Rick v. Harpstead

Full title: DARRIN SCOTT RICK, Petitioner, v. JODI HARPSTEAD, Commissioner, Minnesota…

Court: United States District Court, District of Minnesota

Date published: Jun 21, 2023

678 F. Supp. 3d 1068 (D. Minn. 2023)

Opinion

Case No. 19-CV-2827 (NEB/DTS)

2023-06-21

Darrin Scott RICK, Petitioner, v. Jodi HARPSTEAD, Commissioner, Minnesota Department of Human Services, Respondent.

Lance Heisler, Lampe Law Group, LLP, Northfield, MN, for Petitioner. Elizabeth Beltaos, Hennepin County Attorney's Office, Minneapolis, MN, Matthew Hough, Hennepin County Attorney's Office, Adult Services Division, Minneapolis, MN, Emily Beth Anderson, Matthew Frank, Office of the Minnesota Attorney General, Saint Paul, MN, for Respondent.

Nancy E. Brasel, United States District Judge

Lance Heisler, Lampe Law Group, LLP, Northfield, MN, for Petitioner. Elizabeth Beltaos, Hennepin County Attorney's Office, Minneapolis, MN, Matthew Hough, Hennepin County Attorney's Office, Adult Services Division, Minneapolis, MN, Emily Beth Anderson, Matthew Frank, Office of the Minnesota Attorney General, Saint Paul, MN, for Respondent. AMENDED ORDER ON REPORT AND RECOMMENDATION Nancy E. Brasel, United States District Judge

In what the parties agree was a very close case, a Minnesota court civilly committed Petitioner Darrin Scott Rick as a sexually dangerous person in 2004. He has been a patient of the Minnesota Sex Offender Program ("MSOP") ever since. In 2019, Rick petitioned for writ of habeas corpus under 28 U.S.C. Section 2254, seeking his release from custody based on newly discovered evidence. After holding an evidentiary hearing on the petition, United States Magistrate Judge David T. Schultz issued a Report and Recommendation recommending that the Court grant Rick's petition. (ECF No. 94 ("R&R").) Hennepin County¹ objects to the R&R. (ECF No. 100 ("Obj.").) After a de novo review, the Court overrules the objection, accepts the R&R, and grants the petition.

1 Jodi Harpstead, the Commissioner of the Minnesota Department of Human Services, is the named Respondent who is holding Rick in state custody. Rick v. Harpstead, 564 F. Supp. 3d 771, 775 n.1 (D. Minn. 2021). Hennepin County responds on Harpstead's behalf in defense of the commitment order because it petitioned for Rick's commitment. Id.

BACKGROUND

The R&R details the facts and procedural history of the case. (R&R at 2-10.) The Court lays out the facts necessary for context.²

2 The Court cites the R&R and incorporates the citations it contains.

Criminal conviction. In 1993, Rick pled guilty to four counts of criminal sexual conduct involving minors and was sentenced to 180 months in prison. In re Civ. Commitment of Rick, No. A06-1621, 2007 WL 333885, at *1 (Minn. Ct. App. Feb. 6, 2007). While in prison, Rick participated in but eventually withdrew from sex-offender treatment. (R&R at 2.) After the Minnesota Department of Corrections ("DOC") declined to recommend Rick for civil commitment, Hennepin County began its own civil commitment proceedings. (Id.)

2004 commitment trial. During Rick's civil-commitment trial, court-appointed expert psychologists Dr. Thomas Alberg and Dr. Roger Sweet testified that Rick met Minnesota's statutory definition for a "sexually dangerous person" ("SDP").³ (Id. at 3.) But they also concluded that the MSOP program was unnecessary for Rick's treatment. (Id. at 3-4.) In their opinions, less-restrictive alternatives would suffice. (Id.) The DOC's Civil Commitment Review Coordinator testified that she believed Rick should not be civilly committed. (Id. at 4 (citing Ex. 1⁴ at 10).) Contradictory testimony came from Hennepin County's retained psychologist Dr. James Alsdurf, who testified that the only treatment appropriate for Rick was at the MSOP. (Id. at 4.) The trial court found that Rick's "moderate risk of recidivism combined with not completing sex offender treatment . . . proved by clear and convincing evidence that there is a likelihood of [Rick] reoffending." (Id. at 5; see Ex. 1 at 6-7, 11, 17.) The trial court then determined that Rick met the criteria for commitment to the MSOP. (Ex. 1 at 14.) But the court stayed Rick's commitment so long as Rick complied with certain conditions, including completing an outpatient sex-offender treatment program. (Id.) Hennepin County appealed, and the Minnesota Court of Appeals reversed the stay, both because Hennepin County had not agreed to the stay as required by law, and because record evidence did not establish that an outpatient treatment program had accepted Rick. (R&R at 5.)

3 Under Minnesota law, an "SDP" is a person who:

(1) has engaged in a course of harmful sexual conduct as [statutorily defined];

(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 as [statutorily defined].

Minn. Stat. § 253B.02, subdiv. 18c (2004) (now codified at Minn. Stat. § 253D.02, subdiv. 16(a) (2020)). Minnesota courts interpret the statute to require a person to be "highly likely" to engage in acts of harmful sexual conduct. In re Linehan (Linehan III), 557 N.W.2d 171, 180 (Minn. 1996), vacated sub nom., Linehan v. Minnesota, 522 U.S. 1011, 118 S.Ct. 596, 139 L.Ed.2d 486 (1997), aff'd sub nom., In re Linehan, 594 N.W.2d 867 (Minn. 1999); see also In re Civ. Commitment of Ince, 847 N.W.2d 13, 20-22 (Minn. 2014) (reaffirming the "highly likely" interpretation from Linehan III).

4 The exhibits from the habeas evidentiary hearing are identified as "Ex." The first ten exhibits are also found at ECF No. 5.

2006 commitment hearing. On remand, the trial court held a hearing at which it considered less restrictive treatment alternatives than commitment for Rick. (Id. at 6.) This time, the evidence showed that an outpatient treatment program had accepted Rick, but Dakota County (the county in which Rick planned to live) required Rick to live in a halfway house for 90 days. (Id.) The DOC had approved Rick for only a 60-day stay in a halfway house, and the house refused to accept Rick as a "private pay" client for the remaining 30 days. (Id. at 6-7.) Because the DOC had not approved funding for the full 90 days, the court found that Rick had not shown that a less-restrictive plan was "presently available." (Id. at 7.) On these slim margins, the court committed Rick to the MSOP indefinitely. (Id.) Rick sought post-commitment relief, but he did not succeed. (Id. (citing Rick, 2007 WL 333885).)

New evidence. At the time of Rick's commitment in 2004, the DOC used an actuarial sexual-recidivism risk-assessment tool called the Minnesota Sex Offender Risk Assessment Tool-Revised ("MnSOST-R"). (Ex. 7 at 3.) MnSOST-R predicted that low-risk offenders had a six-year recidivism rate of 12% and moderate-risk offenders had a six-year recidivism rate of 25%. (Id. at 4.) In 2012, new data changed those predictions significantly. Research on Minnesota sex offenders released from the DOC between 2003 and 2006 concluded that the recidivism rate for low-risk offenders was 3% (not 12%), and that the recidivism rate for moderate-risk offenders was 6% (not 25%). (Id. at 4 n.2 (citing G. Duwe & P.J. Freske, Using Logistic Regression Modeling to Predict Sexual Recidivism: The Minnesota Sex Offender Screening Tool-3 (MnSOST-3), Sexual Abuse: Journal of Research and Treatment, 24(4), 350-377 (2012) ("2012 Study")).) Other new research showed that a failure to complete sex-offender treatment did not increase recidivism. (Id. at 7-8 (citing G. Duwe & R. A. Goldman, The Impact of Prison-Based Treatment on Sex Offender Recidivism: Evidence from Minnesota, Sexual Abuse: Journal of Research and Treatment, 21(3), 279-307 (2009) ("2009 Study")).)

In 2019, Rick's counsel consulted forensic psychologist Dr. Amy Phenix, who evaluated Rick, reviewed his commitment case, and determined that the actuarial data used to commit Rick was no longer scientifically reliable. (R&R at 7; see generally Ex. 7.) Dr. Phenix concluded that although the actuarial risk-assessment tools used at the time of Rick's commitment "were appropriate for that period in time," later research showed that the tools overestimated the general probability of sex-offender recidivism and thus overestimated Rick's risk of recidivism as well. (R&R at 8; Ex. 7 at 6-8.) Applying this new research, Dr. Phenix concluded that Rick's current risk of recidivism was between 5.6% and 6.8% in five years. (Ex. 7 at 22, 26.)

Drs. Alberg and Sweet, who had testified in support of Rick's commitment, reviewed Dr. Phenix's report. They concluded that, contrary to their testimony at Rick's commitment trial, Rick did not meet the statutory criteria for commitment in 2004. (Exs. 4, 6.) Dr. Alberg stated that "Rick's commitment as an SDP was inappropriate in 2004." (Ex. 4 at 2.) Dr. Sweet similarly stated that "Rick did not meet the statutory criteria necessary for commitment" as an SDP in 2004. (Ex. 6 at 4 (emphasis omitted).)

Habeas petition. Before the Court is Rick's petition for habeas relief under 28 U.S.C. Section 2254. (ECF No. 1.) Rick contends that his commitment is a fundamental miscarriage of justice because the Minnesota court relied on now-discredited risk-assessment evidence and expert opinions, thus violating his right to due process under the Fourteenth Amendment. (ECF No. 45 ("2d Am. Pet.") at 3-4.) Hennepin County moved to dismiss the operative petition as barred by AEDPA's statute of limitations. (R&R at 12-13.) Judge Schultz recommended that the motion be denied. (Id.) Over the County's objection, this Court accepted Judge Schultz's report and recommendation and ordered a hearing on the issue of whether Rick satisfied the actual-innocence exception to overcome AEDPA's statute of limitations. Rick v. Harpstead, 564 F. Supp. 3d 771, 787-88 (D. Minn. 2021). If Rick satisfied the actual-innocence exception, Judge Schultz was to conduct an evidentiary hearing on the merits of Rick's due-process claim. Id. The parties agreed to present both the actual-innocence issue and the merits of the petition at the same hearing. (R&R at 14.)

Habeas hearing. In December 2021, Judge Schultz held a three-day evidentiary hearing at which five witnesses testified and 21 exhibits were presented. (Id. at 14-22.) Drs. Phenix, Sweet, and Alberg testified on Rick's behalf. (Id. at 14-18.) Hennepin County called Dr. Alsdurf and forensic psychologist Dr. Harry Hoberman. (Id. at 19-22.) Dr. Hoberman was not involved in Rick's 2004 commitment proceedings; he testified about whether the evidence on which Rick's petition was based is reliable under the actual-innocence gateway test. (Id. at 20-21.)

The R&R. Having heard and seen the evidence, Judge Schultz in the R&R concluded that Dr. Phenix's hearing testimony matched her 2019 report and that the testimony of Drs. Phenix, Sweet, and Alberg was credible.⁵ (Id. at 14-17.) Judge Schultz determined that Drs. Sweet and Alberg "recanted their 2004 opinions that Rick met the criteria to be civilly committed." (Id. at 19.) Judge Schultz also found Dr. Alsdurf's testimony "less credible than Drs. Sweet and Alberg regarding whether Rick met the SDP criteria in 2004." (Id. at 20.) He found Dr. Hoberman credible but less persuasive, and he concluded that Dr. Hoberman's testimony about sex-offender recidivism research "was not directly germane to the precise legal issues" at hand. (Id. at 22.)

5 Because Hennepin County objects to Judge Schultz's witness-credibility findings, the Court addresses their testimony in greater detail below.

Judge Schultz determined that Rick satisfied the actual-innocence exception to AEDPA's statute of limitations because he "demonstrated that the research and examiners' recantations are new and reliable evidence and that it is more likely than not that no reasonable jurist would have civilly committed Rick." (Id. at 1.) As for Rick's due-process claim, Judge Schultz concluded that Rick established that "the state court's substantial reliance on the court-appointed examiners' now-recanted testimony rendered his trial fundamentally unfair and Rick's commitment a miscarriage of justice." (Id.) Thus, a reasonable probability existed that "the reliance placed on the now-recanted opinions affected the outcome of Rick's commitment trial because it is unlikely a reasonable jurist considering the new evidence would commit Rick." (Id. at 1-2.) On this basis, Judge Schultz recommends granting Rick's habeas petition. (Id. at 2.)

ANALYSIS

Hennepin County's objection to the R&R addresses both the actual-innocence exception and the due-process claim. On the actual-innocence exception, the County objects to the R&R's conclusion that the new scientific evidence is reliable, including the standard for reliability that the R&R applied. The County also objects to the R&R's evaluation of the credibility of the expert witnesses as their testimony relates to the actual-innocence exception. On the due-process claim, the County objects to the R&R's conclusion that the commitment court substantially relied on the now-recanted examiners' opinions. And the County argues that the R&R ignored several Minnesota cases and erred in finding a reasonable probability that the recanted opinions affected the outcome of Rick's commitment trial.

The Court reviews the portions of the R&R to which Hennepin County objects de novo, including factual findings. 28 U.S.C. § 636(b)(1); D. Minn. L.R. 72.2(b)(3). As part of that review, the Court has examined the record, including the transcripts and exhibits from the habeas hearing, the parties' proposed findings of fact, and their final arguments.

I. Actual-Innocence Exception: Reliability and Credibility

In its prior order, this Court determined that a civilly committed person may invoke the actual-innocence exception to overcome the expiration of AEDPA's one-year statute of limitations for filing federal habeas petitions. Rick, 564 F. Supp. 3d at 781-83; see 28 U.S.C. § 2244(d)(1). Application of the actual-innocence exception should "remain 'rare' " and "only be applied in the 'extraordinary case.' " Schlup v. Delo, 513 U.S. 298, 321, 115 S.Ct. 851, 130 L.Ed.2d 808 (1995). For the exception to apply, Rick must persuade the Court that, "in light of new reliable evidence, it is more likely than not that no reasonable jurist would have ordered the person's civil commitment." Rick, 564 F. Supp. 3d at 783 (first citing McQuiggin v. Perkins, 569 U.S. 383, 386, 133 S.Ct. 1924, 185 L.Ed.2d 1019 (2013); and then citing Schlup, 513 U.S. at 324, 327, 115 S.Ct. 851).

The Court must make its actual-innocence determination "in light of all the evidence," Schlup, 513 U.S. at 328, 115 S.Ct. 851, including evidence that became available only after Rick's commitment trial. See House v. Bell, 547 U.S. 518, 539, 126 S.Ct. 2064, 165 L.Ed.2d 1 (2006) (noting that Schlup "requires a holistic judgment about all the evidence, and its likely effect on reasonable jurors applying the [applicable] standard" (cleaned up, citation omitted)). Evidence of innocence must be "so strong that a court cannot have confidence in the outcome of the trial unless the court is also satisfied that the trial was free of nonharmless constitutional error." McQuiggin, 569 U.S. at 401, 133 S.Ct. 1924 (citation omitted). An actual-innocence claim must be based on new, reliable scientific evidence not presented at the original commitment trial. Schlup, 513 U.S. at 324, 115 S.Ct. 851.

The actuarial tools considered at Rick's commitment trial in 2004 were based on recidivism data from the 1980s and 1990s, which found Rick's recidivism risk to be low to moderate. (See Ex. 1 at 11-12.) At the habeas hearing, the witnesses agreed that recidivism rates for sex offenders had declined before Rick's commitment trial. (See, e.g., Hr'g Tr. II at 80 (Dr. Hoberman agreeing with Dr. Phenix and Dr. Alsdurf that recidivism rates dropped), Hr'g Tr. I at 204 (Dr. Alsdurf agreeing); Hr'g Tr. III at 72 (Dr. Sweet agreeing).)

According to Hennepin County, the R&R used the wrong standard to evaluate the reliability of the new studies, and it made witness-credibility findings erroneously, without considering the entire record.⁶ (Obj. at 2-10.)

6 Hennepin County also objects to the R&R's statement that the County "focuses solely on the 2009 and 2012 Studies and fails to present any argument regarding Rick's assertion that Dr. Phenix's expert opinion and Drs. Sweet's and Alberg's recantations are also new reliable evidence for purposes of the actual innocence gateway." (R&R at 25 (citing ECF No. 90 at 3-18).) Hennepin County maintains that the R&R should have also considered its proposed findings. (Obj. at 9 (citing ECF No. 91).) The Court reviews the R&R de novo and finds that the statement is immaterial to its decision.

A. Reliability of 2009 and 2012 Studies

As noted, Rick must support his actual-innocence claim "with new reliable evidence—whether it be exculpatory scientific evidence, trustworthy eyewitness accounts, or critical physical evidence—that was not presented at [his commitment trial]." Schlup, 513 U.S. at 324, 115 S.Ct. 851. The R&R reasoned that "reliable evidence 'is simply evidence that is trustworthy enough to be admissible under the rules of evidence.' " (R&R at 28) (citing United States v. KT Burgee, 988 F.3d 1054, 1060 (8th Cir. 2021).)

Hennepin County asserts that "trustworthiness" is the wrong standard: "Applying a standard based on trustworthiness—in the context of evidentiary admissibility—was erroneous given that Rick's claim is premised on exculpatory scientific evidence that forms the basis of all opinions derived therefrom and does not adequately assess the strength of the evidence of actual innocence." (Obj. at 9 (emphasis in original).) The County argues that because the 2009 and 2012 Studies are exculpatory scientific evidence, the R&R should have done more than find that they have "several indicia of reliability." (R&R at 29.) The County contends that the Court must instead ask if the Studies are scientifically valid. (Obj. at 9-10) (citing Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 590 n.9, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993)).

When considering scientific evidence, "evidentiary reliability means trustworthiness." Kuhn v. Wyeth, Inc., 686 F.3d 618, 625 (8th Cir. 2012) (citing Daubert, 509 U.S. at 590 n.9, 113 S.Ct. 2786). As the County points out, evidentiary reliability of scientific evidence should be based on its "scientific validity." Daubert, 509 U.S. at 590 n.9, 113 S.Ct. 2786. To assess scientific validity, the Court may consider the Daubert factors: "(1) whether the theory or technique can be (and has been) tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) the known or potential rate of error; and (4) whether the theory has been generally accepted" in the relevant scientific community.⁷ Lauzon v. Senco Prods., Inc., 270 F.3d 681, 686-87 (8th Cir. 2001) (quotation marks and citation omitted); see Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) ("[T]he test of reliability is 'flexible,' and Daubert's list of specific factors neither necessarily nor exclusively applies to all experts or in every case.").

7 Courts have considered the Daubert factors in determining the reliability of new evidence of innocence. See, e.g., Fields v. White, No. 15-38-ART, 2016 WL 7425291, at *3 n.1 (E.D. Ky. Dec. 22, 2016) (noting that if a "petitioner clears the first obstacle blocking the actual-innocence gateway—by presenting 'new' evidence—the Daubert factors would help to verify the reliability of that evidence"); Kuenzel v. Allen, 880 F. Supp. 2d 1205, 1223 (N.D. Ala. 2011) (recognizing that "the actual innocence assessment is not limited to strictly admissible evidence" and finding that an expert opinion was not reliable under Daubert because "[e]ven when the strict rules of evidence do not apply, the lack of scientific reliability undermine[d] the weight" of the medical expert's opinion).

While the R&R did not list the Daubert factors or use the phrase "scientific validity," its analysis is thorough, and Judge Schultz uses some of the considerations listed in Daubert.⁸ (E.g., R&R at 29 (noting that the Studies had been published a peer-reviewed academic journal).) This Court determines that the expert opinions presented are scientifically valid, both for the reasons stated in the R&R and its own de novo analysis that includes consideration of the Daubert factors.⁹

8 The Court is granted "the same broad latitude when it decides how to determine reliability as it enjoys in respect to its ultimate reliability determination." Kumho Tire, 526 U.S. at 142, 119 S.Ct. 1167 (emphasis in original).

9 Hennepin County does not object to the R&R's finding that Drs. Sweet's and Alberg's written statements are new for purposes of the actual-innocence exception. (R&R at 26-28.) It is an expert witness's "opinion itself, rather than the underlying basis for it, which is the evidence presented." Souter v. Jones, 395 F.3d 577, 592 (6th Cir. 2005) (citing Fed. R. Evid. 702). Because Drs. Sweet and Alberg changed their opinions, "the evidence itself has changed, and can most certainly be characterized as new." Id. "As a result, the [recantations] can be consider[ed] 'new reliable evidence' upon which an actual innocence claim may be based." Id. at 593.

According to the County, the R&R ignored evidence that the 2009 and 2012 Studies are "not considered authoritative," and have not been cited in "significant literature" on sexual recidivism. (Obj. at 2; see id. at 8-9 (arguing that the Studies are not generally accepted in the field of forensic psychology and do not repudiate earlier studies).) The R&R concluded otherwise, and the Court agrees with that conclusion. The Court reviews the methodology in each study, but first, it notes that both were written by recognized DOC researchers and published in a peer-reviewed academic journal. (See Exs. 16-17; R&R at 29 (citing Aims & Scope, Sexual Abuse: J. Rsch. & Treatment, https://journals.sagepub.com/aims-scope/SAX (last visited May 17, 2023) ("The [Sexual Abuse] journal publishes rigorously peer-reviewed articles . . . .")).) Moreover, the 2009 Study has been cited in three "recent meta-analyses and/or systematic reviews of sex offender treatment." (Hr'g Tr. II at 54.)

The 2012 Study. The 2012 Study concluded that recidivism rates were lower for sex offenders released from Minnesota prisons between 2003 and 2006 than the rates presented at Rick's commitment trial. (See Ex. 17 at 357-59; Ex. 7 at 4 (comparing the recidivism rates).) In challenging the 2012 Study's reliability, Hennepin County notes that the Study was not a sexual-recidivism study, but an actuarial assessment instrument for the DOC.¹⁰ (Obj. at 2.) The County argues that the 2012 Study's import is limited for a variety of reasons, including because its follow-up period was shorter than other studies and because it measured sexual recidivism by convictions, not arrests, thereby underestimating the risk of re-offense. (See ECF No. 90 at 5-6; Hr'g Tr. II at 20, 23, 34-35 (Dr. Hoberman's testimony); Hr'g Tr. I at 37 (Dr. Phenix explaining that "the longer the follow-up the better," but "sometimes we don't have the data that . . . goes out five, ten or fifteen years"); Hr'g Tr. III at 78-79 (Dr. Sweet acknowledging that the parameters used to measure recidivism may impact the base rate); see also ECF No. 90 at 4-5, 12-13 (arguing the 2012 Study is not reliable because, among other things, it was not cross-validated or independently cross validated, and its sample size included non-sexual offenses).)¹¹

10 The 2012 Study resulted in the development of the MnSOST-3, which the DOC then began to use to assign a risk level to offenders. (Hr'g Tr. I at 76-77.) Dr. Phenix never used the MnSOST-3 apart from this case. (Id.)

11 The County also asserts that the makeup of the sample in the 2012 Study was deficient because it did not include civilly committed individuals. It argues that the recidivism base rate of 3% cited in the 2012 Study does not apply to Rick because he was committed at the time of the study. (ECF No. 90 at 7, 13; see also Hr'g Tr. II at 18-27; Hr'g Tr. III at 80.) This limitation does not render the Study unreliable for purposes of Rick's actual-innocence claim. It is not clear that the recidivism base rate that applied to civilly committed individuals (10.5%), (Ex. 17 at 366), is appropriately applied to Rick because the rate would have been used to determine whether to commit Rick. Even if the appropriate base rate is 10.5%, that rate is significantly lower than the 25% recidivism rate used in the MnSOST-R. (See Ex. 7 at 7 (noting that MnSOST-R accounted for withdrawal from treatment, and the moderate risk level had an associated recidivism rate of 25%).)

These limitations do not detract from the 2012 Study's reliability. Dr. Phenix testified that the 2012 Study's conclusion about recidivism rates being lower at the time of Rick's commitment has been borne out in other studies.¹² (Hr'g Tr. I at 38-39.) And the fact that all three forensic psychologists who participated in Rick's commitment proceeding agree that recidivism rates had declined up to and after the time of Rick's commitment suggests that this conclusion is generally accepted in the field. (See Hr'g Tr. I at 176; Hr'g Tr. II at 67-68; Hr'g Tr. III at 33 ("[B]y 2008, 2009, it was very clear that the base rates were consistently . . . going down. There wasn't just a blip in the curve.").)

12 For example, Rick cites a study entitled Sex Offender Recidivism in Minnesota published by the DOC in 2007 ("2007 Study"). (See generally Ex. 21.) The 2007 Study reached a similar conclusion about the decline of recidivism rates. (See, e.g., id. at 3 ("Since 1990, the sexual recidivism rates in Minnesota has dropped precipitously, as the three-year sexual conviction rate for 2002 releases was 3 percent compared to 17 percent for 1990 releases.").)The R&R did not address this study, noting that "numerous recidivism studies since 2004 could be brought to bear on the question of Rick's likelihood of committing future acts of harmful sexual conduct," but that Drs. Sweet's and Alberg's recantations are "the only new evidence that can be used by this Court to address, in a principled way, the second prong of the actual innocence gateway test." (R&R at 35 n.9.) The Court agrees with this position and only considers the 2007 Study when assessing the reliability of the 2012 Study.

The 2009 Study. The 2009 Study compared offenders who had not participated in treatment with those who had completed treatment and, relevant here, found that "[d]ropping out of treatment did not significantly increase the risk of recidivism . . . ." (Ex. 16 at 26.) At the time of Rick's commitment trial, the court-appointed examiners believed that that Rick's failure to complete treatment increased his risk of reoffending. (See Ex. 3 at 11; Ex. 5 at 26; see also, e.g., Hr'g Tr. III at 27 ("At the time that was pretty much the mantra. People who refused treatment or failed treatment or drop out of treatment are going to be more likely to re-offend.").)

Hennepin County presses that the R&R ignored evidence in the 2009 Study that the risk of re-offense increases when a person fails to complete sex offender treatment. (Obj. at 2.) The study found that this conclusion was "not statistically significant," (Ex. 16 at 22), so the Court puts no weight on the R&R's failure to address it.

At the habeas hearing, Dr. Hoberman claimed that "virtually every other study that exists in the published and unpublished world says that people who don't complete treatment have higher sex offense recidivism rates."¹³ (Hr'g Tr. II at 104.) The court-appointed examiners do not agree. For example, Dr. Sweet testified that the research conflicts on the impact and importance of treatment, and he listed studies questioning the efficacy of sex-offender treatment in his report. (Hr'g Tr. III at 48-50; Ex. 6 at 3). And Dr. Alberg explained that the effect of sex-offender treatment on recidivism is still being studied, noting that "there isn't definite evidence [that] sex offender treatment is absolutely necessary in order to reduce recidivism." (Hr'g Tr. I at 177-78.)

13 Dr. Hoberman did not elaborate on this point, but his report identifies other studies that allegedly found that dropping out of treatment correlated with higher sexual-offense recidivism rates. (See Ex. 20 at 94-96.)

Summary. The standard for judging the evidentiary reliability of expert evidence is "lower than the merits standard of correctness." Kuhn, 686 F.3d at 625 (citation omitted). All sex offender-related studies have limitations. (See Hr'g Tr. II at 52-71.) The 2009 and 2012 Studies acknowledge their limitations, as did the witnesses who relied on them. (See Ex. 16 at 2-3; Ex. 17 at 514-17; see Hr'g Tr. II at 56-57.) The R&R acknowledged these limitations and nevertheless found that the Studies were trustworthy "on their own merits, and in light of the pre-existing evidence regarding the closeness of Rick's case."¹⁴ (R&R at 29.) Having reviewed the hearing testimony and evidence, the Court concludes that the 2009 and 2012 Studies, although not dispositive, are reliable. Hennepin County's objection is overruled.

14 Hennepin County also contends that the R&R "[h]eavily weighed the Studies' origination from Minnesota" even though Drs. Sweet and Hoberman found this was insignificant. (Obj. at 2.) The R&R did not afford particular weight to the origin of the studies; it merely acknowledged—as Dr. Hoberman did—that "research specific to Minnesota provides a more accurate basis for understanding recidivism in Minnesota than does research from other states or countries." (R&R at 14; see Hr'g Tr. II at 14 (testifying that "one is better off if one has recidivism data that's based on the particular location").)

B. Witness Credibility

Hennepin County objects to the R&R's findings about the credibility of the witnesses who testified at the habeas hearing. The Court reviews these credibility findings de novo. See United States v. Lothridge, 324 F.3d 599, 600-01 (8th Cir. 2003) (remanding because the district court failed to review de novo a magistrate judge's credibility findings); Taylor v. Farrier, 910 F.2d 518, 521 (8th Cir. 1990) ("[O]nce a party makes a proper objection to a magistrate's finding, including a credibility finding, the district court must make a de novo determination of that finding." (emphasis in original)).

1. Dr. Phenix

Forensic psychologist Dr. Phenix was asked to "address the methods and procedures of the Sexual Psychopathic Personality/[SDP] evaluations that anchored the 2004 and 2006 judgments" that led to Rick's civil commitment "in the context of past and current research and professional practices and standards." (Ex. 7 at 2.) She determined that some actuarial data used to commit Rick was no longer scientifically reliable. Her report explains that later research—specifically the 2012 Study—showed that the recidivism tools used at Rick's commitment overestimated the general probability of sex-offender recidivism and thus overestimated Rick's risk of recidivism. (Ex. 7 at 6-7.) She determined that Rick's actual risk of recidivism in five years was between 5.6% and 6.8%. (Id. at 22, 26.) Dr. Phenix's report also explained that other new research—the 2009 Study—showed that the failure to complete sex-offender treatment did not increase the risk of recidivism. (Id. at 7-8.) Dr. Phenix testified about her report and conclusions at the habeas hearing. (See generally Hr'g Tr. I.)

Dr. Phenix testified about the strengths and limitations of the 2009 and 2012 Studies, as well as sex-offender research more generally. (See generally id. at 6-124.) She explained that since the early 1990s, sexual recidivism rates had consistently declined in Minnesota, but that the actuarials used in 2004 did not reflect that decline. (Id. at 76; see Ex. 7 at 4.) No witness disputes this fact. Nor does the County.

Hennepin County objects to the R&R's finding that Dr. Phenix's testimony was "consistent with her report and . . . very credible." (R&R at 14.) But the County does not cite any testimony by Dr. Phenix that contradicts the substance of her report.¹⁵ After reviewing Dr. Phenix's testimony, the Court finds that her testimony tracked her report.

15 Hennepin County cites Dr. Phenix's acknowledgment that her curriculum vitae identified the wrong title of the chapter on sexual recidivism that she co-authored. (Obj. at 4; see Hr'g Tr. 1 at 65-66 (explaining she co-authored a different chapter of the book).) This error does not impinge Dr. Phenix's credibility.

In challenging Dr. Phenix's credibility, Hennepin County asserts that Dr. Phenix's reliance on the 2009 and 2012 Studies was flawed. As discussed, the Court has reviewed the Studies and considered their various limitations, and it finds that they are reliable for purposes of Rick's actual-innocence claim.

Hennepin County also challenges Dr. Phenix's calculations of sexual-recidivism rates. Dr. Phenix contended that, given the 2012 Study's determination that four-year reconviction base rates for the 2003-06 release cohort were 3%—down from 12.3% for the 1992 release cohort—"the estimated probability for a moderate risk individual on the MnSOST-R would have dropped from 25% to about 6%, using a new charge for a sexual offense as the criterion for sexual recidivism." (Ex. 7 at 8.) She explained that:

In Minnesota, for example, 4-year sexual reconviction rates dropped from 12.3% for sexual offenders released in 1992 to 3% for sexual offenders released between 2003 and 2006. This represents a 76% decline in 4-year sexual reconviction rates over a span of about 12 years. Assuming a comparable drop in 6-years [sic] sexual charge rates for new sexual offense charges, this would suggest that a moderate risk assignment for an offender released between 2003 and 2006 would be associated with about a 6% rate of sexual recidivism, defined as a new sexual offense charge, rather than the 25% rate in the older normative sample.

(Id. at 4 (emphasis in original) (citing the 2012 Study).) The County challenges her calculation of the percentage decline in sexual recidivism. (Obj. at 5.) At the habeas hearing, Dr. Phenix acknowledged that she calculated the 76% number herself, and that the MnSOST-R measured sexual recidivism by arrest over a six-year period, whereas the 2012 Study measured sexual recidivism by conviction over a four-year period. (Hr'g Tr. I at 88-89.) Hennepin County suggested, and Dr. Phenix agreed, that applying this percentage to an individual on the MnSOST-R was an apples-to-oranges comparison. (Id. at 90.) Even so, the 2007 Study found that by 2002 the recidivism rate for sex offenders rearrested was 3.8%, and reconviction rate was 3%. (Ex. 21 at 20-21.) For purposes of Rick's claim, the .8% difference is minimal. Dr. Phenix's extrapolation appears to accurately depict the significant difference between recidivism rates used in the actuarial tools for Rick's commitment, and the rates that actually applied to sex offenders in 2004 when Rick was committed.

Hennepin County also notes that Dr. Phenix had not read Minnesota Statute Section 253D in years and was not familiar with Minnesota commitment caselaw. (Obj. at 4; see Hr'g Tr. I at 67-68, 73-74.) Dr. Phenix's credibility about accuracy of the recidivism rates used at Rick's original commitment trial and the effect of dropping out of treatment are fact-based, and do not require an analysis of Minnesota's commitment law. Her lack of legal knowledge does not diminish her credibility.

2. Dr. Alberg

Dr. Alberg was one of the court-appointed examiners who interviewed Rick in connection with his commitment in 2004. (Hr'g Tr. I at 137-38.) At that time, Dr. Alberg concluded that Rick "probably does meet the criteria" of an SDP. (Ex. 3 at 11; see Hr'g Tr. I at 156-57; but see Ex. 11 at 388 (not using the term "probably").) In reaching that opinion, Dr. Alberg found Rick's offense pattern and disordered arousal to be "critical," and his failure to complete sex-offender training to be a "major factor" contributing to his likelihood to offend. (Hr'g Tr. I at 138, 170; see id. at 152-53, 169-71.) At the time, Dr. Alberg understood that "[p]eople who withdraw from treatment are at higher risk" of recidivism, and he "gave that factor a great deal of significance." (Id. at 169, 181; see id. at 136 (explaining that a study had "indicated that not completing treatment was a significant factor in predicting recidivism"); Ex. 3 at 11 ("Research has shown that people who have . . . withdrawn from [sex-offender treatment] programs are at a much higher risk to reoffend than people who have been untreated.").)

In 2019, after reviewing Dr. Phenix's report and research articles regarding current recidivism rates and the effect of treatment on those rates, Dr. Alberg concluded that Rick should not have met the SDP criteria back in 2004. (Ex. 4 at 2; Hr'g Tr. I at 154.) He agreed with Dr. Phenix that "treatment has only a modest effect on whether a person is likely to reoffend, and that Mr. Rick's commitment probably was based on an overreliance on his dropping out of treatment." (Ex. 4 at 2.) Dr. Alberg testified to this effect at the habeas hearing, explaining that studies differ on the importance of completing treatment to recidivism rates, and that he possibly gave too much weight to this factor in his 2004 evaluation of Rick. (Hr'g Tr. I at 139.)

Since the 1990s, Dr. Alberg had performed about 200 to 250 initial commitment evaluations and several hundred evaluations on committed individuals seeking relief. (Id. at 125-26.) He could not recall a case resulting in commitment when two court-appointed examiners recommended against commitment and a retained expert recommended it, as with Rick. (Id. at 188.)

Hennepin County argues that the R&R erred when it found Dr. Alberg's testimony "credible and persuasive" and gave it "great weight." (R&R at 17; Obj. at 6-7.) The Court has reviewed Dr. Alberg's testimony and finds it highly credible and persuasive. Dr. Alberg reviewed Dr. Phenix's report and research articles (presumably the Studies) before deciding that Rick did not meet the SDP criteria in 2004. (Hr'g Tr. I at 173-74; Ex. 4 at 2.) He also considered several factors in Rick's record, including his history of sexual offenses and harmful sexual behavior, and his pedophilia and personality-disorder diagnoses. (Hr'g Tr. I at 152-53, 171.) Even on this record, Dr. Alberg "didn't think [Rick] strongly met the criteria" for an SDP in 2004, so his opinion in 2019 that Rick did not meet the SDP criteria "didn't change a lot but it changed a little." (Hr'g Tr. I at 154.) Hennepin County's other arguments—that Dr. Alberg was retained by Rick and testified that his 2004 opinion was not false, (Obj. at 6)—do not persuade the Court that Dr. Alberg is less credible.

3. Dr. Sweet

Dr. Sweet was the other court-appointed examiner for Rick's commitment proceeding in 2004 and interviewed Rick for that proceeding. Dr. Sweet believed the case was "[e]xtremely close," but he concluded that Rick met Minnesota's SDP criteria at that time, in part because Rick withdrew from treatment. (Ex. 5 at 27; Hr'g Tr. III at 26, 29.) At the time, Dr. Sweet believed that anyone who dropped out of treatment would be more likely to reoffend. (Hr'g Tr. III at 26-27.) In 2019, after reviewing Dr. Phenix's report, Dr. Sweet concluded that Rick did not meet the SDP criteria when committed because "improved actuarial results" would show Rick's risk of recidivism was "significantly lower than earlier measures." (Id. at 30-31; see Ex. 6 at 4.)

Dr. Sweet has performed four or five sex-offender evaluations annually for about fifteen years. (Hr'g Tr. III at 16.) As with Dr. Alberg, Dr. Sweet testified that that he had never been involved in a case that resulted in commitment after two court-appointed examiners counseled against commitment, but a retained expert recommended it. (Id. at 90-91.)

Hennepin County contends that the R&R erred in finding that Dr. Sweet's hearing testimony was "consistent with his 2019 written statement and . . . quite credible." (R&R at 15.) Having reviewed the record, the Court finds that Dr. Sweet's testimony aligns with his 2019 statement. The Court also finds Dr. Sweet's testimony credible. Dr. Sweet recanted his earlier opinion that Rick qualified as an SDP in 2004 after reviewing Dr. Phenix's report. (Hr'g Tr. III at 72, 82.) Although Dr. Sweet did not review the 2009 and 2012 Studies before authoring his 2019 statement, he did review them before the habeas hearing and did not change his position. (Id. at 70-71, 81-83.) The remaining issues raised by Hennepin County, including that Dr. Sweet was retained by Rick, testified that his 2004 opinion was not false, and believed that Minnesota's commitment statute was overly broad, (Obj. at 5-6), do not sway the Court as to Dr. Sweet's credibility.

4. Dr. Alsdurf

Dr. Alsdurf interviewed Rick in connection with his criminal proceedings in 1994. (Hr'g Tr. I at 192.) A decade later, Hennepin County retained Dr. Alsdurf in support of Rick's commitment as an SDP. (Id.) Dr. Alsdurf concluded that Rick met the SDP criteria. (Ex. 9 at 16.) After reviewing his original report, the commitment trial transcript, the 2009 and 2012 Studies, and the reports by Drs. Phenix, Alberg, and Sweet, Dr. Alsdurf continues to believe that Rick met the SDP criteria in 2004. In in support of his conclusion, Dr. Alsdurf relied on "the presence of [Rick's] sexual deviance, the persistence of his sexual deviance, his age, and the fact that [Dr. Alsdurf] did not see any real evidence of change over a several year period of time." (Hr'g Tr. I at 208; see id. at 196, 206 (listing these and other factors such as age and gender as "significant factors" in clinical override).) He does, however, acknowledge that this was a close case. (Hr'g Tr. at 210; see id. at 207 ("[T]his is not a slam-dunk case.").)

According to Hennepin County, the R&R erred in finding Dr. Alsdurf "less credible than Drs. Sweet and Alberg regarding whether Rick met the SDP criteria in 2004." (R&R at 20; see Obj. at 7-8.) The R&R found Dr. Alsdurf to be less credible in part because he "was overconfident in his assertions and refused to concede minor points," (R&R at 20), and criticized his own 2004 report, stating he "would write this report differently today" to emphasize Rick's violence. (Hr'g Tr. I at 196.)

Having reviewed the record, the Court agrees that Dr. Alsdurf's testimony about whether Rick met the SDP criteria was less credible than Drs. Alberg and Sweet. Besides the issues above, the Court did not find his testimony persuasive because although Dr. Alsdurf found the 2009 and 2012 Studies "interesting" and "helpful," he did not explain why he did not find them "persuasive in the sense that [they] changed [his] opinion." (Hr'g Tr. I at 194-95.) Dr. Alsdurf also testified that a record of sex-therapy treatment is a "very significant issue," but he did not grapple with the studies indicating that dropping out of treatment has an insignificant effect on future recidivism.¹⁶ (Hr'g Tr. I at 196; see id. at 194-97.)

16 Dr. Alsdurf also appears to have weighed other factors more heavily than the trial court and the court-appointed examiners. For example, he noted Rick's history of abuse was "really deviant," "really serious," and "really violent." (Hr'g Tr. I at 197.) As discussed, he also considered the presence and persistence of Rick's sexual deviance, Rick's age, the lack of change over several years, and gender. This evidence was presented and considered at the commitment trial, and the court initially ordered outpatient treatment for Rick. (See generally Ex. 1.) The Court defers to the commitment court's weight of such evidence. See 28 U.S.C. § 2254(e)(1) ("[A] determination of a factual issue made by a State court shall be presumed to be correct").

5. Dr. Hoberman

As discussed, Dr. Hoberman did not testify at the original commitment trial; he is a forensic psychologist retained by Hennepin County to testify at Rick's habeas hearing. (See generally Hr'g Tr. II.) His testimony was limited to assessing the reliability of the 2009 and 2012 Studies. (ECF No. 81 (order limiting Dr. Hoberman's testimony).) The Court finds that Dr. Hoberman is well qualified to testify about sex-offender research and recidivism rates.

The R&R found that Dr. Hoberman's testimony was "credible, [but] was less persuasive . . . because he at times attempted to avoid answering questions directly and gave non-responsive information in a convoluted way." (R&R at 22 (citing Hr'g Tr. II at 82-88).) "Mostly, however, his testimony was not directly germane to the precise legal issues with which this court must grapple." (Id.) Hennepin County objects to this finding because it "fails to evaluate his testimony in light of the reliability of the Studies." (Obj. at 8.) As part of its de novo review, this Court has considered Dr. Hoberman's testimony—which it finds to be credible, but not more credible than Drs. Alberg and Sweet—in determining that the 2009 and 2012 Studies are reliable for consideration of the actual-innocence gateway exception.

6. Summary

For the reasons above, the Court overrules the County's objections on reliability and credibility. It also concludes that, based on new reliable evidence, it is more likely than not that no reasonable jurist would have ordered Rick's civil commitment in 2004. Thus, the actual-innocence exception to AEDPA applies, and the Court therefore turns to Rick's due-process claim.

II. Due-Process Claim

Rick's due-process claim is based on the evidence discovered after his commitment that "undermined and discredited the critical risk assessment tools and clinical assumptions" on which the state court based his commitment, rendering that commitment "a fundamental miscarriage of justice." (2d Am. Pet. at 3-4.) To prevail on this claim, Rick must show that "the alleged improprieties were so egregious that they fatally infected the proceedings and rendered his entire trial fundamentally unfair." Rousan v. Roper, 436 F.3d 951, 958-59 (8th Cir. 2006) (citation omitted). Hennepin County contends that the R&R erred in concluding that Rick met this standard. (Obj. at 10.)

Recantations. Hennepin County objects to the R&R's characterization of the "recantation evidence" of Drs. Alberg and Sweet. (Obj. at 10.) The County asserts that Drs. Alberg and Sweet did not recant all their prior testimony, and that the R&R erred in not considering other testimony. (Id. at 10-11.)

The Court agrees that Drs. Sweet and Alberg did not recant their entire prior testimony, and after reviewing the R&R, finds that the R&R concluded similarly. The R&R states that "Drs. Sweet and Alberg have recanted their 2004 opinions that Rick met the criteria to be civilly committed," (R&R at 19), and thus Judge Schultz gave "no consideration to the now-recanted testimony." (Id. at 34.) By "now-recanted testimony," the Court understands Judge Schultz to mean Drs. Sweet's and Alberg's opinions that Rick met the SDP criteria. The R&R is based on "all the evidence" and "the entire habeas record," (id. at 41, 42), which includes Drs. Alberg's and Sweet's testimony at the habeas hearing. At that hearing, Drs. Alberg and Sweet testified about the various factors each had considered in reaching their original and revised opinions about Rick's qualification as an SDP. (See Hr'g Tr. I at 159-65 (Dr. Alberg testifying about Rick's offense pattern and treatment history, among other factors); Hr'g Tr. III at 61-63, 66 (Dr. Sweet testifying about factors contributing to Rick's likelihood to reoffend).)

The 2009 and 2012 Studies. Hennepin County also argues that because it disputed the reliability of the evidence presented, the literature relied upon by Rick's experts—presumably, the 2009 and 2012 Studies—are not inherently exculpatory. (Obj. at 11-12.) Courts have denied habeas relief when a petitioner presented evidence that did not repudiate previous evidence, but rather suggested there was debate about its validity within the scientific community. See, e.g., Feather v. United States, 18 F.4th 982, 987 (8th Cir. 2021) ("[I]f Feather's new medical and recantation evidence would have been presented at trial it would have established, at most, conflicting testimony and thus a reasonable juror considering all the evidence, old and new, could still convict Feather." (quotation marks omitted)); Gimenez v. Ochoa, 821 F.3d 1136, 1145 (9th Cir. 2016) (denying habeas relief where petitioner "presented literature revealing not so much a repudiation of [a scientific theory], but a vigorous debate about its validity within the scientific community"). Rick presents more than just literature suggesting a debate about the risk of recidivism. Both court-appointed examiners recanted their opinions that Rick met the SDP criteria. Although Hennepin County maintains that the totality of the evidence supports a finding that Rick was highly likely to reoffend, these examiners—who were questioned about a multitude of risk factors at the habeas hearing—disagree.

Minnesota caselaw. Hennepin County contends that the R&R erred in finding a "reasonable probability that the now-recanted opinions affected the outcome of Rick's commitment trial" because it failed to apply Minnesota caselaw properly. (Obj. at 12 (capitalization omitted).) The County bullet-points (without analysis) the holdings of several cases it contends that the R&R ignored. (Id. at 13-14.) A review of those cases shows that none warrants a conclusion different from the one reached by the R&R.

For example, the County argues that the R&R ignored caselaw holding that expert testimony is not a prerequisite to commitment, nor is the state district court bound by an expert opinion. (Obj. at 13); see In re Civ. Commitment of Miles, No. A14-0795, 2014 WL 4798954, at *4 (Minn. Ct. App. Sept. 29, 2014) (unpublished) ("Under the commitment statute, the district court, not the expert, must determine whether the statutory legal standards are met, although the assistance of experts may be required."); In re Commitment of Luhmann, No. A07-912, 2007 WL 2417341, at *5 (Minn. Ct. App. Aug. 28, 2007) (affirming commitment despite experts' opinions to the contrary). The Court sees nothing in the R&R indicating that Judge Schultz ignored these directives. Instead, he applied the correct standard, which is based on "a multi-factor analysis" using "the benefit of all the relevant and reliable evidence." Ince, 847 N.W.2d at 23; (see R&R at 36-37 (listing the factors)).

Similarly, the County's contention that the R&R ignored caselaw in which "[c]ommitment has been upheld where no base rate statistics were considered," or suggesting that "[b]ase rate information cannot be solely relied upon in assessing risk," is of no moment: the base rate information is not the sole evidence on which the Court is granting the habeas petition, nor was it the sole evidence considered by the R&R. (Obj. at 13 (citing In re Pirkl, 531 N.W.2d 902, 909 (Minn. Ct. App. 1995); Linehan III, 557 N.W.2d at 190).) The remaining caselaw cited by the County in this section of objections does not apply to—or change—the Court's analysis.

Due-process analysis. Rick's initial commitment proceeding was undeniably a very close case. The state court addressed Rick's withdrawal from treatment and Drs. Alberg's and Sweet's opinions that Rick had a moderate risk of recidivism, and it concluded that Rick's "moderate risk of recidivism combined with not completing sex offender treatment . . . proved by clear and convincing evidence that there is a likelihood of [Rick] reoffending." (Ex. 1 at 17.) The court also considered Dr. Alsdurf's conclusion that Rick met the criteria of an SDP in part because Rick "had not completed treatment." (Id. at 13.) And it considered Dr. Alsdurf's conclusions that Rick had a high likelihood of reoffending, and that he should be committed to the MSOP. (Id. at 13.) Based on this evidence, the court in 2004 found that Rick met the criteria for an SDP. (Id. at 14.)

Almost twenty years later, Drs. Alberg and Dr. Sweet have recanted their opinions that Rick met the criteria for an SDP at the time of his commitment and now conclude that he did not meet the criteria. In light of this new evidence, the Court agrees with the R&R that "it is very likely the state court would have ascribed considerably less weight to Rick's 'record with respect to sex therapy' at the time of his commitment." (R&R at 38 (citing In re Linehan, 518 N.W.2d 609, 611 (Minn. 1994))); see generally Lewis v. Erickson, 946 F.2d 1361, 1362 (8th Cir. 1991) ("Recanted testimony . . . is grounds for relief from a conviction when it either bears on a witness's credibility or directly on the defendant's guilt."). Indeed, the Court is hard-pressed to believe that, in such a close case, a reasonable court would reject the testimony of both court-appointed examiners and the DOC's Civil Commitment Review Coordinator that Rick should not be civilly committed, (see Ex. 1 at 10), in favor of the testimony of a privately retained expert that he does. It is more likely than not that, considering the new reliable evidence described above, no reasonable jurist would have found by clear and convincing evidence that Rick met the standard for commitment. Because Rick has shown that the alleged improprieties were so egregious that they fatally infected his commitment proceeding and rendered the proceeding fundamentally unfair, the Court grants his habeas petition.

CONCLUSION

Based on the foregoing and on all the files, records, and proceedings herein, IT IS HEREBY ORDERED THAT:

1. Hennepin County's Objection (ECF No. 100) is OVERRULED;

2. The Report and Recommendation (ECF No. 94) is ACCEPTED;

3. Rick's Second Amended Petition under 28 U.S.C. § 2254 for a Writ of Habeas Corpus (ECF No. 45) is GRANTED; and

4. Respondent is ORDERED to release Rick from detention and confinement in the MSOP program.

LET JUDGMENT BE ENTERED ACCORDINGLY.

REPORT AND RECOMMENDATION

DAVID T. SCHULTZ, United States Magistrate Judge

INTRODUCTION

Petitioner Darrin Scott Rick has been a patient in the Minnesota Sex Offender Program since 2004, when a Minnesota state court civilly committed him as a Sexually Dangerous Person. In 2019, Rick petitioned for a federal writ of habeas corpus, seeking to be released from custody and to invalidate his initial commitment based on newly discovered evidence. Rick challenges his civil commitment, arguing that research published after his commitment caused the two court-appointed examiners to recant their 2004 testimony and to state that, had they known of this research at the time of Rick's trial, they would have testified he did not meet the legal standard for civil commitment.

Rick has demonstrated that the research and examiners' recantations are new and reliable evidence and that it is more likely than not that no reasonable jurist would have civilly committed Rick. He has also established the state court's substantial reliance on the court-appointed examiners' now-recanted testimony rendered his trial fundamentally unfair and Rick's commitment a miscarriage of justice. There is a reasonable probability the reliance placed on the now-recanted opinions affected the outcome of Rick's commitment trial because it is unlikely a reasonable jurist considering the new evidence would commit Rick. Therefore, this Court recommends that Rick's petition for a writ of habeas corpus be granted.

FINDINGS OF FACT

The Court has previously detailed the facts and procedural history of this case and incorporates by reference its earlier summaries. Rick v. Harpstead, 564 F.Supp.3d 771 (D. Minn. 2021) (accepting as modified 2d R&R, Dkt. No. 53), Dkt. No. 58.

I. Background Facts

A. Criminal conviction

In 1993 Rick pleaded guilty to four counts of criminal sexual conduct involving minors and was sentenced to 180 months in prison. Resp. App. 3, Dkt. No. 41. While in prison, Rick participated in a sex offender treatment program and in a faith-based therapy program, though he did not complete either. Id. at 6-7. In contemplation of Rick's scheduled release, the Minnesota Department of Corrections (MNDOC) evaluated whether to recommend Rick for civil commitment and determined it would not do so. Id. at 1-2. Instead, MNDOC planned to conditionally release Rick on an intensive-supervised basis with GPS monitoring and out-patient sex offender treatment at Project Pathfinder. Id. at 1, 10. Notwithstanding MNDOC's decision, Hennepin County petitioned to civilly commit Rick as a Sexually Dangerous Person (SDP).¹⁷ Id. at 1.

17 Hennepin County also sought to commit Rick as a Sexually Psychopathic Personality (SPP), but all three testifying experts opined—and the state court concluded—that Rick did not meet the criteria to be committed as an SPP. Resp. App. 13-14. That determination is not at issue here.

B. The 2004 civil commitment trial

To be civilly committed as an SDP, the state court had to find Rick (1) had engaged in a course of harmful sexual conduct, (2) had manifested a sexual, personality, or other mental disorder or dysfunction, and (3) as a result, was highly likely to engage in acts of harmful sexual conduct in the future.¹⁸ Id. at 14, 16. As part of the commitment process, the court appointed two psychologists, Dr. Roger Sweet and Dr. Thomas Alberg, to evaluate whether Rick met the statutory criteria to be committed. Id. at 2. Both court-appointed examiners conducted in-person forensic interviews with Rick, evaluated his records, and personally visited the Project Pathfinder out-patient treatment program to determine whether it would be appropriate for Rick. Id. at 10-12. Drs. Sweet and Alberg both submitted independent written reports to the state court and testified at the 2004 commitment trial. Id. at 2.

18 Rick's commitment came down to the question whether he was highly likely to commit future acts of harmful sexual conduct as the evidence clearly established Rick had engaged in past acts of harmful sexual conduct and manifested a sexual and/or personality disorder. It is the prediction of Rick's future dangerousness that lives at the heart of this case.

Dr. Sweet concluded Rick had "only a moderate risk for recidivism" but met the statutory criteria for commitment as an SDP. Id. at 11. Dr. Sweet found there was a less restrictive alternative to in-patient commitment and recommended Rick follow the MNDOC intensive-supervised release plan. Id. Similarly, Dr. Alberg concluded Rick was at a low risk of reoffending but determined Rick "probably" met the statutory criteria as an SDP. Heisler Aff. 41, Dkt. No. 5. He based his conclusion, at least in part, on his understanding that:

Research has shown that people who have been in sex offender treatment programs and have failed these treatment programs or withdrawn from these programs are at a much higher risk to reoffend than people who have been untreated. Mr. Rick has been in one inpatient sex offender treatment program and he withdrew from this program. This is certainly a concern and a factor which would indicate a higher likelihood of Mr. Rick offending again in the future.

Id. Like Dr. Sweet, Dr. Alberg believed Rick did not need the intense security of the Minnesota Sex Offender Program (MSOP) to receive appropriate treatment. Id.

In support of its commitment petition, Hennepin County retained psychologist Dr. James Alsdurf. Resp. App. 13. Dr. Alsdurf's opinion was based on his review of Dr. Sweet's written report, as Dr. Alsdurf did not conduct a forensic interview with Rick. Id. In his report, Dr. Alsdurf noted Rick had not completed treatment and opined that Rick had a high likelihood of reoffending. Id. Thus, Dr. Alsdurf testified, Rick met the criteria for commitment as an SDP. Id. Unlike the two court-appointed examiners, however, Dr. Alsdurf felt the only appropriate treatment for Rick was inpatient treatment at MSOP. Id.

Penny Zwecker, MNDOC's Civil Commitment Review Coordinator, also testified at Rick's 2004 commitment hearing. Id. at 10. She stated that while she had some concerns, she did not believe Rick met the statutory criteria to be civilly committed. Id. at 10; State Tr. 324-25, Dkt. No. 43. She noted the "political climate" at the time favored civil commitment. Id. The state court found that Zwecker was experienced, professional, and thoughtful, that she had a very difficult job, and that her experience and demeanor led the court to give her opinion great weight. Resp. App. 10.

The state court also heard testimony from Rick, his current and former therapists, his mother and stepfather, his pastor, and three other witnesses. Id. at 2. Based on the trial testimony and the three experts' opinions, the state court found Rick was at a moderate risk of reoffending, noting specifically that Rick "started but did not complete sex offender treatment program while incarcerated." Id. at 17. The court concluded his "moderate risk of recidivism combined with not completing sex offender treatment . . . proved by clear and convincing evidence that there is a likelihood of [Rick] reoffending," and cited the SDP statutory language that a person be "likely to engage in acts of harmful sexual conduct." See Minn. Stat. § 253B.02, subd. 18c (2002) (now codified at Minn. Stat. § 253D.02, subd. 16 (2022)). The court did not reference or cite In re Linehan, in which the Minnesota Supreme Court specifically held that, despite the statute's use of "likely," the state and federal constitutions require that a person be "highly likely" to engage in acts of harmful sexual conduct in order to be civilly committed. In re Linehan, 557 N.W.2d 171, 180 (Minn. 1996), vacated and remanded, 522 U.S. 1011, 118 S.Ct. 596, 139 L.Ed.2d 486 (1997), aff'd as modified, 594 N.W.2d 867 (Minn. 1999).

When someone is committed as an SDP, "the court shall commit the [person] to a secure treatment facility unless the [person] establishes by clear and convincing evidence that a less restrictive treatment program is available that is consistent with the [person's] treatment needs and the requirements of public safety." Minn. Stat. §§ 253B.18, subd. 1, 253B.185, subd. 1 (2002) (now codified at Minn. Stat. § 253D.07, subd. 3 (2022)). The state court determined MNDOC's conditional release plan was a less restrictive alternative to treatment at MSOP. Id. at 18. The court stayed Rick's commitment to MSOP so long as Rick complied with his conditional release terms. Id. at 14. Hennepin County appealed the stay. Id. at 54-55. In an Order Opinion, the Minnesota Court of Appeals reversed, reasoning that Hennepin County had not agreed to a stay as required by statute and that the record did not establish the outpatient treatment program had accepted Rick. In re Rick, No. A04-1475 (Minn. App. Jan. 25, 2005) (order op.). The Minnesota Court of Appeals remanded the matter to the trial court to determine the availability and appropriateness of the outpatient treatment program. Id.

C. The 2006 hearing

On remand, the trial court held a hearing to determine whether a less-restrictive treatment program was available to Rick. Resp. App. 57. At the hearing, several witnesses testified including Hennepin County's retained expert Dr. Alsdurf and the court-appointed examiners Drs. Sweet and Alberg. Id. at 63-64. Dr. Sweet again testified he believed Rick could be adequately treated in an outpatient program and did not need to be at MSOP. Id. at 63. The court again found Dr. Sweet to be highly credible. Id. Like Dr. Sweet, Dr. Alberg continued to believe Rick did not need to be committed to MSOP. Id. at 63-64. Unlike the two court-appointed examiners, Dr. Alsdurf believed Rick should obtain treatment at MSOP. Id. at 63. Even so, Dr. Alsdurf stated that if Rick "had completed his sex offender treatment program [while incarcerated, he] most likely would not be before the court on this commitment proceeding." Id. at 63. The court found Dr. Alsdurf's observation "accurate." Id.

Based on the testimony of the experts and four other witnesses, the state court found Project Pathfinder "is not willing to accept [Rick] for treatment because of social and political pressure." Id. It also found an outpatient sex offender treatment program at the University of Minnesota had agreed to accept Rick, but his supervising county would not allow him to live with his parents upon release from MNDOC. Id. at 62. The supervising county instead required Rick live in a halfway house for at least 90 days, but MNDOC had approved funds for Rick to stay only 60 days. Id. Though Rick's parents were willing to personally fund the entire cost of Rick's lodging and treatment, the halfway house would not accept Rick as a "private pay" client for the 30 days not approved by MNDOC. Id.

In its decision, the state court noted Rick was unable to prove, "through no failure of his own," that funding existed for the halfway house. Id. at 64. It therefore concluded that although "it is a very close question," Rick had not shown by clear and convincing evidence that the proposed less-restrictive plan was "presently available." Id. In short, but for the refusal of the halfway house to accept payment from Rick's parents, Rick would have been treated in an out-patient program. Instead, Rick was committed to MSOP indefinitely, where he has been ever since. Id. at 65.

Rick appealed his commitment, arguing there was insufficient evidence to establish he was highly likely to reoffend, and that he had proven by clear and convincing evidence a less-restrictive alternative program was available. In re Rick, No. A06-1621, 2007 WL 333885 (Minn. App. Feb. 6, 2007). The Minnesota Court of Appeals rejected both arguments and affirmed the commitment order. Id. The Minnesota Supreme Court denied further review. Resp. App. 129. Rick has remained at MSOP for the past 18 years.

D. New evidence

In 2019, Rick's counsel consulted forensic psychologist Dr. Amy Phenix. Heisler Aff. 76. She evaluated Rick, reviewed his commitment case, and determined that the actuarial data used to commit Rick was no longer scientifically reliable. Id. at 103. Dr. Phenix issued a forensic report explaining her conclusions in April 2019. Id. at 76. The report addressed "(1) whether appropriate instruments, tools, methods, procedures, and assumptions were used in [Rick's] evaluations based on research and standards of practice at the time of the evaluations, and (2) whether more contemporary research and standards continue to support or question those instruments, tools, methods, procedures, and assumptions" that led to Rick's civil commitment. Id. at 78. Dr. Phenix concluded that while the actuarial risk assessment tools used at the time of Rick's commitment "were appropriate for that period in time," subsequent research has shown those tests vastly overestimated the general probability of sex offender recidivism. Id. at 81-101.

Dr. Phenix's report explained that one actuarial risk assessment tool used during Rick's commitment proceedings was developed from studies examining recidivism¹⁹ of sex offenders released from MNDOC custody in 1988, 1990, and 1992. Id. at 79. At the time this data was collected, offenders were released with no structured, external controls on sexual behavior and no supports for pro-social behavior. Id. Based on those samples, the studies predicted a base rate recidivism of 23%. Id. Low-risk offenders had a six-year recidivism rate of 12%, moderate-risk offenders had a six-year recidivism rate of 25%, and high-risk offenders had a six-year recidivism rate of 52%. Id.

19 Defined as the occurrence of a formal criminal charge for a new sexual offense within six years of release.

Research published in 2012 demonstrated that Minnesota sex offenders released from MNDOC between 2003 and 2006 (the period when Rick was to be released) had lower recidivism²⁰ rates than the offender sample used to develop the actuarial assessments analyzed in Rick's commitment proceedings. Id. (citing G. Duwe & P.J. Freske, Using Logistic Regression Modeling to Predict Sexual Recidivism: The Minnesota Sex Offender Screening Tool-3 (MnSOST-3), Sexual Abuse: J. Rsch. & Treatment, 24(4), 350-77 (2012)). This new research demonstrated that recidivism rates for low-risk offenders dropped from 12.3% for offenders released in 1992 to 3% for offenders released between 2003 and 2006. Id. From that study, Dr. Phenix extrapolated that a moderate-risk offender would have a four-year recidivism rate of 6% rather than 25%. Id. As part of that calculation, Dr. Phenix defined sexual recidivism as "a new charge for a sexual offense" and included offenders who had no treatment, completed treatment, quit treatment, or were kicked out of treatment. Id.

20 Defined as a sexual reconviction within four years of release.

In discussing research regarding sex offender treatment as it relates to recidivism, Dr. Phenix also noted that Dr. Sweet had cited a 1998 article that stated failure to complete treatment led to increased levels of recidivism. Id. at 82. Dr. Phenix explained that considering Rick's withdrawal from treatment in addition to the actuarial assessment used by Dr. Sweet (which already included his withdrawal from treatment in its calculations) "inappropriately double-counts this variable." Id. She explained further that research published in 2009 concluded that withdrawal from treatment is not strongly associated with an increased risk of sexual recidivism. Id. (citing G. Duwe & R. A. Goldman, The Impact of Prison-based Treatment on Sex Offender Recidivism: Evidence From Minnesota, Sexual Abuse: J. Rsch. & Treatment, 21(3), 279-307 (2009)). Research studies published since Rick's commitment have concluded that the failure to complete sex offender treatment does not produce a statistically significant increase in the risk of recidivism. Id. at 83. Based on the 2009 and 2012 Studies examined in her report, Dr. Phenix concluded that the court-appointed examiners and Hennepin County's expert all placed "undue negative weight on Mr. Rick's withdrawal from sex offender treatment and used actuarial risk assessment tools that subsequent research has documented as outdated and inappropriately high." Id. at 99. Therefore, she asserted, the evidence presented in Rick's commitment proceedings was based on, and led to, an inflated assessment of Rick's risk of recidivism. Id. at 98-101.

Both court-appointed examiners from Rick's commitment proceedings reviewed Dr. Phenix's 2019 report and concluded that, contrary to their previous testimony, Rick did not meet the statutory criteria for commitment in 2004. Id. at 42-43, 72-75. Dr. Sweet asserted that he "could have avoided the frustrating and exasperating process of arguing for a Stay of Commitment, because [he] would have opined that Mr. Rick did not meet the statutory criteria necessary for commitment." Id. at 75 (emphasis in original). Similarly, Dr. Alberg stated that he had been "reluctant to recommend Mr. Rick be civilly committed back in 2004," that Rick's failure to complete sex offender treatment was a "very significant factor" in his decision to recommend commitment, and that "it was inappropriate for Mr. Rick to be civilly committed as an SDP." Id. at 42-43.

II. Procedural History

In October 2019, Rick petitioned for a writ of habeas corpus under 28 U.S.C. § 2254, asserting two grounds for habeas relief. Pet., Dkt. Nos. 1. First, he argued new evidence shows that his initial commitment violated his right to due process under the Fourteenth Amendment (Ground One). Id. He asserted that the 2009 and 2012 Studies, Dr. Phenix's report, and the court-appointed examiners' recantations are new and reliable evidence that discredits the reliability of the actuarial assessments and clinical assumptions about sexual recidivism that led the three experts to testify Rick met the commitment criteria. Pet. Mem. 6-23, Dkt. No. 6. Rick contended that because the state court relied so heavily on the expert opinions and the evidence regarding the impact of treatment withdrawal on recidivism rates and because the court-appointed examiners have now recanted their testimony, his commitment violated his right to due process under the Fourteenth Amendment and is a fundamental miscarriage of justice. Id. Second, even if his initial commitment was constitutional, Rick asserted the new evidence renders his continued commitment unconstitutional (Ground Two). Id. at 23-25.

Rick brought his habeas petition against Jodi Harpstead, the Commissioner of the Minnesota Department of Human Services, and the officer holding Rick in state custody. Rick, 564 F.Supp.3d at 775-76. Harpstead moved to dismiss Rick's petition, with Hennepin County defending Ground One and the Minnesota Attorney General's Office (AGO) defending Ground Two. Dkt. Nos. 12, 15, 16, 55. Hennepin County argued Ground One is time barred and no equitable tolling or actual innocence exception applies to Rick. 1st Dismiss Mem. 7-15, Dkt. No. 15. It asserted the actual innocence gateway should not apply to civil commitments and Rick does not qualify for the actual innocence gateway on the merits. Id. at 15-36. The County also argued Rick failed to exhaust his state court remedies because he could seek discharge from commitment by petitioning the Minnesota Special Review Board (SRB) and the Commitment Appeal Panel (CAP). Id. at 36-41 (citing Minn. Stat. § 253D.27-31 (2019)). The AGO argued Ground Two is time-barred and unexhausted because Rick brought the petition outside the one-year limitation period and he could petition the SRP and CAP for full discharge. AGO Answer 12-15, Dkt. No. 16.

This Court concluded Rick had not exhausted Ground One "only because there is no process for him to do so. Harpstead acknowledges 'Rick could not present to the [judicial appeal panel] the precise claim he makes . . .' because both the panel and the Minnesota Court of Appeals have held the panel's jurisdiction does not include challenges to the original commitment." 1st R&R 8-9, Dkt. No. 35 (alteration in original) (citations omitted). The Court concluded Rick failed to exhaust Ground Two and stayed the matter to allow Rick to either exhaust his claim or amend his mixed petition to remove it. Id. at 9-11, 15. The Court recommended that the motion to dismiss Ground One be denied. Id. at 15. Neither Hennepin County nor the AGO objected to the Report and Recommendation (R&R). Order Accepting R&R, Dkt. No. 37. Finding no clear error, the District Judge accepted the R&R and denied the motion to dismiss. Id.

Rick amended his petition to remove Ground Two. Dkt. Nos. 35, 45. Hennepin County again moved to dismiss the petition, arguing Rick's claim was time barred and no equitable tolling or actual innocence exception applied. Resp. Mem. 8-15, Dkt. Nos. 36, 40, 42. The County asserted the actual innocence gateway should not apply in civil commitment cases and that Rick does not qualify for the actual innocence gateway on the merits. Id. at 12-36. Hennepin County also asserted Rick should be required to exhaust the state remedy available to him, arguing the SRP and CAP "may grant Rick the relief that federal habeas would grant him" through full discharge. Resp. Mem. 37-38 n.15.

This Court concluded that Hennepin County argued Rick could petition for discharge from MSOP, not that he could obtain the relief he seeks under Ground One—invalidation of his commitment. 2d R&R 23. The Court reasoned that Rick challenges his original commitment trial as fundamentally unfair and that while the state procedure—if successful—could result in his release, that release may be provisional and would not invalidate his original commitment. Id. at 23-24; see Karsjens v. Piper, 845 F.3d 394, 399 (8th Cir. 2017) (explaining reduction in custody relief). In contrast, this petition—if successful—would invalidate his original commitment. Id. at 24; see Magwood v. Patterson, 561 U.S. 320, 332, 130 S.Ct. 2788, 177 L.Ed.2d 592 (2010) (explaining habeas petitioner "seeks invalidation (in whole or in part) of the judgment authorizing . . . confinement"). This Court determined Rick's habeas claim is distinct from the state administrative discharge process and cannot be exhausted through the SRP and CAP. Id. (citing Simpson v. Norris, 490 F.3d 1029, 1035 (2007)). The Court also concluded that either Rick's petition was timely or, if untimely, the actual innocence gateway is applicable to civil commitments and Rick had presented sufficient evidence to warrant an evidentiary hearing on the gateway innocence argument. 2d R&R 15-22. This Court further found that Rick had stated a plausible due process claim. Id. at 28. The Court therefore recommended that Hennepin County's motion to dismiss be denied. Id. at 32.

Hennepin County objected to the second R&R, arguing Rick's petition is untimely, the actual innocence gateway does not apply to civil commitments, Rick failed to state a cognizable due process claim, and the Linehan line of Minnesota Supreme Court decisions render Rick's claim meritless. Obj., Dkt. No. 56. The District Judge carefully considered Hennepin County's objections but concluded the actual innocence gateway may apply in the civil commitment context and Rick had presented sufficient evidence to warrant an evidentiary hearing on the issue. Rick, 564 F.Supp.3d at 781-85. The District Judge also determined Rick's petition raised important questions about the fundamental fairness and reliability of his commitment hearing and the Linehan decisions did not render Rick's claim meritless. Id. at 785-87. Accordingly, the District Judge denied Hennepin County's motion to dismiss and ordered an evidentiary hearing to determine whether Rick meets the actual innocence gateway criteria, and if so, whether Rick has established his due process rights were violated at the 2004 commitment trial. Id. at 787-88. The parties agreed to have the actual innocence gateway issue and the merits of the petition presented in the same evidentiary hearing.

III. Evidentiary Hearing

In December 2021, the Court held a three-day evidentiary hearing at which five witnesses testified and 21 exhibits were admitted into evidence. Dkt. Nos. 82-84, 86, 88, 89. As each witness testified, this Court observed the witnesses' demeanor and evaluated their credibility.

A. Rick's witnesses

Rick called three witnesses. Dr. Phenix testified first. Tr. I, 6, Dkt. No. 88. The Court finds her testimony is consistent with her report and is very credible. She is an experienced forensic psychologist and is well qualified to testify about sex offender recidivism research. Id. at 6-8. Dr. Phenix testified regarding the strengths and limitations of the 2009 and 2012 Studies, as well as sex offender research in general. Id. at 6-124. In particular, the Court finds persuasive her testimony that research specific to Minnesota provides a more accurate basis for understanding recidivism in Minnesota than does research from other states or countries. Dr. Phenix also testified that since the early 1990s there has been a consistent decline in recidivism rates among sex offenders, the reasons for which are "multi-faceted." Id. 76, 103. She also testified that in 2004, actuarial assessments did not reflect these declining rates of recidivism. Id. at 76. The Court finds recidivism rates in general have been declining since the early 1990s, but the actuarial assessments utilized in 2004 did not reflect the lower recidivism rates in real time.

Dr. Phenix further testified regarding standards and oversight of professionals who evaluate sex offenders. Id. at 14-42. Dr. Phenix noted that evaluators are not required to use only empirically verified factors to determine whether someone meets the SDP criteria and is unaware of any process for evaluators to disclose bias, personal experience, or political considerations in connection with a recommendation. Id. Moreover, she testified, there is no standard for the level of education or training required of an evaluator. Id. She also testified that an evaluator may consider an offender's treatment history without regard to the substance of the treatment programs. Id. at 17, 24-25, 47.

Dr. Phenix testified that there was a 76% decline in four-year sexual reconviction rates over a span of 12 years. Id. at 87-90, 117. She arrived at that number by analyzing different studies and calculating an estimated percent decline. Id. While the exact percent is subject to interpretation, the Court finds there was a significant decline in recidivism base rates for offenders released in Minnesota between 2003 and 2006, as compared to those released in 1992. Lastly, Dr. Phenix opined that when the reliance the examiners and the state court placed on Rick's treatment history is excluded, the record could not support a determination that Rick was highly likely to reoffend. Id. at 60.

Dr. Sweet, the court-appointed First Examiner, also testified. The Court finds his testimony is consistent with his 2019 written statement and is quite credible. He became a psychologist in 1969 and began working on sex offender civil commitments in the early 1990s. Tr. III, 6-7, Dkt. No. 86. Dr. Sweet performed about four or five sex offender evaluations a year for about 15 years and stated he likely recommended commitment in "four out of five" cases. Id. at 16-17, 90.

Dr. Sweet remembered Rick's case "quite well," in part because MNDOC did not recommend Rick be civilly committed. Id. at 17-18, 91. He could not think of another case in which a county petitioned for commitment without MNDOC's recommendation; it was "very unusual." Id. at 17-18. Dr. Sweet stated that although Rick's actuarial scores "were pretty much in the low to moderate range," he placed great emphasis on Rick's withdrawal from treatment because he believed that anyone who refused, failed, or dropped out of treatment would be more likely to re-offend. Id. at 25-27. Rick's withdrawal from treatment was a significant factor that led him to opine in 2004 that Rick met the SDP criteria. Id. at 29.

Dr. Sweet stopped performing sex offender evaluations around 2010 and did not stay current with sex offender literature thereafter. Id. at 16, 59. He became aware of the 2009 and 2012 Studies after reviewing Dr. Phenix's report. Id. at 70, 82-83. Though he did not read the Studies when they were published, he has read them since. Id. The Court finds that, while Hennepin County's skillful cross-examination elicited somewhat ambiguous testimony from Dr. Sweet, he had read both Studies before testifying. Id. Dr. Sweet further testified that, despite his 2019 written statement's reliance on Dr. Phenix's report regarding the 2009 and 2012 Studies, he independently knew that sex offender recidivism rates declined between 2003 and 2012. Id. at 72.

Dr. Sweet testified he had never been involved in a case in which someone was committed when two court-appointed examiners recommended against commitment, but the retained expert recommended commitment. Id. at 90-91. He described Rick's case as "an extremely close case" and that he felt "back then" Rick was "just over the wire." Id. at 30, 91. He testified that because improved actuarial assessments would show Rick's risk of recidivism is "significantly lower than earlier measures," Rick did not meet the standard for commitment as an SDP in 2004. Id. at 30-31. The Court finds this testimony to be truthful and significant.

Dr. Alberg, the court-appointed Second Examiner, also testified. The Court gives his testimony great weight as he is credible and persuasive. Dr. Alberg began doing mental health commitment work in Minnesota in the 1980s. Tr. I, 125-26. He has performed about 200-250 initial commitment evaluations and several hundred evaluations on committed individuals seeking relief. Id. Dr. Alberg is highly qualified and experienced regarding civil commitment matters. Id. at 125-27. He testified that he recommends civil commitment in roughly 75% of the cases in which he has been an examiner. Id. at 184. Like Dr. Sweet, he could not think of a case in which an individual was committed when the two court-appointed examiners recommended against commitment and the retained expert recommended commitment. Id. at 188.

Dr. Alberg testified that although he opined in 2004 that Rick met the SDP criteria, that opinion was not strong at the time. Id. at 128-56. He had based his 2004 opinion on Rick's various risk factors, actuarial assessments, record information, and impressions from his interview with Rick. Id. 125-82. Dr. Alberg believes the impact of sex offender treatment on recidivism rates "is not nearly as clear as I thought it was" in 2004 and is a factor he may have weighed too heavily in Rick's case. Id. at 139, 185. At the time of his 2004 testimony, he had placed great significance on Rick's sex offender treatment history. Id. at 181, 185.

Dr. Alberg described Rick's case as "uniquely close." Id. at 186. In 2004, he believed Rick needed treatment but did not need to be in a locked facility such as MSOP. Id. However, because of the political environment at the time there were not many treatment options available. Id. at 143-44, 147-48.

Lastly, Dr. Alberg testified that after reviewing the current information regarding sexual recidivism rates and the effect of treatment on recidivism, he does not believe that in 2004 Rick met the statutory criteria to be committed as an SDP. Id. at 154. He described that change in opinion to be a "little" change because even in 2004, he did not think Rick "strongly met the criteria." Id. As with Dr. Sweet, the Court finds this testimony to be truthful and significant.

Hennepin County previously objected to this Court's characterization of Drs. Sweet's and Alberg's 2019 written statements as "recantations" because neither Rick nor the court-appointed examiners used that term. Obj. 1 n.1. In discussing the recantations, Rick stated they were a "reversal of opinion" by Drs. Sweet and Alberg. Pet. Mem. 15. That the word "recantation" was not previously used does not preclude this Court from finding the court-appointed examiners have recanted their testimony. Recant means "to withdraw or renounce (prior statements or testimony) formally or publicly," Recant, Black's Law Dictionary (11th ed. 2019), or "to withdraw, retract, renounce, or disavow (a former statement, opinion, belief, action, etc.) as erroneous or heretical, esp. formally or publicly," Recant, v.1., Oxford English Dictionary, https://www.oed.com/view/Entry/159342?rskey=Z7W6D3&result=1&isAdvanced=false#eid (last visited May 16, 2022). Here, Drs. Sweet and Alberg have formally and publicly withdrawn as erroneous their prior reports and testimony in which they had opined that in 2004 Rick met the statutory criteria to be civilly committed. Heisler Aff. 42-43, 72-75; Tr. I, 154; Tr. III, 54; see Santos v. Thomas, 779 F.3d 1021, 1026 (9th Cir. 2015) (determining statements were "recantations" because they directly contradicted or otherwise challenged witnesses' own initial statements) rev'd en banc on other grounds, 830 F.3d 987 (9th Cir. 2016); Souter v. Jones, 395 F.3d 577, 583-84 (6th Cir. 2005) (stating forensic pathologist's trial testimony that victim's injuries "may well have been inflicted by" bottle was recanted by his affidavit stating "it is unlikely" bottle caused wounds). Thus, this Court finds Drs. Sweet and Alberg have recanted their 2004 opinions that Rick met the criteria to be civilly committed.

B. Hennepin County's witnesses

Hennepin County called two witnesses, with Dr. Alsdurf testifying first. Tr. I, 190; Tr. II, Dkt. No. 89. Dr. Alsdurf has been licensed as a psychologist since 1984 and focuses on forensic evaluations. Tr. I, 190. In 2004, Hennepin County retained Dr. Alsdurf to support its petition seeking Rick's commitment. Id. at 192-93.

Dr. Alsdurf testified that the 2009 and 2012 Studies, though interesting and helpful, have not changed his opinion regarding Rick. Id. at 194-95. Though he agrees that recidivism rates in general were much lower in 2004 than the then-current actuarial assessments indicated, he continues to believe that in 2004 Rick met the statutory criteria to be committed. Id. at 204. He found Rick's violent behavior in the underlying criminal offense to be significant. Id. at 196. He testified that Rick's record regarding sex offender treatment was a significant issue, calling it a "very solid piece of data." Id. at 196-98. Dr. Alsdurf stated that if Rick's treatment record contained evidence that Rick "had come to terms with [his sexual deviance,] even at a moderate, maybe even a less-than-moderate level, I would have been much more inclined to recommend" that Rick did not meet SDP criteria. Id. at 198. Dr. Alsdurf also stated that had Rick completed sex offender treatment, "he would never have been referred" for civil commitment and that "everyone was conflicted about this case; this is not a slam-dunk case." Id. at 203, 207.

The Court finds Dr. Alsdurf has substantial experience conducting sex offender evaluations, but he is less credible than Drs. Sweet and Alberg regarding whether Rick met the SDP criteria in 2004. At times he was overconfident in his assertions and refused to concede minor points. Id. at 201. When cross examined, Dr. Alsdurf recognized this when he stated, "I don't mean to sound so bombastic." Id. at 199.

The Court also finds his testimony less persuasive because he had not personally interviewed Rick in connection with the 2004 commitment proceedings or in preparation for the evidentiary hearing. Id. at 200. When testifying to the risk factors Rick had in 2004, Dr. Alsdurf spoke as though he had a personal treatment history with Rick in 2004, but he in fact had only reviewed Dr. Sweet's report and Rick's records. Id. at 201. Dr. Alsdurf did not view his lack of interview with Rick as limiting his ability to assess Rick's risk of recidivism. Id. He testified that while he "might have understood things better" had he interviewed Rick, the interview would not have changed his opinion. Id. at 211. He also criticized his own 2004 written report, stating he would have written it differently today (approximately 18 years after his review of Rick's record) to indicate he believed Rick posed greater risk than he had portrayed in his 2004 report. Id. at 196, 205. These examples illustrate the basis for the Court's finding that Dr. Alsdurf's testimony is less credible and persuasive than that of Drs. Sweet and Alberg.

Next, Dr. Hoberman testified. Tr. II, 2-112. He had not been involved in Rick's 2004 commitment proceedings but was retained by Hennepin County to testify at the habeas evidentiary hearing. Id. at 8. As such, his testimony was limited to whether the new evidence on which Rick's petition is based is reliable under the actual innocence gateway test.²¹ Order, Dec. 13, 2021, Dkt. No. 81.

21 Hennepin County originally submitted a 103-page report by Dr. Hoberman in which he addressed not only whether the evidence was new and reliable but also whether in his opinion, Rick met the standard for commitment in 2004 and whether Rick currently meets the standard for commitment. Dkt. No. 77. Those issues are beyond the purview of the evidentiary hearing and beyond what this Court must decide as to Rick's petition. Dr. Hoberman's report also purported to address whether Rick had satisfied his burden of proof on the second prong of the actual innocence gateway test and the due process analysis. The Court excluded these portions of Dr. Hoberman's initial report as irrelevant and invading the province of the Court. Dkt. No. 81.

Dr. Hoberman is very well credentialed in the area of sex offender commitment. See, generally, Tr. II, 3-8. In 1986, he obtained his Minnesota license to practice psychology and in 1991 he was court-appointed to work on sex offender cases. Id. at 4. In the late 1990s, he received training in forensic psychology. Id. at 3. Dr. Hoberman reviewed the 2009 and 2012 Studies and discussed their various limitations. Id. at 15-68. For example, he believes the 2012 Study's four-year follow-up period was too short and that including in the study individuals with non-sexual index offenses was unusual and underestimated the true recidivism rates regarding sex offenders. Id. at 18-20. Dr. Hoberman stated the 2009 Study, although included in other studies regarding sex offender treatment, is "not particularly" significant and he is aware of studies with the opposite conclusion. Id. at 54-55, 61-62. He does not believe the 2009 and 2012 Studies, because of their limitations, are applicable to Rick. Id. at 15-30, 42, 56-62.

Dr. Hoberman agrees with the other witnesses that the reported (or "detected") recidivism rates of sex offenders decreased significantly from 1990 to 2005, Id. at 67-68, 80, but he believes that 80 to 90% of sex offenses are undetected and unaccounted for in risk assessment tools. Id. at 68-71. Thus, he opines, all actuarial tools underestimate the true rates of sexual recidivism. Id. at 69. Dr. Hoberman does not weigh the undetected factor when he completes his assessments of sex offenders but does include information regarding undetected offenses in his reports. Id. Dr. Hoberman generally views sex offender research with skepticism but believes risk assessment tools "remain reliable instruments for assessing recidivism, particularly when considered collectively." Id. at 52.

The Court finds Dr. Hoberman is an experienced forensic psychologist and is well qualified to speak about sex offender research and recidivism rates. Dr. Hoberman though credible, was less persuasive to this Court because he at times attempted to avoid answering questions directly and gave non-responsive information in a convoluted way. See id. at 82-88. Mostly, however, his testimony was not directly germane to the precise legal issues with which this court must grapple. Id.

IV. Arguments

Because this is a complex habeas corpus case, it is appropriate to summarize the parties' arguments as presented in the petition, the parties' briefing and exhibits, the evidentiary hearing, and the parties' final arguments. Rick argues the Court may decide his habeas petition despite any procedural bar because he has presented new, reliable evidence that establishes his commitment is a miscarriage of justice. Pet. Final Br., Dkt. No. 92. He bases his habeas claim on new evidence in the form of the 2009 and 2012 Studies, Dr. Phenix's opinions, and Drs. Sweet's and Alberg's recantations. Id. Rick contends that because his commitment was based on the state court's substantial reliance on now-recanted testimony (which was in turn based on actuarial assessments developed from outdated research) "he has proven by clear and convincing evidence that no reasonable fact finder would have found [in 2004] that he met the criteria as a sexually dangerous person," and that the flawed testimony undermined the fundamental fairness of his trial. Id. at 24-31.

Hennepin County argues Rick has failed to establish that the actual innocence gateway applies and therefore his petition is untimely. Resp. Mem. 8-39. It contends Rick's evidence is neither reliable, nor establishes that it is more likely than not that no reasonable jurist would have committed him. Resp. Final Br. 18-19, Dkt. No. 90. Hennepin County also asserts that Rick's due process claim is without merit because the state court considered evidence of risk factors other than Rick's treatment history and did not rely solely on actuarial instruments or base rate evidence. Id. Hennepin County argues that, based on the record as a whole, Rick's trial was fundamentally fair. Id. at 19.

CONCLUSIONS OF LAW

I. Habeas Legal Standard

In reviewing a habeas petition, federal courts presume state court factual determinations are correct, although the petitioner can rebut that presumption by clear and convincing evidence. 28 U.S.C. § 2254(e)(1). If the state court did not adjudicate a claim on the merits, federal courts apply a de novo standard of review to determine the facts. Porter v. McCollum, 558 U.S. 30, 39, 130 S.Ct. 447, 175 L.Ed.2d 398 (2009). The state court has not evaluated the 2009 and 2012 Studies, Dr. Phenix's opinions, or the court-appointed examiners' recantations. Thus, while the factual determinations from Rick's commitment proceedings are presumed correct, the unadjudicated facts and conclusions regarding the new evidence are determined de novo.

Generally, a petitioner may not obtain federal habeas relief unless the petitioner has exhausted available state court remedies and brought the petition within the one-year statute of limitations. Duncan v. Walker, 533 U.S. 167, 179, 121 S.Ct. 2120, 150 L.Ed.2d 251 (2001) (applying 28 U.S.C. § 2244(d)(1), (2)). The Court has determined Rick has exhausted the available state court remedy identified by the Court, but has failed to prove his petition is timely because the factual predicate of Rick's claim is the research advancements underlying the 2019 opinions, not the opinions themselves. Rick, 564 F.Supp.3d at 777-783 n.12, 780.

A federal court may consider the merits of an untimely habeas claim if the petitioner has presented new reliable evidence that demonstrates it is more likely than not that no reasonable jurist would have civilly committed him. Id. at 781, 783-84, 786; see House v. Bell, 547 U.S. 518, 537-38, 126 S.Ct. 2064, 165 L.Ed.2d 1 (2006). This is often referred to as the "actual innocence gateway" because an adequate showing of actual innocence may serve as a gateway for a petitioner's otherwise procedurally barred habeas claim. Rouse v. United States, 14 F.4th 795, 801 (8th Cir. 2021) (citing McQuiggin v. Perkins, 569 U.S. 383, 386, 133 S.Ct. 1924, 185 L.Ed.2d 1019 (2013)). This gateway standard does not require absolute certainty. House, 547 U.S. at 537-38, 126 S.Ct. 2064. The actual innocence gateway allows a federal court to consider the petitioner's constitutional claim when the new evidence is "so strong that a court cannot have confidence in the outcome of the trial unless the court is also satisfied that the trial was free from non-harmless constitutional error." McQuiggin, 569 U.S. at 401, 133 S.Ct. 1924.

Thus, Rick must assert both a legitimate constitutional claim and a credible claim of actual innocence. See Amrine v. Bowersox, 128 F.3d 1222, 1227 (8th Cir. 1997) (citing Schlup v. Delo, 513 U.S. 298, 316, 115 S.Ct. 851, 130 L.Ed.2d 808 (1995)). Rick has asserted a legitimate due process claim—that the state court's reliance on the experts' now-recanted testimony undermined the fundamental fairness of his commitment trial. Rick, 564 F.Supp.3d at 784-86. To consider the merits of Rick's due process claim, the Court must first decide whether Rick has established that he meets the actual innocence gateway, that is, whether, in light of new and reliable evidence, it is more likely than not that no reasonable jurist would have ordered Rick's civil commitment.

II. Actual Innocence Gateway

A. New reliable evidence

A credible claim of actual innocence must be supported with "new reliable evidence . . . that was not presented at trial." House, 547 U.S. at 537, 126 S.Ct. 2064 (quotation omitted). This Court must first determine the nature and scope of the evidence to consider under the actual innocence gateway analysis. Hennepin County contends Rick failed to present new reliable evidence in the form of sex offender recidivism research because the research, it argues, has significant deficiencies and is not widely accepted in the field of forensic psychology. Resp. Final Br. 16. Hennepin County focuses solely on the 2009 and 2012 Studies and fails to present any argument regarding Rick's assertion that Dr. Phenix's expert opinion and Drs. Sweet's and Alberg's recantations are also new reliable evidence for purposes of the actual innocence gateway. See id. at 3-18.

To begin, while the Court previously found Rick's petition untimely because the factual predicate of his claim could have been discovered more than one year before he filed his petition, this finding does not limit the scope of the evidence the Court may consider in conducting the actual innocence gateway analysis. See McQuiggin, 569 U.S. at 399, 133 S.Ct. 1924. A credible claim of actual innocence requires "new reliable evidence." House, 547 U.S. at 537, 126 S.Ct. 2064 (quoting Schlup, 513 U.S. at 324, 115 S.Ct. 851). That evidence must be new only in the sense that it was "not available at trial through the exercise of due diligence." Kidd v. Norman, 651 F.3d 947, 953 (8th Cir. 2011). Any alleged delay or lack of diligence in pursuing a claim of actual innocence does not bar the claim but is among the factors that the court may consider in assessing the merits. McQuiggin, 569 U.S. at 388-400, 133 S.Ct. 1924.

In McQuiggin, the Supreme Court distinguished the 28 U.S.C. § 2244(d)(1)(D) factual predicate analysis from the actual innocence gateway analysis. 569 U.S. at 389-99, 133 S.Ct. 1924 (clarifying that actual innocence claim "seeks an equitable exception to § 2244(d)(1), not an extension of the time statutorily prescribed" (emphasis in original)). The Supreme Court rejected the argument that a petitioner asserting actual innocence must prove the same diligence required under 28 U.S.C. § 2244(d)(1)(D). Id. at 399, 133 S.Ct. 1924. In doing so, it made clear that determining "new, reliable evidence" under the actual innocence gateway requires a different analysis than determining whether a petitioner acted with due diligence from the date of the "factual predicate" of his claim. Id.

The Eighth Circuit has recently reiterated that "evidence is 'new' if it was not available at the time of trial through the exercise of due diligence." Jimerson v. Payne, 957 F.3d 916, 927 (8th Cir. 2020) (finding petitioner discovered facts of actual innocence claim when someone else confessed); see Amrine v. Bowersox, 238 F.3d 1023, 1028-29 (8th Cir. 2001) (finding affidavit recanting trial testimony was "new" evidence); see also Johnson v. Norris, 170 F.3d 816, 818 (8th Cir. 1999) (questioning whether evidence was new when there was "no showing that [officer] would not have testified at trial the same way that he did at the habeas hearing had he been asked the right questions"). Here, the facts necessary to determine Rick's actual innocence claim not only come from the 2009 and 2012 Studies, but also from Dr. Phenix's report analyzing those studies and, most significantly, Drs. Sweet's and Alberg's recantations that stem from their review of the 2009 and 2012 Studies and Dr. Phenix's report. Thus, Hennepin County's assertion that Rick's new evidence is only the 2009 and 2012 Studies is incorrect. Under the actual innocence gateway analysis, the evidence this Court considers is the 2009 and 2012 Studies, Dr. Phenix's opinions, and the court-appointed examiners' recantations.

Hennepin County previously argued the evidence is not "new" because some of the research fundamental to the Studies and Dr. Phenix's report was available at the time of Rick's commitment hearing, but it has since abandoned that argument. See id. at 3-18. Nonetheless, the Court briefly examines whether this evidence is new. That the 2009 and 2012 Studies may have contained information available before the 2004 trial does not mean the Studies, as published, were available. Although other research studies available at the time of trial may have included similar data, research methodology, or conclusions, the Court finds the 2009 and 2012 Studies themselves are the evidence Rick presents. Because a 2009 Study and a 2012 Study could not be available in 2004, the Court finds they are "new" evidence. For the same reasons, the Court finds Dr. Phenix's 2019 report and her opinions analyzing those studies are "new" evidence.

Regarding Drs. Sweet's and Alberg's recantations, Hennepin County had argued they are not "new" as they are based in part on recidivism data that was available in 2004. That argument mischaracterizes the nature of Drs. Sweet's and Alberg's trial testimony. The court-appointed examiners were testifying at trial as expert witnesses. Therefore, it is their opinions, rather than the underlying basis for them, which are the evidence presented. Souter, 395 F.3d at 592 (comparing new expert opinion to eyewitness subsequently remembering additional details to find the opinion was "new"). If Drs. Sweet and Alberg have changed their expert opinions, the evidence itself has changed, and can most certainly be characterized as "new." Id. Moreover, it is abundantly clear from the record in this case that it was the 2009 and 2012 Studies and Dr. Phenix's 2019 report that caused them to recant their prior opinions and testimony.

Because determining whether the petitioner has presented "new, reliable evidence" under the actual innocence gateway requires a different analysis than determining "due diligence from the date of the factual predicate" under the timeliness analysis, McQuiggin, 569 U.S. at 399, 133 S.Ct. 1924, the Court looks at whether the evidence was available at the time of trial through the exercise of due diligence. Jimerson, 957 F.3d at 927. The evidence Rick presents did not exist at the time of trial; therefore, it was not available. Thus, the Court finds the 2009 and 2012 Studies, Dr. Phenix's 2019 report, and the 2019 recantations of the court-appointed examiners are "new" evidence.

Next, the Court must decide whether this new evidence is reliable. House, 547 U.S. at 537, 126 S.Ct. 2064. Courts may find evidence is reliable if the evidence is credible. Kidd, 651 F.3d at 952 n.5; Amrine, 238 F.3d at 1029. Generally, reliable evidence "is simply evidence that is trustworthy enough to be admissible under the rules of evidence." United States v. KT Burgee, 988 F.3d 1054, 1060 (8th Cir. 2021) (citing Kuhn v. Wyeth, Inc., 686 F.3d 618, 625 (8th Cir. 2012)) ("The Supreme Court explained that evidentiary reliability means trustworthiness."). However, courts are "not bound by the rules of admissibility that would govern a trial" when determining the reliability of evidence for actual innocence purposes. Schlup, 513 U.S. at 328, 115 S.Ct. 851. Instead, courts determine the reliability of all the new evidence submitted, both admissible and inadmissible, and determine whether it is trustworthy by considering it both on its own merits and, where appropriate, in light of the pre-existing evidence in the record. See id. at 327-28, 115 S.Ct. 851.

The 2009 and 2012 Studies have several indicia of reliability, such as being authored by recognized Minnesota Department of Corrections researchers and being published in a peer-reviewed academic journal. Resp. Exs. 16, 17; Aims & Scope, Sexual Abuse: J. Rsch. & Treatment, https://journals.sagepub.com/aims-scope/SAX (last visited May 16, 2022) ("The journal publishes rigorously peer-reviewed articles . . . ."). The 2009 Study has also been included in at least three meta-analyses or systemic reviews of sex offender treatment. Tr. II, 53-54. Moreover, at least three experts on sex offender recidivism have relied on the Studies as the basis for their sworn testimony opinion and the Studies are the type of information experts ordinarily use when determining sex offenders' risk of recidivism. Heisler Aff. 79-82; Tr. I, 6-124, 139-85; Tr. III, 72-76.

Hennepin County argues the 2009 and 2012 Studies are not reliable because they do not present "new science" regarding sex offenders and are not definitive or authoritative pieces of literature. Resp. Final Br. 4-15. In essence, it contends the Studies are unreliable because they have limitations, but this is true for all sex offender related studies, as Dr. Hoberman testified. Tr. II, 52-71. Dr. Hoberman's critiques (i.e., that the 2012 Study's four-year follow-up period is short and the inclusion of individuals with non-sexual index offenses underestimates the true recidivism rates regarding sex offenders) may impact the conclusions one can draw from the Studies, but do not render the Studies unreliable. After a thorough review of the 2009 and 2012 Studies, taking into consideration the limitations to which both Dr. Phenix and Dr. Hoberman testified, the Court finds these Studies are trustworthy both on their own merits, and in light of the pre-existing evidence regarding the closeness of Rick's case. Schlup, 513 U.S. at 327-28, 115 S.Ct. 851. This Court finds the Studies are—although not dispositive of Rick's claims and not as compelling as Rick asserts—reliable evidence this Court may consider in determining whether Rick's petition may pass through the actual innocence gateway.

Hennepin County similarly challenges Dr. Phenix's opinions as unreliable. Resp. Final Br. 7-8. In House, the Supreme Court concluded the petitioner presented new, reliable evidence to establish an actual innocence gateway. 547 U.S. at 554, 126 S.Ct. 2064. The petitioner had retained an expert witness to review his case and provide the habeas court with expert opinion regarding blood evidence. Id. at 542, 126 S.Ct. 2064. In doing so, the expert's testimony challenged the prosecution expert's trial testimony. Id. The Supreme Court found that even though there was conflicting expert testimony, the petitioner's expert witness's testimony was credible and "called into question" evidence connecting the petitioner to the crime. Id. at 546, 126 S.Ct. 2064. As with House, Dr. Phenix was retained by Rick to review the trial evidence and provide the habeas court with her expert opinion regarding the new research and how it relates to Rick's 2004 commitment trial. As this Court noted above, Dr. Phenix is well qualified to inform the Court about sex offender research and her testimony is quite credible. Hennepin County attempted to undermine Dr. Phenix's opinion with Dr. Hoberman's testimony, however the Court finds that Dr. Hoberman's testimony does not contradict Dr. Phenix's opinions. Rather, Dr. Hoberman explained limitations on sex offender research in general and used the 2009 and 2012 Studies and Dr. Phenix's opinion to exemplify these general limitations. Dr. Phenix also acknowledged limitations of the Studies and her own analysis.

Dr. Phenix testified that in 2004 the base rates of sex offender recidivism were lower than the actuarial assessment data demonstrated, and that sex offender treatment history did not impact risk of recidivism—if at all—as significantly as previously believed. In light of this, Dr. Phenix opines that, given the closeness of Rick's case, the new evidence demonstrates that Rick did not meet the SDP criteria in 2004. Tr. I, 60. Drs. Hoberman and Alsdurf disagree, but the fact that Hennepin County's retained experts disagree does not mean that Dr. Phenix's opinion is unreliable. It is abundantly clear that Dr. Phenix's opinion would be admissible at trial, at which the trial court would be tasked with deciding whose opinion carries the day. That is not what the Court is tasked with here, however. Instead, this Court must merely determine if the new evidence is reliable. Dr. Hoberman's critiques of and disagreement with Dr. Phenix's opinions regarding a matter of difficult professional judgment do not render her opinion unreliable. Hennepin County sets too high a bar for reliability; the evidence must simply be trustworthy on its own or when considering all the pre-existing evidence. Schlup, 513 U.S. at 327-28, 115 S.Ct. 851. The Court has closely considered Dr. Phenix's report and testimony and finds that both are reliable.

Lastly, the Court turns to the recantations. Both Drs. Sweet and Alberg are qualified to opine on whether Rick met the SDP criteria in 2004, as is readily apparent by their roles as court-appointed examiners in Rick's commitment proceedings. Each gave credible testimony regarding their perceptions of Rick's case, sex offender evaluations, sex offender recidivism research, and the basis for their recantations. Their recantations are more than sufficiently reliable because they are not merely based on a re-examination of the case, but rather, on an enhanced understanding of sex offender recidivism research. Souter, 395 F.3d at 592-93. This Court has considered not only the recantations, but the motives prompting them, the timing, any possible motive for the original opinions, any inconsistencies in the testimony or between the testimony and other evidence, the plausibility of inferences or assumptions that crediting the recantation would require, as well as all other factors generally considered in assessing witness credibility. The Court has heard directly from Drs. Sweet and Alberg and observed their demeanor as their recantations were offered on direct examination and tested on cross-examination. Hennepin County does not argue—and the record is devoid of any indication—that the court-appointed examiners have any motive to be insincere in their recantations. Dr. Alsdurf's disagreement with Drs. Sweet's and Alberg's opinions does not render their testimony unreliable. Therefore, the Court finds Drs. Sweet's and Alberg's recantations are reliable.

The evidentiary hearing testimony primarily focused on the experts' use of sex offender recidivism data and the meaning of such data in predicting future dangerousness. There was no disagreement that actuarial assessments based on sex offender re-arrest and/or reconviction data over a limited period of time underestimate the true rate of sex offender recidivism. The degree to which the data underestimates the true risk of recidivism and how recidivism data ought to impact expert opinions is a debate that has raged within the psychological community and will continue to do so. This Court will not resolve that debate; nor should it, as the question before the Court is whether the 2009 and 2012 Studies and the opinions based on those Studies are reliable. The answer to that question is "yes."

B. Reasonable jurist

The Court must now consider whether, in light of the new reliable evidence, it is more likely than not that no reasonable jurist would have ordered Rick's civil commitment in 2004. House, 547 U.S. at 537, 126 S.Ct. 2064. In applying the actual innocence analysis, the ordinary rule is that Courts are not limited to the trial record or the reliable new evidence. Id. Rather, the Court conducts a comprehensive assessment of any evidence probative of whether the petitioner is actually innocent. Schlup, 513 U.S. at 327-28, 115 S.Ct. 851. Courts may even consider evidence previously excluded, as courts are not bound by the rules of evidence that would govern at trial. Id. "[A]ll the evidence, old and new, incriminating and exculpatory" may be considered. House, 547 U.S. at 538, 126 S.Ct. 2064 (quotations omitted).²² There is precious little guidance, however, as to how the Court should weigh the new evidence it considers. All we are told, generally, is that the Court should decide whether, in light of the new evidence, it is more likely than not that no reasonable fact finder would have found the petitioner "guilty."

22 Even if the Court determined the 2009 and 2012 Studies were not "new, reliable" evidence, the Court could nonetheless consider the studies to determine whether it is more likely than not that no reasonable jurist would have civilly committed Rick. House, 547 U.S. at 538, 126 S.Ct. 2064 (quotations omitted).

In this case, the Court is adopting and applying the actual innocence gateway to the context of a civil commitment rather than to a criminal conviction. Adapting this criminal standard and applying it in the civil commitment context creates analytical complexities for which there is virtually no guidance. In a criminal context, the operative event is the petitioner's conviction offense. In that context the petitioner would have to establish, for example, that it is more likely than not that no reasonable juror would vote to convict the petitioner of murder. This inquiry is made simpler by the fact that it has both a temporal component—the offense happened in the past—and that it is a matter of fact rather than opinion—the defendant either did or did not commit, e.g., the homicide. Establishing the actual innocence gateway in a civil commitment context is fundamentally different and far more complex. The operative event is not a past occurrence but rather a prediction of future events—how likely is it that the petitioner will commit harmful sexual conduct in the future. The Court does not analyze whether reasonable jurors would conclude the petitioner did or did not do something, but must analyze whether a reasonable jurist would predict the likelihood of future misconduct to be high. This is not a matter of fact, but rather a predictive opinion.²³ In re Civil Commitment of Ince, 847 N.W.2d 13, 24 (Minn. 2014) (stating decision of commitment court relies heavily on expert opinions).

23 Parties may also call expert witnesses in criminal proceedings; however, the inquiry still focuses on determining historic facts. Here the experts opine about whether volitional events will occur in the future.

Therefore, in applying the gateway innocence analysis to this unique endeavor (i.e., whether it is more likely than not that no reasonable jurist would have civilly committed Rick) the Court must determine the scope of evidence it will consider and how it will assess that evidence. In this case the Court finds that the only principled way to make this determination is to consider only the impact of the recantations on the evidence admitted at Rick's commitment trial, giving no consideration to the now-recanted testimony. The Court does not consider either Drs. Phenix's or Hoberman's opinion because neither opinion assists this Court in resolving the issue whether in light of the new evidence no reasonable jurist would have committed Rick in 2004. This is so because—as is readily apparent from this habeas action—each party here could have presented any number of opinions by experts who did not testify in 2004, to support its position. Moreover, this Court is not well positioned to decide which new opinion—Drs. Phenix's or Dr. Hoberman's—carries the day on the issue whether Rick met the criteria for commitment in 2004. The determination whether someone does or does not meet the criteria for commitment under Minnesota statutes is uniquely the province of the state courts, who are not only charged with making such determination, but have developed the expertise to do so. Therefore, the Court does not consider either Drs. Phenix's or Hoberman's opinions regarding whether Rick met the SDP criteria in 2004.²⁴ Rather, this Court's decision is based on the state court record in which Drs. Sweet's and Alberg's now-recanted 2004 opinions are replaced with their current opinions,²⁵ and Dr. Alsdurf's opinions—then and now—are weighed in the balance. Put another way, this Court has used the 2009 and 2012 Studies, and the opinions of Drs. Phenix and Hoberman to test whether the recantations are reliable. As noted, the Studies are new and reliable evidence. Dr. Phenix's opinion is new and reliable evidence. Together, and despite the critiques by Drs. Hoberman and Alsdurf, this evidence establishes that the recantations by Drs. Sweet and Alberg are reliable new evidence that must now be run through the gateway innocence analysis.

24 For this reason, this Court limited the testimony of Dr. Hoberman to the question only whether Rick's evidence was new and reliable. As noted, based on all the new evidence, this Court has found that Drs. Sweet's and Alberg's recantations are reliable.

25 In a similar fashion, the Court does not weigh the 2009 and 2012 Studies in determining whether no reasonable jurist would have committed Rick in 2004. Again, it is abundantly clear that numerous recidivism studies since 2004 could be brought to bear on the question of Rick's likelihood of committing future acts of harmful sexual conduct. As with the opinions of Drs. Hoberman and Phenix, the Studies were helpful to the Court in assessing whether Drs. Sweet's and Alberg's new opinions are reliable. That evidence—the recantations—is the only new evidence that can be used by this Court to address, in a principled way, the second prong of the actual innocence gateway test.

With this evidentiary record in mind, the Court now determines whether, in light of the new recantation evidence, it is more likely than not that no reasonable jurist would have found in 2004 that Rick met the criteria to be civilly committed. To succeed under the actual innocence gateway, Rick's new evidence must convincingly undermine Hennepin County's 2004 commitment case. Larsen v. Soto, 742 F.3d 1083, 1096 (9th Cir. 2013). Definitive, affirmative proof that Rick did not meet the SDP criteria is not strictly required. Id. The Court's task is to "make a probabilistic determination" about what a reasonable jurist would do based on a total record. House, 547 U.S. at 538, 126 S.Ct. 2064. A "reasonable jurist" is one who fairly considers the evidence presented and conscientiously obeys Minnesota law requiring clear and convincing evidence that Rick is highly likely to commit harmful sexual conduct in the future. See Schlup, 513 U.S. at 299-300, 115 S.Ct. 851. The Court must assess how that jurist "would react to the overall, newly supplemented record" and if necessary, this may include consideration of "the credibility of the witnesses presented" at Rick's commitment trial. House, 547 U.S. at 538-39, 126 S.Ct. 2064 (quotation omitted).

Rick argues that the court-appointed examiners' recantations so undermine the remaining expert testimony from his commitment hearing that it eliminates the basis for the state court's decision to commit him. In response, Hennepin County asserts that Rick's new evidence does not show that it is "more likely than not" that no jurist would have committed Rick because other evidence in the state record could support a judgment of commitment.

Under Minnesota state law, a petitioner seeking to civilly commit someone as an SDP must establish by clear and convincing evidence that the individual has a psychological disorder making the person highly likely to commit harmful sexual conduct in the future. Minn. Stat. § 253D.02, subd. 18. When predicting whether a person poses such a danger to the public, courts should consider all relevant and reliable evidence, including the "Linehan factors" which are:

(a) the person's relevant demographic characteristics . . . ; (b) the person's history of violent behavior . . . ; (c) the base rate statistics for violent behavior among individuals of this person's background . . . ; (d) the sources of stress in the environment . . . ; (e) the similarity of the present or future context to those contexts in which the person has used violence in the past; and (f) the person's record with respect to sex therapy programs.

In re Linehan, 518 N.W.2d 609, 611 (Minn. 1994). "[T]he district court must make a good faith attempt to isolate the most important factors in predicting harmful sexual conduct." Ince, 847 N.W.2d at 23 (alteration and quotation omitted). As the trier of fact, the state court is "in the best position to determine the weight to be attributed to each factor," and performs the "critical function" of evaluating the credibility of witnesses in cases that "rely so heavily on the opinions of experts." Id. at 23-24 (citations omitted). "Where the findings of fact rest almost entirely on expert testimony, the district court's evaluation of credibility is of particular significance." In re Civil Commitment of Duvall, 916 N.W.2d 887, 895 (Minn. App. 2018) (quotation and alteration omitted).

This Court's analysis begins with the recognition that even in 2004 Rick's case was uniquely close. In deciding that Rick met the SDP criteria, the state court considered Rick's record in therapy programs and the experts' opinions on how his treatment record impacted his likelihood of reoffending.²⁶ Resp. App. 6-7, 11, 13, 17. At the commitment trial, Dr. Alberg opined that Rick's withdrawal from treatment suggested a higher risk of recidivism. Likewise, Dr. Sweet stated that Rick's decision to withdraw from treatment was a "significant factor" in his 2004 opinion. Drs. Sweet and Alberg, on whom the state court heavily relied, now recant their testimony and conclude Rick did not meet the criteria as an SDP when committed. Their recantations are based on research relating to sex offender recidivism rates and research regarding sex offender treatment programs' impact on recidivism rates. In light of this new evidence, it is very likely the state court would have ascribed considerably less weight to Rick's "record with respect to sex therapy" at the time of his commitment.²⁷ Linehan, 518 N.W.2d at 611.

26 Though this Court is not reviewing the state court decision for error, but rather making a probabilistic determination about what a reasonable jurist would do; House, 547 U.S. at 538, 126 S.Ct. 2064, it is helpful to discuss the evidence presented to the state court for its consideration.

27 The experts' consideration of the treatment program factors likely "double counted" that factor because the actuarial assessments already accounted for it. See Ince, 847 N.W.2d at 24 (cautioning trial courts "to be wary of the potential factor repetition that can result from considering the Linehan factors in addition to multiple actuarial assessments that use different approaches based on factors that are the same as or similar to the Linehan factors").

In Souter, the petitioner sought habeas relief based on newly discovered evidence. 395 F.3d at 583-84. The petitioner had been convicted of murder on the theory that he used a whiskey bottle to inflict two fatal blows to the victim, who was found dead on a road. Id. at 582-83. At trial, one pathologist testified the victim's wounds were inconsistent with having been hit by a passing car, and that instead the victim's injuries were inflicted by a sharp edge on a whiskey bottle like the one petitioner admitted to having earlier that evening. Id. Two other expert witnesses concluded that the wounds could have been inflicted by the bottle, supporting the first pathologist's conclusion. Id. The petitioner presented one defense witness who opined the bottle did not cause the injuries. Id. at 583. The petitioner was convicted and later filed an untimely habeas petition, relying on evidence discovered after trial. The new evidence included recantations by the two expert witnesses, previously unavailable photos that undermined the pathologist's testimony, and evidence that the whiskey bottle could not have had a sharp edge. Id. at 583-84. The appellate court concluded that the totality of this evidence undermined the testimony of the pathologist who concluded the wounds were inflicted by the bottle and held this was a rare and extraordinary case in which the actual innocence gateway applied. Id. at 590-91.

As in Souter, this Court concludes that Rick's case is rare and extraordinary. In his habeas petition, Rick has presented new evidence that raises sufficient doubt about whether he met the statutory criteria to be civilly committed and that undermines confidence in the result of his commitment trial. The evidence relied on most by the state court was the testimony of the court-appointed examiners and Dr. Alsdurf, who all testified that Rick met the statutory criteria to be civilly committed as an SDP. Since that time Drs. Sweet and Alberg—both of whom interviewed Rick—have recanted their trial testimony and unambiguously stated that had they known then what they know now, they would not have opined he met the commitment criteria. Dr. Alsdurf, who did not interview Rick, and who unpersuasively doubled down on his initial trial testimony, nonetheless acknowledged that even in 2004 Rick's was a very close case. Unlike in Souter, where a trial witness initially testified in the petitioner's favor that the bottle had not caused the injuries, in Rick's case, none of the expert witnesses initially opined that he did not meet the SDP criteria. This meant that, unlike the Souter trial court, the commitment court did not have contradictory evidence to consider. Thus, the recantations in this case undermine Rick's commitment even more completely than did the evidence in Souter's case undermine his conviction.

Additionally, the weight given by the state court to the testimony of the experts cannot be ignored. The state court apparently credited Drs. Sweet's and Alberg's testimony more than Dr. Alsdurf's because the court followed their recommendation that Rick's commitment be stayed so he could obtain outpatient treatment. Resp. App. 12, 14. By ordering Rick's civil commitment stayed, the state court rejected Dr. Alsdurf's opinion that Rick could only obtain appropriate treatment as a patient at MSOP. Id. at 13. The state court also noted that Dr. Sweet was persuasive, that Dr. Alberg's opinion aligned with Dr. Sweet's opinion, that Dr. Alsdurf was retained by Hennepin County, and that, unlike Drs. Sweet and Alberg, Dr. Alsdurf did not personally meet with Rick. Id. at 10-13. Dr. Alsdurf's testimony at trial was based on a review of Rick's commitment records without any in person clinical observations. Id. at 14. The state court also found the testimony of Penny Zwecker—who testified against commitment—highly credible. Id. at 10. Dr. Zwecker had testified the "political climate" favored commitment, but a reasonable jurist would not base a decision to commit Rick on political pressure to do so.

In considering the new evidence, for a court to conclude in this uniquely close case that Rick met the SDP criteria it would have to reject the testimony of its own two court-appointed examiners as well as the testimony of MNDOC's Civil Commitment Review Coordinator, all in favor of the testimony of an expert privately retained by an interested party.²⁸ It is unlikely any reasonable jurist would do so.

28 Even if a reasonable jurist discounted both Drs. Alberg's and Alsdurf's opinions as biased (Dr. Alberg was appointed at the suggestion of Rick's counsel; Dr. Alsdurf was retained by Hennepin County), the two independent experts (Zwecker and Dr. Sweet) opine Rick did not meet the SDP criteria in 2004.

Furthermore, newly presented evidence may call into question the credibility of the commitment trial witnesses. See Schlup, 513 U.S. at 330, 115 S.Ct. 851. In such a case, the habeas court may make credibility assessments. Id. A witness recanting his trial testimony certainly falls under that scope. Souter, 395 F.3d at 593 n.8. That the recantation may be cumulative to Rick's evidence does not minimize its effectiveness in weakening Hennepin County's case and calling into question Dr. Alsdurf's credibility. Id. Considering all the evidence, particularly that the two court-appointed examiners recanted their expert opinions that Rick met the SDP criteria in 2004, a reasonable jurist would have difficulty adopting Dr. Alsdurf's opinion to the contrary. It is more likely than not that, in light of Rick's new reliable evidence, no reasonable jurist would have found by clear and convincing evidence that Rick met the standard for commitment.

Even if this Court were to consider Drs. Phenix's and Hoberman's opinions, the result would be the same. Dr. Phenix's report unambiguously calls into question the competency of Dr. Alsdurf, the only remaining expert who stands behind his previous opinion that Rick met the SDP criteria. Heisler Aff. 99 (stating she found Dr. Alsdurf's report "less coherent" than other reports and parts of his report were "categorically untrue"). Dr. Phenix's opinion supports this Court's conclusion, while Dr. Hoberman's opinion supports Hennepin County's position. Even though this Court has found Dr. Hoberman's testimony to be credible, it was less persuasive because he at times attempted to avoid answering questions directly and at times gave non-responsive argumentative testimony in a convoluted way. Dr. Hoberman also testified that out of the 120 cases in which he was appointed a court-appointed examiner, he has only recommended the person not be committed in 10-12 cases. This means Dr. Hoberman recommends commitment at least 90% of the time he serves as a court-appointed examiner. Tr. II, 117-18. Dr. Hoberman did not testify that Rick's case was not a close case. His testimony focused primarily on the limitations of research, which this Court fully appreciates and has factored into this decision. Therefore, even if this Court considered Drs. Phenix's and Hoberman's opinions, it would find that it is more likely than not that no reasonable jurist would have civilly committed Rick under the required clear and convincing evidence standard.

Hennepin County argues, citing the Linehan factors, that there was sufficient evidence in the record for the state court to civilly commit Rick. Resp. Final Br. 18-19. The Supreme Court has made clear, however, that the actual innocence gateway differs from the criminal sufficiency-of-the-evidence standard. House, 547 U.S. at 538, 126 S.Ct. 2064. Unlike the actual innocence gateway, the sufficiency standard requires a court to determine whether the trial evidence, viewed in the light most favorable to the prosecution, could allow any reasonable trier of fact to find a charged crime proved beyond a reasonable doubt. Id. (emphasis added) (citing Jackson v. Virginia, 443 U.S. 307, 326, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979)). If the sufficiency standard were enough to defeat an actual innocence claim, the actual innocence gateway doctrine would be meaningless. Larsen, 742 F.3d at 1099. In fact, the Supreme Court has expressly held a "petitioner's showing of innocence is not insufficient solely because the trial record contained sufficient evidence to support the jury's verdict." Schlup, 513 U.S. at 331, 115 S.Ct. 851. Rick's new reliable evidence casts doubt on his commitment "by undercutting the reliability of the proof of guilt," whether or not it "affirmatively prov[es] innocence." Sistrunk v. Armenakis, 292 F.3d 669, 673 (9th Cir. 2002). The question this court must decide is not whether the record is sufficient to commit Rick. The question is whether, considering all the evidence including Rick's new evidence, a reasonable jurist would likely have committed Rick. This Court has made a probabilistic determination about what a reasonable jurist would do based on the entire habeas record and determined that it is more likely than not, in light of the pertinent new reliable evidence, that no reasonable jurist would have civilly committed Rick in 2004. House, 547 U.S. at 537-38, 126 S.Ct. 2064.

When someone under state commitment challenges that commitment in federal court, we must be "careful to limit the scope of federal intrusion into state . . . adjudications and to safeguard the States' interest in the integrity of [state] proceedings." Williams v. Taylor, 529 U.S. 420, 436, 120 S.Ct. 1479, 146 L.Ed.2d 435 (2000). "Congress passed AEDPA, including its stringent statute-of-limitations provision, to further these principles of comity, finality, and federalism." Larsen, 742 F.3d at 1099 (quotation and alteration omitted). Even so, "the Supreme Court has repeatedly recognized, in appropriate cases these principles of comity and finality . . . must yield to the imperative of correcting a fundamentally unjust incarceration."²⁹ Id. (quotation omitted). This is such a case even though it is not a case of conclusive exoneration.³⁰ Some evidence—Dr. Alsdurf's testimony—may still support civil commitment. Yet the central commitment evidence—Drs. Sweet's and Alberg's testimony—has changed, and Rick has put forward substantial evidence supporting his position that he did not meet the statutory requirements to be committed in 2004. The Court concludes that it is more likely than not that no reasonable jurist considering all the evidence, including Rick's reliable new evidence, would have found by clear and convincing evidence that Rick met the SDP statutory criteria to be civilly committed in 2004.

29 See House, 547 U.S. at 548-54, 126 S.Ct. 2064 (finding actual innocence gateway where evidence raised inference another could have been murderer); Larsen, 742 F.3d at 1099 (petitioner established actual innocence gateway); Souter, 395 F.3d at 581-84, 591-92, 596 (same); Carriger v. Stewart, 132 F.3d 463, 465, 471, 478 (9th Cir. 1997), cert. denied, 523 U.S. 1133, 118 S.Ct. 1827, 140 L.Ed.2d 963 (1998) (same); Lisker v. Knowles, 463 F. Supp. 2d 1008, 1018-28 (C.D. Cal. 2006) (same); Garcia v. Portuondo, 334 F. Supp. 2d 446, 455-56 (S.D.N.Y. 2004) (same); Schlup v. Delo, 912 F. Supp. 448, 451-55 (E.D. Mo. 1995) (same); but see Lee v. Lampert, 653 F.3d 929, 943-46 (9th Cir. 2011) (finding no actual-innocence exception where evidence was insufficient to overcome otherwise convincing proof of guilt); Sistrunk, 292 F.3d at 675-77 (same).

30 In contrast to a criminal conviction where the adjudication of guilt is based on the presence or absence of an historical fact, it is difficult to conceive of an exoneration in the context of a civil commitment, which is based on predictive opinions as to a person's likely future behavior.

III. Due Process

The Court now turns to the merits of Rick's due process claim. Rick asserts that "facts discovered subsequent to his commitment hearing, . . . have undermined and discredited critical risk assessment tools and clinical assumptions upon which his judgment of commitment was based, and which have rendered his initial commitment a fundamental miscarriage of justice." 2d Am. Pet. 3-4. Rick argues that his commitment was based on disputed expert testimony that has now been recanted, undermining the fundamental fairness of the proceedings. Pet. Final Br. 24-31. Hennepin County argues that Rick's claim is without merit because the state court considered evidence other than Rick's treatment history and actuarial instruments and because the record as a whole support civil commitment. Resp. Final Br. 18-19.

Under 28 U.S.C. § 2254(d), "any claim that was adjudicated on the merits in State court proceedings is entitled to deference by the federal courts." Worthington v. Roper, 631 F.3d 487, 495 (8th Cir. 2011) (quotation omitted). Here, the state court did not adjudicate Rick's habeas claim on its merits and, therefore, 28 U.S.C. § 2254(d) does not apply. Taylor v. Bowersox, 329 F.3d 963, 968 (8th Cir. 2003). When federal habeas grounds have not been "adjudicated on the merits" by the state court, the pre-AEDPA standard for habeas review governs. Gingras v. Weber, 543 F.3d 1001, 1003 (8th Cir. 2008). The pre-AEDPA standard requires federal courts to review conclusions of law de novo, and to give the state court's factual findings a presumption of correctness. Stringer v. Hedgepeth, 280 F.3d 826, 829 (8th Cir. 2002); see also 28 U.S.C. § 2254(e)(1). To succeed on the merits, the petitioner must establish a "reasonable probability that the error complained of affected the outcome of the [proceedings]." Robinson v. Crist, 278 F.3d 862, 865-66 (8th Cir. 2002).

Under the pre-AEDPA standard for habeas review, "[n]ewly discovered evidence relevant to constitutionality of a state prisoner's detention is a ground for federal habeas relief." Lewis v. Erickson, 946 F.2d 1361, 1362 (8th Cir. 1991). Federal courts "grant habeas relief based on newly discovered evidence if the evidence would probably produce an acquittal on retrial." Id. (quotations omitted) (citing Sanders v. Sullivan, 863 F.2d 218, 224-25 (2d Cir. 1988) (failure to cure conviction after credible recantation of material testimony violates due process if recantation "would most likely affect the verdict")). Recanted testimony is grounds for relief when it either bears on a witness's credibility or directly on the defendant's guilt. Id.

The Court turns to whether Rick's newly discovered evidence would most likely affect the commitment judgment.³¹ The Court focuses on Drs. Sweet's and Alberg's now-recanted testimony that Rick met the statutory criteria to be committed. The recantations bear directly on whether Rick could, under Minnesota law, be civilly committed and therefore detained at MSOP. Because this is analogous to a defendant's guilt, the recantations may be grounds for relief if they "would probably produce [no commitment judgment] on retrial." Id.

31 The Court does not consider Dr. Hoberman's report or testimony to determine Rick's due process claim because it is not Rick's "newly discovered evidence." Lewis, 946 F.2d at 1362. Even if considered, however, this evidence would not change the Court's conclusion.

The state court's 2004 Order indicates the experts' now-recanted opinions, which were based in part on actuarial assessments derived from old research and a certain understanding of how treatment impacts recidivism, played a central role in Rick's commitment. See Ince, 847 N.W.2d at 24 (noting Minnesota courts rely heavily on expert opinions in civil commitment cases). The state court specifically addressed Rick's withdrawal from treatment and Drs. Sweet's and Alberg's opinions, concluding that Rick's "moderate risk of recidivism combined with not completing sex offender treatment . . . proved by clear and convincing evidence that there is a likelihood of [Rick] reoffending." Resp. App. 17. The state court also addressed Dr. Alsdurf's conclusion that Rick met the criteria as an SDP, "had not completed treatment," had a high likelihood of reoffending, and should be committed to MSOP. Id. at 13. Based on this evidence, the court found that Rick met the criteria as an SDP. Id. at 14.

The record shows that this was a very close case. The 2019 recantations, which rely on the 2009 and 2012 Studies and Dr. Phenix's report, are critical because Drs. Sweet and Alberg now conclude that Rick did not meet the criteria as an SDP at the time of his initial commitment. The court-appointed experts' now-recanted testimony in support of civil commitment overwhelmed the fairness of Rick's trial because the state court recognized that Drs. Sweet and Alberg strongly supported Rick remaining in the community, that the court-appointed examiners were beholden to the court, and that they personally interviewed Rick. For these reasons and those outlined in the actual innocence analysis, the Court does not believe another reasonable jurist would commit Rick. See Lewis, 946 F.2d at 1363. Thus, there is a reasonable probability the reliance placed on the court-appointed examiners' now-recanted opinions affected the outcome of Rick's trial. Robinson, 278 F.3d at 865-66.

The County cautions this Court about disturbing the finality of judicial proceedings and the comity afforded to state judgments. Rick's petition is not about whether the state court that committed him reached the proper decision. It is instead about the right to a fair trial, which is "the most fundamental of all freedoms." Estes v. Texas, 381 U.S. 532, 540, 85 S.Ct. 1628, 14 L.Ed.2d 543 (1965). The state court's substantial reliance on the court-appointed examiners' now, persuasively-recanted testimony renders his trial fundamentally unfair and Rick's commitment a miscarriage of justice. Accordingly, the Court recommends Rick's writ of habeas corpus be granted.

RECOMMENDATION

For the reasons set forth above, the Court RECOMMENDS THAT: Petitioner Darrin Scott Rick's Second Amended Petition [Dkt. No. 45] be GRANTED.

NOTICE

Filing Objections: This Report and Recommendation is not an order or judgment of the District Court and is therefore not appealable directly to the Eighth Circuit Court of Appeals. Under Local Rule 72.2(b)(1), "a party may file and serve specific written objections to a magistrate judge's proposed finding and recommendations within 14 days after being served a copy" of the Report and Recommendation. A party may respond to those objections within 14 days after being served a copy of the objections. LR 72.2(b)(2). All objections and responses must comply with the word or line limits set for in LR 72.2(c).