abstract. While on the Supreme Court, Justice Kennedy wrote several opinions that shifted the Court’s Eighth Amendment sentencing jurisprudence for people who have committed crimes before the age of eighteen. Many read these cases to herald a fundamental change in the way juveniles are treated in the criminal justice system. But the better reading is more modest. In important respects, these cases have not lived up to their promise. Instead, they force us to ask: what do we really mean when we say that youth is relevant to the determination of a just prison sentence?

Introduction

On June 24, 2018, Justice Kennedy left the Supreme Court seat he had occupied for three decades. Since Justice Kennedy’s retirement and the nomination of Judge Kavanaugh for his vacant seat, commentators have discussed what a Supreme Court with now-Justice Kavanaugh will mean for contentious issues like abortion.1 By contrast, what Justice Kennedy’s absence will mean for the Court’s Eighth Amendment jurisprudence—particularly a series of cases about the constitutionality of punishments for people convicted of violent crimes committed before their eighteenth birthday—has been ignored. This Essay addresses that gap. Part I reviews the Supreme Court’s evolving Eighth Amendment jurisprudence on capital sentences in cases involving juveniles. These cases coincide with Justice Kennedy’s tenure on the bench, and deeply reflect how he shaped the Court’s jurisprudence in this area. Part II asks if these cases have lived up to their promise. The Essay concludes that the cases have been most important not because of their holdings, but because they have forced scholars, practitioners, states, and courts to think harder about what proportionality in prison sentences really means.

I. the cases

A. The Death Cases: From Thompson to Roper

The Supreme Court’s current Eighth Amendment doctrine on punishments for crimes committed when criminal defendants were under the age of eighteen began to develop contemporaneously with the start of Justice Kennedy’s tenure on the Court. William Wayne Thompson was fifteen years old when a trial court convicted him and three others of murdering Thompson’s former brother-in-law on January 23, 1983.2 On November 9, 1987, three months before the U.S. Senate confirmed Justice Kennedy to the Supreme Court, the Justices heard oral argument in Thompson’s case. With a plurality opinion by Justice Stevens, the Court held that Thompson’s death sentence violated the Eighth Amendment of the Constitution, invalidating for the first time a sentence that would be constitutional if applied to a defendant who was older at the time of the crime.

Thompson’s attorneys argued that his death sentence was unconstitutional because the sentence made “no measurable contribution to acceptable goals of punishment,” “the capacity of the young for change, growth and rehabilitation makes the death penalty particularly harsh and inappropriate,” and the Court must prevent the execution of children below a specific age “to vindicate American traditions of special treatment of juvenile offenders.”3

In his opinion, Justice Stevens largely agreed with Thompson’s attorneys, reviewing, among other sources of authority, Oklahoma and other state laws distinguishing between fifteen-year-olds and those who are sixteen and older.4 Those laws revealed that state legislatures recognized a distinction between juveniles and adults. Surveying the Court’s precedent, Justice Stevens found that “the Court has already endorsed the proposition that less culpability should attach to a crime committed by a juvenile than to a comparable crime committed by an adult,”5 and concluded that retribution and deterrence, the two principal social purposes of capital punishment, do not warrant a death sentence for so young an offender.6 In ruling that the execution of people under the age of sixteen violates the Eighth Amendment’s prohibition against cruel and unusual punishment, Justice Stevens argued that “the experience of mankind, as well as the long history of our law, [teaches] that the normal 15-year-old is not prepared to assume the full responsibilities of an adult.”7 The questions raised in Thompson would frame two decades of Eighth Amendment jurisprudence addressing the constitutionality of sentences for criminal defendants under the age of eighteen. Although Justice Kennedy did not sit for the oral argument, he would later adopt much of the plurality’s reasoning.

A year later, the Court granted certiorari in a case asking the Court to determine if executing people who committed their crimes at the age of sixteen or seventeen violated the Eighth Amendment.8 Kevin Stanford was seventeen years and four months old when he and an accomplice raped, sodomized, and murdered Barbel Poore after they robbed the gas station where she worked.9 In the neighboring state of Missouri, sixteen-year-old Heath Wilkins and an accomplice stabbed Nancy Allen to death in a convenience store robbery.10 The consolidated case asked if the ruling in Thompson should be extended to sixteen- and seventeen-year-olds, marking the second time in as many years that the Court grappled with where the line should be drawn when it comes to youth culpability in the context of the death penalty. The Court refused to extend its holding in Thompson. Justice Kennedy joined with Chief Justice Rehnquist and Justices White, O’Connor, and Scalia in a five-to-four decision holding that the Eighth Amendment did not prohibit the execution of sixteen- and seventeen-year-olds.11

Attorneys for Stanford argued that “the judicial system and state legislatures have recognized the particular vulnerability of children as well as their inherent inability to function in the same responsible manner as is expected of adults.”12 Seeking to establish that “youth is more than a chronological fact,”13 the petitioners brought to the Court’s attention an array of state laws and regulations that drew a distinction between childhood and adulthood “not only to protect children from themselves but also to protect society from the consequences of their immature judgment and impulsive and irresponsible behavior.”14 The implication of Stanford’s argument was that because state legislatures often prohibit juveniles from acquiring a license before turning sixteen (and require a provisional license before turning eighteen), prohibit them from purchasing alcohol or tobacco products, and bar them from voting, they have provided a rationale for why executing people for crimes committed before their eighteenth birthday offends the Constitution. Writing for a four-Justice plurality that included Justice Kennedy, Justice Scalia responded:

[E]ven if the requisite degrees of maturity were comparable, the age statutes in question would still not be relevant. They do not represent a social judgment that all persons under the designated ages are not responsible enough to drive, to drink, or to vote, but at most a judgment that the vast majority are not. These laws set the appropriate ages for the operation of a system that makes its determinations in gross, and that does not conduct individualized maturity tests for each driver, drinker, or voter. The criminal justice system, however, does provide individualized testing.15

The plurality dismissed the notion that the cognitive development of juveniles had any relevance to the constitutionality of the death penalty for juveniles: “[S]ocioscientific, ethicoscientific, or even purely scientific evidence is not an available weapon.”16 Moreover, the plurality’s position was that the Justices lacked the “power under the Eighth Amendment to substitute our belief in the scientific evidence for the society’s apparent skepticism.”17 The Thompson plurality had rendered the inquiry a line-drawing question; the Stanford Court made clear that the line drawn prohibiting the execution of persons for crimes committed at age fifteen should move no further.

Roughly fifteen years later, in 2005, the Supreme Court decided Roper v. Simmons.18 In that case, Justice Kennedy would adopt arguments that he had rejected in Stanford, and he would begin to map out a dramatically different understanding of what protections the Eighth Amendment provided to juvenile defendants. Writing for a majority that included Justices Stevens, Souter, Ginsburg, and Breyer, Justice Kennedy overruled Stanford and held that executing people under eighteen years of age at the time of their crime violates the Eighth Amendment’s prohibition against cruel and unusual punishment.19 Kennedy’s vote changed the law, and for the next thirteen years his voice would shape jurisprudence on this issue.

In Roper, the Court emphasized three distinctions between juveniles and adults that called into question the constitutionality of the death penalty for those under eighteen years old. First, in confirming the significance of science to its analysis, the Court accepted a series of arguments rejected by the Stanford majority:

[A]s any parent knows and as the scientific and sociological studies respondent and his amici cite tend to confirm, “[a] lack of maturity and an underdeveloped sense of responsibility are found in youth more often than in adults and are more understandable among the young. These qualities often result in impetuous and ill-considered actions and decisions.”20

Second, Justice Kennedy noted that juveniles are more “susceptible to negative influences and outside pressures.”21 The Court explained that “the character of a juvenile is not as well formed as that of an adult.”22 In emphasizing that “[y]outh is more than a chronological fact”23Kennedy spoke to the still-forming character of the juvenile and agreed with an argument advanced by Stanford’s attorneys, explaining that childhood is a time of vulnerability and immaturity that differs dramatically from adulthood. And unlike the Stanford Court, he concluded that these facts “render suspect any conclusion that a juvenile falls among the worst offenders.”24

These arguments were rooted in advances in neurological science suggesting that the brain development of young people made them less culpable.25 But Justice Kennedy went further. Not only did the science and our understanding of what it means to be a juvenile make them incapable of being the worst of the worst, but, “[f]rom a moral standpoint[,] it would be misguided to equate the failings of a minor with those of an adult, for a greater possibility exists that a minor’s character deficiencies will be reformed.”26 From a distance, it would have been hard to recognize the Justice who signed onto the majority opinion in Stanford. Justice Kennedy had instead adopted the reasoning first advanced by Justice Stevens in Thompson and later expanded upon in the petitioner’s brief in Stanford.

B. The Life Imprisonment Cases: Graham and Miller

The Court’s opinion in Roper opened an opportunity to challenge sentences that, like the death penalty, had previously been held to be constitutional for juveniles. Encouraged by the success of Roper, attorneys for Terrance Graham appealed his sentence of life without parole (LWOP) to the Supreme Court.27

In July 2003, Graham and three other juveniles had attempted to rob a restaurant in Jacksonville, Florida.28 They were armed with one metal bar, which one of Graham’s accomplices used to strike the restaurant manager on the back of the head.29 After his arrest, Graham’s prosecutor charged him as an adult with armed burglary with assault or battery, a first-degree felony carrying a possible life without parole sentence, and with attempted armed robbery, a second-degree felony carrying a maximum fifteen-year sentence. In exchange for Graham’s guilty plea, the trial court withheld adjudication of guilt for both charges and sentenced Graham to concurrent three-year probation periods.30

Five months after the first offense, Graham and two friends committed an armed home invasion robbery, holding the homeowner and his friend at gunpoint for thirty minutes as they searched the house for money.31 The State of Florida alleged that later that night, Graham and his two accomplices attempted another robbery in which one of Graham’s accomplices was shot.32 After a high-speed chase, Graham crashed into a telephone pole, and police apprehended him. Inside the car, police found three handguns. At the time of his arrest, Graham was thirty-four days shy of his eighteenth birthday.33

Eleven days later, Graham’s probation officer asked the court to revoke his probation.34 The trial court found that Graham violated his probation by committing new crimes. At the resentencing hearing, the requested sentences varied widely. Graham’s attorney asked for a five-year sentence. Graham’s presentencing report, prepared by the Florida Department of Corrections, recommended that Graham receive at most a four-year sentence. The prosecutor, meanwhile, asked for fifteen years on the attempted armed robbery count and thirty years for the armed burglary. The same trial judge who previously gave Graham probation did not follow either party’s recommendations. Instead, the judge sentenced him to the maximum term for each offense. Graham received fifteen years for the attempted armed robbery and life imprisonment without parole for the armed burglary.35

The Graham Court confronted a straightforward question. Does the Eighth Amendment prohibit LWOP sentences in nonhomicide cases for persons convicted of crimes committed before their eighteenth birthday? The Court held that it does, declaring juvenile LWOP sentences for nonhomicide offenses unconstitutional.36 Prior to Graham, the Court distinguished between two general categories of Eighth Amendment challenges that alleged unconstitutionally disproportionate punishments: challenges to “term-of-year” sentences, and challenges to capital sentences, which had yielded “certain categorical restrictions on the death penalty.”37 Until Graham, few challenges to “term-of-year” sentences had been successful;38 the Supreme Court had generally declined to hold any prison sentence unconstitutional because of its length. But in Graham, the Court’s meaningful proportionality scrutiny applied in capital cases found its way into the Court’s noncapital jurisprudence.39

Yet the Graham Court left considerable room for its decision to be construed narrowly. While maintaining that juveniles were inherently less culpable than adults and therefore did not merit the same sentences, the Court also stated that “[a] State is not required to guarantee eventual freedom to a juvenile offender convicted of a nonhomicide crime.”40 According to the Court, states only had to “give defendants like Graham some meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation.”41

Still, the case seemed revolutionary. Justice Kennedy acknowledged that juvenile LWOP implicated “an entire class of offenders who have committed a range of crimes.”42 Reading the opinion, one is compelled to consider the question: if a life sentence offended the Eighth Amendment, did a seventy-year sentence?

Two years after Graham, in Miller v. Alabama,43the Court, in another majority opinion joined by Kennedy, held that the Eighth Amendment prohibited mandatory sentences of life without parole for people who committed homicide before the age of eighteen. Four years later, in Montgomery v. Louisiana,44 yet another majority opinion penned by Justice Kennedy, the Court held that Miller was a retroactive decision, applying both to those incarcerated after the Miller decision and those before.

Justice Kennedy, by moving away from the conservative justices on this issue, changed the landscape of an area of law. Over a thirty-year period, he served as the fifth vote in Stanford, Roper, Miller, and Montgomery. And while Graham was a 6-3 decision in which Kennedy and the liberal Justices were joined by Chief Justice Roberts, Roberts specifically limited his concurring opinion to the case of Terrance Graham, rejecting what he called the Court’s “new constitutional rule of dubious provenance.”45 There too, Justice Kennedy’s vote proved decisive. Given Justice Kennedy’s retirement, one might worry that much of this progress may disappear with a single decision. But, worse still, for all their celebration, it’s not clear how significant an effect these cases have had on the lives of those who ostensibly are covered by them.

II. have these cases lived up to their promise?

Many have lauded the trio of Roper, Graham, and Miller as a watershed development in American juvenile justice.46 This seems a mistake. The cases represented a shift towards more sensible sentencing policies for juveniles in the adult system, but even a cursory review of the state decisions shows that, for many, these cases have not resulted in a significant reduction in the length of juvenile prison sentences.47 From the perspective of friends of mine who are still in prison—young men who have been in some of the most wretched places I know for the past twenty or more years—these decisions remind them of how far removed the reform efforts that dominate the public discourse are from the prospect of their release. The fundamental problem is that from the perspective of many lower court judges, Graham, the opinion most likely to touch on lengthy sentences, can be cabined to LWOP sentences; certainly, the opinion failed to articulate what other prison term for a juvenile might be unjust.

Several months ago, I took a tour of the Louisiana State Prison, called Angola by most everyone who has ever heard of the place. The tour guide, a former corrections officer, reeled off facts about Angola as we traveled a small fraction of the acreage in a large bus. Angola sits on 18,000 acres.48 As of 2016, 6,000 prisoners were incarcerated there; 4,000 were serving life sentences.49 Our guide told us several disturbing facts: the average prison sentence in Angola is 88 years. Surrounded by the Mississippi River on three sides, the state has cross-bred German Shepherds with a breed of wolf-dog to patrol just outside the fences, as budget costs have reduced the number of correctional officers.

Our visit was facilitated by Norris Henderson and Calvin Duncan, two men who had spent more than two decades each in Angola, much of that time in the law library, and our first stop was in the prison law library. There were probably fifty of us on the tour, and we crammed inside the small library, sitting in plastic chairs. About twenty prisoners surrounded us. They were Angola’s legal support system—nearly two dozen inmate lawyers. For more than an hour, they talked to us about their cases and the challenges of doing legal work with severely restricted access to legal research platforms like Westlaw and Lexis Nexis. Secondary sources and treatises were also unavailable. One of the inmate lawyers asked us about Montgomery. Though held up as a huge win and a significant advance in Eighth Amendment jurisprudence, the win hadn’t trickled down to Angola.

Despite the 2016 ruling that mandatory life sentences for juveniles violate the Constitution’s Eighth Amendment prohibition against cruel and unusual punishment, Henry Montgomery, incarcerated in 1963, is still locked up in Angola. According to the Alabama Department of Corrections inmate search tool, Evan Miller still serves a life sentence, too. Christopher Simmons serves a life-without-parole sentence in Missouri. Terrance Graham—whose crime never warranted a sentence approaching LWOP in the first place—was resentenced to twenty-five years and is still incarcerated in Florida.

The post-Graham cases offer the best lens through which to see the challenges with implementing the Graham, Miller, and Montgomery holdings on the ground. Florida provides a good illustration. Five years passed before the Florida State Supreme Court addressed the question of what term of length sentence offended the Graham ruling. This occurred only after several courts routinely resentenced people to exceedingly long prison terms—sometimes in the hundreds of years.50 For example, after Daryl Thomas’s concurrent life sentences for armed robbery and aggravated battery were modified, he was resentenced to concurrent fifty-year sentences.51 The sentencing judge argued that since he would be released in his late sixties and had a life expectancy of 70.2 years, the sentence did not violate the prohibition of life for nonhomicide offenses established in Graham.52

Only in 2015, after having the question certified to the court multiple times,53 did the Florida Supreme Court provide direction to courts seeking to determine what term-of-years was constitutional in a Graham resentencing. In Gridine v. State, the court held that a seventy-year sentence denies a juvenile “a meaningful opportunity for early release” and therefore violates Graham.54 Gridine relied on the rationale the court articulated in a case that came down on the same day. In Henry v. State, the court held that a ninety-year sentence violated the prohibition against LWOP sentences articulated in Graham.55The Henry court began to carve out the outer limits of permissibility in sentencing. Still, while Henry asserts that Graham confirmed “that juveniles constitute a category of offenders that are not as capable of engaging in conduct that is as ‘morally reprehensible’ as adults and, therefore, cannot be reliably ‘classified among the worst offenders,’”56 it remains difficult to determine just what kind of sentence is appropriate for a juvenile who has committed a nonhomicide offense, let alone a homicide.

* * *

At sixteen, I faced a life sentence in prison for carjacking and robbery. Instead, in 1996, the judge sentenced me to nine years. Most of the men I served time with, myself included, were guilty. I met Rojai during the fall of 1997. He slept in the cell that adjoined mine, a cell on a small corridor filled with twenty similar cells and a single shower. We were all under the age of twenty-one, and many of us were under the age of eighteen. Fats was seventeen, like me. He’d been convicted of murder a few months earlier in Richmond, Virginia. The judge sentenced him to fifty-three years in prison. During my nine-year stretch in prison, Rojai was the only person so adamant about his innocence. And I’ve never forgotten him.

Years later, when I was a third-year law student, Rojai reached out to me through another friend. In his letter to me, he told me that Deanthony M. Doane, another prisoner incarcerated in Virginia, had written an affidavit admitting to the murder for which Rojai was serving fifty years. His story reminded me of The Shawshank Redemption.57 As I read the affidavit, I wasn’t sure if an attorney would take it seriously, let alone if a court would consider it. Still, I did what I could. I contacted lawyers, including Deirdre Enright at the Virginia Innocence Project. The story was compelling, but at that stage, the case seemed thin.

Then, in September 2016, Staunton News Leader reporter Brad Zinn published a story that changed everything.58 Zinn unveiled startling facts. For one thing, the arresting officers had never interviewed Rojai. Worse still, Rojai was an easy target to set up—raised by his mother and older sister, they’d struggled with homelessness and had little support to challenge an unjust arrest. That the evidence strongly pointed to other suspects didn’t matter to the arresting officer; no other suspects were seriously pursued. I forwarded Zinn’s article to Enright, and soon the Virginia Innocence Project had taken on Rojai as a client.

Rojai’s case may seem only tangentially related to Graham and Miller. Many courts would argue that he does not have a life sentence, and that even if his sentence is considered a life sentence, he did not receive a mandatory life sentence. And the State of Virginia would argue that geriatric parole, available to Virginia prisoners when they reach the age of 65, satisfies the requirement of hope articulated in Graham.59 Still, Rojai’s case makes it plain that the Supreme Court decisions broadly, and Justice Kennedy’s decisions specifically, have helped shift the way some think about people alleged to have committed crimes as juveniles. What may secure Rojai’s release is a prosecutor who seems to have adopted the logic of Kennedy’s majority opinions in a letter he’s written to the Governor supporting Rojai’s release from prison.

Brian Wainger, the prosecutor in Rojai’s case, the other living person involved, has written:

I never imagined, approximately 18 years after leaving the field of criminal prosecution and commencing a career in civil law, that I would be confronted with the possibility that a person I prosecuted may well be innocent of the charges for which he was convicted. That none of us will ever know with certainty the guilt or innocence of Mr. Fentress does not alone justify his incarceration. Such a default is not acceptable to me.60

Wainger’s letter makes plain that Rojai’s case forces us to think about is how much time is enough. Concluding his letter, Wainger writes,

[t]aking into consideration all of the facts and circumstances of this case, I recommend that the Governor grant Mr. Fentress’s Petition for a Conditional Pardon. If, on the other hand, Mr. Fentress committed the crimes for which he was convicted, he has already served a fair sentence of 23 years, and deserves a second opportunity at life, now as an adult.61

While our state and federal policy priorities may run contrary to the underlying logic of these Supreme Court cases, it cannot be denied that the understanding of childhood advanced in these cases has moved us in the direction of asking and answering more critical questions about how much time in prison is enough.

* * *

What we fail to admit about the influence of Justice Kennedy on these juvenile cases is that, as Justice Thomas emphasized in his dissent,62 the difficult question is the one left unanswered in Graham: is twenty years appropriate? Is thirty? Moreover, what if the foundation of these decisions, the reliance on brain science development, can’t carry the water? At sixteen, I remember believing that though my crimes were wrong, prison could not be the outcome. I’d graduated from high school a year early and had never been considered anything less than intelligent. Maybe this is why Scalia’s dissent in Roper challenges me. What if we’d gotten an individual assessment? Is there still a reason for us to have been given a break?

In the wake of the Supreme Court’s shift in Eighth Amendment jurisprudence around sentences for juveniles convicted of violent crimes, some scholars have acknowledged that the argument regarding the science of brain development—the foundation for why juveniles should be treated differently—is tenuous.63 On the surface, Justice Scalia’s assertion in the Stanford opinion that “[s]ocioscientific, ethicoscientific, or even purely scientific evidence is not an available weapon”64 has proven false. In Rethinking Juvenile Justice, Elizabeth S. Scott and Laurence Steinberg argue that scientific knowledge about adolescent development “should be the foundation of the legal regulation of juvenile crime.”65 But as Terry Maroney points out, the effect of this highly touted approach has been limited. At best, the successes have been at the margins. But even conceding that there has been significant change in what occurs at the margins, the question remains: was a swing vote ever a good reason to fail to develop other arguments for why young people who have committed serious crimes deserve a break?

Gideon Yaffe’s new book The Age of Culpability presents such an argument.66 The book deserves more careful attention than I can give it here. It is worth pointing out that in nearly 30 years of Supreme Court decisions circling around these issues, the reasons for and against giving kids a break have not fundamentally changed. But Yaffe shows us that the dominant arguments for why kids should be treated more leniently are unpersuasive. He begins by taking as a given that young people, juveniles under the age of eighteen when they commit their crime, deserve a break. His intention, then, is to identify what “underlying philosophy of the nature of childhood, the nature of responsibility, the nature of crime, the justification of punishment, and the justification of the disenfranchisement of kids . . . supports and justifies us in giving kids a break?”67 And his conclusion is less complicated than the hidden machinations of the adolescent brain. Yaffe argues that “kids should be given a break because they are disenfranchised, denied as much say over the law as adults, and so denied an equal role in authoring the law’s demands.”68 But the challenge, even if one agrees with Yaffe, is a question that he leaves unanswered in his book: what break do children deserve?

One response to this is to return to Justice Kennedy. Maybe the Justice, in his opinions, was writing to an ideal: hope as the horizon that should comfort. Yet our society has done little to address the lengthy sentences given to teenagers throughout the 1990s. Advocates have been equally silent on what to do about the aging men still serving out lengthy prison sentences. We should not ignore the successful challenges to life without parole, but by anchoring reform efforts to the longest, most egregious sentences, have we decided to kick the question of what the break should be down the road?

Justice Kennedy’s retirement opens the possibility that the Court will retreat from its more expansive interpretation of the protections the Eighth Amendment affords to juvenile defendants. But how much would it matter? Graham, the most significant of the Justice Kennedy opinions made no real claims about proportionality in sentencing. We should not consider these cases, which fail to address the root of the most pervasive problems, as the centerpiece successes of juvenile justice reform. In the case of Rojai Fentress, even innocence is not enough to swing the prison doors open—at least not yet. In the case of other men incarcerated during their teenage years, a wide array of mechanisms prevents the Graham cases from affecting their lives in any consequential way.

What matters to the people I know—the men who were given twenty- and thirty-year sentences as teenagers—is the right to a just sentence that includes a real possibility of parole. A simple start would place a ceiling of twenty-five years on prison sentences. The number, though arbitrary, accounts for the research that suggests that crime peaks during the late teenage years, after which it gradually declines.69 Moreover, there is too much temptation for judges and prosecutors alike to turn the expectation of a lengthy sentence into a default. If state systems coupled a twenty-five year maximum sentence with the availability of parole, even in the absence of robust programming, prisoners would find ways to demonstrate their rehabilitation. As it stands, men like Rojai, serving out fifty-year sentences, are often still striving to rehabilitate themselves, but have no forum to appeal to for an early release.

Conclusion

In 1893, an early Supreme Court case addressed the question of youth and culpability.70 A jury convicted Allen, a fifteen-year old African-American boy, of shooting and killing eighteen-year-old Philip Henson during a fight. The Supreme Court overturned Allen’s death sentence because of an incorrect jury instruction.71 The case is illustrative of the questions that would plague the Court for a century and will become again uncertain now that Justice Kennedy has retired.

During Allen’s trial, the trial judge incorrectly told the jury that under Arkansas law the presumption against culpability ended at eleven years of age; instead, it ended at fourteen. The Court recognized that the Arkansas legislature and Arkansas Supreme Court maintained that those under fourteen could not be guilty of a criminal offense.72 Still, Allen, two months past his fifteenth birthday, was not entitled to a presumption against culpability. But the Court was not “persuaded that the consequences of want of accuracy [in the jury instruction] ought to be assumed to have been harmless.”73 Justice Brewer, writing for the dissent, disagreed. “Strike from the case the testimony as to age, and there is nothing in the story of the homicide, whether as told by the witnesses for the prosecution or those of the defendant, which suggests either youth, immaturity, or mental unsoundness.”74 The central issue here, according to the Justice Brewer, was the relevance of youth.

Over the next three years, Allen would have his case returned to the Supreme Court two more times. The first time, it would again be kicked back down to the trial court.75 The next time, his death sentence would be affirmed.76 More than a hundred years later, it is hard to shake the notion that we’ve been winning just to lose in the end. If we fail to think harder about what a just prison sentence for a child really is, Justice Kennedy’s jurisprudence will not spare us this fate.

Reginald Dwayne Betts holds a J.D. from Yale Law School and is currently a Ph.D. in Law candidate at Yale. His latest collection of poetry, Bastards of the Reagan Era, received the 2016 PEN New England Award in Poetry. He is an Emerson Fellow at New America and a 2019 recipient of a National Endowment for the Arts fellowship for poetry.