Doe, Sex Offender Registry Board No. 291554 v. Sex Offender Registry Board

Wolohojian, J.

(dissenting). When creating the sex offender registration scheme, the Legislature determined that sex offenders should be classified by the Sex Offender Registry Board (board) into three levels depending on whether they present a low, moderate, or high risk of reoffense. Although these terms are undefined, they must remain connected to our ordinary understanding of the quantitative concepts underlying them. Thus, a low risk of reoffense is “not merely a hypothetical or speculative potential risk.” Doe, Sex Offender Registry Bd. No. 24341 v. Sex Offender Registry Bd., 74 Mass. App. Ct. 383, 388 (2009) (Doe No. 24341). “Rather, it is a risk that is ‘cognizable’ — i.e., ‘perceptible’; ‘[cjapable of being known, perceived, or apprehended by the senses or intellect.’ ” L.L. v. Commonwealth, 470 Mass. 169, 179 (2014) (footnote omitted). A moderate level of risk must, by virtue of logic, common sense, and fairness, be something quantifiably more than a low level of risk. Likewise, a high level of risk must, for the same reasons, be appreciably greater than a moderate one. The inquiry is prognostic; the hearing examiner is required to assess not only what (i.e., another sex offense) might happen in the future, but also its quantifiable likelihood (i.e., low, moderate, or high) of occurring. Because human behavior is unpredictable, and predictions of future behavior are inherently *216 difficult and complex, both prongs of the inquiry must rest on something other than guesswork, conjecture, fear, or surmise.

This is not to say that the Legislature intended the three levels of classification to be a purely statistical analysis. The hearing examiner is entitled to, and should, examine qualitative factors that may also bear on the person’s future risk of reoffense. There must, however, be an objective quantitative basis for the particular classification. Otherwise, if we allow the terms “low,” “moderate,” and “high” to become separated from their quantitative core, a person could be classified as a level three (high risk) sex offender regardless of whether he presents a five percent risk of committing another sex offense or a seventy-five percent risk. This is unacceptable.

I respectfully dissent because the board did not prove, by a preponderance of the evidence, the quantum of risk Doe presents of reoffense. We do not know even in a general way, let alone with any precision — nor did the hearing examiner find — where on the spectrum of risk (from zero to one hundred percent) the plaintiff, John Doe, falls. Nor did the hearing examiner explain why she found Doe’s risk of reoffense to be high rather than moderate. Indeed, the board concedes that it cannot explain why Doe presents a high risk as opposed to a moderate risk. It argues, instead, that the discretion afforded to the hearing examiner permitted her to classify Doe as either a level two (moderate risk) or a level three, and that neither classification would have been error based on the facts. If accepted, this argument would essentially eliminate the very structure of the three-tier classification system, which is premised on distinct and different levels of risk. It would also make even arbitrary decisions of hearing examiners unreviewable.

It is not enough to establish that a person presents some risk of reoffense. To avoid arbitrary and capricious application of the law, one “must be able to [articulate] and [describe] based on affirmative evidence,” why and how a person presents a particular risk of reoffense (whether low, moderate, or high) as opposed to any other.” Ibid. See Doe No. 24341, supra at 387-388. The terms low, moderate, and high cannot and should not be interpreted so amorphously as to permit arbitrary classification. The board was required to prove by a preponderance of the evidence that Doe presents a cognizable, perceptible risk of reoffense that is greater than low, and also greater than moderate. This it did not do. There was, for example, no expert evidence presented on the risk of reoffense by a person who, like Doe, has no criminal history or *217 history of sex offenses. We are left to guess what predictive value the characteristics of Doe’s acts (while serious) might have on the quantum of risk Doe presents in the future. The record contains no empirical or other scientific information or data concerning the recidivism rates for a person who has committed acts of the nature at issue here. And it already has been noted that “there is reason for some concern as to whether the guidelines [upon which the board relies] continue to reflect accurately the current state of scientific knowledge.” Doe, Sex Offender Registry Bd. No. 68549 v. Sex Offender Registry Bd., 470 Mass. 102, 116 (2014).

For these reasons, I would vacate the denial of Doe’s motion for judgment on the pleadings and remand the matter to the hearing examiner for further proceedings to determine Doe’s level of risk of reoffense based on affirmative evidence.