Indiana Law Review Intoxication Roadblocks R. George Wright* In State v. McLaughlin,^ the Indiana Court of Appeals was con- fronted with a fourth amendment issue arising from the defendant's motion to suppress evidence connected with his arrest at a roadblock for driving while intoxicated. The defendant driver had been arrested after failing a breathalyzer test administered at a police roadblock estab- lished '^primarily for the purpose of detecting and apprehending drunk drivers."^ In other words, the defendant had not been stopped primarily because he had aroused suspicion; the roadblock indiscriminately led to the gathering of the evidence. The court of appeals employed the balancing test used by the United States Supreme Court in Brown v. Texas? Under this balancing test, the reasonableness of a seizure upon suspicion not focused on a particular individual — as opposed to a situation where an individual is stopped because he or she has aroused suspicion — requires weighing "the gravity of the pubHc concerns served by the seizure, the degree to which the seizure advances the pubHc interest, and the severity of the interference with individual liberty.""^ For guidance in further explicating the balancing test in Brown, the court turned to a decision of the Kansas Supreme Court. ^ The Kansas court had, through its examination of a number of Supreme Court cases in various contexts, arrived at a list of thirteen explicitly non-exhaustive factors to be weighed somehow in the balance.^ The factors listed in State V. Deskins include: (1) The degree of discretion, if any, left to the officer in the field; (2) the location designated for the roadblock; (3) the time *Bigelow Fellow and Instructor of Law, University of Chicago. A.B., University of Virginia, 1972; Ph.D. Indiana University, 1976; J.D.; Indiana University School of Law—Indianapolis, 1982. '471 N.E.2d 1125 (Ind. Ct. App. 1984), transfer denied. May 3, 1985. McLaughlin has been quoted extensively in a recent New Jersey decision, State v. Kirk, 493 A.2d 1271, 1280-82 (N.J. Super. 1985). Ud. at 1129. The seizure occurred prior to any individualized suspicion falling upon the defendant. Id. at 1128-29. M43 U.S. 47, 50-51 (1979). Brown did not involve a drunk driving roadblock stop, and no other United States Supreme Court case has addressed the major issue raised in McLaughlin. ^Id. This test shares its obviously utilitarian foundations with the procedural due process test enunciated in Mathews v. Eldridge, 424 U.S. 319, 334-35 (1976). 'State V. Deskins, 234 Kan. 529, 673 P.2d 1174 (1983). 'Id. at 542, 673 P. 2d at 1185. 33 34 INDIANA LAW REVIEW [Vol. 19:33 and duration of the roadblock; (4) standards set by superior officers; (5) advance notice to the pubhc at large; (6) advance warning to the individual approaching motorist; (7) maintenance of safety conditions; (8) degree of fear or anxiety generated by the mode of operation; (9) average length of time each motorist is detained; (10) physical factors surrounding the location, type and method of operation; (11) the availability of less intrusive methods for combating the problem; (12) the degree of effec- tiveness of the procedure; and (13) any other relevant circum- stances which might bear upon the test.^ While it is clear that, in Indiana, the reasonableness of the search does not require that every factor be found favorable to the state, ^ the relative weight to be accorded each individual factor was left largely unspecified. The court of appeals did indicate in dicta that extreme or "unbridled discretion" of field officers, as opposed to supervisory of- ficers, would by itself dictate a finding of the unreasonableness of the search.^ Field discretion, however, is probably not unique in its potential decisiveness. For example, unreasonably and unnecessarily lengthy de- tentions of motorists subject to no individualized suspicion would sim- ilarly dictate a finding of unreasonableness. '^ If certain of the thirteen specified factors have the potential to be individually decisive of the case adversely to the state, it is arguable that others do not. The mere availability of less intrusive alternative procedures may be one example. The Supreme Court has recently ob- served in a somewhat different context that [a] creative judge engaged in post hoc evaluation of police conduct can almost always imagine some alternative means by which the objectives of the poHce might have been accomplished. But '*[t]he fact that the protection of the public might, in the abstract, have been accomplished by 'less intrusive' means does not, in itself, render the search unreasonable." The question is not simply whether some other alternative was available, but whether the police acted unreasonably in failing to recognize or to pursue it." Id., quoted in State v. McLaughlin, 471 N.E.2d 1125, 1135-36 (Ind. Ct. App. 1984), transfer denied. May 3, 1985. 'See 471 N.E.2d at 1136. "Id. (citing Delaware v. Prouse, 440 U.S. 648 (1979)). '"See United States v. Sharpe, 105 S. Ct. 1568, 1575 (1985). "Id. at 1576 (quoting Cady v. Dombrowski, 413 U.S. 433, 447 (1973)) (citations omitted). 1986] ROADBLOCKS 35 Without further elaboration of the weighing process, the court in McLaughlin went on to consider the thirteen factors in the context of the three-part balancing test in Brown. ^^ The court in effect undertook an elaborate balancing-within-a-balancing. The court concluded that Despite the gravity of the public concern for identifying and apprehending drunk drivers and the moderately low level of interference with individual liberty occasioned by the roadblock procedure, the state failed to present any evidence that the roadblock procedure advanced the public interest to a greater degree than would have been achieved by traditional methods of drunk-driving law enforcement, which are to be preferred because they are based upon a requirement of individualized suspicion.'^ McLaughlin was thus in a way an easy case. The state raised only a detection-and-apprehension interest, and defaulted on its burden of showing the superiority of the roadblock procedures in this regard. The case therefore provides little guidance for future cases in which some quantum of evidence is presented, or in which long or short-term de- terrence of drunk driving is the public interest alleged, or where a more easily demonstrable, but perhaps less crucial, public interest is raised, such as the expression of public concern over drunk driving, or the calUng of community attention to the drunk driving problem. The court in McLaughlin began its three-part analysis by finding the public concern served, or purportedly served, by the seizure to be *'very grave indeed.'*"* The court took judicial notice of the extent of '^See All N.E.2d at 1135-36. *^Id. at 1141. This formulation of the court's holding is reassuring in that it explicitly recognizes that the proper inquiry is not into "the degree of effectiveness of the procedure," see infra text accompanying note 57, but into the relative effectiveness of the particular roadblock technique, as contrasted with the efficacy of the available alternatives. The formulation is less satisfactory insofar as it may suggest that detection and apprehension of drunk drivers, as opposed to deterrence of drunk driving, must be the primary state interest weighed in the balance; it must be remembered, however, that the state in McLaughlin did not allege or attempt to show a deterrence effect. Id. at 1129. It should also be noted that both Brown and McLaughlin appear to equate "public concern" with "the public interest." See Brown, 443 U.S. at 50-51; McLaughlin, 471 N.E.2d at 1135, 1141. Wherever the court is to look to determine the public interest, it is at least arguable that it may look to other sources to attune itself to public concerns. Nor is it clear that every public concern is fully consistent with the public interest, in the sense of being either well-advised or constitutionally legitimate. 'M71 N.E.2d at 1137. 36 INDIANA LAW REVIEW [Vol. 19:33 the drunk driving problem," and cited Indiana legislative responses along with judicial recognition of the magnitude of the problem.'^ The court entered into its detailed discussion of the thirteen factors in examining the second element of the Brown balancing test — the degree to which the particular roadblock seizure advanced the public concerns involved.'" The court's crucial conclusion in this regard was that "[t]he "While one court has, in State ex rel. Ekstrom v. Justice Court of State, 136 Ariz. 1. 5, 663 P. 2d 992, 996 (1983) (en banc), declined to take comparable judicial notice, the McLaughlin court's willingness to do so seems sound. See id. at 8 n.2, 663 P.2d at 999 n.2 (Feldman, J., specially concurring); State v. Superior Court, 143 Ariz. 45, 48, 691 P. 2d 1073, 1076 (1984) (en banc) (minimizing the import of Ekstrom in this and other respects). '*471 N.E.2d at 1136 (citing South Dakota v. Neville, 459 U.S. 553 (1983), along with Ruge v. Kovach, 467 N.E.2d 673, 681 (Ind. 1984), in addition to recent Indiana statutory changes). '471 N.E.2d at 1137. The second and third Brown criteria, which subsume most of the thirteen factors discussed originally in Deskins and adopted in McLaughlin, are treated variously in the increasing number of state cases involving intoxication roadblocks. For a discussion of a number of these cases, which continue to reach divergent results on varying facts and states of the record, see Note, The Constitutionality of Roadblocks Conducted to Detect Drunk Drivers in Indiana, 17 Ind. L, Rev. 1065 (1984). See also Annot., 37 A.L.R.4th 10 (1985) (collecting routine roadblock cases without limitation to drunk driving stops, but already dated until supplemented). A sampling of additional recent law review articles includes: Note, Curbing the Drunk Driver under the Fourth Amendment: The Constitutionality of Roadblock Seizures, 71 Geo. L.J. 1457 (1983); Comment, The Prouse Dicta: From Random Stops to Sobriety Checkpoints'}, 20 Idaho L. Rev. 127 (1984); Comment, Filling in the Blanks after Prouse: A New Standard for the Drinking-Driving Roadblock, 20 Land & Water L. Rev. 241 (1985); Comment, Sobriety Checkpoint Roadblocks: Are They Constitutional in Light o/ Delaware v. Prouse?, 28 St. Louis U.L.J. 813 (1984); Comment, The Fourth Amendment Roadblock Against Detecting Drunk Drivers, 18 Suffolk U.L. Rev. 475 (1984). Probably the most thoughtful of the articles not dealing specifically with Indiana law is Grossman, Sobriety Checkpoints: Roadblocks to Fourth Amendment Protections, 12 Am. J. Crim. L. 123 (1984). A number of the cases in this area are too recent to have been included in the McLaughlin opinion. Among these cases are State v. Superior Court, 143 Ariz. 45, 691 P. 2d 1073 (1984) (en banc) (upholding DWI roadblock on fourth amendment challenge, thereby minimizing the import of State ex rel. Ekstrom v. Justice Court of State, 136 Ariz. 1, 663 P. 2d 992 (1983) (en banc), in which excessive on-site police discretion and a lack of evidence of the roadblock's superiority in apprehending drunk drivers led to a finding of unconstitutionality); Jones v. State, 459 So. 2d 1068 (Fla. Ct. App. 1984) (DWI roadblock unconstitutional due largely to inadequate factual record and, in particular, lack of evidence of comparative effectiveness in apprehending drunk drivers); State v. Golden, 171 Ga. App. 27, 318 S.E.2d 693 (1984) (driver's license checkpoint case; defend- ant's motion to suppress sobriety evidence denied on appeal in light of minimal field discretion, minimal delay, clear identification by signs as a police checkpoint); State v. Cloukey, 486 A.2d 143 (Me. 1985) (driving on revoked license case; roadblock for traffic safety check; roadblock constitutional despite only modest supervisory involvement in advance planning stages); Little v. State, 300 Md. 485, 479 A.2d 903 (1984) (checkpoint established both to detect and deter drunk driving; upheld on constitutional grounds based on Deskins factors despite extremely dubious and equivocal statistical and anecdotal evidence 1986] ROADBLOCKS 37 State . . . presented absolutely no evidence regarding the availability of less intrusive methods of law enforcement for combatting the drunk driving problem."'^ Similarly, the state, at the hearing on the defendant's motion to suppress, offered no evidence "of the inadequacy of the traditional method of enforcing DWI laws, nor of the superiority of the roadblock method of identifying and apprehending drunk drivers."''^ Two questions are raised by the court's analysis on this point. First, if it is legitimate to take judicial notice of the severity of the drunk driving problem, is it also legitimate for the court simply to infer the inadequacy of traditional methods from the magnitude and persistence of the problem?^^ Probably so, but this is of limited help to the state's case, since the superiority of the roadblock method in achieving any of various possible purposes is clearly inapt for resolution by judicial notice. 2' for the superior detection and deterrence effect of the checkpoint program); People v. ScoU, 63 N.Y.2d 518, 473 N.E.2d 1, 483 N.Y.S.2d 649 (1984) (DWI roadblock upheld despite rapid shifts in location; state required to show only a reasonable basis for an asserted deterrence or detection effect and not required to separate out the incremental deterrent effect of the roadblock from that of other components of the state's broader effort against drunk driving); People v. Torres, 125 Misc. 2d 78, 478 N.Y.S.2d 771 (Crim. Ct. 1984) (DWI roadblock permissible where established in non-arbitrary manner and discernible need established; decided without benefit of court of appeals opinion in Scott); State V. Smith, 674 P.2d 562 (Okla. Crim. App. 1984) (roadblock to detect license and registration violations; dismissal of DWI charge affirmed based on absence of statutory authority; exceptionally broad holding not apparently dependent upon any particular remediable set of roadblock procedures); State v. Schroeder, 66 Or. App. 754, 675 P.2d 1111, petition for review denied, 296 Or. 648, 678 P.2d 1227 (1984) (en banc) (denial of motion to suppress evidence affirmed; illuminating dissent to denial of petition for review by former professor and now Justice Linde). Thus, the differences not merely in result, but in the standards to be imposed upon the state, are evident even in the state cases post-dating McLaughlin. In light of cases such as Scott and Smith, it can easily be argued that the cases since McLaughlin have polarized even further along crucial dimensions, and that Indiana must at some point take a position on the key questions not reached in McLaughlin. It should be noted that there is essentially no helpful prior Indiana authority, although the court in Irwin v. State, 178 Ind. App. 676, 681, 383 N.E.2d 1086, 1089 (1978), suggested in dicta that "no one questions the right of law enforcement officers to establish a roadblock to conduct a routine traffic check of all vehicles and drivers passing through that point during a given period of time." '«471 N.E.2d at 1137. ^°This would not itself be an instance of judical notice, as judicial notice is typically thought of in terms of fact, rather than evaluation of fact. See, e.g.. Glover v. Ottinger, 400 N.E.2d 1212, 1214 (Ind. Ct. App. 1980). ^'Matters such as the detection or deterrence value, or the superiority of intoxication roadblocks are subject to reasonable dispute. See H. Ross, Deterring the Drinking Driver 110 (1982). As such, they are to be established by evidence rather than judical notice. See, e.g., Daar v. Yellow Cab Co., 67 Cal. 2d 695, 716, 433 P.2d 732, 747, 63 Cal. Rptr. 724, 739 (1967) (en banc). 38 INDIANA LAW REVIEW [Vol. 19:33 Second, it must be asked whether the court is in fact moving toward simply assuming the adequacy of traditional means of detecting intox- ication. The court observed that ''[n]o doubt, police officers are well trained to identify such indicators as weaving between lanes, failing to signal a turn, speeding, etc."-^ The assumption that the police are capable of non-intrusively detecting drunken drivers with some accuracy is cer- tainly widely held.-^ This confidence, however, is probably misplaced. ^^ No legal harm follows from any such false assumption, though, if detection of drunk drivers is no easier at roadblocks than through ordinary patrolhng.^^ The court in McLaughlin concluded its analysis of the second Brown element by determining that a potential state interest in deterring, as opposed to detecting, drunken driving was "not present in this case."^^ Certainly, the state made no substantial attempt to establish empirically the magnitude of any deterrent effect. The court should, however, have found the relative deterrent effect of the roadblock program in question to have been unproven, rather than apparently assuming the deterrence effect to be non-existent. While it is true that the police did not widely publicize in advance the roadblock program's general features^^ in order "471 N.E.2d at 1137. '^See, e.g.. State ex rel. Ekstrom v. Justice Court of State, 136 Ariz. 1, 5, 663 P. 2d 992, 996 (1983) (en banc); People v. Hartley, 125 111. App. 3d 575, 578, 466 N.E.2d 346, 348 (1984) ("[a]n intoxicated motorist can be easily discerned by a trained officer without having to stop all traffic at a roadblock"), criticized. People v. Conway, 135 111. App. 3d 887, 482 N.E.2d 437 (1985), rev'd. People v. Hartley, No. 60593 (111. Nov. 21, 1985); State V. Deskins, 234 Kan. 529, 545, 673 P.2d 1174, 1187 (1983) (Prager, J., dissenting); State V. Cloukey, 486 A.2d 143, 147 (Me. 1985). '"See Grossman, Sobriety Checkpoints: Roadblocks to Fourth Amendment Protections, 12 Am. J. Crim. L. 123, 159 n.l99 (1984). See also id. at 158 (estimate that 7-8% of those driving at night are legally intoxicated), & 162 (estimate that for every 2,000 trips taken by drunk drivers, only one will result in single arrest). It is difficult to imagine how the probability of arrest could be 0.00044 if signs of driving while intoxicated were readily detectable by trained persons using non-intrusive means. See M. Ross, Deterring THE Drunk Driver 107 (1982). ^'See, e.g., State ex rel. Ekstrom v. Justice Court of State, 136 Ariz. 1, 2, 663 P. 2d 992, 993 (1983) (en banc) (5,763 vehicles stopped at roadblocks; 13 DWI arrests); State V. Deskins, 234 Kan. 529, 545, 673 P.2d 1174, 1187 (1983) (between 2,000 and 3,000 vehicles stopped; 15 DWI arrests) (Prager, J., dissenting); People v. Scott, 63 N.Y.2d 518, 523, 528 n.3, 473 N.E.2d 1, 2, 5 n.3, 483 N.Y.S.2d 649, 650, 653 n.3 (1984) (evidence that ten percent of drivers intoxicated during prime weekend late evening hours, but roadblocks conducted largely during these hours resulted in fewer than 1/10 of one percent being arrested for DWI). The ratio was somewhat better in Jones v. State, 459 So. 2d 1068, 1079 (Fla. Ct. App. 1984), where five or six DWI arrests resulted from the roadblock stop of between only 100 and 200 cars. This ratio was roughly matched by the particular roadblock at issue in McLaughlin, where three DWI arrests resulted from the stopping of 115 cars over the period of an hour. 471 N.E.2d at 1137. ^"471 N.E.2d at 1138. ^'Id. 1986] ROADBLOCKS 39 to maximize deterrence, it is certainly conceivable that a series of * 'eight to ten roadblocks in Tippecanoe County during September, 1982"^*^ would, without additional official publicity, generate at least a localized, temporary deterrent effect based solely on word of mouth communication. All that may be required for deterrence is the perceived possibility of the roadblock's recurrence, not official advance notice.^^ The remainder of the dozen factors listed in Deskins were discussed, finally, in connection with the third element of the Brown test, the severity of the interference with the defendant's liberty caused by the seizure in question. -'^ The court did not find this factor to be decisive. The court found that the objective intrusion on the detained motorists' fourth amend- ment rights was relatively low; the subjective intrusion caused by the physical characteristics of this roadblock was somewhat higher, due to its isolated location, questionable lighting, and absence of warning signs; and the perceived and actual discretion left to the officers in the field was adequately controlled. . . .^' It does seem clear that the presence or absence of these factors should not be examined with equal degrees of scrutiny. There is, for example, no excuse for failure to post illuminated warning signs indicating the precise purpose of the impending stop. The cost in money, or in detection or deterrence, is certainly minimal, and the benefit in reduced driver anxiety is clearly significant.^^ On the other hand, factors such as the most expeditious location for the roadblock, or even the appropriate degree of lighting, should be judicially reviewed only with great restraint, out of deference to the professional and technical expertise of the police. The police have no obvious incentives to locate the roadblock frivolously in unproductive locations. They may well be confronted with a delicate tradeoff between the productivity of the roadblock and the safety of all those concerned." -'Id. at 1137 n.6. -^While advance notice of the precise time and place of roadblocks maximizes neither apprehension nor deterrence, advance publicity of a more general nature may maximize deterrence while resulting in a lesser reduction in apprehension. Some degree of tradeoff between deterrence and apprehension may be inevitable, and the state should be wary of designing a roadblock program in an effort to accomplish both purposes, lest it be unable to demonstrate the superiority of roadblocks in either respect. See Jones v. State, 459 So. 2d 1068, 1076 (Fla. Ct. App. 1984). ^«471 N.E.2d at 1138. ''Id. at 1141. ^The court in State ex rel. Ekstrom v. Justice Court of State, 136 Ariz. 1, 5, 663 P. 2d 992, 996 (1983) (en banc) noted a lack of warning signs or advance flashing lights and inferred an unnecessary degree of driver suprise. "Safety is the seventh Deskins factor cited in McLaughlin, 471 N.E.2d at 1135, and 40 INDIANA LAW REVIEW [Vol. 19:33 Judicial attempts at second-guessing such determinations would seem generally unnecessary.'''^ The court in McLaughlin determined that the level of actual and perceived operational discretion involved fell within acceptable Hmits because the roadblock was conducted pursuant to "previously specified neutral criteria"-' and because "every car that arrived at the roadblock site was stopped. "^^ The balance of the third Brown factor was thus not itself adverse to the state. ^^ However, the overall balance of the three factors in Brown dictated the affirmance of the trial court order granting the defendant's motion to suppress. The state in McLaughlin was held to have "failed to meet its burden of proving the reasonableness of the warrantless seizure of defendant. . . ."^^ There was a fatal absence of evidence for the su- periority, in terms of detection or deterrence, of the roadblock method. ^^ The court concluded by warning in dicta that a showing by the state of a deterrent effect of roadblocks generated by advance publicity should take account of the possibility that a similar publicity blitz associated with traditional procedures to detect and deter drunk drivers used in an intensified manner would be equally productive/*^ a concern for safety may account for any "isolation" of the roadblock, as found in id. at 1141. ^The Supreme Court has said in a somewhat different context that "[w]e may assume that . . . officials will be unhkely to locate a checkpoint where it bears arbitrarily or oppressively on motorists as a class." United States v. Martinez-Fuerte, 428 U.S. 543, 559 (1976). '-471 N.E.2d at 1141 n.8, in accordance with Delaware v. Prouse, 440 U.S. 648, 662 (1979). M71 N.E.2d at 1139. A number of the cases involve what could easily appear to be arbitrary selection from the driver's standpoint, such as stopping every third or fifth car, or allowing enough cars to pass to avoid undue congestion, but such practices are generally not condemned. See, e.g.. People v. Scott, 63 N.Y.2d 518, 526, 473 N.E.2d 1, 4, 483 N.Y.S.2d 649, 652 (1984) (citing authority). "It should be noted that the courts often further subdivide their inquiry into the subjective intrusiveness of the roadblock by considering such questions, inspired by Mar- tinez-Fuerte, 428 U.S. at 558, as whether the driver could "see visible signs of the officers' authority." See McLaughlin, 471 N.E.2d at 1139. See also State v. Golden, 171 Ga. App. 27, 29, 318 S.E.2d 693, 695 (1984); People v. Torres, 125 Misc. 2d 78, 82, 478 N.Y.S.2d 771, 774 (Crim. Ct. 1984). Until poHce roadblocks are conducted in unmarked police vehicles, or unauthorized highway brigandage becomes more prominent, this factor may safely be dropped from consideration. '"471 N.E.2d at 1141. See also State v. Goins, 16 Ohio App. 3d 168, 172, 474 N.E.2d 1219, 1222 (1984) (inquiry into police "reasonableness" under all of the circum- stances of the checkpoint). '471 N.E.2d at 1141. *"Icl. at 1142. It is not settled whether a proper balancing would compare the effects of a roadblock program against those of a particularly "concentrated effort" using traditional techniques. Presumably, no roadblock will be as effective as the police's ignoring all other tasks, but using only traditional means, to detect drunk drivers. 1986] ROADBLOCKS 41 In view of the state's failure to adduce any evidence on the question of the relative efficacy of roadblocks, or to advance any public interest other than detection and apprehension, it is clear that McLaughlin raises far more questions than it answers. Crucial questions, such as the kinds of evidence that are acceptable, and the nature of the weighing process involved in passing on questions of relative effectiveness, remain unanswered. It is unclear how the courts will react to masses of conflicting statistical data, and how they will review trial court findings on such potentially complex empirical ques- tions. As the perceived difficulty of evaluating the statistical evidence of the relative efficacy of roadblock detentions threatens to become un- manageable, though, the courts are under greater pressure simply to cut, rather than painstakingly unravel, the Gordian knot of statistical evidence. The First District of the Indiana Court of Appeals has recently taken just this tack, disagreeing with McLaughlin, in State v. Garcia^^ In Garcia, the court of appeals reversed the trial court's granting of the defendant driver's motion to suppress evidence of his driving while intoxicated where the evidence was obtained through a procedurally well-conducted temporary roadblock. "^^ Most of the numerous Deskins factors'*^ were found favorably to the state, and during the two-hour operation of the roadblock, the stopping of approximately 100 vehicles netted seven DWI arrests. "^^ The essential difference between Garcia and McLaughlin is the un- willingness of the Garcia court to consider the relative efficacy of the roadblock in question as contrasted with more traditional and less in- trusive means of enforcing drunk driving laws.'*^ The court in Garcia was concerned largely with the presence or absence of "unbridled discretion and standardless conduct of an officer in the field'"*^ and saw the crucial constitutional task as simply one of striking "a balance between the public interest in highway safety, which includes ridding the roads of drunk drivers, and the individual's right to personal security from arbitrary interference by law officers. "-^^ What the court in Garcia did not acknowledge is that the concern for "the degree to which the seizure advances the public interest" in ^'481 N.E.2d 148 (Ind. Ct. App. 1985). *^Id. at 154. The roadblock in Garcia may have had better advance general publicity than that involved in McLaughlin, cf. supra note 27 and accompanying text, but the court in Garcia did not rest its analysis on this distinction. ^-See 481 N.E.2d at 152-53 in the context of supra note 7 and accompanying text. ''Id. at 150. ''See id. at 153-54. ''Id. at 152. "Id. 42 INDIANA LAW REVIEW [Vol. 19:33 reduced highway carnage derives ultimately from the language of the United States Supreme Court/^ and that a fair reading of this requirement leaves the limitation of police arbitrariness and discretion as a "central," but not exclusive, concern/^ No one denies that roadblock procedures are more intrusive than traditional procedures under which a driver whose conduct gives no grounds for reasonable articulable individualized suspicion may generally proceed unmolested by the authorities. The concern of the McLaughlin court, as opposed to the Garcia court, is that this greater intrusiveness on fundamental individual constitutional rights be counterbalanced by a showing that the roadblock in question has some compensating ad- vantage. The logic of McLaughlin in interpreting Brown is that there is no point in countenancing exceptionally burdensome law enforcement techniques if those techniques are no more efficacious, or even less efficacious, than less burdensome techniques. Defenders of the Garcia approach can offer several reasonable re- sponses in addition to noting the potential difficulties inherent in having the courts grapple with competing statistical analyses. There is undoubt- edly a distinction that should be respected between dictating police techniques and dictating constitutional limitations on police techniques. ^° It is also possible to argue that the "obvious failure of the so-called traditional methods"^' should give rise to some presumption that alter- native techniques, including roadblocks, will be more effective. A plea for time to develop adequate statistics for comparative purposes can be made," although this controversially assumes that the welter of currently available statistics is legally inadequate to settle the issue, even tem- porarily, until better statistics can be developed to resolve the empirical questions more definitively. What should clearly be resisted, though, is any temptation to eyeball the number of arrests per hour," or even per officer per hour, or the phenomenon of drivers' avoiding the roadblock by turning around,^"* and assume that superior deterrence has been shown or that it can be inferred that the McLaughlin approach results in unnecessary highway tragedies. ^^ ^"Brown v. Texas, 443 U.S. 47, 50-51 (1979). It is unclear how roadblocks would advance the public interest at all if it were to be conceded that they were less effective than equally intense use of traditional techniques. '"Id. at 51. '"Cf. Garcia, 481 N.E.2d at 152. ''Id. ''Id. ''Id. at 154. ''Id. "See id. 1986] ROADBLOCKS 43 The ultimate criticism of the Garcia approach remains that it imposes obvious costs in civil Uberties without any assurance of compensations in increased public safety. The issue is not the magnitude or difficulty of the drunk driving problem, but the avoidance of incurring potentially pointless losses of basic civil liberties. The Indiana Supreme Court may eventually adopt the Garcia approach based on practical considerations. This Article, however, will not assume so, as this would be a virtually unique instance of applying what amounts to merely a due process ''rational basis" test with respect to the substantial burdening of fun- damental rights and civil liberties. ^^ It is sometimes suggested that either the legislature or the attorney general should promulgate uniform roadblock operational standards." The value of this step is unclear, given the unsettled state of Indiana law in this regard. The thirteen factors enumerated in Deskins provide some guidance. ^^ Adherence to statutory or administrative guidelines would not guarantee constitutionality, and departure from the guidelines would not necessarily result in an unconstitutional seizure. In the absence of decisive, methodologically unimpeachable, uncon- tradicted, and plainly applicable controlled studies, it is also unclear whether the state should abandon its detection and apprehension rationale and focus instead on building a technically sound case for the relative deterrence value, at least over the short term,^^ of its roadblock program. While the cases do not permit a rigorous comparison of the cost- effectiveness of roadblocks and of routine patrolling, the productivity in terms of detection and apprehension per police officer-hour of road- blocks may well not exceed that of conscientious routine patrolling, other things being equal. ^^ This is not to suggest that the unique deterrence value of roadblocks is easy to establish, although some courts have been satisfied in this regard.^' Indiana has not yet determined how deeply the courts will ^'^This is, of course, not to suggest that there is no authoritative support for the Garcia approach in this particular area. One of the most recent such qases is State v. Martin, 496 A.2d 442 (Vt. 1985), in which the Vermont Supreme Court simply assumed some level of deterrence attributable to the roadblock, and seemed disinclined to require any sort of a showing of any greater deterrence flowing from the roadblock technique than from less intrusive techniques. Id. at 447-48. ''See, e.g.. State v. Deskins, 234 Kan. 529, 543, 673 P.2d 1174, 1185-86 (1983); Commonwealth v. McGeoghegan, 389 Mass. 137, 449 N.E.2d 349, 353 (1983). '^Deskins, 234 Kan. at 541, 673 P.2d at 1185. ^'For a variety of drunk driver programs, longterm effects have been hard to demonstrate. See M. Ross, Deterring the Drunk Driver 103 (1982). ^'See the examples cited at supra note 25. ""'See, e.g.. State v. Superior Court, 143 Ariz. 45, 48, 691 P. 2d 1073, 1076 (1984); Little V. State, 300 Md. 485, 504, 479 A.2d 903, 913 (1984). The dissenting opinion in Little is cogently argued in this regard. 44 INDIANA LAW REVIEW [Vol. 19:33 plunge into the task of sifting out methodologically flawed statistical studies from the welter of inconsistent claims. Even such preliminary tasks as determining whether a study from another time and jurisdiction should be accorded substantial weight as evidence of a deterrent effect of an Indiana program present formidable difficulties. Just as the courts may wish to discourage litigation based on only minimal departures from an optimally conducted roadblock, so they may wish to avoid having every roadblock case turn into a battle of statisticians. From this standpoint, it is tempting to adopt New York's "reasonable basis" test,^' rather than rigorously pursuing the difficult question of the roadblock's identifiable deterrence value. The temptation to simplify the inquiry by moving closer to a min- imum scrutiny equal protection or substantive due process standard^^ should be resisted. In roadblock cases, it should be recalled that the state is ultimately seeking to impose a criminal sanction and to engage in procedures which specially burden the privacy and travel rights of numerous innocent drivers. ^'^ It is doubtful that such an intrusion should be permitted merely because there is a "reasonable basis"^^ for believing that there is some practical reason for doing so. Unfortunately, the process of avoiding extreme and obviously flawed solutions may not help the courts to crystallize a uniquely justifiable moderate approach. A moderate standard, though, should include judicial deference to most limited departures from allegedly optimal roadblock procedures. ^^ Beyond this obvious step to discourage undue litigation, the decisions could reasonably take various directions. One approach to the knotty problem of duly weighing conflicting statistical evidence on the claimed superior deterrence effect of roadblock programs would start by generally admitting into evidence otherwise competent studies from jurisdictions other than Indiana. ^^ Methodological criticisms of such studies should generally go to their evidentiary weight, and not to their admissibility.^^ Techniques should be devised to minimize ''See People v. Scott, 63 N.Y.2d 518, 529, 473 N.E.2d 1, 6, 483 N.Y.S.2d 649, 654 (1984). "'See, e.g., Lindsley v. Natural Carbonic Gas Co., 220 U.S. 61, 78 (1911); Linde, Due Process of Lawmaking, 55 Neb. L. Rev. 197, 201 (1976); Note, Legislative Purpose, Rationality, and Equal Protection, 82 Yale L.J. 123, 124-25 (1972). '"See generally Grossman, Sobriety Checkpoints: Roadblocks to Fourth Amendment Protections, 12 Am. J. Crim. L. 123 (1984). See also State v. Schroeder, 296 Or. 648, 649, 678 P. 2d 1227, 1228 (1984) (Linde, J., dissenting to denial of petiton for review). '''See supra note 49 and accompanying text. '^See supra notes 32-34 and accompanying text. '•'National, as opposed to exclusively New York-based, statistics were apparently examined and relied on for at least limited purposes in People v. Scott, 63 N.Y.2d 518, 526-27, 473 N.E.2d 1, 4, 483 N.Y.S.2d 649, 652 (1984). '^See McCoRMicK on Evidence §§ 208, 209 (E. Cleary 3d ed. 1984). For a general 1986] ROADBLOCKS 45 any necessity for continually relitigating the validity of leading studies as roadblock cases continue to arise nationally. The critical superior deterrence effect issue might best be resolved through the following judicial standard: if the statistical evidence as to the alleged detection or deterrence superiority of intoxication roadblocks remains substantially irreconcilable, conflicting, or inconclusive after full opportunity for cross-examination and adversary commentary, or if the state's own evidence by itself is judicially deemed to be unduly im- pressionistic, anecdotal, or materially deficient, the state has not dis- charged its burden of justifying the special intrusiveness of the intoxication roadblock. ^^ One can only hope that additional empirical studies will simplify, rather than further complicate, the question. introduction to some of the statistical techniques relevant to adjudicating claims of de- terrence value, see Fisher, Multiple Regression in Legal Proceedings, 80 Colum. L. Rev. 702 (1980). While competent statistical analysis cannot by itself resolve the underlying issues of public policy, McCormick, supra at 646 ("Deciding what level of confidence is appropriate in a particular case ... is a policy question and not a statistical issue"), some courts in other contexts are beginning to insist upon statistical methodologies more sophisticated than those commonly employed thus far in roadblock cases. See, e.g., Moultrie V. Martin, 690 F.2d 1078, 1082 (4th Cir. 1982) ("in all cases involving racial discrimination, the courts of this circuit must apply a standard deviation analysis . . . before drawing conclusions from statistical comparisons"). Of course, it is possible that roadblocks either invariably or never have superior apprehension or deterrence effects, where no such firmly established broad principle can be applied in individual racial discrimination cases. "^The point is in part to control the demands on judicial technical expertise while avoiding merely speculative or intuitive conclusions on the relative effectiveness of the roadblocks.