- www.texas-justice.com
- www.texas-justice.com
Denver University Law Review 1997
Symposium on Coercion: An Interdisciplinary Examination of Coercion, Explotation, and the Law
II. Coerced Confessions
*1135 MISSING THE FOREST FOR THE TREES: A RESPONSE TO PAUL CASSELL'S BALANCED APPROACH" TO THE FALSE CONFESSION PROBLEM
In his Comment, [FN1] Paul Cassell ignores what our article [FN2] is about: the development and illustration of a decision model that analyzes and explains how modern methods of psychologically-based interrogation lead both to true confessions from the guilty and false confessions from the innocent. Our central point is that false confessions come about when commonplace interrogation methods are used improperly, inappropriately or ineptly. Rather than address the subject of our article, Cassell narrowly limits his commentary to the last eight pages of our 143 page paper, principally criticizing the assertion that wrongful convictions based on false confessions are numerous and declaring our policy recommendations unsound.
In this brief response, we will argue (1) that it is presently not possible to quantify the number and frequency of false confessions or the rate at which they lead to miscarriages of justice; and (2) that Cassell misunderstands our suggestions for improving the quality of contemporary interrogation practices and increasing the reliability of confession evidence. We commend Cassell for endorsing mandatory taping of interrogations, [FN3] and we agree with him that each miscarriage of justice arising from a false confession is a grave tragedy. [FN4] However, the bottom line for Cassell appears to be that if the pervasiveness of false confessions cannot be demonstrated, then it is really not that important of a policy problem. Consequently, Cassell need not address the content of our article or worry about the predictable effects of psychological coercion inside contemporary interrogation rooms.
*1136 I. Are False Confessions Pervasive or Isolated?
Because Cassell's comments can be insightful, we are disappointed that he chose not to address the substance of our article. Our aim was to explicate with unimpeachable data (i.e., excerpts from recorded interrogations) how modern interrogation tactics influence the decision-making of the guilty and the innocent. By illustrating our theoretical analysis with excerpts from interrogations that produced both true and false confessions, we sought to analyze the methods of modern psychologically based interrogation in a manner that has never before been done. Our goal here was to describe how interrogations in America today are conducted and to explain why they sometimes produce false confessions. [FN5]
Instead of addressing the topic we chose to analyze, and instead of correcting any mistakes or developing the argument in ways that we may have missed, Cassell first criticizes us by fastening onto the fringe question of whether false confessions are pervasive or isolated and then criticizes our policy recommendations. Accusing us of "reason(ing) dramatically, not quantitatively," [FN6] Cassell asserts that we do not adequately support our statement that wrongful convictions stemming from false confessions are numerous. Cassell declares that "the frequency claim is not proven," [FN7] even though our article did not seek to establish any frequency claims. [FN8] While we will take these issues up with Cassell in more depth in another forum, [FN9] several points are worth emphasizing here.
There are at least three reasons why at present it is not possible to devise an empirical study to measure, quantify or estimate with any reasonable degree of certainty the incidence of police induced false confessions or the number of wrongful convictions they cause.
*1137 First, American police typically do not record interrogations in their entirety. When they do record, only the confession statement is likely to be memorialized. Without a record of the interrogation process, it is difficult to evaluate how an initially resistant suspect was motivated to confess. It is also difficult to assess the reliability of a confession statement because of the possibility that a careless investigator educated the suspect about the facts of the crime. Any attempt to reconstruct what occurred during an interrogation and what a suspect independently knew is necessarily undermined by the lack of a record: human memory for conversation is limited and recall is selective due to the position bias of the participant. In many disputed confession cases, then, it may not be possible ever to ascertain whether the confession was voluntary and/or reliable with any reasonable degree of certainty.
Second, because no criminal justice agency keeps records or collects statistics on the number or frequency of interrogations in America, no one knows how often suspects are interrogated or how often they confess, whether truthfully or falsely. It is untenable to randomly sample interrogations to find out how many lead to false confessions not only because it is currently not possible to identify the universe of cases under study, but also because it may not be possible to determine the ground truth (i.e., what really happened) in a particular case. Nor is it possible to extrapolate the number of interrogations occurring annually from a single non-random study in one non- representative jurisdiction. Unless and until a criminal justice agency begins to collect such statistics, it simply will not be possible to estimate or provide an answer even to the elementary question of how often interrogations occur, much less how often they lead to false confessions.
Third, many cases of false confession are likely to go entirely unreported and therefore unacknowledged and unnoticed. There are no central databanks or yearly criminal justice handbooks documenting the annual numbers even of detected false confessions. Typically the only public record is in local court papers and perhaps in the local press, sources not easily accessible to most scholars. Even when such records documenting a disputed confession case are available, it is often difficult to unequivocally establish the ground truth about the crime, especially since most confessors will be arrested, charged, prosecuted and/or convicted. [FN10] Because it is not possible to reach valid or reliable estimates of the incidence of false confessions, it is also not possible to estimate how often false confessions lead to wrongful convictions.
Nevertheless, the empirical record does not caste doubt on the likely occurrence of false confessions. Rather, there is compelling and abundant evidence that false confessions occur regularly. Cassell overlooks much of the empirical record when he suggests that readers of our article are "left with a promissory note that this proof will be supplied in the future." [FN11] In fact, social scientists, researchers and journalists have documented numerous cases of psychologically induced false confession in recent years. [FN12] They have also *1139 demonstrated that police-induced false confessions often lead to lengthy and unjust deprivations of liberty and miscarriages of justice. [FN13] Because false confessions are likely to go unnoticed, however, it is reasonable to assume that the reported cases represent only the proverbial tip of the false confession iceberg.
In sum, it is presently not possible to know, estimate or quantify the frequency of false confessions or the annual rate or number of miscarriages of justice they cause. It appears, however, that police induced false confessions occur often and are highly likely to lead to the wrongful arrest, prosecution, conviction, and/or incarceration of the innocent. [FN14] In some instances, police-induced false confessions have led to the execution of the innocent, the ultimate miscarriage of justice. [FN15]
Even if it were possible to quantify the rate at which police-induced false confessions occur and the rate at which they lead to wrongful conviction, Cassell does not provide any meaningful theoretical context within which to understand the importance of such figures. Because not all facts are created equal, social scientists typically work from paradigms that order and give meaning to the facts. Without a paradigm, though, it is not possible to ascertain from which base population the rates of false confession and confession-driven miscarriages of justice should be computed and how they should be interpreted. Should the base population, for example, be the number of interrogations police conduct annually? The number of index crimes? The most serious index crimes, such as murder and rape? And what would such figures tell us? Without a theory or paradigm to inform us which numbers are important and why, it may not be worth the effort and expense necessary to overcome the formidable barriers to quantification, especially since there appears to be widespread agreement that false confessions and miscarriages of justice occur sufficiently often to warrant the abiding concern of legal scholars, jurists, and legislators.
Our experience with a few hundred cases of false confession leads us to conclude that it is far more important to study the conditions under which they occur, the characteristics of such cases and why they lead to deprivations of liberty and miscarriages of justice than it is to attempt to quantify what currently cannot be known. Our research indicates that the phenomenon of false confession in America today is, for the most part, explained by the inappropriate, inept and/or illegal use of contemporary interrogation methods, not the dispositions or attributes of particular individuals. In virtually all of the cases *1140 we have studied, the crime under investigation was murder; the police were under enormous pressure to solve it; they had not been able to identify a suspect based on solid detective work or forensic evidence, but instead relied on pseudoscientific interpretations of behavior, gut hunches, and/or the suspect's relationship to the victim (e.g., spouse, boyfriend, etc.); there existed little or no credible evidence against the suspect, either before or after the confession; and police resorted to inappropriate and/or illegal interrogation methods such as the maximization/minimization technique to elicit the false statement(s). [FN16]
If our analysis is correct, then the important question is not whether false confessions are pervasive or isolated. Rather, it is: In high profile murder cases, why do police investigators so often select the wrong suspect, resort to inappropriate and illegal interrogation tactics, and automatically consider a case closed once they have elicited a confession? Why do police, prosecutors, judges and juries so readily assume that a confession must be true when it was obtained by psychologically coercive methods, is internally inconsistent, does not lead to corroboration and/or is contradicted by the facts of the case? Why do police, prosecutors, judges and juries so readily presume a suspect's guilt based on a questionable confession statement when the sum of evidence should lead them to conclude that the suspect is innocent beyond any reasonable doubt? And what can be done to prevent police-induced false confessions in high profile cases and their admission into evidence?
Legal scholars, jurists and legislators seeking answers to these questions will be disappointed by Cassell's "balanced approach" to the false confession problem.
II. Are the Proposed Remedies Sound Public Policy?
Cassell appears to misunderstand our suggestions for improving contemporary interrogation practices and the quality of confession evidence. He states that we advocate that "police questioning should be restricted to prevent frequently occurring miscarriages of justice from false confessions" [FN17] and suggests that we fail to consider "a carefully crafted balance designed to fully protect both the defendant's and society's interest." [FN18] These assertions are incorrect.
While we have argued for better police training, better pre-interrogation selection procedures and better analysis of the probative value of post- admission narratives, we have not suggested any restrictions on police interrogation procedures. Cassell recognizes this when criticizing Professor Alschuler for suggesting prohibitions on interrogators' ability to lie to and trick suspects, [FN19] *1141 something we do not propose. A careful reader of our article will be hardpressed to find any suggestion that the Fifth, Sixth or Fourteenth Amendments--the current doctrine's regulating American interrogation practices--be employed to restrict the investigative powers of police.
Cassell also suggests that we fail to take into account society's interest. This too is incorrect. It goes without saying that police interrogation is an important and legitimate criminal investigation strategy in America. Cassell does not fully appreciate that our research here is neutral with respect to whether the rules of interrogation should be modified to advantage the state or the suspect. Our analysis of the causes and consequences of false confessions should not be misconstrued as either pro-prosecution or pro-defense. Instead, we take the existing legal rules as given, and suggest how police can improve the quality of interrogation practices, obtain more reliable confession evidence, and detect false confessions early on while staying within these rules. We also empirically illustrate the dangers of violating prescriptions that were written into law to reduce false confessions and protect the innocent. For example, we have criticized the maximization/minimization technique as inherently coercive because, as laboratory and field researchers have demonstrated, [FN20] it relies on threats of harm and promises of leniency for its efficacy and is likely to elicit false confessions.
We make two simple policy recommendations: mandatory electronic taping of the entirety of the interrogation process in all felony cases, [FN21] and a requirement that a confession demonstrates its reliability before being admitted into evidence. [FN22] These are hardly controversial proposals. Indeed, Cassell has also recommended taping interrogations, though he suggests that this will be acceptable to law enforcement only if it is coupled with scaled back Miranda warnings (i.e., eliminating waiver and questioning cut-off requirements). [FN23] While we neither endorse nor reject Cassell's proposal, a taping requirement is logically independent of whether to keep, scale back or eliminate Miranda's warning and waiver requirements. The two issues need not be conflated and should not be confused. Cassell offers this proposal as a kind of political horse trade [FN24] that, he believes, represents a reasonable compromise to all sides. [FN25] There is no evidence, however, that the law enforcement community and criminal justice policy-makers will accept taping requirements only if pitched as an alternative to Miranda. [FN26]
*1142 Cassell also misunderstands our second policy recommendation: that confession statements should be evaluated with respect to their degree of reliability before they are admitted into evidence. We are not suggesting that confession statements should be admitted only if they are corroborated by independently obtained evidence. Rather, we are suggesting that since a suspect's admission that "I did it" can be either true or false, the confession that follows may turn out to be either evidence of guilt or evidence of innocence. To determine which, courts should examine the fit of the details (both dramatic and mundane) of the suspect's post-admission narrative with the facts of the crime to make an evaluation of whether the suspect appears to possess or lack the psonal knowledge that would be expected from the true perpetrator. If the confessor's description of the crime fits the verifiable crime facts and is not inconsistent with the physical evidence, the confessor must possess the personal knowledge that would be known to the perpetrator and thus his post-admission narrative provides sufficient indicia of reliability to warrant its admission into evidence. If, however, the suspect's answers fit poorly with the known crime facts and therefore demonstrate an apparent ignorance of the crime (e.g., his answers about missing evidence are proven wrong, he inaccurately describes verifiable crime facts, etc.), then the post- admission narrative has failed to provide sufficient indica of reliability to warrant its admissibility.
In making this commendation we are not suggesting that judges decide whether a particular confession is true or false. Rather, because confession evidence is so exceptionally persuasive--so prejudicial--it should be assessed to determine whether its probative value outweighs its immense prejudicial effect. [FN27] No evidence can be judged to have probative value if it lacks reasonable indicia of reliability. We propose that there should be a standard of demonstrable reliability below which no confession statement should be admitted into evidence. To possess suicient indicia of reliability, a confession statement must fit the facts of the crime to a reasonable degree. Indeed, American law enforcement routinely acts on this principle when screening volunteered confessions; judges routinely act on this principle when deciding whether to admit hearsay testimony. If it is not sufficiently reliable, a confession's prejudicial effect will likely be high and may greatly confuse and mislead the jury. [FN28] Accordingly, consistent with the judge's traditional role as gatekeeper to exclude untrustworthy evidence that has a strong tendency to mislead the jury, the judge should not admit confessions without finding that the confession meets a minimum standard of reliability.
*1143 III. Conclusion
Paul Cassell's Comment avoids the heart of our article and instead criticizes our assertions about the frequency of false confessions and our policy recommendations. These criticisms are unwarranted. It is well- established that false confessions occur regularly and thus with troubling, if unquantifiable, frequency. The policy recommendations that Cassell attacks would not restrict interrogation practices or disadvantage police investigators. Rather, the mandatory taping of interrogations in their entirety and the minimal reliability requirements that we propose should improve the quality of police practices and therefore result in better confession evidence. These proposals would benefit both law enforcement and innocent defendants. [FN29] More importantly, they would improve the truth finding function of the criminal process.
It is well-documented that false confessions lead to significant deprivations of liberty and miscarriages of justice. [FN30] In some instances, false confessions have even led to the wrongful execution of innocents. [FN31] The conventional wisdom that people do not confess to crimes they did not commit and that the American criminal justice system does not wrongfully convict innocents permits these injustices to occur. The first step toward preventing wrongful *1144 deprivations of liberty and miscarriages of justice is to acknowledge that the conventional wisdom is wrong. Rather than dispel this myth, Paul Cassell's "balanced approach" to the false confession perpetuates it.