When the Supreme Court handed down Herring v. United States, 555 U.S. 135 (2009), a case about the Fourth Amendment’s exclusionary rule, there was an interesting blog debate about how much it mattered. Tom Goldstein thought it was a revolutionary case; I thought it was not. Almost two years later, I think it’s fair to say that the disagreement has been shared by lower court judges. Some courts have interpreted Herring as changing very little. Others have seen it as drastically reshaping the exclusionary rule. I wanted to flag one of the latter readings of Herring that I think is pretty clearly wrong, but that some courts have adopted or strongly suggested: Weighing costs and benefits of the exclusionary rule on a case-by-case basis.
First, a bit of context. The Supreme Court adopted a cost/benefit approach to the exclusionary rule over 35 years ago in United States v. Calandra, 414 U.S. 338 (1974). The cost/benefit analysis is programmatic: Courts have to weigh the costs and benefits of having an exclusionary rule over the category of cases in which the case before the court resides. For example, the category might be searches made in violation of the knock-and-announce rule (as in Hudson v. Michigan). Or perhaps the category is searches made in reliance on a warrant that had minor defects (as in United States v. Leon). In each case, the issue is how the costs and benefits of the exclusionary rule applies over that defined category of cases. I see Herring as a relatively routine application of that principle: Herring weighs the costs and benefits of an exclusionary rule over the general category of arrests made as a result of merely negligent errors in police databases.
Of course, the tricky part of this programmatic weighing is that it’s never really clear ex ante what the category should be over which the costs and benefits must be measured. If you can manipulate the category, you can manipulate the cost/benefit balance. But that’s a broader problem of the doctrine, and of Fourth Amendment law more generally, that we need to bracket for now. (I’m working on an article about this problem, actually, as it’s a critical but largely unappreciated problem.)
Now let’s turn to what lower courts have done with Herring. Some courts have treated it as a relatively minor case. But some courts have treated it as a blockbuster, and in a very curious way. Specifically, without much (or any) analysis, some lower courts have construed Herring as now requiring a case-by-case weighing of costs and benefits. The category has been reduced to a single case — the case before the judges. In these courts, after finding a Fourth Amendment violation, the court must ask if the deterrent benefit of excluding the evidence in that one specific case is worth freeing that one specific defendant. See, e.g., Delker v. State, — So.3d –, 2010 WL 3911329 (Miss. 2010) (exclusionary rule not justified when officer conducted DUI stop outside jurisdictional boundary line because the defendant was a “recalcitrant, multiple-DUI offender” who could have killed someone if he had not been arrested, while the benefit getting officers to recognize the jurisdictional boundaries of their authority is comparatively modest). See also United States v. Julius, 610 F.3d 60 (2d Cir. 2010) (“Herring requires careful consideration by district courts of whether the goal of deterring violations of the Fourth Amendment outweighs the costs to truth-seeking and law enforcement objectives in each case.”).
The exact clarity by which lower courts have reduced the category to a single case varies from court to court. To be candid, I don’t think most judges even realize the category problem: Most judges seem to just weigh interests that occur to them, without thinking carefully about the specific category over which they are balancing. But the trend at least among some courts is to reduce the category to the single case before them. Those judges think Herring requires them to ask whether the benefit of applying the exclusionary rule in that one case will be greater than the social cost of freeing the one defendant in the litigation before them.
In my view, reducing the category to a single case just has to be wrong. Assessing the impact of the exclusionary rule for a single case is a metaphysical sort of question. On the deterrent side, it seems to depend on how many people read the opinion and how much press attention it gets. On the cost side, it would seem to defend on how bad the one defendant is. Weighing the costs and benefits, the rule would seem to become that the exclusionary rule doesn’t apply whenever the cops catch a really really bad person — at least unless the case get picks up by the legal blogs and then makes its way to the New York Times and then gets read by a lot of cops. But how would you even know that? Do you have media consultants provide expert testimony as to how the press, the blogs, and police academies will react to different hypothetical decisions? Would the outcome hinge on whether the opinion is published, unpublished, or just a one line statement affirming or reversing that leaves its rationale unknown?
I can see the disagreement about the proper scope of Herring, but I just don’t see how it can be read to collapse the balancing test to a case-by-case balancing for each individual decision. The scope of the exclusionary rule has traditionally been about balancing costs and benefits over categories — balancing that leads to outcomes, and outcomes that lead to rules for whether the exclusionary rule applies to that category. It strikes me as a very different enterprise, and not one called for by Herring, to replace that system with case-by-case balancing.