Is a DNA Identification Database in Your Future?

By David H. Kaye, Michael E. Smith, and Edward J. Imwinkelried*

This essay appeared in the journal of the American Bar Association Section on Criminal Justice, Criminal Justice, vol. 16, no. 3, Fall 2001, pp. 5-11 & 19

When law enforcement authorities began building DNA databases in the early 1990s, only individuals convicted of serious sexual crimes were included. Before long, though, many states authorized the sampling of DNA from those convicted of murder, and then other violent felonies. Some states now collect DNA from all felons and certain misdemeanants. Bills have been introduced in several states to expand the databases beyond convicted offenders, to include arrestees. Louisiana’s statutes already authorize it.

These computer-searchable DNA databases help solve cases that have baffled police for decades, and they identify previously convicted offenders who commit new crimes. Examples abound. In Virginia, there was the rapist who blew out a candle before attacking his victim. The candle had his saliva. There was the burglar who wore a pair of socks on his hands and left no fingerprints. The discarded socks contained skin cells. There was the bank robber who dropped his ski mask. All were identified by checking the DNA profiles in these traces against the state’s database of convicted felons. (See Profile: Use of a DNA Data Bank to Catch Criminals in Virginia, Nat’l Public Radio Morning Edition, Mar. 8, 2001, available at 2001 WL 9326731.)

Britain has assembled a large, national database by taking DNA samples from virtually everyone arrested for significant offenses. (Britain used to delete the identifying profile only when no conviction followed, but a recent act of Parliament allows retention even without a conviction.) Not only did the DNA profiles of arrested property offenders often yield database “hits” in rape cases, but the growing database produced even more hits in burglaries and vehicle thefts. The British experience made it clear that expanded use of the technology could have powerful crime prevention effects — not only through apprehension and incapacitation of offenders, but also by deterring from crime those who know their DNA profiles will routinely be compared to DNA evidence found at crime scenes.

Each DNA database expansion has triggered sharp debate. Some scholars and advocacy groups condemned the early database laws and their gradual expansion as “unfettered government-sponsored bioinvasion,” “surveillance creep,” and a “dangerous erosion of privacy.” The ACLU, which originally maintained that the convicted-offender databases were unconstitutional -- a position that no appellate court accepted -- now urges that the line be drawn at conviction and that samples be destroyed. In a number of articles, though, we suggest that a truly comprehensive database -- one that covers the entire population -- may represent a fairer and more effective accommodation of the interests in public safety and civil rights and liberties than the current system of piecemeal expansion. (See Edward J. Imwinkelried & D.H. Kaye, DNA Typing: Emerging or Neglected Issues, 76 Wash. L. Rev. 413 (2001); D.H. Kaye & Michael Smith, DNA Databases for Law Enforcement: The Coverage Question and the Case for a Population-Wide Database, in The Technology of Justice: The Use of DNA in the Criminal Justice System (David Lazer ed. forthcoming); D.H. Kaye, Two Fallacies About DNA Databanks for Law Enforcement, 67 Brook. L. Rev. 181 (forthcoming 2001).) Here, we describe one vision of a population-wide database and respond to various objections that might be raised against it. Our vision is futuristic —- indeed, it grew out of discussions within the Legal Issues Working Group of the National Commission on the Future of DNA Evidence. Moreover, it is incomplete and tentative. Many details remain to be worked out. Nevertheless, it appears to us that in comparison to the public safety benefits of a population-wide database, the true threat to privacy posed by such a database is modest and manageable.

A population-wide database

Creating a national identification database all at once would be prohibitively expensive today, even if we had the laboratory capacity to do it. But DNA typing technology is advancing at a pace reminiscent of the exponential growth in computer microprocessing power that has made the “personal computer” a fixture on every desk. Soon it will be feasible to create a DNA identification record for everyone, at least prospectively. For example, it would be easy to extract identification profiles as an adjunct to the existing public health programs that for many years have screened DNA samples from almost all newborns, to identify infants with treatable genetic diseases. The identification profiles could be transmitted to a single, secure, national database. The genetic locations (“loci” is the technical term) used for those identification profiles would be strictly limited to sequences that have no implications for health or other significant physical or mental traits. Furthermore, access to the database would be limited to law enforcement personnel investigating specific crimes in which DNA trace evidence already has been found. Law enforcement agencies would not need -- and should not be permitted -- to handle, much less retain, the samples.

The system would resemble, to the greatest extent [----6----] possible, a digitalized collection of identifying features comparable to ordinary fingerprints. However, it would be far more useful in deterring potential offenses, in generating investigative leads, and in exonerating the innocent. There would be no need to resort to inefficient "DNA dragnets" of entire neighborhoods, as have been conducted in California, Florida, and elsewhere. (See Philip P. Pan, Pr. George's Chief Has Used Serial Testing Before; Farrell Oversaw DNA Sampling of 2,300 in Fla., Wash. Post, Jan. 31, 1998, at B1.) Not only would a comprehensive database be valuable to the criminal justice system, but it also would be useful in identifying remains after natural disasters, mass accidents, and terrorist attacks. Such a database is, we believe, socially advantageous. But we would be the first to acknowledge that this belief is surely debatable, and a panoply of questions must be considered.

Does all DNA reveal our most private information?

In the popular mind, "DNA" has achieved iconic status as the essence of an organism. Metaphors like "the blueprint of life" and "the future diary" convey the impression that the molecule tells reveals everything important there is to know about a person. Of course, this kind of "genetic determinism" and "genetic essentialism" is naive. There is much more to life than DNA. Still, there can be no doubt that the genome contains a vast amount of sensitive, personal information. That is why so many states have passed laws prohibiting unauthorized acquisition, dissemination, or use of genetic testing by health care providers, employers, insurers, or others. (See, e.g., William F. Mulholland II & Ami S. Jaeger, Comment, Genetic Privacy and Discrimination: A Survey of State Legislation, 39 Jurimetrics J. 317 (1999); Helen R. Davis & Janis V. Mitrius, Note, Recent Legislation on Genetics and Insurance, 37 Jurimetrics J. 69 (1996).) Presumably, it is also why President Bush has announced his support for comparable federal legislation. (More comprehensive and balanced legislation protecting medical records would be preferable to the existing potpourri of ad hoc laws, but that debate is another story. See, e.g., Lawrence O. Gostin & James G. Hodge, Jr., Genetic Privacy and the Law: An End to Genetics Exceptionalism, 40 Jurimetrics J. 21 (1999).)

In the database system we envision, the information that the government is allowed to have is very different from the types that have sparked debates over medical privacy. The preponderance of the human genome consists of sequences that have no medical importance or social significance. Much of the genome is "noncoding" -— these sequences are not translated into the proteins that are the machinery of cells -- and most of them are not genetically "linked" to any coding sequences. Even in the coding regions, many DNA sequences merely code for traits such as gross features of fingerprints or the pattern of hair follicles in the skin that have no stigmatizing potential. Consistent with current practice, we would limit the database loci to such regions. Consequently, the genetic information included in the database would be no more invasive of privacy than an image of the ridges and whorls in a fingerprint or of the blood vessels in the retina of the eye.

Could the Samples Be Used by Insurers or Employers?

Closely related to the question of the strength of the interest in secrecy is the concern that DNA information could be used to deprive an individual of opportunities for employment or health insurance. Although the records of nonpredictive, nonstigmatizing DNA sequences that would be stored in a population-wide DNA identification database could pose no such threat, the molecules of DNA extracted from the samples of blood, saliva, or epithelial cells do contain the full complement of each person’s genes --and the continued existence of the tissue samples indexed by name would pose a genuine threat.

For this reason, the system we envision keeps almost all samples out of the hands of law enforcement officials. Recall that the initial typing would be done by health workers, not police, as part of neonatal screening. No samples would be sent to law enforcement agencies -— they would receive only the biometric genotypes that have no use except for identification. To the extent that additional sampling, say, of immigrants or citizens born abroad, would be necessary to cover as much of the population as possible, the sample could be destroyed as soon as the typing is complete. In fact, an instrument could be built that would extract an identifying profile and destroy the sample at the same time. Proper procedures for sampling the DNA, extracting the identifying profile, and immediately destroying the sample would protect everyone's genetic privacy -— to the extent we have any when private hospitals and HMOs keep samples of our blood and other tissue together with information far more sensitive than the random bits of DNA that identify us. The government officials maintaining the database could neither invade privacy nor enable insurers or employers to do so.

Would It Be Fair?

Even if the threat to privacy is contained, a population-wide database would include DNA from people who have not been convicted of any crimes and who are not even suspected of criminal misconduct. Is it fair to insert even the limited records of DNA sequences needed for identification of all these innocent people into the databases without their consent? Some critics of the fairness of DNA databases argue that people can be falsely suspected of criminal activity, either because more than one individual’s DNA matches the crime scene DNA at the recorded [-----7--8-----] loci or because of human errors in labeling the records.

The first type of "error" is likely to be occur very rarely, but to the extent it does, the concern argues for a truly comprehensive database. Although identical profiles will be uncommon, to say the least, when typing is done at 13 or more loci, if several individuals in the population do match, the task for the police will be to zero in on the one possibly guilty party. They will seek evidence that eliminates the coincidentally matching individuals as plausible suspects. This evidence might be obtained by traditional methods of investigation, such as questioning witnesses and suspects, or by acquiring new DNA samples from suspects for testing at additional loci. Often, investigators will be able to eliminate false leads with little effort and little burden on the DNA-based suspects. Moreover, the total number of people subjected to investigation, whether intrusive or not, frequently will be less with the population-wide database because many individuals who might have been suspects under traditional methods of investigation will have been excluded by the initial database search. With the existing, smaller databases, however, it is more likely that an innocent person will become the focus of investigators' attention -— and even be convicted. This is so because when potentially incriminating DNA evidence found at a crime scene cannot be matched to any known individual, suspicions will fall on one or more individuals for reasons both good and bad. Investigators who must rely exclusively on other, nongenetic evidence will follow inferences from such evidence as well as their hunches based on no evidence at all. Many cases in which death-row and life-sentence inmates have been exonerated by DNA evidence are good illustrations of how an investigator’s hunch grows into a certainty for which evidence is gathered selectively -- perhaps because conflicting evidence simply does not seem relevant. Thus, a population-wide identification database would reduce the ever-present risks of false accusation as well as the risk of false conviction.

The second source of error also poses no major obstacle if, when a database match leads to a person's arrest, a fresh sample of that person's DNA is tested and compared to the database profile that matched the crime scene DNA evidence. If the individual arrested does not have the DNA types recorded in the database, his or her innocence will be immediately obvious.

There is yet another sense in which a population-wide database would be fairer. In this country, it is a given that any system that places only the DNA profiles of those arrested for or convicted of crime will generate a racially skewed collection of profiles. African-American males, in particular, will be burdened and stigmatized, and whites will be less likely to be identified when they commit crimes and less likely to be deterred from crime by the knowledge that their DNA profiles would be compared to DNA left at crime scenes. White men commit heinous crimes, and there are many more of them.

Criminal justice statistics collected over the past decade make it apparent that any identification system limited to those arrested or convicted will be egregiously skewed along racial lines. According to the Bureau of Justice Statistics’ analysis of 1991 incarceration rates, an African-American man is six times more likely to be imprisoned at some point in his lifetime than is a white man. (See Thomas P. Bonczar & Allen J. Beck, Lifetime Likelihood of Going to State or Federal Prison (Bureau of Justice Statistics 1997.) Nearly two-thirds of all men sent to prisons that year were from African-American, Hispanic, or other minority groups. In some African-American neighborhoods, virtually all adult males have been arrested by police, for something, at some point in their lives. The reasons for the disparities range from racial prejudice shaping the discretionary decisions of some police officers and prosecutors, to race-neutral police deployment patterns that lead to more aggressive policing in minority communities than others. But it seems clear -— particularly to minority youth — that racial minorities are over-represented among both arrestees and convicts. If they are or believe themselves to be wrongfully over-represented, construction of exclusively convict or arrestee DNA databases could be a huge step in the wrong direction, exacerbating racial divisions.

In contrast, a comprehensive DNA database would help correct the racial distortions of our criminal justice system. DNA evidence does not care about race. A database profile either does or does not match a crime-scene sample. With a population-wide database, the identity of any matching individual would be known—no matter what the race. Routine production of a short list of individuals with DNA matches would be an antidote to the presence and perception of racism in the investigation of crime. When a [---- 9 ----] person is arrested and crime-scene DNA points to another person whose DNA profile is in the database, prompt exoneration and release of the innocent are likely to follow -- no matter what the initial suspect's race or status. If an innocent defendant does go to trial in such a case, the crime-scene DNA evidence and the results of the database search would be available, regardless of the defendant's race, to raise reasonable doubt about guilt. In short, moving to a population-wide DNA database would help to combat the appearance and reality of racism in the criminal justice process. It is certainly pertinent to ask whether it would be fair to include everyone in the database, but, in the final analysis, the real question must be whether doing so would be fairer than relying on today's more limited databases that over-represent minorities.

Would it be constitutional?

Although a population-wide database seems fairer and more efficacious than offender and arrestee databases, it might be beyond the power of government to implement. The most significant constitutional barrier is the Fourth Amendment’s prohibition on unreasonable searches and seizures. But despite centuries of cases interpreting this clause, the meaning of "search and seizure" and "unreasonable" is hardly pellucid, and whether the Supreme Court would uphold a population-wide database that incorporates significant privacy protections is difficult to predict.

Blood samples taken from infants by governmental edict would "constitute searches of 'persons,' and depend antecedently upon seizures of 'persons.'" (Schmerber v. California, 384 U.S. 757, 767 (1966) (dealing with a blood sample taken from a driver being treated in a hospital for injuries received in an automobile accident).) But what if the DNA were acquired by applying a sticky pad to the infant’s skin to acquire some exfoliating, epidermal cells without even puncturing the skin surface? Would the reduced level of bodily invasion and the fact that these cells are constantly exposed to the public lead a court to hold that no search is involved? In Palmer v. State, 679 N.E.2d 887 (Ind. 1997), the Indiana Supreme Court reasoned that the warrantless acquisition of defendant’s fingerprints during his trial did not constitute a seizure forbidden by the Fourth Amendment because "fingerprints are an identifying factor readily available to the world at large." (Id. at 891.) Other courts have held that shining an ultraviolet lamp on an arrestee’s skin to expose chemicals transferred from stolen money is not a search because the flourescent material "may be compared to a physical characteristic, such as a fingerprint or one’s voice, which 'is constantly exposed to the public.'" (State v. Holzapfel, 748 P.2d 953 (Mont. 1988); see also United States v. Richardson, 388 F.2d 842, 845 (6th Cir. 1968) ("We do not regard the examination of appellant's hands under the ultraviolet light as a search within the meaning of the Fourth Amendment.").)

One might have hoped that the Supreme Court's latest encounter with defining a "search" in Kyllo v. United States, 121 S.Ct. 2038 (2001), would clarify the viability of this "public exposure" theory. In Kyllo, a federal agent used an infrared detector to find that “the roof over the garage and a side wall of petitioner's home were relatively hot compared to the rest of the home and substantially warmer than neighboring homes in the triplex." (Id. at 2041.) "Based on tips from informants, utility bills, and the thermal imaging, a federal magistrate judge issued a warrant authorizing a search of [Kyllo's] home, and the agents found an indoor growing operation involving more than 100 plants." (Id.) Before trial, Kyllo moved to suppress the evidence on the ground that the thermal imaging required a warrant. When the motion was denied, he entered a conditional guilty plea and appealed. The Ninth Circuit Court of Appeals ultimately affirmed, reasoning that defendant had neither a subjective nor an objectively reasonable expectation that "amorphous 'hot spots' on the roof and exterior wall" would go unobserved. (Id.) In other words, according to the Court of Appeals, there was no "search" under the standard announced in Katz v. United States, 389 U.S. 347 (1967).

A sharply divided Supreme Court reversed. At first blush, this reversal seems to foreclose the argument that inspecting materials on the surface of the body is not a search. After all, if the use of an instrument to capture infrared rays coming from the surface of a house is a search, arguably, so is the use an instrument to capture and analyze DNA on the surface of the body. However, the rationale of Kyllo is quite limited. Justice Scalia’s opinion for a majority of five Justices looks to the historically recognized zone of privacy in which government surveillance is prohibited. Apparently assuming that Eighteenth Century constables would have had to enter the house to detect heat sources — a trespass that is the very paradigm of a search -- the majority announced that the infrared scan also was a search. As the Court put it, "obtaining by sense-enhancing technology any information regarding the interior of the home that could not otherwise have been obtained without physical 'intrusion into a constitutionally protected area,' constitutes a search -- at least where (as here) the technology in question is not in general public use." (Id. at 2043.) In other words, Kyllo holds only that technology that is functionally equivalent to trespassing into a home to acquire information is a search. This result, the Court suggested, was necessary for “the preservation of that degree of privacy against government that existed when the Fourth Amendment was adopted.” (Id.) To hold otherwise, the majority insisted, would “permit police technology to erode the privacy [originally] [---- 10 ----] guaranteed by the Fourth Amendment.” (Id.)

In contrast, protections against searches (as opposed to seizures) of the person lack "roots deep in the common law . . . ." (Id.) As the Schmerber Court observed, in "dealing with intrusions into the human body rather than with state interferences with property relationships or private papers — 'houses, papers, and effects' —- we write on a clean slate." (384 U.S. at 767–78.) Therefore, Kyllo's historical functional equivalence test does not dictate the conclusion that it is a "search" to take from the surface of a person's skin cells that are constantly being shed and to analyze the DNA they contain. Unlike infrared scanning that, in effect, places the police in the interior of a house, DNA sampling and analysis is not functionally equivalent to any Eighteenth Century practice proscribed by the Fourth Amendment.

Despite the limited scope of Kyllo, let us assume that even noninvasive DNA sampling is a search under the Fourth Amendment. (See D.H. Kaye, The Constitutionality of DNA Sampling on Arrest, 10 Cornell J. L. & Pub. Pol’y __ (forthcoming 2001) (arguing for this conclusion).) The sampling still might be permissible, since "[t]o hold that the Fourth Amendment is applicable to the . . . testing . . . is only to begin the inquiry into the standards governing such intrusions." (Skinner v. Railway Labor Executives' Ass’n, 489 U.S. 602, 618–19 (1989.) "Neither a warrant nor probable cause, nor, indeed, any measure of individualized suspicion, is an indispensable component of reasonableness in every circumstance." (National Treasury Employees Union v. Von Raab, 489 U.S. 656, 665 (1989).) In Von Raab, the Court explained that "our cases establish that where a Fourth Amendment intrusion serves special governmental needs, beyond the normal need for law enforcement, it is necessary to balance the individual's privacy expectations against the Government’s interests to determine whether it is impractical to require a warrant or some level of individualized suspicion in the particular context." (Id. at 665–66.)

Inasmuch as the blood presently taken from infants is collected for medical purposes, the acquisition of the DNA in it "serves special governmental needs" that have nothing to do with "normal law enforcement." The many "special needs" cases like Skinner and Von Raab entail a balancing of government and private interests when a search or seizure serves a purpose other than ordinary law enforcement. Hence, the constitution may well be found to permit "piggybacking" on top of the routine genetic screening for treatable diseases an analysis at the loci useful only for future identification. There is no incremental detention of the person, no additional bodily invasion, and no further acquisition of medically or socially revealing information. Thus, such a law enforcement add-on would involve, at most, a tightly circumscribed intrusion on privacy.

Under two other Supreme Court cases decided this Term, however, one might question whether "special needs" balancing could be invoked if a Fourth Amendment challenge were raised to the procedure outlined above. In City of Indianapolis v. Edmond, 121 S.Ct. 447 (2000), the Court declined to apply the exception to a program in which police used dogs to sniff for drugs in vehicles pulled over in groups at fixed roadblocks. Distinguishing sharply between "highway safety interests and the general interest in crime control” (id. at 453), the majority reasoned that "[b]ecause the primary purpose of the Indianapolis narcotics checkpoint program is to uncover evidence of ordinary criminal wrongdoing, the program contravenes the Fourth Amendment." (Id. at 454.) Likewise, in Ferguson v. City of Charleston, 121 S.Ct. 1281 (2001), the Court invalidated a program in which a state university hospital tested urine samples from pregnant women for cocaine and reported positive results to the police. The hospital informed the police so that those women, fearing prosecution, would be induced to participate in substance-abuse counseling offered as an alternative to criminal prosecution. Again, the majority of the Court emphasized "the relevant primary purpose" -- which was said to be "the arrest and prosecution of drug-abusing mothers." (Id. at 1290 (internal quotation marks omitted).)

Yet, Edmond and Ferguson are distinguishable. Constructing an identification database would be an ancillary rather than the “primary purpose” of continuing to extract DNA from newborns. The primary purpose always has been, and would remain screening for correctable genetic conditions. Furthermore, even were Edmond and Ferguson extended to programs whose secondary purpose is law enforcement — undermining the entire line of special needs cases p— it is far from easy to predict what the Supreme Court would do if given the opportunity to recognize a new “biometric identification exception” to the warrant requirement. (See Kaye, The Constitutionality of DNA Sampling on Arrest, supra.)

This is not to minimize the seriousness of the Fourth [---- 10--19 ----] Amendment question. The constitutionality of a population-wide database surely is open to a powerful challenge. The issue is complex, and the caselaw is subject to varying interpretations. But one cannot summarily dismiss a population-wide database as manifestly unconstitutional.

Would It Be Useful?

The final question that we consider is whether the game is worth the candle. Suppose, for the sake of illustration, that the cost of the additional DNA testing for law enforcement were one dollar per sample and that the marginal cost of operating the larger database were 25 cents per sample. At present, some four million babies are born every year in the United States. Even assuming that population growth and were to as much as double the number of births by the time a decision to implement the comprehensive database were made, the cost in 2001 dollars would be on the order of $10 million per year. Considering that last December, Congress authorized $170 million to help states process DNA evidence, the incremental cost of assembling a population-wide database prospectively does not seem excessive.

Substantial benefits would likely flow from this investment. We know that many offenders commit other crimes—sometimes many other crimes—before they are first convicted. We also know that with only 135,000 convicted-offender samples, Virginia’s database averages one hit per week. Many more hits would result from a truly comprehensive, national system. When one considers the personal and social costs of the crimes that might be prevented, the balance might well favor the most inclusive database. Of course, it is nearly impossible to document and quantify the net benefits, since we cannot know how many crimes would be prevented or solved by any given expansion of the national DNA identification database. We cannot even predict the extent to which those who are bent on crime will try to outsmart the technology by taking care not to leave any DNA at the scenes of their crimes or to obscure the evidence they do leave. (See Richard Willing, Criminals Try to Outwit DNA, USA Today, Aug. 28, 2000, at 1).)

In the end, the unavoidable question is whether the threat to privacy from a population-wide database is so great that we must be content to build and rely on DNA identification systems that exacerbate racial divisions and deprive us of potential public safety advantages. The prospect of a population-wide database poses a difficult choice for those who are committed to fighting the racial imbalances that sap the legitimacy of law, but who also prize privacy. Like the double helix of the DNA molecule, privacy and equality are intertwined in complex ways. When they are untangled and evaluated, the case for a population-wide DNA database is strong. Perhaps it will prove compelling. At the very least, this is the debate in which the nation must engage.

* David H. Kaye is regents’ professor and fellow at the Center for the Study of Law, Science, and Technology, Arizona State University. Edward J. Imwinkelried is professor of law at the University of California (Davis). Michael E. Smith is professor of law at the University of Wisconsin. Smith chaired the Legal Issues Working Group of the National Commission on the Future of DNA Evidence, on which Kaye and Imwinkelried served.