The United States Court of Appeals for the Second Circuit has overturned a six and a half year sentence imposed by a federal district court in 2009, after the judge in that case impermissibly relied on a genetic theory to support the argument that the man, Gary Cossey, was likely to re-offend. Professor Douglas Berman at Sentencing Law and Policy has the details borrowed from the decision (warning PDF) and The New York Times.
The facts of the case are not particularly unusual for a federal child pornography charge. In the fall of 2008, Mr. Cossey was charged of two counts of possession of child pornography, both violations of 18 U.S.C. § 2252A(a)(5)(B). Both counts charged him with possessing the prohibited images between September of 2003 and October of 2005. Before those dates, in 2001, the Federal Bureau of Investigation informed Mr. Cossey that he was the subject of a child pornography investigation. Although his decision to continue viewing images despite the FBI visit would have been a proper reason to find that Mr. Cossey was at a high risk of re-offending, that is not what the district court relied on. From the Second Circuit's decision:
The Government is correct in asserting that considering Cossey’s return to viewing child pornography after the FBI investigation would be an appropriate basis for the court’s finding that Cossey is at a high risk for re-offending. However,the court, citing "a gene you were born with," and "not a gene you can get rid of," stated that "[y]ou are what you’re born with. And that’s the only explanation for what I see here. Because . . . there’s no other explanation for the fact that the FBI is knocking on the door telling you that you are the subject of investigation for precisely what you end up prosecuted for. You walk away from it; when you think they’re no longer there, you’re right back at it." The court then stated that the problem is that "[i]t is something you could not control." After explaining its personal views regarding Cossey’s conduct and the reasons behind it, the court concluded "[f]or all of those reasons . . . I am persuaded that the advisory guideline here, at the minimum, is the appropriate sentence."
It is undisputed that it would be impermissible for the court to base its decision of recidivism on its unsupported theory of genetics. For Cossey’s challenge to survive, there must be error and it must be plain. Where a district court relies on its own scientific theories of human
nature to sentence a defendant, as it does here, a finding of plain error is warranted. The court’s belief that Cossey was genetically incapable of controlling his urges affected the court’s decision to sentence him to imprisonment, to impose a prison term that is lengthy, and to order him to
submit to supervised release for life, all of which affect Cossey’s substantial rights. Once plain error affecting substantial rights has been established, an appellate court may exercise its discretion to correct it if it seriously affects the fairness, integrity, or public reputation of judicial proceedings. See Doe, 297 F.3d at 82. It is uncontroversial to conclude that a sentencing decision that relies on factual findings that were unsupported in the record, and thus could not
possibly have been established by a preponderance of the evidence, seriously affects the fairness, integrity, and public reputation of judicial proceedings.
An interesting factual and political aside: The New York Times reports that Judge Sharpe was appointed by President Bush in 2004, while the Northern District of New York's website indicates that he was appointed in 1997. The NY Times reporting appears to be based on a Library of Congress entry that indicates Judge Sharpe replaced Judge Thomas McAvoy. Just an aside, as I think that this error could have been just as easily committed by a Democrat or a Republican.
Neither attorney wanted to discuss this case with the Times, but looking at the plea agreement, some idea of how the negotiations might have gone comes to mind, if only because this case is fairly typical. The defendant in this case was charged with possession, and not receipt. What's the difference? Receipt of child pornography is covered by 18 U.S.C. § 2252A(a)(2), and even without any priors the charge carries a mandatory minimum of five years imprisonment. See 18 U.S.C. § 2252A(b)(1) ("Whoever violates, or attempts or conspires to violate, paragraph (1), (2), (3), (4), or (6) of subsection (a) shall be fined under this title and imprisoned not less than 5 years and not more than 20 years..."). Although the elements of possession and receipt are not identical (and in fact possession is a lesser included offense, similar to the relationship between simple possession and possession with intent to distribute narcotics charges), the overwhelming majority of current child pornography defendants could be charged with both. We no longer live in the 1970s or 1980s; before the advent of the internet, a defendant who wanted to procure prohibited images would need to go out of his way to obtain them. Today, he only needs to purchase a computer and download file sharing software.
With this in mind, the plea agreement that might strike others as a bad bargain makes much more sense. Given the prospect of a five year mandatory minimum, 57 months is not that bad; the maximum sentence is ten years, and the advisory federal sentencing guidelines presumably called for more (at the very least, 78 months, or six and a half years, was within the range). It also preserved his right to appeal a sentence beyond 57 months which, it turns out, was a damn good thing for Mr. Cossey.
Let’s start by acknowledging that the district court ignored two psychological evaluations in this case. Mr. Cossey was referred to the first one by Pretrial Services, the probation arm of the U.S. courts. The second evaluation was apparently conducted on the defendant’s own initiative. Both of these evaluations apparently concluded that the defendant was at a low to moderate risk of re-offending. This is important because 18 U.S.C. § 3553(a) lists several factors that courts can and must consider when imposing a sentence. Among the factors listed is the risk of re-offending, via the route of specific deterrence and protection of the public, along with the need to consider the particular characteristics of the offense and the defendant at issue.
So with this in mind, why did the district court reject the psychological evaluations? Because according to the court, psychology, as a field, is all over the map on sex offender issues. I have a lot of sympathy for the court’s decision here. There does not appear to be any consensus on causality for sexual attraction to prepubescent children, or pedophilia, apart from a general sense of relief that the condition itself appears to be quite rare because the prospect of treatment is spotty at best. Additionally, although many pedophiles are interested in child pornography, the offense of possession of child pornography ensnares many more people (primarily men) than the condition of pedophilia or child molestation. As Professor Hamilton of the University of South Carolina notes in a recent publication, the research to date does not really support this view:
In general, the literature supports the view that while child molesters may possess child pornography, those that possess child pornography are generally not likely to engage in contact offenses against children. Instead, child molesters are merely a small subset of child pornographers. Or, another way to look at it is that child molesters and child pornography offenders are two groups with occasional overlap in membership. There are strong reasons to differentiate the groups. Research shows that contact sex offenders with child victims vary in significant and risk relevant ways from child pornography offenders. In a study comparing a group of convicted internet child pornography offenders with convicted child molesters, researchers found that child molesters were significantly more likely (6 times) to have committed prior sexual offenses against children prior to the offense of conviction. Notably, the child molester group committed no prior child pornography crimes. The same study showed differences from child pornography offenders where the child molester group held more assaultive attitudes, experienced higher levels of psychopathy, were far more likely to recidivate, and were more likely to otherwise engage in sexually risky behavior. Other researchers also comparing groups found that child molesters tested at greater levels of overassertiveness and cognitive distortions as to whether children enjoy sexual contact and are not harmed by it. The latter researchers concluded that these variations could explain why child pornography offenders do not progress to being contact offenders.
Melissa Hamilton, THE EFFICACY OF SEVERE CHILD PORNOGRAPHY SENTENCING: EMPIRICAL VALIDITY OR POLITICAL RHETORIC?, Draft Copy, 22 STANFORD LAW & POLICY REVIEW, available online at the Social Science Research Network.
In fact, there is some research that suggests the focus on pornography may actually be counterproductive. Child pornography defendants list multiple reasons for using the contraband; their motives include arousal, imitation and catharsis. In one study, the offenders who listed catharsis as a motivation for their offense also contended that the availability of the contraband keeps them from acting out against live children. Of course, so does prison, and I would not expect this argument to get a defendant very far in front of a judge given that one of the two rationales used to support criminalization of child pornography in the face of First Amendment challenges was the argument that the demand for the images creates a market for child sexual abuse, but it is interesting to consider the possibility that a more relaxed sentencing regime for possession of child pornography could also reduce rates of physical child sexual abuse.
The reason for the judge’s confusion in this case, however, is easy enough to identify. Proponents of severe sentencing for child pornography possession often rely on questionable social science data. Professor Hamilton refers to "two of the most commonly cited studies that severe sentencing supporters use to bolster their positions," studies that "were produced by two Federal Bureau of Prisons psychologists working with the institution‘s residential sex offender treatment program at Butner, North Carolina." Hamilton, supra at 39. The two studies, conducted in 2000 and 2009, respectively, contained several flaws: questionable interpretation of results, biased sampling and methodological problems. In fact, the Bureau of Prisons refused to allow the 2009 study to be published because the authors refused to make recommended changes regarding generalizability; as one BOP official put it, "We believe it is unwise to generalize from limited observations gained in treatment or in records review to the broader population of persons who engage in such behavior.." Hamilton, supra at 45. In fact one of the study’s authors backed away from its use "to fuel the argument that the majority of [child pornography] offenders are indeed contact sexual offenders and, therefore, dangerous predators," stating that "[t]his simply is not supported by the scientific evidence."
Judging by the proliferation of sex offender legislation in the last two decades, there has been very little need or desire for scientific evidence. Moral panic has led Congress and state legislatures to support an ever-expanding regime that broadens the scope of sex offense laws, ranging from child pornography to residency restrictions and chemical and physical castration, even lifetime monitoring and civil commitment for first time offenders. The federal and state judiciaries have been reluctant to impose limits on any of this activity; with the exception of Kennedy v. Louisiana, a 2008 Supreme Court case that struck down Louisiana’s law allowing the imposition of the death penalty for extreme cases of child sexual abuse, there has been virtually no decision limiting this regime. The bipartisan repudiation of that 2008 case illustrates the essential problem; President Obama, then a senator running for the presidency, condemned the Supreme Court’s decision. His solicitor general, Elena Kagan, now a justice herself, argued that civil commitment of first time pornography offenders was necessary "to run a criminal justice system that does not itself endanger the public." Extended Civil Commitment of Sex Offenders Is Upheld, NY Times, May 17, 2010, available at http://www.nytimes.com/... (Last accessed January 29, 2011). In fact, the civil commitment legislation does not require a prior conviction, only Bureau of Prisons custody. The federal government has even attempted to civilly commit alien detainees destined for deportation and military prisoners; the only impediment to those commitments has been a finding that the defendants were under the legal custody of Immigration and Customs Enforcement (ICE) and the U.S. Army, respectively. See United States v. Hernandez-Arenado 571 F.3d 662 (7th Cir. 2009) and United States v. Joshua, 607 F.3d 379 (4th Cir. 2010).
The prevailing criminal justice regime in the United States is unheard of outside its borders. According to a 2007 Human Rights Watch report, state and federal law is truly an outlier:
Sexual violence and abuse against children are, unfortunately, a worldwide problem. Yet the United States is the only country in the world that has such a panoply of measures governing the lives of former sex offenders. It is the only country Human Rights Watch knows of with blanket laws prohibiting people with prior convictions for sex crimes from living within designated areas. To our knowledge, six other countries (Australia, Canada, France, Ireland, Japan, and the United Kingdom) have sex offender registration laws, but the period required for registration is usually short and the information remains with the police. South Korea is the only country other than the United States that has community notification laws.
Officials in Australia, Ireland, and the United Kingdom have considered and in each case rejected the adoption of universal community notification laws (although in some cases, police are authorized to notify the public about the presence of a convicted sex offender in the neighborhood). After reviewing the experience of the United States, they concluded that there is little evidence that community notification protects the public from sex crimes, and that such laws are often accompanied by vigilante violence against registrants.
Human Rights Watch, No Easy Answers: Sex Offender Laws in the United States, September 2007 Report, available at Human Rights Watch website, Link (PDF).
There has been little in the way of a national debate and discussion of this issue, and there does not appear to be any significant impetus for reform. Neither political party has any desire to tackle this issue, and has instead relied on questionable or nonexistent data to support draconian legislation that has overshadowed more probing underlying questions about treatment, recidivism and public safety.
Senator Jim Webb of Virginia has proposed a National Criminal Justice Commission Act, which would "create a blue-ribbon, bipartisan commission of experts charged with undertaking an 18-month top-to-bottom review of the nation’s criminal justice system and offering concrete recommendations for reform." Although the commission enjoys bipartisan support and passed through the House of Representatives last year, it is unclear what its fate will be in this Congress. It is also unclear whether or not the commission will tackle the sensitive issue of sex offender policy, given the bipartisan path to the current crisis.
I support Senator Webb’s proposal, but I think that it is clear that sex offender policy needs to be examined in its own right. The proposed NCJC is not aimed at any particular class of offense, but the issues surrounding sex offender legislation pose unique scientific, legal and policy questions that cannot be easily subsumed under the more generalized NCJC proposal. If the issue is addressed by the NCJC, it will probably be in the form of a panel recommendation, and those recommendations will not be binding on the NCJC under the proposed legislation. Moreover, the commission will be specialized, but not on sex offender issues.
There are several reasons to call for a national, specialized commission on sex offender issues. First, much of the legislative activity surrounding sex offender legislation was conducted without the guidance of empirical research. The recent Second Circuit decision is illustrative but not particularly unique; although there is no credible research that indicates child pornographers have compulsions rooted in genetics, this assumption (which is itself built on a questionable assumption about recidivism rates) motivates much of the innovation we see in sex offender policy, including community notification, civil commitment and chemical castration. In order to implement effective policy, legislators need sound research, or at the very least a credible assessment of the state of current research.
Second, the internet and new technology both pose serious problems for this area of the law, whether discussing adult or juvenile offenses. Several high profile "sexting" cases illustrate the problem. Teenagers and preteens with access to cell phones and computers are not only at risk of predation, they are at serious risk of becoming sex offenders themselves. Additionally, the easy availability of child pornography on the internet creates the risk of creating an offender class that might not ordinarily exist in the absence of that access. We do not know enough about the underlying social and behavioral reality to make sober policy judgments, however, and our failure to address these questions is creating intolerable and often lifelong human misery.
Finally, a credible expert commission will probably propose legislative changes that politicians will have a difficult time swallowing. A bipartisan commission will provide them with some insulation from political backlash.
In an age of declining revenue, deficit reduction and fiscal austerity, the United States cannot afford to ignore the structural problems created through decades of bad and bipartisan policy choices. The NCJC is an important first step in this process; it has been prompted in part by a national discussion over the so-called "War on Drugs" and the deleterious impact it has had on state and federal sentencing and incarceration. It is time to address a much more taboo and unpopular subject.