Eric A. Johnson: Assistant Professor of Law, University of Wyoming College of Law. I am grateful to Joshua Dressler and Diane Courselle for their helpful comments on an earlier draft; to the George Hopper Faculty Research Fund for its generous support of this research; and to Matthew Lynch of the Iowa Law Review for his thoughtful editorial guidance. All errors are my own.
Conventional wisdom teaches that the general rule governing factual cause in criminal law "is strict, straightforward, and virtually without exception"1: a criminal defendant's conduct will qualify as a "cause" of a prohibited result only if "it is an antecedent but for which the result in question would not have occurred."2 This conventional wisdom is not borne out by the case law, however. In a number of recent cases, appellate courts have upheld the homicide convictions of defendants whose conduct merely reduced the victim's chances of surviving another injury or illness that might well have caused the victim's death anyway. In these cases, it could not be said that the victim would not have died but for the defendant's conduct; rather, it could only be said that the victim might not have died but for the defendant's conduct.3 None of these decisions is explained, moreover, by Page 62 the rule that "one who hastens the victim's death is a cause of his death."4Nor do any of them fall within the special exception for cases where two persons simultaneously and independently inflict mortal wounds on a third person.5
What makes these cases surprising is not just their number but also their variety. They have little in common apart from the nature of the causal connection between the defendant's conduct and the result; apart, that is, from the fact that the defendant's conduct merely reduced the victim's chances of surviving another injury or illness. Not surprisingly, some of the cases involved culpable omissions by the defendant.6 In People v. Knapp,7 for example, it was the defendant's abandonment of his injured companion in a secluded area that reduced the companion's chances of surviving a head injury. And in State v. Muro,8 it was a mother's failure to obtain medical care for her daughter that reduced the daughter's chances of surviving a skull fracture inflicted by her father. Some of these cases, however, involved affirmative acts by the defendant. In State v. Montoya,9 for example, it was the defendant's kidnapping of the victim that reduced the victim's chances of surviving an earlier gunshot wound. And in Armstrong v. State,10 it was the Page 63 defendant's battery of the victim that reduced her chances of surviving her extreme intoxication.
Though this class of cases has gone wholly unnoticed among scholars of criminal law, the cases' civil counterparts are well known to scholars of tort law, among whom they are referred to as "lost chance" or "loss of chance" cases.11 Tort scholars have recognized,12 as have the courts in tort cases,13that proof of a lost chance-proof that the plaintiff might have survived but for the defendant's actions-does not suffice to establish but-for causation. Victims of lost chance nevertheless have had a measure of success in civil actions, "primarily in medical malpractice cases involving improper diagnosis or treatment."14 Some courts have accommodated these cases by relaxing the test for factual cause.15 Others have accommodated these cases by "reconceptualizing the harm," i.e., by recognizing the "lost chance" itself Page 64 as a compensable injury.16 All of these courts, though, have recognized that some modification of traditional principles is necessary if the tort plaintiff is to satisfy his or her burden of proof on the issue of causation in these cases.
Given the uniformity of this recognition in the tort cases and tort scholarship, it is striking to what degree the courts in similar criminal cases have failed to perceive any issue whatsoever. Even as these courts have upheld the homicide convictions of defendants whose only role in the victim's death was a small reduction in her chances of surviving another injury, the courts have generally displayed no recognition that they were departing from traditional principles of causation.17 Indeed, in several of these cases, the courts insisted that they were merely applying the but-for test as they always had.18 Moreover, with two exceptions, the courts arrived at these results without adverting to any other case where a court had reached the same result on similar evidence.19 These cases do not, then, represent an emerging "line of authority." Rather, courts widely scattered around the country have arrived at eerily similar results without any cognizance of one another and without any rationale to guide them. The cases reflect only a remarkable uniformity of intuition.
The purposes of this Article are, first, to draw attention to these decisions and, second, to attempt to make sense of the courts' shared intuitions. (The Article could take its motto from jazz pianist Bill Evans, who said "Intuition has to lead knowledge, but it can't be out there on its own."20) In service of the second of these purposes, the Article will argue tentatively that what defines these cases is the complementary relationship Page 65 between the defendant's culpable act or omission and another non- background causal factor, such as a grave injury inflicted by another person. Only in the presence of this complementary relationship-only, that is, when the defendant's conduct increases the risk posed by some other non- background causal factor-does it make sense to say, as the courts have said in the lost-chance homicide cases, that the defendant deprived the victim of a "chance" of surviving.
This complementary relationship explains, in turn, why the courts have intuitively treated lost chance as a species of causation. Tort scholars have long recognized that the essence of the causal relationship lies in the sufficiency of the causal antecedent to produce the result, rather than in its necessity.21 Accordingly, they have taken the position-most recently in the tentative draft of the Restatement (Third) of Torts-that a defendant's conduct will qualify as a "cause" if it was part of any "causal set" sufficient to bring about the result.22 The second part of this Article explains why the principle of "multiple sufficient causal sets" appears to encompass every homicide where the defendant contributed to the cumulative effect of a causal process that led to the victim's death-why, in other words, this principle appears to encompass the lost-chance homicide cases.
As it turns out, this connection explains not just the lost-chance cases but another category of cases that criminal law scholars have long treated as anomalous: namely, cases where the defendant is prosecuted as an accomplice.23 Courts have never required as an element of accomplice liability that the accomplice's aid or encouragement qualify as a but-for cause of, say, the victim's death. What they have required instead, as Part III of this Article will argue, is that the accomplice "be shown to have put the deceased at a disadvantage, to have deprived him of a single chance at life."24 This connection between lost-chance liability and accomplice liability answers an argument raised by Joshua Dressler in 1985-and until now unanswered-that the law's treatment of accomplices "deviates from the normal rules of criminal liability" and that this supposed deviation should be corrected by extending the rule of but-for causation to accomplices.25
Finally, Part IV of this Article will address the policy implications of imposing criminal liability for the loss of a chance. Specifically, it will argue that the imposition of liability for the loss of a chance satisfies whatever rationale lies behind the decisions of state legislatures to assign weight to causation in the classification of crimes. Page 66
According to the prevailing orthodoxy, the issue of causation in criminal cases raises two distinct questions, one easy and one very hard.26The "initial factual inquiry" is whether the defendant's conduct was a but-for cause of the result; that is, whether the result would not have occurred "but for" the defendant's conduct.27 The second, policy-bound inquiry is whether the causal connection between the defendant's conduct and the result satisfies the additional requirements that are generally classified under the heading of "proximate cause."28 This orthodox treatment of but-for causation as a minimum threshold requirement is...