In a decision handed down last week, United States v. Thompson, Judge Michael McConnell of the Tenth Circuit had an apt observation about the Supreme Court test for when a person is "seized" for Fourth Amendment purposes:

  According to formal legal doctrine, an encounter between an individual and the police is consensual when "a reasonable person would feel free 'to disregard the police and go about his business.'" Florida v. Bostick, 501 U.S. 429, 434 (1991) (quoting California v. Hodari D., 499 U.S. 621, 628 (1991). In addressing this question, however, we must be guided by the Court's decisions in similar cases rather than our own experience regarding how reasonable people actually respond to police investigations.[Fn1]

  [FN1: It might bring greater clarity to this area of the law if the test were framed in terms of whether the officer's behavior is coercive rather than whether, under the circumstances, the reasonable person would feel "free to disregard the police," which we suspect is unrealistic.]

  I think Judge McConnell is basically right here. The Supreme Court's stated test is artificial: the Justices have a sense of the kind of circumstances that they want to be regulated by the Fourth Amendment, and they haven't done a very good job of articulating that sense through a legal standard. The Court ended up adopting the "reasonable person would feel free to leave" test because it does cover a number of useful cases: If the police order someone to stay, then they have exerted control over the situation and that should be regulated, while if the police haven't made a showing of force, then that is something that the Fourth Amendment should sensibly not regulate. But the "reasonable person" formulation ends up being artificial, as Judge McConnell suggests, because (as you might guess) most people just don't feel free to walk away from a police officer.

   The Supreme Court has tried to address this problem a bit by clarifying that the standard is of a reasonable innocent person, apparently on the theory that a reasonable innocent person has no reason to fear the police. But again, that's unrealistic: most people don't feel free to walk away even if they feel they're totally innocent. I end up advising my students along the lines of what Judge McConnell suggests: Apply the test based on analogy to decided cases applying it rather than by relying on the natural meaning of its words.

  But even then, judges — and Justices — sometimes split on this. An interesting example of the split is Yarborough v. Alvarado, 541 U.S. 652 (2004), an AEDPA case where the Supreme Court divided on whether a state court had reasonably applied the law to whether the suspect was in custody for Miranda purposes (an inquiry that is very similar to when a person is seized for Fourth Amendment purposes). The majority, in an opinion by Justice Kennedy, applied the test by analogy to the facts of decided cases: From that perspective, the state court's application of the law was reasonable. The dissent, in an opinion by Justice Breyer, applied the test by relying on its formal terms: From that perspective, the state court's application of the law was unreasonable. Whether the state court application of the law was reasonable depended on whether you interpreted the doctrine by analogy to case-by-case application or as the formal test, and the Court split 5-4 on that.

  Thanks to FourthAmendment.com for the link.