¶ 16. concurring. I concur in the opinion for the Court that the records in issue are not exempt under 1 Y.S.A. § 317(c)(5), the exemption for “records dealing with the detection and investigation of crime,” because they are records “reflecting the initial arrest of a person.” I write separately, however, because I believe that there is a more direct and simple rationale for reaching the same result, and it is more in keeping with the intent behind the Public Records Act (PRA). In this case, there was no crime. For this reason, I conclude that the records concerning the incident do not fall within the basic exemption.
¶ 17. We have issued a number of decisions interpreting the PRA. Whatever the result, these decisions have rarely created bright lines that make it easy for governmental officials to determine what records are publicly accessible and what records are not. For example, the decision in this case calls upon the record-holder to apply an eight-factor test to determine whether detention amounted to an arrest. See ante, ¶ 13. A decision of this complexity will likely have to be made by an agency lawyer, not the record-holder, and is likely to lead to more court appeals. Because the right answer is highly debatable, the natural inclination for the agency will be to deny the request for access and leave it to the court to make the decision. Rather than the prompt production or expeditious review the statute contemplates, see 1 V.S.A. § 318(a), the final decision may take months, or even years, *179to finally appear. In too many instances the overarching policy “to provide for free and open examination of records,” id. § 315, would be wholly undermined. What is sorely needed in this area are simple and predictable rules that can be applied by the custodians of records and, where necessary, trial courts.
¶ 18. This case presents an opportunity to provide a clear rule for future applications of exemption five. The public records request in this case arises entirely ex post. This is not a case involving a request for records that are part of an ongoing criminal investigation, or a completed investigation where criminal proceedings may be ongoing, or even a completed investigation where no charges are pending that might someday be lighted anew by additional evidence. As a subject of investigation, the incident is completely behind us. From this retrospective stance, we can, with the benefit of hindsight, affirmatively state that there was no crime.
¶ 19. In cases like this, when we know that there never was any crime to be investigated or detected, the plain language of exemption five does not provide clear guidance. There are essentially two ways that one can read § 317(c)(5)’s reference to “records dealing with the detection and investigation of crime.” On the one hand, it can be read to include essentially all records of police investigation. As long as there was an investigation, the exemption would apply. On the other hand, it can be read to include only investigatory records of crimes. Where there was no crime, the exemption does not apply. The former interpretation would make the records in this case exempt; the latter would not.4
¶ 20. As between these two ways to read the statutory language, I believe that the rule yielding no exemption is the legally correct *180choice. Where statutory language is open to multiple readings, we rely on context and .consequences to select the appropriate interpretation. See Shea v. Metcalf, 167 Vt. 494, 498, 712 A.2d 887, 889 (1998). In the context of the PRA and in light of the felicitous consequences, I would not interpret § 317(c)(5) to cover the situation where there was no crime as in this case. Four general reasons support this choice.
¶21. First, exemptions from public access are to be read narrowly. See 1 V.S.A. § 315 (stating that the PRA “shall be liberally construed” in order to implement a policy of free and open government); see also Trombley v. Bellows Falls Union High Sch. Dist. No. 27, 160 Vt. 101, 106-07, 624 A.2d 857, 861 (1993). We have consistently understood this directive to mean that, where there is doubt or ambiguity, records should be accessible to the public as part of our democratic right. See, e.g., Wesco, Inc. v. Sorrell, 2004 VT 102, ¶ 10, 177 Vt. 287, 865 A.2d 350 (“We construe [Public Record Act] exceptions strictly against the custodians of records and resolve any doubts in favor of disclosure.”); Caledonian-Record Publ’g Co. v. Walton, 154 Vt. 15, 20, 573 A.2d 296, 299 (1990) (“[T]he public interest clearly favors the right of access to public documents and public records.”). Given that the language of exemption five is ambiguous, we should follow the Legislature’s explicit directive and interpret the exemption narrowly and in favor of disclosure.
¶ 22. Second, the policy rationales for the exemption at issue are essentially absent where hindsight undeniably shows that there was no crime. The exemption for investigatory records serves the critical purpose of ensuring that police can carry on their investigations without interference. See Walton, 154 Vt. at 21, 573 A.2d at 300 (“[T]he state has significant interests in protecting the public from criminal activity, prosecuting those who commit crimes, and protecting the privacy rights of individual citizens. These interests may, at times, override the interest in public disclosure.”). The state’s interests in nondisclosure are at their apex where there is an open criminal investigation or ongoing criminal proceedings that might be impaired. See id. at 23, 573 A.2d at 300-01 (“[Disclosure of this type of information may endanger the state’s position in criminal prosecutions since the *181material may be used to the disadvantage of the prosecution and . . . may reveal the names of informants and threaten to intimidate potential -witnesses.”).
¶ 23. The state’s interests in nondisclosure are substantially weaker when an investigation or prosecution is no longer pending, but they are not absent. See Rutland Herald v. Vt. State Police, 2012 VT 24, ¶ 12, 191 Vt. 357, 49 A.3d 91 (finding no support for “a time-based limitation ... in the plain language of § 317(c)(5)”). The residual reasons against disclosure even after an investigation’s conclusion — articulated by a number of federal appeals courts in the early 1970s — include concerns about chilling citizen cooperation in future cases or revealing investigatory tactics. See Aspin v. Dep’t of Def., 491 F.2d 24, 30 (D.C. Cir. 1973) (“It is clear that if investigatory files were made public subsequent to the termination of enforcement proceedings, the ability of any investigatory body to conduct future investigations would be seriously impaired. Few persons would respond candidly to investigators if they feared that their remarks would become public record after the proceedings. Further, the investigative techniques of the investigating body would be disclosed to the general public.”); Frankel v. SEC, 460 F.2d 813, 817-18 (2d Cir. 1972) (“If an agency’s investigatory files were obtainable without limitation after the investigation was concluded, future law enforcement efforts by the agency could be seriously hindered. The agency’s investigatory techniques and procedures would be revealed. The names of people who volunteered the information that had prompted the investigation initially or who contributed information during the course of the investigation would be disclosed. The possibility of such disclosure would tend severely to limit the agencies’ possibilities for investigation and enforcement of the law since these agencies rely, to a large extent, on voluntary cooperation and on information from informants.”).
¶ 24. Even these residual interests, however, are vanishingly small where, as here, there incontrovertibly was never any crime. Where no criminal activity has occurred, there can be little fear of a chilling effect on citizens reporting crime or on informants giving tips about crime. Indeed, one would hope that citizens would not report crime where there is none.5 The concern that *182disclosure might reveal police techniques is also minimal. Because there was no crime, the police records necessarily cannot provide an example of how the police successfully detect crime. Furthermore, sophisticated surveillance or interrogation tactics are unlikely to be involved where there actually was no criminal activity whatsoever. What may be revealed — general facts about how police respond to 911 calls — are such a far cry from the case-specific details of an ongoing investigation, concern for which undergirds § 317(c)(5), that I doubt that the Legislature intended the exemption be read so broadly.
¶ 25. Third, the absence of any crime is objective, and its consideration therefore avoids inquiring into the subjective intent of the record’s creator. It is tempting to assume that these are records of the investigation of crime based on the fact that the police thought that they were investigating a crime when the incident occurred. But, it is unlikely that the Legislature would intend that the character of a record would turn on the subjective intent of the record’s creator at the time the record was created. Cf. John Doe Agency v. John Doe Corp., 493 U.S. 146, 155 (1989) (rejecting an interpretation of the federal exemption for “records or information compiled for law enforcement purposes” that would depend on the original purpose for which the record is compiled, focusing instead on the status of the documents “when the response to the [public records] request must be made”). In many instances, the records custodian will have no ability to determine the intent of the record-creator. For example, we could not expect a record custodian to determine whether particular parts of the “investigation” described in a record were pretextual and not done for the investigatory purpose asserted. The subjective approach would also mean that the result in this case turns entirely on the fact that the 911 caller misidentified the situation. Had the caller requested the police in their community caretaking capacity and sought a welfare check on the homeowner, then the records of the incident would not fall within exemption five.6 For these reasons, *183the subjective-intent approach is, I believe, unworkable. Where we can affirmatively state that there was no crime, we should, rely on this objective fact rather than on the subjective perception of the incident at the time.
¶ 26. This brings me back to my introductory comments, which I restate here as the fourth reason in favor of this interpretation of the exemption. An interpretation that turns on whether there was actually a crime is simple and easy to apply — no crime, no exemption. This rule resolves this case cleanly, and it provides an uncomplicated way to handle analogous cases in the future. It can be used by records custodians who are not trained in the law. Cf. State v. Wood, 148 Vt. 479, 490, 536 A.2d 902, 908 (1987) (discussing the value of adopting a rule that “simplifies the analysis that must be made by courts and those who follow our decision”). In saying this, I recognize the rule has limited applicability to circumstances, as present here, where the absence of a crime can be clearly determined. But the introduction of clear, objective rules, even for the small part of the overall public records involved, is a step forward and perhaps a model as we decide other PRA issues.
¶ 27. In this case, nothing criminal occurred, and I therefore would hold that the police records of the incident are not records dealing with the detection and investigation of crime. This result is simple, workable, and true to the objectives of the Public Records Act.
¶28. I am authorized to state that Judge Zonay joins this concurrence.
Burgess, J.,¶ 29. dissenting. I respectfully dissent from construing “arrest” as virtually the opposite of what the statute plainly says. The Legislature expressly provided at the time this incident arose that records of criminal investigations were not accessible to the public under the Public Records Acts (PRA), except for “records reflecting the initial arrest of a person and the charge.” 1 V.S.A. § 317(c)(5) (2010) (emphasis added). It could hardly be clearer that the Legislature intended to withhold information on criminal investigations and investigative detentions not resulting in charges, while mandating disclosure of arrests accompanied by a formal criminal charge.
*184¶ 30. Instead of this straightforward legislative standard, the plurality chooses to impose a variable, or floating, test for public access of police records. Now disclosure depends on whether the temporary detention of a suspect amounts to an arrest for purposes of Fourth Amendment protection, even when, as here, no such claim of unconstitutional invasion is at issue. Custodians of police records must now puzzle over “de facto” arrest versus investigative detention not amounting to arrest — a moving target worthy of countless and diverse court decisions. See D. Dinger, Is There a Seat for Miranda at Terry’s Table?: An Analysis of the Federal Circuit Court Split Over the Need for Miranda Warnings During Coercive Terry Detentions, 36 Wm. Mitchell L. Rev. 1467, 1485-87 (2010) (noting that one of most-litigated issues is point at which temporary investigative detentions become de facto arrests). No police department can reasonably be expected to respond correctly to PRA requests when the answer is so situational and confounding to courts.
¶31. By importing Fourth Amendment jurisprudence into this case rather than relying on the plain language of § 317(c)(5) (“arrest of a person and the charge”), the plurality guarantees litigation over whether, under the circumstances at the time, police intervention amounted to a “de facto” arrest rather than a temporary detention incident to investigation. See State v. Chapman, 173 Vt. 400, 403-04, 800 A.2d 446, 449 (2002) (setting forth totality-of-circumstances test and accompanying factors). As the plurality acknowledges, ante, ¶ 13, there is barely a dim distinction between investigative detentions and de facto arrests. Chapman, 173 Vt. at 403, 800 A.2d at 449 (“There is no bright line rule differentiating an arrest from a detention supportable by less than probable cause.”); see also Allen v. City of Los Angeles, 66 F.3d 1052, 1056 (9th Cir. 1995) (stating that there is “no bright line rule for determining when an investigatory stop crosses the line and becomes an arrest”) (quotations omitted); United States v. Tilmon, 19 F.3d 1221, 1224 (7th Cir. 1994) (noting that “[s]ubtle, and perhaps tenuous, distinctions exist between a Terry stop, a Terry stop rapidly evolving into an arrest and a de facto arrest”).
¶ 32. Thus, not only will municipalities be puzzled, but courts will vary in their results when called upon to review decisions to withhold records of detentions deemed by custodians to be less than de facto arrests. Whether a detention amounts to an arrest will be decided case by case, depending on the circumstances. *185Chapman, 173 Vt. at 403, 800 A.2d at 449. De facto arrest records now classified by the plurality as public information are not always easily recognizable. See id.-, see also United States v. Sharpe, 470 U.S. 675, 685 (1985) (recognizing that its Fourth Amendment precedents “may in some instances create difficult line-drawing problems in distinguishing an investigative stop from a de facto arrest”); Morelli v. Webster, 552 F.3d 12, 19 (1st Cir. 2009) (noting that not only can it be challenging to distinguish between temporary detentions and de facto arrests, but line between them “can shift in the course of a single encounter so that what starts out as an investigatory stop may morph into a de facto arrest”). So much for timely resolution of time-sensitive PRA requests.7
¶ 33. In Caledonian-Record Publishing Co. v. Walton, 154 Vt. 15, 23, 573 A.2d 296, 301 (1990), our principal precedent on the issue raised in this case, this Court concurred with the “holdings of the Ohio and Texas courts that arrest records are not records dealing with the investigation and detection of crime, but rather are the product of such an investigation.” Walton established that a citation was the equivalent of an arrest for purposes of the PRA, and described what underlies an accessible arrest record under the PRA. Both an arrest and a citation turn upon a police finding of probable cause that a crime has been committed and signal a “commencement of a criminal proceeding based on that finding.” Id. at 26, 573 A.2d at 302. That this meant formal arrest on a charge is clear from the assertion that “[l]ike an arrest, a citation must be based upon probable cause to believe that the person has committed a crime.” Id. at 27, 573 A.2d at 303. Again, the statute’s reference to formal, rather than de facto, arrest was reiterated in the Court’s observation that, for purposes of *186§ 317(c)(5), “[o]ne who has been arrested or issued a citation in lieu of arrest is more than just a suspect under investigation” because arrest and citation converts the person from suspect to an accused “summoned into court.” Id. at 26-27, 573 A.2d at 302-03; cf. State v. Lancaster Police Dep’t, 528 N.E.2d 175, 178 (Ohio 1988) (noting that investigation of suspect, unlike arrest or issuance of citation, involves “no public action”). Thus, “for the purposes of the public access statutes,” both an arrest and the issuance of a citation “crosses the line between a police investigation and the judicial process.” Walton, 154 Vt. at 27, 573 A.2d at 303. The line arrived at in Walton was bright enough, and was not crossed here.
¶ 34. Then, as now, the distinction in Walton between an inaccessible criminal investigation file and publicly available records of a later formal arrest or citation accompanied by a charge makes sense. The PRA’s wording and objective requires no extension of “arrest” to temporary detentions incidental to a criminal investigation. That this is what the Legislature intended is reflected in its express connection of the public arrest record to a public charge in § 317(c)(5).8
¶ 35. Limiting public disclosure under § 317(c)(5) to incidences involving only formal arrest, rather than temporary de facto arrests, is also consistent with principles of common law access to such records to prevent secret arrests and provide basic information to the public regarding individuals subjected to the judicial process. See Cnty. of Los Angeles v. Superior Court, 22 Cal. Rptr. 2d 409, 416 (Ct. App. 1993) (observing that analogous public-record-act provision was intended “only to continue the common law tradition of contemporaneous disclosure of individualized arrest information in order to prevent secret arrests and to mandate the continued disclosure of customary and basic law enforcement information to the press”). When, as here, an investigation is closed without charge, formal arrest or citation due to an ultimate *187police determination of no criminal activity, there is no secret arrest, no arrestee answerable to judicial process, and no reason for the police to publicize the identity and circumstances of its investigation of a private citizen.9 Cf. State ex rel. Jenkins v. City of Cleveland, 613 N.E.2d 652, 660 n.6 (1992) (explaining that not allowing release of records concerning investigation of suspects before arrest or filing of charges prevents interference with ongoing investigation and “protect[s] people’s reputations”).
¶ 36. Finally, even assuming an arrest occurred here, the plurality calls for release of not only the arrest record, but of all documents related to the criminal investigation exempted from disclosure under § 317(c)(5). Regarding criminal investigations, the statute permits disclosure only of records “reflecting the arrest.” If not explicit, this legislative verb choice fairly implies a narrow exception to the broad exclusion of criminal investigative files from public access. By its terms, an arrest record does not automatically encompass documents of the underlying investigation.
¶ 37. Nevertheless, the plurality opinion orders disclosure, beyond the statute’s command, of all of the records “in question,” ante, ¶ 1, when plaintiff seeks all of “the Town of Hartford’s records related to the incident, including audio recordings of the incident, the witness’s 911 call, officers’ reports, the dispatcher’s log, and written witness statements.” Ante, ¶ 5. Most of those records are outside of the scope of the arrest record, and are not publicly available under the statute. Broader disclosure may be a sound public policy choice, but it was not the option adopted by the Legislature.
¶ 38. Accordingly, I respectfully dissent from extending public access to records of de facto arrests without an accompanying criminal charge — in contradiction of legislative intent — and from expanding that access to records beyond the arrest documents as restricted by the statute.
The ambiguity of the exemption’s text relates to the general ambiguity in referring to something that does not exist — in this case, the crime that is the subject of the investigation. See generally B. Russell, On Denoting, 14 Mind 479 (1905) (confronting the ambiguity in references to things that do not exist); RF. Strawson, On Referring, 59 Mind 320 (1950) (criticizing Russell for attempting to assign a definite truth or falsity to sentences with such references). On the one hand, one might plausibly say that the library of Salem, Massachusetts, contains records dealing with the investigation of witches. One can implicitly ignore the fact that witches don’t exist. On the other hand, one might equally well reject the claim that there are any records of the investigation and detection of witches. After all, witches don’t exist. In short, when the object of investigation or detection does not exist, a reference to records dealing with its detection and investigation can be read either as denoting the records of the misguided enterprise or as denoting no records at all. As a result, § 317(c)(5)’s reference to “records dealing with the *180detection and investigation of crime” is ambiguous between covering and not covering the situation where no crime exists.
The dissent claims that the concerns about chilling citizen tips are equally strong when the tips turn out to be unsubstantiated. Post, ¶ 32 n.1. I disagree. Tips that *182turn out to be unsubstantiated not only lack any positive value, but can actually be counterproductive or harmful. Citizens should be aware that, if they give information that turns out to be unsubstantiated, their anonymity will not be guaranteed.
That the subjective approach is misguided is further indicated by the fact that a record might be created initially as community caretaking under the impression that no crime was present, only to discover a crime subsequently. We would, I *183believe, view such a record as a record of a criminal investigation. This situation here is the reverse, and the result should be the reverse.
Nor is Justice Dooley’s proposed hindsighted “bright line” approach, which would deny that a criminal investigation occurred when no crime was detected, ante, ¶ 16, consistent with § 317(c)(5) or its purpose. It is true, as posited, that police files on witchcraft investigations would not be records of criminal investigations since witchcraft is no crime. The same cannot be said, however, of records of police response to a burglary complaint, and it is no less a “record dealing with the detection and investigation of crime” when the complaint, like the one in this case, turns out to be unfounded. The same “concerns about chilling citizen cooperation in future cases or revealing investigatory tactics” cited by the concurrence as justifying denying public access to closed criminal investigations under § 317(c)(5), ante, ¶ 23, also generally apply to records of police responses to tips and complaints of possible, although ultimately unsubstantiated, criminal activity.
As the plurality points out, § 317(c)(5) was amended effective July 1, 2011, to refer to records of arrests and records of charges in separate clauses. The pertinent reference to records of “initial arrest” includes “any ticket, citation, or complaint issued for a traffic violation,” and goes on to say that “records reflecting the charge of a person shall be public.” As a whole, the amendment does not undercut the point that § 317(e)(5) refers to arrests associated with charges, or at least an intent to charge — in other words, a “formal” arrest accompanied by an official accusation.
Citizens are, of course, free to disclose whatever they want about an interaction with the police and, in the event of claims of unreasonable or abusive police conduct, may publieally pursue their remedies and records discovery through the courts.