DNA evidence in criminal law: new developments.

The first execution of a defendant convicted on the basis of DNA evidence took place last April when Timothy Spencer was electrocuted in Virginia.(1) DNA tests had linked Spencer to semen found on the clothing of four women who had been raped and murdered. His execution was hailed by prosecutors as proof of the value of DNA evidence and as a milestone in its growing acceptance in the legal system.

But defense lawyers questioned whether the DNA evidence against Spencer had been adequately evaluated and accurately explained to the jury during the 1988 trial.(2) Spencer's trial lawyers had done little to oppose the use of DNA evidence. They failed to obtain full disclosure of the laboratory work that incriminated their client and presented no experts to challenge it.

Shortly before the execution, Spencer's appellate lawyers asked independent experts to review the portion of the lab work that had been disclosed. They noted some disturbing anomalies.

* In one case, a lab had tested a stain that purportedly came from the bathrobe of one of Spencer's victims. The swab contained DNA from a male (presumably the rapist) and a female (presumably the victim). The independent experts' review of this evidence showed that the "profile" of the male DNA was indeed consistent with Spencer's, but the "profile" of female DNA did not match that of the victim.

* In another case, there were small but troubling inconsistencies between Spencer's DNA "fingerprint" and those of the evidentiary stain.

Despite these findings and other concerns about the quality of the DNA evidence and the accuracy of what the jury was told about it, last-minute motions to delay the execution to allow a reevaluation of the laboratory work were denied by the courts.

The controversy over Spencer's execution reflects in microcosm the broader dispute over DNA evidence. First introduced by two commercial laboratories in 1987, forensic DNA testing has gained widespread acceptance among forensic scientists.

DNA testing is offered by the FBI and by several dozen state and local crime laboratories and is used to analyze evidence in thousands of cases each year. It has been hailed as a tremendous boon to law enforcement. But there are lingering concerns about whether testing procedures have been adequately validated, whether the laboratory work is of good quality, and whether the strengths and limitatios of the tests are being adequately explained to juries.(3)

Admissibility of DNA Test Data

When DNA tests were first introduced, relatively little validation had been done on them. None of the tests met the standards later set forth by the National Research Council (NRC).(4) Nevertheless, the tests were embraced by the courts. The great majority of early cases were like Spencer's: Prosecutors presented DNA evidence; defense lawyers retained no experts to examine it and presented no witnesses to challenge it.

Doubts about the new tests began to arise in 1989. Defense counsel in a handful of cases retained qualified experts to analyze the tests, and those experts expressed serious concerns about laboratory methods. The methods for interpreting DNA evidence, for determining whether DNA prints match, and for statistical estimation were all criticized, although the statistical issues got the most attention. These concerns were first aired in the courtroom, but they eventually found their way into scientific journals.

A pivotal event occurred in December 1991. Two prominent geneticists published an article in Science that analyzed the statistical methods used by forensic laboratories and declared them "unjustified and generally unreliable."(5) In this article, Harvard professors Richard Lewontin and Daniel Hartl questioned the standard forensic assumption that the bands in a DNA print are statistically independent. Human populations may be "structured," they argued, in such a way that certain DNA patterns are particularly common in people of the same ethnic, religious, or geographic subgroup.

If there is a significant amount of structure in U.S. populations, the standard method for calculating the likelihood of a coincidental DNA match--multiplying the individual band frequencies as they are found in a database--could greatly underestimate the frequency of specific DNA profiles within a particular group.

By analogy, suppose a population survey showed that 10 percent (1 in 10) of Europeans have blond hair, 10 percent have blue eyes, and 10 percent have fair skin. Multiplying these frequencies yields a figure of .001 (1 in 1,000) for the frequency of Europeans with all three traits. But this estimate would be inaccurate because these traits tend to occur together among Nordics. The frequency of blond, blue-eyed, fair-skinned Europeans is much higher than 1 in 1,000.

Even greater errors could occur if one tried to estimate the frequency of people with all three traits among Scandinavians, but based the calculations on the frequency of each trait in an inappropriate database--Sicilians, for example. Forensic databases are typically "convenience samples," often drawn from a single blood bank, rather than random samples of the population they supposedly represent.

Concerns about population structure had previously been raised by other experts in court hearings. The Lewontin and Hartl article brought the issue to prominence, igniting a furious debate that continues to this day. The debate pits academic supporters of the forensic laboratories against those who side with Lewontin and Hartl.

The fundamental issue is how much "structure" there is in human populations. However, much of the debate has focused on arcane technical issues like the appropriate research methods and statistical techniques to use for analyzing population structure.

The debate slowed the judicial landslide favoring DNA evidence. By 1991, a few courts had held DNA evidence inadmissible.(6)

Most of these courts were in jurisdictions that apply the Frye standard, which allows evidence derived from a novel scientific technique to be presented to a jury only if it has "gained general acceptance in the particular field in which it belongs."(7) The statistical method used by the forensic labs was not generally accepted, so these courts concluded that the laboratories' statistical estimates were inadmissible.

A few courts allowed testimony that DNA prints matched, despite the absence of statistical data.(8) But most courts viewed DNA evidence as meaningless and potentially misleading without statistics on the rarity of matching profiles.(9) They therefore excluded DNA evidence altogether after finding the results of the statistical procedure inadmissible.

The trend against admissibility in Frye jurisdictions gained momentum with the April 1992 publication of a major report by the NRC.(10) The report was based on a two-year study by a distinguished panel of scientists, judges, and lawyers.

The report endorsed the general principles behind forensic DNA testing and said it should continue. However, it acknowledged problems with implementing the new technology and called for improvements. The report made recommendations on procedures for interpreting DNA evidence and ensuring the quality of laboratory results. Many of these recommendations were at odds with the procedures used in most forensic laboratories.

Without taking sides in the controversy over population structure, the report acknowledged the importance of the dispute. The report concluded that the statistical methods used by forensic laboratories should be abandoned in favor of a more conservative approach called the ceiling principle, which produces less extreme match estimates.

Since the publication of the NRC report, appellate courts in six states, the District of Columbia, and Guam have held DNA evidence inadmissible under the Frye standard.(11) Indeed, the only appellate courts that have affirmed its admissibility under Frye "either failed or deliberately refused to consider the NRC report before rendering a decision."(12)

In non-Frye jurisdictions, however, courts have generally continued admitting DNA evidence. The NRC report has had little impact there.(13)

Ceiling Approach: Failed Compromise?

The NRC's proposed "ceiling approach" for computing the frequency of DNA profiles has spawned a lively debate. The courts that rejected the statistical methods of the forensic laboratories under Frye often suggested that the alternative approach proposed by the NRC--the ceiling approach--might be admissible. In several instances, these courts remanded cases for a determination of whether this approach was generally accepted and, if so, how different the ceiling statistics were from the statistics that had been proffered against the defendant.(14)

Under the ceiling approach, laboratories may continue to estimate gene frequencies by multiplying the frequencies of the individual DNA bands found. But they must use band frequencies conservative enough to allow for the possibility that certain bands or combinations of bands may be more common in some populations than in others.

The NRC recommended that random samples of 100 people should be drawn from each of 15 to 20 distinctive and genetically homogeneous groups. When computing the frequency of a given DNA profile, the frequency assigned to each band should be a "ceiling frequency"--the highest that is observed for that band in any of the population groups, or 5 percent (1 in 20), whichever figure is higher.(15)

Pending completion of the population sampling study, which no one has yet undertaken, the NRC recommended use of a "modified ceiling approach" for computing the frequency of DNA prints.

First, the laboratory should check all its existing databases to determine whether they contain DNA prints that match the one in question. The frequency of such matches (and the size of the databases) should be reported to the trier of fact.

Second, the laboratory may compute the frequency of the DNA print by multiplying "modified ceiling frequencies" assigned to each band. The modified ceiling frequency of a band is the largest observed frequency of that band in any existing database (adjusted upward to take into account sampling error in the databases) or 10 percent (1 in 10), whichever is higher.

This approach was intended to provide a reasonable, broadly acceptable basis for making statistical estimates in the face of uncertainty over population structure. According to Eric Lander, one of the authors of the NRC report, the goal was to "ensure the admissibility of this important technology [by defining] a common ground, namely, a standard of practice so conservative as to ensure that there would be no serious scientific argument that the evidence could be said to overstate the case against a defendant."(16)

But the modified ceiling approach is not yet an acceptable compromise. Some scientists have attacked the ceiling principle as unnecessarily conservative,(17) some have said it is unclear and subject to conflicting interpretations,(18) some have defended it as a workable compromise,(19) and some have argued that it may not be conservative enough.(20)

Science or Compromise?

A major theme of critics from all perspectives is that the "ceiling principle" is not a principle of science. It is an arbitrary policy statement and can be accepted or rejected only as such.(21)

Another problem is that a dispute has arisen over how to apply the NRC's modified ceiling approach.(22) Several points are in contention. One issue is which databases (and how many databases) should be consulted when estimating gene frequency.

Some experts read the NRC recommendations to mean that all existing databases for each DNA typing system should be consulted. Others say only databases for "major races" should be consulted. The difference is often significant because bands that are rare in the "major race" databases might be quite common, for example, in databases of Native Americans or Southeast Asians.(23)

Another issue is whether, in return for adopting the NRC's "conservative" approach, labs may change other features of their estimation procedure to make them less conservative. For example, when reporting "modified ceiling" numbers, the FBI sometimes uses a method for counting the frequency of bands that many experts believe produces severe underestimates. The FBI uses this controversial procedure only in conjunction with the ceiling approach.(24)

What is at stake, obviously, is just how conservative the modified ceiling numbers will be, and how much flexibility forensic laboratories will have in implementing the NRC's recommendations. These interpretive issues may have a dramatic effect on the results of the frequency calculations.

In one case, an FBI agent-examiner purported to follow the modified ceiling principle while computing the frequency of the defendant's DNA profile six different ways.(25) The results ranged from 1 in 1.26 million to 1 in 877. But even this did not represent the full possible range. A defense expert who consulted a larger number of databases put the frequency at 1 in 84.(26)

Hence, for courts considering the admissibility of statistical estimates, the issue is not just whether the NRC's modified ceiling estimates are admissible, but which versions, if any, of these numbers are admissible.

Reaction on the Bench

Judicial reaction to the modified ceiling approach has been mixed. Some courts, noting the conservatism of the National Research Council's approach, have found that it passes muster under Frye. Other courts are still troubled by the controversy that surrounds the NRC's recommendations.

For example, a California Court of Appeal speculated in 1992 that the NRC's statistical method would gain general acceptance. Then late last year, the court noted that the NRC report had come under fire in academic publications and concluded that "recent developments have shown that general acceptance may not be easily achieved."(27)

In the meantime, research on DNA statistics continues to appear in the scientific literature. Studies in which millions of pairwise comparisons are made among thousands of DNA profiles in existing databases have been reassuring. Although multiple-band matches do occur, their frequency within existing databases appears not to be appreciably higher than that predicted by multiplying individual band frequencies.

Critics respond that there may nevertheless be structuring in the U.S. population that is undetected because members of distinctive subgroups do not appear in the databases or because they appear in numbers too small for their differences to be noticed.

Genetic studies of isolated and distinctive groups like Native Americans have found some striking differences in band frequencies between groups and strong evidence of homogeneity within groups. In a sample of approximately 50 Karitiana from Brazil, for example, about 20 percent of the individuals matched the DNA profile of another individual at four or more loci.(28)

On the other hand, data complied by the FBI from crime laboratories and blood banks worldwide showed considerably less variability from one sample to another and much greater diversity with in groups. The FBI's scientific staff concedes that the worldwide data show some population structure but argue that its effects are unlikely to be "forensically significant."(29) Obviously, the final word on this is yet to be heard.

The National Research Council made plans in 1993 for a new study that would make a second attempt to resolve the statistical debate. The study will draw on new data that have emerged since the publication of the first NRC report.(30)

The NRC's new effort was prompted by a request from former FBI Director William Sessions, who wrote that debate over the ceiling principle had "created a climate of confusion for the courts."(31) Sessions said that since the release of the NRC report, 11 of 30 appellate decisions had ruled DNA evidence inadmissible and that "courts in Canada, Australia, and the United Kingdom began hearing challenges to DNA evidence."(32) Calling this situation "a crisis," Sessions urged the NRC to "act quickly to resolve the controversy."(33)

So far, however, the new NRC effort awaits funding from various government agencies including the FBI. The funding decision is partly political. At one point, the FBI threatened to withhold funds unless the NRC limited its inquiry to the statistical methods for estimating the frequency of DNA profiles and eschewed a broader inquiry into how to describe the strength of DNA evidence in ways useful to courts. The broader inquiry presumably would have focused on the potential for laboratory errors and misinterpretations to undermine the value of the evidence.

General Acceptance

Despite all the controversy, the NRC's modified ceiling approach seems so far to have a stronger claim to "general acceptance" than any alternative. Much of the criticism of this approach comes from supporters of the forensic labs who believe that this approach is too conservative--that is, unduly beneficial to the accused. Only a handful of scientists have argued that it may not be conservative enough.

So there may well be general agreement within the meaning of Frye that the modified ceiling numbers represent an upper limit on the frequency of DNA profiles. Courts that review the admissibility of DNA statistics under Frye might reasonably conclude that presenting frequency numbers that are too conservative is better than presenting no numbers at all.

In any event, the Frye standard will be used less in the future due to the U.S. Supreme Court's decision in Daubert v. Merrell Dow Pharmaceuticals, Inc.,(34) which abrogated the Frye standard in federal trials. Daubert held that Frye had been superseded by the Federal Rules of Evidence, which do not require that scientific evidence meet the test of general acceptance to be admissible.

Although Daubert is binding only on federal courts, it is likely to be persuasive at the state level. The New Mexico Supreme Court already has followed the federal example and adopted the Daubert standard.(35)

However, at least two other state high courts recently decided to stick with Frye for the time being.(36) The Arizona Supreme Court suggested that Frye's general acceptance test is a particularly appropriate standard for evaluating the admissibility of forensic DNA evidence because its complexity makes it difficult for judges and juries to evaluate.

Daubert requires trial judges to assess the validity of scientific evidence and how well it "fits" the issues of the case. The Supreme Court recognized that scientific evidence may be valid for one purpose and not another. A key factor for judges to consider is whether there is "a valid scientific connection to a pertinent inquiry."(37)

After ascertaining the purpose of the scientific testimony in a case, the judge must conduct a "flexible" assessment of its scientific validity for that purpose, focusing on several factors, none of which is dispositive.

These factors include whether the technique is testable and has been tested, whether there is a known or potential rate of error, whether established performance standards have been followed, whether the technique has been subject to peer review, and (echoing Frye) whether it is generally accepted within the scientific community.

Different Factors

Commentators have suggested that courts evaluating DNA evidence under the Daubert standard are likely to focus on different factors than they would under Frye.(38)

The Frye inquiry begins and ends with general acceptance of the method. The key questions for courts under this analysis are "What is the method?" "What is the relevant scientific community?" and "Is the method generally accepted in the relevant community?"

Under Frye, concerns about a laboratory's rate of error or its compliance with professional standards are tangential to whether "the method" is accepted. Thus, these concerns are considered issues of the evidence's weight, to be argued to the jury, rather than issues of admissibility, to be argued to the judge.

Under Daubert, however, these issues may loom larger with regard to admissibility because they affect the "fit" of the scientific testimony to the issues raised in the case.

Consider, for example, the issue of laboratory error. DNA tests can produce wrong results due to switching, mixing, or cross-contaminating the samples in the laboratory. These errors occur often enough to be of concern. For example, in one proficiency test involving about 100 DNA samples, a commercial DNA laboratory had two sample-handling errors, each of which resulted in a false match between samples from different people.(39) According to one commentator, the rate of such errors in forensic DNA laboratories is 1 to 4 percent.(40)

Sample handling errors have also been documented in case work.(41) Hence false incriminations are far more likely to occur because of laboratory error than because two people happen to have the same DNA profile.

The purpose of statistical estimates is to show the likelihood of finding a match between two samples that are not from the same person. The "pertinent inquiry" is thus the likelihood of a false match. Yet the fact that only one other person in a billion would have a matching DNA profile may be irrelevant if the probability of a false match due to laboratory error is 1 in 50.

Given the potential for false matches due to laboratory error, the population frequency of the DNA profile, even if valid in itself, arguably may lack the "valid scientific connection ... to a pertinent inquiry" required by Daubert because it does not tell jurors what they really need to know--the probability of a false match. Worse yet, the frequency might give jurors an exaggerated impression of the strength of the DNA evidence and thereby prove prejudicial.

In the Courtroom

So far, only a handful of courts have reviewed DNA evidence under Daubert, and all have found it admissible. The most closely watched case was United States v. Bonds,(42) the only post-Daubert case in which the defendant had mounted a significant challenge to the DNA evidence.

The major issue was, as usual, the validity of the laboratory's statistical methods. The court, however, construed the statistical debate as an issue going to the weight, not the admissibility, of the statistical estimates (a position at odds with most other appellate courts that have considered the issue). Hence, the court did not come to grips with the question of the validity and "fit" of the statistical estimates within the Daubert framework.

Moreover, the long-term viability of the decision in the Bonds case is weakened by the court's refusal to consider the National Research Council report, which was published after the trial court ruled. The Sixth Circuit held that this report was not part of the record and thus had no bearing on the case. But "in future cases, where [the NRC report] is introduced into evidence, it arguably could lead to different results."(43)

While the first generation of forensic DNA tests was debated, the development of new procedures continued. New tests are already beginning to appear in the courtroom. Our experience with the first generation of tests should teach us to be cautious when evaluating newer DNA procedures.

Each new method should be carefully validated to ensure that it is reliable. As the NRC report explained,

A given DNA typing method might or might not be scientifically appropriate for forensic use. Before a method can be accepted as valid ... it must be rigorously characterized in both research and forensic settings to determine the circumstances under which it will and will not yield reliable results.(44)

Moreover, we should verify that the testing methods were properly applied in each case, so problems like those belatedly noticed in the Timothy Spencer case can be identified and addressed before it is too late.

DNA typing is a powerful weapon in law enforcement's war on crime, but like most weapons, it should be used cautiously and responsibly.

Notes

(1)Murderer Put to Death in Virginia: First U.S. Execution Based on DNA Tests, N.Y. Times, Apr. 28, 1994.

(2)See Spencer v. Murray, 18 F.3d 229 (4th Cir. 1994).

(3)See William C. Thompson & Simon Ford, DNA Testing: Debate Update, TRIAL, Apr. 1992, at 52; William C. Thompson & Simon Ford, DNA Typing, TRIAL, Sept. 1988, at 56.

(4)COMMITTEE ON DNA TECHNOLOGY IN FORENSIC SCIENCE, NATIONAL RESEARCH COUNCIL, DNA TECHNOLOGY IN FORENSIC SCIENCE (1992) (hereafter NATIONAL RESEARCH COUNCIL).

(5)Richard C. Lewontin & Daniel Hartl, Population Genetics in Forensic DNA Typing, 254 SCI. 1745, 1750 (1991).

(6)Commonwealth v. Curnin, 565 N.E.2d 440 (Mass. 1991); State v. Despain, No. 15589 (Ariz., Yuma County Super. Ct. Feb. 12, 1991); State v. Fleming, Nos. 90-CR-2716 and 90-CR-5546 (Ill., Cook County Cir. Ct. Mar. 12, 1991); State v. Passino, No. 185-1-90 (Vt., Franklin Cir. Ct. May 13, 1991); United States v. Porter, 120 Daily Wash. L. Rep. 477 (D.C. Super. Ct. 1991). For reviews, see William C. Thompson, Evaluating the Admissibility of New Genetic Identification Tests: Lessons From the "DNA War," 84 J. CRIM. L. & CRIMINOLOGY 22 (1993); David H. Kaye, DNA Evidence: Probability, Population Genetics and the Courts, 7 HARV. J.L. & TECH. 101 (1993).

(7)Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923).

(8)See Rivera v. State, 340 P.2d 933 (Wyo. 1992); State v. Schwartz, 447 N.W.2d 422 (Minn. 1989).

(9)Commonwealth v. Curnin, 65 N.E.2d 440, 443 (Mass. 1991); People v. Barney, 10 Cal. Rptr. 2d 731, 742 (Ct. App. 1992); State v. Cauthron, 846 P.2d 502, 516 (Wash. 1993).

(10)NATIONAL RESEARCH COUNCIL, supra note 4.

(11)State v. Bible, 858 P.2d 1152 (Ariz. 1993) (en banc); People v. Wallace, 17 Cal. Rptr. 2d 721 (Ct. App. 1993); People v. Pizarro, 12 Cal. Rptr. 2d 436 (Ct. App. 1992); People v. Barney, 10 Cal. Rptr. 731 (Ct. App. 1992); Vargas v. State, Nos. 92-556, 92-557, 92-558 (Fla. Dist. Ct. App. June 1, 1994); Commonwealth v. Lanigan, 596 N.E.2d 311 (Mass. 1992); State v. Vandebogart, 616 A.2d 483 (N.H. 1992); State v. Anderson, 853 P.2d 135 (N.M. Ct. App. 1993), cert. granted, 848 P.2d 531 (N.M. 1993); United States v. Porter, 618 A.2d 629 (D.C. 1992); People v. Atoigue, No. CR 91-95A, 1992 WL 245628, at [sup.*]1 (D. Guam App. Div. Sept. 11, 1992).

(12)Barry C. Scheck, DNA and Daubert, 15 CARDOZO L. REV. 1957 n.20 (1994).

(13)See Thompson, supra note 6, Kaye, supra note 6, and Scheck, supra note 12 (reviewing cases).

(14)Vandebogart, 616 A.2d 483; Porter, 618 A.2d 629.

(15)NATIONAL RESEARCH COUNCIL, supra note 4, at 82-85.

(16)Eric S. Lander, DNA Fingerprinting: The NRC Report, 260 SCI. 1221 (1993).

(17)See Peter Aldous, Geneticists Attack NRC Report as Scientifically Flawed, 259 SCI. 755 (1993); Bernard Devlin et al., Statistical Evaluation of DNA Fingerprinting: A Critique of the NRC's Report, 259 SCI. 748 (1993).

(18)Bruce Weir, DNA Fingerprinting Report, 260 SCI. 473 (1993).

(19)Lander, supra note 16; Dan Krane et al., Genetic Differences at Four DNA Typing Loci in Finnish, Italian, and Mixed Caucasian Populations, 89 PROC. NAT'L ACAD. SCI. 10583 (1992).

(20)Jennifer R. Slimowitz & Joel E. Cohen, Violations of the Ceiling Principle: Exact Conditions and Statistical Evidence, 53 AM. J. HUM. GENETICS 314 (1993).

(21)In an affidavit filed recently in a case in Washington state, Prof. Elizabeth Thompson, chair of the Department of Statistics at the University of Washington, derided the ceiling principle as "data-driven, interest-ridden, voodoo, pseudo-statistical, ad hoc methodology to which no statistician (or scientist) should be a party." Responding to the question of whether the ceiling principle is generally accepted, she said that it and other ad hoc adjustment methods "are not principles and are not science. Hence it is useless to ask concerning acceptance of these 'principles' by the 'scientific community.'" State v. DeFroe & Hollis, Nos. 92-1-03699-8 and 92-104-603-9 (Wash., King County Super. Ct. June 3, 1993) (ruling on motion to exclude DNA evidence).

(22)See Thompson, supra note 6, at 79-82.

(23)Id. at 80-81.

(24)Id.

(25)Affidavit of Harold Deadman, filed in Anderson, 853 P.2d 135. Available from author.

(26)Affidavit of Laurence Mueller, filed in Anderson, 853 P.2d 135 (on file with author).

(27)Wallace, 17 Cal. Rptr. 2d 721, 725.

(28)Laurence D. Mueller, The Use of DNA Typing in Forensic Science, 3 ACCOUNTABILITY IN RES. 55-67 (1993).

(29)Bruce Budowle et al., The Assessment of Frequency Estimates of Hoe III-Generated UNTR Profiles in Various Reference Databases, 39 J. FORENSIC SCI. 319 (1994).

(30)Telephone interview with Eric Fischer, Project Director, National Research Council (Sept. 15, 1993).

(31)Letter from William S. Sessions, Director, Federal Bureau of Investigation, U.S. Department of Justice, to Dr. Frank Press, President, National Academy of Sciences (Apr. 16, 1993) (on file with author).

(32)Id. at 1.

(33)Id. at 3.

(34)113 S. Ct. 2786 (1993).

(35)State v. Alberico, 861 P.2d 192 (1993). This ruling postdated Anderson, 853 P.2d 135, where the New Mexico Court of Appeal had rejected DNA statistics under Frye.

(36)Bible, 858 P.2d 1152, 1183; Fishback v. State, 851 P.2d 884 (Colo. 1993) (en banc).

(37)Daubert, 113 S. Ct. 2786, 2795.

(38)See Scheck, supra note 12; Edward J. Imwinkelried, The Next Step After Daubert: Developing a Similarity Epistemological Approach to Ensuring the Reliability of Non-Scientific Expert Testimony, 15 CARDOZO L. REV. 1958 (1994); Bert Black & John A. Singer, From Frye to Daubert: A New Test for Scientific Evidence, 1 SHEPARD'S EXPERT & SCI. EVIDENCE Q. 19 (1993).

(39)See William C. Thompson & Simon Ford, The Meaning of a Match: Sources of Ambiguity in the Interpretation of DNA Prints, in FORENSIC DNA TECHNOLGY 93, 142-45 (Mark A. Farley & James J. Harrington eds., 1991); Richard Lempert, Some Caveats Concerning DNA as Criminal Identification Evidence: With Thanks to the Reverend Bayes, 13 CARDOZO L. REV. 303, 323 n.56 and accompanying text (1991).

(40)Jonathan J. Koehler, DNA Matches and Statistics: Important Questions, Surprising Answers, 76 JUDICATURE 222, 228 (1993); Jonathan J. Koehler, Error and Exaggeration in the Presentation of DNA Evidence at Trial, 34 JURIMETRICS 21, 26 (1993).

(41)NATIONAL RESEARCH COUNCIL, supra note 4, at 88.

(42)12 F.3d 540 (6th Cir. 1993).

(43)Bert Black & Donna Rodney, Case Law Developments in Expert and Scientific Evidence, 1 SHEPARD'S EXPERT & SCI. EVIDENCE Q. 379, 403 (1994).

(44)NATIONAL RESEARCH COUNCIL, supra note 4, at 51-52.

Criminal Law: Helping Each Other Is the Key

ATLA's Criminal Law Section, whose mission is to assist ATLA in preserving and defending the rights of the accused, provides networking opportunities for members working in criminal law.

The section's membership is increasing, and to foster communication among lawyers, the section publishes an annual directory that includes names, addresses, and phone and fax numbers of members as well as information on practice specialties. A computer database of cases is being created for members' use. In addition, members can tap into ATLA's online bulletin board system to share information on hot topics, trends, and cases in criminal law.

Cheryl Whipper Hamilton, the 1993-94 first vice chair, said she has benefited greatly from networking with section members across the country. "The associations I have made with other lawyers are invaluable," said Hamilton, who practices in North Charleston, South Carolina. "I've found the [members] to be very willing to assist. We're not too busy to help each other."

The section works with other lawyers' groups like the National Association of Criminal Defense Lawyers and appropriate ATLA litigation groups. At ATLA's 1994 Annual Convention in Chicago, Hamilton and Joan MacDonald from ATLA's Domestic Violence Litigation Group joined forces to lead a seminar on issues common to family law attorneys and criminal defense lawyers.

Also at this year's convention, the section presented a daylong educational program that culminated in a session featuring Jack Zimmermann, who represented David Koresh during the Branch Davidian standoff with federal agents in Waco, Texas. The session also featured David Lewis, attorney for the defendants in the New York City World Trade Center bombing. The attorneys focused on how to handle high-profile cases, according to Matthew Smith, 1993-94 chair of the Criminal Law Section.

In the coming year, Hamilton said, she would like to see the section begin regular production of a newsletter to give lawyers an opportunity to share information about their cases. She added, "I think one of the most important things is staying abreast of new developments in the area of criminal law."

For information about the section, call ATLA staff liaison Susan Poland at [phone], ext. 421. For information about the bulletin board system, call Peter Larsen at [phone], ext. 308. To join, call the membership department at [phone]. Dues are $25 a year.