Back to Main Page

By Gary T. Marx

The Wall Street Journal, April 20, 1998

If forensic scientists had been in on the Creation and been asked to develop an ideal system of personal identification, they might have recommended one in which the identification would be indelible, unalterable and --unlike an ID card --part of the individual. Every cell in the body would be marked with the same bar code; no two individuals could have the same code. Identity could be inferred form minute samples of specimens such as blood, semen, or hair follicles that might unwillingly be left at the scene of a crime. Material that could be analyzed even years after the crime occurred and identity could be determined in a way that didn't invade privacy or require the individual's cooperation.

In fact a system with these characteristics --DNA "fingerprinting" --is now available and its use is spreading rapidly. California and Massachusetts are among a number of states considering plans to create a computerized generic data base using blood and saliva samples of all those convicted of violent crimes. The FBi is seeking to develop a standard national classification system and is helping states adopt the technology. Several congressional committees have held hearings recently on the issue of a national standard.

The technology offers many advantages. It can provide new means of identifying the guilty and of freeing the innocent: In a recent British case, a suspect who confessed to a rape was later shown conclusively not to be the culprit. It can greatly help in determining paternity and maternity. And it's likely to be particularly useful in rape cases since it offers a much more powerful technique for identification than is currently available.

Yet it's necessary to proceed with caution. New technologies that provide answers also raise new questions. Which of several competing types of analysis should become the standard? What statistical confidence levels and laboratory standards should be set for the legal admission of DNA evidence? At what rate do biological specimens deteriorate and how does this effect test results? Is it possible to frame someone by leaving his DNA sample at a crime scene?

Who should control information about DNA patterns, and when should individuals be required to reveal such information about themselves? Should DNA identifying information receive the same privacy protections as medical records, or no more protection than a home address? Would it violate the Fourth Amendment to use blood collected in a routine medical examination in a later DNA match, if there are no grounds for suspecting the individual?

Current proposals call for the mandatory provision of a DNA sample --and creation of a data bank --only for those convicted of violent crimes. But through a process of "surveillance creep," will this spread to those convicted of non-violent crimes? And will mandatory genetic "fingerprinting" eventually be required of the population at large? There are other justifications for creating a DNA record on the entire population. Missing children, or amnesia, Alzheimer's and homicide victims could be easily identified.

Once DNA analysis comes to be seen as a familiar and benign crime control tactic, will the way be paved for more controversial uses--for example denial of certain types of employment or insurance, or even the right to have children in those whose genetic makeup indicates they may be prone to particular illnesses or forms of anti-social behavior? It's possible a numerically expressed national standard DNA pattern could find its way onto the hundreds of documents that make up one's "data image" in computers. Perhaps it will replace the Social Security number that now (against Congress's original intentions) serves that function.

Police and prosecutors welcome the technique: juries are impressed with scientific evidence. But no matter how sophisticated forensic science may become, there are no "techno-fixes" when it comes to matters of judgement.

Just because no two people, other than identical twins, have the same DNA patterns, it doesn't follow that the results of a match between the DNA material of a suspect and that found at a crime scene will be accurate. That depends on an unbroken chain of custody for the evidence and t he competence with which the test was performed. Compared with a fingerprint match, DNA analysis requires greater care and skill. Samples can be switched, and unsanitized lab equipment, or the perspiration of the person collecting the sample, can distort the findings. And persons analyzing the materials might be corrupted.

Even an accurate match done under exemplary lab conditions doesn't necessarily offer proof of legal guilt. Legal guilt depends on the context, on the interpretation of motives and the meaning of behavior, and on following correct procedure. In matters of criminal justice, the Lone Ranger will always be the only one with the silver bullet.

Back to Main Page | Top

You are visitor number to this page since November 1, 1999.