My co-bloggers’ posts about the Fourth Amendment test for when a person is “seized” brings up some really interesting questions about the nature of Fourth Amendment law and criminal procedure more generally. In particular, I think provides a helpful demonstration of the pragmatic nature of Fourth Amendment doctrine.
As my co-bloggers point out, Fourth Amendment doctrine teaches that a person is “seized” by a police officer when a reasonable person in his situation wouldn’t feel free to leave. But the courts always apply this test in a very weird way: the reasonable person is assumed to be innocent person who has no fear or even anxiety about interacting with the police. As a result, the courts say someone is not “seized” even when we all know that a reasonable person in their situation would not feel free to leave. In the cases, a reasonable person would only feel they can’t leave if they are stopped, or the officer tells them they can’t go, the officer threatens to use force, or something like that. What gives?
I think the explanation for the odd test is consequentialist, and reveals a lot about the pragmatic nature of criminal procedure law. The background legal rule is that the police can only conduct a “seizure” with some kind of cause, such as reasonable suspicion or probable cause: if the police seize something without cause, the fruits of the seizure are suppressed. With that background test in mind, let’s try to come up with a test for when a person is “seized.”
A textualist approach might be to say that a person is seized when he is physically grabbed, such as by arresting him. But is you take that narrow definition, you run into a problem: it means that the police can do anything to stop someone’s movement short of actually grabbing them without triggering any legal oversight. No seizure would occur, so the police could do it whenever they wanted. That gives the police too much power, right? In particular, it doesn’t account for all the times that a person is functionally stopped by a reasonable fear of police action: If officer A points a gun at suspect B and yells “freeze!,” a person might reasonably feel like they are “grabbed” even though no force has been used against them. Does that mean they weren’t actually “seized” at that point? Seems weird to say so.
So what other test could be used other than an “actual grabbing” rule? Well, you could look at things from the standpoint of a reasonable person, and ask whether that reasonable person would feel free to leave. That covers the “freeze” case, but it also presents a problem: by its terms, it probably covers nearly all police-citizen interactions. Most people feel like they have to do what a police officer says; most feel that if a police officer walks up to them and starts asking them questions, they’re not allowed to just walk away. Sure, it would be crystal clear if the officer pulled out his gun and yelled “freeze!”, but for most people the mere fact that the person is a uniformed officer is enough to get them to feel they need to comply.
Why is that a problem? It’s a problem because it would make it unconstitutional for the police to just walk up to people on the street and start asking them questions. Most people feel they can’t leave in such situations, so they would be constitutionally “seized” without a warrant; asking questions of someone without at least reasonable suspicion would be an unlawful seizure. But that’s a bad rule, because we want the police to be able to ask questions to try to solve crimes; it’s an essential technique, and although you might want to limit it (as in Miranda) you wouldn’t want to end it outright.
So what to do? One approach would be to articulate a very specific test that encompasses the desired results. For example, you could say that a person is seized when an officer “physically impedes the person’s motion, shows a weapon or otherwise threatens force, or verbally indicates to an individual that they cannot leave.” In other words, you could actually catalog the set of circumstances in which you think Constitutional protection should apply, and cause should be required. But this gets messy pretty quickly. The facts of cases are endless diverse, and the list of circumstances would have to be really long to be remotely complete. It’s not so easy to write that out as a constitutional rule.
Again, what to do? The Supreme Court’s solution to this problem was to take the simplicity of the “reasonable person” approach and tweak it to get the desired results. Formally speaking, the the test is whether a reasonable person would feel free to leave. But then the Justices invented an odd sort of imaginary “reasonable person.” Formally speaking, the Court’s way of getting to this result is to say that “the reasonable person test presupposes an innocent person,” Florida v. Bostick, 501 U.S. 429 (1991), but of course the test does more than that: This imaginary reasonable person generally feels free to leave unless the police show force, tell him to leave, or physically get in his way. Who is this reasonable person? Cynically speaking — although not inaccurately, I think — he’s the person who needed to be imagined to create a useful set of legal rules governing police conduct.
These sort of dynamics explain why just as there are no atheists in foxholes, there are no constitutional theorists in criminal procedure.** For the most part, the doctrine is not arising out of text, or history, or natural law, or First Principles of Immutable Justice. Rather, it’s arising out of a need to impose a set of reasonable limits on police practices given a few basic guideposts of doctrine and the common facts of police investigations. This dynamic occasionally leads to doctrinal tests that sound like they should work one way but actually work quite differently in practice.
** To be clear, there are in fact both atheists in foxholes and theorists in criminal procedure. My point is that experience with the law in this area tends to make it hard to be a theorist, not that it’s impossible.