*356 Defendants Leonel Contreras and William Rodriguez were convicted in a joint trial of kidnapping and sexual offenses they committed as 16 year olds. Rodriguez was sentenced to a term of 50 years to life, and Contreras was sentenced to a term of 58 years to life. We granted review to determine whether the sentences imposed on these juvenile nonhomicide offenders violate the Eighth Amendment as interpreted in People v. Caballero (2012) 55 Cal.4th 262 , 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ( Caballero ) and Graham v. Florida (2010) 560 U.S. 48 , 130 S.Ct. 2011 , 176 L.Ed.2d 825 ( Graham ). We hold that these sentences are unconstitutional under the reasoning of Graham .

*357 I.

On September 3, 2011, Jane Doe 1 and Jane Doe 2 attended a birthday party for Doe 1's uncle in the Rancho Peñasquitos area of San Diego County. Doe 1 was 16 years old, and Doe 2 was 15 years old. In the evening, Doe 1 and Doe 2 went for a walk to a greenbelt nearby and sat near a tree to talk. Two teenagers, later identified as Contreras and Rodriguez, walked past them dressed in dark clothing and with their hoods up. Shortly thereafter, defendants walked up behind Doe 1 and Doe 2, tackled them, and forced them to walk across the street, up an embankment, and into a vegetated area. Contreras held a knife to Doe 1's neck and told her to tell Doe 2 to "shut the fuck up" multiple times. Rodriguez covered Doe 2's mouth with his hand, tied a bandana around her mouth, and threatened to hurt her if she screamed. Doe 2 repeatedly tried to get away, fell once from struggling, and at one point bit Rodriguez's hand.

Rodriguez raped and sodomized Doe 2. Contreras raped Doe 1 and forced her to orally copulate him. Rodriguez then raped and sodomized Doe 1 and forced her to orally copulate him. Contreras put a knife to Doe 2's neck, raped her, and forced her to orally copulate him. Rodriguez forced *251 Doe 2 and then Doe 1 to orally copulate him. Defendants then told Doe 1 and Doe 2 to get dressed. Rodriguez told Doe 1 and Doe 2 not to tell anyone what happened. One of the defendants said they would follow Doe 1 and Doe 2 home and come after them and one of Doe 1's family members if they told anyone what had happened. Doe 1 and Doe 2 walked to the street and saw Doe 1's parents, who had been searching for them.

In 2012, defendants were charged as adults under former Welfare and Institutions Code section 707, subdivisions (d)(1) and (d)(2)(A) (amended by Prop. 57, § 4.2, eff. Nov. 9, 2016) and were jointly tried before separate juries. A jury convicted Contreras of conspiracy to commit kidnapping and forcible rape ( Pen. Code, § 182, subd. (a)(1) ; all undesignated statutory references are to this code), rape by foreign object (§ 289, subd. (a)(1)(A) ), two counts of kidnapping (§ 207, subd. (a) ), seven counts of forcible rape (§ 261, subd. (a)(2) ), eight counts of forcible oral copulation (§ 288a, subd. (c)(2)(A) ), and two counts of sodomy by use of force (§ 286, subd. (c)(2)(A) ). The jury found true allegations that Contreras committed the crimes with use of a knife (§ 12022.3, subd. (a) ) as **447 well as allegations that many of the sexual assault crimes were committed during a kidnapping, against more than one victim, and with a knife within the meaning of subdivisions (d)(2), (e)(1), (e)(3), and (e)(4) of section 667.61, the "One Strike" law.

On the same day, a jury convicted Rodriguez of two counts of kidnapping (§ 207, subd. (a) ), two counts of forcible rape (§ 261, subd. (a)(2) ), four *358 counts of forcible oral copulation (§ 288a, subd. (c)(2)(A) ), and two counts of sodomy by use of force (§ 286, subd. (c)(2)(A) ). The jury found true allegations that Rodriguez had committed the sexual assault crimes during a kidnapping and against multiple victims within the meaning of subdivisions (d)(2) and (e)(4) of section 667.61.

At defendants' sentencing hearings, the parties and the trial court agreed that the court could not impose the statutory maximum sentences of several hundred years, as those sentences would fall outside of defendants' natural life expectancies. At Rodriguez's hearing, defense counsel noted that Rodriguez had no criminal history, and the court acknowledged his "very difficult upbringing." But the court said, "I have to weigh that against the horrible scars that you have left on these two girls." The court then sentenced Rodriguez to two consecutive terms of 25 years to life. The court observed that it was required to sentence Rodriguez to additional consecutive terms of 25 years to life under section 667.61, subdivision (i) but reasoned that doing so would violate Graham and Caballero .

At Contreras's hearing, defense counsel noted that Contreras had no arrests and one prior misdemeanor for vandalism. The court said, "I think that Mr. Rodriguez was a follower. Mr. Contreras was the shot caller." The trial judge identified the "brutal and callous and ruthless" nature of the crimes and expressed skepticism about Contreras's ability to rehabilitate: "I think his brain is developed into who he is ...." Based on these factors, among others, the court stated, "I think that it's only appropriate that he suffer the same punishment that Mr. Rodriguez did and plus he used a knife, so he should get a little bit more." The court sentenced Contreras to two consecutive terms of 25 years to life in addition to two four-year terms and imposed many additional concurrent or stayed sentences. The trial judge concluded by noting, "If I could sentence you to 640 years to life, I would have. ... Because you were a minor, you were spared that sentence."

*252 Defendants appealed their convictions and sentences on multiple grounds. The Court of Appeal affirmed the convictions but reversed defendants' sentences. It held that the sentences "preclude any possibility of parole until [defendants] are near the end of their lifetimes" and thus "fall[ ] short of giving them the realistic chance for release contemplated by Graham ." The Court of Appeal remanded the matter to the trial court for resentencing, with instructions to consider the circumstances of the crimes, including the existence of multiple victims, together with all mitigating circumstances, and to impose a parole eligibility date consistent with the holding in Graham .

We granted review and deferred briefing pending our decision in People v. Franklin (2016) 63 Cal.4th 261 , 202 Cal.Rptr.3d 496 , 370 P.3d 1053 *359 Franklin ). In Franklin , we held that juvenile homicide offenders may not be sentenced to the functional equivalent of life without parole (LWOP) without certain protections afforded by the Eighth Amendment as interpreted in Miller v. Alabama (2012) 567 U.S. 460 , 132 S.Ct. 2455 , 183 L.Ed.2d 407 ( Miller ). ( Franklin , at p. 276, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) The defendant there had been sentenced to 50 years to life for first degree murder, and he claimed that his sentence was the functional equivalent of LWOP and was imposed in violation of Miller . We held that because section 3051 entitles Franklin to a youth offender parole hearing during his 25th year of incarceration, his sentence "is neither LWOP nor its functional equivalent" and thus gives rise to "no Miller claim." ( Franklin , at p. 280, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .)

A youth offender parole hearing is not available to juveniles convicted under the One Strike law, as defendants were here. (§ 3051, subd. (h).) Because Franklin does **448 not resolve this case, we ordered briefing to address whether Rodriguez's sentence of 50 years to life or Contreras's sentence of 58 years to life violates the Eighth Amendment.

II.

The Eighth Amendment ban on cruel and unusual punishment "flows from the basic ' "precept of justice that punishment for crime should be graduated and proportioned to [the] offense. [Citation.]" ' " ( Roper v. Simmons (2005) 543 U.S. 551 , 560, 125 S.Ct. 1183 , 161 L.Ed.2d 1 ( Roper ).) "By protecting even those convicted of heinous crimes, the Eighth Amendment reaffirms the duty of the government to respect the dignity of all persons." ( Ibid . ; see Robinson v. California (1962) 370 U.S. 660 , 667, 82 S.Ct. 1417 , 8 L.Ed.2d 758 [Eighth Amendment applies to the states].)

The United States Supreme Court has interpreted the Eighth Amendment to impose unique constraints on the sentencing of juveniles who commit serious crimes. This case law reflects the principle that "children are constitutionally different from adults for purposes of sentencing." ( Miller , supra , 567 U.S. at p. 471, 132 S.Ct. 2455 .) "From this principle, the high court has derived a number of limitations on juvenile sentencing: (1) no individual may be executed for an offense committed when he or she was a juvenile ( Roper , [ supra ,] 543 U.S. at p. 578, 125 S.Ct. 1183 ); (2) no juvenile who commits a nonhomicide offense may be sentenced to LWOP ( Graham , supra , 560 U.S. at p. 74, 130 S.Ct. 2011 ); and (3) no juvenile who commits a homicide offense may be automatically sentenced to LWOP ( Miller , at p. [465] )." ( Franklin , supra , 63 Cal.4th at pp. 273-274, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 ; see *253 Montgomery v. Louisiana (2016) 577 U.S. ----, ----, [ 136 S.Ct. 718 , 734], 193 L.Ed.2d 599 ( Montgomery ) [" Miller announced a substantive rule of constitutional law" that applies retroactively].) The second limitation is relevant here: Because Contreras and Rodriguez committed *360 nonhomicide offenses, the Eighth Amendment does not permit them to be sentenced to LWOP. Although they may be punished with long sentences, they must have "some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." ( Graham , at p. 75, 130 S.Ct. 2011 .)

In Caballero , we held that a juvenile defendant's sentence of 110 years to life for three counts of attempted murder was the functional equivalent of LWOP and, under Graham , violated the Eighth Amendment. ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) We rejected the argument that Graham 's prohibition on LWOP does not apply to aggregated sentences for distinct crimes where each sentence individually provides for the possibility of parole within a juvenile's expected lifespan. ( Id. at pp. 267-268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) We said: " Graham 's analysis does not focus on the precise sentence meted out. Instead, ... it holds that a state must provide a juvenile offender 'with some realistic opportunity to obtain release' from prison during his or her expected lifetime." ( Id . at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .)

Graham and Caballero together hold that the Eighth Amendment does not allow juveniles who commit nonhomicide crimes to be sentenced to LWOP or to a term of years well in excess of natural life expectancy. But neither Graham nor Caballero considered whether a lengthy sentence short of LWOP or its equivalent would likewise violate the Eighth Amendment in this context. The question here is whether Rodriguez's sentence of 50 years to life or Contreras's sentence of 58 years to life for nonhomicide offenses violates the same Eighth Amendment principles that bar the imposition of LWOP for their crimes.

A.

The Attorney General says we "should adopt the following rule: any term of imprisonment that provides a juvenile offender with an opportunity for parole within his or her expected natural lifetime is not the functional equivalent of LWOP ...." The Attorney General urges us to determine natural life expectancy by looking to a report published by the Centers for Disease Control (CDC), based on 2010 data, providing the life expectancies of various age and gender cohorts **449 living in the United States. (See Arias, United States Life Tables, 2010, National Vital Statistics Reports, vol. 63, no. 7 (Nov. 6, 2014) p. 1 (2010 Life Tables).) According to that report, a 16-year-old boy in the United States is expected to live an additional 60.9 years, for a total life expectancy of 76.9 years. ( Id. at p. 11, table 2.) Noting that "Rodriguez will be 66 years old when first eligible for parole, and Contreras will be 74 years old when first eligible for parole," the Attorney General contends that "[b]ecause it affords appellants an opportunity for parole within their expected natural lifetimes, a sentence of 50 years to life *361 and 58 years to life is not the functional equivalent of LWOP and therefore may be constitutionally imposed." As explained below, this actuarial approach urged by the Attorney General is practically and conceptually problematic.

As an initial matter, we find unpersuasive the Attorney General's claim that we already decided in Caballero that a term-of-years sentence does not violate the Eighth Amendment if it allows the possibility of parole at some point during the juvenile offender's natural life expectancy. Caballero held that "sentencing a juvenile offender for a nonhomicide offense to a *254 term of years with a parole eligibility date that falls outside the juvenile offender's natural life expectancy constitutes cruel and unusual punishment in violation of the Eighth Amendment." ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) But the defendant in Caballero challenged a sentence allowing for parole eligibility "over 100 years from now." ( Ibid. ) In that context, it was enough to note that the parole eligibility date "falls outside the juvenile offender's natural life expectancy." ( Ibid. ) We had no occasion to consider whether a term-of-years sentence violates the Eighth Amendment only if it exceeds a juvenile defendant's natural life expectancy. (See Kinsman v. Unocal Corp. (2005) 37 Cal.4th 659 , 680, 36 Cal.Rptr.3d 495 , 123 P.3d 931 [" 'It is axiomatic that language in a judicial opinion is to be understood in accordance with the facts and issues before the court. An opinion is not authority for propositions not considered.' "].)

Taken on its own terms, the Attorney General's actuarial approach gives rise to a tangle of legal and empirical difficulties. In defining life expectancy, the Attorney General relies on our statement in Caballero that "the term 'life expectancy' means the normal life expectancy of a healthy person of defendant's age and gender living in the United States." ( Caballero , supra , 55 Cal.4th at p. 267, fn. 3, 145 Cal.Rptr.3d 286 , 282 P.3d 291 , italics added.) But this passing statement was unnecessary to our decision because the 110-years-to-life sentence at issue clearly exceeded the defendant's life expectancy under any definition. Although a gender-specific approach to determining life expectancy reflects the reality that females generally live longer than males (see 2010 Life Tables, supra , at p. 2 ["The difference in life expectancy between the sexes was 4.8 years in 2010 ...."] ), we did not examine in Caballero whether it would be constitutional to authorize lengthier sentences for girls than for boys in determining the parameters of lawful punishment for juvenile nonhomicide offenders.

"We long ago concluded that discrimination based on gender violates the equal protection clause of the California Constitution (art. I, § 7, subd. (a) ) and triggers the highest level of scrutiny. ( Sail'er Inn, Inc. v. Kirby (1971) 5 Cal.3d 1 , 17-20, 95 Cal.Rptr. 329 , 485 P.2d 529 .)" ( *362 Catholic Charities of Sacramento, Inc. v. Superior Court (2004) 32 Cal.4th 527 , 564, 10 Cal.Rptr.3d 283 , 85 P.3d 67 .) "In order to satisfy that standard, the state must demonstrate not simply that there is a rational, constitutionally legitimate interest that supports the differential treatment at issue, but instead that the state interest is a constitutionally compelling one that justifies the disparate treatment prescribed by the statute in question. [Citation.]" ( In re Marriage Cases (2008) 43 Cal.4th 757 , 847, 76 Cal.Rptr.3d 683 , 183 P.3d 384 .) And "the state must demonstrate that the distinctions drawn by the statute (or statutory scheme) are necessary to further that interest. [Citation.]" ( Id. at p. 848, 76 Cal.Rptr.3d 683 , 183 P.3d 384 .)

It is unclear whether sentencing juveniles based on gender-specific life expectancies **450 would satisfy strict scrutiny. But assuming it would, there would then be no reason why the definition of life expectancy should not also account for well-documented racial differences, since racial classifications are evaluated under the same constitutional standard. (See Johnson v. California (2005) 543 U.S. 499 , 505, 125 S.Ct. 1141 , 160 L.Ed.2d 949 ; Coral Construction, Inc. v. City and County of San Francisco (2010) 50 Cal.4th 315 , 337, 113 Cal.Rptr.3d 279 , 235 P.3d 947 .) According to the CDC report on which the Attorney General relies, life expectancy in 2010 was *255 83.8 years for Hispanic females, 81.3 years for non-Hispanic white females, 78.0 years for black females, 78.7 years for Hispanic males, 76.5 years for white males, and 71.8 years for black males. (2010 Life Tables, supra , at p. 5.) These differences present a conundrum: Although persons of different races and genders are not similarly situated in terms of life expectancy, it seems doubtful that considering such differences in juvenile sentencing would pass constitutional muster.

Moreover, were we to adopt the Attorney General's proposed rule, it is not obvious why the definition of life expectancy should ignore other group-based differences that may be relevant to a particular juvenile defendant. The Pacific Juvenile Defender Center (PJDC), as amicus curiae, notes that life expectancy is affected by many " 'variables that have long been studied by social scientists but are not included in U.S. Census or vital statistics reports-income, education, region, type of community, access to regular health care, and the like ....' " (See Cummings & Colling, There is No Meaningful Opportunity in Meaningless Data: Why It Is Unconstitutional to Use Life Expectancy Tables in Post- Graham Sentences (2014) 18 U.C. Davis J. Juvenile L. & Policy 267, 282.)

Defendants and PJDC highlight the relevance of one variable in particular: incarceration. PJDC cites studies showing that incarceration accelerates the aging process and results in life expectancies substantially shorter than estimates for the general population. (See Patterson, The Dose-Response of *363 Time Served in Prison on Mortality: New York State, 1989-2003 (2013) 103 Am. J. Pub. Health 523, 526 [finding each year of incarceration correlated with a 15.6 percent increase in odds of death for parolees and a two-year decline in life expectancy]; U.S. Dept. of Justice, Nat. Inst. of Corrections, Correctional Health Care: Addressing the Needs of Elderly, Chronically Ill, and Terminally Ill Inmates (2004) pp. 9-10 (Correctional Health Care) [stresses of incarceration intensify the health problems of elderly inmates and accelerate aging processes]; see also Spaulding et al., Prisoner Survival Inside and Outside of the Institution: Implications for Health-Care Planning (2011) 173 Am. J. Epidemiology 479, 484 [currently and formerly incarcerated individuals in Georgia have "overall heightened mortality ... over 15 years of follow-up relative to the general Georgia population," with significant differences by race, gender, and time incarcerated].) One state high court has taken such evidence into account in determining whether a term-of-years sentence violates the Eighth Amendment. (See Casiano v. Commissioner of Correction (Conn. 2015) 317 Conn. 52 , 115 A.3d 1031 , 1046 ( Casiano ).)

On the other hand, it has been suggested that inmates who "have aged in place are generally the best adapted to prison life because they have been in prison since their youth and have adjusted to it." (Correctional Health Care, supra , at p. 10.) Further, although incarceration has its stresses, it may shield inmates from other stresses that would afflict them outside of prison, including violence, accidents, and poor access to health care. (See Spaulding et al., supra , at pp. 482-485; Rosen et al., All-Cause and Cause-Specific Mortality Among Black and White North Carolina State Prisoners, 1995-2005 (2011) 21 Ann. Epidemiology 719, 725-726 [average death rates for currently incarcerated black men in North Carolina prisons are significantly lower than for the black population in the state overall, but currently incarcerated white men have slightly higher average death rates than white men in the state].) In addition, the Attorney General asserts that although race, region, and economic status may affect death rates outside prison, *256 such findings are not necessarily true "for those inside prison, where living conditions, medical treatment, and wealth are roughly the same for all." **451 The record in this case contains no findings by the trial court on these matters. At sentencing, the prosecution introduced evidence of statistical life expectancies, and neither defendant presented evidence demonstrating shorter life expectancy in prison. But we decline to adopt a constitutional rule that employs a concept of life expectancy whose meaning depends on the facts presented in each case. Determining the validity of lengthy term-of-years sentences under the Eighth Amendment through a case-by-case inquiry into competing evidence of the life expectancy most pertinent to a particular juvenile defendant would lead to problems of disparate sentencing. Moreover, even if there were a legally and empirically sound approach to estimating life expectancy, it must be noted that a life expectancy is an average. (2010 Life *364 Tables, supra , at p. 2.) In a normal distribution, about half of a population reaches or exceeds its life expectancy, while the other half does not. Under Graham , juvenile nonhomicide offenders must be given "some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 ; see id. at 130 S.Ct. 2011 [the state must give juvenile nonhomicide offenders "some realistic opportunity to obtain release before the end of [a life] term"].) An opportunity to obtain release does not seem "meaningful" or "realistic" within the meaning of Graham if the chance of living long enough to make use of that opportunity is roughly the same as a coin toss. (Cf. dis. opn. of Cantil-Sakauye, C. J., post, at 229 Cal.Rptr.3d at p. 281, 411 P.3d at p. 472.) Of course, there can be no guarantee that every juvenile offender who suffers a lengthy sentence will live until his or her parole eligibility date. But we do not believe the outer boundary of a lawful sentence can be fixed by a concept that by definition would not afford a realistic opportunity for release to a substantial fraction of juvenile offenders.

B.

In addition to raising legal and empirical difficulties, the actuarial approach proposed by the Attorney General is misguided at a more fundamental level. When evaluating a sentence that clearly exceeds natural life expectancy, like the 110-years-to-life sentence in Caballero , it is straightforward to conclude that the sentence is "functionally equivalent" to LWOP as an actuarial matter. ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) But the issue of functional equivalence in this context is not limited to determining whether a term-of-years sentence is actuarially equivalent to LWOP. Although the Attorney General trains his inquiry on that question, there is a separate and distinct question whether a lengthy term-of-years sentence, though not clearly exceeding a juvenile offender's natural lifespan, may nonetheless impinge on the same substantive concerns that make the imposition of LWOP on juvenile nonhomicide offenders impermissible under the Eighth Amendment. This latter notion of functional equivalence-that a term-of-years sentence may function like LWOP with respect to the Eighth Amendment concerns that constrain lawful punishment for juvenile nonhomicide offenders -is what we must address in this case. (See State v. Null (Iowa 2013) 836 N.W.2d 41 , 71 ["[W]e do not believe the determination of whether the principles of Miller or Graham apply in a given case should turn on the niceties of epidemiology, genetic analysis, or actuarial sciences in determining precise mortality dates."].) To resolve this question, the proper starting point is not a *257 life expectancy table but the reasoning of the high court in Graham .

The defendant in Graham , at age 16, was charged in Florida as an adult for armed burglary with assault or battery, which carried a maximum sentence of LWOP, and attempted armed robbery, which carried a maximum sentence of 15 years. ( Graham , supra , 560 U.S. at pp. 53-54, 130 S.Ct. 2011 .) Graham pleaded guilty to *365 both charges and, in a letter to the trial court, said " 'this is my first and last time getting in trouble' " and " 'I've decided to turn my life around.' " ( Id. at p. 54, 130 S.Ct. 2011 .) The trial court withheld adjudication of guilt and sentenced him to probation. ( Ibid. ) Less than six months later, 34 days before his 18th birthday, Graham participated in a home invasion robbery and afterward admitted he had violated his probation conditions. ( Id. at pp. 54-55, 130 S.Ct. 2011 .) At that point, the trial court found Graham guilty of the earlier **452 armed burglary and attempted armed robbery. ( Id. at pp. 55-57, 130 S.Ct. 2011 .)

At sentencing, the trial court said: " 'Mr. Graham, as I look back on your case, yours is really candidly a sad situation. You had, as far as I can tell, you have quite a family structure. You had a lot of people who wanted to try and help you get your life turned around including the court system, and you had a judge who took the step to try and give you direction through his probation order to give you a chance to get back onto track. And at the time you seemed through your letters that that is exactly what you wanted to do. And I don't know why it is that you threw your life away. ... [¶] But you did, and that is what is so sad about this today .... [¶] ... [¶] And I don't understand why you would be given such a great opportunity to do something with your life and why you would throw it away. The only thing that I can rationalize is that you decided that this is how you were going to lead your life and that there is nothing that we can do for you. And as the state pointed out, that this is an escalating pattern of criminal conduct on your part and that we can't help you any further. We can't do anything to deter you. This is the way you are going to lead your life .... [¶] ... [¶] ... I don't see where any further youthful offender sanctions would be appropriate. Given your escalating pattern of criminal conduct, it is apparent to the Court that you have decided that this is the way you are going to live your life and that the only thing I can do now is to try and protect the community from your actions.' " ( Graham , supra , 560 U.S. at pp. 56-57, 130 S.Ct. 2011 .) The trial court sentenced Graham to the maximum penalty for both crimes: LWOP for the armed burglary and 15 years in prison for the attempted armed robbery. ( Id. at p. 57, 130 S.Ct. 2011 .) The high court held that the Eighth Amendment categorically "prohibits the imposition of a life without parole sentence on a juvenile offender who did not commit homicide." ( Id. at p. 82, 130 S.Ct. 2011 .)

Central to the high court's analysis was its "consideration of the culpability of the offenders at issue in light of their crimes and characteristics, along with the severity of the punishment in question." ( Graham , supra , 560 U.S. at p. 67, 130 S.Ct. 2011 .) As for culpability, the high court reiterated its observations in Roper that "[a]s compared to adults, juveniles have a ' "lack of maturity and an underdeveloped sense of responsibility" '; they 'are more vulnerable or susceptible to negative influences and outside pressures, including peer pressure'; and their characters are 'not as well formed.' [Citation.] These salient characteristics mean that '[i]t is difficult even for expert psychologists *366 to differentiate between the juvenile offender whose crime reflects unfortunate yet transient immaturity, and the rare juvenile offender whose *258 crime reflects irreparable corruption.' " ( Graham , at p. 68, 130 S.Ct. 2011 , quoting Roper , supra , 543 U.S. at pp. 569-570, 573, 125 S.Ct. 1183 .) Further, the high court underscored that "defendants who do not kill, intend to kill, or foresee that life will be taken are categorically less deserving of the most serious forms of punishment than are murderers. ... Although an offense like robbery or rape is 'a serious crime deserving serious punishment,' those crimes differ from homicide crimes in a moral sense." ( Graham , at p. 69, 130 S.Ct. 2011 , citations omitted.)

As for the punishment, the high court noted that a sentence of LWOP "deprives the convict of the most basic liberties without giving hope of restoration." ( Graham , supra , 560 U.S. at pp. 69-70, 130 S.Ct. 2011 ; see id. at 130 S.Ct. 2011 ["this sentence 'means denial of hope; it means that good behavior and character improvement are immaterial ....' "].) In addition, "[l]ife without parole is an especially harsh punishment for a juvenile. Under this sentence a juvenile offender will on average serve more years and a greater percentage of his life in prison than an adult offender. ... This reality cannot be ignored." ( Ibid. )

The high court then evaluated the sentence against the penological goals of "retribution, deterrence, incapacitation, and rehabilitation." ( Graham , supra , 560 U.S. at p. 71, 130 S.Ct. 2011 .) Linking retribution to culpability, the high court said " 'the case for retribution is not as strong with a minor as with an adult' " and "becomes even weaker with respect to a juvenile who did not commit homicide."

**453 ( Ibid. ) As for deterrence, the high court said that "[b]ecause juveniles' 'lack of maturity and underdeveloped sense of responsibility ... often result in impetuous and ill-considered actions and decisions,' [citation], they are less likely to take a possible punishment into consideration when making decisions." ( Id. at p. 72, 130 S.Ct. 2011 .)

As for incapacitation, the high court acknowledged that "[r]ecidivism is a serious risk to public safety, and so incapacitation is an important goal." ( Graham , supra , 560 U.S. at p. 72, 130 S.Ct. 2011 .) But the "characteristics of juveniles" make it "questionable" to conclude that a juvenile offender is incorrigible; indeed, " 'incorrigibility is inconsistent with youth.' " ( Id. at pp. 72-73, 130 S.Ct. 2011 .) A sentencing authority may not make a judgment "at the outset" that a juvenile nonhomicide offender will "be a risk to society for the rest of his life." ( Id. at p. 73, 130 S.Ct. 2011 .) This was true even for Graham, who had violated the terms of his probation "despite his own assurances of reform" and had engaged in "what the trial court described as an 'escalating pattern of criminal conduct.' " ( Ibid. ) "A life without parole sentence improperly denies the juvenile offender a chance to demonstrate growth and maturity." ( Ibid. )

The high court then discussed rehabilitation and explained that LWOP "forswears altogether the rehabilitative ideal. By denying the defendant the *367 right to reenter the community, the State makes an irrevocable judgment about that person's value and place in society. This judgment is not appropriate in light of a juvenile nonhomicide offender's capacity for change and limited moral culpability." ( Graham , supra , 560 U.S. at p. 74, 130 S.Ct. 2011 .) The high court also noted that inmates sentenced to LWOP "are often denied access to vocational training and other rehabilitative services that are available to other inmates," making "all the more evident" the disproportionality of LWOP when imposed on "juvenile offenders, who are most in need of and receptive to rehabilitation." ( Ibid. ) "In sum," Graham concluded, "penological theory is not adequate to justify life without parole for juvenile nonhomicide offenders." ( Ibid. ) *259 C.

What emerges from Graham is not a constitutional prohibition on harsh sentences for juveniles who commit serious crimes. ( Graham , supra , 560 U.S. at p. 71, 130 S.Ct. 2011 ["Society is entitled to impose severe sanctions on a juvenile nonhomicide offender to express its condemnation of the crime and to seek restoration of the moral imbalance caused by the offense."].) Nor does Graham "require the State to release [a juvenile nonhomicide] offender during his natural life. Those who commit truly horrifying crimes as juveniles may turn out to be irredeemable, and thus deserving of incarceration for the duration of their lives." ( Id. at p. 75, 130 S.Ct. 2011 .) But Graham "does prohibit States from making the judgment at the outset that those offenders never will be fit to reenter society." ( Ibid. ) "What the State must do ... is give defendants like Graham some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." ( Ibid. )

While finding LWOP impermissible for juvenile nonhomicide offenders, the high court did not define the maximum length of incarceration before parole eligibility that would be permissible in light of the concerns it set forth in Graham . But the high court made clear the nature of its concerns: A lawful sentence must recognize "a juvenile nonhomicide offender's capacity for change and limited moral culpability." ( Graham , supra , 560 U.S. at p. 74, 130 S.Ct. 2011 .) A lawful sentence must offer "hope of restoration" ( id. at 130 S.Ct. 2011 ), "a chance to demonstrate maturity and reform" ( id. at 130 S.Ct. 2011 ), a "chance for fulfillment outside prison walls," and a "chance for reconciliation with society" ( ibid. ). A lawful sentence must offer "the opportunity to achieve maturity of judgment and self-recognition of human worth and potential." ( Ibid. ) A lawful sentence must offer the juvenile offender an "incentive to become a responsible individual." ( Ibid. )

Although the Attorney General says a penalty is not invalid under Graham unless it "is tantamount to [a] sentence of **454 death," he does not *368 seriously contend that a term-of-years sentence with parole eligibility at any point before the end of life expectancy-whether it is one year, one month, or one day-would satisfy the Eighth Amendment. Even assuming defendants' parole eligibility dates are within their expected lifespans, the chance for release would come near the end of their lives; even if released, they will have spent the vast majority of adulthood in prison. We agree with the Court of Appeal that these sentences "tend to reflect a judgment Rodriguez and Contreras are irretrievably incorrigible" and "fall[ ] short of giving them the realistic chance for release contemplated by Graham ."

Several considerations support this conclusion. First, although the high court has not defined what it means for a juvenile offender "to rejoin society" ( Graham , supra , 560 U.S. at p. 79, 130 S.Ct. 2011 ), the language of Graham suggests that the high court envisioned more than the mere act of release or a de minimis quantum of time outside of prison. Graham spoke of the chance to rejoin society in qualitative terms-"the rehabilitative ideal" ( id. at 130 S.Ct. 2011 )-that contemplate a sufficient period to achieve reintegration as a productive and respected member of the citizenry. The "chance for reconciliation with society" ( id. at 130 S.Ct. 2011 ), "the right to reenter the community" ( id. at 130 S.Ct. 2011 ), and the opportunity to reclaim one's "value and place in society" ( ibid. ) all indicate concern for a measure of belonging and redemption that goes beyond mere freedom from confinement. It is also significant that *260 Graham said juvenile nonhomicide offenders should not be denied access to "vocational training" and "education," among other rehabilitative services. ( Id. at pp. 74, 79, 130 S.Ct. 2011 .) Presumably one purpose of such programming is to enable a juvenile offender to hold a job or otherwise participate as a productive member of society if released. Graham 's directive that "[t]he juvenile should not be deprived of the opportunity to achieve ... self-recognition of human worth and potential" implies that the juvenile may someday have the opportunity to realize that "potential." ( Id. at p. 79, 130 S.Ct. 2011 .) For any individual released after decades of incarceration, adjusting to ordinary civic life is undoubtedly a complex and gradual process. Confinement with no possibility of release until age 66 or age 74 seems unlikely to allow for the reintegration that Graham contemplates.

Second, in underscoring the capacity of juveniles to change, Graham made clear that a juvenile offender's prospect of rehabilitation is not simply a matter of outgrowing the transient qualities of youth; it also depends on the incentives and opportunities available to the juvenile going forward. (See, e.g., Graham , supra , 560 U.S. at p. 79, 130 S.Ct. 2011 [prison system may "become[ ] complicit in the lack of development" of a juvenile offender by "withhold[ing] counseling, education, and rehabilitation programs"].) Importantly, Graham said "[a] young person who knows that he or she has no chance to leave prison before life's end has little incentive to become a responsible individual." ( Ibid. ) We believe the same is true here: A young person who knows he or she has *369 no chance to leave prison for 50 years "has little incentive to become a responsible individual." ( Ibid. )

Third, a sentence of 50 years to life imprisonment bears an attenuated relationship to legitimate penological goals under the reasoning of Graham . Such a sentence, though less harsh than LWOP, is still "an especially harsh punishment for a juvenile," who "will on average serve more years and a greater percentage of his life in prison than an adult offender." ( Graham , supra , 560 U.S. at p. 70, 130 S.Ct. 2011 .) It is also a highly severe punishment for a juvenile nonhomicide offender who, "when compared to an adult murderer," has "a twice diminished moral culpability." ( Id. at p. 69, 130 S.Ct. 2011 ; cf. § 190, subd. (a) [penalty for adult who commits first degree murder simpliciter is 25 years to life].) The retributive case for a 50-years-to-life sentence, as for LWOP, is weakened by the juvenile nonhomicide offender's "age ... and the nature of the crime." ( Graham , at p. 69, 130 S.Ct. 2011 .) As for deterrence, Graham 's observation that juveniles have limited ability to consider consequences when making decisions ( id. at 130 S.Ct. 2011 ) applies to a sentence of 50 years to life just as it does to a sentence of LWOP. And as for incapacitation, a judgment that a juvenile **455 offender will be incorrigible for the next 50 years is no less "questionable" than a judgment that the juvenile offender will be incorrigible "forever." ( Id. at pp. 72-73, 130 S.Ct. 2011 ; see Montgomery , supra , 577 U.S. at p. ----, [136 S.Ct. at p. 736] [" Miller 's central intuition" is "that children who commit even heinous crimes are capable of change"].) Finally, as noted, a sentence of 50 years to life "cannot be justified by the goal of rehabilitation" because it offers a juvenile offender "little incentive to become a responsible individual." ( Graham , at pp. 74, 79, 130 S.Ct. 2011 .)

Fourth, our conclusion that a sentence of 50 years to life is functionally equivalent to LWOP is consistent with the decisions of other state high courts. Setting aside courts that have disagreed with our case law holding that Graham and Miller apply *261 to aggregated sentences (see Franklin , supra , 63 Cal.4th at p. 276, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 ; Caballero , supra , 55 Cal.4th at pp. 267-268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ), we are not aware of any state high court that has found incarceration of a juvenile for 50 years or more before parole eligibility to fall outside the strictures of Graham and Miller . (See State v. Zuber (2017) 227 N.J. 422 , 152 A.3d 197 , 212 [110-year sentence with parole eligibility after 55 years "is the practical equivalent of life without parole"]; Casiano , supra , 115 A.3d at p. 1044 [same for 50-year sentence] ; Bear Cloud v. State of Wyoming (Wyo. 2014) 334 P.3d 132 , 142 [same for 45-years-to-life sentence]; Null , supra , 836 N.W.2d at p. 71 [same for 75-year sentence with parole eligibility after 52.5 years]; but cf. Collins v. State (Fla.Ct.App. 2016) 189 So.3d 342 , 343 [55-year sentence with parole eligibility after 52 years does not violate Graham ]; United States v. Mathurin (11th Cir. 2017) 868 F.3d 921 , 934-936 [57-year sentence, which defendant could reduce to a near-50-year sentence by earning good-time credits, does not violate Graham ].) *370 Finally, our conclusion is also consistent with state legislation adopted in the wake of Graham and Miller , assuming that the parole hearings in these statutory schemes provide for meaningful consideration of the inmate's age at the time of the offense and demonstrated maturity and rehabilitation. (See Ark. Code Ann. § 16-93-621 (a)(1) [juvenile nonhomicide offenders eligible for parole after 20 years]; Colo. Rev. Stat. Ann. § 18-1.3-401 (4)(c)(I)(B) [juvenile offenders sentenced to LWOP for a crime other than first degree murder resentenced to life with opportunity for parole after 40 years]; Conn. Gen. Stat. Ann. § 54 -125a(f)(1) [juvenile offenders sentenced to over 50 years eligible for parole after 30 years, and juvenile offenders sentenced to between 10 and 50 years eligible for parole after the greater of 12 years or 60% of the sentence]; Del. Code Ann. tit. 11, § 4204A(d) [juvenile offender convicted of a crime other than first degree murder eligible for resentencing after 20 years]; D.C. Code Ann. § 24-403.03 (a) [juvenile offenders eligible for sentence reduction after 20 years]; Fla. Stat. Ann. § 921.1402 (2)(d) [juvenile offenders convicted of offenses other than murder entitled to review of sentence after 20 years]; La. Rev. Stat. § 15:574.4(D)(1) [juvenile offenders sentenced to life for crimes other than first or second degree murder eligible for parole after 30 years]; Sen. Bill No. 16 (La. 2017 Reg. Sess.) [juvenile offenders sentenced to life for crimes other than first or second degree murder eligible for parole after 25 years, effective August 2017]; Mo. Ann. Stat. § 558.047 (1) [juvenile offenders sentenced to LWOP eligible for review of sentence after 25 years]; Nev. Rev. Stat. Ann. § 213.12135 [juvenile nonhomicide offenders eligible for parole after 15 years]; House Bill No. 1195 (N.D. 2017 Reg. Sess.) [juvenile offenders eligible for sentence reduction after 20 years]; W.Va. Code § 61-11-23(b) [juvenile offenders eligible for parole after 15 years]; Wyo. Stat. Ann. § 6-10-301 (c) [juvenile offenders sentenced to life eligible for parole after 25 years]; but see Wash. Rev. Code § 9 .94A.730(1) [juvenile offenders eligible for release after 20 years, except for those serving sentences for aggravated first degree murder or certain sex offenses].) In enacting these sentencing reforms, these state legislatures observed that sentencing juvenile nonhomicide offenders to 50 or more years of incarceration without parole eligibility is not consistent with Graham . (See, e.g., Sen. Bill No. 294 (Ark. 2017 Reg. Sess.) § 2; **456 Colo. Rev. Stat. Ann. § 16-13-1001 ; Sen. Judiciary Com., Summary of Sen. Bill No. 796 (Conn. 2015 Reg. Sess.) § 1; Synopsis of Sen. Bill No. 9 (Del. *262 2013-2014 Reg. Sess.); House Judiciary Com., Crim. J. Subcom., Analysis of Sen. Bill No. 384 (Fla. 2014 Reg. Sess.) Jan. 3, 2014, pp. 1-4; Resume Dig. for Sen. Bill No. 317 (La. 2012 Reg. Sess.); Resume Dig. for Sen. Bill No. 16 (La. 2017 Reg. Sess.).) *371 D.

The Chief Justice criticizes our decision today as an "unwarranted extension of Graham ." (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 273, 411 P.3d at p. 465.) She observes that " Graham ... invalidated a narrowly defined, specific type of sentence" for juvenile nonhomicide offenders-namely, life without parole, " ' "the second most severe penalty permitted by law." ' " ( Id. at p. ----, quoting Graham , supra , 560 U.S. at p. 69, 130 S.Ct. 2011 .) Our decision, she contends, ignores "the limited nature of the holding in Graham " and disregards the " 'clear line' " that Graham drew in demarcating the type of sentence that violates the Eighth Amendment. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 273, 277, 284, 284 & fn. 7, 286, 411 P.3d at pp. 465, 469, 474, 475 & fn. 7, 476, quoting Graham , at 74, 130 S.Ct. 2011 ).

But what exactly is the "clear line" that Graham drew? Here is the passage where those words appear in Graham : "[P]enological theory is not adequate to justify life without parole for juvenile nonhomicide offenders. This determination; the limited culpability of juvenile nonhomicide offenders; and the severity of life without parole sentences all lead to the conclusion that the sentencing practice under consideration is cruel and unusual. This Court now holds that for a juvenile offender who did not commit homicide the Eighth Amendment forbids the sentence of life without parole. This clear line is necessary to prevent the possibility that life without parole sentences will be imposed on juvenile nonhomicide offenders who are not sufficiently culpable to merit that punishment. Because '[t]he age of 18 is the point where society draws the line for many purposes between childhood and adulthood,' those who were below that age when the offense was committed may not be sentenced to life without parole for a nonhomicide crime." ( Graham , supra , 560 U.S. at pp. 74-75, 130 S.Ct. 2011 .)

The Chief Justice reads the phrase "clear line" to distinguish between LWOP and other types of sentences. But in context, the phrase more sensibly refers to two other distinctions: (1) between homicide and nonhomicide offenses, and (2) between juvenile and adult offenders. The "line" that Graham made "clear" is that LWOP may not be imposed on juveniles who commit nonhomicide offenses, even if it may be imposed (rarely) on juveniles who commit homicide offenses or on adults who commit nonhomicide offenses. In drawing this line, the majority in Graham was rejecting Chief Justice Roberts's view that the Eighth Amendment does not support a "categorical rule that juveniles may never receive a sentence of life without parole for nonhomicide crimes" and instead "allow[s] courts ... to consider the particular defendant and particular crime at issue." ( Graham , supra , 560 U.S. at pp. 89, 86, 130 S.Ct. 2011 (conc. opn. of Roberts, C. J.); see id. at pp. 93-95, 130 S.Ct. 2011 [arguing that some juvenile nonhomicide offenders may deserve an LWOP sentence].) Graham does not hold or suggest that only LWOP sentences, and *372 no sentences other than LWOP, violate the Eighth Amendment when imposed on a juvenile nonhomicide offender.

Indeed, our dissenting colleagues do not contend that the reasoning of Graham is limited to LWOP sentences, for we have already rejected that proposition in Caballero . The Attorney General argued in Caballero that "a cumulative sentence for *263 distinct crimes does not present a cognizable Eighth Amendment claim .... In addition, the Court of Appeal reasoned that Graham applied a categorical rule specifically limited to juvenile nonhomicide offenders receiving an explicitly designated life without parole sentence ...." ( Caballero , supra , 55 Cal.4th at p. 267, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) At the time we decided Caballero , several appellate courts had held that Graham applies **457 only to LWOP sentences and not to any individual or aggregate term-of-years sentences. (See Bunch v. Smith (6th Cir. 2012) 685 F.3d 546 , 552 ; Henry v. State (Fla.Ct.App. 2012) 82 So.3d 1084 , 1089 ; State v. Kasic (Ariz.Ct.App. 2011) 228 Ariz. 228 , 265 P.3d 410 , 415.) Notwithstanding these arguments and authorities, we unanimously held that Graham 's reasoning applies to a "term-of-years sentence that amounts to the functional equivalent of a life without parole sentence." ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ; see id. at pp. 271-273, 145 Cal.Rptr.3d 286 , 282 P.3d 291 (conc. opn. of Werdegar. J.).)

As the Chief Justice acknowledges, the "line" that Graham actually drew between lawful and unlawful sentences for juvenile nonhomicide offenders is not between LWOP and other sentences, but between sentences that do and sentences that do not provide " 'some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation.' " (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 277, 411 P.3d at p. 469, quoting Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 .) Whatever "abstraction," "vagueness," or "subjectiv[ity]" (dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 285, 411 P.3d at p. 475) there may be in analyzing whether a particular sentence provides "some meaningful opportunity to obtain release" ( Graham , at p. 75, 130 S.Ct. 2011 ), we are obligated to apply the rule stated by the high court, and that is what our opinion today does.

The Chief Justice would hold that a sentence provides a meaningful opportunity for release if it allows for parole eligibility within a defendant's life expectancy. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 280-281, 411 P.3d at pp. 471-472.) This approach is problematic for reasons we have explained above. ( Ante , at pp. ---- - ----.) The Chief Justice does not dispute that the life expectancy tables she relies on show significant disparities by race and gender. Her response is that apart from race, sex, and custodial status, "juvenile defendants belong to a nearly infinite number of cohorts" with varying life expectancies. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 282, 411 P.3d at p. 472.) She then says: "Given that a defendant could be placed within any of many peer groups for purposes of assessing his or her life expectancy, and given as well *373 the need to use some conception of life expectancy as a benchmark, reliance on general population life expectancies makes good sense as providing an administrable rule of decision that is consistent with Graham ." ( Id. at p. ----.) This is a non-sequitur. Why does reliance on general-population life expectancies make good sense when it is acknowledged that life expectancies vary by race, sex, custodial status, and other traits as well? Such an approach seems quite arbitrary.

Even if general-population life expectancies were relevant to evaluating whether a particular sentence provides a meaningful opportunity for release, the Chief Justice does not answer the crucial question of how many years before the end of a defendant's life expectancy must parole eligibility be provided in order to satisfy Graham . The Chief Justice believes five years is sufficient. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 281, 282, 411 P.3d at pp. 472, 473 [parole eligibility at age 74 falls "well within" the general life *264 expectancy of 79 years for 15- to 16-year-olds].) But why is five years sufficient? Why not require 10, 15, or 25 years? And if five years is sufficient, then what about four years? three? two? or one?

Ultimately, any line-drawing must depend on a considered judgment as to whether the parole eligibility date of a lengthy sentence offers a juvenile offender a realistic hope of release and a genuine opportunity to reintegrate into society. Reasonable minds may disagree on such judgments, but it is specious to contend that an approach based on life expectancy would avoid "subjective and quite likely divergent assessments of what constitutes adequate reintegration into society, and the time necessary to accomplish this reentry." (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 264, 411 P.3d at p. 476.) In the end, the Chief Justice's conclusion that defendants' sentences are lawful rests on her view that "profound life experiences still may lie ahead of someone **458 released from prison at age 66 or 74." ( Id. at p. ----.) Whatever the merits of this view, the analysis that underlies it is not more "objective," more "workable," or more conducive to drawing a " 'clear line' " ( id. at pp. ----, ----) than the analysis set forth in our opinion today. Indeed, the Chief Justice's approach calls for the very sort of line-drawing she purports to disavow: Under her approach as under ours, the controlling inquiry is not simply whether defendants' sentences provide for parole eligibility within their life expectancies, but whether the sentences "impinge on the same substantive concerns that make the imposition of LWOP on juvenile nonhomicide offenders impermissible under the Eighth Amendment." ( Ante , at p. ----.)

III.

After oral argument in this case, the Governor on October 11, 2017, signed into law Assembly Bill No. 1448 and Senate Bill No. 394. Assembly Bill *374 1448 codifies the Elderly Parole Program, under which prisoners age 60 or older who have served at least 25 years in prison are entitled to a parole hearing. (Assem. Bill No. 1448 (2017-2018 Reg. Sess.) § 3.) Senate Bill 394 extends eligibility for a youth offender parole hearing after 25 years of incarceration to a person who was convicted of certain controlling offenses committed before 18 years of age and sentenced to life without the possibility of parole. (Sen. Bill No. 394 (2017-2018 Reg. Sess.) § 1.) In addition, upon the passage of Proposition 57 in the November 2016 elections, the California Department of Corrections and Rehabilitation (CDCR) issued new regulations governing the ability of inmates to earn custody credit to advance their parole dates. We vacated submission of this case and ordered supplemental briefing from the parties on what bearing, if any, Assembly Bill 1448, Senate Bill 394, or the regulations codified at sections 3043, 3043.2, 3043.3, 3043.4, 3043.5, and 3043.6 of title 15 of the California Code of Regulations have on the question presented.

The Chief Justice contends that regardless of whether defendants' original sentences are valid, the recent legislation authorizing elderly parole means "both defendants will have an opportunity for parole at age 60," and "[a] sentence offering an opportunity for parole no later than age 60 is not invalid under Graham ." (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 286-287, 411 P.3d at p. 477.) Further, she asserts, "even without the Elderly Parole Program, Rodriguez may be eligible for parole when he is 57 years old, simply by earning good-conduct credits" ( id. at 229 Cal.Rptr.3d at pp. 286-287, 411 P.3d at p. 477), and "Contreras could advance his initial parole date to age 64 through good *265 conduct" ( id. at 229 Cal.Rptr.3d at p. 294, 411 P.3d at p. 482). As explained below, we decline to resolve whether the newly enacted legislation and regulations affect the validity of defendants' sentences and instead leave these novel issues for the lower courts to address in the first instance.

A.

The elderly parole statute provides that when considering the release of an eligible inmate, the Board of Parole Hearings (Board) "shall give special consideration to whether age, time served, and diminished physical condition, if any, have reduced the elderly inmate's risk for future violence." (§ 3055, subd. (c).) A key question is whether an elderly parole hearing offers a juvenile offender a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation ." ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 , italics added.)

The legislative history of Assembly Bill 1448 indicates that the legislation's main purpose was to curb rising medical costs of the geriatric inmate population and to provide a "compassionate" release for those elderly individuals. (Assem. Concurrence in Sen. Amends. to Assem. Bill No. 1448 (2017-2018 *375 Reg. Sess.) Sept. 11, 2017.) In contrast to the statute authorizing youth offender parole hearings, the text of the elderly parole statute does not mention youth-related considerations or rehabilitation. (Compare § 3051, subd. (f)(1) with § 3055.)

The Attorney General contends that elderly parole hearings are governed by section 4801, subdivision (c) and are thus required to consider youth-related factors associated **459 with the controlling offense. Section 4801, subdivision (c) says: "When a prisoner committed his or her controlling offense, as defined in subdivision (a) of Section 3051, when he or she was 25 years of age or younger, the board, in reviewing a prisoner's suitability for parole pursuant to Section 3041.5, shall give great weight to the diminished culpability of youth as compared to adults, the hallmark features of youth, and any subsequent growth and increased maturity of the prisoner in accordance with relevant case law." Noting that the provisions for parole hearings set forth in section 3041.5 apply to "all hearings for the purpose of reviewing an inmate's parole suitability" (§ 3041.5, subd. (a) ), the Attorney General argues that "they necessarily therefore apply to parole consideration under the Elderly Parole Program."

But it is questionable whether the Board is reviewing an inmate's suitability for parole "pursuant to Section 3041.5" (§ 4801, subd. (c) ) when it conducts an elderly parole hearing. The elderly parole statute contains a provision that makes applicable section 3041.5, subdivision (b)(3)'s schedule for a subsequent parole hearing in the event of a parole denial (§ 3055, subd. (f) ) and another provision stating that "when considering a request for an advance hearing pursuant to subdivision (d) of Section 3041.5, the board shall consider whether the inmate meets or will meet the criteria [for the Elderly Parole Program]" (§ 3055, subd. (d) ). These provisions, which appear to treat section 3041.5's parole procedures as separate and distinct from those in section 3055, suggest that an elderly parole hearing is conducted pursuant to section 3055, not pursuant to section 3041.5.

The Chief Justice does not endorse the Attorney General's interpretation of the statute and instead asserts that "the decision whether to grant elderly parole is concerned with the same question of public safety that governs conventional parole hearings." (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at p. 287, 411 P.3d at p. 477.) At conventional parole hearings, " '[a]ll relevant, reliable information *266 available to the panel shall be considered in determining suitability for parole. Such information shall include the circumstances of the prisoner's: social history; past and present mental state; ... past and present attitude toward the crime; ... and any other information which bears on the prisoner's suitability for release.' ( Cal. Code Regs., tit. 15, § 2281, subd. (b).)" ( Id. at 229 Cal.Rptr.3d at p. 287-288, 411 P.3d at p. 477, fn.omitted.) She contends that "[a]lthough in an elderly parole hearing 'special *376 consideration' is given to the three factors specified in section 3055, subdivision (c), there is no suggestion that these 'special' considerations somehow skew the basic question before the panel." ( Id. at 229 Cal.Rptr.3d at p. 288, 411 P.3d at p. 478.)

But the Chief Justice's interpretation is not the only plausible reading of the elderly parole statute, and we decline to issue a definitive interpretation less than five months after the statute's enactment, before any Court of Appeal has filed a published opinion applying it in the context of juvenile sentencing, and before CDCR has adopted any implementing regulations. We are not certain, for example, that the statute would preclude CDCR from adopting regulations that focus the Elderly Parole Program on identifying those inmates who no longer pose a risk of future violence primarily because of their age, illness, or other physical incapacitation, while leaving all other inmates age 60 or older who may be suitable for parole to the ordinary parole process. Such an interpretation does not appear foreclosed by the statutory text, and it seems consistent with the Legislature's purpose of reducing costs of geriatric care and providing compassionate release for elderly inmates. Yet it is questionable whether such a parole hearing would provide juvenile offenders with a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation ." ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 , italics added.) The record before us contains no information on how the Elderly Parole Program actually operates or what considerations, apart from the "special considerations" set forth in the statute (§ 3055, subd. (c) ), guide the Board's determination of suitability for elderly parole. This information may be developed on remand.

**460 The Chief Justice says such development is unnecessary, noting that we required no similar information before finding the availability of a youth offender parole hearing sufficient to moot the Eighth Amendment claim in Franklin . (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 290-291, 411 P.3d at pp. 479-480, citing Franklin , supra , 63 Cal.4th at pp. 284-286, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) But Franklin addressed legislation whose explicit and specific purpose is "to establish a parole eligibility mechanism that provides a person serving a sentence for crimes that he or she committed as a juvenile the opportunity to obtain release when he or she has shown that he or she has been rehabilitated and gained maturity, in accordance with the decision of the California Supreme Court in [ Caballero ] and the decisions of the United States Supreme Court in Graham ... and Miller .... It is the intent of the Legislature to create a process by which growth and maturity of youthful offenders can be assessed and a meaningful opportunity for release established." (Sen. Bill No. 260 (2013-2014 Reg. Sess.) § 1.) As noted, the statute expressly mandates consideration of youth-related factors in youth offender parole hearings. (§§ 3051, subd. (e), 4801, subd. (c).) For this reason, and because the statutes contemplate that "juvenile offenders [must] have an adequate opportunity to make a record of factors, including youth-related factors, relevant to the eventual *377 parole determination," we were assured "at this point" that a juvenile offender eligible for *267 such a hearing has a meaningful opportunity for release within the meaning of Graham . ( Franklin , at p. 286, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) Neither the text nor history of the elderly parole statute contains any indication that the Legislature intended elderly parole hearings to be responsive to the Eighth Amendment concerns raised by lengthy juvenile sentences.

Even assuming that elderly parole hearings consider normal parole factors, it is not clear that elderly parole eligibility after 44 years in prison would provide the 16-year-old nonhomicide offenders in this case with the "hope of restoration" and realistic opportunity to reintegrate into society that Graham requires. ( Graham , supra , 560 U.S. at p. 70, 130 S.Ct. 2011 .) The Chief Justice notes that Bear Cloud v. State , supra , 334 P.3d 132 invalidated a 45-year sentence for a 16-year-old nonhomicide offender, but that three other state high courts have held that parole eligibility at or around age 60 passes constitutional muster. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 291-293, 411 P.3d at pp. 481-482.) Among them, only Angel v. Commonwealth (2011) 281 Va. 248 , 704 S.E.2d 386 ( Angel ) concluded that a geriatric release program for inmates who are 60 or older satisfies Graham . The Virginia Supreme Court's holding was premised on its understanding that "the factors used in the normal parole consideration process apply to conditional release decisions under [Virginia's geriatric release] statute." ( Angel at p. 402,.)

Notably, in Virginia v. LeBlanc (2017) 582 U.S. ----, [ 137 S.Ct. 1726 ], 198 L.Ed.2d 186 ( LeBlanc ), the high court considered on habeas review whether Virginia's geriatric release program provides a meaningful opportunity for a juvenile nonhomicide offender to obtain release based on demonstrated maturity and rehabilitation. The trial court in LeBlanc , relying on Angel , rejected the defendant's Eighth Amendment challenge, and the high court held that the trial court's ruling was not objectively unreasonable. ( LeBlanc , at p. ---- [ 137 S.Ct. at p. 1729 ].) In so doing, the high court emphasized that it was applying the deferential standard of review required by the federal Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA) ( 28 U.S.C. § 2254 (d)(1) ) and that " '[t]here are reasonable arguments on both sides.' " ( LeBlanc , at p. ---- [ 137 S.Ct. at p. 1729 ].) On one hand, because Virginia's geriatric release program considers "normal parole factors," it "could allow the Parole Board to order a former juvenile offender's conditional release in light of his or her 'demonstrated maturity and rehabilitation.' " ( Ibid. ) On the other hand, there were concerns "that the Parole Board's substantial discretion to deny geriatric release deprives juvenile nonhomicide offenders a meaningful opportunity to seek parole and that juveniles cannot seek geriatric release until they have spent at least four decades in prison." ( Ibid. ) The high **461 court thus recognized there is a reasonable argument that even an elderly parole process that considers normal parole factors could, in practice, fail to provide a meaningful opportunity for *378 release and that incarcerating a juvenile nonhomicide offender for 40 years or more without parole eligibility is simply too long under Graham .

Defendants here raise an additional concern: Juvenile offenders for whom the Elderly Parole Program provides the first opportunity for release will invariably spend more time in prison before parole eligibility compared to adult inmates who committed the same crime and served at least 25 years before age 60-a result at odds with the high court's "conclusion in Roper v. Simmons , 543 U.S. 551 , 125 S.Ct. 1183 , 161 L.Ed.2d 1 (2005), that juvenile offenders are generally less culpable than adults who commit the same crimes." ( *268 Graham , supra , 560 U.S. at p. 86, 130 S.Ct. 2011 (conc. opn. of Roberts, C. J.); see Roper , supra , 543 U.S. at p. 570, 125 S.Ct. 1183 .) In Graham , the high court reasoned that "[l]ife without parole is an especially harsh punishment for a juvenile" because "a juvenile offender will on average serve more years and a greater percentage of his life in prison than an adult offender." ( Graham , at p. 70, 130 S.Ct. 2011 .) Defendants contend the same reasoning applies to a sentence of more than 40 years without parole eligibility. ( Id. at p. 71, 130 S.Ct. 2011 ["This reality cannot be ignored."].)

These issues are novel and substantial, and we leave them for the lower courts to address in the first instance. Like the high court in LeBlanc , we decline to resolve in this case whether the availability of an elderly parole hearing at age 60 for a juvenile nonhomicide offender satisfies the Eighth Amendment concerns set forth in Graham .

B.

Apart from defendants' eligibility for elderly parole, the Chief Justice claims that "simply by maximizing the good-conduct credits that are available" to them under Proposition 57, Rodriguez can advance his initial parole date to age 57 and Contreras can advance his initial parole date to age 64. (Dis. opn. of Cantil-Sakauye, C. J., post , at 229 Cal.Rptr.3d at pp. 293-294, 411 P.3d at pp. 482-483.) But as with elderly parole, no Court of Appeal has filed a published opinion addressing the relevance of Good Conduct Credit to the constitutionality of a juvenile sentence, and the regulations, promulgated less than one year ago, remain in emergency form. ( Cal. Code Regs., tit. 15, § 3043.2.) In addition, the record before us contains no information on how Good Conduct Credit operates in practice.

The Chief Justice rests her calculations on defendants' ability to earn the maximum amount of Good Conduct Credit, but neither she nor Justice Kriegler makes any mention of the myriad ways inmates can lose such credit. Good Conduct Credit is subject to forfeiture upon "a finding of guilt of a serious rule violation in accordance with section 3323." ( Cal. Code Regs., tit. 15, § 3043.2, subd. (c).) The activities that can constitute a "serious rule violation" span a *379 broad range of conduct. ( Id. , §§ 3315, 3323.) A "credit forfeiture of 61-90 days" is assessed for, among other violations, "[l]ate return from a temporary community leave" or "[f]ighting." ( Id. , § 3323, subds. (f)(7), (f)(9).) A "credit forfeiture of 31-60 days" is assessed for, among other violations, "damage to ... state property valued at less than $400," "[p]ossession of alcoholic beverages or intoxicating substances in a community-access facility under the jurisdiction of CDCR," or "[g]ambling." ( Id. , § 3323, subds. (g)(1), (g)(2), (g)(5).) A "credit forfeiture of 0-30 days" is assessed for, among other violations, "[m]isuse, alteration, unauthorized acquisition, or exchange of personal property, state funds, or state property" or "[h]arassment of another person, group, or entity." ( Id. , § 3323, subds. (h)(4), (h)(11); see also id. , § 3315, subd. (a)(3) [listing 27 offenses that qualify as a "serious rule violation," including "(G) Possession of five dollars or more without authorization" and "(H) Acts of ... disrespect which by reason of intensity or context create a potential for violence ...."].)

In positing an initial parole date at age 57 for Rodriguez and at age 64 for Contreras, our dissenting colleagues assume that correctional authorities will not revoke any Good Conduct Credit that defendants earn while **462 incarcerated for 40-plus years, citing select cases of inmates who have demonstrated good prison behavior (though none of them served anything close to 40 years). (See dis. opn. of Kriegler, J., post , at 229 Cal.Rptr.3d at pp. 300-301, 411 P.3d at pp. 488-489.) But the *269 record before us contains no information on how likely it is that an inmate can achieve a spotless prison record over a span of four or more decades. Nor is it clear that Graham 's requirement of a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation" ( id. at p. 75, 130 S.Ct. 2011 ) would be satisfied by a parole eligibility date that is contingent upon a perfect or near-perfect record in prison. (See Graham , supra , 560 U.S. at p. 70, 130 S.Ct. 2011 ["the remote possibility" of release does not satisfy the Eighth Amendment].) As with elderly parole, we leave these novel issues for the lower courts to address in the first instance.

IV.

For the reasons above, we agree with the Court of Appeal that defendants' sentences violate the Eighth Amendment under the standards articulated in Graham . We affirm the judgment of the Court of Appeal and remand these matters for resentencing. The sentencing court is directed to consider, in light of this opinion, any mitigating circumstances of defendants' crimes and lives, and the impact of any new legislation and regulations on appropriate sentencing. The sentencing court is further directed to impose a time by which defendants may seek parole, consistent with this opinion.

*380 Justice Kriegler says this disposition "is likely to leave the trial judge mystified" because the trial court already considered any mitigating circumstances of defendants' crime and lives in imposing their original sentences. (Dis. opn. of Kriegler, J., post , at 229 Cal.Rptr.3d at pp. 296-297, 411 P.3d at p. 485.) But the trial court did not undertake its sentencing analysis with the benefit of our opinion today. In addition, the trial court appeared to stray from the fundamental teaching of Graham when it said at Contreras's sentencing: "So somebody with that kind of psychology is not somebody I feel confident is going to rehabilitate, change, and become a different person regardless of his brain development. I think his brain is developed into who he is and who he was demonstrated on that whole event where he raped those two girls." (Cf. dis. opn. of Kriegler, J., post , at 229 Cal.Rptr.3d at p. 298, 411 P.3d at p. 486 [asserting that defendants' crimes "reveal[ ] the actions of violent sexual predators, not that of rogue youths misbehaving on a lark"].)

The trial court in Graham had similarly concluded that the 16-year-old defendant, a recidivist felon, was not capable of rehabilitation: " 'I don't see where I can do anything to help you any further. You've evidently decided this is the direction you're going to take in life, and it's unfortunate that you made that choice. [¶] ... Given your escalating pattern of criminal conduct, it is apparent to the Court that you have decided that this is the way you are going to live your life and that the only thing I can do now is to try and protect the community from your actions.' " ( Graham , supra , 560 U.S. at p. 57, 130 S.Ct. 2011 .) But the key holding of Graham is that "in light of a juvenile nonhomicide offender's capacity for change and limited moral culpability" ( id. at 130 S.Ct. 2011 ), no sentencing court is permitted to render a judgment "at the outset" that a juvenile nonhomicide offender is incorrigible ( id. at pp. 73, 75, 130 S.Ct. 2011 ). On remand, the sentencing of each defendant must be guided by the "central intuition" of the high court's case law in this area-"that children who commit even heinous crimes are capable of change." ( Montgomery , supra , 577 U.S. at p. ----, [136 S.Ct. at p. 736] ; see Miller , supra , 567 U.S. at p. 473, 132 S.Ct. 2455 ["none of what [ Graham ] said about children ... is crime-specific"].)

In so holding, we do not minimize the gravity of defendants' crimes or their lasting impact on the victims and their families.

*270 No one reading the disturbing facts of this case could disagree with the trial court that the crimes were "awful and shocking." The Court of Appeal was correct to observe that "[w]hatever their final sentences, Rodriguez and Contreras will need to do more than simply bide their time in prison to demonstrate **463 parole suitability. ... The record before us indicates Rodriguez and Contreras have much work ahead of them if they hope to one day persuade the Board they no longer present a current danger to society and should be released on parole."

Our dissenting colleagues further assert that our decision today provides "virtually no guidance" (dis. opn. of Cantil-Sakauye, C. J., *381 post , at 229 Cal.Rptr.3d at p. 273, 411 P.3d at p. 465) and "not a whiff of direction" (dis. opn. of Kriegler, J., post , at p. ----) on what length of sentence below 50 years will satisfy Graham . But in this context, we find it prudent to follow a "cardinal principle of judicial restraint-if it is not necessary to decide more, it is necessary not to decide more." ( PDK Laboratories Inc. v. U.S. Drug Enforcement Admin. (D.C. Cir. 2004) 362 F.3d 786 , 799 (conc. opn. of Roberts, J.).)

Today's decision, building on Caballero , elucidates Graham 's applicability to a term-of-years sentence, and our reasoning will inform the application of Graham by California courts going forward. Our disposition takes the approach we took in Caballero , where we unanimously declared the defendant's 110-years-to-life sentence unconstitutional and remanded for the sentencing court to "consider all mitigating circumstances attendant in the juvenile's crime and life ... so that it can impose a time when the juvenile offender will be able to seek parole from the parole board." ( Caballero , supra , 55 Cal.4th at pp. 268-269, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ; see id. at p. 273, 145 Cal.Rptr.3d 286 , 282 P.3d 291 (conc. opn. of Werdegar, J.).) No member of this court suggested that we should provide further guidance on what would constitute a lawful sentence. Instead, the court's opinion expressly stated that "we will not provide trial courts with a precise timeframe for setting these future parole hearings in a nonhomicide case." ( Id. at p. 269, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .)

As it turns out, our restraint in Caballero proved well-advised. Our opinion concluded with a footnote "urg[ing] the Legislature to enact legislation establishing a parole eligibility mechanism that provides a defendant serving a de facto life sentence without possibility of parole for nonhomicide crimes that he or she committed as a juvenile with the opportunity to obtain release on a showing of rehabilitation and maturity." ( Caballero , supra , 55 Cal.4th at p. 269, fn. 5, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) The Legislature responded the following year with Senate Bill No. 260. This legislation made it unnecessary for us to decide Eighth Amendment challenges to sentences of 25 years or more for a broad range of juvenile homicide and nonhomicide offenses; juvenile offenders serving such sentences are now entitled to a youth offender parole hearing during their 25th year of incarceration. (§ 3051, subd. (b)(3); see Franklin , supra , 63 Cal.4th at pp. 277-280, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 ; cf. id. at pp. 284-286, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 [leaving undecided whether youth offender parole hearings, "in practice," will conform to applicable statutory and constitutional law].) In addition, whereas Senate Bill No. 260 made youth offender parole hearings available for juveniles who committed their controlling offense before age 18 (Stats. 2013, ch. 312, § 5), the Legislature has since amended the age threshold to age 23 (Stats. 2015, ch. 471, § 2) and now to age 25 (Stats. 2017, ch. 684, § 2.5 [eff. Jan. 1, 2018] ). Moreover, the Legislature's enactment of Senate Bill 394 just a few months ago *271 extended youth offender parole hearings in the 25th year of incarceration to juveniles serving an LWOP sentence. (§ 3051, subd. (b)(4).) One Strike offenders remain ineligible for youth offender parole hearings. (§ 3051, subd. (h).) But in light of *382 the changing statutory landscape, we see no reason to opine here on constitutional and statutory issues that may be rendered moot by further legislative action.

Finally, we note defendants' contention that the current treatment of juvenile One Strike offenders is anomalous given that juveniles convicted of special circumstance murder and sentenced to LWOP are now eligible for parole during their 25th year in prison. This scheme appears at odds with the high court's observation that "defendants who do not kill, intend to kill, or foresee that life will be taken are categorically less deserving of the most serious forms of punishment than are murderers. ... Although an **464 offense like robbery or rape is 'a serious crime deserving serious punishment,' those crimes differ from homicide crimes in a moral sense." ( Graham , supra , 560 U.S. at p. 69, 130 S.Ct. 2011 , citations omitted.) In the death penalty context, the high court has said "there is a distinction between intentional first-degree murder on the one hand and nonhomicide crimes against individual persons, even including child rape, on the other. The latter crimes may be devastating in their harm, as here, but 'in terms of moral depravity and of the injury to the person and to the public,' they cannot be compared to murder in their 'severity and irrevocability.' " ( Kennedy v. Louisiana (2008) 554 U.S. 407 , 438, 128 S.Ct. 2641 , 171 L.Ed.2d 525 , citation omitted.)

The parties point to no other provision of our Penal Code, and we are aware of none, that treats a nonhomicide offense more harshly than special circumstance murder. (Compare § 190.2 [prescribing penalty of death or LWOP for special circumstance murder] with § 667.61 [prescribing maximum penalty of 25 years to life or, when the victim is under age 14, LWOP for aggravated rape offenses].) We are also unaware of any other jurisdiction that punishes juveniles for aggravated rape offenses more severely than for the most aggravated forms of murder. Further, we note the concern raised by amicus curiae PJDC that if defendants had killed their victims after the sexual assaults and had been sentenced to LWOP, they would have been eligible for a youth offender parole hearing after 25 years of incarceration. (Cf. Kennedy v. Louisiana , supra , 554 U.S. at p. 445, 128 S.Ct. 2641 ["[B]y in effect making the punishment for child rape and murder equivalent, a State that punishes child rape by death may remove a strong incentive for the rapist not to kill the victim."].)

Defendants contend that this treatment of juvenile One Strike offenders violates principles of equal protection and the Eighth Amendment. There is also a colorable claim that it constitutes "unusual punishment" within the meaning of article I, section 17 of the California Constitution. As with the other issues arising from new legislation, we decline to resolve these contentions here. It suffices to note, as we did in Caballero , that the current penal scheme for juveniles may warrant additional legislative attention.

*383 CONCLUSION

We affirm the judgment of the Court of Appeal and remand these matters for resentencing. The sentencing court is directed to consider, in light of this opinion, any mitigating circumstances of defendants' crimes and lives, and the impact of any new legislation and regulations on appropriate sentencing. The sentencing court is further directed to impose a time by which *272 defendants may seek parole, consistent with this opinion.

WE CONCUR:

CHIN, J.

CUÉLLAR, J.

KRUGER, J.

CANTIL-SAKAUYE, C. J.

DISSENTING OPINION BY CANTIL-SAKAUYE, C. J.

I respectfully dissent. The majority's erroneous interpretation and extension of Graham v. Florida (2010) 560 U.S. 48 , 130 S.Ct. 2011 , 176 L.Ed.2d 825 ( Graham ) yield a result the Graham court did not intend-the categorical condemnation of all sentences in which juvenile offenders convicted of nonhomicide crimes will serve a term of 50 years or greater. At the same time, the majority fails to properly account for legislation and regulations that afford defendants William Rodriguez and Leonel Contreras an initial opportunity for parole no later than when they reach the age of 60. These measures take defendants' sentences outside of Graham 's purview even under the majority's mistaken approach to that decision. Defendants' sentences do not violate the Eighth Amendment to the United States Constitution, and I would so hold.

In Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , the high court invalidated a particular type of prison sentence-one of life imprisonment without the possibility of parole (life without parole)-when imposed upon a juvenile **465 convicted only of a nonhomicide crime or crimes. The court took great care in describing the type of sentence it considered "cruel and unusual" under the Eighth Amendment. ( U.S. Const., 8th Amend.) The majority in Graham characterized life without parole as " 'the second most severe penalty permitted by law.' " ( Graham , at p. 69, 130 S.Ct. 2011 , quoting Harmelin v. Michigan (1991) 501 U.S. 957 , 1001, 111 S.Ct. 2680 , 115 L.Ed.2d 836 (conc. opn. of Kennedy, J.).) A life without parole sentence, the court stressed, "alters the offender's life by a forfeiture that is irrevocable. It deprives the convict of the most basic liberties without giving hope of restoration, except perhaps by executive clemency." ( Graham , at pp. 69-70.) Such a sentence " 'means that ... [the convict] will remain in prison for the rest of his days.' " ( Id., at 130 S.Ct. 2011 , quoting Naovarath v. State (Nev. 1989) 105 Nev. 525 , 779 P.2d 944 , 944.) "Life in prison without the possibility of parole," the Graham court emphasized, "gives no chance for fulfillment outside prison walls, no chance for reconciliation with society, no hope." ( Graham , at p. 79, 130 S.Ct. 2011 .)

Today, the majority declares unconstitutional a range of sentences that most certainly are not the second most severe penalty permitted by law; that do *384 offer hope of restoration of basic liberties; that do not necessarily mean that defendants will remain in prison for the rest of their days; and that do give a chance for fulfillment outside prison walls, do give a chance for reconciliation with society, and do offer hope. In short, the majority extends Graham to invalidate an array of sentences that are qualitatively different from the sort of punishment that Graham was concerned with.

The majority asserts, unconvincingly, that behind Graham 's cautious and consistent phrasing lies a more far-reaching intent to invalidate all sentences that do not provide juvenile offenders convicted of nonhomicide crimes with an opportunity for parole at an age when release would, in the majority's view, be sufficiently conducive to their full reintegration into society. This reading of Graham is flawed on several *273 levels. It is inconsistent with the careful, incremental approach the high court has taken when addressing categorical Eighth Amendment challenges to sentencing practices. It defies the Graham court's articulations of its subject and holding, and represents an inadequately justified extension of that decision. It departs from this court's prior description of Graham as demanding that a juvenile offender convicted of a nonhomicide crime must receive " 'some realistic opportunity to obtain release' from prison during his or her expected lifetime ." ( People v. Caballero (2012) 55 Cal.4th 262 , 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ( Caballero ), italics added, quoting Graham , supra , 560 U.S. at p. 82, 130 S.Ct. 2011 .) And it unnecessarily premises a constitutional rule on the majority's subjective and speculative views regarding the time frame necessary to have a meaningful postcustodial life. The result is a dubious judicial incursion into the legislative sphere, pitched at such a high level of abstraction that it provides sentencing courts with virtually no guidance for determining whether a lengthy prison sentence of less than 50 years will be held lawful.

The majority's rendering of Graham is not only wrong; it is also unnecessary. The majority's analysis assumes that defendants will first become eligible for parole at ages 66 and 74, after serving terms of 50 and 58 years, respectively. That assumption is incorrect. Both defendants will be eligible for parole no later than age 60 under the Elderly Parole Program recently codified by the Legislature. (See Pen. Code, § 3055.) 1 Defendants may be eligible for parole even sooner due to recently expanded programs for earning good conduct and other credits. ( Cal. Code Regs., tit. 15, §§ 3043.2 [Good Conduct Credits], 3043.3 [Milestone Completion Credits], 3043.4 [Rehabilitative Achievement Credits], 3043.5 [Educational Merit Credits].) A sentence that affords a meaningful opportunity for parole at age 60 or earlier cannot properly be characterized as a sentence of life without parole or its functional equivalent, even under the majority's unwarranted extension of Graham .

**466 *385 In sum, the majority opinion gives short shrift to the limited nature of the holding in Graham , to our prior understanding of that decision, and to the steps California has taken toward ensuring that juvenile offenders convicted of nonhomicide offenses receive "some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 .) And in significantly expanding the Graham rule, the majority ultimately condemns as unconstitutional sentences that are materially different from the ones defendants actually will serve. Therefore, I respectfully dissent.

I. FACTUAL AND PROCEDURAL BACKGROUND

The Court of Appeal offered this recitation of the facts presented at trial regarding the brutal series of sexual assaults that led to the sentences before us:

"[Jane] Doe 2, then 15, accompanied [Jane] Doe 1, then 16, and Doe 1's parents to a party for one of Doe 1's relatives. The party was at the relative's house. At dusk, while the party was still going on, the girls went for a walk and sat down by a tree in an open space area. Contreras, then 16, and Rodriguez, then 16, walked past them. Both boys wore dark clothing with hoods covering their heads. Rodriguez wore a red and black cap, a dark-colored Padres T-shirt, and a long-sleeve, plaid or checkered jacket with a gray hood. Contreras *274 wore a long-sleeve, dark-colored, hooded jacket.

"A short time later, Contreras and Rodriguez tackled the girls from behind. Contreras tackled Doe 1 and Rodriguez tackled Doe 2. Both boys wore bandanas covering their noses and mouths. Contreras held a knife to Doe 1's throat. One of boys asked for the girls' cell phones.

"The boys pulled the girls up and started taking them toward a street. Rodriguez covered Doe 2's mouth with his hand as she struggled to get away. Contreras repeatedly told Doe 1 to tell Doe 2 to 'shut the f-k up.' The boys forced the girls to walk across the street, up an embankment, and into a wooded area. As they started going up the embankment, Doe 2 continued to struggle and threw her weight backward, causing both her and Rodriguez to stumble. Doe 2 bit Rodriguez's hand and tried to get away. However, Doe 1, at Contreras's direction, told Doe 2 to be quiet and stop resisting.

"When Doe 2 got up off the ground, Rodriguez tied his bandana around her mouth and told her he would hurt her if she screamed. He took her to a clearing. Contreras took Doe 1 to a different location nearby. The area was not lighted and was not visible from the street.

"Rodriguez took off Doe 2's shorts and underwear. He told her to get down. As she lay on her back, he got on top of her, put his penis in her vagina, and *386 started thrusting in and out. He pulled down the bandana and kissed her, putting his tongue in her mouth. He told her not to scream or he would hurt Doe 1. He asked her if she liked what he was doing. She was wearing a purity ring and had never had sexual intercourse before. His actions were painful and caused her to wince.

"After what seemed like a long time to Doe 2, Rodriguez made her flip over. As she lay on her stomach, he put his penis in her anus and started thrusting in and out.

"As Rodriguez was assaulting Doe 2, Contreras had Doe 1 lay down. He took off her shorts, underwear, and shoes, had her help him take off her dress, and had her take off her bra. He touched her breasts and tried to push his penis into her vagina, but his penis was soft. He asked her whether she was a virgin and she told him she was. He put his fingers in her vagina for a couple of seconds, which was painful for her. He told her to keep her legs open and pushed his now erect penis into her, which was also painful for her. He then started thrusting in and out.

"After awhile, he took his penis out of her vagina, stood up, told her to suck it, and warned her he did not want to feel any teeth. He put his penis in her mouth and pushed her head back and forth. She gagged and threw up. He then pushed his penis back into her vagina. He told her to keep quiet and keep her legs open. She tried to keep quiet, but made some noise because she was uncomfortable. He told her to shut up. He kept the knife in his pocket during the sex acts.

**467 "Around this time, Rodriguez called over to Contreras and the two boys switched places. Rodriguez kissed Doe 1 and bit her cheek and neck. He put his penis in her vagina and thrust in and out. He then put his penis in her mouth and pushed her head back and forth. She gagged and threw up again. He lay down on the ground, had her get on top of him, pushed his penis into her anus, and had her "hump" him by moving up and down. After a couple of minutes, he had her sit back down. He put his penis in her mouth again and pushed her head back and forth. She gagged and threw up again.

"As Rodriguez was engaging in sex acts with Doe 1, Contreras took off Doe 2's dress and had her help him take off her *275 bra. Once all of her clothes were off, he had her lay on her back. While holding the knife to her neck, he told her to open her legs 'really wide.' He then put his penis into her vagina and started thrusting. The action was painful to her. He asked whether she was a virgin and she told him she was. He also asked whether she had a boyfriend and where she went to school. She told him she did not have a boyfriend and what school she attended. *387 "After some period of time, Contreras moved further up on Doe 2. While holding the knife in his hand, he put his penis in her mouth and told her to suck it. She turned her head away and told him she could not breathe. He put his penis back in her mouth and told her to try. She turned her head away again. He changed their positions so he lay on his back and she was on top of him. He told her to put his penis in her vagina. She told him she did not know how, so he put it in himself. He told her to jump up and down, but she did not know what he meant. He thrust up and down while fondling her breasts. His knife was on the ground nearby. When they were in this position, Contreras's bandana slipped and Doe 2 got a good look at his face.

"At some point, Contreras asked Doe 2, 'Did [Rodriguez] f-k your mouth?' She told him no. Rodriguez then brought Doe 1 over to the same place as Doe 2. Once more, Rodriguez put his penis in Doe 1's mouth and pushed her head back and forth. Once more, she threw up. Afterwards, the two boys switched again.

"Rodriguez had Doe 2 get on her back and he put his penis in her mouth. She turned her head away and told him she could not breathe, but he put his penis back in her mouth. While this was occurring, Contreras put his penis in Doe 1's mouth. He moved her head back and forth and warned her he did not want to feel any teeth. She gagged yet again. Neither Contreras nor Rodriguez wore a condom during any of the sex acts.

"When the boys decided to stop, they had the girls put their clothes back on. As Doe 2 was getting dressed, Rodriguez kissed Doe 2, touched her legs, put his finger in her vagina, and told her she was beautiful. Before Doe 1 got dressed, Rodriguez also kissed her and asked her if she liked what had happened. He told her she was beautiful and that, if they had known each other before, she would have been his girlfriend.

"Meanwhile, Contreras pulled a bicycle from the bushes. The boys then directed the girls which way to go and told them not to say anything to anyone. One of the boys said they would follow the girls home and come after the girls if they ever told anyone. Contreras also threatened to find and hurt one of Doe 1's young relatives.

"The girls walked down the slope and across the street, where they met up with Doe 1's parents, who had been looking for them. They got in Doe 1's parents' car and left. Doe 1's mother asked where they had been and what had happened to them.

"At first, the girls did not say anything. Doe 2 did not say anything because she thought the boys were still close by and she just wanted to get away.

*388 However, Doe 1's mother asked them directly if they had been raped and they acknowledged they had been. Doe 1's parents took them back to Doe 1's relative's home, where someone called the police."

The case was tried before two juries. One convicted Rodriguez of two counts of forcible rape (§ 261, subd. (a)(2) ), two counts of kidnapping (§ 207, subd. (a) ), four counts of forcible oral copulation (§ 288a, subd. (c)(2)(A) ), and two counts of sodomy by use **468 of force (§ 286, subd. (c)(2)(A) ). The jury also found true allegations *276 that Rodriguez had committed the sexual assault crimes during a kidnapping and against multiple victims (§ 667.61, subds. (d)(2) & (e)(4) ). The other jury convicted Contreras of seven counts of forcible rape (§ 261, subd. (a)(2) ), conspiracy to commit kidnapping and forcible rape ( § 182, subd. (a)(1) ), rape by foreign object (§ 289, subd. (a)(1)(A) ), two counts of kidnapping (§ 207, subd. (a) ), eight counts of forcible oral copulation (§ 288a, subd. (c)(2)(A) ), and two counts of sodomy by use of force (§ 286, subd. (c)(2)(A) ). This jury returned true findings on allegations that Contreras committed the crimes with use of a knife (§ 12022.3, subd. (a) ), as well as other allegations bringing Contreras's case, like Rodriguez's, within the purview of the "One Strike" law for sentencing purposes (§ 667.61, subds. (d)(2), (e)(1), (3), & (4) ).

These convictions and findings meant that under the One Strike law, defendants faced sentences whereby their first opportunity for parole would not arise until long after their natural lifespans had elapsed. (See §§ 667.6, subd. (d), 667.61, subd. (i).) At the time of sentencing, however, the trial court recognized that in Caballero , supra , 55 Cal.4th 262 , 145 Cal.Rptr.3d 286 , 282 P.3d 291 , this court had construed Graham as directing that a juvenile offender convicted of a nonhomicide crime receive " 'some realistic opportunity to obtain release' from prison during his or her expected lifetime." ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 .) The court advised Rodriguez that had he been an adult, it would have had "no problem" sentencing him to the maximum term of 200 years to life. The court observed, however, that it "couldn't give [Rodriguez] 75 years to life because that would probably take him outside of this life expectancy. ... So probably the most I could give him is 50 to life," which Rodriguez's attorney conceded was a lawful sentence under Caballero . The court imposed this sentence on Rodriguez, sentencing him to two consecutive terms of 25 years to life on the two forcible rape counts, and running the terms on all other counts concurrently. With regard to Contreras, the court acknowledged a prospective statutory sentence of 620 years to life. To comply with Graham , the court imposed a sentence of 58 years to life. This sentence was comprised of two consecutive terms of 25 years to life on two forcible rape counts and an eight-year term on the knife enhancement, with all other terms to run concurrently.

*389 II. DISCUSSION

As explained below, the majority adopts a faulty, overbroad construction of Graham , and extends that decision well beyond the boundaries marked by the high court. And it does so needlessly, because the sentences here are quite different from the ones condemned by the majority. Defendants will become eligible for parole not at ages 66 and 74, as the majority generally assumes, but no later than age 60. These sentences comport with the Eighth Amendment even under the majority's unjustified extrapolation from Graham , making it unnecessary to announce a general standard in today's decision.

A. The Majority Misconstrues Graham

1. Graham is concerned only with sentences of life without parole and functionally equivalent sentences

In Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , the United States Supreme Court considered whether the Eighth Amendment to the United States Constitution absolutely prohibits the imposition of a sentence of life without the possibility of parole on a juvenile offender convicted only of a nonhomicide offense. In resolving *277 this question, the court applied its "categorical" strain of Eighth Amendment jurisprudence. (See id. , at pp. 60-62, 130 S.Ct. 2011 .) This approach evaluates whether a particular type of punishment is "cruel and unusual" ( U.S. Const., 8th Amend.) in all of its applications, or is categorically prohibited with regard to a certain class of offenders. ( Graham , at pp. 60-61.) Prior to Graham , the high court had applied this form of analysis only to sentences of death. ( Id., at 130 S.Ct. 2011 .)

As befits the categorical approach, Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , ultimately invalidated a narrowly defined, specific type of sentence-one that does not **469 afford a juvenile offender convicted of a nonhomicide crime "some realistic opportunity to obtain release." ( Id., at 130 S.Ct. 2011 .) 2 Again and again in its analysis, the Graham court stressed the distinctive characteristics of a sentence of life without parole that made it vulnerable to an Eighth Amendment challenge. *390 The court described life without parole as " 'the second most severe penalty permitted by law,' " and observed that "life without parole sentences share some characteristics with death sentences that are shared by no other sentences." ( Id., at 130 S.Ct. 2011 .) A life without parole sentence, the court emphasized, "alters the offender's life by a forfeiture that is irrevocable. It deprives the convict of the most basic liberties without giving hope of restoration, except perhaps by executive clemency." ( Id. , at pp. 69-70, 130 S.Ct. 2011 .) Such a sentence " 'means that ... [the convict] will remain in prison for the rest of his days.' " ( Id., at 130 S.Ct. 2011 .) "Life in prison without the possibility of parole," the court emphasized, "gives no chance for fulfillment outside prison walls, no chance for reconciliation with society, no hope." ( Id., at 130 S.Ct. 2011 .) Even if the defendant in Graham were to spend "the next half century attempting to atone for his crimes and learn from his mistakes," the court observed, his "sentence guarantees he will die in prison." ( Ibid. )

Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , concluded that for a juvenile offender convicted of a nonhomicide crime, a sentence that guarantees death in prison was unjustified by any prevailing penological rationale, be it retribution, deterrence, incapacitation, or rehabilitation. ( Id. , at pp. 71-74, 130 S.Ct. 2011 .) The court thus believed it necessary to draw a "clear line" that prohibits the imposition of life without parole sentences on juvenile offenders who commit only nonhomicide offenses. ( Id., at 130 S.Ct. 2011 .) It articulated this line as follows: "[a] State is not required to guarantee eventual freedom to a juvenile offender convicted of a nonhomicide crime. What the State must do, however, is give defendants like Graham some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation. It is for the State, in the first instance, to *278 explore the means and mechanisms for compliance." ( Id., at 130 S.Ct. 2011 .) Later, the court reiterated, "A State need not guarantee the offender eventual release, but if it imposes a sentence of life it must provide him or her with some realistic opportunity to obtain release before the end of that term." ( Id., at 130 S.Ct. 2011 .)

Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , was an extension of the Supreme Court's prior Eighth Amendment jurisprudence, but a limited one. The restrained nature of the Graham holding, and the deference it afforded states to "in the first instance ... explore the means and mechanisms for compliance" ( id., at 130 S.Ct. 2011 ), were consistent with the careful, incremental approach the high court has taken when addressing Eighth Amendment questions. The court has been properly mindful that it is the legislature, not the judiciary, that the public anticipates will define the parameters of permissible criminal sentences. (See Rummel v. Estelle (1980) 445 U.S. 263 , 274, 100 S.Ct. 1133 , 63 L.Ed.2d 382 ["one could argue without fear of contradiction by any decision of this Court that for crimes concededly classified and classifiable as felonies, that is, as punishable by significant terms of imprisonment in a state penitentiary, the length of the sentence actually imposed is purely a matter of legislative *391 prerogative"].) We have expressed similar views. ( **470 People v. Wingo (1975) 14 Cal.3d 169 , 174, 121 Cal.Rptr. 97 , 534 P.2d 1001 ["The doctrine of separation of powers is firmly entrenched in the law of California, and a court should not lightly encroach on matters which are uniquely in the domain of the Legislature. Perhaps foremost among these are the definition of crime and the determination of punishment."].)

2. Subsequent judicial application of Graham

Some courts have regarded the Graham holding as very narrowly circumscribed. To these courts, Graham 's reach does not extend to aggregate sentences arising out of convictions for multiple nonhomicide crimes imposed as a specific term of years, or a specific term of years to life, even if the initial opportunity for parole appears outside of the juvenile offender's life expectancy. (E.g., Bunch v. Smith (6th Cir. 2012) 685 F.3d 546 , 552 ; Lucero v. People (Colo. 2017) 394 P.3d 1128 , 1133 ; State v. Brown (La. 2013) 118 So.3d 332 , 342 ; Willbanks v. Dept. of Corrections (Mo. 2017) 522 S.W.3d 238 , 246-247.) 3

Other courts-including our own court-have concluded that a juvenile offender convicted of a nonhomicide crime or crimes does not have the "realistic opportunity to obtain release" ( Graham , supra , 560 U.S. at p. 82, 130 S.Ct. 2011 ) that Graham requires when he or she is sentenced to a term of years in which the initial opportunity for parole plainly arises outside of normal life expectancy, even when multiple convictions are involved. ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ; see also Budder v. Addison (10th Cir. 2017) 851 F.3d 1047 , 1059 ; Moore v. Biter (9th Cir. 2013) 725 F.3d 1184 , 1192 ; Henry v. State (Fla. 2015) 175 So.3d 675 , 679-680, State v. Boston (Nev. 2015) 363 P.3d 453 , 458-459.) Like a sentence explicitly imposed as "life without parole," an aggregate sentence of a term of years in which the initial opportunity for release certainly will come only after the inmate's death-in other words, one that is *279 the functional equivalent of life without parole-" 'means that ... [the convict] will remain in prison for the rest of his days.' " ( Graham , at p. 70, 130 S.Ct. 2011 .)

When this court adopted the latter interpretation of Graham , we related our view of what that decision holds. In Caballero , supra , 55 Cal.4th 262 , 145 Cal.Rptr.3d 286 , 282 P.3d 291 , we concluded that a sentence of 110 years to life fell within Graham 's strictures. We observed that "[d]efendant in the present matter will become parole eligible over 100 years from now. [Citation.] Consequently, he would have no opportunity to 'demonstrate growth and maturity' to try to secure his release, in contravention of Graham 's dictate. [Citations.] Graham 's analysis does not *392 focus on the precise sentence meted out. Instead, as noted above, it holds that a state must provide a juvenile offender 'with some realistic opportunity to obtain release' from prison during his or her expected lifetime ." ( Id. , at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 , italics added.) We later reiterated, "Consistent with the high court's holding in Graham [ ], we conclude that sentencing a juvenile offender for a nonhomicide offense to a term of years with a parole eligibility date that falls outside the juvenile offender's natural life expectancy constitutes cruel and unusual punishment in violation of the Eighth Amendment." ( Ibid. ) Significantly, these descriptions of Graham in Caballero represented a positive articulation of the Supreme Court's holding, not merely an application of Graham to a particular sentence that left our view regarding the scope of that ruling unclear.

Caballero thus interpreted Graham in a manner comporting with the high court's focus and phrasing-unlike the majority here. The language used within Graham itself establishes, and our precedent has recognized, that the Graham court was concerned with prohibiting a relatively discrete class of sentences that do not afford a prisoner " 'some realistic opportunity to obtain release' from prison during his or her expected lifetime" ( **471 Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 , quoting Graham , supra , 560 U.S. at p. 82, 130 S.Ct. 2011 ). These, and only these, sentences involve " 'the second most severe penalty permitted by law.' " ( Graham , at p. 69.) These, and only these, sentences "share some characteristics with death sentences that are shared by no other sentences." ( Ibid. ) And these, and only these sentences mean that a defendant " 'will remain in prison for the rest of his days.' " ( Id., at 130 S.Ct. 2011 .)

3. The majority offers an overbroad construction of Graham

Compare the careful and consistent language used in Graham with the holding today. The majority provides that "[a] lawful sentence must recognize 'a juvenile nonhomicide offender's capacity for change and limited moral culpability.' [Citation.] A lawful sentence must offer 'hope of restoration' [citation], 'a chance to demonstrate maturity and reform' [citation], a 'chance for fulfillment outside prison walls,' and a 'chance for reconciliation with society' [citation]. A lawful sentence must offer 'the opportunity to achieve maturity of judgment and self-recognition of human worth and potential.' [Citation.] A lawful sentence must offer the juvenile offender an 'incentive to become a responsible individual.' " (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 259, 411 P.3d at p. 454, quoting Graham , supra , 560 U.S. at pp. 69-70, 74, 79, 130 S.Ct. 2011 .) 4

*280 *393 The majority thus invalidates sentences in which an initial opportunity for parole (or another possible avenue for release) arises even well within a defendant's life expectancy. What Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , meant to say, the majority professes-notwithstanding the limiting language interwoven throughout that opinion-is that a lawful sentence must provide more than a "meaningful" ( id., at 130 S.Ct. 2011 ) or "realistic opportunity to obtain release" ( id., at 130 S.Ct. 2011 ). According to the majority, the state also must structure prison sentences to offer an initial opportunity for release at a juncture that affords sufficient time for the inmate to fully reintegrate into society. Although the majority declines to explain what constitutes an adequate postcustodial buffer, today's ruling makes clear that in the majority's view, an initial opportunity for release at age 66 or 74 does not provide enough time.

Today's ruling thus declares unconstitutional a range of sentences that are qualitatively **472 different from the sentences of life without parole that Graham addressed. Neither Rodriguez's sentence of 50 years to life nor Contreras's sentence of 58 years to life represents " 'the second most severe penalty permitted by law.' " ( Graham , supra , 560 U.S. at p. 69, 130 S.Ct. 2011 .) Neither sentence ensures an "irrevocable" forfeiture of the inmate's liberties "without giving hope of restoration." ( Id. , at pp. 69-70, 130 S.Ct. 2011 .) Neither sentence means that the defendant " 'will remain in prison for the rest of his days.' " ( Id., at 130 S.Ct. 2011 .) *394 Neither sentence "gives no chance for fulfillment outside prison walls, no chance for reconciliation with society, no hope." ( Id., at 130 S.Ct. 2011 , italics added.) Neither sentence "guarantees" the defendant "will die in prison." ( Ibid. ) *281 On the contrary, the sentences here afford defendants a "meaningful" ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 ) and "realistic" ( id., at 130 S.Ct. 2011 ) opportunity for parole within their lifetimes. Both defendants will be eligible for parole well within prevailing life expectancies for people their age. (Nat. Vital Statistics System, U.S. Dept. of Health & Human Services, United States Life Tables, 2010 (Nov. 6, 2014) p. 9 (National Vital Statistics System Study) [projecting an average life expectancy of approximately 79 years for persons aged 15-16 in the United States as of 2010], < https://www.cdc.gov/nchs/data/nvsr/nvsr63/nvsr63_07.pdf> [as of Feb. 26, 2018].) Furthermore, extrapolating a median age at death from average life-expectancy figures, as the majority does (maj. opn., ante , at p. ----), in fact significantly underestimates the likelihood that a person will live to a certain age. (See Nat. Vital Statistics System Study, at pp. 2, 9-10 [providing data and associated interpretive guidance forecasting that as of 2010, a 15-year-old member of the general public has a greater than 57 percent chance of surviving to age 80, and a greater than 50 percent chance of surviving to age 82].)

The majority refuses to consider these or any other empirical data for purposes of determining when a sentence affords a "meaningful" ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 ) or "realistic" ( id., at 130 S.Ct. 2011 ) opportunity for release. The majority expresses concern that use of such data would entail a choice between, on the one hand, disadvantaging members of a cohort that may in the aggregate have a lower life expectancy than that of the general public; or on the other, improperly relying on race, gender, or other characteristics in assessing whether a sentence falls sufficiently within a defendant's life expectancy. (Maj. opn., ante , at pp. ---- - ----.)

There are three responses. First, some reliance on lifespan data is not merely recognized by our precedent ( Caballero , supra , 55 Cal.4th at p. 268, 145 Cal.Rptr.3d 286 , 282 P.3d 291 ), but is unavoidable when determining whether a sentence affords a " realistic opportunity to obtain release." ( Graham , supra , 560 U.S. at p. 82, 130 S.Ct. 2011 , italics added.) In People v. Franklin (2016) 63 Cal.4th 261 , 202 Cal.Rptr.3d 496 , 370 P.3d 1053 ( Franklin ), for example, this court also considered an Eighth Amendment challenge to a sentence of 50 years to life imposed on a juvenile, who characterized the sentence as the functional equivalent of life without parole. We found this challenge mooted by the Legislature's then-recent enactment of a system of youth offender parole hearings (see § 3051) that provides for a parole hearing no later than an eligible offender's twenty-fifth year of *395 incarceration. ( Franklin , at pp. 279-280, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) 5 Under this program, the defendant in Franklin would be eligible for parole at the age of 41 years. ( Id. , at p. 279, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) A sentence affording a meaningful opportunity for parole at such a juncture, we concluded, was not the functional equivalent of a sentence of life without parole. ( Ibid. ) To have drawn this conclusion, we must have mapped the defendant's sentence against some conception of his life expectancy. And, truth be told, the majority here must have engaged in comparable benchmarking. In invalidating defendants' sentences on the ground that they provide insufficient time for reintegration into society **473 *282 upon early parole, the majority must have some notion of defendants' life expectancy in mind. The majority, however, does not disclose this figure.

Second, although the majority emphasizes its concerns with life expectancies based on race, sex, and custodial status, juvenile defendants belong to a nearly infinite number of cohorts. Some of these groups may have longer life expectancies than the general population, others shorter. To assign more importance to a defendant's membership in one cohort than to his or her presence in another would be speculative. Given that a defendant could be placed within any of many peer groups for purposes of assessing his or her life expectancy, and given as well the need to use some conception of life expectancy as a benchmark, reliance on general population life expectancies makes good sense as providing an administrable rule of decision that is consistent with Graham .

Third, and most fundamentally, the majority's concerns derive from its fundamental mischaracterization of what Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , requires. The majority appears to impose upon the People the burden of showing that defendants do not belong to any cohort in which the average member lacks a high probability of surviving until well past the ages of 66 or 74. That is not what Graham holds, and is also inconsistent with the general principle that the defendant bears a " 'considerable burden' to show a punishment is cruel and unusual." ( People v. Meneses (2011) 193 Cal.App.4th 1087 , 1092, 123 Cal.Rptr.3d 387 , quoting People v. Wingo , supra , 14 Cal.3d at p. 174, 121 Cal.Rptr. 97 , 534 P.2d 1001 .) As discussed ante , Graham requires only a "meaningful" ( Graham , at p. 75, 130 S.Ct. 2011 ) or "realistic" ( id., at 130 S.Ct. 2011 ) opportunity for parole, not a certain one (which would be impossible to guarantee); and it does not require the very substantial postcustodial period that the majority demands. A sentence that *396 offers an initial parole hearing at age 66 or 74, well within prevailing public life expectancies, offers the sort of opportunity that Graham contemplates. 6

B. The Majority Provides No Persuasive Rationale for Extending Graham

The preceding discussion establishes that there is a basic disconnect between Graham itself, and the majority's interpretation of that decision. Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , condemned one type of sentence; the majority, another altogether. To bridge this gap, the majority justifies its holding as a logical extension of aspects of Graham 's reasoning. But the majority's analysis on these points is unpersuasive.

1. The majority's discussion of penological objectives does not support its expansion of Graham

The majority's principal justification for extending Graham to the sentences here is the cursory survey it conducts of the four penological rationales for sentencing practices that Graham considered. (See Graham , supra , 560 U.S. at pp. 71-74, 130 S.Ct. 2011 .) The majority perceives from this review inadequate justification for the sentences here. (Maj. opn., ante , at p. ----.) But the majority's discussion of these penological *283 objectives proves both too much and too little.

The discussion proves too much, in that majority's vague critiques of the prison terms imposed on defendants as insufficiently justified by reference to these penological objectives could be read to forbid any lengthy sentence imposed upon a juvenile offender. We are told that "[t]he retributive case for a 50-years-to-life sentence, as for [life without parole], is weakened by the juvenile nonhomicide offender's 'age ... and the nature of the crime.' [Citation.] As for deterrence, Graham 's observation that juveniles have limited ability to consider consequences when making decisions [citation] applies to a sentence of 50 years to life just as it does to a sentence of [life without parole]. And as for incapacitation, a judgment that a juvenile offender will be incorrigible for the next 50 **474 years is no less 'questionable' than a judgment that the juvenile offender will be incorrigible 'forever.' [Citations.] Finally, as noted, a sentence of 50 years to life 'cannot be justified by the goal of rehabilitation' because it offers a juvenile offender little 'incentive to become a responsible individual.' [Citation.]" (Maj. opn., ante , at p. ----.) On each of these points, the majority offers no limiting principle that would *397 establish why similarly broad criticisms could not be lodged against the sentence we upheld as lawful in Franklin , supra , 63 Cal.4th at pages 279-280, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 , which afforded an initial opportunity for parole only after 25 years of incarceration.

Meanwhile, a more careful analysis establishes that the majority's survey of penological objectives proves too little, because the sentences here are better justified by reference to penological aims than the life without parole sentences addressed in Graham were. With regard to retribution, the Graham court was concerned with a perceived lack of proportionality between a nonhomicide crime and imposition of "the second most severe penalty" on a juvenile. ( Graham , supra , 560 U.S. at p. 72, 130 S.Ct. 2011 ; see also id., at 130 S.Ct. 2011 .) But the proportionality analysis is different here. The sentences here are not as severe as one that "guarantees" the defendant "will die in prison." ( Id., at 130 S.Ct. 2011 .) A sentence that withholds any hope of release signifies a final determination that the juvenile will never again be fit to reenter society. A sentence that affords some hope of parole within prevailing life expectancies does not send a similar message. Such a sentence manifests a belief that the offender can change. Consistent with this belief, it offers the prospect of release. Likewise, a sentence that offers a "meaningful" ( id., at 130 S.Ct. 2011 ) and "realistic" ( id., at 130 S.Ct. 2011 ) chance of parole within the offender's lifespan, as the sentences here do, does not utterly foreswear the rehabilitative ideal, or demand incapacitation forever, regardless of whether the inmate remains a threat to public safety. (See § 3041, subd. (b)(1) [describing the standard for a grant of parole].) Instead, such a sentence recognizes that the offender may become an improved person while in prison, which may give him or her the possibility of release.

Finally, Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , perceived the fourth penological objective it discussed, deterrence, as an insufficient justification for a sentence of life without parole for a juvenile offender convicted only of a nonhomicide crime. ( Id., at 130 S.Ct. 2011 .) The Graham court believed that juveniles may not be deterred by the prospect of a lifelong prison term, particularly given how rarely such a term had been imposed for a *284 nonhomicide crime. ( Ibid. ) But Graham did not categorically cast lengthier terms of incarceration as having no marginal deterrence value for juveniles, relative to shorter terms. Nor did the court suggest that deterrence, together with other penological rationales, would not provide an adequate justification for a sentence that does offer an opportunity for parole within prevailing lifespans. (See ibid. [noting that "any limited deterrent effect provided by life without parole is not enough to justify the sentence"].) *398 In short, a proper review of the penological objectives of sentencing further establishes that the majority has improperly extended Graham to an array of sentences that are materially different from the type of sentence condemned by the Supreme Court.

2. Graham did not endorse an approach as vague as the majority's

Lastly, regardless of whether the majority is better described as adopting an erroneous interpretation of Graham , or as an improper extension of that decision, its holding fails to heed the Supreme Court's guidance regarding the need for workable, objective rules in the Eighth Amendment sphere.

In appropriate instances, the Supreme Court has drawn clear lines for the administration of a constitutional rule. (See, e.g., County of Riverside v. McLaughlin (1991) 500 U.S. 44 , 56, 111 S.Ct. 1661 , 114 L.Ed.2d 49 [specifying 48 hours as the maximum period to fulfill the judicial presentment and probable-cause determination requirement of Gerstein v. Pugh (1975) 420 U.S. 103 , 95 S.Ct. 854 , 43 L.Ed.2d 54 ]; cf. Maryland v. Shatzer (2010) 559 U.S. 98 , 110, 130 S.Ct. 1213 , 175 L.Ed.2d 1045 .) The court has regarded **475 such an approach as preferable to a "vague standard" that fails to provide "sufficient guidance," particularly when adoption of a rule would avoid having "judges in the role of making legislative judgments." ( County of Riverside v. McLaughlin , at p. 56, 111 S.Ct. 1661 .) It would represent a logical application of this general principle to rely on life expectancies in ascertaining whether a sentence comports with Graham , particularly given the Graham court's express avowal that it was drawing a "clear line" with its decision. ( Graham , supra , 560 U.S. at p. 74, 130 S.Ct. 2011 .) 7

The majority's approach, in contrast, turns on highly subjective impressions regarding matters such what adequate postcustodial reintegration into society entails, and the time necessary to accomplish this assimilation. It thus runs counter to the high court's stated view that " 'Eighth Amendment judgments should not be, or appear to be, merely the subjective views of individual Justices; judgment should be informed by objective factors to the maximum possible extent.' " ( Rummel v. Estelle , supra , 445 U.S. at pp. 274-275, 100 S.Ct. 1133 , quoting *399 Coker v. Georgia (1977) 433 U.S. 584 , 592, 97 S.Ct. 2861 , 53 L.Ed.2d 982 (plur. opn. of White, J.).) Today's decision instead announces precisely the sort of "vague standard" involving "legislative judgments" ( *285 County of Riverside v. McLaughlin , supra , 500 U.S. at p. 56, 111 S.Ct. 1661 ) that the Supreme Court has told us to avoid.

To repeat, the majority holds that under the Eighth Amendment, "[a] lawful sentence must recognize 'a juvenile nonhomicide offender's capacity for change and limited moral culpability.' [Citation.] A lawful sentence must offer 'hope of restoration' [citation], 'a chance to demonstrate maturity and reform' [citation], a 'chance for fulfillment outside prison walls,' and a 'chance for reconciliation with society' [citation]. A lawful sentence must offer 'the opportunity to achieve maturity of judgment and self-recognition of human worth and potential.' [Citation.] A lawful sentence must offer the juvenile offender an 'incentive to become a responsible individual.' " (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 259, 411 P.3d at p. 454.) One could regard all of these as worthwhile objectives, and certainly Graham condemned sentences of life without parole , as imposed on juvenile offenders who committed only nonhomicide crimes, on grounds that included the perception that they offered no hope of freedom, no chance to demonstrate that they had matured, and no opportunity for fulfillment outside prison. But this aspect of Graham simply makes the Supreme Court's limiting language, which the majority omits in relating its holding, all the more important. What the Supreme Court in Graham appreciated-but today's decision does not-is the need for coherent rules for application in specific cases.

The courts of this state, capable though they are, undoubtedly will struggle to apply standards presented at the majority holding's high level of abstraction. The inevitable disagreements will be resolved only by another set of highly subjective judgments on appeal, and so forth. Even as applied here, the vagueness inherent in the majority's approach makes it unclear that defendants' sentences are unlawful. We know that the sentences are unconstitutional only because the majority tells us as much. Yet I anticipate that even the majority would concede that profound life experiences still may lie ahead of someone released from prison at age 66 or 74. The majority describes these ages as falling "near the end" of a person's life, language that suggests that fulfillment at such a juncture is well-nigh impossible. (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 259, 411 P.3d at p. 454.) The millions of productively employed senior citizens would beg to differ (see **476 State v. Smith (Neb. 2017) 295 Neb. 957 , 892 N.W.2d 52 , 66 ["in today's society, it is not unusual for people to work well into their seventies"] ), as would the millions more who have retired from the workforce, or perhaps never entered it, but represent valued contributors to their families and communities. And, I anticipate, many inmates who are freed from custody at these ages also would disagree with the assessment that they are "near the end" of their lives. True, prisoners who are released from prison after serving lengthy terms will need to adjust to their changed circumstances. *400 But substantial fulfillment-whether in the form of rapprochement or reunions with friends and family, community service, continuing education, employment, or otherwise-does not necessarily arrive only after many years outside of custody, particularly for those who already have demonstrated maturity and the capacity to reform.

Given the degree of subjectivity entailed in applying the majority's approach to sentences of 50 years to life and 58 years to life, how these standards apply to sentences of less than 50 years to life presents even more difficult questions. (See, e.g., People v. Bell (2016) 3 Cal.App.5th 865 , 208 Cal.Rptr.3d 102 , review granted Jan. 11, 2017, S238339.) Here again, I doubt this is what Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , intended: a series of judicial decisions upholding or invalidating sentences affording an opportunity for parole *286 at age 65, 64, 63, 62, or younger, based on judges' subjective and quite likely divergent assessments of what constitutes adequate reintegration into society, and the time necessary to accomplish this reentry. The Graham court said it was drawing a "clear line." ( Id., at 130 S.Ct. 2011 .) I would not obfuscate what the high court sought to clarify.

The majority opinion asserts that using life expectancy as a measure for the constitutionality of a sentence under Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , implicates as much vagueness and subjectivity as its own approach does. (Maj. opn., ante , at 229 Cal.Rptr.3d at pp. 263-264, 411 P.3d at pp. 457-477.) This false equivalence once again mischaracterizes Graham . The majority asserts that both approaches "depend on a considered judgment as to whether the parole eligibility date of a lengthy sentence offers a juvenile offender a realistic hope of release and a genuine opportunity to reintegrate into society." ( Id. , at 229 Cal.Rptr.3d at p. 286, 411 P.3d at p. 457.) But, as the foregoing text makes clear, only the first half of this rule comes from Graham . The second half ("and a genuine opportunity to reintegrate into society") is the majority's own creation. ( Ibid. ) This modification effectively displaces the relatively straightforward and objective Graham inquiry into whether sentence affords a "meaningful opportunity to obtain release" ( Graham , at p. 75, 130 S.Ct. 2011 ), with a far more idiosyncratic inquiry into whether a sentence offers what the majority considers a sufficiently meaningful period of release. 8

The majority's revision of the Graham rule also infiltrates its errant assessment that the "crucial question" in this case is how long a defendant can expect to live after his or her first opportunity for parole arrives (maj.

*401 opn., ante , at 229 Cal.Rptr.3d at p. 263, 411 P.3d at p. 457), and its attempt to characterize the disagreement here as concerned only with the length of this period ( ibid. ). The truly crucial question, of course, is what Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 , holds. As discussed above, and as recognized in Caballero , the core of the Graham holding is that a defendant must receive a "meaningful" ( id., at 130 S.Ct. 2011 ) and "realistic" ( id., at 130 S.Ct. 2011 ) opportunity to obtain release. A defendant made eligible for parole at an age within general population life expectancies receives such an opportunity. Many defendants who earn parole at such a juncture will have a robust postcustodial period of freedom. Some will not, as would be true of any sentence. But it is the opportunity for release, not the precise length of postcustodial **477 period, that lies at the heart of the Graham ruling. The majority errs in shifting the law toward a different position.

C. Even Under the Majority's Approach, the Sentences Here Satisfy Graham

The majority's holding is doubly misguided because it presumes that defendants will not have a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation" ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 ) until they reach the ages of 66 and 74. But this too is wrong. Under the state's Elderly Parole Program for prison inmates ( § 3055 ), both defendants will have an opportunity for parole at age 60. Furthermore, even *287 without the Elderly Parole Program, Rodriguez may be eligible for parole when he is 57 years old, simply by earning good-conduct credits. ( Cal. Code Regs., tit. 15, § 3043.2.) A sentence offering an opportunity for parole no later than age 60 is not invalid under Graham , even under the majority's flawed construction of that decision.

1. The Elderly Parole Program offers defendants a meaningful opportunity for parole at age 60

In 2014, the State of California instituted the Elderly Parole Program in response to a long-running prison-population lawsuit in federal court (Case No. 3:01-cv-01351-JST, N.D. Cal.), which now bears the title Brown v. Plata . The program was codified by the Legislature last year. (See Assem. Bill No. 1448 (2017-2018 Reg. Sess.).) Aside from certain exceptions not pertinent here, the program is available to any state inmate who is "60 years of age or older and has served a minimum of 25 years of continuous incarceration on his or her current sentence." ( § 3055, subd. (a).)

Under the Elderly Parole Program, an eligible inmate "shall meet with the [Board of Parole Hearings] pursuant to subdivision (a) of Section 3041. If [the] inmate is found suitable for parole under the Elderly Parole Program, the [Board of Parole Hearings] shall release the individual on parole as *402 provided in Section 3041." ( § 3055, subd. (e).) The elderly parole statute also directs that "[w]hen considering the release of an inmate specified by subdivision (a) pursuant to Section 3041, the [Board of Parole Hearings] shall give special consideration to whether age, time served, and diminished physical condition, if any, have reduced the elderly inmate's risk for future violence." ( § 3055, subd. (c).) 9

As reflected in the statutory reference to an inmate's "risk for future violence" ( § 3055, subd. (c) ), the decision whether to grant elderly parole is concerned with the same question of public safety that governs conventional parole hearings. (See § 3041, subd. (b)(1) ["The panel or the board, sitting en banc, shall grant parole to an inmate unless it determines that the gravity of the current convicted offense or offenses, or the timing and gravity of current or past convicted offense or offenses, is such that consideration of the public safety requires a more lengthy period of incarceration for this individual."]; Cal. Code Regs., tit. 15, § 2281, subd. (a) ["[r]egardless of the length of time served, a life prisoner shall be found unsuitable for and denied parole if in the judgment of the panel the prisoner will pose an unreasonable risk of danger to society if released from prison"].) 10 In making this determination, "[a]ll relevant, reliable information available to the panel shall be considered in determining suitability for parole. Such information **478 shall include the circumstances *288 of the prisoner's: social history; past and present mental state; ... past and present attitude toward the crime; ... and any other information which bears on the prisoner's suitability for release." ( Cal. Code Regs., tit. 15, § 2281, subd. (b).) 11 *403 Although in an elderly parole hearing "special consideration" is given to the three factors specified in section 3055, subdivision (c), there is no suggestion that these "special" considerations somehow skew the basic question before the panel. In other words, there is no indication that within the elderly parole process, an inmate for whom "consideration of the public safety" does not require "a more lengthy period of incarceration" (§ 3041, subd. (b)(1) ) would nevertheless be denied parole because he or she is too healthy or robust. On the contrary, the statutory reference to "special consideration" being given to "time served" in Elderly Parole Program proceedings corroborates that these hearings are to take into account the enhanced maturity that may come from time in custody, along with all other relevant facts. ( § 3055, subd. (c).) 12

The Elderly Parole Program thus offers a meaningful vehicle for juvenile offenders who have been sentenced to lengthy terms to secure their release at age 60. Inexplicably, even though we requested and received supplemental briefing on this program, the majority declines to address its impact on defendants' Eighth Amendment claims. The majority instead remands the matter for the sentencing court and the parties to develop a record "on how the Elderly Parole Program actually operates," along with other matters. (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 266, 411 P.3d at p. 459.) This remand is both regrettable and wholly unnecessary.

The majority's rationale for remanding the matter is not entirely clear. Defendants express concerns that in practice, the Elderly Parole Program may not give "great weight to the diminished culpability of youth as compared to adults, the hallmark features of youth, and any subsequent growth and increased maturity of the prisoner in accordance with relevant case law." (§ 4801, subd. (c).) But as noted, "[a]ll relevant, reliable information" ( Cal. Code Regs., tit. 15, § 2281, subd. (b) ) is to be considered in a parole hearing, including an elderly parole hearing. There is no reason to believe that salient facts regarding the diminished culpability of juveniles, hallmark features of youth, and an inmate's subsequent growth and increased maturity, where pertinent, are somehow excluded from consideration in an elderly parole hearing, or given short shrift. Unless the prospect of parole at age 60 comes too late to satisfy the Eighth Amendment-a point discussed below-the Constitution *289 requires no more. 13 **479 *404 Nor is a remand necessary for any other reason. Again, the majority seeks to develop a record concerning "how the Elderly Parole Program actually operates." (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 266, 411 P.3d at p. 459.) Yet there is nothing in the record to suggest that elderly parole hearings function differently from how they have been described above. 14 Even Contreras, in his supplemental brief, acknowledges that parole decisions under the Elderly Parole Program are based on an assessment of whether the inmate's release would threaten public safety. Likewise, in a filing with the federal court overseeing the Brown v. Plata litigation, the State has explained that in an elderly parole hearing, the Board of Parole Hearings "will give special consideration to eligible inmates' advanced age, long-term confinement, and diminished physical condition, if any. The board will also consider all other relevant information when determining whether or not there is a reasonable likelihood that consideration of the public and victim's safety does not require the additional period of incarceration of the inmate ." (Board of Parole Hearings, Elderly Parole Program (June 16, 2014) p. 1, < https://www.cdcr.ca.gov/BOPH/docs/Policy/Elderly_Parole_Program_Overview.pdf> [as of Feb. 26, 2018], italics added.) A remand order should be based on something more substantive than an inchoate concern that a duly enacted government program isn't what the relevant statutes and regulations say it is, and what the parties tell us it is. *405 Similarly, the majority speculates that the California Department of Corrections and Rehabilitation someday might adopt "regulations that focus the Elderly Parole Program on identifying those inmates who *290 no longer pose a risk of future violence primarily because of their age, illness, or other physical incapacitation, while leaving all other inmates age 60 or older who may be suitable for parole to the ordinary parole process." (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 266, 411 P.3d at p. 459.) But this a strawperson argument, for no such regulations exist, or are on the horizon. Although one can always conjure up what-if scenarios about future changes in the law, such conjecture does not provide a basis for ignoring our responsibility to interpret the law as it presently stands.

In fact, we have declined to indulge this sort of speculation under similar circumstances. In Franklin , supra , 63 Cal.4th 261 , 202 Cal.Rptr.3d 496 , 370 P.3d 1053 , an amicus curiae asserted that the youth offender parole hearing program (§ 3051) would not operate in practice as the governing statutes said it would, and therefore would not provide the defendant and those similarly situated with a meaningful opportunity for release. ( Franklin , at pp. 284-285, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .) Unlike here, however, we did not treat such a possibility as providing a basis to decline to apply a statute as written. Instead, in concluding that the youth offender parole hearing program mooted the defendant's Eighth Amendment challenge, we noted the "absence of any concrete controversy in this case concerning" the actual functioning of the program. ( Id. , at p. 286, 202 Cal.Rptr.3d 496 , 370 P.3d 1053 .)

**480 The majority claims that the situation in Franklin differed from the one here in that the "explicit and specific purpose" of the statute that created the youth offender parole hearing program at issue in Franklin was to provide an early opportunity for juvenile offenders to seek parole. (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 266, 411 P.3d at p. 460.) Here, by comparison, "[n]either the text nor history of the elderly parole statute contains any indication that the Legislature intended elderly parole hearings to be responsive to the Eighth Amendment concerns raised by lengthy juvenile sentences." ( Ibid .) But this purported distinction, which says nothing about how the Elderly Parole Program actually functions, does not provide a basis to avoid our duty to construe the law. 15 If the majority takes the view that the Elderly Parole Program does not provide *406 juvenile offenders with a "meaningful opportunity to obtain release" under Graham , supra , 560 U.S. at page 75, 130 S.Ct. 2011 , it should simply say so, and explain why, rather than engage in statutory interpretation in order to avoid statutory interpretation. 16 *291 2. A sentence that provides a juvenile offender convicted of a nonhomicide crime a meaningful opportunity for release at age 60 is constitutional under Graham

The majority's decision to remand this matter means that it does not consider whether Graham prohibits a sentence that offers an opportunity for parole no later than age 60. I would address this question, and conclude that it does not. As explained below, even under the majority's view that a sentence that affords an initial opportunity for parole at the age of 66 or 74 is unlikely to provide a juvenile offender with a sufficient period to adequately reintegrate into society, and is therefore unconstitutional (maj. opn., ante , at pp. ---- - ----), the same cannot be said of a sentence that affords an opportunity for parole at age 60. 17

**481 *407 A sentence affording an opportunity for parole at age 60 offers a juvenile offender a substantial likelihood of spending not just a few, but many productive years outside of custody, if he or she demonstrates sufficient maturity to secure parole. During this time, a juvenile offender who has been released on parole because his or her personal development confirmed Graham 's intuitions can participate in the workforce, 18 develop interpersonal relationships, and otherwise seek and obtain the degree of personal fulfillment contemplated by the majority.

Indeed, many of the majority opinion's arguments for invalidating sentences that afford an initial opportunity for parole at *292 ages 66 and 74 lose their force, or cut in the opposite direction, when applied to sentences that afford an initial opportunity for parole at age 60. For example, the majority opinion relies on the fact that all state high courts to have considered sentences of 50 years to life or longer, when imposed on juvenile offenders convicted of nonhomicide crimes, have struck those sentences as unconstitutional. (Maj. opn., ante , at p. ----.) 19 But the balance of the case law from even this highly refined subset of courts shifts when what is being considered is a sentence that affords an opportunity for parole at age 60. The weight of authority regards such a sentence as passing muster under Graham . (See State v. Smith , supra , 892 N.W.2d at pp. 64-66 [holding that a nonhomicide sentence *408 affording an opportunity for release at 62 comports with Graham ]; Angel v. Commonwealth (2011) 281 Va. 248 , 704 S.E.2d 386 , 401-402 [rejecting an Eighth Amendment claim in light of a state geriatric release program affording an opportunity for release at 60, where "the factors used in the normal parole consideration process apply to conditional release decisions under [the] statute"]; cf. State v. Charles , supra , 892 N.W.2d at p. 921 [finding a sentence lawful under Miller , noting that "[b]ecause [defendant] has the opportunity for release at age 60, his sentence does not 'guarantee[ ] he will die in prison without any meaningful opportunity to obtain release' "]; but see Bear Cloud v. State (Wyo. 2014) 334 P.3d 132 , 147 [regarding a 45-year sentence with parole eligibility at age 61 as subject to Miller ].)

Recognizing the lack of authority for its position, the majority searches for support **482 from an unlikely source: LeBlanc , supra , 582 U.S. ----, [ 137 S.Ct. 1726 ], a recent high court decision that denied habeas corpus relief under circumstances similar to those present here. The court in LeBlanc determined that lower federal courts had overstepped their authority under the federal Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA) ( 28 U.S.C. § 2254 (d)(1) ) in granting habeas corpus relief to a petitioner who claimed that his sentence, which offered an opportunity for geriatric parole at age 60, violated Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 . *293 The Supreme Court stated that it was expressing " 'no view on the merits of the underlying' Eighth Amendment claim," but concluded that a Virginia state court's determination that the inmate's sentence comported with Graham was not "objectively unreasonable in light of this Court's current case law." ( LeBlanc , at p. ---- [137 S.Ct. at p. 1729 ].)

The court in LeBlanc , supra , 582 U.S. ----, [ 137 S.Ct. 1726 ] also noted that " '[p]erhaps the logical next step from' Graham would be to hold that a geriatric release program does not satisfy the Eighth Amendment, but 'perhaps not.' " [Citation.] ( Id. , at p. ---- [ 137 S.Ct. at p. 1729 ].) The court observed that " '[T]here are reasonable arguments on both sides.' [Citation.]" ( Ibid. ) With respect to the state, these arguments included the fact that "the geriatric release program employed normal parole factors," consideration of which "could allow the Parole Board to order a former juvenile offender's conditional release in light of his or her 'demonstrated maturity and rehabilitation.' " ( Ibid. ) With respect to the habeas corpus petitioner, the arguments to the contrary included "the contentions that the Parole Board's substantial discretion to deny geriatric release deprives juvenile nonhomicide offenders a meaningful opportunity to seek parole and that juveniles cannot seek geriatric release until they have spent at least four decades in prison." ( Ibid. )

Properly understood, LeBlanc , supra , 582 U.S. ----, [ 137 S.Ct. 1726 ] undermines the majority's position. First, the high court's analysis further *409 confirms that the employment of " normal parole factors " in the parole process ( id. , at p. ---- [ 137 S.Ct. at p. 1729 ] ), as the Elderly Parole Program does, affords a juvenile offender a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." ( Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 .) Second, LeBlanc perceived any invalidation of a sentence such as the one imposed upon the habeas corpus petitioner before it as a potential " ' next step ' " from Graham -but not compelled by Graham itself. ( LeBlanc , at p. ---- [ 137 S.Ct. at p. 1729 ], italics added.) Our job is not to anticipate the infinite array of possible next steps that the Supreme Court may take that would break new ground in the law, but to apply the law as it stands. Third, to the extent that the court identified " 'reasonable arguments' " suggesting that the sentence before it might be subject to close review as a " 'next step' " from Graham , this discussion was dicta, 20 as underscored by the LeBlanc court's reminder that it was expressing no view on the merits of the issue. ( Ibid. ) Fourth, and finally, the LeBlanc court's determination that the procedural posture of that case meant that there was no need for it to resolve the substantive merits of the habeas corpus petitioner's Eighth Amendment claim provides no support for this court avoiding its own responsibility to decide the Eighth Amendment issue before it.

3. Defendants' eligibility for conduct credits further establishes that their sentences are lawful

The majority also refuses to discuss the impact that conduct credits will have on defendants' sentences. (Maj. opn., ante , at 229 Cal.Rptr.3d at pp. 267-268, 411 P.3d at pp. 462-461.) In fact, 60 represents the *294 latest age at which defendants will become eligible for parole. Rodriguez has it wholly within his power to advance his parole hearing to age 57 simply by maximizing the good-conduct **483 credits that are available to him under state law. Contreras could advance his initial parole date to age 64 through good conduct. (See Cal. Code Regs., tit. 15, § 3043.2.) Both defendants could receive even earlier parole hearings by earning other types of conduct credits. 21 Although the majority declines to acknowledge the impact *410 of any of these programs, the availability of these credits provides an additional, independent basis for concluding that defendants are not serving unlawful sentences.

In considering whether a juvenile offender is serving a life sentence under Graham , it is appropriate to assume that the juvenile will maximize available good-conduct credits. After all, good conduct in prison merely substantiates Graham 's intuitions regarding the possibility of maturation and redemption. In Mathurin , supra , 868 F.3d 921 , for example, the court described as an "important additional factor that is absolutely pivotal to [the] inquiry" into the lawfulness of a sentence under Graham the fact that the defendant could "shorten his sentence by earning good-time credit." ( Id. , at p. 934.) The court in Mathurin acknowledged that "[i]t is true that [d]efendant may not receive all of the [available] good-time credit if he misbehaves and thereby forfeits some of that credit." ( Id. , at p. 935.) That fact notwithstanding, the court determined that it was proper to take the credits into account because "it is totally within [d]efendant's own power to shorten the sentence imposed." ( Ibid. ) Furthermore, the court stressed that "good-time credits provide a potent rehabilitative incentive for juvenile offenders subject to lengthy sentences, which, according to the Supreme Court's rationale in Graham is an important objective. ... Similar to parole, the ability to earn good-time credits ... [gives] the juvenile offender a reason to pursue and exhibit 'maturity and rehabilitation.' " ( Id. , at p. 935, quoting Graham , supra , 560 U.S. at p. 75, 130 S.Ct. 2011 .) Consistent with Mathurin , other courts have similarly factored an assumption of maximized good-conduct credits into an assessment of a sentence's length, for purposes of determining whether the sentence comports with Graham . (E.g., State v. Smith , supra , 892 N.W.2d at p. 974 ; People v. Evans (Ill.Ct.App. 2017) 416 Ill.Dec. 769 , 86 N.E.3d 1054 , 1057 ; see also Steilman v. Michael (Mont. 2017) 389 Mont. 512 , 407 P.3d 313 [taking good-time credits into account in concluding that a sentence imposed upon a juvenile for a homicide crime did not violate Miller ]; cf. U.S. v. Tocco (2d Cir. 1998) 135 F.3d 116 , 132 [taking maximized good-conduct credits into account in determining whether a sentence represented *295 a life term under the 18 U.S.C. § 34 ]; but see Johnson v. State (Fla. 2017) 215 So.3d 1237 , 1242 [declining to consider the sentence-reducing effect of "gain time" in connection with a Graham claim].)

I too would take the availability of good-conduct credits into account in determining whether defendants' sentences violate the Eighth Amendment. Maximizing these credits, by itself, would not advance Contreras's initial *411 parole hearing before age 60, but it would make Rodriguez eligible for parole at age 57. No plausible argument exists that such a sentence is tantamount to a life term, or would offer an inadequate time for reconciliation with society under the majority's reasoning. And although good-conduct credits, on their own, would not advance Contreras's first opportunity for parole before the time of his initial elderly parole hearing at age 60, the majority has not adequately justified **484 its failure to give effect to that recent legislation, nor has it explained how its reasoning would apply to a sentence that affords a juvenile offender the possibility of parole at age 64.

III. CONCLUSION

Today's decision opens the door to ill-advised and ill-informed incursions into sentencing questions that have, to this point, properly been understood as the Legislature's domain. Had the Supreme Court in Graham directed this type of judicial intervention, that would be one thing. But it did not, and the majority errs in expanding Graham well beyond the more limited and more reasonable boundaries marked by the high court. Moreover, the decision today does not even resolve the lawfulness of the sentences that defendants actually will serve. I would not remand this matter for the resolution of phantom issues of fact, or to punt the legal issues involved to other courts. The victims of brutal and senseless crimes such as those committed by defendants deserve better; so too do the trial courts of this state, the Legislature, and defendants themselves. Therefore, I respectfully dissent.

WE CONCUR:

CORRIGAN, J.

KRIEGLER, J. *

KRIEGLER, J. *

DISSENTING OPINION BY KRIEGLER, J.

A trial court may reasonably expect that a reviewing court will (1) not direct it to hold a hearing and make findings it has already made, and (2) provide some guidance explaining how the trial court can avoid error upon remand. The disposition in this case requires the trial court to consider issues it has already ruled on, and at the same time, provides not a whiff of direction on how the lower court is expected to cure the purported error. I respectfully dissent.

The dissent of the Chief Justice, which I join without reservation, correctly analyzes whether the sentences imposed on defendants Lionel Contreras and William Steven Rodriguez violate the Eighth Amendment's prohibition on cruel and unusual punishment as interpreted in People v. Caballero (2012) 55 Cal.4th 262 , 145 Cal.Rptr.3d 286 , 282 P.3d 291 ( Caballero ) and *412 *296 Graham v. Florida (2010) 560 U.S. 48 , 130 S.Ct. 2011 , 176 L.Ed.2d 825 ( Graham ). I write separately on three points. First, the factual issues on which the remand is based (consideration of mitigating factors in the crimes or defendants' lives) have already been resolved by the trial court, and remand for those purposes is a futile act. Second, the alternative grounds for remand-to create a factual record regarding the operation of the recently enacted Elderly Parole Program ( Pen. Code, § 3055 ) 1 and the 2017 regulations issued by the Department of Corrections and Rehabilitation pursuant to article I, section 32 of the California Constitution, as added by Proposition 57 2 -are issues this court can resolve de novo as a matter of statutory interpretation. The new statute and Regulations provide two ways to resolve this case with certainty, without prolonging the pain and suffering of the young women horribly victimized by these defendants. Third, the remand without meaningful guidance as to what the majority believes to be the constitutional limits of one strike sentences imposed on juvenile sexual predators regrettably sets the stage for years of continuing litigation on an issue that can and should be resolved in this appeal. **485 The reality is that since the time review was granted in this case, defendants' sentences have been substantially altered by the new Elderly Parole Program ( § 3055 ) and the conduct credit regulations generated under the authority of section 32, subdivision (b) of article I of the California Constitution, as added by Proposition 57. At the time of sentencing, it was understood that Contreras would receive his first parole eligibility hearing at age 74, and Rodriguez's first hearing would be at age 66. These dates are no longer accurate, although the majority addresses the case as if the dates continue to be operative. Under the elderly parole program, as a matter of law, defendants will have their first parole suitability hearing at age 60. In addition, defendants are eligible for a variety of conduct credits, which allow them to reduce their sentences by as much as 50 percent per year. If defendants take advantage of the new Regulations, they have the potential to advance their initial parole suitability hearings to a date prior to age 60. I would hold that a parole eligibility at age 60 or younger is not a de facto life sentence and does not otherwise violate the Eighth Amendment as to juvenile violent sexual offenders. *413 A. The remand order requires the sentencing court to consider factors it has already taken into account .

The language of the disposition is likely to leave the trial judge mystified. The majority commands "[t]he sentencing court ... to consider ... any mitigating circumstances of defendants' crimes and lives, and the impact of any new legislation and regulations on appropriate sentencing. The sentencing court is further directed to impose a time by which defendants may seek parole, consistent with this opinion." (Maj. opn., ante , at 229 Cal.Rptr.3d at p. 269, 411 P.3d at p. 462.) As to the first portion of the remand order, the trial court at the original sentencing hearings has already thoroughly considered "any mitigating circumstances of defendants' crimes and lives." ( Ibid. ) No claim is made by defendants that the court failed to consider any *297 mitigating factors as to the crimes and their life experiences. There is nothing left for the trial court to consider on this subject other than to repeat itself.

The trial court conducted separate sentencing hearings for the defendants, beginning with Rodriguez. The court "read and considered the probation report" and "read and considered the psychological evaluations" submitted on behalf of Rodriguez. The court considered argument from counsel for Rodriguez, who emphasized that her client fully acknowledged his responsibility for the crimes and has been "remorseful about it from the beginning." Counsel noted Rodriguez felt "a tremendous sense of shame and guilt for what he did and for what he did to these girls," pointing out that a psychological evaluation stated Rodriguez would carry that shame and guilt for the rest of his life. Counsel asked the court to consider Rodriguez's age at the time of the offenses (16), all the mitigating circumstances of his life ("unrelenting abuse throughout his childhood," as described by one reporting doctor), and the scientific evidence relating to the development of the brain.

The sentencing court acknowledged that Graham and Caballero were the controlling cases and that it "cannot give a juvenile offender the equivalent of life without parole." The court expressly recognized that the full statutory sentence required under the one strike sentencing scheme (§ 667.61) was inconsistent with the Eighth Amendment as to juvenile one strike offenders, and to impose a constitutional sentence, it would have to disregard the mandatory consecutive sentences otherwise required by the one strike sentencing scheme. The court agreed Rodriguez's "background is terrible," but tempered that comment with, "this crime is terrible." After considering argument from the prosecutor, the court observed that Rodriguez "was not a passive participant. He was a very active participant." The court believed it could not constitutionally impose a sentence of 75 years to life, "so probably the most I could give him is 50 to life." The court repeated that it had read the psychological report, which showed that "Mr. Rodriguez has had a very *414 difficult upbringing." But the court recounted that "it is awful and shocking how long this incident lasted even though **486 these girls were protesting and the great lengths Mr. Rodriguez and Mr. Contreras went to get these girls to a secluded place so they could have their way with them." The court described one victim's "heartbreaking" testimony that she asked the doctor performing the sexual assault examination if she could still wear her chastity ring. The court was understandably adamant that concurrent sentences were inappropriate, "because in my thinking, you don't get a free victim." The court stated it would have had no problem imposing a sentence of 200 years to life, "but the law says I can't do that." Rodriguez was sentenced to two consecutive terms of 25 years to life.

Given this record, there is no reason to remand Rodriguez's case for consideration of "any mitigating circumstances of defendants' crimes and lives." There is no mitigating evidence attendant to Rodriguez's crimes. Rodriguez has never had the audacity to suggest there is anything remotely mitigating about the crimes. The majority offers no clue as to what the mitigating evidence relating to the crimes might be. The victims will undoubtedly be shocked by the suggestion that there may be some aspect of Rodriguez's crimes that is mitigating. The court also considered the mitigating circumstances personal to Rodriguez, but found them dwarfed by the enormity of the offenses he committed. The court's findings are amply supported by the record. Rodriguez's crimes were not the product of youthful indiscretion. The brutality of the defendants' conduct *298 (see, dis. opn. of Cantil-Sakauye, C. J., ante , at 229 Cal.Rptr.3d at pp. 273-276, 411 P.3d at pp. 465-468) reveals the actions of violent sexual predators, not that of rogue youths misbehaving on a lark.

The record of the sentencing hearing as to Contreras essentially followed the same pattern as that of Rodriguez. The court stated that it read all of Contreras's "submissions including the two psychological reports." The court acknowledged it could not impose, under decisions of the United States and California Supreme Courts, the maximum sentence on the 21 guilty verdicts suffered by Contreras, which would have generated a sentence of as much as 620 years to life. The prosecutor argued that a minimum sentence of 50 years to life complied with Caballero , pointing out that Caballero leaves the actual number of years up to the trial court. With remarkable foresight, and anticipating this appeal, the trial court replied, "They are just going to tell us, 'you figure it out.' Then they are going to tell us, 'you are wrong' when it goes up to the Court of Appeals [ sic ]."

The court expressed its understanding and agreement with the research on the development of the juvenile brain. But the court questioned the honesty of Contreras, who denied responsibility, despite the overwhelming evidence of his guilt. The court discounted the value of the diagnoses of the psychologists, because they were based on statements of a defendant who was not *415 telling the truth. The court considered Contreras the "shot caller" in the crimes because "[h]e was definitely the guy in charge of this particular event. It was brutal and callous and ruthless." The court pointed to Contreras's manipulative attitude during his interview with law enforcement as an indication that "his brain is developed into who he is [and] who he was demonstrated on that whole event where he raped those girls. [¶] So he used a knife. He threatened them. I don't-I am not confident that people with that kind of psychology are rehabilitatable." The court imposed the same 50-year-to-life sentence given Rodriguez, but enhanced it by eight years for Contreras's use of a knife, again stating, "you don't get a free victim." Although the court felt that Contreras deserved the full term required by law "based on your attitude and your behavior in this case," he "was spared that sentence" under the Eighth Amendment.

As with Rodriguez, there are no mitigating circumstances relating to the crimes committed by Contreras for the trial court to consider on remand. The court considered the psychological reports on Contreras, but understandably found them of little value since he denied culpability. A remand to examine the mitigating circumstances of Contreras's crimes and his life experiences is an exercise in futility.

As to the first portion of the order on remand, the disconnect between the majority **487 opinion and the reality of what has already occurred in the trial court is startling. The trial court has made its findings on these issues. Those findings are supported by substantial evidence and are unchallenged. There is nothing left for the trial court to consider on these issues.

B. This court can resolve the issues relating to the Elderly Parole Program by statutory construction.

Because the trial court has already considered, and rejected, the notion of mitigating circumstances as to the crimes and defendants' lives, as a practical matter all that is left of the remand order is for the sentencing court "to consider ... the impact of any new legislation and regulations on appropriate sentencing," and "to impose a time by which defendants may seek parole, consistent with this opinion." (Maj. opn., *299 ante , at 229 Cal.Rptr.3d at p. 269, 411 P.3d at p. 462.) The application of the new section 3055 presents a legal question, not a factual one. No remand is needed.

The majority is unwilling to address whether an initial parole hearing for these defendants at age 60 violates the Eighth Amendment. 3 The reluctance is *416 understandable from the majority's point of view, because neither Graham , supra , 56 U.S. 48 , 130 S.Ct. 2011 (which prohibits life without parole for nonhomicide juvenile offenders) nor Caballero , supra , 55 Cal.4th 262 , 145 Cal.Rptr.3d 286 , 282 P.3d 291 (which prohibits a sentence in years exceeding a juvenile's life expectancy) support a categorical ban on initial parole suitability hearings for sexually violent juvenile offenders at age 60.

As the dissent of the Chief Justice demonstrates, the parole board at an elderly parole hearing will consider all relevant circumstances, including defendants' youth and the attributes of youth, in determining parole suitability. The majority is uncertain how the Elderly Parole Program will operate. But how the various parole statutes work in pari materia is a legal issue which we address de novo. (See Lexin v. Superior Court (2010) 47 Cal.4th 1050 , 1072, 1090-1091, 103 Cal.Rptr.3d 767 , 222 P.3d 214 ["on issues of statutory interpretation, our review is de novo," and "[i]t is a basic canon of statutory construction that statutes in pari materia should be construed together so that all parts of the statutory scheme are given effect"].) There is no reason for this issue of law to be decided in the first instance by the trial court.

The majority is unwilling to resolve this (and other issues) because of the "novel issues" (maj. opn., ante , at 229 Cal.Rptr.3d at p. 265, 411 P.3d at p. 458) associated with it. There is nothing novel about the interpretation of the statutes relating to the evidence that may be considered at an initial parole hearing. Certainly no evidentiary hearing is required to resolve that issue in this case.

C. The Proposition 57 Parole Regulations afford defendants an opportunity for an initial parole hearing prior to age 60.

The Proposition 57 Parole Regulations adopted by the Department of Corrections and Rehabilitation permit defendants to earn credits that approach 50 percent annually. As the Chief Justice correctly notes, Rodriguez may reduce his initial parole suitability date to age 57 simply by behaving in prison. (Regs., § 3043.2.) There are abundant additional credits defendants may earn, including: (1) milestone completion credit of 12 weeks per 12 month period (Regs., § 3043.3 ); (2) rehabilitative achievement credit of four weeks per year (Regs., § 3043.4 ); and (3) educational merit credit in increments of 90 days for a high school diploma or GED, and 180 days for the "Offender Mentor Certification Program," associate of arts or science degree, bachelor of arts or science degree, or post-graduate degree (Regs., § 3043.5 ).

*417 While it may not be possible for defendants to earn the full amount of credits, the fact **488 remains both have the ability to reduce their initial parole suitability date to below age 60.

There is no reason to remand to the trial court to determine how the credits will be awarded by prison officials. The Regulations *300 have the force of law, and we should presume that official duty will be regularly performed by the Department of Corrections and Rehabilitation. ( Evid. Code, § 664.) The majority's characterization of how the system of credits will operate as "novel" (maj. opn., ante , at 229 Cal.Rptr.3d at p. 265, 411 P.3d at p. 458) is again odd, considering that conduct credits have long been a component of California's sentencing law, and this court has addressed entitlement to conduct credits as a matter of law in various cases. ( In re Martinez (2003) 30 Cal.4th 29 , 34-37, 131 Cal.Rptr.2d 921 , 65 P.3d 411 ; In re Cervera (2001) 24 Cal.4th 1073 , 1077-1080, 103 Cal.Rptr.2d 762 , 16 P.3d 176 ; People v. Thomas (1999) 21 Cal.4th 1122 , 1125-1130, 90 Cal.Rptr.2d 642 , 988 P.2d 563 ; People v. Sage (1980) 26 Cal.3d 498 , 502-506, 165 Cal.Rptr. 280 , 611 P.2d 874 .) The current Regulations were drafted to aid in rehabilitation of inmates and reduction in the prison population through incentives. There is no reason to doubt, at this point, that most if not all of the conduct credit programs will be available to defendants. According to the Department of Corrections and Rehabilitation, "the credit-earning opportunities for Milestone Completion, Rehabilitative Achievement, Educational Merit, and Extraordinary Conduct, [ ] went into effect on August 1, 2017." (< https://news.cdcr.ca.gov/news-releases/2017/11/29/cdcr-issues-amended-proposition-57-regulations> [as of February 26, 2018].) The department has stated that it "gives inmates a strong incentive to participate in and complete rehabilitative programs." ( Ibid .) The program is in place, and no evidentiary hearing is required at this point to explore its operation.

The majority faults the failure of the two dissents to consider that Contreras and Rodriguez may commit misconduct in prison and forfeit their good conduct credits, suggesting this is a reason why the Regulations do not help to solve the Eighth Amendment issue presented. (Maj. opn., ante , at pp. ---- - ----.) According to the majority, "the record before us contains no information on how likely it is that any inmate can achieve a spotless prison record over a span of four or more decades." (Maj. opn., ante, at 131 Cal.Rptr.2d 921 , 65 P.3d 411 .) There is no need for an evidentiary record on this point, because abundant California case law-including decisions of this court-demonstrate that inmates convicted of the most serious offenses are capable of being incarcerated for extended periods with no disciplinary actions, or only trivial violations not resulting in loss of conduct credits. ( In re Shaputis (2008) 44 Cal.4th 1241 , 1249, 82 Cal.Rptr.3d 213 , 190 P.3d 573 ["Petitioner has remained discipline-free throughout his incarceration" spanning two decades];

*418 In re Rosenkrantz (2002) 29 Cal.4th 616 , 630, 682, 128 Cal.Rptr.2d 104 , 59 P.3d 174 [petitioner convicted of second degree murder engaged in no disciplinary misconduct in prison over 16 years]; In re Morganti (2012) 204 Cal.App.4th 904 , 909, 139 Cal.Rptr.3d 430 [inmate "has functioned without behavioral problems for almost 20 years"]; In re McDonald (2010) 189 Cal.App.4th 1008 , 1017, 118 Cal.Rptr.3d 145 [inmate "had been a model prisoner who had never been disciplined for serious misconduct in prison"]; In re Cerny (2009) 178 Cal.App.4th 1303 , 1305, 101 Cal.Rptr.3d 200 [inmate convicted of second degree murder in 1981 "has not been disciplined for violating prison rules"]; In re Scott (2004) 119 Cal.App.4th 871 , 898, 15 Cal.Rptr.3d 32 [inmate convicted of second degree murder in 1986 " 'has been disciplinary-free' in prison"]; see also In re Lawrence (2008) 44 Cal.4th 1181 , 1199, 82 Cal.Rptr.3d 169 , 190 P.3d 535 ["petitioner had been counseled eight times for misconduct, including as recently as 2005, but ... she has not been subject to any disciplinary actions"]; In re Stoneroad (2013) 215 Cal.App.4th 596 , 605, 155 Cal.Rptr.3d 639 *301 ["petitioner has an exemplary prison history; his only disciplinary citation was for 'leaving an unattended hotpot in his cell' in 1990"].) 4 **489 I disagree with the majority's speculative proposition that Contreras and Rodriguez will suffer a forfeiture of credits due to misconduct. They have every reason to comply and remain discipline free. The Regulations create an opportunity for inmates to demonstrate rehabilitation and advance the initial parole suitability date, a point the majority makes by citing Graham , supra , 560 U.S. at page 79, 130 S.Ct. 2011 , for the proposition that rehabilitation "depends on the incentives and opportunities available to the juvenile going forward." (Maj. opn., ante , at p. ----.) If Contreras and Rodriguez forfeit conduct credits due to serious misconduct, they will demonstrate a lack of parole suitability. ( In re Reed (2009) 171 Cal.App.4th 1071 , 1085, 90 Cal.Rptr.3d 303 .) But speculation as to their potential for misconduct in prison has no bearing on an Eighth Amendment analysis, because as the majority recognizes, Contreras and Rodriguez may be held in prison for life, and it is up to them to earn the right to release. (Maj. opn., ante , at pp. ---- - ----.)

I would address the applicability of the Regulations now, rather than deferring to some undefined fact finding hearing in the trial court.

*419 D. The remand order provides no guidance to the trial court on how the resentencing hearing should be conducted or how the court might formulate a sentence that does not violate the Eighth Amendment.

The trial court predicted the result in this case. The court worked to craft sentences that complied with Graham and Caballero , and now has been told it was wrong, but the majority offers no description of what would solve the problem it perceives. The trial judge did a commendable job performing the unpleasant assignment of presiding over a case involving violent sexual assaults on young women. He is entitled to some suggestions as to how the majority wants to remedy the problem it sees, particularly since any reduction of defendants' sentences will trample on the Legislature's authority to fix the punishment for crimes.

The Legislature has repeatedly determined that one strike juvenile offenders are not entitled to a youth offender parole hearing under section 3051. An early version of section 3051 did not exclude juvenile one strike offenders from a youth offender parole hearing (Legis. Counsel's Dig., Sen. Bill No. 260 (2013-2014 Reg. Sess.) as amended June 27, 2013, p. 5), but the legislation was amended several months later to specifically exclude this class of offenders (Legis. Counsel's Dig., Sen. Bill No. 260 (2013-2014 Reg. Sess.) as amended Sept. 3, 2013, p. 9). Subsequent amendments to the statute have maintained the exclusion of one strike juvenile offenders from section 3051 hearings. Instead, the Legislature has provided for a parole hearing for one strike juvenile offenders at age 60 under section 3055. Establishing a longer period of incarceration before parole suitability hearings for juvenile one strike offenders is consistent with the state's long-standing policy recognizing the unique danger of recidivism posed by violent sexual offenders. (See §§ 290 [registration requirement for sex offenders]; 6600 et seq. [civil commitment for sexually *302 violent predators]; Evid. Code, § 1108 [in a prosecution for a sexual offense evidence of defendant's commission of another sexual offense is not inadmissible to prove a disposition to commit the charged crime].) Case law from this court is replete with examples of recidivism by sex offenders. (See People v. Davis (2009) 46 Cal.4th 539 , 602-603, 94 Cal.Rptr.3d 322 , 208 P.3d 78 ; People v. Falsetta (1999) 21 Cal.4th 903 , 909-910, 89 Cal.Rptr.2d 847 , 986 P.2d 182 ; People v. Frank (1990) 51 Cal.3d 718 , 724-725, 274 Cal.Rptr. 372 , 798 P.2d 1215 .)

Any reduction in sentence in this case, or alteration of parole dates, will be inconsistent with statutory law. If existing law must be ignored in order to satisfy the Eighth Amendment as to an entire body of offenders, that is a policy decision best made by this court rather than a single trial court judge, whose ruling will not be binding, or even citable, in any other court of the *420 state. But at a minimum, the trial court is entitled to some vision of how to accomplish the result desired by the majority.

The majority's nonspecific remand order sets the stage for an extended Socratic dialogue between the trial court and the appellate **490 court, in which the trial court whittles away a de minimis portion of a one strike juvenile sentence, awaiting a response from the appellate court. It is not difficult to imagine this case going through several cycles of sentencing hearings and further remands on appeal. In the meantime, the victims of these 2011 offenses endure additional delay, uncertainty, and a lack of finality, a result inconsistent with the plain language of the California Constitution. "Victims of crime are entitled to finality in their criminal cases. Lengthy appeals and other post-judgment proceedings that challenge criminal convictions, frequent and difficult parole hearings that threaten to release criminal offenders, and the ongoing threat that the sentences of criminal wrongdoers will be reduced, prolong the suffering of crime victims for many years after the crimes themselves have been perpetrated. This prolonged suffering of crime victims and their families must come to an end." ( Cal. Const., art. I, § 28, subd. (a)(6).)

The unguided remand also has the potential to lead to arbitrarily disparate parole suitability dates for similarly situated one strike juvenile offenders. One judge might order a parole suitability hearing at age 45, another based on identical commitment offenses might order a hearing at age 50, and yet another might select age 55. The potential for disparate parole dates for similar offenses is not only unfair to defendants and an administrative nightmare for prison officials, it is inconsistent with the categorical requirements of Graham , supra , 560 U.S. 48 , 130 S.Ct. 2011 .

If a parole suitability hearing for juvenile one strike offenders at age 60 violates the Eighth Amendment, this court should say so now, and explain the contours of what the Eighth Amendment requires for this class of offenders. I respectfully dissent.

WE CONCUR:

CANTIL-SAKAUYE, C. J.

CORRIGAN, J.