Scientific Evidence

In any event, what is clear is that reliance on scientific proof has increased. One survey concluded: "About one quarter of the citizens who had served on juries which were presented with scientific evidence believed that had such evidence been absent, they would have changed their verdicts—from guilty to not guilty" (Peterson et al., p. 1748).

Voiceprints. During the last quarter of the twentieth century, each decade seemedto produce its own "novel" technique for identification. In the 1970s the admissibility of voice identification by spectrographic examination of speech samples, often referred to as "voiceprint" evidence, divided the courts. The technique was designed to identify a person's voice, often an important fact in telephone threats, extortion, and kidnaping cases. Even the name "voiceprint," which was initially used to describe this technique, sparked controversy because it suggested an unwarranted comparison with fingerprint identification. Some speech scientists concluded that the differences between these techniques "seem to exceed the similarities" (Bolt et al., p. 599). Although numerous early cases had admitted voiceprint evidence, a National Academy of Sciences report undermined the scientific basis for the technique: "Estimates of error rates now available pertain to only a few of the many combinations of conditions encountered in real-life situations. These estimates do not constitute a generally adequate basis for a judicial or legislative body to use in making judgments concerning the reliability and acceptability of aural-visual voice identification in forensic applications" (National Research Council, 1979, p. 60). This report, however, has not rendered voiceprint evidence inadmissible in all jurisdictions (State v. Coon, 974 P.2d 386 (Alaska 1999)).

Hypnotically-refreshed testimony. In the early 1980s hypnotically enhanced testimony became the focal point of the admissibility battles. The courts had rejected hypnotic evidence when first offered at trial (People v. Ebanks, 49 P. 1049, 1053 (Cal. 1897)). The issue remained dormant until the 1970s when "a dramatic rise in the use of hypnotism as an aid in criminal investigations" occurred (People v. Lucas, 435 N.Y.S.2d 461, 462 (Sup. Ct. 1980)). After traumatic crimes, victims often experience difficulties recalling the details. Police decided to resort to hypnosis to help victims retrieve the memories. Indeed, it was "hailed as a major breakthrough in police investigation" (Orne et al., p. 171).

Several factors may have accounted for this trend. First, the leading case permitting hypnotically refreshed testimony was decided in 1968. (Harding v. State, 246 A.2d 302 (Md. App. 1968), cert. denied, 395 U.S. 949 (1969)). Second, by the 1960s professional organizations, notably the American Medical Association and the American Psychiatric Association, had recognized the validity of hypnosis for therapeutic purposes, such as psychotherapy, treatment of psychosomatic illnesses, and amnesia. Third, hypnotic induction is easily learned. "A police officer can become a reasonably skilled hypnotist in a few hours of practice, with or without formal instruction" (Diamond, p. 314). Fourth, a number of books and articles on hypnosis had advocated its use, often claiming that hypnosis yielded valuable leads in many investigations: "In 77% of cases, important information was elicited that had not been available by routine interrogation" (Reiser and Nielson, p. 76).

The early judicial decisions that considered the admissibility of hypnotically refreshed testimony took the position that hypnosis affects the credibility of the testimony but not its admissibility. Under this view, the use of hypnosis to stimulate recall is treated no differently than other methods of refreshing recollection. Accordingly, cross-examination, the presentation of expert testimony on the dangers of hypnosis, and cautionary instructions are thought adequate to both protect the defendant against unreliable evidence and enable a jury to evaluate the credibility of previously hypnotized witnesses. In retrospect, there seems little dispute that this approach rested on dubious scientific grounds, and by the early 1980s courts began to more closely scrutinize this type of testimony. In rejecting hypnotically enhanced testimony in State v. Mack, 292 N.W.2d 764, 768 (Minn. 1980), the Minnesota Supreme Court recognized that for hypnosis to be "therapeutically useful, it need not produce historically accurate memory." In therapy, the mental health expert is often more concerned about what the patient believes than in what actually occurred.

Two years later, the California Supreme Court identified several dangers of hypnotically enhanced testimony (People v. Shirley, 641 P.2d 775, cert. denied, 459 U.S. 860 (1982)). These dangers include hypersuggestibility and hypercompliance. In addition, there is an inability to distinguish accurate from inaccurate recall; neither the subject nor the hypnotist can differentiate between true memories and pseudomemories. Moreover, "a witness who is uncertain of his recollections before being hypnotized will become convinced by that process that the story he told under hypnosis is true and correct in every respect" (p. 803). There is, however, no correlation between confidence and accuracy in this context. The jury may find the testimony convincing precisely because the witness displays (unjustified) confidence in his or her demeanor on the stand. Finally, the danger of confabulation—filling in details from the imagination in order to make a memory more coherent and complete—is present. After much litigation, the majority of courts concluded that the dangers outweighed any possible benefit and adopted a per se rule of exclusion, albeit with some exceptions. Other courts left the admissibility decision to the trial judge, under what became known as the "totality of the circumstances" approach.

Many of these issues resurfaced in the 1990s with "repressed memory" cases. Perhaps the most publicized case was Franklin v. Duncan, 884 F. Supp. 1435 (N. Cal.), aff'd, 70 F.3d 75, 78 (9th Cir. 1995), where the defendant was convicted of a 1969 murder based on the testimony of his daughter. She did not recall the event until twenty years later following hypnosis. The conviction was overturned on constitutional, not evidentiary, grounds.

Social science research. During the 1980s, expert testimony based on social science research found its way into American courtrooms. The first inroad was evidence of battered wife syndrome (BWS). While being a battered woman by itself is no defense to homicide, the syndrome may assist a jury in understanding two elements of a self-defense claim: (1) the defendant's subjective fear of serious injury or death, and (2) the reasonableness of that belief. For example, the evidence may explain why a battered woman has not left her mate. According to the New Jersey Supreme Court, "[o]nly by understanding these unique pressures that force battered women to remain with their mates, despite their longstanding and reasonable fear of severe bodily harm and the isolation that being a battered woman creates, can a battered woman's state of mind be accurately and fairly understood" (State v. Kelly, 478 A.2d 364, 372 (N.J. 1984)).

Another illustration is rape trauma syndrome (RTS), a phrase coined by Burgess and Holmstrom to describe the behavioral, somatic, and psychological reactions of rape and attempted rape victims. Other studies elaborated on the initial research, sometimes confirming the earlier studies and occasionally providing additional insights. "Subsequent research, which is much more rigorous, conceptualizes rape trauma in terms of specific symptoms rather than more general stages of recovery" (Frazier and Borgida, p. 299).

While BWS was primarily used by defense attorneys, RTS became an important tool for prosecutors. By the early 1980s prosecutors had begun to use this research in rape trials. RTS evidence was offered at trial for two different purposes: (1) to prove lack of consent by the alleged victim, and (2) to explain postincident conduct by a victim, such as delayed reporting of the incident, that a jury might perceive as inconsistent with the claim of rape and therefore impeaching. The courts divided over the first use but generally accepted the second use. In 1982 the Kansas Supreme Court in State v. Marks, 647 P.2d 1292, 1299 (Kan. 1982), became the first state supreme court to uphold the admission of RTS evidence. The court concluded: "An examination of the literature clearly demonstrates that the so-called 'rape trauma syndrome' is generally accepted to be a common reaction to sexual assault." Many other courts followed this precedent.

In contrast, courts rejecting RTS as proof of lack of consent disputed the scientific validity of the syndrome when offered for the first purpose. For example, in People v. Bledsoe, 681 P.2d 291 (Cal. 1984), the California Supreme Court noted that "rape trauma syndrome was not devised to determine the 'truth' or 'accuracy' of a particular past event—i.e., whether, in fact, a rape in the legal sense occurred—but rather was developed by professional rape counselors as a therapeutic tool, to help identify, predict and treat emotional problems experienced by the counselors' clients or patients" (p. 300). Thus, according to the court, although generally accepted by the scientific community for a therapeutic purpose, expert testimony on RTS was not generally accepted as a fact-finding technique "to prove that a rape, in fact, occurred" (p. 301).

Although the California Supreme Court rejected RTS evidence offered to prove lack of consent, it approved prosecution use of RTS evidence where the defendant attacked the victim's credibility and suggested to the jury that the victim's conduct after the incident was inconsistent with the claim of rape. In this situation, the court wrote, "expert testimony on rape trauma syndrome may play a particularly useful role by disabusing the jury of some widely held misconceptions about rape and rape victims, so that it may evaluate the evidence free of popular myths" (p. 298). Most courts accepted this position, admitting expert testimony to account for a victim's (1) passive resistance during a rape, (2) delay in reporting the crime, (3) failure to attempt to escape, and (4) calm demeanor after an attack. RTS evidence has also been introduced to explain that "in the context of a trust relationship, such as a doctor-patient relationship, some victims may return to the trusted relationship for further contact with the perpetrator of the assault" (Commonwealth v. Mamay, 553 N.E.2d 945, 951 (Mass. 1990)).

DNA profiling. We turn now to the late 1980s and 1990s. DNA (deoxyribonucleic acid) is a chemical messenger of genetic information, a code that gives both common and individual characteristics to people. Except for identical twins, no two individuals share the same DNA pattern. DNA is found in every body cell with a nucleus. Red blood cells lack a nucleus. However, blood may still be used as evidence because white cells and other components of blood have DNA. With few exceptions, DNA does not vary from cell to cell. Each cell contains the entire genetic code, although each cell reads only the part of the code that it needs to perform its job. Thus, blood obtained from a suspect can be compared with semen or hair cells from a crime scene. In the World Trade Center bombing prosecution, an F.B.I. expert matched saliva on an envelope sent by the terrorists to the New York Times with the DNA of one of the defendants.

In effect, the United States imported DNA typing from the United Kingdom. In 1985 Dr. Alec Jeffreys of the University of Leicester, England, recognized the utility of DNA profiling in criminal cases. Its first use in American courts came the following year. The initial appellate case, Andrews v. State, 533 So. 2d 841 (Fla. App. 1988), rev. denied, 542 So. 2d 1332 (Fla. 1989), was reported in 1988. By January 1990, forensic DNA evidence had been admitted in at least 185 cases by 38 states and the U.S. military. At the close of the twentieth century, DNA evidence, in one form or the other, was admissible in every state and federal circuit. These developments are remarkable. No other scientific technique had gained such widespread acceptance so quickly. No other technique had been as complex or evolved so rapidly. DNA profiling raised issues at the cutting edge of modern science. New DNA technologies were introduced even as cases litigating the older procedures worked their way through the court system; there have already been three generations of tests—Restriction Fragment Length Polymorphism, Polymerase Chain Reaction, and the current state of the art, Short-Tandem Repeat (STR) analysis.

Finally, no other technique has been as potentially valuable. One court called DNA evidence the "single greatest advance in the 'search for truth'. . . since the advent of cross-examination" (People v. Wesley, 533 N.Y.S.2d 643, 644 (Co. Ct. 1988), aff'd, 633 N.E.2d 451 (N.Y. 1994)). Even its critics acknowledged that "[a]ppropriately carried out and correctly interpreted, DNA typing is possibly the most powerful innovation in forensics since the development of fingerprinting in the last part of the 19th Century" (Lewontin and Hartl, p. 1746).

Yet, much of the initial euphoria that accompanied the introduction of DNA was dispelled by People v. Castro, 545 N.Y.S.2d 985, 996 (Sup. Ct. 1989), the first reported case to successfully challenge DNA evidence. In Castro the court accepted the general validity of the DNA technique but ruled that the test results in Castro were inadmissible. The court found fault with the specific manner in which the analysts applied the technique and conducted the test. Later, the F.B.I.'s top DNA expert conceded these deficiencies: "The initial outcry over DNA typing standards concerned laboratory problems: poorly defined rules for declaring a match; experiments without controls; contaminated probes and samples; and sloppy interpretation of autoradiograms. Although there is no evidence that these technical failings resulted in any wrongful convictions, the lack of standards seemed to be a recipe for trouble" (Lander and Budowle, p. 735).

After Castro was decided, the National Academy of Science's National Research Council convened a committee to study the forensic use of DNA. That committee's 1992 report recommended stringent laboratory procedures—written laboratory protocols, objective and quantitative procedures for identifying patterns, clearly defined procedures for declaring a match, and methods for identifying potential artifacts (National Research Council, 1992, pp. 52– 55). The report bemoaned the fact that there was no mandatory proficiency testing. The NAS report added: "No laboratory should let its results with a new DNA typing method be used in court, unless it has undergone such proficiency testing via blind trials" (p. 55). The publication of the National Academy of Science report did not resolve all the issues. Indeed, it sparked a heated controversy, which an experienced prosecutor would later describe as follows: "[S]cientists fighting for principle displayed an intensity, even a savagery, unmatched by the most aggressive lawyers" (Levy, p. 106). The report recommended statistical techniques designed to ensure that any random match probabilities quoted in court were relatively conservative. Critics charged that these techniques were nothing more than policy judgments masquerading as science. The criticisms of its first report were so forceful that the NAS commissioned a second report, published in 1996. The 1992 "report did not eliminate all controversy. Indeed, in propounding what the committee regarded as a moderate position [for dealing with the statistical issues] the report itself became the target of criticism from scientists and lawyers on both sides of the debate on DNA evidence in the courts" (National Research Council, 1996, p. 1).

This episode illustrates the difficulties courts face when trying to understand complicated scientific procedures and their suitability for forensic use. In 2000, DNA evidence is routinely admitted at trial. The importance of DNA profiling was underscored by a Department of Justice report that discussed the exoneration of twenty-eight convicts through the use of DNA technology—some of whom had been sentenced to death (Connors et al.). By mid-1999, more than seventy convicts had obtained postconviction relief based on exculpatory DNA test results.