Crimes go unsolved as DNA tool ignored
Genetic profiles in rapes, slayings not sent to FBI
By Maurice Possley and Steve Mills; Chicago Tribune
October 26, 2003
On the same day he pardoned Lesly Jean, a former Marine who had served nine years for a rape he did not commit, North Carolina Gov. Mike Easley asked law-enforcement officials to reopen the case in hopes of finding the real rapist.
Easley asked authorities to begin by submitting the DNA profile of the victim's attacker to state and national databases to compare it with the genetic profiles of more than 1.5 million convicted criminals, an investigative step that requires just a few keystrokes on a computer.
Today, almost three years later, law-enforcement officials concede the profile was never sent. The investigation was not reopened. The case remains unsolved.
"I don't understand it," said G.Dewey Hudson, district attorney for Onslow County, where the rape occurred. "Damn, there is a real rapist who could be committing more crimes."
In the past 15 years, DNA testing has emerged as a revolutionary forensic tool, providing an unprecedented exactitude in identifying and convicting criminals and freeing prisoners convicted of crimes they did not commit.
Yet a Tribune investigation shows that, as in the Jean case, law-enforcement authorities are still struggling--and in some cases refusing--to fully exploit DNA technology.
The newspaper examined every case in the U.S. where DNA testing has freed a convicted inmate, focusing on 115 murders and rapes where the release left a crime unsolved. In some of those cases, DNA was quickly used to link known suspects to the crime.
But in 44 of the remaining 97 cases, or just under half, authorities have not followed up by submitting the genetic profile of the suspected perpetrator to the FBI's national DNA database, a well-established law-enforcement tool that has been used to link more than 8,600 convicted felons to other crimes.
"There's absolutely no excuse for not putting the sample into any database to find out who did it," said Robert McCulloch, president of the National District Attorneys Association, which represents more than 2,300 chief prosecutors nationwide. "You don't want the wrong guy in prison and the right guy walking the streets."
The failure to seek a DNA match is all the more surprising given that in the cases where DNA was submitted, genetic profiling identified the real criminal more than 40 percent of the time.
Sometimes there are legal bars to prosecution because the crimes are too old, or DNA profiles developed by outdated methods are not retested and sent to the database.
But, the Tribune review found, the search for the real culprit also has been thwarted by bureaucratic infighting that bars from the database test results from one of the nation's premier DNA scientists.
Further, law-enforcement authorities have refused to submit DNA profiles because they continue to believe that freed defendants are guilty.
Robert Pino, director of the Massachusetts State Police crime lab, acknowledged that some law-enforcement officials do not want to submit DNA profiles from exoneration cases to databases because they have difficulty believing the wrong person was convicted.
"Putting a profile into a database is part of an effort to move the immovable force--the police officer--toward the truth," Pino said.
The sight of inmates walking free, out of prison gates or down courthouse steps, after DNA tests prove their innocence has become almost commonplace. But following these exonerations, there has been little scrutiny of whether the original crimes are ever reinvestigated.
Refusing to believe
Most troubling are those cases where DNA clears a defendant of a conviction, but law-enforcement officials refuse to consider him innocent or to pursue new suspects through database matching.
In February 2002, Bruce Godschalk, convicted in 1987 of two rapes in a suburb of Philadelphia, was exonerated by DNA in tests performed by two laboratories--one hired by prosecutors and one by defense attorneys.
Ultimately, Godschalk was released after Montgomery County District Atty. Bruce Castor Jr. decided not to take him to trial again.
"I came to the conclusion that he could not be proven guilty beyond a reasonable doubt," Castor said in an interview.
Castor said the main obstacle to prosecuting Godschalk again was the DNA test results. But, he said, "There is an equally huge body of evidence" against Godschalk.
He said that one of the two victims still insists Godschalk was her attacker and that Godschalk gave a 33-minute audiotaped confession. Castor also said the DNA tests in one case were misleading because one victim contended Godschalk had not ejaculated, and in the other case, the genetic material tested was found on a carpet, not on the victim.
Castor will not say that Godschalk was exonerated.
Recently, Montgomery County officials agreed to pay Godschalk $740,000 to settle part of a wrongful-conviction lawsuit. Godschalk is still suing detectives, alleging they coerced him to confess.
Faced with DNA results that undermine their conviction, police and prosecutors often point to evidence such as witness testimony or a disputed confession to make the case that they got the right man. But the DNA sometimes forces them to change their theory of how the crime was committed.
Charles Irvin Fain was convicted in 1983 in Idaho of the abduction, rape and murder of a 9-year-old girl and sentenced to death. At his trial, an FBI expert said three hairs found on the girl's body were similar to Fain's.
But after DNA testing proved the hairs were not Fain's, the conviction and death sentence were vacated in 2001 and he was released from prison. Although authorities had always contended that Fain acted alone, Canyon County Sheriff Gerald Nourse said Fain remains under investigation.
"He could still be involved," Nourse said. "The hair could have belonged to an accomplice."
Prosecutors in Jefferson County, Alabama, still believe Ronnie and Dale Mahan are guilty of abduction and rape, even though both men were released from prison after DNA tests.
The Mahan brothers were arrested after an 18-year-old woman told police in 1983 that she was kidnapped from a Bessemer shopping mall by two men wearing stocking masks. She said the two drove her to a wooded area, where she was forced to smoke marijuana, then was raped four times by one man while the other held her down.
Six weeks after the attack, the victim identified the Mahan brothers as her assailants, saying she recognized them because both had lifted their masks during the attack. Both men were convicted and Ronnie Mahan got a life sentence while Dale Mahan was sentenced to 35 years in prison.
Eleven years later, DNA tests excluded both men as the source of the semen recovered after the assault, and they were granted a new trial. At that point, authorities returned to the victim, and she gave a new version of events to explain why the semen did not come from either defendant. She said that she had sex with her husband earlier that day, something she had not disclosed for more than 13 years.
So prosecutors changed their theory to one in which the rapist had not ejaculated--a tactic frequently taken by prosecutors when confronted with such an evidentiary situation, according to the Tribune review.
But the case became more complicated when DNA tests excluded the husband. That's when the victim provided a third version, saying that she had sex with a boyfriend that day. And indeed, the genetic profile from the boyfriend, who had since died, matched the profile from the evidence in the assault case, authorities said.
Although the woman's credibility had been severely strained, prosecutors began preparing to retry the Mahan brothers and sent a hair recovered from the victim's clothing for DNA tests.
DNA testing of hair is less precise than DNA testing of fluids. It has the power to exclude someone but not to identify an individual with the same degree of accuracy.
That is the primary reason the FBI will not accept DNA profiles obtained from hair for comparison in its database.
The test results on the hair came back with a genetic profile that was not the husband, not the boyfriend, and not the Mahan brothers. The more authorities worked to find evidence to convict the defendants, the more the case fell apart.
Finally, with a defense lawyer arguing that the hair must have belonged to the rapist, the case was dropped and the Mahan brothers were released, maintaining they had been exonerated.
But that's not what authorities say.
Jefferson County Assistant District Atty. Arthur Green said he wanted to take the case to trial again with the revised theory that the attacker had not ejaculated, but the victim did not want to testify. Green believes that the hair had nothing to do with the attack and that the Mahan brothers were involved.
"These sons of bitches are guilty as sin," Green said in an interview. "There's no question in my mind. This is not a case of innocence. ... These two bastards are guilty. I just can't prove it."
Genetic `fingerprint'
Though the existence of DNA (deoxyribonucleic acid) has been known for more than 100 years, the first profiling test was not developed until 1985 in England by Sir Alec Jeffreys. A year later, the test was used to clear a suspect and pinpoint the perpetrator in a double murder in the English Midlands.
DNA tests, which compare the unique genetic profiles of individuals, can discriminate between billions of people and conclusively tie a person to a crime.
In 1990, the FBI established its database, known as the Combined DNA Index System, containing genetic profiles from unsolved crimes as well as from convicted offenders, primarily men convicted of sex offenses. It began as a pilot project with 14 laboratories connected by computer to find and compare DNA profiles, much like the Automatic Fingerprint Identification System that contains millions of prints. The database became operational nationally in 1998.
The convicted offenders database has grown considerably as many states, including Illinois, have broadened the group of prisoners required to provide a sample of their DNA.
Virginia was the first state to develop a database and remains a leader in the field, last year reporting its 1,000th match of a convicted offender to a genetic profile developed from evidence at a crime scene. Paul Ferrara, director of the Virginia Division of Forensic Science, said more than 350 of the crimes solved were murders and rapes.
By 2003, the FBI database contained profiles from 65,868 unsolved cases and 1,507,278 convicted offenders, according to the agency. Since the inception of the FBI database, there have been 8,675 DNA matches.
Only two states, Mississippi and Rhode Island, are not connected to the FBI database, and those states are being certified.
The first DNA exoneration in the United States occurred in Chicago when Gary Dotson was cleared of the 1977 rape of Cathleen Crowell Webb. Dotson had been convicted in 1979, but the case began to unravel six years later when Webb came forward to say she had concocted her rape allegation because she feared a boyfriend might have gotten her pregnant.
In 1988, defense lawyers for Dotson hired forensic scientist Ed Blake, who performed the DNA tests that corroborated Webb's recantation, and Dotson was freed a year later. Police were left with no crime to investigate because Webb insisted there was no rape.
In the ensuing years, Blake, head of Forensic Science Associates in Richmond, Calif., has come to be recognized as one of the nation's most sought-after DNA testing experts.
He has been responsible for more than 40 post-conviction DNA exonerations in the United States and has played a key role in exonerations in Canada as well. In these cases, he often has worked with Barry Scheck and Peter Neufeld of the Innocence Project at the Benjamin Cardozo School of Law in New York City, a non-profit organization that has been involved in more than half of the nation's DNA exonerations.
During that same time, Blake also has performed work for prosecutors more than 100 times, with his tests confirming the involvement of defendants and aiding the prosecution case.
Scientist locked out
Despite Blake's reputation as one of the best DNA scientists in the world, and although his test results are consistently accepted by judges and prosecutors as accurate, federal legislation prohibits the FBI from putting his profiles into its database.
FBI lab director Dwight Adams said in an interview that legislation creating the database set out strict requirements proposed by a panel of experts that must be met before a DNA profile can be submitted for comparison.
The legislation allows the acceptance of profiles only from accredited government-funded labs or accredited private labs on contract to a government-funded lab.
"These are laws which Congress has enacted for us to follow," Adams said. "It is not our choice that any profile can or cannot be put into the database."
Blake refuses to be accredited by the American Society of Crime Laboratory Directors-Laboratory Accreditation Board, which inspects and audits labs and provides the certification the FBI accepts.
Blake said his credentials should stand on their own and, citing problems and scandals in accredited labs, maintained that the outside review panel's stamp of approval is meaningless.
"Just because a lab has credentials doesn't mean it produces accurate work," he said. "What the FBI and crime laboratory managers are trying to do is say that the only thing that we will trust are those people we tell you we should trust."
Virginia lab director Ferrara knows Blake and respects his work. "Just because you're not accredited doesn't mean your data isn't fine," he said. "But it's the law."
The Tribune review identified 12 cases where Blake developed a DNA profile that led to the release of a wrongly convicted defendant, but the profile of the perpetrator has never been submitted to the FBI database--raising the possibility that a dangerous criminal remains free or is in prison awaiting release.
The statute of limitations has expired in eight of those cases, but four could still be prosecuted if the true assailant could be identified.
Some of the cases involving Blake-generated profiles have ultimately been submitted to the FBI database, but only after the evidence in those cases was retested by an FBI-approved laboratory that duplicated Blake's results.
One of the 12 Blake cases that has not been submitted to the database occurred in West Virginia, where Glen Dale Woodall was convicted in 1987 of kidnapping and raping two women and sentenced to two life terms in prison, plus 325 years. In 1992, Blake was hired to perform DNA tests, and the results identified a profile other than Woodall's, prompting his release.
West Virginia crime lab director Ted Smith said he has asked law enforcement to submit the evidence for retesting, but nothing has been sent.
Cabell County sheriff's investigator James Scheidler said he continues to investigate the case. "I am the sole investigator and until the day I leave here, I won't stop looking," he said. "I would love to submit it to the lab."
Scheidler said he has asked Cabell County Prosecuting Atty. Christopher Chiles to authorize submitting the evidence to the lab for retesting, but more than a decade later, Scheidler is still waiting for the test to be conducted.
"In my opinion, he feels that Glen Woodall is guilty," Scheidler said. "He says he is still thinking about it."
Indeed, Chiles said the DNA "did not necessarily" exonerate Woodall. But last week--11 years after Woodall was released--Chiles said he would ask that the evidence from the case be sent for tests "out of an abundance of caution."
The Tribune review identified 16 of Blake's cases where authorities said profiles were ultimately submitted to the FBI database. In 12 of the cases, there were no matches. In three of the cases, the profiles were matched to new suspects, but no charges were brought because the statute of limitations had expired.
In the last case, the DNA profile from a murder case was matched to a man serving a prison sentence for rape, and he was charged with the murder.
Blake's cases aren't the only ones that remain dormant over lack of lab accreditation. For example, three other crimes--a murder and two sexual assaults--are still unsolved in West Virginia. The DNA profiles that freed defendants in those cases were developed by private labs that lack FBI approval.
"It's a question of whether the test results meet the criteria for entering into the database," Smith said. "I don't think any of them has met the criteria."
A case unfolding in Philadelphia provides the most recent example of the fallout from this bureaucratic freeze-out of labs whose work is accepted in court to free a defendant, but not by the FBI for submission to its database.
Nicholas James Yarris, who was sentenced to death for the 1981 abduction, rape and murder of Linda Craig, a 32-year-old woman from Boothwyn, was recently granted a new trial and his death sentence was set aside after DNA tests performed by Blake eliminated him as the source of evidence.
Earlier this year, by agreement of Yarris' lawyers and Delaware County prosecutors, the evidence in the case--gloves found in the victim's car, which prosecutors said were the killer's, a swatch of cloth from the victim's underwear, and scrapings from the victim's fingernails--was sent to Blake for testing.
Blake identified two DNA profiles from the victim's underwear, neither of which were Yarris' or the victim's husband. One matched a profile developed from skin cells from the gloves as well as a profile found in the fingernail scrapings.
The profiles cannot be submitted to the FBI database because they were developed by Blake. Assistant District Atty. Sheldon Kovach conceded, "There is a problem. We are hoping we can check it against the Pennsylvania State Police lab database. I have an optimistic view that we can at least get that done."
There is no doubt that "the real killer is out there," said Michael Wiseman, one of Yarris' attorneys. "We have two DNA profiles that are not Mr. Yarris.
"That is an outrage," Wiseman said. "The real killer might be in some databank. As a citizen who lives in the same county where this crime occurred, I'd like to know who the killer is."
`A no-brainer'
When a defendant is cleared by DNA tests, and a crime once considered solved becomes an open investigation, the responsibility for submitting the profile to a database falls on the shoulders of law enforcement. But, as the Lesly Jean case demonstrates, there frequently is poor communication between prosecutors and police about which agency should make sure the genetic profile is entered into a database.
In Jean's case, after Onslow County District Atty. Hudson expressed shock that nothing had been done, Jacksonville Deputy Police Chief Sammy Phillips conceded that police had failed to reinvestigate the case, despite Gov. Easley's request.
"We're not making excuses by any means," Phillips said in an interview. "Putting it into the database--that's a no-brainer."
A spokesman for Easley said the governor was unaware of the county's inaction and would renew his request.
Though failure to take full advantage of DNA is sometimes an oversight, there also can be technical reasons for not submitting a sample.
For example, the DNA profile that exonerated Steven Linscott of murder in Oak Park and led to Linscott's release was obtained using an older method of testing that cannot be submitted to the database.
Oak Park police said that they are stymied in pursuing that avenue because no evidence remains that could be tested with the method accepted by the FBI.
And authorities in Kentucky said the DNA profile that exonerated William Gregory of a rape was obtained through mitochondrial testing of hair--a method that is not compatible with the FBI database. Further, authorities in Louisiana said the DNA profile that freed Calvin Willis was sufficient to rule him out as the perpetrator of the rape for which he had been convicted, but it is not compatible with database requirements that mandate a certain number of identifying markers.
In cases where the tests must be redone because the profile is outdated, these tests, even if requested, frequently are given low priority in already backlogged crime labs, translating into delays of months, even years.
Race against clock
Another reason that prosecutors and police cited for not submitting a profile to the database after a DNA exoneration is the expiration of the statute of limitations.
In most states, rape cases carry a time limit for prosecutors to bring charges, and this can be as short as three years. There is no statute of limitations for murder in any state.
The Tribune review shows that of the 44 cases where the profiles were never submitted to a database, 21 could not be prosecuted at the time of exoneration because the statute of limitations had expired, and the statute has since expired in five others.
That leaves 18 cases where DNA testing still could help find the real assailant because the statute of limitations has not expired. Yet the genetic profiles have not been submitted to the database.
Some prosecutors disagree with colleagues who believe the expiration of the statute of limitations is a good reason to sit on DNA evidence.
"That is a stupid answer," said Marie Munier, a prosecutor in Houston. "A hit might be good for somebody else. It could always be used as punishment evidence in a new case that could be brought against a defendant.
"Maybe you can't prosecute someone for the crime where the statute has expired," she said, "but if a person is out committing new crimes, you could use it to enhance a sentence for another conviction."
McCulloch, of the district attorneys association, agreed. "I don't think the expiration of the statute of limitations is a good excuse for not putting a profile into a database," he said. "That profile may hit on another case--an unsolved case where the statute has not expired. Rapists tend not to rape just one time."
He added that if there is a database match and there are no other cases that could be prosecuted, "at least you know you have a rapist on your hands, you know who he is, and where he is, or at least you should know where he is."
In 2000, after A.B. Butler was exonerated of kidnapping and rape by DNA and released in Smith County, Texas, the DNA profile of the rapist was never sent to a database. Edward Marty, a Smith County assistant district attorney, said they could have done it, but did not.
The statute of limitations had expired, the victim still insisted Butler was her attacker, and, in part, prosecutors were simply reluctant to submit the profile to the database.
"There was even a little, `Well, maybe we don't want to know,'" Marty said.
Several states, including Illinois, have enacted legislation to allow for longer statutes of limitation in cases where DNA evidence is involved.
Attempts by some states to lengthen statutes of limitation retroactively were struck down this year by the U.S. Supreme Court. The ruling set aside a number of convictions in California, which had enacted a law to allow for prosecuting out-of-date cases that involved newly discovered DNA evidence.
The ruling applied to all states, such as Oklahoma, where under a similar law, authorities had indicted Edward Alberty last year in a 1987 rape, even though the statute of limitations had expired in 1994.
Arvin McGee Jr. had been convicted of the rape in 1989 and sentenced to 398 years in prison. When DNA testing exonerated McGee last year, the genetic profile from the evidence was submitted to the FBI database, authorities said, and it was matched to Alberty, who was then indicted.
But after the Supreme Court ruling, the charges against Alberty had to be dropped.
Other states, such as New York, are racing to file charges before current statutes of limitations expire by obtaining grants to test evidence in unsolved cases and indict specific DNA profiles when the identities of the individuals are unknown.
This year, in Wisconsin, authorities won an important legal victory in their use of such DNA profiles. In this case, a rape suspect was identified and convicted through DNA after the statute had expired. The Wisconsin Supreme Court in April upheld use of a nameless genetic profile to get an arrest warrant on file in the case before the statute of limitations expired.
In September, members of Congress predicted swift movement on a bill designed to provide greater access to DNA tests for Death Row inmates and in rape cases. More than $750 million is earmarked to clear the backlog of what some authorities estimate to be several hundred thousand untested rape evidence kits.
Retesting pays off
The 1984 murder of 9-year-old Dawn Hamilton is the most recent example of the value of submitting DNA profiles to databases.
Kirk Bloodsworth, an ex-Marine security guard, was convicted of the girl's murder and sentenced to death. Nearly a decade later, he was exonerated when he was excluded as the source of semen found on the girl's clothing.
Pardoned in late 1993, Bloodsworth went about the task of putting his life together, while Baltimore County police were left with an unsolved murder. Over the years, police identified alternative suspects and their DNA was compared with the profile from the evidence, but there was no match. The profile was never sent to either the Maryland database or the federal database because it had been prepared by Ed Blake.
In vain, Barry Scheck, attorney for Bloodsworth, begged Baltimore County authorities for more than five years to send the evidence to be retested to obtain a profile that the FBI database would accept.
"At the heart of this case, and a lot of others, is the issue of pride," Scheck said. "There is the feeling that they have already let criminals go free and in many of these cases they prefer to say they don't know if the released individuals are guilty or not, rather than try to find out who the real criminals are."
Ann Brobst, an assistant state's attorney who prosecuted Bloodsworth, said in an interview that she also had been urging police to send a profile to the database.
Police said that nothing was done until this year because they were waiting for the additional funding to process DNA evidence on cold cases. Finally, in the summer, a new profile was obtained from an accredited lab and submitted to the database.
In September, authorities announced new charges in the case, after the profile matched a prison inmate who had been convicted of rape after Dawn Hamilton's murder.
"I cannot accept the proposition," Blake said, "that a crime lab would not investigate the rape and murder of a child for which an innocent man has been convicted and sentenced to death until they receive a grant."
William Sessions, a former FBI director who is a member of the Constitution Project's Death Penalty Initiative, noted that for a decade after the DNA exoneration, prosecutors in Baltimore County still considered Bloodsworth their chief suspect. He said the case is a "vivid example" of the need for law-enforcement authorities to embrace DNA testing technology more actively.
"I don't think that prosecutors should stand by and say, `I don't want to look at this'--to stand by and say, `We had a conviction,'" Sessions said. "Why isn't it a priority to actually go back and investigate?"
FINDINGS
A Tribune analysis of 115 rape and murder cases in which inmates were released after DNA testing did not link them to the crime revealed:
18 cases in which new suspects were identified at the same time DNA profiles exonerated defendants.
23 cases in which new suspects were identified after DNA profiles were submitted to a national FBI database.
30 cases in which DNA profiles were submitted to the database, but no new suspects were identified.
44 cases that authorities have not followed up on by submitting DNA profiles to the database.