415 F.3d 88
D.C. Cir.
2005

Lead Opinion

Opinion for the Court filed by Circuit Judge ROGERS.

Concurring opinion filed by Circuit Judge EDWARDS.

Dissenting opinion filed by Circuit Judge ROBERTS.

ROGERS, Circuit Judge.

This appeal challenges the district court’s denial of a motion to suppress evidence on the ground that the police lacked probable cause to search the trunk of a car stopped for a traffic violation. . The question before the court is whether the evidence would have led a.“ ‘prudent, reasonable, cautious police- officer’ to believe that there was a reasonable likelihood the trunk contained contraband” or evidence of a crime. United States v. (Monte) Brown, 374 F.3d 1326, 1328 (D.C.Cir.2004) *90(quoting United States v. Davis, 458 F.2d 819, 821 (D.C.Cir.1972)); see also Illinois v. Gates, 462 U.S. 213, 238, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983). Upon de novo review, see Ornelas v. United States, 517 U.S. 690, 699, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996), we hold that the police lacked probable cause to search the trunk, and accordingly, we reverse the judgment of conviction.

I.

At approximately 1:00 a.m. on May 4, 2002, United States Park Police Officers Jeffrey Garboe and Wayne Johnson observed a 1988 Mercury Marquis without a functioning tag light. The officers initiated a traffic stop based on the absence of the tag light. Prior to approaching the car, they conducted a records check that indicated the car’s temporary license tags had been reported stolen from Fairfax County, Virginia. The officers arrested the driver for the stolen tag offense. When the driver was unable to produce a registration or a driver’s license, the officers conducted a further records check that indicated that his driving privileges had been suspended in Virginia. The officers also checked the vehicle identification number in a computer database, and it yielded an “old listing” from Virginia, meaning that the car had once been registered there but that it was not currently registered. There was no report that the car had been stolen.

After handcuffing the driver and securing him inside their cruiser, the officers searched the passenger compartment of the car, including the glove compartment, for documentation of ownership. They did not find any documentation, contraband, or evidence of criminal activity. Nevertheless, the officers searched the trunk, based on their prior experience of finding “real tags” and “other identifying information about the vehicle” there. Although the officers again did not find the “real tags” or any identifying information, they did find a loaded .25 caliber pistol and ammunition inside a child-sized backpack within the trunk. The officers then transported the driver for booking, leaving the car parked on the public street. According to Officer Johnson, while being transported for booking the driver indicated that the car belonged to his girlfriend. Officer Garboe also remembered the driver making such a statement, including that his girlfriend had purchased the car at an auction a month before, but he somewhat inconsistently could not recall at what point the driver made the statement. The district court did not make a finding on when the officers received this information.

The driver of the car, Tarry M. Jackson, was indicted for one count of unlawful possession of a firearm and ammunition in violation of 18 U.S.C. § 922(g) (2000). The district court denied his motion to suppress the evidence seized from the car trunk. While noting that this case was “a pretty close call” and that the officers’ testimony about why they searched the trunk was “confused,” the district court concluded that “there was a fair probability that a search of the trunk and the backpack would produce evidence related to Jackson’s use of a stolen tag- — perhaps the ‘real’ tags — or information indicating that Jackson was not the owner or authorized user of the vehicle.” The district court acknowledged that “[t]he vehicle had not been reported stolen,” but observed that “the information available to the officers [did not] establish that it was not stolen.” Jackson then conditionally pled guilty to the unlawful possession charge, preserving his right to appeal the denial of the suppression motion. He was sentenced to twenty-one months of incarcera*91tion, three years of supervised release, and a special assessment, and he now appeals.

II.

The Fourth Amendment provides, “The right of the people to be secure in their ... effects, against unreasonable searches and seizures, shall not be violated.” In most instances, searches must be supported by a warrant obtainable upon a showing of probable cause. See, e.g., California v. Carney, 471 U.S. 386, 390-91, 105 S.Ct. 2066, 85 L.Ed.2d 406 (1985); Mincey v. Arizona, 437 U.S. 385, 390, 98 S.Ct. 2408, 57 L.Ed.2d 290 (1978). “It remains a cardinal principle that searches conducted outside the judicial process, without prior approval by judge or magistrate, are per se unreasonable under the Fourth Amendment — subject only to a few specifically established and well-delineated exceptions.” California v. Acevedo, 500 U.S. 565, 580, 111 S.Ct. 1982, 114 L.Ed.2d 619 (1991) (internal quotation marks omitted).

One exception allows the police to search a vehicle’s passenger compartment, including the glove compartment, incident to the lawful arrest of the vehicle’s occupant. New York v. Belton, 453 U.S. 454, 460, 101 S.Ct. 2860, 69 L.Ed.2d 768 (1981); see also Thornton v. United States, 541 U.S. 615, 124 S.Ct. 2127, 158 L.Ed.2d 905 (2004). The rationale behind the exception is that “articles inside the relatively narrow compass of the passenger compartment of an automobile are in fact generally, even if not inevitably, within ‘the area into which an arrestee might reach in order to grab a weapon or evidentiary ite[m].’ ” Belton, 453 U.S. at 460, 101 S.Ct. 2860 (quoting Chimel v. California, 395 U.S. 752, 763, 89 S.Ct. 2Ó34, 23 L.Ed.2d 685 (1969)) (alteration in original). The officers conducted a search of the passenger compartment, .and Jackson raises no objection to that search. Jackson’s arrest for traffic violations and stolen tags, however, did not automatically permit the officers to search the car’s trunk. See id. at 461 n. 4, 101 S.Ct. 2860; see also Pennsylvania v. Labron, 518 U.S. 938, 940, 116 S.Ct. 2485, 135 L.Ed.2d 1031 (1996). Rather, although a search warrant was not required, the officers could constitutionally search the trunk (and the containers therein) only if. they had probable cause to believe that the trunk contained contraband or evidence of a crime. See Acevedo, 500 U.S. at 579-80, 111 S.Ct. 1982.'

Probable cause is synonymous with “fair probability,” Gates, 462 U.S. at 238, 103 S.Ct. 2317, and it is an objective standard requiring an analysis of the totality of the circumstances and the facts known to the officers at the time of the search, Ornelas, 517 U.S. at 695-96, 116 S.Ct. 1657; Gates, 462 U.S. at 230-31, 103 S.Ct. 2317; cf. United States v. Arvizu, 534 U.S. 266, 274, 122 S.Ct. 744, 151 L.Ed.2d 740 (2002). “Subjective intentions play no role in ordinary, probable-cause Fourth Amendment analysis,” Whren v. United States, 517 U.S. 806, 813, 116 S.Ct. 1769, 135 L.Ed.2d 89 (1996), and the officers’ “actual motives for conducting the search [are] not relevant as long as [their] actions were objectively reasonable.” United States v. (Rocky Lee) Brown, 334 F.3d 1161, 1172 n. 8 (D.C.Cir.2003) (quoting United States v. Christian, 187 F.3d 663, 670 (D.C.Cir.1999)) (internal quotation marks omitted); see also Devenpeck v. Alford, — U.S.-,-, 125 S.Ct. 588, 593-94, 160 L.Ed.2d 537 (2004); United States v. Holmes, 385 F.3d 786, 790 (D.C.Cir.2004). The court has recognized that “the discovery of contraband in the passenger compartment of a car is a factor that strongly supports the lawfulness of a trunk search.” (Rocky Lee) Brawn, 334 F.3d at 1173. And so long as probable cause exists to search the trunk, police *92officers may also search any of the trunk’s contents “that may conceal the object of the search.” United States v. Ross, 456 U.S. 798, 825, 102 S.Ct. 2157, 72 L.Ed.2d 572 (1982); see also Wyoming v. Houghton, 526 U.S. 295, 300-02, 119 S.Ct. 1297, 143 L.Ed.2d 408 (1999).

Both in the district court and in this court upon de novo review, “the burden is on those seeking the exemption [from the warrant requirement] to show the need for it.” United States v. Jeffers, 342 U.S. 48, 51, 72 S.Ct. 93, 96 L.Ed. 59 (1951); see also Chimel, 395 U.S. at 762, 89 S.Ct. 2034; In Re Sealed Case, 153 F.3d 759, 764 (D.C.Cir.1998). In this instance, it is the government’s burden to demonstrate that there was a fair probability that the car trunk would contain contraband or evidence of a crime. Because the totality of the circumstances known to the officers at the time of the search of the car trunk, including that the driver likely had committed several traffic violations and had received stolen property, do not support a determination of probable cause, the government has not carried its burden. .

According to the government, at the time of the search of the car trunk, the officers had probable cause to believe that the driver had committed several offenses: three traffic violations for which he received citations — driving with a suspended license, D.C. Code Ann. § 50-1403.01 (2001), operating an unregistered motor vehicle, id. § 50-1501.04(a)(l)(A), driving without required vehicle identification tags, id. § 50-1501.04(a)(l)(B) — and two criminal offenses — receiving stolen property (the stolen tags), id. § 22-3232, and unauthorized use of a motor vehicle, id. § 22-3215. Therefore, the government maintains, the search of the trunk was objectively reasonable because the officers had probable cause to believe that they would find “additional contraband, such as one or more additional stolen tags” in the trunk, Br. of Appellee at 14, or that “further evidence concerning this range of probable criminal activity might well have been concealed in the trunk of the car,” id. at 13. The government does not rely expressly on the district court’s determination that the search was permissible based on the fair probability of finding the car’s “real tags,” but, consistent with the district court’s ruling, it does maintain that the officers had probable cause to search for documentation of ownership, assuming, as it does, there was probable cause to believe that the driver was an unauthorized user of the car.

The government’s first justification— that the officers had probable cause to search for contraband — is readily dismissed. There can be no serious argument that the existence of stolen tags affixed to a car gives rise to probable cause to believe that additional contraband, particularly additional stolen tags, would be in the car trunk. Contrary to the government’s contention, this case is not similar to (Monte) Brown, 374 F.3d 1326, where a fraudulent credit card, personal check, and driver’s licenses found in a vehicle’s passenger compartment gave rise to probable cause to believe that items fraudulently obtained would be found in the trunk. In so holding, the court determined that the evidence in the passenger compartment led a “ ‘prudent, reasonable, cautious police officer’ to believe that there was a reasonable likelihood the trunk contained contraband.” Id. at 1328 (quoting Davis, 458 F.2d at 821). This was so, the court explained, because of the significant correlation between what was found in the passenger compartment indicating a desire to engage in fraudulent transactions and the likely fruits of acting on that desire that could be in the trunk. Here, on the other hand, finding stolen tags affixed to the car *93provides no similar correlation that items related to the use of the stolen tags would be located in the trunk. In fact, it is difficult to conceive of what contraband would be associated with stolen tags, wherever found, that is of a nature similar to fraudulent documents and the resulting fraudulent purchases. Stolen tags affixed to a car are an end in and of themselves, and they do not point-to related contraband that may be present in the trunk.

This search is also easily distinguishable from the searches found to be permissible in (Rocky Lee) Brown, 334 F.3d 1161, and United States v. Turner, 119 F.3d 18 (D.C.Cir.1997). In (Rocky Lee) Brown, the court relied on, among other things; the empirical connection between guns discovered in a car’s passenger compartment and the presence of other contraband in the trunk. 334 F.3d at 1171 (citing cases). Similarly, in Turner, the court relied on the established connection between drugs in the passenger compartment and the presence of additional drugs in the trunk. 119 F.3d at 20-21 (citing cases). In the instant case, there is no similar established connection between stolen tags displayed on a vehicle and additional contraband. Nor does the presence of displayed stolen tags suggest a fair probability that there is an additional stolen tag(s) in the trunk, especially where, as here, Officer Garboe testified at the suppression hearing that the car contained both front and back tags and therefore was not missing any tags.

As to the second justification for searching the trunk, the government is unable to explain what further- evidence pertaining to the driver’s probable criminal activity, viewed -cumulatively, would be located there. It is entirely implausible that the trunk would contain additional evidence to support the charges of driving on a suspended license, operating an unregistered vehicle, and driving without required vehicle identification tags. Similarly, the charge of receiving stolen property for the stolen tags did not necessitate a search of the trunk because evidence of the offense was displayed on the car and, as noted, there was not probable cause to believe that the trunk contained additional, related contraband. In fact, at the suppression hearing, Officer Garboe acknowledged that prior to the search of the trunk he had all of the evidence that he needed to arrest the driver for the above offenses and to impound the car based on the records check and the visual inspection of the car. Our analysis thus is consistent with Knowles v. Iowa, 525 U.S. 113, 118, 119 S.Ct. 484, 142 L.Ed.2d 492 (1998), where the Supreme Court declined to extend the “bright-line rule” for the search-ineident-to-arrest exception and reversed the denial of a motion to suppress evidence found in a car’s passenger compartment, where the police, having “stopped [a car] for speeding and issued a citation [to the driver, had] all the evidence necessary to prosecute that offense.”

Turning to the officers’ professed need to search the trunk for evidence of ownership, the district court justified the search on the grounds that “the information available to the officers [did not] establish that [the car] was not stolen.” The lack of evidence indicating that the car was stolen cuts against, rather than supports, a finding of probable cause to search the trunk because the result would authorize officers to search a vehicle anytime -it is unregistered, a proposition the government does not advance. The officers must have probable cause to believe that documentation demonstrating that the driver was not authorized to drive the car would be in the trunk; searching the trunk for documentation establishing or confirming that the driver properly possessed the car would not constitute contraband or evidence of a crime as is-required under the probable *94cause standard. See Acevedo, 500 U.S. at 579-80, 111 S.Ct. 1982. But that is exactly what the officers did here.

The officers’ computer records checks did not indicate that the car was stolen, nor did the records checks indicate to whom the car was formerly registered. Although our dissenting colleague characterizes the result of the records checks as “unusual,” Dissenting Op. at 102, the result was simply an “old listing” that, for whatever reason, did not include the former registrant’s name; the record does not suggest that such a result was suspicious. Similarly, the search of the passenger compartment did not reveal any evidence that the driver was an unauthorized user of the car. For reasons already discussed, considered commutatively, other circumstances that the officers encountered — a driver with a suspended license driving late at night with a broken tag light and without a registration — do not affect the probability that the driver was an unauthorized user. Otherwise, borrowing a Mend’s car becomes a very risky undertaking. This leaves the stolen tags as the critical feature of this traffic stop.

Under the circumstances, we can conceive of only three reasons of varying likelihood why stolen tags would be on the car, and the government has not suggested any others. First, stolen tags may be placed on an otherwise lawfully used car without tags to give the appearance of legitimate tags and therefore to reduce the risk that the police will initiate a traffic stop for lack of tags. Second, stolen tags may be used to replace expired tags on an otherwise lawfully used vehicle, again in the hope of avoiding immediate detection. The lack of registration and the absence of a report that the car was stolen are consistent with these first two rationales, which suggest that the driver was an authorized user of the ear. Third, stolen tags may be used to conceal the fact that a vehicle is stolen by replacing the stolen vehicle’s “real tags.” However, as Jackson’s counsel pointed out during oral argument, the government has not demonstrated a significant correlation between the presence of stolen tags and the vehicle itself being stolen, and a case that our dissenting colleague cites, United States v. Barlow, 41 F.3d 935, 939 (5th Cir.1994), illustrates the government’s problem: using stolen tags to obscure the fact that a vehicle is stolen at best may momentarily delay police discovery that the car is stolen while a records check is made of the car’s tags, for in Barlow, the officer determined through a records check that the vehicle possessed stolen tags before determining that the vehicle itself was stolen. Because the officers here were confronted with three possible explanations for the presence of the stolen tags on the car, two of which suggested authorized use and were consistent with the lack of registration and the absence of a report that the car was stolen, and only one of which supported an inference of unauthorized use, the officers lacked probable cause to search the trunk for documentation that the driver was an unauthorized user of the car.

While the existence of probable cause does not depend on the elimination of all innocent explanations for a situation, Gates, 462 U.S. at 243 n. 13, 103 S.Ct. 2317, our dissenting colleague, although acknowledging the values underlying the Fourth Amendment, see Dissenting Op. at 105, posits the most incriminating interpretation of the circumstances, as though the existence of countervailing probabilities was irrelevant. Were that the law, then the government’s burden would be considerably eased, for the particular circumstances causing the police to make a traffic stop could often be viewed most negatively without regard to a citizen’s *95Fourth Amendment protections. The Fourth Amendment requires a different analysis, as the concurring opinion of Judge Edwards makes clear. See Concurring Op. at 98, 101. That the Fourth Amendment places a heavy burden on the government is apparent from our car-trunk search cases, which carefully articulated the substantiality of the connection between information known to the officers and the likelihood of contraband in the car trunk. See, e.g., (Monte) Brown, 374 F.3d at 1328-29; (Rocky Lee) Brown, 334 F.3d at 1171. Here the government attempts to elide that burden by ignoring the explanations indicating authorized use and instead hastily asserting that there was a fair probability that Jackson was an unauthorized user of the car.

Our ' dissenting colleague emphasizes that Officer Garboe also testified that on six or seven occasions he had encountered vehicles with stolen tags that had “real tags” or other identifying information in' the trunk, see Dissenting Op. at 102, but the government on appeal does not embrace the aspect of the district court’s ruling that justified the search of the trunk on the possibility of finding “real tags.” There are multiple sensible reasons for the government’s approach that, consequently, undercut the dissent’s position. Even if there was probable cause to believe that the trunk would contain the car’s expired “real tags,” these tags, like a tool kit, are neither contraband nor evidence of a crime because there is nothing illegal about having such tags in the trunk of an unregistered car. In overlooking this point, our dissenting colleague, see Dissenting Op. at 102, posits an evidentiary inference based on finding “real tags” in the trunk that is irrelevant in the absence of probable cause to believe that the trunk contained contraband or evidence of a crime. Further, the record does not indicate that the car’s expired “real tags” would provide the officers with any additional information regarding the ownership of the car because a records check based on the vehicle identification number indicated only an “old listing.” In any event,- even if “real tags” or identifying information could in some instances constitute contraband or evidence of a crime, the officer’s prior experience of finding such information in a vehicle trunk, while relevant, see (Monte) Brown, 374 F.3d at 1328, is unhelpful here because his testimony is devoid of , the critical circumstances of those searches, including whether the identifying information revealed that the vehicle was stolen. Without this essential detail, it cannot be said that Officer Garboe’s past experience revealed that trunks of vehicles with stolen tags often contain contraband or evidence of a crime in the form of identifying information, as opposed merely to containing ownership information confirming that the driver is an authorized user.

The government’s difficulty in demonstrating why the officers’ experience is pertinent here, and in explaining why the presence of expired “real tags” in the trunk is at all relevant to the probable-cause inquiry, may be due in part to what the district court characterized as the officers’ confusion about why they searched the trunk and to the changing testimony of Officer Garboe. Officer Garboe had testified" before the grand jury that only one stolen tag was on the car, and the government accordingly argued in opposing Jackson’s motion to suppress that the officers had probable cause to believe the. other stolen tag or the “real tags” might be in the car trunk. Indeed, the district court, in denying the motion to suppress, continued to refer to Jackson’s use of “a stolen tag.” By the time Officer Garboe testified at the suppression hearing, however, he acknowledged there were stolen tags on both the front and back of the car, thus *96eliminating a strong strain of the government’s argument.

In searching the trunk, the Fourth Amendment makes clear that the officers stopped their investigation too soon. While the fact that a car has stolen tags may, in some instances, suggest that the car itself is stolen and therefore may provide probable cause to search for documentation of ownership, no such inference could be drawn here. Instead of establishing probable cause and justifying a search of the trunk, the lack of information about the driver’s authority to use the car and the ownership of the car should -have served to prompt further inquiry. See Bigford v. Taylor, 834 F.2d 1213, 1218-19 (5th Cir.1988); cf. United States v. Mayo, 394 F.3d 1271, 1276 (9th Cir.2005). Our dissenting colleague conveniently ignores that nothing in the record indicates that, at the time the officers searched the car, the driver, who was secured inside the cruiser in handcuffs, had reason to believe that he was suspected of being an' unauthorized user of the car;' thus he had no reason to volunteer an explanation for his use of the car. The cases cited by the government illustrate that a prudent, cautious, and experienced officer would seek information from the driver precisely because the driver’s responses to police inquiries can clarify the situation and may provide probable cause for a further search of a vehicle. See, e.g., United States v. Maher, 919 F.2d 1482 (10th Cir.1990); United States v. Owens, 346 F.2d 329 (7th Cir.1965). H.ad the officers, for example, inquired of the driver about the ownership of the car and how he came to be driving it, then based on his answers and demeanor they may have been able to establish probable cause to believe contraband or evidence of a crime was in the trunk, or the driver’s response could have confirmed the lack of-probable cause. And, contrary to the view of our dissenting colleague, see Dissenting Op. at 104 -105, the officers’ questions would not have been futile, as they had “ready means” of verifying ownership of the car: they could have called the purported owner and had her come to the scene with proof, much like Officer Garboe testified he does upon finding verified proof of ownership in a vehicle.

This is not a case where the officers inquired of the driver about who owned the car, and the driver was unable to provide a reasonable or consistent explanation. Officer Johnson testified that Jackson did identify the owner, but only while he was being driven to the police station after the car trunk had been searched; Officer Garboe confirmed that fact and while he was uncertain about when Jackson identified the owner, he never suggested the identification Jackson provided was unverifiable. Nor is this a case where ownership documents in the passenger compartment were inconsistent with the driver’s explanation for being in possession of the car. Rather, the ambiguity of the circumstances presented the officers, like the officers in the cases on which the government relies, with the need to continue their investigation. While courts are not to dictate proper investigative techniques for law enforcement officers, see United States v. Montoya de Hernandez, 473 U.S. 531, 542, 105 S.Ct. 3304, 87 L.Ed.2d 381 (1985), and we do not presume to identify “preferred investigative procedures” here, Dissenting Op. at 105, those cases suggest that the officers could have continued their investigation here by asking the driver a few questions to determine whether it would be reasonable to conclude -that documentation of the driver’s unauthorized use of the ear would be in the trunk. Instead, lacking any such indication that the driver was an unauthorized user of the car and lacking any documentation in the passenger compartment *97suggesting that he did not have authorization, the officers nonetheless proceeded to search the trunk.

This also is not a case where the officers were faced with an uncooperative or non-communicative person and therefore were unable to obtain even basic information regarding, ownership or identity. Cf. Hiibel v. Sixth Judicial Dist. Court, 542 U.S. 177, 124 S.Ct. 2451, 159 L.Ed.2d 292 (2004). Rather, the officers testified that the driver did not resist in any way, that he was cooperative, and that he provided his name and date of birth upon request. His interactions with the officers indicate that he was lucid. Nor was this a fast-moving, quickly evolving situation where the officers were forced to process the facts before them rapidly and to make a hasty decision regarding the search of the trunk. Cf. Graham v. Connor, 490 U.S. 386, 396-97, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989). To the contrary, at the time the officers opted to search the trunk, they had secured the driver and spent additional minutes searching the passenger compartment. The circumstances did not indicate that the officers were concerned about their personal safety at the time they searched the trunk, cf. Belton, 453 U.S. at 460, 101 S.Ct. 2860, nor that they had reason to search the trunk in order to preserve evidence of a crime, cf. Schmerber v. California, 384 U.S. 757, 770-71, 86 S.Ct. 1826, 16 L.Ed.2d 908 (1966), much less to remove dangerous contraband, see Cady v. Dombrowski, 413 U.S. 433, 447-48, 93 S.Ct. 2523, 37 L.Ed.2d 706 (1973). Finally, there were nó circumstances from which the officers could imply that the driver had consented to the search of the trunk, and the officers did not testify that he had, much less that they ever had asked for his consent.

Instead, the conduct of the officers demonstrated a lack of appreciation for the distinction between a permissible search of the passenger compartment incident to a lawful arrest, see Belton, 453 U.S. at 460, 101 S.Ct. 2860, and an unconstitutional search of a car trunk in the absence of probable cause, see id. at 460 n. 4, 101 S.Ct. 2860; Acevedo, 500 U.S. at 579-80, 111 S.Ct. 1982. The exception to the warrant requirement imposes a duty on law enforcement officers to gather sufficient evidence for a determination of probable cause. Id. Otherwise, the exception for searches of passenger compartments incident to arrest would become an exception for car trunks as well, a proposition the Supreme Court has not endorsed, cf. Knowles, 525 U.S. at 118, 119 S.Ct. 484, and that is inconsistent with the Court’s rationale for the passenger-compartment exception, see Belton, 453 U.S. at 460, 101 S.Ct. 2860. Although Officer Garboe testified that he and his partner desired to find documentation of ownership so as to not have to leave the car parked on the public street, in the absence of probable cause to search the car trunk for documentation that the driver is an unauthorized user, the constitutional route is to impound the car, cf. South Dakota v. Opperman, 428 U.S. 364, 96 S.Ct. 3092, 49 L.Ed.2d 1000 (1976), or, as the officers did here, to leave it on the public street.

Of course, the fact that a car has not been reported stolen is not determinative of whether an officer has probable cause to believe that it is stolen and to search the trunk for evidence of ownership. Cf. United States v. Brigham, 382 F.3d 500, 509 (5th Cir.2004). But the cases upon which the government and our dissenting colleague rely are manifestly dissimilar to the circumstances here. In fact, the government does not cite a single case where a vehicle with stolen tags and without any documentation of ownership, alone, provided probable cause to search the trunk. In *98United States v. Owens, 346 F.2d 329 (7th Cir.1965), for instance, the police had probable cause to believe a vehicle was stolen because the driver did not recognize the name of the vehicle’s owner. Similarly, contrary to the dissent’s characterization of United States v. Maher, 919 F.2d 1482 (10th Cir.1990), see Dissenting Op. at 104, the driver’s inability to provide the complete name and address of the person whom he claimed sold him a trailer assisted the court in finding probable cause to arrest the driver. But here, the officers did not ask the driver about ownership prior to searching the trunk and therefore could not evaluate the plausibility of his explanation. Further, in Botts v. United States, 310 A.2d 237 (D.C.1973), the inspection sticker was expired, and the police could not check whether the vehicle ha$ been reported stolen because their computer was inoperable. Here, the officers conducted a computer records check and determined that the car had not been reported stolen, leaving the circumstances ambiguous and not implausibly inconsistent with authorized use. The reliance by the government and the dissent, see Dissenting Op. at 104 -105, on isolated dicta in a footnote in United States v. Robinson, 471 F.2d 1082, 1104 n. 38 (D.C.Cir.1972) (en banc), rev’d, 414 U.S. 218, 94 S.Ct. 467, 38 L.Ed.2d 427 (1973), indicating that “some courts have held that when a car has no license plates, or fictitious plates, and the driver cannot produce proof of ownership, probable cause exists to believe that the car may have been stolen,” is of no moment. The court neither indicated that it endorsed that approach nor provided any analysis of the constitutional requirements for probable cause; moreover, here the officers searched the car trunk for documentation of ownership even though neither a search of the passenger compartment nor any discussion with the driver provided an indication that the car was stolen.

Accordingly, because the officers lacked probable cause to search the car trunk for additional contraband, such as additional stolen tags, other evidence concerning the driver’s probable criminal activity, or documentation that the driver was an unauthorized user of the car, the district court erred in denying the motion to suppress the evidence seized from the trunk, and we reverse the judgment of conviction.





Concurrence Opinion

HARRY\T. EDWARDS, Circuit Judge,

concurring.

The needs of law enforcement stand in constant tension with the Constitution’s protections of the individual against certain exercises of official power. It is precisely the predictability of these pressures that counsels a resolute loyalty to constitutional safeguards. It is well to recall the words of Mr. Justice Jackson, soon after his return from the Nuremberg Trials:
“These [Fourth Amendment rights], I protest, are not mere second-class rights but belong in the catalog of indispensable freedoms. Among deprivations of rights, none is so effective in cowing a population, crushing the spirit of the individual and putting terror in every heart. Uncontrolled search and seizure is one of the first and most effective weapons in the arsenal of every arbitrary government. ” Brinegar v. United States, 338 U.S. 160, 180, 69 S.Ct. 1302, 93 L.Ed. 1879 (Jackson, J., dissenting).

Almeida-Sanchez v. United States, 413 U.S. 266, 273-74, 93 S.Ct. 2535, 37 L.Ed.2d 596 (1973) (alteration in original).

* * * * Hs

“It [is] a cardinal principle that searches conducted outside the judicial process, without prior approval by judge or magis*99trate, are per se unreasonable under the Fourth Amendment — subject only to a few specifically established and well-delineated exceptions.” California v. Acevedo, 500 U.S. 565, 580, 111 S.Ct. 1982, 114 L.Ed.2d 619 (1991) (internal quotation marks omitted). The officers in this case obtained no warrant before searching the trunk of the car Tarry Jackson was driving. The Government invokes the so-called “automobile exception” established by the Court in Carroll v. United States, 267 U.S. 132, 45 S.Ct. 280, 69 L.Ed. 543 (1925), which, because of the exigency arising out of an automobile’s likely disappearance, permits warrantless searches of moving vehicles. Although that exception reheves officers of the duty to obtain a warrant, it retains the requirement that there exist “probable cause to believe that the vehicle contain[s] evidence of crime.” Acevedo, 500 U.S. at 569, 111 S.Ct. 1982. Accordingly, the question before us is whether, “given all the circumstances” preceding the search, there existed “a fair probability that contraband or evidence of a crime [would] be found” in the trunk. Illinois v. Gates, 462 U.S. 213, 238, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983) (reaffirming an objective “totality-of-the-circumstances analysis” in probable cause determinations).

******

The police officers in this case were faced with an unlicensed driver operating a car with a broken tag light and stolen tags. The car itself was unregistered, and a records check revealed that it had not been reported stolen. The District Court concluded that a fair probability existed that the officers might find the car’s “real tags” in the trunk. Wisely, the Government does not defend this argument before us, for, as Jackson points out, there is no reason to believe that an unregistered car would have any legitimate tags. .Br. for Appellant at 9-10. Instead, the. Government presents two arguments in support of its claim that the facts here gave rise to probable cause to search the trunk. Neither argument is persuasive.

First, the Government argues that the presence of stolen tags on the car creates a fair probability that' “additional contraband, such as one or more additional stolen tags, might well be in the trunk too.” Gov’t Br. at 14. This is a preposterous contention. The fact that a car has stolen tags affixed to it creates no good reason to believe that additional contraband will be found in the trunk. The Government’s reliance on United States v. Rocky Brown, 334 F.3d 1161 (D.C.Cir.2003), and United States v. Turner, 119 F.3d 18 (D.C.Cir.1997), is unavailing. Those cases stand only for the unremarkable proposition that finding drugs or guns in the passenger compartment of a car gives rise to probable cause to search the trunk for additional contraband. Finding stolen license plates affixed to a car, however, is fundamentally different from finding drugs or guns in the passenger compartment.

Unlike in Brown and Turner, where we noted the empirical connection between the existence of guns and drugs in a passenger compartment and additional contraband in the trunk, there is no evidence before us to suggest that a person who affixes stolen tags to a car is likely to have additional stolen tags or related contraband in the trunk. And the Government fails to provide any specific explanation why such an inference should be drawn.

The Government’s reliance on United States v. Monte Brown, 374 F.3d 1326 (D.C.Cir.2004), is similarly misplaced. In that case, we held that a fraudulent credit card and fraudulent driver’s licenses found in the pássenger compartment of a car created a fair probability that items fraudulently purchased with the documents would be found in the trunk. . Finding *100stolen tags on a car creates no similar inference that related contraband will be located in the trunk. One is hard pressed even to imagine what contraband could be associated with stolen tags in a manner similar to the way in which fraudulent purchases are associated with fraudulent credit cards and identification.

At bottom, the Government’s position would require us to accept the view that any time contraband is found in a person’s possession, a fair probability exists that additional contraband will be found in other areas under the person’s control. In other words, the Government would have us hold that evidence indicating that a person may have committed one crime, without more, invariably gives rise to probable cause to believe that he has committed others. This position is clearly untenable under our Fourth Amendment jurisprudence.

The Government’s second contention is that the stolen tags gave the police officers reason to believe that the car was stolen, creating a fair probability that evidence related to the theft would be found in the trunk. This might be a tenable argument in some cases, because it is not entirely implausible to assume that people who steal cars may replace the stolen car’s real tags with stolen tags in order to conceal the true identity of the car. In this case, however, such an inference cannot be easily drawn, because the police officers knew that the car was unregistered and that it had not been reported stolen.

If a car is not registered, then it has no legitimate tags. The most reasonable inference to be drawn in this situation is that the owner has placed stolen tags on the car to avoid being stopped for driving without tags, while avoiding the expense attendant to registering the car and obtaining legitimate tags. In other words, if a car has no legitimate tags because it is unregistered, then police officers have no good reason to assume that the stolen tags are intended to conceal the true identity of the vehicle. Moreover, the explanation Jackson gave to the officers — that he had borrowed the car from his girlfriend who had recently purchased it at an auction — is reasonable on its face and comports with the inference that the stolen tags were affixed not to conceal that the car was stolen but instead because the car had no legitimate tags. Accordingly, considering the totality of the circumstances — including the known and undisputed facts that the car was unregistered and had not been reported stolen — the officers lacked probable cause to search the trunk for evidence that the car was stolen.

In short, before the officers conducted their search of the vehicle’s trunk, there was no fair probability that contraband or evidence of a crime would be found. Therefore, the District Court erred in holding that the police officers were engaged in objectively reasonable law enforcement activity when they searched the trunk of the car that Jackson was driving.

* * * * * *

There is always a temptation to turn a blind eye to invasions of citizens’ Fourth Amendment rights in the face of potentially inculpatory evidence. Judges are not immune from the burdens of human nature, so we are invariably tested when asked to exclude evidence that tends to prove a defendant’s guilt. “The cost to the truth-seeking process of evidentiary exclusion invariably is perceived more tangibly in discrete prosecutions than is the protection of privacy values through deterrence of future police misconduct.” James v. Illinois, 493 U.S. 307, 319, 110 S.Ct. 648, 107 L.Ed.2d 676 (1990). But judges must resist the temptation to ignore unconstitutional conduct by police officers, because it is our sworn obligation to *101show “jealous regard for maintaining the integrity of individual rights” and “ ‘resist every encroachment upon rights expressly stipulated for in the Constitution by the declaration of rights.’ ” Mapp v. Ohio, 367 U.S. 643, 647, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961) (quoting I Annals of Cong. 439 (1789) (remarks of James Madison)).

In applying the exclusionary rule, courts “must focus on systemic effects ... to 'ensure that individual liberty from arbitrary or oppressive police conduct does not succumb to the inexorable pressure to introduce all incriminating evidence, no matter how obtained, in each and every criminal case.” James, 493 U.S. at 319-20, 110 S.Ct. 648. “The occasional suppression of illegally obtained yet probative evidence,” distasteful though it may seem in the context of a particular case, “has long been considered a necessary cost of preserving overriding constitutional values.” Id. at 311, 110 S.Ct. 648. It is our duty to maintain the sanctity of the constitutional right to privacy free from unreasonable government intrusion.





Dissenting Opinion

ROBERTS, Circuit Judge,

dissenting.

The question for the court is whether the circumstances of the stop and arrest of Tarry Jackson presented “a fair probability that contraband or evidence of a crime [would] be found” in the trunk of the- car he was driving. Illinois v. Gates, 462 U.S. 213, 238, 103 S.Ct. 2317, 76 L.Ed.2d 527 (1983). Like the district court, I conclude that they did, and therefore dissent.

The officers who stopped Jackson had good grounds for believing that the ear he was driving had been stolen and that relevant evidence could be found in the. trunk. It was late at night and the tag light was out — suggesting from the beginning of the encounter that Jackson was attempting to obscure the car’s license plates. Once Jackson was pulled over, the . officers learned there was indeed something to hide: the temporary tags affixed to the car had been stolen and altered to match the car’s make, model, and vehicle identification number.

Stolen tags often accompany stolen cars. See, e.g., United States v. Rhind, 289 F.3d 690, 692 (11th Cir.2002) (defendants traveled in stolen car with stolen plates); United States v. Rose, 104 F.3d 1408, 1411 (1st Cir.1997) (same); United States v. Barlow, 41 F.3d 935, 939 & n. 6 (5th Cir.1994) (same). The reason is obvious: by replacing the real tags with stolen tags, the thief makes it impossible for police to identify a stolen vehicle by sight. A stolen vehicle will normally be described by its make, model, and license plate number. An officer cruising the streets cannot readily identify a particular Mercury Marquis as the stolen Mercury Marquis if the original tags have been replaced. See Turner v. United States, 623 A.2d 1170, 1172 (D.C.1993) (officer noting that “a lot of time if a crime was to go down, say like a stolen vehicle, ... license plates can easily be switched” (alteration in original)). So the stolen tags raised a suspicion that the car may have been stolen as well.

The officers’ records check on the vehicle did nothing to dispel this suspicion. Had Jackson been able to produce the car’s registration or had the records check indicated that it was his car, the police would have been reasonably certain they were dealing only with stolen tags, a broken tag light, and a driver with a suspended license. But no such reassurance was forth-coming: Jackson himself could produce no license and no proof of registration, and the records check, which revealed only an old listing for the vehicle, did not show Jackson as the owner.

My colleagues seem to believe that the inconclusive records check somehow dissipated any suspicion that the car was sto*102len. See Op. at 93. To the officers on the scene, however, the failure of the records check to resolve ownership of the vehicle was unusual. See Hr’g Tr., June 9, 2003, at 9-10 (“Normally if you run [a registration check] haying already run the operator, it’ll tell you that it comes back with an expired listing to that operator. And that was not the case in this case.”) (emphasis added) (Officer Garboe). The fact that Jackson was not listed on the car’s last registration could reasonably have heightened the officers’ suspicion: now they were dealing not only with a car with stolen tags, but with a car that had no recorded connection to Jackson.

Given the cumulation of suspicious circumstances suggesting the ear may have been stolen, the officers, reasonably in my view, turned their attention to the trunk. Why the trunk? One of the officers at the scene would later testify that he had made about ten previous vehicle stops involving stolen tags. Hr’g Tr., June 9, 2003, at 13-14 (Officer Garboe). On six or seven of those occasions, he had found the vehicle’s real tags in the trunk. Id. at 19; see Ornelas v. United States, 517 U.S. 690, 700, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996) (“our cases have recognized that a police officer may draw inferences based on his own experience in deciding whether probable cause exists”); United States v. Brown, 374 F.3d 1326, 1328 (D.C.Cir.2004) (“ ‘probable cause’ is evaluated not only from the perspective of a ‘prudent man,’ but also from the particular viewpoint of the officer involved in the search or seizure” (citation omitted)).1 This is not especially surprising: the trunk is certainly a convenient place to stash the real tags once they have been removed from the back of the vehicle. See Brown, 374 F.3d at 1329 (“The trunks of automobiles store items; they also conceal them.”). Real tags in the trunk would clearly be probative evidence suggesting the car was stolen, for the reason just noted: car thieves replace the real tags with stolen ones to help avoid detection.

The real tags were not the only piece of evidence police could have been looking for in the trunk. Given the ample grounds to suspect the car was stolen, the officers certainly had a reasonable basis for supposing that the trunk would contain other items that might have confirmed their suspicion — such as identification or belongings of the real owner — or items that helped connect Jackson to stealing it. Such items could have included “equipment that’s used to steal a car,” such as “a crowbar, or whatever tools one uses to punch an ignition or ... open a locked door.” See Oral Arg. Tr. at 2:20-:27, 3:00-:16 (Jackson’s counsel acknowledging that, had the car been reported stolen, officers might have had reason to search the trunk for such items).

My colleagues offer several reasons why they believe the officers lacked probable cause to search the trunk for evidence that the car was stolen:

1. The concurrence contends that because the car was unregistered, there could be no real tags or other documents for the officers to discover in the trunk. See Cone. Op. at 99. Not so. A vehicle’s *103license plates do not simply disappear once its registration lapses; many cars roam the roads bearing plates from expired registrations. People get tickets for that all the time, but they are usually able to show that they are the owner listed on the expired registration. Jackson was not able to do that, heightening, the suspicion that he had no legitimate connection to the car.

Moreover, contrary to the majority’s suggestion, finding the expired “real tags” would have provided police with additional evidence of criminal activity. As explained, switching tags is a common ploy of car thieves. And under the majority’s own analysis, see Op. at 94, finding the real tags would have ruled out the possibility the stolen tags were being used only to drive a vehicle that otherwise had no tags, making it more likely that the vehicle had been stolen.

2. The majority doubts the rationale for replacing a stolen vehicle’s real tags with stolen tags and therefore discounts the inference that the car might have been stolen. Op. at 94. But lawyers learn early on that “a page of history is worth a volume of logic.” New York Trust Co. v. Eisner, 256 U.S. 345, 349, 41 S.Ct. 506, 65 L.Ed. 963 (1921) (Holmes, J.). Officer Garboe’s history with stolen tags had confirmed that they, more often than not, led to real tags in the trunk. The reported cases confirm that criminals often use stolen tags on stolen cars. This history is enough to support the officers’ inferring from the stolen tags and the lack of any registration (current or expired) linking Jackson to the car that the car might well have been stolen.

Moreover, replacing a stolen vehicle’s original tags makes sense: it prevents the vehicle from being readily identified as stolen by a passing police cruiser. Of course, the tags may have been reported stolen, and the police can check that, too. But a busy police officer is not going to run a check on the tags of every Mercury Marquis he passes, and people are likely to be much less diligent-about reporting sto-' len tags — particularly temporary ones— than stolen cars.

3. The majority reasons that the officers lacked probable cause because they “were confronted with three possible explanations for the presence of the stolen tags on the car, two of which suggested authorized use and were consistent with the lack of registration and the absence of a report that the car was stolen, and only one of which supported an inference of unauthorized use.” Op. at 94.

Even assuming for the moment the validity of this approach, one of the majority’s “possible explanations” — that stolen tags might be used to conceal an expired registration — does not strike me as probable at all. Using stolen license plates is a serious offense. See D.C.Code Ann. § 22-3232(c) (2001) (receipt of stolen property punishable by up to 180 days in prison if value of property is less than $250, up to seven years if greater than $250). It is unlikely that someone would run so great a risk merely to avoid getting stopped for an expired registration — a steep ticket, to be sure, but not likely to lead to hard time.

The more serious problem is that probable cause does not depend on eliminating other innocent (or, here, less incriminating) explanations for a suspicious set of facts. See United States v. Gagnon, 373 F.3d 230, 236 (2d Cir.2004) (“the fact that an innocent explanation may be consistent with the facts as alleged does not negate probable cause”) (citation omitted); United States v. Funches, 327 F.3d 582, 587 (7th Cir.2003) (“the mere existence of innocent explanations does not necessarily negate probable cause”); see also Gates, 462 U.S. at 243 n. 13, 103 S.Ct. 2317 (“probable cause requires only a probability or sub*104stantial chance of criminal activity, not an actual showing of such activity ... [and] therefore, innocent behavior frequently will provide the basis for a. showing of probable cause”). Of course, considering alternative explanations is “often helpful,” Funches, 327 F.3d at 587, but the officers were not required, before searching the trunk, to negate the possibility that the stolen tags were used only to drive an unregistered car. This is particularly so here, where any plausible explanation for the circumstances of Jackson’s stop — the broken tag light, the stolen tags, Jackson’s lack of registration, and the failure of the records check to connect Jackson with the vehicle — suggested that Jackson was deliberately trying to conceal unlawful activity involving the car itself.

4. Like the majority, I see no reason why the fact that a vehicle has not been reported stolen should preclude probable cause. See Op. at 98. It may take time for a vehicle’s owner to learn that his car has been stolen. A car being driven at 1:00 a.m. by someone without a license and with no registration, bearing stolen tags, may not have been reported stolen because its owner had retired for the night and would be blissfully unaware of his loss until he awoke the next morning. See United States v. Brigham, 382 F.3d 500, 509 (5th Cir.2004) (en banc) (officer’s suspicion that vehicle was stolen was reasonable “because in his experience, the fact that a vehicle has not yet been reported stolen does not necessarily mean that the vehicle has not actually been stolen”); United States v. Maher, 919 F.2d 1482, 1486 (10th Cir.1990) (probable cause to arrest suspect for stealing trailer where “the trailer bore a stolen license plate; the trailer was unregistered; [suspect] was carrying no ownership documents for the trailer; and [suspect] was unable to provide a complete name or address of the person who allegedly sold him the trailer”); see also United States v. Robinson, 471 F.2d 1082, 1104 n. 38 (D.C.Cir.1972) (en banc) (“some courts have held that when a car has no license plates, or fictitious plates, and the driver cannot produce proof of ownership, probable cause exists to believe that the car may have been stolen, and the officer may ... search [the] car ... for evidence of ownership and •identity”).

The majority nevertheless purports to distinguish Jackson’s case on the ground that “the officers did not ask the driver about ownership prior to searching the trunk and therefore could not evaluate the plausibility of the driver’s explanation.” Op. at 97 - 98. This accords with the majority’s suggestion that the officers might have gathered facts amounting to probable cause if only they had “ask[ed] the driver a few questions” and not “stopped their investigation too soon.” Op. at 95, 96. This is a hazardous approach to assessing probable cause.

The officers who stopped Jackson had no ready means of verifying ownership of the vehicle at the scene. Jackson himself had no license and no proof of registration, and the car had stolen tags. The officers’ records check not only failed to resolve the question of ownership but raised more suspicion: the car itself was not registered and Jackson’s name was not on the old listing. The majority suggests that the officers should have called “the purported owner and ha[d] the owner come to the scene with proof.” Op. at 96. But this assumes that the officers had nothing better to do while on night patrol than linger roadside, tracking down exculpatory leads .for suspects.2

*105The officers could have reasonably concluded that further questioning would have yielded nothing more than the usual story any suspect in Jackson’s situation would be expected to deliver. See Williams ex rel. Allen v. Cambridge Bd. of Educ., 370 F.3d 630, 637 (6th Cir.2004) (“law enforcement is under no obligation to give any credence to a suspect’s story ... nor should a plausible explanation in any sense require the officer to forego arrest pending further investigation if the facts as initially discovered provide probable cause” (internal quotation marks omitted)). The best Jackson could do was tell the officers — as he did, at some point — that the car belonged to his girlfriend.3 Sometimes a car being driven by an unlicensed driver, with no registration and stolen tags, really does belong to the driver’s friend, and sometimes dogs do eat homework, but in neither case is it reasonable to insist on checking out the story before taking other appropriate action. Even if Jackson had provided contact information for his girlfriend in response to inquiries from the officers, and even if the officers had been able to reach the girlfriend and she were responsive to their questions, I cannot see any conceivable value in the over-the-phone testimony of a suspect’s apparent girlfriend — someone unknown to the officers, whose number was given to them by the suspect himself — that an unregistered car with stolen tags, driven by an unlicensed driver, was indeed hers and was being used with her permission.

Finally, my colleagues’ insistence that police should have further questioned Jackson amounts to prescribing preferred investigative procedures for law enforcement. We have neither the authority nor the expertise for such an enterprise. See United States v. Montoya de Hernandez, 473 U.S. 531, 542, 105 S.Ct. 3304, 87 L.Ed.2d 381 (1985) (“creative judges engaged in post hoc evaluations of police conduct can almost always imagine some alternative means by which the objectives of the police might have been accomplished” (internal quotation marks omitted)). In the end, I would leave the judgment as to what lines of inquiry ought to be pursued to the officer himself, and judge probable cause on the facts as they are, rather than on what they might have been had the officer pursued a different course.

* * * * :!< tfc

I wholeheartedly subscribe to the sentiments expressed in the concurring opinion about the Fourth Amendment’s place among our most prized freedoms. See Cone. Op. at 98, 101. But sentiments do not decide cases; facts and the law do. There is no dispute here on the law: if the officers had probable cause, they did not need a warrant; if they did not have probable cause, no warrant would issue in any event. As for the facts, the officers en*106countered at 1:00 a.m. an unlicensed driver operating an unregistered car with a broken tag light and stolen tags. The experienced district court judge concluded — and I agree — that “the circumstances were suspicious enough to amount to probable cause to search the trunk.” Memorandum Order, at 5. Right or wrong, nothing about that determination reflected insensitivity to constitutional values, any more than a contrary determination would have reflected insensitivity to the needs of law enforcement.

I respectfully dissent.

Notes

. The majority dismisses the officer's testimony out of hand because it "is devoid of the critical circumstances of those searches, including whether the identifying information revealed that the vehicle was stolen.” Op. at 95. I fail to see how this undermines the relevance of the officer’s experience. Whatever the circumstances, the officer noted a strong correlation between stolen tags on a vehicle and the presence of the vehicle’s real tags in the trunk. If Jackson’s attorney had thought that the surrounding circumstances of those stops might undermine the probative value of the officer’s experience, the attorney had ample opportunity to question the officer about them on cross-examination.

. Officer Garboe testified that, if police find proof of ownership they can verify, they ,usu*105ally call the "registered owner” of the vehicle to ask if he or she would "like to come and get it or ... to have it towed.” Hr’g Tr. at 62. This is very different from the majority's suggestion that police could verify ownership by waiting around for a suspect’s girlfriend to meet them on a District street at one o’clock in the morning. See Op. at 96.

. Because Jackson did not argue below that the police should have conducted a more elaborate investigation, the district court did not make any factual finding as to precisely when the suspect told officers that the car belonged to his girlfriend. Compare Hr'g Tr., June 9, 2003, at 62 (“I don't recall exactly when the conversation took place in which he ... informed us that his girlfriend had purchased the vehicle at an auction.”) (redirect of Officer Garboe) with id. at 70 (Q. "Did you have any information regarding this defendant or anyone else's possible ownership of this particular vehicle prior to the search?” A. "No.”) (Officer Johnson).