State v. Medina

2014 VT 69

State v. Medina (2012-087), State v. Hewitt, Jr. (2012-101), State v. Goodrich (2012-102), State v. Ramos (2012-103), State v. Abernathy (2012-207), State v. Hartz (2012-231), State v. Gerrow (2012-309)

2014 VT 69

[Filed 11-Jul-2014]

NOTICE:  This opinion is subject to motions for reargument under V.R.A.P. 40 as well as formal revision before publication in the Vermont Reports.  Readers are requested to notify the Reporter of Decisions by email at: JUD.Reporter@state.vt.us or by mail at: Vermont Supreme Court, 109 State Street, Montpelier, Vermont 05609-0801, of any errors in order that corrections may be made before this opinion goes to press.

2014 VT 69

Nos. 2012-087, 2012-101, 2012-102, 2012-103, 2012-207, 2012-231 & 2012-309
State of Vermont	Supreme Court
v.	On Appeal from
	Superior Court, Addison Unit,
Ronald Medina	Criminal Division
State of Vermont	March Term, 2013

v. Superior Court, Rutland Unit,

Criminal Division

Douglas J. Hewitt, Jr.

State of Vermont Superior Court, Rutland Unit,

Criminal Division

     v.

Shane T. Goodrich

State of Vermont Superior Court, Rutland Unit,

Criminal Division

     v.

Ricardo Ramos

State of Vermont Superior Court, Chittenden Unit,

Criminal Division

     v.

William Abernathy

State of Vermont Superior Court, Windsor Unit,

Criminal Division

     v.

Tyler J. Hartz

State of Vermont Superior Court, Orleans Unit,

Criminal Division

     v.

Jeffrey Gerrow, et al.

Helen M. Toor, J. (12-087)

Cortland Corsones, J. (12-101, 12-102 & 12-103)

Alison S. Arms, J. (12-207)

M. Patricia Zimmerman, J. (12-231)

Robert R. Bent, J (12-309)

William H. Sorrell, Attorney General, and Bridget C. Asay and John Treadwell, Assistant

  Attorneys General, Montpelier, for Plaintiff-Appellant.

Matthew F. Valerio, Defender General, and Joshua S. O’Hara, Appellate Defender, Montpelier,

  for Defendants-Appellants.

PRESENT:  Reiber, C.J., Dooley, Skoglund, Burgess and Robinson, JJ.

¶ 1. DOOLEY, J.   Defendants in these consolidated cases challenge as unconstitutional a recent amendment to Vermont’s DNA-database statute that, as of July 1, 2011, mandates warrantless, suspicionless DNA collection and analysis from anyone arraigned for a felony after a determination of probable cause.  20 V.S.A. § 1933(a)(2). All five of the trial courts in these cases found that the amendment authorizes unconstitutional searches and seizures, either under the Vermont Constitution, Chapter I, Article 11, or under the Fourth Amendment to the U.S. Constitution, or both.  We affirm, addressing only the compliance of the statute with the requirements of Article 11 of the Vermont Constitution.

¶ 2. We repeat at the outset that our holding today pertains only to the Vermont Constitution and not to the U.S. Constitution.  After the trial courts in these cases issued their opinions addressing both constitutions, the U.S. Supreme Court decided that a similar Maryland statute—one that authorized warrantless, suspicionless DNA collection from persons arrested for violent crimes or burglary[1]    —is constitutional under the Fourth Amendment.  Maryland v. King, 133 S. Ct. 1958, 1980 (2013).  We delayed our ruling in these cases to consider the import of King[2]     and to allow additional briefing on the matter.  Having done so, we recognize that there are two possible bases to differentiate this case from King: (1) the Vermont statute sufficiently differs from the Maryland statute involved in King to produce a different result under the Fourth Amendment;[3]     and (2) the heightened standards and requirements of Article 11 of the Vermont Constitution compel a different result.  We have examined the second basis and determined that the result is different.  We have not analyzed the first possible basis in depth, although differences are noted as we encounter them.  Nor does the outcome of the Fourth Amendment analysis determine compliance with the Vermont Constitution, as we have firmly established that Article 11 is more protective in this area than its federal counterpart.  State v. Cunningham, 2008 VT 43, ¶ 16, 183 Vt. 401, 954 A.2d 1290 (“We have consistently held that Article 11 provides greater protections than its federal analog, the Fourth Amendment . . . .” (citing State v. Berard, 154 Vt. 306, 576 A.2d 118 (1990))); see generally State v. Jewett, 146 Vt. 221, 500 A.2d 233 (1985) (expounding on the necessity of state constitutional analysis that is independent from federal constitutional analysis, with particular reference to Article 11 as distinct from the Fourth Amendment).  We strike down the amendment to 20 V.S.A. § 1933 as unconstitutional solely under Vermont Constitution Chapter I, Article 11. Although we discuss King, it is only to determine whether we would adopt parts of its reasoning in our Article 11 analysis.

¶ 3. With that preamble, we begin with a discussion of the evolution of the statute and its DNA-collection mandate.  We then turn to a summary of our own Article 11 jurisprudence as it currently exists regarding the special-needs doctrine.  Next, we examine and distinguish King, as well as a handful of other Fourth Amendment decisions that we find to be helpful in determining the contours of Article 11.  Finally, we apply our Article 11 special-needs doctrine to the case at hand.

I.

¶ 4. As an initial matter, Vermont’s statutory scheme creates both a DNA data bank, which contains the DNA samples, and a DNA database, which contains the DNA records (also known as “profiles”) derived from the DNA samples.  20 V.S.A. §§ 1932(10)-(11), 1938(c)-(d).  In 1998, Vermont created the statewide DNA data bank and database and began populating them by collecting and analyzing DNA from those convicted of any statutorily defined “violent crime.”  1997, No. 160 (Adj. Sess.), § 1 (codified at 20 V.S.A. § 1932(12), which defined “violent crime,”[4]     and § 1933(a), which required a DNA sample from any person convicted of a violent crime).  In 2005, the Legislature expanded the statutory mandate to require a DNA sample and profile from all those convicted of any felony or attempted felony.  2005, No. 83, §§ 7, 8 (codified as amended at 20 V.S.A. §§ 1932(12), 1933).[5]      We upheld this expansion as constitutional under Article 11 in State v. Martin, 2008 VT 53, ¶ 35, 184 Vt. 23, 955 A.2d 1144.[6]     The most recent amendment, enacted in 2009, expands further those subject to DNA sampling by adding the following language: “The following persons shall submit a DNA sample: . . . A person for whom the court has determined at arraignment there is probable cause that the person has committed a felony in this state on or after July 1, 2011.”[7]      2009, No. 1, § 24 (codified at 20 V.S.A. § 1933(a)(2)).  It is this requirement, expanding mandatory DNA sampling to those merely charged with a felony, but not yet convicted, that defendants challenge here.[8]      Hereinafter, we refer to felony charges on which probable cause has been found as “qualifying charges.”

¶ 5. The current laws governing the data bank and database are codified at 20 V.S.A. §§ 1931-1946.  Other than the expansion, described above, of the classes of people subject to DNA sampling under the scheme, the law remains essentially unchanged since 1998.  The policy section of the database and data bank law, § 1931, reads as follows:

It is the policy of this state to assist federal, state, and local criminal justice and law enforcement agencies in the identification, detection, or exclusion of individuals who are subjects of the investigation or prosecution of crimes.  Identification, detection, and exclusion may be facilitated by the DNA analysis of biological evidence left by the perpetrator of a crime and recovered from the crime scene.  The DNA analysis of biological evidence can also be used to identify missing persons.

The law allows analysis of DNA samples only “for law enforcement identification purposes,” “to assist in the identification of human remains,” and, “if personal identifying information is removed, for protocol development and administrative purposes.”[9]      Id. § 1937(a).  It also provides that DNA records “in appropriate circumstances . . . may be used to identify missing persons.”  Id. § 1941(b).  The statute specifically prohibits analysis “for identification of any medical or genetic disorder.” Id. § 1937(b).

¶ 6. The DNA sample is analyzed to produce a record, or profile, of identification information from the DNA loci specified for the Combined DNA Index System (CODIS), the national DNA repository maintained by the Federal Bureau of Investigation.  Id. § 1932(4).  Both the sample and its associated record may “be provided to law enforcement agencies for lawful law enforcement purposes.”  Id. § 1938(a).  The tissue or fluid from which the DNA is extracted “may be provided to law enforcement agencies only for DNA sample analysis for use in any investigation and prosecution.”  Id. § 1938(b).  The Vermont database shares its DNA profiles with the national CODIS database.  Id. §§ 1936, 1938(e), 1939(b).

¶ 7. The Legislature included several provisions to safeguard the integrity of the database and data bank and the privacy of the personal information contained therein.  The statutes contain a general confidentiality requirement, id. § 1941(a), impose criminal penalties for breach of that requirement, id. § 1941(c), and allow a private right of action for equitable relief and damages, including punitive damages and reasonable attorney’s fees, id. § 1941(d).  Criminal and civil penalties also attach to tampering or attempted tampering with DNA samples. Id. § 1945.  Additionally, for those convicted of a qualifying offense, DNA records must be expunged and samples destroyed if the qualifying offense is pardoned, or reversed and dismissed.  Id. § 1940(a)(1)-(2).  For those whose DNA is collected after arraignment on a qualifying charge, DNA records must be expunged and samples destroyed if the qualifying charge is dismissed or pled down to a nonqualifying charge, or if the qualifying charge is acquitted or downgraded to a nonqualifying charge at trial.  Id. § 1940(a)(3)-(5).  If, before the record is expunged, it yields a match with another record in the state or federal system, the record of that match is retained even though the sample itself and the original record are destroyed.  Id. § 1940(d). 

¶ 8. In its implementation of the database and data bank law, the State incorporates further safeguards to protect DNA privacy and minimize the intrusion on the individual.  The law provides for the DNA sample to be extracted from a blood draw unless a “less intrusive means” of collection is available. Id. § 1934.  The State’s current practice is to collect the sample via a cheek swab.  See Martin, 2008 VT 53, ¶ 22.  For purposes of this analysis we will assume, as we did in Martin, that the State uses only cheek swabs to collect DNA.[10]      Id.  We do not analyze any other method of collection, nor have the parties asked us to do so.

¶ 9. The State maintains a separate database for convicted-offender records, arraignee records, and unknown forensic-sample records (i.e., unsolved crime samples).  The DNA samples, the DNA records generated from the samples, and the identifying information of the subject—name, date of birth, fingerprints, height and weight—are all kept in separate files, with only a common numeric identifier to link them together. 

¶ 10. DNA profiling is accomplished by analyzing a DNA sample at thirteen standard loci within a subject’s chromosomes to determine which genetic variations are present at each location.  The thirteen loci were originally selected as a national standard because they are highly variable among individuals and because they were thought to have no known associations with disease or other personal medical information.  When a DNA sample is analyzed, the variations at these locations are identified as a series of numbers and letters.  This string of numbers and letters constitutes the “profile” that is uploaded to the state and federal (CODIS) data banks. 

¶ 11. Defendants in these cases have all been arraigned on qualifying charges and subsequently refused to give a DNA sample.  The State moved to compel them to do so, and they have each had a hearing on the issue pursuant to 20 V.S.A. § 1935(b).  At their sampling hearings, each claimed that the statute violates the Vermont Constitution.  See State v. Wigg, 2007 VT 48, ¶ 5 n.3, 181 Vt. 639, 928 A.2d 494 (mem.) (stating that constitutional issues may be raised at such hearings despite statutory proscriptions limiting the nature of the hearing).  As discussed below, see infra ¶¶ 26-27, all five trial courts[11]     agreed with defendants that mandatory DNA sampling prior to conviction runs afoul of Article 11, albeit each with slightly different reasoning.

II.

¶ 12. Article 11 of the Vermont Constitution protects against unlawful searches and seizures.  Vt. Const. ch. I, art. 11.  We have previously determined that the DNA sampling mandated by § 1933 constitutes two distinct searches under Article 11:  “The initial taking of the DNA sample, either by blood draw or by buccal swab, and the subsequent analysis, storage and searching of the DNA profile are independent intrusions upon personal security that merit scrutiny under Article 11.”  Martin, 2008 VT 53, ¶ 14 (discussing the 2005 amendment to § 1933, mandating DNA sampling from all convicted felons).

¶ 13. Article 11 states as follows:

That the people have a right to hold themselves, their houses, papers, and possessions, free from search or seizure; and therefore warrants, without oath or affirmation first made, affording sufficient foundation for them, and whereby any officer or messenger may be commanded or required to search suspected places, or to seize any person or persons, his, her or their property, not particularly described, are contrary to that right, and ought not to be granted.

Vt. Const. ch. I, art. 11. Article 11 thus requires both a “warrant” and “oath or affirmation . . . affording sufficient foundation”—also known as probable cause.[12]     See State v. Berard, 154 Vt. 306, 310-11, 576 A.2d 118, 120-21 (1990) (stating that Article 11 contains warrant and probable cause requirements).  “The language of Article 11 does not expressly limit its protection to ‘unreasonable’ searches and seizures as does the Fourth Amendment to the United States Constitution.  This Court, however, has consistently interpreted Article 11 as importing the ‘reasonableness’ criterion of the Fourth Amendment.”  Id. at 309, 576 A.2d at 120 (citing State v. Jewett, 148 Vt. 324, 328, 532 A.2d 958, 960 (1986)).  “Although warrantless searches are sometimes permitted under Article 11, these exceptions must be ‘jealously and carefully drawn.’ ”  State v. Savva, 159 Vt. 75, 85, 616 A.2d 774, 779 (1991) (quoting Jewett, 148 Vt. at 328, 532 A.2d at 960).  Warrantless searches are thus per se unreasonable.  State v. Meunier, 137 Vt. 586, 588, 409 A.2d 583, 584 (1979).  “The warrant requirement favors decisionmaking by the judicial branch, a neutral and detached magistrate, rather than by the executive branch, the officer engaged in the often competitive enterprise of ferreting out crime.”  Savva, 159 Vt. at 86, 616 A.2d at 780 (quotation omitted).  The warrant requirement likewise favors judicial decisionmaking over legislative decisionmaking—that is, evaluations made on a case-by-case basis, with particularized suspicion, rather than on the issuance of “general warrants”—or laws that may essentially function as general warrants.  State v. Record, 150 Vt. 84, 86, 548 A.2d 422, 424 (1988).  We generally afford legislative enactments the presumption of constitutionality.  See State v. Hance, 2006 VT 97, ¶ 6, 180 Vt. 357, 910 A.2d 874.  In this case, however, the presumptive unconstitutionality of warrantless searches and seizures trumps our baseline deference to the Legislature.  See State v. Birchard, 2010 VT 57, ¶ 17, 188 Vt. 172, 5 A.3d 879 (“Where defendant had an expectation of privacy . . . the burden then shifts to the State to show a warrantless search is not prohibited under Article 11.”).  Defendants, like the rest of us, have an expectation of privacy in their oral cavity and in the information contained in their DNA. See Martin, 2008 VT 53, ¶ 21 (identifying the two privacy intrusions).  The State therefore has the burden to prove the constitutionality of 20 V.S.A. § 1933(a)(2).  Id. ¶ 9.

¶ 14. With that said, “Article 11 does not contemplate an absolute prohibition on warrantless searches or seizures.”  Id. (quoting Welch, 160 Vt. at 78-79, 624 A.2d at 1110).  But we do not depart from the standard lightly.  As we have stated before: “this Court will abandon the warrant and probable-cause requirements, which constitute the standard of reasonableness for a government search that the Framers established, only in those exceptional circumstances in which special needs, beyond the normal need for law enforcement, make the warrant and probable-cause requirement impracticable.”  Berard, 154 Vt. 306 at 310-11, 576 A.2d at 120-21 (quotation omitted).  If we find a special need, our next step is to “turn to a balancing of the competing public and private interests at stake.”  Martin, 2008 VT 53, ¶ 21.

A.

¶ 15. We announced the “special needs” exception to the warrant requirement of Article 11 in the context of approving random, suspicionless searches of prison inmates’ cells.  Berard, 154 Vt. at 311, 576 A.2d at 121.  We determined that “[r]equiring the State to demonstrate that it has special needs for a warrantless, suspicionless search or seizure ‘focuses attention on the nature and extent of those needs and allows the courts . . . to pursue the necessary balancing test in a manner calculated to interfere least with preservation of [individual] rights.’ ”  Martin, 2008 VT 53, ¶ 9 (quoting Berard, 154 Vt. at 311, 576 A.2d at 121). Once the State has proven its special need, beyond the ordinary needs of law enforcement, we balance the strength of the State’s need against the privacy intrusion.  Id. 

¶ 16. In the prison context in Berard, we based our conclusion that the State had met its burden of proving special needs “in part on the inexorable nature of prison governance in general and in part on the particular circumstances of the facts found by the trial court.”  154 Vt. at 312, 576 A.2d at 121.  As to the nature of prison governance in general, we found that “if the prisoners’ right to privacy prevented random prison cell searches, it would be impossible to accomplish the objectives of guarding against drugs and other contraband, like illicit weapons, thwarting escape, and maintaining a sanitary and healthful environment.”  Id. at 312, 576 A.2d at 121-22.  As to the particular circumstances found by the trial court, we noted the trial court’s statements that “[p]ossession of contraband by inmates is an ongoing concern among correctional officials,” that the commissioner of corrections is statutorily obligated to “maintain security, safety and order,” and that “it is difficult to see how the department [of corrections] could fulfill its primary objective of the disciplined preparation of offenders for their responsible roles in the open community, if it did not have an effective procedure for detecting contraband.”  Id. at 313, 576 A.2d at 122 (quotations and citations omitted).

¶ 17. We also found a public-safety-related special need in the context of a warrantless seizure of a gun from a car that was about to be impounded after a driving-while-intoxicated stop in State v. Richardson, 158 Vt. 635, 636, 603 A.2d 378, 379 (1992) (mem.).  The seizure was justified by the “obvious prudence” of removing the gun, in comparison to the “unacceptable danger to the public at large” if the gun were left in place.  Id. at 635-36, 603 A.2d at 379.

¶ 18. In the case we have often cited as the precursor to our special-needs jurisprudence, we upheld random roadside sobriety checkpoints as compliant with Article 11.  Record, 150 Vt. at 90, 548 A.2d at 426.  Such seizures were acceptable because they “enabled the police to apprehend intoxicated drivers who may have otherwise posed a serious threat to society.” Id. at 86, 548 A.2d at 424.  In addition, Article 11’s proscription on general warrants was mitigated because “the written police guidelines prevented arbitrary police conduct, and the scope of the roadblock was narrowly drawn.”  Id. 

¶ 19. In the probation context, although we did not rely on the special needs test per se, we found that a warrantless (but not suspicionless) search of a probationer’s home was acceptable in part because of the “special needs of the state in administering its probation program” and in part because “if a probation term provides for warrantless searches and the terms of the probation are narrowly tailored to fit the circumstances of the individual probationer, the Griffin ‘reasonable grounds’ standard strikes the proper balance between probationer privacy rights and public protection concerns.”  State v. Lockwood, 160 Vt. 547, 559, 632 A.2d 655, 663 (1993) (relying on the reasoning of Griffin v. Wisconsin, 483 U.S. 868, 875-76 (1987), which held that the special needs of probation supervision allowed for warrantless searches justified by “reasonable grounds” rather than by probable cause).  Our reasoning in that case relied on the public-protection goals of the probation system, as balanced against the diminished privacy rights of probationers. Id. at 559-60, 632 A.2d at 663. 

¶ 20. Finally, in the context of DNA sampling of convicted felons under 20 V.S.A. § 1933 in Martin, we found special needs beyond the normal needs of law enforcement.  We borrowed from the reasoning of the New Jersey Supreme Court under the Fourth Amendment and the New Jersey state analog and found that “the central purposes of . . . DNA testing are not intended to subject the donor to criminal charges,” 2008 VT 53, ¶ 16 (quoting State v. O’Hagen, 914 A.2d 267, 278 (N.J. 2007)), despite the defendants’ insistence that the policy objectives listed in § 1931 stated otherwise.[13]     Rather, we decided that the purposes of “creat[ing] a DNA database and . . . assist[ing] in the identification of persons at a crime scene should the investigation of such crimes permit resort to DNA testing of evidence” are beyond ordinary law enforcement purposes.  Martin, 2008 VT 53, ¶ 16 (quoting O’Hagen, 914 A.2d at 279).  Such a “long-range special need . . . does not have the immediate objective of gathering evidence against the offender.”  Id. (quoting O’Hagen, 914 A.2d at 278).  Recognizing that using DNA to determine who committed a past crime is fulfilling an ordinary law enforcement purpose, Martin drew a distinction between past and future crimes in the paragraph expressing its central holding:

We conclude that the O’Hagen reasoning also applies under Article 11, and that DNA sampling and analysis to assist in identifying persons at future crime scenes is a special need beyond normal law enforcement.  Vermont’s DNA database statute has as its stated purpose “to assist federal, state and local criminal justice and law enforcement agencies in the identification, detection or exclusion of individuals who are subjects of the investigation or prosecution of violent crimes.”  20 V.S.A. § 1931.  These purposes are distinct from the normal law-enforcement activities of investigating particular people for crimes already committed.

Id. ¶ 19 (emphasis added).  We also pointed to the secondary statutory purpose of “identifying missing persons” as beyond normal law enforcement.  Id. ¶ 20.  Finally, we found that sampling and indexing DNA from convicted felons may serve to deter recidivism.  Id.

B.

¶ 21. The second step in the special needs analysis is balancing the public and private interests at stake.  In Berard, we upheld random prison cell searches by considering “the State’s paramount interest in institutional security” versus “the inmates’ residuum of privacy rights.”  154 Vt. at 313, 576 A.2d at 122 (quotation omitted).  We acknowledged that, while an inmate’s “rights may be diminished by the needs and exigencies of the institutional environment, a prisoner is not wholly stripped of constitutional protections.”  Id. (quoting Wolff v. McDonnell, 418 U.S. 539, 555-56 (1974)).  We found several factors germane to protecting the inmates from abusive or arbitrary searches: “(1) the establishment of clear, objective guidelines by a high-level administrative official; (2) the requirement that those guidelines be followed by implementing officials; and (3) no systematic singling out of inmates in the absence of probable cause or articulable suspicion.”  Id. at 314, 576 A.2d at 122.  We held that the inmates’ diminished privacy and possessory interests were not unreasonably burdened by such “routine, random and warrantless search[es],” in the face of the State’s great need for promoting “institutional security.”  Id. at 318, 576 A.2d at 124.

¶ 22. In Martin, we analyzed the two different privacy intrusions separately: “(1) the initial sampling by buccal swab, and (2) the subsequent analysis, indexing, and searching of the information obtained.”  2008 VT 53, ¶ 21.  As to the cheek swab, we found that the invasion was “minimally intrusive,” relying mainly on our precedent upholding DNA sampling by buccal swab under a nontestimonial order based on reasonable suspicion. Id. ¶ 23 (citing In re R.H., 171 Vt. 227, 762 A.2d 1239 (2000)).

¶ 23. Most of our analysis in Martin therefore focused on the second intrusion—the “analysis, indexing, and searching” component of § 1933.  We rejected arguments that the state DNA data bank presages “an inexorable march . . . to a dystopian future of eugenics, gene-based discrimination, and other horribles worthy of Aldous Huxley.”  Id. ¶ 25.  As with the prison search policies at issue in Berard, we looked to individual privacy safeguards and restrictions on state actors built into the DNA-sampling statute to conclude that such totalitarian fears are unfounded.  For example, we determined that the thirteen DNA loci used to create an individual database profile “are not associated with any known physical trait” and are used by the State merely to create “a unique alphanumeric identifier,” useful only to “establish identity.”  Id. ¶ 26 (quotation omitted).  We further noted that the statutory scheme provides remedies for wrongful use or disclosure of confidential information and that we are obligated to presume that the government follows its own rules unless presented with evidence to the contrary.  Id. ¶¶ 28-29.  Finally, we recognized that the searches were “subject to clear administrative guidelines and . . . performed uniformly on all felons subject to them.”  Id. ¶ 30.  The DNA-sampling scheme therefore does not provide opportunity or pretext for the kind of individual harassment proscribed by Article 11.  Id. 

¶ 24. With these limitations in mind, we concluded that “the post-sampling intrusion on protected privacy interests is closely akin to that occasioned by the retention and searching of fingerprint records. . . . The data retained in the database serve only to prove identity.”  Id. ¶ 31.  We ultimately held that “the DNA sampling statute does not offend Article 11 as applied to nonviolent felons . . . .  The statute serves special needs beyond normal law enforcement and advances important state interests that outweigh the minimal intrusions upon protected interests.”  Id. ¶ 35.

¶ 25. In addition to our decision in Martin, we examine three other sources of information helpful to our decision.  While no source is dispositive, all three help frame our analysis and our evaluation of the arguments of the parties.  They are: decisions from our trial courts, decisions from other appellate courts around the country, and the decision of the U.S. Supreme Court in King.  In looking at the last two sources, we stress that these decisions do not purport to apply the unique standards of Article 11, but nevertheless are helpful in our analysis.

¶ 26. First we look at the five superior court decisions that are on appeal in this case.  Each of the five lower court opinions consolidated here distinguished arraignee DNA sampling from convicted felon DNA sampling, which we upheld in Martin.  All held that under the special-needs balancing test, as explained in Martin, the state’s need for the DNA samples at the time of arraignment was outweighed by the defendant’s privacy interest.  One held that we need not reach the balancing test because the State did not show a sufficient special need.  On a more complete evidentiary record, three held that defendant’s privacy interest was enhanced because the court found that the DNA sample could be used to show more than identification.

¶ 27. The thorough analysis of the trial courts informs our balancing analysis as discussed infra.  These decisions are in “stark contrast,” post, ¶ 89, with the dissent’s assertion that Martin “almost entirely controls the instant case.”  Post, ¶ 71.  Because we base our decision on the balancing analysis, we do not revisit the existence of a special need as explained in Martin, except to respond to the dissent’s attempt to enlarge the special need recognized in that decision.  Nor do we rest our decision on the conclusion of the Chittenden Superior Court, adopted by the Windsor and Orleans Superior Courts, that the “CODIS loci provide information beyond mere identity.”  State v. Abernathy, No. 3599-9-11 Cncr, at 24 (Vt. Super. Ct. June 1, 2012).  All parties have included in their briefs extensive arguments about this conclusion.  We do not reach those arguments.   

III.

¶ 28. Mandatory preconviction DNA sampling is a trending legal topic across the country; many states have recently passed legislation analogous to § 1933(a)(2),[14]     and resulting constitutional challenges are proliferating.  As discussed earlier, the U.S. Supreme Court has now settled the question of Fourth Amendment challenges to such laws in King, 133 S. Ct. 1958.  King, a 5-4 decision, was decided under a standard of reasonableness, weighing the governmental interest against the degree to which the search intrudes on privacy.  133 S. Ct. at 1970.  The Court judged the privacy interest invaded by the buccal swab to be minimal, concluding that “[t]he fact that an intrusion is negligible is of central relevance to determine reasonableness.”  Id. at 1969.  Moreover, the Court reasoned that a person arrested for a felony, the initial trigger under the Maryland law, has a reduced expectation of privacy.  Id. at 1978. Although the Court did not hold that the analysis of the defendant’s DNA is a second search, as we did in Martin, it noted that the analysis is only of loci “that do not reveal the genetic traits of the arrestee.”  Id. at 1979.  The Court also concluded that under the circumstances, intrusion on the privacy interests involved “does not require consideration of any unique needs that would be required to justify searching the average citizen.”  Id. at 1978.  Thus, it held that “the [federal] special needs cases, though in full accord with the result reached here, do not have a direct bearing on the issues presented.”  Id.

¶ 29. The Court found one primary government interest: “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.”  Id. at 1970.  Included in this interest is the right to search incident to an arrest.  Id. at 1970-71.  The Court explained that the search for identity is broader than determining the arrestee’s name and social security number.  Drawing on the use of fingerprints, the Court noted that identity includes the arrestee’s criminal history and involvement in unsolved crimes.  Id. at 1971-72.  It includes whether the custody of the person—for example, because of a history of violence—creates a risk for detention facility staff, for other detainees, or for the arrested person.  Id. at 1972.  It includes whether a defendant presents a risk of flight because of the disclosure of other crimes.  Id. at 1972-73.  Finally, it includes the extent to which the arrestee is dangerous to the public.  Id. at 1973.

¶ 30. The result of the Court’s balancing is contained in the last paragraph of the majority opinion:

In light of the context of a valid arrest supported by probable cause respondent’s expectations of privacy were not offended by the minor intrusion of a brief swab of his cheeks.  By contrast, that same context of arrest gives rise to significant state interests in identifying respondent not only so that the proper name can be attached to his charges but also so that the criminal justice system can make informed decisions concerning pretrial custody. Upon these considerations the Court concludes that DNA identification of arrestees is a reasonable search that can be considered part of a routine booking procedure.  When officers make an arrest supported by probable cause to hold for a serious offense and they bring the suspect to the station to be detained in custody, taking and analyzing a cheek swab of the arrestee’s DNA is, like fingerprinting and photographing, a legitimate police booking procedure that is reasonable under the Fourth Amendment. 

Id. at 1980.    

¶ 31. King resolved the Fourth Amendment challenge to arrest-based DNA-sampling statutes.  It could not, of course, determine the constitutionality under state constitutional provisions.  Despite the extensive adoption of preconviction DNA-sampling requirements, prior to this case, no court has decided the issue independently based on a state constitution. The Minnesota Court of Appeals in In re Welfare of C.T.L. held, in the context of a juvenile delinquency action, that a state statute requiring the juvenile defendant to submit to the taking of a DNA sample where the court has found probable cause that the juvenile committed a felony violates the Fourth Amendment to the U.S. Constitution and Article I, Section 10 of the Minnesota Constitution.  722 N.W.2d 484, 492 (Minn. Ct. App. 2006).  That decision, however, came before the decision of the Minnesota Supreme Court in State v. Bartylla, which upheld the Minnesota statute requiring DNA samples from convicted felons under the Fourth Amendment and held that the Minnesota Constitution offered no greater protections in this context than the Fourth Amendment.  755 N.W.2d 8, 18-19 (Minn. 2008).  Given Bartylla, we cannot view Welfare of C.T.L. as an independent state constitutional decision.  See also State v. Johnson, 813 N.W.2d 1, 6-7 (Minn. 2012) (explaining that, in Bartylla, the court interpreted the Minnesota Constitution coextensively with the Fourth Amendment).  The remaining state court decisions are based on the Fourth Amendment and not on an independent state constitutional ground.  See Mario W. v. Kaipio, 281 P.3d 476 (Ariz. 2012); People v. Lowe, 165 Cal. Rptr. 3d 107 (Ct. App. 2013) (following the U.S. Supreme Court decision in King), petition for review granted by People v. Lowe, 320 P.3d 799 (Cal. 2014); State v. Franklin, 76 So. 3d 423 (La. 2011); King v. State, 42 A.3d 549 (Md. 2012), rev’d sub nom. Maryland v. King, 133 S. Ct. 1958 (2013); Anderson v. Commonwealth, 650 S.E.2d 702 (Va. 2007).

¶ 32. A number of these out-of-state decisions are helpful to our analysis, even though they are based on the Fourth Amendment, because they analyze many of the arguments we face in our state constitutional decision.  We are particularly influenced by the Maryland Court of Appeals opinions in King v. State and the opinions in the closely divided en banc decision of the U.S. Court of Appeals for the Third Circuit in United States v. Mitchell, 652 F.3d 387 (3d Cir. 2011), upholding the federal DNA sample statute.  We have also drawn from Welfare of C.T.L. and Mario W. in the decision that follows.

¶ 33. The State urges us to deny the constitutional challenge based on the rationale of King—that the DNA sample requirement is part of a search incident to an arrest, and more particularly a booking search, and is constitutionally indistinguishable from a fingerprint requirement. Instead, for three main reasons, we reject that King’s analysis controls whether the Vermont statute complies with Article 11.  The third reason—the difference between the requirements of Article 11 and the Fourth Amendment—is the fundamental one, but the first two provide important context for that analysis.

A.

¶ 34. First, the Maryland statute challenged and upheld in King is triggered by arrest, and the Court’s rationale is based on the right of law enforcement officers to search incident to an arrest, particularly where the alleged criminal will be incarcerated pending trial.  King, 133 S. Ct. at 1970-71.  Thus, the Court emphasizes the need to identify accurately the person who is being arrested and charged with a crime; to know that person’s history; to know the risks the person presents for facility staff, other detainees and the arrested person; to ensure the arrestee will appear for trial; and to assess the danger the arrestee presents to the community.  Id. at 1971-73.  The Vermont statute is triggered by the judicial finding of probable cause “at arraignment” for a felony, a determination that normally follows the defendant being brought to court and the preparation of an information.  20 V.S.A. § 1933(a)(2); see V.R.Cr.P. 5(c).  The defendant may never be arrested, and even if arrested, may have been released from pretrial detention.  Not surprisingly, identification of arrested persons is not included in the stated purposes for the Vermont DNA-collection law.  See 20 V.S.A. § 1931.  Nor is it a special need identified in Martin.

¶ 35. The mismatch between the triggers in the Maryland and Vermont laws means that the Supreme Court’s rationale in King applies to only some of the defendants covered by Vermont’s statute.  Further, limitations contained in the Vermont law—the requirement of a judicial probable-cause determination, the limitation to felonies and the expungement of the DNA evidence where defendant is not convicted—are insignificant under the Supreme Court decision.[15]    

B.

¶ 36. Second, identification of the arrestee, even if the defendant is arrested and continued in pretrial detention, is tangentially accomplished by post-arraignment DNA collection and analysis.  The current system of photographs and fingerprints fully responds to the need for identification of the defendant.  In the many cases now consolidated in this appeal, the State has identified none in which there is a need for more accurate identification.  By the time that DNA is analyzed, the risks connected with a defendant have been determined and reflected in pretrial detention provisions.  As the Rutland Superior Court observed, “it is unlikely that a DNA sample, taken post-arraignment, will be of much assistance” to ensure the accurate identity of the person arrested.  For this reason, identification of the defendant was not included as a special need in Martin.

¶ 37. The Supreme Court acknowledged this limitation but answered that (a) identification information, even if untimely, could become useful, and (b) improvements in technology will make DNA analysis quicker and more timely.  King, 133 S. Ct. at 1974, 1977.  The Court reasoned that “[r]egardless of when the initial bail decision is made, release is not appropriate until a further determination is made as to the person’s identity in the sense not only of what his birth certificate states but also what other records and data disclose to give that identity more meaning in the whole context of who the person really is.”  Id. at 1973-74.  The Court went on to note that it takes considerable time to arrange pretrial release, between 27 and 112 days in the federal system.  During this time period, the Court observed that “more about the person’s identity and background can provide critical information relevant to the conditions of release and whether to revisit an initial release determination.”  Id. at 1974.  These statements appear to be highly inapplicable in a jurisdiction like Vermont, where nonarrest or pretrial release is the norm and release pending trial is often accomplished in a matter of hours or a few days. See 13 V.S.A. § 7554(a) (stating that “[a]ny person charged with an offense, other than a person held without bail . . . [for life imprisonment felonies and certain other violent felonies] shall at his or her appearance before a judicial officer be ordered released pending trial in accordance with this section,” with highest priority placed on release on personal recognizance or unsecured appearance bond).  Moreover, King’s search for relevant information appears to have no boundaries in the determination of “who the person really is,” thus stripping the charged individual of all privacy interests.  Thus, the King rationale is an invitation to broader DNA collection and arrests for purposes of obtaining a DNA sample rather than only DNA collection for determining risk associated with a need for detention.  This rationale would justify a full analysis of the DNA sample, not only an analysis of the loci limited to identification.

¶ 38. We are unimpressed by King’s first answer—that identification information, even if untimely, could become useful.  In State v. Handy, 2012 VT 21, 191 Vt. 311, 44 A.3d 776, an Article 11 decision after Martin, we considered the constitutional validity of a statute that requires defendants convicted of a sex offense to submit to testing for sexually transmitted diseases.  See 13 V.S.A. § 3256.  The ostensible purpose of the requirement was to enable the victim to determine whether he or she was at risk of contracting a sexually transmitted disease from the perpetrator.  On this basis, the trial court had upheld the requirement without an evidentiary hearing.  We found from the legislative history of the statute that it had been enacted to obtain federal funding and the testimony presented to the legislative committees indicated that the requirement provided “no medical benefit for victims” because the testing was too late to respond to the need.  Handy, 2012 VT 21, ¶¶ 19-21.  We indicated that if obtaining federal funding were the sole governmental interest supporting the testing requirement, it would be constitutionally “suspect.”  Id. ¶ 22.  We held, however, that the purpose to give victims “peace of mind” was sufficient to uphold the statute in the balancing analysis.  Id. ¶ 23-24.  The relevant point of Handy is that our Article 11 analysis requires a valid and timely governmental interest, whatever the superficial governmental interest that is asserted.  We do not find here that untimely information gathering is a valid interest.

¶ 39. As to King’s second answer—that improvements in technology will make DNA analysis quicker and timelier—we are unwilling to speculate on the different functionality that improvements in technology and systems will bring, just as we are unwilling to predict what information can be found in the DNA analysis even where the prediction is supported by expert testimony, as in the record of the Chittenden Superior Court hearing.[16]      Based on current experience, we understand that the chemistry of the decision we are making may change in multiple directions.  The technology may better accomplish law enforcement goals but the invasion of privacy may change and increase, and the rules in place to protect legitimate privacy interests may become more fragile. Further, the Legislature may again amend the statute to redefine the government interests and methods as well as the protection of privacy interests.  For example, under the current statutory scheme, the State holds indefinitely the DNA sample even though the current authorized use is very specific and limited—but, in the future, the Legislature may authorize a broader use of the samples.

¶ 40. We must hold that our decision defining and enforcing Article 11 values and policy is based on the circumstances we encounter when the decision is made and on the statute before us.  If those circumstances change substantially, we will no doubt encounter this question, or a variation of it, again and view it anew.

C.

¶ 41. The third and most important reason to reject the King analysis is that it is inconsistent with precedents defining the underlying policies of Article 11.  King represents a large expansion of law enforcement’s power to search incident to an arrest.  As that doctrine developed in Chimel v. California, the power to search incident to an arrest, without a warrant, is limited to the person arrested and the area within the immediate control of the person arrested.  395 U.S. 752, 763 (1969).  In United States v. Robinson, the Supreme Court explained that Chimel provided a bright-line rule and a search incident to arrest “requires no additional justification.”  414 U.S. 218, 235 (1973).  In United States v. Edwards, the Court explained that property available to be searched at the time of arrest can be searched later at the place of detention, even if the search does not actually occur for some time after the arrested person’s detention.  415 U.S. 800, 807-08 (1974).  In reaching this conclusion, the Court quoted from the U.S. Court of Appeals decision in United States v. DeLeo, 422 F.2d 487, 493 (1st Cir. 1970):  “While the legal arrest of a person should not destroy the privacy of his premises, it does—for at least a reasonable time and to a reasonable extent—take his own privacy out of the realm of protection from police interest in weapons, means of escape, and evidence.”  Edwards, 415 U.S. at 808-09.  Also relevant in this line of cases is New York v. Belton, which held that “when a policeman has made a lawful custodial arrest of the occupant of an automobile, he may, as a contemporaneous incident of that arrest, search the passenger compartment of that automobile,” including any containers therein.  453 U.S. 454, 460-61 (1981) (footnotes omitted), modified by Arizona v. Gant, 556 U.S. 332, 344 (2009) (holding that searching the passenger compartment of a vehicle incident to arrest is only lawful if the area is within reaching distance of the arrestee).

¶ 42. In State v. Bauder, 2007 VT 16, 181 Vt. 392, 924 A.2d 38, and State v. Neil, 2008 VT 79, 184 Vt. 243, 958 A.2d 1173, this Court rejected much of this federal search-and-seizure jurisprudence in applying Article 11 of the Vermont Constitution.  Bauder specifically rejected the bright-line approach of Belton as “an adequate basis for relaxing the fundamental limitation on governmental power represented by the warrant requirement.” 2007 VT 16, ¶ 20.  Thus, it required that in order to avoid the warrant requirement, the officers must demonstrate that the search is needed because of exigent circumstances—to secure the safety of the officers or to preserve evidence of a crime.  Id. ¶ 22.

¶ 43. In Neil, the defendant was arrested on an outstanding arrest warrant issued because the defendant failed to pay a small fine.  After the arrest, the officers patted the defendant down and found a rolled-up bill with white powdery residue on it and a closed black pouch which they opened to find cocaine.  In holding that Article 11 precluded a warrantless search of the pouch, this Court rejected the bright-line holding of United States v. Robinson as applicable to Article 11.  We held that “the police must get a search warrant before searching a closed container unless ‘exceptional’ circumstances—risk of undue delay, destruction of evidence, or danger to officers—make getting a warrant impracticable.”  2008 VT 79, ¶ 7.  We explained why the case did not meet the standard of “exceptional circumstances”:

[T]he exigency must be factually and narrowly tied to the circumstances that rendered a warrant application impracticable.  Here, the officers knew defendant and knew he had no history of violent behavior or carrying weapons.  The evidence does not show, nor is it argued, that the officers subjectively believed that the circumstances necessitated a warrantless search.  The State concedes that the pouch was not threatening or suspicious.  With defendant in custody, once the officers seized the pouch, any danger to the officers or the public was eliminated.

Id. ¶ 13 (citation omitted). 

¶ 44. We find a broad warrantless-search authorization, under the theory that it is a search incident to an arrest, to be inconsistent with the requirements of Article 11 as we have developed them.  To the extent we have recognized the validity of a warrantless search incident to an arrest, it has been in cases where exigent circumstances were present.  While it is possible that the fruits of a DNA search will produce information bearing on conditions of release or confinement with respect to a particular defendant, that possibility alone is insufficient to justify a warrantless DNA search of every defendant, with no distinction among those who will be searched.

¶ 45. In reaching this conclusion, we recognize that we have never held that a warrantless booking search of a detainee’s person or property is inconsistent with Article 11 or that routine fingerprinting of arrestees is prohibited by Article 11.  However we decide the validity of these routine practices under Article 11, they do not justify the DNA sample capture involved here.  We do not equate a procedure that takes a visible image of the surface of the skin of a finger with the capture of intimate bodily fluids, even if the method of doing so is speedy and painless.  More important, despite the occasional usefulness of DNA samples for ordinary identification as described in King, the real functionality, and statutory purpose, is to solve open criminal cases or ones that may occur in the future.  While part of this functionality may respond to a special need as we held in Martin, it is far afield from the immediate concern for the protection of arresting officers or the destruction of evidence, the concerns underlying our search-incident-to-arrest doctrine.  The real expansion of warrantless search power in King is “its reimagination of the idea of ‘identity’ to include criminal history and other information.”  E. Murphy, License, Registration, Cheek Swab: DNA Testing and the Divided Court, 127 Harv. L. Rev. 161, 177 (2013).  Despite the assurances of the Court in King, it is difficult to see any limit on what information may be gathered about an arrestee and the effect of that information gathering on the decision whether to arrest.

¶ 46. The Court in King held that its decision was not grounded on its special-needs jurisprudence, and that there need not be a special need to justify the search under the Maryland statute.  133 S. Ct. at 1978. Our conclusion from the foregoing analysis of the King rationale is that the Vermont DNA-collection statute can be upheld under Article 11 only if it meets the requirements of our special-needs doctrine, as defined primarily in Martin.

IV.

¶ 47. Our special-needs test requires that (1) the statute fulfills a special need, beyond the normal needs of law enforcement, and that (2) the balance between public interests and private interests at stake weighs in favor of allowing the search or seizure.  Martin, 2008 VT 53, ¶ 9.  We turn to that analysis.  We focus, as we did in Martin, on the search involving the “analysis, indexing, and searching of the information obtained” from the DNA sample, rather than the search involving the cheek swab itself.  Id. ¶¶ 21-22.

¶ 48. The State has articulated no special need for preconviction DNA sampling beyond the special needs we found in Martin for postconviction DNA sampling.  We held in Martin that the main special need for postconviction DNA sampling is that the report will “assist in identifying persons at future crime scenes” and this need is “a special need beyond normal law enforcement.”  Id. ¶ 19.  We distinguished this from “normal law-enforcement activities of investigating particular people for crimes already committed.”  Id.  We also identified secondary special needs of identifying missing persons, 20 V.S.A. § 1931, and deterrence, because the convicted person knows his or her DNA record is available for comparison to any evidence left at a future crime scene.  Martin, 2008 VT 53, ¶ 20.

¶ 49. As noted earlier, despite the attenuation of the special needs identified in Martin, we do not ground our decision on the absence of a special need.  As with Martin¸ our decision rests on the balancing of interests in the second prong of the test.  We do, however, recognize the more limited special need in the balancing of interests that follows.  In this case, each defendant’s privacy interest is greater because he or she has not been convicted.  At the same time, the State’s interest is less weighty.  The issue then is whether the greater privacy interest of the individual and the lesser interest of the State are such that the balance is sufficiently different from that in Martin to invalidate the statute.

A.

¶ 50. We address first the State’s interests.  In order to determine its weight, we first define the State’s interests in collecting the DNA information at arraignment and, as a result, expanding the class of persons from whom DNA is collected.  This class includes two subclasses: (1) persons who will be subject to the collection requirement on conviction, but under the statutory amendment will be subject to the requirement sooner; and (2) persons who will be subject to the requirement at arraignment but will not be convicted and would not have been subject to a collection requirement before the amendment to the statute.   

¶ 51. Before we address the subclasses, we stress two points.  First, irrespective of whether a defendant is ultimately convicted, the statutory amendment advances in time the point at which DNA is collected but generally produces the same result as the original statute at the time that the criminal case is over.[17]     Although the end result may be the same for both subclasses, there will be a differential overall impact on the two subclasses, as discussed below. The State’s interest that we are weighing involves only this limited period in which the case is pending.

¶ 52. Second, the special needs we recognized in Martin do not include “investigating . . . [defendant] for crimes already committed.”[18]      2008 VT 53, ¶ 19.  While we have no doubt that the State will investigate past crimes through DNA collected from defendants, the ability to do so does not represent a legitimate interest that we will weigh in the balancing process.  The interests to be weighed are those in investigating future crimes, deterring criminal conduct, or identifying missing persons.  Id. ¶ 20.  Two of these special needs, including the main one—investigating future crimes—involve a defendant’s conduct after arraignment on the charge that creates the DNA-sampling requirement.  Because of our first point, that conduct must occur between the time of arraignment and the time the criminal case is over.  The State’s interests we are weighing involve only investigating or deterring criminal conduct during that period or identifying missing persons.

¶ 53. With those points in mind, we address the subclasses.  For the former subclass, the issue is only the timing of the requirement.  Under the statutory amendment, the requirement to give the DNA sample will come sooner than conviction.  The State’s interest is thus primarily in having the DNA profile earlier, to deter criminal conduct of a type where DNA would be helpful to determine the perpetrator, occurring between arraignment and the end of the criminal case, and for comparison with DNA left at a crime scene during this period.  Secondarily, the State’s interest is in earlier DNA comparison with that of a missing person.  Of course, with respect to the primary interest in comparing DNA with that from future crime scenes, the effect of the statutory amendment is only timing, because the preexisting statute would allow DNA collection on conviction.  We recognize that the State also has an interest in accurate identification of persons who are subject to conditions of release, or those who are incarcerated pretrial—but as discussed above, the need for more accurate identification is rare and apparently has not arisen among the large numbers of defendants joined in this case.  We also note that conditions of release or pretrial incarceration generally impair the ability of a defendant to commit future crimes, and the weight of the State’s interest in solving and preventing future crimes through DNA collection is necessarily lower.  Moreover, as a general observation, the State has not shown why quicker access to the DNA is a weighty interest, and we cannot find it to be so.

¶ 54. The situation with respect to the second subclass of persons—those not convicted of a qualifying crime—is different.  The collected DNA sample is expunged if the criminal case ends without a conviction for a qualifying crime, although the record of any match is retained.  Therefore, without the amendment, the State loses the ability to use the DNA to solve crimes committed after arraignment and before the end of the criminal case, as well as the deterrent effect with respect to those crimes.  The State also loses the ability to match defendant’s DNA to that of missing persons.  It is in these situations that the State’s interest in the amendment has the greatest weight.

¶ 55. The dissent rejects the above analysis, arguing that “the State’s special need is the same” for felony convicts as it is for arraignees.  Post, ¶ 73.  In response, we note that this litigation is solely about the period between arraignment and conviction or nonconviction—nothing in our decision will affect DNA-collection requirements at conviction.  Further, we have not required the State to provide an additional special need beyond those recognized in Martin.  What we have examined, as we are required to do under the Martin analysis, is the weight of the special needs with respect to arraignees in the context of a law that will require DNA collection if they are ultimately convicted of felonies.  We would have conducted the same analysis if we had combined the consideration in this decision and in Martin—if the Legislature had adopted in the first instance a DNA-collection requirement that applied to arraignees—because the privacy interests of arraignees are stronger than those of convicted defendants.

¶ 56. Finally, we note that the Legislature has recognized the limited weight of the State’s interest in these samples by requiring expungement of the sample and profile when the adjudication is completed without a conviction of a qualifying crime.[19]     Thus, it has truncated the main special need found in Martin by requiring expungement where there is no conviction for a qualifying crime and preventing comparison with DNA gathered at future crime scenes.  20 V.S.A. § 1940(a).  As the Minnesota Court of Appeals held in Welfare of C.T.L.:  “This requirement [of expungement] suggests that the legislature has determined that the state’s interest in collecting and storing DNA samples is outweighed by the privacy interest of a person who has not been convicted.”  722 N.W.2d at 491.

B.

¶ 57. Next, we turn to defining the privacy interest to be weighed in the balancing test.  The privacy interest of the preconviction defendant is greater than the interest of one who has been convicted because a preconviction defendant has a presumption of innocence.  13 V.S.A. § 6502; see State v. Camley, 140 Vt. 483, 488, 438 A.2d 1131, 1133-34 (1981) (explaining operation of 13 V.S.A. § 6502 at trial); see also Coffin v. United States, 156 U.S. 432, 453 (1895) (“The principle that there is a presumption of innocence in favor of the accused is the undoubted law, axiomatic and elementary, and its enforcement lies at the foundation of the administration of our criminal law.”).  Indeed, a substantial percentage of persons from whom DNA samples will be taken will never be convicted of a qualifying offense.  The judicial finding of probable cause at arraignment is no substitute for a criminal conviction—the watershed moment that strips a defendant of the presumption of innocence and related privacy protections.  In saying this, we do recognize that pretrial arraignees have a legitimate expectation of privacy that is less than the population as a whole.  The lesser expectation may result in temporary incarceration or conditions of release that result in limitations on privacy.

¶ 58. The restrictions we place on the liberties of pretrial defendants, however—through pretrial detention, bail, or conditions of release—while at times substantial, are all tailored to ensure the State’s need for the defendant’s presence in court and the State’s need to reduce immediate risks to public safety.  E.g., State v. Roessell, 132 Vt. 634, 636, 328 A.2d 118, 119 (1974) (per curiam) (“[W]here there is sufficient evidence to demonstrate a substantial risk that a defendant will not show up for trial, conditions, monetary or otherwise, to insure his return are indicated[] and . . . where there is a danger to the public, conditions for the protection of public safety are appropriate.”); see also State v. Webb, 132 Vt. 418, 422, 320 A.2d 626, 629 (1974) (“The imposition of physically restrictive conditions of release pending trial upon a defendant whose release has been determined to constitute a danger to the public weighs heavily against the presumption of innocence.”).

C.

¶ 59. This brings us to a balancing of the interests of the defendant and the State.  In doing so, we start with the generalization, adopted by the U.S. Supreme Court in King and the dissent in this case, that the interests are the same for DNA collection as for fingerprinting and, if anything, the State’s interest in DNA is greater than for fingerprinting because of DNA’s greater utility.^[20]     Putting aside that one involves a bodily invasion and the other does not, we do not now believe we can equate fingerprinting and DNA retrieval. Fingerprints can show only identification, and they have limited functionality in solving old cases.  DNA samples provide a major new tool for investigation of open and future crime cases, as well as correcting wrongful convictions in closed cases.[21]     It is also important to note that the DNA samples being seized provide a massive amount of unique, private information about a person that goes beyond identification of that person.  See Martin, 2008 VT 53, ¶ 24.  Martin framed the DNA sampling scheme in terms of two searches: “[1] the initial taking of the DNA sample, and [2] the subsequent analysis, storage, and searching of the DNA profile.”  Id. ¶ 14 (emphasis added).  Martin is silent on the storage of the sample, which is retained by the State, apparently indefinitely.  Even after the identification profile is created from the DNA sample and is made part of the national database, the State retains the DNA sample.  Neither the statutory purposes, nor the State’s asserted justifications for the law, provide any rationale for retaining the DNA sample once the profile has been created.  While current law limits use of the sample, that law can be amended to allow greater use; the retention of the DNA sample suggests that expanded use is possible in the future.  We assume for purposes of this opinion that restrictions on use will be enforced, but the current restrictions are only a partial answer to the extent of the invasion of privacy as long as the State continues to hold the DNA samples.^[22]     Contrary to the dissent, we read defendants’ arguments as raising this feature of the statute as a factor weighing in the balancing test.

¶ 60. The point of the comparison is that DNA collection and use under the statute is a significantly greater invasion of the defendant’s privacy than that involved in fingerprinting, even if the DNA samples were expunged in all circumstances after the DNA profile is taken.  We do not accept the widespread use of pretrial fingerprinting of defendants as deciding this case.

¶ 61. From the foregoing analysis, the main weight of the State’s interest involves cases where the defendant is not convicted of a qualifying crime and the State never has the opportunity to gather a DNA sample.  In these cases, however, the defendant’s privacy interest is the greatest.  The Legislature recognized that interest by providing for expungement of DNA samples where the criminal case has ended without conviction for a qualifying crime.  The State has an interest in identifying defendants in its custody, as King held, but that interest is of little weight here. Traditional methods have identified defendants in the cases involved here, and many are not incarcerated in any event. 

¶ 62. Because of the limited weight of the State’s interest in the expansion of the DNA sampling requirement to defendants on arraignment for a qualifying crime, and the greater privacy interest of the defendant at that stage of the adjudication, we—like the Minnesota Court of Appeals in Welfare of C.T.L.—conclude that the balance tips to the defendant.  We also concur in the analysis of the Arizona Supreme Court that “[h]aving a DNA profile before adjudication may conceivably speed . . . investigations [of other crimes]. . . .  But one accused of a crime, although having diminished expectations of privacy in some respects, does not forfeit [constitutional] protections with respect to other offenses not charged absent either probable cause or reasonable suspicion.”  Mario W., 281 P.3d at 483.

¶ 63. The marginal weight of the State’s interest in DNA collection at the point of arraignment, balanced against the weight of the privacy interest retained by arraignees prior to conviction, persuades us to hold that 20 V.S.A. § 1933(a)(2), and associated sections, which expand the DNA-sample requirement to defendants charged with qualifying crimes for which probable cause is found, violates Chapter I, Article 11 of the Vermont Constitution.

Affirmed.

		FOR THE COURT:
		Associate Justice

¶ 64. REIBER, C.J., dissenting.   In holding that 20 V.S.A. § 1933(a)(2) violates Article 11 of the Vermont Constitution, the majority overstates the privacy interests of felony arraignees and understates the government’s important interests in identifying perpetrators and excluding the innocent—objectives that DNA analysis can accomplish with unparalleled accuracy.  Consequently, the majority unduly restricts the State’s ability to make good on its fundamental duty to do justice through enforcement of the law.  I therefore respectfully dissent.

¶ 65. This case pertains to DNA identification of felony arraignees.  In doing so, it provokes the deeper question of how to balance law enforcement efforts and privacy concerns in an age of rapidly evolving technology. This Court’s decision in State v. Martin provided a blueprint for preserving these competing interests.  2008 VT 53, ¶ 35, 184 Vt. 23, 955 A.2d 1144.  Martin recognized that, because DNA contains a tremendous amount of indisputably private physiological information, those accused, arraigned and convicted have a reasonable expectation of privacy in DNA—apart from any initial physical intrusion—and therefore an Article 11 search analysis applies.  Id. ¶ 14.  Martin’s acknowledgment of the privacy interests, couched within the language of the statute, led to our consequent focus on the scope of the search as a limiting principle.  This approach is prescient for future cases at this thorny intersection of law and technology.

¶ 66. Martin accomplished a sensible balancing that the majority would now uproot.  The majority says it is applying the analysis in Martin to the facts before us, but creates ambiguity where none exists.  Simply stated: for the same purposes articulated in the statute and under Martin’s precedent, felony arraignees have a reduced privacy expectation compared to the general population based on the fact that a neutral magistrate has found probable cause that they committed a serious crime.  Although no one disputes that arraignees have a reasonable expectation of privacy in personally intimate DNA, the scope of the search here is strictly limited by law to identifying information.  Considering the explicitly limited scope of the search under the statute, it is a stretch for the majority to construe Martin’s holding otherwise.

¶ 67. The majority employs reasoning that overlooks the common-sense distinctions between felony arraignees and the population at large and between personal DNA information and merely identifying DNA information.  The majority’s restriction on the State’s ability to use the best means at its disposal for identification, DNA evidence, is unnecessary and to the detriment of the criminal justice system as a whole. 

I.  

¶ 68. Article 11 presumptively requires a warrant, supported by probable cause, as a prerequisite to search or seizure.  State v. Berard, 154 Vt. 306, 310-11, 576 A.2d 118, 120-21 (1990).  A departure from this requirement is warranted “only in those exceptional circumstances in which special needs, beyond the normal need for law enforcement, make the warrant and probable-cause requirement impracticable.”  Id. (quotations omitted).  Once the State has proven a special need, “we balance the need served against the privacy intrusion at stake.”  Martin, 2008 VT 53, ¶ 9. 

¶ 69. In Martin we found adequate reason for such a departure, holding that the expansion of Vermont’s DNA database by statute from violent to nonviolent convicted felons complied with the requirements of Article 11. Id. ¶ 35.  We determined that the purpose of DNA identification—“to create a DNA database and to assist in the identification of persons at a crime scene should the investigation of such crimes permit resort to DNA testing of evidence”—was a special need beyond normal law enforcement.  Id. ¶ 16 (quoting State v. O’Hagen, 914 A.2d 267, 279 (N.J. 2007)).  We held that the Vermont statute was not concerned with ordinary law enforcement because it sought “to use DNA to accurately and efficiently identify persons in a variety of contexts, including subsequent criminal prosecutions.”  Id. ¶ 20.  In making this distinction, we emphasized that “[a]lthough the enumerated purposes [of the statute] may involve law enforcement to some degree, the central purposes of the DNA testing are not intended to subject the donor to criminal charges.”  Id. ¶ 16 (quoting O’Hagen, 914 A.2d at 278).  We endorsed the rationale given by the Second Circuit Court of Appeals, which in approving a similar statute found it “crucial that the state, in collecting DNA samples, is not trying to determine that a particular individual has engaged in some specific wrongdoing.  Although the DNA samples may eventually help law enforcement identify the perpetrator of a crime, at the time of collection, the samples in fact provide no evidence in and of themselves of criminal wrongdoing, and are not sought for the investigation of a specific crime.”  Id. ¶ 18 (quoting Nicholas v. Goord, 430 F.3d 652, 668-69 (2d Cir. 2005)).  Finally, we noted that the statute at issue in Martin did not perpetrate “the principal evil sought to be remedied by Article 11,” which is the issuance of general warrants that vest state officers “with unlimited discretion to intrude upon the privacy interests of particular individuals of their choice without particularized suspicion, in the hope of immediately discovering wrongdoing.”  Id. ¶ 18.  Having found a special need, we balanced the privacy interests of felony convicts in both the initial buccal swab and the DNA profiling against the State’s interest in accurate identification of future criminal perpetrators.  We concluded that the State’s important interests outweighed felons’ minimal privacy interests.  Id. ¶ 35.

¶ 70. In segmenting the steps employed to extract information from the designated group, and the objective to be accomplished from the information collected, Martin’s approach began to address the complexities introduced by technological developments in search and seizure capabilities.  See United States v. Weikert, 504 F.3d 1, 16-17 (1st Cir. 2007) (“[I]t may be time to reexamine the proposition that an individual no longer has any expectation of privacy in information seized by the government so long as the government has obtained that information lawfully.”).  Martin established, in accord with many courts, that there is a reasonable expectation of privacy in the personal information contained within DNA,[23]     and that DNA analysis is a search under Article 11.  It confirmed that the Article 11 inquiry is driven by the scope of the search—here, the identifying information of felony arraignees—and not by potential informational uses prohibited by statute.  2008 VT 53, ¶ 25 n.10.  Beyond identity, there is, of course, other information contained in DNA samples that is protected, where personal interests are heightened and privacy rights attach.  Such additional analysis is not addressed in the statute or the case and controversy before us and should not be presumed constitutional under Martin’s holding.  See Maryland v. King, 133 S. Ct. 1958, 1979 (2013) (“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.”). 

II.

¶ 71. Though the majority purports to rule today in the name of privacy, it does not engage the possibility that the approach in Martin and similar cases might better balance law enforcement prerogatives and privacy rights in our search and seizure jurisprudence.  Indeed, one would have to be forgiven for not realizing that Martin almost entirely controls the instant case—the only difference being that the population targeted by the DNA identification statute has been expanded from felony convicts to felony arraignees.  The minimal intrusion of the buccal swab, combined with recognition that advances in genetics might someday allow scientists to glean additional information from so-called “junk” or noncoding DNA,[24]     informed Martin’s central consideration that DNA samples are analyzed for the limited purpose of confirming or excluding identity, and that use for other purposes is illegal.  2008 VT 53, ¶ 25-32. The majority acknowledges Martin’s holding, but then proceeds to dislodge its careful balancing of public and private objectives.  Where the majority does not rehash issues decidedly foreclosed by Martin, its analysis of the State’s special need and its balancing of the State and private interests is flawed in several respects.

A.

¶ 72. First, the majority misconstrues the State’s special need.  The majority underestimates the value of DNA evidence and its superiority to other methods of identification, particularly as applied to felony arraignees, who, in contrast to arrestees, are subject to a neutral magistrate’s finding that probable cause exists to support a charge of a serious crime.  This is a finding denoting a matter of substantial public interest.

¶ 73. DNA identification fulfills the State’s special need to identify perpetrators at crime scenes, exonerate the innocent, deter crime, and identify missing persons.  Id. ¶¶ 15-20.  The primary purpose of the DNA sampling—creation of a database with identification information—signifies a “long-range special need that does not have the immediate objective of gathering evidence against the offender.”  Id. ¶ 16 (quoting O’Hagen, 914 A.2d at 279).  Because DNA identifying information is not inherently incriminating, and is not collected as part of an ongoing investigation that specifically targets defendants, the DNA sampling here does not fulfill ordinary law enforcement purposes. Although this case goes one step further than Martin, the State’s special need is the same, and applies to felony convicts as well as arraignees. 

¶ 74. It is axiomatic that Article 11 requires a relation between the State’s special need and the nature and extent of the intrusion.  See id. ¶ 9; Berard, 154 Vt. at 311, 576 A.2d at 121.  The mere fact that a population has a reduced privacy interest does not automatically justify suspicionless searches.  See Goord, 430 F.3d at 667 (noting that suspicionless searches cannot be justified merely on basis of reduced privacy, especially “in light of the wide swath of the general population who at one point or another has had a reduced expectation of privacy”); see also King, 133 S. Ct. at 1979 (cautioning that Court does not “suggest that any search is acceptable solely because a person is in custody” and explaining that the level of intrusion must be taken into account). 

¶ 75. Here, felony arraignment is a watershed event that signals that “probable cause exists to remove an individual from the normal channels of society and hold him in legal custody.”  King, 133 S. Ct. at 1971.  The court’s independent probable-cause finding is a judicially determined and constitutionally sufficient conclusion that, as a matter of the required probability analysis, there is sufficient evidence to support a charge that an arraignee has committed a serious crime.  Such a conclusion thus strengthens the nexus between the State’s need for DNA identification and the class subject to the intrusion.[25]     Such a finding also distinguishes DNA profiling of felony arraignees, in which the State has a strong interest, from profiling of the general population or those who have been arrested.  See State v. Lockwood, 160 Vt. 547, 559, 632 A.2d 655, 658 (1993) (holding that a warrantless search program, as applied to probationers, “permits a degree of impingement upon privacy that would not be constitutional if applied to the public at large” (quotation omitted)). Given the judicially-confirmed probable cause beyond mere suspicion, the purpose of the statute—to identify persons who have committed crimes in the past and those at future crime scenes—has ample relation to the class of persons sampled.      

¶ 76. The State’s need to identify defendants here is particularly important in light of the consequences of felony arraignment.  The majority rejects King’s reasoning that, under the Maryland statute, DNA identification is necessary to accurately process arrestees into custody, because it contends that such identification is unnecessary by the time a defendant reaches the arraignment stage.  Ante, ¶¶ 36-39.  But the need for accurate identification is more pressing at arraignment than at arrest, not less. Under long-settled law, once arraigned, an accused may be detained with or without bail, placed in home detention, or released subject to various conditions.  See 13 V.S.A. §§ 7551-7554.  Despite the majority’s assurance that “nonarrest or pretrial release is the norm” in Vermont, ante, ¶ 37, this exercise of the State’s police power must nevertheless be undertaken carefully and judiciously.  Proper identification of criminal defendants, using the most accurate means available, is fundamental to the State’s responsible use of its power.  See Hiibel v. Sixth Judicial Dist. Ct. of Nev., Humboldt Cty., 542 U.S. 177, 191 (2004) (“In every criminal case, it is known and must be known who has been arrested and who is being tried.”).

¶ 77. The majority’s attempt to denigrate the State’s special need does not stand against logic or policy clearly articulated by the Legislature.  The special needs identified in Martin—to identify perpetrators at crime scenes and exonerate the innocent—apply with equal force to felony arraignees as to convicts.

B.

¶ 78. The majority compounds its errors in its balancing of the State and private interests, the second step of our special needs analysis.  See Martin, 2008 VT 53, ¶ 9 (explaining that the special needs test requires courts “to pursue the necessary balancing test in a manner calculated to interfere least with preservation of [individual] rights” (quotation omitted)).  In order to reach its outcome, the majority understates the State’s interest and overstates defendants’ privacy interest.

¶ 79. In Martin, we held that the State’s interest in identifying persons at crime scenes is an important factor, given that “DNA is more accurate and far less susceptible to the various methods of deception employed by wrongdoers.”[26]     2008 VT 53, ¶ 33.  We concluded that “[t]his accuracy, and DNA’s concomitant ability to conclusively exonerate the innocent, weigh heavily in favor of the statute.”  Id.  This reasoning applies equally to felony arraignees.  The public’s faith in the criminal justice system to treat the accused fairly is bolstered through the use of identification techniques that lend greater accuracy to the process.  As explained above, this statute was not enacted to simplify or expedite police and prosecution procedure toward resolution of particular charges against an accused felon, but rather to further the integrity and accuracy of the criminal justice system.  As the Third Circuit Court of Appeals explained in upholding a statute requiring DNA identification of arrestees:

To the extent that DNA profiling assists the Government in accurate criminal investigations and prosecutions (both of which are dependent on accurately identifying the suspect), it is in the Government’s interest to have this information as soon as possible.  Collecting DNA samples from arrestees can speed both the investigation of the crime of arrest and the solution of any past crime for which there is a match in CODIS.  Moreover, use of CODIS promptly clears thousands of potential suspects—thereby preventing them from ever being put in that position, and advancing the overwhelming public interest in prosecuting crimes accurately. 

Mitchell, 652 F.3d at 414-15 (quotations and citations omitted).  The Mitchell court’s reasoning applies with equal force here.

¶ 80. The majority cannot adequately distinguish Martin’s holding, and thus whittles down the State’s important interest using artificial and arbitrary metrics.  The majority begins by framing the State’s interest as “involv[ing] only [the] limited period in which the case is pending,” on the basis that the arraignees’ DNA will be collected upon conviction in any event.  Ante, ¶ 51.  The time period between arraignment and conviction is not the question—it is the moment of the search.  The strength of the State’s interest at the time of sampling is not properly evaluated based on the passage of time.  Moreover, the majority’s reasoning that the expungement provisions diminish the strength of the State’s interest turns the part of the statute designed to limit the intrusion on privacy into a constitutional defect, as this statute is surely on stronger constitutional ground than it would be without an expungement requirement.  No court of which I am aware has required the State to prove a “special need for preconviction DNA sampling beyond the special needs . . . for postconviction DNA sampling,” as the majority does here.  Ante, ¶ 48 (emphasis added). Rather, the issue is the strength of the State’s interest in identifying the targeted population—felony arraignees.  See Mitchell, 652 F.3d at 413 (rejecting defendant’s argument that postconviction DNA samples would serve equally well as preconviction samples and holding that “collecting identifying information to aid law enforcement . . . applies with equal force to arrestees and pretrial detainees [as to convicts]”).  The majority has not identified any practical distinction between the uses of physically identifying DNA at conviction—identified in Martin as special needs beyond ordinary law enforcement—and the identical uses of this DNA at arraignment.  

¶ 81. The majority further understates the State’s interest based on a strained reading of Martin.  The majority claims that “the special needs we recognized in Martin do not include ‘investigating . . . [defendant] for crimes already committed.’ ” Ante, ¶ 52 (quoting Martin, 2008 VT 53, ¶ 19).  Aggregating this claim with its earlier contention that the State’s interest only has weight until the moment of conviction, the majority concludes that the State’s interest is limited to “investigating or deterring criminal conduct” during the period between defendants’ arraignment and conviction.  Ante, ¶ 52.  The problem with the majority’s conclusion is that Martin did not hold that the State’s interest is in investigating future crimes, but rather future crime scenes. See Martin, 2008 VT 53, ¶ 19 (holding that “DNA sampling and analysis to assist in identifying persons at future crime scenes is a special need beyond normal law enforcement” (emphasis added)).  Our emphasis in Martin on “future crime scenes” was not to define a literal timeline, but to again highlight that the creation of the DNA database is “a long-range special need that does not have the immediate objective of gathering evidence against the offender.”  Id. ¶ 16 (quotation omitted).  As the King court recognized, identifying DNA does not serve as direct proof of a crime.  133 S. Ct. at 1972 (“[T]he 13 CODIS loci are not themselves evidence of any particular crime, in the way that a drug test can by itself be evidence of illegal narcotics use.”).  Thus, DNA collection is ultimately about identification of defendants, to the end of increasing accuracy in the criminal justice system as a whole.  It is not about resolution of a particular crime, whenever committed.  O’Hagen, 914 A.2d at 279 (finding a special need “because the testing is not for the immediate investigation of a specific crime”).  Moreover, by the majority’s logic, only crimes committed after defendants’ DNA is entered in CODIS could constitutionally be matched to defendants’ DNA, a conclusion that is not only impracticable but ignores that evidence of a crime may only surface many years after the fact.

¶ 82. As to defendants’ privacy interest, Martin’s holding dictates that, like felony convicts, felony arraignees’ privacy interest in their identity is minimal if not nil.  It was settled before Martin that the buccal swab was so minimally invasive to protected privacy interests as to not require probable cause under Article 11.  In re R.H., 171 Vt. 227, 234, 762 A.2d 1239, 1244 (2000).  Martin confirmed the swab to be a minimal intrusion, comparing it to photography and fingerprinting at arrest, and held that “the initial sampling, taken alone, does not violate Article 11.”  2008 VT 53, ¶ 23.  The majority unsettles these well-established precedents.  In claiming that DNA sampling is unlike fingerprinting because “one involves a bodily invasion and the other does not,” ante, ¶ 59, the majority does not explain how the “light touch on the inside of the cheek” occasioned by the buccal swab is materially distinguishable from pressing a finger against a surface.  King, 133 S. Ct. at 1969.     

¶ 83. Regarding the second search, Martin held that “[i]n light of the statutory limits on the analysis of genetic information, the post-sampling intrusion on protected privacy interests is closely akin to that occasioned by the retention and searching of fingerprint records.”  2008 VT 53, ¶ 31.  The intrusion here is limited to identifying information and is thus minimal.  Like felony convicts, defendants, once arraigned, no longer have a legitimate privacy interest in their identifying information.  Jones v. Murray, 962 F.2d 302, 306 (4th Cir. 1992) (“[W]hen a suspect is arrested upon probable cause, his identification becomes a matter of legitimate state interest and he can hardly claim privacy in it.”).  As the King court explained, “DNA is another metric of identification used to connect the arrestee with his or her public persona, as reflected in records of his or her actions that are available to the police. . . .  [DNA is] a different form of identification than a name or fingerprint, but its function is the same.”  133 S. Ct. at 1972.    The majority has not justified its assertion that the privacy interest in merely physically identifying information is greater for felony arraignees than for convicted felons.  To the contrary, the majority’s reasoning highlights the logical contradiction in its contentions that, on the one hand, defendants’ privacy interests outweigh the need to identify them using DNA, and, on the other hand, that DNA identification is unnecessary because “[t]he current system of photographs and fingerprints fully responds to the need for identification of the defendant.”  Ante, ¶ 36.  How can defendants’ privacy interests preclude their identification using DNA when defendants have already been identified?  

¶ 84. Moreover, once lawfully obtained, the government’s matching of an “identification record against other records in its lawful possession does not infringe on an individual’s legitimate expectation of privacy.”  Boroian v. Mueller, 616 F.3d 60, 67 (1st Cir. 2010); see also Jones, 962 F.2d at 306 (“[T]he identification of suspects is relevant not only to solving the crime for which the suspect is arrested, but also for maintaining a permanent record to solve other past and future crimes.”). Defendants have no constitutionally protected privacy interest in their identifying information left at a past or future crime scene, and thus no constitutional right to keep that information from authorities.  United States v. Post, No. 3:13-CR-20, 2014 WL 345992, at *3 (S.D. Texas Jan. 30, 2014) (noting that hypothetical criminal perpetrator would not be able to suppress DNA found in his clothing left at crime scene “because he left the clothing in a public place and lost any expectation of privacy he had in it, regardless of how he contemplated that clothing could be used”). Once arraigned, defendants’ privacy interest in their physically identifying information is nonexistent. 

¶ 85. The majority further reasons that “DNA samples being seized provide a massive amount of unique, private information about a person that goes beyond identification of that person.”  Ante, ¶ 59.  But there is a crucial distinction between a DNA sample, which contains an individual’s entire genome, and the DNA profile derived from the thirteen CODIS loci, which is only used to obtain identifying information.  See Weikert, 504 F.3d at 16 (distinguishing DNA profiles from “the samples from which those profiles are created,” which “have the potential to reveal information about an individual’s health, propensity for certain diseases, and, perhaps, sexual orientation and propensity for certain conduct”).   Martin considered the second search, the DNA analysis, to encompass the “analysis, storage, and searching of the DNA profile.” 2008 VT 53, ¶ 14 (emphasis added).  We did not consider the separate intrusion in the long-term storage of the DNA sample; the majority departs from this framework without explanation. 

¶ 86. Defendants here have not directly challenged the seizure implicated by the State’s indefinite storage of their DNA samples, and the complex issues potentially raised by such a seizure have not been adequately briefed. Other courts in similar procedural postures have expressed concern that retention of DNA samples beyond portions necessary for identification may implicate additional privacy concerns, but declined to address the issue. Mitchell, 652 F.3d at 412 (noting distinction between DNA sample and profile, but holding that because defendant’s DNA sample had not yet been collected, he was “not in a position to challenge the retention of his sample”); Boroian, 616 F.3d at 70-71 (upholding retention of DNA profile and recognizing separate challenge to retention of DNA sample, but declining to address it based on defendant’s failure to raise the issue on appeal). 

¶ 87. I am mindful of the majority’s concern that technological advancements have enabled the government to analyze immense amounts of personal information.[27]      If the majority’s contention is that long-term retention of the DNA sample may be overly broad in accomplishing the State’s narrow goal of identification through the DNA profile, I share this concern.  In addressing this concern, however, not only does the majority consider issues not raised before us, but it also “go[es] back on our statement in Martin that we will not ground our decision on arguments about DNA that invoke speculation about massive incursions on privacy.”  See ante, ¶ 59 n.22.  Rather, we recognized in Martin that strict statutory limitations ensure that the DNA sampling is to be used only for identification purposes.  2008 VT 53, ¶¶ 26-27.  We noted that a “potential search, particularly one that is explicitly prohibited by statute, cannot be the subject of a case or controversy ripe for decision by this Court.”  Id. ¶ 25 n.10.  As in Martin, there is no reason to expect that the State will not comply with the limitations in the statute, id. ¶ 28; as the majority acknowledges, we assume “that restrictions on use will be enforced.”  Ante, ¶ 59; see also Judicial Watch, Inc. v. U.S. Dep’t of Health & Human Servs., 27 F. Supp. 2d 240, 243 (D.D.C. 1998) (“The Court must presume . . . that the Executive Branch is aware of its duty . . . to faithfully execute the law as enacted.”). 

¶ 88. The State’s strong interest in accurate identification of offenders, balanced against defendants’ negligible privacy interest in their identifying information, weigh in favor of the statute here.  Yet the majority balances what it considers “[t]he marginal weight of the State’s interest in DNA collection at the point of arraignment” against “the weight of the privacy interest retained by arraignees prior to conviction,” and concludes to the contrary.  Ante, ¶ 63.  The majority weighs the State’s interest too lightly and defendants’ privacy interest too heavily, erroneously tipping the scale in favor of defendants.  Though the presumption of innocence weighs in favor of defendants, this presumption has never—before today—justified blocking the State from using the most accurate means to identify the persons in its charge.

¶ 89. The majority’s attempt to distinguish this case from Martin through the balancing of the public and private interests rests on a series of cascading logical failures.  Reliance on Martin, while declining to apply its holding to the practically indistinguishable facts of this case, lends confusion to the jurisprudence.  The majority’s approach is in stark contrast with other courts, which have recognized the potential of DNA analysis to transform the criminal justice system at a time when other forensic identification methods have repeatedly and forcefully been called into doubt.  The stakes are high for both defendants and the State, and compel a decision that places emphasis on proper identification of the accused. For these reasons, I respectfully dissent.

¶ 90. I am authorized to state that Justice Burgess joins this dissent.

		Chief Justice

---

[1]     See King v. State, 42 A.3d 549, 566 (Md. 2012), rev’d sub nom. Maryland v. King, 133 S. Ct. 1958 (2013).  Under the Maryland statute, although the collection occurs at arrest, the processing of the DNA sample cannot begin until the arrestee is arraigned.  King v. State, 42 A.3d at 559.

[2]     Our references to King throughout refer to the U.S. Supreme Court case Maryland v. King, 133 S. Ct. 1958, 1980 (2013), not the underlying Maryland Court of Appeals case.

[3]     For example, defendants urge us to find that, because the Vermont statute authorizes DNA collection from all felony arraignees rather than just from arraignees charged with crimes of violence, the Vermont statute is unconstitutional under the Fourth Amendment while the Maryland statute is not.  Compare 20 V.S.A. § 1933(a)(2), with Md. Code Ann., Pub. Safety § 2-504(a)(3)(i). 

[4]     Statutorily defined violent crimes included murder, manslaughter, aggravated forms of assault and robbery, kidnapping, first-degree unlawful restraint, maiming, first-degree aggravated domestic assault causing serious bodily injury, sexual assault and aggravated sexual assault, lewd and lascivious conduct with or without a child, sexual exploitation of a child or an elderly or disabled adult, burglary, unlawful trespass of a residence, or attempt of any of the above-listed crimes.  1997, No. 160 (Adj. Sess.), § 1.

[5]     The amendment also required a DNA sample where a defendant is charged with an offense, conviction for which would trigger a DNA-sample requirement, and (1) probable cause was found for this offense, and (2) as part of a plea agreement there is a requirement for the defendant to give a DNA sample.  2005, No. 83, § 7 (effective June 28, 2005), amending 20 V.S.A. § 1932(12)(C) (codified at § 1932(12)(E)).

[6]     This decision did not address the expansion for plea agreements contained in 20 V.S.A. § 1932(12)(C), now § 1932(12)(E).

[7]     Unlike the statute considered in King, the Vermont statute does not allow collection of the sample upon arrest of the criminal defendant. Both collection of the sample, and its analysis, occur after judicial finding of probable cause and arraignment.  20 V.S.A. § 1933(a)(2).

[8]     The 2009 amendment also enhanced the list of crimes for which a sample is required on conviction to include domestic assault, pursuant to 13 V.S.A. § 1042, and “any crime for which a person is required to register as a sex offender pursuant to subchapter 3 of chapter 167 of Title 13.”  20 V.S.A. § 1932(12)(B), (C).  Since felonies were already covered, this part of the amendment added certain misdemeanors.  This enhancement is not involved in this case.

[9]     Examples of such purposes include “quality control” and “development of a population database.”  20 V.S.A. § 1937(a)(2)(A), (C).

[10]     The statute, 20 V.S.A. § 1934(a), states that the DNA sample “shall be obtained by drawing blood, unless the Department [of Public Safety] determines that a less intrusive means to obtain a scientifically reliable sample is available, in which event such less intrusive means shall be used.”  From the record, we assume that the department has determined that cheek swabs provide an acceptable less intrusive means.

[11]     Trial courts in other Vermont counties have stayed arraignee sampling and related hearings pending the outcome of this appeal.  The parties have represented that no pre-conviction DNA sampling is occurring under the current statute, pending this decision. 

[12]     By way of comparison, the Fourth Amendment to the U.S. Constitution states:

  The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

The requirements of a warrant and probable cause contained in each constitution, although similar, are not the same.  State v. Welch, 160 Vt. 70, 76, 624 A.2d 1105, 1108 (1992) (“[T]he right to be free from unreasonable government intrusions into legitimate expectations of privacy . . . is the same under both the United States and Vermont Constitutions, [but] our Article 11 jurisprudence has diverged from the United States Supreme Court’s analysis of the Fourth Amendment.”).

[13]     The first sentence of § 1931 stated, at the time: “It is the policy of this state to assist federal, state and local criminal justice and law enforcement agencies in the identification, detection or exclusion of individuals who are subjects of the investigation or prosecution of violent crimes.”  1997, No. 160 (Adj. Sess.), § 1, adding 20 V.S.A. § 1931.  The current language of § 1931 is the same except for the removal of the word “violent.”  20 V.S.A. § 1931.

[14]     E.g., Ala. Code § 36-18-24; Alaska Stat. § 44.41.035; Ariz. Rev. Stat. Ann. § 13-610; Ark. Code Ann. §§ 12-12-1006, 1105; Colo. Rev. Stat. § 16-23-103; Fla. Stat. § 943.325; Kan. Stat. Ann. § 21-2511; La. Rev. Stat. Ann. § 15:609; Mich. Comp. Laws § 750.520m; Mo. Rev. Stat. § 650.055; N.M. Stat. Ann. § 29-3-10; N.C. Gen. Stat. § 15A-266.3A; N.D. Cent. Code § 31-13-03; Ohio Rev. Code Ann. § 2901.07; S.C. Code Ann. § 23-3-620; S.D. Codified Laws §§ 23-5A-5.2, 23-5A-1; Tenn. Code Ann. § 40-35-321; Tex. Gov’t Code Ann. § 411.1471; Utah Code Ann. § 53-10-403; Va. Code Ann. § 19.2-310.2:1.

[15]     These or similar limitations are in the Maryland law, but the Supreme Court did not rely on them under its rationale.

[16]     Curiously, the dissent appears to be most concerned with the evidence presented in the Chittenden Superior Court case that the DNA loci analyzed may contain other personal information.  Post, ¶ 71 n.24.  It concludes that “[s]uch rapid technological developments dictate the need for a more agile search and seizure doctrine . . . where the evidence sought by law enforcement is entangled with vast amounts of private information, as with DNA.”  Id.  Despite that statement, the dissent fails to address the conclusion of that court that the statute is unconstitutional under the balancing test because the privacy invasion is much greater than that recognized in Martin.  Accepting the Chittenden Superior Court’s analysis would, of course, lead to the same result as this majority opinion although on different grounds.

[17]     We recognize that here and in the later discussion we have summarized and simplified the statute to make the explanation easier.  While generally the difference between the amended statute and the original statute involves the period between arraignment and the end of the criminal case, there are exceptions to this simplification.  If the original felony charges are dismissed but defendant is convicted of “domestic assault pursuant to 13 V.S.A. § 1042 or a sex offense for which registration is required pursuant to 13 V.S.A. § 5401 et seq.,” the DNA sample is not expunged.  20 V.S.A. § 1940(a)(3).  Also, even though the felony charges that caused the DNA-sampling requirement are dismissed, the court can order that expungement of the DNA sample not occur where the prosecutor “can show good cause why the sample should not be destroyed.”  Id. § 1940(a)(5). Finally, a pardon by the Governor can result in expungement after the criminal case is concluded.  Id. § 1940(a)(2).  None of these exceptions change the analysis that follows.  

[18]     The dissent argues that Martin actually stands for the proposition that investigating who perpetrated past crimes by comparing a defendant’s DNA with DNA found at crime scenes is a special need beyond normal law enforcement.  It argues that the language of the Martin decision, holding that identifying persons at future crime scenes is a special need but identifying persons at past crime scenes is not a special need, was not intended “to define a literal time frame.”  Post, ¶ 81.  The dissent’s position is a distortion of Martin, turning it into what the dissenting Justices now wish it held, not what it actually held.  We respond clearly and definitively that collecting DNA in order to determine whether the person from whom it is collected committed any of the unsolved past crimes contained in the national DNA database is a normal need of law enforcement and not a special need that will justify involuntary collection of the DNA.  This is the precise holding of Martin, to which we adhere.

[19]     To be clear, we do not view the expungement provisions as a “constitutional defect” in the DNA collection statute, as appellant claims.  Rather, we view them as a constitutionally justified recognition from the Legislature that the privacy interest of someone not convicted of a crime—the average citizen—generally outweighs the state interest in that person’s DNA sample.

[20]     This argument is addressed in detail in C. Preston, Note, Faulty Foundations: How the False Analogy to Routine Fingerprinting Undermines the Argument for Arrestee DNA Sampling, 19 Wm. & Mary Bill Rts. J. 475 (2010).  We have included some but not all of the points raised in this article in the text of this decision.

[21]     The dissent cites to the three hundred persons who have been exonerated by DNA evidence as a demonstration of the need for involuntary DNA collection. Post, ¶ 79 n.26.  Any criminal defendant who believes he or she has been wrongfully identified as the perpetrator of a crime can voluntarily give a DNA sample in the hope it will exonerate him or her. This case is about involuntary DNA collection from an unwilling defendant.  It is not about exoneration, and the prospect of exoneration, if any, does not support the law we are considering.

[22]     Despite the dissent’s protestations, we do not by our observations intend to go back on our statement in Martin that we will not ground our decision on arguments about DNA that invoke speculation about massive incursions on privacy.  2008 VT 53, ¶ 25.  As we stated earlier, we will not speculate on the power of future technology or the policy decisions that will be made in light of that technology.  But we must take note of the scope of the search and seizure in this case.  As Judge Rendell observed in her dissent in Mitchell, 652 F.3d at 424:  “It is akin to saying that if the Government seizes personal medical information about you but can only use the subset of that information that serves to identify you, your privacy interest in the information taken is confined to a mere interest in your identity.  Nothing could be further from the truth.”  This question about the scope of the search has become a major public policy issue with the disclosure that the National Security Agency has collected all the telephone records of Americans so it can quickly get access to specific phone records with court approval during a specific investigation.

[23]     See, e.g., United States v. Mitchell, 652 F.3d 387, 407 (3d Cir. 2011) (“The second ‘search’ at issue is, of course, the processing of the DNA sample and creation of the DNA profile for CODIS.  This search also has the potential to infringe upon privacy interests.”); Weikert, 504 F.3d at 12 (“Importantly, [the defendant’s] privacy is implicated not only by the blood draw, but also by the creation of his DNA profile and the entry of the profile into CODIS.”); Goord, 430 F.3d at 670 (holding that DNA analysis “is potentially a far greater intrusion than the initial extraction of DNA, since the state analyzes DNA for information and maintains DNA records indefinitely”).

[24]      In Martin, we noted that any information that was captured in a swab sample along with the thirteen identifying CODIS loci was “junk” DNA that was “not associated with any known physical trait” and had “no known function, except to accurately and uniquely establish identity.”  2008 VT 53, ¶ 26 (quotation omitted).  We recognized the possibility that the CODIS loci might eventually reveal more private information, but stressed that such analysis would be precluded under the Vermont statute.  Id.

Martin’s caution in this regard is instructive.  Accord Weikert, 504 F.3d at 16 (“The ongoing evolution in our understanding of DNA warrants particular caution in determining what is constitutionally permissible.”).  In the few short years since Martin issued, scientific understanding of noncoding DNA has dramatically advanced, as the record below reflects.  It now appears, with emerging science and new DNA analysis techniques, that these loci may contain other personal information.  Such rapid technological developments dictate the need for a more agile search and seizure doctrine—one that does not rely on quickly-outdated assumptions about the reach of new technologies.  This is particularly true where the evidence sought by law enforcement is entangled with vast amounts of private information, as with DNA.  I urge legislative participation, for “[i]n circumstances involving dramatic technological change, the best solution to privacy concerns may be legislative.  A legislative body is well situated to gauge changing public attitudes, to draw detailed lines, and to balance privacy and public policy in a comprehensive way.”  United States v. Jones, 132 S. Ct. 945, 964 (2012) (Alito, J., concurring) (citation omitted). 

[25]     Vermont’s statute provides even greater protection to the accused than the statute upheld in King, as the probable cause finding is not merely determined by a law enforcement officer but must be confirmed by an independent magistrate.  20 V.S.A. § 1933(a)(2).

[26]      Perhaps the most jarring aspect of the majority’s special needs analysis is its apparent ease with requiring the State to rely on indisputably less-accurate identification techniques.  As to the majority’s assertion that “[t]he current system of photographs and fingerprints fully responds to the need for identification of the defendant,” ante, ¶ 36, other courts would beg to disagree.  See, e.g., King, 133 S. Ct. at 1976 (“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.”); Banks v. United States, 490 F.3d 1178, 1190 (10th Cir. 2007) (“While fingerprint evidence might often be sufficient, we have always recognized the Government’s compelling need to accurately identify offenders.”). 

Even more powerful than its capacity to identify, DNA profiling has an “unparalleled ability both to exonerate the wrongly convicted and to identify the guilty.  It has the potential to significantly improve both the criminal justice system and police investigative practices.”  King, 133 S. Ct. at 1966 (quotation omitted).  Even if individuals can voluntarily submit their DNA to attempt to exonerate themselves, use of the DNA data bank “promptly clears thousands of potential suspects—thereby preventing them from ever being put in that position, and advancing the overwhelming public interest in prosecuting crimes accurately.”  United States v. Kincade, 379 F.3d 813, 839 n.38 (9th Cir. 2004) (quotation omitted).  At a time when the inadequacies of fingerprinting and eyewitness testimony have been starkly revealed, the majority’s stubborn objection to DNA identification of arraignees on serious charges is difficult to justify.  See Innocence Project, DNA Exoneree Case Profiles, available at http://www.innocenceproject.org/know/ (last visited July 3, 2014) (explaining that over three hundred people in the United States have been exonerated using DNA evidence and that in almost half of these cases the actual perpetrator was identified using DNA).  Like other courts that have examined these issues, I do not interpret our Constitution’s privacy protections as requiring the State to eschew the most advanced and accurate identification techniques. 

[27]      Martin held that there is a reasonable expectation of privacy in the information contained in one’s DNA, and that—apart from the physical intrusion required for extraction—any analysis of DNA that results in, or is intended to result in, the collection of new information constitutes a search triggering constitutional analysis and protection.  2008 VT 53, ¶ 14.  The majority’s failure to acknowledge that this holding strictly limits the scope of personal intrusion is regrettable, as Martin may also inform other jurisprudential developments not before us here now, particularly the third-party doctrine. 

While the majority asserts its position is not inconsistent with our decision in Martin, its “if it opens the door, it must be wrong” approach moves us away from a position of embracing new technology with the potential to benefit institutions like the criminal justice system while preserving individual rights.  The majority’s broad brush analysis is therefore a step backward in this regard.