UNITED STATES of America, Plaintiff-Appellee, v. Matthew Henry WEBER, Defendant-Appellant.
Penile plethysmograph testing is a procedure that “involves placing a pressure-sensitive device around a man's penis, presenting him with an array of sexually stimulating images, and determining his level of sexual attraction by measuring minute changes in his erectile responses.” Jason R. Odeshoo, Of Penology and Perversity: The Use of Penile Plethysmography on Convicted Child Sex Offenders, 14 Temp. Pol. & Civ. Rts. L. Rev.. 1, 2 (2004). Although one would expect to find a description of such a procedure gracing the pages of a George Orwell novel rather than the Federal Reporter, plethysmograph testing 1 has become routine in the treatment of sexual offenders and is often imposed as a condition of supervised release. We address the procedures that must be followed before a district judge may impose such a requirement on a criminal defendant.
In May of 2001, an electronics store technician discovered several images of child pornography on the hard drive of a computer that the defendant, Matthew Henry Weber, had brought in for repairs. The manager of the store informed the Los Angeles Police Department of the images, which contacted the FBI. When Weber arrived to pick up his computer, he was interviewed by an FBI agent about the images. Weber claimed to be unaware of the child pornography images on his computer. The FBI seized Weber's computer and conducted a full forensic examination of the hard drive, uncovering hundreds of images depicting children engaged in sexually explicit activity.
On January 17, 2003, a grand jury in the Central District of California returned a one-count indictment charging Weber with possession of child pornography in violation of 18 U.S.C. § 2252A(a)(5)(B).2 Weber subsequently pleaded guilty to the single count in the indictment, pursuant to a plea agreement with the U.S. Attorney's Office. On March 4, 2005, the district court sentenced the defendant to twenty-seven months imprisonment and three years of supervised release.
In preparing the presentence report (PSR), the Probation Office proposed that twenty special conditions be imposed as specific terms of Weber's supervised release. Among them was Condition Nine, the requirement that Weber
participate in a psychological/psychiatric counseling and/or a sex offender treatment program, which may include inpatient treatment, as approved and directed by the Probation Officer. The defendant shall abide by all rules, requirements, and conditions, of such program, including submission to risk assessment evaluation(s), and physiological testing, such as polygraph, plethysmograph, and Abel testing,[ 3] and shall take all prescribed medication.
As justification for the proposed conditions of supervised release, the PSR stated:
During the period of supervised release, it is imperative that the defendant, who has mental health issue [sic], continue to receive mental health treatment and counseling. Further, it is recommended that the defendant continue sex offender treatment, and to be subject to intensive supervision to monitor the defendant's progress. Meanwhile, these special conditions are necessary to protect the public as the defendant undergoes treatment․ Conditions Nos. 3 to 5, and 8 to 19 have been recommended as a result of the instant offense involving the possession of child pornography, which was collected and stored using his computer, and the history and characteristics of the defendant.
In his written objections to the PSR and orally at the sentencing hearing, Weber objected to only one aspect of his supervised release-the requirement that he submit to plethysmograph testing.4 The district court declined to strike that condition, stating:
Now, in terms of [Condition] number nine, the particular testing, what I-if you felt for whatever reason and could support those reasons that whatever test was requested was medically not necessary, you could certainly ask-express that to the probation officer and ask for a hearing, but I intend to keep the condition; but you certainly, as in any condition, probation-or for supervised release, you would have the ability to request a modification.
The district court overruled Weber's objection and incorporated all of the proposed conditions into the judgment and commitment order. Weber timely appealed.
Before turning to the merits of Weber's appeal, we consider whether Weber's claim is ripe for review. Although neither party raises the issue of ripeness, because “[t]he constitutional component of ripeness is a jurisdictional prerequisite,” United States v. Antelope, 395 F.3d 1128, 1132 (9th Cir.2005), we are obligated to address the matter on our own motion “to ensure that proper subject matter jurisdiction exists to hear the case,” Poland v. Stewart, 117 F.3d 1094, 1104 (9th Cir.1997).
Condition Nine requires Weber to participate in a sexual offender treatment program and submit to various tests, including plethysmograph testing, as a part of that program. There is nothing in the record indicating that Weber has yet been ordered to undergo plethysmograph testing and it is not certain that he will ever be ordered to do so.5 That determination will presumably be made by Weber's probation officer in consultation with the appropriate treatment personnel. Weber's refusal to submit to plethysmograph testing once ordered would place him in violation of the terms of his supervised release.
A defendant need not refuse to abide by a condition of supervised release to challenge its legality on direct appeal from the imposition of sentence. In United States v. Williams, 356 F.3d 1045, 1049-51 (9th Cir.2004), the defendant objected to the condition of his supervised release that required him to take psychotropic and other medications prescribed for treatment of his mental illness. Although there was no evidence that the defendant had refused to take any such medications, we rejected the government's argument that adjudication of the propriety of the condition was premature. Id. at 1051. Rather, we held the jurisdictional prerequisite of ripeness does not require “violation of a specified supervised release condition to permit appellate review.” Id.
Relying on Williams, we recently rejected a similar argument by the government that a challenge to a supervised release condition that depended on several contingencies was unripe for appellate review. See United States v. Rodriguez-Rodriguez, 441 F.3d 767, 771-72 (9th Cir.2006). In Rodriguez-Rodriguez, the defendant was convicted of illegal reentry following deportation and was sentenced to a prison term of seventy-seven months, to be followed by a three-year term of supervised release. Id. at 769. Among the conditions of his supervised release was a requirement that he report to the probation officer within seventy-two hours of his release from custody or reentry into the United States. Id. The government claimed Rodriguez-Rodriguez's challenge to the condition was not ripe because it depended on a number of contingencies that had yet to occur, in particular, the completion of his prison term, deportation, and illegal reentry into the United States. Id. at 771. Unpersuaded, we held that the defendant could raise a facial challenge to the reporting condition as it was a part of the sentence imposed-a final judgment subject to immediate appeal pursuant to § 3742(a). Id. at 771-72.
The same is true here. The silence in the record as to whether Weber has previously had to undergo plethysmograph testing or will have to do so at some point in the future does not make this case unripe for review. A term of supervised release, even if contingent, is part and parcel of the defendant's sentence and can be challenged on direct appeal. Accordingly, there is no jurisdictional barrier to our consideration of the merits.
We begin our merits analysis with a discussion of several governing principles.
A. Statutory Framework
Although the consideration of plethysmograph testing as a term of supervised release is a question of first impression in this circuit, we are guided in our analysis by the statutory requirements governing the imposition of conditions of supervised release and by our prior case law interpreting those requirements. We have repeatedly held that a district court enjoys significant discretion in crafting terms of supervised release for criminal defendants, including the authority to impose restrictions that infringe on fundamental rights. See United States v. T.M., 330 F.3d 1235, 1239-40 (9th Cir.2003); United States v. Bee, 162 F.3d 1232, 1234 (9th Cir.1998). In fashioning conditions of supervised release, a district court “has at its disposal all of the evidence, its own impressions of a defendant, and wide latitude.” Williams, 356 F.3d at 1052. In light of this “wide latitude,” we give considerable deference to a district court's determination of the appropriate supervised release conditions, reviewing those conditions deferentially, for abuse of discretion. Id.
A district court's discretion in this regard is not, however, boundless. The principal statute governing a district court's ability to impose conditions of supervised release is § 3583. Section 3583(c) states:
The court, in determining whether to include a term of supervised release, and, if a term of supervised release is to be included, in determining the length of the term and the conditions of supervised release, shall consider the factors set forth in section 3553(a)(1), (a)(2)(B), (a)(2)(C), (a)(2)(D), (a)(4), (a)(5), (a)(6), and (a)(7).
The cross-referenced § 3553(a) factors that are of particular relevance here direct a court to consider:
(1) the nature and circumstances of the offense and the history and characteristics of the defendant;
(2) the need for the sentence imposed-
(B) to afford adequate deterrence to criminal conduct;
(C) to protect the public from further crimes of the defendant; and
(D) to provide the defendant with needed educational or vocational training, medical care, or other correctional treatment in the most effective manner.
In addition to setting forth certain mandatory conditions of supervised release, § 3583(d) permits a district court to impose any condition it deems appropriate,6 so long as the discretionary condition
(1) is reasonably related to the factors set forth in section 3553(a)(1), (a)(2)(B), (a)(2)(C), and (a)(2)(D);
(2) involves no greater deprivation of liberty than is reasonably necessary for the purposes set forth in section 3553(a)(2)(B), (a)(2)(C), and (a)(2)(D); and
(3) is consistent with any pertinent policy statements issued by the Sentencing Commission pursuant to 28 U.S.C. 994(a).[ 7]
Under this statutory scheme, then, conditions of supervised release 8 “are permissible only if they are reasonably related to the goal of deterrence, protection of the public, or rehabilitation of the offender.” T.M., 330 F.3d at 1240. “Conditions of supervised release must relate to these purposes, but may be unrelated to one or more of [them], so long as they are sufficiently related to the others.” Bee, 162 F.3d at 1235 (alteration in original) (internal quotation marks omitted). In addition, a supervised release condition need not relate to the offense of conviction, as long as it satisfies one of the above goals. See T.M., 330 F.3d at 1240. Finally, even if a proposed condition otherwise meets the statutory requirements of § 3553(a), it still must “involve ‘no greater deprivation of liberty than is reasonably necessary for the purposes' of supervised release.” Id. (quoting § 3583(d)(2)).
B. Burden of Justification
Although our case law has repeatedly explained the statutory framework governing the imposition of supervised release terms, we have not had occasion clearly to delineate which party bears the burden of demonstrating that a discretionary supervised release condition is appropriate in a given case. We think the answer to this question is fairly evident in light of the above statutory requirements and our case law discussing the burden of proof at sentencing generally.
As we have explained, supervised release conditions “are permissible only if they are reasonably related” to the goals of deterrence, public protection, and rehabilitation. T.M., 330 F.3d at 1240 (emphasis added). In addition, the condition must “involve[ ] no greater deprivation of liberty than is reasonably necessary” to meet those purposes. § 3583(d)(2) (emphasis added). In light of the statute's recognition that supervised release conditions put a defendant's liberty at stake, the burden should fall on the government to demonstrate that the statutory standards have been met.
The determination that the burden to justify a condition of supervised release should rest with the government is all the more apparent when viewed in light of our case law allocating the burden for other aspects of sentencing. We have held that the government bears “the burden of proving the facts necessary to establish the base offense level” under the United States Sentencing Guidelines. United States v. Howard, 894 F.2d 1085, 1090 (9th Cir.1990). That conclusion was premised on the fact that it is “the government [that] is initially invoking the court's power to incarcerate a person.” Id. Similarly, it is the government that bears the burden of establishing that the offense level should be raised through enhancements (although the defendant bears the burden when he seeks to have the offense level lowered through downward adjustments). Id. In support of our holding in Howard, we relied on the Third Circuit's statement that “[o]ne who affirmatively seeks special favor at sentencing has the burden of proving why it should be bestowed.” United States v. McDowell, 888 F.2d 285, 291 (3d Cir.1989) (alteration in original) (quoting United States v. Garcia, 544 F.2d 681, 685-86 (3d Cir.1976)).
We hold that the same rule applies with regard to a government-supported imposition of a discretionary condition of supervised release. We have long held that a term of supervised release is part of a defendant's sentence, see United States v. Soto-Olivas, 44 F.3d 788, 790 (9th Cir.1995), and, like imprisonment, restricts a defendant's liberty and fundamental rights, see Williams, 356 F.3d at 1052-53; United States v. Bolinger, 940 F.2d 478, 480 (9th Cir.1991). As a result, when the government seeks to restrict a defendant's liberty through a term of supervised release, it shoulders the burden of proving that a particular condition of supervised release involves no greater deprivation of liberty than is reasonably necessary to serve the goals of supervised release.9
C. Procedural Requirements
In applying the substantive statutory standards governing the imposition of supervised release conditions, we have had occasion to consider the procedural steps a district court must take before imposing certain conditions on a criminal defendant as a term of his supervised release. In United States v. Rearden, 349 F.3d 608 (9th Cir.2003), we held that a district court is not generally required “to articulate on the record at sentencing the reasons for imposing each condition.” Id. at 619. Our holding was premised on the circumstances that in that case, (1) the defendant was on notice of the conditions that would be imposed, and (2) the PSR had adequately “spelled out the relationship between them and the factors set forth in § 3583(d) in detail.” Id.; see also United States v. Dupas, 419 F.3d 916, 922 (9th Cir.2005) (noting Rearden's general rule that “an explicit statement of reasons is not absolutely required by 18 U.S.C. § 3583(d)”), cert. denied, 547 U.S. 1011, 126 S.Ct. 1484, 164 L.Ed.2d 261 (2006).10
We subsequently fashioned an exception to Rearden's general rule that a district court need not articulate specific reasons for imposing a condition of supervised release. In Williams, we considered a term of supervised release that required a defendant to “take such psychotropic and other medications prescribed for him.” 11 356 F.3d at 1047 (footnote omitted). We held in Williams that, in light of the particularly significant liberty interest at stake, a district court is required, before ordering such a condition, to “make on-the-record, medically-grounded findings that court-ordered medication is necessary to accomplish one or more of the factors listed in § 3583(d)(1).” Id. at 1057. We also concluded that a district court must make an explicit finding that the condition “involves no greater deprivation of liberty than is reasonably necessary.” Id. (quoting § 3583(d)(2)). Although we recognized in Williams that our previous decision in Rearden does not ordinarily require a district court to make such specific findings before imposing a term of supervised release, we held that when a supervised release condition involves an especially significant liberty interest, such as a requirement to take psychotropic medications, Rearden's general rule does not apply. Id. at 1055.
Our conclusion in Williams was guided by the Supreme Court's decisions in Washington v. Harper, 494 U.S. 210, 110 S.Ct. 1028, 108 L.Ed.2d 178 (1990), Riggins v. Nevada, 504 U.S. 127, 112 S.Ct. 1810, 118 L.Ed.2d 479 (1992), and Sell v. United States, 539 U.S. 166, 123 S.Ct. 2174, 156 L.Ed.2d 197 (2003), each of which addressed the law governing forced medication. See Williams, 356 F.3d at 1053-56. In light of those decisions, we concluded that “an order compelling a person to take antipsychotic medication is an especially grave infringement of liberty.” Id. at 1055. The due process liberty interest at stake in Williams, we noted, is grounded in the dual notions that such drugs interfere with an individual's personal autonomy and that such drugs have the potential for serious negative side effects. Id. at 1054. We concluded that, because of the severity of the infringement, “a thorough inquiry is required before a court may issue” an order making forced psychotropic medication a condition of supervised release. Id. at 1055.
We were careful to note in Williams that our holding did not preclude a district court from ordering, as a term of supervised release, that a defendant take certain prescribed medications. Id. at 1055-56. Rather, our holding was that before mandating such a condition, a district court must make a specific “finding of overriding justification and a determination of medical appropriateness.” Id. at 1056 (internal quotation marks omitted) (quoting Riggins, 504 U.S. at 135, 112 S.Ct. 1810).
In light of Rearden and Williams, the procedural requirements governing a district court's imposition of a condition of supervised release may be summarized as follows: We will generally not require a district court to articulate the reasons behind imposing a certain condition. If, however, the condition implicates a particularly significant liberty interest of the defendant, then the district court must support its decision on the record with record evidence that the condition of supervised release sought to be imposed is “necessary to accomplish one or more of the factors listed in § 3583(d)(1)” and “involves no greater deprivation of liberty than is reasonably necessary.” Williams, 356 F.3d at 1057 (internal quotation marks omitted) (quoting § 3583(d)(2)).
In light of these governing principles, we turn our attention to the specifics of penile plethysmograph testing. Weber argues that the requirement that he submit to plethysmograph testing should be vacated because such testing (1) is not reasonably related to the purposes of deterrence, rehabilitation, or protection of the public, and (2) even if it does satisfy one of the above purposes, the testing requirement results in a greater deprivation of liberty than is reasonably necessary.12 To properly assess these claims, we consider both the nature of the testing at issue and the reception it has received among courts, psychologists, and academics.13
A. The Nature of Plethysmograph Testing
As noted at the outset, penile plethysmograph is a test designed to measure a man's sexual response to various visual and auditory stimuli. More precisely, the male “places on his penis a device that measures its circumference and thus the level of the subject's arousal as he is shown sexually explicit slides or listens to sexually explicit audio ‘scenes.’ ” Berthiaume v. Caron, 142 F.3d 12, 13 (1st Cir.1998); see also Doe ex rel. Rudy-Glanzer v. Glanzer, 232 F.3d 1258, 1262 (9th Cir.2000) (“A penile plethysmograph is a test that measures, through electric wires attached to a man's penis, the reactions that a man has when presented with certain visual stimuli․”). The following account spells out how plethysmograph testing works in practice:
Prior to beginning the test, the subject is typically given instructions about what the procedure entails. He is then asked to place the device on his penis and is instructed to become fully aroused, either via self-stimulation or by the presentation of so-called “warm-up stimuli,” in order to derive a baseline against which to compare later erectile measurements. After the individual returns to a state of detumescence, he is presented with various erotic and non-erotic stimuli. He is instructed to let himself become aroused in response to any of the materials that he finds sexually exciting. These stimuli come in one of three modalities-slides, film/video clips, and auditory vignettes-though in some cases different types of stimuli are presented simultaneously. The materials depict individuals of different ages and genders-in some cases even possessing different anatomical features-and portray sexual scenarios involving varying degrees of coercion. The stimuli may be presented for periods of varying length-from mere seconds to four minutes or longer.
Changes in penile dimension are recorded after the presentation of each stimulus․
Odeshoo, supra, at 8-9 (footnotes omitted).
Initially developed by Czech psychiatrist Kurt Freund as a means to study sexual deviance, plethysmograph testing was also at one time used by the Czechoslovakian government to identify and “cure” homosexuals. David M. Friedman, A Mind of Its Own: A Cultural History of the Penis 232 (2001). Today, plethysmograph testing has become rather routine in adult sexual offender treatment programs, with one survey noting that approximately one-quarter of adult sex offender programs employ the procedure. Odeshoo, supra, at 8. Another survey has placed the relative incidence of the test among adult sexual offender programs at fifteen percent, a somewhat lower, yet still considerable, level. See D. Richard Laws, Penile Plethysmography: Will We Ever Get It Right?, in Sexual Deviance: Issues and Controversies 82, 97 (Tony Ward et al. eds., 2003).
B. The Significance of the Liberty Interest
Courts have previously recognized that plethysmograph testing “can [be] help [ful] in the treatment and monitoring of sex offenders.” Glanzer, 232 F.3d at 1266. At the same time, the First Circuit has noted, putting it mildly, that plethysmograph testing is likely to “strike most people as especially unpleasant and offensive.” Berthiaume, 142 F.3d at 16. Although we agree that “there are plenty of ordinary medical procedures that are disagreeable or upsetting to the patient,” id., this test is not a run-of-the-mill medical procedure. Plethysmograph testing not only encompasses a physical intrusion but a mental one, involving not only a measure of the subject's genitalia but a probing of his innermost thoughts as well. See Odeshoo, supra, at 23.
Moreover, plethysmograph testing is exceptionally intrusive in nature and duration. As one commentator has noted:
It is true that cavity searches and strip searches are deeply invasive, but [plethysmograph testing] is substantially more invasive. Cavity searches do not involve the minute monitoring of changes in the size and shape of a person's genitalia. Nor do such searches last anywhere near the two or three hours required for penile plethysmography exams. Nor do cavity or strip searches require a person to become sexually aroused, or to engage in sexual self-stimulation.
Id. (footnote omitted). We note also that “[t]he degree of privacy afforded to subjects during the procedure varies considerably.” Id. at 8. Sometimes the test is conducted by placing the patient in a private room away from the clinician, other times the two are separated by a curtain or one-way mirror. Id.
As these descriptions of plethysmograph testing indicate, the procedure implicates a particularly significant liberty interest. In reaching this conclusion, we follow the reasoning of the First Circuit in Harrington v. Almy, 977 F.2d 37, 44 (1st Cir.1992). Harrington determined that a government employee had raised sufficient questions as to his due process interest in refusing his employer's demand that he submit to plethysmograph testing to warrant a jury trial on the question whether the requirement violated substantive due process. Id.
Harrington considered the strength of the plaintiff's liberty interest claim in refusing to submit to plethysmograph testing in light of Rochin v. California, 342 U.S. 165, 72 S.Ct. 205, 96 L.Ed. 183 (1952), and Winston v. Lee, 470 U.S. 753, 105 S.Ct. 1611, 84 L.Ed.2d 662 (1985), cases in which the Supreme Court considered the constitutional interest inherent in avoiding “unwanted bodily intrusions or manipulations.” Harrington, 977 F.2d at 43-44. As the First Circuit observed in Harrington, the governing case law indicates that “nonroutine manipulative intrusions on bodily integrity will be subject to heightened scrutiny to determine, inter alia, whether there are less intrusive alternatives available.” Id. at 44. Applying that standard, the First Circuit concluded:
A reasonable finder of fact could conclude that requiring the plethysmograph involves a substantive due process violation. The procedure, from all that appears, is hardly routine. One does not have to cultivate particularly delicate sensibilities to believe degrading the process of having a strain gauge strapped to an individual's genitals while sexually explicit pictures are displayed in an effort to determine his sexual arousal patterns. The procedure involves bodily manipulation of the most intimate sort. There has been no showing regarding the procedure's reliability and, in light of other psychological evaluative tools available, there has been no demonstration that other less intrusive means of obtaining the relevant information are not sufficient.
Although, given the supervised release context, we are not considering the same substantive due process question at issue in Harrington, Harrington rests on the premise that the strong liberty interest in one's own bodily integrity is impaired by the plethysmograph procedure.14 We find the First Circuit's analysis persuasive in this regard.
Similarly, Coleman v. Dretke, 395 F.3d 216, 223 & n. 28 (5th Cir.2004), cert. denied 546 U.S. 938, 126 S.Ct. 427, 163 L.Ed.2d 325 (2005), supports the conclusion that plethysmograph testing implicates a particularly significant liberty interest. In that case, the Fifth Circuit considered a sex offender treatment program which included plethysmograph testing and was imposed by Texas on criminal defendants released on mandatory supervision or parole. Id. Referring specifically to plethysmograph testing and citing Harrington, Coleman held that “due to its highly invasive nature, Texas's sex offender therapy program is ‘qualitatively different’ from other conditions which may attend an inmate's release” and that the Due Process Clause “provides [an individual] with a liberty interest in freedom from the stigma and compelled treatment on which his parole was conditioned” sufficient to require especially stringent procedural protections. Id. at 223.
C. Reactions to Plethysmograph Testing
Our concerns with plethysmograph testing do not rest solely on the invasive nature of the test itself. In addition, the accuracy and reliability of penile plethysmograph testing have been severely questioned. The American Psychiatric Association has expressed reservations about the procedure, observing: “The reliability and validity of this procedure in clinical assessment have not been well established, and clinical experience suggests that subjects can stimulate response by manipulating mental images.” am. psychiatric ass'n, diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR 567 (4th ed.2000); see also W.L. Marshall & Yolanda M. Fernandez, Phallometric Testing with Sexual Offenders: Limits to Its Value, 20 Clinical Psychol. Rev.. 807, 810-13 (2000) (questioning the reliability of phallometric testing, such as plethysmograph testing); Odeshoo, supra, at 10-13 (detailing a number of problems with plethysmograph testing, among them inconsistent results as to the test's reliability and validity).
A predominant concern with plethysmograph testing is its susceptibility to manipulation via faking. Several studies have acknowledged that subjects can control their sexual arousal during the test, thereby posing a threat to the validity of plethysmograph testing. See, e.g., James G. Barker & Robert J. Howell, The Plethysmograph: A Review of Recent Literature, 20 Bull. Am. Acad. Psychiatry & L. 13, 21-23 (1992). According to one source, “[s]everal studies have shown that normal subjects can significantly inhibit their arousal by using mental activities to distract themselves, despite a clear indication that they were attending to the stimuli.” Marshall & Fernandez, supra, at 810. Because “it appears virtually impossible to prevent or detect dissimulation ․ faking will always constitute some undetermined degree of threat to the validity of phallometric assessments.” Id. at 811; see also Walter T. Simon & Peter G.W. Schouten, The Plethysmograph Reconsidered: Comments on Barker and Howell, 21 Bull. Am. Acad. Psychiatry & L. 505, 510 (1993) (“The vulnerability of the plethysmograph to voluntary control has been widely documented and is a major concern in the use of the test with offenders.”).
Plethysmograph testing has also been sharply criticized as lacking “uniform administration and scoring guidelines.” See Simon & Schouten, supra, at 510; see also Odeshoo, supra, at 12-13 (noting a lack of standardization in administration of plethysmograph testing). One researcher noted well over a dozen potential sources of variation among different assessments, including the type of measuring device and stimuli that are used, the characteristics of the test, and the setting in which it is conducted. See Laws, supra, at 87-88. The lack of standard procedures governing plethysmograph testing has led one pair of commentators to conclude that “research data as well as individual findings derived by plethysmograph must be considered idiosyncratic[and] unamenable to normative comparisons, if not impossible to interpret from a traditional psychometric perspective.” Simon & Schouten, supra, at 511. The lack of uniform standards is compounded by reports that indicate a lack of formal training for clinicians administering the test. See Laws, supra, at 87 (characterizing as “truly appalling” one survey's findings that seventy-six percent of plethysmograph technicians received one week or less of training and eighteen percent received no training whatsoever).
The supporters of plethysmograph testing acknowledge its limitations. See Barker & Howell, supra, at 13, 22-23 (noting that while some research supports the notion that plethysmograph testing “is a reliable and valid method of objectively measuring and assessing the erectile response in male sexual offenders,” the propensity for faking and lack of standards poses a challenge to accurate use of such testing). In addition, at least one former advocate of the procedure has since changed his tune. See Laws, supra, at 82-84, 99 (explaining the author's account of why his former faith in plethysmograph testing has subsequently been “seriously eroded”).15
Despite these criticisms, plethysmograph testing has been recognized by some psychologists and researchers as a useful technique in the treatment of sexual offenders. “The ideal application for the plethysmograph is the assessment and treatment of known sex offenders.” Barker & Howell, supra, at 18. Its role in a treatment program is to aid in identifying whether an individual exhibits a sexual response to deviant stimuli and determining whether a prescribed course of behavior modification therapy is effective in promoting “non-deviant arousal.” Id. In particular:
The plethysmograph can help in identifying offenders who manifest high levels of arousal to stimuli depicting inappropriate sexual activity, or those showing very low levels of arousal to stimuli that would be considered portraying appropriate sexual activity. The plethysmograph can help determine and enhance specialized behavior therapy for these offenders and evaluate therapeutic efficacy without the normal distortion evident in the subject's self-report.
Another researcher has differentiated between the appropriate and inappropriate uses of plethysmograph testing: Among the accepted uses of the procedure, according to that account, are the “[u]se of erection responses to indicate the need to target deviant sexual arousal for treatment and to monitor the effectiveness of that treatment” and the “[u]se of erection responses to confront subjects who deny having deviant arousal.” Laws, supra, at 98. That same analysis, however, cautioned that plethysmograph testing should not be used to “determine or make statements about whether someone has committed a specific sexual offense or whether someone ‘fits the profile’ of a sexual offender.” Id. Also inappropriate would be using plethysmograph testing “as a sole criterion to decide someone's release from custody or from a treatment program” or “to screen general populations in search of potential sex offenders.” Id.
D. Plethysmograph Testing as a Condition of Supervised Release
In light of these observations by courts and commentators alike, we cannot say categorically that, despite the questions of reliability, plethysmograph testing can never reasonably promote at least one, if not all three, of the relevant goals laid out in § 3553(a)(2)-namely, deterrence, public protection, and rehabilitation. As the Fourth Circuit, the only circuit to address the permissibility of plethysmograph testing as a condition of supervised release, has held, plethysmograph testing is regarded as “useful for treatment of sex offenders” in appropriate circumstances and thus can be “reasonably related” to “treatment, fostering deterrence, and protecting the public.” United States v. Dotson, 324 F.3d 256, 261 (4th Cir.2003) (internal quotation marks omitted) (quoting United States v. Powers, 59 F.3d 1460, 1471 (4th Cir.1995)); see also Walrath v. United States, 830 F.Supp. 444, 446-47 (N.D.Ill.1993) (upholding plethysmograph testing as a condition of parole against Fourth and Fifth Amendment challenges, concluding that “[t]he fact that two recommended institutions require the plethysmograph as an evaluative tool suggests that it serves a useful function in the treatment of sexual deviance”); State v. Riles, 135 Wash.2d 326, 957 P.2d 655, 668 (1998) (upholding plethysmograph testing as part of a treatment program for a sexual offender in light of the observation that such testing is “an effective method for diagnosing and treating sex offenders”).
To so conclude, however, is not the end of the story. First, although we recognize that plethysmograph testing can reasonably promote the goals of supervised release, the question of whether it will promote those goals in a particular case must be an individualized determination. Section 3583(d)(1) requires that conditions of supervised release be “reasonably related” to “the nature and circumstances of the offense and the history and characteristics of the defendant.” See §§ 3583(d)(1), 3553(a)(1). This tailoring requirement is all the more important in cases such as this, where a particularly strong liberty interest is at stake. To satisfy the standard that a supervised release condition be “reasonably related” to the statutory goals in the particular circumstances, a district court must consider whether, given the level of intrusion required by the test, its noted flaws, and its downsides,16 plethysmograph testing is sufficiently likely, given a defendant's specific characteristics, to yield sufficiently useful results. Only a finding that plethysmograph testing is likely given the defendant's characteristics and criminal background to reap its intended benefits can justify the intrusion into a defendant's significant liberty interest in his own bodily integrity.
Second, conditions of supervised release must also “involve ‘no greater deprivation of liberty than is reasonably necessary for the purposes' of supervised release.” T.M., 330 F.3d at 1240 (quoting § 3583(d)(2)). There are alternatives available in the treatment of sexual offenders that are considerably less intrusive than plethysmograph testing and may be sufficiently accurate. See Laws, supra, at 99; Marshall & Fernandez, supra, at 817; Odeshoo, supra, at 13-16.
For example, sexual offenders are often treated through self-reporting interviews, during which the subject is asked about his sexual preferences. Odeshoo, supra, at 14. Other sexual offender programs rely on a card-sorting test, which involves asking the individual to sort cards depicting sexual images into deviant and non-deviant categories. Id. Although these techniques have been criticized for their susceptibility to faking on the part of the subject, see id., plethysmograph testing, as we have observed, is not immune from this criticism. The effectiveness of these procedures in the treatment of sexual offenders is disputed among the experts, with one commentator noting that “some researchers believe that basic self-reporting ․ is as effective as [plethysmograph testing] or other techniques,” id., and another study concluding that “the psychometric data on these alternative approaches is far less satisfactory than for phallometrics,” Marshall & Fernandez, supra, at 817.
Another non-physiological test which also appears to enjoy routine use in sexual offender programs is Abel testing. Abel testing, which was also required in this case but is not challenged by Weber, involves exhibiting photographs to an individual and measuring the length of time he looks at each picture. See Odeshoo, supra, at 13. This procedure is much less intrusive into the body and somewhat less intrusive into the mind of a defendant than plethysmograph testing. Much like plethysmography, the effectiveness and reliability of Abel testing is the subject of some debate. See id. at 14; Marshall & Fernandez, supra, at 817. One researcher, however, has deemed Abel testing to be a “promising development.” Laws, supra, at 99. Given that Abel testing is not properly before us, we do not set forth any opinion as to its propriety in this, or any other case. We discuss the procedure only to point out the existence of a less intrusive alternative to plethysmograph testing that enjoys similar, if not more, support among researchers. The appropriateness of Abel testing in a particular case should, of course, be left to the district court judge and probation officer, with appropriate expert consultation.
Ordinary polygraph testing is another possible viable alternative to plethysmograph testing that can be considered by district courts as they fashion supervised release conditions. Already more common in sexual offender treatment programs than plethysmograph testing, polygraph testing is much less costly to administer and “appears to be at least as valid and reliable as the plethysmograph (if not more so).” Odeshoo, supra, at 14-15. Most importantly, a polygraph examination “may well be preferable by virtue of its less intrusive and controversial character.” id. at 16.17
The existence of non-physiological, less-intrusive alternatives to plethysmograph testing, including interviews, card-sorting, and Abel and polygraph testing, is, self-evidently, highly relevant to the question of whether plethysmograph testing “involves no greater deprivation of liberty than is reasonably necessary” to serve the purposes of supervised release. § 3583(d)(2); see also T.M., 330 F.3d at 1240. As we have indicated, imposing such testing as a condition of supervised release implicates a liberty interest sufficiently weighty to trigger the enhanced procedural requirements established in Williams. When viable and effective alternatives exist to plethysmograph testing, a procedure that involves intrusion on an especially significant liberty interest, a district court should be hesitant to impose that procedure as a supervised release condition and may do so only after explaining on the record why the alternatives are inadequate.
We conclude that, just as the particularly significant liberty interest at stake in Williams meant that “a thorough inquiry is required” before a district court may impose forced medication as a condition of supervised release, including “on-the-record medically-grounded findings,” Williams, 356 F.3d at 1055-57, so the particularly significant liberty interest in being free from plethysmograph testing requires a thorough, on-the-record inquiry into whether the degree of intrusion caused by such testing is reasonably necessary “to accomplish one or more of the factors listed in § 3583(d)(1)” and “involves no greater deprivation of liberty than is reasonably necessary,” given the available alternatives. Id. at 1057.
One critical determination that must guide a district court's inquiry as to whether the government has met its burden to show that plethysmograph testing is a necessary condition of a defendant's supervised release is whether such testing is reasonably necessary in that particular case to promote the goals “of deterrence, protection of the public, or rehabilitation of the offender.” T.M., 330 F.3d at 1240. Making such a determination requires consideration of evidence that plethysmograph testing is reasonably necessary for the particular defendant based upon his specific psychological profile.18 We expect that the probation officer or the district court will ordinarily consult the views of a psychologist or other expert as to the propriety of plethysmograph testing for the particular defendant, although there may be circumstances in which it is not necessary to do so. Cf. Williams, 356 F.3d at 1056 (requiring findings based on a “medically-informed record” before antipsychotic medication could be required as a term of supervised release).
Additionally, when engaging in this inquiry the district court must consider the particular sexual offenses committed by the defendant, as well as related offenses likely to be committed if he is not treated. Weber objects to the imposition of plethysmograph testing on the ground that his crime, possession of child pornography, does not warrant such a procedure, contending that plethysmograph testing is appropriate only for individuals who have committed, or attempted to commit, sexual acts directly against children. The district court is not, however, restricted to the crime of conviction in applying the “reasonably related” standard.19 Still, a generalized assessment based on the class of sex offenders generally, rather than on the particular sex offenses a defendant has committed or related offenses he is likely to commit if not treated, cannot fulfill the mandate that a term of supervised release satisfy the “reasonably related” standard.20
In response to Weber's objection to the plethysmograph testing requirement, the district court noted that if, in the future, Weber thought that such testing “was medically not necessary,” he could “ask for a hearing” or “request a modification.” As we have explained, however, the burden is on the government, not the defendant, to establish at the time of sentencing that plethysmograph testing is both reasonably necessary “to accomplish one or more of the factors listed in § 3583(d)(1)” and “involves no greater deprivation of liberty than is reasonably necessary.” 21 Williams, 356 F.3d at 1057 (internal quotation marks omitted). On remand, if the government continues to seek submission to plethysmograph testing as a condition of supervised release, then it must meet its burden of justifying the requirement, and the district court must make on-the-record findings that it has done so.
We note that our holding does not displace Rearden's general rule that, so long as the PSR adequately explains the relationship between proposed conditions of supervised release and the purposes those conditions are designed to serve, a district court usually need not specifically articulate those reasons on the record.22 As we noted in Williams, however, that general rule is subject to limited exceptions. Today, we recognize that the imposition of plethysmograph testing implicates a sufficiently significant liberty interest to require heightened procedural protections similar to those established in Williams. Again, as in Williams with regard to forced medication, we are not holding that a district court may never impose plethysmograph testing as a condition of supervised release, only that “a thorough inquiry is required” before a court may do so. 356 F.3d at 1055.
The requirement that Weber submit to plethysmograph testing as part of his sex offender treatment program was imposed without the necessary evidentiary record, justification, and findings we now hold are required. Accordingly, we vacate the condition and remand for further proceedings consistent with this opinion.
VACATED and REMANDED.
Judge Berzon's excellent opinion is deserving of support. I would, however, go beyond it to hold the Orwellian procedure at issue to be always a violation of the personal dignity of which prisoners are not deprived. The procedure violates a prisoner's bodily integrity by affecting his genitals. The procedure violates a prisoner's mental integrity by intruding images into his brain. The procedure violates a prisoner's moral integrity by requiring him to masturbate.
By committing a crime and being convicted of it, a person does not cease to be a person. A prisoner is not a mere tool of the state to be manipulated by it to achieve the purposes the law has determined appropriate in punishment. The prisoner retains his humanity and therefore has purposes transcending those of the state. A prisoner, for example, cannot be forced into prostitution to aid the state in securing evidence. A prisoner, for example, cannot be made to perjure himself in order to assist a prosecution. Similarly, a prisoner should not be compelled to stimulate himself sexually in order for the government to get a sense of his current proclivities. There is a line at which the government must stop. Penile plethysmography testing crosses it.
1. In addition to penile plethysmograph testing, there is a corresponding procedure for women, known as “vaginal plethysmography.” See Odeshoo, supra, at 2 n. 9. All references in this opinion to the general term “plethysmograph testing” cover only penile plethysmograph testing.
2. All statutory references in this opinion are to Title 18 of the United States Code, unless otherwise indicated.
3. Abel testing, another procedure used in sexual offender treatment programs, “involves presenting individuals with non-erotic pictures of children and adults and determining sexual interest by measuring how long a person spends viewing each picture.” Odeshoo, supra, at 13.
4. We address Weber's additional challenges to his supervised release conditions, raised for the first time on appeal, in a memorandum disposition filed concurrently with this opinion.
5. At present, Weber has completed his prison sentence and is serving his term of supervised release.
6. In addition to permitting a district court to design any supervised release condition it deems appropriate, § 3583(d) specifically enumerates as permissible conditions a number of discretionary conditions of probation catalogued at § 3563(b). Relevant here is § 3563(b)(9), which provides that a district court may require a defendant to “undergo available medical, psychiatric, or psychological treatment, including treatment for drug or alcohol dependency, as specified by the court, and remain in a specified institution if required for that purpose.”
7. The United States Sentencing Guidelines Manual contains a policy statement that recommends the imposition of sex offender treatment programs for defendants convicted of a sex offense. See U.S.S.G. § 5D1.3(d)(7)(A). The Guidelines do not, however, specify the particulars of such treatment programs and do not mention the plethysmograph testing at issue here.
8. As mentioned earlier, § 3583(d) provides for a number of mandatory conditions of supervised release that are to be imposed by the district court in certain circumstances. The statutory framework we have catalogued, however, is applicable to discretionary conditions that a district court may choose to impose on a defendant. This case deals with such a condition. Any reference in this opinion to the standards governing “supervised release conditions” should be understood as referring to only discretionary supervised release conditions.
9. Our conclusion as to the proper allocation of the burden of proof at the imposition of a condition of supervised release is further supported by case law considering which party bears the burden of proof in other proceedings concerning supervised release. If a defendant violates a condition of his supervised release, a district court may revoke his supervised release and impose a term of imprisonment. See § 3583(e)(3). We have held that, in such a proceeding, it is the government that bears the burden to demonstrate that a defendant has violated a condition of his supervised release. See United States v. Turner, 312 F.3d 1137, 1142 (9th Cir.2002); Soto-Olivas, 44 F.3d at 792. Likewise, a district court can terminate supervised release early and discharge the defendant, provided he has served at least one year of his supervised release term. See § 3583(e)(1). In that situation, courts have required the defendant, as the party receiving the benefit of early termination, to demonstrate that such a course of action is justified. See United States v. Weintraub, 371 F.Supp.2d 164, 167 (D.Conn.2005); United States v. McKay, 352 F.Supp.2d 359, 361 (E.D.N.Y.2005).
10. We note that some of our sister circuits are in disagreement with our general approach, holding instead that § 3553(c), which requires a district court to “state in open court the reasons for its imposition of the particular sentence,” requires an explanation for each supervised release condition. See United States v. Loy, 191 F.3d 360, 371 (3d Cir.1999) (remanding to allow the district court the opportunity to articulate reasons for imposing specific conditions, so as to ensure effective appellate review); United States v. Edgin, 92 F.3d 1044, 1049 (10th Cir.1996) (same).
11. We read the term “psychotropic” narrowly, so as to “encompass only ‘antipsychotic’ or ‘neuroleptic’ drugs.” Williams, 356 F.3d at 1047 n. 2.
12. We have previously approved of a condition of supervised release that required a defendant to participate in a sexual offender treatment program and “follow all other lifestyle restrictions or treatment requirements imposed by defendant's therapist.” United States v. Fellows, 157 F.3d 1197, 1203-04 (9th Cir.1998). We held that the requirement that the defendant abide by the treatment conditions imposed by his therapist was neither overly broad nor an improper delegation of the district court's authority, reasoning that “[t]he court cannot be expected to design and implement the particularities of a treatment program” and that the defendant's therapist “is in the best position to know what lifestyle restrictions are necessary to enhance his treatment and reduce the likelihood that he will re-offend.” Id. at 1204. We also stated that “[u]nless [the defendant] is required to comply with those restrictions, he will not receive the full benefit of the treatment program.” Id. Fellows had no reason to, and did not, consider the propriety of any specific aspect of sexual offender treatment programs, including plethysmograph testing, and does not inform our inquiry here.
13. In the proceedings before the district court, both Weber and the government introduced a number of secondary sources, including psychological journals, in support of their respective positions on the propriety of plethysmograph testing. The discussion that follows relies on some of those sources, as well as others, to describe the nature of plethysmograph testing and the corresponding praise and criticism it has received. The broad propositions for which we rely on these sources for support were all litigated before the district court, although not all of the psychological reports were cited by the parties. Relying on such sources is not novel. In considering the liberty interest of an individual to be free from forced medication, the Supreme Court relied on various literature from the psychological field. See Harper, 494 U.S. at 234 n. 13, 110 S.Ct. 1028. We have also previously relied on secondary sources to inform our judgment in similar circumstances. See Williams, 356 F.3d at 1054-55 (citing secondary sources for the harm caused by antipsychotic medication); Hernandez v. Ashcroft, 345 F.3d 824, 836-38 (9th Cir.2003) (reviewing secondary sources in the field of domestic violence); Lopez-Galarza v. INS, 99 F.3d 954, 962-63 (9th Cir.1996) (reviewing psychological studies and other sources on the harm caused by rape). As detailed later, we rely on these sources not to come to a conclusion as to the propriety of plethysmograph testing in this case or any other, but to explain why district courts must conduct a thorough inquiry on an evidentiary record before imposing such a requirement as a term of supervised release.
14. As stated above, Weber has objected to the requirement that he submit to plethysmograph testing on statutory grounds-that such testing is not reasonably related to the goals of supervised release and would result in a greater deprivation of liberty than necessary. See § 3583(d). He has not claimed that plethysmograph testing as a condition of supervised release is a violation of his substantive due process rights. Because the issue is not before us, we express no opinion on the question whether requiring plethysmograph testing as a condition of supervised release amounts to a substantive due process violation.
15. Courts have uniformly declared that the results of such tests are “inadmissible as evidence because there are no accepted standards for this test in the scientific community.” Glanzer, 232 F.3d at 1266; see also United States v. Powers, 59 F.3d 1460, 1470-71 (4th Cir.1995) (concluding that a district court did not abuse its discretion in denying the admissibility of expert testimony on plethysmograph testing under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), because, although “useful for treatment of sex offenders” there was “extensive, unanswered evidence weighing against the scientific validity of the penile plethysmograph test”); Laws, supra, at 95-97 (noting the unlikelihood that plethysmograph testing can satisfy several of the Daubert criteria). But see Alex Kozinski, Brave New World, 30 U.C. Davis L. Rev. 997, 1008-09 (1997) (suggesting that in certain circumstances plethysmograph testing might meet the requisites of Daubert ). Furthermore, although the First Circuit noted that “the plethysmograph is widely used in the scientific community for the treatment of pedophilia,” that court also stated that “its use for screening is debatable and the scientific community is not of one mind.” Berthiaume, 142 F.3d at 17; see also Simon & Schouten, supra, at 511 (plethysmography's “scientific status remains that of an experimental technique”).
16. One of these detriments derives from the fact that plethysmograph testing involves presenting subjects with various erotic images which, according to some accounts, can include child pornography. See Odeshoo, supra, at 33. “Treating” a sexual offender, like Weber, who was convicted of possession of child pornography by presenting him with images of the very sort that lead to his conviction is a little like handing a pyromaniac a lighted match but cautioning him not to use it. The consideration that the test involves exposing the sexual offender to the very stimuli that other conditions of supervised release may discourage him from possessing is one more reason why a careful, on-the-record assessment of the individual's circumstances, most usefully informed by expert opinion, is necessary to sort out whether the likely impact of plethysmograph testing, is, on balance, reasonably related to the statutory purposes of supervised release.
17. We recently considered a challenge to a supervised release condition mandating polygraph testing for a sexual offender as part of his treatment program. See Antelope, 395 F.3d at 1133-41. The defendant in that case challenged the polygraph requirement on the grounds that it violated his Fifth Amendment right against compelled self-incrimination. Id. at 1133. While we acknowledged the rehabilitative purpose behind the polygraph questioning, we held that requiring, as a condition of supervised release, that a defendant answer questions about potential criminal activity in a polygraph examination was significantly incriminating and coercive to violate the Fifth Amendment. Id. at 1134-39. That conclusion, however, did not doom the condition. Rather, we held that a defendant retains his right against self-incrimination during the required polygraph testing and can refuse to answer any incriminating questions unless he is granted use and-derivative-use immunity under Kastigar v. United States, 406 U.S. 441, 92 S.Ct. 1653, 32 L.Ed.2d 212 (1972). See Antelope, 395 F.3d at 1141 & n. 5. After Antelope, then, a district court may require, as a term of supervised release, that a defendant submit to polygraph testing, provided such a condition comports with the requirements of § 3583(d), but a defendant retains his Fifth Amendment rights during any such testing.
18. We note that, in this case, following the investigation into Weber's possession of child pornography and the initial seizure of Weber's computer, it was later discovered that he continued to possess child pornography on a second computer. Although this fact alone would be insufficient to warrant plethysmograph testing, Weber's apparent inability to control his desire for child pornography, even after detection by authorities, might be relevant to a determination of the appropriateness of such testing.
19. A supervised release term need not be related to the offense of conviction and instead can “look forward in time to crimes that may be committed in the future.” See United States v. Wise, 391 F.3d 1027, 1031 (9th Cir.2004).
20. We note that Weber's claim that plethysmograph testing may have limited applicability to individuals like him-those who have not committed an act of sexual abuse-has some support from researchers in the field. Although studies claim that plethysmograph testing is appropriate for treatment of “sex offenders,” see Barker & Howell, supra, at 18, one study of plethysmograph testing has specifically limited the definition of sexual offenders to “mature males who either coerce an adult female to have sex with them, or have sex with a child,” Marshall & Fernandez, supra, at 809. Although recognizing that there are individuals who commit other sexually offensive acts such as exhibitionism or voyeurism (of which viewing child pornography is a variety), the authors of that study stated that “phallometry has either had limited or no application to these populations.” Id.
21. Should circumstances arise after sentencing, while a defendant is serving his term of supervised release, that would impact the decision of whether plethysmograph testing is proper, it is possible that on an appropriate showing the district court could modify the defendant's conditions of supervised release. See § 3583(e)(2); United States v. Miller, 205 F.3d 1098, 1100-01 (9th Cir.2000).
22. We note that even if Rearden were the governing precedent, its requisites almost surely were not met. The PSR did put Weber on notice of the proposed plethysmograph testing. It did not, however, spell out “in detail” the relationship between that condition and the statutory requirements for conditions of supervised release. See Rearden, 349 F.3d at 619. In support of its recommended conditions of supervised release, the PSR in this case contained only a general bare bones statement:Meanwhile, these special conditions are necessary to protect the public as the defendant undergoes treatment[. ] Pursuant to 18 USC 3583(d), conditions of supervised release must be reasonably related to the nature and circumstances of the offense and the history and characteristics of the defendant. Conditions Nos. 3 to 5, and 8 to 19 have been recommended as a result of the instant offense involving the possession of child pornography, which was collected and stored using his computer, and the history and characteristics of the defendant.
Opinion by Judge Berzon; Concurrence by Judge Noonan