A long-awaited federal report on a powerful new genetic technique for identifying criminal suspects says it should not be allowed in court in the future unless a more scientific basis is established.
In principle, the technique, known as DNA fingerprinting or DNA typing, allows a person to be identified from the tiniest scraps of body tissue at the scene of a crime, whether a drop of dried semen, a strand of hair or a spatter of blood. The technique relies on the uniqueness of each person's DNA, the genetic material.
The new report, by the National Academy of Sciences, says that courts should cease to admit DNA evidence until laboratory standards have been tightened and the technique has been established on a stronger scientific basis.
The report says that though the forensic technology is valid in principle, serious questions have been raised about the way it has been used. It adds that the method is potentially too powerful and too important for its development and use to be left solely in the hands of prosecutors and law-enforcement officials.
Instead, the report says, it must be regulated and controlled by scientists and federal agencies that have no stake in the method's success or failure.
The report, by a 12-member panel of scientific and legal experts headed by Dr. Victor A. McKusick, a geneticist at Johns Hopkins Hospital in Baltimore, is to be issued by the academy on Thursday.
Ever since DNA typing was introduced in U.S. courts in the late
1980s, it has been the subject of bitter disputes. The technique has been vigorously embraced by some prosecutors because of its scientific basis and persuasiveness to juries. But defense lawyers have often fought it because they fear it could persuade jurors to overlook any other evidence.
Most courts have agreed to admit DNA evidence, and hundreds of cases that relied on DNA fingerprinting have been tried. But critics have argued that the method is not as conclusive as it appears and that it is not yet ready to serve as evidence in court.
Some legal specialists said that the report would have an immediate impact. "Any defense lawyers who are worth their salt will ask appellate courts to take a look at this," said Dr. William Thompson, a lawyer and molecular biologist at the University of California at Irvine. He said the report "raises serious concerns about whether DNA tests as they are currently done meet the standards for admissibility in court."
Dr. P. Michael Conneally, a geneticist at Indiana University who has testified for the prosecution in DNA cases, said that he saw no problem with the way DNA evidence had been acquired or used and that the committee's precautions were unnecessary.
The academy formed the committee two years ago as disputes over DNA typing were getting more and more heated. The panel included scientists, a judge and law-enforcement officials.
The committee's report accepts the scientific basis of DNA typing: that each person's DNA is unique, that DNA patterns can be detected in the laboratory and that DNA can be used to establish whether cells from a crime scene are likely to have come from a suspect.
But, the report said, "before a method can be accepted as valid for forensic use, 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."
The group cautioned against overselling DNA typing. "DNA evidence is not infallible," the report said. "All laboratory work is subject to error; and, given current population data banks and laboratory protocols, a witness or prosecutor will seldom (if ever) be justified in stating that the probability that a reported DNA match involves someone other than the suspect is so low as to make the possibility entirely implausible. Claims that treat DNA identifications as though they are as reliable as fingerprint identifications in the typical rape or murder case are unjustified; until technology and data banks improve, they are likely to remain so."
Paul Gianelli, a law professor at Case Western Reserve University in Cleveland who studies science in the courtroom, said he expected that the report would set off fundamental changes in forensic science in general.
"I think this is going to lead the way," he said. "It will be very difficult to say that if you have proficiency testing for DNA that you shouldn't have it for gunshot residues or fingerprints or firearms identification or drug cases. DNA is a more complicated procedure. If you can do it for DNA, you can do it for the others."
One problem, the panel said, is that forensic laboratories have not been required to show proficiency or even to publish their methods. But, the report says, the sketchy evidence of laboratory errors in DNA typing that has emerged indicates that as many as one time in 50 cases a lab can blunder so badly that an innocent person would seem to be guilty.