(dissenting with whom Liacos, J., joins). The court properly acknowledges that the stopping of each defendant’s motor vehicle was a seizure under the Fourth and Fourteenth Amendments to the United States Constitution. Ante at 86. See Delaware v. Prouse, 440 U.S. 648, 653-654 (1979). The court also properly states that the Commonwealth bears the burden of proof on the issue of lawfulness of the seizure. Ante- at 90. See Commonwealth v. McGeoghegan, 389 Mass. 137, 144 (1983). The court has failed, however, to appreciate the full implications of this burden. The Sunderland roadblock resulted in an intrusion into the right of several hundred people to be free from arbitrary interference by law enforcement officials. If such an intrusion can ever be said to be reasonable in a constitutional sense, it can only be after the Commonwealth has met its burden of showing that “[sjuch a procedure [achieves] a degree of law enforcement and highway safety that is not reasonably attainable by less intrusive means.” Commonwealth v. McGeoghegan, supra at 143-144. The court apparently assumes that roadblocks are a reasonable mechanism for advancing the public interest in curbing drunk driving, and examines merely whether the guidelines at issue and their application in this case are reasonable. Since I am unable to find any basis in the record for concluding that roadblocks are *99more effective than less intrusive mechanisms for curbing drunk driving,1 I respectfully dissent.
“No method of spot checking motor vehicles to discover drunk drivers, in the absence of probable cause or articulable suspicion, has been held by the Supreme Court of the United States to be constitutionally permissible.” Commonwealth v. McGeoghegan, supra at 139. That Court has repeatedly emphasized that a factor to be weighed in determining whether a roadblock is constitutional is “the absence of practical alternatives” for protecting the public interest. See, e.g., Delaware v. Prouse, supra at 655, quoting United States v. Brignoni-Ponce, 422 U.S. 873, 881 (1975). Even when a vital State interest is implicated, “[t]he question remains . . . whether in the service of these important ends the [roadblock] is a sufficiently productive mechanism to justify the intrusion upon Fourth Amendment interests which such stops entail.” Delaware v. Prouse, supra at 659. There, the Court concluded, “Given the alternative mechanisms available, both those in use and those that might be adopted, we are unconvinced that the incremental contribution to highway safety of the random spot check justifies the practice under the Fourth Amendment.” Id. The Court stated, furthermore, that “[t]he foremost method of enforcing traffic and vehicle safety regulations ... is acting upon observed violations.” Id.
Nor has this court previously decided that the Commonwealth had met its burden to show the reasonableness of a roadblock conducted under the circumstances of the one in question here. Nevertheless, the court relieves the Commonwealth of its burden to show that the roadblock was more effective in advancing the public interest in curbing drunk driving than alternative, less intrusive mechanisms. Other courts have recognized that the State must make this showing, *100and have struck down roadblocks partly on this basis. See, e.g., State v. Koppel, 127 N.H. 286 (1985);2 Jones v. State, 459 So. 2d 1068, 1077 (Fla. Dist. Ct. App. 1984); State v. McLaughlin, 471 N.E.2d 1125, 1141 (Ind. Ct. App. 1984); People v. Bartley, 125 Ill. App. 3d 575 (1984); State ex rel. Ekstrom v. Justice Court, 136 Ariz. 1, 5 (1983). Cf. State v. Superior Court, 143 Ariz. 45, 48-49 (1984) (upholding roadblock where State official offered evidence that there were no less intrusive alternatives). See also State v. Deskins, 234 Kan. 529, 544-545 (1983) (Prager, J., dissenting) (“The most pressing question before us is the degree to which this roadblock checkpoint actually promoted the public interest in deterring drunk drivers. ... In my judgment, the trial court correctly concluded that the State failed in its burden of proof in establishing that the roadblock checkpoint promoted the public interest in light of available less drastic alternative measures which could have been used by the officers to combat the problem, without setting up a roadblock and stopping between 2,000 and 3,000 motorists.”)
The record in this case clearly raises the question whether the Sunderland roadblock of July 2-3, 1983, was as effective in curbing drunk driving as traditional, less intrusive alternatives. During approximately two and one-half hours, State police officers stopped 503 vehicles. Although the roadblock claimed the services of thirteen officers and caused an intrusion into the Fourth Amendment rights of several hundred law-abiding citizens, it resulted in only eight arrests. There is no showing in the record that the roadblock was more effective than less intrusive mechanisms would have been.3
*101The Commonwealth has argued that, even though fewer than two per cent of the drivers stopped at the Sunderland roadblock were ultimately arrested, the roadblock program •nevertheless serves as a deterrent and has contributed to a decline in highway fatalities. As an appendix to its brief, the Commonwealth has presented a graph showing a decline in highway fatalities during the period July 1 through Labor Day weekend for the years 1981-1983. Absent a clear showing of a causal relationship, the Commonwealth cannot claim this decline resulted solely from the use of roadblocks rather than from increased public and media attention to the problem, as well as more emphasis on the more traditional, less intrusive means of enforcement. The decline in highway fatalities in 1982 and 1983 could also reflect a change in public perception of drunk driving due to the imposition of stricter criminal penalties for driving under the influence.4 The Commonwealth has failed to demonstrate that the decline in highway fatalities is attributable to institution of a roadblock program, and has also not shown that the same or greater declines could not have been achieved by use of less intrusive mechanisms.5
The court’s holding that the seizures at the Sunderland roadblocks were reasonable contrasts sharply with other decisions of this court concerning the scope of protection afforded by the Fourth Amendment. See, e.g., Commonwealth v. Piet-rass, 392 Mass. 892, 898-900 (1984) (even though there was probable cause to arrest the defendant on charges of aggravated rape and other violent crimes, the Commonwealth failed to meet its burden of showing exigency, so the police were not justified in making a warrantless entry of a dwelling to arrest *102him); Commonwealth v. Thibeau, 384 Mass. 762, 763-764 (1981) (even though defendant was riding a bicycle, a known means of transporting illegal drugs, and made a sudden left turn as police cruiser approached, police lacked probable cause to arrest him and seize envelopes containing narcotics that were sticking out of his pocket); Commonwealth v. Bacon, 381 Mass. 642, 645-646 (1980) (police were not justified in making investigatory stop of relatively expensive motor vehicle operated by two youthful appearing men, even though one attempted to conceal his face, presumably from police view). I am unable to reconcile the court’s ruling that the Commonwealth has sustained its burden in this case with the traditional concern this court has expressed for protecting Fourth Amendment rights. Here, the court concludes that the warrantless stop of 503 vehicles was justified even though the police lacked a basis for suspecting that any particular operator was or had been engaged in criminal activity. The court gives no satisfactory explanation for why it has suddenly lowered the threshold showing which the Commonwealth has traditionally been required to make to justify a warrantless stop or arrest. I would answer “no” to question 2 and decline to answer questions 1 and 3.
At least one objective study supports the contrary proposition and demonstrates that a roadblock program was an ineffective technique by which to deter and detect drunk driving. See Little v. State, 300 Md. 485, 516-517 (1984) (Davidson, J., dissenting). It is not enough to say, as does the court, that this issue is not raised by the reported questions because without such ■a showing the roadblock could not pass constitutional muster.
The court’s holding, however, rested solely on an interpretation of the New Hampshire Constitution. Although the language of Part I, article 19, the relevant section of the New Hampshire Constitution, is “similar to that in the Federal Constitution,” article 19 was construed to provide greater protection for individual rights than the Fourth Amendment. State v. Koppel, supra at 289.
The police could have patrolled the roadways and stopped any motor vehicle where the driver’s conduct raised a reasonable suspicion of intoxication. See Terry v. Ohio, 392 U.S. 1 (1968). Also, the police could have conducted a surveillance of local bars and restaurants.
In 1982, the Legislature amended G. L. c. 90, § 24 (1) (a) (1), to impose stricter penalties for operating a motor vehicle while under the influence of intoxicating liquor. The minimum fine was increased from $35 to $100. Furthermore, first offenders may now be imprisoned for up to two years. See G. L. c. 90, § 24 (1) (a) (1) (1984 ed.).
Although I do not reach the question, I have some doubts concerning the court’s conclusion that the promulgation of the roadblock guidelines need not have been in compliance with the requirements for adopting regulations under G. L. c. 30A, §§ 2 & 3 (1984 ed.).