dissenting. If the informant in this case had identified herself, I would enthusiastically join the majority opinion. Although anonymous accusations are not per se unreliable, the disclosure of the accuser’s identity shows a willingness to bear the consequences of the accusation. See Brown v. United States, 590 A.2d 1008, 1015 (D.C. 1991); 13 V.S.A. § 1754 (knowingly making a false report to a law enforcement officer “to implicate another” is a crime). Many persons who provide information anonymously are public-spirited and want to prevent criminal conduct, but are unwilling to get involved in the process of investigation and conviction. Others may have different motives. We instinctively are concerned about giving a central place to anonymous accusations in law enforcement because we cannot know the motive of the accuser or judge the accuser’s reliability.
In this case, the anonymous nature of the “tip” was wholly unnecessary, assuming the tip was, in fact, anonymous.* The caller reported a domestic dispute and gave her location. I doubt that she thought she could remain anonymous; the simple request that she give her name would probably have been honored and would have ended this dispute.
*204Unfortunately, the majority has used this case to stretch beyond recognition the limits necessarily imposed to ensure the reliability of anonymous tips and reduced any incentive for law enforcement to obtain the name of tipsters. As a result, this case proves that the danger predicted by Justice Stevens’ dissent in Alabama v. White, 496 U.S. 325, 333 (1990), is real: “Anybody with enough knowledge about a given person to make her the target of a prank, or to harbor a grudge against her, will certainly be able to formulate a tip about her” like the one present here. I do not believe that the curtailment of the personal liberty of innocent Vermonters that will result from today’s decision is justified by the small increase in DUI enforcement it might produce. Accordingly, I dissent.
I agree that the critical precedent is White, 496 U.S. 325. In describing White, however, the majority has left out the central parts of its holding. First, the Supreme Court described the facts in White as a “close case.” Id. at 332. Second, the Court particularly relied on the fact that the tip specified that the defendant would drive a specified route to “Dobey’s Motel,” a drive that involved several turns, and the officers stopped the vehicle just short of the motel. Thus, the Court reasoned:
We think it also important that. . . “the anonymous [tip] contained a range of details relating not just to easily obtained facts and conditions existing at the time of the tip, but to future actions of third parties ordinarily not predicted.” . . . The fact that the officers found a car precisely matching the caller’s description in front of the 235 building is an example of the former. Anyone could have “predicted” that fact because it was a condition presumably existing at the time of the call. What was important was the caller’s ability to predict respondent’s future behavior, because it demonstrated inside information — a special familiarity with respondent’s affairs. The general public would have had no way of knowing that respondent would shortly leave the building, get in the described car, and drive the most direct route to Dobey’s Motel. Because only a small number of people are generally privy to an individual’s itinerary, it is reasonable for police to believe that a person with access to such information is likely to also have access to reliable information about that individual’s illegal activities. . . . When significant aspects of the caller’s predictions were verified, there was reason to believe not only that the caller *205was honest but also that he was well informed, at least well enough to justify the stop.
Id. In summary, White was a close case where the Court was able to find reasonable suspicion because the anonymous tip predicted future behavior, that is, a four-mile journey to Dobey’s Motel.
The majority glosses over this aspect of White, suggesting here that the tip predicted defendant’s itinerary and gave “significant portions of information not privy to the general public.” This is a gross exaggeration of the contents of the tip in this case. The tip stated only that defendant would get into his automobile and leave through the end of a short dead-end street that was open. This was not a prediction because, as the majority states, “the informant’s call was contemporaneous with defendant’s departure.” All the officer corroborated was that defendant was driving out the end of the dead-end street. Indeed, when the officer arrived twelve minutes later, defendant had gone only a few feet and claimed to still be in the driveway.
The majority argues, however, the “prediction” takes on added significance because defendant was in a rural area. Unless we are prepared to say that fewer people know of their neighbors’ (or enemies’) affairs in a rural area, an assumption that is probably directly contrary to reality, this is a red herring. The fact is that there is no prediction here, and what facts were disclosed in the tip were available to others. See, e.g., State v. Altieri, 951 P.2d 866, 868 (Ariz. 1997) (anonymous tip that stated that vehicle contained drugs, gave license plate number, and added that it could be found going in a certain direction on an identified highway did not provide reasonable suspicion even when police observed vehicle going by on highway); VonLinsowe v. State, 445 S.E.2d 371, 372-73 (Ga. Ct. App. 1994) (same); Johnson v. State, 659 N.E.2d 116, 119 (Ind. 1995) (tip that named individual would be found transporting drugs in described automobile “in a particular area of town” was insufficient to create reasonable suspicion even though officer had arrested individual for drug possession in past and described vehicle was found where tipster predicted it would be); Commonwealth v. Lyons, 564 N.E.2d 390, 393 (Mass. 1990) (tip that included description of vehicle, even when confirmed, is not sufficiently predictive to provide reasonable suspicion); State v. Miller, 510 N.W2d 638, 644-45 (N.D. 1994) (tips that give only description and location of vehicle provide only easily obtained facts available to general public; corroboration of this type of information does not increase reliability of tip).
*206Nor am I persuaded that the fact that the officer in this case had “processed” defendant for DUI in the past justified the stop. I agree that in the totality of the circumstances an officer may rely on knowledge of a suspect’s prior criminal activity. See United States v. Harris, 403 U.S. 573, 583 (1971). I don’t believe, however, that this factor can create reasonable suspicion where it does not otherwise exist. See 4 W LaFave, Search & Seizure § 9.4(f), at 192 (3d ed. 1996) (those with prior arrests and convictions must be allowed to live without risk of constant harassment based on their record); cf. Parish v. State, 939 S.W2d 201, 204 (Tex. Ct. App. 1997) (where anonymous tip of drug selling is inadequately corroborated, fact that defendant has a criminal record is not sufficient to allow issuance of search warrant).
Despite the purported adherence to White, the real holding of this case is that the corroboration deemed necessary in White is unnecessary here because defendant was accused of driving while intoxicated. Contrary to the majority’s argument, there is no statement in White that the nature of the crime is balanced against the reliability of the anonymous tip to determine whether it creates reasonable suspicion. Although White was a drug possession case, most of the anonymous tip cases have involved drug sales. If the principle the majority espouses is that uncorroborated anonymous tips are acceptable if they include an allegation of on-going criminal conduct that will injure third parties, these cases would qualify for the exception. The exception the majority espouses is neither narrow nor containable.
In adopting this broad exception, the majority has uncritically accepted decisions that do adopt a narrow and reasonable exception that is not applicable here. In United States v. Bold, 19 F.3d 99, 104 (2d Cir. 1994), the court adopted its exception because reasonable law enforcement tools to confirm the criminal conduct claimed were too dangerous to the law enforcement personnel involved and others. See Commonwealth v. Hawkins, 692 A.2d 1068, 1071 (Pa. 1997) (noting that carrying licensed firearm is lawful in Pennsylvania, court rejected argument that anonymous tip was reliable because it claimed that defendant possessed gun in course of criminal conduct). The DUI cases generally involve situations where the anonymous caller reports having seen dangerous driving, consistent with the alcohol impairment of the operator, and describes the vehicle, its location, and its direction of travel. As the Supreme Court of New Hampshire emphasized in State v. Melanson, 665 A.2d 338, 340 (N.H. 1995), this is a report of publicly-displayed criminal conduct, much harder to *207fabricate, and suggesting that the officer could not safely observe the defendant without threatening his own safety and that of the public.
In this case, no one saw defendant operate a vehicle unsafely; the claim was generally that defendant was intoxicated, not that he had been driving dangerously; and the officer arrived on the scene before defendant had traveled more than a few feet. Thus, unlike Melanson, the information was private, not public. Unlike Bold, there is no specific reason to believe that the officer could not have safely followed defendant and stopped him when he observed erratic operation, a routine law enforcement action that has been reported scores of times in our cases. Given the BAC level in this case, I have no doubt that defendant’s driving would have demonstrated impairment and would have established that he was guilty of the crime, rather than the subject of a false accusation.
It is this latter concern that makes the majority’s broad rule unacceptable to me. If the anonymous informant calls and says defendant is carrying drugs, the police may not stop the vehicle. If the informant calls and says that defendant is carrying drugs, while driving drunk, the police may stop the vehicle. The dishonest informant will learn to tailor the accusation to bring about the desired result. Exactly the consequences feared by Justice Stevens are made easier by the majority’s relaxed standard of reliability.
Finally, I find this case an inappropriate vehicle to fashion the broad rule on which the majority relies. As discussed infra, the theory used by the majority is not that used by the trial court, and there are serious consequences of the shift in theory. The trial court found the stop valid because the caller reported a domestic disturbance. The majority has not attempted to affirm the decision on the trial court’s theory. I do not think it can be affirmed on that theory because the tip did not claim that the domestic dispute involved criminal conduct and it did report that defendant had left the scene. See State v. Caron, 155 Vt. 492, 499, 586 A.2d 1127, 1131 (1990) (facts and inferences from them must “warrant a reasonable belief that a suspect is engaging in criminal activity”).
There were serious grounds to question the reliability of the anonymous informant in this case. She called during a domestic dispute with defendant, suggesting the possibility of an improper motive. During the pretrial hearing, the officer testified that he determined at the scene that Leni Briggs was the complainant. Leni Briggs testified at trial, and said she believed that Lila Dickinson made the call to the police. Ms. Briggs and Ms. Dickinson were *208friends and lived in the same house, apparently in separate apartments. When Ms. Dickinson testified at trial, she was never asked whether she called the police, but testified that she assumed defendant was waiting for his brother to drive him home.
Although the caller said defendant was leaving as she called, he was stopped within a few feet of the house twelve minutes later. In fact, the defense at trial was that defendant was stopped within the private driveway of the house he was visiting, and virtually all of the testimony went to that issue. There were inaccuracies in the caller’s description of the car.
The majority glosses over the discrepancies as unimportant because one of the women called the police and both had access to the information. In fact, the “ambiguity” in the testimony, as the majority calls it, points out why reliance on anonymous tips is dangerous. Leni Briggs denied making the call, and Lila Dickinson’s testimony was inconsistent with the contents of the call because she did not believe defendant was driving away. The testimony does not support the majority’s conclusion that one of the two women made the call. Because no one takes responsibility for the contents of the call, it is impossible to test its veracity.
Defendant raised the legality of the stop in a pretrial motion, as he was required to do. The motions judge denied the motion based on the theory that the caller reported that defendant was intoxicated and that the officer had once processed defendant for DUI. In reaching this conclusion, the motions judge stated “the court will take the testimony of the officer as true in weighing the evidence at hand.”
Defendant renewed his motion at trial, as he is required to do under State v. Senecal, 145 Vt. 554, 558, 497 A.2d 349, 351 (1985). The trial court refused to uphold the stop on the theory adopted by the motions judge and the majority here, holding that the complaint of a domestic dispute justified the stop. Although there was additional evidence bearing on the motion, the trial judge made no findings. The State has urged affirmance here on the theory that defendant was reported to have been involved in a domestic disturbance.
We considered a situation similar to this case in State v. Bruno, 157 Vt. 6, 595 A.2d 272 (1991), where the motions judge had found a DUI stop to be constitutionally lawful based on one theory, with supporting findings, and the trial judge found it lawful based on a different theory, supported by new findings inconsistent with those of the motions judge. We held that the action of the trial judge controlled and found this judge’s rationale to be valid. See id. at 9-11, 595 A.2d *209at 274-75. Justice Morse dissented, in part, because he believed there were no real findings supporting the trial judge’s decision. See id. at 13, 595 A.2d at 276.
While I disagreed with Justice Morse that there were inadequate trial judge findings in Bruno, I do not disagree that adequate findings are required. The trial judge decided to reconsider the pretrial motion decision, and there were discrepancies between the motion evidence and the trial evidence. The State’s theory was that Leni Briggs made the anonymous complaint, and that she and defendant had had the argument. The motions judge found that the caller reported that she and defendant had had an argument. Leni Briggs testified that she did not make the call. The evidence suggests that the content of the call may have been different from that found by the motions judge, and the basis of the caller’s knowledge was weaker. Because the trial judge adopted a different theory to determine the motion, there are no findings based on the full evidence before the court. At a minimum, we should remand for findings. It may be that the court will disbelieve Leni Briggs, or it will affirm the content of the tip as found by the motions judge, despite the fact that the caller was not the person in a domestic dispute with defendant. That determination should be made by the trial court in the first instance.
With the current emphasis on increasing DUI enforcement, an emphasis with which I agree, it is tempting to relax the standards police must follow to make vehicle stops based on suspicion that the operator is impaired. Where the information behind a stop comes from an anonymous tip and there is no significant, independent observation that the tip is reliable or that the person to be stopped is operating illegally, I would resist this temptation. We are greatly increasing the likelihood that dishonest tips will be effective with little improvement in DUI enforcement.
I would reverse. I am authorized to state that Justice Johnson joins in this dissent.
The findings of the motions judge indicate that she believed the caller gave her name, but the prosecution failed to offer any evidence of this. The State had the burden of production at the suppression hearing and later at the trial. See United States v. Longmire, 761 F.2d 411, 417 (7th Cir. 1985); 5 W LaPave, Search & Seizure § 11.2(b), at 38 (3d ed 1996). In the absence of evidence on the point, we must assume that the tip was anonymous. I can’t help believe, however, that this case is before us because the State never obtained evidence from the dispatcher who could have provided more information on the tip.