¶ 36. dissenting. A generation ago, the late Justice William 0. Douglas presciently stated: “[T]he privacy and dignity *47of our citizens is being whittled away by sometimes imperceptible steps. Taken individually, each step may be of little consequence. But when viewed as a whole, there begins to emerge a society quite unlike any we have seen — a society in which government may intrude into the secret regions of man’s life at will.” Osborn v. United States, 385 U.S. 323, 343 (1966) (Douglas, J., dissenting and concurring). This case, involving as it does the need to maintain constitutional values of privacy while allowing law enforcement to take full advantage of new technologies for solving crimes, starkly raises the concern addressed by Justice Douglas.
¶ 37. The statute under review authorizes the State to extract and to maintain forever a DNA sample from any person convicted of a felony. All defendants in this case have been duly convicted of such crimes and are, therefore, subject to the statute. They have refused to give a sample of their DNA, asserting that the statute violates their privacy rights under Article 11 of the Vermont Constitution. As convicted felons, defendants may evoke little sympathy when they seek to challenge the pains and penalties that stem from their guilt. And it is well-established that conviction for a felony necessarily results in certain diminutions in one’s privacy rights. These considerations, however, must not deter us from our duty to fully uphold the constitutional rights of all of the people, for, as Justice Douglas so insightfully observed, if we fall into the habit of tolerating small assaults on our privacy, on the ground that in and of itself each seems inconsequential to society at large, we will eventually reach a condition in which we have little or no privacy at all.
¶ 38. We are asked to decide whether the taking, the perpetual maintenance and the future analysis of DNA samples, involuntarily taken from individuals for the sole reason of their conviction of certain crimes, violates Article 11 of the Vermont Constitution. Conceptually, this is a matter of the law of search and seizure. In my judgment, the majority has given us the wrong answer to the question presented, and its analysis is wholly inadequate for the important task before the Court. Because I believe that our jurisprudence plainly requires a different result, I respectfully dissent.
¶ 39. Before turning to the precise legal question involved, it is important to bear in mind the context in which the DNA database statute was passed. The availability of DNA evidence has enhanced forensic investigations of certain kinds of violent crimes, *48and most notably, has served to exonerate people wrongfully convicted through processes that appellate courts have nevertheless adjudged as fair. See P. Monteleoni, DNA Databases, Universality, and the Fourth Amendment, 82 N.Y.U. L. Rev. 247, 253 (2007). Although its significance as evidence in other crimes is doubtful,12 DNA evidence has come to be seen as a panacea for solving crimes, especially past crimes where leads have grown cold. Hence, federal and state statutes have been passed permitting the collection of DNA from convicted felons and, in some cases, even mere arrestees, without requiring any showing of the individualized suspicion ordinarily necessary before the government may constitutionally undertake a search. See T. Simoncelli & S. Krimsky, A New Era of DNA Collections: At What Cost to Civil Liberties'!, at 4-6 (2007) available at http://www.acslaw.org/ node/5338.
¶ 40. The history of DNA database statutes reveals that, while databases were first limited, as they were in Vermont, to violent felons, many databases have been expanded to include all felons, juveniles, some misdemeanants, and even arrestees. Rothstein & Talbott, supra, at 153 (“[T]he success of [DNA] databases in solving violent crimes provided the impetus for Congress and state legislatures to expand the scope of the databases with little critical examination of each expansion’s value to law enforcement or cost to privacy and civil liberties.”). Indeed, the statute before us in this case expanded DNA collection from violent felons to all felons, and a bill has recently been introduced in the Vermont Legislature to extend the reach of the database to arrestees. See H.181, 2007-2008 Gen. Assem., Bien. Sess. (Vt. 2007) (expanding DNA database to certain classes of arrestees). If that bill is enacted, anyone arrested on felony charges, including those persons who may be entirely innocent or whose identity has been mistaken, would be subject to the invasion of privacy now sanctioned against all convicted felons.13 Whether such an expanded *49database would be constitutional under Article 11 of the Vermont Constitution depends on the same analysis that the majority misapplies today to the statute requiring all felons to provide DNA samples. That is precisely why this case is so important to us all.
¶ 41. The State, and every court to consider the issue, concedes that the taking of a DNA sample involves an initial search, that the subsequent analysis of the individual’s DNA material is another search, and that both of these searches are subject to constitutional requirements. To pass constitutional muster, a search must either be supported by probable cause and a warrant or be subject to one of the limited exceptions to the warrant rule. State v. Bauder, 2007 VT 16, ¶ 14, 181 Vt. 392, 924 A.2d 38; State v. Savva, 159 Vt. 75, 86, 616 A.2d 774, 780 (1991). As we said in Savva,
as a matter of constitutional policy, a warrant requirement is not a starting point for deriving exceptions that balance citizens’ interest in privacy against law enforcement’s interest in expeditious searches. Rather, it is the balance reached by the constitutional drafters ....
159 Vt. at 85-86, 616 A.2d at 780. Because the warrant and probable cause requirements set a standard of reasonableness, we have departed from these basic principles only when exigent circumstances were present or when special needs beyond the ordinary needs of law enforcement demanded an exception. See State v. Berard, 154 Vt. 306, 310-11, 576 A.2d 118, 120-21 (1990) (“[Tjhis Court . . . abandon[s] 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.” (quotation omitted)); see also State v. Welch, 160 Vt. 70, 78-79, 624 A.2d 1105, 1110 (1992). In other words, if the government is engaged in the ordinary business of law enforcement, such as crime detection, it must have a warrant to infringe on individual privacy rights.
¶ 42. Thus, the constitutional problem presented by the DNA database statute is how to fit the statute within an exception to *50the warrant requirement. In my view, it cannot be done.14 Some courts have done it by finding the government’s interest in obtaining the DNA of persons within the corrections system to be a special need beyond the ordinary detection of crime. See, e.g., Nicholas v. Goord, 430 F.3d 652, 668-69 (2d Cir. 2005); Green v. Berge, 354 F.3d 675, 677-79 (7th Cir. 2004); United States v. Kimler, 335 F.3d 1132, 1146 (10th Cir. 2003). Many courts, including the overwhelming majority of federal circuit courts, have been unable to overlook the “centrality of law enforcement objectives” in these statutes and have analyzed the question using a totality of circumstances test to justify departure from the warrant rule. United States v. Weikert, 504 F.3d 1, 8-10 (1st Cir. 2007) (recognizing that majority of federal circuit courts apply a “totality of the circumstances” test to DNA database statutes, including the Third, Fourth, Fifth, Eighth, Ninth, Eleventh and D.C. Circuits). Under that test, and after the United States Supreme Court decision in Samson v. California, the privacy rights of prisoners, probationers and conditional releasees are reduced to nil such that there is nothing left to balance against any articulated government interest.15 547 U.S. 843 (2006). Under our own jurisprudence, however, there must first be a special need *51beyond general crime control to allow courts to engage in balancing individual privacy interests against the government’s interest, and second, there must be a nexus between the government’s interest and the restriction on privacy rights. See State v. Lockwood, 160 Vt. 547, 556, 632 A.2d 655, 661 (1993) (probation conditions allowing for warrantless searches must be “supported by the findings and [be] narrowly tailored to fit the circumstances of the individual probationer” with respect to the probationer’s rehabilitation goals and public safety concerns (quotation omitted)). The DNA statute, as applied to the nonviolent felons in this case, does not meet either prong of our test.
¶ 43. I begin with the statute and its purpose, as articulated by the Legislature in its preamble to the original statute:
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. Identification, detection and exclusion may be facilitated by the DNA analysis of biological evidence left by the perpetrator of a violent crime and recovered from the crime scene. The DNA analysis of biological evidence can also be used to identify missing persons.
20 V.S.A. § 1931 (effective April 29, 1998). The original purpose remains unchanged, but the law now requires that every person convicted of a designated crime, including nonviolent felonies, whether in the custody of the Department of Corrections, on probation, parole or serving a supervised community sentence, submit a DNA sample. Id. § 1933 (Cum. Supp. 2007). Designated crimes include felonies, an attempt to commit a felony, or any other offense resolved by plea agreement, if as part of the plea agreement, the original charge was a designated crime and probable cause was found by the court. Id. § 1932(12). Thus, under *52the statute it is the felon’s status that compels the sample rather than any individualized suspicion that a particular felon has committed a crime in the past or will commit a crime in the future.
¶ 44. The primary purpose of the statute is straightforward — to identify suspects who may have committed crimes or may at some indefinite future point commit crimes. Indeed, this purpose was obvious in the legislative history, during which the Legislature heard from the FBI, local law enforcement, and parents of a violent crime victim, among others. Vermont’s statute is based on the federal DNA Act, which authorizes extraction of DNA samples from individuals convicted of a qualifying federal offense for entry into the FBI’s Combined DNA Index System (CODIS) — “a massive, centrally managed database including DNA profiles from federal, state, and territorial DNA collection programs, as well as profiles drawn from crime-scene evidence.” Weikert, 504 F.3d at 3-4. The mission statement of CODIS, which is incorporated by reference in the Vermont DNA database statute, explicitly states that it is an “effective tool for solving violent crimes” and that it was formed in 1990 “for law enforcement purposes.” Id. at 10 (quotation omitted). The Vermont bill was promoted by its advocates as an important national law enforcement tool, and the Legislature’s statement of purpose reflects that goal.16 Although the State has articulated other legitimate uses of the database under the statute, it cannot reasonably be argued that the primary purpose of collection is other than crime detection. And while the legislative purpose is an obviously valid one, the means to the end must nonetheless pass constitutional muster.
¶45. Nothing in our Article 11 jurisprudence overcomes the basic fact that the State cannot articulate a special need to collect DNA samples from all felons. Previously, we have found special needs in two contexts that are relevant here — invasions on the *53privacy rights of prison inmates and of probationers.17 Our controlling precedent is State v. Berard, 154 Vt. 306, 576 A.2d 118 (1990), in which we recognized for the first time that a special need going beyond ordinary law enforcement could support a different constitutional analysis for determining the reasonable*54ness of a search undertaken without probable cause or a warrant. See O’Connor v. Ortega, 480 U.S. 709, 741 (1987) (Blackmun, J., dissenting) (quoting New Jersey v. T.L.O., 469 U.S. 325, 351 (1985)). In Berard, a prison inmate was charged with possession of drugs after prison officials conducted a routine, random and suspicionless search of his cell without a warrant. In upholding the search, we adopted a two-part test, in which a balancing test was substituted for the warrant and probable-cause requirement.
¶46. To trigger the balancing test, the State first had to demonstrate that its ability to maintain institutional security presented a special need that rendered the warrant and probable-cause requirements impracticable. Berard, 154 Vt. at 311-12, 576 A.2d at 121-22. We recognized that the prison environment was unique, and that the State’s objective in operating an institution free of contraband of all kinds, including dangerous weapons, was likely to be substantially hindered by the inability to conduct random searches.18 Id. at 312-13, 576 A.2d at 121-22.
¶ 47. Once we determined that a significant special need allowed a balancing test to be substituted for a warrant and probable cause, we proceeded to weigh the State’s “ ‘paramount interest in institutional security’ against the inmates’ residuum of privacy rights.” Id. at 313, 576 A.2d at 122. We rejected the State’s position that we analyze the constitutional question as if prison inmates had no reasonable expectation of privacy, such that any official conduct was reasonable. Id. at 310-12, 576 A.2d at 120-21; cf. Hudson v. Palmer, 468 U.S. 517, 526 (1984) (holding that under Fourth Amendment, prisoners have no reasonable expectation of privacy in their cells). Although privacy rights of prisoners are diminished, we noted they are not nonexistent. Berard, 154 Vt. at 313, 314 n.4, 576 A.2d at 122, 123 n.4 (“though his rights may be diminished by the needs and exigencies of the institutional environment, a prisoner is not wholly stripped of constitutional protections” (quotation omitted)). Nonetheless, the privacy interest of the defendant in Berard was overcome by the State’s interest because the searches were undertaken as part of a random policy that served broad objectives, and did not arbitrarily or specifically target him.
*55¶ 48. We applied a similar line of reasoning in State v. Lockwood, 160 Vt. 547, 632 A.2d 655 (1993), in considering a blanket probation condition that permitted warrantless searches of a probationer’s home and person. We found the probation condition overbroad in that it did not recognize the probationer’s right of privacy, one that was greater than that of the prison inmate in Berard, but upheld the search because the officers had “reasonable grounds” to support it prior to the search, and the probation condition restricting the probationer’s constitutional rights was individually and narrowly tailored to meet the goals of his rehabilitation and the protection of the public. Lockwood, 160 Vt. at 558, 632 A.2d at 662. We stated that “if a probation term provides for warrantless searches and the terms of probation are narrowly tailored to fit the circumstances of the individual probationer, the . . . ‘reasonable grounds’ standard strikes the proper balance between probationer privacy rights and public protection concerns.” Id. at 559, 632 A.2d at 663. Therefore, outside the prison walls, the government’s burden to overcome even the diminished privacy rights of a probationer demands a nexus between the special need and the warrantless intrusion, and individualized suspicion.
¶ 49. The majority relies on the special-needs doctrine, arguing that the collection of DNA for inclusion in a law enforcement database is not an ordinary law enforcement purpose, and that the protections of Article 11 may therefore be overlooked as impracticable. Although our law is considerably different from federal law and that of other states, the majority reaches this conclusion by accepting the analysis of the Supreme Court of New Jersey in State v. O’Hagen, 914 A.2d 267 (N.J. 2007). In O’Hagen, the court concluded that the primary purpose of the DNA tests is to create a DNA database to assist in the future resolution of crimes, and because it is a long-range special need that does not have the “immediate” objective of gathering evidence against the offender, the statute does not impermissibly target individuals without reasonable suspicion. 914 A.2d at 278. The majority also accepts the special-needs analysis of the Second Circuit Court of Appeals in Nicholas v. Goord, interpreting the New York statute as presenting a more “nuanced approach to law-enforcement concerns.” 430 F.3d at 668. The court in Goord concluded that DNA samples are simply an information-gathering tool that may eventually help law enforcement identify the perpetrator of a crime, *56but because they do not do so immediately at the time of collection, the collection is not sought for the investigation of a specific crime. Id. at 667-68.
¶ 50. First, the reasoning on which the majority opinion relies completely overlooks that DNA collection from persons recently convicted may identify — immediately, or as soon as the analysis and database check is completed — a suspect in an old, unsolved crime, and that such person will be prosecuted as a result of the identification. This is ordinary detection of crime for which the majority mounts no special-needs justification.
¶ 51. Second, even if the DNA collected is analyzed and awaits a possible match with a future crime, the time distinction increases the constitutional problem, rather than decreases it. If no current crime needs solving, the government’s need to use extraordinary procedures that abandon the warrant and probable-cause requirement is difficult to justify. The State’s interest is obviously weaker if it is gathering information for crimes that have not yet been committed. Cf. Illinois v. Lidster, 540 U.S. 419 (2004) (upholding roadblock on grounds that stop’s objective was to help find the perpetrator of a specific and known crime, not of unknown crimes of a general sort). Contrary to the majority’s assertion, cataloguing DNA for future prosecutions, for crimes not yet committed, is precisely the kind of evil sought to be remedied by Article 11 — the intrusion on the privacy of individuals for whom no reasonable suspicion of wrongdoing can be articulated.19 Collecting DNA for crimes that may never be committed, to further some sort of general deterrence goal, undeniably serves a general law enforcement purpose.20
¶52. Third, the fact that DNA statutes merely allow the collection of data for identity purposes begs the special-needs question. See Goord, 430 F.3d at 669 (defining governmental interest in DNA database as “obtaining identifying information from convicted offenders and keeping a record of such informa*57tion”). The question is whether the identifying information is sought for ordinary law enforcement purposes.21 Our constitution makes no distinction between evidence gathered for identification of suspects and other kinds of evidence. If the police decided to conduct a dragnet by demanding the DNA of every white male in a certain neighborhood in an attempt to “merely identify” the perpetrator of a violent crime, that action could not be carried out constitutionally as a special need.
¶ 53. In short, the DNA statute allows the State to collect DNA from individuals for the express purpose of identifying them as either the perpetrator of a past crime, or the perpetrator of a future crime, so that they may be prosecuted. As many courts and commentators have now recognized, calling the collection of DNA under the database statutes a special need of law enforcement is like trying to fit a square peg into a round hole. See, e.g., Weikert, 504 F.3d at 9-10; J. Rikelman, Justifying Forcible DNA Testing Schemes Under the Special Needs Exception to the Fourth Amendment: A Dangerous Precedent, 59 Baylor L. Rev. 41, 65-68 (2007); Monteleoni, supra, at 262-67.
¶ 54. The majority’s analysis plainly fails to meet the first requirement of our own precedent in Berard — to demonstrate that a credible special need exists to abandon the warrant and *58probable-cause requirement. Its analysis is inadequate to the important task before the Court.
¶ 55. Even if I were to accept that the collection of DNA is a special need, however, the majority has not shown that there is a nexus between the State’s special need and the warrantless intrusion on appellants’ rights, or that the intrusion is narrowly tailored to accomplish those ends. See Lockwood, 160 Vt. at 556, 632 A.2d at 661. Thus, its conclusion that the State’s interest in the DNA database statute outweighs the important privacy interests that nonviolent felons maintain in their DNA is entirely unsupported.
¶ 56. The State has the burden to demonstrate a nexus between the collection and profiling of DNA samples from all convicted, nonviolent felons and its interest in identifying perpetrators of future crimes, deterring criminal activity, and preventing recidivism. As we have held, the State’s interest in both rehabilitating those in the corrections system and protecting the public from convicted criminals is reasonably furthered by activities such as random prison-cell searches and searches of probationers’ homes based on reasonable suspicion, and thus, there is a sufficient nexus to depart from the warrant and probable-cause requirements. Here, there is no similar connection between the State’s interest in penalizing, controlling, or rehabilitating convicted felons and collecting, profiling, and repeatedly searching their unique genetic material, nor does the State even attempt to argue that there is.
¶ 57. The State’s general interest in preventing crime is no more furthered by collecting the DNA of all convicted felons than it would be by sampling the DNA of all Vermont citizens. As the District Court, Judge Levitt presiding, saliently concluded in its ruling on the constitutionality of the DNA database statute:
[Tjhere is a question of “fit.” Identity is never at issue in the type of crimes in which the above-named defendants were convicted, that is false pretenses, drug possession, or driving while intoxicated. It is crimes against persons where DNA identification and exclusion is most relevant, not the non-contact crimes involving possession, intoxication or larceny.
To reach the majority’s conclusion, one therefore has to accept the proposition that, although identity is generally not at issue in *59nonviolent felonies, the fact that the State holds the DNA of a nonviolent felon will deter that person from committing crimes in the future. Not even the majority finds that a viable proposition. See ante, ¶ 20.
¶ 58. Given the weak nexus between the State’s means and its ends, the majority has not shown that the State’s interest in solving or deterring future crimes outweighs defendants’ privacy interests in their DNA. See Berard, 154 Vt. at 311, 576 A.2d at 121; Lockwood, 160 Vt. at 556-59, 632 A.2d at 661-63. Even the warrantless search in Lockwood, which intruded only on possessory interests, was supported by “reasonable grounds” in light of the residuum of privacy accorded to a probationer. 160 Vt. at 558-59, 632 A.2d at 662-63. Unlike Lockwood, the majority here gives short shrift to the significant privacy interests involved.
¶ 59. Each of us has a reasonable expectation of privacy in our unique genetic make-up. If we did not, there would be no constitutional question before us. While a convicted felon’s privacy interest in his home, or possessory interest in his belongings, may be temporarily diminished given the constraints and goals of the corrections system, he cannot be said to have a similarly lessened expectation of privacy in his unique genetic profile. DNA analysis has the potential to reveal a panoply of private genetic information, including physical and medical characteristics, genealogy, and predisposition to disease — none of which bears any relation to the penal, rehabilitative or administrative goals of the corrections system.
¶ 60. To be sure, the privacy rights of convicted felons within the corrections system, whether prisoners, probationers or parolees, are necessarily diminished.22 As the majority points out, those currently within the corrections system cannot be said to have a reasonable expectation of privacy in their physical identity. Indeed, convicted criminals make their identity a public concern by *60committing a crime, and the government must have identifying information to administer the criminal justice system. This lack of privacy in the mere fact of one’s identity, however, does not reasonably extend to the extraction, analysis, indefinite storage and repeated law enforcement searches of convicted felons’ DNA without seriously undercutting the constitutional right of citizens to be free in their persons.
¶ 61. The majority attempts, as have other courts, to liken DNA profiling to fingerprinting, and thereby deemphasize the important privacy interests implicated by the DNA database statute. There are, however, critical differences between the two that we must bear in mind in our analysis of the database statute. To begin, fingerprints are regularly exposed to the outside world and cannot be manipulated to reveal private information beyond identity. Within the corrections system, fingerprinting is undertaken as part of the routine booking procedure to establish the identity of the person taken into custody, ascertain the person’s criminal history and outstanding warrants, if any, and aid in the apprehension of escaped prisoners. United States v. Olivares-Rangel, 458 F.3d 1104, 1113 (10th Cir. 2006).
¶ 62. DNA samples serve a different purpose. They must be analyzed to create a unique profile useful to law enforcement in identifying criminal suspects, and in the process of that analysis, the intimate details of one’s genetic make-up are revealed, exposing felons to the type of governmental abuse of privacy that Article 11 is intended to thwart. The majority claims that the statute simply authorizes “the creation, storage, and searching of a unique alphanumeric identifier based on analysis of thirteen locations on DNA that are not associated with any known physical trait.” Ante, ¶ 26. As scientific advancements in the field of genetics continue, however, the likelihood that increasingly private information can be gleaned from the seemingly innocuous DNA profiles expands. In fact, while the thirteen loci at which DNA samples are measured were once widely believed to be “junk DNA,” scientists have in recent years discovered that DNA profiling “can reveal probabilistic information about one’s ethnicity and gender.” Monteleoni, supra, at 256. Furthermore, there is strong scientific evidence to suggest that the profiles may carry information about individuals’ genetic predisposition to certain *61diseases.23 See id.; D. Concar, What’s in a Fingerprints, New Scientist, May 5, 2001, at 9. Thus, DNA profiling involves a significantly greater intrusion of privacy than fingerprinting and other methods of criminal detection and identification.
¶ 63. Even if I agreed that DNA profiling, like fingerprinting, provides no more information than a unique identifier, the critical difference is in the way the identifying information is used. No one here is contending that the State collects DNA samples solely for administrative purposes, such as those underlying the routine fingerprinting by corrections. Rather, the explicit purpose of collecting DNA samples from felons — potentially long after their identity has been established by fingerprinting — is to use them indefinitely to investigate past or as-yet-uncommitted future crimes.
¶ 64. Finally, the majority’s reliance on In re R.H. for the proposition that DNA sampling, by saliva swab, involves no intrusion greater than that engendered by fingerprinting is misplaced. 171 Vt. 227, 238, 762 A.2d 1239, 1247 (2000). In that case, the view that a cheek swab was a minimal intrusion of privacy was appropriate in light of the safeguards present — reasonable suspicion of the defendant’s involvement in a crime and judicial review leading to a nontestimonial order (NTO). Furthermore, we recognized in R.H. that protections such as destruction of nonmatching profiles analyzed pursuant to an NTO might be prudent, and requested that the Advisory Committee on the Rules of Criminal Procedure consider adding such safeguards. Id. at 238-39, 762 A.2d at 1247-48. While the Advisory Committee determined that such protections were more appropriate for legislative action and the Legislature has as yet failed to act on the matter, the privacy issues implicated were, in the very least, brought to the forefront. See Reporter’s Notes, 2006 Amendment, V.R.Cr.P. 41.1. Today, the majority fails entirely to consider the implications of indefinitely storing DNA profiles which, unlike profiles analyzed pursuant to an NTO, are extracted without *62individualized suspicion or judicial oversight and based only on the status of the individuals providing the samples.
¶ 65. In comparison to the significant privacy rights at risk in DNA profiling and data-banking, the State’s general interest in adding to its arsenal of investigative tools is weak at best. To the extent that the DNA database is intended to solve future crimes that have not yet been committed, the State has little interest in invading the privacy rights of nonviolent felons. To the extent that it is used to solve past crimes, there is already a constitutional mechanism in place for obtaining DNA samples from suspects on the basis of some individualized suspicion and judicial review. See R.H., 171 Vt. at 238-39, 762 A.2d at 1247-48. Nor is the statute narrowly tailored to meet its goals, as any future crimes likely to be committed by recidivist nonviolent offenders are unlikely to be solved by DNA evidence.
¶ 66. In any event, there are no exigencies involved in the use of DNA profiling to investigate past or future crimes. In the case of searches of prison cells or probationers’ homes, contraband or other evidence of criminality could be concealed or destroyed in the time it takes to obtain a warrant. DNA, on the other hand, can neither be destroyed nor concealed. The privacy interests of convicted felons in their unique genetic code far outweigh the governmental interest in collecting, storing and searching the DNA of an entire class of individuals for the investigation of as-yet-uncommitted crimes. Thus, the State’s inability to demonstrate that its interests in creating and maintaining the investigative tool should overcome the important privacy interests at stake, and its wholesale failure to prove a sufficient nexus between its goals and its methods — sampling of all nonviolent felons — further militate against abandoning Article 11 protections.
¶ 67. It is no answer to the invasion of privacy permitted by the majority’s opinion that the statute provides both criminal and civil remedies for wrongful disclosure of private information obtained pursuant to the DNA database statute. See 20 Y.S.A. § 1941. The delicate nature of the genetic information at stake here and its availability not only to state but federal law enforcement officials dictates against after-the-fact remedies and in favor of pre-search constitutional safeguards. Unfortunately, our nation’s history provides stringent warnings against unabashedly entrusting government with sensitive information about our citizenry. Whether intentionally or accidentally, both our state and federal govern*63ments have in the past failed to protect individuals’ privacy rights when tasked with maintaining large stores of personal information. Only recently, state officials announced a major breach of the Vermont Offices of Child Support computer system, providing hackers with access to the names, social security numbers and bank account information of nearly 70,000 state residents. D. Gram, State Was Warned of Potential Computer Security Breach, Assoc. Press, January 30, 2007. In response, a computer security expert noted the state’s carelessness in storing the data on a computer with such limited security protection. Id. Regardless of the state’s diligence in maintaining the state DNA database, the fact remains that the statute at issue allows for the dissemination of profiles for storage and use in the federal CODIS. The federal government’s track record where protection of sensitive information is concerned is even less reassuring than the state’s. Not only have its security measures been compromised on numerous occasions, there is ample evidence of its affirmative misuse of private data for political purposes. For example, as recently as March 2006, the Justice Department reported that the FBI had found hundreds of violations of its own wiretapping and other intelligence-gathering procedures. E. Lichtblau, Justice Dept. Report Cites Intelligence-Rule Violations by F.B.I., N.Y. Times, March 9, 2006, at A21. More disturbingly, only a year later, the Justice Department’s inspector general issued a scathing report criticizing the FBI’s unauthorized use of administrative subpoenas to obtain thousands of telephone, business and financial records of American citizens under the guise of national security. D. Johnston & E. Lipton, U.S. Report to Fault Wide Use of Special Subpoenas by F.B.I., N.Y. Times, March 9, 2007, at Al. Given this history, I am not persuaded that we should rely on statutory penalties as a meaningful substitute for constitutional safeguards.
¶ 68. The majority scoffs at the notion that DNA database statutes presage futuristic societies like that portrayed in the movie Gattaca (Columbia Pictures Corp. 1997), in which the government maintains a DNA database of all citizens and classifies them as “valid” or “invalid,” takes cheek swabs at roadblocks for instantaneous identity verification, and privileges those citizens whose DNA has the qualities it considers desirable. But I would not take this threat to our liberties so cavalierly. Compulsory DNA testing has steadily expanded in scope. Proposals for further expansions are under active consideration, including expansions *64that target entirely innocent citizens. More ominously, there is nothing in the logic of the majority opinion that will prevent the state from requiring DNA testing of every single one of us. After today, one may ask, what is now the constitutional objection to testing every child at birth in an effort to aid in the identification or recovery of missing children? Every applicant for a state or federal job, or for a job with a government contractor? Every person arrested, whether guilty of some infraction or entirely innocent of wrongdoing? As the majority now defines special need, one can easily think of legitimate reasons for testing virtually everyone; and what we have heretofore required as a valid connection between the special need and the privacy invasion involved in a warrantless search may now apparently be satisfied by the flimsiest of justifications, if any is needed at all.
¶ 69. It is too easy to dismiss a challenge to privacy presented by convicted felons as one that deserves only cursory consideration. The narrow context of the inquiry tends to mask the larger threat to liberty. But even if we are justified in holding a lessened concern for these individuals, they threaten soon to become everyman. We will then have only ourselves to blame for falling within the injunction voiced by one of the most revered of the nation’s founders: “They that can give up essential liberty to obtain a little safety deserve neither liberty nor safety.” Benjamin Franklin, Historical Review of Pennsylvania (1759).
¶ 70. I am authorized to state that Judge Devine joins this dissent.
As commentators have noted, there has been virtually no scientific analysis of the overall effectiveness of DNA databases in solving or preventing crimes or comparison of the databases’ effectiveness to that of other forensic techniques. M. Rothstein & M. Talbott, The Expanding Use of DNA in Law Enforcement: What Role for Privacy?, 34 J.L. Med. & Ethics 153, 154-55 (2006).
At the federal level, the collection and retention of DNA from arrestees and non-U.S. persons detained under federal authority is already sanctioned. 42 U.S.C. § 14135a(a)(l)(A).
While I recognize that every appellate court to have considered the constitutionality of DNA database statutes thus far has upheld them, the issue is not without controversy in the courts, and the constitutionality of DNA databases is not a foregone conclusion — particularly under our unique Article 11 jurisprudence — as implied by the majority. See, e.g., United States v. Stewart, 468 F. Supp. 2d 261 (D. Mass. 2007) (holding that federal DNA Act is unconstitutional as applied to probationer); United States v. Sczubelek, 402 F.3d 175, 189 (3d Cir. 2005) (McKee, J., dissenting); United States v. Kincade, 379 F.3d 813, 842 (9th Cir. 2004) (Reinhardt, J., dissenting).
The United States Supreme Court’s recent Fourth Amendment decision allowing warrantless, suspicionless searches of parolees based solely on a general balancing of interests is inapposite to our own analysis of the DNA database statute’s constitutionality under Article 11. In Samson, the Court relied on the defendant’s status as a person within the corrections system as justification for its departure from the traditional special needs analysis. 547 U.S. at 848-49. As the Samson dissent duly recognized, this is circular reasoning: a defendant’s status in the corrections system makes him subject to warrantless searches, which undermines his reasonable expectation of privacy, which allows the Court to abandon constitutional protections in favor of a general balancing of interests, which necessarily results in any government interest outweighing the parolees’ effectively nonexistent privacy interests. Id. at 857-58. Under this regime, the constitutional protections against unreasonable search and seizure cease to apply to citizens within the corrections system, whether prisoners, probationers or parolees.
*51The Supreme Court’s abandonment of the special needs doctrine stands in sharp contrast to our own jurisprudence. We have never held that an individual’s status alone can so undermine his right to constitutional protections that any articulated governmental interest may justify a warrantless, suspicionless search of that individual. As such, I believe the “totality of the circumstances” test has no place in our Article 11 jurisprudence and reiterate that only a special need, beyond normal law enforcement needs, can justify warrantless, suspicionless searches such as those sanctioned by the DNA database statute at issue.
In fact, introduction of the DNA database statute in the Vermont Legislature was largely motivated by the efforts of the federal government to integrate state DNA databases with CODIS. The state forensic lab director at the time the bill was in committee testified that the state stood to gain $172,000 in federal funding to improve its operations, conditioned on Vermont passing its own DNA database law. State DNA Databank: Hearing on H.89 Before House Judiciary Comm., 1997-1998 Bien. Sess. (Vt. Jan. 22, 1997).
There are other decisions in which we have overlooked the warrant and probable cause requirements of Article 11 that are inapposite to our analysis today. In State v. Richardson, we upheld a warrantless search in which a police officer removed a gun in plain view from the defendant’s car. 158 Vt. 635, 603 A.2d 378 (1992) (mem.). The immediate threat of the exposed gun if left unattended was a special need justifying a departure from Article 11 protections. Id. at 636, 603 A.2d 379. Thus, while the Richardson decision is instructive on the “special need” prong of the Article 11 exception, it differs from the instant ease because the search was based on individualized suspicion of criminal wrongdoing. Likewise, we departed from the warrant and probable-cause analysis in State v. Welch, where we upheld a warrantless inspection of the defendant’s pharmaceutical prescriptions under the “pervasively regulated industry” exception. 160 Vt. 70, 80-81, 624 A.2d 1105, 1111 (1992). Although I advocated for a special needs analysis in my dissenting opinion, the majority reached its holding based on the regulatory search exception to the warrant rule. See id. at 90-91, 624 A.2d at 1116 (Johnson, J., dissenting). In addition, the search in Welch was conducted on the basis of individualized suspicion. Thus, the Welch analysis is inapplicable here.
Prior to Berard, in State v. Record, we upheld DUI roadblocks conducted under strict guidelines, finding that the government’s interest in immediately removing intoxicated drivers from the roadway outweighed the minimal intrusion presented by the roadblock. 150 Vt. 84, 548 A.2d 422 (1988). While Record is relevant to our current inquiry to the extent that it involved suspicionless, warrantless searches, it predated the governing special needs exception and relied on a simple balancing of interests to overcome Article ll’s individualized suspicion and warrant requirements. As the dissent in Record, noted, there was an important analytical oversight in the majority’s analysis of the constitutionality of DUI roadblocks. Id. at 97, 548 A.2d at 430 (Hill, J., dissenting). If Article ll’s protections could be overcome anytime the governmental interest in a search outweighed individual privacy interests, the rule requiring a warrant and probable cause for constitutional reasonableness would be swallowed up by the exception. Thus, the Record dissent posited that the threshold question was “whether a DUI roadblock presents the sort of exceptional circumstances that . . . justify abandonment of the warrant and probable cause requirements,” and concluded that it did. Id. Only then did the dissent go on to conduct a balancing of the respective interests of the government and the individuals subject to the roadblock. Two years later, we decided Berard and adopted the dissent’s reasoning in Record, holding that the State must show “special needs, beyond the normal need for law enforcement” to overcome Article ll’s probable cause and warrant requirements. Berard, 154 Vt. at 311, 576 A.2d at 121. The Court acknowledged that it had not previously announced the “special needs” standard, but maintained that “extraordinary factors” necessitated the search in Record and retrospectively recognized “[t]he special needs of the state in a DUI roadblock.” Id. at 311, 314 n.3, 576 A.2d at 121, 122 n.3.
Even the defendant conceded that random searches were necessary for this purpose, but advocated for a warrant to be issued on something less than probable cause.
To be sure, the evil resulting from unlimited discretion accorded to officers who held general warrants was an additional factor in the adoption of Article 11, but it is not one that overcomes the principal purpose of Article 11: to protect the right of the people to privacy against government interference.
The articulation of “other purposes,” such as the identity of missing persons, or the exoneration of the innocent, are not sufficiently dominant or supported to provide the special needs necessary to meet the State’s burden and serve only as makeweight in the majority opinion.
Although we adopted the special-needs doctrine from federal case law, our Article 11 special-needs jurisprudence has since developed differently from the federal Fourth Amendment test, particularly with respect to prisoners, probationers and others within the corrections system. Nevertheless, recent United States Supreme Court decisions rejecting or upholding the application of special needs, while not controlling, are instructive on the question of what is a special need. The term most certainly does not embrace either society’s general interest in crime control or the targeting of individuals for prosecution. Compare City of Indianapolis v. Edmond, 531 U.S. 32, 41-47 (2000) (invalidating a roadside checkpoint designed to detect illegal drugs), and Ferguson v. City of Charleston, 532 U.S. 67, 77-84, 83 n.20 (2001) (holding unconstitutional a public hospital’s forwarding results of nonconsensual drug testing of maternity patients to police — “[i]n none of our previous special needs cases have we upheld the collection of evidence for criminal law enforcement purposes”), with Lidster, 540 U.S. at 427 (upholding checkpoint, but individuals not targeted for prosecution), Bd. of Educ. v. Earls, 536 U.S. 822, 829 (2002) (upholding random drug testing of students, but no prosecution allowed), and Nat'l Treasury Employees Union n Von Raab, 489 U.S. 656, 663, 677 (1989) (upholding drug testing of urine of customs officials for positions requiring use of firearms or involving interdiction of illegal drugs, but official who tested positive could be dismissed but not prosecuted criminally).
The majority, while acknowledging the diminished privacy interests of those within the corrections system, fails to analyze the spectrum of privacy rights retained by prisoners, probationers, parolees, and others within the system, respectively. By its reasoning, parolees who have already served their time and are released into free society have no greater privacy interests than prisoners confined to a cell. This is in contravention of our jurisprudence. See Lockwood, 160 Vt. at 559, 632 A.2d at 663 (stating that “privacy rights of probationers are arguably greater than those of inmates”); see also Weikert, 504 F.3d at 12 (acknowledging that “different statuses may diminish an individual’s expectation of privacy to different degrees”).
The majority takes us to task for our reliance on articles outside of the record for the proposition that so-called “junk DNA” can reveal private medical information, when the real problem with this case is that there is nothing approaching a record on this issue and yet the majority decides the question as if there were. It is the majority’s reliance on the State’s mere assertion that no weighty privacy interest is involved in profiling “junk DNA” that compels the response that there are contrary views on the issue.