¶ 1. Defendant John Cunningham appeals from a district court order denying his motion to suppress certain evidence obtained after two traffic stops in May of 2005. He argues that the actions taken by the police on both days violated his state and federal constitutional rights. We conclude that both days’ events offend the Vermont Constitution. The district court’s order denying the motion to suppress is reversed, and the case is remanded for further proceedings consistent with the views expressed herein.

I. Facts

a. May 5, 2005

¶ 2. The facts of the first traffic stop are undisputed. Defendant was driving through Vergennes, Vermont, on the afternoon of May 5, 2005, when Officer Rodney Trudeau saw his vehicle and radioed dispatch to request a registration check. The check revealed that defendant owned the vehicle and that his license was suspended for failure to maintain automobile insurance. See 23 V.S.A. § 802(a).

*405¶ 3. The officer, who was on foot, approached defendant’s vehicle at a stop sign and asked if he was John Cunningham. Defendant replied that he was, and the officer asked him to pull over. This occurred at 3:04 p.m. and the incident was labeled at that time in the police radio log as “drugs.” Defendant was asked to produce a driver’s license, vehicle registration, proof of insurance, or other form of identification. He produced none of these, in violation of several provisions of Vermont law. See 23 V.S.A. § 611 (“Every licensee shall have his or her operator’s license certificate in his immediate possession at all times when operating a motor vehicle.”); id. § 676 (operating vehicle after license suspended for failure to maintain insurance is a civil violation); id. § 800 (prohibiting motor-vehicle operation without current automobile liability insurance). These offenses were all civil in nature. When asked, defendant told the officer that he owned the vehicle and gave his true identity. Defendant does not appear to have made any attempt to conceal either his identity or the ownership of the vehicle.

¶ 4. The officer first called a tow truck, and then called dispatch and requested that defendant’s name be run through the computer-aided dispatch (CAD) system. The CAD search revealed, according to the officer’s affidavit, that defendant had “one prior drug involvement.” The officer had also heard “through other sources” that defendant was a cocaine' dealer. These sources were wholly anonymous. The officer then asked defendant if he had drugs in his vehicle; defendant said that he did not. Defendant did not consent when the officer asked to search his vehicle. When the officer asked defendant why he, a Middlebury resident, was in Vergennes that day, defendant responded that there was no particular reason. The officer reported that defendant was acting nervous throughout the stop and subsequent conversation, but that there was no sign of drug intoxication.

¶ 5. The officer then requested backup from the Vergennes police. When the other officers arrived, Officer Trudeau requested that a canine unit also respond to the scene. The closest available canine unit was based in Hinesburg; the officer called the canine unit and then began writing the four traffic tickets he planned to issue to defendant. By the time the canine unit arrived, more than forty minutes had elapsed since defendant was pulled over. The canine-unit officer ordered defendant out of the vehicle, expressing concern that the drug dog, “Tiger,” might otherwise be aggressive *406towards him. When defendant exited the vehicle, the officer patted him down “for weapons” and found cocaine residue, drug paraphernalia, and $263 in cash. Defendant attributed the cash to a construction job, but was unable to identify where or for whom he worked. Defendant was handcuffed, and the canine-unit officer led Tiger through an external sniff of the car.1 Tiger “alerted” to the seams between the front and rear doors on both sides of defendant’s vehicle. Forty-six minutes elapsed between the initial stop and Tiger’s alert.

¶ 6. Defendant was then detained at the police station in Vergennes while the officers applied for a warrant, which they served on defendant at approximately 9:00 p.m. Upon executing the warrant and searching defendant’s clothing and his vehicle, the officers discovered one gram of crack cocaine, some purple pills in an unmarked bottle, various drug paraphernalia, and additional cash.

b. May 17, 2005

¶ 7. The facts of the second traffic stop are also largely undisputed. At approximately 8:00 p.m. on May 17, 2005, Officer Trudeau, who was driving east on South Maple Street in Vergennes, received an anonymous telephone call complaining of suspicious activity at a residence on nearby King Street. The caller reported that “suspicious persons” were carrying packages in and out of the building, that the caller suspected drug activity, and that defendant was leaving the residence in a maroon car and was, like the officer, driving east on South Maple Street. The caller, like the “other sources” from whom the officer had heard before the May 5 stop, was entirely anonymous. The officer followed defendant’s vehicle to an intersection, where defendant applied the brakes, revealing a malfunctioning brake light. After stopping the vehicle, Officer Trudeau recognized the driver as defendant and the passenger as someone the officer had heard was involved with cocaine. Defendant again could not provide a driver’s license, proof of registration, or proof of insurance.

¶ 8. The officer asked defendant if he had drugs in the vehicle; defendant said that he did not. During the stop, the officer noted that both defendant and his passenger appeared “very nervous” *407and “very impatient.” Two vehicles that the officer had seen earlier at the King Street residence drove by repeatedly during the stop. At 8:10 p.m., the officer requested a canine unit from Burlington and made arrangements for defendant’s car to be towed at a later time. The officer began to write the four tickets he intended to issue to defendant. The canine unit — Officer Radford and his dog Stoney — arrived approximately twenty-eight minutes after being summoned, at about 8:38 p.m., by which time Officer Trudeau was writing the third of the four tickets. During the canine sniff of the car, defendant and his passenger were given the option to remain in the vehicle. Defendant chose to remain, while his passenger chose to exit. The passenger was patted down and allowed to leave the scene when nothing incriminating was found on his person. Officer Radford then had Stoney sniff the vehicle. Stoney “alerted” to the vehicle, and defendant was asked to exit the vehicle, which was seized and impounded.

¶ 9. Defendant was placed in custody pending Officer Trudeau’s application for a search warrant covering both defendant and his vehicle. Officer Trudeau obtained the warrant and served it on defendant after midnight. The subsequent search of defendant’s person revealed nothing, but the search of the vehicle disclosed a total of approximately eighteen grams of cocaine, some of it loose and the rest divided between several individual “paper folds” and a plastic bag.

II. The proceedings below

¶ 10. Defendant was charged with two counts of possession of cocaine, one a misdemeanor and the other a felony, based on evidence obtained on May 5 and 17, 2005. See 18 V.S.A. § 4231(a)(1), (2) (Cum. Supp. 2006). Prior to trial, defendant moved to suppress all of the evidence obtained on both days, alleging that both detentions were impermissible under Chapter I, Article 11 of the Vermont Constitution and the Fourth and Fourteenth Amendments to the United States Constitution. Defendant’s motion was denied by written order, and he subsequently pled guilty, conditioned on the outcome of this appeal. V.R.Cr.P. 11(a)(2).

¶ 11. The district court first noted that defendant did not contest the validity of either day’s initial stop, but challenged only his extended detention and the use of the canine sniff. Citing our decision in State v. Emilo, 144 Vt. 477, 481, 479 A.2d 169, 171 *408(1984), the district court stated that a police officer must have a reasonable and articulable suspicion of criminal activity before detaining a person for further investigation after a routine traffic stop.2 Further, the court noted dicta in our decision in State v. Sprague, 2003 VT 20, ¶ 17, 175 Vt. 123, 824 A.2d 539, supporting the proposition that an officer may extend the “scope and length” of a stop beyond its original purpose when the officer has “sufficient justification” for doing so. The court went on to characterize this sufficient justification as “sufficient reasonable suspicion,” and noted that “[officers are not barred from making observations, or asking a few routine questions, or using information radioed from [CAD].”

¶ 12. The district court concluded that Officer Trudeau had a sufficient basis to detain defendant on May 5 based on four “objective facts.” According to the district court: (1) the officer “knew that defendant had prior involvement with drugs from [CAD] and from other sources”; (2) defendant “appeared nervous”; (3) defendant could not produce a driver’s license, proof of registration, or proof of insurance; and (4) “could not explain why he was in Vergennes that day.”3 The court also noted that the officer’s prolonged detention of defendant on May 5 resulted in only a “minimal” additional seizure because the officer had only *409finished writing three of the four tickets by the time the canine arrived.4

¶ 13. As to the May 17 traffic stop, the district court concluded that the objective facts supporting the officer’s post-traffic-stop detention were: (1) “a complaint that identified defendant as being involved with drugs at the Maple Street residence,” (2) the officer’s “observation of suspicious vehicles patrolling the stop scene,” and (3) the officer’s “knowledge of defendant’s prior involvement with drugs.”

¶ 14. Our review of a decision on a motion to suppress involves two steps. State v. Freeman, 2004 VT 56, ¶ 7, 177 Vt. 478, 857 A.2d 295 (mem.). Our first task is to review the trial court’s factual findings for clear error. Id. “If the trial court’s findings are not clearly erroneous, we will then review the legal issues . . . de novo.” Id. Here, defendant does not take issue with the trial court’s findings, and would have us find error only in its conclusions of law. Defendant also concedes that there was probable cause for both days’ initial traffic stops; he challenges his subsequent detention on both days, the May 5 exit order and patdown search, and the use of drug dogs to sniff his car without a warrant. Defendant argues that the district court erred in concluding that these events did not violate his rights under Article 11, and in denying his motion to suppress all evidence obtained during the stops and subsequent investigations. We begin our review with the post-stop detentions on May 5 and 17.

¶ 15. Under both the Fourth Amendment and Article 11, a traffic stop is a seizure and must be supported by a reasonable *410suspicion of criminal activity. United States v. Sharpe, 470 U.S. 675, 682 (1985); Sprague, 2003 VT 20, ¶ 17. We also inquire into “whether [the subsequent investigation] was reasonably related in scope to the circumstances which justified the interference in the first place.” Terry v. Ohio, 392 U.S. 1, 20 (1968); see also State v. Chapman, 173 Vt. 400, 402, 800 A.2d 446, 449 (2002). An investigative stop, based at its inception on a reasonable suspicion, may reveal further information that justifies greater restrictions on a suspect’s liberty, up to and including arrest. State v. Gray, 150 Vt. 184, 192, 552 A.2d 1190, 1195 (1988) (initial stop on suspicion of DUI justifiably escalated into seizure for field-dexterity tests and, ultimately, into arrest based on failure of dexterity and breathalyzer tests).

¶ 16. We have consistently held that Article 11 provides greater protections than its federal analog, the Fourth Amendment:

Whatever the evolving federal standard, when interpreting Article Eleven, 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.

State v. Berard, 154 Vt. 306, 310-11, 576 A.2d 118, 120-21 (1990) (citation omitted).

¶ 17. Accordingly, we have declined to follow the more permissive Fourth Amendment jurisprudence governing exit orders. In Sprague, we rejected on Article 11 grounds the United States Supreme Court’s holding, in Pennsylvania v. Mimms, 434 U.S. 106, 111 (1977), that police officers making routine traffic stops may order motorists to exit their vehicles without having any particularized suspicion of criminal activity or danger to the officer. Sprague, 2003 VT 20, ¶¶ 14-17. We held in Sprague that, while “the police may stop and temporarily detain a vehicle based on little more than a reasonable and articulable suspicion of wrongdoing,” id. ¶ 17, the “police intrusion [must] proceed no further than necessary to effectuate the purpose of the stop.” Id. We cited with approval the leading Massachusetts case in this *411area: “‘Citizens do not expect that police officers handling a routine traffic violation will engage, in the absence of justification, in stalling tactics, obfuscation, strained conversation, or unjustified exit orders, to prolong the seizure in the hope that, sooner or later, the stop might yield up some evidence of an arrestable crime.’ ” Id. (quoting Commonwealth v. Gonsalves, 711 N.E.2d 108, 112 (Mass. 1999)).

¶ 18. Our analysis begins with the question of whether defendant was detained for too long, and with too little justification, before the canine units arrived, to pass muster under Article 11. Because we conclude that he was, and that his motion to suppress should have been granted on that basis, we do not reach his other claims. As noted supra, n.2, the State stipulated that, if we conclude that there was no reasonable suspicion upon which to detain defendant, the motion to suppress' should have been granted.5

III. May 5, 2005

¶ 19. The initial traffic stop on May 5 was justified, as defendant concedes, by the officer’s knowledge that defendant’s license was suspended. See 23 V.S.A. § 674(a) (prohibiting operation of motor vehicle while operator’s license is suspended). Defendant also takes no issue with the officer’s subsequent request that he produce proof of insurance and ownership. Accordingly, under Sprague, the officer could have properly detained defendant for long enough to “effectuate the purpose of the stop,” 2003 VT 20, *412¶ 17, which was to write citations for the traffic violations committed. See also State v. Hewey, 144 Vt. 10, 15, 471 A.2d 236, 239 (1983) (detention for additional nine minutes to verify that defendant had valid license permissible when defendant failed to produce registration certificate). The officer’s expansion of the stop into a drug investigation required a reasonable, articulable suspicion that defendant was committing a drug-related crime. We conclude that the officer here did not have a reasonable, articulable suspicion of drug activity on May 5, and that the district court therefore erred in denying defendant’s motion to suppress the evidence obtained on that day.

¶20. According to the district court, the first “objective fact” that justified Officer Trudeau’s extended detention of defendant was that the officer “had heard that defendant had prior involvement with drugs from the [CAD] system and from other sources.” These other sources were not named, and the CAD entry relied upon did not disclose any detail regarding the information’s reliability, the nature of defendant’s purported involvement, or the identity of the source from whom police heard of the involvement. We first consider the information that the officer had heard from unnamed “other sources” identifying defendant as “a known Cocaine Dealer in the area.”

¶21. We have had occasion to consider the weight to be given to tips in reviewing a warrantless search of an automobile, and that analysis is pertinent here. State v. Langlois, 164 Vt. 173, 667 A.2d 46 (1995). In Langlois, we concluded that an officer did not have probable cause to search an automobile based on a telephone informant who stated that the defendant, Langlois, was driving around downtown Bennington in a 1989 pickup truck with fresh front-end damage and a bag of marijuana behind the front seat. Id. at 178, 667 A.2d at 49. We concluded that “the information . . . provided [about the vehicle] was readily available to any member of the public who could observe defendant’s vehicle” and noted that “[t]here was nothing particularized or predictive about the information.” Id. at 177, 667 A.2d at 49. Here, the information from “other sources” was even less reliable than the tip we rejected in Langlois. First, the informant in Langlois did provide a name, albeit one unknown to police, while the “other sources” here were wholly anonymous. Second, the Langlois informant provided at least some specific information beyond a mere statement of wrongdoing, while here the record reflects that “other *413sources” merely accused defendant of dealing drugs and provided no corroborating information at all, much less any unique information that could form a basis to determine the reliability of the information. While Langlois concerned probable cause, and here defendant’s prolonged detention could be justified based on a lesser showing of reasonable suspicion, the accusations by anonymous “other sources” do not surmount even that lower threshold.

¶22. Similarly, the information from the CAD system does not support a reasonable suspicion that a drug-related crime was afoot. The information was also derived from an anonymous source, which undercuts its reliability. And the mere fact that the information was contained in this particular database does not greatly increase its value as a basis for reasonable suspicion; there is nothing in the record to suggest that information undergoes any sort of vetting prior to inclusion in the database. As was noted at oral argument before this Court, the “prior drug involvement” could have arisen from an incident as innocuous as a neighbor’s hypothesis and unsubstantiated assertion that defendant was involved in some way with drugs, or from defendant’s mere association with someone suspected to be involved with drugs. Based on the nonspecific information provided here, the officer had no way to know reliably what the CAD entry meant. Article 11 does not permit prolonged detention based on an officer’s having heard what amounts to little more than a rumor of wrongdoing.

¶ 23. Indeed, many courts have held that even a prior arrest or conviction does not support a reasonable suspicion of present illegal activity. See, e.g., United States v. Jones, 234 F.3d 234, 242 (5th Cir. 2000) (“[Ajrrest alone does not amount to reasonable suspicion.”); United States v. Sandoval, 29 F.3d 537, 542 (10th Cir. 1994) (“[K]nowledge of a person’s prior criminal involvement (to say nothing of a mere arrest) is alone insufficient to give rise to the requisite reasonable suspicion” to expand a traffic stop into a drug investigation). The CAD entry and the “other sources” do not provide a sufficient basis for a reasonable suspicion of drug activity. But the State also offers, and the trial court found, other bases for the officer’s reasonable suspicion. We turn next to those justifications.

¶24. The trial court noted that defendant appeared nervous. The defendant’s nervousness provides only meager sup*414port for a reasonable, articulable suspicion of drug activity. It is certainly not uncommon for a citizen stopped by police to be nervous, and many courts have found nervousness to be “of limited significance in determining reasonable suspicion.” United States v. Fernandez, 18 F.3d 874, 879 (10th Cir. 1994); see also, e.g., United States v. Tapia, 912 F.2d 1367, 1371 (11th Cir. 1990) (driver’s visible nervousness and shaking hands, even in combination with other asserted grounds for reasonable suspicion of drug activity, “fail to suggest . . . any criminal activity other than speeding on the highway”); People v. Haley, 41 P.3d 666, 675 (Colo. 2001) (“[I]t is not uncommon for most citizens — whether innocent or guilty — to exhibit signs of nervousness when confronted by a law enforcement officer.”) (citation omitted). Although generalized “nervousness” is a weak basis for an officer’s reasonable suspicion of particular criminal activity, such nervousness may be considered together with other, more substantial factors. See, e.g., United States v. Givan, 320 F.3d 452, 458-59 (3d Cir. 2003) (officer justified in prolonging stop based in part on driver’s nervousness where driver had rented vehicle twenty-four hours before, had driven from Michigan to New York without stopping, and officer knew drug dealers commonly made the trip in this manner). Here, however, the other factors offered in support of the detention are insufficient, even in combination with defendant’s nervousness, to support a reasonable suspicion of present illegal activity beyond the motor-vehicle violations.

¶ 25. The district court also concluded that defendant’s prolonged detention was justified in part because defendant did not prove ownership of the vehicle, produce identification, or show proof of insurance. While these violations did justify a detention long enough to write citations for them, without more they provide inadequate support for the officer’s suspicion of drug activity at the May 5 stop. Cf. United States v. Gonzalez, 328 F.3d 755, 758 (5th Cir. 2003) (reasonable justification found for detention and search where defendant was “extremely nervous,” lied about not having driver’s license, was not on plausible route to claimed destination, and had been previously arrested for drug trafficking).

¶ 26. Finally, the district court noted that defendant, “a Middlebury resident, also could not explain why he was in Vergennes that day.” Defendant told the officer that there was “no *415reason” he was in Vergennes on May 5. Under different circumstances, a person’s failure to provide a plausible reason for his presence in a particular place might contribute strongly to an officer’s reasonable suspicion, and might even be a factor in determining probable cause to arrest. See, e.g., United States v. Carrillo, 902 F.2d 1405, 1412 (9th Cir. 1990) (officers had reasonable suspicion for Terry stop — and probable cause to arrest — where defendant was found hiding alone behind a bush in the desert in the middle of the night within walking distance of an airstrip where 700 pounds of cocaine had been intercepted that night; defendant’s claim that he had been “partying” with friends was implausible). The May 5 stop presented circumstances profoundly different from those in Carrillo and cases like it, of course. Defendant here, unlike the defendant in Carrillo, was not hiding, disheveled and exhausted, in a place that no person had a reason to be, in the dead of night. Defendant was simply present — openly and in broad daylight — on a public thoroughfare in a town about fifteen miles from his home. That he told the officer that there was “no particular reason” he was there does not bear any weight in establishing a reasonable suspicion of criminal activity.

¶ 27. The totality of the circumstances known to the officer at the time of the stop on May 5 did not support a reasonable suspicion of present illegal drug activity. And at no time prior to the canine’s alert did the officer learn new information sufficient to support that suspicion. Accordingly, we must conclude that the officer’s detention of defendant for forty-six minutes to await the arrival of the dog violated Article ll.6

¶28. Defendant also finds constitutional fault with two other aspects of the May 5 stop: the order that he exit the car, and the subsequent search of his person. In light of the State’s concession, however, defendant’s forty-six-minute detention alone provided a *416basis to grant the motion to suppress the May 5 evidence, and we need not consider the constitutionality of the exit order or the search of defendant’s person.

IV. May 17, 2005

¶29. The district court denied defendant’s motion to suppress the evidence obtained on May 17 based on the anonymous tip, the purportedly suspicious vehicles driving by the traffic stop, and the officer’s knowledge of defendant’s drug possession during the encounter on May 5. We consider each factor in turn, while acknowledging that the three asserted justifications must ultimately be considered together to evaluate whether the district court erred in denying the motion to suppress.

¶ 30. We have previously evaluated an anonymous tip as support for an officer’s reasonable suspicion to believe a driver was engaged in criminal activity, such that the officer was justified, under the United States Constitution and Terry, 392 U.S. at 21-22, in making an initial investigatory stop. State v. Boyea, 171 Vt. 401, 765 A.2d 862 (2000); State v. Lamb, 168 Vt. 194, 720 A.2d 1101 (1998). While Lamb concerned the reasonable-suspicion standard in the context of an initial stop, the analysis therein is useful to inform our inquiry into whether the officer had a reasonable suspicion of drug activity sufficient to justify the post-stop investigative expansion in this case. That expansion is an additional seizure under Article 11, and therefore must — like an initial stop — be supported by a reasonable, articulable suspicion of wrongdoing. Chapman, 173 Vt. at 403, 800 A.2d at 449.

¶ 31. As we noted in Lamb, “ ‘[a]n informant’s tip, if it carries enough indicia of reliability, may justify a forcible stop.’ ” 168 Vt. at 196, 720 A.2d at 1102 (quoting State v. Kettlewell, 149 Vt. 331, 335, 544 A.2d 591, 594 (1987)). We cited with approval a United States Supreme Court decision, Alabama v. White, in which the Court upheld a stop based on an anonymous tip stating that the defendant possessed cocaine and would soon leave a specific building, get into a particular car, and proceed to a named location along a specified route with numerous turns. 496 U.S. 325, 332 (1990).

¶ 32. Accordingly, in Boyea, we upheld a traffic stop premised on an anonymous tip that the driver of a “blue-purple Volkswagen Jetta with New York plates” was operating erratically *417on a named stretch of Interstate 89. 171 Vt. at 402, 765 A.2d at 863. We noted that, although it was a “close case,” the anonymous tip was sufficiently reliable to support a “temporary and brief detention that is exposed to public view.” Id. at 409-10, 765 A.2d at 868 (citation omitted). We held that “an anonymous report of erratic driving must be evaluated in light of the imminent risks that a drunk driver poses to himself and the public.” Id. at 402, 765 A.2d at 863.

¶ 33. Similarly, in Lamb, we upheld a stop based on an anonymous tip that alerted officers that “defendant was upset and intoxicated and was driving away from a residence on ... a dirt road ... in a rural area.” Lamb, 168 Vt. at 197, 720 A.2d at 1103. Like the White Court, we held — in a divided opinion — that the tip, because it predicted behavior of which only a small number of people would be aware, possessed sufficient indicia of reliability to support the traffic stop. Id. at 201-02, 720 A.2d at 1105-06. Assessing the totality of the circumstances, we held:

Considered in the light of [the cited] authorities, the circumstances here were more than sufficient to justify the stop. These circumstances included not only the trooper’s rapid verification of the information supplied by the informant, but also the virtual impossibility that such information could have been supplied by anyone but a knowledgeable insider, the officer’s personal knowledge of defendant’s prior DUI arrest, and the potential danger — to himself as well as to others — posed by an intoxicated driver.

Id. at 202, 720 A.2d at 1106. Here, by contrast, the information supplied by the anonymous informant was not uniquely available only to “knowledgeable insider[s].” Id. The fact that a maroon vehicle had already left a particular residence and was proceeding in a particular direction along a named street would have been apparent to any member of the public who happened to be in the area. The tip here was quite similar to the tip rejected by the New Hampshire Supreme Court in State v. Kennison, 590 A.2d 1099, 1100-01 (1991), a case we cited with approval in Langlois, 164 Vt. at 178-79, 667 A.2d at 49-50. In Kennison, the court suppressed evidence obtained from an investigative stop that was based on an anonymous informant’s statement that he had seen marijuana in the trunk of the defendant’s blue Cadillac, and on *418the informant’s statement that the defendant would leave a named workplace at a particular time. Kennison, 590 A.2d at 1100. The Kennison court held that an anonymous tip predicting nothing more remarkable about the defendant’s future behavior than that she would come home from work in the afternoon and go out later in the evening did not meet the standard announced in White and so did not support a reasonable suspicion of illegal activity. Id. at 1102.

¶ 34. Here, at the relevant time — during defendant’s prolonged detention waiting for the canine unit to arrive — there was no “potential danger” akin to that posed by the drunk drivers in Boyea and Lamb. The initial stop had already been made, and the informant’s allegation that defendant had illegal drugs in the car did not raise the specter of imminent danger that would have loomed had the tipster told police that defendant was, for example, driving while impaired by drugs or alcohol. See Boyea, 171 Vt. at 409, 765 A.2d at 867; cf. Florida v. J.L., 529 U.S. 266, 272-74 (2000) (declining to adopt a “firearm exception” to the warrant requirement; noting that possession of a firearm, like possession of narcotics, does not pose an imminent danger). As we said in Boyea, anonymous tips alleging possessory offenses, whether of guns or drugs, do not state a risk of imminent harm such as would justify a relaxed reasonable-suspicion standard. Id. As in the concealed-gun context in J.L., “the police could . . . [instead] surreptitiously observe the individual for a reasonable period of time without running the risk of death or injury with every passing moment.” Boyea, 171 Vt. at 409, 765 A.2d at 867.

¶ 35. The second basis cited by the district court for denying defendant’s second motion to suppress was that suspicious vehicles repeatedly drove by the scene of the stop. According to the officer’s affidavit, “[b]oth vehicles were at the same King Street address earlier. Their level of obvious concern seemed suspicious to me.” We do not agree, however, that “obvious concern” by suspected acquaintances supports a particularized suspicion of criminal drug activity. Defendant’s acquaintances would, presumably, have been interested to observe what transpired after the officer pulled defendant over, regardless of whether a crime was afoot. There is nothing in the record to suggest that the other vehicles or their drivers were doing anything other than driving by the scene of the stop. This is *419behavior that might just as commonly be engaged in by the companions of an innocent driver as the friends of a guilty one, and accordingly does not support the officer’s suspicion that defendant was engaged in criminal activity requiring prolonged detention. It certainly does not support a particularized suspicion of drug activity requiring a lengthy wait for a canine unit.

¶ 36. The district court also relied on the officer’s knowledge of defendant’s drug possession on May 5 as a basis for the officer’s suspicion that defendant possessed illegal drugs on May 17. Although none of our cases are closely on point, many courts have concluded that the fact of a prior arrest does not support a reasonable suspicion of current criminal activity. See, e.g., Jones, 234 F.3d at 242 (no reasonable suspicion for prolonged detention awaiting canine unit despite defendant’s admission of prior arrest for crack cocaine possession); United States v. Lee, 73 F.3d 1034, 1040 (10th Cir. 1996) (“Both Defendants have extensive criminal histories, but knowledge of a person’s prior criminal involvement (to say nothing of a mere arrest) is alone insufficient to give rise to the requisite reasonable suspicion to justify a shift in investigatory intrusion from the traffic stop to a . . . drugs investigation.”) (citation omitted), overruled on other grounds by United States v. Holt, 264 F.3d 1215, 1226 n.5 (10th Cir. 2001). Accordingly, the officer’s knowledge of defendant’s prior arrest did not — even in combination with the other justifications discussed above — satisfy Article 11.

¶ 37. The district court also based its denial of the motion to suppress on the following passage from a treatise we have cited in our search-and-seizure cases: “if the suspect’s explanation needs to be checked out . . . there is reason to continue the detention somewhat longer while the investigation continues.” 4 W. LaFave, Search & Seizure § 9.2(f), at 61 (3d ed. 1996). But the record does not reflect any effort to verify an inconsistent or puzzling explanation. Rather, the only justification offered for defendant’s prolonged roadside detention was the officer’s suspicion of drug activity, which apparently was present at the outset of the stops and was not founded on an infirmity in “the suspect’s explanation,” as the treatise contemplates. Id. This is not a case like those cited in LaFave where the circumstances required the officer to detain a motorist to investigate whether the vehicle was stolen, see, e.g., United States v. Pena, 920 F.2d 1509, 1511, 1514 (10th Cir. 1990) *420(car lock was “punched out,” and defendant was unable to supply any proof that he was entitled to operate car), or where the motorist’s story was facially false or inconsistent, see, e.g., State v. Noel, 628 A.2d 692, 695 (N.H. 1993) (in robbery-prone area, driver and passenger gave inconsistent explanation of origin of furniture in pickup truck; detention for forty-two minutes permissible to investigate their stories). Here, by contrast, the officer does not claim to have been investigating or checking out defendant’s explanations or the ownership of the vehicles.

¶ 38. For the foregoing reasons, defendant’s motions to suppress should have been granted.7

Reversed and remanded.

¶ 39. concurring. I agree with the result reached by the majority. However, while the majority describes the issue presented as “whether defendant was detained for too long, and with too little justification, before the canine units arrived, to pass muster under Article 11,” ante, ¶ 18, I would identify the issue as whether there was sufficient support to expand a simple motor-vehicle-violation stop into a drug investigation. I would adopt the analysis the majority applies to the facts of the May 5 and the May 17 stops, but rule that the officer did not have a reasonable, articulable suspicion of drug activity to support the enlargement of the scope of the stop. That is to say, I would hold that there was no reasonable suspicion to support the expansion of a stop for operating with a license suspended due to a failure to maintain automobile insurance — the May 5 stop — and a stop for operating with a malfunctioning brake light — the May 17 stop — into investigations involving a drug-sniffing dog and the pre-arrest detention of the motorist.

¶ 40. A traffic stop is a seizure and must be supported by a reasonable suspicion of criminal activity. United States v. Sharpe, 470 U.S. 675, 682 (1985); State v. Sprague, 2003 VT 20, ¶ 17, 175 Vt. 123, 824 A.2d 539. Any challenge to the constitutionality of a stop and detention must also evaluate “whether [the subsequent investigation] was reasonably related in scope to the circumstances which justified the interference in the first place.” Terry v. Ohio, 392 U.S. 1, 20 (1968); see also State v. Chapman, 173 Vt. 400, 403, 800 A.2d 446, 449 (2002) (holding that State must provide justification for expanding the scope of an investigative detention into a full-scale arrest); State v. Gray, 150 Vt. 184, 191, 552 A.2d 1190, 1195 (1988) (emphasizing that a stop was constitutional partly because “the initial stop and seizure [for DUI did not] include matters outside the original suspicion of DUI”). In Sprague, we held that any “police intrusion [must] proceed no further than necessary to effectuate the purpose of the stop.” 2003 VT 20, ¶ 17; cf. Chapman, 173 Vt. at 402-03, 800 A.2d at 449 (“An investigative detention employs ‘the least intrusive means reasonably available to verify or dispel the officer’s suspicion in a short period of time.’ ” (quoting Florida v. Royer, 460 U.S. 491, 500 (1983)). Thus, in this case, any expansion of the stop into a drug investigation required a reasonable, articulable suspicion that defendant was committing a drug-related crime. As Justice Ginsburg aptly put it in Illinois v. Caballes, “[e]ven if [a] drug sniff is not characterized *422as a . . . ‘search,’ [a] sniff surely broaden[s] the scope of [a] traffic-violation-related seizure.” 543 U.S. 405, 421-22 (2005) (Ginsburg, J., dissenting) (citation omitted).8 Here, there was no valid, reasonable basis for the officer’s decision to convert the processing of simple, civil, ticketed offenses into full-blown criminal investigations.

¶ 41. It is not so much the delay while waiting for the dog to appear as it is the lack of any constitutionally supportable reason to begin a criminal investigation in the first place that transgressed Article 11 here. As the United States Court of Appeals for the Tenth Circuit held in United States v. Sandoval, “knowledge of a person’s prior criminal involvement (to say nothing of a mere arrest) is alone insufficient to give rise to the requisite reasonable suspicion” to expand a traffic stop into a drug investigation. 29 F.3d 537, 542 (10th Cir. 1994). And, as the majority properly concludes, the officer’s other purported reasons for expanding the stop — as relied upon by the trial judge — fall well below any standard of reasonable suspicion and thus fail to provide sufficient justification for the expansion. I am authorized to state that Justice Johnson joins in this concurrence.

¶ 42. concurring and dissenting. In general, we should not decide constitutional questions if we can avoid them, especially novel and difficult questions like whether dog sniffs are searches under Article 11 of the Vermont Constitution. State v. Bauder, 2007 VT 16, ¶ 27, 181 Vt. 392, 924 A.2d 38. Here, of course, the plurality decision9 has not avoided constitutional adjudication; it has simply resolved the case on a different constitutional ground than the one stressed by defendant. More importantly, the path the plurality has taken to avoid the primary *423constitutional question here ignores the fact-finding of the trial court and contradicts the undisputed facts of this case. The plurality justifies this course of action based on a supposed concession of the State. I cannot read the concession of the State as broadly as the plurality does, and question whether we should ignore a legal question squarely presented to us by relying on such a concession in the first place. Thus, the plurality has not resolved this case on the basis of an “adequate lesser ground.” Id. Nor can it do so. Because I believe there is no principled way to avoid the primary constitutional question presented in this appeal, I would hold that we must decide whether the dog sniff is a search under Article 11 of the Vermont Constitution and direct the State to file a brief on this question.10

¶ 43. By contrast, the plurality’s resolution of this case depends on a conclusion that, in both instances where defendant’s car was stopped, the period of detention was extended only in order to enable the drug-sniffing dog to reach defendant’s vehicle. The problem with this conclusion is that it lacks any factual support. Even if we put aside the issues related to the officer’s suspicion that defendant was transporting drugs, defendant was, on both occasions, driving with a suspended license, without a registration, without insurance, and without identification. These were the circumstances in which defendant traveled on May 5 and May 17. Because of these circumstances, the officer impounded the car in each instance and directed that it be towed to a nearby garage. Under the circumstances, defendant would not have driven the car away, at least for hours, if not for days. Even if the towing occurred before the dogs reached the vehicles, the dogs could have sniffed the vehicles at the garages to which they were taken.

*424¶ 44. This case involves a routine application of the “inevitable discovery” rule. Nix v. Williams, 467 U.S. 431, 444 (1984). Under the' rule, if the preponderance of the evidence establishes “that the information ultimately or inevitably would have been discovered by lawful means,” the exclusionary rule does not apply. Id. Here, it was the officers’ decision to detain defendant that was unlawful. But the drugs in the car would have inevitably been discovered by the dogs even if defendant had been released, because, in each instance, the police were lawfully holding the car in which contraband was discovered. The plurality’s answer — that we have no idea “what might actually have happened had defendant gone on his way before the dog arrived” ante, n.7 — is clearly wrong. We do know, in this case, what the result of the dog sniff would have been. Moreover, all inevitable discovery cases involve some degree of prediction. See Nix, 467 U.S. at 444 (explaining that the prosecution has to prove by a preponderance of the evidence that evidence would have been discovered by lawful means).

¶ 45. The proceeding discussion assumes that the plurality is right that officers in both cases extended the detention of defendant until the dogs arrived. This conclusion, however, depends upon appellate fact-finding that is directly contrary to the fact-finding of the trial court. Defendant’s noncompliance with the motor vehicle operating laws meant that he committed four offenses for which the officer had to issue him citations and after which the officer had to arrange for the towing and storage of defendant’s car. These activities necessarily lengthened the stop. Thus, in both instances, the dog arrived while the officer was still writing tickets. With respect to the second stop, the trial court found that “any seizure resulting from delay in the Burlington canine’s arrival was minimal; the officer had four tickets to write and had completed only three when the canine arrived.” The court’s conclusion that the delay was minimal is plainly at odds with the plurality’s conclusion that the officer unduly extended the stop in order to enable the dog to arrive.

¶46. The plurality explains its rejection of the trial court’s finding by asserting that the court’s conclusions as to the length of the stop are internally inconsistent. I find this criticism of the court to be nitpicking, but however one views the court’s finding on point, it directly contradicts the plurality’s finding of fact that the delay was substantial. The plurality has based a wholesale *425rejection of the trial court’s fact-finding on a mere quibble over choice of terminology.

¶ 47. The plurality is aware of these facts; they are discussed in footnote 7 of the plurality decision.11 It answers that the State conceded that “defendant’s detention was prolonged by waiting for the canine unit.” Ante, n.2. I do not believe the State made the concession that the plurality attributes to it. Although the State agreed that reasonable suspicion was required to justify the search of the car, the State also stated that “[djefendant makes much of the fact that Officer Trudeau was still writing tickets when the canine units arrived but provides no basis for the assertion that this was a stalling tactic on the part of Officer Trudeau.” The oral argument in this case proceeded as much on the theory that reasonable suspicion was required by Article 11 for dog sniffs generally as on the theory that some defect in the officer’s actions before the dog sniffs triggered that standard. Indeed, the best explanation for the State’s position is that it agrees with the trial court, which held that:

An officer must have a reasonable and articulable suspicion of drug activity before summoning a canine unit and allowing a traffic stop to extend into a longer investigative pre-arrest detention. Here, Officer Trudeau had specific and articulable suspicions pointing to drug activity and on that basis properly summoned the canine unit.

Under this theory, the decision to call for the dog required reasonable suspicion even if the dog appeared very quickly. In adopting this decision, the trial court did not shy away from *426ruling on the constitutionality of the dog sniffs. I would surmise that the plurality has not analyzed this theory because it is inextricably intertwined with the question of the constitutional validity of dog sniffs.

¶ 48. The far more significant answer to the plurality’s concession theory is that the concession attributed to the State is irrelevant. It does not matter that the detention was extended if the dogs would have sniffed the vehicles in any case. Indisputably, the dogs would have done so here.

¶ 49. I am also troubled by the effect that the plurality has given to the State’s concession as it construes it. It is one thing to concede disputed facts; it is quite another to war with undisputed facts. We can affirm a trial court decision on any grounds. Bloomer v. Gibson, 2006 VT 104, ¶ 26 n.4, 180 Vt. 397, 912 A.2d 424. We should at least determine whether to do so by following the undisputed facts and consider whether the sniff by the dogs was a search.

¶ 50. I recognize that the State has not only made a vague concession, but it has also failed to brief and argue the main constitutional question presented by defendant. Ordinarily, I would say that we should decide the question without the benefit of the State’s position or argument. Certainly, the State has defaulted on its obligation to defend its position before this Court.12

¶ 51. In this case, however, the question before us is so important, both practically and theoretically, that we should not decide it without adversary briefing. See State v. Jewett, 146 Vt. 221, 229, 500 A.2d 233, 238 (1985) (explaining that it is the responsibility of this Court to order rebriefing “when the briefs [on state constitutional issues] do not pass muster”). In Illinois v. Caballes, 543 U.S. 405, 409 (2005), the United States Supreme Court ruled that a dog sniff is not a search under the Fourth Amendment. Defendant argues that we should not follow this precedent because it was wrongly decided and is, in any event, inconsistent with our jurisprudence under Article 11. In similar cases, where this Court has been asked to define the meaning of a search under Article 11 in an area where the United States *427Supreme Court has ruled, this Court has usually divided, producing multiple opinions, each offering complex analysis. See State v. Costin, 168 Vt. 175, 179, 720 A.2d 866, 869 (1998) (interpretation of Article 11 in the context of warrantless video surveillance of area beyond curtilage of defendant’s home); State v. Morris, 165 Vt. 111, 126, 680 A.2d 90, 100 (1996) (whether Article 11 recognizes expectation of privacy in the contents of opaque trash bags); State v. Kirchoff, 156 Vt. 1, 10, 587 A.2d 988, 994 (1991) (whether property owner has a reasonable expectation of privacy in open fields when various indicia would suggest to a reasonable observer that the area is private). Apart from the significance of the case before the Court to the parties involved, this decision is very important because it will help to develop our Article 11 jurisprudence and greatly influence future decisions with different facts. We must decide such cases carefully and cautiously.

¶ 52. The State’s default puts us in the uncomfortable position of doing the research on the State’s side of the issue, with no help from the State. I would not do this work for the State but would instead direct it to file a brief on whether dog sniffs are searches and decide that issue under Article 11, as the facts of this case demand.

¶ 53. In voting as I have, I am also not adopting the position of the concurrence. The concurrence concludes that a reasonable suspicion of drug-related criminal conduct is required before the police may call for the dog in the first instance. Whether we adopt the position of defendant, or the position of the concurrence, the result is exactly the same: dog sniffs require reasonable suspicion. I would require the State to brief that question before reaching that conclusion.