Issue/Holding: Where the officer handcuffed the defendant and told her she was under arrest for an ordinance violation, but also told her that she would “be released if she continued to be cooperative,” there was no arrest in fact and therefore the fruits of an ensuing search incident to (a non-existent) arrest were suppressible:
¶27 In sum, neither party has cited to a case sufficiently analogous to the facts of this case to guide its outcome.  Contrary to the parties’ assertions, no case establishes a bright-line rule as to when an arrest has been effected. Instead, each case focuses on the totality of the circumstances in the record to determine whether a reasonable person in the defendant’s position would have believed he or she was under arrest. Here, the record reveals conflicting circumstances: Ben-Ami told Marten-Hoye she was under arrest, but also that she would be issued a citation for a municipal ordinance violation and would be free to go. She placed Marten-Hoye in handcuffs but did not place her in a squad car, instead conducting the entire interaction in public. While Ban-Ami searched Marten-Hoye, another officer was writing out the citation that would have ended in Marten-Hoye’s release.
¶28 Considering all of the circumstances of the situation, we conclude that a reasonable person in Marten-Hoye’s position would not have believed he or she was “in custody” given the circumstances present here. First, we do not agree with the State that the fact that Ben-Ami told Marten-Hoye that she was under arrest necessarily establishes an arrest. Ben-Ami did not unequivocally tell Marten-Hoye that she was under arrest. Significantly, immediately after Ben-Ami told Marten-Hoye that she was under arrest, she also told her that she would be issued a citation and then would be free to go.  Although the statements by Ben-Ami are contradictory, we conclude that the assurance by Ben-Ami that Marten-Hoye would be issued a citation and released would lead a reasonable person to believe he or she was not in custody, notwithstanding the early statement that he or she was “under arrest.”
¶29 Next, we do not agree that police use of handcuffs transformed the interaction here into an arrest. In this case, Marten-Hoye’s being placed in handcuffs is associated with the fact that she was being loud and uttering profanities rather than indicating that she was being placed in police custody. Additionally, the entire interaction between Ben-Ami and Marten-Hoye was in public and Marten-Hoye was never transported to any other location. Although Ben-Ami’s statements conflict, we are persuaded that in their totality they would not lead a reasonable person to believe he or she was “in custody.” Accordingly, the search of Marten-Hoye is not justified as a search incident to an arrest.
A bit of chicken-roosting: over the years a sort of elastic test has evolved for determining when an arrest has occurred, typically when a Terry stop has evolved into a full-blown arrest. An all-time Case Summaries favorite exemplar is U.S. v. Vega, 72 F.3d 507, 515 (7th Cir. 1995), which in the process of observing that “the line between a lawful Terry stop and an unlawful arrest is not bright,” holds that drawing guns on a suspect and “asking” him to get into a police car for transport to another site didn’t amount to an arrest. Nor, similarly, is handcuffing someone necessarily enough to establish an arrest, U.S. v. Stewart, 388 F3d 1079 (7th Cir. 2004) (“The permissible scope of a Terry stop has expanded in recent years to include the use of handcuffs and temporary detentions in squad cars.”); Jewett v. Anders, 7th Cir No. 06-2982, 4/11/08 (using force to bring suspect to ground, handcuffing him and detaining him in squad car 20 minutes or longer didn’t transform stop into arrest). And, if the detention, intrusive though it might be, doesn’t amount to an arrest then a search of the person more intensive than a pat-down can’t be justified as incident-to, which is the essence of the current case — those are the chickens coming home to roost.
What takes the case somewhat out of the norm is the idea that no arrest occurred even though Marten-Hoye was expressly told she indeed was under arrest. However, she was simultaneously told that her status as an arrestee was entirely conditional; if she continued to be cooperative she would not in fact be arrested. (A slight stretch of an analogy: it’s almost as if the cop imposed a sort of remedial contempt, whereby Marten-Hoye herself held the keys to her potential incarceration; cooperative behavior and she walks.) The court’s holding, in short, represents a narrow, fact-specific result. The State has an elegant solution to this little dilemma, ¶15 n. 9: “The State asserts that to answer whether a defendant was arrested, we must look to who is asserting that an arrest occurred, the State or the defendant, citing 3 Wayne R. Lafave, Search & Seizure § 5.1(a) (4th ed. 2004).” This approach strikes Case Summaries as what we like to term, “result-oriented.” Nor does LaFave seem to support the State’s unqualified view that the court must (which is to say, always) take into account who is arguing what. In any event, the court goes on to suggest that Knowles v. Iowa, 525 U.S. 113 (1998) (search incident to mere non-criminal traffic citation impermissible) takes this case out of that context. Does it matter that Marten-Hoye’s “arrest” was not only conditional but was for a local ordinance violation? You can perhaps make that argument, stressing that concerns for officer safety are much different in that context than arrest for a crime. It may also be that, as a practical matter, you’re much more likely to get a catch-and-release in the context of citation-offenses than where a criminal offense is involved. But: it is settled nonetheless that Wisconsin law does allow arrest (and search-incident) on a noncriminal offense such as an ordinance violation, State v. Robert J. Pallone, 2000 WI 77, ¶43 (“arrests for civil forfeitures are not per se unconstitutional. … Consequently, the Fourth Amendment does not preclude searches incident to arrests for noncriminal violations.”). A policy-based argument, then, that arrest is unsupported for an ordinance violation would seem to be an exercise in futility. It may be that Knowles has changed this landscape, but recall that in that case there was no attempt to arrest on the traffic violation, instead an unsuccessful attempt to create a search incident to mere citation rationale. More: Pallone (¶46) seems to have explicitly rejected such a possibility (“Because this was a search incident to an arrest [for an ordinance violation], not a search incident to the issuance of a traffic citation with no arrest, the Knowles rule does not apply to this case.”)
The particular facts, that is, very much matter. If Marten-Hoye had been told unconditionallyshe was under arrest, the court presumably would have perceived no impediment to an ensuing search. But that is not what happened. Just because the cop could have effectuated an arrest doesn’t mean that she in fact did. Though the court of appeals’ rationale leaves a bit to be desired, the result as indicated above is best seen narrowly: once the cop in effect told Marten-Hoye that she would not remain under arrest but instead would be released with issuance of a citation, the situation became controlled by Knowles. If the cop is going to issue a citation for an ordinance violation without an arrest, a search-incident isn’t permissible.
This isn’t to say questions don’t remain. The Certification perceived the following doctrinal tension:
As we previously explained, the facts in this case meet at the intersection of Knowles and Swanson, and highlight a possible overlapping of the bright-line rules established in these cases. Were the police prohibited from performing a full field search of Marten-Hoye based on the search incident to arrest exception to the warrant requirement because they told her that she was going to be given a citation and released (Knowles)? Were the police allowed to perform a full search incident to arrest because they handcuffed Marten-Hoye and told her she was under arrest, circumstances under which a reasonable person may have believed they were under arrest (Swanson)?
Perhaps. If so, the tension is relieved simply by saying that a reasonable person wouldn’t see him or herself as being under arrest if told a release is imminent. In other words, there’s no arrest under the Swanson test. But this isn’t to say that significant underlying tension is indeed absent. To see the source of this tension, though, take a look at U.S. v. Powell, DC Cir No. 05-3047, 4/17/07, which concludes, after elaborate discussion, that a search-incident may precede an arrest, so long as the arrest is effectuated straight-away:
Powell and our dissenting colleague also contend our decision is inconsistent with Knowles v. Iowa, 525 U.S. 113 (1998) (holding an officer may not conduct a search incident to arrest when, although the officer has probable cause to make an arrest, he issues a citation instead of arresting the suspect). But that is not correct either. Had the officers failed to arrest Powell and merely issued him a citation, then indeed the search would be invalid under Knowles. 525 U.S. at 117 (“The threat to officer safety from issuing a traffic citation … is a good deal less than in the case of a custodial arrest”). That, of course, is not what happened, and we do not say that having probable cause to arrest is by itself sufficient to bring a search within the Belton exception to the warrant requirement. Rather, it is the “fact of the arrest” that makes all the difference. Id. (quoting United States v. Robinson, 414 U.S. 218, 234 n.5 (1973) (“The danger to the police officer flows from the fact of the arrest, and its attendant proximity, stress, and uncertainty”)); see also Washington v. Chrisman, 455 U.S. 1, 7 (1982) (“Every arrest must be presumed to present a risk of danger to the arresting officer”). As we have recently noted: “The key point in Knowles … was not that the officer had a lawful ground for arrest upon which he did not rely, but that he did not arrest the defendant at all.” United States v. Bookhardt, 277 F.3d 558, 566 (D.C. Cir. 2002).
Note that Wisconsin caselaw seems to be to aligned with Powell, State v. Michael D. Sykes, 2005 WI 48; that is, search-incident need only be “contemporaneous” with, as opposed to preceding, arrest. Powell’s analysis is entirely unconvincing, because the plain fact is that the cop did arrest Knowles, albeit after conducting a search (taking into account the court’s big production of saying that it simply doesn’t matter that the search precedes the arrest). Some other point of distinction, then, must be found. Powell (Sykes, too, for that matter) had been detained though not under arrest prior to the search. But Knowles of course had also been detained, so the mere fact of detention can’t be meaningful either. Perhaps the point of departure is that without some explicit statement from the cop, Knowles (the subject of a simple traffic stop) could clearly assume that he would be released; neither Powell nor Sykes could say the same. Traffic stops thus might be seen as a subset of Terry-type detentions, in that without more there is a presumption of release (though to be sure Knowles doesn’t explicitly say as much). Perhaps that distinction also informs the Powell court’s otherwise inexplicable idea that Knowles wasn’t arrested “at all.” And if that is so, then M-H can say the same, albeit in a non-traffic setting: she wasn’t arrested “at all,” because she had been bluntly told that continued cooperation would lead to her immediate release. She no less than Knowles could assume imminent release from custody without arrest. But the only thing that might be said with confidence is that we probably haven’t seen the end of this discussion.
Contrast Virginia v. Moore, USSC No. 06-1802, 4/23/08 (arrest for minor offense based on probable cause but prohibited under state law nonetheless supports search-incident; in other words, mere violation of state arrest law doesn’t ipso facto violate 4th A).