Hersey Lelaind knew he was in trouble — just not how much trouble. He and a housemate had been on a drive, and Lelaind had been smoking pot. When they returned to their home in Vacaville, California, the sheriff’s department was waiting.
“We pull into the driveway. They draw down guns on us and tell us to get out of the car,” Lelaind said in an interview with BuzzFeed News last May. Both men were living with other parolees in a house that had been targeted for a sweep to check for violations. The cops led Lelaind, handcuffed, through the front door of the house, which had already been searched. They made him give a urine sample.
Lelaind knew he’d test positive for marijuana. But after spending the past few years bouncing in and out of jail, mostly for minor parole violations, he wasn’t afraid of another 30-day stretch. I can do that standing on my head, he thought.
That was in 2006, when Lelaind was 26 years old. He’s been kept under lock and key ever since. His problem wasn’t the drug bust itself. But the bust prompted the authorities to review Lelaind’s checkered past. As a teenager, he had been convicted for sexual abuse against a minor — and had served his time.
That fact, along with other aspects of his criminal and life history, were entered into the “Static-99,” a little-known but highly influential questionnaire that critics contend is being tragically misused. The test spit out a score that set him on the path to being locked up in a state psychiatric facility. Why? Because he might commit another crime in the future. He doesn’t know if he will ever be released.
Lelaind is one of thousands of men detained under so-called “civil commitment” laws enacted by 20 states, Washington, D.C., and the federal government. Under these laws, offenders who have been judged to be a high risk can be detained indefinitely — not as criminals, but as psychiatric patients.
The Static-99 helps decide which offenders are the riskiest, and looms large over civil commitment proceedings. It weighs a variety of facts about a sex offender’s past in order to predict the likelihood of future offenses.
More precisely, what the Static-99 predicts — with modest accuracy, at best — is the risk that men within a group of sex offenders will commit a new sex offense, compared to other members of that group. Experts agree that it’s a useful tool for managing sex offenders in prison — assessing which of them need higher levels of security, for example. But the way the test is used in civil commitment — to help make high-stakes decisions about offenders’ liberty after they have served their criminal sentences — is highly controversial.
Civil commitment is “a ridiculous solution on multiple levels,” Raymond Knight of Brandeis University in Waltham, Massachusetts, who studies sexual aggression, told BuzzFeed News. He doubts that it will ever be possible to devise a test that is predictive enough to justify locking individuals away indefinitely for something they might do in future.
In civil commitment trials, Static-99 scores are often converted into precise-sounding rates of reoffending. But over the last few years, as the test’s powers of prediction have been studied in diverse groups of released sex offenders, two big problems have emerged.
First, the original group of Canadian and British criminals used to validate the test came from an earlier era in which sexual violence was more common than it is today. Even for their time, these criminals seem to have been particularly dangerous. This means that the reoffending rates quoted in many civil commitment trials have been significantly overestimated.
What’s more, reoffending rates vary widely among the different groups of men in which the Static-99 has been studied — making it extremely hard to know how to convert a Static-99 score to a meaningful prediction of an individual’s future behavior. To which group should he be compared? Prosecution and defense experts tend to disagree on the answer, and consequently wind up quoting widely differing estimates.
These are not minor issues. And yet the Static-99 remains a cornerstone of civil commitment.
“It’s being relied upon so heavily,” Karen Franklin, a forensic psychologist in El Cerrito, California, who spoke for the defense at Lelaind’s civil commitment trial, told BuzzFeed News. “I think it’s an attempt to make this whole process seem more scientific than it is.”
Lelaind grew up in the projects of Bayview-Hunters Point, a mostly black neighborhood of San Francisco where unemployment and poverty were rife. He was the middle of three boys, all of whom were removed from their crack-addicted mother by the city’s Department of Social Services in 1987, when Lelaind was 6.
The boys moved into their grandfather’s home, and this helped, at first. But later social services documents describe a deteriorating situation in which Lelaind’s increasingly disruptive behavior was met with harsh physical punishment. By 1990, Lelaind was continually getting into fights at school, and had started what he called “bootie-bumping” other children, grinding his crotch against their buttocks.
Most of these incidents were with his younger brother, but one social services document states that it also happened with an 18-month-old baby who was living in the home. Lelaind’s grandfather responded to what he called this “punkish and sissy” behavior with more beatings. By then the old man was running into financial and marital problems, and was in poor health. He could no longer cope.
So Lelaind, then aged 10, and his younger brother were placed in a foster home across the San Francisco Bay, in Oakland. Lelaind continued grinding himself against his brother, and was once found in bed with his foster parents’ 11-year-old adoptive daughter. Lelaind was removed from the placement and for the remainder of his childhood lived in a series of group homes.
When he was 15, criminal charges were filed against him twice. Once he tried to hit a group home worker with a vacuum cleaner. The second incident involved a sexual assault against a 10-year-old boy in his group home. The boy accused Lelaind of trying to remove his pants and threatening to rape him. Lelaind denied it, but was ordered by a juvenile court to participate in sex offender treatment.
Lelaind's offenses continued into young adulthood. When he was 18, he was convicted of sexually abusing a 10-year-old female relative. The abuse happened on a number of occasions, though the specifics are murky. Lelaind denied that there was any penetration, but pleaded “no contest” to the charge of “lewd and lascivious acts” with a minor and was sentenced to three years in prison.
When he was released on parole, in April 2002, the state did not pursue civil commitment. California then required qualifying offenses of sexual violence against two or more victims. Lelaind had just one, his young relative.
Back on the streets, Lelaind, now 21, couldn’t keep out of trouble. Six times he violated parole, each time spending months back inside. The first was for possession of pornography. Others were for minor infractions such as violating curfew or talking to some 17-year-old boys while out riding his bike. Another time a parole officer found he was dating a woman with a young child, breaching the terms of his release. Lelaind also had sexual relations with two 17-year-old girls. One of these cases resulted in a new two-year sentence, after he pleaded guilty to unlawful sexual intercourse.
Yet throughout this time, there is no evidence that Lelaind repeated his original offense, no record of any sexual interest in prepubescent children, and none of the violent outbursts that blighted his youth.
By the time of the parole sweep, in November 2006, Lelaind was living in a transitional home with other sex offenders in Vacaville. As well as testing positive for marijuana, he was again busted for possessing pornography. Records from this arrest show that the offending items were pictures of seminude women and copies of King magazine, a mainstream title aimed at young black men featuring rappers, urban fashion, and scantily clad models.
Even with an additional parole violation for pornography, Lelaind expected to spend just three months inside. But when the time for his release came around, nothing happened. When Lelaind asked why he hadn’t been let out, he was told he had a 45-day hold from something called the “SVP” (for sexually violent predator) program. Lelaind’s sex with his 17-year-old girlfriends did not count as sexual violence. But in 2006, California had amended its civil commitment law, so now one victim was enough.
In February 2007, Lelaind was referred to two state-appointed psychologists, who each filled out the Static-99 as part of their assessments. Both concluded that he met the criteria to be judged a sexually violent predator, setting the stage for his civil commitment trial.
Civil commitment laws came about in response to high-profile cases in which released sex criminals carried out horrendous acts of violence. The first, passed by Washington state in 1990, followed the conviction of Earl Shriner, who raped a 7-year-old boy, cut off his penis, and left him for dead in the woods. Shriner had a history of sadistic assaults, including the killing of a 15-year-old girl. He was also known to fantasize about kidnap and torture.
The value of keeping monsters like Shriner locked away is clear. But few sex offenders are as obviously dangerous as he is. In general, rates of reconviction are low: Only about 5% of sex offenders are convicted of a new sex crime within five years of release. Some are undoubtedly more dangerous than others, but forensic psychologists are remarkably bad at predicting which ones pose the greatest risk. Their “expert” judgment, studies have shown, is not much better than tossing a coin.
If civil commitment was to be anything more than a crapshoot, the courts would need a better prediction method. One emerged, in the late 1990s, from the work of two psychologists who were wrestling with a subtly different problem.
Karl Hanson, working for Canada’s solicitor general in Ottawa, and David Thornton, then at the London headquarters of Her Majesty’s Prison Service, wanted to put sex offenders into categories, based on their risk of reoffending, so that they could be managed better while in prison. As well as helping to determine the security level under which prisoners should be held, the researchers wanted a better way of prioritizing which groups of offenders should be considered for release. “It was very much about the triage of prison populations,” Thornton said. “It wasn’t really about the forensic assessment of individual offenders.”
Both researchers were studying groups of male sex offenders who had been released in their respective countries, looking for factors associated with higher rates of conviction for new sex crimes. Hanson also reanalyzed the results of 61 prior studies, about half of them conducted in the U.S., which collectively involved more than 23,000 released offenders. He tested the predictive power of everything that was recorded — a long list of factors such as current sentence length, the age of any child victims, and whether any victims had been physically injured.
Not surprisingly, the most useful factor in predicting future sex offending was the number of prior sex offenses. Hanson also zeroed in on whether any victims were male or not related to the offender — both of which seemed to increase the risk. Thornton’s list of risk factors was longer, including prior nonsexual violence, a high number of prior sentencing dates, and having never been married.
None of these additional variables, individually, were very strong predictors. Some lacked a clear theoretical explanation of why they were linked to sexual reoffending. But they were the best of the bunch, and improved the overall predictive power of the tests the two researchers were developing if they were added into the mix, alongside the tally of prior sexual offenses.
By the late 1990s, the two researchers decided to combine their respective checklists into a single test. They called it Static-99, after the year of its birth and to emphasize that it measured unchanging facts about an offender’s life history, rather than psychological traits that might shift over time.
The result was a simple 10-item checklist that is usually filled out by reviewing an offender’s files. Since its debut, the Static-99 has assumed an oracle-like status in the world of sex offender management.
Up to three points are scored for prior sex charges or convictions, and one point more if any victims were male. Another point is added for a high number of prior sentencing dates — both for sex offenses and other crimes. And so it continues, through each risk factor. It’s much like calculating auto insurance premiums based on your accident history and where you park your car at night.
The highest possible score is 12, and those who score more than 6 are considered to have a “high” risk of reoffending. Even a score of 4 or 5, in the “moderate-high” range, can mean the difference between release back into society and being considered for indefinite detainment.
Static-99 scores do not predict the severity of potential future offenses, however. Rapes involving extreme violence and the abuse of young children are lumped together with crimes like voyeurism and indecent exposure.
The most recent survey of U.S. civil commitment programs, conducted in 2014, found that at least 4,658 offenders were held under civil commitment. (Five programs failed to provide any numbers, so the true number is higher.) Another 829 were detained, awaiting a decision on their status. The average Static-99 score for civilly committed offenders across the nine states that reported those numbers was 5.
The idea behind the Static-99 is similar to medical tests that try to predict patients’ health outcomes. Such tests can be evaluated using a statistical method that assigns a score of 1 to a perfect test, which always flags serious problems and never gives a false alarm. A worthless test — one that predicts no more accurately than random guessing — gets 0.5. The Static-99, when tested on some 1,300 Canadian and British sex offenders, scored 0.7.
By the standards applied to medical tests, that’s a mediocre score — about as good as judging someone’s likelihood of getting heart disease by weighing known risk factors such as age, smoking, and blood cholesterol.
The Static-99’s predictions may be imperfect, but in the land of the blind, the one-eyed man is king.
The Static-99’s predictions may be imperfect, but in the land of the blind, the one-eyed man is king. So it’s perhaps no surprise that U.S. psychological evaluators eagerly embraced the new tool to assess candidates for civil commitment.
Thornton remembers their interest in his presentations on the Static-99 at scientific conferences. “I think that for a lot of evaluators, it seemed to bring a certain amount of order, and a rational way to proceed,” said Thornton, who is now research director at the Sand Ridge Secure Treatment Center, where Wisconsin’s civilly committed sex offenders are held.
Although he devised the test for different purposes, Thornton now supports its use in civil commitment evaluations. “Given that you’re asked to make a judgment about this, it forms one important input,” he said.
A high Static-99 score isn’t the only factor weighed in civil commitment cases. With rulings in 1997 and 2002, the U.S. Supreme Court decided that offenders who are civilly committed must have a mental abnormality that affects their self-control, predisposing them to acts of sexual violence.
Still, the Static-99 is widely used to screen sex offenders who are nearing release from prison, to help determine who should be considered for civil commitment. Prosecutors then use the scores, together with psychiatric diagnoses, to argue that offenders pose an unacceptable public danger.
