Concept
Timing of the Search
Search incident to lawful arrest
Timing of the Search
Chimel fixed the two places that an arrest opens to search, the arrestee's person and the area within the arrestee's immediate control, and that settles where officers may look. It leaves open when they may look, and the answer comes from the same two dangers that justify the exception. A weapon can be seized, and evidence destroyed, only while the arrestee is present and able to act, so the authority to search is tied to the moment of the arrest and to the place where it happens. The Court has expressed that tie by requiring that a warrantless search incident to arrest be “substantially contemporaneous with the arrest,” which gives the exception a limit in time as well as in space.1 This section begins with the case that announced the limit, then takes up two variations: a search that comes just before the arrest, and a search delayed until after the arrestee is in custody.
The Preston Rule
The limit comes from Preston v. United States, which holds that a valid search incident to arrest must be substantially contemporaneous in time and place with the arrest, and that a delayed or remote search following impoundment falls outside the exception and needs some other justification.2 The requirement has two parts that can fail independently. The time part asks whether the search came while the arrest was still unfolding, and the place part asks whether it happened where the arrest happened. The search in Preston failed on both.
Four Newport, Kentucky officers answered a 3 a.m. complaint about men sitting in a parked car, found three men whose answers they considered evasive, arrested them for vagrancy, searched them for weapons, and took them to police headquarters.3 The car was not searched at the scene. An officer drove it to the station, from which it was towed to a garage, and only after the men had been booked did officers go to the garage, where they found two loaded revolvers in the glove compartment and later entered the trunk through the back seat.4 The Court did not decide whether the arrest was valid, because it held that the search was not.5
The Court called the search too remote in time or place to be incident to the arrest.6 Its reasoning began with the purposes of the rule: contemporaneous searches are allowed to seize weapons that might be used to assault an officer or to escape, and to prevent the destruction of evidence, and those dangers exist when the weapon or evidence is on the accused or under the accused's immediate control.7 At the garage neither danger remained. The men were at the station, none of them could use a weapon in the car or destroy anything in it, and the car sat in police custody with no chance of being driven out of the jurisdiction.8
Later decisions treat Preston as the source of this timing limit. Chimel described it as a case in which three men arrested in a parked car had the car towed to a garage and searched, and the search was held unlawful despite the contention that it was incident to a valid arrest; Chimel presented Preston as an application of the principle that the scope of a search must be tied to the circumstances that made it permissible at the outset.9 Gant restated the point in general terms: if there is no possibility that an arrestee could reach the area officers want to search, both justifications for the exception are absent and the rule does not apply, and it cited Preston for that proposition.10
Each half of the rule follows from the dangers that justify the exception. Time matters because a weapon can be grabbed or evidence destroyed only while the arrestee is free to act, and once the arrest is complete and the arrestee is in custody, that freedom is gone. Place matters because the dangers attach to what lies within the arrestee's reach at the scene, and a garage across town holds nothing the arrestee can seize or destroy. A practical reason sits underneath both. When the dangers have passed, officers have the same opportunity as anyone else to ask a magistrate for a warrant, so excusing the warrant requirement buys nothing. Without the limit, an arrest would operate as a standing license to search whatever property was connected with it, whenever officers chose. The requirement does not forbid the later search; it requires that the later search rest on some justification other than the arrest.
New York v. Belton supplies the matching case on the other side of the line. A trooper who had stopped a speeding car ordered the four occupants out, arrested them for possessing marijuana, patted each one down, and separated them along the roadway, and then searched each man and the passenger compartment, where a jacket on the back seat held cocaine.11 Only afterward did the trooper drive the arrestees to a nearby station.12 Belton treated that roadside search as a contemporaneous incident of the arrest.13 Both cases involve a search of a car after its occupants were arrested, so the fact that moved the answer on timing is when and where the search occurred: at the roadside during the arrest in Belton, and at a garage after booking in Preston. Whether the back seat was within the arrestees' reach is a separate question, taken up in the vehicle sections below.
Students lose this issue when they reason that a lawful arrest and a connection to the crime are enough to carry the exception, because the same arrest can support a search at the roadside and fail to support one at the garage. The question to ask of any search is how long after the arrest, and how far from it, the search happened.
Searches Before the Arrest
The ordinary sequence is arrest first and search second, but the timing rule does not insist on that order in every case. A search of the person that comes just before the formal arrest qualifies as incident to the arrest if the arrest followed quickly on the heels of the search and rested on facts other than those the search uncovered.15 In Rawlings v. Kentucky the petitioner admitted ownership of a large quantity of drugs found in a companion's purse, which gave the police probable cause to arrest him; they searched his person, finding money and a knife, and then arrested him, and the Court treated the search as incident to the arrest because the arrest came so soon after it.16 The order mattered little there because a search before arrest intrudes no more on the person than a contemporaneous search would, so long as the grounds to arrest already existed when the search began.17
The requirement of grounds independent of the search does the real work. A search cannot create its own justification: if the only basis for the arrest is what the search turned up, then calling the search incident to that arrest would let officers justify any search after the fact. Where officers have no present intent to arrest and no prompt arrest follows, the search is not properly called incident to arrest at all, though it may still be lawful on probable cause when limited to what is necessary to preserve highly evanescent evidence, as with the fingernail scrapings in Cupp v. Murphy.18 Rawlings is the exception rather than the rule, because outside such narrow circumstances the arrest must precede the search.19
Delayed Searches After Custody
Delay after the arrest affects the arrestee's person differently from the arrestee's property. Courts have generally upheld delayed searches of the person arrested, such as those made on the way to the station or at the station, on one of two theories: police control of the person by arrest is so substantial that it carries a continuing right of search, or the police are entitled to inventory what is found on a person before placing the person in a cell.21 In United States v. Edwards the police waited ten hours before searching the defendant's clothes for paint chips because it took that long to find substitute clothing, and the search was upheld because there was a legitimate reason for the delay.22 The body and clothing stay in custody with the arrestee, so the same custody that gave the authority to search continues, and that authority does not lapse the way it does for separate property.
Containers and other separate property follow the stricter rule. A bag or briefcase found next to the arrestee must generally be searched during the arrest, and a search made after a wait, such as an hour, is not a valid search incident to arrest.23 United States v. Chadwick shows why. Agents arrested the defendants beside an open trunk, took the defendants, the car, and a double-locked footlocker to the federal building, and opened the footlocker there without a warrant, and the Court held that once officers have reduced property not immediately associated with the person to their exclusive control and the arrestee can no longer reach it, a search of that property is no longer an incident of the arrest.24 With the footlocker in police hands and the arrestees in custody, neither the weapon danger nor the evidence danger remained, which is the same reasoning that defeated the garage search in Preston.
When the search-incident exception fails on timing, the analysis does not end, because the failure removes only one justification. Preston itself shows this. The Court assumed, as the government urged, that the police could have searched the car when they first came on the scene, either because the arrests were valid or because they had probable cause to think the car stolen, and it faulted only the later search at the garage.25 A delayed search of a car or container can still stand on a warrant, on probable cause combined with exigent circumstances, or on a standardized inventory procedure, which exists to ensure that police are not storing dangerous or illegal items.26 Those doctrines carry their own requirements and are taken up with the other warrantless-search exceptions.