Galberth v. United States

ROGERS, Chief Judge: In these consolidated appeals appellants Russell Galberth and Jay Taylor appeal their convictions on the ground that the trial court erred in denying their motions to suppress evidence that was discovered after they were stopped at police traffic roadblocks. 1 Guided by Supreme Court precedent, we balance the government interest served by the roadblocks against the liberty interests of the individuals seized. We therefore conclude that appellant Gal-berth’s conviction must be reversed, given Judge Dixon’s finding that the roadblock at which Galberth was stopped was designed to combat violence and illegal drug activity. We further conclude that Judge Huhn’s findings do not clearly indicate the principal purpose of the roadblock at which appellant Taylor was stopped. We therefore remand for further findings in appellant Taylor’s case. I. The two roadblocks at issue were established in connection with Operation Clean *992 Sweep, a police department program instituted in August 1986 to address illegal drug trafficking, particularly sales in open air markets, and accompanying violence in the District of Columbia. The program was governed by the Field Operations Bureau which developed a manual for each police district containing guidelines for the various law enforcement techniques employed by Operation Clean Sweep. The facts of the two roadblocks differ: Galberth roadblock. Appellant Galberth was stopped at the roadblock at Montello Avenue and Queen Street, N.E., on December 5, 1989. Officer Cephas, a field officer assigned to the roadblock, testified that he received orders that all cars were to be stopped from 9:30 a.m. until 11:30 a.m. and the drivers asked for their licenses and registrations. Eight to ten officers were present, in uniform, and some were wearing visibility jackets; 5 to 6 marked police cars were also on the scene. Officer Cephas’s instructions were that if any questions arose, then he was to run a Washington Area Law Enforcement Service (WALES) computer check to see if the driver had a valid license. No barriers, signs, or traffic cones were used to mark the roadblock, but Officer Cephas estimated that the police officers could be seen from about midway up the block. Officer Cephas testified that persons were stopped for approximately 2 to 3 minutes if there was no problem with their license. The roadblock at Montello and Queen was part of a special operation in the Trinidad area of the city announced by the Mayor to “cut down on the violence and homicides and narcotics trafficking in the Trinidad area, [by] attacking it from three different angles,” — roadblocks, undercover operations, and high visibility “saturation” patrols. Captain O’Donnell, the coordinator for Operation Clean Sweep for the Fifth District, testified that the roadblock program was designed to deal with “traffic problems such as speeding automobiles, vehicles frequenting the area committing traffic violations to purchase narcotics, stolen automobile traffic in the neighborhood, and things of that nature.” As he explained, a high volume of narcotics trafficking brings with it a similarly heavy amount of automobile traffic, since many people come to the open air drug markets by car. Captain O’Donnell’s commanding officer delegated to him the responsibility to select a site for the roadblocks, to notify his commanding officer of his selection, to assign a supervisor on the scene, and to periodically inspect the roadblock while it was in operation. O’Donnell selected the Montello Avenue site because he knew from personal observation as well as citizen complaints that the area was plagued by heavy traffic created when individuals looking to buy drugs parked illegally for a brief period, and then sped away after making their purchase. The evidence he offered (in addition to his own observations) relating to traffic congestion, however, was that in the past three years there had been over 100 citizen complaints about drugs, traffic, abandoned cars, and trash in the area of Montello and Queen. O’Donnell instructed Sergeant Bullock to operate the roadblock from 9:30 a.m. to 11:30 a.m., and then to conduct a high visibility saturation patrol in the Trinidad area. Captain O’Donnell inspected the roadblock while it was in operation, and explained that although the roadblock guidelines suggest using flares, flashing warning lights, or flashlights to increase the roadblock’s visibility, 2 he saw no need for them in this case because it was daytime and the police could be seen 100 yards before a car entered the intersection. In addition, an escape route was available, allowing cars to turn down Holbrook Terrace and avoid passing through the roadblock. Sergeant Bullock’s report indicated that 226 cars, coming from three directions, were stopped, four arrests were made (three for no permit and one for operating *993 after suspension), and seventeen traffic violation citations were issued. Appellant Galberth was stopped at the roadblock and asked by Officer Cephas for his driver’s license and registration. Gal-berth told the officer that he did not have his driver’s license and that he must have forgotten to bring it. Cephas asked for Galberth’s social security number, and the resulting computer check indicated that Galberth’s license had expired in 1983. Ce-phas ordered Galberth out of the van and placed him under arrest for not having a valid operating permit. As Officer Cephas was patting Galberth down, he felt an object in Galberth’s front coat that he thought might be a gun. Reaching into the pocket, Cephas found a .38 caliber revolver. Galberth testified that he did not see the police until he reached the stop sign at the intersection, and that there was no way that he could avoid being pulled over by the police. Mr. Lucas, a passenger in the van that Galberth was driving, testified that he too first noticed the police as the van approached the intersection about 50-75 feet before they came upon the police at the intersection. Lucas saw no visibility jackets, cones or flares, although he did see police officers talking to people in other cars in the area. Judge Dixon denied the motion to suppress the gun. The judge found that there was a substantial police presence and that the officers were wearing visibility jackets. He further found that an organized and defined procedure had been used in setting up the roadblock and that no discretion had been left in the hands of individual police officers in the field. In addition, Judge Dixon found that the intrusion on each motorist’s liberty was minimal. He also found, however, that the primary purpose of the roadblock was to deter drug traffic and violence in connection with Operation Clean Sweep, and that while relieving the traffic problems at Montello and Queen may have been one factor that went into the determination about where to locate the roadblock, traffic alone would not have caused the police to set up a roadblock. He nevertheless ruled that the roadblock sufficiently advanced the government’s interest. Taylor roadblock. Lieutenant Denesio testified that he selected sites at 14th and Quincy Streets, N.W. and 14th and Allison Streets, N.W., for roadblocks on the evening of February 16, 1988, because of high drug activity in the neighborhood that involved heavy vehicular traffic along the public streets. Since being assigned to the Fourth District in September 1987, he had received numerous complaints and eyewitness reports that resulted in identification of 40 large volume street drug sale locations within the Fourth District. He explained that site selection had two purposes: traffic enforcement and disruption of narcotics trafficking. Traffic enforcement was important in his view because “a lot of people from the suburbs ... come into the city to purchase their drugs.” Drug arrests were not the goal of the roadblocks; rather, their primary purpose was to disrupt drug traffic. Lieutenant Denesio testified that usually within a week or two before the roadblock was to be set up, he filed an enforcement activity report with Captain Stewart informing him of the planned roadblock locations and procedures. His instructions for the on-site supervisor left the supervisor no discretion about the location of the roadblock. Although Denesio conceded that no advance publicity warned the public that an Operation Clean Sweep roadblock was going to be operating in the “general neighborhood at around that particular time,” Assistant Chief Fulwood testified that the police department had given the public ample notice of the roadblocks through “almost weekly” news conferences and community meetings attended by the mayor, the police chief and Assistant Chief Ful-wood, as well as information disseminated through district commanders and the Fourth District Advisory Council, and “untold numbers of public announcements about ‘Operation Clean Sweep’ and its impact. ...” Captain Stewart was the Clean Sweep Coordinator for the Fourth District, and thereby responsible for carrying out the *994 roadblock in accordance with the terms set by the Field Operations Bureau of the Police Department. On February 16, 1988, there were 8 to 10 uniformed officers in visibility vests, with two support vehicles and a transport vehicle, who were stopping southbound vehicles on 14th Street at Quincy. Because the field officers had previously conducted Operation Clean Sweep roadblocks, Stewart saw no need to give additional instructions that night; however, he typically instructed the supervising sergeant that the license tag of every car stopped was to be run through the WALES computer. 3 In general, the supervising sergeant had discretion to determine the pattern by which cars would be stopped. In this instance, Stewart thought, the supervising sergeant had decided that every second car should be stopped. Sergeant Canty, who supervised the roadblock at 14th and Quincy Streets, N.W., on February 16, 1988, testified that he received instructions from Captain Stewart to run the roadblock from 9:00 p.m. until 10:30 p.m., and to stop every third car going in both directions on 14th Street at Quincy. Canty testified that all officers were in uniform, used flashlights, wore bright orange visibility jackets, and that the two police cars on 14th Street were marked. The officers stood in the middle of the two-way street and waved cars over to the curb. The officers were instructed to identify themselves, inform those stopped that a roadblock was being conducted, and ask the driver for his or her license and vehicle registration. Forty-two cars were stopped between 9:15 p.m. and 10:10 p.m., and two arrests were made (appellant Taylor and his passenger) for unauthorized use of a vehicle. 4 Canty testified from personal observation that the area was a high drug area, and that individuals frequently double parked their cars while conducting drug transactions. While the roadblock was in operation, however, Canty testified that he saw no drug activity. Appellant Taylor attempted to turn his car around as he approached the intersection, according to Officer Gray, but was directed to stop since his was the third car to approach the intersection. In response to the officer’s request for his driver’s license and car registration, Taylor handed the officer the high school identification card of his passenger, Timothy Minor. A computer check revealed that no license existed for Minor. Taylor then told the officer his name, and while waiting for a second computer check, this one on the license plate of the car Taylor was driving, he told the officer that the car belonged to his sister. Taylor and Minor were arrested after the computer check on the car license plate indicated that the car had been reported stolen. The arrests occurred between five to ten minutes after the car was stopped. According to Officer Gray, the computer check alone took approximately five minutes. Judge Huhn credited the testimony of Assistant Chief Fulwood that the primary purpose of the roadblock was to control traffic congestion associated with open air drug markets. The Assistant Chief had testified that he had decided that traffic roadblocks should be part of Operation Clean Sweep because large open air drug markets and traffic problems go hand in hand: “what I mean ... is cars double, tripled park[ed], large numbers of pedestrians moving through the neighborhood, to the point that it disrupts the lives of people who ordinarily live there who are not involved in any kind of criminal activity of any kind.” Judge Huhn also credited the Chief’s testimony that in most of the open air drug market operations there were motorists driving without a driver’s permit or after revocation of a permit, and motorists whose ability to drive was impaired. *995 Judge Huhn found that the public danger created to pedestrians and other motorists in the area was a sufficient law enforcement purpose to justify the roadblock procedure, that the area was characterized by the traffic problems that Operation Clean Sweep roadblocks were designed to address, and that the roadblock “would disrupt drug trafficking in the area because potential drug purchasers would avoid the area after observing the presence of the police.” The judge also found that the police expected to find motorists operating without permits, operating after revocation and operating with impaired ability to drive. Judge Huhn found that “[i]n fact, there was a reduction in the apparent street level drug sales during the night of the ... roadblock as evidenced by a reduction in the observation of double parked cars and persons walking over to doubled [sic] parked ears.” The judge was satisfied that there had been sufficient notice to the public as a result of weekly news conferences and various community meetings. In rejecting the argument that the roadblock was unconstitutional because the officers had unfettered discretion, because the publicity and advance warning were insufficient, because lighting at the scene was inadequate, or because the time of the detention was too long, the judge relied on Little v. State, 300 Md. 485, 479 A.2d 903 (1984), and United States v. McFayden, 275 U.S.App.D.C. 207, 865 F.2d 1306 (1989), adopting the analysis in the latter opinion. II. Appellants contend that the trial judges erred in denying their motions to suppress evidence under the Fourth Amendment. 5 The government concedes, as it must, that appellants were seized within the meaning of the Fourth Amendment when they were stopped at the roadblocks. See Michigan Dep’t of State Police v. Sitz, — U.S. —, 110 S.Ct. 2481, 2485, 110 L.Ed.2d 412 (1990) (“a Fourth Amendment ‘seizure’ occurs when a vehicle is stopped at a checkpoint”); United States v. Martinez-Fuerte, 428 U.S. 543, 556, 96 S.Ct. 3074, 3082, 49 L.Ed.2d 1116 (1976) (“checkpoint stops are ‘seizures’ within the meaning of the Fourth Amendment”); United States v. Montgomery, 182 U.S. App.D.C. 426, 561 F.2d 875, 878 (1977) (“The stop of a moving vehicle — even if the period of detention is brief — involves a ‘seizure’ within the meaning of the fourth amendment”). Because the Fourth Amendment proscribes only those seizures that are unreasonable, however, United States v. Sharpe, 470 U.S. 675, 682, 105 S.Ct. 1568, 1573, 84 L.Ed.2d 605 (1985), these appeals turn on whether the stops of appellants in connection with Operation Clean Sweep were reasonable. Generally, a seizure is reasonable only if the police have particularized suspicion that the seized individual is involved in criminal activity. Terry v. Ohio, 392 U.S. 1, 21 & n. 18, 88 S.Ct. 1868, 1880 & n. 18, 20 L.Ed.2d 889 (1968). This general rule is applicable to spot checks of vehicles stopped by the police, as the Supreme Court held in Delaware v. Prouse, 440 U.S. 648, 99 S.Ct. 1391, 59 L.Ed.2d 660 (1979). In Prouse the Court held unconstitutional the random stop of an automobile and the detention of its driver, even for the brief time necessary to check his driver’s license and car registration, as unreasonable under the Fourth Amendment in the absence of articulable suspicion that the occupant was subject to seizure for a violation of law. Id. at 663, 99 S.Ct. at 1401. The Court contrasted the roving random patrols that looked for illegal aliens and smuggling activity in United States v. Brignoni-Ponce, 422 U.S. 873, 95 S.Ct. 2574, 45 L.Ed.2d 607 (1975), 6 with *996 the permanent checkpoint stops in United States v. Martinez-Fuerte, 428 U.S. 543, 96 S.Ct. 3074, 49 L.Ed.2d 1116 (1976), in which the Border Patrol slowed or stopped each vehicle, and at the discretion of the agents referred only certain vehicles to an area for “secondary inspection.” The Court noted that it had upheld the constitutionality of the latter type of search because “the subjective intrusion — the generating of concern or even fright on the part of lawful travelers — is appreciably less in the case of a checkpoint stop.” Delaware v. Prouse, supra, 440 U.S. at 656, 99 S.Ct. at 1397 (quoting Martinez-Fuerte, supra, 428 U.S. at 558, 96 S.Ct. at 3083). In particular, the Court explained that “[a]t traffic checkpoints 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.” Id. 440 U.S. at 657, 99 S.Ct. at 1398 (quoting Martinez-Fuerte, supra, 428 U.S. at 558, 96 S.Ct. at 3083). In contrast, although the Court agreed that states have “a vital interest in ensuring that only those qualified to do so are permitted to operate motor vehicles, that these vehicles are fit for safe operation, and hence, that licensing, registration, and vehicle inspection requirements are being observed,” id. 440 U.S. at 658, 99 S.Ct. at 1398, the Court found no “empirical data” which demonstrated that, given the existence of alternative methods, the discretionary spot check “is a sufficiently productive mechanism to justify the intrusion upon Fourth Amendment interests which such stops entail.” Id. at 659, 99 S.Ct. at 1399. Subjecting every occupant of a motor vehicle to a seizure at the “unbridled discretion of law enforcement officials” involved the “kind of standardless and unconstrained discretion [which] is the evil the Court has discerned when in previous cases it has insisted that the discretion of the official in the field be circumscribed, at least to some extent.” Id. at 661, 99 S.Ct. at 1400. The Court nevertheless left open the possibility that methods could be developed to conduct spot checks that “involve less intrusion or that do not involve the unrestrained exercise of discretion,” suggesting questioning all oncoming traffic at “roadblock-type stops” as one possibility. Id. at 663, 99 S.Ct. at 1401. A decade later, Michigan Dep’t of State Police v. Sitz, — U.S. —, 110 S.Ct. 2481, 110 L.Ed.2d 2481 (1990), held that “stops of motorists on public highways,” Sitz, supra, 110 S.Ct. at 2485, which are “carried out pursuant to a plan embodying explicit, neutral limitations on the conduct of individual officers,” Brown v. Texas, 443 U.S. 47, 51, 99 S.Ct. 2637, 2640-41, 61 L.Ed.2d 357 (1979), may be reasonable in certain circumstances, even in the absence of particularized suspicion of crime. Sitz upheld the constitutionality of the initial stop of a motorist passing through a highway sobriety checkpoint. 110 S.Ct. at 2488. The checkpoint was established in accordance with guidelines developed by an advisory committee on site selection, checkpoint operations, and publicity. Id. at 2484. These guidelines left virtually no discretion to the officers stopping motorists. 7 Id. at 2484. The Court noted that sobriety checkpoints, unlike the random stop in Prouse, do not pose the risk of unconstrained discretion by the field officer, and are sufficiently effective in achieving the states’ legitimate interest in preventing accidents caused by drunk driving. Id. at 2487-88. To determine the reasonableness of roadblock seizures, Sitz reaffirmed the test applied in Delaware v. Prouse, supra, requiring courts to strike a “balance between the public interest and the individual’s right to personal security free from arbitrary interference by law officers.” 8 Brown v. Texas, supra, 443 U.S. at 50, 99 S.Ct. at 2640 *997 (quoting Pennsylvania v. Mimms, 434 U.S. 106, 109, 98 S.Ct. 330, 54 L.Ed.2d 331 (1977)), cited in Sitz, supra, 110 S.Ct. at 2484-85. This balance focuses on three factors: “[1] the gravity of the public concerns served by the seizure, [2] the degree to which the seizure advances the public interest, and [3] the severity of the interference with individual liberty.” Brown v. Texas, supra, 443 U.S. at 50-51, 99 S.Ct. at 2640; see also Sitz, supra, 110 S.Ct. at 2485 (applying the Brown v. Texas balancing test to sobriety checkpoints). Applying the Brown v. Texas balance to the instant cases, we first consider the “public concerns served by” the roadblocks. We agree with the government that there clearly were several purposes served by the roadblocks, and that the police may constitutionally benefit from the “spin-off” effect of an otherwise constitutional law enforcement program. See United States v. Villmonte-Marquez, 462 U.S. 579, 584 n. 3, 103 S.Ct. 2573, 2577 n. 3, 77 L.Ed.2d 22 (1983) (upholding the constitutionality of a routine ship inspection, despite the fact that the officers were following an informant’s tip); McFayden, supra, 275 U.S.App.D.C. at 213, 865 F.2d at 1312. We must nonetheless consider the principal purpose underlying each of the roadblocks to determine their constitutionality. See McFayden, supra, 275 U.S.App.D.C. at 213, 865 F.2d at 1312 (principal purpose “to regulate vehicular traffic by allowing police to check drivers’ licenses and vehicular registration,” and not a subterfuge for detecting crimes unrelated to licensing, which might infringe upon Fourth Amendment rights). Because the trial judges’ findings differed as to the principal purposes of the two roadblocks, we strike a different balance in each of the respective cases. A. The Galberth roadblock The trial judge in appellant Gal-berth’s case made the following finding about the government’s purpose in establishing the roadblock: Based on the evidence that the Court has received the Court is not really overwhelmed with the idea of traffic problems at Montello and Queen. That may have fit into the equation which caused the decision to set up the roadblock at that location, but the primary purpose, because the Court does not want to taint the evidence which was really perceived, the primary purpose had to do with Operation Clean Sweep, violence, drugs and guns. (emphasis added). We must accept this finding of fact, which, as the government concedes, is supported by the record and not clearly erroneous. Appellant Galberth urges us to read the trial judge’s statement as a finding that the police stopped his car in order to detect crimes related to “violence, drugs and guns.” Under another plausible reading, the finding could mean that the police established the roadblock in order to disrupt the open air drug market. We conclude that under either reading of Judge Dixon’s finding, the roadblock cannot survive the Brown v. Texas balance, and therefore reverse appellant Galberth’s conviction. 9 First, it is clear that the police may not use a roadblock in order to seek evidence of drug-related crimes. The Supreme Court has allowed suspicionless roadblock stops in two contexts, both of which involved government interests distinct from general law enforcement. In United States v. Martinez-Fuerte, supra, the Court upheld the constitutionality of permanent immigration checkpoints located near the nation’s borders. The Court noted the “formidable” problem posed by “the flow of illegal [immigrants] from Mexico,” id., 428 U.S. at 552, 96 S.Ct. at 3080, and upheld the use of checkpoint stops only after concluding “that maintenance of a traffic-checking program ... is necessary because the flow *998 of illegal aliens cannot be controlled effectively at the border” using any other mechanism. Id. at 556, 96 S.Ct. at 3082. Most recently, in Sitz, supra, the Supreme Court upheld the constitutionality of sobriety checkpoints, noting the importance of the government’s unique interest in combating drunk driving. Sitz, supra, 110 S.Ct. at 2486 (“[n]o one can seriously dispute the magnitude of the drunken driving problem”). The Supreme Court has never upheld, by contrast, a police roadblock designed to promote general law enforcement purposes. Indeed, the Court has indicated that the police must have individual suspicion before they can seize someone for general law enforcement purposes. See Delaware v. Prouse, supra, 440 U.S. at 659 n. 18, 99 S.Ct. at 1399 n. 18 (the governmental interest in controlling automobile thefts “is not distinguishable from the general interest in crime control” and is therefore insufficient to justify suspicionless stops). In Carroll v. United States, 267 U.S. 132, 153-54, 45 S.Ct. 280, 285, 69 L.Ed. 543 (1925), the Court stated that “[i]t would be intolerable and unreasonable if a prohibition officer were authorized to stop every automobile on the chance of finding liquor, and thus subject all persons lawfully using the highways to the inconvenience and indignity of such a search.” 10 Thus, if Judge Dixon meant that the police established the Galberth roadblock in order to detect evidence of drugs or other crimes, the roadblock would be unconstitutional. See 3 W. LaFave, Search and Seizure: A Treatise on the Fourth Amendment § 9.5(b), at 551 (2d ed. 1987) (“a general roadblock ... established on the chance of finding someone who has committed a serious crime” would “quite clearly” be unconstitutional) (citation omitted); see also Ingersoll v. Palmer, 43 Cal.3d 1321, 1327-28, 743 P.2d 1299, 1303 (1987) (“if the primary purpose of the [roadblock] were to detect crime or gather evidence of crime, we would agree with the contention that an individualized suspicion of wrongdoing is required”); Meeks v. State, 692 S.W.2d 504, 508 (Tex.Crim.App.1985) (“The mere asking for a driver’s license will not validate the stopping of an automobile if it is clear that the driver’s license check was not the reason for detention”). Furthermore, if we were to read Judge Dixon’s finding more broadly, as indicating that the police established the roadblock in order to disrupt the open air drug market, we would still conclude that the roadblock was unconstitutional. Under such a justification, the government purportedly hoped that by setting up a roadblock near a high-crime area, people would be discouraged from driving into the area to buy drugs. Thus, the roadblock supposedly would deter drug crimes. We conclude, however, that the government’s general deterrence interest is not substantial enough to outweigh the seized individuals’ liberty interests. When the Supreme Court has upheld sus-picionless roadblock seizures, the government’s purpose in establishing the roadblock had some logical connection to the individuals likely to be in the vehicles that were stopped. Thus, the police set up permanent checkpoints because the flow of illegal aliens could not otherwise be effectively controlled and the checkpoints could reasonably be expected to lead to the arrest of “smugglers and illegal aliens who succumb to the lure of such highways.” Martinez-Fuerte, supra, 428 U.S. at 554, 96 S.Ct. at 3081. Similarly, the government addressed the immense drunk driving problem by establishing sobriety checkpoints designed to prevent and apprehend impaired drivers. See Sitz, supra, 110 S.Ct. at 2487, 2488. The purported deterrence rationale for the Galberth roadblock, by contrast, was addressed to problems of general law enforcement, namely deterring drug traffic and violence and preventing *999 “violence, drugs and guns,” not to problems predictably associated with persons who are stopped at the roadblock. Such a justification is antithetical to the Fourth Amendment. See Delaware v. Prouse, supra, 440 U.S. at 659 n. 18, 99 S.Ct. at 1399 n. 18 (“the general interest in crime control” is insufficient to justify suspicionless stops). Moreover, there is no empirical evidence that the roadblock technique itself effectively promoted the government’s interest in deterring drug crimes. Indeed, common sense (see Sitz, supra, 110 S.Ct. at 2487), as well as the police use of “saturation” patrols, suggests that any disruption resulted from the highly visible police presence during the roadblock, and that any law enforcement technique involving a substantial police presence would have had a similar effect. The government offered no evidence to negate this common sense notion. 11 See Sitz, supra, 110 S.Ct. at 2487 (describing Delaware v. Prouse, supra, as a case in which “no empirical evidence indicated that such stops would be an effective means of promoting” the government’s purposes, and noting that “common sense” indicated to the contrary) (emphasis in original). Accordingly, in view of the insubstantiality of the government’s deterrence interest and the lack of any record evidence indicating that the Galberth roadblock furthered the government interest, the Brown v. Texas test weighs in favor of the individual’s liberty interest. We therefore hold that the Galberth roadblock violated the Fourth Amendment, and that his convictions must be reversed. B. The Taylor roadblock The trial judge in appellant Taylor’s case found: The primary purpose of the roadblock was to control traffic congestion associated with open-air drug markets such as cars double and triple parked and a large number of pedestrians moving through the neighborhood. This activity disrupts the lives of the people who are residents of the neighborhood and creates a safety hazard to them. The police expect to find motorists operating without permits, operating after revocation and operating with impaired ability to drive. Chief Fulwood stated, and the Court credits his testimony on the purposes of the roadblocks, that it is primarily a traffic enforcement program with certain other benefits clearly being derived from it. As in appellant Galberth’s case, we must accept this finding of the government purpose for the roadblock, which is supported by the record and has not been shown to be clearly erroneous. 12 The trial judge’s finding refers, however, to two different government interests. The trial judge first mentions as the “primary purpose” the concern with neighborhood traffic congestion associated with open air drug markets. In the second paragraph, however, the judge describes the problems of those who drive without a permit, after revocation, or with impaired capacity. The judge finally states that the roadblock was “primarily a traffic enforcement program.” We conclude that this finding leaves unclear whether the primary purpose of the Taylor roadblock was (1) to address neighborhood problems of traffic congestion associated with open air drug markets, or (2) to check driver’s licenses and automobile registrations. Because the latter purpose would support the constitutionality of the roadblock, while the *1000 former would not, we remand to the trial judge for a more specific finding. The trial judge relied heavily on the decision in McFayden, supra, 275 U.S.App.D.C. 207, 865 F.2d 1306, in which the United States Court of Appeals for the District of Columbia Circuit upheld a roadblock that was part of Operation Clean Sweep. In concluding that the roadblock had been conducted in a systematic and nondiscriminatory fashion, the court discussed both of the government interests found by Judge Huhn here. The McFayden court upheld the constitutionality of a roadblock, however, by relying on a finding that “the principal purpose of the roadblock was to regulate vehicular traffic by allowing police to check drivers’ licenses and vehicle registrations.” 275 U.S.App.D.C. at 213-14, 865 F.2d at 1312-13. The court noted the importance of this finding in assessing the legality of a police traffic roadblock, observing that “[i]t is possible that a roadblock purportedly established to check licenses could be located and conducted in such a way as to indicate that its principal purpose was the detection of crimes unrelated to licensing.” Id., 865 F.2d at 1312 (emphasis in original) (citing 4 LaFave, supra, § 10.8(a) at 63-64). “Such a subterfuge might result in an infringement of Fourth Amendment rights, and some courts have so held.” Id., 865 F.2d at 1312; see also Webb v. State, 739 S.W.2d 802 (Tex.Crim.App.1987) (holding unconstitutional a roadblock purportedly established to check for licenses and registrations which was in fact established for general law enforcement purposes). The first interest described by Judge Huhn is “to control traffic congestion associated with open-air drug markets such as cars double and triple parked and a large number of pedestrians moving through the neighborhood.” While the government may well have a substantial interest in reducing traffic congestion which involves more than general law enforcement, we are unable to conclude that the Taylor roadblock advanced this interest. Although police officers testified that the traffic congestion was alleviated during the roadblock’s operation, this effect seems most plausibly accounted for by the highly visible police presence during the roadblock. 13 Indeed, to the extent the roadblock itself had an effect separate from the police presence, it could only have increased the traffic congestion. See Commonwealth v. McGeoghegan, 389 Mass. 137, 143-45, 449 N.E.2d 349, 353 (1983) (commenting that vehicles were backed up two-thirds of a mile at a sobriety checkpoint). Finally, the arrest records from the roadblock do not indicate any arrests or tickets for offenses that relate to traffic congestion. Thus, there is “no empirical evidence [which] indicate[s] that [the Taylor roadblock] would be an effective means” of alleviating traffic congestion. Sitz, supra, 110 S.Ct. at 2487 (referring to Delaware v. Prouse, emphasis in original). If the principal interest behind the Taylor roadblock was to solve problems of traffic congestion, we would therefore conclude that the roadblock violated the Fourth Amendment. 14 The other government purpose to which the trial judge referred, however, is to check for expected problems of “motorists operating without permits, operating after revocation and operating with impaired ability to drive.” This purpose, as the Supreme Court has made clear, is substantial and legitimate. See Delaware v. Prouse, supra, 440 U.S. at 658, 99 S.Ct. at 1398 (“the States have a vital interest in ensuring that only those qualified to do so are permitted to operate motor vehicles [and] that these vehicles are fit for safe operation”). Although the Supreme Court has not yet upheld the constitutionality of roadblock stops for the purpose of checking licenses and registrations, dictum in Delaware v. Prouse suggests that such roadblocks would be constitutional. See id. at 663, 99 S.Ct. at 1401 (“Questioning of all on-coming traffic at roadblock-type stops is one possible alternative” not precluded by *1001 the opinion). The lower federal and state courts have generally upheld such roadblocks. See, e.g., United States v. Corral, 823 F.2d 1389, 1392 (10th Cir.1987), cert. denied, 486 U.S. 1054, 108 S.Ct. 2820, 100 L.Ed.2d 921 (1988); United States v. Miller, 608 F.2d 1089, 1093 (5th Cir.1979), cert. denied, 447 U.S. 926, 100 S.Ct. 3020, 65 L.Ed.2d 1119 (1980); Smith v. State, 515 So.2d 149, 152 (Ala.Crim.App.1987); People v. Meitz, 95 Ill.App.3d 1033, 1037-39, 51 Ill.Dec. 561, 565, 420 N.E.2d 1119, 1123 (1981); State v. Cloukey, 486 A.2d 143, 146-47 (Me.1985); Miller v. State, 373 So.2d 1004, 1005 (Miss.1979); State v. Shankle, 58 Or.App. 134, 137-38, 647 P.2d 959, 961-62 (1982). 15 Furthermore, the arrest rates for the Taylor roadblock indicate that it was sufficiently productive of the licensing and registration interest. Sergeant Canty testified that forty-two cars were stopped at the Taylor roadblock over the ninety minute period, and two arrests were made for unauthorized use of an automobile (namely, appellant Taylor and his passenger), meaning that arrests occurred involving approximately 2.4 percent of the stopped vehicles. This would appear to satisfy the requirement that there be something more than “a complete absence of data” indicating effectiveness. Sitz, supra, 110 S.Ct. at 2487 (holding that sobriety checkpoints at which “approximately 1.5 percent of the drivers passing through the checkpoint were arrested” were sufficiently productive); Delaware v. Prouse, supra, 440 U.S. at 659, 99 S.Ct. at 1399 (requiring “some” empirical evidence); McFayden, supra, 275 U.S.App.D.C. at 214, 865 F.2d at 1313. Finally, the evidence supports the trial judge’s findings that the roadblock was established in accordance with supervisory guidelines that removed discretion from the field officer and that the roadblock was readily visible. While prominent signposts appear more in line with what was contemplated by the police department guidelines, Operation Clean Sweep Manual, Pt. 1(B)(3), in view of the evidence before Judge Huhn about the publicity in the Fourth District regarding roadblocks and the judge’s finding regarding the roadblock’s visibility, we are satisfied that the minimum notice requirement was met. Nor was Taylor detained at the roadblock for an impermissibly long period of time. 16 Thus, under Brown v. Texas, the roadblock may not have been impermissibly intrusive of individual liberty. If the principal purpose of the roadblock were to check drivers’ licenses and vehicle registration, we would conclude that the roadblock was constitutional. Accordingly, we remand the record in appellant Taylor’s case for further findings as to whether the principal purpose of the roadblock was to check driver’s licenses and automobile registration or to alleviate neighborhood traffic congestion. The judgments of conviction of appellant Galberth are reversed. The record in appellant Taylor’s case is remanded for further findings. So ordered. Footnotes [1] Appellant Galberth pleaded guilty to carrying a pistol without a license, D.C.Code § 22-3204 (1989), possession of an unregistered firearm, id. § 6-2311, and unlawful possession of ammunition, id. § 6-2361. Appellant Taylor was found guilty, following a stipulated trial, of first-degree theft, id. §§ 22-3811, 3812(a), and unauthorized use of a vehicle, id. § 22-3815. We agree with appellant Taylor that his convictions for unauthorized use of a vehicle and first-degree theft merge. Contrary to the government’s contention, we are bound by a long line of cases beginning with Arnold v. United States, 467 A.2d 136 (D.C.1983), which hold that a conviction for unauthorized use of a motor vehicle merges with a conviction for theft of the same car. See Kirk v. United States, 510 A.2d 499 (D.C.1986); Garris v. United States, 491 A.2d 511 (D.C.1985); Arnold, supra. The government argues that the analysis of several United States Supreme Court cases, Garrett v. United States, 471 U.S. 773, 778-79, 105 S.Ct. 2407, 2411-12, 85 L.Ed.2d 764 (1985), Albernaz v. United States, 450 U.S. 333, 344, 101 S.Ct. 1137, 1145, 67 L.Ed.2d 275 (1981), and Whalen v. United States, 445 U.S. 684, 688-89, 100 S.Ct. 1432, 1435-36, 63 L.Ed.2d 715 (1980), requires us to reject Arnold. None of these decisions is controlling, however. “[T]he question of what punishments are constitutionally permissible is not different from the question of what punishments the Legislative Branch intended to impose.” Alb ernaz, supra, 450 U.S. at 344, 101 S.Ct. at 1145. Arnold effectively held that the legislature did not intend to punish someone for unauthorized use and theft of the same car. See Arnold, supra, 467 A.2d at 139. Even if, as the government contends, Arnold relied on a mistaken analysis, we are bound by the prior decision in the absence of a change in governing law. M.A.P. v. Ryan, 285 A.2d 310 (D.C.1971). See Byrd v. United States, No. 89-804 (D.C. Jun. 28, 1990) (petition for rehearing en banc granted) (whether Arnold and its progeny are inconsistent with Supreme Court decisions). [2] The guidelines also advise police officers to hand out cards to stopped motorists, explaining the procedures involved in the roadblock. O'Donnell had ordered these cards, but they were out of stock and could not be obtained in time for the roadblock. [3] A special system was set up in the Fourth District station whereby one person performed the WALES checks on a dedicated channel, thereby reducing the time required to make the check. [4] Canty testified that no arrests were made at 14th and Allison during the roadblock that had concluded at 9:00 p.m., just before the second roadblock at 14th and Quincy was set up. [5] The Fourth Amendment provides: The right of the people to be secure in their persons, houses, papers and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and persons or things to be seized. U.S. Const, amend. IV. [6] In Brignoni-Ponce, involving traffic stops to inquire about resident status, the Court held that some showing of a reasonable suspicion that a vehicle contained illegal aliens was required before a stop would be permissible, not *996 withstanding the modest nature of the intrusion. 422 U.S. at 881, 95 S.Ct. at 2580. [7] Under the guidelines, all vehicles were stopped, their drivers briefly examined for signs of intoxication, and where the officer detected such signs, the motorist would be directed to a location out of the flow of traffic where the officer would check the driver’s license and registration, and if warranted conduct further sobriety checks. Sitz, supra, at 2484. [8] This balancing test originated in Camara v. Municipal Court, 387 U.S. 523, 87 S.Ct. 1727, 18 L.Ed.2d 930 (1967). [9] We cannot agree with the government’s contention that Judge Dixon found that the "violence, drugs and guns" purpose was “intertwined” with "the traffic justification” for the roadblock. The judge was “not overwhelmed with the idea of traffic problems” near the roadblock, and in no way indicated that such problems were a purpose for the roadblock; quite the contrary, as he found that traffic alone would not have caused the police to set up the roadblock. [10] Although the Supreme Court has subsequently modified its statement in Carroll that vehicles on public highways have “a right to free passage without interruption” that cannot be violated absent probable cause, 267 U.S. at 154, 45 S.Ct. at 285, the passage quoted in the text retains vitality. See United States v. Ramsey, 431 U.S. 606, 618, 97 S.Ct. 1972, 1979-80, 52 L.Ed.2d 617 (1977) (quoting Carroll, supra). [11] Indeed, Assistant Chief Fulwood testified in the Taylor case that although the roadblocks were primarily part of a traffic enforcement program, there were certain other benefits, including the disruption of the open air drug markets as a result of the presence of "10 to 12 police officers in uniform with marked cars and flashlights and visibility jackets....” [12] Appellant Taylor mistakenly contends that the purpose of the roadblock should be treated as a question of law, reviewable de novo. The trial judge observed several witnesses who testified about the "who, what, where, when and how details of the” government’s purpose, which characterize questions of fact. Davis v. United States, 564 A.2d 31, 35 (D.C.1989) (en banc). Of course, the question whether the government's purpose satisfies the Brown v. Texas test is reviewable de novo. See Sitz, supra, 110 S.Ct. at 2485-86. [13] See note 11, supra. [14] We need not decide whether a police roadblock which was shown to alleviate traffic congestion could be constitutional. [15] Roadblocks have been found unconstitutional under state constitutions, however. See, e.g., State v. Koppel, 127 N.H. 286, 499 A.2d 977 (1985); State v. Farms, 523 So.2d 1293 (La.1988); Commonwealth v. Tarbert, 348 Pa.Super. 306, 502 A.2d 221 (1985); see also Webb v. State, supra, 739 S.W.2d 802 (relying on both the Texas and federal constitutions). [16] The instructions to the field officers who conducted the Taylor roadblock called for computer checks to be run on all of the vehicles that were stopped. This may have been an imper-missibly intrusive general practice. See Sitz, supra, 110 S.Ct. at 2485 ("Detention of particular motorists for more extensive testing may require satisfaction of an individualized suspicion standard”). As the instructions for computer checks were applied to appellant Taylor, however, who at the time he was stopped did not have a driver’s license, detention for the additional five minutes required for the expedited computer run was reasonable. Taylor’s appeal therefore does not raise the question whether the police may constitutionally run a computer check on a motorist who provides a valid license and registration. See Barrows v. Jackson, 346 U.S. 249, 255, 73 S.Ct. 1031, 1034, 97 L.Ed. 1586 (1953) (“Ordinarily, one may not claim standing ... to vindicate the constitutional rights of some third party”).