State v. Davis

SUPREME COURT OF ARIZONA En Banc STATE OF ARIZONA, ) Arizona Supreme Court ) No. CR-01-0423-PR Appellee, ) ) Court of Appeals ) Division One v. ) No. 1 CA-CR 99-1028 ) ) Maricopa County ANTHONY CHARLES DAVIS, ) Superior Court ) No. CR 99-90084(B) Appellant. ) ) O P I N I O N __________________________________) Appeal from the Superior Court in Maricopa County The Honorable Dennis W. Dairman, Judge AFFIRMED IN PART; REVERSED IN PART; REMANDED Court of Appeals, Division One Memorandum Decision (filed Oct. 16, 2001) AFFIRMED IN PART; REVERSED IN PART JANET A. NAPOLITANO, ARIZONA ATTORNEY GENERAL Phoenix by Randall M. Howe, Chief Counsel, Criminal Appeals Section and Joseph T. Maziarz, Assistant Attorney General and Katia Mehu, Assistant Attorney General Attorneys for Appellee JAMES J. HAAS, MARICOPA COUNTY PUBLIC DEFENDER Phoenix by Anna M. Unterberger, Deputy Public Defender Attorneys for Appellant B E R C H, Justice ¶1 We granted review in this case to decide whether sentencing a twenty-year-old defendant to a mandatory minimum sentence of fifty-two years without the possibility of parole for having voluntary sex with two post-pubescent teenage girls is so grossly disproportionate to the crime as to violate the Eighth Amendment’s prohibition against cruel and unusual punishment. We hold that it is. FACTS1 ¶2 In January 1999, thirteen-year-old T.E. and her stepsister, C.M., snuck out of their house at night to meet nineteen-year-old Jason in a local park. Jason drove the girls to the home of Defendant Anthony Davis, where they met Davis and two other young men. The six young people socialized for some time. During the conversation, Davis told the girls that he was twenty, and T.E. said she was fourteen. Later that night, Davis and T.E. had sex. This was not T.E.’s first sexual encounter; she testified that she had first had sex when she was twelve years old, and she knew what Davis was doing. ¶3 T.E. and C.M. visited Davis at his house several more times that month, usually late at night. The second victim, P.T., accompanied the girls on January 20, 1999. During that visit, P.T. learned that Davis was twenty; she testified that “everyone knew” she was fourteen. Later that evening, Davis and 1 We review the facts in the light most favorable to sustaining the verdict. State v. Gallegos, 178 Ariz. 1, 9, 870 P.2d 1097, 1105 (1994). In stating the facts, we have used initials to protect the identity of the minor victims. 2 P.T. had sex. They also had sex on two other occasions during the following two weeks. ¶4 Still later in January, Davis, Jason, and another young man visited P.T. and C.M. while P.T. was babysitting. The homeowners returned early, found the men there, ordered them to leave, and contacted the girls’ parents. ¶5 After the babysitting incident, T.E. and C.M. ran away to Davis’s house. P.T. went with them, but returned home the same evening. The next morning, February 1, the police found T.E. and C.M. at Davis’s house with Davis and Jason. ¶6 When the police questioned her, T.E. denied having had sex with either Davis or Jason. The next day, however, T.E.’s mother called the police to tell them there had been sexual contact between T.E. and Davis. On February 3, P.T. and T.E. were examined by a doctor for signs of sexual abuse. The examining doctor found signs indicating that T.E. had had sexual intercourse within the week. In an interview with police on February 4, T.E. conceded that she had sex with Davis two or three times, the last time over the weekend of January 29-31. ¶7 The State charged Davis with four counts of sexual misconduct with a minor, in violation of Arizona Revised Statutes (“A.R.S.”) section 13-1405 (1997): count one with T.E. on January 18, 1999, count two with P.T. on January 20, 1999, count three with P.T. on January 25, 1999, and count four with 3 P.T. on January 29, 1999. Each act was alleged as a dangerous crime against children. See A.R.S. § 13-604.01 (Supp. 1998). ¶8 Throughout the case, Davis denied ever having sex with T.E., but admitted having sex with P.T. on three occasions. He claimed that P.T. initially told him she was eighteen, but told him she was sixteen the day after the babysitting incident. Davis testified that he did not have sex with P.T. after learning that she was only sixteen. ¶9 The jury convicted Davis on all four counts. Upon discovering that the minimum sentence that Davis could receive was fifty-two years, all twelve jurors submitted a note to the trial judge stating their belief that “the punishment for the crime is excessive.” Two jurors submitted individual letters expressing their dismay and strong belief that the potential sentences for Davis were too harsh. The probation officer who prepared Davis’s pre-sentence report acknowledged that while Davis should be held accountable for his crimes, the mandatory sentence was not warranted. The pre-sentence report also noted that neither T.E.’s mother nor P.T.’s mother wished to see Davis sentenced to a long prison term. Even the prosecutor recommended a mitigated prison sentence and agreed that Davis should be allowed to petition the Board of Executive Clemency for a commutation of his sentence. ¶10 The trial judge apparently agreed because he stated at 4 sentencing that all of the charges were “legally non-dangerous and non-repetitive offenses notwithstanding the nomenclature set forth in the charges.” He also entered a special order allowing Davis to petition the Board of Executive Clemency for a commutation of sentence within ninety days of sentencing. See A.R.S. § 13-603(L) (Supp. 2002) (allowing such an order if the judge believes “that a sentence that the law requires the court to impose is clearly excessive”). Nonetheless, as required by statute, the trial judge sentenced Davis to fifty-two years in prison. DISCUSSION A. Proportionality of the Sentence to the Crimes Committed ¶11 Davis argues that given the circumstances of his offenses, the four flat, consecutive, thirteen-year sentences violate the prohibition against cruel and unusual punishment found in the Eighth Amendment to the United States Constitution and Article 2, Section 15 of the Arizona Constitution. He contends that the mandatory fifty-two-year sentence, without the possibility of parole, is so grossly disproportionate to his offenses as to be unconstitutional. ¶12 We asked the parties to brief whether Article 2, Section 15 of the Arizona Constitution provides greater protection against cruel and unusual punishment than does the 5 Eighth Amendment to the United States Constitution. See, e.g., People v. Bullock, 485 N.W.2d 866, 872-74 (Mich. 1992) (finding Michigan’s “cruel or unusual” punishment provision broader than the United States Constitution’s corollary provision). Although we do not follow federal precedent blindly, after considering the issue we do not find in this case a compelling reason to interpret Arizona’s cruel and unusual punishment provision differently from the related provision in the federal constitution. See State v. Noble, 171 Ariz. 171, 173, 829 P.2d 1217, 1219 (1992). 1. Proportionality Review. ¶13 The Eighth Amendment to the United States Constitution and its corollary, Article 2, Section 15 of the Arizona Constitution, prohibit punishments that are cruel and unusual. While originally reserved for review of corporal punishments, the Eighth Amendment has been applied to lengthy sentences of incarceration. Lockyer v. Andrade, ___ U.S. ___, ___, 123 S. Ct. 1166, 1173 (2003). We must decide whether Davis’s fifty- two-year sentence is so excessively long as to be cruel and unusual.2 2 In section B of this opinion, we reverse one conviction. Because Davis was sentenced to fifty-two years and may be retried on the charge, we analyze gross disproportionality based on the sentence imposed. See infra ¶¶ 50-66. 6 ¶14 More than a decade ago, this court wrestled with the problem of a very long sentence in a case exhibiting facts remarkably similar to those at issue before us. See State v. Bartlett, 164 Ariz. 229, 792 P.2d 692 (1990) (Bartlett I). In Bartlett I, a twenty-three-year-old defendant was convicted of sexual conduct with two fourteen-year-old girls. Id. at 231, 792 P.2d at 694. Because both victims were younger than fifteen, the charges were considered dangerous crimes against children, although the girls considered themselves Bartlett’s “girlfriends.” Id. at 233, 792 P.2d at 696. Under the mandatory sentencing provisions of the Dangerous Crimes Against Children Act, Bartlett received the minimum sentence of forty years in prison without the possibility of early release. Id. As does Davis, Bartlett argued that his sentence was so disproportionate to his crimes that it violated the constitutional prohibition against cruel and unusual punishment. Id. Because the facts and circumstances of Bartlett I are so similar to those in the case at bar, we set forth in some detail our reasoning in that case. ¶15 In Bartlett I, we analyzed the constitutionality of the defendant’s sentence under the prevailing test at the time, which had been enunciated by the United States Supreme Court in Solem v. Helm, 463 U.S. 277, 292, 103 S. Ct. 3001, 3011 (1983). Solem set forth a three-part test to determine whether a 7 sentence was disproportionate to the crime and therefore violated the Eighth Amendment’s prohibition against cruel and unusual punishment. Id. The test required examination of the following factors: (a) the severity of the penalty as compared to the seriousness of the offense, (b) the jurisdiction’s penalties for crimes that are more serious than the offense at issue (the intra-jurisdictional analysis), and (c) the sentences other jurisdictions impose for the same crime (the inter- jurisdictional analysis). Id. ¶16 In comparing the penalty with the offense, this court first examined the nature of the crime, including the type of harm threatened or inflicted, and the defendant’s culpability. Bartlett I, 164 Ariz. at 234, 792 P.2d at 697. The court found several factors important: (a) the absence of either the threat or the commission of violence to induce the victims to engage in sex; (b) the victims’ willing participation in the acts; (c) Bartlett’s lack of a criminal record, including any crime against children; (d) his immaturity; (e) the sociological fact that “sexual conduct among post-pubescent teenagers is not uncommon”; and (f) the broad scope of the governing statute. Id. at 234-36, 792 P.2d at 697-99. While observing that the severe mandatory punishment required by the Dangerous Crimes Against Children Act may be justified for other, more heinous crimes, the court concluded that it was “not justified under the 8 specific circumstances of this case.” Id. at 236, 792 P.2d at 699. ¶17 Application of the second and third prongs of the Solem test, the intra- and inter-jurisdictional analyses, provided further support for that conclusion. ¶18 The intra-jurisdictional analysis showed that in Arizona, many more serious offenses qualified for equal or lesser punishments than the punishment Bartlett received. Id. For example, one guilty of second degree murder or forcible sexual assault of a minor could receive the same sentence Bartlett received. Id. ¶19 For the inter-jurisdictional analysis, the court conducted a nationwide survey to determine how other jurisdictions punished defendants guilty of two counts of sexual conduct with minors. Id. at 238-40, 792 P.2d at 701-03. The survey showed that three years was the longest minimum sentence for the first offense, and in all but one state, ten years was the longest minimum sentence for the second offense. Id. at 238, 240, 792 P.2d at 701, 703. All sentences in all jurisdictions were far less than the forty years Bartlett was required to serve. Notably, the court did not find in any other jurisdiction “the particularly harsh combination of provisions present here, including both mandatory consecutive sentencing and nonavailability of parole.” Id. at 240, 792 P.2d at 703. 9 ¶20 In concluding its opinion, this court cautioned “that successful challenges to the proportionality of particular sentences are ‘exceedingly rare.’” Id. (quoting Solem, 463 U.S. at 289-90, 103 S. Ct. at 3009). But in a narrow holding, carefully restricted to the facts of the case, the court determined that Bartlett presented that “exceedingly rare” case in which the application of the provisions of the Dangerous Crimes Against Children Act rendered the sentence unconstitutionally long. Accordingly, we remanded Bartlett I for resentencing without the application of the dangerous crimes against children enhancement. Id. at 242, 792 P.2d at 705. ¶21 On review by certiorari, the United States Supreme Court vacated the opinion in Bartlett I and remanded the case for reconsideration in light of its decision in Harmelin v. Michigan, 501 U.S. 957, 111 S. Ct. 2680 (1991). Arizona v. Bartlett, 501 U.S. 1246, 111 S. Ct. 2880 (1991). ¶22 In Harmelin, a plurality of the Supreme Court recognized a “narrow proportionality” principle embodied in the Eighth Amendment that prohibits sentences that are “grossly disproportionate” to the crime. 501 U.S. at 997, 111 S. Ct. at 2702-03 (Kennedy, J., concurring, joined by O’Connor and Souter, JJ.). Justice Kennedy determined that the Solem three-part analysis remained useful, but a reviewing court should consider the second and third factors – that is, the intra- and inter- 10 jurisdictional analyses – only if “a threshold comparison of the crime committed and the sentence imposed leads to an inference of gross disproportionality.” Id. at 1005, 111 S. Ct. at 2707. ¶23 The five opinions in Harmelin left unclear the contours of the test to be applied on remand in Bartlett’s case. We concluded, however, that Justice Kennedy’s modified Solem analysis provided the proper standard by which to reconsider the sentence in Bartlett I. State v. Bartlett, 171 Ariz. 302, 305, 830 P.2d 823, 826 (1992) (Bartlett II). ¶24 The modified Solem analysis required the court to first examine whether Bartlett’s sentence was “grossly disproportionate to his crimes.” Id. at 306, 830 P.2d at 827. Using the same factors we considered in Bartlett I, we again concluded that a sentence of forty years without possibility of parole for consensual sex with post-pubescent teenagers reaches the threshold of gross disproportion, given the fact that the people, through their legislature, have adopted statutes under which the courts impose comparable punishment by imprisonment for crimes such as violent rape, second degree murder, and brutal assault of children. Id. at 309, 830 P.2d at 830. ¶25 We validated our initial impression of gross disproportionality by conducting intra- and inter-jurisdictional analyses, which showed that little change had occurred in the 11 two years that had passed since Bartlett I was decided. Id. at 310, 830 P.2d at 831. These analyses confirmed that Bartlett’s sentence was grossly disproportionate to his crimes, both because it was harsher than sentences imposed in Arizona for more severe crimes and because it was far harsher than the minimum sentence imposed for a similar offense in any other jurisdiction. Id. ¶26 Two justices dissented in Bartlett II, 171 Ariz. at 311-16, 830 P.2d at 832-37. Although the dissenting justices concurred in the majority’s assessment that Justice Kennedy’s plurality opinion provided the proper framework for a proportionality review, they disagreed as to what that analysis entailed. Id. They believed, as does our dissenting colleague in this case, that the court was prohibited from reviewing the specific circumstances of Bartlett’s crimes. Id. Rather, they reasoned, Harmelin permitted an analysis only of “the threat posed to the individual and to society by the commission of that crime.” Id. at 312, 830 P.2d at 833 (Corcoran, J., dissenting) (citing Harmelin, 501 U.S. at 1001-05, 111 S. Ct. at 2705-07). ¶27 Just four years later, the dissent’s position became the majority. In State v. DePiano, 187 Ariz. 27, 926 P.2d 494 (1996), a three-member majority held that disproportionality must be measured by the nature of the crime and not by the facts and circumstances of any particular defendant’s case. Id. at 12 30, 926 P.2d at 497. ¶28 In DePiano, the defendant attempted to commit suicide by carbon monoxide poisoning by running her car in her closed garage. Id. at 28-29, 926 P.2d at 495-96. She placed her two young sons in the car with her. Id. All survived, and DePiano was sentenced to two flat, consecutive, seventeen-year sentences for two counts of child abuse. Id. In affirming the convictions, this court rejected DePiano’s disproportionality claim “without looking at the particular circumstances surrounding her offense.” Id. at 30, 926 P.2d at 497. ¶29 As the divergent analyses in Bartlett II and DePiano show, this court has struggled to interpret the guidance from the Supreme Court. That Court has acknowledged that it has “not been a model of clarity” and has not “established a clear . . . path for courts to follow.” Lockyer, ___ U.S. at ___, 123 S. Ct. at 1173. Indeed, the Supreme Court has conceded that the only principle to emerge from its splintered Eighth Amendment jurisprudence is that a “gross disproportionality principle” does apply to non-capital cases. Id. Our task today is to decipher the Court’s jurisprudence and determine the standards that govern a proportionality review. The proper application of those standards should resolve the point of divergence between Bartlett II and DePiano: whether a reviewing court conducting a proportionality review may examine the specific facts of the 13 case or whether it may only view the crime generally. ¶30 We are aided in this task by the Supreme Court’s recent decisions in Lockyer and Ewing v. California, ___ U.S. ___, 123 S. Ct. 1179 (2003), companion cases in which the Court attempted to clarify the appropriate analysis for Eighth Amendment proportionality review. In each, the Court reviewed a lengthy sentence imposed under California’s “three-strikes” law. Lockyer, ___ U.S. at ___, 123 S. Ct. at 1170; Ewing, ___ U.S. at ___, 123 S. Ct. at 1182. Although the opinions in each case are sharply divided, a majority of the Court now appears to have adopted the analysis set forth in Justice Kennedy’s plurality opinion in Harmelin. See Ewing, ___ U.S. at ___, 123 S. Ct. at 1187 (stating that “[t]he proportionality principles in our cases distilled in Justice Kennedy’s concurrence [in Harmelin] guide our application of the Eighth Amendment” proportionality review); see also id. at ___, 123 S. Ct. at 1191-92 (Stevens, Souter, Ginsburg, and Breyer, JJ., dissenting, agreeing that a proportionality principle applies). ¶31 In conducting its analysis in each case, the Supreme Court reviewed the specific facts and circumstances of the offense that led to the imposition of the three-strike enhancement, as well as reviewing each defendant’s prior record. Lockyer, ___ U.S. at ___, 123 S. Ct. at 1169-70; Ewing, ___ U.S. 14 at ___, 123 S. Ct. at 1189-90.3 In Ewing, the specific facts of Ewing’s crime revealed more than simply “shoplifting”; by stealing three golf clubs worth nearly $1200, he had committed grand theft. ___ U.S. at ___, 123 S. Ct. at 1189. Compounding the seriousness of the acts, Ewing committed his crime while on probation and “after previously having been convicted of at least two ‘violent’ or ‘serious’ felonies.” Id. These individual circumstances, Justice O’Connor concluded, rendered Ewing’s sentence “justified by the State’s public-safety interest in incapacitating and deterring recidivist felons, and amply supported by [Ewing’s] own long, serious criminal record.” Id. at ___, 123 S. Ct. at 1190. Thus, in conducting its proportionality review, the Court examined the specific facts and circumstances of the defendant’s crime. ¶32 Other courts have also reviewed individual circumstances when conducting proportionality reviews. See Henderson v. Norris, 258 F.3d 706, 709, 712 (8th Cir. 2001) (reviewing the specific facts of the defendant’s case, “the harm caused or threatened to the victim or to society and the 3 Lockyer was before the Court on federal habeas corpus review. ___ U.S. at ___, 123 S. Ct. at 1171. Because the Court was examining whether the state had failed to apply “clearly established federal law,” the analysis in that case provides little guidance for our inquiry here, except insofar as the Court considered the defendant’s specific facts and circumstances. We therefore do not further discuss the Lockyer opinion. 15 culpability and degree of involvement of the defendant” in finding a sentence disproportionately long); Hawkins v. Hargett, 200 F.3d 1279, 1283 n.2 (10th Cir. 1999) (concluding that the defendant’s culpability is relevant in Eighth Amendment proportionality review); United States v. Harris, 154 F.3d 1082, 1084 (9th Cir. 1998) (considering, in upholding a ninety-five- year sentence, the facts of a bank robbery that was “undeniably violent[;] employees and bystanders were threatened and occasionally harmed”). These recent decisions demonstrate the utility and appropriateness of analyzing the specific facts and circumstances of the offenses when determining if a sentence is grossly disproportionate to the crime committed. ¶33 One other factor motivates us to review the specific circumstances of Davis’s case: The legislature permits this court to reduce lengthy sentences when “the punishment imposed is greater than under the circumstances of the case ought to be inflicted.” A.R.S. § 13-4037(B) (2001) (allowing imposition of “any legal sentence”). Although Davis’s sentence fell within the legal sentencing range, if we find a sentence excessive, A.R.S. § 13-4037(B) imposes on us the duty to review the circumstances of the case to determine whether the sentence imposed is in fact unwarranted. ¶34 Our opinion in DePiano, which prohibited such an individualized analysis, while an understandable attempt to 16 comply with the Supreme Court’s splintered Eighth Amendment jurisprudence of the day, now appears to be a rivulet diverting from the mainstream analysis, and we overrule it.4 Subsequent guidance from the Supreme Court suggests that, in assessing the constitutionality of a sentence, the reviewing court should examine the crime, and, if the sentence imposed is so severe that it appears grossly disproportionate to the offense, the court must carefully examine the facts of the case and the circumstances of the offender to see whether the sentence is cruel and unusual. 2. Proportionality Review of Davis’s Sentence. ¶35 Against this background, we analyze the sentence imposed on Davis. We must first determine whether an inference of gross disproportionality between Davis’s offenses and his sentence can be drawn. 4 We do not lightly overrule a decision that is barely eight years old. But stare decisis “is not ‘an imprisonment of reason,’” Guardians Ass’n v. Civil Serv. Comm’n, 463 U.S. 582, 618, 103 S. Ct. 3221, 3241 (1983) (Marshall, J., dissenting) (quoting United States v. Int’l Boxing Club of N.Y., 348 U.S. 236, 249, 75 S. Ct. 259, 266 (1955) (Frankfurter, J., dissenting)), or an “inexorable command,” Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 405, 52 S. Ct. 443, 446-47 (1932) (Brandeis, J., dissenting), and when later opinions of the Supreme Court show our constitutional interpretations to be incorrect, we must overrule them and bring our decisions into conformity with Supreme Court precedent. See U.S. Const. art. VI, cl. 2 (supremacy clause). 17 a. Gross Disproportionality. ¶36 Davis was sentenced to serve fifty-two years in prison for having non-coerced sex with two post-pubescent teenage girls. This strikes the court – as it did the jurors, the trial judge, the pre-sentence report writer, and the girls’ mothers – as an extraordinarily long sentence. Many of the factors deemed important in determining that Bartlett’s sentence was disproportionate to his crimes are also present here: (1) Davis’s sexual relations with the girls involved neither actual nor threatened violence; in each instance the girls knew what they were doing and willingly participated. Indeed, the victims sought Davis out; all acts occurred after the victims went voluntarily to Davis’s home. (2) Davis does not have an adult criminal record, nor has he committed any previous crimes against children. (3) Post-pubescent sexual conduct appears to be no less common today than it was in 1990.5 (4) There is evidence in the record that Davis’s intelligence and maturity 5 A 1997 study conducted by the Center for Disease Control and Prevention showed that nearly 17% of eighth graders had previously engaged in sexual conduct, and of those, 46% had done so with three or more partners. J.V. Fetro, K.K. Coyle, P. Pham, Health-Risk Behavior Among Middle School Students in a Large Majority-Minority School District, 71(1) Journal of School Health 30-37 (2001). Another study found that 25% of fifteen- year-old females reported having had sex at least once. Elizabeth Terry and Jennifer Manlove, Trends in Sexual Activity and Contraceptive Use Among Teens, Child Trends Research Brief (2000), available at http://www.childtrends.org/onlinecart/ product.cfm?id=628. 18 level fell far below that of a normal young adult. (5) Like Bartlett, Davis was caught in the very broad sweep of the governing statute, which makes any sexual conduct with a person younger than fifteen years old by a person older than eighteen years old a “dangerous crime against children,” whether the offense is a rape-incest by a step-parent who forces sex on a trusting ward or a pedophile who uncontrollably preys upon young children, see State v. Taylor, 160 Ariz. 415, 773 P.2d 974 (1989), or the more benign boyfriend-girlfriend situation in which one party is older than eighteen and the other younger than fifteen. ¶37 We recognize society’s strong interest in protecting children and understand and appreciate that it is the legislature’s province to assess the appropriate punishment for crimes against children. But we cannot say that all incidents of sexual conduct are of equal seriousness and pose the same threat to their victims or to society. The broad range of offenses encompassed by the statute under which Davis was charged, coupled with the legislature’s command in A.R.S. § 13- 4037(B) and the Supreme Court’s Eighth Amendment jurisprudence, impose on us the duty to apply the law to the specific facts of the cases that come before us to determine the constitutionality of sentences imposed. After conducting that review, we conclude that Davis’s conduct was swept up in the broad statutory terms, 19 which, in turn, triggered the mandatory sentences imposed. The trial judge, the jury, the pre-sentence report writer, and even the victims’ mothers all recognized the injustice of sentencing Davis to a fifty-two-year prison sentence with no possibility of early release for the crimes at issue in this case. We cannot ignore that injustice. While recognizing that many sex crimes against children may well justify such a sentence, others do not. We conclude that given the circumstances of Davis’s offenses, the sentence imposed in this case appears to be grossly disproportionate to his crimes. b. Intra-jurisdictional Comparison. ¶38 Once an inference of gross disproportionality has been found, the Supreme Court suggests that a reviewing court validate that impression by conducting an intra- and inter- jurisdictional analysis. 6 Harmelin, 501 U.S. at 1005, 111 S. Ct. at 2707 (Kennedy, J., concurring); Solem, 463 U.S. at 291-92, 103 S. Ct. at 3010. In the intra-jurisdictional analysis, we compare the sentences imposed in Arizona for crimes more serious than those committed by Davis to see whether those sentences validate our tentative conclusion of gross disproportionality. 6 We recognize that the intra- and inter-jurisdictional analysis is not required. See Dissent, ¶ 102. But we agree with the Supreme Court’s suggestion that such an inquiry might validate the court’s initial impression of gross disproportionality. 20 See Bartlett I, 164 Ariz. at 242, 792 P.2d at 705. ¶39 Little has changed since this court conducted a similar analysis in 1990 in Bartlett I. Id. at 236-37, 792 P.2d at 699-700. It continues to be the case that those guilty of more serious crimes, such as second degree murder, sexual assault, or continued sexual abuse of a minor under fifteen years of age receive the same presumptive sentence that Davis did. See A.R.S. § 13-604.01(C). It also remains true that dangerous crimes against children such as kidnapping, child abuse, aggravated assault, or commercial sexual exploitation of a child, all seemingly more dangerous crimes than Davis’s, carry a lesser presumptive sentence and, with mitigation, those who commit such crimes are eligible for a ten-year minimum sentence for each count, less than the thirteen-year minimum sentence for each count for Davis’s crime. A.R.S. § 13-604.01(D). ¶40 Additionally, other serious felony offenses not involving children receive significantly less severe sentences in Arizona. Those guilty of kidnapping a person older than fifteen, A.R.S. § 13-1304 (2001), sexual assault of a victim older than fifteen, A.R.S. § 13-1406 (2001), first degree burglary of a residential structure, A.R.S. § 13-1508 (2001), and arson of an occupied structure, A.R.S. § 13-1704 (2001), are all eligible for more lenient sentences, and the sentencing judge may order that multiple sentences be served concurrently. 21 See A.R.S. § 13-708 (2001). ¶41 In response, the State cites five decisions affirming extremely harsh sentences imposed against those convicted of sex crimes against children. Taylor, 160 Ariz. at 422, 773 P.2d at 981 (eighty-five consecutive life sentences for a total of 2975 years for sexual exploitation and sexual conduct with minors); State v. Jones, 188 Ariz. 534, 937 P.2d 1182 (App. 1996) (six consecutive twenty-five-year sentences for sexual assault); State v. Hamilton, 177 Ariz. 403, 868 P.2d 986 (App. 1993) (three consecutive twenty-year sentences for child molestation); State v. Ross, 166 Ariz. 579, 804 P.2d 112 (App. 1990) (twenty- five-year sentence for one count of sexual assault); State v. Crego, 154 Ariz. 278, 742 P.2d 289 (App. 1987) (forty-year sentence for child molestation). The State argues that these cases show that Davis’s sentences are not grossly disproportionate given the way other similar crimes have been punished in Arizona. ¶42 But even a cursory review of these five cases reveals enormous differences in the nature of the crimes, the harm to the victims and to society, and the culpability of the defendants. In Jones, the victim was the defendant’s daughter, whom he raped and molested, coerced by threats of violence, for nearly ten years, from the time the child was five years old until she turned fourteen. Jones, 188 Ariz. at 537, 539, 937 22 P.2d at 1185, 1187. The five known victims who were the subjects of the eighty-five counts on which Taylor was convicted ranged from three to eight years old, and many unknown victims were never located. Taylor, 160 Ariz. at 417, 423, 773 P.2d at 976, 982.7 The defendant, who had three prior felonies involving sexual conduct with minors, had a large collection of photographs of the young victims engaged in various sex acts with each other and with him. Id. at 417, 422, 773 P.2d at 976, 981. In Hamilton, the victims were the defendant’s girlfriend’s twelve- and nine-year-old daughters, who had been subjected to years of sexual abuse and threats of violence when left in the defendant’s care. Hamilton, 177 Ariz. at 405, 868 P.2d at 988. In Ross, the defendant and two other males abducted a fourteen- year-old girl whom they sexually assaulted. Ross, 166 Ariz. at 582, 804 P.2d at 115. Finally, the defendant in Crego was convicted of molesting two children under the age of fifteen only months after being released from prison for similar conduct. Crego, 154 Ariz. at 279-80, 742 P.2d at 290-91. ¶43 Thus, the facts of these cases stand in stark contrast to the facts of the case before us, in which the post-pubescent 7 Although a sentence of nearly 3000 years is extreme, it is the functional equivalent of a life sentence without the possibility of parole. 160 Ariz. at 423, 773 P.2d at 982. Such a sentence is not cruel and unusual given the facts of Taylor. Id. 23 victims sought Davis out and willingly participated in the criminal acts.8 Indeed, we draw two conclusions from the cases the State cites. First, Davis’s fifty-two-year sentence is grossly disproportionate to his crimes when compared with the sentences imposed for the crimes described in the five cases cited by the State. Second, these cases vividly demonstrate why, when considering the proportionality of a sentence imposed, this court must look beyond the nomenclature of the crime charged and consider the facts of each particular case. c. Inter-jurisdictional Comparison. ¶44 The inter-jurisdictional analysis requires comparison of punishments imposed for the same crime in other states. Bartlett II, 171 Ariz. at 310, 830 P.2d at 831. This court conducted such a comparison in Bartlett I. 164 Ariz. at 237-41 & nn.4-5, 792 P.2d at 700-04 & nn.4-5. What was true in 1990 when we decided Bartlett I remains true today; the sentence 8 We do not mean to blame the victims or to imply that they “asked for” what they got. See Dissent, ¶ 78. Nor do we mean to justify Davis’s crimes. See id. We cite these facts to show that Davis did not threaten violence or use force to effectuate his crimes. But in analyzing the constitutionality of the sentence before us, we are required to assess “the sentence-triggering criminal conduct,” which includes “the offender’s actual behavior or other offense-related circumstances.” Ewing, ___ U.S. at ___, 123 S. Ct. at 1194 (Breyer, J., dissenting) (citing Rummel v. Estelle, 445 U.S. 263, 276, 100 S. Ct. 1133, 1140 (1980), and Solem, 463 U.S. at 290-303, 103 S. Ct. at 3010-16). These “offense-related circumstances” show that the Defendant did not threaten violence or use force to effectuate his crimes. 24 Davis received is much more severe than the minimum possible sentence a defendant could receive in any other state.9 See supra ¶ 19. ¶45 The minimum sentence in Arizona for an offender who has no criminal history, but has been convicted of the offenses at issue here, is four thirteen-year sentences, which must run consecutively and for which there is no possibility of parole, for a total of fifty-two years. See A.R.S. § 13-604.01. In no other state would a defendant in similar circumstances face a minimum possible sentence exceeding twenty years, and in the few states in which a twenty-year sentence is possible, the sentencing judge has the discretion to reduce the sentence. E.g., R.I. Gen. Laws §§ 11-37-8.1 to -8.2 (2002). In nearly all states, a defendant guilty of similar crimes could receive concurrent sentences totaling fewer than five years’ imprisonment. E.g., Cal. Penal Code §§ 261.5, 288(a) (West 9 In applying this analysis from Bartlett I to Davis, we note that Bartlett was twenty-three when he committed his crimes, while Davis was only twenty. The dividing line for criminal responsibility in many states is twenty-one years of age. Because Davis was only twenty, he would receive more leniency in some states than Bartlett would have received. On the other hand, some states increase penalties for sexual conduct with one younger than fourteen. In Bartlett, both victims were fourteen. Davis was charged with one count of sexual conduct with T.E., who was thirteen. In jurisdictions that draw the line at fourteen, Davis might have received a stiffer penalty than he would in jurisdictions that do not so distinguish, but still would not receive a sentence of fifty-two years in any other jurisdiction. 25 1999), 1170 (West 1985 & Supp. 2003), 1203 (West 1982 & Supp. 2003), 3000 (West 2000 & Supp. 2003) (setting the minimum sentence that could be imposed for Davis’s offenses at three years, with the possibility of and eligibility for probation and parole); N.M. Stat. Ann. §§ 30-9-11(F) (Michie 1997 & Supp. 2002), 31-18-15(A)(6), 31-18-15.1, 31-20-3, 31-20-5, 31-21-10 (Michie 2001) (providing for a sentence of eighteen months, probation- and parole-eligible). ¶46 In this case, the trial judge and the members of the jury thought Davis’s sentence was “clearly excessive.” The pre- sentence report recommended a sentence in the range of five years, a proposal with which the victims’ mothers agreed. Even the prosecutor recommended that Davis be eligible to immediately apply for clemency. But the trial judge was statutorily bound to impose a flat fifty-two-year sentence. In no other state would a sentencing judge be required to impose such a severe sentence. Davis’s sentence, therefore, fails the third prong of the gross disproportionality test. d. Consecutive Sentences. ¶47 Although this court normally will not consider the imposition of consecutive sentences in a proportionality inquiry, this case cries out for departure from that general rule. See Bartlett I, 164 Ariz. at 239 n.6, 792 P.2d at 702 n.6. It is in part because judges in Arizona have no discretion 26 regarding the minimum sentence and must impose consecutive sentences that this sentence fails the proportionality test.10 See A.R.S. §§ 13-604.01(C) (requiring minimum sentence), 13- 604.01(K) (requiring that sentences be served consecutively). Therefore, to ignore the requirement that the sentences be served consecutively would be to ignore one of the causes of the disproportionality. We recognize the legislature’s right to impose a thirteen-year minimum sentence for dangerous crimes against children and to require that the sentences be served completely. We also recognize the legislature’s right to require consecutive sentences for this type of offense. We cannot, however, uphold a sentence that becomes unconstitutionally disproportionate to the crimes committed because the sentences are mandatorily lengthy, flat, and consecutive. ¶48 Accordingly, while we recognize that Davis committed crimes worthy of severe punishment, we nonetheless find that the application of the mandatory sentencing provisions of the 10 In Lockyer and Ewing, the Court considered important that the prosecutor had the discretion to charge the crimes as misdemeanors or felonies and, in both cases, had chosen to charge the crimes as felonies. Lockyer, ___ U.S. at ___, 123 S. Ct. at 1170; Ewing, ___ U.S. at ___, 123 S. Ct. at 1183. Even after conviction, the trial judge could have declined to impose the sentence enhancement by defining the triggering offenses as misdemeanors rather than felonies. Lockyer, ___ U.S. at ___, 123 S. Ct. at 1170; Ewing, ___ U.S. at ___, 123 S. Ct. at 1183. The trial judge had no such discretion in Davis’s case. 27 Dangerous Crimes Against Children Act creates an unconstitutionally disproportionate punishment in light of the specific facts and circumstances of Davis’s offenses. We therefore vacate the sentences and remand this case to the trial court for resentencing for the offenses for which the convictions are affirmed as class two felonies, non-dangerous, pursuant to A.R.S. §§ 13-702, 13-702.01 and 13-702.02. ¶49 Let us be clear that we do not find Davis’s sentence disproportionate simply because it “seems too long.” Nor are we merely substituting our judgment for the legislature’s considered determination of the appropriate punishment for illegal sexual conduct with a minor. Sex forced on a minor by an adult may be a dangerous crime and the legislature may punish it as such. But we recognize that the Supreme Court has construed the Eighth Amendment to impose on a reviewing court the duty to examine a sentence claimed to be cruel and unusual in light of the specific facts and circumstances under which it is imposed. Having done so in this case, we abide by this court’s determination that when a punishment is “so severe as to shock the conscience of society,” it “violates the constitutional mandate.” State v. Davis (Randal), 108 Ariz. 335, 337, 498 P.2d 202, 204 (1972). That Davis’s fifty-two-year sentence shocks the societal conscience is apparent from the reactions of the trial judge, the jurors, and the victims’ 28 mothers. Accordingly, our conclusion that Davis’s sentence violates the Eighth Amendment’s prohibition against grossly disproportionate punishments is limited to the specific facts and circumstances in the record before us and is based on our determination that it is so disproportionate to the offenses that it shocks the moral sense of the court and community. B. Lack of a Specific Offense Date in Counts One and Four ¶50 At trial, Davis objected to the instruction informing the jury that the prosecutor need not prove the exact date on which the offenses occurred, but that it was “sufficient if the evidence shows beyond a reasonable doubt that the crime was committed by the defendant on or about these dates charged in the indictment.” Davis argues that the State’s failure to specify a date for counts one and four impermissibly amended the indictments on those counts. This, he claims, resulted in a duplicitous charge, which created the possibility of a non- unanimous jury verdict as to count one and nullified his alibi defense on counts one and four. We affirm Davis’s conviction on count four, but reverse and remand for a new trial on count one. 1. Count One. ¶51 Count one charged Davis with having sex with T.E. “on or about the 18th day of January, 1999.” At trial, T.E. testified that she had sex with Davis twice, once around the middle of January and once again during the weekend of January 29 29-31. The State presented the doctor who examined T.E. in early February. She testified that the examination revealed signs consistent with T.E. having had sexual intercourse “within the past week,” which would have been the end of January. Although the prosecutor mentioned briefly in closing that the event charged occurred on the 18th, she also argued that the doctor’s testimony supported a finding that the act occurred on Super Bowl weekend. Thus, the jury heard evidence that Davis had sex with T.E. on two separate occasions, occurring at least eleven days apart. The verdict form for count one, sexual conduct with T.E., did not specify the date of the offense. ¶52 Davis contends that because the trial judge informed the jury that the exact dates were not important, the verdict form for count one contained no date, and there was evidence that he had sex with T.E. on two separate occasions, count one effectively and impermissibly charged two crimes.11 Some jurors might have concluded that he had sex with T.E. on January 18, 11 The court of appeals found it “clear that the state presented evidence aimed at proving the first act of intercourse between T.E. and defendant.” State v. Davis, 1 CA-CR 99-1028, slip op. at ¶ 25 (Ariz. App. Oct. 16, 2001) (mem. decision). The record shows, however, that the State also presented evidence and argued in closing that T.E. engaged in sexual conduct with Davis during the Super Bowl weekend. Specifically, T.E. testified that she engaged in sexual conduct on two occasions and the doctor testified that an early February exam showed that T.E. had engaged in sex one week before, around the end of January, during Super Bowl weekend. 30 while others, relying on the doctor’s testimony, might have concluded that he had sex with her the weekend of January 29-31, but the jury was not unanimous on either date. ¶53 We agree with Davis that such a possibility existed. a. Charging more than one act in Count One of the Indictment. ¶54 “[E]ach offense must be charged in a separate count” in an indictment or complaint. State v. Whitney, 159 Ariz. 476, 480, 768 P.2d 638, 642 (1989) (citing Ariz. R. Crim. P. 13.3(a)). Charging more than one act in a single count is forbidden because it does not provide “adequate notice of the charge to be defended, . . . present[s] a hazard of a non- unanimous jury verdict, and . . . make[s] a precise pleading of prior jeopardy impossible in the event of a later prosecution.” Id. (citations omitted). ¶55 We considered an argument similar to the one Davis now makes in Spencer v. Coconino County Superior Court (State), 136 Ariz. 608, 667 P.2d 1323 (1983). In Spencer, a father was charged with one count of incest and one count of molestation for a series of offenses occurring over the course of nearly four years. Id. at 609-10, 667 P.2d at 1324-25. We concluded that the defendant was put in jeopardy in the two counts charged “for any one of what could be one hundred separate criminal offenses.” Id. at 611, 667 P.2d at 1326. Charging in this 31 manner does not give a defendant “clear notice of the crime with which he is charged,” and does not allow him to mount an adequate defense. State v. Martin, 139 Ariz. 466, 471, 679 P.2d 489, 494 (1984). Although the scale is smaller in Davis’s case, the principle applies. ¶56 But the State asserts, and the court of appeals agreed, that State v. Schroeder, 167 Ariz. 47, 804 P.2d 776 (App. 1990), disposes of Davis’s claim. In Schroeder, the defendant was charged with only one count of sexual abuse, but the victim testified to seven distinct acts of fondling that occurred one evening. Id. at 51, 804 P.2d at 780. The defendant argued that the charge was duplicitous because it failed to specify the particular act on which the charge was based. Id. ¶57 The court in Schroeder held the charge not duplicitous because it determined the question to be simply one of credibility; “only if the jury unanimously accepted the victim’s version of the events could they find defendant guilty of sexually abusing her.” Id. at 53, 804 P.2d at 782. ¶58 The acts alleged by T.E., on the other hand, occurred eleven days apart, and, unlike the defendant in Schroeder, Davis offered more than one defense. Davis offered an alibi defense for the charge that he engaged in sexual conduct with the victims over the Super Bowl weekend and contended that he did 32 not have sex with T.E. at all. He also attempted to show that T.E. could have engaged in sex with another person staying at his house during the Super Bowl weekend. Thus, the two distinct acts here differ from the evening’s events in Schroeder, and call for a different conclusion. ¶59 We conclude that Davis’s case is more like Spencer’s. Spencer, 136 Ariz. at 611, 667 P.2d at 1326. We think it possible that some jurors may have believed Davis’s alibi defense and convicted him for an offense on January 18, while other jurors may have convicted him for the Super Bowl weekend offense, based on the doctor’s testimony that T.E. had engaged in sex within the preceding week. Because we cannot be certain which offense served as the predicate for the conviction, we conclude that the real possibility of a non-unanimous jury verdict exists. ¶60 The State contends, however, that the charge is not duplicitous because the date of the offense is not an element of the crime of sexual conduct with a minor. See Jones, 188 Ariz. at 543, 937 P.2d at 1191. While we agree with the general proposition, the case on which the State relies is inapposite to the point. In Jones, the evidence showed and the jury found the same number of sexual assaults as alleged in the indictment. Id. Here, Davis was convicted of one count, based on proof of two acts. 33 ¶61 If the indictment, the evidence, the jury instructions, and jury forms reflect the same number of offenses, the State does not need to prove the exact date of the offenses. When, however, “the evidence shows, or tends to show, that several acts of intercourse have occurred between defendant and prosecuting witness, it is incumbent upon the prosecution to elect which one of such acts it relies upon for a conviction.” Hash v. State, 48 Ariz. 43, 50, 59 P.2d 305, 308 (1936). Failure to do so results in a duplicitous charge. That is what happened here, and the resulting risk that the jury returned a non-unanimous verdict constituted error. b. Waiver. ¶62 The court of appeals found, and our review of the record confirms, that Davis did not object at trial to the jury form for count one. Failure to object at trial waives an error unless the error is fundamental. Ariz. R. Crim. P. 21.3(c); State v. Gendron, 168 Ariz. 153, 154, 812 P.2d 626, 627 (1991). The court of appeals concluded that the failure to specify the date on the form was not fundamental error. ¶63 While we agree that failure to object to a jury form will rarely result in reversal, that is not the gravamen of Davis’s complaint. Davis complains that the verdict form allowed the jury to choose between two acts on which to convict him on count one, creating a duplicitous charge. Therefore, the 34 heart of Davis’s complaint regarding count one is not an incorrect jury form, but rather a duplicitous charge, which led to a non-unanimous jury verdict. ¶64 Article 2, Section 23 of the Arizona Constitution guarantees a defendant the right to a unanimous jury verdict in a criminal case. A violation of that right constitutes fundamental error. State v. Woods, 141 Ariz. 446, 456, 687 P.2d 1201, 1211 (1984); see also State v. Counterman, 8 Ariz. App. 526, 531, 448 P.2d 96, 101 (1968); cf. United States v. Ullah, 976 F.2d 509, 512 (9th Cir. 1992) (stating that under federal law a unanimous jury verdict is nonwaivable). ¶65 In Counterman, a defendant who fired his weapon twice during a domestic dispute argued that the prosecution presented evidence of two separate assaults and had to elect which event supported the charge, otherwise the charge would be duplicitous. 8 Ariz. App. at 530, 448 P.2d at 100. The court noted that a duplicitous charge creates the danger that the jury will not return a unanimous verdict and failure to do so is fundamental error. Id. at 531, 448 P.2d at 101; see also Woods, 141 Ariz. at 456, 687 P.2d at 1211. While in Counterman the court upheld the defendant’s conviction because it determined that the series of events formed a single transaction, such is not the situation in Davis’s case. The two acts that the State relied upon to support the charge involving T.E. occurred approximately eleven 35 days apart and were not part of a single transaction. ¶66 Because the State offered evidence of more than one offense to support its first charge against Davis and the events were not part of a single transaction, we conclude that the charge was duplicitous. And because the jury determination may have been other than unanimous, we find the error fundamental. Accordingly, we vacate Davis’s conviction as to count one and remand for a new trial. Because we vacate Davis’s conviction for count one based on duplicity, we need not consider the effect of the jury instruction on his alibi defense for that count. 2. Count Four. ¶67 Count four charged Davis with having sex with P.T. “on or about January 29.” At trial, the State attempted to prove that P.T. had sex with Davis on three separate occasions: January 20, 25, and 29 (counts two, three, and four). Davis presented an alibi for January 29, and thus asserted that he could not have committed the crime charged in count four. The trial court instructed the jury that the exact date was not an element of the offense and that the State need only prove that the crime was committed “on or about” the date alleged. ¶68 Davis argues that because the State was not required to prove the specific date for count four, he was effectively prevented from presenting an alibi defense. We disagree. 36 ¶69 Davis testified that he worked all day on January 29, adding that he spent the night in his truck at the job site. Davis’s boss confirmed that Davis worked at the job site that day. Davis’s trial counsel argued in closing that Davis was not at home on January 29 and thus could not have committed the crime on the 29th. The trial judge gave an alibi instruction before the jury began its deliberations. Thus, Davis presented evidence of an alibi, his counsel argued the alibi defense in closing, and the trial judge instructed the jury regarding the alibi defense. The jury simply did not believe the defense. ¶70 A defendant’s mere assertion of an alibi defense “cannot compel the state to elect an exact day.” State v. Simmering, 89 Ariz. 261, 264, 361 P.2d 4, 6 (1961). In State v. Verdugo, for example, the defendant was charged with two counts of rape occurring “on or about the month of August.” 109 Ariz. 391, 392, 510 P.2d 37, 38 (1973). The defendant put forth an alibi defense, offering evidence that he was out of town during the summer and fall of that year and presenting witnesses who testified that he worked with them in Nashville, Tennessee until “Halloween” that year. Id. This court concluded that the defendant was not deprived of his alibi defense; rather, the jury “apparently failed to believe” his defense. Id. ¶71 In this case, the State presented evidence that count four occurred on January 29, but it also presented evidence that 37 the offense occurred sometime during that weekend. The jury heard from P.T. that she could not remember the exact dates of her sexual encounters with Davis, but she narrowed the date of the third encounter to Super Bowl weekend, which included January 29. Perhaps most damaging to Davis’s defense was his own admission that he had sex with P.T. on three separate occasions in January. Thus, the jury either rejected Davis’s alibi defense outright or concluded that, although he may have been at work on the 29th, he nevertheless had a sexual encounter with P.T. “on or about” January 29th. In either circumstance, there was sufficient evidence to support the conviction on count four of the indictment. CONCLUSION ¶72 We vacate Davis’s conviction as to count one, affirm all other convictions, vacate the sentences imposed, and remand the case to the superior court for retrial and sentencing pursuant to A.R.S. §§ 13-702, 13-702.01 and 13-702.02. _______________________________________ Rebecca White Berch, Justice CONCURRING: ______________________________________ Charles E. Jones, Chief Justice 38 ______________________________________ Nanette M. Warner, Judge* *Due to the vacancy created by the retirement of Justice Thomas A. Zlaket, the Honorable Nanette M. Warner, Judge of the Pima County Superior Court, was designated to sit with the court and participate until the final disposition of this matter. Ariz. Const. art. 6, §§ 3, 20; A.R.S. § 38-813 (2001). NOTE: Justice Stanley G. Feldman sat for oral argument but retired before the filing of the opinion and therefore did not participate in the opinion. M c G R E G O R, Vice Chief Justice, concurring in part and dissenting in part: ¶73 I agree with much of the majority opinion. I agree that Davis received a very long sentence following his conviction of four counts of sexual misconduct with a minor, based upon four instances of sexual intercourse with girls under fifteen years of age. Op. ¶ 36. I agree that the United States Supreme Court has now made clear, through its decision in Ewing v. California, ___ U.S. ___, ___, 123 S. Ct. 1179, 1187 (2003), that in deciding whether a sentence is so grossly disproportionate as to violate the Eighth Amendment, we should apply the approach first announced in Justice Kennedy’s concurring opinion in Harmelin v. Michigan, 501 U.S. 957, 996- 1001, 111 S. Ct. 2680, 2702-05 (1991) (Kennedy, J., concurring). 39 Op. ¶ 30. I also agree that applying the Harmelin principles allows us to consider the facts of the crime involved. Op. ¶ 31. ¶74 For several reasons, I dissent from the majority’s holding that the Eighth Amendment prohibits the sentence imposed in this case. First, the majority interprets Harmelin too broadly and bases its proportionality analysis not only upon the facts of the crime but also upon its view of the relative culpability of the defendant and relative blameworthiness of the victims. Second, while the majority states that it will apply the proportionality principles of Harmelin, the opinion instead relies upon the approach defined in Solem v. Helm, 463 U.S. 277, 103 S. Ct. 3001 (1983), an approach that produces a result different from that reached under the Harmelin analysis. Finally, the majority relies upon factors that, in my view, should not apply to and do not further an Eighth Amendment analysis. I. ¶75 My first point of departure from the majority involves its decision to conduct its proportionality analysis by considering not only the facts of Davis’s crimes but also its subjective characterization of the actions of the victims and the defendant. In Ewing v. California, the Supreme Court considered whether the Eighth Amendment prohibits imposing a 40 life sentence under California’s recidivist statute, commonly referred to as the “three strikes and you=re out” law. ___ U.S. at ___, 123 S. Ct. at 1182. As the majority notes, the Ewing Court discussed the nature of the offenses of which Ewing had been convicted prior to engaging in its proportionality review. Id. at ___, 123 S. Ct. at 1189-90. The Court had followed a similar practice in earlier Eighth Amendment proportionality 12 decisions. In those decisions, the Court described, although it seemed not to attach great importance to, the crimes committed by the defendants. In this case, adopting that practice would justify our considering more than the language of the statutes under which the jury convicted Davis. That is, we can take into account not simply that this case involves an adult charged with four counts of sexual misconduct with two minors, but also that the jury convicted Davis, a twenty-year- old man, of four counts of sexual misconduct, occurring on four separate occasions, and involving two girls, aged thirteen and fourteen. ¶76 The majority, however, takes a far more expansive view of the nature of the offenses and also considers the relative 12 See, e.g., Harmelin, 501 U.S. at 961, 111 S. Ct. at 2684; Solem, 463 U.S. at 279-84, 103 S. Ct. at 3004-06; Hutto v. Davis, 454 U.S. 370, 370-72, 102 S. Ct. 703, 703-04 (1982) (per curiam); Rummel v. Estelle, 445 U.S. 263, 264-68, 100 S. Ct. 1133, 1134-36 (1980). 41 culpability of Davis and of the victims. That approach leads to the most disquieting feature of the majority opinion. ¶77 The majority finds significant the victims’ prior sexual history and their “consent” to sexual misconduct by Davis. The opinion discusses, for instance, the fact that T.E. 13 had engaged in sexual activity when she was twelve years old. Op. ¶ 2. The majority also finds it important to note that the victims “sought Davis out,” Op. ¶¶ 36, 43, that the victims “knew what they were doing,” Op. ¶ 36, and that they “willingly participated” in the sexual conduct. Op. ¶¶ 36, 43. ¶78 I thought we long ago had put aside any approach that blames victims for sexual crimes. The notion that a young victim “asks for it” or deserves to become the victim of sexual misconduct because he or she does not object to the sexual activity must be discarded. Our laws permit no justification defense that relies upon the argument that a young victim seduced his or her violator or consented to the violation, and this court should not suggest that such a justification plays any role in evaluating the culpability of a defendant. ¶79 The majority’s decision to examine the personal culpability of the defendant also leads it into a fact-finding 13 For some reason, the majority does not comment upon P.T.’s sexual history, although three of the counts of which Davis was convicted involved her. 42 role for which this court is not suited. The majority states that “[t]here is evidence in the record that Davis’s intelligence and maturity level fell far below that of a normal young adult.” Op. ¶ 36. That finding does not inhere in the jury’s verdict, and we have no finding from the trial judge or the jury that this statement accurately describes Davis. We should not base constitutional analysis upon factual findings made by this court while acting in its capacity as an appellate court. ¶80 Although this court should not engage in characterizing the relative culpability of Davis and the victims, our justice system does provide a method for undertaking the type of analysis engaged in today by the court. If a trial judge concludes that a sentence required by law is clearly excessive, the judge can enter a special order allowing the defendant to petition the board of executive clemency (the Board) for a commutation of sentence within ninety days after a defendant is committed to custody. Arizona Revised Statutes (A.R.S.) § 13-603.L (Supp. 2002). The trial judge followed that procedure in this matter. Op. ¶ 10. If the Board finds by clear and convincing evidence that the sentence is clearly excessive and that a substantial probability exists that the offender will conform to the law, the Board can recommend that the governor commute the sentence. A.R.S. § 31-402.C.2 (2001). 43 The Board makes its findings, however, only after holding a hearing at which the victim, county attorney, and presiding judge receive a chance to be heard. Id. Rather than rely on that existing procedure, the majority today, without the benefit of a hearing, simply substitutes its judgment for that which the Board might make. ¶81 The majority’s decision to consider the relative culpability and blameworthiness of the defendant and the victims as part of an Eighth Amendment proportionality analysis suffers from another shortcoming: It inserts considerable subjectivity into the court’s analysis. The Supreme Court has instructed that the Eighth Amendment proportionality analysis should rely upon objective factors. Ewing, ___ U.S. at ___, 123 S. Ct. at 1186 (citing Harmelin, 501 U.S. at 1001, 111 S. Ct. at 2705). The majority’s approach makes following that direction impossible. Would the outcome today differ if the court regarded the defendant as more mature or if the defendant had been twenty-two rather than twenty years of age or if at least one of the victims had not revealed prior sexual activity? Such subjectivity, it seems to me, has no place in the constitutional analysis this case requires the court to make. II. ¶82 The majority concludes, and I agree, that we should apply the principles of Harmelin in conducting an Eighth 44 Amendment proportionality analysis. Op. ¶ 30. Rather than analyze this case under the Harmelin principles, however, the majority actually applies the analysis used in Solem v. Helm, as interpreted by this court in State v. Bartlett, 171 Ariz. 302, 830 P.2d 823 (1992) (Bartlett II), and State v. Bartlett, 164 Ariz. 229, 792 P.2d 692 (1990) (Bartlett I). That approach skews and affects the outcome of the analysis. ¶83 In Ewing v. California, the Court announced that the “proportionality principles in our cases distilled in Justice Kennedy=s concurrence [in Harmelin] guide our application of the Eighth Amendment . . . .” ___ U.S. at ___, 123 S. Ct. at 1187. The Court recognized four principles of proportionality review identified by Justice Kennedy in Harmelin: “[T]he primacy of the legislature, the variety of legitimate penological schemes, the nature of our federal system, and the requirement that proportionality review be guided by objective factors”Bthat “inform the final one: The Eighth Amendment does not require strict proportionality between crime and sentence. Rather, it forbids only extreme sentences that are ‘grossly disproportionate’ to the crime.” Justice Kennedy’s concurrence [in Harmelin] also stated that Solem “did not mandate” comparative analysis “within and between jurisdictions.” Id. at ___, 123 S. Ct. at 1186-87 (quoting Harmelin, 501 U.S. at 1001, 1004-05, 111 S. Ct. at 2705, 2707). ¶84 In applying the Harmelin/Ewing analysis, we must be 45 mindful that the Eighth Amendment contains only a “narrow proportionality principle” that “applies to noncapital sentences.” Harmelin, 501 U.S. at 996-97, 111 S. Ct. at 2702- 03. “Outside the context of capital punishment, successful challenges to the proportionality of particular sentences have been exceedingly rare.” Rummel v. Estelle, 445 U.S. 263, 272, 100 S. Ct. 1133, 1138 (1980). To demonstrate the extremely narrow parameters of the proportionality principle in noncapital cases, the Supreme Court has explained that the principle would “come into play in the extreme example . . . if a legislature made overtime parking a felony punishable by life imprisonment.” Id. at 274 n.11, 100 S. Ct. at 1139 n.11. This case does not fit within the narrow scope of the proportionality principle that the court purports to apply. A. ¶85 Under the Harmelin/Ewing analysis, we should first consider the primacy of the legislature and focus our analysis upon the public policy judgments made by the legislature when it enacted the Dangerous Crimes Against Children Act (the Act), which includes A.R.S. section 13-604.01, the sentencing statute at issue here.14 14 For a comprehensive description of the various provisions adopted as part of the Dangerous Crimes Against Children Act, see State v. Williams, 175 Ariz. 98, 854 P.2d 131 (1993). 46 ¶86 We have previously considered the legislature’s intent in adopting the enhanced penalties of section 13-604.01: The legislature’s purpose in enacting the Dangerous Crimes Against Children Act can be surmised. Protecting the children of Arizona and punishing severely those who prey upon them certainly are two legislative goals. In addition . . . the legislature is attempting to address the problem of recidivism alleged to exist in this category of offender. State v. Wagstaff, 164 Ariz. 485, 490-91, 794 P.2d 118, 123-24 (1990). The language of section 13-604.01 reflects the legislature’s decision to protect, punish, and deter. ¶87 Rather than defer to the legislature’s public policy choices reflected in the Act, the majority identifies five factors it deems important in concluding that Davis’s sentence, although required by the clear language of the Act, was grossly disproportionate to the crimes he committed. I find none of the majority’s justifications for departing from the statutory language persuasive. ¶88 The majority first concludes that the crimes “involved neither actual nor threatened violence” because the victims consented to the sexual acts, noting that the girls “willingly participated.” Op. ¶ 36. In other words, the crimes involved “non-coerced sex with two post-pubescent teenage girls.” Id. But the legislature has already made the policy decision that children under the age of fifteen cannot give consent to sexual 47 intercourse.15 ¶89 Our society and legal system recognize that, when a child falls below a particular age, that child lacks capacity to consent to harmful acts committed against the child. Because children cannot protect themselves, society and the law assume responsibility for protecting children in those instances in which a child’s lack of capacity prevents him or her from giving consent: The state has a recognized interest in the welfare of its citizens who, by reason of age or physical or mental disability, cannot care for themselves. So it is with children of tender years. The conclusive presumption that children less than sixteen years of age are unable to consent to sex acts is but a further extension of the protective arm of government which is universally followed. Payne v. Commonwealth, 623 S.W.2d 867, 875 (Ky. 1981); see also Charles A. Phipps, Children, Adults, Sex and the Criminal Law: In Search of Reason, 22 Seton Hall Legis. J. 1, 33-34 (1997) (noting that protecting children is one of the most consistently expressed reasons for prohibiting sexual conduct with children). ¶90 We do not enforce contracts entered into by children 15 Under A.R.S. section 13-1407, an accused can defend a charge under section 13-1405, the statute under which the jury convicted Davis, by asserting his lack of knowledge of the victim’s age if the victim’s lack of consent is based on incapacity to consent and the victim is fifteen, sixteen or seventeen years of age. A.R.S. § 13-1407.B (2001). The legislature expressly excluded this defendant from being able to argue that these victims could actually consent to his actions. 48 because they lack capacity to consent, Worman Motor Co. v. Hill, 54 Ariz. 227, 231, 94 P.2d 865, 866 (1939), and we surely must not place responsibility for determining whether a child should engage in sexual activity with an adult upon the child. Yet the majority adopts just such a view today and makes the victims of these crimes at least partially responsible because they “willingly participated.” ¶91 Determining the age of consent requires that a line be drawn. Presumably all would agree that a ten-year-old child could not consent to sexual relations, no matter what words the child used to express consent, and even if he or she had become sexually active at eight years of age. The majority may disagree with the placement of the line drawn for consent by the legislature, but unless the age selected by the legislature is irrational, this court should not engage in re-drawing the lines involving age of consent.16 ¶92 Nor can we take comfort from the fact that these victims were “post-pubescent,” another fact afforded 16 The statutes defining sexual offenses against children frequently involve distinctions based on age. See, e.g., A.R.S. § 13-1405.B (2001) (class of felony depends upon whether victim is under fifteen years of age); A.R.S. § 13-1407.F (2001) (defense to prosecution if victim fifteen, sixteen or seventeen and defendant less than nineteen or attending high school and no more than twenty-four months older than victim and conduct is consensual); A.R.S. § 13-1410.A (2001) (child molestation involves sexual contact with child under fifteen years of age). 49 significance in the majority opinion. Op. ¶¶ 36, 43. The age of consent involves mental, not physical, maturity. A profoundly retarded adult could not consent to sexual relations, regardless of his or her physical maturity. A.R.S. § 13- 1401.5(b) (2001) (“without consent” includes victim incapable of consent by reason of mental disorder or mental defect). Whether these children were pre- or post-pubescent, therefore, does not affect their capacity to consent. ¶93 The majority then observes that Davis “does not have an adult criminal record, nor has he committed any previous crimes against children.” Op. ¶ 36. The record supports that statement, although Davis, of course, did commit more than one crime against children; the jury convicted him of four crimes against children. ¶94 The majority also finds support for its holding by observing that post-pubescent sexual conduct appears no less common today than it was in 1990, the year we decided Bartlett I. I do not doubt that the majority’s statement is accurate, but the legislature also presumably knew of “post-pubescent sexual conduct” when it adopted the Act and imposed harsher punishment for those who engage in sexual misconduct with children under the age of fifteen. ¶95 The majority further states that “[t]here is evidence in the record that Davis’s intelligence and maturity level fell 50 far below that of a normal young adult.” Op. ¶ 36. As previously mentioned, this factor relies upon fact-finding undertaken by this court, without the benefit of a hearing, and provides no reliable basis for conducting a proportionality review under the Eighth Amendment. ¶96 The next factor considered in the majority’s proportionality review involves its conclusion that “Davis was caught in the very broad sweep of the governing statute, which makes any sexual conduct with a person younger than fifteen years old by a person older than eighteen years old a ‘dangerous crime against children . . . .’” Id. The statute, by its terms, generally operates precisely as described.17 We previously have recognized, however, that the legislature used broad language because it intended that an adult who targets a child “assumes the risk that the victim will turn out to be within a protected age group.” State v. Williams, 175 Ariz. 98, 103, 854 P.2d 131, 136 (1993). If the court regards the statute as applying to conduct that the state cannot regulate, we should overturn the statute because it is overbroad. See Purcell v. 17 The legislature has recognized some exceptions to the general language of the statute. For instance, A.R.S. section 13-1407.F allows a defendant to assert, as a defense to prosecution, that he is less than nineteen or attending high school and that the victim is between fifteen and seventeen and consented to the sexual conduct. The majority seeks to add further exceptions to the language of the statute. 51 Super. Ct., 111 Ariz. 582, 584, 535 P.2d 1299, 1301 (1975) (finding that a statute that applies to conduct the state cannot regulate is overbroad). Absent such a determination, we should apply the statute as written to further the legislature’s intent in adopting the Act. ¶97 The majority might prefer that different ages apply, both for defendants and victims, Op. ¶ 44 n.9, and the Arizona Legislature certainly could have selected older or younger ages. Another legislature might well conclude that sexual conduct with a fourteen-year-old child is not a dangerous crime against children unless the adult is twenty-one or twenty-five or thirty years old, or unless the victim is less than fourteen, or less than thirteen, or less than twelve years of age. But we generally leave to the legislature the task of drawing such lines, which must be placed somewhere if we regard dangerous crimes against children as deserving punishment more severe than crimes committed against adults. ¶98 We should not conduct a proportionality review without giving full consideration to the policies underlying a challenged sentencing scheme. In Ewing v. California, the Court concluded: “When the California Legislature enacted the three strikes law, it made a judgment that protecting the public safety requires incapacitating criminals who have already been convicted of at least one serious or violent crime. Nothing in 52 the Eighth Amendment prohibits California from making that choice.” Ewing, ___ U.S. at ___, 123 S. Ct. at 1187. In adopting the Dangerous Crimes Against Children Act, the Arizona Legislature made a judgment that public safety requires severe punishment of adults who engage in sexual conduct with children under the age of fifteen. Our laws have long granted special protection to children, and I do not think the Eighth Amendment prohibits the state from choosing to afford extra protection to children under the age of fifteen when they become victims of sexual misconduct by adults. Although the Act required the trial judge to impose very long sentences in this instance, I would find no gross disproportionality between the crimes and the sentences. B. ¶99 The second principle of proportionality review defined by Harmelin/Ewing, which the majority does not address, directs us to consider the variety of legitimate penological schemes that a legislature may choose to adopt. This court has recognized that, in adopting the Act, the legislature intended to protect children, to punish severely those who prey upon children, and to address the problem of recidivism alleged to exist in this class of offender. Wagstaff, 164 Ariz. at 490-91, 794 P.2d at 123-24. Protection, punishment, and deterrence constitute appropriate penological schemes. Applying this 53 factor of the Harmelin/Ewing proportionality review, then, also indicates that this case does not fall within the narrow range of those sentences that offend the Eighth Amendment. C. ¶100 The final Harmelin/Ewing principle applicable here18 requires that “proportionality review be guided by objective factors.” Ewing, ___ U.S. at ___, 123 S. Ct. at 1186. I cannot discern the objective factors applied by the majority. Rather than consider objective factors, the majority has elected to base its decision upon factors other than those the legislature emphasized in adopting the Act. I am uncertain how to apply the approach adopted today, which echoes that of Bartlett I and Bartlett II, except to conclude that when an adult under some undefined age engages in sexual conduct with a child above some undefined age, we will find the sentences required by Arizona law grossly disproportionate to the crimes committed, particularly if the victim is post-pubescent and admits to some sexual experience. Those factors do not, in my view, provide the sort of objective analysis envisioned by Harmelin/Ewing. 18 The third principle of proportionality review involves the nature of the federal system. Ewing, ___ U.S. at ___, 123 S. Ct. at 1186. That factor does not apply to our analysis of state legislation, although we should afford the same deference to legislative decisions involving sentencing schemes as do the federal courts. 54 ¶101 The majority opinion implicitly recognizes the lack of standards developed for a proportionality analysis by limiting its decision “to the specific facts and circumstances in the record before us,” Op. ¶ 49, the same approach taken in Bartlett I and Bartlett II. Bartlett II, 171 Ariz. at 308-09, 830 P.2d at 829-30 (“As we said before . . . ‘[a]lthough such a harsh penalty may be justified in the context of other, more heinous crimes . . . it is not justified under the specific circumstances of this case.’” (quoting Bartlett I, 164 Ariz. at 236, 792 P.2d at 699)); Bartlett I, 164 Ariz. at 241, 792 P.2d at 704 (“In this case, however, we hold that the statutory range of sentencing . . . cannot constitutionally be applied to defendant’s crimes under the facts and circumstances of this case.”). The state, defendants, victims and trial courts should be able to rely upon clear principles of analysis that apply to all cases, rather than being required to consider the specific facts of each case without discernible guidelines. The principles of proportionality review set out in Harmelin/Ewing provide such guidelines, and I would follow them. III. ¶102 Although Harmelin/Ewing does not mandate a comparative analysis within and between jurisdictions, the majority devotes much of its opinion to engaging in just such an analysis. Op. ¶¶ 38-43. I disagree with the premise underlying that analysis. 55 ¶103 The majority’s intra-jurisdictional comparison relies upon the premise that those convicted of “crimes more serious” receive the same or a lesser presumptive sentence when compared with that given Davis. Op. ¶ 38. But that conclusion assumes the question. The legislature has determined that sexual conduct with children unable to give consent is a very serious crime. The majority may regard kidnapping of a person older than fifteen, first degree burglary of a residential structure, or arson of an occupied structure as more serious than sexual conduct with a minor, Op. ¶ 40, but the legislature did not. Although the majority assumes that the harm to society and victims for the crimes involved here is relatively slight, the legislature could, and did, decide otherwise.19 ¶104 After considering other decisions that upheld extremely harsh sentences for those convicted of dangerous 19 Children involved in sexual conduct are not engaged in harmless fun. Numerous studies point to serious difficulties that result from such activity. For instance, one study found that the frequency of sexual activity during adolescence is related to coerced sex, or sexual victimization, later in life. Janine M. Zweig et al., Adolescent Risk Factors for Sexual Victimization: A Longitudinal Analysis of Rural Women, 17 J. Adolescent Res. 586, 595-96 (Nov. 2002). Another study found that adolescents who initiate health-risk behaviors such as sexual intercourse at an early age frequently experience poorer health later in life, lower educational attainment, and less economic productivity than their peers. Kathleen Mullan Harris et al., Evaluating the Role of “Nothing to Lose” Attitudes on Risky Behavior in Adolescence, 80 Soc. Forces 1005, 1008 (Mar. 2002). 56 crimes against children,20 Op. ¶¶ 41-42, the majority concludes that those decisions involve “enormous differences in the nature of the crimes, the harm to the victims and to society, and the culpability of the defendants.” Op. ¶ 42. In reaching that conclusion, the majority again relies upon the notion that these “post-pubescent victims sought Davis out and willingly participated in the criminal acts.” Op. ¶ 43. That statement troubles me on several levels. First, nothing these victims did constituted a criminal act; only Davis committed criminal acts. Second, once again the statement imposes responsibility for an adult’s actions upon victims too young to give consent. Much younger children also “consent” to predatory acts. I am uncertain just where this line of analysis leads us, and I am unwilling to conclude, without the benefit of generally applicable principles, that particular crimes do not merit the 20 Because Arizona has adopted harsh sentencing laws, many other examples of extremely long sentences exist, both in cases involving crimes against children and in cases involving other crimes. See, e.g., State v. Jonas, 164 Ariz. 242, 792 P.2d 705 (1990) (forty-six years for trafficking in stolen property and sale of one marijuana cigarette to a minor under fifteen); State v. Hummer, 184 Ariz. 603, 911 P.2d 609 (App. 1995) (three consecutive life sentences and three concurrent twelve-year sentences for convictions of two counts of sexual conduct with a minor, three counts of furnishing obscene materials to a minor, and one count of molestation of a minor, with two prior felony convictions); State v. Freeland, 176 Ariz. 544, 863 P.2d 263 (App. 1993) (life without possibility of parole for twenty-five years for aggravated assault while on probation); State v. Jobe, 157 Ariz. 328, 757 P.2d 604 (App. 1988) (fifty-five years for two counts of child molestation). 57 punishment selected by the legislature. IV. ¶105 In addition to disagreeing with the majority’s failure to apply the Harmelin/Ewing principles, I cannot agree with several other conclusions reached in the majority opinion. ¶106 First, while recognizing that we usually do not consider the imposition of consecutive sentences as part of a proportionality review, the majority finds reason to depart from our general rule. Op. ¶ 47. Although the majority does not make its conclusion explicit, I gather that the majority does not regard a sentence of thirteen years as grossly disproportionate to the crime of sexual misconduct with a child. That is, the majority would not reach the conclusion that the sentences required by law for Davis violate the Eighth Amendment if Davis could have served the sentences concurrently. If that is so, then today’s analysis should focus upon A.R.S. section 13-604.01.K, which requires consecutive sentences, rather than upon section 13-604.01.C. An analysis of section 13-604.01.K would require that we consider the legislature’s reasons for requiring consecutive sentences rather than the facts surrounding the underlying convictions. ¶107 I also cannot agree with the majority’s conclusion that we can determine that Davis’s sentence “shocks the societal conscience” because of “the reactions of the trial judge, the 58 jurors, and the victims’ mothers.” Op. ¶ 49. Surely a test of constitutional validity cannot depend upon such subjective factors. Would the result required by the Eighth Amendment be different if these victims’ mothers had been less understanding of Davis’s conduct or more protective of their children? If the trial judge or jurors had not made their comments, would this case pass constitutional muster? Moreover, why should we look to these few individuals to define the societal conscience? The statute adopted by the legislature also reflects societal norms. Why should we regard the personal reactions of those involved in this particular case as more reflective of society’s norms than is the legislation enacted by a representative body? I cannot agree that constitutional analysis should depend upon the subjective factors relied upon by the majority. ¶108 Finally, although I heartily agree with the majority that the inordinately long sentence involved here derives partly from the fact that judges in Arizona have no discretion in deciding whether to impose consecutive sentences under these circumstances, I do not agree that the lack of discretion should be considered in our proportionality analysis. As the majority avers, in Ewing v. California, the Supreme Court referred to the facts that the crimes involved were “wobblers” and that both the prosecutor and the trial judge could exercise discretion in the treatment of such crimes. The Court expressly stated, however, 59 “That grand theft is a ‘wobbler’ under California law is of no moment.” Ewing, ___ U.S. at ___, 123 S. Ct. at 1189. We cannot, therefore, rely upon Ewing v. California to support the majority’s discussion of the importance of judicial or prosecutorial discretion in conducting this proportionality review. V. ¶109 For the foregoing reasons, I respectfully dissent from paragraphs 13 through 49 of the majority opinion. I concur in the remainder of the decision. _______________________________________ Ruth V. McGregor, Vice Chief Justice 60