Sexual Predators

Emergence. In the late 1980s and early 1990s, a confluence of factors produced a set of innovations that supplemented the traditionally reactive governmental role in addressing sexual violence. The main tools of the new approach were registration, notification, and civil commitment. Underlying notification and civil commitment is a fourth tool, systematic risk assessment. All of these tools recast the role of government, placing upon it the affirmative task of prevention through the systematic identification and control of individuals who pose a danger of future sexual harm.

The beginnings of the second generation of sex offender commitment laws took root in about 1989. Task forces in the states of Washington and Minnesota proposed a renewed use of civil commitment as a tool for containing sexually violent individuals. Both states were addressing public outrage and intense media coverage of heinous crimes committed by sex offenders recently released from prison. Both states faced the same dual-faceted problem: First, in the early 1980s Washington and Minnesota had replaced indeterminate sentencing with determinate schemes under which incarceration terms were set at sentencing. This change deprived the states of their power to exercise long-term, flexible control over offenders deemed too dangerous to release from prison. Second, both states replaced indeterminate sentences with a system of presumptive sentencing guidelines. By the late 1980s, society's increased attention to sexual assault had produced a change in its perceived seriousness. The sentencing guidelines were seen as inadequate punishment. By the late 1980s and early 1990s, individuals imprisoned under the new sentencing guidelines became eligible for release. This exposed a gap in the states' ability to protect against recidivist sexual violence.

Contemporary Western thought endorses the view that the state has a strong obligation to protect its citizens from violence. This protection imperative manifested itself in the form of a principle of incapacitation—the notion that public protection against violence is best accomplished by locking up criminals. Thus, the need to close the gap in sexual-violence protection produced a strong push to extend the systems for the incapacitation of sex offenders.

Key constitutional constraints prevented the states from addressing this incapacitation imperative by simply increasing criminal punishments for sexual violence. Retroactive increases would have run afoul of prohibitions on ex post facto legislation and double jeopardy prosecutions. Prospectively, the states could—and did—increase sentence lengths for sex offenders. But both Washington and Minnesota rejected a return to long indeterminate sentences for sex offenders, at least in part because such a scheme would be inconsistent with the principle of desert. Finally, the Constitution prohibited the state from imposing criminal punishment simply for the status of being mentally disordered and dangerous, or for the punishment of future predicted crimes (Robinson v. California, 370 U.S. 660, 666 (1962); Lafave and Scott).

Searching for an escape from these constraints on the criminal law, both states turned to civil commitment to close the "gap" in social control. Neither state addressed in any careful way the failure of the earlier sex psychopath laws, nor did either state attempt to assess whether commitment was a wise allocation of treatment and prevention resources. Soon, Wisconsin, Kansas, and California followed suit.

The new use of civil commitment differed in one critical respect from the first generation of laws. Instead of being addressed to persons considered "too sick" for punishment, the new laws would aim at those considered "too dangerous" to release from prison. Though the statutory language differed from law to law, the structure of all was similar.

Sex offender commitment laws typically contain four substantive elements and two procedural characteristics. Substantively, the laws aim at individuals who have (1) a history of sexually violent or dangerous conduct and (2) a mental or personality disorder or "abnormality" which (3) "causes" or "predisposes" the individual to (4) likely future dangerous sexual conduct. The statutes provide little guidance for either the type of mental disorder that qualifies or the risk-threshold for dangerousness. Some courts, however, have suggested that a high probability of future harm is constitutionally required. The laws specify procedural protections that are typically higher than standard civil litigation standards, but may fall short of the protections required in a criminal prosecution. The laws generally allow the state to force the respondent to testify, and they abrogate the rules of privilege and confidentiality to allow the admission of relevant treatment records without the consent of the individual. Some, but not all, states require proof beyond a reasonable doubt and allow for a jury trial. Although details vary, committed individuals can expect to remain committed until the conditions giving rise to the commitment—either the presence of a mental disorder or the prediction of future dangerousness—are ameliorated. Commitment statutes generally require the provision of treatment.

Constitutional challenges. The new use of civil commitment inspired swift constitutional challenges. In badly split opinions, the supreme courts of Washington, Minnesota, and Wisconsin held the laws to be constitutional. A federal court in Washington and the Kansas Supreme Court held the laws unconstitutional.

In 1997, reversing the Kansas Supreme Court, the U.S. Supreme Court upheld the constitutionality of the Kansas commitment law in Kansas v. Hendricks, 521 U.S. 346 (1997). The court disposed of constitutional challenges on two main fronts. The substantive due process challenge argued that the laws extend civil commitment beyond its proper scope. The "punishment" challenge asserted that the laws had a punitive purpose, and therefore violated the double jeopardy and ex post facto prohibitions of the U.S. Constitution.

Substantive due process imposes limits on state power to restrict the liberty of individuals. In a series of cases, the Supreme Court has defined some of the constitutional limits on civil commitment. In O'Connor v. Donaldson, 422 U.S. 563 (1975), the Court suggested that it would be unconstitutional for a state to confine a harmless though mentally ill person. In Addington v. Texas, 441 U.S. 418 (1979), the Court said that commitment can rest on both parens patriae and police power, but the Court did not clarify whether both powers are necessary, or either is sufficient, to support civil commitment. Some commentators read United States v. Salerno, 481 U.S. 739, 748 (1987), a case upholding a statutory system of limited preventive detention of persons charged with criminal offenses, as approving preventive (noncriminal) detention based on dangerousness alone. This theory was rejected, however, in Foucha v. Louisiana, 504 U.S. 71 (1992), where the Court struck down dangerousness-based civil commitment in the absence of mental illness.

These cases provided the context for the constitutional challenges in Hendricks. Citing Foucha, Hendricks argued that the Kansas law exceeded the state's constitutional powers because it failed to require a "mental illness" as a prerequisite to commitment. Challengers argued that terms defining the commitment class either were not "medically valid" (e.g., "mental abnormality") or, though medically valid, were not "mental illnesses" (e.g., "personality disorder"). The Court rejected these arguments. While acknowledging that commitment requires some "mental disorder" predicate, the Court held there was no particular constitutional significance to the term "mental illness," thus personality disorders were not categorically ineligible to be constitutional predicates for commitment. The Court stated that the mental-disorder predicate for civil commitment is a legal, as opposed to a medical, term. Thus, the use of "medically valid" categories is not constitutionally required.

What, then, is the legal definition of "mental disorder" in the commitment context? In Addington, the Court focused on mental illness as a foundation for the assertion of the state's parens patriae power. But the state's parens patriae power was not alluded to in Hendricks, no doubt because the subjects of sex offender commitments are not incompetent, and the commitments are justified by reference to the public's interest, rather than the individual's. Thus, the question posed in Hendricks is the role played by mental disorder in commitments that rest solely on the state's police power.

Although Hendricks is not entirely clear, the Court's holding suggests that the role of "mental disorder" in this context is to maintain the primacy of the criminal law in the fight against antisocial violence. The constitutional role of the "mental disorder" requirement is to limit civil commitment to a small portion of the social control function of the state, and to provide a justification for the use of preventive confinement as a supplement to the criminal law. As the Supreme Court in Hendricks has stated, mental disorder limits civil commitment "to a small segment of particularly dangerous individuals," and "adequately distinguishes [persons subject to civil commitment] from other dangerous persons who are perhaps more properly dealt with exclusively through criminal proceedings" (521 U.S. at 360). This interpretation reinforces the Court's statement in Foucha v. Louisiana that states must justify their departure from the "ordinary criminal processes . . . the normal means of dealing with persistent criminal conduct" (504 U.S. 71, 83 (1992)).

The Supreme Court in Hendricks stated that the Kansas law passed this constitutional test by focusing on the individual's inability to control his dangerous sexual behavior. Several state supreme courts have interpreted this language as imposing a constitutional "volitional dysfunction" test for sex offender commitments.

Lack of volitional control is sometimes associated with the insanity defense. Limiting commitments to criminally excused persons would satisfy the Court's requirement of a distinct and narrow application for civil commitment and would preserve the primacy of the criminal law. But, as noted below, this interpretation is contrary to actual practice, where almost all persons committed have been held responsible for their prior crimes. In light of this, many commentators argue that "volitional dysfunction" is an essentially meaningless concept. They argue that the courts' use of volitional dysfunction simply serves to limit commitments to repeat offenders—those whose past course of criminal conduct indicates a failure to control sexually violent behavior.

Double jeopardy and ex post facto. The double jeopardy clause prevents the state from "punishing twice, or attempting a second time to punish criminally, for the same offense" (Witte v. United States, 515 U.S. 389, 396 (1995). The ex post facto clause "forbids the application of any new punitive measure to a crime already consummated" (Lindsey v. Washington, 301 U.S. 397, 401 (1937). Challengers claimed that sex offender commitment laws are, in actuality, punishment masquerading as civil commitment, whose true purpose is to extend the confinement of sex offenders who have served their prison terms. The Supreme Court in Hendricks rejected this argument on the ground that incapacitation of dangerous individuals is a well-accepted purpose of civil commitment.

Challengers also argued that the failure to provide effective treatment to committed individuals is evidence of a punitive intent. Doubts about the efficacy of treatment were central to the critique of the earlier sex offender commitment laws. Studies of sex offender treatment in the 1980s were taken as suggesting that "nothing works." Indeed, the legislative preambles to a number of the statutes acknowledged that treatment of sex offenders was problematic. The sex offender commitment laws were thus attacked on the ground that incarceration, rather than treatment, was the primary goal of the legislation.

One branch of this argument asserted that there is no proven effective treatment for the "condition" of being a sexual predator, and that commitment is thus tantamount to lifetime incapacitation. Hendricks stated that the availability of effective treatment is not a constitutional prerequisite for civil commitment, explaining that a contrary holding "would obligate a State to release certain confined individuals who were both mentally ill and dangerous simply because they could not be successfully treated for their afflictions" (521 U.S. at 367). An alternative argument asserted treatment is possible, but that Kansas failed to provide it. Most courts acknowledge, under the influence of Youngberg v. Romeo, 457 U.S. 307 (1982), that the Constitution requires states to provide available treatment, at least up to some minimum level. In Hendricks, the Supreme Court suggested that Kansas' rather "meager" treatment program was not fatal because of the newness of the commitment law. But the laws have since matured. A federal district court has enjoined as constitutionally inadequate conditions and treatment in Washington State's Special Commitment Center (Turay v. Seling, 108 F.Supp.2d 1148 (WD Wash. 2000). In a separate case, the Supreme Court held that such conditions do not entitle an inmate to release on a writ of habeas corpus pursuant to the ex post facto and double jeopardy clauses (Seling v. Young, 121 S.Ct. 727 (2001)). The Seling case did not decide whether persistent unconstitutional conditions might entitle an inmate to release under other legal theories, such as substantive due process.

Commitment statutes in practice. Prior to the Supreme Court's Hendricks decision, five states and the District of Columbia had active sex offender commitment programs. In the years after the Hendricks decision approximately ten additional states have enacted laws. About twelve additional states had legislation under consideration as of the year 2000.

The sex offender commitment programs follow a common pattern. Officials are required to assess the risk posed by sex offenders who are about to be released from prison. This assessment process identifies a group for referral to the commitment process. Minnesota and Kansas refer about 10 percent, whereas Arizona referred 25 percent and California about 3 percent. Within the judicial system, prosecutors exercise discretion in deciding whether to file a formal petition, and some cases may be dismissed at a probable cause hearing, or after a full-blown trial on the merits. Though judging net commitment rates is difficult, programs appear to commit at rates ranging from less than 1 percent to about 6 percent of sex offenders about to be released from prison.

The targets of contemporary sex offender commitment laws are most often people whose abilities to reason and perceive are not psychotically impaired, who can care for themselves and are properly held responsible for their behavior. These individuals have mental disorders that are characterized by strong "deviant" sexual desires (paraphilias), or maladaptive, persistent personality patterns (personality disorders). A large proportion also have diagnoses of chemical abuse or dependency. Almost all of the commitments are of individuals who have been held criminally responsible for their conduct.

Commitment under these laws is to secure treatment facilities, mostly operated by departments of human services or mental health. In a few states, the facilities are operated by departments of corrections, or are located within correctional facilities. One state's law (Texas) provides for out-patient commitment.

The annual cost for commitment programs ranged (in 1999) from a low of about $70,000 per person to a high of $164,000. As of the year 2000, almost one thousand persons were under commitment nationwide. Commitments are long-term. Only a handful of individuals have been released from commitment, even in those states with relatively mature programs. Thus, state officials predict continued growth in both the population under commitment and the consequent cost of the programs. A Minnesota study predicted growth in cost of 450 percent between 1998 and 2010, and California projects a growth of about 400 percent over the same time period. Legislators and policymakers in some states are alarmed by the growth, and are seeking ways to reduce the use of civil commitment as a tool to prevent sexual violence.

Evaluating sex offender commitment schemes. As indicated above, sex offender commitment laws are controversial. Most of the debate has centered on the constitutionality and morality of the laws. Though the broad constitutional questions have been resolved, serious questions remain about the actual manner in which these statutes are applied. Many of these questions implicate moral concerns, but others raise social policy issues that can be properly resolved only through empirical research. This section summarizes the critiques and social policy concerns.

The civil liberties critique. The civil liberties critique worries that sex offender commitment laws sweep too broadly and arbitrarily, and therefore exceed their proper role as a narrow exception to the primacy of the criminal justice system. Civil confinement is proper only for those dangerous individuals who are "mentally disordered." But researchers report that a majority of prisoners in general, and of sex offenders in particular, are diagnosable with some form of mental disorder, if that term includes personality disorders. Thus, a broad mental-disorder requirement allows commitments to sweep within their reach most of the criminal population, undercutting the argument that commitment schemes are limited and principled exceptions to the primacy of the criminal justice system.

As indicated above, some courts claim they apply commitment only to persons whose mental disorders render them unable to control their sexual behavior. But courts routinely apply the inability-to-control standard to sex offenders who have been held fully responsible for their behavior. Critics contend this is a contradiction that renders the inability-to-control standard a meaningless legal fiction. In practice, commitment requires nothing more than a history of sexual violence and a diagnosed "mental disorder." Such a definition, rather than distinguishing civil commitment from criminal confinement, allows commitment virtually to swallow the criminal law.

Critics also contend that prediction is so inaccurate that it renders commitment essentially arbitrary, and therefore unconstitutional. This is a "facial" challenge the courts have rejected, in large measure because prediction is central in a broad range of legal settings—from the death penalty to child custody.

Critics also make an "as applied" argument. At least in their rhetoric, some courts and legislatures assert that commitment statutes target only the "most dangerous" offenders who are "highly likely" to reoffend. Challengers argue that the prediction evidence offered in individual cases does not satisfy these thresholds. However, some scholars report advances in the ability to predict sexual recidivism among released sex offenders. Researchers have developed actuarial instruments that employ factors related to sexual recidivism (such as sexual offense history, age, antisocial personality or psychopathy, physiologically measured sexual deviance, and patterns and persistence of victimization) and statistically derived weights that reflect their relative importance. Researches apply these tools to populations of released sex offenders and measure the correlation between the resultant scores and the population's recidivism. In assessment, the test score for an individual is interpreted to indicate a risk of reoffense equal to the proportion of identically scoring study subjects who committed another sex crime. Some researchers conclude that careful risk assessment of this sort can reliably identify a small group of sex offenders with a risk of sexual reoffense of at least 50 percent, and possibly as high as 70 to 80 percent.

Despite these apparent advances, critics urge caution in the use of these risk assessment tools, especially where the consequence is a deprivation of liberty. Samples may not be representative of the present prison-release group, thus introducing a sampling error. Apparent correlations may shrink when tests are cross-validated on other populations. Human error will introduce an error of measurement into the results. High probabilities of sexual reoffense assume a follow-up period of decades, and so may be of little value in making judgments about risk management in the immediate future. At present, most of the instruments take either no, or minimal, account of dynamic or changeable risk factors, such as treatment completion or level of community supervision. These factors can significantly affect risk levels. Finally, some of the test instruments have not been subject to publication in peer reviewed journals, and may not comport with professional standards for psychological testing.

The resource allocation critique. Sex offender commitment programs consume resources hugely disproportionate to the numbers of individuals confined. For example, in a recent year, Minnesota spent about $17 million per year on the 150 men committed as sex offenders, while spending only about $1.2 million per year on sex offender treatment for the eleven hundred men confined to state prisons, and $1.1 million on community-based treatment for sex offenders. Proponents often point to the incapacitation of committed individuals as proof of the efficacy of commitment laws. But commitments confine only a small percentage of all recidivist sex offenders. Most recidivists will be released from prison at the end of their sentences, and the vast majority of future sex offenders are in the community or in the correctional system. Given a finite pool of resources, critics argue that research is needed to determine whether an expensive commitment program that focuses on a very small number of dangerous individuals is the most effective way to spend prevention and treatment dollars. They argue for careful consideration of alternative programs, such as intensive "containment" approaches in the community, beefed-up treatment and supervision programs for offenders under correctional control, and public-health-style campaigns to change widespread attitudes about sexual violence.

Treatment disincentives. Most commentators are at least mildly optimistic about the efficacy of sex offender treatment. But success depends on full investment by the treatment subject, including disclosure of past sexual misconduct. Commitment laws abrogate confidentiality for precommitment treatment disclosures, and permit prosecutors to base commitment cases on information gleaned from treatment sessions. Thus, critics urge study as to whether the specter of commitment may serve as a treatment disincentive for sex offenders in the correctional or community setting.

Public policy distortion. Critics contend that a "psychological" model for sexual violence is central to civil commitment laws and that adoption of this model can distort public policy. Sex offender commitment laws are based on the theory that the sexual violence is caused by a mental abnormality. This model suggests that sexual violence is aberrational rather than systemic, in that sex offenders are different in kind (they have a "disorder") from "normal" individuals. This model suggests that the greatest danger of sexual assault arises from the deranged predator attacking a stranger-victim. Competing explanations characterize sexual violence as relatively ubiquitous, the manifestation of widespread societal attitudes about women, children, and sex. This competing model suggests that the biggest danger is sexual assault committed by an acquaintance or family member, which constitutes 75 percent of all sexual victimizations. By characterizing sexual violence as aberrational and deranged, the psychological model tends to absolve the larger society of its responsibility to change the conditions that produce this violence.

The civil commitment model also undercuts the individual's responsibility for his own sexual behavior. Most commentators characterize sexual violence as a behavior that is chosen by the individual. They view this choice-model as an important heuristic for changing sexually harmful behavior. The civil commitment laws may undercut and confuse this model by reviving the discredited notion that women can trigger uncontrollable sexual impulses in men.