Harvard Law Review Harvard Law Review Harvard Law Review

Constitutional Law

Kansas v. Glover

The Supreme Court once considered “reasonable suspicion” to be “one of the relatively simple concepts embodied in the Fourth Amendment.”1×1. United States v. Sokolow, 490 U.S. 1, 7–8 (1989). Yet this ostensibly simple concept has eluded lower courts ever since Terry v. Ohio.2×2. 392 U.S. 1 (1968); see also Andrew Guthrie Ferguson, Big Data and Predictive Reasonable Suspicion, 163 U. Pa. L. Rev. 327, 338–40 (2015). Last Term, in Kansas v. Glover,3×3. 140 S. Ct. 1183 (2020). the Supreme Court held that a police officer had reasonable suspicion to stop a vehicle after he learned that the owner of the vehicle had a revoked license.4×4. Id. at 1186. The Court based its holding on common sense and discussed empirical evidence only in dicta. In so doing, the Court confused — rather than clarified — the ongoing debate about the role of empirical analysis in reasonable suspicion determinations. In the meantime, it offered lower courts no alternative method for judging the “common sense” of an officer’s inference under similar facts.

While on patrol in April 2016, Deputy Mark Mehrer ran the license plate number of a 1995 Chevrolet pickup truck through the Kansas Department of Revenue database.5×5. State v. Glover, 422 P.3d 64, 66–67 (Kan. 2018); accord State v. Glover, 400 P.3d 182, 184 (Kan. Ct. App. 2017). He learned that the vehicle’s registered owner, Charles Glover, Jr., had a revoked driver’s license.6×6. Glover, 422 P.3d at 66–67; accord Glover, 400 P.3d at 184. Deputy Mehrer initiated a traffic stop, assuming that Glover was driving the truck, even though he did not observe the driver commit any traffic violation or try to confirm the driver’s identity.7×7. Glover, 422 P.3d at 66–67; accord Glover, 400 P.3d at 184.

Glover was indeed the driver and was charged with driving as a habitual violator — driving repeatedly without a valid license.8×8. Glover, 140 S. Ct. at 1186–87 (citing Kan. Stat. Ann. § 8-285(a)(3) (2001)). Glover moved to suppress all evidence seized during the traffic stop.9×9. Id. at 1187. Instead of testifying at the hearing, the parties entered into a brief seven-paragraph stipulation of facts.10×10. Id.; see also id. at 1196 (Sotomayor, J., dissenting). The trial court granted Glover’s motion to suppress because Deputy Mehrer did not have reasonable suspicion to stop his truck.11×11. Glover, 400 P.3d at 184–85. The Kansas Court of Appeals reversed, holding that Deputy Mehrer had reasonable suspicion to initiate the traffic stop because he “kn[ew] the registered owner of the vehicle ha[d] a suspended license” and was “unaware of any” contradictory “evidence or circumstances” indicating that the driver was not the vehicle’s owner.12×12. Id. at 188.

The Kansas Supreme Court reversed.13×13. State v. Glover, 422 P.3d 64, 66 (Kan. 2018). The court rejected the “owner-is-the-driver presumption” in part because it turned on “two unstated assumptions.”14×14. Id. at 69. The court noted that even if the two assumptions were accepted as valid inferences, Kansas law would not allow stacking one assumption on another. Id. at 70. First, it was erroneous to assume that the registered owner of a vehicle was also its primary driver when common community experience suggests people often drive vehicles registered to their family members.15×15. See id. at 69. Second, it was impermissible to assume that a vehicle’s owner had a “broad and general criminal inclination,” and, in this case, that he would likely continue to drive in disregard of the revocation or suspension order.16×16. Id. at 70. In addition, the court held that the owner-is-the-driver presumption impermissibly shifted the burden of proof to Glover. Id. at 70–71.

The U.S. Supreme Court reversed and remanded.17×17. Glover, 140 S. Ct. at 1191. Writing for the Court, Justice Thomas18×18. Every member of the Court save Justice Sotomayor joined the majority opinion. held that an officer has reasonable suspicion to stop a vehicle when he “learn[s] that the registered owner has a revoked driver’s license” and “lacks information negating an inference that the owner is the driver of the vehicle.”19×19. Glover, 140 S. Ct. at 1186. The Court began its analysis with a brief overview of precedent, noting that the reasonable suspicion standard is considerably less onerous than proof by a preponderance of the evidence or probable cause.20×20. Id. at 1187. The standard does not “demand scientific certainty . . . where none exists” and allows officers to make “commonsense . . . inferences about human behavior.”21×21. Id. at 1188 (quoting Illinois v. Wardlow, 528 U.S. 119, 125 (2000)).

Guided by precedent, the Court turned to the facts of the case, which it concluded permitted Deputy Mehrer to draw the “commonsense inference” that Glover was the driver of the vehicle and gave him “more than reasonable suspicion” to stop the vehicle.22×22. Id. The Court argued that the fact that the registered owner is not always the driver did not alter the analysis because the same logic could have applied to all reasonable inferences.23×23. Id. Nor did Glover’s revoked license negate a finding of reasonable suspicion, as “common experience readily reveals,” and empirical studies have confirmed, that drivers with revoked licenses often keep driving.24×24. Id. Moreover, because Kansas law revokes licenses for drivers “who have already demonstrated a disregard for the law,” the Court said, it was all the more reasonable to infer that Glover might continue to drive.25×25. Id. at 1188–89 (citing scattered sections of the Kansas Statutes Annotated that authorize license revocation upon a driver’s conviction for certain felonies or various other offenses).

In reaching its decision, the Court addressed and rejected two claims.26×26. See id. at 1189. First, it rejected the claim that an officer needs to call on law enforcement training or experience to justify a stop.27×27. See id. In contrast, the Court had repeatedly recognized that an officer need not draw inferences based solely on his law enforcement training and experience.28×28. Id. Here, Deputy Mehrer’s inference was a reasonable one “made by ordinary people on a daily basis.”29×29. Id. If professional expertise were the only justification for an inference, courts would “impose on police the burden of pointing to specific training materials or field experiences . . . for [a] myriad [of] infractions”30×30. Id. at 1190. and “impermissibly tie a traffic stop’s validity to the officer’s length of service.”31×31. Id. (citing Devenpeck v. Alford, 543 U.S. 146, 154 (2004)). Second, the Court made clear that its holding drew on common sense, not just probabilities.32×32. See id. Though officers may “rely on probabilities,” in this case Deputy Mehrer also “used common sense to form a reasonable suspicion.”33×33. Id. (emphasis added).

Finally, the Court stressed the narrow scope of its holding.34×34. Id. at 1191. Because an officer’s traffic stop “must be ‘justified at its inception’”35×35. Id. (quoting Hiibel v. Sixth Jud. Dist. Ct., 542 U.S. 177, 185 (2004) (internal quotation marks omitted)). when “tak[ing] into account the totality of the circumstances,”36×36. Id. (quoting Navarette v. California, 572 U.S. 393, 397 (2014)). “the presence of additional facts might dispel reasonable suspicion.”37×37. Id. (citing Terry v. Ohio, 392 U.S. 1, 28 (1968)). For example, an officer would lack reasonable suspicion if he knew the owner was an elderly male but observed the driver was a young female.38×38. See id. Because Deputy Mehrer did not possess such “exculpatory information,” the Court held the stop reasonable.39×39. Id.

Justice Kagan40×40. Justice Kagan was joined by Justice Ginsburg. concurred to highlight the decision’s narrow scope.41×41. See Glover, 140 S. Ct. at 1194 (Kagan, J., concurring). She joined the Court’s opinion for two reasons. First, Deputy Mehrer knew about Glover’s “proclivity for breaking driving laws” because “Kansas almost never revokes a license except for serious or repeated driving offenses.”42×42. Id. at 1192. Second, this case “strange[ly]” rested on a “barebones stipulation,” in which Glover offered no additional facts for rebuttal.43×43. Id. at 1194. Justice Kagan would have treated Glover differently had Kansas suspended, rather than revoked, Glover’s license.44×44. Id. at 1192. “[N]o similar evidence” could have then indicated his “penchant for ignoring driving laws” because “most license suspensions do not relate to driving at all.”45×45. Id. Even when an initial presumption of reasonable suspicion is warranted, a defendant may still challenge that presumption with additional information, such as “the attributes of the car” or “statistical evidence.”46×46. Id. at 1193.

Justice Sotomayor dissented.47×47. Id. at 1194 (Sotomayor, J., dissenting). She characterized the majority opinion as “dramatically alter[ing] both the quantum and nature of evidence” a state may produce to prove reasonable suspicion.48×48. Id. First, she argued that the majority flipped the burden of proof by permitting traffic stops unless officers have information suggesting that a vehicle’s unlicensed owner is not its driver.49×49. Id. at 1195. In addition, she explained that reasonable suspicion requires an officer’s — as opposed to a layperson’s or a judge’s — common sense.50×50. Id. at 1195–96. The majority’s reliance on judicial inferences “promotes broad, inflexible rules that overlook regional differences.”51×51. Id. at 1196. Justice Sotomayor also disapproved of the majority’s use of empirical studies because statistics do not capture the individualized evidence requisite for any reasonable suspicion inquiry.52×52. See id. at 1197. She thus warned that the Court “paved the road to finding reasonable suspicion based on nothing more than a demographic profile.”53×53. Id. Moreover, Justice Sotomayor criticized the majority’s approach for “mak[ing] scant policy sense.”54×54. Id. at 1198. She argued that if states can decline to list the considerations that contributed to a reasonable suspicion determination, defendants will have a difficult time raising contrary evidence to challenge a stop.55×55. See id. at 1197–98. Finally, she noted that “the majority’s distinction between revocation and suspension may not hold up in other jurisdictions.”56×56. Id. at 1198.

By doubling down on common sense and referencing empirical data only in passing, Glover increased confusion about the appropriate role of empirical data in reasonable suspicion determinations without explaining how else lower courts should evaluate the “common sense” of an inference. Glover was a clear opportunity for the Court to weigh in on the scholarly debate about whether empirical evidence should be used to evaluate reasonable suspicion. But the Court declined to do so. Instead, it introduced additional uncertainty about when and to what extent lower courts may use such evidence to decide reasonable suspicion cases. At the same time, it failed to explain why Deputy Mehrer’s inference was common sense, leaving lower courts with no easy way to apply Glover to similar factual circumstances.

The Court’s reasonable suspicion jurisprudence has long rested on a totality-of-the-circumstances approach that lacks clarity and fails to provide needed guidance to lower courts. The Court has crafted an “‘I know it when I see it’ jurisprudence,”57×57. Andrew Manuel Crespo, Probable Cause Pluralism, 129 Yale L.J. 1276, 1276 (2020). There, Professor Andrew Crespo “uses the phrase ‘probable cause’ to include . . . ‘reasonable suspicion.’” Id. at 1279 n.1 (citing Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 58 Minn. L. Rev. 349, 414 (1974)). describing “reasonable suspicion” as requiring more than a “mere ‘hunch’” but “considerably” less than “preponderance of the evidence.”58×58. United States v. Arvizu, 534 U.S. 266, 274 (2002) (citing United States v. Sokolow, 490 U.S. 1, 7 (1989)). But such a commonsense approach is “ever-malleable” and “put[s] very little meat on the bones of the analysis necessary for determining” suspicion “in a particular cluster of facts.”59×59. David Rudovsky & David A. Harris, Terry Stops and Frisks: The Troubling Use of Common Sense in a World of Empirical Data, 79 Ohio St. L.J. 501, 504 (2018). Largely ignoring similar criticisms, the Court has maintained that the concept of reasonable suspicion cannot be reduced to a “neat set of legal rules.”60×60. Arvizu, 534 U.S. at 274 (quoting Ornelas v. United States, 517 U.S. 690, 695–96 (1996)). In fact, the Court has insisted that “[a]rticulating precisely what ‘reasonable suspicion’ . . . mean[s] is not possible,” because it is a “commonsense, nontechnical conception[]”61×61. Ornelas, 517 U.S. at 695. that is “incapable of precise definition or quantification into percentages.”62×62. Maryland v. Pringle, 540 U.S. 366, 371 (2003) (describing the probable cause standard, which similarly rests on a totality-of-the-circumstances approach).

Against this jurisprudential backdrop, scholarly debate continues on the use of empirical evidence in reasonable suspicion determinations. On one side of the debate is an expanding cohort of scholars who recognize the utility of empirical studies in helping quantify the probabilistic determinations that underlie many reasonable suspicion — or more broadly, probable cause — questions.63×63. See, e.g., Crespo, supra note 57, at 1291, 1310 (arguing that “any evidentiary claim [based on] a single fact (or a very small set of interrelated facts) is . . . inherently quantitative,” id. at 1291 (emphasis omitted), while defending the use of a qualitative, “holistic, totality-of-the-circumstances approach . . . in cases presenting narrative mosaics,” id. at 1310); Erica Goldberg, Getting Beyond Intuition in the Probable Cause Inquiry, 17 Lewis & Clark L. Rev. 789, 794 (2013) (arguing for a numerical legal standard establishing the probability level that triggers probable cause in cases where suspicion is created through a mechanistic process); Tracey L. Meares & Bernard E. Harcourt, Foreword: Transparent Adjudication and Social Science Research in Constitutional Criminal Procedure, 90 J. Crim. L. & Criminology 733, 735 (2000) (contending that in order to “mak[e] criminal procedure decisions more transparent,” courts ought to “place[] empirical and social scientific evidence at the very heart of constitutional adjudication”). These scholars criticize the Court’s current commonsense approach for saying barely anything about “either the methodology or the substance”64×64. Crespo, supra note 57, at 1280. of reasonable suspicion determinations and for relying on “unexamined assumptions.”65×65. Rudovsky & Harris, supra note 59, at 505 (emphasis omitted). On the other side of the debate, critics of empirical analysis object that its use contradicts the context-dependent nature of the original reasonable suspicion test. Some scholars fear that the use of quantifiable information may be too rigid to address the “practical complexities of life,”66×66. Amsterdam, supra note 57, at 375. But Professor Anthony Amsterdam notes that if “some discipline is not enforced, . . . we shall have a [F]ourth [A]mendment with all of the character and consistency of a Rorschach blot.” Id. will block and replace the intuitions of judges and police officers,67×67. See Orin Kerr, Why Courts Should Not Quantify Probable Cause, in The Political Heart of Criminal Procedure 131, 132, 139 (Michael Klarman, David Skeel & Carol Steiker eds., 2012). and may potentially run counter to the “individualized” requirement of reasonable suspicion inquiries.68×68. See, e.g., Crespo, supra note 57, at 1294.

Glover presented an ideal opportunity for the Court to definitively stake out a position on whether empirical studies should play a role in reasonable suspicion determinations. The factual stipulations in Glover confined its record to a closed universe,69×69. See Glover, 140 S. Ct. at 1194 (Kagan, J., concurring). so the Court had no occasion to second-guess the lower courts’ factfinding. That record boiled down to the “single, simple fact” of the owner’s revoked license.70×70. Id.; see also id. (Sotomayor, J., dissenting) (“The stop at issue here . . . rests on just one key fact . . . .”). Moreover, Kansas could have easily produced the empirical evidence needed to answer the primary question of how likely it is that a car registered to an unlicensed owner is indeed being driven by the owner.71×71. See Andrew Manuel Crespo, The Unavoidably Empirical Fourth Amendment: A Case Study of Kansas v. Glover, 1 Cts. & Just. L.J. 217, 225, 234 (2019). Given this somewhat unique situation, some have argued that Glover presented a “straightforward, empirical question.”72×72. Id. at 218.

Yet the Court declined the chance to resolve the debate, discussing empirical evidence only in dicta.73×73. See Glover, 140 S. Ct. at 1188. This comment does not attempt to argue what approach the Court should take, especially in light of the fact that this normative question has been thoroughly discussed and debated. See, e.g., sources cited supra notes 63–68. The “heart of the opinion” consists of “just a single sentence”74×74. Orin S. Kerr, Commonsense Suspicion: Thoughts on Kansas v. Glover, Reason: Volokh Conspiracy (Apr. 7, 2020, 5:48 AM), https://reason.com/2020/04/07/commonsense-suspicion-thoughts-on-kansas-v-glover [https://perma.cc/84FC-5CXP]. — “Deputy Mehrer drew the commonsense inference that Glover was likely the driver of the vehicle” from the stipulated facts.75×75. Glover, 140 S. Ct. at 1188. The Court did not use empirical evidence to assess the reasonableness of this inference, nor did it reject its potential application. The Court did make a passing reference to empirical studies but only to rebut the argument that Glover’s revoked license made it unreasonable to infer that he was driving.76×76. See id. (“Glover’s revoked license does not render Deputy Mehrer’s inference unreasonable either. Empirical studies demonstrate what common experience readily reveals: Drivers with revoked licenses frequently continue to drive . . . .”). Glover’s core holding rested solely on “common sense.”77×77. See id. Naturally, without taking a clearly discernible position on the use of empirical evidence, the Court missed the opportunity to clarify whether (or when) the use of empirical evidence would run afoul of the Fourth Amendment’s requirement that a traffic stop must be individualized. See Glover, 140 S. Ct. at 1195 (Sotomayor, J., dissenting). The Court briefly remarked that Deputy Mehrer had individualized suspicion, because “a particular citizen was engaged in a particular crime.” Glover, 140 S. Ct. at 1190 n.1 (majority opinion).

Even if the Court’s mention of empirical studies in dicta constituted an endorsement of their use, Glover created confusion as to when lower courts should rely on empirical evidence. In Glover, nowhere in the trial record — a mere stipulation of facts — did the prosecution introduce the key empirical studies the Court cited in dicta.78×78. See Brief of Professor Andrew Manuel Crespo as Amicus Curiae in Support of Affirmance at 20, Glover, 140 S. Ct. 1183 (2020) (No. 18-556). In the past, the Court has suggested that it is incumbent upon the officer to “point to specific and articulable facts” to justify the initial search79×79. Terry v. Ohio, 392 U.S. 1, 21 (1968). — a burden that should be met during the trial stage. For example, in United States v. Brignoni-Ponce,80×80. 422 U.S. 873 (1975). the Court explicitly declined the government’s post-trial request to consider the location of the Terry stop in addition to the defendants’ Mexican ancestry, the latter being the only factor the government relied on at trial to justify the reasonableness of the stop.81×81. See id. at 885–87, 886 n.11. The Court deemed the government’s new proffered evidence an “after-the-fact justification” when “[a]t trial the officers gave no reason for the stop except the apparent Mexican ancestry of the car’s occupants.”82×82. Id. at 886 n.11.

Although the prosecution in both Brignoni-Ponce and Glover introduced new evidence on appeal, the Glover Court neither distinguished the former83×83. The Court did cite Brignoni-Ponce but not for the issue of belated evidence. See Glover, 140 S. Ct. at 1190 (citing Brignoni-Ponce, 422 U.S. at 882). nor explicitly rejected the after-the-fact empirical studies that Kansas presented.84×84. The Court touched upon post hoc justifications briefly, explaining that “it is the information possessed by the officer at the time of the stop, not any information offered by the individual after the fact, that can negate the inference” of reasonable suspicion. Id. at 1191 n.2. Lower courts may interpret Glover as authorizing them to consider empirical studies proffered after trials.85×85. One way to reconcile Brignoni-Ponce and Glover is to emphasize that the Glover Court did not ultimately rely on the empirical studies to conclude that Deputy Mehrer had reasonable suspicion. But that reasoning fails to (1) explain why the Court considered the empirical studies relevant enough to discuss or (2) provide lower courts with more clarity about when, if ever, they should consider empirical data. Suppose, for example, that an officer stops a defendant upon the defendant’s unprovoked flight. Could the officer retroactively justify the stop with empirical evidence showing that the alleged flight happened to occur in a high-crime area, a fact to which the officer did not testify at trial?86×86. This hypothetical case resembles Illinois v. Wardlow, 528 U.S. 119 (2000), in which the Court held that Wardlow’s presence “in an area of heavy narcotics trafficking” and “his unprovoked flight” provided reasonable suspicion for a Terry stop. Id. at 124–25. But, unlike Deputy Mehrer in Glover, the officer who stopped Wardlow “testified that he . . . had gone to the area because it was ‘one of the areas in the 11th District that’s [sic] high narcotics traffic.’” People v. Wardlow, 678 N.E.2d 65, 66 (Ill. App. Ct. 1997), rev’d, 701 N.E.2d 484 (Ill. 1998), aff’d, 528 U.S. 119. The Court’s reluctance to address the tension between its precedent and Glover has made the answer to this question unclear.

Lastly, the Court did not clarify whether empirical evidence could independently support a finding of reasonable suspicion in the absence of a commonsense basis for the stop. Granted, the Court cited empirical studies to support the proposition that drivers with revoked licenses routinely continue to drive.87×87. See Glover, 140 S. Ct. at 1188. But, in claiming that these studies “demonstrate what common experience readily reveals,”88×88. Id. the Court gave empirical evidence a purely supporting role. Even more confusingly, the Court’s use of empirical studies was secondary. That is, the Court cited the studies to refute the claim that Glover’s revoked license rendered Deputy Mehrer’s inference unreasonable rather than to answer the primary question of how likely it is that the driver of a vehicle registered to an unlicensed owner is in fact the owner.89×89. See id. Consider the following extreme hypothetical situation. Suppose that, on average, one hundred people share a single car in a community and that each of them uses it for roughly the same amount of time. Then, even if the owner of the car who has a revoked license continued to drive, the probability of the driver being the owner is still only one percent.  Without also knowing, among other things, how often people share cars in a community, the fact that drivers with revoked licenses continue to drive does not cut to the core of the issue. That said, it is not necessary to know the answers to either of the secondary questions in Glover. Technology such as “mobile data terminals” has made it much easier to record “automotive information” to analyze directly the primary question at issue, that is, how likely it is that the driver of a car registered to an unlicensed owner is indeed the owner. Crespo, supra note 71, at 235. Therefore, even if empirical studies are appropriate to consider as a general matter, it is uncertain whether lower courts could use empirical analysis to quantify the probable basis for reasonable suspicion independently of a commonsense justification for the stop.

Glover’s noncommittal stance on empirical evidence is especially problematic because the Court failed to explain what made the inference at issue “common sense.” Rather, the Court simply asserted that it was so.90×90. See Glover, 140 S. Ct. at 1196 (Sotomayor, J., dissenting); see also Kerr, supra note 74. The Court could have taken a qualitative approach by focusing, for example, on community experience. With such an approach, the Court could have argued, rather than merely asserted, that it is either common or uncommon for somebody other than the registered owner to drive that vehicle.91×91. Compare Glover, 140 S. Ct. at 1189 (“The inference that the driver of a car is its registered owner . . . is a reasonable inference made by ordinary people on a daily basis.”), and Kerr, supra note 74 (“For the most part, people drive the cars that they own.”), with State v. Glover, 400 P.3d 182, 185 (Kan. Ct. App. 2017) (quoting the trial court judge, who stated, “I think that’s true for a lot of families that if there are multiple family members and multiple vehicles, that somebody other than the registered owner often is driving that vehicle”). Alternatively, the Court could have substantiated common sense or the lack thereof with support based on law enforcement training or experience.92×92. See Glover, 140 S. Ct. at 1195–96 (Sotomayor, J., dissenting). Instead, the Court relegated to a supportive, secondary role,93×93. As discussed earlier, the Court relegated empirical evidence to a supporting role. ignored,94×94. The Court did not address at all the issue of community experience. or rejected95×95. See Glover, 140 S. Ct. at 1189 (“Nothing in our Fourth Amendment precedent supports the notion that . . . an officer can draw inferences based on knowledge gained only through law enforcement training and experience.”). additional rationales that might have provided content to the concept of common sense.

The narrow scope of Glover’s holding only aggravates the problem of relying exclusively on common sense. If the Court sought to avoid issuing broad bright-line rules and to give lower courts leeway to decide individual cases based on different facts, it would be all the more important to provide interpretive metrics so that lower courts can weigh those facts in a consistent and non-arbitrary manner.96×96. Cf., e.g., Tara Leigh Grove, The Structural Case for Vertical Maximalism, 95 Cornell L. Rev. 1, 31 (2009) (advocating for broad rulemaking but noting that “a rigid focus on rules would obscure the central point,” which “is not whether the Court has chosen a rule or a standard but rather the extent to which its precedent serves to bind — and thus to guide — the lower courts”); id. at 29–30 (arguing that even if narrow rules bind some cases, they do not provide adequate guidance to lower courts). Bald assertions that a particular inference is common sense leave courts with little objective guidance on how to decide cases with slightly different factual permutations. For instance, should lower courts reach a different conclusion if the defendant instead had a suspended license? On the one hand, license suspensions differ from license revocations as a matter of Kansas law.97×97. As Justice Kagan noted, Kansas “suspends licenses for matters having nothing to do with road safety.” Glover, 140 S. Ct. at 1192 (Kagan, J., concurring). For that reason, the majority’s claim that drivers with revoked licenses “have already demonstrated a disregard for the law,” id. at 1188–89 (majority opinion), does not apply with equal force to drivers with suspended licenses. On the other hand, Glover noted that any inference drawn from Kansas state law was not necessary to its ultimate holding.98×98. Only Justice Ginsburg joined Justice Kagan’s concurrence, and the majority also emphasized that Kansas law merely reinforced the reasonableness of Deputy Mehrer’s inference and that common sense alone sufficed to justify the inference. See id. at 1188. Without the need to resort to Kansas law, the Court’s rationale under revoked-license cases seems applicable to suspended-license cases. But, because the Court’s narrow opinion decided only the revoked-license question, lower courts will need to use other tools such as empirical evidence to decide the suspended-license question. If Glover had either approved the use of empirical studies or rejected them in favor of an alternative method for establishing that a reasonable suspicion determination is common sense, the opinion would have provided more guidance than it does.

The recent national spotlight on police brutality and structural racism within the criminal justice system has made the need for such guidance all the more apparent. The commonsense approach to evaluating police action grants officers broad interpretive leeway that facilitates the very police abuses racial justice activists decry.99×99. See, e.g., L. Song Richardson, Arrest Efficiency and the Fourth Amendment, 95 Minn. L. Rev. 2035, 2059–73 (2011) (explaining how intuition and common sense in law enforcement can serve to exacerbate racial bias). More problematically, many have criticized “common sense” itself as reflecting racial bias and unconscious prejudice, insofar as the line between intuitive judgments and racial profiling is thin at best and ephemeral at worst.100×100. See, e.g., Floyd v. City of New York, 959 F. Supp. 2d 540, 580–81 (S.D.N.Y. 2013) (“[P]sychological research has shown that unconscious racial bias continues to play an objectively measurable role in many people’s decision processes.” Id. at 580.); Tracey Maclin, The Decline of the Right of Locomotion: The Fourth Amendment on the Streets, 75 Cornell L. Rev. 1258, 1320–26 (1990) (“The fundamental point is that police officials should not be free to effect seizures based upon factors allegedly possessed by those engaged in criminal conduct, but also shared by a significant percentage of innocent persons particularly when those factors concern characteristics like race and age.” Id. at 1324.). See generally Devon W. Carbado, From Stop and Frisk to Shoot and Kill: Terry v. Ohio’s Pathway to Police Violence, 64 UCLA L. Rev. 1508 (2017) (arguing that stop and frisk enables police violence against Black people). The growing prevalence of automated license-plate readers that allow the police to conduct license searches on a mass scale101×101. See Ferguson, supra note 2, at 357 & n.154; see also Josh Kaplan, License Plate Readers Are Creeping into Neighborhoods Across the Country, Slate (July 10, 2019, 7:30 AM), https://slate.com/technology/2019/07/automatic-license-plate-readers-hoa-police-openalpr.html [https://perma.cc/5NQD-YSX6]. exemplifies the broad trend of surveillance policing that perpetuates inequality under the law.102×102. Cf. Utah v. Strieff, 136 S. Ct. 2056, 2069–71 (2016) (Sotomayor, J., dissenting) (“This Court has given officers an array of instruments to probe and examine you. When we condone officers’ use of these devices without adequate cause, we give them reason to target pedestrians in an arbitrary manner. We also risk treating members of our communities as second-class citizens.” Id. at 2069.). With so much at stake, the Court could have said more.