In the United States Court of Appeals For the Seventh Circuit ____________________ No. 13-2995 UNITED STATES OF AMERICA, Plaintiff-Appellee, v. LARRY BENTLEY, Defendant-Appellant. ____________________ Appeal from the United States District Court for the Central District of Illinois. No. 10-10108-001 — James E. Shadid, Chief Judge. ____________________ ARGUED DECEMBER 11, 2014 — DECIDED JULY 28, 2015 ____________________ Before WOOD, Chief Judge, and FLAUM and MANION, Circuit Judges. WOOD, Chief Judge. A great many police departments rely on trained dogs to detect hidden drugs (or other substances, including explosives, blood, and human remains). Nagging questions remain, however, about the accuracy of the dog’s performance, especially when a dog’s alert provides the sole basis for a finding of probable cause to search or arrest someone. 2 No. 13-2995 In Larry Bentley’s case, a police officer initiated a traffic stop after observing Bentley’s vehicle cross into another lane on an Illinois highway without signaling. After stopping Bentley, the officer decided to call for a drug-detection dog named Lex. Once on the scene, Lex alerted, and the officers found close to 15 kilograms of cocaine in the vehicle. But what if Lex alerts every time he is called upon? The fact that drugs are (or are not) found would have nothing to do with his behavior. That, in essence, is what Bentley is arguing here. The evidence Bentley was able to gather suggests that Lex is lucky the Canine Training Institute doesn’t calculate class rank. If it did, Lex would have been at the bottom of his class. Nevertheless, in light of the Supreme Court’s decision in Florida v. Harris, 133 S. Ct. 1050 (2013), which addressed the use of drug-detection dogs, we conclude that the district judge did not err when he decided that Lex’s alert, along with the other evidence relating to the stop, was sufficient to support probable cause. Bentley’s other two challenges based on the traffic stop and his alleged lack of knowledge of the cocaine in the vehicle also fail. We thus affirm his conviction. I On October 14, 2010, Aaron Veerman, an officer with the Bloomington (Illinois) Police Department, ran a license check on a Chrysler Pacifica he observed at a Circle K convenience store. Veerman learned that the car was registered to Tonya Smith of Kankakee, Illinois, but that Smith’s driver’s license had expired 18 years earlier. Officer Veerman followed the car onto I-55 and stayed with it until the driver committed a lane violation. At that point, he signaled for the car to pull over. When it did, he learned that Tonya Smith was not driv- No. 13-2995 3 ing the car; Larry Bentley was. Veerman explained to Bentley that he had seen the car commit lane violations and wanted to make sure he was all right. Bentley first informed Veerman that he was driving from Chicago to his home in St. Louis and that Smith was his girlfriend. He then fumbled, said that Smith lived with him in St. Louis, and then changed his story again and said that she actually lived in Kankakee, but often stayed with him in Missouri. Bentley handed over a valid driver’s license and some documents (including proof of insurance) from the glove compartment. Another officer, Nikolai Jones, arrived during this exchange and with a flashlight observed a spare tire in the back seat of the car. The two officers spoke and then radioed for a drugdetection dog. Officer Justin Shively responded and brought Lex, a trained drug dog, to the scene. Bentley agreed to get out of the car and to allow the officers to search him while Lex was sniffing. The officers found $1,699 in cash in Bentley’s pockets—far more than the “couple hundred” to which Bentley had admitted. The officers also found a cell phone, nine money orders for a total of $5,600, and a wallet that contained three more money orders adding up to $900. Meanwhile, Lex alerted to drugs in the car. Sure enough, officers found nearly 15 kilograms of cocaine in a trap compartment. Bentley was later charged by a federal grand jury with possession with intent to distribute more than five kilograms of cocaine. 21 U.S.C. § 841(a)(1). After an evidentiary hearing, the district court denied Bentley’s motion to suppress. It found that Officer Veerman had both probable cause and reasonable suspicion to stop Bentley, because even though the officer did not see who the driver was, the officer knew 4 No. 13-2995 that the owner of the vehicle did not have a valid license and the officer observed a lane violation. The court also found that Lex was reliable enough as a drug-detection dog to establish probable cause. At the conclusion of the trial, the jury convicted Bentley. The district court denied Bentley’s Rule 29 motion for a judgment of acquittal and sentenced him to 240 months, followed by 10 years of supervised release. This timely appeal followed. II Bentley offers three challenges to his conviction. He begins with the contention that the officer did not have reasonable suspicion to make the initial stop. Even if the stop survives scrutiny, he argues next that Lex’s alert was not sufficiently reliable to support probable cause. Finally, he maintains that even if the stop and search were valid, the government failed to present sufficient evidence to establish his guilt beyond a reasonable doubt. We address his challenges in that order. A When a criminal defendant appeals the district court’s denial of a motion to suppress, we review both legal conclusions and mixed questions of law and fact de novo. United States v. Henderson, 748 F.3d 788, 790 (7th Cir. 2014). Vehicle stops are analyzed using the Fourth Amendment’s reasonableness standard. Whren v. United States, 517 U.S. 806, 810 (1996). We have recognized that it is a mistake to treat “all traffic stops identically.” United States v. Childs, 277 F.3d 947, 952 (7th Cir. 2002) (en banc). Instead, we distinguish between stops based on reasonable suspicion and those based on probable cause. The latter are not subject to the scope and No. 13-2995 5 duration restrictions of Terry v. Ohio, 392 U.S. 1 (1960). Childs, 277 F.3d at 952–53. While Bentley argues that the initial traffic stop was not justified by either probable cause or even a reasonable suspicion, he does not challenge the duration of the stop. As a result, we need consider only whether, under the totality of the circumstances, an officer would have had reasonable suspicion that Bentley had committed a lane violation. Officer Veerman pulled Bentley over for violating 625 ILCS 5/11-709(a). That provision reads, “Whenever any roadway has been divided into 2 or more clearly marked lanes for traffic … (a) A vehicle shall be driven as nearly as practicable entirely within a single lane and shall not be moved from such lane until the driver has first ascertained that such movement can be made with safety.” To establish probable cause for a violation of Section 11-709(a) (a petty offense under Illinois law, see 625 ILCS 5/16-104), “the officer must point to facts which support a reasonable belief that the defendant has deviated from his established lane of travel and that it was ‘practicable’ for him to have remained constant in his proper lane.” People v. Hackett, 971 N.E.2d 1058, 1066 (Ill. 2012). Veerman testified that Bentley’s car committed two lane violations, once on the curved on-ramp and later when he crossed the line dividing a new lane caused by another on-ramp. According to Bentley, both occurred on a curve in the road during or immediately after he merged onto the interstate highway. The video the government offered at the suppression hearing supported Officer Veerman’s testimony. It is hard to see from the video when Bentley’s car supposedly crossed the line the first time, because there is a car between Officer 6 No. 13-2995 Veerman’s squad car and Bentley’s. The video leaves no doubt, however, about the second infraction: it unmistakably captures Bentley’s car crossing the fog line. That evidence, coupled with Veerman’s testimony, is enough to support probable cause for the traffic stop. See, e.g., United States v. Hernandez-Rivas, 513 F.3d 753, 759 (7th Cir. 2008) (“We have held improper lane usage is a legitimate reason for an investigatory stop.”). In the video, one can hear Officer Veerman say to Bentley, “The reason I’m stopping you is you drifted over into the side of the road a couple of times. You passed over the white line here. I’m just making sure you’re all right.” Bentley insists that the video is not so clear and thus does not reveal any violation of the traffic laws. He maintains that the wheels of his car only briefly touched the fog line, maybe crossing it momentarily, while he was driving in the dark and merging onto an interstate highway with faster vehicles passing him. But, especially for probable cause determinations, there is no rule excusing momentary slips. Hackett, 971 N.E.2d at 1065–66; see also People v. Geier, 944 N.E.2d 793, 799 (Ill. App. Ct. 2011). Bentley is right that a car was starting to pass his, but the other car was not moving in such a way that would suggest Bentley was driving in adverse conditions and was unable safely to stay in his lane. Alternatively, Bentley demands that this court interpret the term “practicable” in the statute as ambiguous and resolve the ambiguity in his favor. See City of Chi. v. Morales, 527 U.S. 41, 56 (1999). The problem with this argument is that there were no extenuating circumstances that made it difficult for Bentley to stay in his lane. His case is thus not like others in which courts have found extenuating circum- No. 13-2995 7 stances. See United States v. Freeman, 209 F.3d 464, 466 (6th Cir. 2000) (no probable cause to stop a large motor home crossing into an emergency lane once); United States v. Gregory, 79 F.3d 973, 978 (10th Cir. 1996) (no reasonable suspicion when defendant’s car crossed a fog line on a winding road in windy conditions). In Bentley’s case, the lanes were clearly marked on a well-lit highway. The weather was good with high visibility and little wind. And there was no dispute at the suppression hearing about the quality of the road or the behavior of other drivers. Finally, Bentley points out that Officer Veerman testified that an owner’s suspended license was insufficient for a traffic stop and that he would not have pulled Bentley over but for the lane violation. Officer Veerman was mistaken. He could have stopped Bentley’s vehicle on the ground of the owner’s suspended license alone. See United States v. Jerez, 108 F.3d 684, 693 (7th Cir. 1997). Because the test is an objective one, we ignore Veerman’s mistake. A reasonable officer would have had sufficient facts to justify a reasonable suspicion that the driver of the Chrysler Pacifica was violating Illinois’s traffic law requiring a valid license. 625 ILCS 5/6101(a). Neither Rodriguez v. United States, 135 S. Ct. 1609 (2015), nor Heien v. North Carolina, 135 S. Ct. 530 (2014), to which Bentley referred as supplemental authority, requires a different result. Rodriguez dealt with stops of extended duration, but Bentley waived any challenge to the length of his stop, while Heien held that reasonable suspicion for a traffic stop may rest on a reasonable mistake of law—a holding that if anything undermines Bentley’s position. B We now turn to Bentley’s challenge to the use of the 8 No. 13-2995 drug-detection dog. An alert from an adequately trained and reliable dog is sufficient to give rise to a finding of probable cause. United States v. Washburn, 383 F.3d 638, 643 (7th Cir. 2004). The government conceded that the police lacked probable cause to search Bentley’s vehicle if we disregard Lex’s alert. In so doing, it may have acted too hastily: before Lex alerted, the police already knew that Bentley could not keep his story straight, that the spare tire was in an odd place, and that the search of Bentley’s person (done with his permission) had turned up far more money than Bentley had admitted to having. If Bentley can show that Lex was not adequately trained and reliable, this would weaken the case for probable cause, but we nonetheless would need to consider the totality of the circumstances before finding that the search of the car was unconstitutional. In pressing his challenge to the dog’s alert, Bentley makes two principal points. First, he contends that Lex’s past performance in the field suggests he is particularly prone to false positives (i.e., signaling to his handler that there are drugs in a vehicle when there are not). He has a point. Lex alerts 93% of the time he is called to do an openair sniff of a vehicle, and Lex’s overall accuracy rate in the field (i.e., the number of times he alerts and his human handler finds drugs) is not much better than a coin flip (59.5%). The Supreme Court, however, recently rejected a proposed rule that would have treated the dog’s field record as a “gold standard.” To the contrary, it said, the record is of “relatively limited import.” Florida v. Harris, 133 S. Ct. 1050, 1056 (2013); see also United States v. Funds in Amount of $100,120.00, 730 F.3d 711, 724 (7th Cir. 2013) (recognizing that Harris changes the district judge’s analysis). Instead, “evidence of a dog’s satisfactory performance in a certification or training pro- No. 13-2995 9 gram can itself provide sufficient reason to trust his alert.” Harris, 133 S. Ct. at 1057. In order to assess whether the police adequately trained their dog, the Harris Court instructed trial judges to hold a probable-cause hearing: If the State has produced proof from controlled settings that a dog performs reliably in detecting drugs, and the defendant has not contested that showing, then the court should find probable cause. If, in contrast, the defendant has challenged the State’s case (by disputing the reliability of the dog overall or of a particular alert), then the court should weigh the competing evidence. Id. at 1058. The Court did not, however, suggest what weight courts should give to different types of evidence, nor did it offer any tie-breakers for district courts to use. The district judge dutifully followed the Harris Court’s instructions: he let the government submit evidence about Lex’s training. That evidence included the dog’s success rates in controlled settings as well as testimony from the dog’s handler and the training institute’s founder. The judge also allowed Bentley to challenge those findings, to crossexamine the handler and the Canine Training Institute’s (CTI) founder, and to put on his own expert witness. The judge then weighed all the evidence, decided to credit the government’s experts over Bentley’s, and decided that Lex’s alert was reliable enough to support probable cause. Our review of a district court’s choice between one version of the evidence and another is typically very deferential (even if experts are involved), and we are given no reason to deviate from that approach here. 10 No. 13-2995 We acknowledge that Bentley put on a good case. He presented a fair amount of evidence that Lex was at the back of the pack. The head of CTI, the company that trained Lex, was embarrassed by Lex’s 93% alert rate in the field: “Well, I don’t like to see that he indicated at that high of a rate.” He went on to testify (consistently with the government’s theory) that the dog’s rate is so high because there is embedded bias in his use: Lex is called only when the police already suspect that drugs may be present. He added, “I understand that the way that they are actually deploying the dogs in Bloomington, not to do general interdiction but specifically when there is suspicion that made me feel more comfortable.” Bentley also brought out that long after the 2010 traffic stop, Lex was removed from the field for two weeks in April 2012 after he failed two simulated vehicle searches. Bentley rightly points out that Lex is smart. Shively testified that he rewards Lex every time the dog alerts in the field. Presumably the dog knows he will get a “giftee” (a rubber hose stuffed with a sock) every time he alerts. If Lex is motivated by the reward (behavior one would expect from any dog), he should alert every time. This giftee policy seems like a terrible way to promote accurate detection on the part of a service animal, lending credence to Bentley’s argument that Lex’s alert is more of a pretext for a search than an objective basis for probable cause. Even if we were to ignore Harris and focus on Lex’s 59.5% field-accuracy rate, though, that rate is good enough to support a finding of his reliability and thus to allow his alert to constitute a significant piece of evidence supporting the ultimate conclusion of probable cause. In the past, we have concluded that a 62% success rate in the field is enough to prevail on a preponderance of evidence, and we have No. 13-2995 11 gone on to note that “’probable cause’ is something less than a preponderance.” United States v. Limares, 269 F.3d 794, 798 (7th Cir. 2001) (citation omitted). Other circuits have accepted field detection rates less than Lex’s 59.5%. See, e.g., United States v. Holleman, 743 F.3d 1152, 1157 (8th Cir. 2014) (57%); United States v. Green, 740 F.3d 275, 283 (4th Cir. 2014) (43%). This should not become a race to the bottom, however. We hope and trust that the criminal justice establishment will work to improve the quality of training and the reliability of the animals they use, and we caution that a failure to do so can lead to suppression of evidence. We will look at all the circumstances in each case, as we must. Bentley’s second argument that Lex is an unreliable source of probable cause hinges on the allegedly poor quality of the school that trained him and his handler. This argument cannot get off the ground. Bentley concedes that there are no national standards by which we can judge the training Lex received at CTI. Moreover, there is evidence in the record that CTI modeled its certification standards after the leading national associations in the field. The district judge did not err when he found Lex to be reliable for purposes of contributing to a probable cause determination based on his training records, his 59.5% field rate, and CTI’s curriculum. Lex’s mixed record is a matter of concern, but under Harris’s totality-of-the-circumstances test, we have no reason to override the district court’s determination. C Finally, we turn to Bentley’s “Hail Mary” argument: that the district court erred by denying his motion for a judgment 12 No. 13-2995 of acquittal under Federal Rule of Criminal Procedure 29. Our review is de novo, and we take the evidence in the light most favorable to the government. United States v. White, 698 F.3d 1005, 1013 (7th Cir. 2012). We will affirm Bentley’s conviction if any rational trier of fact could have found him guilty. United States v. Blitch, 773 F.3d 837, 846 (7th Cir. 2014). The government had to prove beyond a reasonable doubt that Bentley “knowingly or intentionally possessed a controlled substance with the intent to distribute it, while knowing that it was a controlled substance.” United States v. Carraway, 612 F.3d 642, 645 (7th Cir. 2010) (citation omitted); see 21 U.S.C. § 841(a)(1). Bentley contends that the government failed to prove that he knew there were drugs in the car. Both Bentley and the government insist that we have never had the occasion to hold that it is permissible to infer knowledge that drugs are concealed in a vehicle solely from proof that there are secret compartments and the driver is alone in the car. We could not find a case to contradict them. Most circuits take the position that more is needed. United States v. Tran, 519 F.3d 98, 105 (2d Cir. 2008) (“We agree with the other courts to have addressed the issue that a defendant’s sole occupancy of a vehicle cannot alone suffice to prove knowledge of contraband found hidden in the vehicle; corroborating evidence … is necessary to prove this element.”); United States v. Stanley, 24 F.3d 1314, 1320–21 (11th Cir. 1994); United States v. Anchondo-Sandoval, 910 F.2d 1234, 1236 (5th Cir. 1990). We see no reason to depart from those holdings. Requiring an additional showing of some evidence tending to show knowledge appropriately prevents a finding of guilt in those cases where a third party “conceal[s] the controlled sub- No. 13-2995 13 stances in the vehicle with the intent to use the unwitting defendant as the carrier in a smuggling enterprise.” United States v. Martinez-Lugo, 411 F.3d 597, 599 (5th Cir. 2005) (quotation marks and citation omitted). A rule that allows a factfinder to infer that a driver knows about a hidden compartment in a vehicle simply because he is driving it goes too far. But this does not help Bentley, because the government did not try to rest its case on that one fact. Unlike the unwitting smuggler contemplated by our sister circuits, there was substantial evidence that gave rise to a reasonable inference that Bentley knowingly possessed the cocaine hidden in the car. First, physical evidence suggested that Bentley had been driving the Chrysler Pacifica for some time. While Bentley was not the owner of the car (his girlfriend Tonya Smith was), Tonya’s license had expired nearly two decades earlier. Moreover, his documents were commingled with hers in the glove compartment, including some dated 21 months before the stop. The police also found a key for a car Bentley had rented a few weeks earlier, and they saw papers connected to his employment. They thought it was peculiar that a spare tire was placed in the backseat of the Pacifica, rather than in the normal place. The fact that the spare tire was not properly stowed suggested that the compartment may have been used for other purposes and that Bentley was aware of this. Even more damning was the evidence that the police collected during the search—evidence that could not be used to support the search, but that was highly pertinent to guilt. The sheer amount of drugs in the car (with a street value of $2.4 million according to the government), the presence of six cell phones, and thousands of dollars in cash or money 14 No. 13-2995 orders all supported an inference of knowledge. See, e.g., United States v. De Jesús-Viera, 655 F.3d 52, 60 (1st Cir. 2011) (“The jury could have drawn the inference that De JesúsViera knew that the over $1.45 million worth of drugs were in his car based on its common sense.”). The government provided testimony to the jury that drug dealers carry multiple phones, particularly prepaid phones that cannot be traced (as four of Bentley’s phones were). See United States v. Hernandez-Mondaza, 600 F.3d 971, 977 (8th Cir. 2010) (concluding that the jury could infer knowledge where there was evidence of four phones in the vehicle and testimony of drug traffickers carrying multiple phones to avoid detection). A large amount of cash also provides circumstantial evidence of a person’s involvement in drug trafficking. See, e.g., United States v. Harris, 585 F.3d 394, 400 (7th Cir. 2009) ($8,900 in cash found in driver’s pockets); United States v. Gamez-Gonzalez, 319 F.3d 695, 698 (5th Cir. 2003) ($1,600 in cash along with testimony about defendant’s nervousness supports knowledge of hidden drugs). Finally, Officer Veerman noted Bentley’s odd behavior during the stop. Bentley could not keep his story straight: he contradicted himself about where his girlfriend, the owner of the car, lived, and he said he was carrying a couple hundred dollars even though he was carrying eight times that amount. Also (although this may have been a function of a general dislike of being stopped), Bentley’s hand trembled when he handed over his license to Officer Veerman. This evidence, taken as a whole, was enough to permit the jury to reject Bentley’s protestation that he was unwittingly driving a car with drugs stashed in a secret compart- No. 13-2995 15 ment. The district court was thus correct to deny Bentley’s Rule 29 motion. III Officer Veerman had reasonable suspicion to stop Bentley’s car. Lex’s alert, though more equivocal than we might prefer, taken together with the other evidence before the officers was enough to support the finding of probable cause for the officers’ search of the vehicle. And the court properly rejected Bentley’s challenge to the sufficiency of the evidence to support his conviction. The district court’s judgment is AFFIRMED.