179 F. App'x 825

BECKER, Circuit Judge,
dissenting.
I would hold that the search of Morris’ vehicle was not a valid inventory search because police officers departed from their own written procedures. Furthermore, because the government did not raise the search incident to arrest and inevitable discovery arguments in the District Court, I would remand the case to allow the District Court to consider these arguments in the first instance and to develop the record further.
I. Inventory Search
The majority concludes that Officer Le-hew performed a valid inventory search under South Dakota v. Opperman, 428 U.S. 364, 96 S.Ct. 3092, 49 L.Ed.2d 1000 (1976), and its progeny because the search complied with standard police procedures. This conclusion, however, is based on a misreading of § 8.14.13(c) of the Wilkins Township Motor Vehicles Inventory Policy, quoted in the majority opinion.
The pivotal question is whether the phrase “if possible” in the first sentence of this provision modifies “remove” or “shall be asked.” If, as the majority concludes, “if possible” modifies “shall be asked,” then Lehew did not have to ask Morris to remove all valuables from his car. On the other hand, if “if possible” modifies “remove,” then the phrase “shall be asked” is unqualified. Under this second reading, Wilkins Township police must always ask a suspect before removing objects from the car. It would follow that Lehew was not permitted to remove objects from Morris’ car without asking him.
Under the last antecedent rule, “a limiting clause or phrase ... should ordinarily be read as modifying only the noun or phrase that it immediately follows.” Barnhart v. Thomas, 540 U.S. 20, 26, 124 S.Ct. 376, 157 L.Ed.2d 333 (2003). “If possible” follows on the heels of “remove,” and therefore modifies “remove” under the last antecedent rule of construction. Therefore, Lehew violated § 8.14.13(c).
This would be the better reading even if the first sentence of § 8.14.13(c) were read in isolation. But in fact the second sentence further supports the conclusion that “if possible” modifies “remove.” That sentence explains what an officer should do in the event that it is not possible to remove an object. Such an explanation only makes sense if the first sentence is read as distinguishing between situations in which objects can and cannot be removed.
The police also violated § 8.14.13’s requirement that the owner of the vehicle “verify the completeness of the inventory by signature.” Officer Lehew admitted as much:
Q. In this case, neither you, nor Brokaw, followed that policy, to have Morris verify by signature or refusal of signature, whichever it may have been? You didn’t have that opportunity in violation of your policy?
A. He was not on scene to be able to sign it; that’s correct.
Q. Is there anything in your policy that provides to waive the signature or to deviate from your policy because a defendant is taken from the scene?
A. No. There is nothing in our policy that states that.
Clearly, then, Lehew did not follow the Inventory Policy in that he did not ask Morris to remove his valuables and did not ask Morris to verify the inventory by signature.
Where, as here, a formal, written policy does exist, and where officers violate that policy, the inventory search is not restricted in the manner required by Colorado v. Bertine, 479 U.S. 367, 107 S.Ct. 738, 93 L.Ed.2d 739 (1987). In United States v. Bridges, 245 F.Supp.2d 1034, 1037 (S.D.Iowa 2003), the District Court held an inventory search invalid for failing to comply with written procedures, even where an officer testified that it was “standard operating procedure ... to impound a vehicle for many unwritten reasons not contained in the policy.” See also United States v. Maple, 348 F.3d 260, 263-65 (D.C.Cir.2003) (invaliding an automobile search where the police failed to follow their own procedures). I therefore con-elude that the search cannot be upheld as a valid inventory search.
II. Inevitable Discovery and Search Incident to Arrest
In the District Court, the government did not raise search incident to arrest or inevitable discovery arguments, and the District Court did not rule on these issues. However, the government now argues, and the majority concludes, that Morris’ cocaine would have inevitably been discovered if the police had followed the Inventory Policy.
Under the inevitable discovery doctrine, “[i]t is the government’s burden to show that the evidence at issue would have been acquired through lawful means.” United States v. Vasquez De Reyes, 149 F.3d 192, 195 (3d Cir.1998). Here, “lawful means” would require compliance with § 8.14.13(c), which mandates that police ask the suspect to remove valuables from the vehicle. It appears that Lehew could have complied with this requirement in two ways. First, he could have asked Morris to inventory the car before Morris was taken to the hospital. Second, Lehew could have decided not to conduct an inventory search. Section 8.14.6 of the Inventory Policy suggests that police are never required to conduct an inventory search: “Legal Authority to Inventory: An authorized member of this agency may conduct a motor vehicle inventory without a warrant or probable cause ...” (emphasis added.)
The majority simply assumes that faced with this choice, Lehew would decide to conduct the inventory search with Morris present, even though this would delay taking Morris to the hospital for a blood-alcohol test. Given the officers’ desire to take Morris to the hospital as soon as possible, I cannot view this decision as inevitable. In fact, Officer Brokaw testified that he was anxious to get Morris to the hospital before his blood-alcohol level began to drop.
While the Inventory Policy suggests that he could have chosen not to perform the inventory, Lehew’s testimony complicates matters: “We have a policy saying when we have someone in custody, there is no other driver with the vehicle, we have to do an inventory report.” (emphasis added.) If Lehew had no choice but to conduct an inventory, then the only way to comply with the Inventory Policy would have been to inventory the car with Morris present, a process that would have led to the discovery of the drugs.
Because it is unclear from the record whether Lehew had discretion not to conduct an inventory search, and because application of the inevitable discovery doctrine turns on this determination, I would remand the case to the District Court for a determination whether the evidence would inevitably have been discovered.
In light of my disagreement with the majority, I must reach the government’s final argument, that the police officers performed a valid search incident to arrest. I have serious doubts that the search incident to arrest exception to the warrant requirement, emanating from New York v. Belton, 453 U.S. 454, 460, 101 S.Ct. 2860, 69 L.Ed.2d 768 (1981), should apply to this case, where Lehew searched Morris’ car while Morris was en route to the hospital with Brokaw.
In Belton, which involved the search of an automobile after the suspects had exited the vehicle, the Supreme Court announced the following rule: “[W]hen a policeman has made a lawful custodial arrest of the occupant of an automobile, he may, as a contemporaneous incident of that arrest, search the passenger compartment of that automobile.” 453 U.S. at 460, 101 S.Ct. 2860. The Courts of Appeals have extended Belton to great lengths, in some cases upholding searches that occur while the suspect is handcuffed, in a police cruiser at the scene of the crime, or both. However, courts are divided on the precise issue in this case, whether a search incident to arrest is valid when, at the time of the search, police have transported the suspect away from the scene of the crime.
In United States v. Lugo, 978 F.2d 631, 635 (10th Cir.1992), the Tenth Circuit held that a search is not contemporaneous with an arrest where the suspect has been transported from the scene of the crime. The court explained:
In this case, when the search of Lugo’s truck began, Lugo was no longer at the scene. He was handcuffed and sitting in the back seat of a patrol car proceeding toward Green River. Once Lugo had been taken from the scene, there was obviously no threat that he might reach in his vehicle and grab a weapon or destroy evidence. Thus, the rationale for a search incident to arrest had evaporated. [The police officer’s] inventory was not contemporaneous because it was remote in time and in place as regards Lugo and his truck being in entirely different locations, and no exigency existed.
Id. at 635.
In United States v. McLaughlin, 170 F.3d 889, 891-92 (9th Cir.1999), the Ninth Circuit reached the opposite conclusion, stressing that the Belton rule is categorical:
Because it is a bright-line rule that may be invoked regardless of whether the arresting officer has an actual concern for safety or evidence, we have held that the applicability of the Belton rule does not depend upon a defendant’s ability to grab items in a car but rather upon whether the search is roughly contemporaneous with the arrest.
The First Circuit decided a similar issue in Doward, where it upheld a search that began while the suspect was on the premises but continued after he was transported away. 41 F.3d at 791. The Court reasoned that Belton established a bright-line rule designed to prevent courts from second guessing police officers’ decisions. Id. at 792-93. The court noted that although there was no real risk in Belton that suspects already outside the car would be able to reach back into the passenger compartment, Belton still upheld the search. Id. at 793.
Subsequent to Doward, Lugo, and McLaughlin, the Supreme Court decided Thornton v. United States, 541 U.S. 615, 124 S.Ct. 2127, 158 L.Ed.2d 905 (2004), which extends Belton, holding that police may conduct a search incident to arrest of the passenger compartment, even when police make contact only after the suspect leaves the vehicle. In Thornton, the suspect was handcuffed and placed in the back of a police car, but, in contrast to this case, the police car had not left the scene of the arrest. Id. at 618, 124 S.Ct. 2127. At all events, the question whether handcuffing the arrestee and placing him in the patrol car affected the applicability of the search incident to arrest doctrine was not within the Court’s grant of certiorari. As the Court emphasized, “[t]he question presented — ‘[w]hether the bright-line rule announced in New York v. Belton is confined to situations in which the police initiate contact with the occupant of a vehicle while that person is in the vehicle,’ — does not fairly encompass” such questions. Id. at 624 n. 4, 124 S.Ct. 2127 (citation omitted).
I am inclined to adopt the Tenth Circuit’s approach and conclude that Lehew did not perform a valid search incident to arrest. Indeed, I have serious doubts that the search incident to arrest doctrine discussed in Belton and Thornton can be stretched so far as to encompass the present situation, in which the suspect has been carried away in handcuffs from the scene of the crime. However, since the issue was not dealt with in the District Court, I would remand the case to allow the District Court to address it in the first instance.
For these reasons, I respectfully dissent.
. The concurrence contends that we should first "determine whether the policy at issue failed to specify the procedure to be employed for the specific factual scenario presented in the case.” Concurring Op. at 4. But the Inventory Policy does not cite any example of specific factual circumstances in which the removal of valuables and signature requirements apply. Thus, if the requirements only apply where the policy "specif[ies] the procedure to be employed for the specific factual scenario presented,” then the requirements would, quite literally, never apply. Such a reading would nullify the Inventory Policy.
Rather than listing every conceivable factual circumstance, the Inventory Policy states a categorical rule. The owner “shall be asked to remove” valuables and he "shall be requested to verify the completeness of the inventory by signature.” This mandatory language contains no exceptions and therefore applies in all circumstances.
To the extent that the concurrence suggests that we should consider the good faith of the officers even if they violated the Inventory Policy, I cannot agree. Where police violate their own procedures in circumstances such as these, courts invalidate the search without any inquiry into subjective motivations. As the D.C. Circuit stated, "the salient point is that whatever his good intentions, the officer failed to follow established police policy and procedures.” Maple, 348 F.3d at 265 (emphasis added); Bridges, 245 F.Supp.2d at 1037 (invalidating an inventory search for failure to follow procedures even though the "officer ... may have been motivated solely by the reasons offered and not by the hope of finding evidence of illegal activity”).
Permitting the police to violate their own policies would effectively replace judicial examination of written procedures, which Bertine requires, with an analysis of the subjective intent of individual officers. See Bertine, 479 U.S. at 374 n. 6, 107 S.Ct. 738 ("Our decisions have always adhered to the requirement that inventories be conducted according to standardized criteria.”) (citing Illinois v. Lafayette, 462 U.S. 640, 648, 103 S.Ct. 2605, 77 L.Ed.2d 65 (1983); Opperman, 428 U.S. at 374-375, 96 S.Ct. 3092). Applying Supreme Court doctrine, I would avoid questions of subjective intent by objectively focusing on compliance with a written policy.
. See United States v. Doward, 41 F.3d 789, 791 n. 1 (1st Cir.1994) (collecting cases).
. The Tenth Circuit has repeatedly reaffirmed this holding. See United States v. Edwards, 242 F.3d 928, 937-38 (10th Cir.2001) (holding that a search that occurred when the suspect was handcuffed in a police car, after being arrested 100-150 feet away from his vehicle, was not a valid search incident to arrest); United States v. Dennison, 410 F.3d 1203, 1209 (10th Cir.2005); United States v. Walker, 33 Fed.Appx. 921 (10th Cir.2002).

Passage view · GavelSight