Biometrics and Fourth Amendment 'overbreadth'

The U.S. and California Supreme Courts are in the midst of deciding whether law enforcement may constitutionally collect DNA samples from persons just because they are arrested for certain crimes. See Maryland v. King, No. 12-207; People v. Buza, S169200. The Supreme Courts' decisions will likely turn on the extent to which they adhere to prior decisions analyzing the suspicionless collection and use of biometrics and whether they embrace or eschew a relatively new concept of Fourth Amendment "overbreadth."

Biometrics are the characteristics that define each of us physiologically. They include everything from our outward physical appearance to our fingerprints to our bodily fluids, which can be analyzed for alcohol, drugs or DNA. Because biometrics do not reveal our inner thoughts, the primary constraint on their use by law enforcement is the Fourth Amendment - not the privilege against self-incrimination. See United States v. Dionisio, 410 U.S. 1, 5-9 (1973).

Some biometrics do not implicate the Fourth Amendment at all. We have no reasonable expectation of privacy in our outward appearance, the sound of our voice or our handwriting; as long as these biometrics are collected from public places or through a grand jury subpoena, there is no "search" or "seizure" to regulate. Dionisio, supra; United States v. Mara, 410 U.S. 19, 21 (1973); cf. Hayes v. Florida, 470 U.S. 811, 816-17 (1985) (arrest to obtain fingerprints is "seizure").

However, obtaining a set of fingerprints, a buccal (or cheek) swab, or a blood sample requires a physical touching; these personal trespasses are "searches." Skinner v. Railway Labor Executives' Ass'n, 489 U.S. 602, 616 (1989); Cupp v. Murphy, 412 U.S. 291, 295 (1973) (obtaining fingernail scrapings is a "search"). The collection of a urine sample is a "search" because the visual or aural witnessing of its production intrudes upon privacy interests. Skinner, supra. The subsequent chemical analysis of these biometrics constitutes a separate "search." Id., at 617-18.

Thus far, the suspicionless collection and analysis of biometrics covered by the Fourth Amendment has turned on a balancing of the government's interests in obtaining the biometric data against the individual's privacy interests. E.g., Skinner, supra; Ferguson v. City of Charleston, 532 U.S. 67 (2001).

These searches are more likely to be upheld if they serve "special [governmental] needs" beyond solving crimes such as supervising parolees, Samson v. California, 547 U.S. 843 (2006); supervising employees in sensitive jobs, National Treasury Employees Union v. Von Raab, 489 U.S. 656 (1989) (U.S. Customs employees); Skinner, supra (railroad workers); or supervising schoolchildren involving in extracurricular activities, Vernonia Sch. Dist. 47J v. Acton, 515 U.S. 646 (1995); Board of Educ. v. Earls, 536 U.S. 822 (2002).

The individual's privacy interests are a function of the volume of information revealed by the biometric and the extent to which the individual has otherwise surrendered his or her expectations of privacy. Not all biometrics reveal the same quantum of information. Our outward physical appearance and fingerprints can be used to identify us and for little else. Our bodily fluids can reveal not only prior drug and alcohol use relevant to certain crimes, but also various medical conditions such as pregnancy, diabetes and epilepsy. Our DNA is in some sense the ultimate biometric because it is the molecular blueprint that chemically defines us as individuals. However, individuals can dilute their privacy interests in these biometrics. Prisoners and parolees, for instance, have a significantly diluted expectation of privacy that extends to their biometrics. United States v. Kincade, 379 F.3d 318, 833-39 (9th Cir. 2004).

Using this rubric, courts have affirmed the constitutionality of statutes authorizing the suspicionless collection of DNA samples from conditionally released convicted criminals. Kincade, supra. Convicts have a diminished expectation of privacy and collecting their DNA serves the "special need" of supervising these individuals and ensuring that they do not re-offend. These considerations obviously do not apply to persons merely arrested for crimes and courts accordingly confront a closer question when analyzing the constitutionality of mandatory DNA testing for persons arrested for all or some felony offenses. 42 U.S.C. Section 14135a(a)(1)(A) (all felonies); Cal. Penal Code Sections 296.1(a)(1)(A), 296(a)(2)(C) (all felonies); Md. Pub. Safety Section 2-504(a)(3)(i) (completed and attempted crimes of violence and burglaries).

By and large, the decisions examining the constitutionality of obtaining DNA from arrestees have hinged on the degree to which they have put DNA on all fours with the longstanding and long-accepted practice of fingerprinting. To be sure, there is room to question whether fingerprinting arrestees is constitutional (as does the defendant in King) given that fingerprinting was ostensibly "grandfathered" in as part of routine booking practice when the Fourth Amendment moved to a privacy-based paradigm in the 1960s. But assuming fingerprinting is upheld, it differs from DNA in both speed and overbreadth.

Currently, fingerprinting yields quicker results than DNA testing. It takes only minutes to obtain a set of prints, run them through a database, and learn whether the arrestee is using a false name, has a criminal record, or is implicated in any unsolved crimes. This means that fingerprinting can be used not only to solve past crimes, but also to serve the "special needs" of making more intelligent bail decisions and better assessing the arrestee's custodial security classification. Accord Florence v. Bd. of Chosen Freeholders, 132 S. Ct. 1510, 1515-22 (2012) (strip searches for jail security purposes). DNA analysis takes weeks. The results of DNA testing thus come too late to substantially serve these "special needs." This distinction is nevertheless unlikely to persist. In another 18 or 24 months, "rapid DNA" technology will be able to produce results in a matter of hours.

Thus, the key difference between fingerprinting and DNA is "overbreadth." Both can be used to identify or exonerate individuals, but a person's DNA (unlike fingerprints) also contains his or her entire genome - all 3.2 billion rungs in the DNA helix's ladder. That ladder contains a vast amount of information. Scientists have already correlated risk of certain cancers with mutations in particular genes and, as will be decided by the U.S. Supreme Court this term, have sought to patent those genes. See Association for Molecular Pathology v. Myriad Systems, No. 12-398. If science is one day able to correlate anti-social or violent behavior with other mutations, it would be possible to use one's DNA to prove future dangerousness in post-custodial civil commitment proceedings, in sentencing proceedings or even in preventative detention hearings. DNA is accordingly overbroad because it contains more information than the identity-establishing data than is constitutionally justifiable.

Congress and state legislatures are aware of this overbreadth and have taken steps to ameliorate it by limiting examination to 13 locations (or "loci") along the DNA strand that are considered "junk DNA" because they are thought not to correspond to any traits. These statutes make it a crime to analyze any other portion of the DNA strand or to use the DNA for anything other than identification. 42 U.S.C. Section 14135e; Penal Code Section 299.5(i)(1)(A).

It is an open question whether these statutory restrictions are sufficient to cure the overbreadth. The 3rd U.S. Circuit Court of Appeals has held that the federal DNA statute's restrictions suffice. United States v. Mitchell, 652 F.3d 387, 407-08 (3d Cir. 2011). This is consistent with precedent affirming blood and urine testing statutorily limited to analyses for drugs and alcohol. See Skinner, supra; Von Rabb, supra; Acton, supra. The depublished California Court of Appeal decision in Buza ruled that the California statute's similar restrictions were insufficient. 197 Cal. App. 4th 1424, depublished.

Where there is no authorizing statute, federal courts have recently started to invalidate overbroad suspicionless searches. In United States v. Jones, 132 S. Ct. 945 (2012), five Justices suppressed all data from a GPS tracker installed without a warrant because it collected too much data. Along the same lines, the 9th Circuit in United States v. Cotterman, 2013 U.S. App. LEXIS 4731 (Mar. 8, 2013), recently held that warrantless forensic analysis of a computer seized at the border was invalid because it collected more data than a cursory search.

The outcome in King and Buza may well come down to whether the Supreme Courts extend this emerging line of precedent to the DNA statutes or instead follow the older cases accepting statutory limitations. Either way, they may dictate whether "overbreadth" becomes a feature of the Fourth Amendment as well as the First.