People of Michigan v. Raymond Curtis Carp , 496 Mich. 440 ( 2014 )

Markman, J.

We granted leave to appeal to address (1) whether Miller v Alabama, 567 US_; 132 S Ct 2455; 183 L Ed 2d 407 (2012), should be applied retroactively — pursuant to either the federal or state test for retroactivity — to cases in which the defendant’s sentence became final for purposes of direct appellate review before Miller was decided and (2) whether the Eighth Amendment of the United States Constitution or Const 1963, art 1, § 16 categorically bars the imposition of a life-without-parole sentence on a juvenile homicide offender. After considering these matters, we hold that the rule announced in Miller does not satisfy either the federal test for retroactivity set forth in Teague v Lane, 489 US 288; 109 S Ct 1060; 103 L Ed 2d 334 (1989), or Michigan’s separate and independent test for retroactivity set forth in People v Sexton, 458 Mich 43; 580 NW2d 404 (1998), and People v Maxson, 482 Mich 385; 759 NW2d 817 (2008). We further hold that neither the Eighth Amendment nor Const 1963, art 1, § 16 categorically bars the imposition of a life-without-parole sentence on a juvenile homicide offender.

I. FACTS AND HISTORY

A. DEFENDANT CARP

Defendant Raymond Carp was 15 years of age when he participated in the 2006 bludgeoning and stabbing of Mary Ann McNeely in Casco Township. He was charged with first-degree murder in violation of MCL 750.316 and tried as an adult. On October 5, 2006, a St. Clair County jury convicted Carp of this offense, and in accordance with the law he was sentenced to life imprisonment without parole. Carp’s conviction was subsequently affirmed by the Court of Appeals, People v Carp, unpublished opinion per curiam of the Court of Appeals, issued December 30, 2008 (Docket No. 275084), and his application for leave to appeal in this Court was denied on June 23, 2009, People v Carp, 483 Mich 1111 (2009). Because Carp did not seek review in the United States Supreme Court, his conviction and sentence became final for the purposes of direct appellate review on June 23, 2009.

In September 2010, Carp sought to collaterally attack the constitutionality of his sentence by filing a motion for relief from judgment pursuant to MCR 6.501 et seq. The trial court denied this motion, concluding that the imposition of a mandatory sentence of life without parole on a juvenile first-degree-murder offender did not constitute cruel or unusual punishment, citing People v Launsburry, 217 Mich App 358, 363-365; 551 NW2d 460 (1996), lv den 454 Mich 883 (1997), and recon den 454 Mich 883 (1997). Carp then sought leave to appeal in the Court of Appeals, which was denied on June 8, 2012. People v Carp, unpublished order of the Court of Appeals, entered June 8, 2012 (Docket No. 307758). Seventeen days later, the United States Supreme Court issued its decision in Miller, leading Carp to move for reconsideration, and the Court of Appeals granted his motion. People v Carp, unpublished order of the Court of Appeals, entered August 9, 2012 (Docket No. 307758). On reconsideration, the Court determined that Miller had created a “new rule” that was “procedural” in nature and therefore not subject to retroactive application under the rules set forth in Teague. People v Carp, 298 Mich App 472, 511-515; 828 NW2d 685 (2012). The Court further held that Miller was not subject to retroactive application under Michigan’s separate test for retroactivity set forth in Sexton and Maxson.¹ Id. at 520-522. This Court subsequently granted Carp leave to appeal with respect to whether Miller should be applied retroactively under either federal or state law. People v Carp, 495 Mich 890 (2013).

B. DEFENDANT DAVIS

Defendant Cortez Davis, age 16 at the time of his offense, and one of his cohorts, while both brandishing firearms, accosted two individuals in Detroit for the purpose of robbery.² Two witnesses testified that when one of the victims attempted to flee, Davis and his cohort fired five or six shots, killing the victim. Davis was charged with felony first-degree murder in violation of MCL 750.316(l)(b) and convicted by a jury in the former Recorders Court for the City of Detroit (now part of the Wayne Circuit Court) on this charge on May 10, 1994.

At sentencing, the trial court initially ruled that Michigan’s statutory sentencing scheme for first-degree murder could not constitutionally be applied to juvenile homicide offenders because it was “cruel and unusual” to impose a sentence of life without parole on a juvenile who was “capable of rehabilitation.” In concluding that Davis was such an individual, the court surmised that Davis’s role in the commission of the offense was that of an aider and abettor, not an actual shooter. The court, however, did not make any finding concerning Davis’s intentions with respect to the fleeing victim or whether he reasonably foresaw the possibility that a life might be taken when he initially engaged in the firmed robbery. The trial court thereupon sentenced Davis to a term of imprisonment of 10 to 40 years.

On appeal, however, the Court of Appeals reversed and remanded for resentencing pursuant to Michigan’s statutory sentencing scheme, People v Davis, unpublished order of the Court of Appeals, entered November 23, 1994 (Docket No. 176985), and at resentencing, the trial court imposed the required sentence of life without parole. Direct appellate review of defendant’s conviction and sentence concluded in 2000. People v Davis, unpublished order of the Court of Appeals, entered June 15, 2000 (Docket No. 224046).³

In 2010, Davis filed his current motion for relief from judgment, contending that Graham v Florida, 560 US 48; 130 S Ct 2011; 176 L Ed 2d 825 (2010), constituted a “retroactive change in the law” in that it categorically barred life-without-parole sentences for juveniles convicted of nonhomicide offenses. Concluding, however, that felony murder is in fact a “homicide offense,” even when the defendant is not the actual shooter but an aider and abettor, the trial court denied this motion. The Court of Appeals denied Davis’s application for leave to appeal. People v Davis, unpublished order of the Court of Appeals, entered November 16, 2011 (Docket No. 304075). While Davis’s application for leave to appeal in this Court was pending, the United States Supreme Court issued its decision in Miller. In light of Miller, Davis’s case was remanded to the trial court for a determination of whether Miller applied retroactively. People v Davis, 492 Mich 871 (2012). On remand, the trial court concluded that Miller did apply retroactively, entitling Davis to be resentenced. The prosecutor then appealed, and the Court of Appeals reversed. People v Davis, unpublished order of the Court of Appeals, entered January 16, 2013 (Docket No. 314080), citing Carp, 289 Mich App 472. Davis again sought leave to appeal in this Court, which we granted to address whether the Eighth Amendment of the United States Constitution or Const 1963, art 1, § 16 categorically bars imposing a sentence of life without parole on a juvenile convicted of felony murder on aiding-and-abetting grounds. People v Davis, 495 Mich 890 (2013).

C. DEFENDANT ELIASON

Unlike Carp and Davis, whose sentences became final for purposes of direct review before Miller was decided, at least 10 defendants were convicted and sentenced before Miller, but their cases were on direct appeal at the time Miller was decided. Dakotah Eliason is one of those defendants. At age 14, Eliason, without provocation and after hours of deliberation, fired a single deadly shot into the head of his stepgrandfather as he slept in his Niles Township home. Eliason was charged with first-degree murder in violation of MCL 750.316(l)(a) in the Berrien Circuit Court, convicted by a jury, and sentenced in October 2010 to life without parole.

While Eliason’s appeal was pending before the Court of Appeals, Miller was decided. In assessing the effect of Miller on Michigan’s sentencing scheme for juvenile first-degree-murder offenders, the Court of Appeals held that a trial court must as a result of Miller perform an individualized sentencing analysis based upon the factors identified in Miller. People v Eliason, 300 Mich App 293, 309-311; 833 NW2d 357 (2013), citing Carp, 289 Mich App at 522-532. Using this analysis, the trial court must then choose between imposing a sentence of life with or without parole. Eliason, 300 Mich App at 310. Eliason sought leave to appeal in this Court, challenging the sentencing procedures and options defined by the Court of Appeals, contending that the trial court should have the further option of imposing a sentence of a term of years. Eliason additionally argued that Const 1963, art 1, § 16 categorically bars the imposition of a life-without-parole sentence on a juvenile. We granted leave to appeal on both issues. People v Eliason, 495 Mich 891 (2013).

II. MICHIGAN STATUTES

Pending our resolution of this appeal, and in response to Miller, the Legislature enacted, and the Governor signed into law, 2014 PA 22, now codified as MCL 769.25 and MCL 769.25a. This law significantly altered Michigan’s sentencing scheme for juvenile offenders convicted of crimes that had previously carried a sentence of life without parole.

A. PRE-MILLER

To understand the full context of defendants’ appeals and the relief each seeks in reliance on Miller, it is necessary first to delineate the pre-Miller statutes that controlled the trial and sentencing of juvenile first-degree-murder offenders in Michigan. Each defendant before this Court was charged with first-degree murder under MCL 750.316. When a juvenile defendant “14 years of age or older” is charged with a felony, the family division of the circuit court would typically possess initial jurisdiction. MCL 712A.4(1). However, when a juvenile is charged with a “specified juvenile violation,” including first-degree murder in violation of MCL 750.316, “the prosecuting attorney may authorize the filing of a complaint and warrant on the charge . . ..” MCL 764.1f. If the prosecutor does so, the circuit court itself, rather than the family division of the circuit court, acquires jurisdiction over the juvenile defendant’s case and must try that person as an adult. See MCL 712A.2(a)(l).

This process has been termed the “automatic waiver process” because the Legislature has vested exclusively in the prosecutor the executive discretion to charge and try a juvenile as an adult when the juvenile stands accused of first-degree murder. People v Conat, 238 Mich App 134, 141-142; 605 NW2d 49 (1999). The prosecutors in the instant three cases filed complaints and warrants placing the cases within the jurisdiction of the circuit court, where each defendant was then tried and convicted as an adult. When this occurs and the offense is included in an enumerated subset of specified juvenile violations (which includes first-degree murder), “[t]he court shall sentence a juvenile... in the same manner as an adult[.]” MCL 769.1(1). Because an adiilt convicted of first-degree murder “shall be punished by imprisonment for life,” MCL 750.316(1), and is not eligible for parole, MCL 791.234(6)(a), defendants were ultimately sentenced to terms of life without parole. Each defendant now seeks resentencing and, pursuant to the statutory response to Miller, would, if granted resentencing, be subject to the new sentencing rules established for juveniles by 2014 PA 22.

B. POST'-MILLER

MCL 769.25, enacted in response to Miller, prescribes a new sentencing scheme for juveniles convicted of violating certain provisions of Michigan laws, such as MCL 750.316, that had previously carried with them a fixed sentence of life without parole. The effect of MCL 769.25 is that even juveniles who commit the most serious offenses against the laws of this state may no longer be sentenced under the same sentencing rules and procedures as those that apply to adults who commit the same offenses. Rather than imposing fixed sentences of life without parole on all defendants convicted of violating MCL 750.316, MCL 769.25 now establishes a default sentencing range for individuals who commit first-degree murder before turning 18 years of age. Pursuant to the new law, absent a motion by the prosecutor seeking a sentence of life without parole,

the court shall sentence the individual to a term of imprisonment for which the maximum term shall be not less than 60 years and the minimum term shall be not less than 25 years or more than 40 years. [MCL 769.25(4) and (9).]

When, however, the prosecutor does file a motion seeking a life-without-parole sentence, the trial court “shall conduct a hearing on the motion as part of the sentencing process” and “shall consider the factors listed in Miller v Alabama . . . .” MCL 769.25(6). Accordingly, the sentencing of juvenile first-degree-murder offenders now provides for the so-called “individualized sentencing” procedures of Miller.

In adopting this new sentencing scheme, the Legislature was clearly cognizant of the issue surrounding whether Miller was to be applied retroactively. In defining the scope of the new scheme, the Legislature asserted that “the procedures set forth in [MCL 769.25] do not apply to any case that is final for purposes of appeal on or before June 24, 2012 [the day before the United States Supreme Court’s decision in Miller].” MCL 769.25a(l). Instead, the Legislature specified:

If the state supreme court or the United States supreme court finds that the decision of the United States supreme court in Miller v Alabama. [567] US_; 183 L Ed 2d 407; 132 S Ct 2455 (2012), applies retroactively to all defendants who were under the age of 18 at the time of their crimes, and that decision is final for appellate purposes, the determination of whether a sentence of imprisonment for a violation set forth in [MCL 769.25(2)] shall be imprisonment for life without parole eligibility or a term of years as set forth in [MCL 769.25(9)] shall be made by the sentencing judge or his or her successor as provided in this section. [MCL 769.25a(2).][⁴]

We now take up the question identified in MCL 769.25a(2) — whether Miller must be applied retroactively.

III. STANDARD OF REVIEW

Whether a decision of the United States Supreme Court applies retroactively under either federal or state retroactivity rules poses a question of law that is reviewed de novo. Maxson, 482 Mich at 387. Whether a statute is constitutional also poses a question of law that is reviewed de novo. Hunter v Hunter, 484 Mich 247, 257; 771 NW2d 694 (2009). When the constitutionality of a statute is brought into question, “[t]he party challenging [it] has the burden of proving its invalidity.” People v Thomas, 201 Mich App 111, 117; 505 NW2d 873 (1993). To sustain its burden, the party challenging the statute must overcome the presumption that a statute is constitutional, and the statute “will not be declared unconstitutional unless clearly so, or so beyond a reasonable doubt.” Cady v Detroit, 289 Mich 499, 505; 286 NW 805 (1939). Furthermore, a “party challenging the facial constitutionality of a statute faces an extremely rigorous standard, and must show that no set of circumstances exists under which the [a]ct would be valid.” In re Request for Advisory Opinion Regarding Constitutionality of 2005 PA 71, 479 Mich 1, 11; 740 NW2d 444 (2007) (citations and quotation marks omitted).

IV ANALYSIS

To determine whether Miller must be applied retroactively, it is helpful to first identify exactly what Miller held by way of understanding what precedents were relied on in forming its rule. Miller is the product of “two strands of precedent,” one requiring a particular form of individualized sentencing before capital punishment can be imposed and the other addressing the constitutionality of imposing specific punishments on juvenile offenders. Miller, 567 US at_; 132 S Ct at 2463-2464. We now consider both strands of precedent with the purpose of identifying what is required by the rules formed from each strand of precedent and then comparing and contrasting what is required by each with what is required by the rule in Miller in order to determine whether the latter rule should be applied retroactively.

A. GENESIS OF MILLER

1. CAPITAL-PUNISHMENT STRAND

In Furman v Georgia, 408 US 238; 92 S Ct 2726; 33 L Ed 2d 346 (1972), the United States Supreme Court decided 5-4 in seven separate opinions that it constituted cruel and unusual punishment in violation of the Eighth Amendment to impose capital punishment pursuant to a sentencing scheme that, in its words, “vested the [sentencer] with complete and unguided discretion to impose the death penalty .. . .” Beck v Alabama, 447 US 625, 639; 100 S Ct 2382; 65 L Ed 2d 392 (1980). In response, some states enacted sentencing schemes requiring the imposition of capital punishment for select crimes by way of the mandatory operation of law. Woodson v North Carolina, 428 US 280, 286-287, 298; 96 S Ct 2978; 49 L Ed 2d 944 (1976). Those sentencing schemes were also challenged on Eighth Amendment grounds in Woodson, with the Court understanding the case as challenging not the state’s ability to impose capital punishment but “the procedure employed by the State to select persons for the ... penalty of death.” Id. at 287 (emphasis added).

In Woodson, the Court, in another 5-4 decision, held that those schemes were unconstitutional. The plurality opinion viewed as unconstitutional sentencing schemes that employed a process that did not permit for “the prevailing practice of individualizing sentencing determinations” as part of the process for imposing capital punishment. Id. at 303-304 (opinion of Stewart, Powell, and Stevens, JJ.). Accordingly, post-Woodson, capital punishment could only be constitutionally imposed after “consideration of the character and record of the individual offender and the circumstances of the particular offense ...Id. at 304. Notably, however, on the same day that the United States Supreme Court decided Woodson, it also declined to categorically bar the imposition of capital punishment. Gregg v Georgia, 428 US 153; 96 S Ct 2909; 49 L Ed 2d 859 (1976).

Following Woodson and Gregg, the United States Supreme Court confronted two additional cases challenging whether the sentencing procedures employed to impose capital punishment complied with Woodson's requirement of individualized sentencing determinations. See Lockett v Ohio, 438 US 586; 98 S Ct 2954; 57 L Ed 2d 973 (1978), and Eddings v Oklahoma, 455 US 104; 102 S Ct 869; 71 L Ed 2d 1 (1982). Both Lockett and Eddings were cited in Miller as part of the capital-punishment strand of precedent that culminated in Miller. Miller, 567 US at_; 132 S Ct at 2467. The plurality opinion in Lockett stated that statutory schemes authorizing capital punishment must permit the sentencer to consider all forms of mitigating evidence relating to two measuring points for determining the propriety of the sentence — evidence relating to the defendant’s “character or record and any of the circumstances of the offense . . . .” Lockett, 438 US at 604 (opinion by Burger, C.J.). Relevantly listed as factors that the sentencer must be permitted to consider were the defendant’s “role in the offense” and the defendant’s “age.” Id. at 608.

In Eddings, the Court, in a 5-4 decision, applied Lockett to a case in which the trial court, in considering mitigating factors before imposing capital punishment, declined to consider either the defendant’s family background, including the physical abuse and neglect he had suffered, or the fact that he suffered from an alleged “personality disorder.” Eddings, 455 US at 112-113. The Court ruled that while a sentencer may “determine the weight to be given relevant mitigating evidence,” the sentencer may not decide to give a piece of relevant mitigating evidence “no weight by [altogether] excluding such evidence from .. . consideration.” Id. at 114-115. Under Lockett and Eddings, in which individualized sentencing is required, not only must statutory procedures for imposing capital punishment permit the defendant to present all relevant mitigating evidence, but the sentencer must also consider and accord some weight to that evidence. Id. at 112-115.

2. JUVENILE-SENTENCING STRAND

The second strand of precedent was developed in two cases, Roper v Simmons, 543 US 551; 125 S Ct 1183; 161 L Ed 2d 1 (2005), and Graham. Roper and Graham were understood by the Court in Miller to have “establish[ed] that children are constitutionally different from adults for purposes of sentencing.” Miller, 567 US at_; 132 S Ct at 2464. This constitutional distinction has resulted in downward alterations in Roper and Graham in the range of punishments that the state may constitutionally impose on juvenile offenders. When the rules from Roper and Graham are considered together, a state may only impose a sentence of life without parole on a juvenile for the commission of an offense that if committed by an adult would constitutionally permit the state to punish the adult by capital punishment.

In Roper, the Court held that the “Eighth and Fourteenth Amendments forbid imposition of the death penalty on offenders who were under the age of 18 when their crimes were committed.” Roper, 543 US at 578. The Court characterized the rule it was adopting as a “categorical rule.” Id. at 572.⁵ The subsequent decision in Graham adopted what the Court again characterized as a “categorical rule,” i.e., that a sentence of life without parole could not be imposed on a juvenile nonhomicide offender. Graham, 560 US at 79. In reaching this conclusion, Graham drew comparisons between a capital sentence for an adult offender and a life-without-parole sentence for a juvenile offender. Id. at 69-70. To justify this categorical rule, the Court relied on the factors identified in Roper that assertedly distinguished juvenile and adult offenders. Id. at 68, citing Roper, 543 US at 569-570. The Court also supported its prohibition of life-without-parole sentences for juvenile nonhomicide offenders by concluding that the goals of punishment (retribution, deterrence, incapacitation, and rehabilitation) are not furthered when a nonparolable life sentence is imposed. Id. at 71-74. Combining strands of precedent that were previously limited to capital sentences and juvenile nonhomicide offenders respectively, and holding for the first time that these separate strands were relevant to noncapital sentences for juvenile homicide offenders, the United States Supreme Court reached its holding in Miller.

3. MILLER v ALABAMA

Miller v Alabama created the rule that Carp and Davis seek to have applied retroactively. Having identified what is required by the rules from each of the two strands of precedent that underlie Miller, we now identify what is required by the rule in Miller in order to determine whether Miller is more like the juvenile-sentencing strand whose rules have applied retroactively under Teague or more like the capital-punishment strand whose rules have not been applied retroactively under Teague. We compare and contrast the rule in Miller in this way because, as discussed later, the “form and effect” of a rule is essential in determining whether a rule is to be applied retroactively under Teague. One form of a rule will produce a single invariable result, or a single effect, when applied to any defendant in the class of defendants to whom the rule is pertinent. Another form of a rule will produce a range of results, or have multiple possible effects, when applied to different defendants in the class of defendants to whom the rule is pertinent. The form and effect of the rules derived from the capital-punishment strand of precedent varies considerably from the form and effect of the rules derived from the juvenile-sentencing strand of precedent, and this variance has markedly different consequences for the question of retroactivity. The capital-punishment strand of precedent prescribed rules that require a sentencer to perform an individualized sentencing analysis resulting in capital punishment being either imposed or not. By contrast, the juvenile-sentencing strand of precedent prescribed rules that categorically bar the imposition of a particular sentence, requiring the sentencer to impose a lesser sentence in every case. The former class of rules does not clearly satisfy the test for retroactivity, while the latter class of rules does. In assessing whether the form and effect of the rule in Miller is more akin to that of the capital-punishment strand of precedent, and therefore less clearly retroactive, or more akin to the juvenile-sentencing strand of precedent, and therefore more clearly retroactive, we find it important to examine what Miller itself stated about the form and effect of its own holding.

Miller held “that the Eighth Amendment forbids a sentencing scheme that mandates life in prison without possibility of parole for juvenile offenders.” Miller, 567 US at_; 132 S Ct at 2469. Within the very same paragraph in which Miller announced this holding, the Court also stated that its decision “require [s] [the sentencer] to take into account how children are different, and how those differences counsel against irrevocably sentencing them to a lifetime in prison.” Id. at _; 132 S Ct at 2469. Miller then provides substantial details regarding what must be considered as part of the individualized sentencing process before a sentence of life without parole can be imposed on a juvenile:

Mandatory life without parole for a juvenile precludes consideration of his chronological age and its hallmark features— among them, immaturity, impetuosity, and failure to appreciate risks and consequences. It prevents taking into account the family and home environment that surrounds him — and from which he cannot usually extricate himself — no matter how brutal or dysfunctional. It neglects the circumstances of the homicide offense, including the extent of his participation in the conduct and the way familial and peer pressures may have affected him. Indeed, it ignores that he might have been charged and convicted of a lesser offense if not for incompetencies associated with youth — for example, his inability to deal with police officers or prosecutors (including on a plea agreement) or his incapacity to assist his own attorneys. And finally, this mandatory punishment disregards the possibility of rehabilitation even when the circumstances most suggest it. [Id. at_; 132 S Ct at 2468 (citation omitted).]

Miller’s summarization of what the trial court must evaluate as part of the new individualized sentencing process tracks in large part the two measuring points about which a defendant must be allowed to present mitigating evidence within the capital-punishment context of Lockett — evidence relating to “the ‘circumstances of the particular offense and [to] the character and propensities of the offender.’ ” Id._n 9; 132 S Ct at at 2471 n 9, quoting Roberts v Louisiana, 428 US 325, 333; 96 S Ct 3001; 49 L Ed 2d 974 (1976) (opinion of Stewart, Powell, and Stevens, JJ.), and citing Sumner v Shuman, 483 US 66; 107 S Ct 2716; 97 L Ed 2d 56 (1987). Although the focus of the rule in Miller — life-without-parole sentences for juvenile offenders — is, of course, distinct from the focus of the rules in capital-punishment cases, the form and effect of the rule in Miller is quite similar to that of the rules in capital-punishment cases. That is, the rule in Miller requires a sentencer to perform an individualized sentencing analysis resulting in a life-without-parole sentence being either imposed or not, very much like the capital-punishment cases require a sentencer to perform an individualized sentencing analysis resulting in capital punishment being either imposed or not.

It is considerably more difficult to draw the same comparison between the rule in Miller and the categorical rules in Graham and Roper. Indeed, the United States Supreme Court itself specifically distinguished the form and effect of these rules:

Our decision does not categorically bar a penalty for a class of offenders or type of crime — as, for example, we did in Roper or Graham. Instead, it mandates only that a sentencer follow a certain process — considering an offender’s youth and attendant characteristics — before imposing a particular penalty. [Miller, 567 US at _; 132 S Ct at 2471.][⁶]

Thus, rather than relying on Graham and Roper to give form and effect to Miller, in the same manner as the capital-punishment decisions, the Court relied on Graham and Roper in Miller only for a generalized “principle” regarding juvenile offenders. Id. at_; 132 S Ct at 2471, 2472 n 11. That is, Miller relied on Graham and Roper for the general principle of law that juveniles possess different mental faculties than adults, so the United States Constitution requires that they be treated differently than adults for sentencing purposes with respect to the imposition of capital punishment and sentences of life without parole. Although this principle of law explains why the United States Supreme Court found it necessary to adopt the rule in Miller, it has no bearing on the actual form and effect of the rule adopted in Miller. Accordingly, because the form and effect of a rule rather than the principle underlying the rule’s formation controls whether the rule must be applied retroactively under federal retroactivity rules, whether Miller must be applied retroactively will center on whether a rule with a form and effect similar to the rules in Woodson, Lockett, and Ed-dings (rather than Roper and Graham) is the type of rule entitled to retroactive application under Teague.⁷ With this in mind, we next define Teague’s federal retroactivity test so as to determine whether the rule in Miller is entitled to retroactive application under that test.

B. FEDERAL RETROACTIVITY

1. GENERAL OVERVIEW

There is a “general rule of nonretroactivity for cases on collateral review” when it comes to applying new constitutional rules to cases that became final before the new rule was announced.⁸ Teague, 489 US at 307 (opinion by O’Connor, J). This default rule is driven by “the principle of finality which is essential to the operation of our criminal justice system.” Id. at 309. Supporting this same principle are concerns arising from the burdens placed on the administration of justice when new rules are applied retroactively, in that “[t]he ‘costs imposed upon the State[s] by retroactive application of new rules of constitutional law on [collateral review] generally far outweigh the benefits of this application.’ ”⁹ Id. at 310, quoting Solem v Stumes, 465 US 638, 654; 104 S Ct 1338; 79 L Ed 2d 579 (1984) (second alteration in original).

For this reason, the first inquiry in which a court must engage when determining whether a rule applies retroactively to cases presented on collateral review concerns whether the rule constitutes a “new rule” as defined by Teague, 489 US at 299-301 (opinion by O’Connor, J.), and Penry v Lynaugh, 492 US 302, 329; 109 S Ct 2934; 106 L Ed 2d 256 (1989). Saffle v Parks, 494 US 484, 487; 110 S Ct 1257; 108 L Ed 2d 415 (1990). Generally speaking, a rule is “new” if the rule announces a principle of law not previously articulated or recognized by the courts and therefore “falls outside [the] universe of federal law” in place at the time defendant’s conviction became final. Williams v Taylor, 529 US 362, 381; 120 S Ct 1495; 146 L Ed 2d 389 (2000) (opinion by Stevens, J.). If a rule is not deemed a “new rule,” then the general rule of nonretroactivity is inapplicable and the rule will be applied retroactively even to cases that became final for purposes of direct appellate review before the case on which the defendant relies for the rule was decided. Whorton v Bockting, 549 US 406, 416; 127 S Ct 1173; 167 L Ed 2d 1 (2007). If, however, a rule is deemed a “new rule,” then the general rule of nonretroactivity does apply. See Saffle, 494 US at 494.

When a rule is deemed a “new rule” and the general rule of nonretroactivity applies, a court must then engage in Teague’s second inquiry, to wit, whether the “new rule” satisfies one of Teague’s two exceptions to the general rule of nonretroactivity for new rules. See id. If the “new rule” satisfies either of Teague’s two exceptions, then it will be applied retroactively. Id. If, however, the “new rule” fails to satisfy either of those exceptions, the rule will only be entitled to prospective application. Id. Whorton succinctly summarized Teague’s two exceptions to the general rule of rionretroactivity as follows:

A new rule applies retroactively in a collateral proceeding only if (1) the rule is substantive or (2) the rule is a “ ‘watershed rul[e] of criminal procedure’ implicating the fundamental fairness and accuracy of the criminal proceeding.” [Whorton, 549 US at 416, quoting Saffle, 494 US at 495, quoting Teague, 489 US at 311 (opinion by O’Connor, J.) (alteration in original).]

2. “NEW RULE”

Turning to the first inquiry of the retroactivity analysis, whether the rule in Miller is “new,” we note that the United States Supreme Court has defined a rule as “new” when the rule “ ‘breaks new ground,’ ‘imposes a new obligation on the States or the Federal Government,’ or was not ‘dictated by precedent existing at the time the defendant’s conviction became final.’ ” Saffle, 494 US at 488, quoting Teague, 489 US at 301 (opinion by O’Connor, J.) (emphasis omitted). Essential to any of these bases for finding that a rule is “new” is the question of whether “all reasonable jurists would have deemed themselves compelled to accept” the rule at the time defendant’s conviction became final. Graham v Collins, 506 US 461, 477; 113 S Ct 892; 122 L Ed 2d 260 (1993) (emphasis added). The fact that a “decision is within the ‘logical compass’ of an earlier decision ... is not conclusive for purposes of deciding whether the current decision is a ‘new rule’ under Teague.” Butler v McKellar, 494 US 407, 415; 110 S Ct 1212; 108 L Ed 2d 347 (1990). In determining whether the rule in Miller is “new,” this Court inquires whether before Miller courts of this state, if presented with a constitutional challenge to our pre-Miller sentencing statutes, would have felt bound to declare those statutes unconstitutional for the reasons expressed in Miller.

It is apparent, in our judgment, that the rule in Miller constitutes a new rule. Miller imposed a hitherto-absent obligation on state and lower federal courts to conduct individualized sentencing hearings before imposing a sentence of life without parole on a juvenile homicide offender. As part of this process, a prosecutor seeking a life-without-parole sentence must now present evidence of aggravating factors relevant to the offender and the offense, juvenile defendants must be afforded the opportunity and the financial resources to present evidence of mitigating factors relevant to the offender and the offense, psychological and other evaluations relevant to the youthfulness and maturity of the defendants must be allowed, and courts must now embark upon the consideration of aggravating and mitigating evidence offered regarding juvenile defendants as a condition to imposing sentences that previously required no such consideration. It thus seems certain as a result of Miller that a considerable number of juvenile defendants who would previously have been sentenced to life without parole for the commission of homicide offenses will have a lesser sentence meted out. Under Teague and Saffle, these new obligations clearly render the rule in Miller a new rule. We are not aware of any statement of this Court by any justice before Miller that argued in support of, or anticipated, the constitutional requirements set forth in that decision. Unless every affirmation by this Court of a sentence of life without parole on a juvenile offender before Miller, including those that followed decisions such as Roper, Graham, Eddings, and Lockett, can be characterized as “unreasonable,” there cannot be serious argument that Miller did not define a “new rule.”

Although Miller may be “within the logical compass” of earlier decisions, and built upon their foundation, cases predating Miller can hardly be read as having “dictated” or “compelled” Miller’s result. Miller undoubtedly broke new ground in that it set forth the first constitutional rule to mandate individualized sentencing before noncapital punishment can be imposed. In this respect, the capital-punishment cases, although providing a model for the form and effect of Miller, would not have required a reasonable jurist to conclude that a life-without-parole sentence for a juvenile could only be constitutionally imposed following an individualized sentencing hearing.

Turning to the juvenile cases, Roper also dealt exclusively with the imposition of capital sentences without discussing the constitutionality of life-without-parole sentences and the need for individualized sentencing hearings. While Graham’s focus was on life-without-parole sentences, its constitutional rule was limited to nonhomicide offenses, and it did not make individualized sentencing the constitutional threshold for imposing a sentence of life without parole. Furthermore, while Graham drew a comparison between life-without-parole sentences for juvenile offenders and capital punishment, which was pivotal in deciding Miller, Graham also stopped well short of finding the two punishments equivalent. See Graham, 560 US at 69. This is evident by Graham’s reference to life without parole as “ ‘the second most severe penalty permitted by law,’ ” id., quoting Harmelin v Michigan, 501 US 957, 1001; 111 S Ct 2680; 115 L Ed 2d 836 (1991) (Kennedy, J., concurring in part), and its description of capital punishment as “ ‘unique in its severity and irrevocability,’ ” id., quoting Gregg, 428 US at 187 (emphasis added). Accordingly, although Roper and Graham could certainly be argued as being part of a longer-term movement toward application of the individualized sentencing capital-punishment cases to life-without-parole sentences for juvenile homicide offenders, Graham itself nowhere compelled or dictated this application. Since before Miller a court of this state could have reasonably rejected a constitutional challenge to Michigan’s preMiller sentencing scheme similar to that raised in Miller, Miller is clearly a “new rule.”

3. PROCEDURE VERSUS SUBSTANCE

Concluding that Miller announced a new rule, we turn to the second inquiry, whether the rule in Miller fits within one of Teague's two “narrow exceptions” to the general rule of nonretroactivity. Saffle, 494 US at 486. At the outset, we note that neither Carp nor Davis advanced any argument before this Court suggesting that Miller should be applied retroactively under the second exception, the “watershed rule of criminal procedure” exception. Accordingly, we consider any argument regarding Miller identifying a “watershed rule of criminal procedure” unpreserved, and we will only consider whether the rule in Miller fits within the first exception to the general rule of nonretroactivity.¹⁰

The first exception differentiates between new substantive rules and new procedural rules, allowing for the retroactive application of only the former. See Whorton, 549 US at 417; Schriro v Summerlin, 542 US 348, 351-352; 124 S Ct 2519; 159 L Ed 2d 442 (2004). The origin of the first exception predates Teague, as that decision drew the contours of this exception from Justice Harlan’s partial concurrence and partial dissent in Mackey v United States, 401 US 667; 91 S Ct 1160; 28 L Ed 2d 404 (1971). Teague, 489 US at 311 (opinion by O’Connor, J.). In speaking of the “general” rule against retroactive application of new constitutional rules, Justice Harlan commented that the Court’s

discussion is written only with new ‘procedural due process’ rules in mind, that is, those applications of the Constitution that forbid the Government to utilize certain techniques or processes in enforcing concededly valid societal proscriptions on individual behavior. New ‘substantive due process’ rules, that is, those that place, as a matter of constitutional interpretation, certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe, must, in my view, be placed on a different footing [and afforded retroactive application], [Mackey, 401 US at 692 (Harlan, J., concurring in the judgments in part and dissenting in part).]

Justice Harlan supported this differentiation by emphasizing that retroactive application of a substantive rule “represents the clearest instance where finality interests should yield” because “[t]here is little societal interest in permitting the criminal process to rest at a point where it ought properly never to repose.” Id. at 693. Contrasting the retroactive application of a substantive rule with that of a procedural rule, Justice Harlan proceeded to offer the observation that the retroactive application of a substantive rule “entails none of the adverse collateral consequences of retrial” certain to follow the retroactive application of a procedural rule. Id. This is because a substantive rule precludes the possibility of retrial given that its application dictates a single result for the class of individuals or type of conduct formerly regulated by the old rule and now governed by the new rule. It is in this sense that categorical rules, such as those derived from the juvenile-sentencing strand of precedent, are substantive because they have a “form and effect” that always results in the imposed punishment being unconstitutional, i.e., they produce a “single result.” Conversely, noncategorical rules, such as those derived from the capital-punishment strand of precedent — and Miller — are procedural because they have a “form and effect” that does not always result in the imposed punishment being unconstitutional, i.e., they do not produce a “single result.” The latter rules merely require a court to perform a new or amended analysis before it can be determined whether a given punishment can be imposed on a particular defendant.

Teague subsequently adopted Justice Harlan’s distinction between procedural and substantive rules, including the definition of when a rule is substantive. Teague, 489 US at 310-311 (opinion by O’Connor, J.). Since Teague, the United States Supreme Court has continued to recognize that the exceptions proposed by Justice Harlan in his opinion in Mackey were adopted in Teague. See, e.g., Danforth v Minnesota, 552 US 264, 273-275; 128 S Ct 1029; 169 L Ed 2d 859 (2008); Penry, 492 US at 329-330; see also Schriro, 542 US at 362 (Breyer, J., dissenting).

Although Teague addressed whether a new rule germane to the trial stage of a criminal case could be applied retroactively, later cases have addressed whether new rules pertaining only to punishments and the sentencing phase are substantive and fit into Teague’s first exception to the general rule of nonretroactivity. In so doing, the United States Supreme Court has provided three descriptions of what makes a new rule “substantive” within the context of a new rule governing the sentencing stage of a criminal case. Each of these, however, can be boiled down to whether the punishment imposed is one that the state has the authority to, and may constitutionally, impose on an individual within the pertinent class of defendants.

First, a new rule has been described as “substantive” when the rule “prohibit[s] a certain category of punishment for a class of defendants because of their status or offense.” Penry, 492 US at 330; see also Saffle, 494 US at 494-495. Put another way, the new rule is “substantive” when the punishment at issue is categorically barred. The requirement that the new rule be “categorical” in its prohibition is the direct product of how Justice Harlan’s first exception has been understood. That is, his first exception permits the retroactive application of “substantive categorical guarantees accorded by the Constitution, regardless of the procedures followed.” Penry, 492 US at 329 (emphasis added); see also Saffle, 494 US at 494.

Second, a new rule has been described as “substantive” if it “alters the range of conduct or the class of persons that the law punishes.” Schriro, 542 US at 353, citing Bousley v United States, 523 US 614, 620-621; 118 S Ct 1604; 140 L Ed 2d 828 (1998). The dissent contends that when a new rule “expand[s] the range of punishments” available to the sentencer, the rule fits within this second description of a new rule as substantive. Post at 545. Although a new rule could potentially be viewed as altering the range of punishments available to the sentencer when the rule makes a previously unavailable lesser punishment available to the sentencer, the United States Supreme Court has adopted a different definition for when a new rule “alters the range” of available punishments. We are bound to abide by that definition when considering the rule in Miller for federal retroactivity purposes. Under that definition, a new rule alters the “range of conduct” that the law can punish when it “placets] particular conduct or persons covered by the statute beyond the State’s power to punish.” Schriro, 542 US at 352 (emphasis added) (citations omitted). In this sense, the new rule transforms the conduct in which the defendant engaged, and which was previously within the state’s power to regulate, into conduct that is no longer subject to criminal regulation. Applied in the context of rules governing sentencing and punishment, it must be the case that under the previous rule, the defendant “faces a punishment that the law cannot [any more] impose upon him” in light of the new rule. Id. In this sense, a new rule only “alters the range” of punishments available to the sentencer if it shifts the upper limits of the range of punishments downward so that the previously most severe punishment to which defendants have been sentenced is no longer a punishment that the sentencer may constitutionally impose.¹¹

Third, a new rule has been described as “substantive” when it “narrow[s] the scope of a criminal statute by interpreting its terms . ...” Id. at 351, citing Bousley, 523 US at 620-621 (emphasis added). This third description addresses situations in which a criminal statute has previously been interpreted and applied beyond the statute’s intended scope so that the “defendant stands convicted of ‘an act that the law does not make criminal.’ ” Bousley, 523 US at 620, quoting Davis v United States, 417 US 333, 346; 94 S Ct 2298; 41 L Ed 2d 109 (1974).¹² Put another way, this description is implicated when a court, rather than a legislature, has criminalized conduct, authorized punishment, or construed a statute to apply more broadly than it is later deemed to apply. See id. at 620-621 (“For under our federal system it is only Congress, and not the courts, which can make conduct criminal.”). In this sense, the state cannot constitutionally impose the punishment at issue because the new rule determines that no lawfully enacted statute has given the state the authority to impose such a punishment.

In distinguishing what makes a new rule substantive, the United States Supreme Court has also afforded considerable direction regarding the qualities and contours of nonsubstantive, or procedural, rules. Simply put, “rules that regulate only the manner of determining the defendant’s culpability are procedural.” Schriro, 542 US at 353. This is because a rule that alters the “manner of determining” culpability “merely raise[s] the possibility that someone convicted with use of the invalidated procedure might have been acquitted otherwise.” Id. at 352. Applying this understanding to new rules governing sentences and punishments, a new procedural rule creates the possibility that the defendant would have received a less severe punishment but does not necessitate such a result. Accordingly, a rule is procedural when it affects how and under what framework a punishment may be imposed but leaves intact the state’s fundamental legal authority to seek the imposition of the punishment on a defendant currently subject to the punishment.

Turning to how the United States Supreme Court has applied this distinction between substantive and procedural rules, in Schriro the Court was confronted with whether the new rule from Ring v Arizona, 536 US 584; 122 S Ct 2428; 153 L Ed 2d 556 (2002), was substantive or procedural. Ring’s rule invalidated Arizona’s capital-punishment sentencing scheme and required that a jury rather than a judge make the determination whether aggravating factors necessary for the imposition of capital punishment had been proved. Id. at 609. Despite the fact that Ring invalidated Arizona’s statutory sentencing scheme authorizing capital punishment, its rule was ultimately deemed “procedural” on the basis that it

did not alter the range of conduct Arizona law subjected to the death penalty. ... Instead, Ring altered the range of permissible methods for determining whether a defendant’s conduct is punishable by death, requiring that a jury rather than a judge find the essential facts bearing on punishment. Rules that allocate decisionmaking authority in this fashion are prototypical procedural rules, a conclusion we have reached in numerous other contexts. [Schriro, 542 US at 353.]

In Saffle, the Court similarly deemed a new rule “procedural” when it would have prohibited anti-sympathy instructions to juries performing the individualized sentencing process as a condition to imposing capital punishment. See Saffle, 494 US at 486. In doing so, Saffle stated that the rule “would neither decriminalize a class of conduct nor prohibit the imposition of capital punishment on a particular class of persons.” Id. at 495. It is with Schriro and Saffle in mind that we turn to the question of whether the rule in Miller is properly viewed as substantive or procedural.

Although the new procedures required by Miller may be more elaborate and detailed than the new procedures at issue in Schriro and Saffle, the basic form and effect is the same. As discussed earlier, Miller requires that the trial court “follow a certain process” before it can impose a sentence of life without parole on a juvenile homicide offender. Miller, 567 US at_; 132 S Ct at 2471. Miller, however, specifically “does not categorically bar a penalty for a class of offenders or type of crime[.]” Id. at_; 132 S Ct at 2471.

Considering Miller’s self-description of its rule, it is clear that the rule is not substantive within the terms of the first description of when a rule is substantive, i.e., when the rule “prohibits] a certain category of punishment for a class of defendants because of their status or offense.” Penry, 492 US at 330; see also Saffle, 494 US at 494. The category of punishment implicated by Miller is a sentence of “life without parole,”¹³ the class of defendants receiving the benefit of Miller are juvenile defendants who are under the age of 18 at the time they commit their offenses, and the types of offenses implicated by Miller are homicide offenses. Accordingly, for Miller to be considered “substantive” under the first description of when a rule is substantive, it must prohibit sentences of life without parole for juvenile offenders under the age of 18 who are convicted of homicide offenses, and clearly Miller does no such thing. Instead, as with the procedural rules in Schriro and Saffle, and the rules from the capital-punishment cases of Woodson, Lockett, and Eddings, Miller creates only the possibility that a defendant may have received a lesser punishment had the trial court employed the new process that is constitutionally required by Miller.

The second description of when a rule is substantive is equally of no avail to Carp and Davis because a rule is substantive under that description only when it alters the range of punishments that a state is permitted to impose by foreclosing the state’s ability to impose the punishment defendant is serving. See Schriro, 542 US at 353. In this sense, a rule is only substantive if it acts to ratchet down the previously most severe punishment possible. Conversely, and contrary to the dissent, a rule will be considered procedural if it merely expands the range of possible punishments that may be imposed on the defendant. Applied to Michigan’s sentencing scheme, Miller now requires the sentencer to consider imposing a sentence of life with the possibility of parole, but it does not require the sentencer to exclude from consideration a sentence of life without parole. Accordingly, Miller does not remove the punishment imposed on Carp and Davis from within the range of punishments the state has the power to impose. Accordingly, the rule in Miller again cannot be viewed as substantive under the second United States Supreme Court description.

The third description of when a rule is substantive is altogether inapplicable to Miller. The decision did not rest on any principle of statutory interpretation, and it did not pertain to a situation in which life-without-parole sentences were being imposed on juvenile homicide offenders absent clear statutory authority to do so. Just as Carp and Davis were sentenced to life without parole in full accordance with Michigan’s statutory sentencing scheme, Miller was sentenced to life without parole in full accordance with Alabama’s statutory sentencing scheme. See Miller, 567 US at_; 132 S Ct at 2462-2643.

Ultimately, the rule in Miller is procedural because, as with the rule in Ring, it merely shifts “decisionmaking authority” for the imposition of a life-without-parole sentence on a juvenile homicide offender.¹⁴ Schriro, 542 US at 353. Whereas Ring shifted decision-making authority for imposing capital punishment from the judge to the jury, Miller shifted decision-making authority for imposing a sentence of life without parole on a juvenile homicide offender from the legislature to the judiciary, by way of its individualized sentencing requirements.¹⁵ Although the process set forth in Miller is undoubtedly more favorable to juvenile homicide defendants as a class, the new process has no effect on Michigan’s inherent authority to lawfully and constitutionally seek the imposition of a life-without-parole sentence on any and every given juvenile homicide offender. Just as no court may impose a sentence of life without parole without conducting an individualized consideration of certain factors, no court relying on Miller may categorically refuse to impose a sentence of life without parole if the individualized sentencing factors do not operate in a defendant’s favor. Accordingly, in contrast to a substantive rule that avoids the adverse collateral consequences of retrial by dictating a singular result, Mackey, 401 US at 693 (Harlan, J., concurring in the judgments in part and dissenting in part), retroactive application of Miller necessarily requires this adverse collateral consequence. In this regard, the rule in Miller in no reasonable way can be said to “represento the clearest instance where finality interests should yield.” Id. (emphasis added). Because Miller continues to permit Michigan to impose a life-without-parole sentence on any juvenile homicide offender (but only after individualized consideration), it must necessarily be viewed as procedural rather than substantive. Therefore, we hold that the rule in Miller does not satisfy the first exception to the general rule of nonretroactivity in Teague.

An additional consideration serves to strengthen this conclusion. In its description of the rule in Miller, the articulation employed by the United States Supreme Court is telling. Teague’s retroactivity analysis distinguishing substantive and procedural rules is in no sense new or novel. Rather, the proposition that “substantive categorical guarantees” should receive retroactive application while “procedural noncategorical guarantees” should only receive prospective application predates Teague. See Penry, 492 US at 329. In the face of this reasonably well-defined and longstanding distinction, Miller, in describing the nature and scope of its rule, repeatedly employs language typically associated with nonretroactive procedural rules. Although fully recognizing that Roper and Graham announced “categorical” bars, Miller twice states that its rule does not create a “categorical” bar. Miller, 567 US at_; 132 S Ct at 2469, 2471. Furthermore, Miller, in straightforward terms, speaks of its rule as one that “mandates only that a sentencer follow a certain process[.Y Id. at_; 132 S Ct at 2471 (emphasis added). It is hard to view these statements as anything other than expressions of continuity in the Court’s understanding of the law of retroactivity, particularly in a circumstance in which the four justices of the Supreme Court who were presumably the least inclined to extend Miller to a broader range of cases — the dissenting justices who had rejected the new rule in the first place — -were absent from the majority opinion.¹⁶

Carp advances three arguments in an effort to overcome our conclusion that Miller does not qualify for retroactive application under Teague. First, he argues that each of the strands of precedent that underlie Miller has been granted retroactive status. While there may be considerable force to the argument that categorical rules like those in Roper and Graham must be applied retroactively under Teague, the same cannot be said for the strand of cases requiring individualized sentencing before capital punishment can be imposed on an adult offender. Despite considerable effort by Carp, including post-oral-argument supplemental briefings, we remain unpersuaded that the United States Supreme Court, or even any federal court of appeals,¹⁷ has declared any of the individualized sentencing capital-punishment cases retroactive under Teague.

In an effort to demonstrate to the contrary, Carp principally cites Sumner, in which the United States Supreme Court held that individualized sentencing was required before capital punishment could be imposed on a defendant, Shuman, who was serving a life-without-parole sentence at the time he committed the capital offense. Sumner, 483 US at 80-81. Carp is correct that Sumner relied on Woodson in creating its rule, id. at 70-75, and is also correct that Sumner involved the review of a state conviction on collateral habeas review, see id. at 68. However, not all cases presenting themselves on collateral review are equivalent for retroactivity purposes. Some cases on collateral review assert that state courts failed to properly apply constitutional rules in effect before the defendant’s conviction became final, while others seek the application or creation of a new rule that was not announced before the defendant’s conviction became final.

If, with respect to the application of Woodson, Sumner fell into the latter category, then we might agree with Carp that Woodson had been applied retroactively. Sumner, as it relates to the application of Woodson, however, falls into the former category of cases presenting themselves on collateral review. Woodson was decided on July 2, 1976, and Shuman’s conviction did not become final for direct review purposes until May 17, 1978, nearly two years after Woodson was decided. See Shuman v State, 94 Nev 265; 578 P2d 1183 (1978). Accordingly, to the extent that Woodson was applied in Sumner, it was simply not applied retroactively to a case that had become final for direct review purposes before Woodson was issued.¹⁸

Apparently anticipating these flaws in the argument that Woodson has been applied retroactively, Carp contends that Sumner itself has been applied retroactively post-Teague. For this proposition, he cites Thigpen v Thigpen, 926 F2d 1003, 1005 (CA 11, 1991). We, however, do not read Thigpen as addressing the question of Sumner's retroactivity. Although the district court below had applied Sumner retroactively to invalidate Thigpen’s sentence, that portion of the district court’s ruling was never appealed and the only issue before the United States Court of Appeals for the Eleventh Circuit was Thigpen’s appeal concerning whether the district court had erred by upholding his conviction. See id.¹⁹

Accordingly, Carp has not succeeded in demonstrating that any of the individualized sentencing capital-punishment cases, i.e., Furman, Woodson, Lockett, Ed-dings, or Sumner, have been applied retroactively under Teague. This failure is pivotal given our earlier conclusion that the rule in Miller is of the same form and effect as the rules in the individualized sentencing capital-punishment cases.

Second, Carp argues that Miller has added “age” and “incorrigibility” as elements of what must be assessed before a life-without-parole sentence can be imposed on a juvenile offender. Carp argues that it follows from this that age and the juvenile offender’s incorrigibility are aggravating factors that raise the mandatory minimum sentence that a defendant could receive under Michigan’s pre-Miller sentencing scheme because they must now be shown by the state before a juvenile offender can be sentenced pursuant to MCL 750.316(1) and MCL 791.234(6). Citing Alleyne v United States, 570 US_; 133 S Ct 2151, 2155; 186 L Ed 2d 314 (2013), Carp notes that “any fact that increases the mandatory minimum is an ‘element’ that must be submitted to the jury.” Accordingly, he argues that the rule in Miller must be viewed as substantive and applied retroactively when it is considered in light of Alleyne because Miller combined with Alleyne substantively alters the way Michigan law defines and sentences juvenile homicide offenders.

Even assuming for the sake of argument that Miller made assessments of “age” and “incorrigibility” necessary elements for imposing a life-without-parole sentence on a juvenile homicide offender, Carp’s argument still fails.²⁰ This is because his argument relies on the new rule adopted in Alleyne and therefore Alleyne itself would need to qualify for retroactive application to have any bearing on the instant case. Carp, however, has failed to even argue, much less persuade this Court, that Alleyne established a substantive rule entitled to retroactive application under Teague. Absent being so persuaded, we treat the rule in Alleyne as a procedural rule entitled only to prospective application.²¹ Accordingly, to the extent that we view Alleyne as establishing a nonretroactive procedural rule, Alleyne may not be bootstrapped onto the rule in Miller to transform the latter from a nonretroactive procedural rule into a retroactive substantive rule.

Third, Carp cites Miller’s companion case of Jackson v Hobbs as evidence that Miller has already been accorded retroactive status, and therefore presumably that the present judicial exercise has been rendered unnecessary. In offering this argument, Carp is correct that Jackson presented itself on collateral review and that the case was remanded for resentencing pursuant to the rule announced in Miller. Miller, 567 US at_; 132 S Ct at 2475. Accordingly, Carp also correctly notes that Jackson received retroactive relief under Miller. Id. at_; 132 S Ct at 2475. That being said, the fact that Jackson received the benefit of Miller being applied retroactively does not lead to the conclusion that Miller must be applied retroactively to any other defendant. This is because the assertion that a rule is nonretroactive is an “affirmative defense,” available to a prosecutor in objection to collateral relief being sought by a defendant. Thompson v Runnels, 705 F3d 1089, 1099 (CA 9, 2013) (noting that Caspari v Bohlen, 510 US 383, 389; 114 S Ct 948; 127 L Ed 2d 236 (1994) held that “ ‘a federal court may, but need not, decline to apply Teague if the State does not argue it,’ but ‘if the State does argue that the defendant seeks the benefit of a new rule of constitutional law, the court must apply Teague before considering the merits of the claim’ ”). As such, the nonretroactivity argument must be affirmatively raised by the state and when it is not raised, it is waived:

Since a State can waive the Teague bar by not raising it, and since the propriety of reaching the merits of a dispute is an important consideration in deciding whether or not to grant certiorari, the State’s omission of any Teague defense at the petition stage is significant. Although we undoubtedly have the discretion to reach the State’s Teague argument, we will not do so in these circumstances. [Schiro v Farley, 510 US 222, 229; 114 S Ct 783; 127 L Ed 2d 47 (1994) (citation omitted).]

In this sense, a defense premised on the nonretroactivity of a new rule is “not ‘jurisdictional’ ” in nature, and the court does not have any duty sua sponte to conduct a retroactivity analysis. Collins v Youngblood, 497 US 37, 41; 110 S Ct 2715; 111 L Ed 2d 30 (1990). Rather, because the question of retroactivity is “grounded in important considerations of federal-state relations,” a state is free to “[choose] not to rely on Teague” without the federal courts’ invalidating that choice. Id. By opting not to raise the defense in Jackson, the defense was waived and the question whether Miller should be applied retroactively was never presented to the United States Supreme Court.²²

Carp, however, contends that “principles of evenhanded justice” dictate that the rule in Miller be applied retroactively in his case since it was applied retroactively in Jackson’s case. He draws his argument from Teague, wherein the United States Supreme Court stated;

We can simply refuse to announce a new rule in a given case unless the rule would be applied retroactively to the defendant in the case and to all others similarly situated. ... We think this approach is a sound one. Not only does it eliminate any problems of rendering advisory opinions, it also avoids the inequity resulting from the uneven application of new rules to similarly situated defendants. We therefore hold that, implicit in the retroactivity approach we adopt today, is the principle that habeas corpus cannot be used as a vehicle to create new constitutional rules of criminal procedure unless those rules would be applied retroactively to all defendants on collateral review .... [Teague, 489 US at 316 (opinion by O’Connor, J.) (all but last emphasis added).]

As evidenced by the very quotation on which Carp relies, the application of the “principles of even-handed justice” only become relevant when the United States Supreme Court has actually undertaken a retroactivity analysis in the course of announcing a new rule. If no such analysis is necessary because of the posture of the case, as here, the Court will obviously not have the occasion to consider whether the new rule can be applied retroactively to all defendants who are situated similarly to the defendant before the Court.²³ Under those circumstances, the idiosyncrasies, strategies, or policies and practices of a single prosecutor, among more than 3,000 throughout the country, cannot possibly be allowed under our system of federalism to determine what “even-handed justice” requires (and what the law does or does not command) of all prosecutors in every jurisdiction throughout the country.²⁴

Having concluded that Miller established a new procedural rule that does not “categorically bar a penalty,” but instead requires “only that a sentencer follow a certain process,” Miller, 567 US at_; 132 S Ct at 2471, and having rejected the arguments in support of the retroactive application of Miller, we hold that the United States Supreme Court’s decision in that case does not require retroactive application under Teague. In light of this holding, we now turn to whether Miller is entitled to retroactive application under Michigan’s separate test for retroactivity.

C. STATE RETROACTIVITY

Although states must apply a new rule of criminal procedure retroactively when the new rule satisfies Teague’s exceptions to the general rule of nonretroactivity, they are permitted to “give broader retroactive effect” to a new rule than is required by Teague. Danforth, 552 US at 288-289. In this sense, Teague provides a floor for when a new rule of criminal procedure must be applied retroactively, with a state nonetheless free to adopt its own broader test for requiring the retroactive application of a new federal or state constitutional rule. See id. at 289-290.

Michigan has adopted its own separate test for when a new rule of criminal procedure should be applied retroactively. See Maxson, 482 Mich at 392-393. Michigan’s test for retroactivity was originally derived from the pre-Teague federal test set forth in Linkletter v Walker, 381 US 618; 85 S Ct 1731; 14 L Ed 2d 601 (1965). See People v Hampton, 384 Mich 669, 674; 187 NW2d 404 (1971).

Despite Michigan’s having adopted its own retroactivity test that may give broader retroactive effect to some new rules than is mandated by the Teague test, Michigan nonetheless still adheres to the general principle of nonretroactivity for new rules of criminal procedure.²⁵ As a result, “Michigan law has regularly declined to apply new rules of criminal procedure to cases in which a defendant’s conviction has become final.” Maxson, 482 Mich at 392-393 (citing several examples of new rules of criminal procedure that this Court declined to apply retroactively under its version of the Linkletter test). With Michigan’s predisposition against the retroactive application of new rules of criminal procedure firmly in mind — in that only the extraordinary new rule of criminal procedure will be applied retroactively under Michigan’s test when retro-activity is not already mandated under Teague — we proceed to evaluate whether the rule in Miller satisfies this state test.

Michigan’s test for retroactivity consists of three factors:

“(1) the purpose of the new rule[]; (2) the general reliance on the old rule[;] and (3) the effect of retroactive application of the new rule on the administration of justice.” [Maxson, 482 Mich at 393, quoting Sexton, 458 Mich at 60-61, citing Hampton, 384 Mich at 674 (second alteration in original).]

The first factor, the purpose factor, assesses the nature and focus of the new rule and the effect the rule is designed to have on the implementation of justice. See People v Young, 410 Mich 363, 366-367; 301 NW2d 803 (1981). Under this first factor, when a new rule “concerns the ascertainment of guilt or innocence, retroactive application may be appropriate.” Id. at 367, citing Hampton, 384 Mich 669 (emphasis added). Conversely, “[w]hen the ascertainment of guilt or innocence is not at stake, prospective application is possible” because “the purposes of the rule can be effectuated by prospective application.” People v Markham, 397 Mich 530, 535; 245 NW2d 41 (1976). Consistent with this standard for when a rule should be applied only prospectively, “a new rule of procedure . . . which does not affect the integrity of the fact-finding process should be given [only] prospective effect.” Young, 410 Mich at 367.

Carp contends that Miller, although not impheating his guilt or innocence, nonetheless, goes to the “integrity of the fact-finding process” because it is essential to evaluating a defendant’s level of culpability when imposing a sentence. In support of this contention, he cites McConnell v Rhay, 393 US 2, 3-4; 89 S Ct 32; 21 L Ed 2d 2 (1968), in which pursuant to Linkletter, the United States Supreme Court retroactively applied a new rule of criminal procedure despite the new rule’s being relevant only to the sentencing phase.²⁶ As Carp correctly observes, McConnell, in effecting its proretroactivity holding, stated that “the right being asserted relates to ‘the very integrity of the fact-finding process.’ ” Id. at 3, quoting Linkletter, 381 US at 639.

Two considerations, however, leave us unpersuaded that this remark necessitates the conclusion that the first factor of Michigan’s test favors the retroactive application of Miller. First, the new rule applied retroactively in McConnell addressed the right to counsel, a right with unique significance both within the context of the criminal proceeding²⁷ and within the context of the United States Supreme Court’s retroactivity jurisprudence.²⁸ Given this extraordinary footing of the right to counsel, we read McConnell’s statement that “the right being asserted relates to ‘the very integrity of the fact-finding process’ ” as concerning specifically the right to counsel rather than all new rules that may expand the fact-finding process at sentencing. For this reason, we do not understand McConnell as necessitating the view that, for retroactivity purposes under the Linkletter test, rules implicating the fact-finding process at sentencing must be placed on equal footing with rules implicating the fact-finding process for guilt or innocence.

Second, even if McConnell supported the expansive view that Carp attributes to it, that view is contrary to how Michigan law describes its own application of the Linkletter test. In every case to date in which this Court has applied the state retroactivity test, the “integrity of the fact-finding process” has always been referred to in the context of determining a defendant’s “guilt or innocence.” Maxson, 482 Mich at 393-394; Sexton, 458 Mich at 62; Young, 410 Mich at 367. To the extent that McConnell may have viewed the “fact-finding process” as continuing throughout sentencing, we respectfully disagree and decline to adopt such an expansive view for purposes of our separate and independent test for retroactivity. It reflects an understanding of retroactivity that is no longer subscribed to by the United States Supreme Court and an understanding to which this Court has never subscribed. There is utterly no obligation on our part to forever maintain the Linkletter test in accordance with every past federal understanding when the test is now defunct for federal purposes and this Court, although initially relying on Linkletter to formulate our state test for retroactivity, has added its own interpretations to that test. Instead, the general principle of nonretroactivity for new rules of criminal procedure, to which Michigan adheres and which informs this state’s retroactivity analysis, is properly served, in our judgment, by applying retroactively only those new rules of procedure that implicate the guilt or innocence of a defendant. We acknowledge that there are circumstances in which our state test may sometimes apply a new rule retroactively in circumstances in which Teague would not apply, but we are not prepared to extend our test beyond the federal test to the degree urged upon us by Carp.

In declining to expand the scope of the first factor of Michigan’s state test for retroactivity, we note again that although our state test is derived from Linkletter, nothing requires this Court to adopt each and every articulation of that test — one that is no longer adhered to by the United States Supreme Court itself. Our state test for retroactivity is supplemental to the current federal test set forth in Teague, and it is separate and independent of the former federal test set forth in Linkletter. See Danforth, 552 US at 289. As the Teague test replaced the Linkletter test for federal purposes, doubtlessly contracting the universe of new constitutional rules that will be applied retroactively,²⁹ it should be unsurprising that this Court would decline to grant retroactive status to a new rule of criminal procedure affecting only the sentencing phase of a criminal case when such a permutation of the defunct test has never before been so applied in this state.³⁰

From our holding that the first factor of our state test for retroactivity focuses on whether a new rule of procedure implicates a defendant’s guilt or innocence, it is apparent that the first factor clearly militates against the retroactive application of Miller. As Miller alters only the process by which a court must determine a defendant’s level of moral culpability for purposes of sentencing, it has no bearing on the defendant’s legal culpability for the offense of which the defendant has been duly convicted.

In light then of our conclusion that the first state factor clearly counsels against the retroactive application of Miller, we find it relevant here to address the interplay between the three factors of the test and the weight that must be given to each before we determine the effect of the second and third factors on Miller’s retroactive application. That a test consists of multiple factors does not logically signify that equal weight must be given to each. The United States Supreme Court, in applying the Linkletter test before it adopted the Teague test, observed that the second and third factors “have been regarded as having controlling significance ‘only when the purpose of the rule in question did not clearly favor either retroactivity or prospectivity.’ ” Michigan v Payne, 412 US 47, 55; 93 S Ct 1966; 36 L Ed 2d 736 (1973), quoting Desist v United States, 394 US 244, 251; 89 S Ct 1030; 22 L Ed 2d 248 (1969). Deductively from this statement, if two of the three factors only control when the first factor does not “clearly favor” retroactivity or prospectivity, it follows that the first factor must be afforded more weight than either of the other two factors when the first factor does “clearly favor” retroactivity or prospectivity. We are persuaded by, and adhere to, Payne’s and Desist’s understanding regarding the heightened weight to be afforded the first factor when it strongly supports one side or the other of the retroactivity question.

Placing such an emphasis on the first factor is fully consistent with this Court’s longstanding practice of dealing with the second and third factors “together.” Young, 410 Mich at 367; Hampton, 384 Mich at 677. In this sense, the second and third factors will generally tend to produce a unified result that either favors or disfavors retroactivity. This is because the subject of the second factor (general reliance on the old rule) “will often have a profound effect on” the subject of the third factor (administration of justice), given that the greater the reliance by prosecutors of this state on a rule in pursuing justice, the more burdensome it will generally be for the judiciary to undo the administration of that rule. Sexton, 458 Mich at 63-64; see also Hampton, 384 Mich at 677-678. In light of the weight to be afforded the first factor when it clearly preponderates against retroactive application, our unified consideration of the second and third factors would need to favor retroactive application to a substantial degree in order for Miller to satisfy the requirements for retroactive application under our state test.

Turning to the inquiry required to evaluate the second and third factors “together,” the second factor — the reliance on the old rule — must be considered both from the perspective of prosecutors across the state when prosecutors faithfully abided by the constitutional guarantees in place at the time of a defendant’s conviction, see Adams v Illinois, 405 US 278, 283-284; 92 S Ct 916; 31 L Ed 2d 202 (1972), and Johnson v New Jersey, 384 US 719, 731; 86 S Ct 1772; 16 L Ed 2d 882 (1996), as well as from the collective perspective of the 334 defendants who would be entitled to resentencing if the new rule were applied retroactively, seeMaxson, 482 Mich at 394. Inherent in the question of reliance by prosecutors across the state is the extent to which the old rule received constitutional approval from the judiciary before the adoption of the new rule. See Tehan v United States ex rel Shott, 382 US 406, 417; 86 S Ct 459; 15 L Ed 2d 453 (1966). When the old rule is merely the result of a “negative implication” drawn by prosecutors, the prosecutors’ good-faith reliance on the old rule is at its most minimal. Brown v Louisiana, 447 US 323, 335; 100 S Ct 2214; 65 L Ed 2d 159 (1980) (opinion by Brennan, J.). Similarly, when the old rule was of “doubtful constitutionality,” the ability of prosecutors across the state to rely on the old rule in good faith is diminished. Id. Conversely, when the old rule has been specifically approved by the courts as passing constitutional muster, prosecutors have their strongest argument for having relied on the old rule in good faith. Tehan, 382 US at 417. Moreover, when prosecutors relied in good faith on the old rule and did so for a lengthier period of time, reliance can be viewed as more significant and the second factor will tend to counsel against retroactive application. Id. As for defendants’ reliance on the old rule, they must demonstrate not only that they relied on the old rule by taking or not taking a specific action, but that they “detrimentally relied on the old rule.” Maxson, 482 Mich at 394 (emphasis added).

The inquiry into reliance will significantly affect any inquiry into the burden placed on the administration of justice because when prosecutors have relied on the old rule, they have presumably taken few, if any, steps to comply with the new rule. The greater the extent of their reliance, and the greater the extent to which the new rule constitutes a departure from the old rule, the more burdensome it becomes for prosecutors to take the steps necessary to comply with the new rule. Similarly, the greater the extent of the departure, the more difficult it becomes for courts to look back and attempt to reconstruct what outcome would have resulted had the new rule governed at the time a given defendant was sentenced. A burden is placed on the administration of justice in the form of time and expense to the judiciary in retroactively accommodating the new rule. Far more importantly, when a new rule is likely to be difficult to apply retroactively, a burden is placed on the administration of justice in the form of compromising the accuracy with which the new rule can be applied and the confidence the public may have regarding judicial determinations in situations in which the new rule is applied to cases that became final many years or even decades earlier.

Applying these considerations in evaluating the second and third factors to Miller, it is apparent that these factors do not sufficiently favor the retroactive application of Miller so as to overcome the first factor’s clear direction against its retroactive application. The old rule permitting life-without-parole sentences on the basis of the pre-Miller sentencing scheme established by the Legislature received in 1996 the specific approval of its constitutionality by our judiciary. Launsburry, 217 Mich App at 363-365. Further, nothing in United States Supreme Court caselaw called into any question life-without-parole sentences for any juvenile offenders until Graham was decided in 2010, and even then Graham was specifically limited in its breadth to juveniles who committed nonhomicide offenses.³¹ Graham, 560 US at 82. Indeed, before Roper in 2005, United States Supreme Court precedent specifically held that it was constitutional to impose capital punishment on juveniles over the age of 16 convicted of homicide offenses. Stanford v Kentucky, 492 US 361, 380; 109 S Ct 2969; 106 L Ed 2d 306 (1989). Accordingly, at the time prosecutors across Michigan sought life-without-parole sentences for 302 of the 334 defendants who would gain a resentencing hearing if Miller were applied retroactively, the Eighth Amendment of the United States Constitution was affirmatively understood as permitting the imposition of not merely life without parole but also the imposition of capital punishment on juvenile first-degree-murder offenders.³²

On the basis of this state of the law, prosecutors across Michigan entirely in good faith relied on the old rule whenever they sought hfe-without-parole sentences for juvenile homicide offenders. Considering the constitutional approval the old rule received from both our judiciary and the United States Supreme Court, as well as the length of time during which the old rule prevailed — dating back to our state’s founding in 1837 — the reliance on the old rule by Michigan prosecutors was significant and justified.³³

Conversely, we note that this is not a situation in which it can fairly be said that, as a group, the 334 defendants who would be entitled to resentencing if the rule in Miller were applied retroactively have “relied” on the old rule to their “detriment.” First, we find it difficult to understand, and Carp and Davis themselves fail to identify, exactly what adverse action the 334 defendants have taken, or opted not to take, in “reliance” on the old rule (except perhaps to recognize and abide by the old rule as the then extant law of this state).³⁴ If such “reliance,” in the sense of merely having to comply with the then extant law, is viewed as sufficiently “detrimental” to satisfy the second state retroactivity factor, then it would almost always be the case that this factor would weigh heavily in favor of retroactivity, since it must be assumed that criminal defendants, or at least their counsel, would almost always rely on existing law in formulating their trial and appellate strategies. There is nothing “detrimental” about that reliance except that the law is not as hospitable to the interests of such defendants as they might like it to be. That the law might have been destined to become more hospitable in the future is of little relevance since it is only because of that development that the issue of retroactivity has arisen in the first place.

Second, even to the extent that any defendants can be said to have taken or foregone some action to their detriment in reliance on the old rule, they still can only be said to have “detrimentally” relied on the old rule if they can establish that they would have obtained a result more favorable to them under the new rule. Maxson, 482 Mich at 394-396. In this sense, defendants can only be said to have “ ‘detrimentally relied’ on the old rule” if they “suffered actual harm from [their] reliance . . . .” Id. at 396. However, a majority of the 334 defendants who would receive resentencing hearings if the rule in Miller were applied retroactively were between 17 and 18 years of age when they committed their homicide offenses. Because Miller requires a sentencing court to give specific consideration to the age and the mental development of a juvenile offender before imposing a sentence of life without parole, when a juvenile most closely approaches the age of majority at the time the juvenile commits a homicide offense, Miller would seem least likely to counsel in favor of sentencing that juvenile with special leniency, given that in only as few as several months the juvenile would be ineligible for any leniency at all.³⁵ In this sense, it is speculative at best to presume that a majority of Michigan’s juvenile offenders serving life-without-parole sentences would gain relief in the form of a lesser sentence if they received a resentencing hearing pursuant to the retroactive application of Miller. Accordingly, juvenile defendants, as a class, are unable to demonstrate with any certainty under the state test that they detrimentally relied on the old rule to such an extent as to outweigh the state’s reliance on the old rule.

As between defendants and the prosecutors of this state, it is further apparent that the latter have relied far more heavily on the old rule, have done so in good faith, and would have relied “detrimentally” on behalf of the people were Miller to be applied retroactively. In particular, in relying on the old rule, prosecutors did not for the purpose of sentencing have any cause at the time to investigate or present evidence concerning the aggravating or mitigating factors now required to be considered by Miller. If Miller were to be applied retroactively, prosecutors would be abruptly required to bear the considerable expense of having to investigate the nature of the offense and the character of the 334 juvenile offenders subject to Miller’s retroactive application. This task, if newly thrust upon prosecutors, would be all the more burdensome and complicated because a majority of the 334 defendants were sentenced more than 20 years ago and another 25% were sentenced between 15 and 20 years ago. And in many, if not most, of those instances, the prosecutor who initially tried the case would likely no longer be available for a resentencing hearing. That is, Miller makes many things relevant to the sentencing process that were simply not relevant at the time of the initial sentencing, and these things would have to be reconstructed, almost impossibly so in some cases, after many years, in order to sustain a criminal sentence that was viewed at the time as the culmination of a full and fair process by which justice was obtained in cases of first-degree murder. There would be considerable financial, logistical, and practical barriers placed on prosecutors to re-create or relocate evidence that had previously been viewed as irrelevant and unnecessary. This process would not, in our judgment, further the achievement of justice under the law because it would require in many instances that the impossible be done, and if it could not be, a heavy cost would be incurred by society in the form of the premature release of large numbers of persons who will not have fully paid their legal debt to society, many of whom as a result might well continue to pose a physical threat in particular to individuals living in our most vulnerable neighborhoods.

Miller requires trial courts to determine a defendant’s moral culpability for the murder the defendant has committed by examining the defendant’s character and mental development at the time of the offense. Even if the myriad evidence could somehow be obtained by the prosecutor, it is fanciful to believe that the backward-looking determination then required of the trial court could be undertaken with sufficient accuracy and trustworthiness so many years after the crime had been committed, the trial completed, and the defendant sentenced. Further, just as the prosecutor might no longer be available to represent the people’s interest, neither might the sentencing judge. We are not confident that the justice achieved by a resentencing process taking place many years after the original trial and sentencing — many years after the victims of the homicide have become little more than historical footnotes to all but their immediate families — and presided over by a judge who can never entirely be situated like the judge who presided over the trial, can effectively replicate the justice achieved at the initial sentencing. Instead, we believe that the trial court’s ability to travel back in time to assess a defendant’s mental state of some 20 years earlier — evidence of which may not even have been gathered at the time — is limited; that the recollection of memories about aggravating and mitigating circumstances — evidence of which may again not even have been gathered at the time — is questionable; and that, as a result, public confidence in the integrity and accuracy of those proceedings will understandably be low.

For these reasons, we find that the second and third factors do not sufficiently favor the retroactive application of Miller so as to overcome the first factor counseling against the retroactive application of Miller. As a result of this analysis, Miller is not entitled to retroactive application under Michigan’s test for retroactivity.

D. CONSTITUTIONAL ISSUES

Defendants raise a series of constitutional challenges arguing that the Eighth Amendment of the United States Constitution or Const 1963, art 1, § 16, or both, categorically bars the imposition of a life-without-parole sentence on a juvenile homicide offender. We consider each challenge in turn.

1. FEDERAL CATEGORICAL BAR

Defendants assert that the Eighth Amendment of the United States Constitution³⁶ categorically bars the imposition of a sentence of life without parole on any juvenile homicide offender, regardless of whether the “individualization” of sentencing is performed before that sentence is imposed. The effect of the categorical rule sought by defendants would not only mandate resentencing for all juvenile defendants sentenced to life without parole under the pre-Miller sentencing scheme, but would also invalidate those portions of MCL 769.25 allowing the state to impose a life-without-parole sentence on particular juveniles following an individualized sentencing hearing in accordance with Miller. See MCL 769.25(2) through (7). Defendants ask this Court to read the United States Supreme Court’s rulings in Roper, Graham, and Miller as necessarily foreshadowing the conclusion that the Eighth Amendment categorically bars life-without-parole sentences for all juvenile offenders. However, the limited nature of each of these rulings does not, in our judgment, necessitate that conclusion. Moreover, the proportionality-review employed by the United States Supreme Court in fashioning the rules in Roper, Graham, and Miller also does not support the categorical rule sought by defendants.

As noted earlier, the holding in Roper was specifically limited to capital punishment in that the “Eighth and Fourteenth Amendments forbid imposition of the death penalty on offenders who were under the age of 18 when their crimes were committed.” Roper, 543 US at 578. Given that capital punishment was only “likened” to life without parole for a juvenile offender, Miller, 567 US at_; 132 S Ct at 2463-2464, rather than deemed equivalent to life without parole for a juvenile offender, neither Roper nor Roper in conjunction with Graham and Miller suggests in any way that the Eighth Amendment must be read as invalidating the state’s ability to impose a life-without-parole sentence on a juvenile homicide offender. Likewise, Graham’s holding was specifically limited so as to categorically bar only the imposition of life-without-parole sentences for juvenile offenders convicted of nonhomicide offenses. Graham, 560 US at 79. Accordingly, Graham also does not compel the invalidation of a state’s ability to impose a sentence of life without parole on a juvenile homicide offender.

Turning lastly to Miller, its rule is specifically limited in that it counsels against the very categorical rule sought by defendants. As discussed earlier, Miller requires that an individualized sentencing hearing occur before a life-without-parole sentence may be imposed, but expressly “does not categorically bar a penalty” or “foreclose a sentencer’s ability” to impose a life-without-parole sentence. Miller, 567 US at_; 132 S Ct at 2469, 2471. Defendants’ proposed categorical rule would therefore read the Eighth Amendment as categorically barring precisely the very punishment that Miller declined to categorically bar and, in so doing, asserted was not categorically barred by the Eighth Amendment.

Defendants alternatively contend that, in light of the manner in which state legislatures reacted to Miller by adjusting sentencing schemes governing juvenile homicide offenders, it is now, pursuant to the proportionality review employed in Roper, Graham, and Miller, cruel and unusual punishment to impose a life-without-parole sentence on a juvenile homicide offender. Within the context of the Eighth Amendment, the United States Supreme Court has used a multipart test to determine if a punishment imposed on a juvenile offender is disproportionate:

A court must begin by comparing the gravity of the offense and the severity of the sentence. “[I]n the rare case in which [this] threshold comparison ... leads to an inference of gross disproportionality” the court should then compare the defendant’s sentence with the sentences received by other offenders in the same jurisdiction and with the sentences imposed for the same crime in other jurisdictions. If this comparative analysis “validate[s] an initial judgment that [the] sentence is grossly disproportionate,” the sentence is cruel and unusual. [Graham, 560 US at 60, quoting Harmelin, 501 US at 1005 (Kennedy, J., concurring in part).]

Starting with the preliminary question whether “the gravity of the offense” is commensurate with “the severity of the sentence,” Graham, 560 US at 60, we note that first-degree murder is almost certainly the gravest and most serious offense that an individual can commit under the laws of Michigan — the premeditated taking of an innocent human life. It is, therefore, unsurprising that the people of this state, through the Legislature, would have chosen to impose the most severe punishment authorized by the laws of Michigan for this offense. Although the individualized sentencing process now required by Miller (and as a necessary response to Miller by MCL 769.25) may perhaps indicate that some juvenile offenders lack the moral culpability and mental faculties to warrant a life-without-parole sentence pursuant to the premises of Miller, when the contrary conclusions are drawn, as they presumably will be in some cases, a sentence of life without parole for first-degree murder will not “lead[] to an inference of gross disproportionality.” Id. Accordingly, defendants have failed to demonstrate that the imposition of a life-without-parole sentence will satisfy the first part of the United States Supreme Court’s test for proportionality. As the first part of this federal test is a necessary requirement for finding that a punishment is “disproportionate,” defendants’ facial challenge fails as they are consequently unable to demonstrate that the Eighth Amendment categorically bars the imposition of a life-without-parole sentence on juvenile homicide offenders.

Even if defendants had satisfied the first part of the federal test for disproportionality, however, they have also failed to satisfy the second part of the test, which compares the life-without-parole sentence defendants seek to invalidate “with the sentences received by other offenders in the same jurisdiction and with the sentences imposed for the same crime in other jurisdictions.” Id. As for other offenders within the state of Michigan, defendants are correct to note that life without parole is the most severe punishment imposed by this state. This fact alone, however, does not persuade us that imposing a life-without-parole sentence on a juvenile homicide offender is disproportionate.

First, as noted in the first part of this test for proportionality, first-degree murder is almost certainly the gravest and most serious offense that can be committed under the laws of Michigan. As with juveniles, adult offenders who commit the offense of first-degree murder face the same sentence of life without parole. Because some juvenile offenders will possess the same mental faculties of an adult so that they are equally able to recognize the consequences of their crimes and form an unequivocal premeditated intent to kill in the face of the consequences, it is not categorically disproportionate to punish at least some juvenile offenders the same as adults.

Second, there are some nonhomicide offenses that may be viewed as less grave and less serious than first-degree murder and for which only adult offenders face a life-without-parole sentence in this state. For instance, an adult who commits successive first-degree criminal sexual conduct offenses against an individual under the age of 13 faces a sentence of life without parole. MCL 750.520b(2)(c). Accordingly, when the commission of a nonhomicide offense by an adult offender may result in the imposition of a life-without-parole sentence, it does not appear categorically disproportionate to impose a life-without-parole sentence on a juvenile offender for committing the gravest and most serious homicide offense.

Third, although this Court is required by Graham to assess the proportionality of a sentence of life without parole imposed on juveniles who commit first-degree murder, we would be derelict if we did not observe that the people of this state, acting through their Legislature, have already exercised their judgment — to which we owe considerable deference — that the sanction they have selected for juvenile first-degree-murder offenders is, in fact, a proportionate sanction. We are not certain that there is a superior test for assessing a determination of proportionality than that a particular sanction is compatible with public opinion and sentiment. Nonetheless, because this Court is required to do so by Graham, we undertake to the best of our ability to exercise independent judgment in analyzing the criminal punishments authorized by our Legislature and assessing their propriety in the light of the crimes for which the Legislature has deemed them proportionate.

Turning to whether Michigan’s sentencing scheme for juvenile first-degree-murder offenders is “disproportionate” to sentencing schemes used in other states, defendants have wholly failed to present relevant data demonstrating that Michigan is an outlier when it comes to permitting the imposition of life-without-parole sentences for juvenile first-degree-murder offenders, even on the assumption that being an “outlier” adversely affects our state’s compliance with the United States Constitution. Defendants in their briefs cherry-pick six states in which sentencing schemes have been altered post -Miller to eliminate life-without-parole as a possible sentence for juvenile offenders. The fact that six states have eliminated life-without-parole sentences for juvenile offenders in response to Miller tells us next to nothing about how Michigan’s choice to impose life-without-parole sentences on juveniles convicted of first-degree murder compares to sentencing schemes across the nation, and defendants have come nowhere close to satisfying their burdens in this regard.

What trend is demonstrated by the actions of these six states alone? How many states at the time of Miller imposed a sentence of life without parole on juvenile homicide offenders? How many of these states responded to Miller in a manner similar to that of Michigan? What is apparent is that at the time of Miller, “26 States ... [made] life without parole the mandatory (or mandatory minimum) punishment for some form of murder, and would apply the relevant provision to 14-year-olds ----” Miller, 567 US at_n 9; 132 S Ct at 2471 n 9. Another 15 states allowed for the discretionary imposition of life-without-parole sentences on juvenile offenders. Id. at_n 10; 132 S Ct at 2472 n 10. Combined therefore, 41 states exercised the authority under at least some circumstances to impose a life-without-parole sentence on a juvenile. If, as defendants assert, six of those states have departed from this practice by eliminating that sentence altogether, can it be concluded that life-without-parole sentences for juveniles are disproportionte when they remain an option of some kind in 35 states in total, or 70% of the states composing the Union?

In summary, we have no evidence that sustains defendants’ burden of demonstrating that Michigan’s statutory scheme is categorically disproportionate to those of other states. As defendants have failed to demonstrate that either part of the federal test for the constitutionality of punishments supports the conclusion that a life-without-parole sentence for juvenile homicide offenders is disproportionate, we decline to hold that the Eighth Amendment of the United States Constitution categorically bars that punishment.

2. STATE CATEGORICAL BAR

Defendants next contend that even if the Eighth Amendment does not categorically bar the imposition of sentences of life without parole on juvenile homicide offenders, Const 1963, art 1, § 16 does mandate such a categorical bar. Whereas the Eighth Amendment proscribes the imposition of “cruel and unusual punishments,” Const 1963, art 1, § 16 states:

Excessive bail shall not be required; excessive fines shall not be imposed; cruel or unusual punishment shall not be inflicted; nor shall witnesses be unreasonably detained. [Emphasis added.]

The textual difference between the federal constitutional protection and the state constitutional protection is of consequence and has led this Court to conclude that Article 1, § 16 provides greater protection against certain punishments than its federal counterpart in that if a punishment must be both “cruel” and “unusual” for it to be proscribed by the Eighth Amendment, a “punishment that is unusual but not necessarily cruel” is also proscribed by Article 1, § 16. People v Lorentzen, 387 Mich 167, 172; 194 NW2d 827 (1972).

This broader protection under Article 1, § 16 against punishments that are merely “unusual” has led this Court to adopt a slightly different and broader test for proportionality than that employed in Graham. See id. at 171-172; see also People v Bullock, 440 Mich 15, 31; 485 NW2d 866 (1992).³⁷ As set forth in Lorentzen and Bullock, the state test for proportionality assesses (1) the severity of the sentence imposed compared to the gravity of the offense, (2) the penalty imposed for the offense compared to penalties imposed on other offenders in the same jurisdiction, (3) the penalty imposed for the offense in Michigan compared to the penalty imposed for the same offense in other states, and (4) whether the penalty imposed advances the penological goal of rehabilitation. Bullock, 440 Mich at 33-34, citing Lorentzen, 387 Mich at 176-181.

At the outset, we note that the LorentzenIBullock test bears a considerable resemblance to the federal test for proportionality because the first three factors combine to effect the same general inquiry as the two-part test employed in Graham. See Bullock, 440 Mich at 33 (“Our analysis in Lorentzen foreshadowed in a striking manner the three-pronged test later adopted by the United States Supreme Court in Solem v Helm, 463 US 277, 290-291; 103 S Ct 3001; 77 L Ed 2d 637 (1983).”). Our conclusion that none of the first three factors supports the inference that a life-without-parole sentence for a juvenile offender is disproportionate under the Eighth Amendment also bears on the first three inquires of the proportionality analysis under the LorentzenIBullock test. Accordingly, only the fourth factor of the LorentzenIBullock test remains to be assessed before weighing these factors and reaching a conclusion about the proportionality of a life-without-parole sentence for a juvenile homicide offender under Article 1, § 16 of our state constitution.

Concerning the fourth factor, we concur with the United States Supreme Court’s assessment that a life-without-parole sentence for a juvenile does not serve the penological goal of rehabilitation.³⁸ Graham, 560 US at 74. As stated in Graham, when life without parole is imposed on a juvenile, “[t]he penalty forswears altogether the rehabilitative ideal. By denying the defendant the right to reenter the community, the State makes an irrevocable judgment about that person’s value and place in society.” Id. Accordingly, the fourth factor of the Lorentzen/Bullock test supports defendants’ contention that a life-without-parole sentence for a juvenile offender is disproportionate. That said, with only one of the four factors supporting the conclusion that life-without-parole sentences are disproportionate when imposed on juvenile homicide offenders, defendants have failed to meet their burden of demonstrating that it is facially unconstitutional under Article 1, § 16 to impose that sentence on a juvenile homicide offender. While the language of the Michigan counterpart to the Eighth Amendment is at some variance from the latter, it is not so substantially at variance that it results in any different conclusion in its fundamental analysis of proportionality.

3. AIDING AND ABETTING

Davis argues that even if the Eighth Amendment does not categorically bar imposing sentences of life without parole on juvenile homicide offenders, it at least categorically bars imposing life-without-parole sentences on juvenile homicide offenders, such as himself, convicted of felony murder ostensibly on the basis of an aiding-and-abetting theory. At the outset of our analysis, we note that our Legislature has chosen to treat offenders who aid and abet the commission of an offense in exactly the same manner as those offenders who more directly commit the offense:

Every person concerned in the commission of an offense, whether he directly commits the act constituting the offense or procures, counsels, aids, or abets in its commission may hereafter be prosecuted, indicted, tried and on conviction shall be punished as if he had directly committed such offense. [MCL 767.39.]

Moreover, the Legislature has enacted a felony-murder statute, which treats the commission of a murder during the course of a robbery as first-degree murder. See MCL 750.316(l)(b).³⁹ These choices by the Legislature must be afforded great weight in light of the fact that Lockett, one of the capital-punishment cases relied on by the United States Supreme Court in forming the rule in Miller, specifically instructs:

That States have authority to make aiders and abettors equally responsible, as a matter of law, with principals, or to enact felony-murder statutes is beyond constitutional challenge. [Lockett, 438 US at 602.]

Davis attempts to overcome this constitutional pronouncement in light of his own proposed categorical rule mandating a lesser maximum penalty for aiders and abettors by asserting that Miller and Graham combine to necessitate such a rule. He advances a two-part argument to this effect: (1) the rule in Miller requires individualized sentencing for juvenile offenders in an effort to account for “their lesser culpability,” Miller, 567 US at_; 132 S Ct at 2463, and (2) Graham has already determined that aiders and abettors are sufficiently less culpable that a sentence of life without parole is never constitutionally appropriate, see Graham, 560 US at 69.

Although the first part of this syllogism is undoubtedly accurate, the same cannot be said of the second part. Graham made two statements pertinent to the second part of Davis’s argument:

The Court has recognized that defendants who do not kill, intend to kill, or foresee that life will he taken are categorically less deserving of the most serious forms of punishment than are murderers.. ..

It follows that, when compared to an adult murderer, a juvenile offender who did not kill or intend to kill has a twice diminished moral culpability. [Id.]

In combination with Miller’s requirement that individualized sentencing account for a juvenile’s “lesser culpability,” it has been argued that a juvenile offender cannot be sentenced to life without parole when the defendant did not kill, intend to kill, or foresee that life would be taken as a result of the offense, even when the offense of which the offender was convicted was felony murder. Just such a contention was advanced by Justice Breyer in his concurrence in Miller, in which, addressing specifically the constitutionality of life-without-parole sentences for juvenile offenders convicted of felony murder on an aiding-and-abetting theory, he stated, “Graham dictates a clear rule: The only juveniles who may constitutionally be sentenced to life without parole are those convicted of homicide offenses who ‘kill or intend to kill.’ ” Miller, 567 US at_; 132 S Ct at 2476 (Breyer, J., concurring).

Assuming for the sake of argument that some categorical rule of this nature is the necessary product of Graham and Miller,⁴⁰ it still does not follow that the rule pertains to and encompasses all instances in which a juvenile aids and abets a felony murder. As recognized by Justice Breyer himself, a juvenile who aids and abets a felony murder may have intended the death of any victim of the offense. Id. at _; 132 S Ct at 2477 (indicating that on remand, the trial court would need to determine if the defendant, who was convicted of felony murder for aiding and abetting the commission of a robbery that resulted in a death, “did intend to cause the clerk’s death”). Further, a juvenile who aids and abets a felony murder may have foreseen that a life might be taken as a result of his offense, but proceeded notwithstanding to engage in the underlying offense with indifference to this risk. Accordingly, when a juvenile can be convicted of felony murder on an aiding- and-abetting theory while either intending to kill or having foreseen the possibility that a life could be taken, any categorical rule gleaned from Graham pertaining to the limited situation in which a juvenile homicide offender lacked the intent to kill and did not foresee the possibility that a life could be taken will once again not categorically bar the imposition of a sentence of life without parole for that offense.⁴¹

This conclusion is entirely consistent with, and arguably dictated by, the individualized sentencing process required by Miller. In seeking to assess a juvenile offender’s moral culpability, Miller instructs trial courts to consider the “ ‘circumstances of the particular offense and the character and propensities of the offender.’ ” Id. at_n 9; 132 S Ct at 2471 n 9, quoting Roberts, 428 US at 333, and citing Sumner, 483 US 66 (emphasis added). A categorical rule altogether foreclosing a trial court from imposing a life-without-parole sentence on a juvenile convicted of felony murder on an aiding-and-abetting theory obviates the necessity for any evaluation of either the circumstances of the individual defendant’s offense or the individual defendant’s character. Such a categorical rule would permit a defendant to avoid a life-without-parole sentence for aiding and abetting a felony murder even if the defendant was closely nearing the age of 18 at the time of the offense, intended the death of the victim by instructing a coconspirator to fire the fatal shot, and had had previous encounters with the criminal justice system that demonstrated a lack of amenability to rehabilitation. Because it is not difficult to imagine such a defendant, and because imposing a life-without-parole sentence on that defendant would be warranted and entirely constitutional under Miller, we reject Davis’s facial challenge and his contention that the Eighth Amendment categorically bars the imposition of a life-without-parole sentence on a juvenile convicted of felony murder on an aiding-and-abetting theory.⁴²

4. RIPENESS

Eliason asserts that Const 1963, art 1, § 16 categorically bars the imposition of a sentence of life without parole on a juvenile homicide offender who is 14 years of age at the time of the offense. For Eliason’s facial challenge to be ripe, there must be “a real and immediate threat... as opposed to a hypothetical one” that a sentence of life without parole will be imposed on him. Conat, 238 Mich App at 145, citing Los Angeles v Lyons, 461 US 95, 101-101; 103 S Ct 1660; 75 L Ed 2d 675 (1983), and Dep’t of Social Servs v Emmanuel Baptist Preschool, 434 Mich 380, 410; 455 NW2d 1 (1990) (CAVANAGH, J.). Put differently, in determining whether an issue is justiciably “ripe,” a court must assess “ ‘whether the harm asserted has matured sufficiently to warrant judicial intervention.’ ” Emmanuel Baptist, 434 Mich at 412 n 48 (citation omitted). Inherent in this assessment is the balancing of “any uncertainty as to whether defendant[] will actually suffer future injury, with the potential hardship of denying anticipatory relief.” Id. at 412, citing Abbott Laboratories v Gardner, 387 US 136, 148-149; 87 S Ct 1507; 18 L Ed 2d 681 (1967).

Eliason was 14 years of age at the time of his offense and was initially sentenced to life without parole. However, because Eliason’s case is on direct review, he is entitled to resentencing pursuant to MCL 769.25(l)(b)(ii). Under MCL 769.25(9), the default sentence for a juvenile convicted of first-degree murder is a sentence of a term of years within specific limits rather than life without parole. A juvenile defendant will only face a life-without-parole sentence if the prosecutor files a motion seeking that sentence and the trial court concludes following an individualized sentencing hearing in accordance with Miller that such a sentence is appropriate. MCL 769.25(2) through (7).

Although the prosecutor has filed a motion seeking the imposition of a sentence of life without parole, it is no more than speculation whether the trial court will depart from the default sentence in response to the prosecutor’s motion and impose a life-without-parole sentence, and it is not apparent that Eliason faces a “real and immediate” threat of receiving a life-without-parole sentence. Furthermore, because he will be facing a minimum sentence of “not less than 25 years,” MCL 769.25(9), to deny on ripeness grounds the relief Eliason seeks will cause him no legally cognizable hardship or harm. If a life-without-parole sentence is imposed at resentencing, Eliason will have more than ample time to appeal and assert either an as-applied or a facial constitutional challenge to his sentence before he completes the minimum possible sentence for his offense. Accordingly, in light of Eliason’s being entitled to resentencing under MCL 769.25, his facial constitutional challenge to life-without-parole sentences for juvenile homicide offenders who are 14 years of age at the time of their offense is no longer justiciable.⁴³

V CONCLUSION

For these reasons, we hold that the rule set forth in Miller should not be retroactively applied under either the federal retroactivity test set forth in Teague or Michigan’s separate and independent retroactivity test set forth in Sexton and Maxson. In so doing, we affirm the judgments of the Court of Appeals in Carp and Davis that Miller should not be applied retroactively. We further hold that neither the Eighth Amendment nor Const 1963, art 1, § 16 categorically bars the imposition of a sentence of life without parole on a juvenile first-degree-murder offender or a juvenile convicted of felony murder on the basis of an aiding-and-abetting theory. Finally, we hold that Eliason’s facial constitutional challenge is no longer ripe and therefore remand his case for resentencing pursuant to MCL 769.25.

YOUNG, C.J., and ZAHRA and VIVIANO, JJ., concurred with Markman, J.

1 The Court of Appeals also opined in dictum how Miller should be applied by trial courts in resentencing juvenile first-degree-murder offenders in cases that were not presented on collateral review. Carp, 298 Mich App at 523-537.

2 At trial, Davis testified that he had not participated in the robbery, but that a third cohort, “Shay-man,” and the other cohort, had committed the offense without Davis's help or encouragement.

3 A federal district court dismissed Davis’s federal habeas petition, expressly rejecting his contention "that there was insufficient evidence to convict him of first-degree felony murder.” Davis v Jackson, unpublished opinion and order of the United States District Court for the Eastern District of Michigan, issued April 30,2008 (Docket No. 01-CV-72747), p 9. The court relied on the surviving victim’s “testi[mony] that both [Davis] and his co-defendant fired their weapons at the decedent.” Id. Davis challenged the credibility of this witness, but the court rejected this assertion because “[t]he testimony of a single, uncorroborated prosecuting witness or other eyewitness is generally sufficient to support a conviction, so long as the prosecution presents evidence which establishes the elements of the offense beyond a reasonable doubt.” Id. at 11. The court later denied Davis’s request for a certificate of appealability. Davis v Jackson, unpublished order of the United States District Court for the Eastern District of Michigan, entered June 4, 2008 (Docket No. 01-CV-72747). The United States Court of Appeals for the Sixth Circuit affirmed this denial, stating that “[a]n eyewitness . . . testified that both Davis and his co-perpetrator fired shots at the decedent.” Davis v Jackson, unpublished order of the United States Court of Appeals for the Sixth Circuit, entered July 14, 2009 (Docket No. 08-1717), p 2.

4 MCL 769.25a(3) contains a similar exception to the prospective application of MCL 769.25 in the event that this Court or the United States Supreme Court holds that Miller applies retroactively to juvenile first-degree-murder offenders convicted on a felony-murder theory under MCL 750.316(l)(b).

5 The Court’s basis for prescribing this rule, distinguishing between adult and juvenile offenders for purposes of constitutional analysis, rested on three factors: (1) juveniles, by way of their “lack of maturity,” tend to engage in “impetuous and ill-considered actions,” (2) “juveniles are more vulnerable or susceptible to negative influences and outside pressures” because they “have less control... over their own environment,” and (3) “the character of a juvenile is not as well formed as that of an adult.” Roper, 543 US at 569-570 (citation and quotation marks omitted).

6 This is but one of several statements from Miller highlighting the limited effect of its rule as it pertains to requiring “a certain process” rather than “categorically bar[ring] a penalty.” In the paragraph in which it describes its holding and addresses the sentencer’s obligations before imposing a life-without-parole sentence, the Court stated, “[W]e do not foreclose a sentencer’s ability to make that judgment in homicide cases . ...” Id. at ,_; 132 S Ct at 2469. Additionally, in discussing the breadth of its holding, the Court stated unequivocally that it has not placed any bar on imposing a life-without-parole sentence on juvenile homicide offenders because it had declined to even reach the question of whether the Eighth Amendment requires such a bar. See id. at_; 132 S Ct at 2469 (“[W]e do not consider Jackson’s and Miller’s alternative argument that the Eighth Amendment requires a categorical bar on life without parole for juveniles . ...”). Indeed, the only opinion in Miller even to entertain the possibility that the Eighth Amendment imposes a categorical bar on life-without-parole sentences for juvenile homicide offenders was Justice Breyer’s concurrence, joined in only by Justice Sotomayor, in which he stated,

Given Graham's reasoning, the kinds of homicide that can subject a juvenile offender to life without parole must exclude instances where the juvenile himself neither kills nor intends to kill the victim. [Id. at_; 132 S Ct at 2475-2476 (Breyer, J., concurring).]

Had the Court itself adopted Justice Breyer’s proposed rule, then Miller might be said to have the same form and effect of the categorical rules adopted in Graham and Roper, but the Court did not. The dissent in this case further errs in its attempt to read the rule in Miller and the rule proposed by Justice Breyer as one and the same. See post at 545. Whereas the rule proposed by Justice Breyer draws a bright line, foreclosing the state’s ability to impose a sentence of life without parole for a juvenile convicted of a homicide offense in which the juvenile offender did not kill or intend to kill, the rule in Miller does not foreclose imposing a life-without-parole sentence on such an offender. This is because the rule in Miller, unlike that proposed by Justice Breyer, requires a sentencer to look at not only the circumstances of the offense, but also at the characteristics of the defendant such that a juvenile homicide offender who did not kill or intend to kill could be sentenced to life without parole if the offender, for example, possessed a prior criminal record, showed no signs of amenability to rehabilitation, and exhibited mental faculties similar to those possessed by an adult offender.

7 The dissent does not appear to dispute that the rule in Miller has the form and effect of the rules from Woodson, Lockett, and Eddings, rather than those from Roper and Graham, when it describes the latter decisions as having “forbade” and “prohibited” specific types of punishments as applied to juveniles while describing Miller as having “struck down a sentencing scheme.” Post at 531.

8 This general rule of nonretroactivity stands in contrast to the general rule requiring the retroactive application of new rules to cases that have not become final for purposes of direct appellate review before the new rule is announced. Griffith v Kentucky, 479 US 314, 328; 107 S Ct 708; 93 L Ed 2d 649 (1987).

9 By our count, Carp and Davis are 2 of 334 defendants currently serving life-without-parole sentences in Michigan for crimes committed before they turned 18 years of age whose sentences became final for purposes of direct review before the Supreme Court’s decision in Miller. To fully understand the effect of applying Miller retroactively, it may be helpful to briefly consider the demographics and case histories of the defendants who would be entitled to resentencing if Miller is applied retroactively. There are at least two reasons why these factors are relevant to the Miller analysis: first, Miller focuses its individualized sentencing analysis on the defendant’s circumstances and personal characteristics at the time of the offense, so any retroactive application of Miller necessarily requires an analysis specific to that time, however long ago it may have been. The older the case generally, the greater the state’s interest in finality and, concomitantly, the more burdensome it is likely to be to accurately reconstruct what characterized the offense and the offender at that time. Second, because Miller identifies age and mental development as two consequential factors in determining whether a life-without-parole sentence is constitutionally permissible for a juvenile offender, that sentence is increasingly likely to be permissible the closer an offender was to 18 years of age at the time of the offense. See note 35 of this opinion.

Of the 334 affected defendants, 4 were 14 years of age when they committed their first-degree-murder offenses, 44 were 15 years of age, 105 were 16 years of age, and 181 were 17 years of age. Of the 181 defendants who were 17 years of age at the time of their offenses, 28 were within two months of turning 18 years of age, with several of those individuals within days of turning 18. As for when the defendants were initially sentenced, 172 of the defendants were sentenced at least 20 years ago, with several sentenced as early as the mid- to late 1970s. Another 83 defendants were sentenced between 15 and 20 years ago, 46 were sentenced between 10 and 15 years ago, 33 were sentenced between 5 and 10 years ago, and none were sentenced within the last 5 years.

10 Nonetheless, we observe that

[i]n order to qualify as watershed, a new rule must meet two requirements. First, the rule must be necessary to prevent an impermissibly large risk of an inaccurate conviction. Second, the rule must alter our understanding of the bedrock procedural elements essential to the fairness of a proceeding. [Whorton, 549 US at 418 (citations and quotation marks omitted).]

In applying this standard, the only rule that the United States Supreme Court has ever identified as a “watershed rule” for purpose of Teague’s second exception is the rule drawn from Gideon v Wainwright, 372 US 335; 83 S Ct 792; 9 L Ed 2d 799 (1963), which established that the Sixth Amendment included the right to appointed counsel at trial for indigent defendants. See Whorton, 549 US at 419. Furthermore, the sentencing rule in Miller has no possible effect in preventing any “impermissibly large risk of an inaccurate conviction” and pertains to no “bedrock procedural elements essential to the fairness of a proceeding.”

11 Although the dissent argues that Schriro’s definition of a rule that alters the range of punishments is “inclusive and not exclusive,” post at 545 n 68, the dissent fails to identify a single Supreme Court decision that classifies a rule as “altering the range” of punishments when the rule requires the sentencer to consider a lesser punishment, hut does not exclude any punishment from the range of punishments that may be considered. Despite no such decision, the dissent would make retroactive a type of rule that the Supreme Court has never before granted retroactive status under Teague’s first exception to the general rule of nonretroactivity.

12 Notable to the scope and application of this third description, both Bousley and Davis involved collateral attacks to federal criminal convictions in which such attacks were dependent on the interpretation of federal law, rather than the development of a new constitutional rule.

13 Carp and Davis argue that the sentence imposed on them was a sentence of “mandatory” life without parole. Regardless of the process by which a defendant is sentenced to life without parole, however, the term that the defendant serves is simply life without parole. Had, for instance, Carp and Davis received all the procedural protections afforded by Miller ■ before being sentenced, the terms they would serve in prison would be identical. The specific manner in which a defendant is sentenced, i.e., by operation of law or as a result of individualized sentencing, does not alter the actual sentence rendered or the length of time the defendant must remain in prison.

14 The dissent asserts that the rule in Miller, although having “procedural implications,” is nonetheless substantive because it invalidated “an entire ‘sentencing scheme.’ ” Post at 540. While the dissent is correct that Miller invalidated Michigan’s sentencing scheme authorizing the imposition of a life-without-parole sentence for a juvenile homicide offender, Ring also invalidated Arizona’s sentencing scheme authorizing the imposition of capital punishment on a homicide offender. As Ring was deemed procedural, it follows that the distinction between substantive and procedural rules does not turn on whether the new rule invalidates a sentencing scheme authorizing a punishment. Instead, the distinction turns on whether the punishment is one that the state may constitutionally impose under any conceivable sentencing scheme governing the class of defendants to which the defendant belongs.

15 The dissent argues that while a shift in decision-making authority from a judge to a jury is procedural, a shift in decision-making authority from the legislature to the judiciary is substantive because it vests new authority (the authority to impose a lesser sentence) in the judiciary. Post at 544-545. Although we acknowledge that there is a difference between these respective shifts in decision-making authority, we do not find the difference pivotal in determining whether a new rule is substantive or procedural. This is because the question at hand is not focused on whether the judiciary’s or the legislature’s or the executive’s authority has changed as a function of the new rule, but inquires only whether the punishment imposed is one that is beyond the state’s or the law’s power to impose. Schriro, 542 US at 352 (defining a rule as substantive when it “placets] particular conduct or persons covered by the statute beyond the State’s power to punish” or means that the defendant “faces a punishment that the law cannot [any more] impose upon him”) (emphasis added). Both before and after Miller the state of Michigan possessed the authority to constitutionally impose a sentence of life without parole on a juvenile homicide offender.

16 One of the critical divides between how this majority resolves the question of Miller’s retroactivity and how the dissent resolves the same question centers on the significance each accords to the words the Supreme Court chose to use in describing the rule in Miller. Despite its many thoughtful arguments, the dissent is unable to explain why the Supreme Court, if it genuinely intended for the rule in Miller to be applied retroactively under Teague, specifically stated that the rule in Miller does not “categorically bar a penalty,” Miller, 567 US at_; 132 S Ct at 2471, when the “categorical bar” versus “noncategorical bar” distinction defines the critical element of the retroactivity analysis in Teague. The dissent contends that by focusing on “categorical” versus “noncategorical” distinction, the majority “muddles” the Teague analysis. Post at 540. However, it is the dissent that misapprehends Teague by its conclusion that the rule in Miller is entitled to retroactive application despite its acknowledgement that Miller did not categorically bar life-without-parole sentences for juveniles. Id. Neither defendants nor the dissent has identified a single Supreme Court decision that has ever concluded that a noncategorical rule is entitled to retroactive application under the first of Teague’s two exceptions to the general rule of nonretroactivity. From this, we can only reason that Teague does not merely stand for the proposition, as the dissent asserts, that a categorical rule is substantive, but also for the proposition that a rule is substantive only when it is categorical.

17 We include federal courts of appeal in our discussion because Carp cites federal courts of appeal decisions for the proposition that the capital-punishment strand of precedent has been applied retroactively.

18 We further note that even if Sumner had applied Woodson retroactively to a case that had become final for direct review purposes before Woodson was announced, it still would not follow that Woodson qualified for retroactive application under Teague. This is because Sumner was decided in 1987 and Teague, in which a plurality of the United States Supreme Court announced the current federal retroactivity test, was not decided until 1989. It is for this same reason that we reject Carp’s contention that the retroactive application of Lockett’s rule in Songer v Wainwright, 769 F2d 1488, 1489 (CA 11, 1985), and Dutton v Brown, 812 F2d 593, 599 n 7 (CA 10,1987), carries any weight with regard to whether those courts applying Lockett retroactively would have done so under Teague. The same can also he said about the significance of the retroactive application of the rule from Furman as acknowledged in Michigan v Payne, 412 US 47, 57 n 14; 93 S Ct 1966; 36 L Ed 2d 736 (1973).

19 In framing the issue before the court, the Eleventh Circuit stated:

On appeal, Thigpen raises only one issue: whether the admission of evidence that he was convicted in 1972 of another first-degree murder and received a death sentence ... rendered his trial so fundamentally unfair that he was convicted without the due process of law. For the reasons set forth below, we affirm the district court’s conclusion that Thigpen’s conviction was constitutional. [Thigpen, 926 F2d at 1005.]

20 Because Carp’s argument fails here, we find it unnecessary to address whether Miller adds the elements of age and incorrigibility to what must be found before a life-without-parole sentence may be imposed on a juvenile homicide offender. We do note that Miller’s repeated statements that individualized sentencing hearings could occur before a “ judge or jury,” Miller, 567 US at_; 132 S Ct at 2460, 2470, 2475, tend to suggest that Miller did not make age or incorrigibility aggravating elements because under Alleyne aggravating elements that raise the mandatory minimum sentence “must be submitted to the jury and found beyond a reasonable doubt,” Alleyne, 570 US at_; 133 S Ct at 2155. (Emphasis added.) However, because Alleyne was decided after Miller, Miller’s reference to individualized sentencing being performed by a “judge or jury” might merely be instructive on the issue but not dispositive. As none of the defendants before this Court asserts that his sentence is deficient because it was not the product of a jury determination, we find it unnecessary to further opine on this issue and leave it to another day to determine whether the individualized sentencing procedures required by Miller must be performed by a jury in light of Alleyne.

21 Treating Alleyne as a procedural rule is consistent with how multiple federal courts have resolved the issue of whether Alleyne is procedural or substantive for federal retroactivity purposes. See, e.g., Simpson v United States, 721 F3d 875, 876 (CA 7, 2013) (comparing Alleyne to the rule from Apprendi v New Jersey, 530 US 466; 120 S Ct 2348; 147 L Ed 2d 435 (2000), which has been held to be procedural); United States v Evans,_ F Supp 2d_(WD Ark, February 25, 2014, Case Nos. 1:11-CR-10012 and 1:13-CV-1025), citing United States v Lara-Ruiz, 721 F3d 554, 557 (CA 8, 2013); Willoughby v United States,_F Supp 2d_(WD NC, September 17, 2013, Case Nos. 3:13-CV-493-FDW and 3:99-CR-24-FDW-6).

22 Tellingly, with regard to the prosecutor’s intentions in Jackson, we further note that on remand the prosecutor conceded the defense of retroactivity, but did so only on the basis “that Jackson is entitled to the benefit of the United [States] Supreme Court’s opinion in his own case.” See Jackson v Norris, 2013 Ark 175, p 6; 426 SW3d 906 (2013) (emphasis added).

23 The dissent similarly acknowledges that the Supreme Court’s application of the rule in Miller to Jackson is “inconclusive” about whether the rule should be applied retroactively and that the relief Jackson received does not mandate the retroactive application of Miller to any other case. Post at 535 n 31.

24 Although the issue was not raised in any way by any of the defendants, the dissent argues that Miller is similar to Atkins v Virginia, 536 US 304; 122 S Ct 2242; 153 L Ed 2d 335 (2002), because “considerable discretion” is left to the states by both rules, so that where Atkins has been applied retroactively, so too should Miller. Post at 547-549. While the dissent is not incorrect to suggest that Miller and Atkins both allow some discretion to the states, it fails to examine this issue with greater precision. Atkins held that the Eighth Amendment bars the imposition of capital punishment on a “mentally retarded offender.” Atkins, 536 US at 321. Atkins, however, left it to the discretion of the states to establish criteria for whether a defendant qualifies as “mentally retarded.” Id. at 317. Accordingly, the discretion left to the states by Atkins pertains to when Atkins applies and which defendants fall within the universe of defendants governed by Atkins. Once a defendant is deemed to be mentally retarded, however, the state’s discretion ceases and Atkins compels the single result that the state is constitutionally prohibited from imposing capital punishment on the defendant. Under Miller, by contrast, all juveniles are entitled to individualized sentencing hearings and accordingly the state has no discretion to determine when, and to which defendants, Miller applies. Instead, the discretionary element oí Miller only comes into play in selecting a sentence for a defendant after it has been determined, per Miller, that the defendant is a juvenile by virtue of being under the age of 18 at the time of the offense. In this regard, the rules announced in Atkins and Miller have both different forms and different effects. That is, Atkins has the form of a categorical rule in that after a state has determined that a defendant is “mentally retarded,” it applies to bar the imposition of capital punishment on that defendant, while Miller has the form of a noncategorical rule in that it requires individualized sentencing before a life-without-parole sentence may be imposed on a juvenile homicide offender but expressly does not bar the imposition of that sentence. Further, the effect of Atkins wifi always produce a single result in invalidating the capital sentence of every defendant who falls within the rule because the defendant is “mentally retarded,” while the effect of Miller will necessarily result in the imposition of a variety of sentences for different offenders, creating only the potential that any given juvenile will receive a sentence other than life without parole.

25 Contrary to Carp’s and Davis’s assertions, and consistently with the general principle of nonretroactivity, this Court does not adhere to the doctrine that an unconstitutional statute is void ab initio. People v Smith, 405 Mich 418, 432-433; 275 NW2d 466 (1979). In rejecting this doctrine, this Court in Smith, 405 Mich at 432, cited Lemon v Kurtzman, 411 US 192; 93 S Ct 1463; 36 L Ed 2d 151 (1973), which, for federal retroactivity purposes, departed from the view that an unconstitutional statute is a nullity ah initio. Smith also quoted Chicot Co Drainage Dist v Baxter State Bank, 308 US 371; 60 S Ct 317; 84 L Ed 329 (1940), for the proposition that a new constitutional rule does not always nullify past application of the old rule when the old rule was understood to have conformed with the Constitution at the time it was applied: “ ‘The actual existence of a statute, prior to such a determination, is an operative fact and may have consequences which cannot justly be ignored. The past cannot always be erased by a new judicial declaration.’ ” Smith, 405 Mich at 432, quoting Chicot Co, 308 US at 374.

26 The new rule made retroactive in McConnell was set forth in Mempa v Rhay, 389 US 128; 88 S Ct 254; 19 L Ed 2d 336 (1967), and held that the Sixth Amendment right to counsel, including the appointment of counsel for indigent defendants, extended to the sentencing phase of a criminal trial. McConnell, 393 US at 2-3.

27 The Sixth Amendment right to counsel has been described as a right “necessary to insure fundamental human rights of life and liberty” with “[t]he Sixth Amendment stand[ing] as a constant admonition that if the constitutional safeguards it provides be lost, justice will not ‘still be done.’ ” Johnson v Zerbst, 304 US 458, 462; 58 S Ct 1019; 82 L Ed 1461 (1938), citing Palko v Connecticut, 302 US 319, 325; 58 S Ct 149; 82 L Ed 288 (1937). In Gideon, 372 US at 344, the Sixth Amendment right to counsel was described as “fundamental and essential to fair trials,” such that indigent criminal defendants facing felony charges are entitled to the appointment of counsel.

28 As McConnell noted, rules extending “a criminal defendant’s right to counsel at trial, Gideon v. Wainwright, 372 U. S. 335 (1963) ; at certain arraignments, Hamilton v. Alabama, 368 U. S. 52 [82 S Ct 157; 7 L Ed 2d 114] (1961) ; and on appeal, Douglas v. California, 372 U. S. 353 [83 S Ct 814; 9 L Ed 2d 811] (1963), have all been applied retroactively.” McConnell, 393 US at 3. In fact, the right to counsel is such a uniquely fundamental right that Gideon remains “the only case that [the United States Supreme Court has] identified as qualifying under the [watershed rule of criminal procedure exception to nonretroactivity from Teague].” Whorton, 549 US at 419.

29 See Sawyer v Smith, 497 US 227, 257-258; 110 S Ct 2822; 111 L Ed 2d 193 (1990) (Marshall, J., dissenting) (“The Court’s refusal to allow Sawyer the benefit of Caldwell [v Mississippi, 472 US 320; 105 S Ct 2633; 86 L Ed 2d 231 (1985)] reveals the extent to which Teague and its progeny unjustifiably limit the retroactive application of accuracy-enhancing criminal rules. Prior to Teague, our retroactivity jurisprudence always recognized a difference between rules aimed primarily at deterring police conduct and those designed to promote the accuracy of criminal proceedings.”).

30 ^e recognize that the prosecutor in Davis and the Attorney General as an intervenor in Carp both assert that this Court should abandon Michigan’s separate test for retroactivity and adopt Teague as our state test. We further recognize the anomalousness of this Court applying new federal rules retroactively pursuant to a standard that is more expansive than that which the United States Supreme Court has directed be applied by federal courts themselves. This anomalousness — at least as it applies to Michigan’s retroactive application of new federal rules — is further heightened when, as in the instant case, (a) the federal rule contradicts the laws of our state as enacted by the Legislature in accordance with the will of the people of Michigan and (b) the Supreme Court has, for purposes of federal court application, specifically rejected the retroactivity test adopted by Michigan. See Teague, 489 US 288. This issue not having been the focal point of briefing or argument, we do not address it further in this case.

31 Interestingly, we note that none of the 334 defendants who would receive resentencing under Miller if it were applied retroactively to cases that had become final before Miller was issued was sentenced after Graham was decided. Therefore, to whatever extent it might he argued that Graham weakened the constitutional foundation of the old rule permitting life-without-parole sentences for juvenile homicide offenders, the argument is of little relevance to the retroactive application of Miller regarding any juvenile defendants currently serving life-without-parole sentences in Michigan.

32 Even with respect to the 34 defendants sentenced post-Roper, there was no cause for prosecutors to believe that the decision had any significant bearing on their ability, on behalf of the people of Michigan, to constitutionally seek a sentence of life without parole or that it brought into question the decision in Launsburry upholding the imposition of life-without-parole sentences.

33 Although Maxson’s analysis of the second factor focused exclusively on whether the defendants in that case had detrimentally relied on the old rule without considering the extent to which prosecutors had detrimentally relied on the old rule, Maxson’s approach to analyzing the second factor is not inconsistent with the approach we use today. When there are two relevant entities, concluding that one of these entities has or has not relied detrimentally on the old rule may be sufficient to reach a conclusion concerning the effect of the second factor on retroactivity. In Maxson, it was clear that the defendants’ detrimental reliance on the old rule was insignificant so it was unnecessary to consider the extent to which prosecutors had relied on the old rule at issue in that case. Although the inverse is largely true here in that the detrimental reliance interests of prosecutors across this state are considerable, we have reviewed what is asserted to be Carp’s and Davis’s detrimental reliance on the old rule and see none. Once again, merely to act in accord with the old rule is not tantamount to detrimental reliance.

34 The dissent similarly struggles to identify what action that would have benefited the 334 defendants was taken or not taken in “detrimental reliance” on the old rule. First, the dissent asserts that trial courts would have engaged in individualized sentencing hearings, but for the old rule. Post at 552. This, however, is an action that courts, not a defendant, would have taken, and essentially asserts nothing more than that Miller has altered the rules. Second, the dissent argues that defendants relied on the old rule by not seeking appellate review of their life-without-parole sentences. Post at 552. In making this argument, the dissent compares this case to Maxson, in which this Court suggested that a defendant’s decision not to pursue an appeal could constitute an action that the defendant opted not to take in reliance on the old rule. Maxson, 482 Mich at 394-395. However, Maxson was addressing the retroactivity of Halbert v Michigan, 545 US 605; 125 S Ct 2582; 162 L Ed 2d 552 (2005), “which held that indigent defendants who plead guilty to criminal offenses are entitled to appointed appellate counsel on direct appeal.” Maxson, 482 Mich at 387. Accordingly, the old rule analyzed in Maxson, that indigent defendants who pleaded guilty to criminal offenses were not entitled to appointed appellate counsel on direct appeal, served as a direct impediment to a defendant’s ability to file an appeal after pleading guilty. In these cases, the pr e-Miller constitutionality of imposing life-without-parole sentences on juvenile homicide offenders by mandatory operation of law did nothing to hinder a defendant’s ability to file an appeal challenging Michigan’s then extant sentencing scheme or its personal application. Furthermore, as Michigan caselaw had specifically upheld the constitutionality of our pr e-Miller sentencing scheme, Launsburry, 217 Mich App 358, it is unclear how defendants’ failures to seek appellate review proved detrimental. While the dissent is obviously correct that their interests were not favored under the old rule to the extent they are under the new rule, that is not the equivalent of having “detrimentally relied” on the old rule.

35 In focusing on the age of the defendants who would receive resentencing if Miller were applied retroactively, we nowhere suggest that age is the exclusive factor that the trial court should consider in imposing a sentence on a juvenile homicide offender, and we agree with the dissent that Miller calls for a “multifaceted” approach to sentencing. Compare page 466 of this opinion with post 553 n 88. However, in light of the other factors that Miller instructs a trial court to consider, it seems apparent that a juvenile’s age at the time of the offense will weigh relatively heavily at sentencing hearings. In most cases, a juvenile’s age will reasonably correspond to his or her mental and emotional development as well as the ability to overcome a difficult family and home life. Additionally, as a juvenile approaches 18 years of age at the time of the offense, and may even turn 18 during the proceedings related to the offense, it follows that the “incompetencies associated with youth” will come to have increasingly less of an effect on the juvenile’s ability to communicate with, and to assist, his or her attorneys in their legal preparations. Accordingly, while age is by no means the only factor to be considered in imposing a sentence pursuant to Miller, an offender’s age is likely to be given significant weight in the court’s deliberations and may well constitute the single best factor for ascertaining whether a Miller-benefited offender would actually gain relief if Miller were applied retroactively.

36 The Eighth Amendment of the United States Constitution reads:

Excessive hail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted. [US Const, Am VIII.]

37 The inclusion of proportionality review under Article 1, § 16 has been the subject of significant disagreement. Bullock, 440 Mich at 46 (Riley, J., concurring in part and dissenting in part) (“I believe that People v Lorentzen .. ., the principle case relied on by the majority to support its conclusion, was wrongly decided and that proportionality is not, and has never been, a component of the ‘cruel or unusual punishment’ clause of this state’s constitution.”); People v Correa, 488 Mich 989, 992 (2010) (Markman, J., joined by Corrigan and Young, JJ., concurring) (“[A]t some point, this Court should revisit Bullock’s establishment of proportionality review of criminal sentences, and reconsider Justice Riley’s dissenting opinion in that case.”). However, because life without parole is not a categorically disproportionate sentence for a juvenile homicide offender, we find it unnecessary in this case to resolve whether proportionality review is rightly a part of the protection in Article 1, § 16 against “cruel or unusual punishment,” instead assuming for the sake of argument that it has a place in an analysis under Article 1, § 16.

38 In accepting this conclusion, this Court, as did the United States Supreme Court, speaks of “rehabilitation” exclusively within the context of a defendant reforming himself or herself for the purpose of reintegration into society. See Graham, 560 US at 74. This, however, is not to foreclose the ability of a person, however long the person is to he incarcerated, to rehabilitate himself or herself in the sense of fully comprehending the nature of the wrong, achieving a greater awareness of and commitment to the elements of moral behavior, attaining a sincere adherence to religious faith, or contributing in positive ways to those with whom the person interacts in whatever environment he or she has been placed.

39 We speak of the felony-murder statute in terms of the underlying felony being a robbery merely because the underlying felony in Davis’s case was a robbery. The reasoning put forth in this part, however, would apply equally when the underlying felony is any one of the other felonies listed in MCL 750.316(l)(b).

40 Although we assume for the sake of argument that such a categorical rule may exist, nothing in this opinion should be understood as actually accepting or adopting such a rule. To the contrary, we note that a categorical rule mandating that a subclass of aiders and abettors be treated differently with respect to what punishments can be imposed would run directly contrary to both the aforementioned statement in Lockett and MCL 767.39. Further, Justice Breyer in his concurrence spoke only for himself and one other justice.

41 To the extent that Graham and Miller might create a categorical rule prohibiting life-without-parole sentences for juveniles convicted of aiding and abetting a felony murder “who do not kill, intend to kill, or foresee that life will be taken,” Graham, 560 US at 69, Davis would not he entitled to relief under that rule. Although the trial court concluded at sentencing that Davis was not the shooter, it did not make an explicit finding regarding Davis’s intentions about the victim’s death, and it made no findings indicative of whether he foresaw the potential that life would he taken as a result of the armed robbery in which he engaged. To go back and attempt to make these findings now would entail engaging in the broader individualized sentencing procedures called for by Miller that we have already determined today need not be engaged in retroactively.

42 This holding carries with it the conclusion that some juveniles convicted of felony murder on an aiding-and-abetting theory might be as morally culpable for their crimes as juveniles who commit premeditated first-degree murder and not simply as legally culpable. A juvenile convicted of felony murder on an aiding-and-abetting theory can be said to have committed as grave an offense as a juvenile who commits premeditated first-degree murder. Accordingly, for the purpose of Davis’s challenge under Const 1963, art 1, § 16, the first two factors of the LorentzenIBullock proportionality test will be resolved in a fashion identical to how they were resolved for life-without-parole sentences generally. Concerning the third factor, Davis fails to present any data specific to how other jurisdictions sentence juveniles convicted of felony murder on an aiding-and-abetting theory, only putting forth a sampling of how a very few states now sentence juveniles convicted of first-degree murder generally. In the absence of evidence to the contrary, we are left to assume that a majority of other states hold aiders and abettors equally responsible for their offenses. Accordingly, the third factor also counsels against a finding of disproportionality. Because only the fourth factor of the Lorentzen/Bullock proportionality test, pertaining to rehabilitation, favors holding life-without-parole sentences for juveniles convicted of felony murder on an aiding-and-abetting theory unconstitutional, Davis’s facial challenge under Article 1, § 16 fails as well.

43 As conceded by the parties at oral argument, Eliason’s other issues on which this Court granted leave to appeal are moot as a result of the enactment of MCL 769.25.