In re Civil Commitment of Ince

Full title: In the Matter of the Civil Commitment of: Cedrick Scott Ince.

Court: STATE OF MINNESOTA IN COURT OF APPEALS

Date published: Mar 18, 2013

A12-1691 (Minn. Ct. App. Mar. 18, 2013)

Opinion

A12-1691

03-18-2013

In the Matter of the Civil Commitment of: Cedrick Scott Ince.

Lori Swanson, Attorney General, Noah A. Cashman, Assistant Attorney General, St. Paul, Minnesota; and David E. Schauer, Sibley County Attorney, Winthrop, Minnesota (for respondent) Kenneth R. White, Law Office of Kenneth R. White, P.C., Mankato, Minnesota; and Anthony F. Nerud, Nerud Law Office, Arlington, Minnesota (for appellant)

HALBROOKS

This opinion will be unpublished and

may not be cited except as provided by

Minn. Stat. § 480A.08, subd. 3 (2012).

Affirmed

Halbrooks, Judge

Concurring specially, Johnson, Chief Judge

Dissenting, Stauber, Judge

Sibley County District Court

File No. 72-PR-11-52

Lori Swanson, Attorney General, Noah A. Cashman, Assistant Attorney General, St. Paul, Minnesota; and David E. Schauer, Sibley County Attorney, Winthrop, Minnesota (for respondent) Kenneth R. White, Law Office of Kenneth R. White, P.C., Mankato, Minnesota; and Anthony F. Nerud, Nerud Law Office, Arlington, Minnesota (for appellant)

Considered and decided by Johnson, Chief Judge; Halbrooks, Judge; and Stauber, Judge.

UNPUBLISHED OPINION

HALBROOKS, Judge

The district court committed appellant as a sexually dangerous person. On appeal, appellant argues that the state failed to prove by clear and convincing evidence that he is highly likely to reoffend and, alternatively, that the district court failed to adequately address the less-restrictive alternative he offered. Because we conclude that the legal standard for whether appellant is highly likely to reoffend was met and the district court adequately addressed the less-restrictive alternative, we affirm.

FACTS

On September 21, 2011, Sibley County petitioned to civilly commit appellant Cedric Scott Ince as a sexually dangerous person (SDP) and a sexual psychopathic personality (SPP). Ince was 21 at the time and one day away from his scheduled intensive supervised release (ISR) from Minnesota Correctional Facility-Lino Lakes. The petition for civil commitment detailed Ince's two criminal-sexual-conduct convictions and his history of alcohol abuse, as well as his prior treatment, psychological reports, and placements.

Criminal Convictions

Ince has two criminal-sexual-conduct convictions, both stemming from incidents involving alcohol and victims who were close to his age at the time of the offense. In February 2007, when Ince was 17, he sexually assaulted a 17-year-old girl, A.L.B., after she had passed out on a couch at a party. When a friend tried to wake A.L.B. to ask if she knew what had happened, she briefly opened her eyes before passing out. The Waseca County Attorney's Office charged Ince with third- and fourth-degree criminal sexual conduct. A year later, Ince pleaded guilty to an amended count of fifth-degree criminal sexual conduct and was ordered to undergo a sex-offender assessment. On September 15, 2008, the district court adjudicated Ince delinquent and placed him on probation.

Three weeks later, Ince violated his probation by breaking into 19-year-old R.V.P.'s house and sexually assaulting her. R.V.P. told police that she awoke on October 5, 2008, to find Ince on top of her. He choked her, making her unable to breathe, and the two struggled for about 20 to 30 minutes before the sexual assault began. Ince raped R.V.P. and digitally penetrated her, stopping after she said to him, "Wouldn't your dad be proud." Before leaving, Ince threatened to shoot R.V.P. if she called the police. The Scott County Attorney's Office charged Ince, then 18, with first-degree burglary, third-degree criminal sexual conduct, and domestic assault by strangulation. In August 2009, Ince pleaded guilty to third-degree criminal sexual conduct. He was sentenced to a downward departure of 48 months in prison, a ten-year conditional-release term, and required to register as a predatory offender.

Ince later told authorities that he was under the influence of alcohol during both instances. Ince identifies himself as an alcoholic. He began drinking when he was 15 years old and drank daily after graduating from high school. In 2009, he claimed that he drank to the point of "blacking out" once per week. Ince also experimented with various drugs, beginning at age 16, and used marijuana a couple of times a month. He possessed cocaine on school property in October 2006, and was cited multiple times for underage consumption.

Ince was released on bail while each of his two cases was pending and did not maintain his sobriety. After the February 2007 incident, Ince tested positive for marijuana use. Following the October 2008 incident, Ince entered a chemical-dependency treatment program but was terminated from the program after consuming alcohol on two occasions.

Ince's Release

Two days after Sibley County petitioned to civilly commit Ince, he was released on ISR until the conclusion of his commitment trial in May 2012. During those roughly eight months, Ince obtained employment but failed to notify his employers of the specifics of his criminal history and that he was prohibited from having contact with minors. Ince initially lived with his mother but later began renting a farm house from his employer. Although Ince reported to CORE Professional Services that he had bought a farm, his employer testified that was not true and that the two had never discussed Ince buying the farm. While living with his mother, Ince had sexual relations with a woman who was an unapproved visitor. His mother did not report the unapproved visitor, which concerned Ince's ISR agent. Beginning in January 2012, Ince entered court-ordered sex-offender treatment through the CORE treatment program.

Commitment Proceedings

Prior to the commitment trial, the district court appointed two psychological examiners, Penny Zwecker, Ph.D., and Peter Marston, Ph.D., to evaluate Ince. A third psychologist, Rosemary Linderman, Ph.D., reviewed their interviews of Ince as well as his records. During a four-day commitment trial in May 2012, the district court received testimony from the three psychologists, Ince, his employer, his ISR agent, and the two victims. The district court found Dr. Marston's opinions to be particularly persuasive and convincing.

Dr. Marston interviewed Ince and conducted risk assessment and psychopathy testing. Ince scored a 30 on the Hare Psychopathy Checklist Revised (Hare PCL-R) Second Edition, which Dr. Marston said indicated that Ince "appears to present most of the features and functioning of a clinical psychopath and [is] sufficient to sustain a provisional diagnosis of psychopathy." For a sexual-recidivism assessment, Dr. Marston administered the Static-99R and also examined dynamic factors that have a significant bearing on recidivism. On the Static-99R, Ince scored a +5, placing him in the moderate-high category of risk compared to a general population of convicted sex offenders. Dr. Marston later elevated that score to a +6 based on a juvenile disorderly conduct conviction stemming from an accusation that Ince masturbated or simulated masturbation on a school bus. Dr. Marston testified that based on his assessments Ince is almost three times more likely to reoffend than the typical sex offender and that there is a 62.4% chance that Ince will reoffend in his lifetime. Ince's assessment also evidenced an elevated risk in seven of nine dynamic factors. Based upon the Static-99R actuarial results showing a moderate-high risk and the dynamic-factors assessment showing an elevated risk, Dr. Marston concluded that the "combined results yield a final risk assessment of High Risk or 'highly likely.'"

Dr. Zwecker interviewed Ince and conducted psychological tests. She used two actuarial instruments to assess the likelihood that Ince would reoffend. On the Rapid Risk Assessment of Sex Offender Recidivism (RRASOR) test, Ince scored a 3, which placed him at a medium range of risk for reoffending. On the MnSOST-3.1, Ince had a predicted four-year probability of sexual recidivism of 7.92%, which is higher than 89.90% of incarcerated offenders in Minnesota, and placed him at a high risk to be reconvicted of a sex offense within four years. The two risk-assessment instruments indicated, according to Dr. Zwecker, that Ince "is at a moderate to high risk to sexually reoffend."

Dr. Linderman reviewed the interviews with Drs. Zwecker and Marston and Ince's records and psychological-assessment results. Based on that information, Dr. Linderman gave Ince a prorated score of 28.4 on the Hare PCL-R assessment, which placed him in the "upper end of the moderate range of psychopathy." Dr. Linderman also conducted actuarial-risk assessments that resulted in Ince scoring a +8 on the Static-99R, making him "just short of being five times more likely to sexually reoffend than the typical sex offender." On the Static-2002 test, Ince scored +11, which estimated a sexual recidivism rate of 55% in the next five years and 64.6% in the next ten years. Like Drs. Zwecker and Marston, Dr. Linderman also concluded that Ince is highly likely to engage in future acts of harmful sexual conduct and meets the criteria for commitment as an SDP.

Drs. Marston, Zwecker, and Linderman all agreed that Ince is highly likely to engage in future acts of harmful sexual conduct and meets the statutory criteria required for commitment as an SDP. The three psychologists also agreed that Ince needs inpatient treatment and should be committed to a structured, residential setting, specifically the Minnesota Sex Offender Program (MSOP). Ince disagreed, testifying that he is doing well in the CORE program and that he believes that he can receive proper treatment there and, upon completion of the program, he "will never sexually assault anyone ever again."

At times, Ince's testimony conflicted with that of the victims and with prior statements and accounts that he had given regarding the incidents. For example, Ince testified that A.L.B. was "easy" and that the two had engaged in consensual sex earlier in the evening before she passed out. A.L.B. rejected both claims. Regarding R.V.P., Ince testified that the two had dated for several years and had sexual intercourse more than 1,000 times. R.V.P. also denied both claims.

The district court concluded that there is clear and convincing evidence that Ince meets the statutory requirements for commitment as an SDP, but that he does not satisfy the requirements for commitment as an SPP. In reaching its conclusion, the district court addressed each of the Linehan factors and stated that testimony supported each factor. The district court indeterminately committed Ince as an SDP to the MSOP. This appeal follows.

DECISION

I.

Ince argues that the district court erred by ordering his commitment as an SDP because the state failed to prove the third requirement of the statute—that he is highly likely to reoffend. A person may be civilly committed as an SDP if the petitioner proves the statutory criteria by clear and convincing evidence. Minn. Stat. §§ 253B.18, subd. 1(a), .185, subd. 1(a) (2010). "Clear and convincing evidence is evidence that is 'more than a preponderance of the evidence but less than proof beyond a reasonable doubt.'" State v. Jones, 753 N.W.2d 677, 696 (Minn. 2008) (quoting Weber v. Anderson, 269 N.W.2d 892, 895 (Minn. 1978)). When reviewing a district court's findings on the elements of the civil-commitment statutes, we review factual findings for clear error. In re Commitment of Stone, 711 N.W.2d 831, 836 (Minn. App. 2006), review denied (Minn. June 20, 2006). And "[w]here the findings of fact rest almost entirely on expert testimony, the [district] court's evaluation of credibility is of particular significance." In re Knops, 536 N.W.2d 616, 620 (Minn. 1995). But whether the evidence is sufficient to meet the statutory requirements for commitment is a question of law, which this court reviews de novo. In re Linehan, 518 N.W.2d 609, 613 (Minn. 1994) (Linehan I); In re Commitment of Martin, 661 N.W.2d 632, 638 (Minn. App. 2003), review denied (Minn. Aug. 5, 2003).

An SDP is a person who "(1) has engaged in a course of harmful sexual conduct . . . ; (2) has manifested a sexual, personality, or other mental disorder or dysfunction; and (3) as a result, is likely to engage in acts of harmful sexual conduct." Minn. Stat. § 253B.02, subd. 18c (2010). A showing that a person has a complete inability to control his or her sexual impulses is not required. Id., subd. 18c(b). Rather, the person must be "'highly likely [to] engage in harmful sexual acts in the future.'" Stone, 711 N.W.2d at 840 (quoting In re Linehan, 594 N.W.2d 867, 876 (Minn. 1999) (Linehan IV)).To determine whether a person is "highly likely" to reoffend, the Minnesota Supreme Court has stated that a district court must consider six factors:

(1) the offender's demographic characteristics; (2) the offender's history of violent behavior; (3) the base-rate statistics for violent behavior among individuals with the
offender's background; (4) the sources of stress in the offender's environment; (5) the similarity of the present or future context to those contexts in which the offender used violence in the past; and (6) the offender's record of participation in sex-therapy programs.

Id. (citing Linehan I , 518 N.W.2d at 614).

Neither low actuarial scores nor any other single factor is determinative of whether someone is highly likely to commit harmful sexual conduct. In re Commitment of Navratil, 799 N.W.2d 643, 649 (Minn. App. 2011) (rejecting argument that commitment was not warranted when base-rate statistics indicated low likelihood to reoffend). The supreme court has declined to modify the Linehan I factors to place more emphasis on the third factor—base-rate statistics—over the other factors. See In re Linehan, 557 N.W.2d 171, 189 (Minn. 1996) (Linehan III), vacated on other grounds, 522 U.S. 1011, 118 S. Ct. 596 (1997), aff'd on remand, 594 N.W.2d 867 (Minn. 1999) (Linehan IV).

Here, the district court concluded that Ince meets each of the three statutory requirements for civil commitment as an SDP and used the six Linehan I factors to find that Ince is highly likely to reoffend. The district court specifically noted that it had found Dr. Marston's testimony "particularly persuasive and convincing." The district court stated that, unless specifically noted, it had accepted the testimony and opinions of Dr. Marston and rejected the testimony of the other experts only to the extent they did not agree with that of Dr. Marston.

Ince argues that the district court's findings "are infected by several fundamental errors, warranting reversal." First, Ince argues that Dr. Marston's testimony should be discredited because he based much of his opinion on Ince's score on the Static-99R test, which he alleges was incorrectly increased from a +5 to a +6. We disagree. Dr. Marston explained that the additional point was added to Ince's test score because of a prior conviction for disorderly conduct stemming from a masturbation incident on a school bus that the district court refused to consider. And when asked about the impact of the increased point, Dr. Marston testified that changing the score of the Static-99R did not ultimately affect his overall opinion:

It simply increases his score past the threshold of 5, which is moderately high category of risk to high risk. So it nudges his score into that category. And I had placed him in that category already, based upon my combined analysis—that is, combining that actuarial score with the dynamic—a structured dynamic risk assessment. So this didn't change where I ended up, but it got me there faster, so to speak.

The district court also relied on more than the Static-99R results, noting Dr. Marston's testimony that Ince's Hare PCL-R score of 30 places Ince in a psychopathic range and indicates a greatly elevated risk to reoffend.

Furthermore, Ince is overstating the district court's conclusion regarding his disorderly conduct conviction, because the district court stated that it gave some credibility to the incident but that the incident did not weigh heavily on the district court's determination. In concluding that Ince is highly likely to engage in future acts of sexual conduct, the district court stated that it relied on Dr. Marston's testimony that Ince has "7 of 9 dynamic or individual factors placing him at high risk" and that the "extreme violence of Ince's last offense confirms the elevated risk." The district court also analyzed each of the six Linehan I factors, specifying evidence from Dr. Marston to support each factor, and found that each indicates that Ince is highly likely to reoffend. Each of the six Linehan I factors is supported by evidence in the record.

First, regarding the offender's demographic characteristics, the district court noted that Ince's age, aggressive behaviors, diagnosis, and gender place him at an increased risk to reoffend. This is supported by Dr. Marston's testimony that Ince's age (21) increased his risk of reoffending and that "[t]he problematic behaviors likely to follow from Mr. Ince's antisocial personality disorder and psychopathic disorder are at their peak at his age and will not begin to substantially diminish until his mid-40s, approximately."

Second, regarding Ince's history of violent behavior, the district court relied on Dr. Marston's conclusions that the offenses were extremely violent, serious, and recent. Further, there was an indication that Ince lacks self-control, given that his second offense occurred only three weeks after he was placed on probation. This is also supported by the record. As Dr. Marston noted, Ince's first offense involved a helpless female who was intoxicated and incapacitated, and his second offense involved breaking into a woman's home, violently attacking her and choking her until she feared for her life and then threatening to kill her if she told police. The second offense occurred just three weeks after Ince was placed on probation for his first offense, leading Dr. Marston to state that "[t]he timing of his second offense couldn't be much more alarming."

Third, regarding the base-rate statistics for violent behavior among individuals with Ince's background, the district court relied on Dr. Marston's examination of Ince, which, consistent with the other two psychological examiners, concluded that Ince is at a "high risk" of reoffending. Dr. Marston not only testified to a 62.4% chance that Ince would reoffend in his lifetime, but also that Ince was nearly three times more likely to recidivate than the typical sex offender. The actuarial results presented by Dr. Marston, while not dispositive, support the district court's analysis of the third Linehan I factor and its ultimate conclusion that Ince is highly likely to reoffend.

Fourth, regarding the sources of stress in Ince's environment, the district court noted that although Ince has support in the community, obtained employment, maintained sobriety, and attended sex-offender treatment, Dr. Marston opined that Ince's antisocial personality disorder places him at a high risk to reoffend. Dr. Marston noted that Ince "will, no doubt, find it increasingly difficult to continue to comply with all the rules and requirements given his past lifestyle and personality disordered functioning and impulsiveness." We note that Ince's case is unusual in the fact that he has had a period of freedom prior to his commitment trial. During that time, he obtained employment, leased a home, purchased a vehicle, and remained sober. However, Ince's period of release was not without problems as he failed to inform his employer about his release restrictions, as required, and he had a one-night stand with a woman who was an unapproved visitor. Dr. Marston specifically noted Ince's past record of multiple supervision failures. Thus, the district court's analysis of this factor is also supported by the record.

Fifth, regarding the similarity of the present or future context to those contexts in which Ince used violence in the past, the district court relied on Dr. Marston's opinion that Ince's history of "beating the system" will lead to problems without a secure setting. This is supported by Dr. Marston's report and Ince's admission during his interview with Dr. Marston. Dr. Marston concluded that "as the intensity of his supervision diminishes, [Ince] could more easily place himself in similar situations where he has offended in the past, though it would be difficult for him to sustain this behavior without being detected." During his interview with Dr. Marston, however, Ince stated that he had learned how to "beat the system" involving probation urine checks and supervision in order to avoid being caught.

Finally, regarding the sixth factor involving Ince's record of participation in sex-therapy programs, the district court acknowledged that while Ince was engaged in treatment, the facility was under the incorrect assumption that he had previously completed sex-offender treatment. The district court further noted that all three psychologists testified that Ince has no adequate relapse prevention plan and has previously been terminated from chemical-dependency treatment. During his time in a chemical-dependency-treatment program that he successfully completed, Ince struggled repeatedly due to his lack of cooperation, inappropriate sexual comments, repeated rule violations, and aggressive reaction to peers who tried to hold him accountable. In his current treatment program through CORE, Ince has repeatedly indicated that he does not like treatment. Dr. Marston also observed that Ince has reported difficulty relating to other members of his current treatment group. Thus, the record also supports the district court's analysis of this factor.

To summarize, we conclude that the district court's analysis of the Linehan I factors is supported by the record. Ince only challenges the third statutory element—that he is highly likely to reoffend—which is analyzed under the six Linehan I factors. Because the six Linehan I factors support a conclusion that Ince is highly likely to reoffend, the state has met its burden of proving the statutory factor by clear and convincing evidence.

Ince also makes multiple arguments about alleged errors that hinge on the district court's credibility determinations. First, he argues that the district court did not consider the importance of his alcohol abuse and Dr. Marston's discussion of Ince's ADHD. But the district court considered Ince's alcohol use and specifically noted Dr. Marston's observation regarding Ince's report that he has had 20 to 30 one-night stands while drinking, which Dr. Marston noted "begs the question of whether there are more victims." Second, Ince argues that the district court "largely ignored the dynamic factors pointing towards Ince avoiding future sex offenses," adding that Ince's testimony regarding his motivation and desire to succeed demonstrates that he has decided to correct his behavior. But as the district court specifically noted, Ince "has 7 of 9 dynamic or individual factors placing him at high risk," and Dr. Marston opined that Ince's "resolve is likely to diminish given his past record of multiple supervision failures." The district court also found that Ince has failed to comply with his ISR by failing to disclose all of his consensual sexual activities and by having sexual intercourse with a female without obtaining approval for her to visit him. Moreover, as the state argues, Ince is asking this court to reweigh the evidence. Because Ince's arguments hinge on credibility determinations by the district court, they are unpersuasive. This court defers to the district court's credibility determinations and its opportunity to weigh the evidence. In re Ramey, 648 N.W.2d 260, 269 (Minn. App. 2002), review denied (Minn. Sept. 17, 2002); Knops, 536 N.W.2d at 620.

Ince asserts that the district court ignored his history following his release from prison, specifically his entering sex-offender treatment and participating in Alcoholics Anonymous. But this argument also fails. First, Ince was ordered to enter the sex-offender treatment program. Second, good behavior under supervision is not determinative of good behavior should Ince be released from commitment. See In re Bobo, 376 N.W.2d 429, 432 (Minn. App. 1985) (reasoning that a period of non-offending is not necessarily indicative of control or good behavior when a person is in an environment that does not present an opportunity to reoffend).

Finally, Ince argues that the district court's conclusion that Ince has an "utter lack of power to control his impulses to engage in harmful sexual conduct" is unsupported by the record because Dr. Marston rejected the idea and the district court specifically stated that it was relying on Dr. Marston unless it indicated otherwise. But such a conclusion is not required. Minn. Stat. § 253B.02, subd. 18c(b), states that "it is not necessary to prove that the person has an inability to control the person's sexual impulses." Thus, the district court was not required to make such a determination to conclude that Ince is an SDP, and a lack of such a finding does not undermine the district court's conclusion.

In sum, the record supports the district court's findings and conclusions that Ince meets the criteria of an SDP. Therefore, the district court did not err in determining that the state met its burden of proving by clear and convincing evidence that Ince should be civilly committed as an SDP.

II.

Ince also argues that the district court failed to adequately address the less-restrictive alternative of CORE treatment that Ince suggested. Under Minnesota law, district courts "shall commit the patient to a secure treatment facility unless the patient establishes by clear and convincing evidence that a less-restrictive treatment program is available that is consistent with the patient's treatment needs and the requirements of public safety." Minn. Stat. § 253B.185, subd. 1(d) (2010). "[T]he burden of proving that a less-restrictive program is available is on the patient." In re Robb, 622 N.W.2d 564, 674 (Minn. App. 2001), review denied (Minn. Apr. 17, 2001). This court will not reverse a district court's findings on the propriety of a treatment program unless its findings are clearly erroneous. In re Thulin, 660 N.W.2d 140, 144 (Minn. App. 2003).

Here, all three psychologists testified that Ince needs inpatient treatment and should be committed to a structured, residential setting, specifically MSOP. Although Ince disagreed with the three psychologists' conclusions and testified that he is doing well in the CORE program, he admitted that CORE was likely under the mistaken impression that he had completed sex-offender treatment through a chemical-dependency treatment program. Ince failed to produce any evidence that CORE would accept a judicially committed sex offender. Moreover, Ince's records from CORE revealed problems with his treatment arrangement there as he was having sex with an unapproved visitor and CORE mistakenly believed he had already participated in sex-offender treatment in prison. As the district court noted, although Ince did not re-offend while on intensive supervised release, "there were several violations of his ISR contract." The district court also noted that Dr. Marston testified that Ince has no adequate relapse prevention plan and Ince has been previously terminated from chemical-dependency treatment. The district court's order demonstrates that it considered but rejected Ince's argument regarding a less-restrictive placement given Ince's history and the testimony of all three psychologists that Ince needs the structured setting that is provided by MSOP. Accordingly, Ince has not met his burden to establish that a less-restrictive treatment program is available.

Affirmed. JOHNSON, Chief Judge (concurring specially)

I concur in the opinion of the court. I believe it is proper to affirm in light of the applicable standard of review, which requires that we consider the evidence in the light most favorable to the district court's findings and judgment. See In re Knops, 536 N.W.2d 616, 620 (Minn. 1995). But I have some doubt as to whether Ince actually is "highly likely" to engage in harmful sexual conduct in the future. I write separately to make four observations concerning the difficulties that may be encountered by district courts and the court of appeals in this type of case.

A.

First, this case suggests that the six-factor Linehan test is either outdated or unduly inflexible in light of subsequent developments in the nature of the expert evidence that typically is introduced in SDP commitment cases.

The six-factor Linehan test was adopted for purposes of the SDP statute in In re Linehan, 557 N.W.2d 171, 189 (Minn. 1996) (Linehan III), vacated, 522 U.S. 1011, 118 S. Ct. 596 (1997); see also In re Commitment of Navratil, 799 N.W.2d 643, 649 (Minn. App. 2011), review denied (Minn. Aug. 24, 2011); In re Commitment of Stone, 711 N.W.2d 831, 836 (Minn. App. 2006), review denied (Minn. June 20, 2006).¹ In Linehan III, the expert witnesses' opinions about Linehan's likelihood of reoffending were based on clinical judgments, not on any statistical methods. Id. at 171, 176-77 & n.2, 180, 189. Linehan sought to convince the supreme court that the evidence was insufficient because findings of likelihood should be based on "'actuarial' methods of prediction founded on base rate recidivism statistics," which he contended were "more accurate than 'clinical' predictions." Id. at 189. But the supreme court rejected that approach on the ground that it would be "contrary to the multi-factor analysis for dangerousness prediction outlined in Linehan I." Id. The supreme court reasoned that "[s]tatistical evidence of recidivism," by which it apparently meant general base-rate statistics, is "one of the six factors." Id. But the supreme court rejected the argument that actuarial evidence is entitled to equal or greater weight. See id.

1 The six-factor Linehan test appears to be the governing law because of this court's opinions in Navratil and Stone. In Linehan, 594 N.W.2d 867 (Minn. 1999) (Linehan IV), the supreme court reiterated and adopted its analysis in Linehan III with respect to constitutional issues. Id. at 871-76. But the supreme court has not prescribed or applied the six-factor Linehan test since Linehan III, which was vacated by the United States Supreme Court, 522 U.S. 1011, 118 S. Ct. 596.

In the 17 years since Linehan III, it appears that expert witnesses in SDP cases have relied on actuarial risk-assessment tools with increasing frequency and emphasis. Actuarial risk-assessment tools use a mathematical equation that reflects the correlation between specified risk factors and sexual violence and then considers the presence of those risk factors in relation to an individual; the result is a probabilistic statement applicable to a group of sex offenders that includes the individual. See John Monahan, A Jurisprudence of Risk Assessment: Forecasting Harm Among Prisoners, Predators, and Patients, 92 Va. L. Rev. 391, 405-06 (2006); see also Linehan III, 557 N.W.2d at 177 n.2. Two commonly used actuarial risk-assessment tools, the Static-99 and the MnSOST, were mentioned in an appellate opinion in Minnesota for the first time in 2001. In re Holden, 2001 WL 683004, *4 (Minn. App. June 19, 2001), review denied (Minn. Aug. 15, 2001). Those tests (including their variations) have been mentioned in 61 subsequent Minnesota appellate opinions. It has been my experience in the past five years that actuarial risk-assessment tools quite often are used to predict the risk of reoffending in SDP cases that are appealed to this court. The literature indicates that risk-assessment methods generally have evolved from unstructured clinical judgments toward actuarial tools and/or structured professional judgment mechanisms, and the literature further indicates that the risk-assessment tools commonly used today are more accurate than those commonly used in the past and more accurate than unstructured clinical judgments alone. See Christopher Slobogin, Risk Assessment, in The Oxford Handbook of Sentencing and Corrections, 196, 198-201, 207-09 (2012); M. Neil Browne & Ronda R. Harrison-Spoerl, Putting Expert Testimony in its Epistemological Place: What Predictions of Dangerousness in Court Can Teach Us, 91 Marq. L. Rev. 1119, 1204 (2008); Monahan, supra, at 408-09; cf. Linehan III, 557 N.W.2d at 189 (noting testimony that "combining actuarial methods with the clinician's experience and knowledge of the peculiar circumstances of a given case may enhance accuracy").

These general trends may be observed in the present case. All three of the expert psychologists relied on actuarial risk-assessment tools as a significant means, if not the primary means, of predicting whether Ince will reoffend. Penny Zwecker, Ph.D., used the Rapid Risk Assessment of Sex Offender Recidivism (RRASOR) tool and the MnSOST-3.1 tool. Rosemary Linderman, Ph.D., used the Static-99R and the Static-2002 tools, and she also discussed the actuarial tools that others had administered to Ince, including the MnSOST-R, the Static-99, the Stabile 2007, and the RRASOR tools. And Peter Marston, Ph.D., used the Static-99R and the MnSOST-3.1 tools. At trial, Dr. Marston testified that actuarial risk-assessment tools generally are more accurate than unstructured clinical judgments and that the Static-99R is the only predictive tool with "consistent predictive reliability." Dr. Marston also testified that he utilized a "structured dynamic risk assessment" as a "secondary" tool "in conjunction with" the Static-99R tool because the psychologists who developed the Static-99R tool recommend that dynamic factors be considered in addition to the fixed, historical factors that are incorporated into the Static-99R tool. In his written report, Dr. Marston made clear that an overall evaluation should be "based upon the consideration and integration of information obtained from a variety of sources" and that his conclusions were based on a combination of actuarial tools and structured professional judgment tools.

If psychologists who are experienced in sex-offender risk assessments increasingly place heavy reliance on actuarial risk-assessment tools, it would seem incongruous for the caselaw to relegate those actuarial tools to the margins of the legal analysis. Linehan III appears to say that a district court must consider statistical evidence only to the extent that it consists of general base-rate statistics, which is one of the six Linehan factors, but that a district court need not consider evidence of the more specific predictions derived from actuarial tools because such evidence is not contemplated by any of the six Linehan factors. See Linehan III, 557 N.W.2d at 189. Since Linehan III, this court sometimes has followed that approach but sometimes has treated evidence derived from actuarial risk-assessment tools as the equivalent of evidence of general base-rate statistics. Compare Navratil, 799 N.W.2d at 649-50 (discussing "statistical analyses" of appellant's "likelihood of reoffending" as part of third Linehan factor), with Stone, 711 N.W.2d at 840-41 (discussing general base-rate statistics as part of third Linehan factor and discussing MnSOST-R and Static-99 results as factors outside six-factor Linehan test). In any event, the caselaw tends to limit the emphasis that may be given to actuarial risk-assessment tools.

At the least, actuarial evidence is de-emphasized because Linehan III requires a multi-factor analysis. See Navratil, 799 N.W.2d at 649; Stone, 711 N.W.2d at 840. But if a district court credits evidence derived from one or more actuarial risk-assessment tools, it would seem logical for the district court to give less weight or no weight to some of the six Linehan factors. For example, the third factor, general base-rate statistics, would be less relevant as a stand-alone factor if a base rate already is incorporated into an actuarial risk-assessment tool. Evidence of general base-rate statistics may be helpful to a district court if no other type of quantitative prediction is in the evidentiary record, but that is not true in this case. In addition, the first factor, demographic factors, would be less relevant as a stand-alone factor if the actuarial risk-assessment tool already has accounted for age. In this case, Dr. Marston scored Ince one point higher on the Static-99R tool because he is younger than 35 years of age. Furthermore, the second factor, the sex offender's history of violent behavior, also would be less relevant as a stand-alone factor if an actuarial risk-assessment tool has accounted for that factor. In this case, Dr. Marston considered Ince's history of violent behavior with the Static-99R tool, which asks whether Ince engaged in violence during his most recent offense or any prior offense and whether he had been convicted of a non-contact sex offense. In short, if a district court relies on evidence derived from actuarial risk-assessment tools, some of the Linehan factors would appear redundant. In that event, if a district court analyzes all Linehan factors, some of the Linehan factors would be double-counted.

In this case, the district court did not discuss any of the expert witnesses' use of actuarial risk-assessment tools in its findings of facts, its conclusions of law, or its accompanying memorandum. The district court did, however, apply each of the six Linehan factors. The district court's analysis is consistent with Linehan III. But it is unclear whether the district court's analysis focused on the evidence most likely to answer the key question, whether Ince is "highly likely" to engage in harmful sexual conduct in the future.

SDP commitment proceedings require findings of fact concerning uncertain future events. The uncertain future events are matters of human behavior. By their nature, these are difficult things to predict. Furthermore, a district court's factual inquiry in a SDP case concerns the behavior of a particular individual, which is more difficult to predict than the behaviors of a group of people (which is all that insurance companies attempt to predict with actuarial methods). Actuarial risk-assessment tools surely are an imperfect means of predicting a particular person's future behavior, but they appear to provide access to the best prediction presently available. Given the difficulty of the fact-finding task, the case law should allow parties as well as the district courts and this court to take advantage of the best available means of prediction. But the legal analysis prescribed by Linehan III does not do so and, thus, appears to be inconsistent with the present-day views and practices of professionals in this particular field of work and study.

B.

Second, this case illustrates the confusion that may arise from the lack of guidance concerning the type of risk that is relevant to whether a person is "highly likely" to engage in harmful sexual conduct.

In his written report, Dr. Marston repeatedly cited a research paper that explains well the distinction between relative risk and absolute risk:

Actuarial risk scales assess two facets of risk: relative and absolute. Relative risk provides information about a particular offender's level of risk relative to other offenders and can be reported in numerous ways, including risk assessment scores (e.g., "this offender is a 3 on Static-99"), nominal risk categories (e.g., "this offender is high risk"), percentiles (e.g., "95% of offenders score higher than this individual"), or relative risk ratios (e.g., "the risk of recidivism for this offender is about ½ the risk of a typical sex offender").
. . . .
Absolute risk, however, refers to the expected probability of recidivism. Although relative risk information is sufficient for most decisions involving the allocation of scarce resources (i.e., treatment and supervision decisions), absolute risk information is required in certain high-stakes evaluations, notably sex offender civil commitment statutes in the United States.
. . . . [I]nterpreting the phrase "moderate risk sex offender" requires knowledge of the risk posed by all sex offenders. If the base rate for sexual recidivism is 70%, then the release of a "moderate" risk sex offender invokes concern. If the base rate is 5%, however, then the release of a "moderate" risk sex offender is much less problematic.

Leslie Helmus, Re-Norming Static-99 Recidivism Estimates: Exploring Base Rate Variability Across Sex Offender Samples (Sept. 2009) (unpublished M.A. thesis, Carleton University) at 7-8 (2009), available at www.static99.org/pdfdocs/helmus2009-09static-99normsmathesis.pdf.

In his written report and his oral testimony, Dr. Marston described Ince's likelihood of reoffending in terms of both relative risk and absolute risk. With respect to the relative risk, Dr. Marston wrote that Ince's score of +5 on the Static-99R "places him into the Moderate High category of risk compared to a general population of convicted sex offenders." Dr. Marston also wrote that, given his score of +5 on the Static-99R, Ince is within "the 81.4 to the 89.7 percentile of risk for sex offenders." Dr. Marston testified at trial that, given the revised score of +6 on the Static-99R, Ince's risk of recidivism is approximately 2.91 times the recidivism rate of the typical sex offender.

But Dr. Marston also wrote in his report that "[a]bsolute degrees of recidivism risk cannot be directly inferred from percentile rankings." Dr. Marston wrote that a person's absolute risk of recidivism, when determined by actuarial tools, depends in part on the base rate of recidivism of a group of comparable sex offenders. Dr. Marston chose to compare Ince to six samples of sex offenders who had a high risk of recidivism and a high level of treatment needs. Accordingly, Dr. Marston testified at trial that Ince's absolute risk of reoffending, determined by the Static-99R tool and the chosen samples, is 31.2% over a five-year period, 41.9% over a ten-year period, and 62.4% over his lifetime.

Despite the fact that Dr. Marston discussed both relative risk and absolute risk, the district court's order and memorandum did not mention any evidence concerning absolute risk. The district court's findings of fact recited only relative-risk statements made by the department of corrections during Ince's imprisonment:

The [DOC] psychologist noted that Ince scored a +8 on the Minnesota Sex Offender Screening Tool-Revised (MnSOST-R), reflecting a high risk for re-arrest for a sex offense. Ince scored a +3 on the Rapid Risk Assessment for Sex Offense Recidivism (RRASOR), reflecting a moderate risk for sex offense reconviction. Ince also scored +5 on the Static-99, reflecting a moderate-high risk for sex offense reconviction.

(Emphasis added.) These descriptions of Ince's relative risk of re-offending are incapable of answering the key question in this case: whether Ince is "highly likely" to reoffend. See Linehan III, 557 N.W.2d at 180. That question does not ask for a comparison of the probability that Ince will reoffend and the probability that another person will reoffend. Rather, that question asks for a comparison of the probability that Ince will reoffend and the probability that he will not reoffend.

C.

Third, this case raises an issue concerning the selection of the most appropriate actuarial risk-assessment tool and the most appropriate group of comparable sex offenders from which to obtain a general base rate of recidivism.

Dr. Marston wrote in his report that "it is ideal to use local norms that are applicable to the group of offenders to which this offender most closely resembles." A local norm apparently was not available to Dr. Marston at the time he issued his report, in April 2012. Dr. Marston relied on studies whose sample populations included sex offenders from throughout the United States, Canada, and several countries in western Europe. But shortly before trial, Dr. Marston used the MnSOST-3.1 actuarial tool, which recently had been released by the Minnesota Department of Corrections, using recent data concerning Minnesota sex offenders. Using the MnSOST-3.1 tool, Dr. Marston calculated Ince's risk of reconviction to be only 7.12% over a four-year period. Dr. Marston's results with the MnSOST-3.1 tool were similar to those of Dr. Zwecker, who calculated Ince's risk to be 7.92%.

The evidence derived from the MnSOST-3.1 tool is notable for at least three reasons. First, consistent with Dr. Marston's expert opinion, the MnSOST-3.1 tool may be a better tool for determining the probability that a Minnesota sex offender will reoffend because it is based on data that concerns other Minnesota sex offenders and more recent data. Second, it seems that evidence derived from an actuarial risk-assessment tool developed by the State of Minnesota might be given some form of special status in a case in which the Minnesota courts are attempting to determine the risk of reoffending attributable to a Minnesota sex offender whose freedom is being restrained by the State of Minnesota, regardless whether the tool was developed by the Department of Corrections or the Department of Human Services. Third, the MnSOST-3.1 tool indicates that Ince's risk of reoffending is considerably lower than is indicated by the Static-99R tool. This appears to be due, at least in part, to a lower base rate of recidivism for the population sampled by the MnSOST-3.1 tool. Dr. Marston testified that the base rate of recidivism in Minnesota has decreased in recent years because of programs such as intensive supervised release and sex-offender counseling. But he also noted that the most recent offender sample does not include persons who were civilly committed after their release from prison, and he further noted that a significant number of sex offenders in the sample were returned to prison because they violated the terms of their supervised release in non-sexual ways.

The district court did not mention the evidence derived from the MnSOST-3.1 tool in its findings of fact or in its memorandum, let alone attempt to reconcile that evidence with other evidence in the record. In future cases, it seems natural to expect that the most recent data concerning recidivism in Minnesota and the MnSOST-3.1 tool will be explored further by expert witnesses and district courts.

D.

Fourth, this case highlights the lack of clarity arising from the existing formulation of the legal standard for reoffending in a SDP case: that the person is "highly likely" to engage in harmful sexual acts in the future. Linehan III, 557 N.W.2d at 180.

The term "highly likely" is inherently imprecise. This court has issued only two published opinions applying that standard, but in neither case did we refine the meaning of the term "highly likely." See Navratil, 799 N.W.2d at 649-50; Stone, 711 N.W.2d at 840-41. As a consequence, an expert witness has no readily available benchmark by which to determine whether the risk of reoffending in a particular case satisfies the legal standard. For example, in one of this court's recent unpublished cases, two expert witnesses had different understandings of the "highly likely" standard. One expert witness testified at trial that the law requires "[b]asically a better than even chance," while the other expert witness testified at trial that "it's higher than more likely than not but there's no specific number attached to it." In re Commitment of Iverson, 2013 WL 402206, *3 (Minn. App. Feb. 4, 2013). Similarly, a district court presiding over a SDP case lacks specific guidance concerning the degree of likelihood that is necessary to justify a finding that a person is "highly likely" to reoffend. Likewise, a reviewing appellate court may have difficulty identifying the applicable legal standard. See Eric S. Janus & Robert A. Prentky, Forensic Use of Actuarial Risk Assessment with Sex Offenders: Accuracy, Admissibility and Accountability, 40 Am. Crim. L. Rev. 1443, 1493-94 (2003) (noting difficulty in articulating and applying legal standards for risk determinations).

The lack of a clear and definite legal standard is in tension with fundamental notions of the rule of law. A statute that may deprive a person of his or her liberty should have "an understandable meaning with legal standards that courts must enforce." Giaccio v. Pennsylvania, 382 U.S. 399, 403, 86 S. Ct. 518, 521 (1966). A law should not be "so vague and standardless that it leaves . . . judges . . . free to decide, without any legally fixed standards, what is prohibited and what is not in each particular case." Id. at 402-03, 86 S. Ct. at 520-21.

Although Linehan III is imprecise, it does contain a few clues as to the meaning of "highly likely." First, Linehan III established that likelihood may be expressed in numerical terms and that the numerical measure of "highly likely" is greater than 50.1%. The supreme court stated that the SDP act "implies that committing courts cannot combine a factual element that requires only 50.1% probability with an evidentiary standard of less-than-certainty." 557 N.W.2d at 180. The supreme court added, "We do not believe that the legislature intended to weaken the standard of likelihood in the SDP Act by combination with a relatively high burden of persuasion - the clear and convincing evidence standard." Id. In fact, the state argued in that case that the requirements of the SDP act may be satisfied merely with proof that re-offending is "more likely than not," but the supreme court rejected that argument. Id.

Second, Linehan III established that the "highly likely" legal standard should be high enough to ensure that risk determinations are more often correct than incorrect. The supreme court noted that "due process concerns under the state and federal constitutions constrain legislative discretion to set standards of likelihood when liberty is at stake." Id. More specifically, the supreme court stated that a person who may be civilly committed under the SDP act "'should not be asked to share equally with society the risk of error.'" Id. (quoting Addington v. Texas, 441 U.S. 418, 427, 99 S. Ct. 1804, 1810 (1979)). The supreme court explained the problem by stating, "If the state were to require only a 10% probability of dangerousness (the fact to be demonstrated) and a clear and convincing evidence standard (say, a 75% degree of certainty), then the demand of due process that the citizen not share equally the risk of error would be undermined." Id. In a SDP case, "the error that due process seeks to avoid is a false prediction of future harmful conduct." Id. The supreme court noted (but did not necessarily endorse) expert testimony in that case to the effect that if base recidivism rates are low (such as 8%), and if predictors of recidivism are accurate in 75% of the cases in which a prediction is made, 79% of predictions would be erroneous. Id. at 177 n.2.²

2 Since Linehan III, risk assessments apparently have become more accurate, though they still may not meet the aspirations of Linehan III. See Slobogin, supra, at 200 (noting early studies showing false positives of "well over 50 percent" and more recent studies showing false positives of between 15 and 50 percent); Monahan, supra, at 406 (reviewing research "indicat[ing] that the predictive validity of actuarial instruments has significantly improved in the past twenty years"). With actuarial risk-assessment tools, social scientists may be able to allocate the "risk of error" of "a false prediction of future harmful conduct," as described in Linehan III, by adjusting variables such as the degree of likelihood to be proved, the clear-and-convincing evidentiary standard, and the statistical confidence interval. See Slobogin, supra, at 201 (noting that modern actuarial tools may be set to achieve targeted false positive and false negative rates). These are empirical matters that were not addressed by the expert witnesses in this case or by the parties' appellate briefs.

Third, linguistic considerations indicate that "highly likely" requires a probability that is significantly greater than 50.1%. The word "high" means "[h]aving a relatively great elevation; extending far upward," "[e]xtending a specified distance upward," and "[b]eing at or near the peak or culminating stage." The American Heritage College Dictionary 640 (3d. ed. 1997). The antonym of "high" is "low." Id. at 803. The boundary between the high part and the low part of the "likely" range is approximately halfway between 50.1% likelihood and 100% certainty, or approximately 75%.

Clarifying the meaning of "highly likely" is unavoidably a matter of line-drawing. In most situations, courts draw lines simply by recognizing them after numerous case-by-case determinations, similar to the way one can see shapes in a Pointillist painting. Seventeen years after Linehan III, however, the precedential case-by-case determinations on this issue are too few in number and too general in their discussion. See Navratil, 799 N.W.2d at 649-50; Stone, 711 N.W.2d at 840-41. Sometimes "[i]t is impractical to leave the answer to [a] question for clarification in future case-by-case adjudications . . . ." Maryland v. Shatzer, 559 U.S. 98, _, 130 S. Ct. 1213, 1222 (2010).

In SDP cases, it is practical, if not essential, for both a district court and an error-correcting appellate court to refer to a bright line when determining whether the evidence in a particular case is sufficient to satisfy the applicable legal standard. It would be preferable for the supreme court to establish a benchmark as a general rule of law or for an expert witness to suggest a standard for a particular case based on the supreme court's previously expressed tolerance for erroneous predictions. Absent further guidance, I would be inclined to interpret Linehan III's "highly likely" standard to require a petitioner to prove, by clear and convincing evidence, that the probability that a person will engage in harmful sexual conduct in the future is 75% or higher.

E.

The district court made the following finding of fact: "There is clear and convincing evidence that . . . it is highly likely that Ince will engage in further harmful sexual conduct . . . ." The evidence in the record is sufficient to justify the district court's finding. In my view, an analysis of the six Linehan factors leads to mixed results. But Dr. Marston testified that an actuarial risk-assessment tool, the Static-99R, indicates that the probability that Ince will engage in harmful sexual conduct is 31.2% over a five-year period, 41.9% over a ten-year period, and 62.4% over his lifetime.³ Dr. Marston also testified that Ince's risk is elevated above the Static-99R results because of the presence of seven of nine "dynamic risk factors" and because of his relatively high score on the Hare PCL-R test, which indicated that he is borderline psychopathic, which suggests a greater disposition toward sexual violence. The district court expressly stated that Dr. Marston's testimony was "particularly persuasive and convincing." The totality of Dr. Marston's expert evidence is sufficient to support an inference that the probability that Ince will engage in harmful sexual conduct in the future is 75% or higher. See Navratil, 799 N.W.2d at 649 (affirming SDP civil commitment in part based on evidence that appellant's likelihood of reoffending was 80% over ten years); cf. Stone, 711 N.W.2d at 840-41 (affirming SDP civil commitment in part based on evidence that appellant's likelihood of reoffending was 59% over six years). Thus, I join in the decision to affirm the district court's civil commitment order. STAUBER, Judge (dissenting)

3 I assume without deciding, for purposes of this case, that it is proper to consider the likelihood of reoffending over the course of a lifetime, rather than a shorter period of time. I also assume without deciding that Dr. Marston's method of determining a lifetime likelihood is reliable. Dr. Marston testified that he determined Ince's lifetime likelihood by using an extrapolation "rule of thumb" that simply doubles the five-year likelihood. He testified that the Static-99R tool had been validated, but he did not testify that the "rule of thumb" had been validated. See Robert A. Prentky et al., Sexually Violent Predators in the Courtroom, 12 Psychol. Pub. Pol'y & Law 357, 379 (2006) (criticizing extension of risk estimates beyond follow-up periods as "untestable" and based on "possibly erroneous assumption"). The admissibility of this part of Dr. Marston's testimony was not challenged.

I respectfully dissent. I would reverse the district court's order indefinitely committing appellant Cedric Scott Ince to the Minnesota Sex Offender Program (MSOP) and would hold that the district court's determination that Ince is a Sexually Dangerous Person (SDP) is clearly erroneous.

Background

Ince is a 23-year-old man who has been incarcerated since the age of 19, with the exception of eight months of successful intensive supervised release (ISR) in the community prior to his civil commitment trial. Ince's incarceration resulted from his rape of a 19-year-old woman when he was 18, and his subsequent conviction of third-degree criminal sexual conduct. Ince had an earlier juvenile adjudication for fifth-degree criminal sexual conduct at age 17 involving a 17-year-old girl. Both parties agree that Ince's substance abuse played a factor in these crimes. The district court's determination is purportedly based only upon Ince's two sexual offenses.

The question posed by this appeal is whether Ince is highly likely to reoffend by engaging in future acts of harmful sexual conduct. See Minn. Stat. § 253B.02, subd. 18c(a) (2010); In re Linehan, 557 N.W.2d 171, 180 (Minn. 1996) (Linehan III).In attempting to determine whether Ince is highly likely to engage in future acts of harmful sexual conduct, the district court reviewed the evidence in the record under the specific criteria set forth in In re Linehan, 518 N.W.2d 609, 614 (Minn. 1994) (Linehan I ). The district court determined that Ince was highly likely to engage in future acts of harmful sexual conduct and therefore qualified as a sexually dangerous person subject to civil commitment. The district court's conclusions with regard to these criteria are unsupported by the evidence in the record.

Linehan Factors

Customarily, several experts are employed to provide evaluation and testimony at civil commitment trials. Here, Dr. Penny Zwecker, Dr. Rosemary Linderman, and Dr. Peter Marston each submitted reports. The court specifically "found the testimony of Dr. Peter Marston particularly persuasive and convincing." The court used this testimony in evaluating Ince under the Linehan factors. These factors include: (1) the person's demographic characteristics, including age, education, etc.; (2) the person's history of violent behavior; (3) the base rate statistics for violent behavior among individuals of this person's background; (4) the sources of stress in the environment; (5) the similarity of the present or future context to those contexts in which the person has used violence in the past; and (6) the person's record with respect to sex therapy programs. Linehan I, 518 N.W.2d at 614 (determining whether an allegedly SDP is highly likely to engage in future acts of harmful sexual conduct).

The concurrence notes that the Linehan factors may be "outdated or unduly inflexible in light of subsequent developments in the nature of the expert evidence that typically is introduced in SDP commitment cases." The concurrence proposes that because of the "heavy reliance on actuarial risk-assessment tools, it would seem incongruous for the caselaw to relegate those actuarial tools to the margins of the legal analysis." While the concurrence may in fact be accurate, for the purpose of this dissent, it seems incongruous to ignore the law established by the Minnesota Supreme Court, which outlines important dynamic factors affecting defendants, in favor of following a practice. Actuarial scores alone are not determinative of whether a person is highly likely to reoffend. In re Commitment of Navratil, 799 N.W.2d 643, 649 (Minn. App. 2011). I therefore employ the Linehan factors to determine whether the district court's conclusions were correct.

Demographic Characteristics

The district court determined that Ince's age, gender, history of violent behavior, and mental-health diagnosis indicate that Ince has an increased risk of re-offense. This finding is based primarily upon Dr. Marston's testimony at trial that Ince's age, aggressive behaviors, diagnosis, and gender all place him at "an increased risk."

Ince is a 23-year-old man who has been incarcerated since the age of 19, with the exception of eight months in ISR prior to the conclusion of his civil commitment trial. He has battled substance abuse since adolescence. His first clinical diagnosis of substance abuse was at the age of 17. Dr. Marston conceded at trial that Ince's substance abuse has played a significant factor in his mental illness and in his violent behavior. Dr. Marston testified that Ince had "a poor behavioral history from day one, but . . . it got quite a bit worse when he began using chemicals and alcohol."

Ince also suffers from various forms of mental illness. Dr. Marston testified that Ince has been diagnosed with ADHD, antisocial personality disorder, and psychopathic disorder. Dr. Marston's written report explains that because Ince is male, not married, and suffers from ADHD, his likelihood of reoffending is higher. How this makes Ince any different from the vast majority of the sex offender population is entirely unclear.

The majority highlights Dr. Marston's testimony that Ince's "antisocial personality disorder and psychopathic disorder are at their peak at his age," indicating that his likelihood for re-offense is higher as a result. However, the majority ignores Dr. Marston's acknowledgment that Ince's mental illness and aggressive behaviors are closely correlated to his substance abuse. Dr. Marston testified that "if we take [Ince's substance abuse] away, my question then becomes - so if we take that away, then are we back to a person who has just ADHD and isn't really a psychopathic person." In other words, remove the substance-abuse problems, and the mental illness and aggressive behavior may diminish as well. In fact, Ince has removed the substance-abuse problems. He has been sober since October 2008.

At its best, Dr. Marston's testimony on this issue is ambivalent and inconsistent. Ince is a 23-year-old man who has been incarcerated since the age of 19. His two relevant offenses were committed at the ages of 17 and 18. Dr. Marston's ambivalence fails to demonstrate that Ince is "highly likely" to reoffend.

History of Violent Behavior

The district court determined that Ince's prior criminal offenses were "recent, serious, and extremely violent" and indicated a lack of self-control. This is the entirety of the district court's conclusions on the second Linehan factor.

The record supports the district court's finding that Ince's prior crimes were recent and serious. The record also supports the finding that Ince's most recent crime was "extremely violent." But the record does not support that Ince's first crime was "extremely violent." As to the lack-of-control factor, as the majority notes, "it is not necessary to prove that the person has an inability to control the person's sexual impulses." Minn. Stat. § 253B.02, subd. 18c(b) (2010). Yet the majority cites Dr. Marston's testimony regarding the lack-of-control factor - the fact that Ince raped his second victim three weeks after being placed on probation for raping his first victim. In other words, Ince raped his second victim approximately one year and eight months after raping his first victim. Dr. Marston clearly testified that this fact supports the lack-of-control factor, which is unnecessary, according to section 253B.02, subdivision 18c(b).

If the lack-of-control analysis is removed from the equation, we are left with the district court's findings that Ince's crimes were recent, serious, and (one was) extremely violent. Ince was adjudicated as a juvenile for the fifth-degree criminal sexual conduct offense, and as an adult, convicted for the third-degree criminal sexual conduct offense. The thought that a defendant may suffer the consequence of MSOP incarceration for the balance of his life based substantially upon a fifth-degree criminal sexual conduct crime he committed as a juvenile is a precedent that I feel uncomfortable setting. Further, after his first offense, Ince was put on probation, but he was not provided the intensive programming and tools that have since been made available to him during his incarceration and ISR, and which would continue to be made available to him were he to continue with ISR. It is also noteworthy that following his conviction, the district court granted a downward sentencing departure.

Ince does not have the history of violent behavior that is required for an SDP civil commitment. He is a young man who, as a teenager, committed two qualifying crimes. The district court's findings on this factor are scant, and Dr. Marston's testimony does not support a conclusion that Ince is "highly likely" to reoffend.

Base Rate Statistics

The district court noted Dr. Marston's testimony that "the base rate statistics indicate Ince is at a high risk compared to the typical sex offender in the community." Dr. Marston further testified that Ince's number of "one night stands" indicates that there could be more victims than have thus far been discovered. This testimony is clearly speculative, without evidentiary basis, and insufficient for the third Linehan factor. The factor, in its entirety, states: "the base rate statistics for violent behavior among individuals of this person's background." Linehan III, 518 N.W.2d at 614. These statistics are "data showing the rate at which rapists recidivate, the correlation between age and criminal sexual activity, etc." Id. The district court merely makes a conclusory statement with regard to this factor, without evidentiary basis, concluding that Dr. Marston testified that this factor was satisfied.

Dr. Marston's statement in his written report that Ince's behavior is associated with a lifetime recidivism rate of 39 percent is based upon one particular study published in 1998. Dr. Marston concludes in his testimony that this is "a high risk" in comparison with sample populations. Dr. Marston also testified that Ince was three times more likely to recidivate than the average sex offender. However, Dr. Marston further testified that because Ince had not reoffended during his eight months of ISR, the likelihood that he ever would reoffend would be lower, because the first year following release is the most crucial in predictive determinations.

Dr. Marston relied upon the results of actuarial assessment tools administered to Ince. The idea that an actuarial test can predict how a person will respond to any number of influences, situations, events, temptations, or obstacles with any level of accuracy is fundamentally unsound. Dr. Marston himself even conceded during testimony that evaluating human behavior is "difficult to quantify." And when asked at trial whether he could "predict with any statistical reliability who will and will not succeed," Dr. Marston responded "I wish I could." Dr. Marston further testified that "[p]art of the problem with our risk assessment is that we don't - you know, when we're trying to calibrate this kind of thing, I think we're out of our depth."

This is exactly the point. These issues are indeed difficult to quantify, and employing a numeric actuarial assessment tool is sorely deficient when dealing with human subjects and the myriad components that constitute a human life. The concurrence proposes the use of actuarial assessment tools to the exclusion of the other Linehan factors because of their popularity with psychiatrists, who believe the tools nicely blend historical facts with professional judgment. But the question becomes, doesn't an actuarial assessment inherently involve judgment and intuition? An actuarial assessment is a test administered by psychiatrists and psychologists that assigns numeric values to aspects of a human life, like a person's criminal history, his mental disorders, and his answers to questions posed in evaluating him. These actuarial assessments are subjective assessments, even though they use numeric values, and they rely heavily upon psychiatrists' judgments and intuition in an attempt to objectify a clearly subjective process.

At a civil commitment trial, the district court then reviews these "judgments" as presented through expert testimony. The district court makes credibility assessments of these expert witnesses, and these assessments are almost wholly protected from appellate review based on our standard of review.

The concurrence offers that the "predictive validity of actuarial instruments has significantly improved in the past twenty years." If accurate, this is good news, but it still means that we as a society are civilly committing hundreds of people - for life - who may never have reoffended had they been released. Actuarial assessment tools simply are not as accurate as they need to be in predicting who needs to be locked up forever in order to protect society. As the concurrence highlights, nowhere in this nebulous area can we get a truly accurate prediction.

Even were I to believe that these actuarial assessment tools were adequate support for committing people to MSOP, the test results in Ince's case are far from unequivocal. Dr. Marston's initial written report on Ince indicated that he scored Ince at a +5 on the Static 99-R test. However, Dr. Marston testified that in reviewing this rating the evening prior to his testimony, he altered his score to a +6, based solely upon his assessment of an incident where Ince simulated masturbating on a school bus in front of other students when he was 14 years old, an incident that the district court stated it was not considering. This last-minute modification is extremely important, if not determinative, in the outcome for Ince, because a score of +5 puts him in the "moderate-to-high" range of likelihood for reoffending - generally not committable - whereas a score of +6 puts him squarely within the "high" range for reoffending, thus more likely committable.

The majority highlights testimony that appears to indicate that Dr. Marston's +6 score exists independent of the bus masturbation incident and squarely places Ince in the "high risk" category. Specifically, the majority provides that "when asked about the impact of the increased point, Dr. Marston testified that changing the score of the Static-99R did not ultimately affect his overall opinion." The majority then quotes Dr. Marston's testimony: "It simply increases his score past the threshold of 5, which is moderately high category of risk to high risk. So it nudges his score into that category. And I had placed him in that category already." Dr. Marston's reference to "that category" is clearly the "moderately high category of risk to high risk," not the purely "high risk" category. The majority further ignores Dr. Marston's testimony that he changed his rating in order to add "an additional point, with reference to the 2004 bus incident." And when asked whether his rating "was changed today because of the 2004 incident," Dr. Marston responded, "[c]orrect."

The district court may have indicated that it gave "some credibility" to the bus masturbation incident, but it also specifically stated that it was not relying upon the incident in its analysis, and that it was only relying upon the two sex offenses to evaluate Ince. Consequently, the district court's finding that Ince is at a high risk of reoffending is not supported by the evidence in the record.⁴ The most accurate evidence clearly demonstrates that Ince has a moderate-to-high risk of reoffending, a far cry from establishing that Ince is highly likely to engage in future harmful sexual conduct.

4 In further support of a conclusion that Ince has nothing more than a moderate-to-high risk of re-offense, the district court found that:

On October 14, 2009, Duncan Anderson, Psy.D., filed a psychosexual evaluation report with the Scott County District Court, described above. Ince's score on the Static-99 placed him in the moderate-high risk pool for sex offense re-conviction. Ince's score on the Stable-2007 placed Ince in the moderate-high risk for sexual re-offense. Sex offender treatment was recommended.

We need to also discuss sex-offender recidivism rates in general, because it appears that these rates may be far lower than what we have been led to believe. "[S]tudies with the strongest methodology show that the recidivism rate for sex offenders is as low, and often lower, than re-offense rates for criminals convicted of non-sexual crimes." Eric S. Janus & Emily A. Polachek, A Crooked Picture: Re-Framing the Problem of Child Sexual Abuse, 36 Wm. Mitchell L. Rev. 142, 162 (2009). In 1994, the U.S. Department of Justice's Office of Justice Programs conducted a study that followed 9,691 sex offenders who had been released from prison across the United States. In the three years following their release, 5.3 percent had recidivated. Matthew R. Durose, et al., U.S. Dep't of Justice, Bureau of Justice Statistics, Recidivism of Sex Offenders Released from Prison in 1994 (2003), available at http://bjs.ojp.usdoj.gov/index.cfm?ty=pbdetail&iid=1136. And, if Dr. Marston's testimony is accurate, and the first year is the most crucial to predicting whether a sex offender will reoffend, then this is an accurate statistic.

Eric Janus discusses a far more compelling study in his 2003 law review article. See Eric S. Janus, Minnesota's Sex Offender Commitment Program: Would an Empirically Based Prevention Policy Be More Effective?, 29 Wm. Mitchell L. Rev. 1083, 1094 (2003). He cites a Minnesota Department of Corrections study that followed a group of individuals who were referred for commitment but were not committed for a period of 6.5 years. See id. One would think that this would be one of the most similarly situated groups of people to those who have been civilly committed, because they are among the individuals the DOC finds most likely to reoffend. The DOC found that in 6.5 years, 18% of that group recidivated. Id. Recidivism rates are skewed in general, thus any reliance upon them in this case is purely speculative.

Sources of Stress

The district court found that since Ince's release from prison, Ince has received "support in the community and has attended [Alcoholics Anonymous], attended sex offender treatment, maintained sobriety, and obtained employment." This finding is bolstered by the fact that during Ince's eight months of ISR prior to the conclusion of his civil commitment trial, he appeared to have arranged a relatively stable lifestyle. He obtained employment, leased a home, purchased a vehicle, and remained sober during those eight months. He now has a strong relationship with his father along with other significant familial support. His grandparents are preparing to move into his residence with him, providing even more stability and support. But most notably, Ince did not reoffend during his eight months of ISR. He may have had a minor violation of his ISR contract, but that is not the issue here - the issue is whether Ince has removed the sources of stress from his environment that indicate that he is highly likely to reoffend.

The majority points to Ince's antisocial personality disorder in this analysis. An antisocial personality disorder should not be considered a source of stress. The full factor for analysis is "sources of stress in the environment," Linehan I, 518 N.W.2d at 614 (emphasis added), and has been interpreted as "sources of stress in the offender's environment," In re Stone, 711 N.W.2d 831, 840 (Minn. App. 2006). An antisocial personality disorder is not an external influence in a person's environment. An antisocial personality disorder should clearly play into a discussion of a person's demographic characteristics or base rate statistical analysis, but not in an analysis of sources of environmental stress.

Ince has removed the sources of environmental stress that prompted him to commit the prior sexual offenses. Any other finding on this factor is unsupported by the record.

Similarity of Future Context to Past Context Containing Violence

The district court noted that Dr. Marston testified that Ince "is not in exactly the same setting he was in" before he committed his offenses. The district court further noted Dr. Marston's testimony that as supervision of Ince disappears, Ince's "disorders [will] take over," and "he will be at high risk to sexually reoffend."

The district court is again relying upon speculative future predictions given by Dr. Marston instead of focusing on the similarities or differences of Ince's current and future context to his past context. In focusing on the comparison of Ince's current/future context with his past context, we are left only with what the district court finds on this specific matter, which is simply that Ince "is not in exactly the same setting."

This statement is clearly correct. Ince has attained and maintained sobriety, he has obtained employment, purchased a vehicle, leased a home, and completed programs to help him overcome his substance abuse and deal with his sexually violent criminal history. The programs, along with his age, maturity, and family support have provided him with tools he did not have at age 18, prior to incarceration. Ince has remained sober since his arrest in October 2008. Dr. Marston conceded that Ince's associations with friends must necessarily be different in the future because of Ince's parole contract. Ince may have struggled with honesty in his treatment, but this is hardly a consideration in determining whether he is highly likely to reoffend. Dishonesty is a human failing, not a failing specific only to sex offenders.

The majority highlights Dr. Marston's testimony that Ince has a history of "'beating the system' and this type of behavior will reassert itself without a secure setting." This begs the question of whether a "secure setting" need necessarily be MSOP. More importantly, Ince's proclivity for "beating the system" is irrelevant to the specific issue of whether Ince's past/future contexts are different or similar. The record shows that Ince's future context will be starkly different from his past context. Any other finding is not supported by the record.

Sex Therapy Programs

The district court found only that "while Ince is engaged in treatment, CORE had operated under the wrong assumption that Ince had participated in treatment before." The district court further found that Ince had no valid Relapse Prevention Plan and that he had been terminated from chemical-dependency treatment. The issue is "participation in sex therapy programs," not chemical dependency, nor relapse prevention plans, nor incorrect assumptions promulgated by Ince. As to that issue, only the district court's first finding (that Ince "is engaged in treatment") is relevant. The record shows that Ince has participated in sex therapy programs since his incarceration. Ince's ISR required him to undergo sex-offender treatment. He did so in January 2012, entering CORE. Ince successfully completed two parts of CORE prior to the conclusion of his trial.

The majority highlights Ince's struggles with chemical-dependency treatment, wherein he struggled due to lack of cooperation. These facts are irrelevant to the issue of Ince's participation in sex therapy programs. We are left with no fact other than that Ince completed two parts of a sex therapy program. Consequently, the district court's finding is unsupported by the record.

"The key to civil commitment, and the argument that makes it 'constitutional,' is the assumption that the dangerous sexual offender can be distinguished from the typical repeat offender." 799 N.W.2d at 652-53 (citing In re Civil Commitment of Ramey, 648 N.W.2d 260, 266 (Minn. App. 2002)), review denied (Minn. Aug. 24, 2011). Nothing in the record supports a conclusion that Ince has been anything more than a typical offender. His crimes were sexual in nature, but civil commitment is not a substitute for the criminal justice system, which is the system that has been vested with punitive powers over all crimes, including sex crimes. Civil commitment should be reserved for people like Dennis Linehan, who committed numerous sexual assaults, pleaded guilty to kidnapping in a case in which he was suspected of first-degree murder, and sexually assaulted a 12-year-old girl shortly after escaping from prison. Linehan I, 518 N.W.2d at 611. Civil commitment should be reserved for the unequivocal case. This is not that case.

Highly Likely Analysis

The concurrence makes an important point when it states that the term "highly likely" to reoffend, as used in the civil commitment process, has never been adequately defined. The concurrence researches possible interpretations of "highly likely" and concludes that "highly likely" should mean a defendant who has at least a 75% chance of reoffending. It is entirely appropriate to interpret the term "highly likely" with the understanding that 678 of the 679 individuals committed to MSOP have yet to be released from "treatment." This understanding comports with the requirement that defendants "should not be asked to share equally with society the risk of error." Linehan III, 557 N.W.2d at 180 (quoting Addington v. Texas, 441 U.S. 418, 427, 99 S. Ct. 1804, 1810 (1979)). "Highly likely" should therefore be interpreted to require a finding that a person is so likely to reoffend that he should be locked away in MSOP for the rest of his life.

District Court's Credibility Determinations

The district court concluded that the testimony of Dr. Marston was most credible, and relied upon it to the exception of other testimony and evidence. The district court found Dr. Marston's testimony to be more credible than the testimony of the other two psychiatrists who testified at the trial, including the court-appointed psychiatrist. The reasons for this are unclear.

Minnesota Department of Corrections' Independent Legal Counsel

The fact that this case even made it to the district court for a commitment hearing is also interesting and strange. The SDP/SPP Screening Committee charged with determining whether to forward the matter to the county attorney to pursue a petition for civil commitment initially determined that Ince's case should not be forwarded to the county attorney. The case was forwarded for commitment only after the Minnesota Department of Corrections' Independent Legal Counsel, who was not a part of the referral process, "trumped" the recommendation of the screening committee.

The record reflects that counsel "participated in the [screening committee] meeting" and reviewed the screening committee's report. Counsel then recommended forwarding Ince's case to the county attorney despite the committee's decision otherwise. Counsel's recommendation acted to override the report of the committee charged with independently evaluating Ince's case.

Counsel's recommendation is supported by a brief memorandum, wherein he languidly analyzes the six Linehan factors, most of which clearly favor Ince, but summarily concludes that Ince's case should be referred to the county attorney. The Department of Corrections has an established practice for determining whether to forward such cases to the county attorney for prosecution under the civil commitment laws, but the Department of Corrections' counsel's involvement clearly subverted and prejudiced this established practice. I find this disturbing.

Conclusion

Given that 678 of the 679 people who have been civilly committed to MSOP remain in MSOP, we must approach these cases with an appreciation for their gravity and an understanding that every case must be scrutinized to ensure absolute compliance with the law. A civil commitment to MSOP is, essentially, a life sentence. Only those people who are truly "highly likely to reoffend" should be committed to the program. Clearly, the courts have failed to follow the "highly likely to reoffend requirement," because, since 2003, the number of sex offenders who have been civilly committed to MSOP has increased by 357%. In 2003, MSOP had 190 "clients." Eric S. Janus, Minnesota's Sex Offender Commitment Program: Would an Empirically-Based Prevention Policy Be More Effective?, 29 Wm. Mitchell L. Rev. 1083, 1089 (2003). Now, MSOP houses 678. This is an unsustainable system.

Yet this system will continue unabashedly so long as district courts are allowed to rely solely upon actuarial assessment tests to attempt to predict whether an individual will reoffend. District courts will rely upon these tests, as communicated to the court through expert testimony, and will make credibility assessments insulated from appellate review. Given the standard of review, it will continue to be almost impossible to overturn even borderline cases such as Ince's case.

"As the guardians of our constitutional order, the courts have promised that they will 'intervene before civil commitment becomes the norm and criminal prosecution the exception.'" Eric S. Janus, Failure to Protect: America's Sexual Predator Laws and the Rise of the Preventive State, at 26 (2006) (citing Linehan III, 557 N.W.2d at 181). We break this promise by affirming a decision to civilly commit a 23-year-old man for two acts of sexual violence committed as a teenager, before he had even fully developed emotionally and intellectually. As the dissent stated, "To affirm the order for indeterminate commitment under these circumstances is to approve the warehousing of another borderline candidate for treatment, with no guarantee that he will receive meaningful treatment." Navratil, 799 N.W.2d at 653. I could not agree more.

And the district court further found that: The psychologist noted that Ince scored +8 on the Minnesota Sex Offender Screening Tool-Revised (MnSOST-R), reflecting a high risk for re-arrest for a sex offense. Ince scored a +3 on the Rapid Risk Assessment for Sex Offense Recidivism (RRASOR), reflecting a moderate risk for sex offense reconviction. Ince also scored a +5 on the Static-99, reflecting a moderate-high risk for sex offense reconviction.