June 14, 2010
Juvenile justice delayed?—A Commentary by Jeffrey Fagan
The following commentary was published in The National Law Journal on June 14, 2010.
Juvenile justice delayed?
By Jeffrey Fagan
Juvenile justice advocates hoping for a big win in Graham v. Florida came away instead with a consolation prize, and a fairly small one at that. Terrance Lamar Graham's supporters hoped to extend the logic of the Roper v. Simmons ban on juvenile executions to mandatory life without parole (LWOP) sentences for minors. And the U.S. Supreme Court in Graham did just that, but only for nonhomicide offenders. Rather than establishing a firm principle of discounted culpability that would cabin harsh sentencing for all minors, Graham instead offers eligible juvenile offenders a "meaningful opportunity to obtain release based on demonstrated maturity and rehabilitation." The "means and mechanisms for compliance" are left up to the states.
This is where Graham's narrowing principle may be illusory. Graham — and Joseph Sullivan, who was 13 at the time of his violent crime and whose case was heard simultaneously with Graham's — will be eligible for parole consideration after a minimum sentence set by the state legislature. So far, so good. But the devil really is in the details, and here's where it gets complicated.
The Court's approach is similar to the one it used in Atkins v. Virginia, the 2002 decision that banned the execution of mentally retarded offenders convicted of capital crimes. In Atkins, the Court left it up the states to decide how to consider, measure and prove mental retardation. Nearly a decade later, there remain bitter and contentious disagreements between the states on the standards and procedures — as well as their accuracy — to define and classify mental retardation. It is never easy to translate a diagnosis based on a spectrum disorder into a legal definition.
What standards will the states develop to diagnose, measure and prove "maturity" or "amenability to treatment" for a diverse group of juvenile offenders? Immaturity is hardly a disease (despite what "any parent knows," as Justice Anthony Kennedy said in Roper). Are these legal categories and definitions — or clinical judgments?
These are complex dimensions of human development with strong disagreements among scientists, and anchored to both shifting social norms and changing social science facts. Even a social-framework approach that seeks consensus in the empirical facts as a way for courts to take judicial notice of complex scientific issues won't be particularly helpful in applying these facts to individual cases, a procedural requirement that was the basis of Chief Justice John Roberts Jr.'s concurrence.
There is no less a challenge in crafting the standards for a "meaningful" opportunity to obtain release based on "demonstrated maturity and rehabilitation." This is strong directive language in the discourse of the Supreme Court, signaling that merely permitting "discretion" won't satisfy the new review standard. The states are now tasked with designing the "means and mechanisms for compliance." But this device inherently relies on procedure more than substance. While striking down LWOP for minors, Graham allows states to deny parole eligibility consideration for periods as long as 40 years or more, as Justice Samuel Alito Jr. notes in his dissent. And Roberts, in his concurrence, points out that just because you get parole consideration doesn't mean you necessarily get out.
In short, the Graham opinion is a very narrow application of the 1983 Solem v. Helm principle: We don't like mandatory LWOP for noncapital crimes, but as long as you're eligible for release, we're not going to prescribe when. States such as South Carolina may no longer be able to use a mandatory LWOP sentence to lock up a two-time 15-year-old burglar for life, as it did in 2002 in Standard v. State. But a state could well decide that it will wait 40 years to even think about parole. The 40-year rule is exactly what Colorado did when it abolished juvenile LWOP in 2006. That still is a harsh sentence, by any measure, and harsher than what most adult felons get now for anything less than murder or a third strike.
But the Graham court also showed some ambivalence about proportionality and seems to dance around its 1991 Harmelin v. Michigan opinion that permitted LWOP sentences for nonmurder crimes and rejected proportionality tests when execution is not at stake. In Graham, the Court reaffirmed the culpability discount that animated Roper and tipped the proportionality test away from execution. But the Court seems to have decided that LWOP (or death in prison) for a minor who commits murder is not out of balance. But then the Court rules that LWOP is out of balance with nonmurder crimes by minors, again invoking Roper's implicit culpability discount, and all the behavioral and social science that drives it. This is a bit dizzying and suggests that something else may be at play in the Court's reasoning.
Perhaps the best way to understand Graham is to see it as results-oriented. The Court denied juvenile LWOP to nonmurderers so that it could save mandatory LWOP for capital crimes and other murders. After the firestorm of criticism over the Roper ban on juvenile executions, this Court would have risked a crisis of legitimacy if it went the next step and banned the second-harshest punishment, no matter how logical that extension might be.
With the Graham rules, the United States will remain outside the nearly universal international norms opposing LWOP for minors, with only the subjectivity of parole boards as the regulator of uniquely American near-life sentences for minors. A better path would be for the states to recognize the constitutional fact of diminished culpability of adolescents by applying a "youth discount" on sentences for juveniles who are sentenced as adults. It reduces the guesswork in parole decision-making and infuses the virtue of even-handedness into the jurisprudence of juvenile crime.
Jeffrey Fagan is a professor at Columbia Law School and a visiting professor at Yale Law School.