Abstract
The confession of a criminal defendant serves as a prosecutor's most compelling piece of evidence during trial. Courts must preserve a defendant's constitutional right to a fair trial while upholding the judicial interests of presenting competent and reliable evidence to the jury. When a defendant seeks to challenge the validity of that confession through expert testimony, the prosecution often contests the admissibility of the expert's opinion. Depending on the content and methodology of the expert's opinion, testimony addressing the phenomenon of false confessions may or may not be admissible. This article outlines the scientific and epistemological bases of expert testimony on false confession, notes the obstacles facing its admissibility, and provides guidance to the expert in formulating opinions that will reach the judge or jury. We review the 2006 New Jersey Superior Court decision in State of New Jersey v. George King to illustrate what is involved in the admissibility of false-confession testimony and use the case as a starting point in developing a best-practice approach to working in this area.
Custodial confessions, in the absence of physical evidence, can be the key to a criminal prosecution. Many criminal suspects, for a variety of reasons, give false or unreliable confessions and challenge their admissibility later at trial.1 Courts have been receptive to such challenges on two bases: the voluntariness of the waiver of rights and the reliability of the content of an alleged false confession. In each instance, the suspect's mental state is placed at issue before the court. Courts are aware of the phenomenon of false confessions and may look to expert witnesses to shed light on a defendant's claim. In this article, we examine the phenomenon of false confessions, explore epistemological and practical concerns regarding the admissibility of expert testimony, and recommend ways to structure and present such testimony.
The Phenomenon of False Confessions
By now, there is little doubt that innocent criminal suspects give incriminating statements. The social science literature suggests that the interrogator-suspect interaction, coupled with the suspect's psychological makeup, explains how an individual can act against his best interest.2–5 These considerations are not novel. Indeed, various dynamics surrounding the quality of confessions have been considered as early as the mid-19th century.6 For example, Wharton7 outlined the following points to bear in mind when considering the validity of a confession: Before, however, a confession should be taken as real, it should be subjected to certain psychological tests. Delusion; a morbid desire to attract attention; a sort of epidemic which sometimes strikes down whole classes with a passionate impulse to insist upon some blood-stain on the conscience, something like the hypochondriac epidemic impulse which insists upon some personal abnormity [footnote omitted]; weariness of life; a propensity to self-destruction through a channel which from its very tortuousness possesses its own fascination; a Lara-like [Count Lara of Lord Byron's 1814 poem] desire to appear mysterious and dark…the existence of such elements as these should be inquired into before a confession is received as absolute [Ref. 7, p 16].
The notion of false confessions, however, may be counterintuitive to the average person, the potential juror. What else but guilt would cause any person to confess? This is the quandary in which criminal defendants find themselves when claiming to have falsely incriminated themselves. Nevertheless, it has been reported that numerous convicts who were subsequently exonerated from their alleged crimes initially provided confessions that supported their wrongful convictions.8
For defendants seeking to challenge the validity of their own words at trial, expert testimony may be essential in casting doubt on the customary view of confessions in the eyes of the jury. This task assumes additional complexity when a defendant has provided a false confession voluntarily in the absence of police misconduct. As Alschuler9 points out, there are no constitutional provisions for excluding legally obtained confessions, and accordingly, the jury must be allowed to weigh such confessions as it does other forms of evidence. Defendants therefore need the assistance of expert testimony to subvert the potent inference of guilt that confessions generate.
In some jurisdictions, defendants may proffer expert testimony to educate the judge or jury as to how a psychiatric condition in the suspect might explain a false confession. It is not enough, however, to testify that false confessions occur in various circumstances. Rather, testimony must be clinically based and dynamically related to the confession. A 2005 article in The Journal called attention to the necessity that expert witnesses know the legal standards underlying the admissibility of testimony.10 Legal scholars have conducted comprehensive reviews of the federal and state case law regarding the judicial treatment of expert testimony on false confessions.11,12 As we discuss later, the admissibility of testimony depends on its helpfulness to the trier of fact and on its scientific basis. The following case, taken from an appellate court in New Jersey and decided subsequent to these case law surveys, encompasses many of the judicial concerns expressed regarding the treatment of expert testimony in the context of false confession claims. It is illustrative of the ongoing struggle that courts face as they manage the interplay between the boundaries of expert testimony and the constitutional requirement that a criminal defendant receive a fair trial.
State of New Jersey v. George King
On March 13, 2003, George King was arrested in Newark, New Jersey, for the murder of Edna Ryan.13 During the interrogation, Mr. King confessed to Ryan's murder, adding, “I might as well tell you about Woodbridge,” referring to the unsolved murder of Meifang Rush in Woodbridge, New Jersey, two months earlier. On the basis of his confessions, he was charged with both murders.
He pleaded guilty to aggravated manslaughter of Ryan, but claimed that his confession to Rush's murder was fabricated. He claimed that he had also attempted to confess falsely to several other murders during his interrogation. Police investigators involved in his interrogation claimed that, although he indicated that he possessed information about other unsolved crimes, he confessed only to the murders of Ryan and Rush. He did not assert that his confession was coerced.
Expert Testimony
Mr. King's defense proffered psychiatric testimony to support the false-confession claim. Dr. Roger Harris, a forensic psychiatrist, concluded that Mr. King had narcissistic and antisocial personality disorders with borderline personality traits, making him predisposed to false confessions. He had exaggerated his criminal history to mental health professionals. Harris concluded that Mr. King's grandiosity and need for admiration were the dynamics of the false confessions.
The state moved to exclude the expert testimony regarding Mr. King's personality disorders, asserting that it was not relevant or admissible, because Harris did not provide a scientific basis linking personality disorders to false confessions. In addressing the state's motion, the trial court stated: [B]y human nature, individuals will neither assert, concede, nor admit to facts that would affect them unfavorably. Consequently, statements that so disserve the declarant are deemed inherently trustworthy and reliable…. Dr. Harris offers a reason the Defendant's confession might be false to contrast the usual supposition that confessions, like statements against interests generally, would not be made if they were not true… [Ref. 13, p 816].
In that light, the trial court found that Harris could testify that Mr. King's personality disorders were consistent with his false-confession claim, but restricted him from testifying about “anything the defendant or anyone else told him about the circumstances surrounding the giving of the confession…” (Ref. 13, p 811). Both the defense and prosecution appealed this ambivalent ruling. The state's appeal challenged the admissibility of Harris's testimony, arguing that Mr. King's personality disorders were immaterial to the alleged false confession. The defense appealed the limitation imposed on Harris's testimony.
Appeal and Opinion
The Superior Court of New Jersey, Appellate Division, reviewed the matter and ruled that Harris's testimony regarding Mr. King's personality disorders was admissible. In addressing the state's appeal, the court initially looked to the U.S. Supreme Court decision in Crane v. Kentucky.14 In that case, a criminal defendant appealed his conviction, arguing that the lower court erred in excluding expert testimony that the duration and nature of the interrogation process brought about a false confession. In reversing the conviction, the Crane Court pointed out that under the Sixth and Fourteenth Amendments to the Constitution, criminal defendants are entitled to provide juries with reliable and competent evidence addressing the credibility of their confessions, especially when the confession is central to the innocence plea and there is no valid state justification for excluding it. The Court held that the physical and psychological environment in which the confession was obtained was relevant to the defendant's claim of innocence.
Scope of Testimony
With Crane in mind, the New Jersey appellate court reviewed cases in other jurisdictions, extending Crane beyond the objective interrogation environment to the subjective psychological makeup of defendants. The appellate court cited People v. Hamilton,15 a Michigan Court of Appeals case that reversed the murder conviction of a defendant after he was barred from presenting expert testimony about how his psychological makeup affected the reliability of his statements to police. The Michigan court in Hamilton, considering Crane, found that although Crane dealt with the external interrogation environment, its principle was equally applicable to the psychological makeup of the defendant. As such, the defendant's psychology was considered relevant to the reliability of his confession.
Persuaded by Hamilton and similar holdings in other jurisdictions, the New Jersey appellate court found that the testimony regarding Mr. King's personality disorders in relation to the reliability of his confession was relevant. Since Mr. King's confession was “the linchpin of the State's case,” the court held that such testimony was necessary to assist the fact finder in assessing the reliability of that evidence. The court, however, pointed out that this testimony could be limited, thus adhering to the judicial interests of presenting competent and reliable evidence to the jury.
Admissibility
In keeping with those judicial interests, the appellate court focused principally on whether the scientific reliability requirement of New Jersey Rule of Evidence 702 was satisfied. The state argued that since Harris did not offer an authoritative source or study linking Mr. King's personality disorders to false confessions, Harris's testimony was not scientifically reliable. For that reason, the state asserted that the trial court erred in admitting the testimony under the general-acceptance standard in Frye v. United States.16 The appellate court, however, disagreed with the state's view.
The court acknowledged that the “general acceptance standard by the relevant scientific community” established in Frye determined the scientific reliability of expert testimony in criminal cases in New Jersey. The court indicated that this standard could be established by expert testimony, authoritative scientific literature, or persuasive judicial opinions. According to the court, all of these methods were satisfied in King.
The appellate court contrasted the instant case with its earlier decision in State v. Free.17 In Free, the Superior Court of New Jersey reversed a trial court's decision to admit social psychologist Kassin's testimony regarding police interrogation techniques challenging the reliability of the defendant's confession. In reversing the decision, the appellate court first found that the trial court erred in applying the Daubert18 standard to the proffered testimony instead of the Frye standard. In New Jersey, the admissibility standard for expert testimony in criminal cases is the Frye test, whereas the Daubert test is applied in civil cases. Under Daubert, the judge/gatekeeper would determine, among other things, whether the testimony would be helpful to the jury and is supported by the literature, not just whether the proposed evidence is generally accepted in the field. In its erroneous application of Daubert, the trial court in Free found the social psychologist's testimony admissible on the basis of specialized knowledge. The appellate court reviewed the testimony under the Frye test and did not find general acceptance of the basis of the psychologist's opinion in the relevant scientific community. The appellate court further found that a major defect in the psychologist's testimony was that he addressed only the theoretical effects of police interrogation techniques on confessions and did not evaluate the defendant clinically.
By contrast, the appellate court in King observed that Harris personally examined the defendant as a basis for the psychiatric diagnoses. The court found that Harris had diagnosed psychiatric disorders in the defendant that were described in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV),19 and that the “[g]eneral acceptance of the DSM in the psychiatric community is beyond dispute” (Ref. 13, p 821). Thus, the appellate court held that Harris's testimony satisfied the Frye test, dismissing the state's argument that his opinions were not scientifically reliable. Furthermore, the court was not persuaded by the state's argument that Harris's opinion was inadmissible, since it was not based on scientific studies showing a causal connection between personality disorders and false confessions. As the court noted, Harris's testimony was offered, not for the purpose of establishing causation but to show that Mr. King's personality disorders were consistent with his false-confession claim. In the court's view, lack of a causal nexus between the influence of Mr. King's personality disorders and his false confession claim went to “the weight of Dr. Harris's opinion, not its admissibility [emphasis added]” (Ref. 13, p 821).
In further support of this view, the court cited Beagel v. State.20 In Beagel, the Alaska Court of Appeals overturned a defendant's manslaughter conviction because the trial court excluded psychiatric testimony that the defendant's self-incriminating statements were caused by confabulation produced by psychogenic amnesia. The Beagel court found that the testimony was supported by diagnostic criteria of the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III-R),21 representing a prima facie case that the testimony was based on generally accepted principles in the field of psychiatry, as required by Frye. The appellate court in King noted that the Beagel court was not swayed by the absence of scientific studies linking the defendant's psychiatric diagnosis to false confessions.
The appellate court then turned to Mr. King's appeal of the trial court's order prohibiting his expert from testifying about “anything the defendant or anyone else told him about the circumstances surrounding the giving of the confession…” (Ref. 13, p 822; emphasis in the original). The defense argued that since Harris relied on Mr. King's statements during his clinical assessment to formulate his diagnosis and opinion, he be permitted to explain their basis. Otherwise, such a sweeping prohibition would leave Harris with a net opinion. The defense also argued that Harris should be allowed to refer to police investigators' statements, as they supported his opinion that Mr. King attempted to confess falsely to other murders.
Addressing these arguments, the appellate court noted that under Rule 703 an expert witness is permitted to base an opinion on facts not otherwise admissible as evidence. The court also cited the New Jersey decision in State v. Farthing,22 recognizing the admissibility of hearsay statements relied on by an expert, not for the purpose of establishing the statement's truth, but to provide the jury with the basis of the opinion. The court observed that psychiatrists had been permitted regularly to testify in New Jersey criminal cases about a defendant's statements if the expert relied on those statements to reach an opinion about a psychiatric condition. Accordingly, the appellate court found that the blanket prohibition barring Harris's testimony about Mr. King's statements during his clinical interviews was improper. However, the appellate court cautioned that an expert may not be used to circumvent the inadmissibility of hearsay statements to put before the jury evidence that it would not otherwise be permitted to consider. To that end, the court upheld the inadmissibility of the expert's testimony related to police investigators' statements about Mr. King's reference to other unsolved crimes during the interrogation. In the court's view, allowing Harris to represent the investigators' statements as confirmation that Mr. King attempted to provide additional false confessions would be prejudicial to the state's case.
False Confessions: The Elephant in the (Court)room
Obtaining evidence in criminal cases takes on different meaning when it is infused with interpersonal dynamics (the interrogation) and the psychological traits or psychopathology of the suspect. We suggest that the problem of false or unreliable confessions is greater than meets the eye and represents significant discomfort for law enforcement and prosecutors. It is the elephant in the room that often gets ignored.
The significance of false confessions becomes apparent when one considers that one-quarter of the over 200 wrongful convictions overturned by DNA evidence in the United States involved a false confession.23 Several studies have attempted to measure the breadth of the problem of wrongful convictions based on such confessions. In the first of these studies, Bedau and Radelet24 identified 350 wrongful convictions in capital and potential capital cases between 1900 and 1985 and found that 49 of them were caused by false confessions. Gross and colleagues25 examined 340 exonerations in the United States from 1989 through 2003, of which 144 were cleared by DNA evidence and found that 15 percent of these individuals had falsely confessed. That study further indicated that 69 percent of the innocent confessors were either mentally ill or mentally retarded. In 2000, Scheck and collaborators26 analyzed 62 cases of wrongfully convicted individuals who had been exonerated by DNA evidence and found that approximately 24 percent involved false confessions. Finally, a 2008 empirical study found that 16 percent of the first 200 individuals exonerated by postconviction DNA testing in the United States had false confessions introduced during their trials.27 Despite the variability of results in these studies, the message is clear: the risk of wrongful conviction based on false confession is genuine. Thus, to the degree that a suspect's mental state plays into the analysis, there is an opportunity for expert testimony.
Classifying False Confessions
The custodial interrogation process is inherently coercive and designed to produce a confession. We do not take the position that all confessions are the product of coercion or that anyone who confesses is mentally disturbed. Yet, criminal defendants' claims of false confessions are, by nature, counterintuitive.12 That is, a judge or jury must consider the strange occurrence of an individual's having made a statement against liberty interest. Despite this apparent oddity, the literature supports not only the existence of the false confession phenomenon, but identifies a confluence of factors that may produce them. For a variety of reasons, ranging from police coercion to a guilt-ridden mind to mental illness or retardation, false or unreliable confessions find their way into criminal cases.1,11,12
Kassin and Keichel3,4 categorized false confessions into three types: voluntary, coerced-compliant, and coerced-internalized. Voluntary false confessions come about in the absence of overwhelming external stressors by law enforcement. Instead, observes Gudjonsson,5 an individual may be seeking notoriety, may have guilty feelings about some other transgression, may have an inability to distinguish reality from fantasy, or may be attempting to protect another person. Coerced-compliant false confessions are offered in an attempt to avoid external pressure or to obtain a reward. However, within this framework, the confessor subjectively does not accept the confession as true. Coerced-internalized false confessions involve an innocent person who, because of confusion and the stress of the interrogation, begins to accept criminal responsibility.
Ofshe and Leo2 propose that the improper use of interrogation techniques can result in four types of false confessions:
Stress-compliant, in which the suspect is overwhelmed by the anxiety of the situation, is physically exhausted, or both, and confesses to bring an end to the stress.
Coerced-compliant, in which the suspect responds to overt or covert threats of harm, to promises of leniency, or to both, believing that by confessing, punishment will be reduced or eliminated.
Persuaded false-noncoerced and -coerced, in which the suspect is persuaded by the interrogator of the probability of guilt, even though the suspect has no memory of the crime. The distinction between the noncoerced and coerced types is the extent to which threats, promises, or other coercive techniques are used.
To bridge the cognitive dissonance of guilt in the face of amnesia, interrogators supply theories such as alcohol- or drug-induced blackout, momentary lapse in consciousness, a repressed memory, or multiple personality.2 Ofshe and Leo2 also illustrate multiple dynamics by which skilled interrogators obtain confessions from innocent suspects—for example, the rhetorical minimization of the recited Miranda warnings, which can induce hopelessness; interpreting the suspect's demeanor (“I'm a human lie detector”); and fabricating eyewitness corroboration, co-perpetrator confession, or scientific evidence (polygraphy, fingerprints, DNA). From the suspect's side, a variety of mental conditions may contribute to the susceptibility to confessing falsely or to confessing without full understanding of the waiver of rights.1 These conditions include mental illness (e.g., psychosis, depression), mental retardation, and drug intoxication or withdrawal.
Social Science and Medical Testimony
A defendant facing the legal consequences of a confession given under coercive circumstances often seeks to challenge its validity. Such challenges may be based on the voluntariness of the confession, the credibility of the confession, or both. Various constitutional principles have been identified in the case law that supports such challenges. The Fifth Amendment right against self incrimination, the Sixth Amendment confrontation clause, and the Fourteenth Amendment right to due process are all safeguards against unreliable confessions.28–31 Yet one may wonder: what, if anything, can psychiatric, psychological, or social scientific testimony bring to the table to aid the trier of fact?
Using the New Jersey cases of Free and King as examples, one can observe that expert testimony on a false-confession claim may come from criminology or clinical science: social science or medical models. In the social science model, in the testimony, based on studies of how humans behave under stress, the expert witness explains how an individual may make statements against interest for reasons stemming from the suspect-interrogator interaction. For example, in the Ofshe and Leo,2 Kassin and Keichel,3,4 and Gudjonssen5 models, we see a typology linking typical police tactics to psychological types. This is not to say that good detective work produces false, involuntary, or unreliable confessions, only that there is a model that explains a suspect's will being overborne or shaped to the demand characteristics of the situation. In the Free ruling, social science testimony was too general to be helpful. Perhaps it would have been admitted if New Jersey had adopted the standard in Daubert rather than that in Frye. In the medical model, the expert witness looks at the individual suspect to see whether, at the time of the confession, there were cognizable psychological or psychiatric conditions present that support or explain the defendant's claim that the statement was false. For example, as we saw in the King case, there were aspects of a personality disorder that contributed to Mr. King's self-incrimination. Similarly, in the schema proposed by Weiss,1 there are many such scenarios, based on medical diagnoses of the suspect, through which to understand the suspect's choices. We are uncomfortable with the appellate court's elevating DSM-IV to the level of implying that an official diagnosis tells us something meaningful about the reliability of a confession. While no psychiatric diagnosis is ever dispositive of the question of voluntariness or reliability, expert testimony may shed light on how the characteristics of an individual suspect under the specified conditions may have led to a declaration against interest.
How Do We Know What We Know?
It is all too easy to assert that false confessions occur, that there are models to explain them, and that they may be linked to psychopathology. The harder questions involve how expert witnesses can be helpful and educational without breaching the boundaries of knowledge. Expert testimony in confessions cases, as in other areas of the criminal justice process, is provided with reasonable medical (or psychological) certainty. Witnesses must be careful not to accept the defendant's version at face value. Rather, all factors, including objective evidence of the interrogation, the defendant's prior experience, and the Axis I or II diagnosis, must be synthesized. Even so, the witness would not be expected to opine on whether a confession was true or false, only whether scientific evidence supports the defendant's current claim that the confession was false. The situation varies somewhat in the scenario of determining whether the statement was given knowingly, intelligently, and voluntarily. The expert may introduce evidence of state or trait characteristics that may have been in play in the suspect's capacity at the time the Miranda warnings were given.1 There is usually interplay between the suspect's mental state (medical model) and the overall interrogation (social science model).
Before discussing ways in which expert testimony can be introduced, we would like to comment on its scientific basis. It is apparent from the cases cited that courts are legitimately concerned about the basis of a professional's saying anything about a suspect's state of mind during a custodial interrogation. It is our view that a best-practice approach includes two elements: knowledge of the social science literature regarding police-suspect interactions and the permutations of how a suspect's will and motivations are shaped by the demand characteristics of the interrogation; and clinical data from personal examinations, psychometric testing, medical records, and school records that inform the court, within reasonable scientific certainty, about how, in the instant case, the suspect acted against liberty interest. This is not to say that the two should be given equal weight, since it is clear from case law that without clinical data the social science approach alone is likely to be excluded.12 Indeed, we suggest that ideal testimony would include an interplay between the two.
Support for this two-pronged approach comes from the philosophy of science, and it is fundamental to medical reasoning. Scientific statements are based on generally accepted laws of nature, coupled with empirical data, forming the distinction between nomothetic and idiographic levels of inference.32 In this dichotomy, nomothetic facts are general, whereas idiographic ones are specific. Thus, the social science model looks generally at the problem, and the medical model specifically at the person under consideration. It is not our intent to put too fine a point on our characterization of social science information as nomothetic in the sense of following a basic law of nature. The modest analogy we propose is that the literature on the social psychology of confessions is sufficiently robust for expert witnesses to assert, for example, “There are certain characteristics of humans in stressful situations that give rise to adaptations that would be considered self-defeating.” There does not appear to be much controversy about whether false confessions occur. The idiographic component, then, paints a picture of the individual in question, whereby the expert witness can testify; for example, “The suspect, with borderline personality disorder and anxiety developed an unrealistic belief that confessing would lead to release from custody.” It is our view that such an approach, reflecting ordinary, first-order medical logic, has a sufficient scientific basis to qualify as supportable under the Frye or Daubert tests.
Further support for this approach is found in a recent review by Chojnacki and colleagues33 of the admission of expert testimony in false-confession cases. They note, consistent with our observations, that testimony will focus on two broad areas: macro-level research on the subject, and dispositional factors and situational factors, which are psychological traits of individuals (e.g., mental illness or retardation) and specifics of the interrogation that predisposed to a false confession (e.g., police-suspect interactions and physical conditions). They acknowledge that whether the admissibility standard is Frye, Daubert, or another, the question of the utility of an expert is reduced to “whether the expert's proffered testimony: (1) will assist the jury in evaluating the credibility of the confession and is therefore admissible; or (2) is already within the common knowledge of the jury, making the testimony superfluous and therefore inadmissible” (Ref. 33, pp 20–21). They proceed to discuss the results of their empirical study of “what jurors know.” Starting from the assumption that we must know what jurors tend to know before deciding whether expert testimony would assist them, they posed a series of questions to 502 jury-eligible citizens in 38 states. The results showed an impressive lack of knowledge about suspects' rights, confessions, and police. Examples include uncertainty about the role of Miranda rights; uncertainty as to whether police can detect lying better than the average person; underestimation of the proportion of suspects who confess; and failure to appreciate that a false confession would lead to a conviction (nearly half of those interviewed). Very few said that one could tell whether a confession is true or false by listening to an audiotape or watching a videotape. Looking at the citizens' self-assessments, the authors found that about 80 percent did not believe that they had sufficient knowledge of confessions to participate in a trial and that an equal number believed that, where the truthfulness of a confession is disputed, they would benefit from expert testimony. Chojnacki et al. concluded that these findings lend support to the practice of admitting expert testimony when the question turns on whether the subject matter is beyond the knowledge of the average citizen. After all, the role of the expert is to educate the court and jury.
Show Me the Data
When weighing the value of expert testimony regarding false confession claims, a critical question that must be addressed is whether the empirical evidence supports the testimony. While the literature regarding the phenomenon of false confessions and the theoretical psychological impact of interrogations on vulnerable populations is abundant, currently there is no empirical evidence that links any particular psychiatric diagnosis with giving false confessions.34 Many studies focus on factors such as individual susceptibility and compliance in groups such as juveniles and the intellectually disabled, who may be exploited during the interrogation process.34 Although these studies implicate particular qualities that may give rise to false confessions within these groups, they do not provide a causal link to false confessions or suggest that members of these groups cannot give valid confessions. Moreover, while susceptibility and compliance studies are particularly relevant in alleged coerced false-confession situations, they are less significant in the context of voluntary confessions.
It is possible that, under certain circumstances, an actively psychotic individual who is unable to distinguish reality from delusional beliefs will voluntarily provide unreliable information to authorities. However, there are no scientific studies that establish a causal link between psychosis and false confessions. The absence of such a correlation is also the case with other psychiatric disorders. What, then, is the value of psychiatric and psychological testimony in these cases? First, the expert may educate the jury and the court regarding the existence of false confessions and dispel the common-sense belief that individuals do not falsely incriminate themselves.35 This point cannot be underestimated when one considers that mock-jury studies have shown reliably that confessions carry more weight than eyewitness and character evidence.36
Second, an expert may testify about the presence or absence of mental illness, relevant features of a particular psychiatric diagnosis, if present, and the defendant's mental state at the time of the confession. While the scientific literature may not establish a causal nexus between a particular psychiatric diagnosis and false confessions, an expert may educate the jury about the features of a diagnosis and thereby assist the jury in its determination whether the defendant's diagnosis has caused or is consistent with a false-confession claim. It is our position that since the empirical evidence linking psychiatric diagnoses to false confessions is lacking, it should be left to the jury to determine causation and whether the diagnostic features as outlined by an expert are consistent with a particular defendant's false-confession claim.
Finally, psychiatric and psychological testimony are valuable in cases in which the defendant asserts that mental illness brought about a false confession. As stated in the second point, while an expert should not testify regarding causation in these matters, testimony can be used regarding the presence or absence of mental illness and the impact that the illness may have had on the defendant's mental state. Such a determination would be critical in assessing the defendant's claim and is clearly beyond the knowledge of the average lay person.
Practical Considerations
Thus far, we have looked at the reasons that expert testimony may aid the adjudication of false-confession claims, ranging from the prevalence of the phenomenon to the social and psychological dynamics of interrogations to an education gap in jurors' understanding of the process. From a practical point of view, defendants may have one or more opportunities to raise issues involving expert testimony.
Three Bites of the Apple
A defendant may challenge the reliability of the statement during both the pretrial and trial phases of the criminal matter, as well as in a postconviction relief application. The ability to make such challenges, of course, would be waived by the defendant in the event of a plea agreement, the most prevalent of outcomes. The following discussion suggests uses of expert testimony in cases in which a false-confession claim is made.
First Bite: Pretrial
During the pretrial phase, a motion to suppress the confession may be initiated by the defense to prevent it from reaching the jury at trial. Such a motion is supported by the defendant's constitutional right to a judicial determination of the voluntariness of the confession.37 In response to this motion, a judicial hearing is held to determine, in light of the totality of the circumstances, whether the statement was voluntary.38 If the judge determines that the preponderance of the evidence supports a finding of voluntariness, then the defendant's motion is denied, and the confession is admissible at the trial phase.39 The importance of the false-confession phenomenon becomes magnified when the confession is the key evidence in the prosecution. It comes as no surprise, then, that expert testimony regarding the defendant's cognitive, emotional, or motivational state may come into play. In our experience, judges in evidence-suppression (Miranda) hearings are reluctant to rule on the reliability of alleged false confessions; the decision is relegated to the jury. Beyond that is the question of what kinds of expert testimony might be helpful to the jury without prejudicing them or invading their domain.
Second Bite: Trial
Notwithstanding the pretrial judicial determination permitting the jury to hear the confession, the defendant may still challenge its reliability through expert testimony at trial (there is no second bite if a jury trial is waived). Under such circumstances, the prosecution customarily presents a motion to exclude the defense expert's testimony. The trial judge, guided by the applicable admissibility standards for scientific expert testimony, then determines whether the expert's testimony is admissible. If the judge finds that the expert's testimony is inadmissible, the jury must weigh the incriminating statement without hearing the expert's testimony. If the judge finds that the defense expert's testimony is admissible, then the jury may consider the defendant's confession in light of the expert's opinion regarding its reliability.
Third Bite: After Conviction
Confessions are so persuasive that appellate courts, in our experience, are reluctant to overturn a conviction by revisiting a Miranda issue, except on procedural grounds (whether or when the suspect was mirandized). Psychiatric and psychological expert opinions may be useful, however, when there was ineffective assistance of counsel by failure to explore bites one and two, and DNA or other scientific evidence needs to be bolstered by a credible explanation of how a false confession came about.
Discussion
The decision in King illustrates the delicate balance that courts must apply when weighing a defendant's right to present a complete defense against the judicial interest of allowing only reliable and competent evidence to reach the eyes and ears of the jury. The King case involved voluntary statements by the defendant and, accordingly, did not require a pretrial determination of voluntariness. Thus, the appellate court in King addressed two major questions: first, whether the psychiatric testimony regarding Mr. King's personality disorders was relevant and material to his false-confession claim and, second, whether the proffered expert testimony satisfied the Frye general-acceptance standard of admissibility. The appellate court answered both of these questions in the affirmative. While the King decision cannot be generalized in regard to the standard for admissibility, we agree with the court that meaningful connections can be made between personality disorders and false-confession susceptibility. As always, testimony in this regard would go to the weight of the statement, not its truthfulness. That is, the jury is free to accept the confession at face value, to reject it as unreliable, or to adjust its importance in light of the totality of evidence.
Under the Sixth and Fourteenth Amendments to the Constitution, criminal defendants are entitled to present a complete defense, including the right to confront the state's evidence. In this case, Mr. King's seemingly voluntary statements were the primary evidence against him—the linchpin in the prosecution's case, as the court noted; no physical evidence linked him to the crime. Accordingly, Mr. King's challenging the reliability of his confession was supported by constitutional principles and due-process rights. Moreover, both Crane and Hamilton established legal precedents that such challenges may address not only the external circumstances surrounding the interrogation (for example, police conduct and conditions of the interrogation), but also the internal psychological makeup of a defendant during the interrogation. As noted by Weiss,1 internal factors may include mental illness, mental retardation, intoxication, and drug withdrawal. In Mr. King's case, personality disorder, a cognizable mental disorder, may have affected the content and reliability of his statements and thus were deemed relevant to his false-confession claim on appeal.
Relevance is one threshold for the admissibility of an expert's opinion. The testimony must also satisfy a scientific reliability threshold, such as the standards in Daubert or Frye. In King, the Frye standard governed the admissibility of expert testimony. Mr. King's expert was required to base his testimony on generally accepted principles in the psychiatric community to be admitted. The general acceptance standard of Frye was satisfied in this case, in major part because Mr. King's personality disorders and their diagnostic features were outlined in DSM-IV.19 In fact, the appellate court pointed out that recognition of a mental disorder by an authoritative source, such as the DSM-IV, represents a prima facie case of general acceptance in the psychiatric community. While it may be true that clinicians generally accept DSM-IV for classification purposes, we do not suggest that the willy-nilly use of medical labels should pass for acceptable testimony. The paucity of scientific studies linking the relevant personality disorders to false confessions did not bar the admission of the expert's testimony in King, which was otherwise admissible under the Frye test. Indeed, had the testimony been merely that individuals with personality disorders are generally susceptible to giving unreliable confessions, it likely would have been barred. Habitual lying would be a dubious example of illustrating such a relationship. One can only imagine the ensuing chaos if the confessions of antisocial individuals were considered unreliable as a matter of law.
Despite the admission of expert testimony in King regarding psychiatric disorders in relation to the defendant's false-confession claim, in the absence of scientific evidence supporting a causal link, a word of caution: while studies clearly show that false confessions occur, as experts we must avoid overstating the current empirical evidence regarding the relationship between mental illness and false confessions. Clearly, without such evidence it would be improper for an expert to testify about causation in these matters. Further, a major barrier to testifying with reasonable medical or psychological certainty that a particular diagnosis caused a false confession is that we cannot know that the confession is in fact false. It is the jury's role, not the expert's, to weigh the validity of a confession.
The court in King attempted to ease this conundrum by prohibiting testimony on causation; instead, it permitted testimony on whether the defendant's psychiatric diagnosis was consistent with his false-confession claim. However, we argue that the consistent-with standard is simply a variation of causation and remains problematic for the reasons previously outlined. Consequently, it is our view that the jury should weigh whether a defendant's psychiatric diagnosis is consistent with his false-confession claim in light of expert testimony regarding the nature of the defendant's psychiatric illness and mental state at the time of the confession. The expert can best assist the jury in making this determination by addressing the presence or absence of a psychiatric diagnosis, on the basis of psychiatric history and clinical evaluation; particular features of the defendant's psychiatric diagnosis that may have affected the defendant's ability to communicate with law enforcement (e.g., inability to distinguish reality from delusions); and the defendant's mental state at the time of the confession.
Providing broad statements that a particular psychiatric diagnosis is consistent with a false-confession claim may ultimately diminish the value of expert testimony to the court; features of many psychiatric diagnoses arguably have the potential to produce false confessions under certain circumstances. While the presence of psychiatric diagnoses may open the courtroom door for expert testimony in these matters, experts must be wary of venturing too far without the support of empirical evidence.
Toward a Best-Practice Model
Valid waivers of Miranda rights must be knowing, intelligent, and voluntary. Voluntariness comes into question when the defendant claims coercion. The reliability of the content of a statement will come under scrutiny when a defendant claims his or her will was overborne, usually by a combination of factors. These elements can have both objective and subjective dimensions, the former reflecting social science concerns (macro-research, nomothetic), and the latter being more conducive to mental health testimony (dispositional-situational, idiographic).33 In the earlier New Jersey case, State v. Free,17 the appellate court found inadmissible sociological expert testimony regarding the theory of false confessions on the grounds that it did not satisfy the Frye standard, because it merely asserted general theoretical concepts about how a false confession could transpire. In contrast, Mr. King's expert not only offered generally accepted psychiatric principles, but also based his opinion on the psychiatric examination and history of Mr. King, together with the specific circumstances of the interrogation. With this in mind, it is clear that the proffered expert testimony in Free fell within the social science model, whereas the medical model underlay the testimony in King. The divergent outcomes of these cases demonstrate courts' disapproval of presenting generalized concepts to a jury in the absence of the specific application of those concepts to a particular defendant. That is, judges want less theory and more of a dynamic explanation of how a suspect acted under the specified conditions. Perhaps they understand intuitively that jurors' fund of knowledge is often deficient, as the empirical study showed.33 Although we understand as psychiatrists that medical reasoning traditionally flows from the general to the specific, when that logic is presented to a jury, it must conform to the prevailing admissibility standard, and the jury must not be left to connect the dots.
Jurisdictions other than New Jersey also appear to favor the medical model in regulating the admissibility threshold. Notably, Soree12 called attention to state jurisdictions that addressed both models but showed a clear preference for the medical model over the social science model. In Kansas, the medical model was admitted in State v. Kleypas,40 but the court in State v. Cobb41 held that the social science model was inadmissible. In an Illinois case, People v. Rivera,42 the medical model was admitted, but the social science model was excluded. As noted, King was decided after the above case law surveys were conducted. Given the inadmissibility of the social science model in the earlier New Jersey decision in Free, the King decision places New Jersey squarely within the category of state jurisdictions showing a preference for the medical model. As Bear and Bresler43 commented, testimony regarding specific psychological disorders relevant to a confession may be less likely to elicit a Daubert or Frye challenge, whereas the value of social science testimony to the jury is often considered by the prosecution and the court as questionable. Hence, we suggest that expert testimony based on the medical model is more likely to result in admissibility in these cases, with the understanding that witnesses must be alert to jurisdictional subtleties and local case law that inform the nature of the testimony.
Accordingly, we offer the following approach to a best-practice model for preparing psychiatric testimony in false-confession cases on the defense side; prosecution-proffered testimony can utilize the model in rebuttal.
Be aware of whether the defendant is claiming that he lacked the capacity to waive Miranda rights rather than that he confessed because of factors in the circumstances of the interrogation or a breakdown in mental faculties.
There is a tendency for attorneys to rely on experts in interrogations to present the theory of false confessions in the service of an argument that the confession may have been false; this theoretical support is not sufficient. Even in jurisdictions that seem to favor the social science model, we recommend clinical correlation.
The expert witness should be prepared to use all available data to educate the court about the connection between the circumstances of the interrogation and the suspect's diagnosis on the one hand and the mechanism that produced the incompetent waiver of rights or false confession on the other. These data can be obtained with validated psychometric instruments,44 so long as they are integrated with other clinical data.
Courts generally look unfavorably on sweeping expert opinions that fail to consider the qualities of a specific defendant and the specific circumstances involved in the matter. As a result, a full psychiatric examination of the defendant and a review of the defendant's psychiatric records are likely to be required. Further, the review of any documentation relating to the defendant's confession is critical in assessing the false-confession claim. Such a process would include a review of the complete, unedited video recording of the defendant's interrogation, if available.45
The proffered testimony is more likely to satisfy the threshold requirement for admissibility if the expert offers a specific DSM-IV diagnosis of the defendant. The lack of scientific studies connecting a psychiatric diagnosis to the making of false confessions should not bar admissibility. As the court in King stated, the absence of such studies is a matter of weight, not admissibility.
Experts addressing these matters are not permitted to offer testimony regarding causation. In other words, an expert may not testify that a defendant's psychological makeup caused a false confession. Certainly, an expert cannot know that a confession is truly false and, therefore, can only offer factors that may increase the risk of an unreliable confession. Causation is a question of fact to be determined by the jury; for an expert witness to comment on it would be tantamount to saying guilty or not guilty. In assisting a jury's assessment of causation, an expert may testify in some jurisdictions that a particular psychiatric disorder is consistent with a false-confession claim, in the sense that the disorder may have been a factor in this suspect's behavior on this occasion. As noted by Soree citing United States v. Shay,46 Federal Rule of Evidence 702 “prohibits an expert to testify that ′another witness is lying or telling the truth…because the opinion exceeds the scope of the expert's knowledge and therefore merely informs the jury that it should reach a particular conclusion'” (Ref. 12, p 228). Consequently, avoiding interference with the province of the jury is imperative and increases the likelihood that the court will admit the expert opinion.
As an exception to the inadmissibility rule of hearsay statements, a psychiatrist may testify about statements made by a defendant during the psychiatric examination if those statements were utilized to formulate the opinion. Essentially, an expert may base an opinion on facts not otherwise admissible as evidence. Prosecutors may view the hearsay exception as a drawback, since it allows the defendant's words to reach the jury without cross-examination. However, by permitting such testimony, this exception facilitates the jury's education about how the expert actually reached the opinion. This perspective becomes obvious when one considers the fundamental role of the expert—assisting the jury to understand subject matter that is beyond the ken of the average citizen. Without the exception, the jury would only acquire the expert's bottom-line conclusion, thereby undermining its ability to assess the evidence within the full context of the case. Exclusion of such testimony would defeat the overall purpose of presenting an expert witness to the jury.
As the U.S. Supreme Court has noted, quoting Justice White's dissent in Bruton v. United States: A confession is like no other evidence. Indeed, “the defendant's own confession is probably the most probative and damaging evidence that can be admitted against him…. [T]he admissions of a defendant come from the actor himself, the most knowledgeable and unimpeachable source of information about his past conduct. Certainly, confessions have profound impact on the jury, so much so that we may justifiably doubt its ability to put them out of mind even if told to do so” [Ref. 31, p 296].
The judicious use of expert testimony, when grounded within the limits of empirical science, has the potential to provide fact-finders with needed perspective.
Footnotes
-
Disclosures of financial or other conflicts of interest: None.
- American Academy of Psychiatry and the Law
