State v. Foster (Concurrence)

• Court: Supreme Court of Connecticut
• Decided: June 19, 2025
• Docket: SC20829
• Status: Published

Opinion

************************************************ The “officially released” date that appears near the beginning of an opinion is the date the opinion will be published in the Connecticut Law Journal or the date it is released as a slip opinion. The operative date for the beginning of all time periods for the filing of postopin- ion motions and petitions for certification is the “offi- cially released” date appearing in the opinion. All opinions are subject to modification and technical correction prior to official publication in the Connecti- cut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports. In the event of discrepancies between the advance release version of an opinion and the version appearing in the Connecti- cut Law Journal and subsequently in the Connecticut Reports or Connecticut Appellate Reports, the latest version is to be considered authoritative. The syllabus and procedural history accompanying an opinion that appear in the Connecticut Law Jour- nal and subsequently in the Connecticut Reports or Connecticut Appellate Reports are copyrighted by the Secretary of the State, State of Connecticut, and may not be reproduced or distributed without the express written permission of the Commission on Official Legal Publications, Judicial Branch, State of Connecticut. ************************************************ Page 0 CONNECTICUT LAW JOURNAL 0, 0

2 ,0 0 Conn. 0 State v. Foster

ECKER, J., concurring in part and concurring in the judgment. The modern approach to supervising the detention, treatment, and ultimate release of individuals deemed not guilty by reason of mental disease or defect, or insanity (acquittees), outside the confines of the cor- rectional system has dual roots in the progressive reforms of the 1960s and 1970s and the backlash that followed highly publicized crimes such as the attempted assassination of President Reagan in 1981 by John Hin- ckley, Jr. See, e.g., M. Norko et al., ‘‘Assessing Insanity Acquittee Recidivism in Connecticut,’’ 34 Behav. Sci. & L. 423, 423–25 (2016); J. Rogers & J. Bloom, ‘‘The Insan- ity Sentence: Oregon’s Psychiatric Security Review Board,’’ 3 Behav. Sci. & L. 69, 70–71 (1985). Connecticut has sought to embrace the best practices in the field, having adopted a version of the highly touted Oregon model in 1985. See M. Norko et al., supra, 424. At its best, this model affords mentally ill offenders humane treatment for their illnesses and ensures appropriate and informed oversight of confinement and discharge decisions, rather than simply imposing indeterminate commitment before releasing them without any assur- ance that they will successfully reenter society. See M. Norko et al., supra, 424–25; J. Rogers & J. Bloom, supra, 71, 84. At its worst, unfortunately, this approach in practice has subjected individuals suffering from men- tal illness to unjustifiably prolonged—and, at times, even abusive—deprivations of liberty based on specula- tive fears unsupported by sufficient evidence that they remain dangerous to themselves or the public. Connect- icut has not entirely avoided these pitfalls.1 1 See, e.g., State v. Cusson, 210 Conn. App. 130, 135, 269 A.3d 828 (describing abuses), cert. denied, 343 Conn. 913, 274 A.3d 114 (2022); see also, e.g., Final Report of the Task Force To Review and Evaluate CVH and WFH, the Psychiatric Security Review Board, and Behavioral Health Care Definitions (December 16, 2021) p. 24, available at https://www.cga.ct.gov/ph/tfs/20190426_ CVH%20Whiting%20Task%20Force/CVH%20Whiting%20Final%20Report.pdf (last visited August 15, 2025) (‘‘[t]here . . . was unanimous concern expressed by the members of the [t]ask [f]orce about the lengthy periods 0, 0 CONNECTICUT LAW JOURNAL Page 1

0 Conn. 0 ,0 3 State v. Foster

Navigating these challenging waters is largely the work of the legislature and the various public and pri- vate agencies tasked with supervising the care and treat- ment of acquittees. See, e.g., Sastrom v. Psychiatric Security Review Board, 291 Conn. 307, 331 and n.24, 968 A.2d 396 (2009). But the courts also play a vitally important role in the process by safeguarding the liberty interests of acquittees who no longer need commitment. Specifically, the United States Supreme Court has made it ‘‘clear that commitment for any purpose constitutes a significant deprivation of liberty that requires due process protection’’ and ‘‘has recognized involuntary commitment to a mental institution, in particular, as involving more than a loss of freedom from confinement . . . due to its stigmatizing consequences, and the potential exposure to invasive, compulsory medical and psychiatric treatment.’’ (Citation omitted; internal quo- tation marks omitted.) State v. Metz, 230 Conn. 400, 412–13, 645 A.2d 965 (1994). In applying those principles to Connecticut’s own system, this court has explained that, ‘‘[d]espite the substantial degree of legislative dis- cretion’’; id., 424; ‘‘due process . . . imposes significant constitutional constraints on involuntary commitments.’’ Id., 413. The acquittee in this case, Franklin Foster, argues that (1) the Psychiatric Security Review Board (board) and the trial court relied on impermissible facts, factors, and assumptions in concluding that he ‘‘would consti- tute a danger to himself or others’’ if discharged from the board’s custody; General Statutes § 17a-593 (c); (2) the danger requirement imposed by § 17a-593 (c) is unconstitutionally vague, insofar as it is so devoid of of commitment placed [on] acquittees found [not guilty by reason of insan- ity]’’); D. Tepfer, CT Settles Psychiatric Hospital Abuse Lawsuit for $9 Million, Conn. Post, June 30, 2022, available at https://www.ctpost.com/news/article/CT- settles-psychiatric-hospital-abuse-lawsuit-for-17276464.php (last visited August 15, 2025). Page 2 CONNECTICUT LAW JOURNAL 0, 0

4 ,0 0 Conn. 0 State v. Foster

standards and guidance as to necessarily result in arbi- trary and discriminatory enforcement; and (3) the trial court’s finding of continued dangerousness was unsup- ported by the record and, therefore, clearly erroneous. As I will explain, I largely agree with the acquittee’s first contention. Much of the trial court’s analysis supporting the acquittee’s ongoing commitment was conclusory or, at best, marginally relevant. With respect to the vagueness challenge, I hope to provide some clarifica- tion of the judicial gloss that this court applied to the statutory scheme in cases such as State v. Putnoki, 200 Conn. 208, 510 A.2d 1329 (1986), and State v. Metz, supra, 230 Conn. 400, to assist the board and trial courts in complying with all statutory and regulatory require- ments and steering clear of the constitutional bound- aries. With respect to the outcome of the present case, I agree with the majority that it ‘‘may well approach the outer limits of when continued commitment to the custody of the board is justified.’’ Text accompanying footnote 14 of the majority opinion. I also agree with the majority that the record is minimally sufficient to support the trial court’s findings and conclusions, in light of our deferential standard of review. Especially given that the decision at issue will soon expire even if upheld; see footnote 7 of the majority opinion; my primary focus is on providing sufficient guidance to ensure that future discharge determinations, with respect to both this acquittee and others, fully comply with the law. I Before I address the acquittee’s case, I offer a few brief observations to better frame my analysis. A First, I do not envy the task of the board and the trial court in these matters. They are charged by statute with predicting a future that defies prognostication. It 0, 0 CONNECTICUT LAW JOURNAL Page 3

0 Conn. 0 ,0 5 State v. Foster

is unrealistic to expect anyone to determine—by clear and convincing evidence, no less—that an individual like the acquittee, who engaged in a single criminal outburst at the age of twenty-four, when his disease had just emerged and was untreated, is likely to reoffend if he is discharged from the board’s custody two dozen years later, after having spent most of his adult life living and receiving treatment in mental health facilities.2 Adding to the difficulty inherent in the task of pre- dicting dangerousness is the sometimes vexing question of whether the danger is causally related to one of the acquittee’s diagnosed psychiatric disabilities or, rather, is a result of, say, immaturity, or low frustration toler- ance, or a self-defeating fear of independence, or low intelligence,3 or any of the myriad other reasons that 2 As the acquittee notes, both science and law have long cast serious doubt on clinicians’ ability to accurately predict the future, violent behavior of acquittees or other populations, concluding that dangerousness is systemically overpre- dicted. See, e.g., State v. Putnoki, supra, 200 Conn. 219–20; C. Slobogin, ‘‘Danger- ousness and Expertise,’’ 133 U. Pa. L. Rev. 97, 98–127 (1984). In fairness, statistically based models and instruments have a better—and improving—track record, with some measures apparently demonstrating a predictive accuracy approaching 70 percent. See M. Ogonah et al., ‘‘Violence Risk Assessment Instru- ments in Forensic Psychiatric Populations: A Systematic Review and Meta- Analysis,’’ 10 Lancet Psychiatry 780, 784 (2023) (meta-analysis of Historical Clini- cal Risk Management-20 version 2 demonstrates pooled accuracy of 69 percent predicting violent recidivism); see also S. Ægisdóttir et al., ‘‘The Meta-Analysis of Clinical Judgment Project: Fifty-Six Years of Accumulated Research on Clinical Versus Statistical Prediction,’’ 34 Counseling Psychologist 341, 368 (2006) (dispar- ity between clinical and statistical predictions is greatest with respect to predic- tions of violence and reoffense); J. Stankowski, ‘‘Acquittal by Reason of Insanity Is a Positive Outcome for Defendants Who Cannot Be Restored,’’ 48 J. Am. Acad. Psychiatry & L. 244, 246–47 (2020) (surveying literature). Statistics aside, the issue in any given case is not whether the assessments are reliable on average, but whether the prediction can be made with the required degree of confidence based on the facts of that particular case. 3 As this opinion will discuss, there is evidence in the record supporting the conclusion that most, if not all, of the ‘‘inappropriate’’ conduct on which the board and the trial court based their respective decisions in the present case resulted from these sorts of character traits rather than from the acquittee’s diagnosed psychiatric disabilities. Department of Mental Health and Addiction Page 4 CONNECTICUT LAW JOURNAL 0, 0

6 ,0 0 Conn. 0 State v. Foster

innumerable sane people commit crimes each day.4 The requirement that an acquittee be found dangerous as a result of his psychiatric disabilities has both statutory and constitutional dimensions. The relevant Connecti- cut statutes and regulations indicate that discharge may be denied only if the acquittee ‘‘has psychiatric disabili- ties . . . to the extent that such acquittee’s discharge . . . would constitute a danger to the acquittee or oth- ers . . . .’’ (Emphasis added.) General Statutes § 17a- 580 (10); see also General Statutes § 17a-580 (11); Gen- eral Statutes § 17a-593 (c); Regs., Conn. State Agencies § 17a-581-2 (a) (11). It seems clear to me that the legisla- ture’s use of the phrase ‘‘to the extent that,’’ rather than ‘‘and,’’ requires a causal nexus between the psychiatric disabilities and the resulting danger. See State v. Cuvel- ier, 175 Conn. 100, 102 and n.2, 109–110, 394 A.2d 185 (1978) (treating prior version of statute, with substan- tially similar language, as requiring causal relationship). From a constitutional standpoint, both the United States Supreme Court and this court have stated that an acquittee’s dangerousness must be causally related to his mental illness to permit continued confinement.5 Services staff testified at the acquittee’s hearings regarding the difficulty of parsing out to what extent, if any, the symptoms of the acquittee’s schizophrenia contribute to difficulties he may have had adjusting to community living. 4 For example, Alec Buchanan, a consulting forensic psychiatrist with the Department of Mental Health and Addiction Services, provided some of the most nuanced testimony at the acquittee’s hearings regarding the difficulties inherent in trying to predict the acquittee’s future dangerousness and likely future medication compliance, given that he committed the index offenses just after his mental illness had emerged and most of the behavioral problems he demonstrated posthospitalization had occurred early on, when his disease was not yet under control. Even Buchanan, however, was unable to address the extent to which the acquittee’s behavioral problems are the result of immaturity and attention issues rather than psychiatric disabilities. 5 ‘‘A finding of dangerousness, standing alone, is ordinarily not a sufficient ground [on] which to justify indefinite involuntary commitment. . . . [A]dded statutory requirements serve to limit involuntary civil confinement to those who suffer from a volitional impairment rendering them dangerous beyond their control.’’ (Citations omitted; emphasis added.) Kansas v. Hendricks, 521 U.S. 346, 358, 117 S. Ct. 2072, 138 L. Ed. 2d 501 (1997). In State v. March, 265 Conn. 697, 830 A.2d 212 (2003), this court confirmed that the constitution imposes the 0, 0 CONNECTICUT LAW JOURNAL Page 5

0 Conn. 0 ,0 7 State v. Foster

I would imagine that these and other difficulties inherent in the task might act as a strong incentive to prompt the board to follow a maximally conservative path and not to authorize discharge until confidence is high that every last demon has been exorcized and any risk of reoffense eliminated. Courts and commentators have recognized as much,6 and Department of Mental Health and Addiction Services (DMHAS) staff members acknowledged in this case that ‘‘we all take risks when we recommend someone for a transition to the commu- nity’’ and that ‘‘the hospital certainly wants to be conser- same requirement on § 17a-593 (c). See id., 716 (‘‘[A]s long as an acquittee has a mental illness that requires confinement for purposes of treatment and protection, his confinement to a psychiatric facility is reasonably related to the purpose of commitment and is, therefore, constitutional. . . . What is important is that the mental illness that the acquittee is currently diagnosed with be of the type of mental illness that might cause the acquittee to be dangerous if dis- charged.’’ (Citations omitted; emphasis added; internal quotation marks omit- ted.)); see also People v. Wilder, 33 Cal. App. 4th 90, 101, 39 Cal. Rptr. 2d 247 (1995) (‘‘requiring the prosecution to prove dangerousness because of mental disease, defect, or disorder . . . [en]sures that the extended commitment bears a reasonable relation to the purpose of commitment’’), review denied, California Supreme Court, Docket No. S046088 (June 14, 1995); Rinne v. Psychiatric Security Review Board, 297 Or. App. 549, 558, 443 P.3d 731 (2019) (proceeding on assumption that there must be ‘‘a causal connection’’ or ‘‘a nexus’’ between qualifying mental disease or defect and dangerousness). 6 See, e.g., State v. Hitt, 179 Or. App. 563, 573, 41 P.3d 434 (2002) (‘‘[I]t may be tempting to err on the side of caution and to continue a person’s commitment because you can never be sure. . . . [T]hat reasoning would, effectively, trans- form [an acquittee’s] commitment into a permanent commitment because no judge can ever be sure that the committed person will not engage in further violence. That reasoning, however, effectively shifts the burden of proof from the state to [the acquittee].’’ (Emphasis omitted; internal quotation marks omit- ted.)); J. Stankowski, ‘‘Acquittal by Reason of Insanity Is a Positive Outcome for Defendants Who Cannot Be Restored,’’ 48 J. Am. Acad. Psychiatry & L. 244, 248 (2020) (‘‘[Some] clinicians, aware of [their] responsibility not only to the patient, but the court and [the] community, may become so [risk averse that] they see little benefit to any release. In sum, although the link between insanity, mental disorder, and dangerousness has been shown to be weak, there are myriad strong pressures that conspire to keep insanity acquittees locked up indeterminately.’’ (Footnote omitted.)); Note, ‘‘The Guilty but Mentally Ill Verdict and Due Process,’’ 92 Yale L.J. 475, 482 (1983) (‘‘officials may wish to avoid releasing a person declared mentally ill and guilty, since the public may hold their institution respon- sible if the released person is subsequently accused criminally’’). Page 6 CONNECTICUT LAW JOURNAL 0, 0

8 ,0 0 Conn. 0 State v. Foster

vative . . . .’’ After all, if an acquittee is discharged after the maximum term of commitment,7 the best out- come will be one that attracts no attention; with the assistance of available services, the acquittee will grad- ually learn to readapt to life in the community and to manage his illness, without incident, on a more indepen- dent basis. The worst outcome, of course, is that he will decompensate and that tragic results will follow, with public opprobrium falling squarely on the board, the trial court, and anyone else whose judgment can be questioned in hindsight. But, while the decision to discharge an acquittee, in the face of such uncertainty, may be a difficult one, that’s the job description. It would, no doubt, be com- forting to many people if previously violent acquittees remain indefinitely committed until the treatment team, the board, and the trial court all can guarantee their future harmlessness. But we cannot allow the under- standable desire for certainty to distort the legal stan- dards established by statutory and constitutional command.8 The specific task for this court in the present 7 General Statutes § 17a-582 (h) creates a presumption that an acquittee will be discharged from the jurisdiction of the board upon the expiration of his ‘‘maximum term of commitment,’’ as established by the trial court, which cannot exceed the maximum sentence that could have been imposed if he had been convicted of the index offenses. See also General Statutes § 17a-582 (e) (1) (A). 8 In this case, one member of the board indicated that, although the board is sensitive to the various roadblocks that the system creates for acquittees seeking to reintegrate into society, he believed that his hands were tied because ‘‘the board’s mission is . . . mitigating risk.’’ That’s not quite right. The members of the board are not underwriters, quality control inspectors, or others whose mission it is to eliminate risk as much as possible given financial and other constraints. Rather, they are legally compelled to discharge an acquittee after the maximum term of commitment unless the state proves by clear and convincing evidence that he poses a danger. A security screener, whose job it is to allow anyone to enter unless they plainly pose a well-defined security risk, would be a more apt metaphor. These principles are fully consistent with subsection (g) of § 17a-593, which provides in relevant part that the trial court, in making its final determination as to whether to discharge the acquittee, must ‘‘[consider] that its primary concern is the protection of society and [that] its secondary concern is the safety and 0, 0 CONNECTICUT LAW JOURNAL Page 7

0 Conn. 0 ,0 9 State v. Foster

case is to articulate and enforce clear, consistent, and administrable standards under § 17a-593. B My second observation is that not only is dangerous- ness notoriously difficult to predict, but the concept of dangerousness itself is very difficult to pin down. See, e.g., State v. Putnoki, supra, 200 Conn. 219; State v. Gates, 198 Conn. 397, 403, 503 A.2d 163 (1986); State v. Krol, 68 N.J. 236, 258, 344 A.2d 289 (1975). What is clear is that both the relevant statutes and regulations, and the due process clauses of the federal and state constitutions, place substantive limits on the ability of the state to perpetually confine a mentally ill individual on the basis of some perceived dangerousness. What has not always been clear is whether this and other courts, in describing the standards and guidelines that govern a dangerousness determination, were purporting to construe the particular statute at issue or to place on that statute a judicial gloss so as to conform it to constitutional requirements. See, e.g., Foucha v. Louisi- ana, 504 U.S. 71, 77–78, 112 S. Ct. 1780, 118 L. Ed. 2d 437 (1992). Nor can expert psychiatric testimony and the constantly evolving scientific research that informs (or should inform) that testimony be disregarded, despite our cryptic statement that the ultimate determination well-being of the acquittee . . . .’’ I read this statement merely to underscore the statutory command that a trial court not discharge an acquittee who continues to pose an imminent danger to society, even if discharge might be in his individual best interest. The statutory guidance comes into play only at the end, after the court has made the relevant findings and determinations regarding the acquittee’s mental condition; see General Statutes § 17a-593 (g); and this court has made clear that the statute cannot be construed in such a way as to place a thumb on the scale in favor of indefinite, involuntary commitment or to otherwise offend due process. See State v. Metz, supra, 230 Conn. 419–26. If the ‘‘primary concern’’ language were construed more broadly, it would not only run afoul of Metz but would, in effect, nullify General Statutes § 17a-582 (h), which provides in relevant part that, by default, ‘‘the acquittee shall be immediately discharged at the expiration of the maximum term of commitment.’’ Page 8 CONNECTICUT LAW JOURNAL 0, 0

10 ,0 0 Conn. 0 State v. Foster

is a legal or societal decision, rather than a medical one. See, e.g., State v. Putnoki, supra, 200 Conn. 219, 221. Because the ultimate determination that the trial court must make is a legal one, it will be helpful to briefly restate the legal principles, factors, and stan- dards that are relevant to a determination of dangerous- ness in this context. First, to avoid constitutional peril, the state must carry its burden of demonstrating ‘‘that the acquittee is currently mentally ill and dangerous to himself . . . or others’’ by clear and convincing evi- dence. State v. Metz, supra, 230 Conn. 425. This burden is a ‘‘heavy’’ one. (Internal quotation marks omitted.) State v. Lenarz, 301 Conn. 417, 438, 22 A.3d 536 (2011), cert. denied, 565 U.S. 1156, 132 S. Ct. 1095, 181 L. Ed. 2d 977 (2012); see also Miller v. Commissioner of Cor- rection, 242 Conn. 745, 795, 700 A.2d 1108 (1997) (‘‘the clear and convincing evidence standard should operate as a weighty caution [on] the minds of all judges, and it [is not satisfied] whenever the evidence is loose, equivocal or contradictory’’ (internal quotation marks omitted)). The trial court cannot, in effect, reverse the burden of proof by requiring the acquittee to demon- strate to the court’s satisfaction that there is little or no risk that he ever will reoffend. See State v. Metz, supra, 421–22 (concluding, in light of policy and consti- tutional considerations, that ‘‘the legislature would not have intended to impose, even on an insanity acquittee, a never-ending burden of rebutting an indefinite pre- sumption of continuing insanity and dangerousness’’). Second, although the acquittee’s index offenses, as well as any other past violent behavior, are relevant factors; see State v. Putnoki, supra, 200 Conn. 221; the primary consideration must be the acquittee’s ‘‘present mental condition.’’ (Emphasis in original.) Id., 223; see State v. Metz, supra, 230 Conn. 426 (‘‘the presumption of the existence of facts warranting the [acquittee’s] commitment does not survive the expiration of the max- 0, 0 CONNECTICUT LAW JOURNAL Page 9

0 Conn. 0 ,0 11 State v. Foster

imum term of criminal sanctions’’); see also Powell v. Florida, 579 F.2d 324, 333 n.10 (5th Cir. 1978) (‘‘[although] an initial commitment of an acquittee may well focus [on] the crime committed, the release of an [acquittee] more properly looks to that individual’s behavior since the crime to determine whether . . . his recent behav- ior demonstrates his continuing dangerousness’’). Fac- tors relevant to the acquittee’s present mental condition include, among other things, ‘‘his present and past diag- noses . . . the need for continued medication and ther- apy, and the prospects for supervision if released.’’9 State v. Putnoki, supra, 221. Likewise, the significance of the acquittee’s index offenses, and other incidents of violence, is greater the more recently and regularly such conduct occurred. See, e.g., State v. Long, 301 Conn. 216, 228–29, 19 A.3d 1242 (acquittee committed approximately fifteen assaults between 1998 and 2007, the latter being one year before his continued commit- ment hearing), cert. denied, 565 U.S. 1084, 132 S. Ct. 827, 181 L. Ed. 2d 535 (2011); State v. Putnoki, supra, 222 (acquittee’s index offense was recent and extremely violent murder of her disabled mother-in-law); State v. Gates, supra, 198 Conn. 403 (‘‘[t]he accuracy of such predictions is much greater where there is a pattern of repetitive assault and violent conduct’’). In addition, although, ‘‘[i]n reaching its difficult decision, the [trial] court may and should consider the entire record avail- able to it’’; State v. Putnoki, supra, 221; evidence of oppositional, noncompliant, or inappropriate behavior is relevant only insofar as it speaks to the acquittee’s psychiatric disabilities and any resulting danger. 9 The trial court’s memorandum of decision did not address the acquittee’s prospects for supervision if released—an important factor in cases of this sort— other than to say that the court was skeptical of the acquittee’s stated intent to voluntarily comply with mandated safeguards. There was undisputed testimony that all of the treatment services available to him would continue to be available upon discharge. Page 10 CONNECTICUT LAW JOURNAL 0, 0

12 ,0 0 Conn. 0 State v. Foster

Third, the state must prove, by clear and convincing evidence, that the acquittee, if discharged, would be imminently dangerous. Although the term ‘‘danger’’ as used in § 17a-593 (c) is not defined in the statute,10 the governing regulation provides: ‘‘ ‘Danger to self or to others’ means the risk of imminent physical injury to others or self, and also includes the risk of loss or destruction of the property of others.’’ (Emphasis added.) Regs., Conn. State Agencies § 17a-581-2 (a) (6). In State v. Harris, 277 Conn. 378, 890 A.2d 559 (2006), in the context of resolving an evidentiary challenge, this court addressed whether the standard of dangerousness applied to acquittees is less stringent than that which applies in the civil commitment context. See id., 388–89. The court in Harris concluded that, because the acquittee must be imminently dangerous to remain com- mitted under § 17a-581-2 (a) (6) of the regulations, the standard is at least as stringent as the civil commitment standard; see id., 389; see also State v. Dyous, 307 Conn. 299, 333 n.19, 53 A.3d 153 (2012); which requires ‘‘a substantial risk that physical harm will be inflicted by an individual upon his or her own person or upon another person . . . .’’ (Emphasis added.) General Statutes § 17a-495 (b). In support of the conclusion that the dangerousness standard applied by § 17a-593 (c) is a ‘‘demanding’’ one; State v. Harris, supra, 389; this court pointed to one dictionary that defined ‘‘ ‘[i]mmi- nent’ ’’ as ‘‘ ‘ready to take place; [especially] hanging threateningly over one’s head . . . .’ ’’11 Id., quoting Merriam-Webster’s Collegiate Dictionary (10th Ed.1993) p. 580. 10 General Statutes § 17a-580 (5) does specify that ‘‘ ‘[d]anger to himself or others’ includes danger to the property of others . . . .’’ 11 Singling out the ‘‘hanging threateningly over one’s head’’ language that Harris chose to highlight; (internal quotation marks omitted) State v. Harris, supra, 277 Conn. 389; is problematic because, like the sword of Damocles, a threat can hang over one’s head for one’s entire life without ever drawing blood. The metaphor captures the proximal connotations of the word ‘‘imminent,’’ less so the temporal ones. 0, 0 CONNECTICUT LAW JOURNAL Page 11

0 Conn. 0 ,0 13 State v. Foster

Notably, although this court briefly discussed § 17a- 581-2 (a) (6) of the regulations in Harris, we have never engaged in a proper statutory construction of that regu- lation, which presumably would entail both a more thor- ough review of ‘‘the commonly approved usage of the language’’; General Statutes § 1-1 (a); as well as a con- sideration of how the legislature has used the phrase ‘‘risk of imminent physical injury’’ in related statutes. See General Statutes § 1-2z; see also Alexandre v. Com- missioner of Revenue Services, 300 Conn. 566, 578, 22 A.3d 518 (2011) (‘‘[a]dministrative regulations have the ‘full force and effect’ of statutory law and are interpre- ted using the same process as statutory construction, namely, under the well established principles of . . . § 1-2z’’). A full analysis of the regulatory language is beyond the scope of this concurrence. However, one brief point regarding the imminence requirement war- rants mention. The meaning of the word ‘‘imminent’’ in the English language always refers to proximity, almost always tem- poral proximity. See, e.g., The American Heritage Dic- tionary of the English Language (3d Ed. 1992) p. 903 (defining ‘‘imminent’’ to mean ‘‘[a]bout to occur; impending’’); The Random House Dictionary of the English Language (2d Ed. Unabridged 1987) p. 957 (defining ‘‘imminent’’ as ‘‘likely to occur at any moment; impending’’). In Harris, this court emphasized that the ‘‘imminent’’ language requires not only that the danger posed by the acquittee be substantial, but also that it be ‘‘ ‘ready to take place . . . .’ ’’ State v. Harris, supra, 277 Conn. 389. This is what the word ‘‘imminent’’ means in common usage, and it is what it means in every other legal context that I have seen in which the legislature has used the term ‘‘imminent physical injury.’’ See, e.g., General Statutes § 52-146q (c) (2) (social worker may disclose confidential records and communications of person consulting social worker only when ‘‘[the] social Page 12 CONNECTICUT LAW JOURNAL 0, 0

14 ,0 0 Conn. 0 State v. Foster

worker determines that there is a substantial risk of imminent physical injury by the person to himself or others’’);12 see also, e.g., Haynes v. Middletown, 314 Conn. 303, 322–23, 101 A.3d 249 (2014) (for purposes of identifiable person, imminent harm exception to gov- ernmental immunity, ‘‘the proper standard for determin- ing whether a harm was imminent is whether it was apparent to the municipal defendant that the dangerous condition was so likely to cause harm that the defendant had a clear and unequivocal duty to act immediately to prevent the harm’’); State v. Parnoff, 160 Conn. App. 270, 279, 125 A.3d 573 (2015) (for purposes of fighting words exception to first amendment, court cited ‘‘ ‘hanging threateningly over one’s head’ ’’ definition of ‘‘ ‘[i]mminent’ ’’ and concluded that imminence compo- nent was missing because defendant was not ‘‘immedi- ately capable of the violent act’’), aff’d, 329 Conn. 386, 186 A.3d 640 (2018). As I will explain, the board’s deci- sion in the present case, as well as the decision of the trial court that upheld the board’s decision, largely ignored the existence and significance of the word ‘‘imminent’’ in § 17a-581-2 (a) (6) of the regulations. II The trial court explained its conclusion that the acquittee would constitute a danger to himself or others if discharged both in its brief memorandum of decision and during the hearings it conducted on the state’s petition for continued commitment and the acquittee’s 12 No one would seriously maintain that, when the legislature authorized a clinical social worker to violate a patient’s therapeutic confidentiality to prevent imminent physical injury to a third party, it meant that the social worker could do so any time he concluded that some patient might experience stressors in her life, those stressors might cause her to stop taking prescribed medications, pausing medications might lead to a relapse, a relapse might cause her to wish to harm others, and she might eventually act on those wishes. See State v. Orr, 291 Conn. 642, 652, 969 A.2d 750 (2009) (scope of ‘‘[the] exception [under § 52- 146q (c) (2)] is precise and limited’’ by language of statute). 0, 0 CONNECTICUT LAW JOURNAL Page 13

0 Conn. 0 ,0 15 State v. Foster

subsequent motion for rectification. I have serious prob- lems with four aspects of the court’s analysis. A First, the trial court indicated that its primary concern was the protection of schoolchildren. It goes without saying that this is a perfectly valid concern. If clear and convincing evidence supported a finding that the acquittee continued to pose an imminent danger to schoolchildren, then discharge would not be appro- priate. But, here, there is no such evidence, certainly not enough to make the required showing. The only evidence the court identified in support of its conclu- sion that the acquittee poses an ongoing danger to schoolchildren was the index offenses themselves, which took place on a single occasion eighteen years before the date of the petition for an order of continued commitment. The offenses occurred in 2001, when the acquittee was twenty-four years old and his schizophre- nia was just beginning to manifest itself. At that time, while experiencing hallucinated auditory commands, he entered a school with the intention of picking a fight with the first person he encountered. He physically lifted a sixth grade girl but put her down unharmed. He then assaulted a sixth grade boy, slapping, punching, and kicking the youth.13 At the time, the acquittee had two small knives in his backpack, but there is no indica- tion that he used them or took them out during the incident. There also is no indication that, in the interven- ing eighteen years between the date of the index offenses and the court’s hearing on the state’s petition, the acquittee ever considered assaulting or harming another child, let alone did so, even though he had spent extensive time in various community placements during that period. 13 There is no indication in the record as to whether either child suffered any permanent physical or psychological injuries. Page 14 CONNECTICUT LAW JOURNAL 0, 0

16 ,0 0 Conn. 0 State v. Foster

Although it is not improper for the trial court to consider the acquittee’s index offenses as one factor, among many, when assessing dangerousness, the only purpose of that inquiry is to assess whether the acquittee is presently dangerous. See, e.g., State v. Putnoki, supra, 200 Conn. 221, 223. When a sufficient amount of time has passed and the maximum term of commitment has terminated, as in the present case, the commission of the index offenses, standing alone, can no longer sup- port a presumption of ongoing dangerousness. See, e.g., State v. Krol, supra, 68 N.J. 247. Giving any significant weight to the index offenses in this case is particularly inappropriate because the acquittee had no criminal history prior or subsequent to the index offenses, and those offenses were misdemeanors or relatively low- level felonies.14 Two decades later, something more than the index offenses is needed to establish present dan- gerousness. Cf. Fasulo v. Arafeh, 173 Conn. 473, 477, 378 A.2d 553 (1977) (‘‘the longer the commitment, the greater the safeguards [that] are required to ensure that no one is deprived of liberty without due process’’). The trial court did not leave us to guess why it focused on the safety of schoolchildren despite there having been no evidence that the acquittee himself posed any risk to schoolchildren in 2019 or thereafter. The court discussed at some length the spate of school shootings that have plagued this country since the Columbine 14 The acquittee was charged with assault in the third degree, a class A misde- meanor; see General Statutes § 53a-61 (b); as well as burglary in the first degree, a class B felony; see General Statutes § 53a-101 (c); risk of injury to a child, a class C felony; see General Statutes § 53-21 (a) (1); and possession of a weapon on school grounds, a class D felony. See General Statutes § 53a-217b (c). To be clear, any act of violence of any sort involving schoolchildren is a very serious matter. My point relates to the statutory assessment of future dangerousness almost two decades later on the basis of child related index offenses that the legislature has chosen to categorize as misdemeanors or class C felonies, particularly in circumstances in which there is no evidence that the acquittee either intended to or did cause serious injuries to the victims. 0, 0 CONNECTICUT LAW JOURNAL Page 15

0 Conn. 0 ,0 17 State v. Foster

High School massacre in 1999: ‘‘[T]he protection of our children . . . [is] in the press every single day. Some- body goes into a school with a firearm, and you know what happens. *** ‘‘There’s an underlying criminal behavior here where children were exposed to what society has become acutely aware of . . . . We all sit there. In our own state, we saw it [at Sandy Hook Elementary School in Newtown] . . . where somebody didn’t mitigate risk, where we . . . fell down. We fell short of what . . . our society requires of us . . . as centurions of the public good. . . . [W]e had dead children on our hands here because somebody didn’t mitigate risk.’’ These observations are true. They are also irrelevant. We all are painfully aware of the ‘‘appalling series of crimes’’ committed at Sandy Hook Elementary School, as well as other mass school shootings. Soto v. Bush- master Firearms International, LLC, 331 Conn. 53, 65, 202 A.3d 262, cert. denied sub nom. Remington Arms Co., LLC v. Soto, U.S. , 140 S. Ct. 513, 205 L. Ed. 2d 317 (2019). As a legal matter, however, the fears expressed by the trial court have nothing to do with the assessment of dangerousness that the court was required to make in this case. There was no evidence from any source, much less clear and convincing proof, suggesting that the acquittee would engage in such behavior in the future, imminently or at any later time. The trial court disregarded the relevant evidence and, on the basis of groundless fears drawn from the evening news, deemed this acquittee dangerous because his index crimes are vaguely and tangentially evocative of far more serious crimes, committed by different perpe- trators, against different victims, by different means, and for fundamentally different reasons. Page 16 CONNECTICUT LAW JOURNAL 0, 0

18 ,0 0 Conn. 0 State v. Foster

B The trial court indicated that its second principal concern was the acquittee’s ‘‘inappropriate behavior toward women,’’ an observation that unfortunately also took on the form of social commentary. The court elabo- rated: ‘‘[T]he other [area the court is concerned with] is the license that [the acquittee has] tak[en] with our women in society today. We have become focused, very clearly, on what do we do to protect women’s rights and controlling, if you will, impulse control of others with regard . . . to the security of their person. . . . Somebody gets on a bus and sits next to a female and strikes up a conversation. Irrespective of what the moti- vation is, he puts his hand on the woman’s leg. You know what that can turn into today in our society . . . . The license that was given . . . thirty or forty years ago simply doesn’t pertain today. It isn’t applicable to today’s societal norms, and it shouldn’t be. *** ‘‘The two areas in our society today . . . we have become focused on [are] protecting our kids, especially in school settings, and protecting our better halves. *** ‘‘And we’ve become acutely aware of the protection of our females in society today. And their rights . . . their God given rights to be secure in their persons, if you will, and maybe secure in . . . their psychological and psychiatric makeup. Because you can do a lot of damage both ways to women, and we’ve become acutely aware of that in our society today.’’ The court went so far as to say that, ‘‘[i]rrespective of what the cause is, I am concerned [that] our women in society be protected from any kind of irrational . . . or inap- propriate behavior.’’ 0, 0 CONNECTICUT LAW JOURNAL Page 17

0 Conn. 0 ,0 19 State v. Foster

The trial court’s analysis of the acquittee’s purported danger to women is at least as troubling as its analysis of the danger to schoolchildren. Once again, it is clear that the court’s intentions were good. I would phrase the issue differently, but there is no question that sexual harassment and sexual violence are serious problems and that women have a right to be free from those incursions, in the workplace, on a hypothetical bus, in society at large. But, as with the school shootings, the trial court’s concern with addressing societal problems, and with misconduct perpetrated by others, led the court far afield from the legal question that it was charged with resolving, namely, whether there is clear and convincing evidence that this particular acquittee poses an imminent danger to others (including women, of course) by virtue of his psychiatric disabilities. No doubt the record discloses some reason to be concerned about the acquittee’s attitude toward and treatment of women. Although he has never been charged with or accused of any crime of sexual vio- lence, one of the victims of his index offenses was a girl. Perhaps more to the point, the acquittee has a history of what might be seen as sexualized or inappro- priate speech and conduct toward female hospital staff members. That said, however, the applicable legal stan- dard for continued commitment requires significantly more specific and severe conduct than ‘‘irrational . . . or inappropriate behavior’’ toward women, irrespective of the cause or motivation, which is all that the trial court apparently required. Under the statute, the court was required to find that whatever danger the acquittee posed to women was both imminent and a result of his psychiatric disabilities. See generally footnote 5 of this opinion and accompanying text. At a constitutional level, the United States Supreme Court has made clear that a ‘‘[l]oss of liberty calls for a showing that the individual suffers from something more serious than is Page 18 CONNECTICUT LAW JOURNAL 0, 0

20 ,0 0 Conn. 0 State v. Foster

demonstrated by [abnormal or] idiosyncratic behavior.’’ Addington v. Texas, 441 U.S. 418, 427, 99 S. Ct. 1804, 60 L. Ed. 2d 323 (1979).

Upon examination, it appears clear to me that the evidence is not there. Indeed, the expert testimony of the state’s own psychiatrists and other hospital staff largely undercuts any suggestion that the acquittee posed any actual and appreciable danger to women on the basis of his ‘‘inappropriate conduct.’’ Although the board’s reports over the years repeatedly have refer- enced and relied on the acquittee’s ‘‘inappropriate boundaries’’ with and behavior toward women, those generic descriptors could encompass a wide range of conduct. The sparse supporting evidence for the board’s conclusion reveals that, although the acquittee’s behav- ior toward women has been unsuitable, especially for a professional setting, his conduct falls on the mild end of the range of inappropriate conduct in context.15 One of the few pieces of supporting evidence I could locate was a 2012 statement from Alec Buchanan,16 a DMHAS consulting forensic psychiatrist, who testified that the acquittee addressed some female staff as ‘‘baby’’ and requested hugs from them. Justin Winkel, the unit psy- chologist, concurred: ‘‘[S]ometimes, [the acquittee] will push things too far in his attention seeking way. He’ll start to joke around and sometimes become . . . mildly inappropriate . . . asking for hugs, making comments, [and] expressing his affection for some staff . . . . And, most of the time, it’s very innocuous, it’s 15 To be crystal clear, I do not condone or excuse any of the conduct at issue, which is clearly unwelcome and entirely inappropriate. My point is that the conduct does not demonstrate the statutory standard of dangerousness selected by the legislature to determine whether to discharge the acquittee from com- mitment. 16 Buchanan’s testimony offers an excellent example of helpful mental health evidence. He focuses on the legally relevant questions and provides sufficiently detailed and nuanced support for his conclusions to be useful to the trial court. 0, 0 CONNECTICUT LAW JOURNAL Page 19

0 Conn. 0 ,0 21 State v. Foster

not a problem whatsoever. But, you know, some of the staff, it will rub the wrong way.’’ Subsequent testimony by other DMHAS staff con- firmed that the acquittee’s ‘‘mildly inappropriate’’ behavior toward women was neither dangerous to them (although some were understandably offended) nor an outgrowth of his psychiatric disabilities. They testified that the acquittee is emotionally immature and has spent almost his entire adult life living in mental health facilities, where there were few, if any, opportunities to pursue healthy adult relationships and where any romantic relationships would require the permission of his treatment providers. Claudia Krasnicki, a hospital psychologist, explained that, under those circum- stances, the acquittee ‘‘tended to treat staff like extended family, given that he entered the hospital at a very young age. The line staff became overwhelmed by his constant neediness and sometimes interpreted [the acquittee’s] desire for affection as sexual in nature.’’17 Daniel Papapietro, the acquittee’s individual psychotherapist since approximately 2007, agreed that his conduct toward women is neither predatory nor symptomatic: ‘‘[W]hat gets labeled as inappropriate touching is . . . not sexualized, in terms of touching . . . inappropriate parts of a woman’s body. It’s touch- ing one on the shoulder, on [the] back. It’s a childlike flirtatiousness. It has not [at] all, in my view, had the tenor of some kind of predatory, sexual, or sexualized behavior. . . . I don’t think he’s done that out in the community. As far as I’m aware . . . he’s managed his boundaries out there . . . appropriately.’’ The acquittee 17 In 2015, the board reported that, ‘‘for administrative reasons’’ that had nothing to do with the acquittee, he had been abruptly transferred to a different unit, leaving him little time to ‘‘terminate with . . . his treatment team’’ of eight years. This sudden loss of what was, in effect, his family contributed to his ‘‘long- standing pessimism’’ that he would ever be permitted to leave. He began to miss sessions of his trauma group, which resulted in his temporary leaves being placed on hold. Page 20 CONNECTICUT LAW JOURNAL 0, 0

22 ,0 0 Conn. 0 State v. Foster

‘‘means well and has never been spontaneously aggres- sive or predatory in any way. . . . His inappropriate juvenile flirtation with female staff through inappropri- ate comments [is] as much about him being lonely and in search of some emotional and physical intimacy as it is about him doing it to . . . play with them, [to] get a rise out of them . . . .’’ (Internal quotation marks omitted.) At the discharge hearings in 2019, the only recent example of the ‘‘inappropriate comments’’ on which the board and the trial court relied was provided by the acquittee’s conditional release supervisor, licensed clin- ical social worker Madeline Rodriguez. She testified that the acquittee ‘‘may say, ‘you look nice today,’ or ‘you’re beautiful.’ Never any threatening remarks . . . .’’ Rodriguez further testified that she had no concerns about the acquittee’s ‘‘boundary issues’’ with respect to how he would interact with the general public. As the finder of fact, the trial court was perhaps free to reject all of the experts’ testimony and to conclude that (1) the acquittee continues to act ‘‘inappropriately’’ toward women as a result of his psychiatric disabili- ties,18 and (2) his inappropriate conduct is not merely 18 The trial court found that ‘‘the state has established by clear and convincing evidence that (1) [the acquittee] suffers from a psychiatric illness diagnosed as: schizoaffective disorder bipolar type; (2) borderline intellectual functioning; (3) inappropriate and impulsive behaviors especially toward females; [and] (4) frus- tration difficulties.’’ The first finding clearly satisfies the statutory requirement that the acquittee must suffer from ‘‘psychiatric disabilities . . . .’’ General Stat- utes § 17a-593 (c). To the extent that the court also was suggesting that frustration difficulties and inappropriate, impulsive behaviors toward women qualify as diagnosable psychiatric disabilities for purposes of the statute, such a finding is without support in the record and would be clearly erroneous. In addition, the acquittee’s intelligence quotient, originally assessed at 74 and later revised to 83 (both within the second standard deviation below the mean), is not low enough to qualify him as ‘‘a person with intellectual disability’’ for purposes of § 17a- 593 (c). See General Statutes § 17a-580 (8) (‘‘ ‘[i]ntellectual disability’ has the same meaning as provided in [General Statutes §] 1-1g’’); see also General Statutes § 1-1g (a) and (b) (defining ‘‘intellectual disability’’ as ‘‘a significant limitation in intellectual functioning,’’ meaning ‘‘an intelligence quotient more than two stan- dard deviations below the mean’’). 0, 0 CONNECTICUT LAW JOURNAL Page 21

0 Conn. 0 ,0 23 State v. Foster

bizarre, distasteful, or offensive, but renders him immi- nently dangerous (or exemplifies his dangerousness). But the board is required by statute to substantiate its findings and conclusions; see General Statutes § 17a- 593 (d); and, certainly, the best practice, if not the constitutional mandate, is for the trial court to do so as well. See, e.g., State v. Dyous, 198 Conn. App. 253, 268–69 n.11, 233 A.3d 1138 (best practice is for state to present evidence that acquittee’s mental illness diag- nosis is based on current Diagnostic and Statistical Manual of Mental Disorders), cert. denied, 335 Conn. 948, 238 A.3d 17 (2020); State v. Corr, 87 Conn. App. 717, 732, 734, 867 A.2d 124 (trial court addressed each Putnoski factor before denying petition), cert. denied, 273 Conn. 929, 873 A.2d 998 (2005). In the present case, if the court was persuaded that the evidence supported these conclusions, it should have made those specific findings and pointed to supporting evidence in the record.19 Conclusory statements regarding inappropri- ate conduct toward women and changing societal mores are not enough. At times, the trial court stepped even closer to error by suggesting that the danger that the acquittee purport- edly posed to these two vulnerable groups, school- 19 As the majority describes, there were a few minor incidents in 2016 and 2017 that, although not clearly dangerous, reasonably could have been perceived as threatening by female staff. In one, the acquittee stood so as to seemingly bar a female staff member’s exit from a utility closet. He later said that he should slap another woman. With respect to the first, he further was alleged to have said, ‘‘ ‘[Rosina] Bandanza told me to watch out for you, that you were out to get me. Well, then, I’m out to get you.’ ’’ There is no way to tell if these events served as the basis for the trial court’s legal conclusions. I will not venture an opinion as to whether these events, which occurred several years prior to the hearing on the state’s petition and did not involve acts of physical violence, provided clear and convincing evidence of imminent, future dangerousness in 2019. I would observe, however, that there was no evidence connecting these events to the acquittees’ psychiatric disabilities. Notably, during that same time period, the acquittee was the victim of several unprovoked physical assaults by a psychotic roommate, to which he never retaliated. Page 22 CONNECTICUT LAW JOURNAL 0, 0

24 ,0 0 Conn. 0 State v. Foster

children and women, meant that he personally did not deserve the benefit of heightened constitutional protec- tion: ‘‘Those are the two things that I’m primarily con- cerned with here. And . . . that’s why . . . intermediate scrutiny is not the applicable standard. Not when it comes to him. . . . When it comes to him, I think the history of the case and the underlying charges indicate quite the opposite.’’ (Emphasis added.) The court concluded: ‘‘I don’t believe, in this particular case, as it relates to this particular [acquittee], that intermedi- ate scrutiny is the appropriate standard of review.’’

The standard that governs a legal challenge sounding in due process or equal protection is dictated by the nature of the constitutional interests involved and whether the individual is a member of a protected class. See part I of the majority opinion. The nature of the individual’s crimes (actual or potential) and the identity of his victims are wholly irrelevant to the question of which level of scrutiny applies.

C

The third aspect of the trial court’s analysis that most concerns me is that it could be read to improperly place on the acquittee the burden of satisfying the court that he poses no danger to society. The following additional facts are relevant.

The board’s July 30, 2018 report to the trial court cited testimony from Alexander Westphal, DMHAS con- sulting forensic psychiatrist, that the acquittee was fully engaged in treatment, compliant with his medication, and clinically stable. Specifically, Westphal testified that the acquittee ‘‘had been free from psychotic symp- toms for several years.’’ Crediting this testimony, the board found that the acquittee had ‘‘remained clinically stable and cooperative with treatment’’ since January, 0, 0 CONNECTICUT LAW JOURNAL Page 23

0 Conn. 0 ,0 25 State v. Foster

2016. Given that he also had demonstrated ‘‘good prog- 20

ress’’ with respect to nonpsychiatric traits such as poor social skills and low frustration tolerance, the board approved his conditional release into the community. By the time the board filed its next report with the court on September 3, 2019, the acquittee had been successfully living in the community for approximately fifteen months and had been enjoying community leave privileges for more than seven years without any inci- dence of violence. At that time, the board reported that his psychiatric symptoms remained controlled with medication and treatment, and that he continued to make progress. Accordingly, when the trial court issued its decision in December, 2019, the undisputed evidence was that the acquittee had been clinically stable and symptom free, had taken his medications and complied with his treatment regimen, and had made steady progress for nearly four years, culminating in a fifteen month period of living in the community without incident. Indeed, there was undisputed testimony that the acquittee had been a leader in his community placement, helping to facilitate groups and assisting other community mem- bers who were struggling. The court appeared to accept the board’s findings, repeatedly noting that the acquittee had made progress and that his release into the community was stable. The court nevertheless seemed to put very little stock in those years of prog- ress, telling defense counsel that ‘‘what I’m concerned about here is [that] you seem to dismiss the underlying criminal behavior and the [acquittee’s] history . . . simply because of the progress he has made over the last, two, three, five years,’’ and that ‘‘you’re asking me to take the . . . leap here, the great leap of faith.’’ 20 Aside from early 2016, when there were some minor issues with medication noncompliance, there had been no issues with violence, psychotic symptoms, or medication noncompliance since 2014. Page 24 CONNECTICUT LAW JOURNAL 0, 0

26 ,0 0 Conn. 0 State v. Foster

As I will discuss, there was enough (perhaps just barely enough) evidence in the record to support the trial court’s ultimate determination that the acquittee continued to pose a danger as a result of his psychiatric disabilities (combined with his substance abuse issues). But, at some point—and two decades is well past that point—a nearly four year track record of steady, symp- tomless progress becomes far more relevant evidence with respect to present dangerousness than the iso- lated, decades old incident that led to his involuntary commitment. And, while we may assume that the trial court applied the proper burden of proof and was speak- ing only metaphorically, we ought not suggest, in any manner, that allowing acquittees back into the commu- nity after their maximum term of commitment requires an impermissible ‘‘leap of faith’’ and that we will not take that leap until we have achieved, as faith has been aptly described, ‘‘the assurance of things hoped for, the conviction of things not seen.’’ Hebrews 11:1 (English Standard Version). By the very nature of the statutory directive, the trial court is required to prognosticate the future, regardless of its disposition. There are no assurances, and it is improper to deny a petition unless the acquittee or his lawyer provides the court with a warranty of performance. The statutory scheme, in fact, imposes the burden on the state to do the opposite. D Fourth, I am struck by the extensive record evidence indicating that many of the behavioral traits and much of the oppositional conduct that seem to create barriers to the acquittee’s release may have an iatrogenic source, that is, result from the conditions of confinement as much (or more) than mental illness. One way to uncover this phenomenon is by examining what the acquittee’s treating doctors say about his behavioral issues. Doing so must lead an observer to reflect on the fact that labels such as ‘‘[i]nappropriate and impulsive behaviors’’ and 0, 0 CONNECTICUT LAW JOURNAL Page 25

0 Conn. 0 ,0 27 State v. Foster

‘‘[f]rustration difficulties,’’ which the trial court used to justify the acquittee’s ongoing commitment, may mask a tragic situation, if not a downright failure of the sys- tem. The acquittee’s treating doctors consistently have expressed the view that the system administered by the board sets acquittees up for failure, places them at the mercy of undertrained staff who may impede their progress, and ultimately traps them in a vicious cycle by which they are conditioned to behave immaturely and irresponsibly and then punished for exhibiting those traits by being deprived of the opportunity to grow. Here is just a sampling of the feedback that the acquittee’s treatment providers repeatedly have provided to the board over the past dozen years, since the completion of his maximum term of commitment: • ‘‘Some of what seems to be going on with [the acquittee] is more characterological . . . attention seeking . . . trying to respond to boredom. Those kinds of things get in the way of [his] being able to . . . move forward in some of his treatment . . . . And we continually . . . try and get the staff to be somewhat accepting of his limitations . . . .’’ (Justin Winkel, December 7, 2012). • ‘‘[The acquittee] has struggled over the months with frustration and lack of any clearly defined goals that could put him back on track. . . . [H]e would express his hopelessness . . . . *** ‘‘Over the years, when he felt staff was . . . ‘rude, just telling people what to do, because they can’ . . . he would act out immaturely, by arguing, calling names, and, if it escalated, he would refuse one or two doses of his medications. . . . [He] has always returned to full compliance with medications in a short period of time.’’21 (Kevin Trueblood, January 8, 2016). 21 At the January 8, 2016 hearing, the acquittee’s inpatient psychiatrist, Marilynn Stuart, agreed, stating, ‘‘[i]t’s just that . . . he gets frustrated and feels powerless, Page 26 CONNECTICUT LAW JOURNAL 0, 0

28 ,0 0 Conn. 0 State v. Foster

• ‘‘During the past few years, [the acquittee’s therapy sessions have] not had anything to do with the crime or his mental illness. The sessions have, instead, focused on his self-defeating behaviors, his immaturity, his adolescent rebellion when he feels like staff are treating him like a child, and how all of that has contrib- uted to keeping him here longer than clinically nec- essary. *** ‘‘He’s been here since he was in his twenties, and, now, he’s in his early forties, but he has to keep working on when to try to be funny . . . .’’ (Kevin Trueblood, June 2, 2017). • ‘‘[W]hen [the acquittee is] treated with respect, he responds appropriately. When he’s treated disrespect- fully, and that happens an awful lot, he responds in a childish and in a hurt way, and in a way that’s not dissimilar perhaps [to] how many of us might respond if we’re treated disrespectfully [in public]. We can’t forget that. . . . [T]he cure for institutionalization is not perpetuating institutionalization. *** ‘‘[T]his is really difficult and politically incorrect to talk about, but there is also a pattern, not just with [this acquittee] . . . where, as patients get closer to moving out . . . [t]here’s the crab in a bucket phenomenon, [that is] when one person starts to leave . . . everyone starts to pull them back down, and patients have a way of provoking other patients to act out, and, sometimes, this is something that spills over into staff . . . .’’ (Dan- iel Papapietro, June 2, 2017). like a lot of the patients do . . . .’’ As the acquittee’s counsel noted, this led the board to conclude in a 2014 memorandum of decision that, ‘‘ ‘[d]espite remaining clinically stable, [the acquittee] was unable to appropriately address his behavior with his treatment team in a timely manner, thus delaying his return to the community.’ ’’ 0, 0 CONNECTICUT LAW JOURNAL Page 27

0 Conn. 0 ,0 29 State v. Foster

• ‘‘[I]n the old days of long-term psychiatric hospital- izations, a great many experts wrote about the regres- sive pull of long-term inpatient psychiatric facilities . . . . I think we’ve . . . gotten away from under- standing just . . . how powerful that [pull] is, and, because of the very nature of the locked doors, the rules and restrictions and the limitations of freedom, that it has, by its very nature . . . an infantilizing pull on our patients . . . . [A]s a result, if what we do has an infantilizing effect on our patients, it has a parentify- ing effect on . . . us . . . . There is an iatrogenic cause, an environmental, external cause for a lot of [the acquittee’s] behaviors, as well as an internal psychologi- cal cause . . . . *** ‘‘Now, it’s not something [the acquittee is] going to grow out of while he’s in our hospital. . . . I think we have sufficient evidence that, except for the one [index offense] some time ago, when he was out in the commu- nity, his behavior has been pretty much appropriate, that it’s on an inpatient unit, with the severe limits, with the fact that the . . . most productive years of his life have been spent here, that there’s a natural level of frustration and a natural regressive pull that is always going to end up in [the acquittee’s] behaving childishly and in immature ways. . . . I haven’t seen any evidence that the treatment environment that we have is going to be any more conducive to him getting better, and I do think that, when he was more involved in the outpatient community, he was getting better. . . . I certainly think that, psychologically speaking, he’s more likely to benefit from a more normalized experience in a com- munity [than] the restrictive and sometimes punitive experience of being in a hospital. *** ‘‘[W]hat I’m saying, and what I believe the hospital and the team [are] saying, is [that] we don’t think those Page 28 CONNECTICUT LAW JOURNAL 0, 0

30 ,0 0 Conn. 0 State v. Foster

behaviors create a risk. We don’t believe that [the acquittee] is going to regress in a dangerous way. . . . This is Groundhog Day for him. . . . [H]e keeps repeating the same things with the same kind of people over and over again. It’s not healthy for him, and our system is not going to change, and it’s not going to allow him to mature and behave in a more appropriate way.’’ (Daniel Papapietro, June 2, 2017). • ‘‘I believe that the staff really need a lot of educa- tion. It’s very difficult for them to understand that a patient, who is so tall and strong, could behave some- times in a rather childish manner, but [the acquittee] is a delight to have on the unit. He is a kindhearted individual, and he really means no harm.’’ (Rosina Ban- danza, June 2, 2017). • ‘‘[The acquittee has] been institutionalized for way too long . . . . We do not feel that his risk would be increased in the community, as long as he is with family or his mom.’’22 (Rosina Bandanza, September 15, 2017). • ‘‘This is a politically challenging conversation to have . . . but I think we have to at least acknowledge [that] sometimes staff don’t always behave in the most appropriate ways, and it’s easy to . . . push the blame off onto the patient. We believe that that, in fact, is a large part of what happened in this case. . . . It’s unpleasant to have to suggest that maybe staff had been somewhat inappropriate, but I can’t, in good con- science, be politically correct and try to avoid the issue and let the blame lie inappropriately [with] a patient . . . . It’s not just unique to that unit. It is a part of the problem and difficulty of long-term hospitalization with staff and patients [who] spend a great deal of time together. . . . I just want to be absolutely clear that 22 Notably, even though the acquittee’s mother was a positive influence on his progress and she had moved closer to facilitate visits with him, staff have restricted his visits with her as a behavior management tool. 0, 0 CONNECTICUT LAW JOURNAL Page 29

0 Conn. 0 ,0 31 State v. Foster

none of us [has] thought that this suggests . . . any symptomatic process from [the acquittee’s] mental ill- ness. This is just . . . part of [the] closed environment, living with people and having some conflict.’’ (Daniel Papapietro, September 15, 2017). Little of this commendably frank commentary has been included in the board’s various reports and recom- mendations to the trial court over the years. None was addressed in the court’s decision, notwithstanding that it was all part of the record before the court. See, e.g., State v. Putnoki, supra, 200 Conn. 221 (trial court ‘‘should consider the entire record,’’ including acquittee’s mental health history). III The foregoing analysis has doctrinal implications that may be useful to summarize for the purpose of providing future guidance in connection with the decisions required under § 17a-593. To summarize, in order to ensure that the statutory requirements and demands of due process are satisfied, it is incumbent on the board to provide, and for the trial court to restrict its analysis to, evidence relevant to whether the acquittee will pose an imminent danger to himself or others by virtue of his mental illness. The state bears the burden of proof to demon- strate, by clear and convincing evidence, that the acquittee poses such a danger. To be legally relevant, the evidence of dangerousness must derive in material part from the acquittee’s psychi- atric disabilities, and not be the result of characterologi- cal or other behavioral traits that would create that danger, even in the absence of the disability. As much as possible, evidence relied on should be detailed, date- specific, and supported by expert testimony grounded in scientific research supporting the conclusion that the observed behavior is indicative of imminent danger- ousness. It is not helpful, for example, for the board or Page 30 CONNECTICUT LAW JOURNAL 0, 0

32 ,0 0 Conn. 0 State v. Foster

the court simply to state that the acquittee has a history of inappropriate conduct or a history of treatment non- compliance.23 That type of label may apply to a long and continuing pattern of physical threats, altercations, and refusals to take prescribed psychiatric medications, or it could refer to nothing more than a few outdated, isolated incidents of juvenile prankishness, off-color remarks, or failures to attend a group therapy session. The former is legally relevant; the latter is not. As a general guideline, the most directly relevant evidence would include a protracted or recent history of violent or criminal conduct; recent uncontrolled symptoms of psychosis or other mental illnesses that are linked to violence; recent incidents of refusal to take prescribed psychotropic medications or significant noncompliance with other recommended treatment modalities; recent incidents of prohibited alcohol or drug use (to the extent that the use of such substances is likely to result in decompensation for the individual in question); a recent history of threatening or aggressive conduct; a significant, recent failure to successfully nav- igate conditional release or other community visita- tion;24 and scores on validated psychological instruments 23 The board’s findings in this case often generalize from a few limited (and sometimes long outdated) incidents and distill highly nuanced testimony into a blunt and conclusory condemnation. For example, one of the few specific exam- ples in the record of the acquittee’s impaired judgment and impulsive behaviors, which the board continued to cite as arguments against discharge in 2023, was an immature but harmless incident in 2012 in which he put his face to the window of a psychotherapy group to distract the coordinator. Indeed, when asked in 2016 for examples of the acquittee’s ‘‘poor frustration tolerance,’’ his inpatient psychiatrist, Marilynn Stuart, initially was unable to provide any exam- ples. She ultimately described one incident in which the acquittee had playfully wrestled with another patient, as well as the fact that he could grow impatient with malfunctioning kitchen machinery. Although we assume that the trial court independently reviews all of the underlying testimony and evaluations, the board can be most helpful to the court by summarizing those assessments in a way that exposes, rather than obscures, the details relevant to the dangerousness inquiry. 24 Nothing in this opinion should be taken to mean that it is wrong to require an acquittee to demonstrate an ability to successfully transition to independent community living through conditional release before being discharged. Imposing 0, 0 CONNECTICUT LAW JOURNAL Page 31

0 Conn. 0 ,0 33 State v. Foster

that are predictive of violent or criminal conduct. This list is not necessarily exhaustive, but, before evidence is used in the prediction of dangerousness, there must be a valid basis to accord predictive value to that evidence. Also relevant are the acquittee’s index offenses, although the weight they carry should be proportionate to the recency, seriousness, and type of crime as those factors relate to the risk of recidivism. Similarly, the court may consider a history of criminal conduct prior to the index offenses, as well as testimony by mental health professionals that, although the acquittee has been generally not involved in the aforementioned con- duct and more or less free of symptoms for a prolonged period of time, in their professional opinion, the such a requirement is consistent with prevailing psychiatric practice, the empirical research establishing a strong negative correlation between successful condi- tional release and recidivism; see, e.g., M. Norko et al., supra, 34 Behav. Sci. & L. 439; and the legislative history of § 17a-593. See, e.g., Conn. Joint Standing Committee Hearings, Judiciary, Pt. 5, 1985 Sess., p. 1513, remarks of Howard Zonana, Associate Professor, Yale Department of Psychiatry (‘‘most hospitals want to have some opportunity to let somebody out gradually over increasing periods of time to see how they handle that responsibility especially after they’ve been either hospitalized or incarcerated for long periods of time’’); id., p. 1519, remarks of Peter Zeman, Chairman of the Advisory and Review Board of Whiting Forensic Institute (similar comments). There are two points, however, at which requiring a successful record of conditional release as a precondition for discharge becomes problematic. First, an acquittee may be deprived of the opportunity to successfully navigate conditional release for reasons that have little or nothing to do with his mental illness or dangerousness. This acquittee, for example, has had the opportunity for condi- tional release or other community visitation withheld as punishment for minor rule violations or unspecified ‘‘problematic interactions with staff’’ while commit- ted. In its July, 2018 report to the court, the board conceded as much, acknowledg- ing that the acquittee’s ‘‘adolescent rebellion . . . [had] contributed to his hospitalization for longer than clinically necessary.’’ (Emphasis added.) Second, we should be wary of imposing arbitrarily long conditional release requirements on the theory that we can never be sure that an acquittee will continue to thrive once the board’s supervision is withdrawn. The evidence is that individuals who have successfully navigated some period of conditional release have quite low recidivism rates, especially for felonies. See M. Norko et al., supra, 34 Behav. Sci. & L. 439–41. Page 32 CONNECTICUT LAW JOURNAL 0, 0

34 ,0 0 Conn. 0 State v. Foster

acquittee would be likely to decompensate, and would be unable or unwilling to promptly seek help, if not under the board’s supervision.25 Finally, there are considerations that are largely or entirely irrelevant. Amorphous fears or general societal ills that are not specifically connected to the acquittee have no place in the analysis. This means that a trial court exercising its authority under § 17a-593 must at all times remember that its role in protecting society is constrained by evidentiary fact (as well as the statu- tory and constitutional framework placing the burden on the state) and cannot be animated by fears emanating from outside of the realm of proof. The danger in think- ing of the statutory process primarily as a risk mitigation exercise is that we may lose track of the law’s command that an acquittee must be discharged after the maximum term of commitment unless the state proves by clear and convincing evidence that he poses an imminent danger to himself or others. Also irrelevant is idiosyncratic, or even offensive, behavior that leaves people feeling uncomfortable but bears no demonstrable relationship to dangerousness. See, e.g., Addington v. Texas, supra, 441 U.S. 426–27; O’Connor v. Donaldson, 422 U.S. 563, 575, 95 S. Ct. 2486, 45 L. Ed. 2d 396 (1975); State v. Krol, supra, 68 N.J. 259. The same can be said of character traits such as oppositionality, impulsiveness, low frustration toler- 25 This last consideration, if not cautiously invoked, can create a catch-22 that traps the acquittee in an institutionalized bind. Paradoxically, the only way an acquittee can disprove the alleged need for ongoing supervision is by being afforded some opportunity for independence. The same concern arises from the board’s statements that it was disinclined to recommend discharge because the acquittee had been committed for so long and, therefore, had limited experience in the community. Although there is research to support the value of successful supervised community release as a predictor of nondangerousness, this sort of reasoning also risks creating a catch-22. The longer an acquittee is committed, the less experience he or she will have in the community, and the more difficult the transition to independence will potentially be. 0, 0 CONNECTICUT LAW JOURNAL Page 33

0 Conn. 0 ,0 35 State v. Foster

ance, and poor social skills, even if linked to dangerous- ness, unless those traits are proven, by clear and convincing evidence, to result in material part from psychi- atric disability.26 More specifically, without sufficient evi- dence of dangerousness, an acquittee cannot be detained indefinitely on the ground that he does not always obey institutional rules, adhere to institutional policies, or defer without complaint to institutional authority. The constitu- tion simply does not allow it. IV Applying these principles to the case at hand, I conclude that much of the trial court’s analysis was simply irrele- vant, as I have discussed. Other considerations addressed by the court and the board, while relevant, are tempered by the passage of time. The acquittee committed the index offenses in 2001, when his schizophrenia was newly emerged and untreated, and he has no other criminal history.27 During his initial years at Whiting Forensic Divi- sion of the Connecticut Valley Hospital, when his mental illness was still uncontrolled, the acquittee engaged in some fights with other patients and had one escape attempt. His most recent physical altercation was in 2014, but the incident occurred while he was clinically stable, and the record does not reveal that his conduct was linked to his illness. Some brief refusals to take his medications were reported in early 2016, but he has been compliant since then. As of 2019, he had not heard voices since 2009, when he was taking a different medication, and he had not experienced other psychotic symptoms during that period. Given the singular and relatively minor nature of 26 See Regs., Conn. State Agencies § 17a-581-2 (a) (5) (‘‘ ‘[m]ental illness’ means any mental illness or mental disease as defined by the current Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association and as may hereafter be amended’’); see also General Statutes § 17a-580 (7) (‘‘ ‘[p]sychiatric disability’ does not include an abnormality manifested only by repeated criminal or otherwise antisocial conduct’’); part II B of this opinion. 27 Family and friends described the acquittee’s commission of the index offenses as being out of character. Page 34 CONNECTICUT LAW JOURNAL 0, 0

36 ,0 0 Conn. 0 State v. Foster

the index offenses, one would think that—in light of the burden of proof, coupled with the clear and convincing standard of proof—a multiyear record of medication com- pliance, a lack of psychiatric symptoms,28 and an absence of violent or criminal conduct, along with fifteen months of successful conditional release, would more than coun- terbalance this limited history of dangerousness. Indeed, this perspective is reflected in a July, 2019 report to the board in which Michael F. Genovese, a licensed clinical social worker commissioned by the state to evaluate the acquittee, concluded that, following a year of conditional release, ‘‘he does not present a significant risk to himself or others at this time.’’29 Still, our standard of review is deferential, and, in my view, the record is sufficient, if barely so, to support the trial court’s conclusions. There was recent, albeit conclu- sory, testimony from the acquittee’s consulting forensic psychiatrist that he would pose a risk to himself or others without ongoing supervision.30 Also, in late 2016 and 2017, 28 In 2019, there was a brief notation in his records that the acquittee sometimes believes that ‘‘bad things happen in the world . . . because of him,’’ but there was no testimony that such beliefs are symptomatic of his mental illness or indicative of dangerous propensities, and this item was not relied on by the trial court. 29 In general, the recidivism rate for discharged acquittees is quite low, much lower than for convicted inmates upon their release from prison. See, e.g., M. Norko et al., supra, 34 Behav. Sci. & L. 425, 439, 441; J. Stankowski, ‘‘Acquittal by Reason of Insanity Is a Positive Outcome for Defendants Who Cannot Be Restored,’’ 48 J. Am. Acad. Psychiatry & L. 244, 246 (2020). Of 196 acquittees discharged between 1985 and 2015 in Connecticut, only 16.3 percent were rear- rested. M. Norko et al., supra, 432, 435–36. The majority of the rearrests were for infractions, misdemeanors or class D felonies. Id., 436. None was for class A felonies, and the average period from discharge to rearrest was nearly six years. Id. 30 I am not comforted by the fact that the board and some of the acquittee’s treatment staff have been repeating generic, conclusory statements of this sort for years, and have used them to justify requests for periods of recommitment of as long as five years. If the board is of the view that, despite years of compliance, nonviolence, and steady progress, the acquittee cannot be trusted to manage his disease independently, then the onus is on the board to explain and justify that belief and, within the bounds of safe patient management, to give him an opportunity to prove the board wrong. 0, 0 CONNECTICUT LAW JOURNAL Page 35

0 Conn. 0 ,0 37 State v. Foster

there had been several incidents in which the acquittee reasonably could be understood to have physically or ver- bally threatened female staff members, although it is ques- tionable that such conduct was related to his mental illness.31 See footnote 19 of this opinion. The acquittee also has periodically expressed skepticism about his need to continue taking psychiatric medications, which, depending on the context of the statement, may raise a legitimate cause for concern. At the end of the day, I am inclined to think that, with one-quarter century having passed since the acquittee committed the index offenses and many years since his last physical altercation, the only substantial questions to be resolved are (1) whether he is likely to continue to take his medications, to seek treatment for his mental illness, and to refrain from alcohol and drug use when he is no longer under the board’s supervision, (2) if he fails to take those measures, whether his symptoms are likely to reemerge in a manner that might realistically inhibit his ability to seek medical attention, and (3) if those symptoms reemerge, whether it is likely that he will engage in behav- ior dangerous to himself or others. See, e.g., State v. Dyous, supra, 307 Conn. 311; State v. Corr, supra, 87 Conn. App. 727–28; see also State v. Putnoki, supra, 200 Conn. 222 (prior to Metz decision that adopted clear and convincing evidence standard, court assumed that trial court must find ‘‘that it was more probable than not that the defendant’s progress was dependent on her medication and therapy’’). To keep the acquittee committed, the state must establish all three propositions by clear and convincing evidence. The focus of future proceedings should be on addressing those questions with as much clarity, specificity, and scien- tific authority as reasonably possible. For these reasons, I join in part I of the majority opinion and join in the result reached in part II. 31 Because one incident followed a failed change to his medication, the trial court reasonably could have concluded that there was such a connection, although it made no such finding.