From the Program in Psychiatry and the Law, Mass. Mental Health Center, Harvard Medical School. The authors thank members of the Program, our anonymous reviewers, and Robert Simon, MD for critical comments and assistance. Some portions of this article appeared first in a President's column for the Newsletter of the American Academy of Psychiatry and the Law; used with permission. Other portions were first written by Dr. Bursztajn for http://www.forensic-psych.com. Reprints from Dr. Gutheil at 74 Fenwood Rd., Boston MA 02115. (617) 626-9659 or Dr. Bursztajn at email@example.com.
Recent Supreme Court decisions emphasize the need to regulate the admissibility of expert testimony by means of standards that require opinions going beyond ipse dixit; that is, that are based on more than the fact that the expert "said it him/herself." The authors discuss subtextual themes underlying this issue and suggest approaches to attaining expert clinical opinions that reduce the likelihood of being mislabeled as ipse dixit contributions; the approach involves providing substantiation of testimony by offering a reliable methodologic basis for communicating the relevant opinion in a thoughtful and intellectually rigorous manner. A model is offered, emphasizing a process approach to opinion formulation and reformulation prior to deposition and trial. This approach addresses not only the Supreme Court's current focus on moving expert opinion beyond ipse dixit, but also such concerns as possible distortions of an expert opinion in the adversarial process. Since judicial determinations may vary depending on many factors, however, even the most careful process of opinion formulation cannot guarantee admissibility. The article assumes a general familiarity among forensic readers with the Federal Rules of Evidence and the recent series of Supreme Court decisions in this area.
The legal climate surrounding admissibility of expert witness testimony at trial has been in a state of significant change. The rapid pace of technological and scientific progress has, in accordance with the ancient Chinese curse, created "interesting times" in the courtroom. New opportunities for authentic and validated scientific and clinical expertise appear simultaneously with an emerging crisis in separating such expertise from "junk science" (1,2), as the gap between the language of experts and of the lay public widens. This tension has triggered a broad spectrum of responses, ranging from a deep distrust, bordering on stigmatization, of all expertise, to the wish to leave all decision making about admissibility in the hands of judges alone or judge-appointed expert panels (3,4,5).
Obvious problems with both of these extremes have led the Supreme Court in the past decade to issue a series of decisions that emphasize the trial judge's role as "gatekeeper" for deciding the admissibility of expert testimony (6,7,8,9,10). These decisions also emphasize the need for experts to go beyond apparently conclusory opinions by articulating their underlying methodology and reasoning and by proffering evidence of the relevance and reliability of their conclusions. The court has referred to an unsupported conclusory opinion, where the expert apparently asks the court to accept that opinion merely on his or her "say-so," as an "ipse dixit," a Latin phrase meaning "He said it himself." The Supreme Court expresses this in Kumho v. Carmichael (9, discussed below) as follows:...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. (9 at 1179)
In our discussion, we assume a general familiarity with the relevant Federal Rules of Evidence and case law regarding expert testimony. Our focus here will first address subtextual issues which are implicit in the text of relevant court decisions (these issues may provide heuristics to aid clinical experts to transcend the appearance of an ipse dixit position and, instead, to formulate opinions which are evidently reliable and relevant to clinically sophisticated judges). After reviewing some background in case law and some implicit subtexts, we attempt to indicate those approaches that would take an expert's opinion "beyond ipse dixit."
Supreme Court Cases
Few cases have had as profound an effect on the practice of expert testimony as the United States Supreme Court case of Daubert v. Merrell Dow, decided in 1993 (6). This case, not without controversy (see, for example, 11), is discussed elsewhere at length (3) and will only be summarized here. The case held that trial court judges should be the "gatekeepers" of the admissibility of expert testimony, which had to meet the standards of reliability and relevance. Reliability, the more ambiguous standard, was to be determined be such tests as error rates, peer reviewed publication, widely accepted methodology and the like (6). The standard of relevance addressed the question of whether the opinion bore on the matter at bar with sufficient applicability to be useful to the fact-finder; this, too, was a matter for judicial gatekeeping.
The original Daubert case and its successors emerged, by their own internal descriptions, as attempts to end what was perceived as a significant influx of "junk science" into the courtroom (1,2,12). "Junk science" was envisioned as one expert's basing an opinion on flawed, factitious or idiosyncratic methodology that would not capture reliable approaches to the problem at bar. Indeed, the opinion in one case used "necromancy," divination from corpses, as a metaphoric example of junk science. Courts mentioned in passing the need for a basis for an expert opinion that was more than an "ipse dixit," as earlier noted. A subsequent case, Kumho v. Carmichael Tire (9,10), extends the standards for expert clinical testimony to apply even to experience-based, non-laboratory science ("soft science"), such as clinical psychiatric expertise -- which might well lack reliable data about known error rates and similar hard-science indicia. Kumho's message is an exhortation to proceed with the intellectual rigor as judged by the standards of the relevant field:The objective of [the Daubert gatekeeping] requirement is to ensure the reliability and relevancy of expert testimony. It is to make certain that an expert, whether basing testimony upon professional studies or personal experience, [should employ] in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field. (9, at 1176)
Responses to Daubert
The Daubert decision and its successors may be seen as part of a larger concern about expert testimony within the law, within the forensic field and within the larger society. Arguably, although the testimony itself was reasonable and buttressed by clear bases on both sides, the trial of John Hinckley was a high water mark for public dissatisfaction with expert testimony, especially psychiatric testimony - a dissatisfaction that was easily generalized into skepticism about all expert testimony in the courtroom. Since then, the media and the public appear to have learned a measure of tolerance, so that strident media outcries about expert testimony are now relatively rare. Regardless, clinical experts have already grasped the need for supporting their opinions with sufficient substrate to weather a cross-examination that now draws upon what may be an overly rigid interpretation of Daubert-based concepts of reliability and relevance.
Efforts by the American Medical Association to open expert consultation and testimony to state Board of Medicine regulation represent another attempt to resist allegedly "junk testimony" from so-called "carpetbagger" experts, who supposedly travel to a different state and attempt to define a standard of care that holds the local doctor to be negligent. If, as the AMA proposes, expert testimony is the practice of medicine (which it is not, on clinical, legal or ethical grounds (13)), it can then allegedly be effectively regulated through complaints to ethics committees and Boards of Registration. Among other unfortunate effects such an approach may hang the specter of administrative complaints over the heads of testifying experts in ways that threaten or suppress testimony or participation in the process or exert subtle pressures toward excessive simplification of testimony.
The above measures may have as subtext the wish to exert some control over the "hired gun" phenomenon, a problem for the forensic fields which is notoriously difficult to control (14,15). Among the problems in resolving the hired gun issue are the challenge of definition (a hired gun sells testimony instead of time and expertise, but is this testimony "sold"?); proof (how can we be sure?); and distinguishing individual variance of standard or opinion from venality (is this opinion lying or merely outlying, or merely contrary to our own convictions?) (14).
From the viewpoint of the ethical expert witness the legal cases and ethical issues above place appropriate burdens on the witness to articulate carefully and thoughtfully the basis of the opinion and the reasoning process in reaching the opinion that are being provided for the court; that is, to go beyond ipse dixit (16). There are, however, two dimensions of the Daubert issue that have not been addressed in those opinions -- dimensions that may be relevant for the expert and thus worthy of discussion here. They are a) the true nature of problematic expert testimony and b) the basic issue of the legal system's trust in the jury's capacities.
Expert testimony: Witness- vs. attorney-centered issues
In actual forensic practice, excluding testimony that is the forensic equivalent of necromancy is not a useful remedial approach to the "hired gun" problem that bedevils our field. Testimony that is grossly deviant from general psychiatric practice is a relatively rare occurrence. A far more common problem is that attorneys often fail to pursue questioning that brings out the basis for expert testimony beyond ipse dixit. Questions are commonly posed in categorical form ("Answer yes or no") or in a form requiring an inherently conclusory answer, rather than in a manner allowing qualification and discussion of underlying reasoning or methodology (17). Misleading, constricted or personally-focused direct or cross-examination may elicit expert testimony that distorts probabilistic reasoning into mechanistic conclusions with inescapably conclusory effect (18). The very nature of testimony, often militating against extended discussion from the witness stand, may produce this result as may the occasional lack of clinical sophistication by harried judges.
These forces together may create a special case of ipse dixit. For example, consider a case of a young man who commits suicide by hanging himself in a hospital, after which the estate sues the treaters. In such a common psychiatric malpractice case both plaintiff's and defense's retained experts may exhibit profound biases or present categorical rather than balanced views. The plaintiff's biased version is the claim that -- since all suicides are foreseeable and preventable - the suicide in the case above must have resulted from malpractice, since proper treatment always prevents suicide. The comparably biased defense posture is the claim that no suicide is ever foreseeable or preventable, since the patient was clearly incompetent when he "did it to himself" and psychiatry is more art than science (18,19). As a result, neither these clinicians, nor any clinicians, are liable. Since both of these extreme positions receive little support from the literature or clinical practice, they may be subsumed under ipse dixit testimony.
In contrast to these extremes, valid testimony is based on a multidimensional forensic exploration of the presence in the instant case of a reasonable assessment of clinical condition, competence, risk factors, and state of the therapeutic alliance, coupled with a fitting treatment plan and clinical response matched to that assessment (19). Thus, a useful framework for post-suicide analysis is based on considering which risk factors are foreseeable and open to reduction or remedy, which patients are potentially treatable, and when a failure to treat appropriately is, in fact, a probable medical "cause" of a patient's suicide. Clearly, experts may disagree and still have reached opposing conclusions by accepted methods of equal intellectual rigor.
Note, however, that -- just as malpractice tribunals may work well to screen out ridiculous cases but not frivolous or meritless ones -- so the various Daubert-based approaches such as using gatekeeper proceedings or defining testimony as medical care may not suffice alone to screen out either venal, conclusory or misleading expert opinion or misleading questioning by attorneys that elicits ipse dixit testimony or and conclusory expert opinion.
Jury trust/distrust: attorney ipse dixit in opening statements
Within the legal community, constituting judges, attorneys and law professors, another subtextual schism appears to exist, based on one's faith in whether the jury can itself winnow expert wheat from expert chaff when both are heard in court. This dilemma goes to the heart of the basis for considerations of admissibility of expert testimony based on its substantiation.
Admissibility of evidence rests on a number of principles; these might include fairness to the parties, constitutional concerns, rules of evidence and the like. Of greatest relevance to our subject here is the balancing test between whether certain evidence will aid the jury in its deliberations or will be highly prejudicial to the case, e.g., by inflaming the jury's emotions. An example from a murder trial might be the issue of whether showing the jury evidence in the form of pictures of the mutilated corpse might inflame the jury into a rush to judgment, rather than permit them more coolly to decide the actual question before them, that is, whether this defendant in fact committed that heinous crime.
How does expert testimony fit into this balancing? An expert witness is defined in Federal Rule of Evidence 702 as someone who, by knowledge, skill, training or experience, can aid the fact finder to understand a fact or issue in evidence (20). The unexpressed converse of this model is the idea that a witness may foist off on the jury idiosyncratic, baseless or tendentious opinions, "cloaked in the mantle of expertise" - that is, the jury is persuaded by the expert ipse dixit. In this model the expert is viewed as exerting a form of "undue influence" on the jury, whereby the jury is swayed from their common-sense rationality into giving inappropriate credence to the witness's opinion.
Underlying this fear is a more basic concern about all expert witnesses. A school of thought within the legal community apparently sees every example of expert testimony as a potentially prejudicial intrusion on the sacred precincts of the jury's decision-making, as "invading the province of the fact-finder." (21). For example, in one mock trial at an AAPL meeting the attorney for one side stated publicly that he retains an expert solely to cancel out the other side's expert, so that he can pitch his case to the jury's gut instincts. In the same vein, Harvard Law School Professor Arthur Miller has said in a televised debate on expert witnesses (infelicitously titled "Hired Guns"!) that the jury should be left to its own wisdom in evaluating evidence without any expert input at all, since the latter encroaches on the province of the jury's native judgment as representatives of the community.
The above dilemma can be portrayed as a basic dichotomy in the view of the jury as either suggestible and capable of being swayed by undue influence from ipse dixit opinions issuing from a witness designated by the court as an expert; or as possessed of the common citizen's supposed canny ability to discern truth and to weed out non-credible, inadequately substantiated testimony, whether issuing from a designated expert or presented in the attorney's influential opening argument.
Unfortunately, optimism about a jury's persisting open-mindedness is contrary to observations that a jury will often make up its mind about a case's merit right after opening statements, the "attorney's ipse dixit." Consider this quote from a trial advocacy publication: "If done well, opening statement may be the greatest single predictor of a favorable verdict." (22).
Such unsystematic observations are corroborated by empirical research in the psychology of decision making (e.g., 23). A robust body of data on the social psychology of judgment corroborates the importance of first impressions; this view is stressed by trial advocacy texts, such as that by Slovic and colleagues (23): "One of the most general of presentation artifacts is the tendency of judgments to be anchored on initially presented values" (23). Moreover, any revision of first impressions tends to be difficult because those impressions are relatively resistant, even to significant information subsequently presented (a phenomenon known as "conservatism").
A model opinion-formulation process
In sum, the best empirical evidence currently available shows that more than 80% of a jury will make up its mind about a case's merit right after opening statements, i.e., after the attorney's ipse dixit (22). One implication of this finding is that a substantial part of an expert's efforts in a case will precede trial testimony and be directed to educating the attorney about the clinical issues involved. This education must, of course, follow on such core forensic practices as a comprehensive evaluation with review and analysis of data that emerge from the discovery process. Those data must be integrated with both clinical experience and the professional literature; access to the latter can be accomplished by provision of specific references to the attorney - an extremely valuable step. The opinion formulated on this substrate should consider alternative scenarios and hypotheses (12, 18, 24) and should display in perspicuous fashion the reasoning behind the analysis, as well as the conceptual or data-driven limitations on that reasoning. The structure described is shared in an ongoing dialog with the retaining attorney throughout the attorney-expert relationship.
It is a sad truth of forensic practice that attorney arrogance, inexperience or ignorance - leading to a refusal to be educated - may preclude or vitiate best use of the ideal model herein described. But it remains true as well that an expert who follows Daubert - inspired principles -- by presenting, not a conclusory ipse dixit, but a clearly reasoned and supported opinion-formulation process whose underlying methodology and thinking are transparent - is likely to be effective as both a consulting and a testifying expert. In practice such formulations of opinions and consultation to the ongoing process will follow certain basic guidelines:
In the post-Daubert era, experts can reduce the likelihood that their conclusions will be mislabeled as ipse dixit opinions by addressing the empirical, conceptual, published, clinical, logical and scientific underpinnings of their opinion testimony and attempting to educate attorneys about these concepts. These approaches will increase the likelihood that relevant questions will be asked to elicit credible, ethical, effective and admissible testimony to aid the fact finder. However, judicial determinations of the admissibility of expert opinions are inherently difficult to predict. Under Daubert, perhaps even more than under Frye, uncertainty prevails as to whether the content of an opinion will in fact be admissible. In the face of such uncertainty, experts can best approach opinion formulation in a process-sensitive manner as suggested here.References
Harold J. Bursztajn, M.D., a Harvard Medical School ('77) and Princeton University ('72) honors graduate, maintains an active patient practice, has a distinguished academic achievement record, and consults nationally. He is currently a senior clinical faculty member at Harvard Medical School and co-Directs the Program in Psychiatry and the Law at the Massachusetts Mental Health Center. Additionally, he consults, teaches, and testifies nationwide as an expert qualified in general and forensic psychiatry. His special areas of interest include medical and psychiatric malpractice, informed consent, managed health care, and employment related issues such as ADA, disability, workers' compensation, sexual harassment, and testamentary capacity.
See his Profile on Experts.com.
©Copyright 2000 - All Rights Reserved
DO NOT REPRODUCE WITHOUT WRITTEN PERMISSION BY AUTHOR.