California Answers Some of the Graham/Miller Questions, Sort Of

As I discussed in a recent post, the United States Supreme Court left many questions unanswered in its two recent decisions on life without parole for juveniles.  In the first case, Graham v. Florida (2010), the Court banned LWOP for juveniles convicted of nonhomicide offenses.  Then, in Miller v. Alabama (2012), the Court banned mandatory LWOP even for juveniles convicted of homicide.  These were important Eighth Amendment decisions, but the lower courts have been left to implement them without much guidance.

Yesterday, the California Supreme Court began to address some of the unanswered questions in People v. Caballero.  I think Caballero got things right, as far as it went, but the case left much open for future litigation. 


The facts of the case were straightforward enough.  Caballero, age 16, shot a gun at three members of a rival gang.  Only one was hit, and he recovered.  Caballero was then convicted of three counts of attempted murder.  The judge imposed consecutive sentences of 40, 35, and 35 years to life on the three counts, for a total of 110 years to life.

Here are the questions addressed by the California Supreme Court.

First, the court held that attempted murder counts as a “nonhomicide” offense, and thus comes within Graham‘s ban on LWOP for juveniles.  As Justice Werdegar noted in a concurring opinion, this holding was not strictly mandated by Graham, which hinted in places that attempted murder might be treated for Eighth Amendment purposes as a homicide offense.  However, the California court decided to treat the result of the crime, rather than the intent, as the controlling variable.

Taken to an extreme, this approach may produce some anamolous results.  For instance, while an attempted murderer would be categorically spared from LWOP, a juvenile convicted of negligent homicide would not be, even though many people (including, I think, the vast majority of criminal-law theorists) would regard attempted murder as a much more serious offense than negligent homicide.  This may be more a theoretical than a real problem because statutory maximums tend to be relatively low for negligent homicide and because sentencing judges will not often be inclined to impose extreme punishments for an accident.  However, the problem may eventually arise in a case involving multiple counts.  Imagine a teenager whose criminally negligent driving causes a bus to crash, involving a half-dozen deaths.  Six times whatever the statutory maximum is for negligent homicide could, depending on the jurisdiction, start to get into the “de facto” life sentence range.  In my view, though, under the proportionality-based logic of Graham, the Eighth Amendment protections should not be any less for negligent homicide than they are for attempted murder.

Second, the court held that a 110-year sentence is a life sentence for Eighth Amendment purposes.  This seems sensible enough; to hold otherwise would invite easy evasion of Graham through the use of very long sentences that would unquestionably consume all that remains of the defendant’s life.  However, the court declined to draw a bright-line rule as to what counts as a life sentence, at least not in the form of “X or more years.”  Instead, the court indicated that the test is whether the defendant’s minimum parole eligibility date is beyond his “natural life expectancy.”  This, in turn, was defined as the “normal life expectancy of a healthy person of defendant’s age and gender living in the United States.”

It’s not clear why these criteria were chosen.  For instance, I wonder, why take into account the defendant’s gender but not his chronic health conditions or socioeconomic status or the fact that he is going to spend the next few decades of his life in prison?  Based on all of the information that gets collected and crunched when one purchases life insurance, I imagine that existing actuarial knowledge could permit considerably more fine-grained determinations of life expectancy without creating litigation quagmires for the courts.  Indeed, I expect that the limited California approach will eventually come under serious pressure with respect to juveniles with serious health conditions who can demonstrate without any real question that their actual life expectancies are far short of those of a healthy person of the same age and gender.

Third, the court held that the Graham rule should be applied to Caballero’s entire sentence, rather than to each count separately.  This seems to make sense, especially given that Caballero’s conviction, while divided into three counts, was really for a single incident.  Would the result be different, though, if the three offenses were more clearly distinct from one another?  What if the three shots were fired a month apart from one another?  What if Caballero was charged for the first shot and was out on bail when the second two occurred?  By common ways of thinking about proportionality, these variations on the facts would render Caballero more blameworthy.  Perhaps, then, he should not get the benefit of having his sentence evaluated in a unitary fashion.

Coming at the issue from the opposite direction, though, one might wonder to what extent prosecutors could evade Caballero by charging separate offenses in separate indictments or informations.  Presumably, the sentences would still be treated in a unitary fashion for Eighth Amendment purposes if the cases were consolidated for trial, but defendants might not seek consolidation for fear that evidence of multiple offenses would prejudice them in the eyes of the jury.

Yet another problem is presented by new offenses committed while a juvenile is under sentence for a prior offense.  What if Caballero, in prison on his three counts of attempted murder, tries to kill another inmate?  Or what if he is given a long suspended sentence, commits another serious crime, and faces both the imposition of the suspended sentence and a long new sentence?

And what if Caballero, in connection with the same three-shot incident that prompts the California prosecution, is also prosecuted federally on a gun charge and receives, say, a ten-year federal sentence while the state case is pending?  Would the California courts have to take the federal sentence into account in deciding how long the state sentence could be?

Fourth, the court indicated that Caballero should be resentenced so that he could get a parole hearing within the period of his life expectancy.  The court directed that the sentencing judge “must consider all mitigating circumstances attendant in the juvenile’s crime and life, including but not limited to his or her chronological age at the time of the crime, whether the juvenile offender was a direct perpetrator or an aider and abettor, and his or her physical and mental development.”  It is not clear, though, what “must consider” means.  Imagine that the sentencing judge says this: “I’ve listened to all of the mitigating evidence presented by Caballero, and I have decided to give it zero weight.  Juveniles must be sent a message not to join gangs and not to use firearms.  Caballero’s life expectancy is age 74.  He can have his parole hearing when he is 73 years and 364 days.”  What then?

Finally, the court stated that other incarcerated juvenile offenders seeking to modify their sentences “may file petitions for a writ of habeas corpus in the trial court.”  It is not clear, though, if this is the only permissible remedy.  If an offender believes that the trial court would be particularly unsympathetic, for instance, could the offender petition the parole board directly for a hearing?

It’s good to see the California Supreme Court trying to give life to Graham, rather than minimizing its import through the sort of crabbed formalism for which the state was arguing.  Whether California will prove a national trendsetter or an aberration remains to be seen.