United States v. Caporale, 701 F.3d 128 (4th Cir. 2012)

OPINION

KING, Circuit Judge:

The government appeals the judgment of the district court directing that Patrick Caporale be freed from the custody of the Bureau of Prisons and granted supervised release. Caporale finished serving his prison sentence for child molestation in 2008, but he has remained incarcerated while the government seeks to have him declared a “sexually dangerous person” pursuant to the civil-commitment provisions of the Adam Walsh Child Protection and Safety Act of 2006 (the “Walsh Act”), Pub.L. No. 109-248, 120 Stat. 587, as specifically set forth in 18 U.S.C. § 4248.

A sexually dangerous person under the Walsh Act means one “who has engaged or attempted to engage in sexually violent conduct or child molestation and who is sexually dangerous to others.” 18 U.S.C. § 4247(a)(5). A person is sexually dangerous to others insofar as he or she “suffers from a serious mental illness, abnormality, or disorder,” and, as a result, “would have serious difficulty in refraining from sexually violent conduct or child molestation if released.” Id. § 4247(a)(6). The parties have never disputed that, as evidenced by his several convictions, discussed infra, Caporale satisfies the first, prior-conduct element of § 4247(a)(5) by having engaged in child molestation.

Following an evidentiary hearing whose scope was thereby limited to the second element of § 4247(a)(5), the district court ruled that, as a matter of law, the government had not proved that Caporale suffers from a serious mental illness, abnormality, or disorder. The court perceived in the alternative that even if Caporale were so afflicted, his commitment was not required because the government had also failed to sufficiently show that Caporale will experience serious difficulty in refraining from sexually violent conduct or child molestation if released.

We conclude that, contrary to the district court’s legal determination and as established by the evidence, Caporale indeed suffers from a qualifying mental impairment. We nevertheless affirm the judgment below, discerning no clear error in the court’s alternative rationale that the government fell short of carrying its burden to demonstrate a relative likelihood that Caporale will reoffend.

I.

A.

Caporale, fifty-nine, has a history of sexual offenses involving minors. In 1980, he pleaded guilty in New York to fourth-degree criminal facilitation for recruiting about twenty boys, aged thirteen to sixteen, to have sex with an adult female acquaintance. Then, in 1984, Caporale pleaded guilty to state charges of second-degree sexual abuse for subjecting a thirteen-year-old boy to sexual contact and for

masturbating another boy, twelve years old. In 1986, Caporale again pleaded guilty in state court to acting in a manner injurious to a child, after having persuaded three boys (age fifteen to sixteen) and a girl (age sixteen) to have sex in his apartment while he watched.

Similar conduct in Maine involving a seventeen-year-old boy and the boy’s underage female acquaintances — this time evidenced by photographs and videotapes — resulted in Caporale’s 1992 federal indictment for child molestation and for possession of child pornography. Caporale pleaded guilty to six counts of using a minor to engage in sexually explicit conduct, for which he was sentenced to eighty-four months in prison; he served most of that sentence before being granted supervised release in August 1998. On December 14, 1999, Caporale was arrested and charged with endangering the welfare of a minor following the discovery by local authorities that he had discussed masturbation with a fourteen-year-old male neighbor. The charge was a parole violation, for which Caporale was returned to prison in September 2000.

Caporale was again released to supervision in June 2001, and, two years later, was once more found to have violated his conditions of release by associating with a felon, a man who Caporale described as a pedophile. On September 12, 2003, federal probation officers arrived unannounced at Caporale’s residence and arrested him upon finding numerous videotapes and photographic depictions of underage boys engaged in sexual activity. Caporale subsequently pleaded guilty in New York to a single count of possessing photographs of an obscene performance of a child.

Caporale’s myriad violations ultimately resulted in the revocation of his supervised release. Consequently, Caporale returned to federal prison in 2003 to serve out the sentence imposed on his 1992 conviction. Caporale completed his sentence on March 21, 2008, and he would have been released but for the government’s filing that day of a certification pursuant to the Walsh Act, which provides, in pertinent part:

18 U.S.C. § 4248(a).

The district court was thus required to convene a hearing to afford the government the opportunity to prove the ultimate truth of its certification, i.e., that Caporale is sexually dangerous. An inmate so declared must be kept in federal custody until the danger has passed, or until a state agrees to assume responsibility for the inmate’s custody, care, and treatment. See 18 U.S.C. § 4248(d). As it was uncontested that Caporale’s criminal history satisfied the first element toward a determination of sexual dangerousness, it was the government’s burden at hearing to establish, by clear and convincing evidence, both prongs of the second element: (1) that Caporale is impaired by a serious mental illness, abnormality, or disorder, such that (2) he would, if released, have serious difficulty refraining from sexually violent conduct or child molestation. See id. § 4247(a)(6).

B.

All proceedings having been stayed during the pendency of constitutional chai

lenges to the commitment proceedings wrought by the Walsh Act, see United States v. Hall, 664 F.3d 456, 461 n. 2 (4th Cir.2012) (reciting litigation history beginning with Supreme Court’s ruling in United States v. Comstock, — U.S. —, 130 S.Ct. 1949, 176 L.Ed.2d 878 (2010), and progressing through our decision on remand, United States v. Comstock, 627 F.3d 513 (4th Cir.2010), and its companion case, Timms v. Johns, 627 F.3d 525 (4th Cir.2010)); see also United States v. Timms, 664 F.3d 436 (4th Cir.2012), Caporale’s hearing at last took place on March 5, 2012. The key witnesses were Dr. Gary Zinik and Dr. Lela Demby for the government, and Dr. Joseph Plaud for the defense, all of whom are licensed psychologists and acknowledged experts. All three reviewed the available records, and Dr. Plaud personally interviewed Caporale on November 28, 2011.

1.

According to Dr. Plaud, Caporale is sexually interested in pubescent to post-pubescent males, “12, 13 up to 16, 17 years old.” J.A. 253.1 Dr. Plaud explained that, inasmuch as the rate of physical maturity may differ depending on the person, stages of development defy rigid definition by age, and that it was more accurate to say that Caporale’s sexual interests correspond approximately to Stages III-V on the five-stage Tanner scale of physical development. See id. at 253-54. Dr. Zinik essentially concurred with Dr. Plaud, noting that Caporale is attracted to “[pubescent boys, ages approximately 11 to 14.” Id. at 157.

In contrast to her two colleagues, Dr. Demby opined that Caporale prefers “prepubescent boys ... who ha[ve] not yet developed the majority of their secondary sexual characteristics ... as young as 11 and 12.” J.A. 207. Asked to elaborate, Dr. Demby noted that, as set forth in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (“DSM”), age thirteen is the presumed ceiling for a diagnosis of pedophilia, a condition in which the afflicted individual is attracted to pre-pubescent children. See id. at 209.

Though not confronted directly with Dr. Demby’s testimony, Dr. Plaud flatly rejected her diagnosis:

J.A. 253. Dr. Zinik agreed with the defense on this point, as confirmed by the district court’s questioning:

THE WITNESS: He’s not interested in prepubescent children.

Id. at 169. Indeed, Dr. Demby acknowledged that “individually, [puberty] can happen at different ages for different children,” and she frankly admitted that it would “be possible for somebody who is say, 12 years old to be pubescent.” Id. at 209. Dr. Demby nonetheless diagnosed Caporale with “pedophilia, sexually attracted to males exclusive type, and personality disorder not otherwise specified of dependent traits.” Id. at 207.

Dr. Zinik, having ruled out pedophilia, concluded that Caporale suffers from “paraphilia ... not otherwise specified” (“paraphilia/NOS”). J.A. 161. Dr. Zinik testified that paraphilia is characterized by “recurrent, intense sexual fantasies, urges or behaviors that ... last at least six months and ... cause a significant impairment or disability in the individual’s life.” Id. at 167. Dr. Zinik occasionally referred to Caporale’s affliction as “hebephilia,” which is not a listed diagnosis in the DSM, but a colloquial term understood by mental health professionals to describe the proclivity of “men whose erotic interest centers on pubescents.” Ray Blanchard et ak, Pedophilia, Hebephilia, and the DSM-V, 38 Archives of Sexual Behavior 335, 336 (2009).

Only eight specific manifestations of paraphilia — pedophilia, transvestic fetishism, exhibitionism, fetishism with respect to inanimate objects, voyeurism, sexual masochism, sexual sadism, and frotteurism (sexual rubbing against non-consenting persons, often in crowded areas) — are documented in the Fourth Edition of the DSM, a treatise accepted as authoritative by all the testifying experts. More generally, however, the DSM lists paraphilia/NOS, which, Dr. Zinik explained, “is a residual category in which really all — any other paraphilia could be included.” J.A. 167. According to Dr. Zinik, a paraphilia’s inclusion within the catch-all should not depend on whether it is diagnosed with a frequency similar to those specifically listed, but should merely consider whether it impairs normal functioning to a similar extent. See id.

In Dr. Zinik’s opinion, Caporale’s paraphilia “has caused serious impairment and dysfunction in his life, because he keeps getting arrested and ... he’s never really done much with his life, because of this obsession.” J.A. 161. Dr. Demby agreed with Dr. Zinik that impairment is the hallmark of a mental infirmity, explaining that “[t]he mere presence of sexual attraction to pubescent children in and of itself does not qualify as a mental disorder. The disorder comes when it causes one to have significant impairment in major spheres of functioning in one’s life.” See id. at 210.

In opposition to the government’s experts, Dr. Plaud opined that Caporale suffers from no qualifying illness, abnormality, or disorder under § 4247(a). Dr. Plaud observed that Caporale’s sexual liaisons with pubescent males “is illegal to a certain extent ... upwards, depending on the jurisdiction. But it is not disordered.” J.A. 251. Emphasizing the lack of a discrete DSM listing for Caporale’s affliction, Dr. Plaud reasoned that “if it’s ... truly a disorder, it should have its own diagnostic criteria. [I]f you have something that’s high frequency and it’s truly considered by the professional community to be a disorder, then you’ve got to have a diagnosis for it.” Id. at 252.

Dr. Plaud took issue with Dr. Zinik’s conclusion that, for purposes of the Walsh Act, hebephilia can be shoehorned into the DSM-listed diagnosis of paraphilia/NOS, explaining that “[fit’s only more recently that this has become something, I think, as a consequence of what’s going on in the larger civil commitment world.... And I

do not believe a paraphilia NOS is ... for that purpose. It’s there for low frequency sexual aberration.” J.A. 253. Cross-examination served to further develop Dr. Plaud’s position. Asked by the government’s counsel whether sexual acts with a thirteen-year-old boy are deviant, Dr. Plaud responded:

Id. at 268.

2.

The experts also differed on whether, upon release, Caporale would have serious difficulty in refraining from child molestation. Dr. Zinik, referring to pornographic stories about teenage boys composed in 2008 by Caporale and other inmates using a prison computer, along with suggestive photographs seized from Caporale’s cell in May 2011, opined that Caporale is “still pretty charged up about pubescent boys, and that he would have, as a result of that, serious difficulty controlling himself in the community if he were in contact with pubescent boys as he has done in the past.” J.A. 179. Using three actuarial scales to compute the risk of recidivism for offenders having similar characteristics, Dr. Zinik recited that two of the scales scored Caporale’s profile “high risk” and the third “moderate high risk.” See id.

Dr. Demby considered the stories and photos “evidence of sexual preoccupation, attitudes that condone sexual contact with children as recently as 2011,” which, in her opinion, “exacerbates Mr. Caporale’s risk.” J.A. 227. Noting that Caporale had, in the past, personally taken photographs of illegal activity, Dr. Demby also took into account Caporale’s more recent writings to differentiate his behavior from that typically exhibited by a passive consumer of pornography. See id. at 228. Dr. Demby conceded that writing such stories was not itself “sexually dangerous” in the sense that it did not equate to molestation or violent conduct, but she offered that “if someone knows that they are at risk ... of molesting children and reengaging in past behavior ... to continue to create and read ... and reiterate these themes ... it’s leading him down that same path of a hands-on offense, of re-doing what ... we already know he’s done in the past.” Id. at 233.

Dr. Plaud, on the other hand, thought it more significant that, since 1992, Caporale had committed only noncontact offenses. Caporale had further demonstrated volitional capacity, according to Dr. Plaud, by having ceased writing stories since May 2011, almost a year prior to the hearing. See J.A. 255, 261, 265-66. Dr. Plaud was reluctant to use actuarial tools to predict Caporale’s probability of recidivism, testifying that such tools more accurately serve to exclude test subjects as candidates to engage in molestation, rather than to pinpoint them as likely offenders. See id. at 258-59. Dr. Zinik also acknowledged the tools’ limitations. See id. at 179. Dr. Plaud emphasized, moreover, that even those tools indicating that Caporale posed a “high risk” of recidivism more particularly defined that risk as amounting to less than a twenty-five percent chance over eight years. Indeed, as Dr. Plaud pointed

out, the actuarial risk decreases to less than ten percent for men sixty and over, an age that Caporale has nearly attained. See id. at 260.

■C.

On April 20, 2012, the district court issued a written decision, designated its “Findings of Fact and Conclusions of Law” (the “Findings”), in which the court set forth its ruling against the government.2 Most significantly, the court concluded that hebephilia — whether elevated to a titular condition or deemed a shorthand way to describe the behavioral predicate underlying a more formal diagnosis of paraphilia/NOS — “is not an appropriate basis for civil commitment under the [Walsh] Act.” Findings 6. This is so, the court reasoned, because whether hebephilia constitutes a serious mental illness, abnormality, or disorder “is a contested issue in the mental health community.” Id. In that respect, the court found “more credible the testimony of Dr. Plaud, who opined ... that Mr. Caporale currently suffers from no sexual disorder.” Id. at 7.

The district court also regarded Dr. Plaud’s testimony more credible concerning the issue of whether Caporale would have serious difficulty refraining in the future from sexually violent conduct or child molestation. The court stressed that Dr. Plaud had focused pertinently on evidence of Caporale’s current condition, more so than Dr. Zinik and Dr. Demby, who based their evaluations in substantial part on Caporale’s historical misconduct. See Findings 9-10.

Consistently with its written decision, the district court entered judgment for Caporale and ordered his release. The government timely appealed, and, contemporaneously therewith, filed an emergency motion for a stay of the district court’s order, which we granted. The parties’ respective arguments having thereafter been fully briefed and argued on an expedited basis, the matter is ripe for our disposition.

II.

To the extent that the district court’s judgment in this Walsh Act proceeding rests on the accuracy of its findings of fact, our review is for clear error. See United States v. Hall, 664 F.3d 456, 462 (4th Cir.2012). The court’s legal conclusions, however, are paid no deference. See id. It is beyond our limited role as an appellate court to disturb the judgment below solely on the ground that we would have decided to the contrary. See id. (citing Anderson v. Bessemer City, 470 U.S. 564, 573, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985)). Precedent dictates that “ ‘[i]f the district court’s account of the evidence is plausible in light of the record viewed in its entirety, the court of appeals may not reverse it even though convinced that had it been sitting as the trier of fact, it would have weighed the evidence differently.’ ” United States v. Wooden, 693 F.3d 440, 451 (4th Cir.2012) (quoting Bessemer City at 573-74, 105 S.Ct. 1504). We must therefore permit the court’s judgment to stand unless, with respect to its factual underpinnings, we are “ ‘left with the definite and firm conviction that a mistake has been committed.’ ” Wooden, 693 F.3d at 451 (quoting Easley v. Cromartie, 532 U.S. 234, 242, 121 S.Ct. 1452, 149 L.Ed.2d 430 (2001)).

III.

A.

1.

The experts at hearing were asked whether Caporale suffers from a serious

mental illness, abnormality, or disorder, and they each strove to answer that question within the strictures of the DSM, which is familiar territory to mental health professionals. Dr. Demby perceived Caporale’s sexual aberrations as suggesting a specific DSM diagnosis, while Dr. Zinik thought the aberrant behavior corresponded more closely to a less rigorous, residual category of diagnoses yet within the DSM. Dr. Plaud was of the opinion that Caporale’s condition conformed to no listed diagnosis, which in turn precluded the possibility that Caporale suffers from any serious mental impairment.

The district court’s resolution of the issue in reliance on Dr. Plaud’s testimony cannot be squared with United States v. Carta, 592 F.3d 34 (1st Cir.2010), another Walsh Act proceeding involving a diagnosis of hebephilia, in which the lower court’s judgment on behalf of the respondent was premised on a legal analysis similar to that under review here. In Carta, the respondent’s expert “asserted that hebephilia was not a generally accepted diagnosis in the mental health community, did not fit within the DSM definition of paraphilia, lacked diagnostic criteria and could not be consistently defined.” Id. at 38. Consistently with that testimony, the district court entered judgment in favor of the respondent, ruling that hebephilia is not a serious mental illness, abnormality, or disorder.

The First Circuit vacated and remanded, observing that the lower court “may have assumed that the statutory concept is delimited by the consensus of the medical community, but this is not so. Further, a mental disorder or defect need not necessarily be one so identified in the DSM in order to meet the statutory requirement.” Carta, 592 F.3d at 39-40. The Carta court’s rationale appears sound, as one will search § 4247(a)(6) in vain for any language purporting to confine the universe of qualifying mental impairments within clinical or pedagogical parameters. The statute could have been drafted to comport with clinical norms, but inasmuch as Congress chose not to do so, it has been left to the courts to develop the meaning of “serious mental illness, abnormality, or disorder” as a legal term of art. See Kansas v. Hendricks, 521 U.S. 346, 359, 117 S.Ct. 2072, 138 L.Ed.2d 501 (1997) (explaining that “the term ‘mental illness’ is devoid of any talismanic significance,” and instructing that legislative development of specialized “[Ijegal definitions ... need not mirror those advanced by the medical profession”).

Such an approach is by no means unprecedented. Under the Black Lung Act, for example, we have acknowledged that legal pneumoconiosis encompasses a more inclusive category of ailments than the purely clinical diagnosis, with the result that a claimant’s entitlement to benefits is evaluated under the broader definition. See Clinchfield Coal Co. v. Fuller, 180 F.3d 622, 625 (4th Cir.1999). When the presence of legal — as opposed to clinical or medical — pneumoconiosis is at issue in a black lung case, eligibility for benefits ultimately depends on whether the disease has caused a worsening of symptoms to the extent that the miner’s respiratory or pulmonary functioning has been impaired. See id.3

The analogy to the Black Lung Act is especially apt here, in light of the government experts having testified without contradiction that mental infirmities in general (and paraphilia in particular) have little meaning except to the extent they are manifested by an impairment in the subject’s ability to function as a normal human being. In large part for that reason, the court in Carta had no difficulty concluding that hebephilia is a qualifying illness, abnormality, or disorder within the meaning of § 4247(a)(6), observing that the respondent “has a decades-long sexual fixation on minors that plainly has caused significant distress or impairment in his life.” 592 F.3d at 40 (internal quotation marks omitted).

The First Circuit in Carta made short shrift of the lower court’s rejection of hebephilia as fulfilling the serious-impairment prong, reasoning that the condition also fits within the rubric of paraphilia/NOS, which “is expressly a DSM-listed disorder.” 592 F.3d at 40. In the case at bar, Dr. Plaud’s testimony cast some doubt that hebephilia may so qualify, but the scope of “illness, abnormality, or disorder” in § 4247(a)(6) is certainly broad enough to include hebephilia, by its own or any other name. The district court’s determination to the contrary was thus incorrect as a matter of law. Though his condition may elude definitive labeling, the evidence at hearing established beyond question that Caporale’s ability to function normally in society has been preempted by his sexual fixation on underage, pubescent boys, such fixation having heretofore so dominated his psyche as to substantially impair and disrupt his life.4

2.

The substantial delays occasioned by the constitutional litigation concerning the Walsh Act, see supra Part I.B, resulted in a backlog of commitment proceedings in the district courts. Those proceedings now appear to be moving forward in earnest, and, as a consequence, Caporale’s is already the fourth merits-based appeal of a § 4245 judgment that we have considered, all within the past year. The three previous appeals having been disposed of via published opinions, and each having required an analysis of the hearing evidence under the serious-difficulty prong of § 4247(a)(6), a review of those prior decisions is in order.

a.

In United States v. Hall, 664 F.3d 456 (4th Cir.2012), we affirmed the district court’s ruling in favor of the respondent on the serious-difficulty prong. The diagnosis in Hall was pedophilia, and it was undisputed. The government’s two experts, including Dr. Demby, based their serious-difficulty conclusions primarily on actuarial

tools, the respondent’s ongoing interest in collecting pornographic drawings and photographs, and his general noncompliance with rules and directions while on supervised release and even following his certification. See id. at 464.

The defense expert, in turn, stressed the respondent’s limited history of offenses and victims, his participation in a treatment program, and his abstention from hands-on offenses during his twenty-eight months of release. See Hall, 664 F.3d at 464. The respondent’s expert pointed to recent research linking possession of pornography with a lower recidivism rate, adding that the stringent release provisions and possibility of a life sentence for further offenses would serve as powerful deterrents. See id. at 465.

The district court found the defense expert to be the most credible of the three, discerning that he had more completely discussed the actuarial tools and psychological tests administered to evaluate the respondent. See Hall, 664 F.3d at 465. These tools assessed the respondent’s recidivism risk as being between ten to thirty-six percent within ten years following release. See id. at 466. The court relied on the defense expert’s testimony regarding the respondent’s control on his prior release and the deterrent effect of the conditions attached to the proposed release. See id. Although the government took issue with the determinative weight afforded the respondent’s lack of offense conduct when released previously, we declined to assign clear error, deferring to the district court’s analysis and its credibility assessments, including that of the respondent himself, who indicated remorse and an understanding of the risks of recidivism. See id. at 465-67.

Not long thereafter, in United States v. Francis, 686 F.3d 265 (4th Cir.2012), we again affirmed the district court’s judgment in favor of the respondent. In Francis, a government expert diagnosed the respondent with paraphilia/NOS (telephone scatalogia), i.e., the propensity to make obscene and threatening phone calls; another diagnosed sexual sadism and antisocial personality disorder. Id. at 271. In supporting the government’s case on the serious-difficulty prong, these two experts relied heavily on actuarial tools and, to a lesser extent, on the respondent’s perceived hostility toward women and his noncompliance with supervised release. Id. The respondent’s risk of recidivism over a five-year period was actuarially projected to be between approximately thirty and sixty percent. Id.

The two experts for the respondent, including Dr. Plaud, either agreed with or did not seriously dispute the government’s diagnosis of paraphilia/NOS (telephone scatalogia). The defense instead primarily contested the serious-difficulty prong, urging the court to consider the respondent’s abstention from reoffending during the seven months of his prior release and contending that the actuarial tools employed by the government’s experts were of no use to predict nonphysical behavior. See Francis, 686 F.3d at 272. Employing his own methodology, one of the defense experts testified that the respondent actually posed a low risk of recidivism. See id.

In evaluating the evidence, the district court assumed without deciding that the government had established the first, pri- or-conduct element. See Francis, 686 F.3d at 272-73. No clear finding was made with respect to the serious-impairment prong of the second element, but the circumstances attendant to the court’s discussion required us to assume that it had implicitly resolved the issue in favor of the respondent. See id. at 276. Be that as it may, the court explicitly ruled in favor of the respondent on the serious-difficulty

prong, according greater weight to the testimony of the defense experts. The court noted particularly those experts’ evaluation of the respondent’s behavior on prior release, and it credited their opinions that the government’s actuarial evidence was of scant probative value. Id. at 273. On appeal, we concluded that the court’s analysis in that regard was sufficiently cogent and thorough. Id. at 276-77.

Most recently, in United States v. Wooden, 693 F.3d 440 (4th Cir.2012), we reversed the district court’s determination in favor of the respondent on both second-element prongs, and we remanded the matter for reconsideration. Like Hall, Wooden involved a diagnosis of pedophilia, but on this latter occasion we observed that the court had improperly relied on the testimony of the defense expert, in that the expert could not square his opinions with the undisputed facts, particularly that the respondent: (1) was continuing to have intense fantasies and urges involving prepubescent children; (2) had offered testimony that was littered with rationalizations, denials, and other cognitive distortions or “thinking errors”; (3) was having difficulty distinguishing between dreams and reality with respect to an incident of molestation that was alleged — but not proved — to have occurred in 2005 during his parole period; (4) had admitted to attempting to molest another child perhaps a year before that; (5) had sent a Christmas card to the child in the alleged 2005 incident, in which he promised that, if he were out of prison, “I would do a whole lot for you”; and (6) had, while on parole, hired himself out as a babysitter for young children. See Wooden, 693 F.3d at 444-45, 452-56.

We observed that the evidence set forth above was relevant to both the serious-impairment and serious-difficulty prongs, see 693 F.3d at 458, and with especial regard to the latter, did not support the defense expert’s fanciful theory that the respondent’s tested reluctance to engage generally in impulsive behavior necessarily meant that he could exercise volitional control of his deviant sexual urges. The expert’s opinion, we noted, did not account for the documented existence of patient, calculating child molesters who groom their victims. See id. at 457. Moreover, the record revealed that, beyond the laboratory context in which the expert had conducted his evaluation, the respondent had in fact acted impulsively during his prior crime, and had also evidenced his impulsiveness through his inability to adhere to institutional rules while incarcerated. Id. at 458.

Other reasons offered by the defense expert to support his opinions suffered from similar logical flaws. For example, the expert cited the respondent’s disinterest in child pornography as evidence of docility, yet neglected to acknowledge the utter lack of evidence that the respondent had ever been so interested. See Wooden, 693 F.3d at 448-49, 455. In addition, the expert failed to account for the respondent’s attempted molestation of one child in 2004, his correspondence thereafter with another child via Christmas card, and his decision to place himself in proximity to other children through babysitting, all of which we explained “speaks directly to the serious-difficulty prong.” Id. at 455, 459. Other relevant considerations included the respondent’s criminal history, see id. at 458, his testimonial admissions, see id. at 459, and any cognitive limitations, see id. at 460.

In light of all the circumstances, we concluded that the district court had clearly erred in relying upon the testimony of an expert who “largely ignored all contradictory evidence” and whose “analysis was internally inconsistent.” Wooden, 693

F.3d at 455. We also determined that the court committed an error of law as to the government’s burden of proof by demanding that it demonstrate, through its actuarial models, not merely that the respondent was at high or very high risk of reoffending within five years, but that the likelihood could be quantified in excess of fifty percent. See id. at 447-48, 450, 456-57, 460-62.

b.

The foregoing recitation illustrates the principle that “while clear-error review is deferential, it is not toothless.” Wooden, 693 F.3d at 452 (citation and internal quotation marks omitted). We may discern clear error when a court “makes findings without properly taking into account substantial evidence to the contrary.” Id. at 451 (quoting Francis, 686 F.3d at 273). In Wooden, we reiterated the familiar principle that such error “occurs when a district court’s factual findings are against the clear weight of the evidence considered as a whole.” Id. at 462 (citation omitted).

In support of reversing the judgment below in favor of Caporale on the serious-difficulty prong, the government contends that the district court misplaced its reliance on the expert testimony proffered by the defense, in that Dr. Plaud failed to adequately consider: (1) one of the boys victimized by Caporale’s 1992 offense, though seventeen at the time, had been sexually involved with Caporale for four years; (2) Caporale’s 1999 discussion of masturbation with a fourteen-year-old boy constituted grooming behavior; and (3) Caporale associated with a pedophile in 2003. The government also suggests that the court’s resolution of the serious-difficulty prong was fatally colored by its misanalysis of the serious-impairment prong. Cf. Wooden, 693 F.3d at 457 (noting government’s argument “that the district court’s error in concluding that Wooden does not suffer from pedophilia prevented the court from properly assessing the serious-difficulty question”).

The defense case was by no means irrefutable, but, after reviewing the entire record, it would be inaccurate for us to say that Dr. Plaud was anything but thoroughly familiar with Caporale’s history of molestation. Dr. Plaud readily acknowledged that Caporale was interested in boys as young as twelve or thirteen, but regarded Caporale’s involvement with older minors in 1992 — though similarly unlawful and perhaps no less reprehensible — as perhaps indicating an evolving interest in those who have reached “an adult type of sexual maturity,” J.A. 266. On cross-examination, the government specifically asked Dr. Plaud whether the 1992 incident justified enhancing his previous estimate of Caporale’s recidivism risk. Dr. Plaud responded, “[I]t argues quite to the contrary.” Id. at 265.5

Although the government is correct that the district court did not address Caporale’s supposed grooming conduct in 1999 or his prohibited association with a pedo

phile in 2003, it is also the case that neither of those episodes featured prominently in the experts’ testimony. Dr. Zinik characterized the 1999 incident as grooming in his answer to a single question on the second round of cross-examination conducted by Caporale’s counsel, see J.A. 194-95, and, apart from the most ephemeral of allusions as part of a question posed to Dr. Demby on the subject of treatment, see id. at 212, the 2003 incident was not touched upon at the hearing, but discussed only in a single paragraph of Dr. Zinik’s written report, see id. at 375. While the court’s analysis would have been more comprehensive had it included a discussion of each secondary point, these particular omissions amount to little more than No. 8 needles in a stack of alfalfa.

The battleground at hearing on the serious-difficulty prong instead proceeded on two fronts: the probative value of the government’s actuarial evidence, and the meaning to be accorded the pornographic materials seized from Caporale’s cell during May 2011. The parties agreed that it was impossible to predict with exactitude whether Caporale would reoffend, but that his profile indicated a risk ceiling of about twenty-five percent, which would tend to decrease as he grew older.6 Under the circumstances, it was not unreasonable for the district court to credit Dr. Plaud’s opinion that the actuarial evidence militated toward a conclusion in favor of Caporale. Given our holding in Wooden that a recidivism risk ceiling of fifty percent did not rule out the respondent’s commitment, see 693 F.3d at 460-61, the court may well have reasonably reached the opposite conclusion. But it was not required to.

Similarly, the evidence of Caporale’s recent activities cut both ways. It is hardly surprising that Caporale’s continued predilection for written pornography and suggestive photographs depicting teenage boys would generate some skepticism among the government’s experts regarding his volitional control, though it might be argued as more relevant to the serious-impairment prong. At the same time, that no such materials had been found in Caporale’s possession during the ten months prior to hearing plausibly supports — albeit to a limited degree — an inference of control. See Francis, 686 F.3d at 276 (relating with approval district court’s characterization of respondent’s conformance with the law for more than seven months on release as “compelling evidence” of his volitional control); Hall, 664 F.3d at 465-66 (rejecting government’s contention that district court clearly erred by crediting respondent’s lawful behavior on release for twenty-eight months as probative of his capacity for control). The failure to discover pornographic contraband in Caporale’s cell after May 2011 might also have supported the inference that he became better at concealing it, but it was manifestly the task of the factfinder to choose between competing reasonable inferences.

The district court’s analysis is not immune from attack, however, most notably its preference for Dr. Plaud’s testimony based on the perception that his opinion on the serious-difficulty prong considered Caporale’s current state more thoroughly than the opposing experts. See Findings 9-10. The court’s emphasis on current information was somewhat of a straw man, inasmuch as the government’s experts were not permitted access to Caporale and were therefore constrained to rely solely on their forensic reviews of the record. Moreover, we do not share the court’s

apparent disdain for historical evidence, having observed that a respondent’s criminal record “may well be a historical factor, but it is by no means a stale or irrelevant one. When the question is whether an inmate ... will have serious difficulty refraining from re-offending if released, consideration of the nature of his prior crimes provides a critical part of the answer.” Wooden, 693 F.3d at 458.7

Lastly, although the district court did not specifically indicate any reliance upon them, other aspects of the case support the underlying judgment. For instance, Caporale’s deposition testimony evidenced a normal intelligence, thus allaying concerns that cognitive impairments could potentially “affect [the respondent’s] ability to refrain from re-offending if released.” Wooden, 693 F.3d at 460. Caporale expressed remorse for his past behavior, and he demonstrated the ability to differentiate between right and wrong. See Hall, 664 F.3d at 466. Caporale also testified that the counseling and treatment he had received for his condition had prepared him for the rigors of release, enabling him to recognize and avoid high-risk circumstances that might otherwise tempt him to engage in wrongdoing. See id.

Here, the relative strength of the parties’ evidence and the logical force of the district court’s analysis paint a picture that is more reminiscent of Hall and Francis than it is of Wooden. Though the government attempts to nitpick Dr. Plaud’s opinions, his testimony, by and large, told “ ‘a coherent and facially plausible story that is not contradicted by extrinsic evidence.’ ” Wooden, 693 F.3d at 452 (quoting Anderson v. Bessemer City, 470 U.S. 564, 573, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985)). It is therefore unlikely that the court clearly erred in relying upon it.8

B.

Although the district court incorrectly ruled that impairment through hebephilia cannot be, as a matter of law, a serious mental illness, abnormality, or disorder within the meaning of the Walsh Act, it did not commit clear error by alternatively crediting Dr. Plaud’s testimony that Caporale possesses sufficient volitional control to prevent him from having serious difficulty refraining from sexually violent conduct or child molestation in the future. Under the circumstances, we are not left with the definite and firm conviction that the court was mistaken in concluding that the government had failed to carry its burden of proving Caporale a sexually dangerous person by clear and convincing evidence.

IV.

Pursuant to the foregoing, we affirm the judgment of the district court.