Getting to Foolproof
by david reich
New legal standards, DNA, and the potential of brain mapping fuel a revolution in scientific evidence.
On January 7 a judge in Philadelphia issued a ruling, and for the first
months of this year it was the talk of the courts, with words like "sea
change" and "revolution" echoing in the marble halls. Then the
judge reversed himself.
The case was US v. Llera Plaza, a run-of-the-mill drug conspiracy. But District Court Judge Louis Pollak's initial response was anything but run of the mill. He excluded certain testimony by an expert in fingerprint identification, long the gold standard for physical evidence in criminal trials, ruling that the expert could testify to similarities between the suspect's fingerprints and fingerprints lifted from a crime scene, but could not offer his conclusions about whether the two sets of fingerprints "matched."
Pollak's opinion was the latest, and perhaps the most spectacular, in a series of rulings that go back to 1993 and the US Supreme Court case Daubert v. Merrell Dow Pharmaceuticals, which set a new test for admissibility of scientific testimony. Since then, long-accepted forensic methods have been successfully challenged in trial and appeals courts, and many lawyers are predicting more challenges soon. Meanwhile, DNA testing has taken its place as the new gold standard for scientific evidence, helping undermine confidence not only in some old forensic methods but also in the accuracy of eyewitness perceptions-and in the justice system generally. An even more impressive technology looms on the horizon, a futuristic device that maps the brain. While still some years away from being ready to apply to legal cases, it could be the basis for a nearly foolproof lie detector.
Pollak wrote in his opinion that fingerprint identifications can't be tested or replicated; the discipline is so subjective, he found, that different experts can and often do draw different conclusions about whether two sets of fingerprints match. He also pointed out that the methods used by fingerprint experts have never been published in a peer-reviewed scientific journal, and that the error-rate for fingerprint identification has never been established.
Pollak was applying the US Supreme Court's Daubert standard. Since the Supreme Court's Daubert ruling, many state courts, including those in Massachusetts, have adopted similar or identical standards.
Daubert replaced a standard, called the Frye test, which, since the 1920s had required federal judges to admit the testimony of any expert whose methods were widely accepted in his or her field of expertise. In some cases Frye could be too permissive and in others too restrictive, says Professor Mark Brodin, who teaches a course on evidence at Boston College Law School. On the one hand, Frye excluded new techniques that were perfectly reliable but had yet to gain acceptance. On the other, it allowed in pseudoscience. "If you went to a convention of astrologers and asked if your birth date could determine your fate, everyone would say yes," says Brodin. So, astrological testimony was technically admissible under Frye. The Daubert standard, in contrast, was more reasonable and flexible, asking judges to weigh the factors Pollak mentioned in his written opinion-peer-reviewed publication, a known (and acceptably low) error rate, and the replicability of the expert's methods-in addition to whether the expert's methods are widely accepted in his or her field. Astrological evidence, presumably, doesn't meet the Daubert standard.
For so technical a legal matter, Pollak's opinion received a lot of notice, with stories in national media such as the Associated Press and National Public Radio. And, of course, the news moved through the legal grapevine. Defense lawyers like Scott Lopez '86 saw the ruling as a sign that long-accepted but suspect forensic methods could be successfully challenged in court. A few days before Pollak's self-reversal, Lopez was declaring his intention to challenge fingerprints in a pending case, if the case went to trial.
Maybe it was all the publicity-and the resulting alarm in police agencies and prosecutor's offices-that prompted Pollak's change of heart. The opinion that accompanied his self-reversal, issued March 13, makes much of the fact that the fingerprint experts proposing to testify in US v. Llera Plaza work for and were trained by the FBI. While fingerprint identifications by other agencies' experts may not satisfy the Daubert standard, Pollak writes in the new opinion, identifications by FBI experts do.
Brodin wonders. "I get very skeptical whenever a judge reverses himself," he says. "From my point of view he was overly impressed with testimony from the FBI about the accuracy of their own methods. I can't help but be suspicious that [Pollak] felt pressure from prosecutorial authorities. My surmise is that he was getting some heat."
Is Pollak's self-reversal the sign of a waning revolution or just a minor setback? In part it depends, in Brodin's words, on "whether other judges are going to follow [Pollak's] first opinion or his second opinion."
But only in part. Well before Pollak's headline-grabbing ruling in January, other judges had been using Daubert to exclude traditional forensic techniques. In 1999, for instance, Judge Nancy Gertner, of the federal district court in Boston, issued a ruling Pollak used as a model for his first opinion. She allowed a handwriting expert to testify to similarities between a defendant's handwriting and the writing on a robbery note but ruled out any testimony about whether the two writing samples came from the same person. In 1995, another federal trial court judge, Frank Seay, of Oklahoma, used Daubert to exclude the testimony of a hair identification expert. And last year the Supreme Court of South Carolina, using its own state's Daubert-like test, reversed a conviction in a murder case because it depended on an expert's testimony that impressions found on a shoe insole were consistent with the defendant's feet. Even before Daubert, in 1991, Spencer Letts, a federal judge in California, excluded fingerprint evidence from a criminal trial under the considerably more lenient Frye test, commenting in court that "the expertise [of fingerprint examiners] is as fragile as [for] any group I've heard hold themselves out as experts."
Observers of the legal scene predict that, as criminal and civil defense lawyers grow more conscious of Daubert and the power it places in their hands, other long-accepted kinds of evidence will go the way of the insole impression. "Fringe experts like guys who testify about how people steal cars, they're going to get slapped," says Scott Lopez. "So will human-factors experts who testify [in premises liability cases] that you have to put certain types of padding on steps or else people will trip."
Longtime public defender Paul McManus, who coaches BC Law's mock trial team, says he expects to see judges exclude testimony about defendants' future dangerousness. "How," he asks rhetorically, "do you devise an error rate for something like that?"
Though Robert Goodale '82 has spent most of his career as a prosecutor, he welcomes Judge Gertner's ruling on handwriting analysis and predicts other judges will follow suit. "If Ronald Reagan's economics were voodoo economics, handwriting analysis is voodoo science," Goodale says.
Meanwhile, veteran Boston criminal defense lawyer Norman Zalkind foresees challenges to all sorts of traditional forensic methods: hair analysis, the paraffin test, ballistics tests, the breathalyzer. "In the twenty-first century, technology is so much more developed than in the first three-quarters of the twentieth century," he says. "They're going to go back and have another look at all of these."
In an interesting twist of history, as Daubert excludes old forensic methods such as hair analysis, a new method has come along to take their place: DNA testing. As methods improve, it is a factor in more and more criminal cases. Nowadays experts can extract DNA from smaller and smaller specimens and not only from blood and semen, as before, but also from hair, saliva, and skin.
DNA testing, even more than Daubert, has had a sweeping-some would say revolutionary-impact on the justice system. For one thing, it has transformed the way the system resolves many criminal cases. "Cases used to be gray, but with DNA, they're black and white," says defense lawyer Zalkind. And McManus contends that if a DNA test excludes a defendant as the source of the DNA, the prosecution will not go forward. End of case.
Defendants whose DNA matches crime-scene DNA can certainly plead things such as insanity or self-defense. In a rape case, they can argue that the sex was consensual. But a positive test at least, in McManus's words, "forecloses the 'I wasn't there' defense."
Most often, a positive test leads to a plea bargain, Zalkind says, even in cases that before DNA might have looked good for the defendant. In one of his own cases, for example, "the victim claimed her father was the father of her child," says Zalkind. "She had a history of emotional problems and making false accusations, but this time DNA proved her right. Years ago, this would have been an easy case for a defendant. Not anymore. You can't cross-examine DNA."
Zalkind's client pleaded out. Generally, defendants whose DNA tests are positive do a lot of time. Prosecutors, knowing the positive test almost guarantees conviction at trial, don't have much incentive to make generous offers in exchange for a guilty plea. Nevertheless, says veteran prosecutor Goodale, some prosecutors mistrust DNA testing, suspecting that tests have been manipulated by defendant-friendly scientific labs-an idea Goodale himself admits is "probably paranoia."
Other effects of DNA testing have been even more significant than the changes in the structure of criminal cases. Using DNA testing, for example, defense lawyers like those at the Innocence Project of New York City's Cardozo Law School, have secured the release of more than 100 innocent men and women convicted of crimes such as rape and murder. (For information on BC Law's involvement with the Innocence Project, see sidebar following this article.) Many of those freed have been death row inmates. In fact, between the middle 1970s and 1999, eighty death row inmates in the US were released from prison, based on DNA-testing and other methods, compared to 553 inmates executed-an astonishing proportion that raises major questions about the justice system.
"There's always been a lack of confidence in the jury system by the defense bar," says public defender McManus. "We've always had innocent guys get convicted. I've seen it happen to my clients, and it's heart-rending. So it's been good for other people to see that it's not just defense lawyers saying the sky is falling. Now there's hard evidence that the system makes mistakes, and it makes mistakes in the most serious cases."
This knowledge has helped effect a small but distinct erosion in public support for the death penalty, which turns mistakes into permanent mistakes. "It's very encouraging," says McManus, who opposes capital punishment. "Opinion polls have shown a majority favoring an end to the death penalty if the alternative is a life sentence without the possibility of parole, though that may have changed since September 11."
In addition to moving the polling numbers on capital punishment, cases taken up by the Innocence Project have shed light on problems such as police misconduct (involved in thirty-eight of the project's first seventy successful cases); prosecution misconduct (thirty-four cases); ineffective counsel (twenty-three cases); false testimony by jailhouse informants (sixteen cases); and false confessions extracted using psychological pressure tactics (fifteen cases). Certain forensic methods, too-including some being attacked under Daubert-have contributed to many of the faulty convictions.
In the conviction of innocent people who were later freed by DNA, mistaken eyewitness identification has proven to be the biggest problem. According to the Innocence Project, inaccurate eyewitness testimony contributed to the original guilty verdicts in sixty-one of the project's first seventy successful cases-a whopping 87 percent of them.
This probably comes as no surprise to psychologists such as Elizabeth Loftus of the University of Washington, who, for twenty-five years, has been called as an expert witness in cases where eyewitness memories may have been degraded by stress or time, or tainted by police investigators. Decades of psychological research, described in Loftus's book Eyewitness Testimony, have shown, for example, that if a crime involves a weapon, the witnesses' vision tends to focus on the weapon rather than on the perpetrator, thus reducing their ability to accurately identify the perpetrator; that in lineups where the true suspect isn't present, witnesses, feeling pressure to come up with results, will often identify the person in the lineup who most closely resembles the perpetrator. Research has also found that "nonexistent objects can be introduced into people's recollections" by asking them suggestive questions; that witnesses' identifications of a person of a different race are less accurate than their identifications of a person of their own race; and that the certainty with which a witness identifies a suspect in court testimony has no correlation with the accuracy of the identification. Finally, even faced with problems like these, juries put more stock in eyewitness testimony than in almost any other kind of evidence.
Goodale agrees that eyewitness testimony is extremely fallible. "The defense has always been able to make those arguments, to say that [the witness] was under stress" during the commission of a crime and thus may not remember details accurately. But now defense lawyers can back up their argument with a DNA test.
What is more, says Loftus, the DNA exonerations, and the problems they revealed with eyewitness testimony spurred the US Department of Justice to issue guidelines on handling eyewitnesses for the country's police agencies. Authored by a committee of defense lawyers, prosecutors, police, and psychologists (including Loftus collaborator Gary Wells of Iowa State University), the 1999 guidelines include such common-sense advice as asking open-ended questions ("What did the car look like?" instead of "Was the car green?") and making sure "fillers" in a lineup somewhat resemble the actual suspect. (Wells's section of the Iowa State University website shows a photo of an actual lineup where the suspect is African American and all the fillers are Latinos!) "Police departments are almost going to have to [follow the guidelines] because in court cases, defense attorneys are going to bring up the deviations from those guidelines," Loftus says.
Not all jurisdictions-perhaps not even many-have adopted the new guidelines yet, but some, like the New Jersey State Police, have not only accepted them but gone farther, adopting advanced techniques like the double-blind, sequential lineup, in which a suspect and fillers are shown to the witness one at a time, by police officers who themselves don't know which member of the lineup is the actual suspect. According to psychologists, this reduces pressure on the witness to come up with a positive identification, and prevents police from inadvertently-or deliberately-tipping off the witness.
Not surprisingly, given the polling data, reverberations from the Innocence Project have reached beyond the courts and police precincts, to the halls of government. Public dismay at the death sentences given to innocent people in Illinois forced Governor George Ryan to stop executions, at least until problems with the courts can be solved. On the national level, US Rep. William Delahunt '67 (D-MA) authored a bill, The Innocence Protection Act, to help death row inmates get DNA testing--this currently costs around $8,000. Introduced three years ago, the bill also sets new standards for public defenders and gives states money to help them meet the standards. While the first part of the bill aims to expose past wrongful convictions, the second part aims to prevent such convictions in the future, according to the congressman's press secretary, Steve Schwadron. Before his election to the House, Delahunt put in twenty-two years as a prosecutor, Schwadron says, and "no one knows better than prosecutors that prosecutors goof sometimes. The difference [with a capital case] is that the consequence of an error is so permanent."
Though the bill isn't scheduled for a hearing, Schwadron predicts eventual passage. He says that because of intensive personal lobbying by Delahunt and US Rep. Ray LaHood (R-IL), the bill has accumulated 200 co-sponsors from both sides of the capital punishment debate. "I don't think there's another piece of legislation other than naming post offices that has 200-plus co-sponsors," Schwadron says. "The last thing a death penalty proponent wants is for an innocent person to be executed while a guilty person goes free to kill again."
BC Law Professor Josephine Ross is one of a handful of Massachusetts lawyers preparing the way for Delahunt's bill by lobbying state legislators for companion legislation. Ross and the others would like to see a state law that insures the preservation of biological evidence; sets up testing mechanisms and allocates money for the tests; and offers automatic compensation for people released after wrongful convictions.
One irony is worth noting here. By casting doubt on old verdicts, DNA testing has so far proved a boon to death penalty opponents--who are among testing's most vocal supporters--but it may damage their cause in the long run. As DNA testing becomes increasingly common, public confidence in verdicts can only increase, Professor Brodin predicts. "The risk in my mind is that that softens opposition to the death penalty."
Professor Dean Hashimoto, a colleague of Brodin, concurs that despite the effect of DNA testing on recent polls, it won't help end capital punishment. In 1986 Hashimoto clerked for US Supreme Court Justice William Brennan and shares his view "that someday we will change on that issue, but it will be for moral reasons and social reasons, not for reasons of accuracy."
Even if DNA and Daubert do add up to a revolution, another development
in the pipeline-- brain-mapping technology, which provides a three-dimensional
picture of the brain--may end up dwarfing their effects on the justice system.
Hashimoto, who is a physician as well as a lawyer, says this technology has
advanced to such a point that, in a recent study of adolescents, researchers
claimed to have used brain mapping to distinguish between feelings of love and
those of lust. If they're right, he says, future advances may allow us to tell
"what the brain looks like when someone is lying, what metabolites appear
in what areas of the brain when someone is feeling guilty." If that happens,
Hashimoto predicts brain mapping could be "a foolproof or near-foolproof
lie detector. Some people say that five years from now, it will be beyond the
research stage. Other people say twenty years. But it's a matter of time."
Of course this will mean huge Fourth and Fifth Amendment, as well as Daubert, issues. "Courts will have to sort out the circumstances under which individuals can be required to submit to such a test," says Brodin.
Hashimoto says that on the one hand, "juries may develop the expectation of having those results, and if they don't, wondering why a defendant wasn't willing to provide that kind of evidence. So the question is whether you should allow it in at all." On the other hand, "it would be hard not to allow in this evidence. The accused has the right to defend his case, and if [brain mapping] turns out to be very probative, how could you deny a defendant the right to take the test?"
The effects of brain mapping could reach beyond questions of mere truthfulness; it might also be used to identify disease or genetic abnormalities. "An important piece of evidence in [Ronald Reagan assailant] John Hinckley's case was a CAT scan, which, according to his defense attorney, showed that Hinckley was schizophrenic," Hashimoto says. The new technology, with its 3-D color images, makes CAT scan results look almost primitive. "You can imagine what an impact that technology, combined with expert testimony, could have on a jury."
Hashimoto thinks that even DNA testing may hold further surprises, envisioning its use, for instance, in toxic tort cases, to distinguish the effects of an environmental toxin from diseases caused by random or genetic effects.
In short, the revolution may have barely begun.
As Brodin puts it, "Given the advances in science we're seeing, are people fifty years from now going to look at evidence we're using today and say it's no better than astrology?"
David Reich is a regular contributor to BC Law Magazine. He last wrote about the late novelist George Higgins.
Students Use DNA to Help the Innocent
Though it opened its doors barely two years ago, the New England Innocence Project (NEIP), has already secured DNA-based exonerations for four men serving time in Massachusetts prisons, a result due in part to the work of interns such as BC Law student Nicole Vitrano '03.
One of three BC Law students involved in the project, Vitrano is working on a number of cases at various stages of development. She reads transcripts and motions from NEIP clients' trials, looks into the existence of DNA evidence that might lead to a client's exoneration, and visits crime labs to evaluate the evidence. She also writes up cases and presents them to the project's case review committee, a group of area lawyers and law professors (including BC Law's Josephine Ross) who determine which cases the project's pro bono lawyers should take.
"Generally speaking, the students are the heart and soul of the project. They're really our only investigators." says attorney Seth Rodner of the Boston law firm Testa Hurwitz & Thibeault, where NEIP is headquartered. The local branch is one of several offshoots of Cardozo Law School's Innocence Project that have sprung up in various regions of the country.
"Reading the transcripts has helped me get a better understanding of trial work-how attorneys create arguments and how judges rule on these arguments. It's also interesting to see that our system has been flawed in some ways," says Vitrano.
"The most rewarding part is that I'm part of this team that's helping to correct it."