Risk Management: Psychological Evaluation of Parolees

The public concern with risk and dangerousness in the context of release always sparks lively debates, much of which focus on the management and containment of sex offenders. It is therefore curious that two recent new items, relating to release practices, have retained relatively low profiles.

The first one is related to psychological evaluation of parolees. Part and parcel of the release of any California inmate is a psychological risk assessment. However, no such risk assessments were required for federal or out-of-state parolees who then served their parole in California. Jaycee Lee Dugard's kidnapping by Philip Garrido, who served time in Nevada before being a California parolee, was the inspiration for this bill, whose full text is here.

This bill would require the Department of Corrections and Rehabilitation to assess every person on parole transferred from any other state or by the federal government to this state who has been, or is hereafter convicted in any other court, including any state, federal, or military court, of any offense that, if committed or attempted in this state, would require the person to register as a sex offender, as specified. The bill would specify that this assessment shall occur no later than 60 days after a determination by the Department of Justice that the person is required to register as a sex offender, as specified. The bill would revise the definition of “eligible person” for the purpose in the paragraph above to include all persons required to register as sex offenders pursuant to the Sex Offender Registration Act.

I wonder if this legislation (which, in itself, is fairly sensible, depending on the reliability of the risk assessment tool) would have really prevented Dugard's kidnapping. The answer to that lies not only in conditions for release, but also with the indicia of dangerousness and rehabilitation that parole agents later look for. Would a man with a steady job, a house, and a seemingly functional family drawn any attention? Would his deep pathologies have revealed themselves through the risk assessment tool? These are very hard questions to answer.

In other (not unrelated) news, Cathleen Gagliani, a Democrat assemblywoman from Stockton, has approached Gov. Schwarzenegger to help prevent the release of Loren Herzog, convicted for multiple murders. More on why this is somewhat of an outlier case from the Sac Bee:

In a letter to Schwarzenegger this week, Galgiani asks the Republican governor to "intervene and request a clinical review and risk assessment of Loren Herzog's mental status" in an attempt to keep him incarcerated as a sexually violent predator.

Under state law, a judge can order a sexually violent predator incarcerated for an additional two-year period, with review once that stint ends, if the person is found by two psychiatrists or psychologists to have a mental disorder that makes reoffending likely.

Galgiani wants Schwarzenegger to target Herzog for such a mental evaluation before his release, saying in her letter that "I believe that Loren Herzog will be found to likely reoffend."

One potential obstacle for Galgiani is that the law permitting continued incarceration of sexually violent predators, beyond their parole date, targets those who have been convicted of sex offenses against two or more victims.

Though Herzog pleaded to voluntary manslaughter, Galgiani said in her letter that he was "believed to have been a participant in two rapes" and that "two women testified against Shermantine and Herzog for alleged rapes they never pressed charges for."

This raises another big issue, namely, that of the extent to which we rely on convictions as a record of the "truth". As many of our readers probably know, the vast majority of criminal cases end in plea bargains, in which defendants plead to more lenient charges than those originally pressed in return for a predictable sentence. It is problematic to rely on these records as a genuine expression of what the defendant is believed to have done, but it is also very problematic to assume the opposite, given the prosecutorial incentives to overcharge for the purpose of bargaining. So, when assessing Herzog's dangerousness, do we take into account only the offenses he was convicted of, or also those for which there was evidence but not charges?