One of the most fundamental challenges in forensic science is individualization: determining the identify of a person (such as the perpetrator of a crime) with a high degree of specificity. Law enforcement officials must be able to say that the person who committed a violent crime or a murder was Mr. A. or Ms. B., not the next-door neighbor, a good friend, or even a close relative. Fingerprinting has long been the most reliable method of individualization, based on the belief that no two individuals in the world have exactly the same set of fingerprint patterns. Fingerprinting poses both theoretical and practical problems, however, in that scientists have not yet proved this underlying assumption, and the collection and interpretation of fingerprints is often difficult.

In 1984, the British geneticist Alec Jeffreys (1950– ) discovered a new method of individualization with the promise of its becoming the perfect method for distinguishing any two humans from each other. The method, based on small variations in the DNA of all humans, was originally called DNA fingerprinting, because of its similarity to traditional fingerprinting. The method is now more commonly called DNA typing, DNA profiling, or DNA patterning.

The acronym DNA is short for deoxyribonucleic acid, the name given to a group of molecules that occur in all cells of all living organisms and that carry that organism’s genetic information. That is, they carry the instructions for making the chemical compounds—proteins—by which cells stay alive, grow, develop, reproduce, and carry out all of the functions that constitute life as we know it.

DNA molecules are very large, complex molecules made, nonetheless, of only a few relatively simple units: a sugar called deoxyribose (D); a combination of phosphorus and oxygen atoms called a phosphate group (P); and four nitrogen bases, adenine (A), cytosine (C), guanine (G), and thymine (T). Nitrogen bases are compounds in which carbon and nitrogen atoms are joined to each other in a ring. The combination of one sugar molecule, one phosphate group, and any one nitrogen base (of the form D - P - A, for example) is called a nucleotide. A complete DNA molecule consists of very long chains of thousands of nucleotides joined to each other, as represented by the following abbreviated formula:

where N1, N2, N3, and N4 represent the four possible types of nucleotides, each having a different nitrogen base joined to a sugar and phosphate. Each DNA molecule actually contains a pair of nucleotide chains twisted around each other, somewhat in the form of a spiral staircase, in a structure called a double helix. When scientists describe a DNA molecule, they usually do so by listing the sequence of nitrogen bases in the molecule. The “backbone” to which the bases are attached is a repetitive sequence of the form - D - P - D - P - D - P - D - P - which provides no unique information about the molecule. Thus, one might want to talk about a specific portion of one specific DNA molecule in which the nitrogen base sequence can be described as - A - G - G - G - A - D - T - T -. If that segment of DNA carries useful genetic information—that is, if it tells a cell how to perform some function—it constitutes or is part of a gene. Genes occur in slightly different forms in organisms known as alleles. For example, the gene that tells a cell how to make hair exists in forms that carry the instructions for black hair, red hair, brown hair, or hair of some other color.

Less than one-tenth of 1 percent of the nitrogen base sequences in DNA molecules carry no genetic information, that is, they carry no known useful information for cells. Scientists call such sequences “junk DNA” or, more formally, introns. An intron is a sequence of nitrogen bases with no known human genetic function interspersed between exons, nitrogen base sequences that do carry information and are, therefore, usually expressed in a cell. Although the exons in humans are all very similar to each other (most humans at birth all have two eyes, two ears, one nose, two arms, similar brains, and other structures in common with each other), their introns differ widely. Because of this wide variability, scientists can use the molecular structure of introns to distinguish between any two members of a species: between any two humans, any two killer whales, or any two English sparrows, for example.

As with fingerprinting, scientists cannot say with absolute certainty that no two humans (or two members of any species) are absolutely unique. (In fact, identical twins do share exactly the same DNA patterns.) It is possible, however, for one to calculate the likelihood that any two persons will have exactly the same sequence of nitrogen bases, the same DNA “fingerprint.” Because DNA is such a large and complex molecule, those probabilities are very small indeed. In forensic cases, it is generally not diffi cult to say that the chance of finding a given DNA pattern in two different individuals is one in 10 million or one in 100 million, one in a billion, or some similar very low frequency. Because of this high level of certainty in identifying a specific individual based on his or her DNA, DNA typing has now replaced digital fingerprinting as the “gold standard” of individualization in forensic science.

DNA typing is superior to digital fingerprinting and other forensic techniques not only because it discriminates between two people better than any other procedure, but also because it can be used with a broader range of sample types. A digital fingerprint can be obtained only from a person’s fingertips (or, less commonly, from the palms, toes, or soles of the feet). But DNA occurs in every cell of the body. All an investigator needs is a drop of blood, a single hair, a flake of skin, or a single cell from any other part of the body to obtain a DNA fingerprint. DNA typing can also be carried out with very small samples and with evidence that is months, years, decades, or even centuries old. Unlike blood and other types of evidence that degrades over time, DNA taken from a cell often remains in perfect condition for virtually unlimited periods of time.

Alec Jeffrey’s great accomplishment was his discovery of a method by which scientists can find the sections of DNA that differ in individuals, snip them out of a DNA chain, and take their “photographs” using radioactive materials. The method he developed is known as restriction fragment length polymorphisms (RFLP), a name that comes from the chemical compounds used to do the snipping (restriction enzymes), the size of the segments snipped out, and the variations (polymorphisms) present in the segments. Today RFLP has largely been replaced by a second method for finding, cutting out, and identifying portions of a DNA molecule. That method, invented by American molecule biologist Kary Mullis (1944– ) in 1986 is called polymerase chain reaction (PCR). Virtually all forensic applications of DNA typing now use the PCR technique because it is faster and can be used with much smaller samples of DNA than can RFLP.

Less than a year after Jeffreys discovered the RFLP method, he had an opportunity to use the technique in solving a practical problem in genetics. The problem concerned a young boy who attempted to enter the United Kingdom from Ghana with a British passport that appeared to officials to have been altered. The boy claimed that he was returning to his mother in England after a visit to Ghana. Immigration officials suspected, however, that the boy was a relative of the person named in the passport, perhaps a cousin, trying to enter the country illegally. At fi rst, Jeffreys thought the problem was beyond the scope of science. “Well, forget it!” he said at first. “This is a jigsaw puzzle with too many pieces missing.”33 He eventually decided to try using RFLP, however, to solve that puzzle. He took blood samples from the boy, the boy’s mother, his father, and three sisters for RFLP analysis. He found the six samples were sufficiently similar that there could be no question as to the relationships claimed for the family. The boy was admitted to the United Kingdom.

About a year after the Ghanaian case was solved, Jeffreys received his first request to become involved in a criminal investigation. The case involved a pair of rape-murders in the small town of Narborough in Leicestershire, one that occurred in 1983, the other in 1986. Police had arrested a local boy named Richard Buckland for the crimes. Buckland confessed to the 1986 crime but denied any involvement in the earlier case. Leicestershire police asked Jeffreys to use his DNA test to confirm Buckland’s guilt in both crimes. Jeffreys examined DNA taken from semen at both murder scenes and DNA from a sample of Buckland’s blood. He confirmed that the same person had committed both rape-murders, but that Buckland was not that person. Buckland was exonerated of both crimes, the first person in history to have been found innocent as a result of DNA typing.

The Leicestershire case had a somewhat bizarre conclusion. Police eventually took blood samples from all males in Narborough and two nearby villages. Jeffreys found no match with DNA taken from the crime scene and any of the more than 400 samples collected by the police. The case appeared to be insolvable, at least by means of DNA typing. The unexpected turn came about a year later when a Narborough woman overhead a fellow worker bragging that he had given a sample of his blood under the name of a friend, Colin Pitchfork. Pitchfork was arrested and his DNA tested. It matched the samples taken from the two crime scenes, and he was convicted of the two crimes.

The success of DNA typing in solving the Leicestershire case soon became widely known. Prosecutors in many countries recognized the power of the new technology and began to use it eagerly. As Ron Fridell has written in his book DNA Fingerprinting: The Ultimate Identity, “[n]ever before in the history of law enforcement had a new technique for analysis of physical evidence been adopted so suddenly and so unreservedly. It seemed as if DNA fingerprinting were foolproof.”

In the United States, the first laboratory established specifically to carry out DNA typing was Cellmark Diagnostics, opened in Germantown, Maryland, in 1987. Cellmark trademarked the phrase “DNA Fingerprinting” to denote the specific details of the technology it had developed for DNA typing. Cellmark’s only competitor in the fi eld of DNA typing in the United States was Lifecodes Laboratories of Valhalla, New York. Lifecodes had been founded in 1982 and began DNA testing in 1987, shortly after Cellmark began operations. Lifecodes was hired in November 1987 by prosecutors in Orange City, Florida, to test semen samples found at the scene of a rape for which Tommy Lee Andrews had been arrested. Lifecodes reported that the DNA found at the crime scene matched Andrews’s DNA with a probability of one in 10 billion. (That is, the company said that the chance of there being some other person with exactly the same DNA sample was one in 10 billion.) Convinced to a large extent by the strong DNA evidence, a jury convicted Andrews of rape. He was sentenced to 22 years in prison. A year later, the Florida District Court of Appeals upheld the lower court’s verdict, giving Andrews the dubious honor of being the first person in the United States to have had his or her conviction upheld by a higher court on the basis of DNA evidence.

The Andrews case was reported and commented on widely in the popular press and in legal journals, opening the floodgates to its use in a number of criminal cases. Within a year of the Andrews decision, DNA typing had also been affirmed for the first time by a highest state court, the West Virginia Supreme Court of Appeals in the case of State v. Woodall (385 S.E. 2d 253; W. Va. 1989). It seemed that DNA testing was on its way to general and enthusiastic acceptance within the law enforcement and legal communities.

Then came People v. Castro. The case arose out of the arrest of a 38-yearold Latino man, Jose Castro, for the murder of a neighbor, Vilma Ponce, and her two-year-old daughter. Mother and daughter had been stabbed to death in their apartment, and blood found on Castro’s watch was thought by police to belong to one or both of the victims. Lifecodes conducted a DNA analysis of the blood sample and compared it with DNA taken from the two victims. The company reported the likelihood of finding the match produced was one in 100 million. As in previous cases, the DNA evidence was so strong that a conviction appeared certain.

At that point, however, attorneys for Castro took a step that no defense attorney had yet used: They challenged the scientific validity of the DNA analysis. The two attorneys, Barry C. Scheck and Peter J. Neufeld (who were later to found the Innocence Project, designed to exonerate individuals falsely convicted of crimes) argued that DNA typing was a new technology that had not yet been adequately tested. They based their argument on the existing standard for the admissibility of scientific evidence in a criminal trial, established 60 years earlier in Frye v. United States. According to the Frye standard, scientific evidence had to meet a number of criteria to be admissible in a criminal trial, one of which was that the procedure must have been widely accepted by the scientific community. Scheck and Neufeld argued that DNA typing had not yet reached that level of approval.

Judge Gerald Sheindlin applied the Frye standard to the Lifecodes DNA typing results and found that they did meet most of the standards imposed by that precedent. He wrote in his decision that the scientific community was in general agreement that DNA testing produces valid results in terms of its ability to discriminate among specifi c individuals. DNA typing is valid also, he said, because it makes use of techniques that were in existence even before DNA typing had been invented.

The problem in this case, however, was that Lifecodes had, purely and simply, made technical errors in carrying out DNA testing of blood samples. Not only had it made technical errors, but it had also misjudged the probability of a match between blood samples. Simply stated, company scientists had said the DNA “fingerprints” from the blood on Castro’s watch and the two victims matched when they did not. The evidence provided by Lifecodes failed the final prong of the Frye test, then, because it was not obtained by the proper application of accepted methods. As a result, Sheindlin excluded the DNA evidence from consideration in the case. (Castro later confessed to the murders and was sentenced to life in prison.) People v. Castro is an especially important court case because it established the general principle for the admissibility of DNA evidence. It said that the theory and technology of DNA typing is well established and widely accepted by the scientific community, producing valid results when testing is properly conducted. Other courts have confi rmed and adopted this principle, and it provides the foundation on which DNA typing rests today. The other part of the Castro principle, however, is that laboratories must follow typing procedures with the greatest care in order to obtain valid results. The only question courts have to answer in considering DNA evidence is whether this part of the standard has been observed and whether testing laboratories have avoided the introduction of human error into their results (as Lifecodes failed to do in Castro).

The fi rst federal case in which these standards were applied and DNA typing validated was United States v. Jakobetz, decided by the U.S. Court of Appeals for the Second District in 1992. The case arose when Randolph Jakobetz, a truck driver, was arrested for the violent rape of a woman he surprised and subdued at a rest stop on Interstate 91 in Westminster, Vermont. In presenting its case, the prosecution offered DNA evidence from a blood sample provided by the accused and semen taken from the woman’s body. The testing laboratory determined that the probability of a match of the two samples was 1 in 300 million. Based in part on this evidence, Jakobetz was convicted of the crime. He appealed the trial court’s decision, arguing that the DNA evidence submitted in the case was the U.S. Supreme Court declined to review. The appeals court decision not only affirmed the lower court’s decision, but noted that future cases should consider Jakobetz as a precedent in dealing with DNA evidence. It wrote that

By 1990, experts and interested observers had begun to take sides on the use of DNA typing as a source of valid evidence in criminal cases. On the one side were most prosecuting attorneys, law enforcement officials, scientists familiar with DNA technology, and, of course, laboratories who conducted DNA testing. These individuals were convinced that DNA typing was, as one judge put it, “the single greatest advance in the search for truth since the advent of cross-examination.” On the other side were many defense lawyers, a large fraction of the media, and many ordinary citizens, many of whom were unfamiliar with DNA technology and unconvinced of the extraordinary claims being made for it. Leading spokespersons for the anti-DNA typing camp were Barry Scheck and Peter Neufeld, defense attorneys in the Castro case. They established the DNA Task Force, a committee within the National Association of Criminal Defense Lawyers to limit or prohibit the use of DNA evidence in criminal cases. The Task Force’s efforts had little effect on the legal system, where judges continued to accept DNA evidence and prosecutors relied more and more on such evidence in their cases. But anti-DNA arguments were often persuasive among representatives of the media and general public who often felt the power of DNA typing was being overstated.

In 1990, the National Research Council of the National Academy of Sciences attempted to resolve this disagreement by creating a commission to study DNA typing and determine its validity and reliability in forensic science. That commission issued its report, DNA Technology in Forensic Science, two years later. The report reviewed the status of DNA typing technology and its applications in forensic science and other fields. It gave special attention to both economic and ethical considerations involved in the use of DNA typing in forensic settings. While supporting the use of DNA typing in general and acknowledging that it was a potentially powerful tool of individualization, the committee hedged its conclusions and recommendations by warning of possible misuses of the technology. Its second recommendation, for example, was that

Prosecutors and defense counsel should not oversell DNA evidence. Presentations that suggest to a judge or jury that DNA typing is infallible are rarely justifi ed and should be avoided.

In an effort to be evenhanded, the NRC commission may have failed to make its support of DNA typing strong enough and clear enough to end the debate among either experts or the general public. Defense lawyers, in fact, sometimes felt confident enough to continue challenging DNA evidence even to the extent of using the NRC report in support of their positions. In the meanwhile, most of the scientific and law enforcement communities continued to have confidence in DNA typing and to develop and refine the technology on which the procedure is based and its application in court cases. In 1994, an article by Eric S. Lander and Bruce Budowle appeared in the journal Nature that appeared to resolve once and for all—at least for scientists—the status of DNA typing. “The DNA fingerprinting wars are over,” the authors of that article wrote. “. . . the public needs to understand that the DNA fingerprinting controversy has been resolved. There is no scientific reason to doubt the accuracy of forensic DNA typing results, provided that the testing laboratory and the specific tests are on a par with currently practiced standards in the field. The scientific debates served a salutary purpose: standards were professionalized and research stimulated. But now it is time to move on.”

The key remaining issue—as it had always been—was human error. DNA tests are often carried out on very small samples of blood, hair, semen, or other materials. Even the smallest mistake can totally invalidate a test. As Eric S. Lander, then professor of biology at the Massachusetts Institute of Technology and director of the MIT Center for Genome Research said in a 1992 talk on DNA typing, “ . . . if I sneeze on something, my DNA is there, too. And so there is tremendous need to avoid contamination.” This fact, Lander pointed out, requires that high standards of operation be established for DNA testing laboratories, and special efforts must be made to ensure that those laboratories abide by those standards.