How a Denver Rape Probe Got the FBI to Change Policy and Release Kinship DNA
Posted Dec 25, 2006 1:49 AM CST
By Mark Hansen
On Sept. 28, 2003, a Denver woman was raped and beaten so savagely her eye had to be surgically removed. Police had only a vague description of her assailant. And DNA evidence recovered from the victim didn’t match any of the more than 50,000 profiles of convicted felons in Colorado’s DNA offender database.
So Denver authorities ran the rapist’s DNA through the FBI’s National DNA Index System, which contains DNA profiles from more than 3.5 million convicted offenders and nearly 44,000 arrestees nationwide.
That search didn’t produce an exact match, either. But it did yield a close match with a convicted felon in Oregon whose DNA profile in that state’s database was similar enough to the Denver rapist’s to strongly suggest that the two men might be closely related.
With the Oregon offender’s name and date of birth, Denver District Attorney Mitch Morrissey figured, police could determine whether the man had a close relative—a father, a brother or a son—in the Denver area at the time of the rape. If he did, and police could somehow get their hands on a sample of that person’s DNA, it would settle the matter right then and there.
But when Morrissey asked the FBI for the identity of the Oregon felon, he says, the FBI—which maintains the national DNA database and oversees the exchange of information between states—flatly refused. He was told the FBI had a policy prohibiting the release of any identifying information about an offender in one state’s database to officials in another state unless the offender’s DNA is an exact match with the DNA evidence found at a crime scene.
It took some persistence, but Morrissey now has the felon’s name and the FBI has a new interim policy.
FBI crime lab director Joseph A. DiZinno says that, under the new policy, the FBI will leave it to the states to decide whether to release any identifying information about an offender whose DNA profile closely matches a crime scene sample from another state. An advisory board of scientists is now reviewing the interim policy and will recommend whether the change should be made permanent within a few months.
FBI officials attribute the change in policy, effective July 14, to information that has recently come to their attention about the possibility that the DNA database may contain leads that could help solve violent crimes. But they refused to comment on any communications they may have had regarding the database with Morrissey or anybody else.
New Policy, Old Software
While the FBI will notify investigators about close matches it finds using its present software, DiZinno says the agency has no plans to search its crime database using new software being developed for so called kinship or familial searches. The software will allow DNA samples from the relatives of missing people to be compared with the DNA profiles of unidentified remains in an FBI database. The British have been using the technique to identify suspects for years, with some success.
In one celebrated case, British authorities solved a 1988 murder in 2003 when crime scene evidence was closely matched with the DNA profile of a 14 year old boy whose uncle later confessed to the crime. In another celebrated case, DNA evidence on a bloody brick that had been thrown through the windshield of a truck in 2003, causing the driver to suffer a fatal heart attack, was traced to the brother of a man who also confessed. In this country, such a search has borne fruit at least once. In 2004, a North Carolina man was convicted of a 1984 rape and murder after crime scene evidence was closely matched in the state’s database with the DNA profile of his brother. Police tailed the suspect, got his DNA from a discarded cigarette butt and matched that sample with the evidence from the 20 year old rape and murder, for which an innocent man had already spent 18 years in prison.
Proponents of familial or kinship searching say it would help maximize the potential of DNA offender databases. They note that it would effectively increase the size of the database three or more times because every profile that is entered into it would contain information about at least two other people—the donor’s parents, along with his or her siblings and children. They also cite studies showing that convicted felons are more likely to have a close relative who has been incarcerated than members of the general population.
‘Cold Hits,’ Hot Issues
Frederick Bieber, a medical geneticist and professor of pathology at Harvard Medical School who has written about the investigatory potential of kinship searching, says the use of such a technique could increase the rate of “cold hits,” or DNA database matches, by as much as 40 percent.
Although he is not a lawyer, Bieber says he’s not aware of any state statute that prohibits the practice. In fact, he doubts that anybody even contemplated the possibility of such a search at the time the state’s DNA database statutes were written. Bieber also says that every legal challenge to DNA data banking to date has failed because the courts have uniformly held that the interests of public safety outweigh an individual’s right to privacy.
Many legal experts agree.
Henry T. Greely, a professor of law and genetics at Stanford University, says the legal and policy arguments for not doing such searches are that they may reveal family secrets; they may violate promises of privacy or confidentiality to those who gave DNA samples voluntarily; they may invade the suspects’s privacy, particularly if the person hasn’t been convicted of any crime; and they may put people at risk of being investigated simply because they have a relative who has been convicted of a crime. And he says those arguments are all quite weak.
Yet the idea of using DNA from offenders to help catch their relatives is a little disconcerting, Greely says, particularly given the fact that blacks, who constitute about 13 percent of the U.S. population, make up about 40 percent of the people in the national DNA database. That would put blacks as a group under much greater investigative scrutiny than whites.
Greely notes, however, that the racial implications of such a technique—and, in fact, the technique itself—would disappear if a universal DNA identification database existed. Familial searches would be unnecessary except for people who had somehow slipped through the cracks of the all inclusive database. And the database itself would be fully representative of the country’s racial and ethnic diversity.
The legal, practical and political obstacles to the creation of such a database are substantial, Greely says. But it may be preferable to the alternative: a large but racially biased database that may be as unwise as it is unfair.
Greely says he has no strong views one way or the other about whether—or how—kinship searching should be done. But he says it’s something that we as a society need to examine, discuss and decide openly after giving all of the possible implications of such a policy a lot of thought.
Privacy rights advocates cite no legal or constitutional prohibitions to the practice. They say their objections are purely ethical.
Tania Simoncelli, a science adviser to the American Civil Liberties Union’s Technology and Liberty Program in New York City, says near match DNA searches amount to a massive invasion of privacy.
Simoncelli points out that not all people with similar DNA profiles are necessarily related. She also says that presumably innocent people shouldn’t be subject to criminal investigation because they happen to be related to somebody who’s been convicted of a crime. “I think that’s the worst type of guilt by association,” she says.
To those who suggest that kinship searches could catch more criminals, Simoncelli says so might television cameras in every living room, although society would never stand for that. And to say that convicted felons are more likely than others to have close relatives who have also been incarcerated is no justification, either, she says. The same might be said of a convicted felon’s friends.
“That’s a slippery slope I don’t think we want to go down,” Simoncelli says. While privacy advocates are alarmed about the implications of such DNA searches, Morrissey sees them as another tool to help investigators put criminals behind bars. It was a tool he didn’t want to ignore in the Denver rape case, and he wouldn’t take no for an answer.
A Rapist on the Loose
Morrissey knew that the similarities between the Oregon offender’s DNA and the Denver suspect’s didn’t necessarily mean that the two men were related. In fact, Morrissey, a widely recognized authority on DNA technology, estimated there was about a 40 percent chance the two men were unrelated.
But he also had a brutal rapist on the loose, a traumatized victim to answer to, a vicious crime to solve and no other viable leads to pursue. And the FBI’s policy made no sense to him.
As far as Morrissey could tell, there were no legal or constitutional barriers to the release of such information. Any family secrets that might be revealed through such a comparison could be kept confidential. The FBI had no corresponding policy prohibiting states from doing the same type of search of their own databases. And law enforcement officials basically do the same thing with fingerprints all the time.
“This is just another potential lead,” Morrissey says of the Oregon offender’s identifying information, which he likens to having a vehicle description and a partial license plate number in a hit and run case. It might go nowhere. It might lead directly to his rapist. But how, he asks, “can I not follow up on a potential lead?”
Morrissey says he argued the point with crime lab director DiZinno and DNA database chief Thomas Callaghan for more than seven months. But his arguments fell on deaf ears until he sat down one day in late June and wrote the two FBI officials a letter describing the situation as he understood it.
Colorado had a new law, set to take effect on July 1, requiring law enforcement officials to update the victims of unsolved violent crimes each year on the status of their case, Morrissey told the two officials. Then he informed them he was planning to meet personally with the rape victim who had lost an eye and let her know that the FBI was refusing to turn over information that could lead to the arrest and conviction of the perpetrator.
“Your policy is not likely to sit well with [the victim] when she understands your roadblocks to our investigation of her case,” Morrissey says, quoting from what he says were the pertinent parts of the letter. He also said there was a good chance the woman, whom he described as “quite assertive,” might take a “proactive approach” to the news. “I cannot predict whether she will choose to share this information with other public officials and the media,” Morrissey concluded.
Within a day or two, Morrissey says, he heard directly from FBI Director Robert S. Mueller, a former U.S. attorney, who told him he fully understood his position on the matter and promised a change in policy.
Two weeks later, the bureau had what it describes as a new interim policy regarding the release of information about partial DNA matches across state lines. And Morrissey had the name and birth date of the Oregon felon whose relative may have brutally raped the woman in Denver three years ago, a lead that his investigators are now pursuing through more conventional means.
Morrissey says FBI officials would have the public believe that partial DNA matches are quite rare. But he says he suspects it happens far more often than the FBI is willing to admit.
His staff is only about halfway through a project that involves retesting DNA evidence in about 1,000 unsolved cases, Morrissey says, and has already turned up two more near matches with offenders in other states. In one case, Arizona officials have already agreed to turn over the name of a man in that state’s database whose DNA is a close match with a Denver rape suspect after doing additional tests that will help confirm whether the two men are related. In the other case, California officials have only recently identified a man in that state’s database whose DNA is a near match with another suspected Denver rapist.
And Morrissey says he has since learned about another case in which DNA evidence from the scene of a Florida murder has been closely matched with that of an offender in Michigan, although he says he doesn’t have any further information about that case.
He credits FBI Director Mueller with getting the old policy changed. “He’s the hero in all of this,” Morrissey says. “He’s the one who got the ball rolling.”
Mark Hansen is a senior writer for the ABA Journal.