In Maryland v. King
the U.S. Supreme Court narrowly rejected a constitutional challenge to the controversial practice of routinely collecting DNA from arrested individuals. Calling the case precedent-shattering and “scary,” four dissenting justices charged that the Court has paved the way for states to collect your
DNA and put its CODIS profile in the FBI’s national database.
The case began when police in Maryland arrested Alonzo King for menacing people with a shotgun. Following the arrest, they took his picture, recorded his fingerprints—and swabbed the inside of his cheeks
. When checked against Maryland’s DNA database, his DNA profile led to the discovery that six years earlier King had held a gun to the head of a 53-year-old woman and raped her. Before the DNA match, the police had no reason to suspect King of that crime. Lacking probable cause, they did not rely on a judicial warrant to swab his cheek. They relied on a state law that mandated collection of DNA from all people charged with a crime of violence or burglary
King appealed the resulting rape conviction. He argued that the DNA collection deprived him of the right, guaranteed by the Fourth Amendment to the Constitution, to be free from unreasonable searches or seizures. Maryland’s highest court agreed with him. It held that, except in the rarest of circumstances where a suspect’s true identity could not be established by conventional methods (the court referred to such situations as face transplants!), forcing an arrestee to submit to DNA sampling was unconstitutional.
The state petitioned the Supreme Court for review. Over and over, the Court had denied requests from convicted offenders and, more recently, from mere arrestees, to address the legality of state and federal laws mandating routine collection of their DNA. But this case was different. Never before had a state supreme court or a federal appellate court deemed any federal or state DNA database law unconstitutional. Even before the Court met to consider whether it would review the case, Chief Justice Roberts stayed the Maryland judgment. His opinion stated that “there is a fair prospect that this Court will reverse the decision below” and found that “the decision below subjects Maryland to ongoing irreparable harm.”
The Chief Justice’s prediction proved correct. But the margin of victory was razor thin, and the dissenting justices importuned the Court “some day” to repudiate its “incursion upon the Fourth Amendment.” “[I]f the Court's identification theory is not wrong,” wrote Justice Antonin Scalia in an opinion joined by Justices Ginsburg, Sotomayor and Kagan, “there is no such thing as error.”
Here, I describe the reasoning of the polarized Court and articulate what the majority opinion (penned by Justice Anthony Kennedy and joined by Chief Justice Roberts and Justices Thomas, Breyer and Alito) established about the constitutionality of DNA sampling of arrestees. Being a law professor predisposed to pontificating, I criticize some parts of the Kennedy opinion and of the Scalia dissent. Both opinions, I further explain, leave room for skirmishes over the constitutionality of some features of DNA database laws.
The Court’s Reasoning
The King Court determined three things: (1) that buccal swabbing is a search subject to the Fourth Amendment; (2) that the constitutionality of this kind of search turns on the balance of state and individual interests; and (3) that this balance favors the state when (a) the swabbing is done after charges have been filed, (b) the loci tested do not reveal sensitive personal information and (c) statutory and administrative privacy safeguards are in place. The first point was not in contention. Previous opinions had held that blood and urine sampling—indeed, even scraping a little debris from beneath a fingernail of a suspect—are searches. The real controversy was over the second two points.
Choosing the balancing test
The majority opinion concluded that “the search ... falls within the category of cases this Court has analyzed by reference to the proposition that the ‘touchstone of the Fourth Amendment is reasonableness, not individualized suspicion.’”
. These are code words in Fourth Amendment jurisprudence. The amendment forbids “unreasonable searches and seizures.” Reasonableness is therefore the touchstone in every case, but a longstanding rule renders searches without probable cause or a warrant automatically unreasonable unless they fall within “a few specifically established and well-delineated exceptions.”
King argued that this per se rule governed and that no existing exception applied.
A court determined to uphold DNA sampling before conviction (DNA-BC) could have dealt with the per se rule in four ways. First, it might have forced DNA-BC into an existing categorical exception. The only plausible exception is for “special needs” or “administrative search” cases. The Court has used this exception many times. Drug testing in schools and in the workplace, roadblocks to find information about a hit-and-run driver and to get drunken drivers off the roads, and periodic searches of jail cells, for example, do not require probable cause and a warrant. The common theme of these cases is that the government’s interests go beyond the production of evidence for use in a criminal investigation or prosecution. In ordinary cases, in which the only point of the search is to generate such evidence or to seize contraband or stolen items, a warrant is required. In cases in which an additional interest is present, the Court asks whether the state’s interests outweigh the invasion of the individual interest in being free from arbitrary or oppressive searches.
Establishing a suspect’s true identity—what is called authentication in biometrics—is a special need distinct from gathering evidence in investigations. It permits jailers, judges and prosecutors to know whether an arrestee has a criminal record—but not just because the record might be evidence of guilt in the current case. Above and beyond that possibility, the existing record is relevant to administrative and judicial decisions about the need for and nature of pretrial confinement. Having a permanent biometric record serves still other interests as well
Yet, both opinions stated that the case did not fall within the special-needs exception. The stumbling block was two cases in which the Court declared that the exception was inapplicable because criminal evidence collection was the “primary purpose” of the program of searches. One case involved road blocks and dogs trained to detect drugs; the other was a hospital program, developed in conjunction with law enforcement, to test the urine of pregnant women for drug metabolites and to refer the women to the police if they refused drug counseling. However, these cases are distinguishable in that the “primary purpose” of ordinary law enforcement was the only purpose. The Court has never decided whether the same result should apply when the program truly serves multiple purposes
. In any event, the King Court merely gestured to the special-needs exception. It did not pursue this line of analysis.
Second, the Court might have created a new categorical exception for certain forms of biometric data collection
. Although Maryland mentioned this possibility in its petition for review, neither side referred to it again, and the Court did not consider it.
Third, the Court could have adopted a new regime in which every case involves a direct inquiry into the reasonableness of the search under all the circumstances. This would resemble the rule in tort cases, in which juries are asked to use their best judgment to decide whether the defendant’s conduct was unreasonable. Although some justices have spoken of Fourth Amendment reasonableness in this manner, King does not leave every search open to such ad hoc balancing.
Rather, the Court chose a fourth option. It tried to confine the per-se-rule-with-defined-exceptions to some types of searches, leaving the kind of search before it to be judged under the balancing standard. Thus, Justice Kennedy described the categorical-exception approach as defeasible in some circumstances “such as ‘[w]hen faced with special law enforcement needs, diminished expectations of privacy, minimal intrusions, or the like.’”
This description is somewhat cryptic. Whether, after King, observers can recognize the set of cases for which balancing is appropriate is a nice question, but the Court was certain that the King case belongs to it.
Applying the balancing test
Having settled on the open-ended balancing standard for ascertaining the reasonableness of the Maryland law, the majority applied it. After praising “DNA technology [as] one of the most significant scientific advancements of our era,” the Court noted “the need for law enforcement officers in a safe and accurate way to process and identify the persons and possessions they must take into custody.” This processing and identification, the Court emphasized, is not limited to the authentication function (in detecting escapees or other individuals who have disguised their identity). The processing also can be investigative: “A DNA profile is useful to the police because it gives them a form of identification to search the records already in their valid possession. In this respect the use of DNA for identification is no different than matching an arrestee's face to a wanted poster of a previously unidentified suspect or matching tattoos to known gang symbols to reveal a criminal affiliation or matching the arrestee's fingerprints to those recovered from a crime scene.” In fact, with respect to trawling a database of crime-scene records, “the only difference between DNA analysis and the accepted use of fingerprint databases is the unparalleled accuracy DNA provides.” Once officials discover a recorded criminal history or apparent involvement in an unsolved crime, whether by fingerprints or DNA, they can make better “critical choices about how to proceed” with respect to the period and nature of confinement.
Missing from this list of the state’s interests is the very thing that the DNA sample in King was used for—to charge and convict him of an unrelated crime. The closest the Court comes to acknowledging the state’s dominant objective of criminal intelligence—of linking arrestees to unsolved crimes—is a short paragraph in which it states that “[f]inally, in the interests of justice, the identification of an arrestee as the perpetrator of some heinous crime may have the salutary effect of freeing a person wrongfully imprisoned for the same offense.” By focusing so intently on “processing,” the opinion effectively forces the practice of DNA-BC into the “special needs” category even while saying that the special needs cases “do not have a direct bearing on the issues presented in this case.” In this way, the Court skirted the assumed “primary purpose” requirement of special needs balancing, yet engaged in exactly the kind of balancing it uses for special needs cases.
Having articulated a subset of all of the state interests in DNA-BC, the Court had to weigh them against the individual interests that underlie the Fourth Amendment. First, the Court depicted the state’s interests as substantial. The “proper processing of arrestees,” it stressed, is not only “legitimate” but also “so important [that it] has consequences for every stage of the criminal process.” Next, it perceived a close link between these interests and “DNA identification,” which “represents an important advance in the techniques used by law enforcement to serve legitimate police concerns.” Indeed, the Court insisted that “DNA identification is an advanced technique superior to fingerprinting in many ways, so much so that to insist on fingerprints as the norm would make little sense to either the forensic expert or a layperson.” But the only superior aspect of DNA profiling for authentication that the opinion pointed out was its power to confound the “suspect who has changed his facial features to evade photographic identification or even one who has undertaken the more arduous task of altering his fingerprints … .” Moreover, the Court did not mention the inability of DNA profiling to distinguish among pairs of monozygotic twins (who represent about 3 in 1000 births). In that regard at least, fingerprints are superior.
On the other side of the ledger, the Court perceived no significant “intrusion upon the arrestee's privacy beyond that associated with fingerprinting … .” The sampling procedure itself—the “cheek swab”—“ is a minimal [intrusion].” Neither do the details of King’s “13 CODIS loci … intrude on … privacy in a way that would make his DNA identification unconstitutional.” After all, “alleles at the CODIS loci ‘are not at present revealing information beyond identification’” and “even if non-coding alleles could provide some information, they are not in fact tested for that end.” Under Maryland’s law, “[a] person may not willfully test a DNA sample for information that does not relate to the identification of individuals … .” Thus, “[i]n light of the scientific and statutory safeguards, … the STR analysis of respondent's DNA pursuant to CODIS procedures did not amount to a significant invasion of privacy” when compared to the value of DNA-BC for pretrial authentication of identity and informed decision-making about arrestees.
The Dissent’s Reasoning
Like the majority, the four dissenters recognized that “free-form” balancing is not generally available, but they drew the line at a different point. The Scalia opinion asserted that “[t]he Fourth Amendment forbids searching a person for evidence of a crime when there is no basis for believing the person is guilty of the crime or is in possession of incriminating evidence. … Whenever this Court has allowed a suspicionless search, it has insisted upon a justifying motive apart from the investigation of crime.” The dissent then argued that, when Maryland expanded its DNA databanking law to encompass arrestees, the legislators were not thinking primarily (if at all) about things like setting bail and catching escapees who may have changed their other biometric features. Those lawmakers were out to enhance the efficacy of the state database in catching criminals by using it to associate the unidentified crime-scene samples with a larger collection of known samples (from arrestees plus the previous base of convicted offenders). And that the constitution forbids—no matter how minor the intrusion on the person and on privacy. Why? Because, in the eyes of the dissenters, that is the way it is—the Fourth Amendment’s "prohibition [on suspicionless searches] is categorical and without exception.”
There are, however, a number of difficulties with this reading of the constitution and the case law. For one, the Court has upheld suspicionless searches whose primary purpose was to develop investigatory leads or find evidence of guilt. The sole purpose of the roadblock in a case known as Illinois v. Lidster
was to discover the identity of a hit-and-run driver. There, Justice Scalia joined the Court’s opinion concluding that the search fell into the special needs category in part because the invasion of the interests of the motorists who were stopped for questioning was minimal. So too, the Court in Samson v. California
, with Justice Scalia’s approval, upheld a police officer’s suspicionless search of the clothing of a parolee. Although these cases might be distinguished from King, they show that the dissent’s “categorical” rule is not “without exception.” A great many judges in state and federal courts alike were convinced that the categorical rule did not apply to DNA databases
Because the dissent dismissed the majority’s “free-form reasonableness” analysis as impermissible, it did not directly question the details of that balancing or argue that, in a sensible and complete balancing of the relevant state and individual interests, the former should prevail. In fact, the dissent objected to the balancing precisely because it thought that such balancing would justify far more than the collection of DNA from arrestees (a matter that I address below).
At the same time, the dissent did maintain that arrestee databasing, as actually practiced in Maryland and elsewhere, was inconsistent with the “legitimate” purposes the majority attributed to DNA-BC. The state did not collect the sample, analyze it and upload the profile as part of the booking process. Days, weeks and months went by for this process to be completed. King’s profile was not checked against the existing offender and arrestee indices to see whether he was who he claimed to be. For the dissent, all that mattered was that “DNA … was [not] used for identification [in the sense of authentication] here.” “[T]he proud men who wrote the charter of our liberties would [not] have been so eager to open their mouths for royal inspection,” and that was that.
The Unresolved Issues
When lawyers describe a judicial opinion, they speak of its “holding” as opposed to its “dicta.” The holding, narrowly stated, depends on the facts of the case that were clearly important under the reasoning of the judges. Other statements about the law are dicta. They do not bind other courts. The facts in King that might limit the reach of the Court’s opinion are that DNA sampling was confined to “serious offenses” (violent crimes and burglaries); that officials had no discretion to pick and choose whose DNA to acquire; that a physical intrusion into the body took place; that the loci tested revealed no sensitive medical information; that the state trawled the forensic index (of unsolved crime-scene profiles) only for matches to the arrestee (and not for partial matches that might point to immediate relatives); and that the profiling and uploading occurred after formal charges. Which of these factors are actually critical to the Court’s conclusion that the Fourth Amendment allows DNA-BC?
A “serious offense” limitation? The Kennedy majority cautioned that “the necessary predicate of a valid arrest for a serious offense is fundamental.” The dissenters could not “imagine what principle could possibly justify this limitation … .” In their view:
If one believes that DNA will “identify” someone arrested for assault, he must believe that it will “identify” someone arrested for a traffic offense. This Court does not base its judgments on senseless distinctions. At the end of the day, logic will out. When there comes before us the taking of DNA from an arrestee for a traffic violation, the Court will predictably (and quite rightly) say, “We can find no significant difference between this case and King.” Make no mistake about it: As an entirely predictable consequence of today's decision, your DNA can be taken and entered into a national DNA database if you are ever arrested, rightly or wrongly, and for whatever reason.
However, one can sensibly suggest that, on average, people arrested for minor traffic offenses are less likely to be hiding their true identities and to have incriminating DNA samples at crime scenes than are people arrested for far more serious matters. Under “free-form” balancing, this is a logically relevant consideration. Whether this difference is significant enough to change the outcome is debatable, of course, but the outcome is not “entirely predictable.”
If the dissent were correct in lamenting that “your DNA can be taken and entered into a national DNA database if you are ever arrested, rightly or wrongly, and for whatever reason,” everyone could be rounded up (arrested on no basis whatever) and typed. If such a case ever arose, however, one could argue that King simply does not apply. It does not follow that just because it is permissible to acquire DNA from properly arrested individuals, the government also is free to retain wrongfully acquired samples and profiles.
But what about a system that acquired DNA from everyone without trampling on their Fourth Amendment right to freedom from arbitrary arrests? What if the state made DNA donation at the time of taking a driving test a condition for issuing a driver’s license? What if neonatal screening tests for genetic conditions now performed as a public health measure (and thus permissible under a special-needs analysis) were expanded to include STR profiling, with the identification profiles being uploaded to law enforcement databases? Police never would touch these samples (or the babies), but a population-wide database could emerge
. Universality would remove the disparate impact by race and class that infuses the current system and would intensify the level of public scrutiny of responses to any abuses in its administration. Although it may be a nonstarter politically, the principled case for a population-wide database is not frivolous
Uniformity of application
A key factor for the majority’s approval of direct reasonableness balancing was that “the search involves no discretion that could properly be limited by the ‘[interpolation of] a neutral magistrate between the citizen and the law enforcement officer.’” In other situations, such as roadblocks for drunken driving, the uniformity or randomness of the impositions on affected individuals has been important. If government officials were left to pick and choose among arrestees or other groups of people, the Court would be less inclined to apply the balancing standard, and the DNA sampling would be impermissible under the usual per se rule for warrantless searches.
Intrusion into the body
All the justices seem to agree that the physical intrusion in buccal swabbing is minor. Obviously, this fact is critical to the majority’s balancing. But suppose that the dissent were correct in proposing that even the most trivial physical intrusion justifies applying its categorical rule. A state could collect DNA even less intrusively. At some point, would not the collection fall below the threshold of oppressiveness required for a “search” to exist? For example, the government might only ask for a hand to be placed on a sticky pad. The dissent might respond that the “proud men who wrote the charter of our liberties would [not] be so eager to [move their limbs] for royal inspection,” but neither would those “proud men” be so eager to have their pictures taken by a royal photographer, and photography itself does not even rise to the level of a search that requires justification under the Fourth Amendment
Noncoding loci and statutory privacy protections
The Court makes much of the fact that the CODIS alleles are not transcribed to RNA that is translated into proteins. Although an amicus brief from a group of scientists cautioned that the term “junk DNA” has a scientific meaning that is different than its colloquial one
, the Court wrote that “[t]he CODIS loci are from the non-protein coding junk regions of DNA, and ‘are not known to have any association with a genetic disease or any other genetic predisposition. Thus, the information in the database is only useful for human identity testing.’” Yet, the fact that STRs may be considered “junk” from an evolutionary perspective and that they do not code for protein does not necessarily mean that they do not affect the quantity or timing of gene expression. Moreover, CODIS loci certainly can be used to make inferences about a few family relationships and to give rough indications of biogeographic ancestry. All this was known to the Court.
Although there are reasons to be skeptical of claims that the particular tetranucleotide sequences used in CODIS-compatible databases are likely to cause or to be strongly predictive of any medical conditions, suppose that these sequences turned out to be clinically relevant. Not surprisingly, the Court implies that this could alter the balance of interests: “If in the future police analyze samples to determine, for instance, an arrestee's predisposition for a particular disease or other hereditary factors not relevant to identity, that case would present additional privacy concerns not present here.”
Nevertheless, the Court suggests that even the use of loci that reveal highly sensitive medical information might not change the constitutional calculus—as long as there are sufficient “statutory protections.” To make this critical point, the Court quoted language from a 1977 medical privacy case, Whalen v. Roe
, about the salutary effect of “a ‘statutory or regulatory duty to avoid unwarranted disclosures’ … .” Because of such protections and a history devoid of privacy breaches, the Whalen Court rejected a constitutional challenge to a New York database of all prescriptions for controlled substances.
Outer-directed database trawling (a.k.a. “familial searching”)
Maryland is almost alone in explicitly prohibiting trawling in a database for the purpose of detecting very close relatives who might be the source of the crime-scene sample. The Court noted this part of Maryland’s laws, but it did not consider whether this ban actually mattered in its balancing.
It seems doubtful that a court should strike down a law providing for arrestee sampling—anymore than it should invalidate one that mandates sampling after conviction—just because a jurisdiction uses or might use “familial searching” software. If a constitutional challenge to this procedure arises, a court can address it. Although such “outer-directed” trawling has been criticized as a grave invasion of the privacy of individuals whose DNA is in the database as well as that of their relatives
, the argument that the practice, if implemented properly, is unconstitutional has been criticized
Profiling before charging
Finally, the King decision is limited to a system that defers DNA analysis until charges have been brought. On the one hand, the dissent complained that this delay undercuts the majority’s claim that DNA-BC is for authenticating the identity of the arrestee. On the other hand, the majority seemed comforted by the existence of a judicial finding of probable cause to believe that the arrestee is guilty of an offense. In particular, Justice Kennedy wrote that “[o]nce an individual has been arrested on probable cause for a dangerous offense that may require detention before trial, however, his or her expectations of privacy and freedom from police scrutiny are reduced.”
Although initially appealing, it is not clear why a line of constitutional magnitude must be drawn at the point of a probable cause determination. If what the Court meant by “reduced” “expectations of privacy and freedom” is that the state may subject a prisoner to other humiliating and privacy-reducing procedures, such as unannounced inspections of cells and even intensive strip searches of the person, then the question becomes why those losses justify additional ones. That a person loses some—even a great deal of—privacy because he is in custody does not automatically reduce his interest in maintaining other aspects of privacy. Police may not rush out and search an arrestee’s apartment without a warrant just because he has numerically fewer “expectations of privacy and freedom” while he is jailed.
Nonetheless, a finding of probable cause to hold an individual for trial does justify continued detention, which makes delayed DNA testing for the pretrial purposes enumerated by the majority more justified. For example, that an individual charged with a crime committed a brutal rape bears on his flight risk, since this person might fear that he will be convicted, which would then lead to a DNA match and, in turn, to an additional conviction for the rape. Furthermore, to the extent people who are validly charged are more likely to have committed DNA-related crimes than are people who have been arrested but not validly charged, the state has a greater interest in trawling the crime-scene profiles for possible matches.
Consequently, a prosecutor’s decision to bring some charges and a judicial finding of probable cause are not irrelevant. But are they decisive? A prosecutor may choose not to pursue a case even when there is probable cause, and a magistrate might mistakenly find that such cause is lacking even when it is actually present. In these situations, the state would lose the opportunity to discover whether the arrestee is linked to other crimes. Taking DNA and completing a database search immediately avoids this loss to the government’s interests. As long as DNA sampling is minimally intrusive and the privacy interests in the identification profiles are weak, the net balance of state and individual interests does not seem to change substantially if the DNA sampling and analysis occur at the initial booking
, as they can with microfluidic technology.
Retention without conviction
Finally, the King opinions do not address the reasonableness of retaining samples or profiles even if a conviction does not follow the arrest. Maryland law provides for the automatic destruction of samples and records in that situation. Some other jurisdictions place the burden of requesting expungement on the individual. Although this difference could have important practical consequences, it seems too slight to tip the scales against constitutionality.
But what about a law that allows the state to retain the samples or profiles indefinitely, even in the absence of a conviction? The majority’s position was that there is a presumption that sensitive data (beyond the individual’s apparent link to a crime) will not be disclosed or used. This presumption, it reasoned, was sufficient to protect all legitimate privacy interests. If that is so, then it seems that continued retention and trawling also are permissible. However, under the direct reasonableness standard, it also can be argued that the state has less reason to trawl crime-scene databases for matches to people who never have been convicted of the offenses that trigger inclusion in offender or arrestee databases and who no longer are subject to pretrial detention or supervision.
As far as legal doctrine goes, Maryland v. King is about the necessary conditions that permit the Court to engage in program-specific balancing as opposed to deciding according to a categorical rule for warrantless, suspicionless searches. The Court did not consider the straightforward step of crafting a new exception, confined to biometric data, to the per se rule of unconstitutionality, and it did not limit the few cases that speak of a “primary purpose” requirement rather than a simpler multiple-needs basis for balancing. As a result, the majority balanced without explicitly assessing all of the benefits of DNA databases. These justices surely understood that DNA may solve an unrelated case, but they justified collecting the DNA on grounds specific to pretrial detention. They amassed as many state interests as they could find—other than the most important one of prosecuting arrestees for other and possibly more serious crimes than those for which they were arrested.
The Court’s inquiry into direct reasonableness (but only for purposes other than criminal investigation) was adequate to resolve the case before it, but it leaves important questions about the design of DNA databases unresolved. Litigation on some of these matters, initiated before the King decision, now will continue
, and still more will follow.
Nearly half the states do not take DNA prior to conviction. King preserves their opportunity to expand their databases. If they choose to do so, they should not necessarily copy Maryland’s law but should consider issues such as whether to continue to tie the system to individuals swept into the criminal justice system; which offenses to select for DNA collection under the current approach; how long to retain samples and profiles; and how best to protect against the disclosure of any powerfully private information.