Why Criminal DNA Records Should Be Public

New technologies in law enforcement are often mismanaged, with the result being waves of collective fear. The FBI released a report in January about the National DNA Index System (NDIS), which concluded that 166 of nearly 13 million DNA profiles in their database contained errors. As noted by the Innocence Project, there have been “312 post-conviction DNA exonerations in United States history.” One might think 166 seems like a small number compared to 13 million, but some scholars involved in DNA research cast doubt on the numbers completely.

The idea that DNA was a building block of life was first considered by Friedrich Miescher in 1869. Over a century later and after years of study and experimentation, DNA was eventually used to convict a criminal for the first time in 1987. A Florida court found Tommy Lee Andrews guity of rape based on semen samples taken from the victim.

But cases like that of Damon Thibodeaux, a Louisiana man exonerated by DNA evidence after 16 years in prison, cause concern among DNA experts. Thibodeaux was sentenced because he confessed to killing 14-year-old Crystal Champagne, after eight and a half hours of interrogation. He initially denied committing the murder.

Cases like this bring up many questions: How many more cases could be altered by new DNA evidence that we’re not yet aware of? How many people have already died in jail from lack of DNA evidence? How well is DNA evidence being taken care of and preserved that could accomplish those goals?

I spoke with Dr. Dan Krane, a biology professor at Wright State University who consults for defense lawyers in cases involving DNA evidence. Krane was the lead author, with 40 other scientists, of a letter to the editor of Science demanding that academics be given access to the FBI’s DNA database, over the objections of the FBI, which has said such access would violate people's privacy.

In the letter, published in 2009, Krane asserts that the FBI, "which controls the database, has published no research derived from NDIS and has declined to disclose these records to academic scholars. The National Research Council recently noted that 'methods developed in crime laboratories to aid in law enforcement' would benefit from the contributions of academic scientists. We believe the time has come for the FBI to release anonymized NDIS profiles to academic scientists for research that will benefit criminal justice.”

He argues that the federal government is constantly asking American citizens to surrender their rights to privacy in the guise of protecting them, but, “this is the only instance of which I’m aware, and I’ve asked this of a whole bunch of legal scholars … where it is the federal government that is a proponent of an individual’s privacy rights.” He notes that there is no need for informed consent, where the DNA holder would have a legal standing against their profile being observed, when the person is a convicted felon in the database, as the vast majority of those in the database are.

"As most research scientists know well, the government frequently releases sensitive information under controlled conditions to verified researchers," Krane wrote in his letter. "Even within the criminal justice context, law enforcement officials have made available data about the age, race, gender, geographic residence, and a wide range of other information about criminal offenders so that researchers can conduct studies aimed at improving and enhancing effective law enforcement.”

The FBI's interest in privacy in this scenario may be virtuous, but critics say it raises eyebrows too. It is conceivable, Krane argues, that the FBI is aware of serious flaws in its system and does not want those to be known, as they could destroy the use of DNA in investigations completely. “It gives them the opportunity to have some high ground,” he told me.

Since the profile cannot be viewed by those that could cast scrutiny on it, and possibly find errors, the number of errors presented in the report may not be accurate. Concerning the claim that there were only 166 errors, Krane says, “That strikes me as a pretty low number.” He referenced findings from Arizona's Department of Public Safety that found a slew of possible errorsin their database, which is a subset of the FBI's database. "At the very least, we know there were 166 missed opportunities to take a suspect off the streets."

In a study done on the DNA database in Victoria, Australia, it was found there was an error rate of one in 300 profiles. In the United Kingdom, a database discovered 1,500 administrative mistakes in 2007.

Even the statement that there are nearly 13 million profiles in the United States’ database is questionable. Since few have access to the database, it may be a grandiose number used to exaggerate the difference between the errors and the total amount.

Krane says that defense lawyers in cases involving DNA are almost never given access to DNA records that they could use in their defense. If a defense lawyer could access DNA profiles and learn that other samples were collected from the same crime scene, they could assert that the other profiles might be possible suspects as well. He also notes, however, that letting a defense attorney bring up such new concepts might give them too many opportunities to redirect the case.

The errors themselves are often simply human error. Illegible handwriting and mistakes made during the testing of DNA can cause such miscalculations. The possibility of two people’s DNA being indiscernible being a statistically unlikely event is not a good defense when there are errors made before the DNA evidence is even presented. Krane says that in many cases there are one or two locations in the DNA, or loci, that tend to offer less accurate information than the others. He claims that of the 13 loci used, for the benefit of statistical defense of using DNA, it might be better to search databases using 10 or 11 that are more reliable. “We’ve known for about 15 years now that some of the loci are less likely to give you a complete set of information compared to others,” he says.

Krane also warned against the use of what he calls a “John Doe warrant.” That's when a suspect is searched out based on matching found DNA, as opposed to their name, address, or other identifiers. In a 2010 research paper for Stanford, law expert Kelly Lowenberg asserts:

Not every DNA sample will be a sufficient basis for probable cause to obtain an arrest warrant. For example, DNA in semen found at a rape scene is more inculpating than a hair found at the scene of a burglary. But if a warrant is obtained, the statute of limitations will be satisfied. And a person could be prosecuted for the crime, regardless of how much time has passed, which may make preparing a defense more difficult.

According to the Council for Responsible Genetics, DNA could be used to identify people involved in a “peaceful protest or dissent,” if the government has a large enough database. Found hairs or tissue could identify recent occupants of a certain area. In the United States, DNA samples have increasingly been procured by those arrested for a crime but not yet convicted. When it came into wide used in the 1990s, DNA evidence seemed to be a blessing to the criminal justice system. It's been the final piece of the puzzle in convicting many treacherous criminals. Even critics of the current system hope to see the continuance of the use of DNA in the courtroom. But those critics are also very aware of the existing flaws in the system to let it lie unchallenged.