United States v. Lamar Johnson

FOR PUBLICATION UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT UNITED STATES OF AMERICA, No. 17-10252 Plaintiff-Appellee, D.C. No. v. 3:16-cr-00251-WHA-1 LAMAR JOHNSON, Defendant-Appellant. OPINION Appeal from the United States District Court for the Northern District of California William Alsup, District Judge, Presiding Argued and Submitted September 11, 2018 San Francisco, California Filed January 9, 2019 Before: J. Clifford Wallace, Johnnie B. Rawlinson, and Paul J. Watford, Circuit Judges. Opinion by Judge Wallace; Concurrence by Judge Watford 2 UNITED STATES V. JOHNSON SUMMARY * Criminal Law The panel affirmed a conviction and sentence for multiple crimes in a case in which the district court denied the defendant’s motion to suppress evidence recovered from the warrantless searches of his person and car and the warrant search of his house. The panel held that the search of the defendant’s person was constitutional. The panel addressed whether two wellestablished principles—(1) that a search incident to a lawful arrest does not necessarily need to follow the arrest to comport with the Fourth Amendment and (2) that an officer’s subjective reasons for making the arrest need not be the criminal offense as to which the known facts provide probable cause—may coincide without violating the Fourth Amendment. The panel explained that the justifications for the search incident to lawful arrest exception do not lose any of their force in the context of a search performed by an officer who has probable cause to arrest and shortly thereafter does arrest; and that so long as the search was incident to and preceding a lawful arrest—which is to say that probable cause to arrest existed and the search and arrest are roughly contemporaneous—the arresting officer’s subjective crime of arrest need not have been the crime for which probable cause existed. The panel held that Knowles v. Iowa, 525 U.S. 113 (1998), does not prevent a search incident to a lawful arrest from occurring before the arrest * This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. UNITED STATES V. JOHNSON 3 itself, even if the crime of arrest is different from the crime for which probable cause existed. As for the defendant’s argument that this standard invites pretextual and discriminatory searches, the panel did not think that this case is materially different from cases where the search precedes the arrest and the arresting officer’s subjective crime of arrest is the same as the crime for which probable cause existed. The panel rejected the defendant’s argument that, even applying this standard, the search of his person was unconstitutional because the officer did not have probable cause to arrest him. The panel wrote that the smell of fresh and burnt marijuana in the defendant’s car, along with plastic baggies in the glove compartment, and the defendant’s unusual search of the glove compartment, indicated a fair probability that the defendant had committed, or was about to commit, the offense of marijuana transportation. The panel held that the search of the defendant’s vehicle was justified under the automobile exception to the warrant requirement because when the officer approached the car, he immediately smelled a combination of burnt and fresh marijuana. The panel rejected the defendant’s challenges to the validity of the magistrate’s warrant to search his house. The panel held that an officer’s affidavit provided the substantial basis for the magistrate to determine that probable cause existed, that the confidential informant’s reliability is largely beside the point, and that any omission from the affidavit was immaterial. The panel rejected the defendant’s argument that the district court erred in increasing his offense level under U.S.S.G. § 3B1.5 for using body armor during the commission of the offense because “use” does not mean 4 UNITED STATES V. JOHNSON simply wearing body armor. The panel wrote that there is no reasonable way to construe the guidelines commentary that would exclude wearing body armor from the definition of “use.” The panel concluded that the district court did not abuse its discretion by determining that the enhancement should apply to the defendant. Concurring, Judge Watford joined the court’s opinion because it faithfully applies the rule adopted in United States v. Smith, 389 F.3d 944 (9th Cir. 2004) (per curiam), which held that a warrantless search that precedes an arrest may nonetheless fall within the search-incident-to-arrest exception if “the search is conducted roughly contemporaneously with the arrest” and probable cause to arrest existed at the time of the search. Observing that many courts have adopted the rule, but some have rejected it in favor of a more circumscribed approach, Judge Watford wrote that Smith falls on the wrong side of this divide and should be overruled. COUNSEL Robin Packel (argued), Research and Writing Attorney; Gabriela Bischof, Assistant Federal Public Defender; Steven G. Kalar, Federal Public Defender; Office of the Federal Public Offender, San Francisco, California; for DefendantAppellant. Philip Kopczynski (argued), Assistant United States Attorney; J. Douglas Wilson, Chief, Appellate Division; Alex G. Tse, Acting United States Attorney; United States Attorney’s Office, San Francisco, California; for PlaintiffAppellee. UNITED STATES V. JOHNSON 5 OPINION WALLACE, Circuit Judge: Lamar Johnson was stopped while driving and subjected to a warrantless search of his person and car. One year later, police conducted a warrant search of his home. The fruits of these searches led to Johnson’s indictment and conviction for multiple crimes. Johnson appeals, challenging his conviction and sentence. We have jurisdiction under 28 U.S.C. § 1291 and 18 U.S.C. § 3742 , and we affirm. I. On August 7, 2015, Lamar Johnson was stopped while driving by Sergeant Clint Simmont of the East Palo Alto Police Department. As Simmont spoke with Johnson, he smelled a combination of burnt and fresh marijuana, which he recognized through his work patrolling East Palo Alto and on the San Mateo County Narcotics Task Force. Simmont asked Johnson for his registration and proof of insurance, to which Johnson responded that he was borrowing the car and did not have registration or insurance information. Simmont asked if Johnson was sure, and Johnson opened the glove box as if to check. Simmont observed empty plastic bags and pill bottles in the glove box and noticed that Johnson “moved his hand around on the few items that were in there, but he didn’t actually manipulate any items.” This manner was “inconsistent with the way someone would genuinely search for paperwork.” Simmont then learned from a police dispatch agent that Johnson had been arrested for parole violations, which indicated to Simmont that Johnson had been convicted of a felony. Simmont asked Johnson to step out of the vehicle and searched his person. Simmont discovered that Johnson was 6 UNITED STATES V. JOHNSON wearing a bulletproof vest and arrested him for being a felon in possession of body armor. After backup police units arrived, Simmont and the other officers searched Johnson’s car and discovered a loaded handgun, a pill bottle containing acetaminophen/hydrocodone pills, plastic bags, scales, and concentrated cannabis. Johnson was transported to a police station, where a second search of his person revealed additional controlled substances. The following year, a separate investigation in San Mateo County linked Johnson to controlled substance distribution. On March 16, 2016, a judge in San Mateo Superior Court issued a warrant to search Johnson, a vehicle allegedly belonging to him, and a residence in east Palo Alto allegedly belonging to him. Detective Christopher Sample subscribed and swore to an affidavit in support of the warrant. According to his affidavit, Sample met with a confidential informant (CI) who purportedly could call a man named “Lamar” at a specific phone number and arrange a sale of cocaine base. The CI called the number and a male voice answered the phone and gave a location to meet. Police observed the CI meet Johnson at that location and exchange items. Sample then tested the substance the CI received from Johnson and identified it as cocaine. Sample followed Johnson from the exchange and stopped him in front of a house for a minor traffic violation. Johnson’s driver license stated he lived at the house where they had stopped, and Johnson told Sample that it was his house. Sample then observed Johnson entering the house before he drove away. Sample then arranged a second buy through the same CI. Again, the CI called the phone number, the man provided a location to meet, and the CI exchanged items with Johnson after they met at that location. Sample tested the substance UNITED STATES V. JOHNSON 7 the CI received from Johnson and it again tested positive as cocaine. Again, police followed Johnson and observed him return to the same home. The first buy occurred within the 20 days preceding the affidavit, and the second buy within 10 days. Sample’s affidavit also provided information about his training and experience. Sample averred that drug traffickers who sold cocaine base often purchased it in bulk quantities and stored it in their cars and homes. Based on the factual information recited above and Sample’s description of his training and experience, the superior court issued a search warrant. The search of Johnson’s home recovered a firearm, ammunition, scales, plastic bags, pills in bottles, and cocaine base. Johnson was indicted on nine counts of drug and firearm offenses. Before trial, Johnson moved to suppress all evidence recovered from the warrantless search of his person and car and the warrant search of his house. The district court denied the motion in two separate orders. Johnson then stipulated to certain facts and the district court held a bench trial. The government dismissed two counts and the district court convicted Johnson on the remaining seven. At sentencing, the district court increased Johnson’s offense level by four levels because he had used body armor during the commission of a drug trafficking crime. Johnson appeals, arguing that the district court erred in denying his motion to suppress and by applying the body armor enhancement to his sentence. II. “We review de novo the district court’s denial of a motion to suppress. We review the district court’s 8 UNITED STATES V. JOHNSON underlying factual findings for clear error.” United States v. Camou, 773 F.3d 932, 937 (9th Cir. 2014) (citations omitted). “We review a district court’s construction and interpretation of the Guidelines de novo and its application of the Guidelines to the facts for abuse of discretion.” United States v. Simon, 858 F.3d 1289, 1293 (9th Cir. 2017) (en banc) (alteration in original removed) (quoting United States v. Popov, 742 F.3d 911, 914 (9th Cir. 2014)). III. Johnson argues that the warrantless search of his person, the warrantless search of his car, and the warrant search of his home all violated the Fourth Amendment. We address each argument in turn. A. The search incident to a lawful arrest exception to the warrant requirement allows a police officer to search an arrestee’s person and the area within the arrestee’s immediate control. Arizona v. Gant, 556 U.S. 332, 339 (2009). It is well-established in this circuit that a search, incident to a lawful arrest, does not necessarily need to follow the arrest to comport with the Fourth Amendment. United States v. Smith, 389 F.3d 944, 951 (9th Cir. 2004) (citing Rawlings v. Kentucky, 448 U.S. 98, 111 (1980)). Instead, probable cause to arrest must exist at the time of the search, and the arrest must follow “during a continuous sequence of events.” Id. If these conditions are satisfied, the fact that the arrest occurred shortly after the search does not affect the search’s legality. UNITED STATES V. JOHNSON 9 It is also well-established that the mindset of an arresting officer is usually irrelevant to a seizure’s legality. See Devenpeck v. Alford, 543 U.S. 146, 153 (2004); Arkansas v. Sullivan, 532 U.S. 769, 771 (2001) (per curiam); Atwater v. City of Lago Vista, 532 U.S. 318, 354 (2001); Whren v. United States, 517 U.S. 806, 813 (1996). Instead, the officer’s state of mind matters only to the extent that probable cause must be based on “the facts known to the arresting officer at the time of the arrest.” Devenpeck, 543 U.S. at 152 . Thus, when the officer’s known facts provide probable cause to arrest for an offense, the officer’s “subjective reason for making the arrest need not be the criminal offense as to which the known facts provide probable cause.” Id. at 153 . The question presented in this case is whether these two well-established principles may coincide without violating the Fourth Amendment. Johnson contends that to do so would create a “search incident to probable cause” rule, allowing officers to search a person whenever probable cause to arrest exists. Johnson argues that the existence of such a rule will cause widespread fishing expeditions that are pre-textual and discriminatory. We conclude that the search of Johnson’s person was constitutional. The search incident to a lawful arrest exception is “based upon the need to disarm and to discover evidence,” but it “does not depend on what a court may later decide was the probability in a particular arrest situation that weapons or evidence would in fact be found upon the person of the suspect.” United States v. Robinson, 414 U.S. 218, 235 (1973). Thus, we do not balance Johnson’s interests in not being searched against Sergeant Simmont’s interest in searching him. Instead we evaluate whether, as a general matter, the justifications for the search incident to lawful 10 UNITED STATES V. JOHNSON arrest exception retain force in the context of a search performed by an officer who has probable cause to arrest and shortly thereafter does arrest. See Knowles v. Iowa, 525 U.S. 113 , 117–19 (1998) (evaluating whether the historical justifications for searches incident to arrest applied to decide whether to recognize a “search incident to citation” exception). The justifications for the exception do not lose any of their force in this context. As explained in Robinson, and reaffirmed in Knowles, when an officer begins an encounter with another person, and probable cause to arrest exists, danger to the police officer “flows from the fact of the arrest, and its attendant proximity, stress, and uncertainty, and not from the grounds for arrest.” Robinson, 414 U.S. at 234 n.5 (emphasis added); see also Knowles, 525 U.S. at 117 (quoting same). As in other contexts then, the precise crime for which an officer has probable cause is irrelevant. Cf. Devenpeck, 543 U.S. at 153 (“[An officer’s] subjective reason for making the arrest need not be the criminal offense as to which the known facts provide probable cause”). So long as the search was incident to and preceding a lawful arrest—which is to say that probable cause to arrest existed and the search and arrest are roughly contemporaneous, Smith, 389 F.3d at 951—the arresting officer’s subjective crime of arrest need not have been the crime for which probable cause existed. Johnson argues that this result is inconsistent with Knowles, in which the Supreme Court held that warrantless searches preceding an arrest but following a citation are unconstitutional. See 525 U.S. at 118 . Knowles is distinguishable. In that case, the issuance of the traffic citation for speeding resolved the encounter’s danger, and “the possibility that an officer would stumble onto evidence UNITED STATES V. JOHNSON 11 wholly unrelated to the speeding offense” was too remote to justify the intrusion. Id. In cases such as the one before us, the danger attendant to the custodial arrest remains until the officer decides to arrest, cite, or warn, and probable cause provides a basis for the officer to search for evidence of that crime. We therefore join our sister circuits in holding that Knowles does not prevent a search incident to a lawful arrest from occurring before the arrest itself, even if the crime of arrest is different from the crime for which probable cause existed. See United States v. Diaz, 854 F.3d 197 , 206–07 (2d Cir. 2017); United States v. Coleman, 458 F.3d 453, 458 (6th Cir. 2006); United States v. Sanchez, 555 F.3d 910 , 921–22 (10th Cir. 2009). As for Johnson’s argument that this standard invites pretextual and discriminatory searches, we do not think that this case is meaningfully different from cases where the search precedes the arrest and the arresting officer’s subjective crime of arrest is the same as the crime for which probable cause existed. The safeguards of probable cause and an actual custodial arrest, here as there, are what protect individuals’ Fourth Amendment rights. To the extent Johnson argues that those safeguards are insufficient, his argument is properly directed at the search-preceding-arrest doctrine more generally, and this panel has no power to overrule circuit precedent, let alone that of the Supreme Court. See Rawlings, 448 U.S. at 111 (“Where the formal arrest followed quickly on the heels of the challenged search of petitioner’s person, we do not believe it particularly important that the search preceded the arrest rather than vice versa”). Johnson’s alternative argument is that, even applying this standard, the search of his person was unconstitutional because Simmont did not have probable cause to arrest. We 12 UNITED STATES V. JOHNSON disagree. “‘[P]robable cause’ to justify an arrest means facts and circumstances within the officer’s knowledge that are sufficient to warrant a prudent person, or one of reasonable caution, in believing, in the circumstances shown, that the suspect has committed, is committing, or is about to commit an offense.” Michigan v. DeFillippo, 443 U.S. 31, 37 (1979). This standard is met when there is a “fair probability” that a crime has been committed. United States v. Smith, 790 F.2d 789, 792 (9th Cir. 1986). The smell of fresh and burnt marijuana in Johnson’s car, along with the plastic baggies in the glove compartment, and Johnson’s unusual search of the glove compartment, indicated a “fair probability” that Johnson had committed, was committing, or was about to commit the offense of marijuana transportation. See CAL. HEALTH & SAFETY CODE § 11360. The search prior to Johnson’s arrest was therefore supported by probable cause. B. When an arrestee is the recent occupant of a vehicle, the arresting officer may search that vehicle if the arrestee is unsecured and within reaching distance of the passenger compartment, or if it is “reasonable to believe evidence relevant to the crime of arrest might be found in the vehicle.” Gant, 556 U.S. at 343 (quoting Thornton v. United States, 541 U.S. 615, 632 (2004) (Scalia, J., concurring in the judgment)). Additionally, under the automobile exception, a police officer may conduct a warrantless search of a vehicle if there is probable cause to believe the vehicle contains evidence of a crime. United States v. Faagai, 869 F.3d 1145, 1150 (9th Cir. 2017). The district court relied on both the search incident-to-lawful-arrest exception and the automobile exception to uphold the warrantless search of Johnson’s car. UNITED STATES V. JOHNSON 13 We conclude that the search was justified under the automobile exception, and therefore do not reach whether the search was also justified as incident to Johnson’s arrest. When Simmont approached Johnson’s car, he immediately smelled a combination of burnt and fresh marijuana. This provided probable cause for Simmont to search the vehicle. See United States v. Barron, 472 F.2d 1215, 1217 (9th Cir. 1973) (“Further, the fact that an agent familiar with the odor of marijuana, smelled such an odor emanating from the automobile when he jumped in to stop it, alone was sufficient to constitute probable cause for a subsequent search for marijuana”). Johnson argues that the search of his car was nonetheless illegal because it was the fruit of the illegal search of his person. But, as we have already explained, that search comported with the Fourth Amendment. There being no poisonous tree, the search of Johnson’s car cannot have been the fruit of an illegal search. C. When reviewing the validity of a search warrant issued by a magistrate, we give “great deference” to the magistrate’s probable cause determination, and will uphold the warrant’s validity if the magistrate was supplied “with a substantial basis for determining the existence of probable cause.” United States v. Leon, 468 U.S. 897 , 914–15 (1984) (first quoting Spinelli v. United States, 393 U.S. 410, 419 (1969), then quoting Illinois v. Gates, 462 U.S. 213, 239 (1983)). “In borderline cases, preference will be accorded to warrants and to the decision of the magistrate issuing it.” United States v. Terry, 911 F.2d 272, 275 (9th Cir. 1990) (quoting United States v. Martinez, 588 F.2d 1227, 1234 (9th Cir. 1987)). Johnson attacks the warrant on three grounds. First, he argues that Sample’s affidavit did not establish probable 14 UNITED STATES V. JOHNSON cause that contraband would be found in his home. Second, he argues that the affidavit did not establish the CI’s reliability. Third, he argues that the affidavit omitted the quantity of cocaine involved in the controlled buys, misleading the magistrate into issuing the warrant. We disagree with these arguments. As to probable cause, this case is controlled by United States v. Terry. In that case, we held that an officer’s “first hand knowledge” of the defendant’s possession of controlled substances, combined with the officer’s “experience with other drug dealers,” provided the “substantial basis” for the magistrate to determine that probable cause existed. Id. at 276. The same holds true here. Sample averred that he had twice observed Johnson distribute cocaine in the 20 days preceding the warrant, including once within 10 days. He also averred that, after the buys, he observed Johnson return to the address listed on the warrant application, which Johnson entered and told police was “his house.” These facts—combined with Sample’s description of how drug traffickers buy cocaine in bulk, sell in small amounts, and use their homes as store caches for the remainder—provided a substantial basis for the San Mateo Superior Court to issue the warrant. See id. at 275–76. As to the CI’s reliability, this argument is largely beside the point. As we have just explained, the basis for probable cause in the affidavit was Sample’s “first hand knowledge” of Johnson’s drug dealing and his “experience with other drug dealers” in how and where a confirmed drug dealer might store contraband. See id. at 276. The warrant was not issued, unlike in other cases where informant credibility is crucial, based on the CI’s tip that drugs would be found in Johnson’s home, but on Sample’s observations of the controlled buys and Johnson’s actions thereafter. The CI’s UNITED STATES V. JOHNSON 15 only role in establishing probable cause was therefore to effectuate the controlled buys, and the CI did establish reliability in this regard because police observed the buys, corroborating the CI’s information. Finally, as to the omission of the size of the cocaine rocks sold, we conclude that any omission was immaterial to the magistrate’s decision. A defendant challenging omissions from a warrant must make a substantial showing that “the affiant intentionally or recklessly omitted facts required to prevent technically true statements in the affidavit from being misleading.” United States v. Stanert, 762 F.2d 775, 781 (9th Cir. 1985). Johnson has not met that standard here. Assuming arguendo that Sample’s omission of the precise sizes of the cocaine rocks was intentional or reckless, it was clear from the affidavit that the controlled buys involved small amounts of cocaine. Therefore, inclusion of those facts would not have changed the meaning of any statement in the affidavit. IV. U.S.S.G. § 3B1.5 provides: “If … the defendant was convicted of a drug trafficking crime or a crime of violence; and … the defendant used body armor during the commission of the offense, in preparation for the offense, or in an attempt to avoid apprehension for the offense, increase by 4 levels.” U.S.S.G. § 3B1.5(1), (2)(B). The district court increased Johnson’s offense level by four levels under this provision because Johnson was wearing a bulletproof vest during the 2015 traffic stop. Johnson argues that the district court erred in doing so because “use” does not mean simply wearing body armor. First interpreting the meaning of the guidelines de novo, we reject Johnson’s construction argument. The 16 UNITED STATES V. JOHNSON commentary to the guidelines states that “‘[u]se’ means (A) active employment in a manner to protect the person from gunfire; or (B) use as a means of bartering. ‘Use’ does not mean mere possession (e.g., ‘use’ does not mean that the body armor was found in the trunk of the car but not used actively as protection).” Id. § 3B1.5, cmt. (n.1). There is no reasonable way to construe this language that would exclude wearing body armor from the definition of “use.” Wearing body armor is the precise means by which a person “employ[s] [the body armor] in a manner to protect the person from gunfire.” Accordingly, Johnson “used” the body armor within the meaning of the guidelines simply by wearing it. Accord United States v. Juarez, 866 F.3d 622, 633 (5th Cir. 2017) (stating “this Court and others have only applied the body-armor enhancement where the defendant committed a crime wearing body armor” and collecting cases (emphasis in original)); United States v. Barrett, 552 F.3d 724 , 727–28 (8th Cir. 2009) (upholding enhancement where defendant wore body armor at party). Next reviewing the district court’s application of the guidelines to the facts of this case, there was no abuse of discretion. Johnson argues that he had an alternative explanation for why he wore the body armor. But Johnson also had cocaine base, heroin, marijuana, and oxycodone on his person while he wore the body armor, as well as a loaded handgun, hydrocodone, plastic bags, scales, and concentrated cannabis in his car. On these facts, the district court did not abuse its discretion by determining that the enhancement should apply because Johnson wore body armor during a drug trafficking offense. AFFIRMED. UNITED STATES V. JOHNSON 17 WATFORD, Circuit Judge, concurring: I join the court’s opinion because it faithfully applies the rule we adopted in United States v. Smith, 389 F.3d 944 (9th Cir. 2004) (per curiam). There, we held that a warrantless search that precedes an arrest may nonetheless fall within the search-incident-to-arrest exception if “the search is conducted roughly contemporaneously with the arrest” and probable cause to arrest existed at the time of the search. Id. at 952 . This rule has not been universally embraced. Many courts have adopted it, but some have rejected it in favor of a more circumscribed approach. See Joshua Deahl, Debunking Pre-Arrest Incident Searches, 106 Cal. L. Rev. 1061 , 1086–87 (2018) (cataloguing the split in authority). In my view, Smith falls on the wrong side of this divide and should be overruled. Generally speaking, the police must obtain a warrant before conducting a search of the “persons, houses, papers, and effects” protected by the Fourth Amendment. The Supreme Court has recognized exceptions to that requirement when circumstances justifying the need to take immediate action are present, such that obtaining a warrant would be impracticable. In the case of searches incident to arrest, the justification for dispensing with the warrant requirement stems from “the fact of custodial arrest.” United States v. Robinson, 414 U.S. 218, 236 (1973). An arrest triggers two important government interests that create the need for an immediate search: protecting officer safety and preventing the destruction of evidence. Id. at 235 . The Court described those interests in Chimel v. California, 395 U.S. 752 (1969): When an arrest is made, it is reasonable for the arresting officer to search the person arrested in order to remove any weapons that 18 UNITED STATES V. JOHNSON the latter might seek to use in order to resist arrest or effect his escape. Otherwise, the officer’s safety might well be endangered, and the arrest itself frustrated. In addition, it is entirely reasonable for the arresting officer to search for and seize any evidence on the arrestee’s person in order to prevent its concealment or destruction. Id. at 762–63. A lawful arrest also results in a reduction in the arrestee’s Fourth Amendment privacy interests. Riley v. California, 134 S. Ct. 2473, 2488 (2014). Together, the need to conduct an immediate search and the suspect’s diminished expectation of privacy render warrantless searches incident to a lawful arrest categorically reasonable under the Fourth Amendment. Robinson, 414 U.S. at 234–35. As the doctrinal underpinnings of the search-incident-toarrest exception suggest, the authority to conduct such a search does not arise until an arrest is actually made. If confirmation of that fact were needed, though, the Supreme Court’s decisions in Cupp v. Murphy, 412 U.S. 291 (1973), and Knowles v. Iowa, 525 U.S. 113 (1998), provide it. In Cupp, the suspect in a murder case came to the police station voluntarily. Although the police had probable cause to arrest him, they did not place him under arrest. 414 U.S. at 293–94. Officers instead temporarily detained the suspect just long enough to allow them to take fingernail scrapings, which he was attempting to destroy in their presence. Given these exigent circumstances, the Court held that a search of this limited scope was reasonable. Id. at 295–96. But the Court made clear that, because the suspect had not been arrested, a more intrusive search of the type permitted incident to arrest likely would have been prohibited. The UNITED STATES V. JOHNSON 19 government interests that justify such a search, the Court explained, were not fully implicated by a seizure that falls short of an arrest: Where there is no formal arrest, as in the case before us, a person might well be less hostile to the police and less likely to take conspicuous, immediate steps to destroy incriminating evidence on his person. Since he knows he is going to be released, he might be likely instead to be concerned with diverting attention away from himself. Accordingly, we do not hold that a full [search incident to arrest] would have been justified in this case without a formal arrest and without a warrant. Id. at 296. In Knowles, the officer had probable cause to arrest the defendant for speeding but chose to issue a citation instead. The officer nonetheless conducted a warrantless search of the defendant’s car, which turned up illegal drugs. The officer then arrested the defendant on drug-related charges. 525 U.S. at 114 . The state courts upheld the validity of the search, “reasoning that so long as the arresting officer had probable cause to make a custodial arrest, there need not in fact have been a custodial arrest.” Id. at 115–16. The Supreme Court unanimously rejected that reasoning. The Court held that issuance of a citation does not trigger either of the government interests that justify a warrantless search incident to arrest. The threat to officer safety is not as pronounced when an officer issues a citation, in part because the officer will not have the extended exposure to the suspect that follows taking him into custody and transporting him to 20 UNITED STATES V. JOHNSON a police facility for booking. Id. at 117 . Nor does the need to preserve evidence arise, for once the officer issues the traffic citation “all the evidence necessary to prosecute that offense” will already have been obtained. Id. at 118 . The Court therefore refused to extend the search-incident-toarrest exception “to a situation where the concern for officer safety is not present to the same extent and the concern for the destruction or loss of evidence is not present at all.” Id. at 119 . 1 I do not think our decision in Smith is consistent with these precedents. We have allowed warrantless searches to be conducted before an arrest is made, provided that the officer has probable cause to arrest “independent of the fruits of the search” and the arrest occurs shortly after the search is conducted. 389 F.3d at 951 . Under that rule, however, the search in Knowles would have been permissible, because the officer had probable cause to arrest the defendant for speeding and he arrested the defendant shortly after the search uncovered illegal drugs. It’s true that in Knowles the officer had already issued a citation by the time the search took place, which some courts have interpreted to mean that a search incident to arrest is permitted so long as the officer has not yet decided whether to arrest or cite the suspect. See, e.g., United States v. Diaz, 854 F.3d 197 , 206–08 (2d Cir. 2017); United States v. Pratt, 355 F.3d 1119 , 1124–25 n.4 (8th Cir. 2004). What those courts have overlooked is that the critical fact in Knowles was not the officer’s issuance of the citation, but rather the absence of an arrest. That absence is key because, as discussed above, the exigency that [1] Officers are not without recourse to protect themselves when interacting with suspects prior to arrest. An officer may still perform a protective “patdown” if the officer has reasonable suspicion to believe that the suspect may be armed and dangerous. Knowles, 525 U.S. at 118 (citing Terry v. Ohio, 392 U.S. 1 (1968)). UNITED STATES V. JOHNSON 21 justifies a warrantless search in this context arises from the fact of arrest, Robinson, 414 U.S. at 236 , not from the existence of probable cause to arrest. The rule we adopted in Smith is doctrinally unsound for another reason: It makes the legality of the search dependent upon events that occur after the search has taken place. Under our rule, the same search conducted on the basis of the same showing of probable cause is valid if the officer arrests the suspect afterward, but becomes illegal if the officer subsequently decides not to make an arrest (say, because the search turns up nothing). See Menotti v. City of Seattle, 409 F.3d 1113, 1153 (9th Cir. 2005). That approach is at odds with the background principle that the reasonableness of a search turns on “whether the officer’s action was justified at its inception, and whether it was reasonably related in scope to the circumstances which justified the interference in the first place.” Terry v. Ohio, 392 U.S. 1 , 19–20 (1968). Beyond that, we should be skeptical of any rule that provides officers with an incentive to make an arrest they would not otherwise have made, solely to insulate themselves from civil liability for violating an individual’s Fourth Amendment rights. The firmest support for the rule we adopted in Smith comes from Rawlings v. Kentucky, 448 U.S. 98 (1980), but even that support is flimsy. The Supreme Court in Rawlings did state that “[w]here the formal arrest followed quickly on the heels of the challenged search of petitioner’s person, we do not believe it particularly important that the search preceded the arrest rather than vice versa.” Id. at 111 . But that language cannot fairly be read as having jettisoned the requirement that an arrest occur before an officer may conduct a search incident to arrest. At the time he was searched, the defendant in Rawlings had plainly been 22 UNITED STATES V. JOHNSON subjected to a Fourth Amendment seizure amounting to an arrest, based on probable cause that existed beforehand. See United States v. Powell, 483 F.3d 836, 846 (D.C. Cir. 2007) (en banc) (Rogers, J., dissenting). As I read Rawlings, the Court merely held that the search was not invalidated by the fact that the “formal arrest” (handcuffing, etc.) occurred shortly after the search took place, rather than before. Moreover, interpreting Rawlings to permit a pre-arrest search, so long as an officer has probable cause to arrest, seems to conflict with the Court’s recent decision in Rodriguez v. United States, 135 S. Ct. 1609 (2015). That case held that a traffic stop may not be “prolonged beyond the time reasonably required to complete the mission of issuing a ticket for the violation.” Id. at 1612 (internal quotation marks and alteration omitted). As part of a traffic stop’s mission, an officer may make traffic-code-related inquiries, such as checking the driver’s license and registration, see id. at 1615 , and may take steps to ensure officer safety, such as asking the driver and any passengers to exit the vehicle. Maryland v. Wilson, 519 U.S. 408 , 414– 15 (1997); Pennsylvania v. Mimms, 434 U.S. 106 , 110–11 (1977) (per curiam). But the Court has indicated that conducting a search of the driver’s person ordinarily falls outside the scope of what a routine traffic stop authorizes. Knowles, 525 U.S. at 117 . It’s not clear how that limitation can be squared with the broad reading of Rawlings we adopted in Smith. In most instances, an officer conducting a valid traffic stop will have probable cause to believe that the driver has committed some traffic infraction, which carries with it the latent authority to arrest even if the infraction is a minor one that could not result in jail time. See Atwater v. City of Lago Vista, 532 U.S. 318, 354 (2001). Thus, as long as the officer has not UNITED STATES V. JOHNSON 23 yet issued a citation, there remains the possibility, however remote, that the driver could be arrested for the infraction. Under our holding in Smith, an officer in these circumstances could presumably conduct a search “incident” to an arrest for the traffic violation, knowing that if incriminating evidence is found he can opt at that point to arrest the driver and thereby validate the search. That regime would authorize full-blown investigatory searches of the driver’s person (and in some instances of the vehicle’s passenger compartment as well) as a normal incident of any traffic stop. Permitting such searches seems to me in obvious tension with Rodriguez’s holding. Requiring that a custodial arrest occur before an officer may conduct a search incident to arrest admittedly raises difficult legal issues, since it’s not always clear when an officer’s interference with a suspect’s Fourth Amendment liberty interests ripens into an arrest. But we have a welldeveloped body of Fourth Amendment case law to guide that inquiry, see, e.g., Florida v. Royer, 460 U.S. 491, 503 (1983) (plurality opinion); Dunaway v. New York, 442 U.S. 200 , 212–13 (1979), which focuses on how a reasonable person in the suspect’s shoes would view the nature of the intrusion, Michigan v. Chesternut, 486 U.S. 567 , 573–74 (1988); Ochana v. Flores, 347 F.3d 266, 270 (7th Cir. 2003). That focus makes sense in this context because the arrestee’s perception that he has been placed under arrest is what triggers the need for an immediate search, to ensure that the arrestee cannot use any weapons to resist or escape and cannot conceal or destroy any evidence on his person. The officer’s subjective intent is of course irrelevant to the existence of a custodial arrest, unless “that intent has been conveyed to the person confronted.” Chesternut, 486 U.S. at 575 n.7. 24 UNITED STATES V. JOHNSON If it turns out that an officer conducts a search prematurely—i.e., before the suspect was in fact arrested for Fourth Amendment purposes—that will not invariably result in suppression of any evidence discovered during the search. The government can still attempt to prove, under the inevitable discovery doctrine, that the officer would have arrested the suspect anyway, without regard to what was found as a result of the search. See, e.g., United States v. Cotnam, 88 F.3d 487 , 495–96 (7th Cir. 1996). Suppression would occur only in cases in which “but for the search there would have been no arrest at all.” People v. Reid, 26 N.E.3d 237, 239 (N.Y. 2014). Those are precisely the cases in which suppression should occur in order to combat the serious potential for abuse that otherwise exists when officers possess unfettered discretion as to whom to target for searches. See Deahl, supra, at 1120–22. As has been noted, “it is no secret that people of color are disproportionate victims of this type of scrutiny.” Utah v. Strieff, 136 S. Ct. 2056, 2070 (2016) (Sotomayor, J., dissenting). So long as Smith remains the law of our circuit, it will only exacerbate this problem.