(dissenting, with whom Ireland, J., joins). Today the court holds that the police may use a drug-sniffing dog to make a public and targeted investigation of an automobile driver pulled over for a routine traffic violation (driving during the day time with only one working headlight), even though the police have no reason to suspect that he is involved in any illicit drug activity. Because the scope of the canine detection in this case was neither reasonable in the circumstances nor proportional to the investigating officer’s actual suspicions, as is required by our settled jurisprudence, I respectfully dissent.
It is black letter law that, to withstand a challenge to evidence *91seized during an investigatory stop, the Commonwealth must show that “it was permissible to initiate” the stop and that “the scope of the seizure was justified by the situation.” Commonwealth v. Williams, 422 Mass. 111, 116, 118 (1996), citing Terry v. Ohio, 392 U.S. 1, 19-20 (1968). I agree with the court that, in this case, the stop was legal. Ante at 75.
Our interpretation under art. 14 of the Massachusetts Declaration of Rights of the permissible “scope” of any ensuing investigation is equally settled. Any broadening in the scope of the investigatory stop must be “reasonable” and “proportional” in light of the “degree of suspicion the police reasonably harbor.” Commonwealth v. Williams, supra at 116, 117. See Commonwealth v. Sinforoso, 434 Mass. 320, 323 (2001) (“In evaluating whether the police exceeded the permissible scope of a stop, the issue is one of proportion”); Commonwealth v. Gonsalves, 429 Mass. 658, 662-663, 665 (1999) (reasonableness requirement in escalation of scope of traffic stops helps prevent arbitrary or harassing police conduct). I cannot agree with the court’s conclusion that the decision by the state trooper to summons a narcotics sniffing dog was “reasonable.” Ante at 80.
It is uncontested in this case that the investigating state trooper had no reason to suspect that the defendant was involved in any drug activity when he ordered the canine search. Rather, as the court candidly recognizes, the trooper had only a generalized suspicion that something might be afoot. What that something was, was “decidedly uncertain,” and “consistent with a number of possibilities.” Ante at 80. There is nothing in the record to indicate that the police “reasonably focuse[d] on the possibility of drug-related criminal activity.” Ante at 86. A motorist’s nervousness and false responses to police questions, without more, does not permit the police to conclude that illegal narcotics activity is afoot.1 In this case, as the Appeals Court *92noted, at the point he decided to summon a police dog trained “to detect marijuana, cocaine, hashish, and heroin,” the trooper had determined only that “something meriting further investigation was afoot” (emphasis added). Commonwealth v. Feyenord, 62 Mass. App. Ct. 200, 203 (2004).
I conclude that the use of a drug-detection dog is not “proportional” or “justified” during a routine traffic stop unless the police have a reasonable suspicion that the detained individual is engaged in unlawful drug-related activities. See Illinois v. Caballes, 543 U.S. 405, 418-422 (2005) (Ginsburg, J., dissenting) (use of drug-sniffing dog during traffic stop “broaden[s] the scope of the traffic-violation-related seizure” and therefore ought to require reasonable suspicion). This is not only required, as I shall explain, by our art. 14 jurisprudence, but also is consistent with the conclusions of numerous other State courts of last resort, examining the limits imposed by their respective State Constitutions on targeted canine drug sniffing. See, e.g., State v. Wiegand, 645 N.W.2d 125, 137 (Minn. 2002) (reasonable suspicion of drug activity required before initiation of canine sniff); State v. Pellicci, 133 N.H. 523, 534 (1990) (same); People v. Dunn, 77 N.Y.2d 19, 26 (1990), cert. denied, 501 U.S. 1219 (1991) (same); Commonwealth v. Johnston, 515 Pa. 454, 465-467 (1987) (same). See also State v. Tackitt, 315 Mont. 59, 69-70 (2003) (“particularized suspicion” of narcotics required to initiate sniff).
A canine drug sniff conducted during the course of a traffic *93stop is intrusive, embarrassing, and anxiety inducing.2 That “ ‘drug dogs are not lap dogs’ . . . changes the character” of an interaction between a police officer and a private citizen, which “becomes broader, more adversarial, and (in at least some cases) longer” when the police officer insists on probing the citizen with a drug sniffing dog. Illinois v. Caballes, supra at 421 (Ginsburg, J., dissenting), quoting United States v. Williams, 356 F.3d 1268, 1276 (10th Cir.) (McKay, J., dissenting), cert. denied, 543 U.S. 852 (2004). See 1 W.R. LaFave, Search and Seizure § 2.2(g), at 537 (4th ed. 2004) (canine search of persons “is embarrassing, overbearing and harassing, and thus should be subject to Fourth Amendment restraints”). Traffic stops occur in public places, and the motorist involved may often be known to others passing by. As Justice Greaney notes, a canine search “sends a clear public signal, not only to the detained motorist, but to all others passing by, that a drug investigation is in progress.” Ante at 89. See 4 W.R. LaFave, Search and Seizure § 9.3(f), at 401 (4th ed. 2004). I cannot agree with the court that detaining a motorist at the side of the road while the police use a dog to sniff his vehicle is merely “potentially embarrassing.” Ante at 80-81. Rather, I share Justice Greaney’s concerns that the use of drug-sniffing dogs, when used during a traffic stop, “is more than a de minimis intrusion into the detained individual’s privacy interests.” Ante at 89. For these reasons, the use of a narcotics detection dog is “intimidating” and “upsetting to the innocent motorist.” Y. Kamisar, W.R. LaFave, J.H. Israel, & N.J. King, Modem Criminal *94Procedure 430-431 n. (11th ed. 2005), quoting 4 W.R. LaFave, supra at § 9.3(f). No individual should be individually probed in this manner unless the police have reasonable suspicion that drug activity is afoot.* *3
In my judgment, that conclusion is compelled by our art. 14 jurisprudence. In Commonwealth v. Gonsalves, supra at 663, we held that a police officer must have “a reasonable belief that the officer’s safety, or the safety of others, is in danger before ordering a [person] out of a motor vehicle.” In other words, even if a police officer has a legitimate basis for extending a traffic stop (in Commonwealth v. Gonsalves, supra at 660, the police suspected some illegal activity because the defendant passenger was “extremely nervous” and “appeared to be breathing heavily”), the officer cannot order a person out of a vehicle unless the officer has a reasonable belief that removal from the vehicle is required for reasons of safety. We explained in that case that our “rule” is rooted in art. 14’s requirement that the police respect the privacy of motorists, avoid unnecessary delay and inconvenience in making traffic stops, id. at 663-664, and act at all times “in a reasoned way.” Id. at 665.4 If our Constitution requires a “reasonable” fear for safety to justify ordering a motorist out of an automobile during a traffic stop, surely it requires no less before a canine drug sniff can be initiated against an individual stopped for a routine traffic violation: *95ordering a person from an automobile typically is less intrusive than conducting a canine narcotics sniff. A particularized “reasonable suspicion” requirement for canine drug detection during traffic stops prevents baseless harassment of citizens, without erecting any barrier to legitimate law enforcement activities. See Commonwealth v. Gonsalves, supra at 662-664.
The targeted use of a narcotics detection dog against a particular individual stopped for a routine traffic stop is entirely distinguishable from properly conducted police efforts to ensure public safety. The use of specially trained dogs in airports, schools, subway stations, or other such public arenas gives rise to issues not present in this case. See, e.g., Illinois v. Caballes, supra at 424-425 (Ginsburg, J., dissenting) (distinguishing search and seizure standards involving “general interest in crime control” from those involving “more immediate threats to public safety”), and cases cited. Reasonable canine detection can be utilized for public safety purposes without violating our carefully developed art. 14 jurisprudence.
The court’s conclusion that canine narcotics detection is permissible in this case means necessarily that drug-sniffing dogs may be called in any time the police sense that the subject of a routine traffic stop is engaged in “some” unspecified wrongdoing. In so holding, the court departs (without reason or explanation) from our settled jurisprudence and deprives our citizens of meaningful protection from an investigatory tactic widely recognized as intrusive and “viscerally disturbing.” Ante at 89.1 respectfully dissent.
Significantly, the judge did not find that the State trooper had reason to suspect the defendants of illicit narcotics activity, and the Commonwealth has never made such a claim. The record confirms that there was nothing about the actions or responses of the defendant that suggested illegal narcotics activity. The trooper did not testify to an odor of drugs, that the defendant was weaving while driving, that his eyes were glassy, or that his speech was abnormal. The trooper who pulled the defendant’s automobile over was in fact *92trained in narcotics interdiction. He did not testify that he suspected that the defendant was engaged in any illicit drug activity when he decided to summons the narcotics-detection dog. He testified only that, at the time he summoned the dog, he had ascertained that the defendant may have been driving without a license, that he had given a false name and appeared nervous, and that there were inconsistencies between the defendant’s responses to routine questions and the statements of the passenger in the automobile. See Commonwealth v. Feyenord, 62 Mass. App. Ct. 200, 211 (2004) (Greenberg, J., dissenting). The officer testified that he was “suspicious” of the defendant and the passenger, but not that he had a particularized suspicion that either of the two men possessed drugs. See State v. Wiegand, 645 N.W.2d 125, 128-129, 137 (Minn. 2002) (no reasonable suspicion sufficient to conduct canine drug sniff where individual stopped for inoperable headlight appeared nervous, had glossy eyes, and denied consent to search car, and officer was suspicious but did not suspect that individual was under influence of drugs).
The court’s opinion pays little, if any, attention to the issue whether targeted canine narcotics detection itself broadens the scope of any investigatory stop. Rather, the court focuses, unnecessarily in my view, on the length of the defendant’s detention at the side of the road. It is settled that, had they completed their legitimate investigation, the police could not (without reasonable suspicion) have detained the defendant further to await the dog’s arrival. Illinois v. Caballes, 543 U.S. 405, 408 (2005), citing People v. Cox, 202 Ill. 2d 462 (2002), cert. denied, 539 U.S. 937 (2003). Today’s holding creates troublesome incentives because “the stopping officer is tempted to engage in stalling regarding his proper function during the stop ... in order to give the appearance that the time for the stop has not expired in the interim.” Y. Kamisar, W.R. LaFave, J.H. Israel, & N.J. King, Modem Criminal Procedure 430-431 n. (11th ed. 2005), quoting 4 W.R. LaFave, Search and Seizure LaFave § 9.3(f) (4th ed. 2004).
Well-trained dogs often “alert” to innocent people. Illinois v. Caballes, 543 U.S. 405, 411-412 (2005) (Sauter, J., dissenting). In one case, school officials initiated a canine sniff of their teenage students. The dogs “alerted” to fifty students, thirty-three of whom later were revealed to have been innocent. A dog alerted to one junior high school girl. She was asked to empty her pockets, which contained no drugs. She was then strip searched, but, again, no drugs were found. Later, it was revealed that the girl had been playing that morning with her own dog, who was in heat. Doe v. Renfrow, 475 F. Supp. 1012 (N.D. Ind. 1979), afPd in part, 631 F.2d 91 (7th Cir. 1980), cert. denied, 451 U.S. 1022 (1981). See 1 W.R. LaFave, Search and Seizure § 2.2(g), at 532-534 (4th ed. 2004). Rare as such occurrences may be, they evidence the potential of canine narcotics detection to create a serious and unwarranted intrusion into a person’s life.
We reached our conclusion despite decisions of the United States Supreme Court to the contrary. See, e.g., Pennsylvania v. Mimms, 434 U.S. 106, 111 (1977) (on routine traffic stop police officer may, without additional basis, order driver from vehicle); Maryland v. Wilson, 519 U.S. 408, 415 (1997) (same, regarding passenger).