Document Type


Publication Date



DNA evidence is often presented as the “gold standard” for forensic science. But this was not always the case. For years, eminent scientists complained that the estimates of the tiny frequencies of DNA types were unfounded. It took scores of research papers, dozens of judicial opinions, and two committees of the National Academy of Sciences to resolve the dispute by the mid-1990s. Since 2000, however, reports have surfaced of shocking numbers of “partial matches” among samples within large DNA databases, and some scientists have complained that the infinitesimal figures used in court to estimate the probability of a random match are no better than alchemy. To study the partial-match phenomenon further, defendants have sought to discover all the DNA records (with personal identifiers removed) kept in offender databases. The FBI has responded by branding the proposed research as useless and the release of the data as an illegal invasion of privacy. The media have reacted by calling for congressional hearings and, possibly, criminal charges against FBI officials.

This Article reviews the existing research findings and considers the scientific, legal, and ethical objections to disclosure of the DNA data. It concludes that the arguments against further research are unpersuasive. At the same time, it finds that the claims of dramatic departures from the expected numbers of partial matches are exaggerated and predicts that new research will not reveal unknown flaws in the procedure for estimating the chance of a match to an unrelated individual. In view of the importance of DNA evidence to the criminal justice system, this Article recommends using the databases for more statistical research than has been undertaken so far. It also calls for dissemination of the anonymized records for this purpose.


This article was originally published at 19 Cornell J. L. & Pub. Pol'y 145.