In this case, the court must address the constitutionality of a sentence potentially subjecting Defendant Daniel Proctor to lifetime postrelease supervision and, in turn, to imprisonment for life without parole if he were later to commit any felony, including a property crime otherwise calling for probation. Proctor faces that prospect because he pled guilty to a sex offense — aggravated indecent solicitation of a child — for which he has received a permissible guideline sentence of probation. For Proctor, a man in his early 20’s, the statutory sentencing scheme could put him behind bars for 50 years if he were to shoplift a $1,000 ring or computer or to write a bad check for them. Given Proctor s circumstances and the peculiarly harsh result that could be inflicted on him, the sentence violates the protections against cruel and unusual punishment contained in § 9 of the Kansas Constitution Bill of Rights and the Eighth Amendment to the United States Constitution. The punishment may be considered grossly disproportionate in that context and incompatible with the general purposes of incarceration as a sanction in the criminal justice system. We, therefore, vacate the sentence imposed on Proctor to *890that extent and remand to the Saline County District Court for resentencing.
The governing statutes create the prospect of an exceptionally severe punishment — life in prison without parole is second only to a death sentence in its extremity — for persons convicted of designated sex offenses who then commit property crimes. For Proctor, the disparity between his criminal conduct and that punishment reflects an imbalance of a magnitude implicating constitutional protections. The Kansas sentencing statutes permit probation for both his underlying offense in this case and property crimes amounting to felonies. But the commission of those two offenses in that order may lead to life in prison with no prospect for release. Controlling authority from the United States Supreme Court and the Kansas Supreme Court construing the federal and state constitutional prohibitions on cruel and unusual punishment cannot be reconciled with that result. The sentencing scheme exacts a punishment harsher than those for murder, kidnapping, and other crimes the Kansas Legislature has designated as more serious than Proctor s. It also appears to be more severe than similar statutes applied to sex offenders in the vast majority of other states. Those are the ingredients of an unconstitutionally disproportionate punishment.
I. The Facts and the District Court Proceedings
The facts about Proctor s offenses are limited based on the disposition of the case in the district court. In 2010, Proctor pled guilty to one count of aggravated indecent solicitation of a child in violation of K.S.A. 21-3511 and two counts of lewd and lascivious behavior in violation of K.S.A. 21-3508(a)(2). Aggravated indecent solicitation entails “enticing or soliciting” a child younger than 14 years old to engage in an unlawful sex act. It is a severity level 5 person felony. Lewd and lascivious conduct requires that the perpetrator, motivated by a sexual urge, expose his or her genitals in the presence of a nonconsenting person. If the nonconsenting individual is younger than 16 years old, the offense is a severity level 9 person felony.
*891At the time of the plea, Proctor was 19 years old and had no criminal history. The factual basis for the plea showed the crimes took place in 2009. The victim was T.C., a 12-year-old boy. Proctor had known T.C. and his family for some time. For reasons that are not entirely clear from the record, Proctor lived with T.C. and his family for several months in 2009 and committed the crimes during that time. Proctor apparently cajoled T.C. into having manual and oral contact with Proctor s penis. Proctor also had manual contact with T.C.’s penis, behind, and anus. The record indicates multiple occurrences of illicit contact, but the precise or approximate number was not specified.
T.C. suffered no physical injuries. He did, however, have ongoing emotional problems and was in counseling. Other than recurrent insomnia, those problems were not detailed in the record. T.C. did not attend Proctor’s sentencing, and nobody appeared on his behalf or submitted a written statement for the district court’s consideration.
Proctor was himself a victim of sexual abuse in his early adolescence but apparently never received counseling. He also had not been through any sort of treatment program for sex offenders. Information submitted to the district court at sentencing indicated Proctor would likely benefit significantly from such a program.
Given his lack of criminal histoiy, Proctor fell in a border box on the sentencing grid for the aggravated solicitation conviction and faced incarceration for between 21 and 34 months. The border box sentences are treated as calling for presumptive incarceration. K.S.A. 2009 Supp. 21-4704(f). But a district court may impose a nonprison sentence on a border-box defendant upon a finding that he or she is amenable to “an appropriate treatment program” and participation in the program would be more effective than incarceration in “reducing the risk of . . . recidivism” consistent with “community safety interests.” K.S.A. 2009 Supp. 21-4704(f)(l)-(3). The district court made that finding based on the availability of sex offender treatment for Proctor. The district court imposed a standard sentence of 32 months in prison on Proctor for the aggravated solicitation conviction, put him on a 36-month probation, and ordered that he be placed in community corrections and participate *892in the treatment program. The district court imposed other restrictions and requirements on Proctor, such as refraining from use of alcohol or illegal drugs, obtaining gainful employment, and reporting as required to court officers supervising his probation. The sentence is not considered a departure. K.S.A. 2009 Supp. 21-4704(f).
The lewd and lascivious convictions were presumptive probation offenses. The district court granted Proctor probation on them. The district court imposed a standard 6-month sentence on each of those counts and ordered that they be run consecutive to one another and to the aggravated solicitation count, yielding a controlling prison term of 44 months. At sentencing, the district court told Proctor that he would be required to register as a sex offender under K.S.A. 22-4901 et seq. and that he would be subject to lifetime postrelease supervision under K.S.A. 2009 Supp. 22-3717(d)(1)(G), (d)(2)(F).
Immediately before sentencing, the district court took up Proctor s motion requesting the lifetime postrelease supervision statute be held unconstitutional as applied to him because it amounted to cruel and unusual punishment under both the state and federal constitutions. Both sides submitted extensive written arguments to tire district court. At die hearing, the district court made findings of fact about Proctor and the offenses and concluded lifetime supervised release could be constitutionally imposed on Proctor. Proctor has timely appealed that ruling.
II. Statutory and Constitutional Provisions
The framework for lifetime postrelease supervision is principally set forth in two statutes. Under K.S.A. 2009 Supp. 22-3717(d)(1)(G), anyone convicted of a defined “sexually violent crime” committed after June 30, 2006, and “released from prison, shall be released to a mandatory period of postrelease supervision for the duration of the person’s natural life.” In K.S.A. 2009 Supp. 22-3717(d)(2), the legislature designated the sexually violent crimes requiring lifetime postrelease supervision, including aggravated indecent solicitation of a child. The legislature empowered the Kansas Parole Board to administer postrelease supervision and *893statutorily set forth certain procedural requirements for the Board in determining violations and specified consequences for particular violations. K.S.A. 2009 Supp. 75-5217. Particularly pertinent to Proctor’s appeal, the statute provides: “If the violation results from a conviction for a new felony . . . the inmate shall serve the entire remaining balance of the period of postrelease supervision even if the new conviction did not result in the imposition of a new term of imprisonment.” K.S.A. 2009 Supp. 75-5217(c). A violation based on a misdemeanor conviction permits the parole board to impose “a period of confinement” up to the full length of the postrelease supervision. K.S.A. 2009 Supp. 75-5217(d).
The statutory language seems plain enough. A defendant sent to prison for one of the statutorily designated sexual offenses is subject to lifetime supervision after his or her release from custody. In turn, a person convicted of any felony while on lifetime supervision must be returned to prison for the rest of his or her life. The return is mandatory, and there is no administrative basis for mitigation or early release.
Through Executive Reorganization Order No. 34, issued January 21, 2011, Governor Sam Brownback abolished the parole board and replaced it with the prison review board. The prison review board now performs all of the duties and functions of the parole board. The change has no direct bearing on the issue raised or the arguments presented in Proctor’s appeal.
Convicts on postrelease supervision are subject to various restrictions impairing their liberty to a lesser degree than incarceration. Those restrictions include traveling without permission of an assigned parole officer, consuming alcohol without permission, and giving advance consent to searches of residences and vehicles and for drug and alcohol testing. See wiuw. doc. ks.gov/victim-services/ information/conditions-of-post-release-superoision (accessed April 24,2012). Proctor mentions those restrictions but does not premise his argument on them. We do not find them dispositive and discount them for purposes of our review.
The Eighth Amendment to the United States Constitution provides: “Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishment inflicted.” Section 9 *894of the Kansas Constitution Bill of Rights is identical in pertinent part except it prohibits “cruel or unusual punishment.” (Emphasis added.) The Kansas Supreme Court has interpreted the protections of § 9 to be the same as those in the Eighth Amendment notwithstanding the slight difference in wording. See State v. Scott, 286 Kan. 54, 93-94, 183 P.3d 801 (2008) (court notes cases so holding but acknowledges its prerogative to reexamine construction of the state constitutional protection in light of tire intentions of the framers of each provision). Proctor pins no argument on a suggestion that § 9 should be construed to afford individuals greater protection than the Eighth Amendment. We have no reason to assume such a variance and have no need to do so in addressing any of the parties’ assertions. We, therefore, follow the lead of Scott and treat the two provisions as substantively equivalent.
III. Standard of Review and Nature of Proctor’s Challenge
In reviewing the issue presented, we owe deference to the district court’s factual findings to the extent they are based on credibility determinations or the reconciliation of other conflicting evidence. This is not such a case. There were no material factual disputes presented to the district court, and no appellate argument rests on some discrepancy in the facts. This court must consider the undisputed facts and the sentencing statutes in light of the provisions in the Kansas and United States Constitutions prohibiting cruel and unusual punishments to determine if Proctor faces a sanction violating those protections. That presents a question of law over which appellate courts exercise unconstrained review. State v. Allen, 283 Kan. 372, 374, 153 P.3d 488 (2007). An appellate court must presume a statute to be constitutional and should reasonably construe its language in a way upholding its constitutionality if at all possible. State v. Laturner, 289 Kan. 727, 735, 218 P.3d 23 (2009).
Constitutional challenges to penal sanctions as cruel and unusual punishment take one of two forms: categorical or case specific. A categorical challenge submits a type or categoiy of punishment to *895be inappropriate in an array of cases based on key characteristics common to them. Historically, categorical attacks have been directed at imposition of the death penalty for certain crimes or on certain groups of offenders. See Roper v. Simmons, 543 U.S. 551, 578, 125 S. Ct. 1183, 161 L. Ed. 2d 1 (2005) (death penalty unconstitutional under the Eighth Amendment for defendants committing crimes as juveniles, i.e., under the age of 18); Coker v. Georgia, 433 U.S. 584, 592, 600, 97 S. Ct. 2861, 53 L. Ed. 2d 982 (1977) (four justices hold death penalty for the crime of raping an adult woman violates the Eighth Amendment and are joined in the judgment by two justices who would find the death penalty unconstitutional in every instance). Expanding on that authority, the United States Supreme Court recently upheld a categorical challenge to the imposition of life sentences without the possibility of parole on juvenile offenders found to have committed offenses other than homicide. Graham v. Florida, 560 U.S. 48, 130 S. Ct. 2011, 2034, 176 L. Ed. 2d 825 (2010).
By contrast, a case-specific challenge contends the imposition of a given sentence, typically incarceration for a term of years or life, is grossly disproportionate to the circumstances of a particular criminal incident and offender and, therefore, violates the prohibition on cruel and unusual punishment. That sort of proportionality attack depends upon the facts of a given case and directly affects only the particular defendant. See Graham, 130 S. Ct. at 2021-22.
Proctor does not assert a categorical challenge and relies solely on the disproportionality of lifetime postrelease supervision as it might affect him based on the circumstances of this case. The criteria for establishing categorical unconstitutionality arguably are more rigorous than those for demonstrating disproportionality of a punishment in a particular instance. See Graham, 130 S. Ct. at 2023 (categorical challenge bottomed on “objective indicia of national consensus”); Roper, 543 U.S. at 602-03 (O’Connor, J., dissenting) (evidence fails to support categorical challenge to execution of juveniles, but capital punishment might be unconstitutional in given cases). We, therefore, neither consider nor decide the constitutionality of lifetime postrelease supervision outlined in the *896Kansas statutes as a category of punishment or as it might be applied to anyone other than Proctor. We also note that, in the main, cases addressing the death penalty as cruel and unusual punishment have little direct relevance here. The courts have consistently acknowledged death to be materially different from any punishment based on incarceration, whatever its duration, and have generally discounted capital cases as inapposite in deciding other Eighth Amendment issues. Solem v. Helm, 463 U.S. 277, 294, 103 S. Ct. 3001, 77 L. Ed. 2d 637 (1983); Rummel v. Estelle, 445 U.S. 263, 271-72, 100 S. Ct. 1133, 63 L. Ed. 2d 382 (1980); see Ring v. Arizona, 536 U.S. 584, 614, 122 S. Ct. 2428, 153 L. Ed. 2d 556 (2002) (Breyer, J., concurring) (“This Court has held that the Eighth Amendment requires States to apply special procedural safeguards when they seek the death penalty.”).
IV. Ripeness
A. The Problem
Before outlining and applying the requisite constitutional analysis, we turn to a matter lurking ever so slightly beneath the surface. Given the sentencing scheme and Proctor’s current circumstances, the challenge to that part of his sentence calling for lifetime post-release supervision seems, in many respects, premature. That is, this court has not really been presented with a concrete legal controversy founded on a fixed set of material facts so much as a hypothetical projection of what could happen. Courts typically refrain from deciding abstract issues because those decisions amount to advisory opinions. See State ex rel. Morrison v. Sebelius, 285 Kan. 875, Syl. ¶ 15, 179 P.3d 366 (2008) (A court will consider neither issues that have become moot nor issues that have yet to ripen into a “fixed and final shape” and, instead, “remain[] nebulous and contingent.”). The State has suggested as much and with fair reason. But, as we explain, we believe we are obligated to press ahead.
As of the oral argument in this court, Proctor remained on probation. So long as he continues as a probationer, he neither is on nor even faces lifetime postrelease supervision under K.S.A. 2009 Supp. 22-3717. That sanction goes into effect only when a convict has been released from prison. Proctor would have to violate the *897terms of his probation. The district court would have to revoke and refuse to reinstate the probation. Proctor would then have to serve his 44-month sentence and be released to trigger the postrelease supervision. The thrust of Proctor’s argument for unconstitutionality, however, goes to the mandatory life-without-parole penalty that would accompany a conviction for a new felony. So not only would Proctor have to wind up in prison, he would have to commit some felony after his release. None of that has happened.
As we discuss below, the nature of the new felony could have a significant effect on the constitutional analysis as Proctor presents the issue and might be outcome determinative. Accordingly, deciding the case in this posture is, to say the least, unusual and, in some respects, troubling because of the artificiality of the circumstances critical to the decision. An appeal would present a tangible, well-formed controversy if it were asserted following the defendant’s conviction on a new felony and the prison commission’s initiation of revocation of supervised release.
But we are reasonably sure Proctor cannot defer his challenge in light of setded authority requiring immediate appeal of sentencing issues. The Kansas appellate courts have recognized that defendants waive most potential legal challenges to their sentences if they do not appeal within the statutorily permitted time. See Bryant v. State, 280 Kan. 2, 8, 118 P.3d 685 (2005) (defendant “did waive a challenge to his sentence by not appealing the sentence”); Wilkerson v. State, 38 Kan. App. 2d 732, 734, 171 P.3d 671 (2007); see K.S.A. 22-3608(c) (defendant has 10 days after judgment to appeal). The statutory provisions establishing lifetime supervised release and mandating revocation plainly affix the resulting incarceration as a punishment for the underlying sexual offense and not for the new felony. See K.S.A. 2009 Supp. 21-4704(e)(2) (In a presumptive imprisonment case, the district court must “pronounce the complete sentence,” including the period of postrelease supervision.). In other words, the return to prison for life results from the violation of a condition subsequent — no new felony convictions — imposed as part of the sentence for the sexual offense. A legal attack on that punishment, however contingent and remote it may be, must be lodged at the time sentence is pro*898nounced. The return to prison for violation of postrelease supervision is not imposed as punishment for the new felony. See K.S.A. 2009 Supp. 75-5217(c) (revocation of supervised release and return to prison mandatory even if criminal receives probation on the new felony). In turn, the constitutionality of the postrelease supervision scheme could not be challenged on an appeal from the new felony regardless of how sensible that might be in developing a complete record of historical facts necessary to adjudicate the issue.
Defendants could not rely on habeas corpus relief under K.S.A. 60-1507 to attack their actual incarceration for life following the commission of a new felony while on supervised release. Because the incarceration arises from the punishment imposed for the original sexual offense and not the new felony, the attack could be considered untimely. See K.S.A. 60-1507(f). While a court might afford relief from the time bar, nothing would require that result. Federal habeas corpus offers a similarly rickety vehicle. Federal courts may deny habeas review to state prisoners based on procedural defaults, such as untimely or successive filings, in their state collateral attacks. See Martinez v. Ryan, 566 U.S. _, 132 S. Ct. 1309, 1316, 1320, 182 L. Ed. 2d 272 (2012) (noting procedural default rule and limited equitable exception if counsel for prisoner in initial state collateral proceeding proved ineffective and not merely negligent in failing to raise a substantial claim of ineffectiveness of prisoner s trial counsel).
Proctor could not attack the constitutionality of the lifetime post-release supervision in a motion to correct an illegal sentence under K.S.A. 22-3504(1), which may be filed “at any time.” State v. Edwards, 281 Kan. 1334, Syl. ¶¶ 1, 2, 135 P.3d 1251 (2006) (defining illegal sentence and holding it does not include “a sentence [that] fails to conform to constitutional requirements”).
In short, while we share the State’s expressed concern about the indefiniteness of Proctor’s challenge — events bearing directly on the resolution of the challenge have yet to take place — we also must conclude that Proctor would have no certain path for asserting his legal claim at a juncture when the essential historical facts will have become fixed rather than conjectural. JProctor cannot be required to consign his assertion that life in prison without parole amounts *899to constitutionally impermissible punishment to the vagaries of judicial discretion exercised, if at all, years from now.
We suppose this appeal could be stayed. Should Proctor successfully complete his probation, he would be discharged without having been sent to prison and, thus, would not be on postrelease supervision. The appeal would then be moot. If Proctor were to fail on probation, he would at some point be released from prison on lifetime postrelease supervision. Were he then to be convicted of a felony, his return to prison for life would be triggered. The controversy would then be concrete, although additional factfinding might well be in order. But, of course, this appeal could be stayed for decades or longer awaiting an event that might never happen. The appeal would become moot upon Proctor s death or legislative modification of the postrelease supervision scheme so that life in prison no longer loomed as an irrevocable consequence. The idea of holding an appeal for 40 years or more seems, in a word, unorthodox. Although it theoretically could be done here in a single-issue appeal, the approach would seem unworkable in a case raising multiple issues. We would not presume such a solution to be appropriate here in the absence of clear precedent to that effect.
B. The Solution
Having determined that resolution of Proctor’s attack cannot be deferred to a more opportune time, we must consider how to deal with the unknowable circumstances that have yet to occur. Obviously, a set of facts must be hypothecated — a fancy euphemism, in this instance, for making something up- — to fill the void. We could, of course, conjure a scenario in which Proctor or some similarly situated defendant kills two people in cold blood during an armed robbery and, thus, is convicted of capital murder as the new felony causing his return to prison. In considering a claim that life in prison without parole would be grossly disproportionate under those circumstances, a court could not very well find for the defendant. If we were to choose forensic facts favoring the State in that way, the issue would never be fairly joined in the sense the hypothetical would preordain the outcome. At the same time, we *900should not suppose a future legislature that, in a paroxysm of law- and-order run amok, criminalizes overtime parking as a felony demanding extended incarceration — a recurrent foil in the United States Supreme Court’s debate of cruel and unusual punishment. See Ewing v. California, 538 U.S. 11, 35, 123 S. Ct. 1179, 155 L. Ed. 2d 108 (2003) (Stevens, J., dissenting); Harmelin v. Michigan, 501 U.S. 957, 963, 111 S. Ct. 2680, 115 L. Ed. 2d 836 (1991); Rummel, 445 U.S. at 274 n.11. That would about as surely compel the opposite result.
To test the rule, the hypothesis must, instead, be constructed to favor the defendant consistent with existing law. We, therefore, assume that Proctor will do something that, while not criminal, violates the terms of his probation. Let’s say he consumes alcohol and admits the violation. The district court finds Proctor’s conduct unacceptable and requires him to serve the 44-month sentence. Proctor would then be released from prison in 2016. At that point, Proctor would be in his mid-20’s, and he would be on lifetime postrelease supervision. See K.S.A. 2009 Supp. 22-3716(e).
Completing the necessary hypothetical, we assume that 2 years later Proctor commits and pleads guilty to a low-level felony. We might say he shoplifted a ring worth $1,100. That would be a severity level 9 nonperson felony. See K.S.A. 2011 Supp. 21-5801(b)(3). Or suppose he forged a check for $100 — a severity level 8 nonperson felony. See K.S.A. 2011 Supp. 21-5823(b)(l). Even with his conviction for aggravated indecent solicitation, Proctor would be looking at presumptive probation sentences for those offenses. Or perhaps he continued to consume alcohol and picks up a third DUI, a felony requiring he be incarcerated for at least 240 hours and imposing other restrictions. Unlike the other offenses, however, the felony DUI entails no criminal intent. It is a strict liability offense.
But die demands of K.S.A. 2009 Supp. 75-5217(c) are unyielding. Proctor would have to go to prison for the rest of his life without any possibility of early release, save for executive clemency or a change in the law. Assuming a normal life expectancy, Proctor would spend almost 50 years behind bars with no meaningful chance of returning to freedom.
*901We might vary the facts some. Suppose Proctor served the 44 months and had no more brushes with the law for 20 years. In his late 40’s, he knowingly writes an insufficient funds check for $1,000 to a medical provider to clear up an arrearage so that he could receive immediate care for a family member. Proctor’s conduct would amount to a severity level 9 nonperson felony. See K.S.A. 2011 Supp. 21-5821(b)(2)(A). Although Proctor would be in line for probation on the bad check charge, the violation of his post-release supervision would mandate his return to prison for the rest of his life — conservatively, some 25 to 30 years.
A scenario of that sort, grounded in realistic possibilities, presents a legitimate forensic tool for evaluating the constitutionality of lifetime postrelease supervision coupled with mandatory incarceration for life for anyone committing a felony while on supervision. We, therefore, use that general model to complete our analysis in this case. Our approach does not conflict with the presumption of constitutionality that attaches to duly enacted legislation. The presumption applies in two ways. First, the party challenging the statute must present evidence demonstrating the constitutional defect. That is, the party attacking the statute bears the burden of proof. As we have outlined, we believe Proctor has come forward with evidence to the extent possible. Second, a reviewing court generally should interpret the statutory language in a way favoring constitutionality, including giving a “narrow” reading to operative words to avoid a finding of unconstitutionality. But a court may not rewrite a statute in the guise of rendering a constitutional interpretation. Here, the relevant language in K.S.A. 2009 Supp. 22-3717 and K.S.A. 2009 Supp. 75-5217 is plain and, thus, not susceptible of a restrictive reading that would avert the constitutional deficiencies asserted.
V. The Eighth Amendment
A. Supreme Court Jurisprudence on Noncapital Sentences
Apart from death-penalty challenges, the United States Supreme Court has considered a series of cases based on Eighth Amendment arguments against lengthy periods of incarceration, typically life, for crimes other than homicide. The resulting jurisprudence *902is, however, a fragmented one. Those cases have been decided over the past 30 years, often on 5-4 votes or with only plurality opinions. Most of the cases have looked at recidivist statutes — so-called three-strikes laws — imposing markedly increased prison terms on defendants with past criminal records convicted of new felonies. Lockyer v. Andrade, 538 U.S. 63, 123 S. Ct. 1166, 155 L. Ed. 2d 144 (2003); Ewing, 538 U.S. 11; Solem, 463 U.S. 277; Rummel, 445 U.S. 263. One case, however, considered a statute imposing life without parole for a first-time felon convicted of drug possession. Harmelin, 501 U.S. 957. And the most recent considered life sentences without parole for juveniles convicted of serious non-homicide felonies. Graham, 130 S. Ct. 2011. Those cases form the backbone of the controlling modem authority applying the Eighth Amendment to punishments short of death. In Van Dyke v. State, 31 Kan. App. 2d 668, 672-77, 70 P.3d 1217 (2003), this court ably surveyed some of that law, most notably Lockyer and Ewing, which had been decided only months earlier. Because the governing cases span 3 decades and do not speak in chorus, we review that progression.
1. Rummel and Weems
The United States Supreme Court ushered in the modem era of Eighth Amendment review of prison terms in Rummel in which a five-justice majority upheld the constitutionality of a mandatory life sentence imposed under Texas law on a defendant for a third felony conviction. Rummel had convictions for fraudulent use of a credit card and for forgery when he was convicted of felony theft. He received a life sentence based on a recidivist sentencing statute that increased punishments for second and third felony convictions. Rummel argued the life sentence violated the constitutional ban on cruel and unusual punishment.
In an opinion written by then-justice Rehnquist, the majority rejected that argument based on several considerations. The majority recognized a state legislature has the authority to impose increasingly lengthy terms of imprisonment on repeat offenders. The State’s interest extended beyond simply punishing specific criminal offenses but also reflected a distinct policy purpose “in *903dealing in a harsher manner with those who by repeated criminal acts have shown that they are simply incapable of conforming to the norms of society as established in its criminal law.” Rummel, 445 U.S. at 276. The authority to determine if recidivists should be punished more harshly than first offenders and to determine the length of any prison term imposed on them rest “largely within die discretion of the punishing jurisdiction.” 445 U.S. at 285. The majority noted that the statute requiring a life sentence for Rummel did so “only after shorter terms of actual imprisonment . . . proved ineffective.” 445 U.S. at 278 n.17. The majority also found no constitutional violation even though all three convictions resulted from nonviolent property offenses. The majority observed: “If nothing else, the three-time offender s conduct supports inferences about his ability to conform with social norms that are quite different from possible inferences about first- or second-time offenders.” 445 U.S. at 283 n.27.
The majority also put considerable emphasis on evidence showing that offenders in the Texas prison system sentenced to life under the recidivist statute commonly received parole in as little as 12 years. While the majority recognized the sentence could not be treated as one for that term, given the uncertainty of parole, it also declined to presume Rummel would be incarcerated for life for purposes of any Eighth Amendment analysis. 445 U.S. at 280-81. And the majority distinguished a Mississippi sentencing statute imposing life without parole for certain recidivists convicted of at least one violent felony and two lesser felonies. 445 U.S. at 281.
Although the majority alluded to earlier caselaw considering punishments that might be fairly deemed “grossly disproportionate” to the offenses as indicative of an Eighth Amendment violation, the Rummel majority did not formally adopt that as a test or definition. 445 U.S. at 271-72. It did, however, acknowledge that “a proportionality principle” could “come into play in [an] extreme example,” such as criminalizing “overtime parking [as] a felony punishable by fife imprisonment.” 445 U.S. at 274 n.11. The ultimate inquiry ought to be guided by objective criteria to avoid judgments based on the “subjective views” or standards of individual justices. 445 U.S. at 275. The majority, however, discounted Rum*904mel’s comparison of his offenses to other conduct criminalized in Texas and recidivist statutes in other states. The majority characterized the interstate comparisons as subtle and not wholly convincing given the wide variance in those sentencing regimens and in classification of criminal offenses from jurisdiction to jurisdiction. 445 U.S. at 279-80. The majority, then, concluded Rummel’s sentence did not violate the Eighth Amendment. Rummel apparently was released from prison less than a year after the court ruled. See Solem, 463 U.S. at 297 n.25.
Writing for himself and three colleagues, Justice Powell dissented in what would serve as a template for his majority opinion 3 years later in Solem that struck down a South Dakota recidivist statute. Justice Powell would have disregarded any opportunity for parole as too speculative and, in turn, considered a mandatory life sentence grossly disproportionate to the circumstances, including Rummers criminal history. Rummel, 445 U.S. at 286-87 (Powell, J., dissenting). Justice Powell found proportionality to be a longstanding consideration in establishing and reviewing punishment under Anglo-Saxon law. In punishing recidivists, the concept of “disproportionality” embedded in the Eighth Amendment rests on “the relationship between the nature and number of offenses and the severity of the punishment inflicted upon the offender. 445 U.S. at 288. In turn, judicial review necessarily takes account of a collective “sense of justice” that asks whether the convict “deserves such punishment” and not simply whether it serves “a utilitarian goal.” 445 U.S. at 288.
Justice Powell identified three factors to be considered: (1) the nature of the offense, including circumstances specific to the offender and his or her criminal history, 445 U.S. at 295; (2) comparable sentencing schemes in other states, 445 U.S. at 296-99; and (3) punishments the same state metes out for other offenses, 445 U.S. at 300-01. Justice Powell viewed those factors and the standard of “gross disproportionality” as sufficient safeguards against excessive intrusion into a state’s legislative autonomy. 445 U.S. at 306. He submitted the courts could not shirk enforcement of the Eighth Amendment in the name of deference. 445 U.S. at 303-04. After applying those standards, Justice Powell concluded *905Rummel’s sentence rose to the level of cruel and unusual punishment. 445 U.S. at 307.
Before moving on, we pause for a historical digression to Weems v. United States, 217 U.S. 349, 30 S. Ct. 544, 54 L. Ed. 793 (1910), something of an odd case cited in both the majority and dissenting opinions in Rummel and, in passing, in the developing law of proportionality of noncapital sentences under the Eighth Amendment. Weems was a government clerk in the Philippines sentenced under its laws for falsifying an official record to show payments to workers at two lighthouses. As the majority in Weems characterized the offense, it required only the making of a false entry in a government record without regard to any financial gain or loss. The clerk received a sentence known as cadena temporal requiring that he be imprisoned for 15 years, shackled continuously at both the hands and feet, and made to perform “hard and painful labor.” Upon release, he was subject to an array of civil disabilities.
Weems argued tire punishment violated the Philippine Constitution’s ban on cruel and unusual punishment, a provision that mirrored the Eighth Amendment. In reviewing the case, the United States Supreme Court drew on its own and other American authority. The majority declared the punishment to be impermissibly cruel and unusual and, in doing so, effectively considered the proportionality of the offense to the sentence. 217 U.S. at 366-67, 380-81. Thus, judicial proponents of Eighth Amendment challenges to terms of imprisonment cite Weems as longstanding authority recognizing the theory of proportionality. See Graham, 130 S. Ct. at 2021; Rummel, 445 U.S. at 289-90 (Powell, J., dissenting). But opponents of the theory or its application in a given case to strike down a sentence isolate Weems as a peculiarly unique decision in which the punishment entailed a significant term of imprisonment combined with conditions — the shackling and painful work — that partake of torture rather than merely deprivation of liberty. See Harmelin, 501 U.S. at 990-92; Rummel, 445 U.S. at 272-74. Ultimately, Weems contributes little to the modem debate and reflects more of a footnote than a point of reference.
*9062. Solem
Three years after deciding Rummel, the United States Supreme Court took up Solem, another recidivist case, challenging a life sentence without parole imposed on a repeat offender for negotiating a $100 check knowing the account to have been closed. The case, thus, took the possibility of parole out of tire constitutional calculus, a point the Rummel majority had weighed in upholding a life sentence. In Solem, a five-justice majority found the South Dakota statute amounted to unconstitutionally cruel and unusual punishment. 463 U.S. at 303. Justice Blackmun switched sides, providing the decisive vote in Solem. But he did so without writing separately, so he did not divulge his reasons. Justice Blackmun never authored an opinion addressing prison sentences under the Eighth Amendment. Between Rummel and Solem, Justice O’Connor replaced Justice Stewart. Each was a vote to uphold the constitutionality of the respective sentencing statutes.
Writing for the majority, Justice Powell largely followed tire path he had laid out in his dissent in Rummel. Defendant Plelm had been convicted of six nonviolent felonies between 1964 and 1979, including burglaiy, grand larceny, and a third DUI. In 1979, apparently during a drinking binge, he passed a check on an account he knew to be closed. Helm pled guilty. Under South Dakota’s recidivist statute, a defendant with three past felonies could be sentenced to life in prison for a fourth conviction. The life sentence carried no possibility of parole. Helm got life.
Justice Powell reiterated tire Eighth Amendment requirement that a punishment be proportionate to the offense and recounted the historical basis for that conclusion. Solem, 463 U.S. at 284-88. Based on that survey of the law, the Solem majority held that while a reviewing court owes “substantial deference to the broad authority that legislatures necessarily possess in determining the types and limits of punishments,” the Eighth Amendment requires “a criminal sentence ... be proportionate to the crime for which the defendant has been convicted” and “no penalty is per se constitutional.” 463 U.S. at 290. To determine the constitutionality, Justice Powell reintroduced the three factors he outlined in his dissent *907in Rummel: (1) “the gravity of the offense and tire harshness of the penalty”; (2) sentences imposed for other crimes within the jurisdiction; and (3) sentences imposed in other jurisdictions for tire same offense. 463 U.S. at 292. As Justice Powell envisioned and applied those standards, a defendant’s personal characteristics and tire particulars of his or her criminal offenses would be folded into the mix. 463 U.S. at 296-97 & n.22, 303 n.32 (noting Plelm to be 36 years old and facing life in prison without parole).
Justice Powell dismissed criticism of the comparative elements as inexact and unwieldy. While applying the Eighth Amendment to punishments based on varied terms of imprisonment for varied offenses may involve line-drawing among less distinct categories than between capital and noncapital punishments, courts regularly undertake similarly challenging differentiation in other contexts. See 463 U.S. at 294-95.
At the time of Helm’s conviction, life without parole was tire harshest punishment that South Dakota could impose on a criminal defendant for any offense. And it was “far more severe” than the punishment considered in Rummel, 463 U.S. at 297. Justice Powell characterized the bad check conviction “as among the less serious offenses.” 463 U.S. at 296. Looking at other crimes in South Dakota, only a few warranted similarly harsh sentences. 463 U.S. at 298-99. And the Supreme Court concluded Helm would not have received as harsh a sentence in virtually any other state. 463 U.S. at 299-300.
Based on those circumstances, the majority found that Helm’s sentence of life without parole violated the Eighth Amendment’s prohibition on cruel and unusual punishment. As we discuss later, the majority’s method of analysis matches the one the Kansas Supreme Court first outlined in State v. Freeman, 223 Kan. 362, 367, 574 P.2d 950 (1978), and continues to apply in determining the constitutionality of punishments under § 9 of the Kansas Constitution Bill of Rights.
In Solem, Chief Justice Burger dissented, writing for himself and Justices White, Rehnquist, and O’Connor. He characterized Rummel as rejecting proportionality review of prison sentences under the Eighth Amendment. 463 U.S. at 307. Chief Justice Burger *908suggested the majority’s position could not be justified if Rumrnel were accorded proper stare decisis effect. 463 U.S. at 311-12. And he suggested there were no material differences between the circumstances in Rummel and Solem. 463 U.S. at 315. History has plainly overtaken Chief Justice Burger’s position. In Harmelin, seven justices, including Justice White and Justice O’Connor, signed off on opinions recognizing some form of proportionality analysis in measuring prison sentences as cruel and unusual punishment. See Harmelin, 501 U.S. at 996-97 (Kennedy, J., concurring); 501 U.S. at 1013-14 (White, J., dissenting).
Chief Justice Burger’s characterization of Rummel appears strained. He cited a passage in Rummel, 445 U.S. at 274, in which Justice Rehnquist launched what appears to be a rhetorical foil declaring that “one could argue” that “the length of the sentence actually imposed is purely a matter of legislative prerogative” without being contradicted by any previous decision of the Court. Solem, 463 U.S. at 307. In other words, Justice Rehnquist submitted no United States Supreme Court decision had struck down a sentence as cruel and unusual punishment based only on its length. But that is a far ciy from holding the Eighth Amendment precludes such a determination. And in a footnote omitted from Chief Justice Burger’s discussion, Justice Rehnquist specifically recognized proportionality might well come into play in extreme circumstances. Rummel, 445 U.S. at 274 n.11; see Harmelin, 501 U.S. at 1013 (White, J., dissenting) (noting that Rummel, 445 U.S. at 272, 274 & n.11, “recognized that the Eighth Amendment contains a proportionality requirement. . . [that] would come into play in some extreme, nonfelony cases.”). To be sure, the thrust of the majority opinion in Rummel emphasized the extraordinaiy uncommonness of those circumstances.
3. Harmelin
In Harmelin, a majority of the United States Supreme Court found that a Michigan statute imposing a sentence of life without parole for possession of more than 650 grams of cocaine did not offend the cruel-and-unusual clause of the Eighth Amendment. Harmelin, 501 U.S. at 994-96. Harmelin received that punishment *909despite having no past felony convictions. The deeply divided Court produced only one brief rationale commanding five votes: For a serious crime, Michigan could enact a severe punishment that expressly declined to take into account any mitigating factors a particular defendant might present, such as the lack of any significant criminal history, without violating the Eighth Amendment. The State’s prerogative to criminalize and punish conduct that way was not diminished even though the penalty — life without parole— was the harshest short of death. 501 U.S. at 994-96 (Scalia, J.).
Justice Scalia, joined only by then-Chief Justice Rehnquist, took the position that even the harshest sentence of imprisonment could never be found cruel and unusual under the Eighth Amendment on the grounds it was disproportionate to the criminal offense. Harmelin, 501 U.S. at 965. He also argued that the standards for determining a disproportionate sentence were sufficiently indefinite and malleable that judges applying them had “an invitation to imposition of subjective values” through which they could improperly override otherwise appropriate legislative decisions. 501 U.S. at 986.
Writing for himself and Justices Souter and O’Connor, Justice Kennedy recognized what he termed a “narrow proportionality principle” in the Eighth Amendment governing noncapital sentences. 501 U.S. at 996-97 (Kennedy, J., concurring). He pointed out the marked deference that the courts must accord a state’s decision in determining conduct to be criminalized and punishment then to be administered. Those decisions may be shaped by differing penological theories, and it is not the courts’ business to direct, let alone dictate, a particular theory. The Eighth Amendment neither imposes some theory nor requires that a criminal sanction promote any particular mix of retribution, deterrence, incapacitation or rehabilitation. 501 U.S. at 999-1000.
In considering proportionality, the courts must be guided, to the greatest extent possible, by objective factors, according to Justice Kennedy. Ultimately, Justice Kennedy distilled those considerations and the developing jurisprudence into a conclusion that “[t]he Eighth Amendment does not require strict proportionality between crime and sentence.” 501 U.S. at 1001. But those “extreme *910sentences that are ‘grossly disproportionate’ to the crime” fail under the Eighth Amendment. 501 U.S. at 1001.
In applying drat standard, Justice Kennedy would first look at the crime committed and the sentence imposed to discern “an inference of gross disproportionality.” 501 U.S. at 1005. In what Justice Kennedy anticipated would be rare instances permitting that inference, tire Court should then consider how the particular state punishes crimes typically regarded as worse that the defendant’s crime of conviction. And the Court ought to consider how other states punish offenses that essentially match the crime of conviction. Those comparisons, drawn from Solem, would then test tire initial inference and validate it if the crime of conviction were punished substantially more harshly. 501 U.S. at 1005. In applying those standards, Justice Kennedy concluded, based on the seriousness of the drug offense alone, Harmelin’s sentence was not grossly disproportionate for Eighth Amendment purposes. 501 U.S. at 1004. He pointed out that the penalty reflected a studied legislative determination to impose an extremely harsh sentence and to deprive the sentencing court of any room to mitigate the result. 501 U.S. at 1006. He and his two colleagues joined in the judgment denying relief to Harmelin.
The four dissenting justices would have found Harmelin’s sentence of life without parole to have violated the Eighth Amendment. They, too, recognized an Eighth Amendment principle of proportionality applicable to sentences of incarceration. 501 U.S. at 1013-14 (White, J., dissenting). The dissenters would have applied the Solem factors, along with other case-specific considerations, including “the defendant’s personal responsibility and moral guilt.” 501 U.S. at 1022-23. The approach operated more in the nature of a totality-of-the-circumstances analysis than the segmented, sequential review of Justice Kennedy’s “narrow proportionality” test. Because Michigan had no death penalty, Justice White noted that life in prison without parole was the most severe penalty the state could inflict on a convicted criminal, although possession of even a large quantity of cocaine could not be placed on a par with the crimes of rape and murder. There appeared to be just one other state imposing such a harsh and unyielding sen*911tence for possession of cocaine and, then, only for quantities some 15 times greater than the Michigan statute. 501 U.S. at 1025-26.
Justice White’s dissent in Harmelin is also noteworthy for its endorsement of the proportionality methodology laid out in Solem, even though Justice White joined Chief Justice Burger’s dissenting opinion in Solem rejecting that approach. In his dissent in Harmelin, Justice White observed: “[T]he Solem analysis has worked well in practice.” 501 U.S. at 1015.
4. Ewing and Lockyer
The United States Supreme Court returned to the constitutionality of state recidivist statutes in Ewing, 538 U.S. 11, and Lockyer, 538 U.S. 63, companion cases challenging California's three-strikes law. A fractured majority in Ewing upheld a sentence of 25-years-to-life in prison for a defendant convicted of stealing three golf clubs worth just under $1,200 — a coda to a career in crime including a felony theft, several burglaries, a robbery, and a host of misdemeanors. Five justices held the sentence did not violate the Eighth Amendment. Justice Scalia reiterated his position from Harmelin that the Eighth Amendment permits no proportionality review of terms of imprisonment and any effort to engage in such review becomes an impermissible judicial foray into “evaluating policy.” Ewing, 538 U.S. at 31-32 (Scalia, J., concurring in judgment). In an aside, he observed, however, that the plurality’s reasoning “does not convincingly establish drat 25-years-to-life is a ‘proportionate’ punishment for stealing three golf clubs.” 538 U.S. at 31. Justice Thomas, who came on the Court after Harmelin had been decided, concluded “the Eighth Amendment contains no proportionality principle” and, therefore, joined in the judgment affirming the sentence. 538 U.S. at 32 (Thomas, J., concurring in judgment). He has essentially aligned with Justice Scalia in rejecting proportionality as a legitimate method of analysis in these cases.
In a plurality opinion, Justice O’Connor, joined by Chief Justice Rehnquist and Justice Kennedy, applied the “narrow proportionality” analysis Justice Kennedy used in Harmelin. Ewing, 538 U.S. at 23-24. The plurality concluded the sentence was not grossly disproportionate for an offender committing a third felony, even an *912expensive shoplifting. It was not, therefore, cruel and unusual. 538 U.S. at 30-31. The opinion relied on the historical recognition of “[rjecidivism ... as a legitimate basis for increased punishment.” Ewing, 538 U.S. at 25. Statutes imposing increased penalties for successive crimes reflect a legislative “policy choice that individuals who have repeatedly engaged in serious or violent criminal behavior, and whose conduct has not been deterred by more conventional approaches to punishment, must be isolated from society in order to protect the public safety.” 538 U.S. at 24.
The plurality also found the California sentencing provisions to be calibrated in ways that diminished any arguably excessive dis-proportionality. A defendant with one past conviction for a serious or violent felony would face a doubled sentence on a second felony conviction. Only on a third felony conviction would a defendant receive a sentence of the length imposed on Ewing for stealing the golf clubs. See 538 U.S. at 16. In addition, the plurality noted that under California’s criminal code, the prosecutor and the sentencing court had considerable latitude in classifying certain offenses so that they would not be treated as felonies counted as one of the strikes requiring enhanced punishment. 538 U.S. at 16-17, 29. Because the plurality found no inference of gross disproportionality, it did not survey punishments for other crimes in California or how other states treat recidivists comparable to Ewing.
In dissent, Justice Stevens would have found the sentence to be constitutionally prohibited cruel and unusual punishment based on “a broad and basic proportionality principle” underlying the Eighth Amendment. 538 U.S. at 35. He was joined by Justices Breyer, Ginsburg, and Souter in that view. Justice Souter’s appearance with the dissenters marked his defection from Justice Kennedy’s analytical approach outlined in Harmelin. Justice Breyer wrote an extended dissent applying the “narrow proportionality” test for forensic purposes and concluded, even under that analysis, the sentence violated the Eighth Amendment. 538 U.S. at 36, 53. He suggested the initial analytical swipe aimed at determining an inference of disproportionality ought to focus on: (1) the actual time the defendant likely will serve in prison; (2) the criminal conduct *913triggering the sentence; and (3) the defendant’s criminal history. 538 U.S. at 37.
In Lockyer, 538 U.S. at 77, a majority of the United States Supreme Court found a state prison inmate was not entitled to federal habeas corpus relief from his sentence under the California three-strikes law on the grounds it amounted to cruel and unusual punishment. The majority suggested the lower courts’ rulings affirming the sentence were neither contrary to nor an unreasonable application of clearly established federal law — gateway requirements for a habeas corpus action. The United States Supreme Court decided the case with the same 5-4 split as Ewing. Justice O’Connor’s majority opinion is most noteworthy, perhaps, for its understated characterization of the Court’s Eighth Amendment treatment of prison sentences as an “area [that has] not been a model of clarity.” 538 U.S. at 72. Those decisions, she observed, “have not established a clear or consistent path for courts to follow.” 538 U.S. at 72.
5. Graham and Miller
The United States Supreme Court again looked at a term of imprisonment as cruel and unusual punishment about 2 years ago in Graham. A majority of the court held that tire Eighth Amendment supported a categorical ban on sentencing juvenile offenders to life without parole for crimes other than homicide. In his majority opinion, Justice Kennedy discussed United States Supreme Court precedent addressing the constitutionality of noncapital sentences and reiterated his narrow proportionality analysis. 130 S. Ct. at 2021-23. He also noted that penological justifications for a punishment based on retribution, deterrence, incapacitation, and rehabilitation figure into an examination of disproportionality, picking up on a consideration he mentioned in Harmelin. 130 S. Ct. at 2028-30.
The decisive factor for Justice Kennedy lay in psychological and other scientific evidence demonstrating juveniles’ thought processes and brain functioning to be less fully developed than those of adults. And that lack of maturation affects the ability of the typical juvenile to assess and appreciate fully consequences of criminal activity. Juveniles generally are, then, less morally culpable *914than adults and have a greater capacity to reform their socially and legally unacceptable behaviors. 130 S. Ct. at 2026-27 (citing Roper, 543 U.S. at 569-70). Accordingly, the Court held a punishment banishing juveniles to lifetime incarceration for crimes other than homicide violates the Eighth Amendment. Justices Stevens, Ginsburg, Breyer, and Sotomayor joined in the opinion.
Because the majority decision turned on characteristics of juvenile offenders as a class and the categorical imposition of an especially harsh punishment on them for crimes less serious than homicide, it has only limited precedential application here, since we address a punishment imposed on an adult. But Graham is of interest in several respects. Chief Justice Roberts and Justices Alito and Sotomayor joined the Court after Ewing, the last major Eighth Amendment case challenging incarceration rather than the death penalty. Thus, Graham provided at least a veiled look at their views. And Justice Kennedy used Graham to actively promote his narrow proportionality test.
In voting with the majority, Justice Sotomayor recognized proportionality review as a component of Eighth Amendment jurisprudence. She and Justice Ginsburg also signed off on a brief concurring opinion from Justice Stevens effectively endorsing a broad proportionality test of the sort favored by the dissenters in Harmelin and Ewing. 130 S. Ct. at 2036 (Stevens, J., concurring). Justice Kennedy’s narrow proportionality test, therefore, failed to gamer unqualified support from a Court majority in Graham.
Chief Justice Roberts joined in the judgment of the Court but declined to decide the case as a categorical challenge. He wrote a separate opinion supporting reversal because the punishment of life without parole was unconstitutionally disproportionate as to Graham alone based on the circumstances of his crime and background. 130 S. Ct. at 2039 (Roberts, C.J., concurring in judgment). Chief Justice Roberts applied a narrow proportionality test of the sort Justice Kennedy has advanced. 130 S. Ct. at 2039. But he did not firmly commit himself, acknowledging both the “serious and thoughtful questions” from some of his colleagues about the propriety of any proportionality review for prison sentences and the lack of clarity in the Court’s jurisprudence addressing the issue. *915130 S. Ct. at 2036-37. Despite his stated reservations, Chief Justice Roberts cited Solem with favor several times in outlining the applicable legal principles. 130 S. Ct. at 2037-39.
Justice Alito resisted venturing into that issue. Justice Alito viewed the question before the Court as presenting only a categorical attack on tire life sentence and declined to consider or comment on the as-applied proportionality doctrine. Graham, 130 S. Ct. at 2058 (Alito, J., dissenting). He joined in that portion of Justice Thomas’ dissent rejecting life without parole for juveniles as categorically disproportionate punishment for noncapital offenses, but he did not subscribe to Justice Thomas’ alternative rationale that proportionality amounts to an illegitimate method of analysis. 130 S. Ct. at 2058.
As noted, Justice Kennedy endeavored to advance his narrow proportionality test in Graham by citing his own plurality opinion in Harmelin as authoritative and characterized it as “the controlling opinion” in that case. 130 S. Ct. at 2021-22. But that appears to overstate the support for narrow proportionality. It has never garnered more than three votes in any given case. In Harmelin, Justice Souter signed off on that position only to step back from it in Ewing. And Justice Kennedy’s approach “controlled” the majority judgment only because other justices would not incorporate any form of proportionality into the Eighth Amendment as applied to sentences of imprisonment.
Justice Kennedy’s concept of proportionality set out in Harmelin reflects the narrowest ground supporting the judgment there and in Ewing and, thus, may be considered controlling in that limited sense. See Marks v. United States, 430 U.S. 188, 193-94, 97 S. Ct. 990, 51 L. Ed. 2d 260 (1977) (When no rationale commands five votes, the holding may be considered “ ‘that position taken by those [justices] who concurred in the judgment[] on the narrowest ground.’ ”). Whether Justice Kennedy’s position should be treated as binding authority is a matter of debate. The Marks approach presupposes the positions of the justices in the majority are compatible insofar as they recognize a common right or mode of analysis but disagree as to scope. 430 U.S. at 194 (When two justices would afford broad First Amendment protection to pornographic *916literature and would reverse a finding that a given book was legally obscene and three other justices adopted a more limited protection but would also reverse the judgment, the plurality of three reflects the controlling decision of the Court on the narrowest ground.). That cannot be said of the judgments in either Harmelin or Ewing, since two justices essential to the majority would reject any form of proportionality analysis, including the limited one Justice Kennedy espouses. See Nichols v. United States, 511 U.S. 738, 745-46, 114 S. Ct. 1921, 128 L. Ed. 2d 745 (1994) (application of Marks to some divided decisions may prove difficult when a “lowest common denominator” is not readily apparent; resulting confusion of that sort suggests “a reason for reexamining” the substantive issue); United States v. Carrizales-Toledo, 454 F.3d 1142,1151 (10th Cir. 2006) (Marks inapplicable when plurality and concurring opinions “take distinct approaches” or “the various opinions supporting the Court’s decision are mutually exclusive”); United States v. Alcan Aluminum Corp., 315 F.3d 179, 189 (2d Cir. 2003).
At the close of its 2011 term, the United States Supreme Court issued a decision holding mandatory sentences of life without parole imposed on juveniles committing murder to be constitutionally cruel and unusual punishment. Miller v. Alabama, 567 U.S. _, 132 S. Ct. 2455, 183 L. Ed. 2d 407 (2012). The decision effectively expanded the principal rationale of Graham — that juveniles as a class are less criminally culpable and more susceptible to rehabilitation than adults — to require a judge or jury to take into account a juvenile offender’s individual characteristics before imposing a sentence of life without parole for murder. Miller, 2012 WL 2368659, at *11. As a categorical challenge built on that premise, Miller is, like Graham, largely uninformative as to how this case should be resolved.
The majority opinion, written by Justice Kagan, made passing reference to proportionality of punishment based on characteristics of the crime- and of the criminal as a component of Eighth Amendment analysis. Miller, 2012 WL 2368659, at *7 (citing Weems, 217 U.S. at 367); Miller, 2012 WL 2368659, at *17. But it tiren never fleshed out the attributes of that proportionality or how it should be measured. Rather, the opinion focused on the differences that *917set juveniles apart from adults, as outlined in Roper and Graham, and concluded those differences may warrant less severe punishment and necessarily require a tailored imposition of an extreme penalty such as life in prison without parole. Miller, 2012 WL 2368659, at *7-11. The majority, for example, distinguished Harrnelin precisely because that decision, upholding mandatory life sentences for certain drug traffickers, “had nothing to do with children and did not purport to apply its holding to the sentencing of juvenile offenders.” Miller, 2012 WL 2368659, at *13. Justices Kennedy, Ginsburg, Breyer, and Sotomayor joined in the opinion.
Justice Thomas, joined by Justice Scalia, dissented, repeating his thesis that proportionality has no place in determining whether a given punishment violates the Eighth Amendment. Miller, 2012 WL 2368659, at *26 (Thomas, J., dissenting). In separate dissenting opinions, Chief Justice Roberts and Justice Alito struck common themes in suggesting the majority opinion rests more on the predilections of its supporting justices than sound Eighth Amendment precedent and due respect for legislative decisionmaking. Chief Justice Roberts argued that the sentencing statutes of the federal government and more than half the states impose mandatory sentences of life in prison for juveniles who kill in certain circumstances and more than 2,000 young people have received such a sentence, thereby evincing broad support for that punishment and undercutting any notion of a national consensus to the contrary. Miller, 2012 WL 2368659, at *20 (Roberts, C.J., dissenting). Absent such a consensus, a punishment cannot categorically violate the Eighth Amendment, according to Chief Justice Roberts’ analysis. 2012 WL 2368659, at *20. Justice Alito dissented because the majority rejected the sentencing determinations of a majority of state legislatures with regard to appropriate terms of incarceration for especially serious offenses, an approach he characterized as incompatible with the Eighth Amendment and fundamental democratic principles. Miller, 2012 WL 2368659, at *33 (Alito, J., dissenting). But Justice Alito’s dissent suggested little about how (or even if) he would apply proportionality in deciding a challenge such as Proctor’s.
*918It is hardly self-evident that Justice Kennedy’s narrow proportionality test would command five votes today given the fractured precedent and the changes on the Court since Ewing, the last time the issue was directly addressed in a noncategorical challenge to a prison sentence imposed on an adult. The Solem decision actually reflects the most recent decision in which five justices agreed on a clear rationale disposing of a case dealing with a term of imprisonment as cruel and unusual punishment of an adult. It is also less than self-evident that Solem would garner five votes today. But Solem has never been overruled, and as Chief Justice Roberts’ separate opinion in Graham suggests, the decision remains at least a viable part of Eighth Amendment law.
B. Settled Principles
From the United States Supreme Court precedent, we may distill some settled principles guiding the evaluation of Proctor’s claim. First and foremost, in applying the Eighth Amendment to punishment entailing incarceration, the courts owe great deference to legislative determinations, especially allowing for differing views on penological policy. The theme of deference runs throughout the authority and was most recently recapitulated by Chief Justice Roberts: The Court has “emphasized the primacy of the legislature in setting sentences.” Graham, 130 S. Ct. at 2037; see Solem, 463 U.S. at 290 (“Reviewing courts . . . should grant substantial deference to the broad authority that legislatures necessarily possess in determining the types and limits of punishments for crimes.”); Van Dyke v. State, 31 Kan. App. 2d 668, Syl. ¶ 3, 70 P.3d 1217 (2003). But, as both Graham and Solem illustrate, judicial deference has limits; it is not merely a euphemism for blind adherence. The ultimate test for judging incarceration as cruel and unusual punishment does not depend on a closely calibrated relationship between the offense and the prison term. That sort of proportionality would contravene the broad authority properly accorded legislatures in setting punishments for crimes based on particularized penological policies and public-safety goals. From a practical standpoint, a constitutional test based on a tailored proportionality between crime and punishment would be difficult to administer given *919the often significant variations in sentencing schemes and classifications of crimes across jurisdictions. The constitutional test, then, requires a “gross disproportionality.” See Van Dyke, 31 Kan. App. 2d 668, Syl. ¶ 1.
In evaluating tire constitutionality of a sentence of imprisonment, those members of the United States Supreme Court recognizing the concept of proportionality generally consider three broad categories of information. First, die Court considers the circumstances of tire convict, the offense, and the punishment. This includes background information on the particular individual being punished. See Graham, 130 S. Ct. at 2037-38 (Roberts, C.J., concurring); Solem, 463 U.S. at 296-97 & n.22. The Court looks at how the particular state punishes offenses it considers more serious. And, finally, the Court looks at how other states punish comparable offenses. Those are considered objective criteria measuring the challenged punishment against intrajurisdictional and interjurisdictional standards.
The United States Supreme Court, however, has not presented a clear statement as to how those criteria should be applied. Justice Kennedy’s narrow proportionality test would treat the initial comparison of crime to punishment in the given case as a threshold standard. Only if that review strongly suggested a gross disproportionality — something he characterizes as a rarity — would the judicial inquiry advance to the comparative criteria of punishments for other crimes within the jurisdiction and the punishments for similar crimes outside the jurisdiction. But, as we have noted, that analytical model has never commanded a majority of sitting justices of the Court. The Solem analysis, the last approach supported by a Court majority, would consider essentially the same three factors together in a totality-of-the-circumstances view.
The United States Supreme Court treats a sentence of life without the possibility of parole differently from imprisonment for a term of years in assessing the constitutionality of the punishment under the Eighth Amendment. See Solem, 463 U.S. at 297 (noting the severity of the punishment). Life without parole does not inherently violate the Constitution, at least when applied to an adult offender, as Harmelin demonstrates. But life without parole re-*920fleets a policy determination that tire offender should be treated as incapable of rehabilitation and, thus, beyond redemption. The punishment, therefore, abandons one of the established purposes for criminal sanctions in a way that incarceration for a fixed term does not. See Graham, 130 S. Ct. at 2029-30; Ewing, 538 U.S. at 52 (Breyer, J., dissenting). Sentences for terms of years, even lengthy ones, presuppose at least some offenders will be returned to society. Although a life sentence without parole necessarily amounts to a variable term — actuarially, it will be much longer for a 30-year-old offender than a 60-year-old offender — it reflects a value judgment that is far harsher than virtually any fixed term. It is a legislative determination that the recipient has relinquished any legitimate claim to liberty and is no longer fit for or deserving of freedom and never will be. In that respect, the punishment bears some similarity to the death penalty in a way that a fixed sentence does not. Those circumstances should be reflected in the constitutional assessment of life without parole as a punishment, particularly under recidivist statutes.
There is, of course, an incontrovertible difference separating the death penalty from any punishment by incarceration: The death penalty, having been administered, is irrevocable. For an inmate serving life without parole there is some possibility, slim though it may be, for executive clemency or a change in the law. Graham, 130 S. Ct. at 2021 (“To determine whether a punishment is cruel and unusual, courts must look beyond historical conceptions to ‘ “the evolving standards of decency that mark the progress of a maturing society.” ’ ”) (quoting Estelle v. Gamble, 429 U.S. 97, 102, 97 S. Ct. 285,50 L. Ed. 2d 251 [1976]); 130 S. Ct. at 2036 (Stevens, J., concurring) (“Punishments that did not seem cruel and unusual at one time may, in the light of reason and experience, be found cruel and unusual at a later time.”). But those possibilities are too remote and uncertain to “mitigate” the punishment for Eighth Amendment purposes. Graham, 130 S. Ct. at 2027 (executive clemency); Solem, 463 U.S. at 303 (clemency).
In considering the current state of Eighth Amendment jurisprudence, the Kansas Supreme Court has acknowledged the viability of proportionality analysis for terms of incarceration. State v. *921Gomez, 290 Kan. 858, 863, 235 P.3d 1203 (2010). And the Gomez decision suggests, without formally deciding, that Justice Kennedy’s narrow test may be the appropriate application of proportionality review. 290 Kan. at 863-64. The narrow proportionality test draws on the same factors outlined in State v. Freeman, 223 Kan. 362, 367, 574 P.2d 950 (1978), but it applies them differently. If application of that test were to render a punishment cruel and unusual, that result would be, at the very least, compatible with how the Kansas Supreme Court views the federal constitutional protection.
C. The Principles Applied
Before applying the United States Supreme Court’s precepts for measuring the constitutionality of a term of imprisonment under the Eighth Amendment, we resolve two threshold matters. First, we apply Justice Kennedy’s narrow-proportionality test. We do so not because we have been persuaded that it is necessarily the better test or truer to the principles embodied in the Eighth Amendment but because it is the more restrictive test accepted among those members of the Court recognizing proportionality as a proper method of analysis. That is, if the lifetime postrelease supervision imposed on Proctor is unconstitutional under Justice Kennedy’s approach, it necessarily must also fail under the broader standard applied in Solem and since promoted by a plurality of the Court. See Ewing, 538 U.S. at 35 (Stevens, J., dissenting) (“[T]he Eighth Amendment’s prohibition of ‘cruel and unusual punishments’ expresses a broad and basic proportionality principle . . . .”). We use Justice Kennedy’s approach for forensic purposes only, much as Justice Breyer did in his dissent in Ewing, 538 U.S. at 36, to test the constitutionality of the punishment at issue in that case.
Second, lifetime postrelease supervision is really neither an invariable determinate sentence set through legislative elimination of judicial discretion of the sort at issue in Harmelin nor a recidivist statute imposing enhanced punishment for the most recent crime committed by a repeat offender of the sort at issue in Rummel, Solem, and Ewing. The distinction would appear to make some difference in that the Harmelin Court upheld a statutorily man*922dated sentence of life without parole for possession of more than 650 grams of cocaine, while the Solem Court struck down the same sentence imposed under a recidivist statute on a repeat felon for knowingly writing a check on a closed account.
On balance, lifetime postrelease supervision operates more in the nature of a recidivist punishment than a procrustean initial sentence. As this case illustrates, a district court retains considerable discretion, even without imposing a departure sentence, in punishing a first-time offender convicted of aggravated indecent solicitation of a child. The legislative design in designating the crime as a severity level 5 offense builds in judicial discretion to consider probation as a nondeparture punishment for some defendants. That is markedly different from tire inflexible sentence challenged in Harmelin. The mandatory punishment in Harmelin — to be imposed without judicial discretion for a demonstrably serious crime — was the product of “die collective wisdom of the Michigan Legislature” in attempting to combat illegal drug use and trafficking. Harmelin, 501 U.S. at 1006-07 (Kennedy, J., concurring). The legislature could and did reasonably conclude that the social ills posed by the possession of large quantities of cocaine to be “momentous enough to warrant the deterrence and retribution of a life sentence without parole.” 501 U.S. at 1003 (Kennedy, J., concurring). That conclusion, thus, comported with the Eighth Amendment despite its undeniable harshness, according to the Hanmelin majority. See 501 U.S. at 994 (“Severe, mandatory penalties . . . are not unusual in the constitutional sense.”).
The potential for life in prison without parole that Proctor challenges does not operate that way. And we are not presented with the same sort of legislative determination. Had the Kansas Legislature required a mandatory sentence of life in prison without parole for aggravated indecent solicitation, we would be looking at a Harmelin-like issue. But the Kansas Legislature has not punished the crime in that manner; it has established a substantially lesser punishment that may, in some instances, permit probation.
Rather, the legislature has provided that if a person sent to prison for aggravated indecent solicitation of a child and then released commits any felony, he or she must be returned to prison *923for life without ever getting out again. That punishment is triggered not by the original crime but by a later felony that may be a comparatively minor property offense. The sentencing scheme effectively functions no differently from a statute requiring a sentence of life in prison without parole for anyone convicted of a felony if he or she has a past conviction for a “sexually violent crime.” Revocation of lifetime postrelease supervision, then, functionally behaves like a recidivist statute and may be best analyzed that way for Eighth Amendment purposes.
1. Threshold Comparison of Offense and Sentence
The crime of conviction here is a serious one, more so than those the United States Supreme Court reviewed in the three-strikes cases. Whether it is on some moral plane with the drug offense in Harmelin is hard to say. It is arguably worse. Proctor’s conviction was for an offense classified as a violent sexual felony but which, by definition, required solicitation or enticement for an illicit purpose — conduct preparatory to physical contact. But the factual basis for Proctor’s plea included the admission of sexual contact with the victim, a I2-year-old boy. The victim suffered no physical injury but understandably has experienced psychological trauma of an unspecified dimension, though apparently not so extreme as to be debilitating or to require inpatient care.
Proctor was 19 years old when he abused the victim. He has no other criminal record. Before sentencing, Proctor had not received any counseling for sex offenders. That counseling was available. He was willing to participate. And there was substantial information indicating the counseling could be effective.
The Kansas Legislature permits a first-time offender convicted of a severity level 5 person felony, as Proctor was, to receive probation if counseling is available and the public interest would be served with a nonprison punishment. The district court made that finding in this case, imposing strict limitations on Proctor and requiring he complete sex offender counseling. If Proctor successfully completes the 36-month probation period, he will not be subject to further direct supervision through the court system. He has registered as a sex offender and must continue to do so until at *924least 2020. If he fails on probation and is sent to prison, he will be placed on lifetime postrelease supervision. A conviction for any felony will automatically return him to prison for the rest of his life without any possibility for early release. That punishment renders the sentencing scheme constitutionally suspect as cruel and unusual, especially insofar as the triggering offense may be a low-level felony.
But as we have acknowledged, Proctor or someone else could commit an especially violent or otherwise heinous offense while on supervised release. If the felony triggering an offender’s return to prison for life were a homicide or kidnapping that would weigh strongly against a finding of cruel and unusual punishment as applied to that offender in those circumstances. In other words, we envision realistic scenarios in which the constitutionality of the punishment measured by a grossly disproportionate standard could pose a much closer question or might plainly favor die State.
In Solem, the United States Supreme Court found a recidivist statute unconstitutional when the defendant received a sentence of life in prison without parole for intentionally writing a check on a closed account. If Proctor were on lifetime postrelease supervision, the same would happen to him for writing such a check. The South Dakota statute at issue in Solera did not prescribe life without parole until a third felony conviction, and the defendant actually had twice that many convictions before he received that sentence. The Court characterized all of those offenses as nonviolent, although some were residential burglaries. But the “triggering” offense directly bringing about the challenged punishment would be equivalent — a felony bad check in Solera and the felony theft we have postulated for Proctor. That suggests a gross disproportionality. The punishments upheld in Rummel and Ewing were constitutionally distinguishable because they entailed tangible prospects for release from prison.
In addition, Proctor’s overall record, as we have constructed it for purposes of resolving the case, would not be indicative of a repeat sex offender or an incorrigible recidivist. As we have indicated, the United States Supreme Court has given considerable weight to incorrigibility as a reason for upholding the constitution*925ality of statutes punishing repeat offenders harshly. See Ewing, 538 U.S. at 24; Rummel, 445 U.S. at 276 (The interest undergirding enhanced and severe punishment for recidivists lies “in dealing in a harsher manner with those who by repeated criminal acts have shown that they are simply incapable of conforming to the norms of society as established by its criminal laws.”). But that interest does not appear to infuse the lifetime postrelease supervision that would be imposed on Proctor. As we have pointed out, he could write a felony bad check 20 years from now with the automatic result of spending the rest of his life in prison. That does not evince incorrigibility of the sort constitutionally justifying the lesser, though undeniable harsh, punishments upheld in Rummel and Ewing, let alone tire identical punishment struck down in Solem.
Under the circumstances here, two offenses that individually permit probation combine to result in lifetime incarceration without possible release. The United States Supreme Court has never upheld that sort of recidivist sentencing scheme and even in Rummel, 445 U.S. at 282 n.27, cast doubt on such a result. Moreover, the Kansas sentencing protocols, reflected in the grid, incorporate enhanced punishments based on criminal histoiy, thus accounting for recidivist behavior. An individual with a person-felony conviction who is then sentenced for felony theft, a severity level 9 offense, should presumptively receive 12 months’ probation with an underlying prison sentence of 12 months. The deviation between the customary punishment the legislature has adopted and the result here seems, at least on its face, startlingly anomalous. A defendant first committing a sexually based person felony and later a low-level nonperson felony, such as theft or forgery, may not be deserving of praise, but the conduct reflects a de-escalation of criminality and antisocial behavior. Why that should result in life behind bars begs an obvious explanation, particularly when the reverse behavior does not have anywhere near the same penal consequences.
Someone with a conviction for felony theft who then commits aggravated indecent solicitation would face presumptive incarceration for a standard term of 41 months and would be on lifetime postrelease supervision after serving the time. But that person *926would have to commit yet another felony — his or her third — to be returned to prison for life. Despite escalating from a property crime to a person felony, that defendant would be substantially more favorably treated than someone committing the same offenses in the reverse order. That also reflects a constitutionally suspect disconnect between offense and punishment.
The result cannot be explained because Proctor s initial offense has been classified as and considered to be sexually violent and, therefore, deserving of stern punishment. If that were so, the Kansas Legislature presumably would have enacted stiffer penalties for aggravated indecent solicitation to be imposed at the outset rather than as the result of a condition subsequent triggered by a new felony conviction. Because any felony triggers revocation of the postrelease supervision and a return to prison for life, the scheme does not punish a continuing disposition to commit sexually violent offenses but general criminality. As such, there is no apparent reason for deviating from the usual sentencing protocols. And, as we have said, the overall criminal history of a defendant convicted of aggravated indecent solicitation followed sometime later by a nonperson property offense fails to display the chronic criminality constitutionally necessary for especially harsh punishment of recidivists.
The Kansas Legislature has adopted special sentencing provisions aimed at repeat sex offenders. K.S.A. 2009 Supp. 21-4704(j); K.S.A. 2009 Supp. 21-4642. Under K.S.A. 2009 Supp. 21-4704(j), a “persistent sex offender,” a person with a conviction for a sexually violent crime who is later convicted of another sexually violent crime, shall receive a sentence double the maximum otherwise provided for the new offense. That recidivist statute already applies to Proctor, since it requires only conviction rather than actual incarceration. If Proctor were to complete his probation and again commit aggravated indecent solicitation or another sex offense, he would have to receive a doubled sentence.
Lifetime postrelease supervision applies to Proctor only if he serves prison time on die first offense. The difference in punishments arising from a second conviction — life in prison for violation of the postrelease supervision and doubled prison time for a new *927sex crime — appears to turn on the fact of incarceration for the first conviction. That is, the sentencing scheme would lock up for life a person committing aggravated indecent solicitation if that person serves time and then commits any felony, but it would merely double the sentence of a person committing the same offense, completing probation, and then committing another sexually violent crime. The distinction lies in one offender going to prison and then reoffending in some way and the other avoiding prison but reoffending in a serious way consistent with his or her past criminal behavior. If the fact of past incarceration warrants such a huge increase in punishment upon a convict’s commission of a new felony, we might presume die legislature would have taken a similar approach with felons other than those committing sexual offenses. But it has not.
The recidivist provisions in K.S.A. 2009 Supp. 21-4704(j) also overlap with the operation of lifetime postrelease supervision in a curious way. If a sex offender serves time in prison for a first offense and then reoffends by committing a new sexually violent offense, both provisions apply. The return to prison for life attaches to the first conviction but is triggered by the second. The second conviction is punished with a doubled term of years, a punishment largely superfluous given the revocation of supervised release. Why the legislature would not simply mandate a sentence of life in prison for a person convicted of a second sexually violent offense when it has mandated that a person serving time for a single such offense who then commits any felony should go to prison for life seems peculiar.
The Kansas Legislature has similarly determined to impose a punishment of life in prison without parole for a third conviction for a sexually violent offense under K.S.A. 2009 Supp. 21-4642. But a first offender, having served time, faces the same punishment for then stealing a $1,000 necklace or writing a bad check for over $1,000. Escalating punishment for repeat sex offenders culminating in life in prison reflects an orderly and understandable penological response to a serious social and criminal problem. But imposing the same sentence — the second harshest possible — for a single sex offense followed by any felony conviction looks to be *928unfocused, especially operating simultaneously with the recidivist statutes targeting repeat sex offenders.
All of that suggests something short of the studied legislative determination the United States Supreme Court recognized in Harmelin as constitutionally permitting an initial life sentence without parole for possession of large quantities of cocaine. Harmelin, 501 U.S. at 1002-03 (Kennedy, J., concurring). The Kansas statutes appear far less coherent and measured, however harshly, in addressing offenders situated as Proctor. The harshness of the punishment without an apparent countervailing legislative design suggests gross disproportionality indicative of unconstitutionality. See 501 U.S. at 1007-08 (Michigan’s mandatory penalty reflected “a recent enactment calibrated with care, clarity, and much deliberation to address a most serious contemporary social ill.”).
Finally, we look briefly at how imposition of lifetime postrelease supervision on Proctor and its revocation for any felony conviction would further retribution, deterrence, incapacitation, and rehabilitation — the recognized penological goals for criminal sanctions. The inquiry bears on Justice Kennedy’s proportionality analysis of prison terms under the Eighth Amendment. See Graham, 130 S. Ct. at 2029 (“A sentence lacking any legitimate penological justification is by its nature disproportionate to the offense.”); Ewing, 538 U.S. at 25, 29-30 (California recidivist statute imposing sentence of 25 years to life may be justified by government “interest in incapacitating and deterring recidivist felons”); Harmelin, 501 U.S. at 999 (Kennedy, J., concurring); see Van Dyke, 31 Kan. App. 2d at 673-74.
If Proctor were on lifetime postrelease supervision for the aggravated indecent solicitation offense, sending him to prison as the result of a felony theft conviction decades later seems to lack targeted retribution. Assuming the indecent solicitation offense warranted substantial retribution that purpose becomes wholly diluted and arguably lost if it is deferred until the offender commits another felony — something that might never occur. Retribution typically would be served through the immediate, certain penalty exacted for the offense. In that context, retribution expresses society’s collective opprobrium for the criminal conduct and imposes a sane*929tion specifically to punish. As we have discussed, retribution for the triggering felony, if it were a low-level offense like theft, could not justify a sanction of life in prison even coupled with the underlying aggravated indecent solicitation conviction. Living under the restrictions of lifetime postrelease supervision may itself serve a retributive function. But that purpose does not require that violations result in lifetime incarceration as opposed to some shorter period of imprisonment or more restrictive conditions of supervision.
Deterrence, apart from incapacitation, serves to prevent the commission of an offense in the first place. In theory, knowing of a harsh punishment for a particular crime, people will not engage in that wrongful conduct. Nothing suggests lifetime postrelease supervision and the consequence for violating its terms are so well known that they effectively deter conduct of the sort Proctor engaged in. To the extent a defendant, such as Proctor, learns of lifetime postrelease supervision as part of the disposition of the case, it comes too late to deter the underlying offense. It might well serve to deter an offender from committing a felony that would trigger his or her imprisonment for life. But if it were to fail in that purpose and an offender were to commit a theft or pass a bad check, deterrence alone would not justify an otherwise unconstitutionally cruel and unusual punishment of life in prison for someone in Proctor’s circumstance. In other words, a life sentence without parole to deter felony theft would be grossly disproportionate, even for someone who had already committed a felony.
Revocation of lifetime postrelease supervision incapacitates but in an untargeted way. A person imprisoned is incapacitated from committing additional crimes against the general populace for the duration of that incarceration. To the extent a person guilty of aggravated indecent solicitation ought to be incapacitated that incapacitation presumably would be made part of a prison term to be served immediately. Just why an extraordinaiy period of incapacitation would be required if the offender later committed a dissimilar nonperson felony, such as theft or forgery, is hardly clear. Those property offenses would not evince some continuing or enhanced danger to the public beyond the circumstances of the *930crimes themselves. And justifying incapacitation for the rest of the offender s life, likely measured in decades, is similarly incompatible with how those crimes are usually treated. Incapacitation for life without parole resulting from tire commission of a single property crime, even after a conviction for aggravated indecent solicitation, would be difficult to reconcile with Solem and the other recidivist cases. Conversely, tire serial commission of sex offenses, given the nature of tire crimes, would constitutionally justify extended incapacitation.
The final consideration is rehabilitation. As the United States Supreme Court has pointed out, the government abandons rehabilitation as a penological goal when it imposes punishment of life in prison without parole. Graham, 130 S. Ct. at 2030 (“The penalty forswears altogether the rehabilitative ideal.”); Ewing, 538 U.S. at 52 (“ ‘rehabilitation’ is obviously beside the point” in considering the penological interests supporting a recidivist statute imposing sentence of 25 years to life) (Breyer, J., concurring). This would be precisely such a situation.
Considering the circumstances, tire punishment exacted if Proctor were placed on lifetime postrelease supervision and then revoked for a nonperson felony conviction would be grossly disproportionate to the triggering offense and to the whole of his criminal history. This would be one of those rare instances in which the operation of the statutoiy scheme would call into question the constitutionality of the resulting punishment as cruel and unusual. In Justice Kennedy’s method of analysis, that determination requires a review of how Kansas punishes more serious crimes and how other jurisdictions punish similar crimes.
Before turning to those considerations, we take note of State v. Baber, 44 Kan. App. 2d 748, 753, 240 P.3d 980 (2010). This court’s decision in Baber, upholding the lifetime supervised release provision against a cruel and unusual punishment challenge, does not control here. In that case, Baber argued only that the duration of the supervision violated his constitutional rights. The court rejected that argument, at least in part, based on his failure to produce sufficient evidence. 44 Kan. App. 2d at 751-52. While the duration of the term of supervision is a component of Proctor’s argument, *931it is only part of that argument. Proctor also points to the unconstitutionahty of imposing mandatory imprisonment for life based on any new felony conviction. Baber asserted no such challenge, and the court didn’t presume to deal with that argument. The Baber decision, therefore, neither addresses nor controls on that score.
2. Proctors Sentence Compared to Sentences for Other Kansas Offenses
The current sentencing regimen in Kansas permits a reliable comparison of the relative severity of criminal offenses and their usual punishments. To impose a degree of uniformity in sentencing, the State’s system rests on a grid constructed to measure the severity of the crime of conviction on one axis and a defendant’s criminal histoiy on the other, thus yielding a statutorily presumptive punishment, expressed in months of incarceration, at their intersection. Offense severity and past convictions are directly proportional to the presumptive term of incarceration. Thus, a defendant with an extensive criminal history generally will receive a substantially longer sentence than a first-time offender for the same crime of conviction. And between two defendants with the same criminal history, the one committing the more severe offense will spend more time in prison. The legislature has assigned crimes severity levels corresponding to their relative wickedness. Two crimes with the same severity level are in the legislature’s view rough moral (or immoral) equivalents.
Proctor’s crime of conviction is a severity level 5 person felony and would be comparably contemptible to offenses similarly classified and less so than offenses assigned higher levels. Some of the worst crimes are “off grid” and carry the harshest sentences. A separate grid governs drug-related offenses. But those refinements do not undercut a comparison based on severity level for Eighth Amendment purposes. We look at one off-grid crime and others on the sentencing grid for nondrug offenses. We presume no dis-positional or durational departures and discard possible reductions in the actual amount of time a defendant might serve based on credit for good behavior in prison. A “good time” reduction would *932be added to and extend the period of postrelease supervision. K.S.A. 21-4722(b).
There are a number of person felonies with higher severity levels than aggravated indecent solicitation, including second-degree intentional murder, severity level 1, K.S.A. 21-3402; voluntary manslaughter, severity level 3, K.S.A. 21-3403; intentional aggravated batteiy causing great bodily harm, severity level 4, K.S.A. 21-3414(a)(1)(a); kidnapping, severity level 3, K.S.A. 21-3420; aggravated kidnapping, severity level 1, K.S.A. 21-3421; and aggravated robbery, severity level 3, K.S.A. 21-3427. There are a number of others that are severity level 5 felonies, including aggravated burglary (person felony), K.S.A. 21-3716, and smuggling illegal drugs into a jail or prison (nonperson felony), K.S.A. 21-3926. The mandated period of postrelease supervision for felonies with a severity level of 4 or higher is 36 months. K.S.A. 2009 Supp. 22-3717(d)(1)(A). For severity level 5 felonies generally, the period is 24 months. K.S.A. 2009 Supp. 22-3717(d)(l)(B). If the sentencing court were to rely on a specific finding the crime had been sexually motivated, the period of postrelease supervision could be extended to 60 months.
As we have noted, Proctor would be on lifetime postrelease supervision, and any felony conviction would require he be returned to prison for life without parole. Thus, if he committed a felony theft shortly after serving the sentences imposed for the aggravated indecent solicitation offense and the lesser charges, he could be back in prison at age 28 or 29 with no prospect for release.
We compare how sentencing would work with a couple of the other offenses. Premeditated first-degree murder is an off-grid felony cariying a life sentence. K.S.A. 21-3401. Absent a finding of aggravating circumstances, the murderer would become parole eligible after 25 years. K.S.A. 2009 Supp. 22-3717(b)(l). A defendant sentenced at age 20 could be released at age 45. By contrast, Proctor, having served his time on the aggravated solicitation conviction and then committing a felony theft, could never be released from prison. At age 45, he would have served roughly 20 years in prison, some less than the murderer, but he would face another 33 years in prison before death might metaphorically set him free. While *933moral equivalence is a tricky business, someone committing a premeditated murder generally seems more blameworthy than someone engaging in aggravated indecent solicitation of a minor and then writing a bad check for over $1,000. In considering Eighth Amendment issues, the United States Supreme Court has consistently recognized murder, as a type of crime, to evince greater depravity than other crimes precisely because it takes the life of the victim. Graham, 130 S. Ct. at 2027 (“Although an offense like robbeiy or rape is ‘a serious crime deserving of serious punishment,’ those crimes differ from homicide crimes in a moral sense.”) (quoting Enmund v. Florida, 458 U.S. 782, 797, 102 S. Ct. 3368, 73 L. Ed. 2d 1140 [1982]).
We also look at examples of crimes included on the sentencing grid. Assume Proctor committed second-degree murder and was sentenced at age 20. Without a downward departure, he could not receive probation on the murder charge. We presume he would not. The standard sentence for a person with no criminal record would be 155 months in prison (just about 13 years) with 36 months’ postrelease supervision. Proctor would be released from prison at age 33 and would be free of any supervision at age 36. Assume he waits a year, gets in a bar fight, and kills again. He pleads guilty to second-degree murder. With one past person felony on his record, Proctor would face a standard sentence of 253 months in prison (just over 21 years) and another 36 months’ post-release supervision. Proctor would be released from prison at age 58 and off supervised release at age 61. If he then committed a felony theft, would he go to prison for the rest of his life? No. Based on the grid, with two person felonies, he would serve a standard sentence of 14 months and be on postrelease supervision for 12 months. He would again be without restrictions in time to celebrate his 64th birthday. Actuarially, he could expect to enjoy another 17 birthdays of freedom before his ultimate judgment day.
Suppose Proctor beat someone with a tire iron, causing the victim great bodily harm and a prolonged recuperation. Proctor would not be eligible for probation absent a downward durational departure. The standard sentence would be 41 months with postrelease supervision of 36 months. Proctor would be free of all restrictions *934in less than 6Vz years — well before he turned 30 years old. If he were then to commit a felony theft of $1,000, he would face a standard sentence of 12 months with a presumption for 12 months’ probation. Another felony theft would call for exactly the same punishment. In theory, based on the sentencing grid, each successive felony theft of that magnitude would similarly require a presumptive probation punishment. At some point, however, a sentencing court presumably would impose an upward dispositional departure to send an aggravated batterer turned chronic thief to prison.
We could repeat this sort of exercise with the other crimes we have identified with similar results. As tire examples illustrate, a defendant committing an offense the legislature has ranked by severity level as worse than Proctor’s crime of conviction does not entail a punishment requiring a return to prison for life without parole based on a later felony conviction. Even murderers are not treated so harshly. A person committing successive second-degree murders could be released and then commit a low-level felony without spending the rest of his or her life in prison.
On the whole, the comparison between Proctor’s circumstances if he were on lifetime supervised release for a severity level 5 offense and those of persons committing crimes with higher offense levels supports the gross disparity between offense and punishment indicative of an Eighth Amendment violation. Whether we treat the result as a means of confirming that disparity, in keeping with Justice Kennedy’s methodology, or as an integrated part of an overall analysis of the sentence’s constitutionality, as the Solern majority and the Kansas Supreme Court have done, we arrive at the same point. This factor weighs against the State.
The State argues that the Kansas Supreme Court has upheld lengthy sentences and 60-month periods of postrelease supervision against constitutional challenges. The authority, however, is unpersuasive.
The State cites State v. McCloud, 257 Kan. 1, 891 P.2d 324, cert. denied 516 U.S. 837 (1995), and State v. Tyler, 251 Kan. 616, 644-46, 840 P.2d 413 (1992), as demonstrating tire court has rejected constitutional challenges to prison sentences with minimum terms *935approaching or in excess of 100 years. Those sentences effectively amounted to life in prison for the respective defendants. But the lengthy duration resulted from sentences on multiple criminal acts imposed consecutively. For example, in McCloud, tire defendant committed 12 separate aggravated robberies during a 4-month crime spree. He received a statutorily proper sentence of 8 years to life on each and was ordered to serve them consecutively, resulting in a minimum sentence of 96 years. The aggregate incarceration of 111 to 330 years imposed in Tyler resulted from consecutive sentences on multiple convictions, including second-degree murder of a police officer executing a search warrant, arising from a single episode and enhanced under a habitual criminal statute then in effect allowing doubling or trebling of prison time for repeat felons. Tyler, 251 Kan. at 644-47. Those cases do not speak to penalties permitted for a single criminal offense more serious than Proctor s — the real point of the comparison. If Kansas imposed a 96-year sentence for one count of aggravated robbery or 100 years for one second-degree murder, those cases, or more particularly those statutory penalties, would have relevance to the inquiry regarding comparative punishments.
The State also cites State v. Cullen, 275 Kan. 56, 60-61, 60 P.3d 933 (2003), and State v. Anthony, 273 Kan. 726, 729, 45 P.3d 852 (2002), for the proposition that an enhanced 60-month postrelease supervision is constitutional. In both cases, the Kansas Supreme Court held that because the factual findings necessaiy to support the enhancement were either found by a jury or admitted by the defendant in entering a plea, the punishment did not violate the Sixth Amendment right to jury trial. The court had no occasion to consider the sentences under the Eighth Amendment. The decisions are, therefore, legally irrelevant.
3. Proctors Sentence Compared to Sentences for Similar Offenses in Other States
Proctor submits that 18 states impose mandatory lifetime post-release supervision for at least some convicted sex offenders. But he says only four, including Kansas, have no mechanism for terminating the supervision. Proctor identifies the other three as Col*936orado, Nebraska, and Oklahoma. Two other states permit the supervision to be lifted under certain conditions. According to Proctor, the remaining 12 states impose lifetime postrelease supervision for more serious offenses than he committed or reserve that punishment for repeat offenders or offenders displaying characteristics making them likely recidivists. In its brief, the State does not dispute the results of Proctor’s survey. The State agrees Kansas is one of only four states without any possibility for early termination of lifetime postrelease supervision for convicted sex offenders.
The State’s rejoinder is to argue that tire conditions of post-release supervision are a matter of legislative prerogative. That, of course, is true until the exercise of that prerogative contravenes a citizen’s constitutional rights. And, as all would agree, those instances arise only rarely. But invoking legislative prerogative neither informs the issue of comparative sentencing practices in other jurisdictions nor decides it. If it did, there would be no point to the analysis, since any result could be trumped with the government’s claim of prerogative.
To the extent the parties have joined the issue, we cannot say the result obviously favors the State and certainly not to the degree that it would overcome the substantial considerations militating for a finding of unconstitutionality. In a study of this sort, one state or group of states necessarily will impose the harshest punishment and some other the most lenient. That cannot be decisive. If all the states were to take a similar approach with only minor variances, an argument for unconstitutionality would be more difficult to press. A single state out of step with that commonality might be susceptible to constitutional challenge for a substantially more severe punishment. The parties have not presented that picture to us. A substantial number of states do not seem to have coalesced around a single approach and certainly not the one Kansas has chosen. Had the numbers been reversed, with 46 states adopting irrevocable lifetime postrelease supervision, we would be looking at a materially different assessment on this factor.
The parties have not zeroed in on the most significant aspect of the Kansas scheme: What triggers a violation of the supervision *937and what consequence follows? They have not indicated whether other states conform to Kansas’ requirement that a felony conviction for someone on lifetime supervised release automatically results in imprisonment for life without any possibility of reprieve. Even were we to assume the three other states identified as having irrevocable lifetime postrelease supervision follow that course, we would not be disposed to treat that outcome any more or less favorably for constitutional purposes.
4. Conclusion
As the discussion to this point foreshadows, the imposition of lifetime postrelease supervision on Proctor violates the Eighth Amendment as cruel and unusual punishment. That is true under both the “narrow proportionality” test Justice Kennedy has formulated and the more holistic proportionality review Justice Powell outlined for the majority in Solem and the Kansas Supreme Court applied in Freeman.
We do not belabor or repeat that discussion. The comparison of Proctor’s hypothecated criminal history with a punishment of life in prison without parole depicts gross disproportionality. The United States Supreme Court has never approved a recidivist statute imposing that sentence for a second conviction and has often discussed the necessity for demonstrable incorrigibility — an ongoing inability to conform to the criminal law — to warrant lengthy imprisonment for repeat offenders. The regimen for lifetime post-release supervision does not satisfy those standards.
The Kansas criminal and sentencing statutes punish other offenses and combinations of offenses more harshly. Particularly telling, we think, if Proctor first committed a felony property offense and then aggravated indecent solicitation, he would not face nearly so unyielding a punishment as he does for committing those crimes in reverse order. Persons released from prison after committing homicide offenses would be treated far more leniently upon committing a felony property crime. And, to the extent the parties have addressed comparable sentences in other states, the evidence places Kansas in a distinct minority. The intrastate and interstate comparisons point toward a constitutionally defective scheme *938whether considered as criteria intended to confirm or discount an initial finding of gross disproportionality or as ingredients in an overall determination of unconstitutionality.
The presumption of constitutionality fails to aid the State on lifetime postrelease supervision. The presumption typically applies as a canon of construction to lend a reading to statutory language that avoids constitutional issues or defects. Clark v. Martinez, 543 U.S. 371, 380-82, 125 S. Ct. 716, 160 L. Ed. 2d 734 (2005) (As between “plausible statutory constructions,” a court should reject one that “would raise a multitude of constitutional problems.”); St. Louis S. W. Ry. Co. v. Arkansas, 235 U.S. 350, 369-70, 35 S. Ct. 99, 59 L. Ed. 265 (1914); Planned Parenthood of Central New Jersey v. Farmer, 220 F.3d 127, 135-36 (3d Cir. 2000). A court, for example, may apply a narrow construction to an otherwise potentially vague statute to supply sufficient specificity to avert a constitutional deficiency. Boos v. Barry, 485 U.S. 312, 330-31, 108 S. Ct. 1157, 99 L. Ed. 2d 333 (1998); United States v. Sutcliffe, 505 F.3d 944, 953 (9th Cir. 2007). Here, the language of tire critical statutes is plain rather than ambiguous or vague. A court may not construe a statute in a way that is irreconcilable with its words to trowel over a constitutional defect. Salinas v. United States, 522 U.S. 52, 60, 118 S. Ct. 469, 139 L. Ed. 2d 352 (1997) (court may not engage “disingenuous evasion” in guise of statutory construction to deflect constitutional challenge); Boos, 485 U.S. at 330-31 (The judicial construction of the statutory language must be “fairly possible.”).
We, therefore, find lifetime postrelease supervision as applied to Proctor to be cruel and unusual punishment violating die Eighth Amendment.
VI. Section 9 of the Kansas Constitution Bill of Rights
The Kansas Supreme Court has consistently held that § 9 of the Kansas Constitution Bill of Rights permits proportionality challenges to die duration of incarceration for specific offenses. State v. Gomez, 290 Kan. 858, 863, 235 P.3d 1203 (2010); State v. McDaniel & Owens, 228 Kan. 172, 185, 612 P.2d 1231 (1980). And die Kansas Supreme Court has so recognized notwithstanding its *939expressed doubts in the past about the viability of those challenges under the Eighth Amendment. McDaniel, 228 Kan. at 184-85. In McDaniel, the Kansas Supreme Court questioned whether Eighth Amendment proportionality review of prison terms survived Rummel. 228 Kan. at 184. Regardless of the accuracy of that assumption, the Kansas Supreme Court found such protections for state citizens in § 9, thereby unequivocally recognizing the scope of the Kansas constitutional provision as a source of rights independent of the Eighth Amendment. 228 Kan. at 185. In light of Graham, 130 S. Ct. at 2021, the Kansas Supreme Court reaffirmed the viability of proportionality challenges to noncapital sentences under the Eighth Amendment. Gomez, 290 Kan. at 863. The two provisions, therefore, afford comparable protections, given how the United States Supreme Court and the Kansas Supreme Court respectively construe the federal and state constitutions today. But it is not because the state constitutional prohibition inherently depends upon the measure of the Eighth Amendment for its own metes and bounds.
In State v. Freeman, 223 Kan. 362, 367, 574 P.2d 950 (1978), the Kansas Supreme Court construed § 9 of the Kansas Constitution Bill of Rights to prohibit “punishment... if it is so disproportionate to the crime for which it is inflicted that it shocks the conscience and offends fundamental notions of human dignity.” The court recognized three factors to be considered in assessing the constitutionality of a given punishment. 223 Kan. at 367. The first factor examines “the nature of the offense and the character of the offender . . . with particular regard to the degree of danger present to society.” 223 Kan. at 367. The examination should take into account the “facts of the crime” and “violent or nonviolent nature of the offense” along with “the extent of [the defendant’s] culpability for” any resulting injuries and the way the punishment serves recognized “penological purposes.” 223 Kan. at 367. The second factor entails comparison of the punishment with sentences for more serious offenses in Kansas. 223 Kan. at 367. If more serious crimes were punished less harshly, then the challenged punishment would be constitutionally “suspect.” 223 Kan. at 367. The *940final factor requires comparison of the punishment to sentences in other jurisdictions for the same offense. 223 Kan. at 367.
The Kansas Supreme Court continues to rely on the Freeman factors as the core component in analyzing disproportionality. Gomez, 290 Kan. at 867. The factors are to be considered collectively, and no one alone looms as outcome determinative. 290 Kan. at 867. The Freeman factors and their application essentially parallel the analysis the United States Supreme Court used in Solem. See Solem, 463 U.S. at 292. But application of the complete narrow-proportionality test Justice Kennedy favors — which we have used for forensic purposes — covers the same territory.
The Kansas appellate courts’ discussion and application of the Freeman factors over the years have shed little light on the issue here. Many of the initial cases, including Freeman, upheld a provision in an earlier version of the criminal code requiring mandatory incarceration for defendants using firearms to commit crimes, a readily distinguishable issue. See, e.g., State v. Coleman, 224 Kan. 447, 452-53, 580 P.2d 1329 (1978). The use-of-a-firearm provision did not lengthen the terms of incarceration for the offenses but effectively eliminated probation as a sentencing option. Those cases afford no meaningful guidance in applying tire Freeman factors to the significantly different factual setting here.
In State v. Harder, 8 Kan. App. 2d 98, 103, 650 P.2d 724 (1982), this court rejected an Eighth Amendment challenge to a recidivist provision increasing a second possession-of-marijuana offense from a misdemeanor to a comparatively serious felony. But the court did so in a single paragraph without any developed rationale or any reference to case authority construing the Eighth Amendment or § 9 of the Kansas Constitution Bill of Rights.
Notwithstanding its detailed compilation and discussion of federal and Kansas authority on cruel and unusual punishment, Van Dyke v. State, 31 Kan. App. 2d 668, 672-77, 70 P.3d 1217 (2003), does not directly aid in deciding this case because the underlying facts were markedly different, and the analysis of proportionality is distinctly fact-bound. In that case, a 79-year-old man pled guilty to one count of attempting to rape his 10-year-old granddaughter and challenged his 55-month sentence as cruel and unusual pun*941ishment. Not surprisingly, this court found no gross disproportionality between the offense and the sentence. Van Dyke, 31 Kan. App. 2d at 677. The sentence was neither “extraordinary” nor “extreme” and could not be characterized as “inhumane, shocking, barbarous” or otherwise incompatible with the protections of the Eighth Amendment or § 9. Van Dyke, 31 Kan. App. 2d at 677, 679. As a result, the court didn’t consider any comparative information on penalties for other crimes in Kansas or the treatment of similar offenses in other jurisdictions. But the decision in Van Dyke upholding the constitutionality of a mainstream sentence helps little in resolving this case.
More recently, the Supreme Court has declined to address life sentences for designated sex crimes or lifetime postrelease supervision because the defendants either failed to develop a sufficient record below or failed to raise the issue at all at the trial level. See State v. Roberts, 293 Kan. 1093, 1095-96, 272 P.3d 24 (2012) (declining to consider challenge); State v. Berriozabal, 291 Kan. 568, 594, 243 P.3d 352 (2010) (remanding for further proceedings). Some of those cases include general observations about the Freeman factors of which we have taken note, but none of them applied the factors to resolve a constitutional challenge.
Before deciding Freeman, the Kansas Supreme Court upheld various iterations of habitual criminal statutes against cruel and unusual punishment challenges. See Cipolla v. State, 207 Kan. 822, 824-25, 486 P.2d 1391 (1971). Those statutes commonly permitted a court to double or treble the sentence of an offender if he or she had one or two previous felony convictions. Those cases obviously did not apply Freeman or the United States Supreme Court authority of the modem era. As a result, they are not especially persuasive or illuminating. But the results would be compatible with, for example, the 25-years-to-life sentence upheld for the recidivist offender in Ewing. See Cipolla, 207 Kan. at 824 (citing sentence of 20 to 42 years imposed on a habitual offender for robbeiy in Scoggins v. State, 203 Kan. 489, 454 P.2d 550 [1969], and the same sentence imposed for the same offense on a second-time felon in State v. Shaw, 201 Kan. 248, 440 P.2d 570 [1968], as illustrative of constitutionally permissible punishment). In short, there is little *942Kansas appellate authority directly applicable to the issue before us.
In concluding Proctors sentence violates § 9 of the Kansas Constitution Bill of Rights, we do not replicate the Eighth Amendment analysis here. Because of the common components in the narrow proportionality test and the Freeman factors, the rationale finding Proctor’s punishment cruel and unusual under the Eighth Amendment is sufficient to require the same result under § 9 of the Kansas Constitution Bill of Rights. We do, however, emphasize that we have separately reviewed and applied the Kansas constitutional prohibition of cruel or unusual punishment in arriving at that conclusion.
VII. Conclusion
We vacate Proctor’s sentence to the extent it calls for lifetime postrelease supervision. That portion of the sentence violates the protections against cruel and unusual punishment in the Eighth Amendment to the United States Constitution and § 9 of the Kansas Constitution Bill of Rights. The result requires remand to the district court so that Proctor may be resentenced to a fixed period of postrelease supervision consistent with Kansas law apart from those statutes pertaining to or requiring lifetime postrelease supervision.
Sentence vacated in part and remanded with directions for re-sentencing.