Professor of Law, Northwestern University; Julius Kreeger Professor of Law and Criminology Emeritus, The University of Chicago. I have profited from the comments of Marshall Shapo, John Stinneford, and Ronald Allen, from conversations with Andrew Koppelman and Steven Art, and from the research assistance ofJulia Rickert.
What does the Fourth Amendment exclusionary rule exclude? Mapp v. Ohio says it bars "all evidence obtained by searches and seizures in violation of the Constitution."1 Nardone v. United States says it excludes "fruit of the poisonous tree."2 Wong Sun v. United States says it "extends as well to the indirect as the direct products" of Fourth Amendment violations.3 United States v. Peltier says it suppresses evidence "gained as a result of" Fourth Amendment violations.4
All of these formulations pose a question of causation. Did a violation of the Fourth Amendment cause the government's receipt of evidence that a defendant now seeks to suppress?
When criminal procedure students consider what the exclusionary rule excludes, they return to a subject they are likely to have examined at length in classes on torts and substantive criminal law. For no apparent reason, however, the vocabulary is different. The opinions speak of "derivative evidence" and "fruit of the poisonous tree" rather than "proximate cause." They ask whether the "taint of the primary illegality has dissipated" rather than whether an "independent intervening cause" has broken the causal chain.
Students, like their teachers and like Supreme Court Justices, may assume that the governing principles are the same. As in the law of torts and crimes, there appear to be two requirements. The first is that the Fourth Amendment violation must be a sine qua non, "but-for cause," or "cause in fact" of the discovery of the challenged evidence. The second is that the violation must also be a proximate cause of this discovery. The first issue is seen as one of fact, and the second is seen as one of policy. The task at the second or proximate-cause stage is to choose from all the conditions without which the discovery of evidence would not have occurred those causes that the courts will treat as causes.5 Page 1744
In Hudson v. Michigan, the Supreme Court spoke in conventional causal language and declared that "but-for causality is only a necessary, not a sufficient, condition for suppression."6 Hudson was not the first Supreme Court decision to declare but-for causation essential,7 and Wayne LaFave, America's preeminent authority on the law of the Fourth Amendment, calls the Court's statement of the need for but-for causality "unassailable."8
LaFave, however, excoriates Hudson's application of this doctrine. The Court assumed in Hudson that the police had violated the Fourth Amendment by conducting a search without appropriately knocking and announcing their presence. It then declared that their "illegal manner of entry was not a but-for cause of obtaining the evidence."9 LaFave calls this statement "dead wrong."10 He writes that Hudson "deserves a special niche in the Supreme Court's pantheon of Fourth Amendment jurisprudence, as one would be hard-pressed to find another case with so many bogus arguments piled atop one another."11
I concur with LaFave that Hudson was wrongly decided, but this Article travels a very different route from his to this conclusion. Endorsing the statement that he calls "dead wrong," it contends that the failure to knock and announce in Hudson was not a but-for cause of the police seizure.
Questions of but-for causation must be resolved by asking whether, if the hypothesized cause were absent, the result would still have occurred. An Page 1745 appropriate analysis of this question requires envisioning a world as much like the existing world as possible apart from the absence of the supposed causal event. In a case like Hudson, the question is whether the police would have obtained the challenged evidence even if they had obeyed the Fourth Amendment. The police, however, might have complied with the Fourth Amendment in more than one way. For one thing, they might have complied by abandoning their search, but that scenario does not seem likely. A more plausible scenario-a "counterfactual" more closely resembling the actual world-is one in which they have complied with the Constitution by knocking as the Fourth Amendment requires. Because the police undoubtedly would have obtained the challenged evidence if they had knocked, Hudson was correct that their failure to knock was not a but-for cause of their seizure.
While this Article defends the statement that LaFave considers dead wrong, it assails the proposition he calls unassailable. The Supreme Court generally has not required a but-for causal relationship between a constitutional violation and the discovery of challenged evidence. Instead, the Court has followed the approach that common law courts took in civil actions at the time of the framing of the Fourth Amendment. When government officers failed to obtain a search warrant or to knock and announce their presence, these courts did not ask whether the officers would have inflicted the same harm if they had obtained the required warrant or knocked. Instead, they treated the officers as trespassers and held them strictly liable for any harm they produced. The courts asked only whether the officers' wrongful presence was a cause of their seizure, and it always was. The causal principles followed in search-and-seizure and other trespass cases differed from those followed in most tort and substantive criminal law decisions.
The approach of the common law begged a large question: Which violations of legal requirements turned officers into trespassers, and which did not? This Article argues in favor of a less ad-hoc standard-one that for the most part produces the same results as the common law and early exclusionary-rule cases but that is substantially more protective of Fourth Amendment rights than the requirement of but-for causation articulated in Hudson.
The appropriate standard in exclusionary rule cases is one of "contributory" rather than "but-for" causation. Courts should ask, not whether a constitutional violation enabled the police to obtain evidence they would not have obtained without it, but whether the constitutional violation facilitated the discovery of this evidence. A violation can facilitate the discovery of evidence either by increasing the likelihood of this discovery or by reducing the work required to make it.
This Article shows that the standard it proposes is in accord with nearly all of the results the Supreme Court has reached in exclusionary-rule cases Page 1746 and that Hudson's standard of but-for causation is not.12 It also contends that the proposed standard better accommodates the dual rationales of the exclusionary rule-a rule that seeks both to vindicate the rights of defendants who have been unlawfully searched and to inhibit future illegality. It contends that, although Hudson's requirement of but-for causation accomplishes the former goal (protecting the rights of defendants), it fails to achieve the latter (providing appropriate incentives for law observance by the police). It seems odd that the Court has disavowed the goal that Hudson achieves while endorsing the one it does not.
The development of this Article's thesis will require an analysis of the rule's objectives and of the ability of the two competing causal standards to achieve them. This analysis will require examining not only Hudson's ruling on causation but also two alternative rationales that the Court advanced for its decision. The Court maintained that the interests protected by the knock-and-announce requirement "would not be served by suppression of the evidence obtained"13 and that the social cost of excluding this evidence would outweigh the deterrent gain.14
Part I of this Article reviews the two sorts of justifications asserted for the exclusionary rule-"rights" justifications and "instrumental" justifications-and considers their differing causal implications. A "rights" theory of the rule seeks (within limits) to restore the victim of a constitutional wrong to the position he would have occupied had the wrong not occurred. This theory implies a standard of but-for causation. Excluding evidence the government would have obtained even in the absence of its wrong would afford the victim of the constitutional wrong...