MICHIGAN DEPARTMENT OF STATE POLICE, ET AL., PETITIONERS V. RICK SITZ, ET AL. No. 88-1897 In The Supreme Court Of The United States October Term, 1989 On Writ Of Certiorari To The Michigan Court Of Appeals Brief For The United States As Amicus Curiae Supporting Petitioners TABLE OF CONTENTS Question Presented Interest of the United States Statement a. Site selection b. Notice to the public c. Safety and evidence of authority d. Contact with motorists Summary of argument Argument: I. Brief nondiscretionary stops of motorists at sobriety checkpoints are a constitutionally permissible means of reducing drunk and drugged driving A. Sobriety checkpoints are effective, together with other measures, in reducing drunk driving B. Sobriety checkpoints intrude only minimally on Fourth Amendment interests C. With respect to sobriety checkpoints, the balance of relevant interests justifies brief stops without individualized suspicion Conclusion QUESTION PRESENTED Whether sobriety checkpoints, at which all vehicles are stopped briefly to enable police officers to observe for signs that the drivers are under the influence of alcohol or drugs, are unlawful under the Fourth Amendment. INTEREST OF THE UNITED STATES This case presents the question whether sobriety checkpoints, at which all motorists are stopped and examined briefly for signs that they are driving under the influence of alcohol or drugs, are unlawful under the Fourth Amendment. The Department of Transportation, acting through the National Highway Traffic Safety Administration (NHTSA), plays a major role in efforts to promote highway safety and, more partiularly, to reduce the incidence of drunk driving. NHTSA's studies have shown that sobriety checkpoints are an effective part of a comprehensive program to combat drinking and driving. Accordingly, it has encouraged States to implement -- and has provided substantial technical assistance and funding for -- sobriety checkpoint programs similar to the one at issue in this case. Under the Highway Safety Act, 23 U.S.C. 408(a) (1982 & Supp. V 1987), NHTSA is charged with administering a grant program that provides funding to States "which adopt and implement effective programs to reduce traffic safety problems resulting from persons driving while under the influence of alcohol or a controlled substance." States qualify for basic and supplemental grants by demonstrating that they have adopted measures prescribed by the statute (23 U.S.C. 408(e)-(f)) and regulations promulgated by NHTSA (23 C.F.R. 1309.5-1309.7). One of the measures that States may adopt to qualify for supplemental grants under this program is the "(u)se of roadside sobriety checks as part of a comprehensive alcohol safety enforcement program." 23 C.F.R. 1309.6(b)(11). /1/ In addition, funds provided by NHTSA to the States under Section 402 of the Highway Safety Act, 23 U.S.C. 402 (1982 & Supp. V 1987), have been used to finance sobriety checkpoint programs. /2/ In addition to encouraging and funding the use of sobriety checkpoints by local law enforcement agencies, the federal government has used sobriety checkpoints on military and other federal reservations, and has an interest in continuing to do so. STATEMENT 1. In 1985, Michigan's Drunk Driving Task Force issued a report recommending, among other things, the use of sobriety checkpoints. The Governor of Michigan then directed the State Police to implement a pilot program, and an advisory committee appointed by the Director of State Police prepared detailed guidelines with respect to checkpoint site selection, publicity, and operations. Pet. App. 3a-5a. See id. at 133a-164a. According to the guidelines, the purpose of sobriety checkpoints was "to impact the drunk driving problem by increasing the perception that the drunk driver will be detected and arrested." Pet. App. 144a. Checkpoints were to be conducted in accordance with the following directives: a. Site selection. -- Criteria for checkpoint site selection were to be approved by the Director of State Police and were to take account of the locations of past arrests and accidents. Pet. App. 148a-149a. b. Notice to the public. -- Although the exact time and location of checkpoints were not to be disclosed in advance, the fact of their use was to "be widely publicized to enhance the desired deterrent effect among potential drunk drivers." Pet. App. 152a. c. Safety and evidence of authority. -- Each checkpoint was to be staffed by a specified number of uniformed police officers. The checkpoints were to be established in safe locations, lit, and marked with signs, flares, reflectors, and/or cones. Pet. App. 149a-150a, 152a-153a. d. Contact with motorists. -- Each vehicle passing through a checkpoint was to be stopped. After identifying himself and explaining the purpose of the checkpoint, an officer was to look for "articulable indications of intoxication," including odor, slurred speech, lack of physical coordination, unusual eye movements, and disorientation. If the officer detected an indication of intoxication, the driver was to be directed to a location out of the path of traffic, asked for a driver's license and registration, and asked to perform field sobriety tests, a breath test, or both. Upon sufficient evidence of intoxication, the driver was to be arrested. Pet. App. 154a-156a, 161a-163a. /3/ 2. Between 11:45 p.m. and 1:00 a.m. on May 17-18, 1986, one checkpoint operation was conducted in accordance with the guidelines. One hundred twenty-six vehicles passed through the checkpoint, with an average delay to each motorist of 25 seconds or less. Field sobriety tests were administered to two drivers who stopped at the checkpoint; one was arrested for driving under the influence of alcohol. Another driver was arrested for driving under the influence after he drove through the checkpoint without stopping. Pet. App. 6a-7a. 3. The day before the first checkpoint was established, respondents initiated an action for a declaratory judgment and an injunction. Their complaint alleged that the checkpoint program violated the Fourth Amendment of the United States Constitution and Article 1, Section 11 of the Michigan Constitution; that the program had been promulgated in violation of the Michigan Administrative Procedure Act, Mich. Comp. Law. Ann. Section 24.207 (West 1981); and that the State lacked authority under state law to conduct the program. Pet. App. 29a-30a. /4/ After hearing evidence, the state trial court held that the sobriety checkpoint program violated both the Fourth Amendment and the state constitution. The court ruled that the program's legality should be determined by reference to the factors enumerated by this Court in Brown v. Texas, 443 U.S. 47, 50-51 (1979); those factors are "the gravity of the public concerns served by the (program), the degree to which the (program) advances the public interest, and the severity of the interference with individual liberty." See Pet. App. 57a-58a, 72a. a. The court observed that it was "largely conceded by all that the State has a keen interest in curbing drunk driving." Pet. App. 73a. Indeed, although noting that witnesses had differed on precisely how many persons are killed or injured by drunk drivers, the trial court concluded that "there is a grave and legitimate State interest in curbing drunk driving." Id. at 74a. b. The court then addressed the extent to which sobriety checkpoints advance that interest -- an inquiry which it understood to encompass "not only * * * how effective the technique is, but also * * * the degree to which it can be said that there are no permissible alternative means available to the police which advance the particular state concern." Pet. App. 75a. Relying on evidence and authorities suggesting that the number of motorists arrested at sobriety checkpoints is a small percentage of the drivers who pass through them, the court concluded that "sobriety checkpoints have not been shown to be an effective means for apprehending drunk drivers." Id. at 83a. The state court turned next to the question whether checkpoints would act as a deterrent to drunk driving, and found that the State's showing on that point was "not of a sufficient persuasive value to support the conclusion that checkpoints have a deterrent effect." Pet. App. 92a. The court credited testimony by respondents' expert that checkpoints would not have "any long term, or lasting deterrent effect." Id. at 92a-94a. Invoking its finding that sobriety checkpoints are not an effective means of apprehending drunk drivers, the court reasoned that once the public became aware of the low arrest rate, "it cannot be reasonably supposed that those who are inclined to drink and drive will perceive a sobriety checkpoint as a significant threat to their being arrested." Id. at 95a-96a. Finally, the court "examine(d) the extent to which there (were) no practical alternative means available to the State to advance its interest other than by implementing sobriety checkpoints." Pet. App. 96a. The court dismissed evidence that routine methods of law enforcement had not reduced the level of alcohol-related accidents, finding that "this data does not demonstrate that checkpoints are the only practical alternative." Id. at 97a. The court also referred to evidence that other programs had reduced alcohol-related accidents; in addition, the court expressed concern that other recommendations by Michigan's Drunk Driving Task Force had not been implemented. Id. at 98a-101a. The trial court ruled "that while drunk driving remains a grave problem, other programs which rely on approved methods of law enforcement can, and do, in fact, have a positive effect in reducing drunk driving accidents and fatalities." Id. at 101a. c. With respect to the intrusiveness of sobriety checkpoints, the court found that the average motorist's encounter with police at a sobriety checkpoint lasts approximately 30 seconds and that "the intensity of the investigation" during that period is "not great." Pet. App. 103a-104a. Accordingly, the court found that the "objective nature of the intrusion (is) minimal." Id. at 104a. The court also concluded that the State's guidelines would "ensure that on-the-scene officers' discretion relative to the operation of sobriety checkpoints will be minimal." Id. at 113a. However, the court found that there were other factors that could contribute to the intrusiveness of the procedure -- the fact that checkpoints would not be situated at fixed locations, the possibility that news media might not publicize plans to establish checkpoints, motorists' likely unawareness of opportunities to avoid checkpoints, and the fact that checkpoints would be used to search for violations of criminal law. Id. at 108a-110a. The court also expressed the view that sobriety checkpoints "(went) beyond the traditionally recognized reasons for stopping motorists in their cars," and suggested that checkpoints were susceptible to abuse. Id. at 115a. On balance, the court concluded that the checkpoints would have "a significant subjective intrusive impact on a person's liberty interest." Id. at 119a. Weighing the Brown factors, the trial court ruled that the checkpoint program violated the Fourth Amendment. See Pet. App. 119a-124a. The court also construed the State's constitution to forbid any seizure of motorists "absent at least an individualized articulable suspicion that some criminal wrongdoing had or was about to occur." Id. at 130a-131a. 4. The Michigan Court of Appeals affirmed with respect to the trial court's Fourth Amendment holding. It agreed that the legality of the sobriety checkpoint plan should be determined by reference to the three factors set out in Brown v. Texas, supra. Pet. App. 12a. Applying the "clearly erroneous" standard to the trial court's rulings on each of those factors, the state appellate court concluded that there was no basis for disturbing the trial court's determinations. Id. at 12a, 13a-22a. The court of appeals held that "(w)hile the goals of the sobriety checkpoint program are laudable, the program fails to qualify as a reasonable seizure under the Fourth Amendment." Id. at 23a. /5/ The Michigan Supreme Court, noting two dissents, denied petitioners' application for leave to appeal to that court. Pet. App. 1a. SUMMARY OF ARGUMENT In 1988, more than 47,000 Americans were killed in traffic crashes. Twenty-five percent of the drivers involved were intoxicated, and alcohol was involved in some form in about half of those fatalities. In an effort to reduce this carnage, a number of States, with encouragement and funding from the National Highway Traffic Safety Administration, have implemented sobriety checkpoint programs. The program at issue in this case has the features of such federally encouraged efforts. Every motorist who reaches a well-lit, well-marked checkpoint is observed, for about 30 seconds, for signs of intoxication. Drivers who do not exhibit signs of intoxication are allowed to leave immediately; the others are asked to take a breath test, field sobriety tests, or both. If the tests provide probable cause to believe that a driver is intoxicated, he or she is arrested. In our view, the state courts erred in concluding that these procedures subject motorists to seizures that are "unreasonable" under the Fourth Amendment. Because this case involves a facial challenge to the checkpoint program, as opposed to a claim that an individual motorist was detained without sufficient cause, we focus on the brief, nondiscretionary stops to which all motorists entering a checkpoint are exposed. The Court's cases make clear that the legality of those stops "is judged by balancing (their) intrusion on the individual's Fourth Amendment interests against (their) promotion of legitimate governmental interests." Delaware v. Prouse, 440 U.S. 648, 654 (1979). In view of the effectiveness and minimal intrusiveness of sobriety checkpoints, they are reasonable means of combatting drunk driving; this is an instance "in which the balance of interests precludes insistence upon 'some quantum of individualized suspicion.'" Id. at 654-655. A. There can be no dispute over the seriousness of the drunk driving problem in this country. In terms of deaths, injuries, economic loss, and human suffering, the problem is of grave dimensions. For years, drunk driving has been stubbornly resistant to a variety of counter-measures. However, experience here and abroad now suggests that sobriety checkpoints provide a promising means of deterring drinking and driving. By virtue of their visibility, sobriety checkpoints contribute to the perception that people who drink and drive will be apprehended. Where sobriety checkpoints have been used, there have frequently been immediate reductions in alcohol-related crashes. Moreover, there is evidence that those reductions can be sustained for substantial periods of time. B. On the other side of the balance, a number of circumstances, taken together, assure that sobriety checkpoints intrude only minimally on Fourth Amendment interests. First, persons driving on public roads have a limited expectation of privacy; that expectation is particularly narrow with respect to measures whose purpose is to enforce traffic safety laws. Traffic stops for that purpose have a long history and have come to be accepted by motorists as incident to highway use. Second, each driver is stopped only briefly, along with other motorists, at a well-marked and well-lit checkpoint. Finally, the discretion of officers in the field is narrowly circumscribed. C. Under these circumstances, no individualized suspicion is required to make sobriety checkpoint stops "reasonable" for purposes of the Fourth Amendment. The effectiveness of sobriety checkpoints derives from their universal, predictable, and visible applicability to all motorists. The resulting intrusion on individual interests is minimal. And, in place of individualized suspicion, rigorous limits on police discretion protect individuals from invidious or arbitrary official action. ARGUMENT I. BRIEF NONDISCRETIONARY STOPS OF MOTORISTS AT SOBRIETY CHECKPOINTS ARE A CONSTITUTIONALLY PERMISSIBLE MEANS OF REDUCING DRUNK AND DRUGGED DRIVING The legal framework to guide the Fourth Amendment analysis in this case is well established. We assume, for purposes of this case, that the stops occurring at sobriety checkpoints constitute "seizures" within the meaning of the Fourth Amendment. /6/ The fundamental question under the Fourth Amendment is whether those "seizures" are reasonable. Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. 1402, 1414 (1989) ("the Fourth Amendment does not proscribe all searches and seizures, but only those that are unreasonable"); United States v. Sharpe, 470 U.S. 675, 682 (1985). What is reasonable "depends on all the circumstances surrounding the search or seizure itself." United States v. Montoya de Hernandez, 473 U.S. 531, 537 (1985). See Wyman v. James, 400 U.S. 309, 318 (1971); New Jersey v. T.L.O., 469 U.S. 325, 340 (1985). The permissibility of a program such as sobriety checkpoints "'is judged by balancing its intrusion on the individual's Fourth Amendment interests against its promotion of legitimate governmental interests,'" Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1414 (quoting Delaware v. Prouse, 440 U.S. at 654). Accord, e.g., Brown v. Texas, 443 U.S. at 50-51; United States v. Martinez-Fuerte, 428 U.S. 543, 555 (1976); United States v. Brignoni-Ponce, 422 U.S. 873, 878 (1975). The reasonableness standard usually requires that "the facts upon which an intrusion is based be capable of measurement against 'an objective standard,' whether this be probable cause or a less stringent test." Delaware v. Prouse, 440 U.S. at 654. See Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1414. The Court has recognized a number of situations, however, "in which the balance of interests precludes insistence upon 'some quantum of individualized suspicion.'" Delaware v. Prouse, 440 U.S. at 654-655. "In limited circumstances, where the privacy interests implicated by the search are minimal, and where an important governmental interest furthered by the intrusion would be placed in jeopardy by a requirement of individualized suspicion, a search may be reasonable despite the absence of such suspicion." Skinner v. Railway Labor Executives Ct. at 1417. See United States v. Martinez-Fuerte, 428 U.S. at 560-561. In those situations, "other safeguards are generally relied upon to assure that the individual's reasonable expectation of privacy is not 'subject to the discretion of the official in the field.'" Delaware v. Prouse, 440 U.S. at 655, 661. See, e.g., Donovan v. Dewey, 452 U.S. 594, 601-604 (1981); Marshall v. Barlow's, Inc., 436 U.S. 307, 320-321 (1978). Under these standards, sobriety checkpoints are reasonable and thus pass constitutional muster. A. Sobriety Checkpoints Are Effective, Together With Other Measures, In Reducing Drunk Driving 1. By many measures, drunk driving is among the most critical threats to the health and well-being of our Nation's citizens. /7/ According to statistics compiled by NHTSA, more than 47,000 people were killed in traffic crashes in 1988. NHTSA, Alcohol Involvement in Fatal Traffic Crashes 1988, at 5 (DOT HS 807 448 July 1989) (hereinafter Fatal Traffic Crashes). Almost 25% of the drivers involved in these crashes had blood alcohol concentrations in excess of 0.10%, the legal standard for intoxication in most States. Id. at 11. /8/ Another 7.9% had a lower level of alcohol in their blood. Ibid. Alcohol was involved in some form in nearly half -- more than 23,000 -- of the fatalities that resulted from traffic crashes. Id. at 5. /9/ In addition to the fatalities, each year approximately 500,000 persons sustain injuries, 40,000 of them serious, in alcohol-related crashes. NHTSA, Drunk Driving Facts 1 (July 1989). Two of every five Americans can expect to be in an alcohol-related crash at some time in their lives. Id. at 3. The impact on the young has been particularly severe. Traffic crashes are the leading single cause of death for every age between 5 and 34; more than half of those fatalities result from alcohol-related crashes. Drunk Driving Facts 1 (July 1989). The costs in terms of pain and suffering imposed on the survivors and caretakers of the spouses, children, parents, and friends who are killed or permanently injured in alcohol-related crashes are simply incalculable. 2. In the past decade, federal studies have uniformly concluded that sobriety checkpoints can play an important role in reducing the incidence of drunk driving. In February 1983, NHTSA promulgated regulations establishing, as one criterion for federal funding under Section 408 of the Federal Highway Act, the "(u)se of roadside sobriety checks as part of a comprehensive alcohol safety enforcement program." /10/ Later the same year, the President's Commission on Drunk Driving recommended "the use of preliminary breath testing devices and judicially approved roadblocks, to achieve a high perception of risk of detection for driving under the influence." Presidential Commission on Drunk Driving, Final Report 14-15 (Nov. 1983). In 1984, the National Transportation Safety Board, an independent agency, concluded that sobriety checkpoints and administrative license revocations were "potentially effective deterrent measures that warrant broader application by the States" and that those measures "should be an integral part of a State's comprehensive alcohol and highway safety program." NTSB, Safety Study: Deterrence of Drunk Driving: The Role of Sobriety Checkpoints and Administrative License Revocations 20 (NTSB/SS-84/01 Apr. 1984). /11/ Recently, during the Surgeon General's Workshop on Drunk Driving, a panel on law enforcement recommended "(i)mplement(ing) DUI checkpoints in those jurisdictions not using them, and expand(ing) their use in jurisdictions using them." Surgeon General's Workshop on Drunk Driving, Proceedings 53 (1988). As demonstrated below, these conclusions are consistent with -- and have been based upon -- a substantial and growing body of research demonstrating that sobriety checkpoints can contribute to reductions in drunk driving. 3. The pervasiveness of drinking and driving in this country has made that problem resistant to traditional law enforcement techniques. During weekend nights, approximately 3% of all drivers on the road will have blood alcohol concentrations exceeding the legal limit for intoxication. See note 8, supra. /12/ As a practical matter, it is impossible for police, with limited manpower, to arrest a substantial number of those individuals and remove them from the road. Even in high enforcement areas, only about 1 in 250 such drivers will be apprehended on any given night. J. Hause, R. Voas & E. Chavez, Conducting Voluntary Roadside Surveys: The Stockton Experience 12 (1980) (available from NHTSA, NTS-20). Indeed, of all drivers involved in fatal alcohol-related crashes in 1988, only 13.3% had ever previously been convicted for driving under the influence. NHTSA, Fatal Accident Reporting System: Alcohol Involvement in Fatal Crashes: Recent Trends: 1982-1988, at 13 (available from NHTSA, NTS-20). From these figures, it is clear that any law enforcement program to reduce drunk driving cannot rely primarily on apprehending those who drink and drive. By necessity, such programs must emphasize general deterrence. Notwithstanding sustained efforts by governments and citizen groups to communicate to the public the danger of drinking and driving, the risk of death or serious injury is not sufficiently immediate to motivate many drinking drivers to change their behavior. Research suggests, however, that some of those individuals will react to a threat that they may perceive as more immediate -- the risk that they will be caught and subjected to reasonably swift and certain sanctions. See H. Ross, Deterring the Drinking Driver 110 (1984) ("The accumulated knowledge in this area indicates that an important element of a deterrence-based program is the presentation to drinking drivers of a subjectively important chance of apprehension should they commit the violation."). Public officials have therefore attempted to deploy their limited resources -- resources that permit apprehension of only a small percentage of drinking drivers -- to maximize the public's perception of that risk. Research supports the proposition that sobriety checkpoints have been effective for that purpose. Several studies reflect that the public perceives that an intoxicated driver is more likely to be arrested in an area in which checkpoints are employed than in otherwise similar areas in which checkpoints are not used. For instance, after police in Charlottesville, Virginia, employed sobriety checkpoints for almost a year, a survey was conducted of residents of that city. More than half reported that a drinking driver was more likely to be arrested in Charlottesville than in other cities in Virginia; more than 70% said that the likelihood of arrest was greater at the time of the survey than before checkpoints were put into use. A substantial number of the drivers surveyed reported that checkpoints were the reason why arrest had become more likely. R. Voas, E. Rhodenizer & C. Lynn, Evaluation of Charlottesville Checkpoint Operations 27-28 (DOT HS 806 989 May 1985). In a control city where checkpoints were not used, comparable figures were lower. /13/ There are several plausible explanations for this phenomenon. Many more drivers on the road during high-risk driving hours are exposed to checkpoints than any other enforcement procedure. Checkpoints thus alert the public to the risk of their being detected if they drink and drive. When they have been employed, checkpoints have typically been the focus of intense media attention. Drivers may believe that, while they can elude a roving patrol by driving "carefully," they are less likely to avoid detection in a face-to-face encounter with a police officer. Finally, checkpoints may produce as many actual arrests, or more, than traditional patrols, measured in terms of officer-hours. /14/ Of course, the purpose of sobriety checkpoints -- and the critical yardstick by which any drunk driving program should ultimately be judged -- is the reduction of alcohol-related crashes. On this point, studies in this country reflect that the systematic use of sobriety checkpoints results in reductions in alcohol-related crashes. In Charlottesville, during the year that sobriety checkpoints were used, the number of accidents in which an officer reported that the driver had been drinking declined by 15% from the level of the prior two years. Charlottesville's share of all traffic crashes statewide in which an officer reported that the driver had been drinking also declined, an indication that it was achieving better results in reducing alcohol-related crashes than areas of the Commonwealth where checkpoints were not being used. R. Voas, E. Rhodenizer & C. Lynn, supra, at 44; see id. at 46 (figures consistent with 10% overall reduction in alcohol-related crashes). Similar results were achieved in other States. /15/ Notwithstanding the evidence that checkpoints produce immediate reductions in alcohol-related crashes, the trial court credited the testimony of respondents' expert that checkpoints do not have a long term, deterrent effect. Pet. App. 92a-93a. However, in our view, the indisputable savings in lives and well-being that accompany even temporary reductions in alcohol-related crashes provide a sufficient basis for the discriminating use of sobriety checkpoints, since those reductions involve many lives that would otherwise have been lost. Further, even if the trial court's approach were appropriate, studies of roadside checkpoints in Bergen County, New Jersey, and in Australia demonstrate that sobriety checkpoints have contributed to substantial long-term reductions in alcohol-related crashes. In Bergen County, sobriety checkpoints led to a 14% to 24% reduction in single-vehicle nighttime crashes, a representative measure of alcohol-related crashes, for the first nineteen months of their use, from May 1983 through December 1984. See D. Levy, P. Asch & D. Shea, An Assessment of DWI Programs in New Jersey 27 (1987). A follow-up study reflected a further reduction in the next year. D. Levy, D. Shea & P. Asch, A Further Assessment of DWI Programs in New Jersey 8 (1987) (reduction by 29.5%). In 1982, authorities in New South Wales, Australia, instituted an intensive program of "random breath testing." R. Homel, D. Carseldine & I. Kearns, Drink-Driving Countermeasures in Australia, 4 Alcohol, Drugs & Driving 113, 114 (No. 2 1987) (hereinafter Australian Countermeasures). /16/ For a five-year period, New South Wales maintained a 22% reduction in fatal crashes. Id. at 132. The average number of "classic' alcohol-related crashes" declined by 35%. Ibid. The number of drivers and riders killed in traffic crashes whose blood alcohol content exceeded 0.05% fell by 36%. Ibid. These effects are both impressive and long-lasting, particularly when viewed in light of the persistence of the drunk driving problem in the United States. /17/ To be sure, sobriety checkpoints are not a panacea. They are effective only when used with a variety of measures, including public education, treatment programs, and streamlined procedures for imposing appropriate sanctions on those who are apprehended. /18/ Moreover, there is debate about how effective sobriety checkpoints, or any other law enforcement technique, can be over the long run. See H. Ross, supra, at 99-115, 122-123. Nevertheless, there is now very substantial evidence that sobriety checkpoints contribute to an immediate reduction in alcohol-related crashes, and that these reductions are sustainable. This evidence provides a more than sufficient basis for using sobriety checkpoints as a promising means of mitigating one of the Nation's most serious public safety problems. 4. Much of the material we have summarized was apparently not before the trial court. See Pet. App. 84a-94a. However, we believe that this Court may properly consider it. This case involves a facial challenge to a sobriety checkpoint program that has been allowed to operate on only one occasion. Its legality cannot be evaluated on its own record. Compare United States v. Martinez-Fuerte, 428 U.S. at 554. The question whether sobriety checkpoints advance the interest in reducing drunk driving is an issue of "legislative fact." See, e.g., Lockhart v. McCree, 476 U.S. 162, 168-169 & n.3 (1986) (analyzing data on the extent to which "death qualification" procedures produce "conviction-prone" jurors); Schmerber v. California, 384 U.S. 757, 770-771 (1966) (effectiveness and intrusiveness of blood test); United States v. Leon, 468 U.S. 897, 927 (1984) (Blackmun, J., concurring) (deterrent effect of exclusionary rule); Karst, Legislative Facts in Constitutional Litigation, 1960 Sup. Ct. Rev. 75, 82-86. As in Lockhart, the fact that other courts have reached the opposite conclusion on precisely the same issue illustrates the difficulty in characterizing the trail court's conclusions as findings of adjudicatory fact. /19/ Unless the legality of sobriety checkpoints is to be determined case-by-case (or stop-by-stop, to the extent individual defendants seek to litigate the matter in criminal trials), this Court should be able to examine all available justifications for sobriety checkpoints. Notwithstanding the standard of review applied by the Michigan Court of Appeals, the issues before the Court are substantially legal in nature. The "reasonableness" of an enforcement technique, as opposed to its application in a particular situation, presents an issue of law. Cf., e.g., Terry v. Ohio, 392 U.S. 1, 20-27 (1968). Similarly, this Court's cases "provide guidance" on the assessment of the factors underlying such a judgment -- the significance of a state interest, the effectiveness of the procedure at issue, and the intrusion on reasonable expectations of privacy. Delaware v. Prouse, 440 U.S. at 656-658. Those factors do not present unmixed questions of historical fact. Cf. Cuyler v. Sullivan, 446 U.S. 335, 341-342 (1980). In their analysis of the effectiveness of sobriety checkpoints, the Michigan courts failed to heed this Court's guidance in two critical respects. First, the primary basis for the state trial court's conclusion that sobriety checkpoints are ineffective, both in terms of actual arrests and deterrence, was its finding that only a small percentage of motorists who are stopped are arrested. Pet. App. 79a-83a, 95a. There is no dispute that the percentage of arrests is small, in keeping with the carefully limited nature of the stop and the fact that even during peak periods of alcohol consumption, when sobriety checkpoints are usually employed, only about 3% of the drivers on the road are intoxicated. See note 8, supra. However, it does not follow that checkpoints must therefore be considered unreasonable for purposes of the Fourth Amendment. This Court's cases have recognized that techniques resulting in far lower rates of detection may be "reasonable" measures. In National Treasury Employees Union v. Von Raab, 109 S. Ct. 1384, 1394-1396 & n.3 (1989), the Court rejected the claim that the small number of positive drug tests resulting from a testing plan -- 5 out of 3600 -- required the plan's invalidation. The Court explained that "(t)he mere circumstance that all but a few of the employees tested are entirely innocent of wrongdoing does not impugn the program's validity." Id. at 1395. Second, the trial court erred in finding sobriety checkpoints unreasonable because the State failed to show that "there are no permissible alternative means available to the police which advance the particular state concern." Pet. App. 75a; see id. at 96a-101a. The Court has never insisted on such a showing in determining whether a program involving a search or seizure is reasonable for purposes of the Fourth Amendment. In Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1419 n.9, the Court refused to invalidate a drug testing program based upon "a list of 'less drastic and equally effective means' of addressing the Government's concerns," explaining: We have repeatedly stated * * * that "(t)he reasonableness of any particular government activity does not necessarily or invariably turn on the existence of alternative 'less intrusive' means." * * * It is obvious that "(t)he logic of such elaborate less-restrictive alternative arguments could raise insuperable barriers to the exercise of virtually all search-and-seizure powers. Accord, e.g., Colorado v. Bertine, 479 U.S. 367, 373-375 (1987); Illinois v. Lafayette, 462 U.S. 640, 647 (1983); United States v. Martinez-Fuerte, 428 U.S. at 556-557 n.12; Cady v. Dombrowski, 413 U.S. 433, 447 (1973). This Court's precedents do not require a promising means of reducing drunk driving to be accompanied by a showing, before the program has even been put into effect, that "there is no practical alternative means available to the State to advance its interest other than by implementing sobriety checkpoints." Pet. App. 96a. Here, implementation of a pilot program of sobriety checkpoints was based upon the considered recommendation of the Michigan Drunk Driving Task Force, a recommendation entirely in keeping with the conclusions of NHTSA, the President's Commission on Drunk Driving, the National Transportation Safety Board, and, more recently, the Surgeon General's Workshop on Drunk Driving. Under these circumstances, "(a)t bottom, (an) insistence on less drastic alternatives would require (the Court) to second-guess (their) reasonable conclusions." Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1419 n.9. B. Sobriety Checkpoints Intrude Only Minimally On Fourth Amendment Interests A number of features of sobriety checkpoints render their intrusion on Fourth Amendment interests minimal. Each motorist's brief stop at a sobriety checkpoint entails only a slight interference with his freedom of movement. Checkpoints are designed in various respects to be reassuring to law-abiding drivers. Finally, the checkpoints are conducted under strict guidelines that provide notice to the driver and restrict the discretion of law enforcement officers in the field, two of the principal functions served by a warrant in other settings. See New York v. Burger, 482 U.S. 691, 703 (1987). In these circumstances, the trial court's conclusion that Michigan's sobriety checkpoints "have a significant subjective intrusive impact on a person's liberty interest" (Pet. App. 119a) cannot be reconciled with this Court's cases. 1. As the Court has repeatedly observed, an individual's "expectation of privacy in an automobile and of freedom in its operation are significantly different from the traditional expectation of privacy and freedom in one's residence." United States v. Martinez-Fuerte, 428 U.S. at 561. See South Dakota v. Opperman, 428 U.S. 364, 367-376 (1976); United States v. Ortiz, 442 U.S. 891, 896 n.2(1975); Cady v. Dombrowski, D13 U.S. at 440-441. The "practice of stopping automobiles briefly for questioning has a long history evidencing its utility and is accepted by motorists as incident to highway use." United States v. Martinez-Fuerte, 428 U.S. at 560-561 n.14. See W. LaFave, Search and Seizure Section 10.8 (2d ed. 1987). In keeping with that practice, a motorist's expectation of free passage is at its most limited with respect to stops for the purpose of enforcing highway safety laws. See New York v. Class, 475 U.S. 106, 112-114 (1986). In principle, there appears to be no difference between the routine practice of stopping trucks at checkpoints to determine whether their weight or brakes constitute a danger to highway safety and stopping cars briefly to determine whether the driver is intoxicated. This is not a case in which stops are designed to further criminal investigations unrelated to the operation of a vehicle. Cf. United States v. BrignoniPonce, 422 U.S. at 883 n.8; id. at 887-888 (Rehnquist, J., concurring); Cady v. Dombrowski, 413 U.S. at 440-441. For that reason, the trial court erred in characterizing sobriety checkpoint stops as "go(ing) beyond the traditionally recognized reasons for stopping motorists in their cars" and as designed to "search for violators of criminal law" or for "investigating purposes." Pet. App. 110a, 114a, 115a. It is a crime to bring a gun or a bomb aboard an airliner, but it would nevertheless be unfair to characterize a routine gate check at an airport as a "search for violators of criminal law" (Pet. App. 110a) or a criminal investigation. See Ingersoll v. Palmer, 43 Cal. 3d at 1330-1331, 743 P.2d at 1306; cf. Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1421 (permitting the government to take "necessary and reasonable regulatory steps" to prevent or deter hazardous behavior that, in some instances, is also criminal). In any event, in Martinez-Fuerte, the Court approved checkpoints in the vicinity of the border to detect illegal immigration even though smugglers who were apprehended were "routinely prosecute(d)." 428 U.S. at 553 n.9; see id. at 560 n.14 (discounting the relevance, for purposes of weighing a law's intrusiveness, of its "administrative," as opposed to criminal, purpose). 2. In United States v. Martinez-Fuerte, 428 U.S. at 557, this Court found that fixed checkpoints used to check immigration documents involved a "quite limited" intrusion on Fourth Amendment interests. Checkpoints of the type at issue here are no more intrusive, in constitutional terms, than those at issue in Martinez-Fuerte. The "objective intrusion" caused by a sobriety checkpoint consists of "only a brief detention of travelers" (id. at 558) -- judging from the experience with the only checkpoint operated under Michigan's guidelines, 25 seconds or less for the average motorist (Pet. App. 6a). "All that is required of the (motorist) is a response to a brief question or two * * *." See United States v. Brignoni-Ponce, 422 U.S. at 880; United States v. Martinez-Fuerte, 428 U.S. at 558; Pet. App. 162a-163a. Thus, as the trial court acknowledged, stops at checkpoints intrude only minimally on a motorist's freedom of movement. Pet. App. 104a. 3. Sobriety checkpoints, like the checkpoints at issue in Martinez-Fuerte, are also established and operated in a manner that minimizes any tendency to engender "concern or * * * fright." See 428 U.S. at 558. For instance, Michigan's guidelines require that checkpoints be established at locations that are well-lit and tha signs clearly advise motorists that they are entering a sobriety checkpoint. See Pet. App. 108a, 152a-153a. Motorists see "that other vehicles are being stopped," and signs, cones, and uniformed officers provide "visible evidence, reassuring to law-abiding motorists, that the checkpoints are duly authorized and believed to serve the public interest." United States v. Martinez-Fuerte, 428 U.S. at 558, 559. /20/ Indeed, this Court in Delaware v. Prouse strongly hinted that such checkpoints are lawful, when it noted the constitutionally significant distinction between "sporadic and random stops of individual vehicles" on the one hand and "those stops occasioned by roadblocks where all vehicles are brought to a halt or near halt, and are subjected to a show of the police power of the community." 440 U.S. at 657. At such "traffic checkpoints," the Court noted, the motorist "can see that other vehicles are being stopped, he can see visible signs of the officers' authority, and he is much less likely to be frightened or annoyed by the intrusion." Ibid. (quoting United States v. Ortiz, 422 U.S. 891, 894-895 (1975)). 4. Finally, under a properly designed checkpoint program, like the one at issue here, the discretion of officers in the field is carefully constrained. Michigan's guidelines require sobriety checkpoints to be located in accordance with established criteria, approved by the Director of the Department of State Police. Pet. App. 148a. The guidelines prescribe that all vehicles are to be stopped and detail what is to be said to each motorist. /21/ Pet. App. 157a-164a. Michigan's procedures expressly require a briefing during which "(u)niform officers at checkpoint locations are to be given instruction regarding the procedure to be followed," including "the exact location of the sobriety checkpoint, legal considerations involved, the procedure to be followed in approaching a vehicle, the initial contact with its driver, and what to do in specific situations." Id. at 150a-151a. These procedures are more than sufficient to assure that each motorist's "reasonable expectation of privacy is not 'subject to the discretion of the official in the field.'" Delaware v. Prouse, 440 U.S. at 655, 661; see Colorado v. Bertine, 479 U.S. at 376-377 (Blackmun, J., concurring). /22/ All of these factors -- the limited expectation that motorists have in being free from stops to enforce traffic safety laws, the brief and limited nature of the initial stop, the State's effort to notify the public of sobriety checkpoints and to assure that they are surrounded by signs and indications of authority reassuring to average motorists, and the restraints on the exercise of discretion by officers in the field -- taken together require the conclusion that sobriety checkpoints of the type at issue in this case intrude only very minimally on Fourth Amendment interests. C. With Respect To Sobriety Checkpoints, The Balance Of Relevant Interests Justifies Brief Stops Without Individualized Suspicion The singular importance of reducing drunk driving, the evident effectiveness of sobriety checkpoints in advancing that goal, and their limited intrusion on Fourth Amendment interests justify an exception to the Fourth Amendment's requirement of individualized suspicion. The effectiveness of sobriety checkpoints derives from their universal applicability; every driver knows that if he encounters a checkpoint he will be exposed to a brief examination for signs of intoxication. In this situation, the goal of reducing the incidence of the profoundly dangerous practice of drinking and driving gives rise to "special needs, beyond the normal need for law enforcement, (that) make the warrant and probable-cause requirement impracticable." Griffin v. Wisconsin, 483 U.S. 868, 873 (1987). See Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1414. That is true even though driving under the influence of alcohol is a crime; in Skinner, the Court determined that the fact that the possession of drugs is a crime did not divest the government of its authority to take reasonable regulatory steps to prevent or deter the hazard resulting from drug use. 109 S. Ct. at 1421. Moreover, the "important governmental interest furthered by the (checkpoints) would be placed in jeopardy by a requirement of reasonable suspicion," and "the privacy interests implicated by (sobriety checkpoints) are minimal," Skinner v. Railway Labor Executives Ass'n, 109 S. Ct. at 1417. This is a situation in which "the Government's need to discover * * * latent or hidden conditions, or to prevent their development, is sufficiently compelling to justify" brief stops for the purpose both of removing intoxicated motorists from the road and deterring others from driving, National Treasury Employees Union v. Von Raab, 109 S. Ct. at 1392. Sobriety checkpoint programs such as those at issue in this case do not exceed the limits that the Fourth Amendment imposes "to prevent arbitrary and oppressive interference by enforcement officials with the privacy and personal security of individuals." United States v. Martinez-Fuerte, 428 U.S. at 554. In place of a warrant and probable cause or individualized suspicion, sobriety checkpoint programs are "carried out pursuant to a plan embodying explicit, neutral limitations on the conduct of individual officers." Brown v. Texas, 443 U.S. at 51. See United States v. Martinez-Fuerte, 428 U.S. at 564-566. They do not, therefore, provide for seizures that are "unreasonble" within the meaning of the Fourth Amendment. CONCLUSION The judgment of the Michigan Court of Appeals should be reversed. Respectfully submitted. KENNETH W. STARR Solicitor General EDWARD S.G. DENNIS, JR. Assistant Attorney General WILLIAM C. BRYSON Deputy Solicitor General STEPHEN L. NIGHTINGALE Assistant to the Solicitor General NOVEMBER 1989 /1/ NHTSA recently promulgated proposed regulations that would govern eligibility for grants under the Drunk Driving Prevention Act of 1988, Pub. L. No. 100-690, 100 Stat. 4521, under which States would be required, as a condition for federal funding, to implement "(r)egularly conducted, peak-hour DWI enforcement efforts consisting of measures, such as roadside sobriety checkpoints or special DWI patrols." 54 Fed. Reg. 26,795 (1989). /2/ The federal government has devoted substantial effort and funding to the goal of reducing the drunk driving problem. For fiscal years 1982 through 1987, NHTSA disbursed approximately $266,000,000 under the Highway Safety Act to combat drunk driving. /3/ These procedures are consistent with the guidelines NHTSA has promulgated informally. R. Compton & R. Engle, The Use of Safety Checkpoints for DWI Enforcement 11-14 (DOT HS 806 476 1983). /4/ The trial court rejected the last two of these claims (Pet. App. 36a-44a), and the Michigan Court of Appeals did not address them. /5/ The court of appeals found it unnecessary to reach the trial court's interpretation of the Michigan Constitution; the court explained that the state constitution "offers at least the same protection as accorded by the Fourth Amendment and (that the court had) already found the checkpoint program to be an unreasonable seizure under the Fourth Amendment." Pet. App. 24a. In view of this holding, it is clear that the Michigan Constitution does not provide an independent and adequate ground for the appellate court's judgment. See Michigan v. Long, 463 U.S. 1032, 1037-1044 (1983). /6/ See Delaware v. Prouse, 440 U.S. 648, 653 (1979); United States v. Martinez-Fuerte, 428 U.S. 543, 546 n.1, 556 (1976). There was testimony in the state trial court that Michigan's sobriety checkpoints were required to be established in areas where it was possible for a motorist to make a legal U-turn and that motorists were permitted to refuse to roll down their windows when approached by police. Pet. App. 109a. However, those aspects of the program were apparently not well publicized. In our view, the legality of the checkpoints is not dependent on whether motorists may be deemed to have consented to a particular stop. Cf. United States v. Lopez-Pages, 767 F.2d 776, 779 n.2 (11th Cir. 1985); United States v. DeAngelo, 584 F.2d 46, 47-48 (4th Cir. 1978), cert. denied, 440 U.S. 935 (1979); United States v. Freeland, 562 F.2d 383, 385-386 (6th Cir.), cert. denied, 434 U.S. 957 (1977). /7/ This Court has taken note of the problem on several occasions. South Dakota v. Neville, 459 U.S. 553, 558 (1983) ("carnage caused by drunk drivers is well documented"); Breithaupt v. Abram, 352 U.S. 432, 439 (1957) ("The increasing slaughter on our highways, most of which should be avoidable, now reaches the astounding figures only heard of on the battlefield."). See also Perez v. Campbell, 402 U.S. 637, 657, 672 (1971) (Blackmun, J., concurring). /8/ By contrast, even between 10 p.m. and 3 a.m. on weekend nights, the peak period for drinking and driving, the most recent roadside survey found that an average of 3.2% of the drivers on the road had blood alcohol concentrations of 0.10% or higher. A. Lund & A. Wolfe, Changes in the Incidence of Alcohol-Impaired Driving in the United States, 1973-1986, Insurance Institute for Highway Safety 25 (Feb. 1989). /9/ Thirty-nine percent of all fatalities from traffic crashes occurred in crashes in which a driver or non-occupant -- i.e., an involved pedestrian or bicyclist -- had a blood alcohol concentration exceeding 0.10%, the legal limit for intoxication prescribed by most States. Fatal Traffic Crashes, at 5. Another 10% of the fatalities involved alcohol at a lower level. These figures are estimates, inasmuch as the blood of each driver, pedestrian, and bicyclist involved in a crash is not actually tested. NHTSA uses validated statistical techniques to estimate alcohol involvement in those cases in which tests have not been performed. /10/ This regulation has since been codified, as amended, at 23 C.F.R. 1309.6(b)(11). /11/ The NTSB has called attention to its recommendation on sobriety checkpoints on a number of occasions, most recently in connection with its investigation of a tragic collision between a pickup truck, whose driver was intoxicated, and a bus, in which 26 people were killed. NTSB, Highway Accident Report: Pickup Truck/ Church Activity Bus Head-On Collision and Fire Near Carrollton, Kentucky, May 14, 1988, at 50-51 (NTSB/HAR-89/01 Mar. 1989). /12/ During those weekend nights, in 1988, 44.1% of the men and 28.4% of the women involved in fatal crashes had blood alcohol levels exceeding 0.10%. Fatal Traffic Crashes at 12. /13/ Other studies have reached similar conclusions. A. Williams & A. Lund, Deterrent Effects of Roadblocks on Drinking And Driving, 3 Traffic Safety Evaluation Res. Rev. 12 (Nov.-Dec. 1984) (surveys of residents in (i) Montgomery County, Maryland, where checkpoints were used, and Fairfax County, Virginia, where they were not, and (ii) Kent and Sussex Counties in Delaware, where checkpoints were used, and Maryland's Eastern Shore, where they were not; majorities of all areas surveyed believed that DUI arrests were more likely in areas in which checkpoints were used); R. Hingson & J. Howland, Public Perceptions About Drunk Driving Roadblocks: Surveys in Massachusetts Where Roadblocks Are Used and New Hampshire Where They Are Prohibited 4-7 (available from NHTSA, NTS-20) surveys of residents in Massachusetts and New Hampshire showed that residents of both States believed (i) that the likelihood of a drunk driver's being tested and charged at a checkpoint was greater than after a stop by a roving patrol, and (ii) that drunk drivers were more likely to be arrested in Massachusetts, where checkpoints were employed, than in New Hampshire, where they were not). See also J. Lacey, J. Stewart, et al., Enforcement and Public Information Strategies for DWI General Deterrence: ARREST DRUNK DRIVING -- The Clearwarte and Largo, Florida Experience 54-58 (DOT HS807 066 Dec. 1986); Missouri Div. of Highway Safety, Sobriety Checkpoint Survey: 1985 Report of Findings 2. Several of those surveys indicated that persons who drink and drive are less likely to do so in areas where checkpoints are present. See J.H. Lacey, J.R. Stewart, et al., supra, at 56 (while checkpoint program in place, increasing percentages of drivers indicated that they had not driven after having too much to drink); R. Voas, E. Rhodenizer & C. Lynn, supra, at 38 (many servers at restaurants and bars reported decreased drinking whole checkpoint program was in place); but see A. Williams & A. Lund, supra, at 17 (limited evidence suggests that checkpoints did not change drinking and driving behavior); R. Hingson & J. Howland, supra, Table 5 (absence of statistically significant figures suggesting such a change). /14/ The study of the use of checkpoints in Charlottesville in 1983-1984, R. Voas, E. Rhodenizer, C. Lynn, supra, at 8, showed that one DUI arrest was made for each 6.5 hours of officer time at checkpoints, as opposed to one arrest for each 7.9 hours of officer time on traditional DUI patrols. /15/ J. Lacey, J. Stewart, et al., supra, at 59-65 (concluding that the combined checkpoint and public information program were effective in reducing alcohol-related crashes); D. Levy, P. Asch & D. Shea, An Assessment of DWI Programs in New Jersey 27 (1987) (available from NHTSA, NTS-20) (checkpoint program achieved a reduction of 14-24% in single-vehicle-nighttime accidents (a representative measure of alcohol-related crashes) in Bergen County, New Jersey, during 1983 and 1984). /16/ Random breath testing involves arbitrarily selected checkpoints, usually on main roads. Motorists are pulled over randomly, without regard to age, sex, or manner of driving. Australian Countermeasures at 114. This program varies from sobriety checkpoints in this country, in that here all motorists or a structured sample (1 in every 5, for instance) are stopped at the checkpoint. /17/ Even critics of sobriety checkpoints have been impressed by this data. Respondents' expert testified in this case in mid-1986. Roughly a year later, he published an article in which he stated (Ross, Reflections on Doing Policy-Relevant Sociology: How to Cope with MADD Mothers, 18 Am. Sociologist 173, 177 (1987): The necessity to change opinions due to new research constitutes another pitfall for the researcher in the politically sensitive area. * * * For example, although the U.S. research has been flawed and unconvincing, recent research in Australia suggests that police roadblocks may be more effective than other kinds of patrol in producing the impression of certain punishment for drunk-driving offenders. Although I have strong personal reservations related to the intrusiveness of these techniques, I have had to modify my testimony in cases concerning their constitutionality after reading the Australian reports. /18/ NHTSA's criteria for the distribution of funds under Section 408 of the Highway Safety Act illustrate the range of measures that may contribute to a solution. 23 C.F.R. 1309.5-1309.7. /19/ In this case, the trial court dismissed as unpersuasive evidence based on Maryland's experience with sobriety checkpoints. Pet. App. 84a-92a. The Maryland Court of Appeals, however, found that much of the same material demonstrated that Maryland's sobriety checkpoint program "has been a moderately effective technique for detecting and deterring the drunk driver." Little v. State, 300 Md. 485, 479 A.2d 903, 913 (1984). The Maryland court found, accordingly, that the program was lawful. Though a substantial majority of state courts has concluded that sobriety checkpoint programs are lawful if accompanied by appropriate safeguards, the courts have reached conflicting conclusions. Compare, e.g., Ingersoll v. Palmer, 43 Cal. 3d 1321, 743 P.2d 1299, 241 Cal. Rptr. 42 (1987); People v. Bartley, 109 Ill. 2d 273, 486 N.E.2d 880, cert. denied, 475 U.S. 1068 (1985); People v. Scott, 63 N.Y.2d 518, 473 N.E.2d 1, 483 N.Y.S.2d 649 (1984); State v. Deskins, 234 Kan. 529, 673 P.2d 1174 (1973), with, e.g., State v. Kopell, 127 N.H. 286, 499 A.2d 977 (1985); State v. Smith, 674 P.2d 562 (Okla. 1984). Most of the adverse state decisions have been based on state constitutional grounds. /20/ The trial court identified several features of sobriety checkpoints that it believed increase their intrusiveness -- the fact that they are "transitory," rather than at fixed locations, and the fact that motorists will not be made aware of their opportunity to avoid the checkpoints. These circumstances provide no substantial basis for distinguishing this Court's assessment of the intrusiveness of the checkpoints at issue in Martinez-Fuerte. Consistent with the sobriety checkpoints' purpose of deterring drunk driving, the guidelines provide that "(a)n integral aspect of the pilot program will be to publicize the deployment of sobriety checkpoints to attain maximum public awareness." Pet. App. 145a. That publicity, together with signs and word of mouth, would undoubtedly suffice to give most motorists notice of the nature and purpose of sobriety checkpoints at least equivalent to that available in Martinez-Fuerte. A motorist's theoretical opportunity to avoid the checkpoints at issue in Martinez-Fuerte does not appear to have been a significant factor in this Court's assessment of their intrusiveness. Indeed, the Court noted that checkpoints in the vicinity of the border "operate on a coordinated basis designed to avoid circumvention by smugglers and others who transport the illegal aliens." 428 U.S. at 552. /21/ Other programs may require a fixed percentage of traffic -- for instance, every fifth car -- to be stopped or require officers to cease operations temporarily when appropriate to avoid a backup of traffic. In our view, because programs of that nature also constrain discretion, they are not constitutionally distinguishable from Michigan's program. See Delaware v. Prouse, 440 U.S. at 664 (Blackmun, J., concurring). /22/ To be sure, officers at a sobriety checkpoint exercise judgment when they examine motorists for "articulable indications of intoxication" -- "an odor of intoxicants about the driver, slurred speech, lack of physical coordination, unusual eye movements, disorientation, or other behavior commonly associated" with driving under the influence of drugs or alcohol. Pet. App. 155a. However, those judgments may be considered on an individual basis if challenged by a motorist who claims that there was an insufficient justification for any subsequent detention and search. In this facial challenge, the Court need not address the nature of the individualized suspicion needed for, for instance, a breath test or field sobriety tests.