 253 Additionally, retroactive application of new substantive rules that affect only the scope of conduct punishable avoids the significant costs to the states of re-prose- cuting individuals because, theoretically, the individual did not commit a crime. Such is not the case with new procedural rules being applied retroactively?the state will likely incur significant costs re-prosecuting the potentially thousands of affected prisoners. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 32 5-MAY-05 11:55 1420 OREGON LAW REVIEW [Vol. 83, 2004] rule did not fit the procedural Teague exception, basing its entire discussion on the fact that judicial factfinding in death-sentence hearings did not significantly decrease the accuracy of the under- lying sentence. However, the Ninth Circuit?s decision not only found that such judicial decision-making significantly decreased the accuracy of the underlying conviction, but also that the Ring rule was a watershed one, which must be applied retroactively. This Part will begin by explaining why such a holding by the Ninth Circuit had the effect of significantly lowering the bar for retroactivity law, and will conclude with a discussion of how the Supreme Court?s opinion, at the least, maintained the pre-Sum- merlin bar for retroactivity. Language used by the Supreme Court in prior retroactivity opinions defining and describing the procedural Teague excep- tion illustrates its narrowness. The new rule must be ?implicit in the concept of ordered liberty,?254 and it must be a ?procedure[ ] without which the likelihood of an accurate conviction [or sen- tence] is seriously diminished.?255 The Court has further re- quired that there be an issue of fundamental fairness.256 Only then is the new rule a watershed one.257 Not only has the Court used language describing this exception as narrow, but in practice it is a very narrow exception as well, ?encompassing only a handful of rulings, at least outside the [Eighth] Amendment area.?258 Further, the Court recently noted that it has never held a rule to apply retroactively under this exception.259 Illustrating the narrowness of this exception, 254 Teague v. Lane, 489 U.S. 288, 307 (1989) (plurality) (quoting Mackey v. United States, 401 U.S. 667, 692 (1971) (Harlan, J., concurring in part and dissenting in part)); accord Bousley v. United States, 523 U.S. 614, 620 (1998); O?Dell v. Nether- land, 521 U.S. 151, 157 (1997) (characterizing second exception as ?apply[ing] only to a small core of rules?); Lambrix v. Singletary, 520 U.S. 518, 527 (1997) (describing both Teague exceptions as ?narrow?); Goeke v. Branch, 514 U.S. 115, 120 (1995) (per curiam). 255 Teague, 489 U.S. at 312-13; see also Bousley, 523 U.S. at 620; Goeke, 514 U.S. at 120; Withrow v. Williams, 507 U.S. 680, 700 (1993) (O?Connor, J., concurring in part and dissenting in part); Penry v. Lynaugh, 492 U.S. 302, 314 (1989). 256 See, e.g., Teague, 489 U.S. at 320-21 (Stevens, J., concurring). 257 See, e.g., O?Dell, 521 U.S. at 167; Lambrix, 520 U.S. at 539-40; Goeke, 514 U.S. at 120; Caspari v. Bohlen, 510 U.S. 383, 396-97 (1994); Gilmore v. Taylor, 508 U.S. 333, 344-45 (1993); Graham v. Collins, 506 U.S. 461, 477-78 (1993); Saffle v. Parks, 494 U.S. 484, 495 (1990). 258 2 HERTZ & LIEBMAN, supra note 66, ? 25.7, at 1121. For an excellent listing of the Supreme Court?s decisions illustrating how narrow this exception has been ap- plied, as well as lower court?s decisions, see id. at 1121-24, n.27. 259 Beard v. Banks, 124 S. Ct. 2504, 2514 (2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 33 5-MAY-05 11:55 Death Anyways 1421 the Supreme Court in O?Dell determined that Simmons v. South Carolina,260 which held that a capital defendant could present evidence of ineligibility of parole to rebut the state?s argument of future dangerousness, did not fit the second Teague exception.261 The Court reasoned that the Simmons rule was not ?on par? with the Court?s decision from Gideon v. Wainwright,262 as the petitioner had argued, and that the ?narrow right of rebuttal that Simmons affords to defendants in a limited class of capital cases? was not a watershed rule, as envisioned by Teague.263 Despite the fact that O?Dell was a capital case, the Court dismissed the petitioner?s argument with little reasoning. Another example of how narrow the Court has interpreted this exception is in Goeke v. Branch.264 In that case the Court held that ?[b]ecause due process does not require a State to provide appellate process at all, a former fugitive?s right to appeal [denied here under a state rule forbidding prisoners who escape following conviction to ap- peal] cannot be said to be so central to an accurate determination of innocence or guilt as to fall within this [second] exception to the Teague bar.?265 There have also been a number of cases the Court has held do not apply retroactively that might have been thought to qualify under the second Teague exception. An example is Sawyer v. Smith,266 where the Court held as non-retroactive the rule from Caldwell v. Mississippi,267 which states that it is constitutionally impermissible for a prosecutor to diminish responsibility of ju- rors in their capital-sentencing duties.268 The Court reasoned that this rule was not within the second Teague exception be- cause it is not an ?absolute prerequisite to fundamental fairness,? although the Court did note the Caldwell rule was ?directed to- ward the enhancement of reliability and accuracy.?269 260 512 U.S. 154 (1994). 261 O?Dell, 521 U.S. at 167. 262 372 U.S. 335 (1963). 263 O?Dell, 521 U.S. at 167. 264 514 U.S. 115 (1995). 265 Id. at 120 (citations and internal quotations omitted). 266 497 U.S. 227 (1990). 267 472 U.S. 320, 328-29 (1985). 268 Sawyer, 497 U.S. at 244. 269 Id. (quoting Teague v. Lane, 489 U.S. 288, 314 (1989)). For a listing of other examples that the Court surprisingly did not apply retroactively, see 2 HERTZ & LIEBMAN, supra note 66, ? 25.7, at 1122 n.27 (citing examples such as Butler v. Mc- Kellar, 494 U.S. 407 (1990), Allen v. Hardy, 478 U.S. 255 (1986), and Solem v. Stumes, 465 U.S. 638 (1984)). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 34 5-MAY-05 11:55 1422 OREGON LAW REVIEW [Vol. 83, 2004] With the fact that the second Teague exception applies to only a ?small core of rules?270 in mind, it is evident that the bar for overcoming the presumption of non-retroactivity is extremely high. The Ninth Circuit?s holding that Ring is one of those rare cases that fits the exception lowered that bar. The finding that a jury conducting capital sentencing increases the accuracy of the underlying sentence, relative to the judge?s finding, essentially devolved the ?improves accuracy? element of the second excep- tion into a policy decision. After Ring, none of the evidence presented will be any different. As mentioned above, none of the aggravating or mitigating factors have been required to be any different.271 The difference between whether a judge or jury is more accurate can only be decided normatively. Further, every argument presented by the Summerlin court that a jury is more accurate than a judge can be countered. Accordingly, it is not surprising that the Ninth Circuit majority and dissent split over this issue.272 For example, the court argued judge-based sentencing phases are too short; the counter is that judges are more efficient and cut to the heart of the issues.273 The court argued the jury better reflects the community; the counter is that the judge is more experienced and does not allow defendants with more emotional appeal to be treated differently than the more or less culpable defendant who has lesser counsel or a less emotional case.274 The point is not that the Summerlin majority was wrong in its normative belief that the jury is more accurate. The concern is that the only way to find the jury is more accurate than a judge is a normative policy belief about the role of the decision-maker. Accordingly, the Summerlin majority?s finding on this issue turned the question of whether the new rule im- proves the ?accuracy of the proceeding? into a policy question, depending on one?s belief about the nature of judge versus jury decision-making. Allowing this aspect of the second Teague ex- ception to turn on a policy determination certainly lowers the bar for retroactivity. 270 Graham v. Collins, 506 U.S. 461, 478 (1993). 271 In response to the Supreme Court?s opinion, the Arizona legislature did not change the substance of its death penalty statute. Rather, it merely changed who decides the necessary aggravating and mitigating factors. See ARIZ. REV. STAT. ANN. ?? 13-703, 13-703.01(S)(1) (West Supp. 2003). 272 See Summerlin v. Stewart, 341 F.3d 1082, 1110-16, 1129-31 (9th Cir. 2003). 273 See supra Part IV.B. 274 See id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 35 5-MAY-05 11:55 Death Anyways 1423 In addition, the Ninth Circuit?s finding that Ring was a water- shed rule appeared to lower the retroactivity bar. On first glance, it may be appealing to believe a new rule affecting prison- ers with death sentences is always a watershed rule. However, in cases such as Simmons and Caldwell, that has not been the case. Rather, even in capital cases the rule must be truly watershed. In Summerlin, the court argued that Ring is watershed because it is sweeping, in that it affects every capital trial in the country.275 But as the Summerlin majority noted, Ring affected less than one-fourth of the states? capital sentencing structures.276 The Ninth Circuit also argued that Ring was sweeping because it af- fected every state that desires to sentence defendants to death, by establishing minimal structural requirements.277 The problem with that claim is that every new constitutional rule is creating some kind of minimal requirement that in some way will affect all states. The Ninth Circuit?s reasoning for why Ring is ?sweep- ing? does not provide any limits for any other new rule of crimi- nal procedure, thereby lowering the bar for when a rule is retroactively applied. In short, it is difficult to see how Ring, a case that essentially stated ?who decides? in capital sentencing, is a watershed ruling on par with decisions like Gideon. It does not affect every crimi- nal trial throughout the country and it only changed the proce- dure in less than half of the states that use capital sentencing. Also, judges are traditionally not seen as so suspect that their sentencing decisions render the quality of the sentence untrust- worthy. Given the cases that the Supreme Court has held do not apply retroactively, holding that Ring is a watershed ruling would have significantly lowered the bar for retroactivity on col- lateral review. Despite these reasons for not finding Ring to be watershed, the Supreme Court, as noted above, held only that the Ring rule did not seriously diminish the accuracy of Summerlin?s death sen- tence. It did not directly discuss the issue of whether Ring is a ?fundamental? rule that is ?implicit in the concept of ordered liberty.?278 But the Court?s focused discussion on the ?accuracy? 275 Summerlin, 341 F.3d at 1119. 276 Id. at 1120. 277 Id. 278 The dissent argued that the majority?s silence on the ?implicit in the concept of ordered liberty? issue is an affirmation that Ring is such a rule. Schriro v. Summer- lin, 124 S. Ct. 2519, 2527 (2004) (Breyer, J., dissenting) (stating ?[t]he majority does \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 36 5-MAY-05 11:55 1424 OREGON LAW REVIEW [Vol. 83, 2004] issue under Teague?s procedural exception appears to have main- tained the limited nature of this exception. Perhaps most obvi- ously, it reaffirmed the ?seriously? language of the ?accuracy? element of the ?watershed? exception to Teague. Upon close reading of the Summerlin majority?s opinion, it seems that the Ninth Circuit used the language that the finding must be that the new rule ?seriously? decreases the accuracy of the underlying sentence.279 However, when the Ninth Circuit majority analyzed the issue, the requirement of not only finding diminished accu- racy under Ring, but seriously diminished accuracy, is nowhere to be found.280 As a consequence of this omission, the Ninth Cir- cuit was able to find that Ring enhanced accuracy merely by cit- ing a few controversial empirical studies favorable to its conclusion.281 Although the Supreme Court?s Schriro opinion apparently only cited back to language that has been part of the Teague analysis since the beginning,282 it maintained the extreme nar- rowness of the Teague procedural exception by increasing the significance of a new rule to be one that really puts into doubt the habeas petitioner?s conviction or sentence. Despite the fact that Schriro did not add substance to what exactly is a rule that seriously decreases the accuracy of the conviction, it put a halt to courts using policy preferences to decide whether something is more accurate or not and hence eligible for retroactive applica- tion. Accordingly, this holding in Schriro will at least preserve not deny that Ring meets the first criterion, that its holding is ?implicit in the concept of ordered liberty??). Supporting the dissent?s contention is a string citation of a number of cases noting that the right to a jury trial is a ?fundamental right,? with a majority of the cited opinions being authored by the Schriro majority?s writer, Jus- tice Scalia. Id. Potentially providing further support for such an assertion was the majority opinion?s last paragraph, stating ?[t]he right to jury trial is fundamental to our system of criminal procedure, and States are bound to enforce the Sixth Amend- ment?s guarantees as we interpret them.? Id. at 2526 (emphasis added). Neverthe- less, because the majority spoke neither here nor there to the issue, it is not clear whether a majority of the Supreme Court would have found Ring announced a rule ?implicit in the concept of ordered liberty.? 279 Summerlin, 341 F.3d at 1109. 280 Id. at 1109-16. 281 Id. These studies are controversial because the dissenting opinion in Summer- lin was able to cite the same number of studies which came to the opposite conclu- sion of those cited by the Summerlin majority. Id. at 1129-31 (Rawlinson, J., dissenting). 282 The phrase ?seriously diminished,? regarding the accuracy of the underlying conviction in light of the new rule, appeared in Teague itself. Teague v. Lane, 489 U.S. 288, 313 (1989) (plurality). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 37 5-MAY-05 11:55 Death Anyways 1425 this Teague exception for those truly watershed rules which throw significant doubt into prisoners? convictions and sentences. Also significant is the Court?s rejection of a balancing ap- proach for new rules of criminal procedure.283 Not only did the Court implicitly reaffirm the entire Teague structure, thereby fur- ther entrenching it in American criminal law, it rejected an ap- proach that is attractive in cases like Summerlin?s. As noted above, Summerlin pushed for the rule the Court had explicitly rejected but eventually embraced. But as a consequence of the Court getting the issue wrong in Walton, Summerlin was sen- tenced to death by a drug-addled judge who may have confused Summerlin?s case with somebody else?s. A balancing test, achieved by looking at the policies implicit in the writ of habeas corpus and non-retroactivity, at least on its face, could possibly work justice in a highly questionable case like Summerlin?s. On the other hand, such an approach would provide tremendous amounts of uncertainty, and issues of retroactivity of criminal rules of procedure can potentially affect unthinkable numbers of convictions. Nevertheless, the Court?s statement that it would re- ject a balancing approach indicates that future retroactivity ques- tions must be addressed within the Teague framework. C. Schriro?s Effect on Determinate Sentencing During the same term the Supreme Court handed down its rul- ing in Schriro, the Court issued another significant ruling of crim- inal procedure spawned from Apprendi.284 In Blakely v. Washington,285 the Court threw determinate sentencing struc- tures at both the state and federal level into doubt. Because the Blakely holding will affect hundreds of thousands of prisoners throughout the country,286 the stakes are extremely high in the retroactivity litigation of this recent decision. This section will briefly describe the Blakely opinion and will then discuss the ef- fects of Schriro on the impending retroactivity controversy over whether Blakely applies retroactively. 283 See supra Part V.B. 284 In fact, the Court decided both Blakely v. Washington and Schriro on exactly the same day. Compare Blakely v. Washington, 124 S. Ct. 2531 (2004), with Schriro v. Summerlin, 124 S. Ct. 2519 (2004). 285 124 S. Ct. at 2531. 286 One article noted that Blakely affects ?as many as 270,000 federal cases alone.? Benjamin Wittes, Suspended Sentencing, THE ATLANTIC MONTHLY, Oct. 2004, at 50. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 38 5-MAY-05 11:55 1426 OREGON LAW REVIEW [Vol. 83, 2004] In Blakely, the Court held that judges could not upwardly de- part from presumed sentencing ranges under a determinate sen- tencing scheme.287 The facts of Blakely make clear the court?s holding. In Blakely, the defendant kidnapped his wife, put her in a box in the back of his pickup truck and carried her across three western states.288 The defendant pleaded guilty to second-degree kidnapping, which carried a maximum sentence of ten years under Washington State law.289 However, under the state?s sen- tencing guidelines, the defendant was eligible for a presumptive sentence of forty-nine to fifty-three months.290 Finding one of the state?s aggravating factors, the trial judge instead imposed a ninety-month sentence.291 The Court found that this upward departure violated the de- fendant?s Sixth and Fourteenth Amendment rights.292 The ma- jority opinion began by noting that its decision would require an application of its Apprendi decision.293 Rejecting the state?s ar- gument that the maximum sentence a judge may impose for pur- poses of the Sixth Amendment is the absolute maximum sentence allowed under state statute, including upward depar- tures, the Court explained that ?for Apprendi purposes . . . the maximum sentence a judge may impose? is limited solely to ?the facts reflected in the jury verdict or admitted by the defen- dant.?294 Accordingly, the maximum sentence a judge may im- pose is not the maximum sentence allowed under state law, but the maximum sentence allowed without requiring the judge to find additional facts not included in the underlying crime.295 Ap- plying this reasoning to the facts of the case, the Court found the judge imposed an unconstitutional sentence because he relied on factual findings not admitted in the defendant?s guilty plea, and hence found facts beyond what was required.296 The Court fur- ther noted that under Washington state law, had the judge not considered facts beyond the guilty plea, his upwardly-departed 287 Blakely, 124 S. Ct. at 2537-38. 288 Id. at 2534. 289 Id. at 2534-35. 290 Id. at 2535. 291 Id. 292 Id. at 2538. 293 Id. at 2536. 294 Id. at 2537 (emphasis omitted). 295 Id. at 2537-38. 296 Id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 39 5-MAY-05 11:55 Death Anyways 1427 sentence would have been reversed.297 The Court?s Schriro opinion will likely have a great impact on the retroactivity analysis of Blakely, specifically on the point of whether Blakely will fit either of the two Teague exceptions.298 The retroactivity issues for Blakely should be quite similar to 297 Id. at 2538. 298 Of course, this entire discussion of whether Blakely fits either of the two Teague exceptions to the bar on retroactive application of new rules of criminal procedure assumes that the Court will find Blakely to be a ?new? rule. The Court has announced that a rule is new if it was not ?dictated by then existing precedent? and ?was apparent to all reasonable jurists.? Lambrix v. Singletary, 520 U.S. 518, 528 (1997). Whether or not Blakely is a new rule is an issue because as the Court noted in Blakely, it was merely applying the rule announced in Apprendi and was not announcing an entirely new line of legal doctrine. Blakely, 124 S. Ct. at 2536. Because of this, it could be argued that Blakely did not announce a new legal rule and that petitioners challenging their sentences on collateral review should not have to show that Blakely fits one of the two narrow Teague exceptions. Instead, they need only show that their convictions were final after the Court announced Apprendi. However, two factors counter strongly against such an argument. First, the Court has reasoned that a strong dissent is relevant evidence indicating that reasonable jurists could have differed as to whether prior law compelled the new rule. Beard v. Banks, 124 S. Ct. 2504, 2512-13 (2004). In Blakely there was a large four-justice group of dissenters. See Blakely, 124 S. Ct. at 2543. Although Justice O?Connor?s dissent criticized part of the reasoning in Apprendi, it noted that Apprendi did not command the outcome in Blakely. Id. at 2547-48 (interpreting Apprendi?s ?statu- tory maximum? as the maximum allowed under the statute rather than the maxi- mum presumed sentence). Accordingly, the Blakely dissent based its reasoning on an interpretation of Apprendi that was contrary to that of the majority, rather than only suggesting that Apprendi be overturned. Second, whether lower courts split over whether to apply the then-existing precedent in a manner consistent with the new rule was held to be a critical issue by the Court in Butler v. McKellar, 494 U.S. 407, 415 (1990). As Justice O?Connor noted in her Blakely dissent, ?only one court had ever applied Apprendi to invalidate application of a guidelines scheme.? Blakely, 124 S. Ct. at 2547 n.1 (citing sixteen federal and state opinions that found Apprendi did not strike down guidelines schemes). Every single federal circuit court considering this issue rejected that Apprendi applied to determinate sentenc- ing schemes the way that Blakely applied. See United States v. Toliver, 351 F.3d 423 (9th Cir. 2003); United States v. Heltn, 349 F.3d 295 (6th Cir. 2003); United States v. Johnson, 335 F.3d 589 (7th Cir. 2003) (per curiam); United States v. Goodine, 326 F.3d 26 (1st Cir. 2003); United States v. Piggie, 316 F.3d 789 (8th Cir. 2003); United States v. Luciano, 311 F.3d 146 (2nd Cir. 2002); United States v. Randle, 304 F.3d 373 (5th Cir. 2002); United States v. Mendez-Zamora, 296 F.3d 1013 (10th Cir. 2002); United States v. DeSumma, 272 F.3d 176 (3rd Cir. 2001); United States v. Sanchez, 269 F.3d 1250 (11th Cir. 2001); United States v. Fields, 251 F.3d 1041 (D.C. Cir. 2001)); United States v. Kinter, 235 F.3d 192 (4th Cir. 2000). Such a weight of opinions contrary to Blakely provides significant evidence that Apprendi did not dictate such a result and that the issue was open for reasonable jurists to disagree. Not surprisingly, two courts considering this issue have found that Blakely is a ?new? rule. Morris v. United States, 333 F. Supp. 2d 759 (C.D. Ill. 2004); Lilly v. United States, 342 F. Supp. 2d 532 (W.D. Va. 2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 40 5-MAY-05 11:55 1428 OREGON LAW REVIEW [Vol. 83, 2004] those of the Ring holding, which were analyzed in Schriro, be- cause both new Supreme Court rules come from the Court?s gen- eral holding in Apprendi and have to do with the Sixth Amendment guarantee of a jury trial in sentencing proceedings. Critical distinctions exist, however, which require a complete analysis of what exactly will be the impact of the Court?s holding in Schriro to the impending retroactivity analysis of Blakely. Regarding the first exception, whether Blakely is a substantive ruling, Schriro appears to have ruled out any possibility that Blakely announced a new substantive rule of criminal law. In order to make an argument that Blakely announced a new rule, a similar argument to the one announced by the Ninth Circuit and overruled by the Supreme Court in Schriro would have to be presented.299 With the Court?s clear pronouncement that the substantive versus procedural distinction will focus on whether the new rule regulates the ?manner of determining the [criminal] defendant?s culpability,?300 and because Blakely does not appear to ?alter[ ] the range of conduct or the class of persons that the law punishes,?301 it will likely not be found to be a new substan- tive ruling under the first Teague exception.302 Schriro could also significantly impact a retroactivity analysis of the second Teague exception to the retroactivity bar for new rules of criminal procedure. With the Court re-emphasizing that a new rule must seriously diminish the accuracy of the underlying conviction, greater emphasis will likely be placed on this ?accu- racy? issue. Interestingly, regarding this issue, there appears to be a critical distinction between Blakely and Schriro, opening the 299 Specifically, the argument would be that, because the aggravating factors nec- essary for an upward departure must be found by a jury and are the functional equivalent of criminal elements, Blakely reordered the structure of the underlying crime into two separate crimes. In Blakely?s case, it was assault for the first crime and ?aggravated? assault, leading to a higher than presumed sentence as the second crime. As noted above, that is exactly the argument that was rejected by the Court in Schriro. See supra Part V.A. 300 Schriro v. Summerlin, 124 S. Ct. 2519 (2004). 301 Id. at 2523. 302 Not surprisingly, a number of district courts have already held as much, specif- ically relying on Schriro?s new formulation for determining whether a new rule is procedural or substantive, to find that Blakely is not a new substantive rule. Lilly, 342 F. Supp. 2d at 537; Rosario-Dominguez v. United States, 2004 U.S. Dist. LEXIS 15995, *26 n.3 (S.D.N.Y. Aug. 16, 2004); United States v. Stoltz, 325 F. Supp. 2d 982, 987 (D. Minn. 2004); United States v. Lowe, 2004 U.S. Dist. LEXIS 15455, *7-8 (W.D. Ill. Aug. 5, 2004); Morris v. United States, 333 F. Supp. 2d 759, 771 n.8 (C.D. Ill. 2004). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 41 5-MAY-05 11:55 Death Anyways 1429 possibility that the Court could find the accuracy of old up- wardly-departed sentences seriously diminished, leading to the possible holding that Blakely must be applied retroactively. Spe- cifically, in Schriro, the burden of proof necessary to find the aggravating factors to impose a death sentence was the same both before and after the Court?s decision in Ring. However, after Blakely the burden of proof will likely be higher for sentences imposed before the Court?s decision. In Schriro, the burden was on the state to prove the aggravating factors to a judge beyond a reasonable doubt,303 which is the same standard of proof necessary for a criminal element that must be found by a jury under the Sixth Amendment.304 Accordingly, the only dif- ference between a person sentenced to death pre-Ring and post- Ring was the fact-finder. This is not the case under Blakely. In Oregon, for example, the rule for imposing an upward departure before Blakely was that the judge must find ?substantial and compelling reasons to upwardly depart, based on a list of aggra- vating factors to consider.305 Post-Blakely, because the Court treats sentencing factors subject to the Sixth Amendment like el- ements, and therefore subject to the beyond a reasonable doubt standard,306 for an upward departure to be imposed a jury would have to find the same enumerated factors beyond a reasonable doubt.307 Although on its face the accuracy issue regarding retroactivity of Blakely appears to be quite similar to Ring, there is a poten- tially critical distinction. Quite unlike Ring, pre-Blakely criminal defendants may have been subject to a lower burden of proof than defendants after Blakely, which could logically lead to a finding that the accuracy of the old sentences is seriously dimin- ished. However, merely having the burden of proof elevated from something like a ?preponderance of the evidence? standard to ?beyond a reasonable doubt? may not be enough for the Court to find the accuracy of the underlying conviction is seri- ously diminished. Blakely is extremely similar to Ring in that the large change from the rule is a different factfinder (judge to 303 Schriro v. Summerlin, 124 S. Ct. 2519, 2521 (citing ARIZ. REV. STAT. ANN. ? 13-703 (West 1978), as amended by Act of May 1, 1979, Ariz. Sess. Laws ch. 144). 304 See In re Winship, 397 U.S. 358, 361-64 (1970). 305 OR. ADMIN. R. 213-008-0001, 213-008-0002 (2003). 306 Blakely v. Washington, 124 S. Ct. 2531, 2536-38 (2004); Apprendi v. New Jersey, 530 U.S. 466, 490 (2000). 307 Blakely, 124 S. Ct. at 2536-38. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 42 5-MAY-05 11:55 1430 OREGON LAW REVIEW [Vol. 83, 2004] jury). The Court?s holding in Schriro that the effect on accuracy from such a change in factfinder is too equivocal to find that the accuracy is seriously diminished may undercut any argument that judges are less accurate factfinders than juries and will, therefore, force the argument to focus almost entirely on the adjusted bur- den issue. Further, the Court?s focusing in Schriro on the fact that for retroactive application there must be a finding of a seri- ous diminishing of accuracy means a slight shift in the burdens may not be enough.308 This illustrates an issue that was not en- tirely addressed by the Court in Schriro?how much of a loss of accuracy must there be before the accuracy of the underlying proceeding is seriously diminished? Schriro did not truly address this issue because it focused on the fact that the evidence is equivocal about whether judges or juries are more accurate fact- finders.309 Unfortunately, the facts in Schriro did not present a situation where the accuracy of the underlying conviction had un- equivocally been lessened, so the Court had no occasion to illus- trate what it meant by seriously diminished accuracy. Whether Ring fits the more general piece of the second Teague exception, the requirement that the new rule of criminal proce- dure be ?implicit in the concept of ordered liberty?310 was not specifically addressed by Schriro.311 However, Justice Breyer?s dissent claims that the majority?s silence on this point indicates an implicit finding that the Court?s new rule from Ring is ?im- plicit in the concept of ordered liberty.?312 If this assertion is true, it is possible that the Court could also find Blakely is ?im- plicit in the concept of ordered liberty.? As noted above, both Ring and Blakely derive from the same holding in Apprendi, which is a protection of individuals? right to a have a jury deter- 308 Recently, in Woodroffe v. Lambert, 2004 U.S. Dist. LEXIS 19457, *40-42 (D. Or. Sept. 23, 2004), a district court in Oregon embraced this argument. In that case, the court held that although the petitioner?s enhanced sentence under pre-Apprendi rules was imposed under a lower standard of proof than after Apprendi, the accu- racy of the conviction or the enhanced sentence was still not seriously diminished. In Woodroffe, the court examined whether Schriro cast doubt on the Ninth Circuit?s determination in United States v. Sanchez-Cervantes, 282 F.3d 664 (9th Cir. 2002) (holding that Apprendi did not fit the second Teague exception). Nevertheless, the court?s discussion of changing burdens of proof for a Teague analysis should apply to Blakely. 309 Schriro v. Summerlin, 124 S. Ct. 2519, 2525 (2004). 310 Teague v. Lane, 489 U.S. 288, 311 (1989) (plurality). 311 Schriro, 124 S. Ct. 2519, 2527 (2004) (Breyer, J., dissenting). 312 Id. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 43 5-MAY-05 11:55 Death Anyways 1431 mine truly factual issues in a criminal proceeding.313 This basic principle appears to be indistinguishable in Ring and Blakely. If a majority of the Schriro Court believed that Ring dealt with a right that is ?implicit in the concept of ordered liberty,? and it found that the accuracy of the underlying proceeding was seri- ously diminished, as described above, then it is possible the Court could find that Blakely applies retroactively under the sec- ond Teague exception. There are at least four reasons why such a finding is un- likely.314 First, it is a large assumption that because Schriro did not address the issue of whether Ring announced a rule that is ?implicit in the concept of ordered liberty,? the Court did not dispute the issue.315 Justice Breyer?s dissent supports this asser- tion by citing to a number of opinions that either directly or indi- rectly state that the Sixth Amendment jury trial guarantee is a fundamental right.316 But other opinions casually noting that the Sixth Amendment jury trial right is fundamental do not necessa- rily equate the rule to being ?implicit in the concept of ordered liberty? for Teague retroactivity purposes. As discussed above, for the Court to find that a new rule of criminal procedure ap- plies retroactively, it must be watershed and comparable in im- portance to Gideon v. Wainwright.317 Therefore, ?implicit in the 313 See Blakely v. Washington, 124 S. Ct. 2531, 2536-37 (2004). 314 In fact, a number of courts have already held that Blakely does not apply ret- roactively. See, e.g., United States v. Swinton, 333 F.3d 481, 485 (3d Cir. 2003); Sepulveda v. United States, 330 F.3d 55, 61 (1st Cir. 2003); Coleman v. United States, 329 F.3d 77, 90 (2d Cir. 2003); United States v. Brown, 305 F.3d 304, 310 (5th Cir. 2002); Goode v. United States, 305 F.3d 378, 382 (6th Cir. 2002); Curtis v. United States, 294 F.3d 841, 844 (7th Cir. 2002); United States v. Mora, 293 F.3d 1213, 1219 (10th Cir. 2002); McCoy v. United States, 266 F.3d 1245, 1256-57 (11th Cir. 2001); United States v. Moss, 252 F.3d 993, 997 (8th Cir. 2001); United States v. Sanders, 247 F.3d 139, 151 (4th Cir. 2001); Jones v. Smith, 231 F.3d 1227, 1236 (9th Cir. 2000); In re Smith, 285 F.3d 6, 9 (D.C. Cir. 2002). 315 The following discussion points out the logical reasons to believe Justice Breyer made a large assumption. When it comes to counting Supreme Court jus- tices, it may not be such a ?large? assumption. It is true, of course, that four other justices signed off on Justice Breyer?s dissenting opinion, implying that at least four justices believe that the Sixth Amendment guarantee underlying Ring and Blakely is a fundamental one, implicit in the concept of ordered liberty. This means that only one justice from the Schriro majority must believe that the new rule from Blakely is implicit in the concept of ordered liberty. 316 Schriro, 124 S. Ct. at 2527 (citing Apprendi v. New Jersey, 530 U.S. 466, 499 (2000) (Scalia, J., concurring)); Blakely, 124 S. Ct. at 2536; Ring v. Arizona, 536 U.S. 584, 610 (2002) (Scalia, J., concurring)); Teague v. Lane, 489 U.S. 288, 313 (1989) (plurality); Duncan v. Louisiana, 391 U.S. 145, 157-58 (1968)). 317 See supra note 83 and accompanying text. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 44 5-MAY-05 11:55 1432 OREGON LAW REVIEW [Vol. 83, 2004] concept of ordered liberty? in the habeas corpus retroactivity context does not mean that a right is generally fundamental. Rather, it seems that the new rule must be extremely rare and special in some way.318 This perspective on the type of rule the Court would apply retroactively seems to be substantiated by the Court?s discussion in Beard v. Banks of the rarity of rules it will find apply retroactively, where the Court noted it has not yet found a rule that fits the second Teague exception.319 Second, and similar to the first problem, Blakely did not an- nounce a rule that is entirely unprecedented and new to criminal procedural jurisprudence.320 In Blakely the Court noted that it was merely applying its holding from Apprendi to the context of a determinate sentencing scheme.321 The Court was quite clear that it was not announcing a previously unheard-of legal doc- trine. Instead, Blakely was another logical, albeit surprising, ex- tension of the principle announced in Apprendi. Although the Court has never announced that an extension (or new applica- tion) of a previously established legal rule cannot be one that is ?implicit in the concept of ordered liberty? which rises to the level of watershed, it is difficult to conceive of the Court describ- ing the second Teague exception as only applying to a ?small core of rules?322 but then finding that an extension of a previ- ously announced concept is such a rule. In other words, because Blakely is a mere variant of Apprendi, it seems unlikely that it could be found to be a watershed ruling of criminal procedure. Third, Blakely will likely be subject to harmless-error analysis, a critical consideration in whether a new rule is one ?implicit in the concept of ordered liberty.? The Ninth Circuit recently ex- plained how Sixth Amendment jury rights being subject to harm- less-error analysis is weighty evidence that such rules do not fit the second Teague exception: Our decisions that subjected Apprendi claims to harmless er- ror analysis or plain error review lend additional support to 318 See, e.g., Beard v. Banks, 124 S. Ct. 2504, 2513-14 (2004); O?Dell v. Nether- land, 521 U.S. 151, 157 (1997); Graham v. Collins, 506 U.S. 461, 478 (1993); Saffle v. Parks, 494 U.S. 484, 495 (1990). 319 Beard, 124 S. Ct. at 2513-14. Note that the Court?s formulation of the second Teague exception in Beard seems to be an especially arduous characterization of the second Teague exception. 320 For the double meaning of ?new? that arises in the Teague context see supra note 298. 321 Blakely v. Washington, 124 S. Ct. 2531, 2536-37 (2004). 322 O?Dell, 521 U.S. at 157. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 45 5-MAY-05 11:55 Death Anyways 1433 our determination that Apprendi is not a bedrock procedural rule. In these cases, we did not consider Apprendi errors to be structural. A structural error is one that necessarily renders a trial fundamentally unfair and therefore invalidates the con- viction. We only review for plain error or assess whether an error is harmless when the error is not structural; in those cir- cumstances, the court must determine whether any substantial rights were prejudiced by the error. By applying harmless er- ror analysis or plain error review to Apprendi claims, we have necessarily held that Apprendi errors do not render a trial fun- damentally unfair.323 As noted above, Blakely is a derivative of Apprendi, and it appears that a Blakely claim will be subject to harmless-error re- view because it is not a structural error that renders the entire trial or sentencing fundamentally unfair. In fact, one Ninth Cir- cuit panel recently conducted harmless-error review of a Blakely error, implying that Blakely, like Apprendi, is subject to such re- view, providing significant evidence that Blakely is not a water- shed rule.324 Fourth, there are pragmatic reasons to believe the Court will avoid finding that Blakely fits the second Teague exception. One commentator reported that Blakely would affect up to 270,000 federal cases.325 In addition, if the Court holds that Blakely ap- plies retroactively, federal habeas corpus statutory law allows both state and federal prisoners to file successive petitions.326 This means that prisoners who have already challenged their con- victions and sentences in federal court and lost will be allowed to bring an entirely new case to have their unconstitutional 323 United States v. Sanchez-Cervantes, 282 F.3d 664, 670-71 (2002). 324 See United States v. Ameline, 376 F.3d 967, 972 (9th Cir. 2004) (holding Blakely error requires reversal and resentencing under either plain-error or harm- less-error standard). In Tyler v. Cain, 533 U.S. 656, 666 (2002), the Supreme Court explained that not even all structural-error rules fit into Teague?s procedural excep- tion. This implies that the second Teague exception is even narrower than the cate- gory of structural-error rules. See also United States v. Sanders, 247 F.3d 139, 150 (4th Cir. 2001) (noting that the fact the claim is subject to harmless and plain error indicates it is not a watershed change in criminal procedure and emphasizing that ?finding something to be a structural error would seem to be a necessary predicate for a new rule to apply retroactively under Teague?). 325 Wittes, supra note 286. 326 See 28 U.S.C. ? 2244(b)(2)(a) (2000), which states A claim presented in a second or successive habeas corpus application under section 2254 that was not presented in a prior application shall be dismissed unless . . . the applicant shows that the claim relies on a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable. \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 46 5-MAY-05 11:55 1434 OREGON LAW REVIEW [Vol. 83, 2004] sentences overturned. Such a scenario would mean that sentences imposed as far back as the early 1980s, when the fed- eral and state governments began implementing determinate sen- tencing structures,327 would be subject to review. This could lead to a massive flood of convicted criminals being let out in one large wave. Because of the potential political and economic ef- fects associated with a finding, it seems unlikely the Court would issue such a decision. CONCLUSION The entire case of Warren Wesley Summerlin was unfortunate, beginning with the crime itself and concluding with the Supreme Court?s overturning itself on an issue Summerlin had argued fif- teen years earlier. The issue presented to the Court, however, was not a factual or policy issue about whether Mr. Summerlin deserved his death sentence. Rather, the issue presented in this case was purely a legal one. More particularly, it was about what the Supreme Court did in Ring v. Arizona.328 In determining whether Ring applied retroactively, the Ninth Circuit applied the well-accepted Teague test. This Teague test has placed a very high hurdle in front of federal habeas corpus petitioners to have a new rule of constitutional criminal proce- dure applied retroactively. Surprisingly, in applying the Teague test the Ninth Circuit held that Ring was a new substantive rul- ing, thus applying retroactively. Alternatively, the court held that if the new rule announced in Ring is seen as procedural, then it was a new watershed rule that improved the accuracy of the underlying proceeding. In so holding, the Ninth Circuit ad- ded to confusion regarding when a new rule is substantive, and it significantly lowered the bar for when a procedural rule is ap- plied retroactively. From a legal standpoint, the Supreme Court fortunately granted certoriari and overturned the Ninth Circuit. In so doing, the Court cleaned up the framework for Teague issues and pro- vided significant clarification on the distinction between substan- tive and procedural rules of criminal law. The Court?s distinction is clear, workable, and perhaps most importantly, follows the spirit and underlying policy rationales for the Teague doctrine of 327 See Apprendi v. New Jersey, 530 U.S. 466, 549-50 (2000) (O?Connor, J., dissenting). 328 536 U.S. 584 (2002). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 47 5-MAY-05 11:55 Death Anyways 1435 retroactivity. The Court also provided clarification for the proce- dural exception to the Teague bar of retroactivity. It held that new rules must not only diminish the accuracy of the underlying proceeding, but new rules must seriously do so to be applied ret- roactively. This holding not only stayed within the bounds of al- ready-existing Teague jurisprudence, but also maintained the narrowness of the Teague procedural exception. Accordingly, the Schriro Court took great strides towards protecting finalized convictions. Its opinion also reserves the finding of watershed rules for the extremely rare cases that are truly ?ground- breaking.?329 This holding will prove important as federal courts will now have to deal with whether to apply Blakely, the Supreme Court?s most recent new rule of criminal procedure, retroactively. Schriro will make it very difficult to find that the rule from Blakely announced a new substantive rule. However, Blakely will provide interesting accuracy issues that will likely force the Court to further define what constitutes a serious diminishing of the accuracy of the underlying proceeding. If it is found that the rule from Blakely seriously diminished the accuracy of prior sentences, the Court will also have to deal with the difficult issue of whether Blakely is ?implicit in the concept of ordered liberty.? Although the possibility is definitely there for the Court to an- nounce such a holding, several legal and pragmatic factors make that an unlikely result. Whatever the result, future retroactivity cases will address Blakely, will be heavily influenced by Schriro?and will further define the factors discussed in the Court?s opinion. 329 See O?Dell v. Netherland, 521 U.S. 151, 167 (1997). \\server05\productn\O\ORE\83-4\ORE407.txt unknown Seq: 48 5-MAY-05 11:55 1436 OREGON LAW REVIEW [Vol. 83, 2004]