Criminal No. CCB-03-0150.
January 31, 2006
The United States charged Gordon Elliott Thomas, III with sexual exploitation of a child and the receipt and possession of child pornography and moved for his pretrial detention, on the grounds that the charged offenses are crimes of violence, as defined by Congress, and that no conditions of release could reasonably assure the safety of the community. The defendant opposed the government's motion.
Detention proceedings were held on April 3, April 22, and April 28, 2003. Having considered the parties' submissions and the evidence presented during those proceedings, the undersigned ruled from the bench on April 28, 2003, denying pretrial detention and releasing the defendant on certain conditions. The government appealed the ruling to the presiding district court judge, who upheld the release decision, but imposed two additional conditions. This opinion memorializes and supplements that bench ruling. Given the paucity of reported decisions in this area, and the necessity for prompt and abbreviated decision-making in detention proceedings, the Court has set forth in greater detail the existing law and scientific research on the assessment of future dangerousness of accused sex offenders such as defendant.
The undersigned imposed the following conditions of release pending trial, in addition to certain standard conditions, requiring the defendant to: (1) reside at the Connecticut home of his mother and step-father; (2) avoid all contact with the child in the video; (3) report to and obey the supervising officer as directed; (4) refrain from possessing a firearm, destructive device, or other dangerous weapon; (5) refrain from excessive use of alcohol and from any use of narcotics or controlled substances; (6) undergo appropriate medical, psychological, or psychiatric treatment; (7) submit to an electronic monitoring program, with the additional condition that he may not
The release order of the undersigned specified, among other conditions, that the defendant was to have no direct contact with female children. The undersigned specifically exempted the six-year old grandson of defendant's mother, one of Mr. Thomas's third-party custodians, who frequently visits her home and is also defendant's nephew. On appeal, the Honorable Catherine C. Blake added a condition that also prohibited defendant from having any unsupervised contact with any persons under the age of eighteen. She also required that departures from the residence for the stated purposes be approved in advance by pretrial services.
I. Background A. Charges
The defendant is charged by indictment with sexual exploitation of children in violation of 18 U.S.C. § 2251(a); possession of material shipped and transported in interstate and foreign commerce depicting minors engaged in sexually explicit leave the residence to which he is restricted except for pre-approved attorney visits, court appearances, and medical and mental health treatment, during which he must be accompanied by one of his two third-party custodians; (8) have no direct contact with female children; (9) have no use of, or access to, a computer or any other device with internet access. conduct in violation of 18 U.S.C. § 2252(a)(4)(B); and attempted receipt of images depicting minors engaged in sexually explicit conduct in violation of 18 U.S.C. § 2252(a)(2).
The defendant's initial appearance was on April 1, 2003, at which time the government moved for detention pursuant to 18 U.S.C. § 3142(f)(1)(A). Detention proceedings were held on April 3, April 22, and April 28, 2003. During the course of these proceedings, the government argued that there were no conditions or combination of conditions of release that would ensure the safety of the community, and that the defendant should thus be detained prior to trial.
C. Facts, Alleged Conduct and Evidence
The charges against the defendant were the result of an investigation in which federal agents assumed the computer identity of a person overseas who had been advertising pornographic videos depicting children on the internet. Following the internet advertiser's arrest, agents continued to correspond with that person's customers, including Mr. Thomas. The government contends that the defendant, in response to one of these internet advertisements, paid $200 in exchange for child pornography videos. Following this transaction, the Postal Inspection Service executed a "controlled delivery" of child pornography to the defendant's home. A search warrant was executed at the defendant's home on January 3, 2003 during which a computer, video camera, CD-ROMs, floppy disks, and a number of video tapes were seized. Subsequent analysis of these items revealed that they contained approximately 16,000 images of child pornography. These images are the basis for the two charges that have been brought against the defendant regarding the receipt and possession of child pornography.
In addition to the images just described, one of the video tapes seized from the defendant's home depicted an adult hand exposing the genitalia of a young girl who appeared to be asleep. The government states that this scene was created and produced by Mr. Thomas, while he was babysitting a friend's nine year old daughter. Mr. Thomas signed a written statement acknowledging that he had been advised of his rights, stating that he produced the scene and the video, but claiming that it was a one-time occurrence and that he never touched either the girl depicted in the video or any other child. This image, and the conduct associated with its creation, is the basis for the charge that has been brought against the defendant regarding the sexual exploitation of a child.
Among the seized videotapes, there is also non-pornographic footage of young girls talking in a stairwell and otherwise present on the grounds of an apartment complex. The camera, however, zooms in on their crotch areas, and these portions of the video were allegedly taped surreptitiously through the peephole of an apartment door and from an apartment balcony. The government thus characterizes these portions of the videotapes as both "child erotica" and "surveillance" video, claiming that this is further evidence of the defendant's dangerousness to young girls in the community.
D. Pre-Trial Release: Factors and Proposed Conditions
The defendant, Mr. Thomas, is twenty-nine years old, single and has resided alone in Maryland since attending college here. He has been employed for the past five years as an environmental field technician. He has no criminal history and denies any use of drugs or alcohol.
The defendant's mother and step-father reside in Connecticut and advised the Court that they would be willing to serve as third party custodians on their son's behalf and allow him to reside in their home, subject to electronic monitoring, pending trial. Defense counsel proposed release on a number of conditions. Those proposed conditions were that defendant would reside with his mother and step-father in their home in Connecticut; that they would serve as third party custodians; that defendant would be confined to the home under electronic monitoring; and that defendant's parents would engage the services of appropriate mental health professionals to provide therapy for defendant to address his pedophilic tendencies. The pretrial services officer, however, recommended detention, as she did not believe there was any condition or combination of conditions that could reasonably assure the safety of the community, but offered no further elaboration.
E. Expert Testimony
Each side presented the testimony of a single expert witness to assist the Court in determining the level of risk posed by the defendant if released pending trial.
1. Neil Blumberg, M.D.
The defendant presented the testimony of Neil Blumberg, M.D., who is board-certified in psychiatry and forensic psychiatry with extensive professional experience and academic credentials in both his specialty and sub-specialty. Dr. Blumberg was offered, without objection, and accepted as an expert in forensic psychiatry. (Tr. 23, Apr. 22). Dr. Blumberg conducted a forensic psychiatric evaluation of Mr. Thomas on April 15, 2003 for over five hours. As part of this evaluation, Dr. Blumberg conducted a mental status examination, interviewed the defendant regarding his medical and psychiatric history, made direct observations of him, and administered psychological testing. Dr. Blumberg also interviewed Mr. Thomas's mother, sister, and former girlfriend by telephone. He did not, however, view the video produced by Mr. Thomas that is the evidentiary basis for the exploitation charge.
Dr. Blumberg testified, however, that defense counsel, who did view the video, had summarized its contents for him by letter prior to his evaluation of Mr. Thomas. (Tr. 24, Apr. 22).
Based on his examination of Mr. Thomas, Dr. Blumberg diagnosed the defendant, to a reasonable degree of medical certainty and in accordance with the criteria set out in the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, with three mental disorders: (1) dysthymic disorder, early onset; (2) pedophilia, attracted to females, non-exclusive type; and (3) personality disorder not otherwise specified with avoidant features. (Tr. 30, Apr. 22).
See Addington v. Texas, 441 U.S. 418, 430, 99 S.Ct. 1804, 1811 (1979) (explaining that "[t]he subtleties and nuances of psychiatric diagnosis render certainties virtually beyond reach in most situations . . . [but that] within the medical discipline, the traditional standard for factfinding is a `reasonable medical certainty.'").
The DSM-IV was developed by the American Psychiatric Association to provide mental health professionals with a diagnostic classification system of mental disorders. Diagnostic criteria were developed following a review of published literature, re-analysis of previously collected data, and issue-focused field trials. See DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS, xiii-xxv (4th ed. 1994) (hereinafter DSM-IV). Thus, the DSM-IV provides a standard, comprehensive diagnostic tool for evaluating mental disorders, and reflects a consensus opinion of the medical community at the time of publication. See id. Additionally, courts have found expert diagnoses of mental disorders, in relation to assessing future dangerousness, reliable in part due to a mental health professional's reliance on the criteria specified by the DSM-IV.See United States v. Barnette, 211 F.3d 803, 816 (4th Cir. 2000).
Dysthymia is a chronic mood disorder manifested by depression and its commonly associated symptoms. 536 STEDMAN'S MEDICAL DICTIONARY (26th ed. 1995). See also DSM-IV, 345-49.
Pedophiles are essentially defined as individuals who are sexually attracted to children younger than the age of thirteen, that is, prepubescent children. A pedophile need not act on these desires to be diagnosed as such. See DSM-IV, 527. The DSM-IV lists the diagnostic criteria for pedophilia as:
A. Over a period of at least 6 months, recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving sexual activity with a prepubescent child or children (generally age 13 years or younger).
B. The fantasies, sexual urges, or behaviors cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.
C. The person is at least age 16 years and at least 5 years older than the child or children in Criterion A. Note: Do not include an individual in late adolescence involved in an ongoing sexual relationship with a 12- or 13-year old.Specify if:
Sexually Attracted to Males
Sexually Attracted to Females
Sexually Attracted to Both
Limited to Incest
Exclusive Type (attracted only to children)
Id. at 528.
The DSM-IV explains that "[s]ome individuals with [p]edophilia are sexually attracted only to children (Exclusive Type), whereas others are sometimes [also] attracted to adults (Nonexclusive Type)." DSM-IV, supra note 7, at 527.
The DSM-IV states that this category is for disorders of personality functioning that do not meet criteria for any specific personality disorder. DSM-IV, supra note 6, at 673.
Id. The DSM-IV lists the general diagnostic criteria for a personality disorder as:
An example is the presence of features of more than one specific [p]ersonality [d]isorder that do not meet the full criteria for any one [p]ersonality [d]isorder ("mixed personality"), but that together cause clinically significant distress or impairment in one or more important areas of functioning (e.g., social or occupational).
A. An enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual's culture. This pattern is manifested in two (or more) of the following areas:
(1) cognition (i.e., ways of perceiving and interpreting self, other people, and events)
(2) affectivity (i.e., the range, intensity, lability, and appropriateness of emotional response)
(3) interpersonal functioning
(4) impulse control
B. The enduring pattern is inflexible and pervasive across a broad range of personal and social situations.
C. The enduring pattern leads to clinically significant distress or impairment in social, occupational, or other important areas of functioning.
D. The pattern is stable and of long duration and its onset can be traced back at least to adolescence or early adulthood.
E. The enduring pattern is not better accounted for as a manifestation or consequence of another mental disorder.
F. The enduring pattern is not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., head trauma).Id. at 633.
Dr. Blumberg stated that he made the first and third diagnoses because the defendant suffered from long-standing depression and feelings of inadequacy as well as other symptoms associated with those conditions. Regarding his second diagnosis of pedophilia, Dr. Blumberg explained that the defendant related that he initially became sexually attracted to female children, primarily between the ages of eight and twelve, in the late 1990s. The defendant began viewing adult pornography on the internet during a period in which his ability to interact with other young adults was limited by the long hours required by his job. Mr. Thomas explained that the adult pornography sites that he visited also included "pop-up" pornographic pictures of children, which is how he became interested in such images. He told Dr. Blumberg that while these images sparked his interest in both child pornography and child erotica, he continued to be aroused primarily by adult heterosexual images. Dr. Blumberg further testified that while the defendant acknowledged having joined child porn chat services, from which he had downloaded many images of child pornography, Dr. Blumberg would not necessarily characterize him as being obsessed with child pornography. This assessment was based on the fact that Mr. Thomas had explained that he had downloaded many images at once as a cost-saving measure and had saved them to look at later, perhaps once a week, thereby providing an explanation other than obsession for the large amount of material that he had collected. Dr. Blumberg also concluded, based on psychological testing that included validity measures that screened for both defensiveness and exaggeration, that the defendant was honest in claiming that he had never acted in a predatory way, except for the one occasion on which he video-taped his friend's daughter in May of 2000, the subject of the molestation charge. (Tr. 27, 39, Apr. 22).
Dr. Blumberg defined the term "erotica" in this context as "nudist and naturalistic pictures of naked children. . . ." (Tr. 33, Apr. 22).
Dr. Blumberg thus found that a diagnosis of pedophilia was warranted because the defendant was both interested in and aroused by child pornography, and had also taken affirmative steps to obtain and collect such illegal material. However, Dr. Blumberg also concluded that because Mr. Thomas had reported that he was only attracted to female children, but not to male children, and that he maintained a primary adult heterosexual orientation, his ultimate prognosis was better than it would be otherwise. Dr. Blumberg found the defendant both remorseful and eager to seek treatment for his pedophilia, which he also believed were positive indications. Consequently, Dr. Blumberg testified that he believed the proposed conditions of release, including restriction of the defendant to his parents' home subject to electronic monitoring, no access to children, and no access to a computer or the internet, were sufficient measures to limit Mr. Thomas's access to potential victims so as to reasonably assure that he would not pose a significant risk to the community. (Tr. 37-38, Apr. 22). 2. James Clemente, J.D.
The DSM-IV states that "[t]he recidivism rate for individuals with [p]edophilia involving a preference for males is roughly twice that for those who prefer females." DSM-IV, supra note 6, at 528. The DSM-IV also notes that "[t]he frequency of pedophiliac behavior often fluctuates with psychosocial stress [and] [t]he course is usually chronic, especially in those attracted to males." Id.
The government presented the testimony of James Clemente. Mr. Clemente is a supervisory special agent ("SSA") with the Federal Bureau of Investigation's National Center for the Analysis of Violent Crimes ("NCAVC"), Behavioral Analysis Unit ("BAU"). SSA Clemente has specialized training in the investigation of criminal sexual exploitation of children, as well as some graduate level education in clinical forensic psychology, criminology, and research methodologies. Additionally, SSA Clemente has analyzed and consulted on over 1000 child sexual exploitation and victimization cases, as well as conducting interviews of sex offenders and related research. SSA Clemente was offered, without objection, and accepted by the Court, as an expert in "criminal investigative analysis" in light of his training and experience in investigating child sex offenders and their characteristics and methods. In his written declaration, SSA Clemente defined the term "criminal investigative analysis" as "a law enforcement tool which utilizes investigative results, forensic findings and victim and offender behavior to assess cases. It is further used in child sexual victimization cases to dispel myths and misconceptions about individuals who sexually exploit or molest children." (Declaration, ¶ 5).
SSA Clemente's curriculum vitae states that the mission of the NCAVC is to provide criminal investigative analysis of violent, sexual and serial crimes to law enforcement agencies worldwide. The NCAVC is comprised of three entities: the Behavioral Analysis Unit; the Child Abduction/Serial Murder Investigative Resource Center; and the Violent Criminal Apprehension Program. The three entities "represent the culmination of nearly 30 years of Criminal Investigative Analysis within the FBI." (Clemente C.V.).
SSA Clemente's curriculum vitae states:
The Behavioral Analysis Unit is staffed by Supervisory Special Agents with an average of 18 years of law enforcement experience focused on violent and sexual criminal investigations. Many BAU members hold advanced degrees in areas including Law, Psychology, Entomology and Criminology. Each member has completed the NCAVC's comprehensive 560-hour training regimen, as well as other advanced and specialized courses.
All BAU members perform case analysis, conduct research, and provide training. Annually they analyze 1,500+ cases, conduct multiple ongoing research projects, and train 10,000+ law enforcement officers, prosecutors and health care professionals worldwide. BAU members also provide on-site Crime Scene Analysis and Expert Testimony.
SSA Clemente also holds a J.D., a law degree, as well as a B.S. in Chemistry. (Clemente C.V.). He was previously assigned to the New York Field Division's Joint FBI/NYPD Sexual Exploitation of Children Task Force. Prior to joining the FBI, he headed the Child Sex Crimes Prosecution Team for the New York City Law Department. (Clemente C.V.).
Defense counsel argued, in a memorandum of law in support of pretrial release, that SSA Clemente's opinions should be afforded no weight on two grounds: (1) because they did not meet the Daubert factors; and (2) because SSA Clemente's declaration indicated "no formal training at all in any of the sciences generally entrusted with risk assessment." However, no objections were raised by the defense at the hearing regarding SSA's Clemente's qualifications.
SSA Clemente explained that the term "child sex offender" is a law enforcement term that applies to those who commit crimes (of a sexual nature) against children under the age of eighteen. He further noted that, although the psychiatric term of pedophile applies to some child sex offenders, it does not apply to all, as the term pedophile is reserved for those individuals who are sexually attracted to prepubescent children. (Tr. 23, Apr. 28).
The FBI NCAVC website states, in part:
The mission of the BAU is to provide behavioral based investigative and operational support by applying case experience, research, and training to complex and time-sensitive crimes typically involving acts or threats of violence . . . [including] Crimes Against Children . . . BAU assistance to law enforcement agencies is provided through the process of "criminal investigative analysis." Criminal investigative analysis is a process of reviewing crimes from both a behavioral and investigative perspective. It involves reviewing and assessing the facts of a criminal act, interpreting offender behavior, and interaction with the victim, as exhibited during the commission of the crime, or as displayed in the crime scene. BAU staff conduct detailed analyses of crimes for the purpose of providing one or more of the following services: crime analysis, investigative suggestions, profiles of unknown offenders, threat analysis, critical incident analysis, interview strategies, major case management, search warrant assistance, prosecutive and trial strategies, and expert testimony.
Federal Bureau of Investigation, Critical Incident Response Group, NCAVC, Mission Statement (2003), available at http://www.fbi.gov/hq/isd/cirg/ncavc.htm (last visited January 4, 2006) (emphasis added).
SSA Clemente further explained that preferential child sex offenders exhibit: (1) long-term and persistent patterns of behavior; (2) children as preferred sexual objects; (3) well-developed techniques in obtaining child pornography and or child victims; and (4) fantasy-driven behavior. (Declaration, ¶¶ 7-11) (Tr. 23-24, Apr. 28). He characterized a collection of child pornography as the most telling sign of a preferential child sex offender. (Declaration, ¶ 14). SSA Clemente asserted that because such offenders are often sexually attracted to children of a particular age group, they must periodically search out and groom new children for victimization, because the children they are sexually interested in grow up and "age out" of their desired age range. SSA Clemente stated that such classification is significant, because preferential offenders are "much more likely" to re-offend than situational offenders, as preferential offenders' behavior is compulsive in nature, and therefore less easily controlled. (Tr. 34-35, Apr. 28)
SSA Clemente's declaration explained that there are preferential and situational child sex offenders:
Situational child sex offenders do not have a dominant sexual attraction to children. As a result, their sexual activities with children tend to result from accidental or circumstantial access to children rather than a methodical effort to pursue children. Their offenses can be described as opportunistic and impulsive in nature. Preferential child sex offenders, on the other hand, find themselves sexually attracted to children, usually of a particular age group or other specific characteristics . . . These offenders must periodically search out new [child victims] because the children they are sexually interested in always grow up and "age out" of their desired age range. Their sexual behavior with children, therefore, is typically repetitive and predatory in nature . . . [and they] tend to engage in highly predictable behavior patterns.
(Declaration, ¶¶ 6-7).
SSA Clemente formed his opinions regarding Mr. Thomas, the defendant in this case, after reviewing interview reports, evidence summaries, statements made by Mr. Thomas, and the images and the video seized during the search discussed previously. He also consulted with the principal investigator and Assistant United States Attorney involved in the prosecution of this case. He did not, however, interview the defendant.
SSA Clemente opined that Mr. Thomas has exhibited all four of the typical hallmarks, or behavioral characteristics, of a preferential child sex offender. First, he equated the defendant's collection of child pornography and erotica, amassed over a period of six years, with a long-term and persistent pattern of behavior. Second, SSA Clemente found that Mr. Thomas's collection, which focused almost exclusively on prepubescent girls between the ages of five and thirteen, demonstrated his interest in children as preferred sexual objects. Third, he characterized the defendant's friendship with the mother of the child he videotaped as an effort to gain access and control over that child and her sister. He further assessed Mr. Thomas's relationship with these two girls as typical "grooming" behavior exhibited by such offenders. SSA Clemente thus concluded that Mr. Thomas utilized two well-developed or sophisticated techniques, that is, his friendship with the girl's mother and grooming behavior, to gain access to child victims. Fourth, SSA Clemente found that both the child pornography collection and the videotaping of his friend's daughter demonstrated fantasy-driven behavior, noting that the collection was very directed and goal oriented, and the level of risk the defendant took to make the tape was high. Additionally, SSA Clemente found it significant that the video created by Mr. Thomas spliced together the non-pornographic images of young girls characterized as surveillance video, adult pornography, and the images of the genitalia of his friend's daughter in such a way as to satisfy himself, thus providing additional insight into Mr. Thomas's predilections. Therefore, SSA Clemente concluded that the defendant's behavior, in combination with his child pornography collection that predominantly featured prepubescent girls, demonstrate that he is a preferential child sex offender.
SSA Clemente testified that although it is impossible to predict future criminal behavior with absolute certainty, the best indicator of future behavior is a past pattern of behavior. He stated that the risk of this defendant reoffending is high because Mr. Thomas's conduct regarding his friend's daughter crossed the line separating fantasy from active molestation. (Declaration, ¶ 20) (Tr. 39-40, Apr. 28). He further assessed that Mr. Thomas's overall conduct, which had progressed from private fantasy, to collection of child pornography, to production of child pornography, to actual molestation of a child by exposing and touching her genital areas, demonstrated an escalation of behavior. (Tr. 40, Apr. 28).
SSA Clemente believed some of the assertions made by Mr. Thomas in his written statement did not indicate acknowledgment of his sexual attraction to young girls or acceptance of responsibility for his behavior, but rather represented an attempt to rationalize his conduct. (Tr. 38, Apr. 28). He also stated that because Mr. Thomas had victimized an unrelated child, that is, an extrafamilial child, he was much more likely to re-offend, and that the literature clearly supported this assertion. (Tr. 40-41, Apr. 28). Specifically, SSA Clemente testified that in his opinion there was a "high" risk, which he further defined as "more likely than not", that Mr. Thomas would reoffend if released, even under the proposed conditions, pending trial. (Tr. 75-77, Apr. 28). However, SSA Clemente also testified that he did not believe Mr. Thomas was at the "highest" possible risk for re-offense if released pending trial. (Tr. 75, Apr. 28).
When questioned by the Court regarding the basis for his opinions, SSA Clemente identified his experience as a criminal investigator, the institutional knowledge of the FBI, and a study of inmates in a Sex Offender Treatment Program at the Federal Correctional Institution in Butner, North Carolina. See Andres E. Hernandez, Psy.D., Self-Reported Contact Sexual Offenses by Participants in the Federal Bureau of Prisons' Sex Offender Treatment Program: Implications for Internet Sex Offenders (2000) ("Hernandez Study"). The primary objective of this study was to "examine the incidence of sexual offending involving contact crimes (e.g., child sexual abuse, rape) of program participants, including those inmates convicted of non-contact sexual offenses (e.g., possession of child pornography)." Id. The results of this study were that 90 subjects convicted of crimes involving the production, distribution, receipt, and possession of child pornography, or the luring of a child and traveling across state lines to sexually abuse a child, had committed an additional 1,622 sexual crimes which had never been detected by the criminal justice system. However, when questioned by the Court regarding the basis for his risk assessment as to this defendant, SSA Clemente was unable to explain how his methods or techniques had been error-checked or peer reviewed for accuracy and reliability. This portion of SSA Clemente's testimony is discussed in more detail infra, and will not be reiterated here.
II. Discussion A. The Bail Reform Act of 1984
The Bail Reform Act of 1984 ("the Act"), 18 U.S.C. §§ 3141 et seq., authorizes a court to order a defendant's detention pending trial in certain circumstances if "no condition or combination of conditions will reasonably assure the appearance of the person as required and the safety of any other person and the community. . . ." 18 U.S.C. § 3142(e). The Act is preventative, rather than punitive, in nature. See United States v. Salerno, 481 U.S. 739, 747, 107 S.Ct. 2095, 2101 (1987). Accordingly, the Supreme Court has stated that "liberty is the norm, and detention prior to trial or without trial is the carefully limited exception." Id., 481 U.S. at 755, 107 S.Ct. at 2105.
The government may move for pretrial detention under Section 3142(e) if at least one of the six categories listed in Section 3142(f) is met. United States v. Byrd, 969 F.2d 106, 109 (5th Cir. 1992). The defendant in this case is charged with two offenses related to the receipt, possession and shipment of child pornography, in violation of 18 U.S.C. § 2252(A), as well as sexual exploitation of a child in violation of 18 U.S.C. § 2251(a). Since these offenses fall within the Bail Reform Act's definition of "crimes of violence" (see 18 U.S.C. § 3156(a)(4)(C)), the government is entitled to move for detention. 18 U.S.C. § 3142(f)(1)(A).
At the detention hearing, the Court is charged with determining "whether there are conditions of release that will reasonably assure the appearance of the person as required and the safety of any other person and the community. . . ." U.S.C. § 3142(g). The factors which the Court must consider in deciding whether to detain or release a defendant, and if released under what condition or combination of conditions, include: (1) the nature and circumstances of the crime charged; (2) the weight of the evidence against the defendant; (3) the history and characteristics of the defendant, including character, physical and mental condition, family ties, employment, financial resources, community ties, past conduct, history relating to drug or alcohol abuse, criminal history, and record concerning appearance at court proceedings; and (4) the nature and seriousness of the danger posed by the defendant to any person or the community. Id.
Concerning the first factor, it is not too dramatic to say that the nature and circumstances of these charges, particularly the alleged sexual exploitation of a child, strike fear into the hearts of all parents. The alleged behavior of Mr. Thomas is both profoundly upsetting and revolting. The incident giving rise to the exploitation charge, however, occurred approximately three years ago, and there was no evidence that Mr. Thomas has engaged in any other behavior that can be characterized as "acting out," or crossing the line that separates fantasy from active molestation. To the contrary, the evidence before the Court regarding this defendant indicates that the single incident that gave rise to the exploitation charge was an aberration, rather than a pattern of such behavior. Additionally, while the Court in no way wishes to minimize the seriousness of Mr. Thomas's conduct, both sides acknowledge that it may fairly be placed towards the less egregious end of the spectrum of molestation, although of course all such behavior is both criminal and reprehensible.
Regarding the second factor, the weight of the evidence is very strong in this case. The evidence before the Court includes a written statement in which Mr. Thomas admitted abusing a friend's trust by exposing and videotaping her young daughter's genitalia while she slept. In addition, the Court has had the opportunity to view portions of that video. Finally, there were approximately 16,000 pornographic images of children found in Mr. Thomas's home on various storage devices. Therefore, the evidence against the defendant is strong and the second factor thus weighs heavily against the defendant.
The third factor, the history and characteristics of Mr. Thomas, weighs significantly in the defendant's favor. While Dr. Blumberg diagnosed Mr. Thomas with the mental disorder of pedophilia, carrying such a diagnosis is not a crime in and of itself, provided that his sexual interest in young females is not acted upon. Mr. Thomas has strong family ties, as demonstrated by the support of his family throughout these proceedings, as well as his mother's and stepfather's willingness to serve as third party custodians. The defendant has a solid employment record and strong connections to both the Connecticut and Maryland communities. There is no evidence that Mr. Thomas currently abuses alcohol or drugs or has ever done so in the past. Perhaps most significantly, the defendant has no prior criminal history whatsoever.
The fourth factor, that is, the nature and seriousness of the danger posed by the defendant to any person or the community, is the most difficult in this case. Before examining the evidence regarding this factor in detail, it is important to note that, at the time of the detention proceeding, the defendant was not charged with an offense giving rise to a presumption against release under the Bail Reform Act. The mere fact that Mr. Thomas is charged with a crime of violence does not trigger a presumption. To support its request for detention, the government must therefore carry its burden of proving danger without the benefit of any such presumption.
Although legislation was pending at the time of these proceedings that would create a presumption against release for the offenses charged in this case, it was not enacted at the time of the Court's ruling on April 28, 2003, and thus is inapplicable to this matter. The law was subsequently amended after the Court's ruling. See Prosecutorial Remedies and Other Tools to end the Exploitation of Children Today Act of 2003 or PROTECT Act, S. 151, 108th Cong., Title II, § 203(2) (2003) (enacted Apr. 30, 2003 as P.L. 108-21) (amending 18 U.S.C. section 3142(e) so as to include sections 2251 and 2252 under Title 18 as offenses for which there is a rebuttable presumption against release). Additionally, the Court notes that even if such a presumption applied as a matter of law to these proceedings, it would have been overcome by the evidence presented by the defendant in this case.
The applicable standard of proof is that of "clear and convincing" evidence. The Bail Reform Act provides that "[t]he facts the judicial officer uses to support a finding . . . that no condition or combination of conditions will reasonably assure the safety of any other person and the community shall be supported by clear and convincing evidence." 18 U.S.C. § 3142(f) (emphasis added). Clear and convincing evidence, of course, is a higher standard than proof by a preponderance, but not as high as the standard applied in a criminal trial, which is proof beyond a reasonable doubt. MODERN FEDERAL JURY INSTRUCTIONS, 73-3 (Sand et al., eds., 2002).
Clear and convincing proof leaves no substantial doubt in [one's] mind. It is proof that establishes in [one's] mind, not only [that] the proposition at issue is probable, but also that it is highly probable. It is enough if the party with the burden of proof establishes his claim beyond any "substantial doubt"; he does not have to dispel every "reasonable doubt."Id. See also Addington v. Texas, 441 U.S. 418, 423-33 (1979) (discussing the three standards of proof). Additionally, the Fourth Circuit has stated that the "clear and convincing" standard may be defined as "highly probable." Direx Israel, Ltd. v. Breakthrough Med. Corp., 952 F.2d 802, 810 (4th Cir. 1992) (citing 9 J. WIGMORE EVIDENCE § 2498 (3d ed. 1940)). Cf. Jones v. Pitt County Bd. of Educ., 528 F.2d 414, 417 (4th Cir. 1975) ("a firm belief or conviction as to the allegations sought to be established"); Carpenter v. Union Ins. Soc'y of Canton, Ltd., 284 F.2d 155, 162 (4th Cir. 1960) ("clear and satisfactory evidence to a reasonable certainty") (internal quotation marks and citations omitted).
B. The Fourth Factor of the Bail Reform Act: Danger to the Community
With the standard of proof applicable in this matter thus firmly in mind, the Court returns to the fourth factor that it is required to consider, the nature and seriousness of the danger posed by the defendant's release to any person or thecommunity. 18 U.S.C. § 3142(g)(4). There is no indication whatsoever that Mr. Thomas poses a danger to any particular child. There is no evidence that he has subsequently stalked or pursued the girl depicted in the video in any way, other than a single, innocuous birthday greeting communicated by electronic mail. Additionally, the isolated nature of the event, as well as the fact that it occurred approximately three years ago, militates against concluding that Mr. Thomas continues to pose a danger to this particular child. Finally, a thorough search and analysis of Mr. Thomas's home, computer, video tapes and digital storage devices has revealed no evidence that he ever actively molested any other child. Therefore, there is no clear and convincing evidence that Mr. Thomas poses a threat to any particular child.
The Court must thus weigh the evidence presented regarding the danger, or lack thereof, posed by Mr. Thomas to children in the community at large. Certainly, both sides seem in agreement that Mr. Thomas poses little or no danger at all to male children, as both his collection, conduct and statements all evince an exclusive attraction to female children. Indeed, the government's position regarding the defendant's dangerousness is predicated upon the assumption that Mr. Thomas is a preferential sex offender, with a specific preference for prepubescent girls. No evidence has been presented that the defendant poses a danger to male children. Consequently, it is left to the Court to consider what danger Mr. Thomas poses to female children in the community at large under the proposed strict conditions of release.
On this point, of course, the two experts are diametrically opposed on the issue of future dangerousness. Dr. Blumberg, a psychiatrist with considerable credentials and experience in criminal forensic psychiatry, testified that he believed the proposed conditions of release, including restriction of the defendant to his parents' home subject to electronic monitoring, no access to children, and no access to a computer or the internet, were sufficient measures to limit Mr. Thomas's access to potential victims so as to reasonably assure that he would not pose a significant risk to the community. SSA Clemente, on the other hand, a respected professional in the law enforcement community with extensive experience as an investigator regarding sex crimes against children, testified that he believed Mr. Thomas, as a preferential sex offender, was at a high risk for re-offense if released, even under the proposed conditions, pending trial.
The rules of evidence, of course, do not apply to the presentation and consideration of information at the detention hearing. 18 U.S.C. § 3142(f). However, the Court must nevertheless ensure that the evidence upon which it relies is reliable. United States v. Goba, 240 F.Supp.2d 242, 247 (W.D.N.Y. 2003) (noting that judicial interest in preventing detention proceedings from becoming mini-trials must be tempered by the court's obligation to ensure the reliability of proffered evidence) (internal citations omitted). See also United States v. LaFontaine, 210 F.3d 125, 131 (2d Cir. 2000) (explaining that magistrate judges must ensure reliability "by selectively insisting upon the production of the underlying evidence or evidentiary sources where their accuracy is in question") (internal citations omitted)); United States v. Acevedo-Ramos, 755 F.2d 203, 206-08 (1st Cir. 1985) (assuming that evidence used for purposes of detention determinations must be reliable); United States v. Accetturo, 623 F.Supp. 746, 755 (D.N.J. 1985) (discussing the traditional requirement that evidence proffered in detention hearings be reliable).
The statute states that "[t]he rules concerning admissibility of evidence in criminal trials do not apply to the presentation and consideration of information at the hearing." 18 U.S.C. § 3142(f). See also, e.g., United States v. Lee, 156 F.Supp.2d 620 (E.D. La. 2001) (testimony based upon hearsay was admissible in defendant's detention hearing where evidence was credible); United States v. Bellomo, 944 F.Supp. 1160 (S.D.N.Y. 1996) (polygraph evidence was properly considered by district court in reviewing pretrial detention order, even though inadmissible in criminal trials, given inapplicability of evidentiary rules to detention hearings under Bail Reform Act);United States v. Terrones, 712 F.Supp. 786 (S.D. Cal. 1989) (rules of evidence do not apply at proceedings for determining whether individual can be detained without bail pending trial);United States v. Wind, 527 F.2d 672 (6th Cir. 1975) (evidence at bail hearing need not conform to rules pertaining to admissibility of evidence in court of law).
While the rules of evidence do not govern the admission of evidence at the detention hearing, the law governing the admission of expert testimony in the trial context is instructive in determining the reliability of the expert testimony offered in this matter, and thus is helpful in determining the weight to accord the evidence presented in the instant case. See United States v. Hammond, 44 F.Supp.2d 743, 746 (D. Md. 1999).
C. Evidentiary Law Regarding Expert Opinion
Federal Rule of Evidence 702 provides:
If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.
In Daubert, the Supreme Court charged trial judges with a gatekeeper function to exclude unreliable expert testimony. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 589, 593-94, 597, 113 S.Ct. 2786 (1993). First, the trial court must ensure that the evidence is reliable. Id. at 592. Second, the court must ensure that the evidence will assist the trier of fact and is thus relevant. Id. The Court articulated five factors that could be used in making a reliability determination, while emphasizing that the analysis should be flexible: (1) whether the theory can be and has been tested; (2) whether the theory has been subjected to peer review and publication; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling its operation; and (5) whether it has achieved general acceptance in the relevant community. Id. at 593-94. In Kumho Tire Co. Ltd. v. Carmichael the Court held that the principles enunciated inDaubert applied to all expert testimony, not merely scientific evidence. 526 U.S. 137, 119 S.Ct. 1167, 1171 (1999).
The Daubert standard replaced the test from Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923). Daubert, 509 U.S. at 584-89, 113 S.Ct. 2786. The Frye test had required that the theory be generally accepted by the relevant community before a court could admit it. 293 F. 1013, 1014.
Accordingly, the Fourth Circuit has stated: A reliable expert opinion must be based on scientific, technical, or other specialized knowledge and not on belief or speculation, and inferences must be derived using scientific or other valid methods.
Oglesby v. General Motors Corp., 190 F.3d 244, 250 (4th Cir. 1999) (citing Daubert, 509 U.S. at 590, 592-93, 113 S.Ct. 2786). Significantly, the Fourth Circuit has further noted that the reliability inquiry "necessitates an examination of . . . the reasoning or methodology underlying the expert's proffered opinion." Westberry v. Gislaved Gummi AB, 178 F.3d 257, 260 (4th Cir. 1999). The Court also explained that "the [trial] court has broad latitude to consider whatever factors bearing on validity that the court finds to be useful; the particular factors will depend upon the unique circumstances of the expert testimony involved." Id. at 261 (citing Kumho Tire Co. at 1175-76).
D. Expert Psychiatric Testimony 1. Antecedent Law
While not without debate, federal courts have long recognized that expert psychiatric testimony is generally admissible on the issue of assessing future dangerousness. Barefoot v. Estelle, 463 U.S. 880, 896-903, 103 S.Ct. 3383, 3396-99 (1983) (superseded by statute on other grounds) (holding that expert psychiatric testimony offered by the government to establish defendant's potential future dangerousness as an aggravating sentencing factor was admissible in a capital sentencing context and did not violate due process); Woomer v. Aiken, 856 F.2d 677 (4th Cir. 1988) (same). Cf. Addington v. Texas, 441 U.S. 418, 429, 433, 99 S.Ct. 1804, 1811 (1979) (holding that the standard of proof for use in commitment for mental illness due to dangerousness was clear and convincing, rather than the higher standard of beyond a reasonable doubt, because "[g]iven the lack of certainty and the fallibility of psychiatric diagnosis there is a serious question as to whether a state could ever prove beyond a reasonable doubt that an individual is both mentally ill and likely to be dangerous").
2. Criticism of Psychiatric Assessments
The government did not specifically challenge the bases of Dr. Blumberg's opinions, or even generally question the utility of psychiatric evaluation as a tool for assessing the defendant's future dangerousness. It must be acknowledged, however, that although the law recognizes the relevance and admissibility of psychiatric evaluation and opinion for the purpose of assessing future dangerousness, the use of such evidence has long been criticized as lacking reliability. See, e.g., Bruce J. Ennis Thomas R. Litwack, Psychiatry the Presumption of Expertise: Flipping Coins in the Courtroom, 62 CAL. L. REV. 693, 695-96 (1974). See also, Eugenia T. La Fontaine, A Dangerous Preoccupation With Future Danger: Why Expert Predictions of Future Dangerousness in Capital Cases are Unconstitutional, 44 B.C. L. REV. 207, 241 (2002). In 1974, the American Psychiatric Association Task Force on Clinical Aspects of the Violent Individual, for example, took the position that "neither psychiatrists nor anyone else have demonstrated an ability to predict future violence or dangerousness." American Psychiatric Association, Task Force on Clinical Aspects of the Violent Individual, Report No. 8, 20 (1974). In 1990, the American Psychological Association Task Force on the Role of Psychology in the Criminal Justice System concluded that "the validity of psychological predictions of violent behavior . . . is extremely poor, so poor that one could oppose their use on the strictly empirical grounds that psychologists are not professionally competent to make such judgments." American Psychological Association, Report of the Task Force on the Role of Psychology in the Criminal Justice System, 73 AMERICAN PSYCHOLOGIST 1099 (1990).
Standard 7-3.9 of the American Bar Association, Criminal Justice Mental Health Standards (approved by the ABA House of Delegates, August, 1984) specifically addresses the admissibility of expert testimony concerning a person's future behavior:
An expert opinion stating a conclusion that a particular person will or will not engage in dangerous behavior in the future should not be admissible in any criminal proceeding or in any special commitment hearing involving a person found not responsible under the criminal law. Expert testimony relating to the person's future mental condition or behavior should be admissible in any criminal proceeding or in any special commitment hearing whenever the testimony is based on and is within the specialized knowledge of the witness and is limited to a description of:
(i) the clinical significance of the individual's personal history and proven past criminal act(s);
(ii) scientific studies involving the relationship between specific behaviors and variables that are objectively measurable and verifiable;
(iii) the possible psychological or behavioral effects of proposed therapeutic or habilitative interventions; or
(iv) the factors that tend to enhance or diminish the likelihood that specific types of behavior could occur in the future.Id. at 117-118. While approved in 1984, these Standards remain in effect today. American Bar Association, Criminal Justice Mental Health Standards (1989), available at http://www.abanet.org/crimjust/standards/mentalhealth_toc.html (last visited January 3, 2006). The commentary on this standard further explains that:
[The standard] advocates severe restrictions on the admissibility of mental health . . . professional testimony on future mental condition or behavior. Descriptive data based on clinical observations and behavioral studies is admissible, but expert witnesses should not be permitted to predict an individual's future dangerousness . . . [T]he Association recommends the adoption of a more restrictive rule than the due process minimum embodied in [Barefoot v. Estelle, 463 U.S. 880 (1983)] [b]ecause the relevant scientific community acknowledges the lack of a valid, reliable predictive technology. . . .Id. at 125. It is notable, however, as well as particularly germane to the instant case, that the commentary also makes clear that the general concern expressed over the potential misuse of psychiatric evaluations to assess future dangerousness is due partially to the fact that "psychiatric procedures and techniques [have] not succeeded in reducing the high rate of "false positive" predictions, i.e., affirmative predictions of future violent behavior that are subsequently proven erroneous." Id. at 126, n. 28 (internal citations omitted) (emphasis added).
3. The Current State of the Law
While courts have recognized the fallibility of psychiatric assessments of future dangerousness, they nevertheless acknowledge the necessary reliance on psychiatry to assist in judicial decisionmaking. Barefoot v. Estelle, 463 U.S. 880, 897, 103 S.Ct. 3383, 3396 (1983). The dissent in Barefoot noted that the American Psychiatric Association, in an amicus curiae brief, estimated that two out of three predictions of long-term future violence made by psychiatrists are wrong. 463 U.S. 880, 920, 103 S.Ct. 3383, 3409 (1983). Nonetheless, the majority approved of such testimony:
The suggestion that no psychiatrist's testimony may be presented with respect to a defendant's future dangerousness is somewhat like asking us to disinvent the wheel . . . [i]t is, of course, not easy to predict future behavior. The fact that such a determination is difficult, however, does not mean that it cannot be made. Indeed, prediction of future criminal conduct is an essential element in many of the decisions rendered throughout our criminal justice system. The decision whether to admit a defendant to bail, for instance, must often turn on a judge's prediction of the defendant's future conduct.Id. at 897, 3396 (internal quotation marks and citations omitted) (emphasis added). See also, e.g., United States v. Williams, 299 F.3d 673 (8th Cir. 2002) (civil commitment);Tigner v. Cockrell, 264 F.3d 521 (5th Cir. 2001) (aggravating factor in capital case); Doe v. Pataki, 120 F.3d 1263 (2d Cir. 1997) (parole of child sex offender and notification requirements pursuant to Megan's Law based on re-offense risk assessment level); Fuller v. Johnson, 114 F.3d 491 (5th Cir. 1997) (habeas appeal in capital sentencing context); United States v. Ecker, 543 F.2d 178 (D.C. Cir. 1976) (civil commitment following criminal acquittal).
4. Fourth Circuit Precedent
Moreover, the Fourth Circuit has held, assuming arguendo that the Daubert evidentiary standard applied to capital sentencing hearings, that expert forensic psychological testimony offered by the government regarding future dangerousness and psychopathy as an aggravating sentencing factor was not only admissible, but reliable, because the expert used the following bases for his opinion: (1) the Diagnostic and Statistical Manual, Fourth Edition; (2) the Psychopathy Checklist Revised; (3) first-hand observations of the defendant's behavior; and (4) an actuarial analysis comparing the defendant to groups of people with characteristics similar to him. United States v. Barnette, 211 F.3d 803, 811, 815-16 (4th Cir. 2000).
Cf. United States v. White Horse, 316 F.3d 769, 775 (8th Cir. 2003) (assessment conducted by psychological expert regarding expert's opinion that defendant did not have a sexual interest in underage boys did not satisfy Daubert admissibility standards due to concerns about how well the psychologist's assessment tool — the "Abel Assessment" — `fit' the facts of the case) ("We also do not believe that the clinical interview and mental status examination could have formed an independent basis for the psychologist's expert opinion.").
In the instant case, the Court finds Dr. Blumberg's opinion reliable for the following reasons. First, a review of Dr. Blumberg's curriculum vitae in evidence before this Court, as well as voir dire during the hearing proceedings, reveal that he is eminently qualified as an expert in forensic criminal psychiatry. United States v. Riggleman, 411 F.2d 1190 (4th Cir. 1969) (holding that clinical psychologist who had doctorate in psychology, had qualified as expert on legal sanity in several states and who had spent most of his professional life working under the aegis of one of leading forensic criminal psychiatrists in the country, was qualified to render opinion as to defendant's capacity to appreciate criminality of his conduct or to conform his conduct to requirement of law); Hidden v. Mutual Life Ins. Co. of New York, 217 F.2d 818 (4th Cir. 1954) (holding that testimony of psychologist who was qualified in his field by academic training and experience was erroneously excluded). Second, Dr. Blumberg conducted a standard forensic psychiatric evaluation during which he conducted a thorough mental status examination of Mr. Thomas, administered psychological testing that incorporated validity scales, made first-hand observations of the defendant's behavior and demeanor, interviewed family and friends of the defendant, and utilized the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, in reaching his diagnostic conclusions. Additionally, Dr. Blumberg employed an actuarial approach to the extent that he compared the defendant to groups of people with characteristics similar to him. Thus, the Court finds Dr. Blumberg's opinions reliable.
Forensic psychiatry is a branch of medicine focusing on the interface of law and mental health and comprises psychiatric consultation and expert testimony, as well as clinical work with both victims and perpetrators. Harold J. Bursztajn, M.D., The Role of a Forensic Psychologist in Legal Proceedings: An Overview of the Function of Forensic Psychology, at http://www.forensic-psych.com/articles/artAskexp01.html (last visited January 9, 2006). A forensic psychiatrist is a medical doctor "who has completed several years of additional training in the understanding, diagnosis, and treatment of mental disorders" with yet "additional training and/or experience related to the interface of mental health (or mental illness) and the law."Id. He or she "integrates clinical experience, knowledge of medicine, mental health, and the neurosciences to form an independent, objective opinion. Relevant data is gathered and analyzed as part of a process of alternative hypothesis testing to form an expert medical/psychiatric opinion." Harold J. Bursztajn, M.D., Medicine Psychiatry, at http://www.forensic-psych.com/(last visited January 9, 2006). This expert opinion is then "communicated by written report, deposition, or courtroom testimony." Id.
Notably, it is clear from Dr. Blumberg's testimony that he did not rely on any one basis or test in making his risk assessment of Mr. Thomas. Rather, Dr. Blumberg's testimony indicated that he took all of the available evidence about Mr. Thomas into consideration and used both clinical assessment and actuarial data in forming his opinions.
The clinical method is based on "observational and personal examination, history-taking, and testing. The clinician reviews the data obtained from the assessment and forms an opinion about the likelihood of the individual engaging in a particular future behavior." A JUDGE'S DESKBOOK ON THE BASIC PHILOSOPHIES METHODS OF SCIENCE 181 (State Justice Institute, 1999).
The challenge in the clinical approach "is translating the observed risk factors into recidivism probabilities. Although offenders with all the risk factors would be considered high risk, and those with no risk factors would be considered low risk, this approach provides no explicit direction on how to gauge the risk of the typical offender who has some risk factors." R. Karl Hanson, RISK ASSESSMENT 3-6 (Association for the Treatment of Sexual Abusers, 2000), available at http:// www.atsa.com/pdfs/InfoPack-Risk.pdf (last visited January 9, 2006).
The actuarial method is based on "assigning statistical probabilities of outcomes from combinations of a number of variables that correlate with the behavior at issue . . . [t]he expert's opinion is a general probability based on given variable percentages." A Judge's Deskbook on the Basic Philosophies Methods of Science 181 (State Justice Institute, 1999).
First, Dr. Blumberg took an extensive medical, legal, and psychiatric history of the defendant. He did not rely simply on Mr. Thomas's self-report as to these matters, but also interviewed his family and former girlfriend. Additionally, he was fully aware of the events leading to the charges against Mr. Thomas, and the strength and substance of the evidence against the defendant, as well as the lack of any prior criminal history. He also considered the psychiatric history of Mr. Thomas's family, as there was an extensive history of depression among certain relatives that pre-disposed Mr. Thomas to dysthymia, as well as other psychiatric implications relevant to the defendant's self-esteem and social adjustment.
Second, Dr. Blumberg administered the Minnesota Multiphasic Personality Inventory-2 ("MMPI-2"). The MMPI-2, consisting of 567 items, is the most widely used and researched psychological test available. RICHARD I. LANYON LEONARD D. GOODSTEIN, PERSONALITY ASSESSMENT 68, 72 (John Wiley Sons, Inc., 3rd ed. 1997). The test is used as a diagnostic aid in medical and psychiatric screening, and serves as "an objective, reliable screening instrument for appraising a person's personality characteristics and symptomatic behavior." JAMES NEAL BUTCHER, THE MMPI-2 IN PSYCHOLOGICAL TREATMENT 23 (Oxford University Press, 1990). The test "provides an objective evaluation of the subject's personality characteristics, symptom patterns, and personal attitudes that have been shown, by numerous research studies . . . to be relevant to many aspects of a clinical profile and prognosis." Id. at 24. The MMPI-2 contains ten clinical scales and several validity scales. RICHARD I. LANYON LEONARD D. GOODSTEIN, PERSONALITY ASSESSMENT 69-70 (John Wiley Sons, Inc., 3rd ed. 1997). It should be noted that the MMPI-2 does not "consistently differentiate the type of offender, particularly the sexual offender, from non-offenders, and the sexual offenders from other types of offenders." GEORGE B. PALERMO, M.D., MARY ANN FARKAS, PH.D., THE DILEMMA OF THE SEXUAL OFFENDER 63 (Charles C. Thomas, Ltd. 2001). However, Dr. Blumberg did not use the MMPI-2 results to conclude that Mr. Thomas's profile indicated that he was not a pedophile or a sex offender. Rather, Dr. Blumberg diagnosed Mr. Thomas with pedophilia, and used the MMPI-2 to assist in his clinical assessment of both the presence and severity of other diagnoses, such as dysthymia and personality disorder with avoidant features, that affect Mr. Thomas's overall prognosis for controlling his pedophilic impulses and tendencies. Additionally, Dr. Blumberg testified that he believed the defendant was generally straightforward during the mental status examination, a clinical assessment reinforced by Dr. Blumberg's interpretation of the MMPI-2 results pertaining to the validity scales. Therefore, it appears that Dr. Blumberg appropriately utilized the results of the MMPI-2, in combination with other tools, in forming his opinions.
The clinical scale categories are: (1) Hypochondriasis; (2) Depression; (3) Hysteria; (4) Psychopathic Deviate; (5) Masculinity-Femininity; (6) Paranoia; (7) Psychasthenia; (8) Schizophrenia; (9) Hypomania; (10) Social Introversion-Extroversion. RICHARD I. LANYON LEONARD D. GOODSTEIN, PERSONALITY ASSESSMENT 69-70 (John Wiley Sons, Inc., 3rd ed. 1997). 28
The validity scales provide a basis on which to screen for a subject's general levels of frankness and defensiveness. Id. at 70.
The authors reviewed a number of tests used to determine the recidivism risk of the antisocial behavior of sexual offenders. The Psychopathy Checklist-Revised (PCL-R) "shows reliability for prediction of violence and recidivism; however, it needs collateral and file information in order to rate the individuals tested." GEORGE B. PALERMO, M.D., MARY ANN FARKAS, PH.D., THE DILEMMA OF THE SEXUAL OFFENDER 63 (Charles C. Thomas, Ltd. 2001). The Rapid Risk Assessment for Sexual Offense Recidivism (RRASOR) consists of four assessment factors: (1) prior sexual offenses; (2) age less than twenty-five; (3) extrafamilial victims; and (4) male victims. The authors rate the predictive accuracy of RRASOR as "moderate." When the Violence Risk Appraisal Guide (VRAG), an actuarial component, is combined with the ASSESS-LIST, a ten-item clinical assessment, its predictive value for sexual offender recidivism is "much higher." Finally, the authors note that the Multiphasic Sex Inventory (MSI) has been more successful than the MMPI-2 in detecting both sexual offenders and sexual offender subtypes. Id. at 63-64.See also Judith V. Becker William D. Murphy, What We Know Do Not Know About Assessing Treating Sex Offenders, 4 PSYCHOL. PUB. POL'Y L. 116, 122-126 (1998).
See generally JAMES N. BUTCHER, A BEGINNER'S GUIDE TO THE MMPI-2 (American Psychological Association, 1999). Additionally, the author comments:
Id. at 7.
While psychologists do not have to swear allegiance to a particular theoretical orientation or school of psychology to incorporate the MMPI-2 into their clinical practice . . . the description often attributed to MMPI users is dust bowl empiricists (a term referring to the arid period of drought in the 1930s during which little was thought to come to fruition but what did emerge was solid and lasting) because of the scientific skepticism and proclivity to disbelieve anything that has not been substantially verified by research data.
Dr. Blumberg also administered the Personality Assessment Inventory ("PAI"). The PAI is similar to the MMPI-2 in purpose and structure and also incorporates validity scales. RICHARD I. LANYON LEONARD D. GOODSTEIN, PERSONALITY ASSESSMENT 76-77 (John Wiley Sons, Inc., 3rd ed. 1997). The PAI was developed in 1991 for the clinical assessment of psychopathology and incorporates numerous features that enhance its utility. Id. Dr. Blumberg testified that he also used the results of the PAI, in combination with those of the MMPI-2, to assist his clinical assessment as previously described.
Notably, Dr. Blumberg did not administer any of the specialized assessment instruments for sexual recidivism, such as Minnesota Sex Offender Screening Tool-Revised (Mn SOST-R), Rapid Assessment for Sexual Offense Recidivism (RRASOR) or Static-99. The Court notes, but does not criticize this aspect of approach to Mr. Thomas' assessment. These instruments are said to measure long-term risk, rather than short and medium-terms, which, of course, describe the pretrial time period. Eric S. Janus and Robert A. Prentky, "Forensic Use of Actuarial Risk Assessment with Sex Offenders: Accuracy, Admissibility and Accountability," 40 Am. Crim. L. Rev. 1443, 1445 (Fall 2003).
To the extent that courts seek to measure the long-term, presumptively stable risk posed by individuals, ARA [actuarial risk assessment] provides the most accurate information. But courts ought to be concerned as well with how risk can be managed and modified in the short and medium-terms, through interventions such as treatment and community supervision. This domain, generally referred to as "dynamic" risk assessment, represents the most recent entree to the scientific literature and will likely be the focus of attention among scientists for the foreseeable future. Given its focus on long-term risk, however, ARA is of less direct relevance, at least given the current state of the art. Therefore, on these important questions of risk management and modification, courts may, for the time being, need to rely more heavily on carefully done clinical assessments, though it is likely that dynamic ARA will eventually complement these assessments as well.
Moreover, "the results from most actuarial risk assessment scales must be interpreted as reporting risk without consideration of treatment or state-of-the-art supervision. Id. at 1481. Finally, there appears to be considerable debate whether this general superiority of actuarial approaches to clinical judgment is applicable to actuarial measures for sexual offense recidivism. R. Karl Hanson, "What Do We Know About Sex Offender Risk Assessment?" 4 Psychol. Pub. Pol'y L, 50, 53 (1998). See Grant T. Harris, Marnie E. Rice Vernon L. Quinsey, Appraisal Management of Risk in Sexual Aggressors: Implications For Criminal Justice Policy, 4 PSYCHOL. PUB. POL'Y L. 73, 90 (1998). See also Eric S. Janus Robert A. Prentky, Forensic Use of Actuarial Risk Assessment with Sex Offenders: Accuracy, Admissibility and Accountability, 40 AM. CRIM. L. REV. 1443, 1445, 1456 (2003) (noting that actuarial studies proved to be superior to clinical prediction in either 33% or 47% of all studies, depending on the analysis used; although these studies focus on long term risk, not the short term risk at issue in the pretrial or medium detention situation, as discussed.)
In the main, experts continue to recognize that use of a combination of clinical methods and actuarial data (whether through use of a specialized actuarial assessment instrument or predictive data based on empirical studies) is necessary to make fully informed risk assessments. See Judith V. Becker William D. Murphy, What We Know Do Not Know About Assessing Treating Sex Offenders, 4 PSYCHOL. PUB. POL'Y L. 116, 121-124 (1998).
Accordingly, in the future if (1) research continues to develop on the superiority of actuarial assessment instruments, such as VRAG, to clinical judgment and if (2) the applicability of these instruments for the measurement of short term risk with supervision, is established, Dr. Blumberg's approach may be seen as insufficient. That is, a risk assessment of a charged sex offender without employment of these specialized actuarial assessment instruments may be viewed as incomplete and unreliable. R. Karl Hanson, What Do We Know About Sex Offender Risk Assessment, 4 Psychol. Pub. Pol'y L. 50, 67 (June 1998). ("As research progresses, actuarial approaches are expected to substantially outperform the guided clinical approaches, but currently each approach has demonstrated roughly equivalent (moderate) predictive accuracy."); John A. Fabian, Examining Our Approaches to Sex Offenders the Law, 29 Wm. Mitchell L. Rev. 1367, 1434 (2003). ("Overall research has increasingly revealed that actuarial risk instruments normed on certain populations of offenders exhibit more predictive reliability and validity than the clinical judgment of psychologists and psychiatrists alone. The current research suggests for a clinician to clinically adjust or modify actuarial risk instruments.") But this question was not neither posed to nor briefed for the Court. Dr. Blumberg's approach, utilizing clinical interview techniques and considering actuarial or empirical data on other similar sex offenders, seems sufficiently reliable and accepted in the psychiatric community, at this time.
The literature reviewing meta-analyses of the empirical research regarding the re-offense risk for sex offenders reflects that there are some validated risk criteria on which to base such assessments. Three consistent predictors of sexual recidivism are the number of previous sexual offenses, the selection of male victims, and the selection of unrelated victims. See Grant T. Harris, Marnie E. Rice Vernon L. Quinsey, Appraisal Management of Risk in Sexual Aggressors: Implications For Criminal Justice Policy, 4 PSYCHOL. PUB. POL'Y L. 73, 86 (1998). Two of the predictors, that is, the number of previous sexual offenses and the selection of male victims, are in Mr. Thomas's favor. The evidence indicates that Mr. Thomas has only committed one molestation offense, which occurred approximately three years ago, and that he selected a female, rather than a male, victim. One of the predictors, however, the selection of an extrafamilial victim, weighs against the defendant. Dr. Blumberg's testimony indicated that he considered these predictive factors when making his risk assessment.
Meta-analysis is a statistical procedure that integrates the results of several independent studies that may be properly combined. Matthias Egger, George Davey Smith, Andrew N. Phillips,Meta-Analysis Principles Procedures, at http://bmj.com/archive/7121/7121ed.htm (last visited January 13, 2006).
Proximate risk factors for the recurrence of pedophilic behavior include co-morbid psychiatric disorders as well as indications of substance abuse. Peter J. Fagan, Ph.D., Tomas N. Wise, M.D., Chester W. Schmidt, Jr., M.D. Fred S. Berlin, M.D., Ph.D., Pedophilia, 288 J. AM. MED. ASSOC. 2458, 2458 (2002). Specific psychiatric risk predictors of sexual offender recidivism include antisocial personality disorder or psychopathy and psychological symptoms such as low self-esteem and anger. R. Karl Hanson, What Do We Know About Sex Offender Risk Assessment?, 4 PSYCHOL. PUB. POL'Y L. 50, 57-59 (1998). However, while a history of general psychological symptoms is not predictive of long-term recidivism, "offenders appear to be at high risk when they are acutely symptomatic. . . ." Id. at 59. Logic dictates that the presence and level of severity of such dynamic psychiatric factors must be diagnosed and determined by a psychiatrist or psychologist by performing a mental status examination as well as whatever psychological testing the clinician believes will best assist in making such determinations. Eric S. Janus Robert A. Prentky, Forensic Use of Actuarial Risk Assessment with Sex Offenders: Accuracy, Admissibility and Accountability, 40 Am. Crim. L. Rev. 1443, 1446.
In the instant case, of course, this is exactly what Dr. Blumberg has done. The Court is not qualified to assess whether Mr. Thomas is acutely symptomatic with regard to either his dysthymia or his personality disorder, and neither is SSA Clemente. However, Dr. Blumberg, who examined Mr. Thomas and who considered all of the available information in making his diagnoses, is professionally qualified to make such determinations. It is for this reason that the Court finds that this clinical approach utilizing the customary interview techniques, etc. but within the framework of empirical studies of predictors of recidivism is a sufficiently reliable method, to analyze the level of risk posed by Mr. Thomas in the short term under the conditions of supervision.
6. The Court's Conclusion as to Dr. Blumberg's Opinion
As noted previously, the government did not challenge the reliability or relevancy of Dr. Blumberg's opinions underDaubert and its progeny. Therefore, the record regarding the bases for his opinion is perhaps less developed than it might have been had such a challenge been made. Nonetheless, while the record consequently does not indicate clearly whether the factors enumerated in Daubert were met, a review of the relevant literature discussed supra demonstrates that there is substantial support from a variety of sources, based in part on empirical data, for the approach taken by Dr. Blumberg that establishes its reliability. Moreover, the Court "has broad latitude to consider whatever factors bearing on validity that the court finds to be useful; the particular factors will depend upon the unique circumstances of the expert testimony involved."Westberry v. Gislaved Gummi AB, 178 F.3d 257, 261 (4th Cir. 1999) (citing Kumho Tire Co., 119 S.Ct. at 1175-76).
Thus, while the Court recognizes the difficulties inherent in extrapolating from clinical assessment, psychological testing, and actuarial data in order to make such risk assessments, the Court finds Dr. Blumberg's opinions reliable, as he appears to have reached them by using all of the appropriate professional tools and acumen reasonably at his disposal. Finally, the factors used by this Court to assess the reliability of Dr. Blumberg's opinion, which he was clearly qualified to render, have also been applied by the Fourth Circuit in evaluating the reliability of a forensic psychologist's opinion regarding a defendant's future dangerousness. United States v. Barnette, 211 F.3d 803, 816 (4th Cir. 2000).
E. Expert Law Enforcement Testimony 1. The Current State of the Law
Courts have admitted expert testimony by law enforcement professionals regarding the modus operandi of criminals in a variety of contexts. See, e.g., United States v. Brewer, 1 F.3d 1430 (4th Cir. 1993) (detective could testify as expert regarding significance of extensive phone traffic between defendant and members of alleged drug ring, as evidence showed that detective had specialized knowledge that would assist trier of fact in understanding evidence).
See also, e.g., United States v. Whitehead, 176 F.3d 1030 (8th Cir. 1999) (check kiting scheme); United States v. Brown, 7 F.3d 648, 652 (7th Cir. 1993) (drug trafficking);United States v. Robinson, 978 F.2d 1554 (10th Cir. 1992) (gang affiliation); United States v. Angiulo, 897 F.2d 1169 (1st Cir. 1990) (poker operation). But see United States v. Castillo, 924 F.2d 1227 (2d Cir. 1991) (admission of expert testimony regarding methods used by drug dealers was reversible error, where government used testimony as foundation for its theory that defendants' guilt could be inferred from behavior of unrelated persons).
More specifically, courts have also admitted expert testimony by SSA Clemente's predecessor at the FBI, Agent Kenneth Lanning, regarding the characteristics and methods of preferential child molesters and child pornographers. United States v. Romero, 189 F.3d 576 (7th Cir. 1999); United States v. Cross, 928 F.2d 1030 (11th Cir. 1991).
Additionally, one court has held that a search warrant for a defendant's residence and computer was supported by probable cause, based on the affidavit of a law enforcement agent containing, in addition to other information pertinent to the particular defendant, SSA Clemente's expert opinion that collectors of child pornography rarely disposed of images and typically maintained them at their residences. United States v. Coye, 2002 WL 31526542 (E.D.N.Y. 2002) (unpublished).
SSA Clemente's testimony and curriculum vitae reflect that he has offered related expert testimony in seven court proceedings and has submitted expert testimony in the form of affidavits in eleven cases. However, the Court's research has revealed that only two other opinions has been issued concerning his expert testimony, both of which are unpublished. See United States v. Coye, 2002 WL 31526542 (E.D.N.Y. 2002); Doyle v. Texas, 2004 WL 2714654 (Tex.App. 2004).
SSA Clemente's curriculum vitae reflects that he has provided expert affidavits in this regard for several hundred search warrant applications.
Significantly, the purposes for which expert testimony by members of the FBI's Behavioral Analysis Unit are offered, and indeed admitted, appear to be expanding. See Donald Q. Cochran,Alabama v. Clarence Simmons: FBI "Profiler" Testimony to Establish an Essential Element of Capital Murder, 23 LAW PSYCHOL. REV. 69 (1999) (hereinafter Cochran). The Simmons trial court admitted expert testimony of a member of the FBI's Behavioral Assessment Unit, based expressly on his expertise in criminal investigative analysis, to establish that a horrifically gruesome murder was committed specifically for sexual gratification, an essential element of the capital murder offense for which the defendant was tried and convicted. Id. Prosecutors successfully argued in Simmons, citing SSA Lanning's testimony regarding preferential child molesters inUnited States v. Cross, 928 F.2d 1030 (11th Cir. 1991), that the BAU agent's testimony should be admitted because it also involved the interpretation of "bizarre and deviant behavior that was unlikely to be within the knowledge of ordinary citizens" and because the agent's experience and training qualified him as an expert in criminal investigative analysis. Id. at 84. The Court of Criminal Appeals of Alabama upheld the trial court's admission of the BAU agent's testimony. Clarence Leland Simmons v. State, 797 So.2d 1134 (Ct.Crim.App. Al. 2000). While the court acknowledged that the Supreme Court held in Kumho Tire Co. that the Daubert factors may apply to the testimony of experts who are not scientists, it also noted that a trial court should only consider the specific factors identified in Daubert where they are reasonable measures of an expert's reliability. Id. at 1153. The court found that the BAU agent's testimony established that victimology and crime-scene analysis, as subsets of criminal investigative analysis, constituted reliable and specialized knowledge under Alabama Rule of Evidence 702, which is identical to Fed.R.Evid. 702. Id. at 1154. The court held that the BAU agent's testimony was reliable, and that it satisfied theDaubert factors, because it was based on published research, interviews, case studies, and investigation of similar crimes.Id. at 1154-56. Therefore, the court concluded that crime-scene analysis and victimology were reliable fields of specialized knowledge and that, based upon his studies and experiences in those fields, he was an expert. Id. at 1156.
The author distinguished criminal investigative analysis from profiling as follows:
Criminal investigative analysis involves a detailed review of all aspects of a particular crime, which may have been committed by either a known or unknown offender. Profiling, on the other hand, is an analysis of a crime or series of crimes committed by an unknown offender which results in a detailed description of the type of person who would have done such a crime or series of crimes. This "profile" of the unknown offender is designed to be used by investigators to assist in catching the offender. As the offender in this case . . . was already known, the case involved the use of Criminal Investigative Analysis, not true "profiling."
Donald Q. Cochran, Alabama v. Clarence Simmons: FBI "Profiler" Testimony to Establish an Essential Element of Capital Murder, 23 LAW PSYCHOL. REV. 69, n. 139 (1999) (internal citations omitted).
The author, who prosecuted the case, noted that "[b]ecause [the FBI's BAU expert] did not link the defendant to the murder in any way, but rather testified only that whoever killed [the adult female victim] did so with a sexual motivation, his testimony did not violate the Alabama prohibition against expert testimony on the `ultimate issue.'" Id. at n. 147.
In Nenno v. State, the trial court admitted SSA Lanning's expert testimony regarding preferential child molesters, based on a hypothetical question about the defendant and the crime, that the defendant would be "difficult to rehabilitate" and that he "would be an extreme threat to society and especially children within his age preference." 970 S.W.2d 549, 552, 560, 561-62 (Tex.Crim.App. 1998), overruled on other grounds by State v. Terrazas, 4 S.W.3d 720 (Tex.Crim.App. 1999) (expert testimony regarding future dangerousness as an aggravating factor in a capital case). The defendant unsuccessfully objected to SSA Lanning's testimony as unreliable on the grounds that there was no evidence that (1) "the theories underlying Lanning's testimony are accepted as valid by the relevant scientific community"; (2) "the alleged literature on the theories supports his theories"; (3) "there are specific data or published articles regarding the area of future dangerousness of prison inmates"; (4) "his theories have been empirically tested"; (5) "he has conducted any studies or independent research in the area of future dangerousness"; or (6) "anyone else has tested or evaluated the theories upon which his testimony was based." Id. The court concluded that, while Daubert's gatekeeper rationale applied to all expert testimony, the Daubert factors did not necessarily apply outside of the hard science context, and that in such cases methods of proving reliability would vary, depending upon the field of expertise.
2. BAU Agents' Expert Opinions Must Be Reliable
In light of this expanding role, and the considerable authority and even mystique in which these experts are cloaked due to their status within the FBI, it is essential to establish the reliability of such expert testimony. See Cochran at 89 (noting that "[b]ecause of the status of the FBI's Profiling and Behavioral Assessment Unit as the only organization in the world that specializes in the investigation of bizarre and brutal crimes, testimony by members of the unit will always be powerful evidence") (emphasis added). See also D. Michael Risinger Jeffrey L. Loop, Three Card Monte, Monty Hall, Modus Operandi "Offender Profiling": Some Lessons of Modern Cognitive Science For the Law of Evidence, 24 CARDOZO L. REV. 193, 251 (2002) (opining that questionable profiling data at issue in the article could have been easily validated or invalidated for the simple reason that the FBI had access to data that would settle the issue of raw accuracy and that the FBI should not benefit from "their own failure to aid the generation of defensible data"); Henry F. Fradella, Adam Fogarty Lauren O'Neill, The Impact of Daubert on the Admissibility of Behavioral Science Testimony, 30 PEPP. L. REV. 403, 444 (2003) (noting that "[t]here appears to be only one area in which Daubert is not being rigorously applied to behavioral science testimony . . . [which is] `expert opinions' offered by law enforcement officers based on their years of experience in the field when they offer opinions with regard to modus operandi or other aspects of the . . . criminal mind" and that "[e]xplorations into their theoretical knowledge base, as well as the validity and reliability of both their methodologies and their conclusions, appear to have escaped Daubert review").
3. SSA Clemente's Opinion Lacked Reliability
(a) Lack of Reliability
The Court recognizes that SSA Clemente has certain expertise in criminal investigative analysis regarding the characteristics and behavioral patterns of child sex offenders, based on specialized knowledge, acquired primarily through experience and anecdotal information. In contrast to Dr. Blumberg's opinion, however, the Court finds SSA Clemente's methodology in reaching his conclusion that there was a "high" risk that Mr. Thomas would re-offend if released pending trial insufficiently reliable.
However, expertise in criminal investigative analysis regarding sexual offenses against children does not, based on the evidence presented, equate necessarily to expertise in risk assessment. Shreve v. Sears, Roebuck Co., 166 F.Supp.2d 378, 391 (D.Md. 2001 (finding that "the fact that a proposed witness is an expert in one area, does not ipso facto qualify him to testify as an expert in all related areas"). See also Gross v. King David Bistro, 83 F.Supp.2d 597, 600-01 (D.Md. 2000) (determining that because the proposed testimony was unreliable, the proposed expert's experience and training were "immaterial").
SSA Clemente's testimony in this case went well beyond the substance of the testimony offered by his predecessor, Agent Lanning, in United States v. Romero, 189 F.3d 576 (7th Cir. 1999), as well as the purposes for which it was admitted in that case. Romero has been cited by the government as persuasive authority in this matter.
SSA Clemente's testimony patently demonstrated that theDaubert factors were not satisfied, with the arguably sole exception of general acceptance within the law enforcement community. The Supreme Court, however, expressly rejected theFrye "general acceptance" standard for scientific evidence, and replaced it with the two-pronged gatekeeping test set forth byDaubert. 509 U.S. at 597, 113 S.Ct. 2786. Kumho Tire Co., of course, later extended Daubert to all expert evidence. 509 U.S. at 584-89, 113 S.Ct. 2786.
First, SSA Clemente was unable to demonstrate that his risk assessment methodology had been (or could be) tested. When asked by the Court whether his methodology had been tested or validated, SSA Clemente, in a series of answers largely unresponsive to the Court's inquiry, asserted that his methodology was "outside of scientific analysis" and failed to identify anything that could be even remotely construed as either testing or validation. (Tr. 75-78, Apr. 28).
SSA Clemente characterized his opinion as a risk assessment, testifying that he "[did] not predict behavior" but rather "evaluate[d] the risk based on . . . criminal investigative analysis . . . [and] provide[d] an opinion as to the level of risk [of] reoffending as it relates to this person and a risk to the community." (Tr. 53, Apr. 28).
Moreover, the typology of a preferential sex offender, to which SSA Clemente repeatedly referred as the foundation of his analysis, apparently is based entirely on anecdotal case studies and interviews. SSA Clemente was unable to offer even any retrospective studies establishing the validity of this typology. (Tr. 82-83, Apr. 28). SSA Clemente admitted that the typology of a preferential sex offender, which is the essential predicate to his risk assessment methodology, as well as re-offense risk assessment regarding such offenders, requires additional research to achieve validation. (Tr. 80, Apr. 28). SSA Clemente therefore acknowledged that his opinions were based on anecdotal case studies and interviews, albeit "a huge amount", and lacked an empirical basis. (Tr. 83, Apr. 28).
A typology is defined as "the study of types, as in a systematic classification." WEBSTER'S II NEW RIVERSIDE DICTIONARY, 1249 (Riverside Publishing Co., 1994).
See KENNETH V. LANNING, CHILD MOLESTERS: A BEHAVIORAL ANALYSIS 26-30 (National Center for Missing Exploited Children, 2001) (describing the typology of a preferential sex offender). The typology developed and described by Mr. Lanning, characterized by four hallmarks, is substantively identical to that described in both SSA Clemente's declaration and his testimony, as discussed supra. See id.
SSA Clemente testified that his risk assessment methodology, which he described as "one aspect of criminal investigative analysis", is based "on largely anecdotal studies of cases [because] it's very difficult to conduct empirical research in this area." (Tr. 71-72, Apr. 28). He later reiterated that it was "based on years of interview[s]." (Tr. 77, Apr. 28). Finally, in answer to a series of questions regarding the basis for his risk assessment, SSA Clemente responded: "Now, have we published studies indicating [that people who meet the four hallmarks of a preferential sex offender do re-offend]." No, we haven't. We're in the process of doing probably the first long-term empirical study on this area." (Tr. 82, Apr. 28). See also KENNETH V. LANNING, CHILD MOLESTERS: A BEHAVIORAL ANALYSIS 1 (National Center for Missing Exploited Children, 2001) (acknowledging that the preferential sex offender typology is based on anecdotal information and "the totality of [the author's] acquired knowledge and expertise").
THE COURT: But even a retrospective study, going back and looking at offenders and then looking back at their characteristics, have you published anything that looked at say 100 . . . re-offenders, identified their characteristics and looked at it that way, retrospectively, and published it?
THE WITNESS: Yeah. I have not. One of the things that I did when I joined the unit was make a push towards empirical studies . . . I'm an attorney, and I know the relative weight that can be applied to anecdotal versus empirical. We had a huge amount of anecdotal information. We didn't have empirical. (Tr. 82-83, Apr. 28).
SSA Clemente testified that the Behavioral Analysis Unit is associated with a longitudinal study, with over 1,000 subjects, that is ongoing. (Tr. 80, Apr. 28). However, the results of that study are not yet available, and thus cannot assist the Court in this matter. Additionally, SSA Clemente testified that the results, analysis, and typology of a preferential sex offender, as promulgated by the BAU, had been published in the Journal of the American Medical Association ("JAMA"). (Tr. 83, Apr. 28). However, the Court's research revealed that a sole JAMA article stated merely that "[r]esearchers in a child sex offender program based in Seattle, Washing[ton], have provided a qualitative study of the attitudes and modus operandi of men who have sexually abused children." Peter J. Fagan, Ph.D., Tomas N. Wise, M.D., Chester W. Schmidt, Jr., M.D. Fred S. Berlin, M.D., Ph.D., Pedophilia, 288 J. AM. MED. ASSOC. 2458, 2460 (2002) (citing Conte JR, Wolf S., Smith T., What Sexual Offenders Tell Us About Prevention Strategies, 13 CHILD ABUSE NEGL. 293-301). The Court was unable to obtain the study cited by the JAMA article, neither of which was offered or even specifically identified by SSA Clemente. However, while the cited reference indicates that the study addressed the "attitudes and modus operandi" of certain abusers, it does not reveal whether it in fact validated the BAU's typology for preferential sex offenders.
While the Court recognizes, as SSA Clemente testified, that collecting empirical data is difficult in this context, this difficulty cannot, by itself, render a risk assessment methodology reliable or exempt it from any sort of testing or validation. Thus, the first Daubert factor was not met.
Second, no specific literature was offered demonstrating that his methodology has been subjected to meaningful peer review analysis. While SSA Clemente testified that the typology of a preferential sex offender has been referenced in a number of publications outside the field of criminal investigative analysis, no studies or publications were offered in which SSA Clemente's peers, that is, other criminal investigative analysts, analyzed or validated the typology. (Tr. 84, Apr. 28). Additionally, while SSA Clemente described a process wherein some oversight is provided to BAU staff by an advisory and research board, it does not appear to comprise meaningful peer review in the present legal context. (Tr. 60, Apr. 28). See United States v. Horn, 185 F.Supp.2d 530, 555 (D.Md. 2002) (explaining that "peer review as contemplated by Daubert andKumho Tire must involve critical analysis that can expose any weaknesses in the methodology or principles underlying the conclusions being reviewed").
SSA Clemente testified that the typology was "accepted" by the International Fellowship of Criminal Investigative Analysis as "a good model for analyzing for criminal purposes the behavior of child sex offenders." (Tr. 84, Apr. 28). However, with no other evidence beyond SSA Clemente's assertion, it does not constitute meaningful peer review. See United States v. Horn, 185 F.Supp.2d 530, 555 (D.Md. 2002).
Third, SSA Clemente was likewise unable to explain what, if any, error rate applied to his risk assessment technique, which he repeatedly referred to as one aspect of criminal investigative analysis. On cross-examination, the following colloquy transpired:
Q: Does your office have any sort of program or procedure in place to assess cases over time to determine whether there is an error rate that applies to your analysis?
A: Absolutely. Our office is set up like a a think tank. We do consultations with a board . . . [a]nd basically, when you're going to testify as an expert, you have to present the testimony you're going to give to this committee and they will approve or disapprove it based on our track record over the years. . . .
Q: What do you meant by your track record? Track record of what?
A: On consulting of cases, of giving expert testimony, the accuracy of it and so forth.
Q: Well, by track record are you saying that in 1990 you gave an opinion that a person was a preferential sex offender who had been actively molesting and you tracked that years later to see if that was true or not?
A: Well, for example — sure. You go through the whole case. The case is not close —
Q: And you have never been wrong?
A: That first —
Q: Other than that example you gave me?
A: That he was a preferential offender?
Q: Yes. That sort of thing. Is there any error control over those sorts of opinions, that a person is an offender who is an active molester, and it turns out years later that he is not? Is there any way to pick that up?
A: Well, the only way I would make that determination is if there were evidence that would amount to that determination. I have never — well, I will be able to tell one case.
(Tr. 58-60, Apr. 28) (emphasis added). Obviously, there is no known error rate regarding the application of the required predicate of his risk assessment methodology, that is, the determination of whether a defendant is a preferential sex offender. Thus, the third Daubert factor was not met.
Finally, as the foregoing discussion demonstrates, there is minimal, if any, evidence that there are any standards controlling the operation of SSA Clemente's methodology. While there is apparently some oversight exercised by the board previously described, SSA Clemente failed to identify any measures that might properly be termed "standards" in this context. Thus, the fourth Daubert factor was also not met.
Certainly, the precise factors articulated by Daubert need not be applied to every expert in every case. See Kumho Tire Co., 119 S.Ct. at 150. However, Daubert and Kumho, in conjunction with FED. R. EVID. 702, mandate that the Court's gatekeeping function must be fulfilled by inquiring into factors that fairly assess reliability. SSA Clemente claimed that his area of expertise and methodology were not "scientific" and that while analogies could be made to the scientific method, it should not be strictly applied to his testimony. (Tr. 77, Apr. 28).
Many social scientists, as SSA Clemente has done, rely primarily on real-world experience to arrive at their conclusions. As one court found, prior to Kumho Tire Co., while the Daubert factors "must be applied when an expert bases his testimony on scientific hypotheses which are capable of being refuted by controlled experimentation . . . [they] may be applied in differing degrees when it comes to non-Newtonian science or other specialized knowledge." United States v. Hall, 974 F.Supp. 1198, 1202 (C.D.Ill. 1997) (internal quotation marks omitted). Nonetheless, the court emphasized that "there must be some degree of reliability of the expert and the methods by which he has arrived at his conclusions." Id. (emphasis added). The court explained that expert social science testimony based on "real-world experience rather than experimentation" must meet certain quantitative and qualitative requirements, including threshold number of experiences, sufficient similarity of those experiences "in nature to form a valid basis for comparison", publication in scholarly journals, and meaningful peer review.Id. at 1202-03. However, as demonstrated previously, SSA Clemente's assessment, with the exception of number of experiences, did not even meet the less rigorous factors sometimes applied to `soft' social sciences in determining reliability.
(b) The Hernandez Study: Lack of `Fit'
In his testimony, SSA Clemente referred to and relied on the Hernandez study for his view on Mr. Thomas' recidivism. SSA Clemente noted that this study found that 54 of 62 inmates who were convicted for the mere possession of child pornography admitted, during the course of the study, which was conducted while they were incarcerated for the child pornography charges, that each of them had actually committed approximately 30 instances of molestation which had been undetected. Thus, SSA Clemente used the Hernandez Study to argue that Mr. Thomas, merely on the basis that he possessed a collection of child pornography, was at a high risk of re-offending, and at a higher risk than the recidivism statistics would indicate, if he were released under the proposed conditions pending trial. The Hernandez Study is so dissimilar to the instant facts that it did not `fit' the purpose for which it was used by SSA Clemente.See United States v. Horn, 185 F.Supp.2d 530, 553 (D.Md. 2002) (explaining that factors establishing reliability and relevance must "fit" the case at issue and that careful examination can "expose evidentiary weaknesses that otherwise would be overlooked"). While SSA Clemente testified that sexual offender recidivism statistics are inaccurate because many such crimes remain undetected, it is speculative to conclude that Mr. Thomas is likely to molest a child, under the proposed strict conditions, during the approximately six week period pending trial.
See GEORGE B. PALERMO, M.D., MARY ANN FARKAS, PH.D., THE DILEMMA OF THE SEXUAL OFFENDER 171-72 (2001). The authors state:
Id. (internal citations omitted).
Sex offenders are portrayed as more likely to reoffend than other types of offenders. The offender is characterized as one who repeats his sexual conduct or is likely to repeat it . . . [y]et research is inconclusive regarding the recidivism rates of sex offenders as compared to other offenders even though improvements have been made using actuarial methods. There is no clear evidence that sex offenders are any more likely to recidivate than any other type of offender and there is no empirical basis to assess which sex offenders present the most immediate risk for reoffending.
The Court notes that neither expert quantified his risk assessment. The relevant literature reflects widely divergent recidivism statistics for sexual offenders, depending on numerous factors. One recent article describes the average sex offender recidivism rate as 10% to 17% after four to five years when untreated. Peter J. Fagan, Ph.D., Tomas N. Wise, M.D., Chester W. Schmidt, Jr., M.D. Fred S. Berlin, M.D., Ph.D., Pedophilia, 288 J. AM. MED. ASSOC. 2458, 2463 (2002). The authors also acknowledge, however, that the actual incidence of pedophilic behaviors is probably under-reported. Id. However, another expert concludes that "even with long follow-up periods [in excess of twenty years] and careful record searches, the average sexual offense recidivism rate rarely exceeds 40%." R. Karl Hanson, What Do We Know About Sex Offender Risk Assessment?, 4 PSYCHOL. PUB. POL'Y L. 50, 67 (1998). See also R. Karl Hanson, Who is Dangerous When are they Safe? Risk Assessment with Sexual Offenders, in PROTECTING SOCIETY FROM SEXUALLY DANGEROUS OFFENDERS 63, 64-65 (Bruce J. Winick John Q. La Fond eds., Am. Psych. Assoc., 2003). Cf. John M. Fabian, Psy.D.,Kansas v. Hendricks, Crane Beyond: "Mental Abnormality," "Sexual Dangerousness": Volitional vs. Emotional Abnormality The Debate Between Community Safety Civil Liberties, 29 WM. MITCHELL L. REV. 1367, 1426-32 (2003) (citing a 1997 study that indicated a 52% recidivism rate for extrafamilial child molesters over a 25 year period). Whatever the true recidivism rates, the results of the recidivism studies demonstrate that the longer the follow-up period, the higher the rates of recidivism. R. Karl Hanson, What Do We Know About Sex Offender Risk Assessment?, 4 PSYCHOL. PUB. POL'Y L. 50, 55 (1998). Thus, it can be conversely inferred that the shorter the period of time at issue, the lower the rate of recidivism will be. Id.
SSA Clemente raised the Hernandez Study in making the point that recidivism statistics are understated because many instances of molestation are never reported to, or detected by, the justice system. While that may indeed be so, the issue is the extent of that understatement in the available recidivism statistics, and how such understatement affects re-offense risk assessment in the context of the typical period of time pending trial, which is generally a few months and certainly less than a year.
SSA Clemente's approach in no way accounted for, or even acknowledged, how the proposed conditions of release in this case might mitigate such a risk. Notably, the Hernandez Study subjects reported having committed molestation offenses when they were under no such strict supervision or conditions as those proposed in the instant case. Nonetheless, SSA Clemente maintained that, because in his opinion Mr. Thomas was at a "high" risk for reoffending pending trial, he should be detained, without explaining why the proposed conditions would or could not mitigate the alleged risk. Therefore, the Court finds that the Hernandez Study simply does not fit the facts of this case, and thus adversely affects the reliability of SSA Clemente's risk assessment. See Gross v. King David Bistro, 83 F.Supp.2d 597, 600-01 (D.Md. 2000) (finding a proposed expert's testimony unreliable because the empirical data cited by the proposed expert was "simply too nascent and tepid to support his conclusion").
(c) Unreliable Application of Principles and Methods
SSA Clemente based his opinion on what he characterized as a "pattern" of past behavior. However, the Court is not convinced that collecting child pornography and the sole incident of molestation in evidence in this case equate to the pattern asserted. The Court, as stated earlier, in no way minimizes the seriousness of the conduct underlying the molestation charge. However, there was no evidence whatsoever of any other acts of molestation, and thus this conduct may, on Mr. Thomas's part, indeed have been an aberration, as he argued.
Notably, a recent publication's authors concluded, following an exhaustive review of the empirical issues relevant to various tests used to determine the dangerousness, antisociality and recidivism risk of sexual offenders, that "it remains an empirical question as to whether all individuals who access, view, and download child pornography, eventually escalate into contacting children and molesting them." GEORGE B. PALERMO, M.D., MARY ANN FARKAS, PH.D., THE DILEMMA OF THE SEXUAL OFFENDER 66-67 (2001) (citing a personal communication with Anthony J. Pinizzotto, Ph.D., of the FBI Academy, for this assertion).
Moreover, while the Court fully appreciates the significance of the sophisticated techniques and grooming behavior used by some serial child molesters to gain access to children, the government's argument that, because Mr. Thomas molested the girl in the video, his friendship with her mother and his occasional babysitting of the victim and her two siblings ipso facto constituted such techniques and grooming. Thus, the Court is not convinced, even applying the typology of a preferential sex offender as defined by SSA Clemente, that Mr. Thomas's behavior actually comprises such grooming and techniques, or that the totality of his conduct, given the evidence to date, can accurately be assessed as the typology of a preferential sex offender.
Additionally, SSA Clemente seemingly failed to apply the actual conclusions derived from actuarial data he purported to use. SSA Clemente claimed that Mr. Thomas's risk of reoffense was higher because his alleged molestation involved an extrafamilial child. Dr. Blumberg, however, testified that Mr. Thomas's risk of reoffense was lower because the alleged victim was a girl, as opposed to a boy. There is support in the relevant literature for both statements. See, e.g., Grant T. Harris, Marnie E. Rice Vernon L. Quinsey, Appraisal Management of Risk in Sexual Aggressors: Implications For Criminal Justice Policy, 4 PSYCHOL. PUB. POL'Y L. 73, 75 (1998). However, the literature also reflects that heterosexual extrafamilial child molesters, such as Mr. Thomas, exhibit an intermediate recidivism rate, while homosexual child molesters have the highest recidivism rate.Id. SSA Clemente conceded that the actuarial factors placed Mr. Thomas, according to unidentified studies, in the intermediate rather than the highest risk group. (Tr. 90, Apr. 28). He nonetheless asserted, based on no more than his "anecdotal experience", that certain characteristics of defendant's pornography collection "tend[ed] to reduce that mitigation a little bit." Id.
The Court recognizes that "[i]n applying Daubert, a court evaluates the methodology or reasoning that the proffered . . . expert uses to reach his conclusion . . . rather than the conclusion itself." TFWS, Inc. v. Schaefer, 325 F.3d 234, 240 (4th Cir. 2003). However, Rule 702 also provides that "a witness qualified as an expert . . . may testify thereto in the form of an opinion [only if] the witness has applied the principles and methods reliably to the facts of the case." FED. R. EVID. 702 (emphasis added). Thus, the Court must examine SSA Clemente's application of his methodology to the facts, at least to some degree, to assess properly the reliability of his opinion. See Id. 4. The Court's Conclusion as to SSA Clemente's Opinion
In Daubert, the Supreme Court held that Federal Rule 702 required that scientific evidence be "not only relevant, but reliable." 509 U.S. 579, 113 S.Ct. 2786, 2795, 125 L.Ed.2d 469, 481 (1993). In determining reliability, the Supreme Court provided four factors that could be used, while noting that the analysis should be flexible. Daubert, 509 U.S. at 594, 113 S.Ct. 2786. In Kumho Tire Co. Ltd. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999), the Court held that the principles enunciated in Daubert applied to all expert testimony, not merely scientific evidence. Although experts with specialized knowledge may extrapolate from existing data, "nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert." Kumho Tire, 526 U.S. at 157 (finding no abuse of discretion in rejecting opinion of expert) (internal citation and quotation marks omitted) (emphasis added). Clearly, Daubert andKumho Tire Co., as well as Rule 702, require that expert opinion evidence be connected to existing data by something more than a chain of dubious inferences that amount to an expert's assertion that "it is so because I say it is so."49 Id.
Failure to satisfy four out of the five Daubert factors, and most of the less rigorous criteria sometimes applied in the social science context, combined with a lack of any other persuasive indicia of reliability, forced the Court to conclude that the principles and methods underlying SSA Clemente's opinions were insufficiently reliable. Moreover, even assuming the reliability of SSA Clemente's methodology, that is, application of the preferential sex offender typology by a non-examining criminal investigative analyst combined with actuarial data, his methodology was not well supported in this case.
In summary, while SSA Clemente repeatedly stated that he could not predict what this defendant would do if released, he nevertheless insinuated that because there is admittedly very strong evidence that the defendant committed one act of molestation, that he most likely has committed others and will undoubtedly commit more. This is circular reasoning at best, and without more, cannot serve as clear and convincing evidence of a reoffense risk in a detention determination.
Although SSA Clemente has strong law enforcement and criminal investigation credentials, they cannot serve as a substitute for a proven, reliable method or technique that can be independently verified when it comes to such risk assessment. As Daubert and its progeny make clear, while experts are not required to beright, their testimony must be based on methods and techniques that constitute a reliable foundation. Moreover, even if SSA Clemente's expert opinion in this regard could be fairly assessed as reliable, it would be in equipoise with Dr. Blumberg's opinion. Thus, even if the Court had found SSA Clemente's conclusions reliable, the government would still not have carried its burden in this case.
Finally, it appears that the risk assessment methodology applied by Dr. Blumberg is preferable for two reasons. First, as discussed supra, the presence or absence, as well as the severity level, of certain mental disorders can impede or enhance a defendant's ability to control his pedophilic impulses or tendencies. Therefore, a risk assessment made in a similar context that does not evaluate or consider such factors is likely to be unreliable. Second, a risk assessment methodology that might result in the deprivation of liberty should incorporate an individualized and complete assessment of the defendant. Cf. Flores v. Johnson, 210 F.3d 456, 460 (5th Cir. 2000) (requiring individualized decision in death penalty context). Therefore, it appears that clinical assessment, which can only be conducted by a mental health professional, considered in the context of available actuarial data, is the more reliable and appropriate risk assessment methodology as compared to that employed by SSA Clemente.
In sum, the factors enunciated in the Bail Reform Act demonstrated that the conditions imposed concerning the release of the defendant in this case reasonably assured the safety of the community. The Court did not "roll the dice" by releasing Mr. Thomas under the proposed conditions, as the government has claimed. It is imperative in such cases that courts evaluate the evidence presented and apply the law, rather than allowing natural revulsion to overcome objective analysis. The Bail Reform Act requires a reasonable assurance, not a guarantee, of safety. United States v. Barnett, 986 F.Supp. 385, 400 (W.D. La. 1997). Moreover, the probability and consequences of the defendant's prospective acts must be balanced against the proposed immediate restraint on the defendant's liberty, as he is presumptively innocent. See 18 U.S.C. § 3142(j). Finally, the government failed to show by clear and convincing evidence that there was no combination of conditions that can reasonably assure the safety of the community. Therefore, while the Court imposed the conditions of release that it found reasonably assured the safety of the community, as specified by separate order, the government's motion for detention was denied.