Monday, August 29, 2005


Scientific advances in fighting crime make forensic evidence the stuff of television dramas.

A single hair can identify a killer. A fingerprint can prove that one witness is lying, the other telling the truth. DNA, preserved for years in drops of dried blood or even a stray skin cell can solve cold cases that once left investigators stumped.
Darryl Hunt won his freedom two years ago after DNA correctly identified another man as the person who had stabbed Deborah Sykes to death in Winston-Salem, a crime for which Hunt spent 18 years in prison.

In the messy world of crime and justice, scientific evidence stands out as pure, objective fact – so much so that prosecutors complain that jurors, schooled in justice by CSI and other shows, often won’t convict without it. If the DNA matches, it must be so.

But there’s a story in forensic science that you won’t see on CSI.

It’s the episode in which the scientists get it wrong.

The Winston-Salem Journal decided to examine forensic science in North Carolina because of the large number of wrongful convictions across the country caused by flaws in forensics or its misapplication. The newspaper tracked North Carolina cases with questionable forensic evidence. The focus was on capital cases – in which the stakes and expectation of certainty are highest. They show how in North Carolina, as in other states across the country, prosecutors have portrayed evidence as scientifically grounded and put prisoners on death row or locked them away for life in spite of disputes over the veracity of that science. Just as significantly, the cases show how errors and omissions in the use of science can undermine chances for successful prosecution.

Take the 1994 trial in Stokes County of Rex Penland, charged in the rape and killing of a Winston-Salem prostitute.

Penland said he was passed out drunk in his pickup when Vernice Alford was stabbed to death, but the physical evidence presented at trial supported testimony by his nephews – twin brothers named Larry and Gary Sapp. They said that Penland raped Alford and had them tie her to a tree. The twins then waited in the truck, they said, and he soon returned with a bloody hunting knife and bragged of “icing” her.

A deputy and an agent with the State Bureau of Investigation testified that a footprint at the scene matched Penland’s snakeskin cowboy boots (*SEE CORRECTION). With the footprint and the knife on their side, not to mention the Sapps’ testimony, prosecutors won a murder conviction and a death sentence.

Now, 11 years later, the science is on Penland’s side.

A report by the state crime lab shows that there was no scientific evidence that the footprint matched Penland’s boots. And new DNA evidence shows that the semen collected in the case wasn’t his. As for the blood on his knife, the DNA doesn’t match Alford’s DNA. Instead, the partial DNA profile matches Penland’s, who testified that he cut himself skinning a deer a few days before Alford’s murder.

The new evidence may not prove innocence, but it was enough for a judge last month to grant Penland a new trial.

“What the science today shows is that the jurors were presented with facts and evidence that we now know were not true,” Steve Royster, one of Penland’s attorneys, said at his hearing.

According to a study of wrongful convictions published this month in Science magazine, faulty forensic evidence was the second-leading factor nationwide, after mistaken eyewitness identification – the factor that led to Hunt’s wrongful conviction.

A recent slew of scandals at crime labs across the country has cast doubt even on DNA evidence – the gold standard of forensic science. DNA evidence, which sounds so irrefutable in court, is subject to interpretation. Analysts can make mistakes, and contamination can skew the results of DNA testing so sensitive that it can read a genetic profile from a single cell.

“What we do know is we’ve had a lot of false convictions in cases that seem to have been caused by exaggerated or misleading presentations of science,” said William Thompson, a professor of criminology at the University of California at Irvine and an expert on the use of DNA testing in criminal cases.

“It’s difficult for jurors to evaluate it based on common sense … if they’re hearing testimony cloaked in the mantle of science, I think they’ll give it a great deal of credibility. They’re going to think it’s accurate unless they have some reason to think otherwise.”

A night of violence

Vernice Alford left her mother’s apartment on Conley Street about 4 p.m. on Nov. 30, 1992.

She told her mother that she would be home that night because the next morning she was expected at workat the convention center, where she was a waitress.

Alford, 29, had a second job. She was a prostitute.

That night, Penland and his 18-year-old nephews drove from King to Winston-Salem in Penland’s green Chevrolet pickup.

According to the Sapp brothers’ testimony, they picked up Alford on Patterson Avenue and headed north to Payne Road, which winds through the southern part of rural Stokes County.

Three days later, two surveyors found Alford’s body in a ditch at the end of a logging road.

She had been stabbed 18 times. A length of nylon cord was tied to a bloodstained tree trunk. Investigators found an empty Monarch cigarette pack at the scene and a cigarette butt.

The arrests were quick.

The Stokes County Sheriff’s Office got an anonymous tip that the Sapp brothers had something to do with the crime. Each gave a statement implicating Penland.

Investigators arrested Penland at his mobile home in Germanton, where they seized his knife, a pack of Monarch cigarettes, a pair of muddy cowboy boots and a pair of black pants, torn in the thigh.

Investigators made a cast of the tire marks in the mud near where Alford’s body was found and one of a footprint.

The tire marks were consistent with Penland’s truck – not surprising because there was never any question that it was his truck at the scene. The question was whether Penland was passed out in his truck as he said, or whether the Sapps told the truth.

That made the footprint a crucial piece of evidence.

Joyce Petzka, an agent in the SBI crime lab, sent a report to prosecutors saying that she couldn’t tell anything about the shoe from the cast made at the crime scene.

She testified at Penland’s trial about the tire tracks, but no one asked her about the footprint. Instead, prosecutors called Lt. Al Tuttle of the Stokes County Sheriff’s Office and Bill Lemons, an agent with the SBI, to testify as expert witnesses about the footprint.

They both said that the print was consistent with the boots they had seized from Penland’s home.

In his closing argument, the prosecutor used the footprint to declare Penland the liar and the Sapp brothers the truthful witnesses.

“If Rex Dean Penland was passed out drunk in the truck, how come his cowboy-boot prints are out there in the sod that night?” argued James Yeatts, the assistant district attorney in Stokes County. “What does that tell you about Larry and Gary Sapp? Tells you he wasn’t passed out. It was just like they said. Mister was outside the truck in his boots.”

After deliberating just 46 minutes, the jury convicted Penland and then sentenced him to death. Two weeks later, prosecutors cut a deal with the Sapp brothers. They pleaded guilty to second-degree murder and related charges and were each sentenced to 15 years. They were released in 1998, after serving less than five years. Larry Sapp is back in prison, on a charge of possession of a firearm by a convicted felon; since his release, Gary Sapp has been convicted twice of assault on a female.

Penland has been trying to prove his innocence.

In a motion filed in 1998, his lawyers argued that prosecutors withheld evidence at his trial by not turning over the SBI report saying that the footprint was unreadable. The motion goes on to argue that prosecutors passed over the valid expert opinion from the SBI lab in favor of a deputy’s opinion.

The DNA evidence won Penland a new trial before the court could consider the dispute about the expert testimony on the bootprint.

The new scientific evidence may not settle whether it was Penland or his nephews who lied about what happened the night that Alford was stabbed to death, but it changes the state’s case against Penland. At trial, the forensic evidence corroborated the Sapps’ version of what happend; now the science corroborates Penland’s version.

Identification by footprint

North Carolina produced one of the most infamous charlatans in forensic science.

In the late 1970s and early ’80s, an anthropologist at the University of North Carolina at Greensboro named Louise Robbins testified in numerous criminal cases that she could identify a suspect through a footprint.

Physical anthropologists use footprints to investigate general characteristics about a tribe or a period in human development. Robbins was the first to testify in court that individuals leave unique footprints that could be analyzed with the accuracy of fingerprint identification.

Crime-lab analysts regularly testify that they can match a footprint to a defendant’s shoes by examining treads, heel proportions and width. Robbins’ claims went much further. She argued that a person’s gait, height, weight – even race – left a unique print. Newspapers and Time magazine touted her claims, but the scientific community ignored her work until defense attorneys began to complain.

In 1987, the American Academy of Forensic Sciences found that there was no scientific basis to her theories.

“It’s just cockamamie stuff,” said Russell Tuttle, a professor of physical anthropology at the University of Chicago who has researched her work. “Why do we allow this kind of rot, this pseudoscience to get into our courts? Her so-called method should never have been tried in the court.”

Robbins died shortly after her work was debunked. Her testimony helped convict at least 10 people, including Vonnie Bullard of Cumberland County.

Bullard was convicted in 1982 in the murder of Pedro Hales, who was shot and stabbed to death in August 1981. The state’s case was largely circumstantial, based on witnesses who said they saw Bullard or his truck near where Hales was believed to have been murdered. The only physical evidence was a bare bloody footprint found at what appeared to be the crime scene.

Robbins testified that the footprint was Bullard’s. A Cumberland County jury convicted him of first-degree murder. In 1984, the N.C. Supreme Court upheld the conviction. The U.S. Supreme Court turned down his final appeal in 2002, and his attorneys say that he has run out of legal remedies. He remains in prison, serving a life sentence.

Critics say that Robbins’ work points out the influence that scientists have on a jury.

“Louise Robbins wasn’t a witness who said, ‘I saw a footprint there and it looks like the same size as the defendant’s foot’ – what anyone else could say,” said Rich Rosen, a law professor at University of North Carolina at Chapel Hill. “She said, ‘I’m a scientist, an objective, neutral observer, and based on scientific principles, I can tell you that was the defendant’s footprint.’ That carries tremendous weight with the jury.”

In the years since Robbins’ work was discredited, other forensic theories also have fallen.

Glass shattered in a spider-web pattern was once thought to be proof of arson; bite marks were once thought to be unique; and the composition of bullet lead was once considered a valid way to match bullets to a crime. All have been debunked as nothing more than theory, and experts continue to raise questions about the reliability of standard methods such as fingerprint and hair analysis.

“All these things from CSI programs have this reputation of infallibility, yet when you scratch the surface, most of these so-called sciences are made of tin,” said David Faigman, a professor at the University of California Hastings College of the Law. “It’s almost an alchemy that’s going on when they turn their experience into scientific opinion when there’s no research underlying it.”

The advent of DNA testing

DNA testing was supposed to be above reproach.

Other forensic sciences were developed by criminal investigators without the rigorous peer review that defines modern science. Forensic DNA analysis is different. It grew out of research into the human genetic code carried out at some of the most prestigious research laboratories in the world.

By the early 1990s, scientists learned how to extract DNA from minute samples of biological material such as blood, semen or saliva, and replicate it many times over so that it could be compared to a suspect’s DNA.

The technology, known as polymerase chain reaction, or PCR, uses markers along the DNA molecule to develop a genetic profile. And because every individual, with the exception of identical twins, has a unique genetic code, DNA testing is the most precise tool that investigators have ever had for identifying criminal suspects.

Virginia’s crime lab was considered one of the best in the country. But in April, an audit by the accrediting agency found that the lab had botched DNA testing in the case of Earl Washington, who was within days of execution for a rape and murder when DNA testing raised doubts about his guilt.

He was eventually pardoned, but his lawyers say that the pardon came seven years late because of the crime lab’s mistakes.

The American Society of Crime Laboratory Directors/Laboratory Accreditation Board cited the Virginia lab for errors in the analysis of Washington’s DNA, and, as significantly, for overlooking a match to a serial rapist who has never been prosecuted in the case. In May, the audit ledGov. Mark Warner of Virginia to order a review of 150 other cases.

Betty Layne DesPortes, a defense lawyer in Richmond and the chairwoman of the jurisprudence section of the American Academy of Forensic Sciences, said that the Washington case shows how even the most respected laboratories can make mistakes.

Among other findings, the audit found that the lab failed to catch its own mistake, and that state DNA analysts gave in to political pressure to draw conclusions that would support Washington’s conviction.

“If forensics is not objectively tested and interpreted by adequately trained scientists, your search for the truth is going to be meaningless,” DesPortes said. “Unless we pay attention, we’re going to wind up executing an innocent person or we’re going to wind up sending a guilty person out there to prey on an innocent person.”

Critics say that state crime labs are prone to error because they don’t follow the most basic tenets of science – openness and neutrality.

In medicine, scientists make every effort to shield their research from bias. In clinical studies of new drugs, researchers do not know the identities of the patients or which ones are taking the drug in question and which ones are taking the placebo.

Forensic science lacks that neutrality.

In North Carolina, for example, the reports used by the SBI identify evidence by a suspect’s name. Many crime-lab scientists are sworn law-enforcement officers, raising questions among defense attorneys about their bias. And unless a defense attorney hires an expert to review a particular test, crime labs are rarely subject to the kind of peer review common in other laboratories.

“I think it’s crazy that right now the SBI has a policy of only employing sworn law-enforcement agents to conduct forensic tests and to testify,” said Mark Rabil, a defense attorney in Winston-Salem who challenged SBI evidence in a capital case in Hickory. Rabil also represented Darryl Hunt in Winston-Salem.

“From what I have seen in 25 years of cross-examining SBI lab personnel,” Rabil said, “they generally act as advocates for the state rather than as objective scientists.”

North Carolina’s crime lab, part of the SBI, covers 120,000 square feet in Raleigh. The SBI also runs a satellite laboratory in Asheville. The labs handle nearly 40,000 cases a year, with DNA cases getting the most attention. In the past five years, the SBI has tripled the number of DNA analysts to 28. Some 78 other scientists and technicians handle fiber analysis, drugs, fingerprinting, footprints and other specialties.

North Carolina’s crime lab has been accredited since 1987 and has always passed with high marks. Outside auditors who review its DNA section also give it high marks.

But critics say that auditors can miss problems. Auditors inspect laboratory equipment and check written procedures, but they don’t re-analyze evidence in cases to double-check the work of the laboratory analysts.

“The fact that they’ve passed an audit doesn’t mean there can’t be serious problems,” said Thompson of the University of California at Irvine. He noted that the Virginia lab also was accredited.

Defense attorneys accept some of the blame for problems resulting from forensic evidence. Too many, they say, lack the training to challenge forensic evidence or even to know what data they’re entitled to ask for.

The N.C. Academy of Trial Lawyers is working to change that. The group’s criminal-defense section formed a forensic-science task force this year to train lawyers and help them review scientific evidence.

“People are finally beginning to realize that they have to attack the science just like you would any other witness,” said Diane Savage, the chairwoman of the task force.

The task force has filed complaints with the accreditation board about the way that the SBI laboratory handled scientific evidence in three murder cases. Ralph Keaton, the executive director of the accreditation board, said that the investigation of the complaints is continuing.

Defense attorneys also offer a list of reforms to ensure more neutral and accurate work.

They say that crime labs should operate independent of law-enforcement agencies, which would allow them to share their data and results evenly between defense attorneys and prosecutors and free the scientists from being pressured by either side.

“Crime labs should be an independent force within the criminal-justice system, not beholden to the prosecutor or the defense,” said Barry Scheck, an attorney with the Innocence Project in New York.

However, Robin Pendergraft, the SBI director, said in a recent interview that having the state crime lab as part of the SBI helps her recruit scientists who want a career in law enforcement, and gives her agency flexibility. For example, she can send chemists on drug busts because they are also trained in law enforcement.

“The SBI lab has had a long history of accuracy and fairness to pursue the truth,” she said.

Critics of crime labs also recommend better oversight by independent commissions.

In New York State, for example, the Commission on Forensic Science oversees public crime laboratories. Texas recently adopted a Forensic Science Commission to investigate allegations of misconduct and incompetence in the state’s crime laboratories.

Critics also suggest that crime labs should move to a system of what’s known in medical research as double-blind testing, in which the scientist does not know the source of the evidence that he is working with.

Finally, in North Carolina, the Commission on Actual Innocence is looking into improvements in the storage and preservation of evidence to prevent contamination. The commission was appointed by the chief justice of the N.C. Supreme Court to find reforms that could prevent wrongful convictions.

Science can work both ways.

“Clearly, if you have a scientist saying, ‘We know to a reasonable degree of certainty this is the guy who did it,’ it’s the best piece of evidence we can get,” said Rosen of UNC Chapel Hill and a member of the innocence commission.

“However, if it’s not correct, if it’s fraudulent, it’s the most dangerous piece of evidence because all of us believe in the accuracy of scientific evidence more than we believe in the accuracy of other kinds of evidence.”