181 A.3d 615 (Del. 2018)
In a girls’ bathroom in a Wilmington high school, a verbal confrontation between two 16-year-old students, Tracy Cannon and Alcee Johnson-Franklin,1 turned physical when Tracy threw Alcee to the ground and started throwing punches. Alcee tried to protect herself from the blows, and the two ended up on the floor of the bathroom, grappling and kicking at each other. It was over in less than a minute, but within two hours of the assault, Alcee was pronounced dead—not from blunt-force trauma, but from a rare heart condition that even Alcee did not know she had. This tragic result prompted the State to charge Tracy with criminally negligent homicide, and, after a five-day bench trial in Family Court, she was adjudicated delinquent.2
Tracy appeals. She contends that no reasonable factfinder could have found that she acted with criminal negligence or, even if she did, that it would be just to blame her for Alcee’s death given how unforeseeable it was that her attack would cause a 16-year-old to die from cardiac arrest. We agree. A defendant cannot be held responsible for criminally negligent homicide unless there was a risk of death of such a nature and degree that her failure to see it was a gross deviation from what a reasonable person would have understood, and no reasonable factfinder could conclude that Tracy’s attack—which inflicted only minor physical injuries—posed a risk of death so great that Tracy was grossly deviant for not recognizing it. And even if—as the Family Court saw it—Tracy should have realized that her attack might have deadly consequences because she carried it out in the close confines of the bathroom (near its tile floor and hard fixtures), Alcee’s death had nothing to do with those risks, and they were too far removed from the way that she died to blame Tracy for her death. We therefore reverse the Family Court’s adjudication that Tracy was delinquent of criminally negligent homicide.
A delinquency proceeding in Family Court is not a criminal prosecution,3 but the burden on the State to prove its case is no lighter. Branding a juvenile “delinquent,” and subjecting her to “the stigma of a finding that [she] violated a criminal law” and the possible deprivation of her liberty, “is comparable in seriousness to a felony prosecution” and cannot be done “on proof insufficient to convict [her] were [she] an adult.”4 So when a juvenile challenges whether the record supports a finding of delinquency, we ask, just as we would were this an adult criminal case, whether the evidence, viewed in the light most favorable to the State, would permit a reasonable judge of the facts to find “the essential elements of the crime beyond a reasonable doubt.”5
A clear picture of the events leading up to Alcee’s death emerged at trial. On the day before her fatal encounter with Tracy, Alcee, a 16-year old sophomore at Howard High School of Technology, sent a relatively innocuous text message to a friend warning her to be careful in her communications with schoolmates, some of whom might not be trustworthy. The text was sent to a group that included Tracy Cannon, one of Alcee’s classmates. Believing that she was the butt of the comment, Tracy took offense and exchanged a series of text messages with Alcee that became increasingly contentious. After one of the messages, Alcee—who was in English class at the time—asked for and received permission to go to the bathroom. Under that pretense, she met Tracy and another girl in one of the school’s bathrooms.
According to one of Alcee’s close friends, who was with her in English class, Alcee went to that bathroom expecting a fight, but she ended up returning to class ten or fifteen minutes later, unrattled and reporting that “nothing happened.”6 Alcee was similarly nonchalant about the encounter while discussing it with another friend later that day, who recalled that Alcee had “brushed it off . . . [and] didn’t make it seem like a big deal.”7 But at trial, a hall monitor who had been near the bathroom that day testified that he had overheard “a lot of screaming . . . [and] loud noise.”8
Unfortunately, the quarrel did not end there. Later that day, the other student who had been in the bathroom with Alcee and Tracy posted a video on Snapchat of a portion of the argument, with the caption, “[Tracy] bouta fight her.” The video also depicted Tracy, this other student, and a third student walking down a hallway at school, bragging that they intended to “get” Alcee.
The next morning, Tracy and her allies made good on their taunt. Before first period, Tracy and the other two students followed Alcee as she walked into a bathroom. A number of other students crowded in behind, including Alcee’s friend from English class. At trial, she described what happened next:
When they got in the bathroom, everyone was having a conversation. I really do not know . . . remember what they were talking about, but everyone was trying to solve the problem that [Alcee] told me was already solved [the day before].
So when we got in there, you know, everyone was talking quietly and having a peaceful conversation, and a crowd started hovering over everyone. And I turned around and it was a lot of people there. People started screaming, and I turned around for one second, and as soon as I turned back around, [Tracy] slammed [Alcee] on the ground. And all three girls before that dropped their bags as the crowd started screaming and yelling.9
The testimony concerning what happened after Tracy “slammed” Alcee to the ground is limited, presumably because the prosecution relied upon a video of the encounter captured by one of the onlooking students, which picks up soon after Tracy attacked Alcee. The video lasts approximately 45 seconds, 20 seconds of which depicts a one-sided attack. Tracy yanked Alcee by her hair and started striking her—by and large ineffectually—with loosely-balled fists, before pulling her again by her hair. Alcee tried to defend herself, and the two of them ended up flailing at each other on the floor. Although the Family Court said that Tracy kicked Alcee in the face, the video shows both girls pushing against each other with their feet as they grappled on the floor, while some of the other students sought to pull them apart.
Around that time, a teacher, hearing the commotion, entered the bathroom and dispersed the crowd. Tracy left with the other students, but the teacher found Alcee sitting on the floor with her hands behind her, her face pale. When she asked Alcee what happened, Alcee said, “they snuck me and then they jumped me.”10 The teacher ran for a nurse, who arrived to find Alcee laying on the floor, moaning and holding her abdomen. Emergency responders were called shortly thereafter, but before they could arrive, Alcee, who had been slipping in and out of consciousness, stopped breathing and lost her pulse. Despite intense efforts to revive her, Alcee was declared dead at the hospital, a little over an hour after Tracy’s attack.
Alcee’s autopsy revealed that none of the blows Tracy inflicted were sufficient to cause her death or, for that matter, serious physical injury. The medical examiner found no outward evidence of trauma, other than some “minor soft tissue injuries,” which included contusions under her eyes, broken fingernails, and abrasions on her arms and one of her knuckles.11 The cause of her death was found to be “sudden cardiac death due to [a] large atrial septal defect and pulmonary hypertension,” with the emotional and physical stress from the assault acting as a “contributing” cause.12
At trial, Tracy’s medical expert provided additional insight into the progression and incidence of Alcee’s heart disease. He explained that her combination of heart conditions—an atrial septal defect coupled with pulmonary hypertension—is known as Eisenmenger Syndrome and is so rare in children that its incidence is difficult to accurately quantify. One study he cited found a combination of that nature in only 2.2 out of every million children (or 0.00022% of the juvenile population).
There was no dispute at trial that Alcee’s heart disease had been unknown to her, her family, or her pediatricians. Indeed, Tracy’s medical expert found nothing in Alcee’s medical records to suggest that she was at risk of sudden cardiac death. And in the absence of such a diagnosis, Alcee’s life-threatening response to Tracy’s attack was impossible to predict, “[even] by her pediatrician and certainly not by a lay person.”13 It is “extraordinarily rare,” Tracy’s expert said, for emotional and physical stress to trigger the death of a 16-year old with no prior indication of cardiac disease.14
Our observations about the feebleness of Tracy’s blows and the lack of impact they had on Alcee’s death are not intended to minimize the wrongfulness of Tracy’s behavior. The record amply supports the Family Court’s portrayal of the encounter as an “attack” and an “act of violence” that Tracy inflicted for nothing more than “a perceived slight on social media.”15 But as will be seen, we are constrained by the undisputed facts about the medical cause of Alcee’s death. The Family Court adjudicated Tracy delinquent of criminally negligent homicide, not assault, so the issue before us is not whether Tracy should have to answer for assaulting Alcee, but whether she can be blamed for Alcee’s death, despite the fact that her blows inflicted little physical harm and that it was not those blows (or even the potentially-dangerous environs of the bathroom) that proved fatal.
To be liable for criminally negligent homicide, as the State charged, Tracy must have, “with criminal negligence, . . . cause[d] the death” of Alcee.16 Those are two distinct requirements, both with special meaning under Delaware law, and both must be satisfied to find that she committed criminally negligent homicide.
The first of those requirements requires proof that Tracy was criminally negligent. That requires more than just proof that she engaged in conduct that posed a risk of death. “All behavior that causes harm to some victim is,” in hindsight, “behavior that imposed some risk of that harm,”17 and the mere fact that someone’s behavior posed some risk of death is too slender a reed for criminal culpability. To be criminally negligent, Tracy’s conduct must have posed a risk of death “of such a nature and degree that [her] failure to perceive it constitutes a gross deviation from the standard of conduct that a reasonable person would observe in the situation.”18 Her failure to recognize that her behavior had deadly consequences must, in other words, have been so “abnormal” that her indifference to the risk of death is markedly disparate from how a reasonable person in her position would have grasped the situation.19
And even a person who creates a risk of death of that degree is not responsible for the victim’s death unless it was her risky behavior that caused it. This second requirement—causation—requires more than just proof that the defendant’s conduct set a chain of events in motion that led to the victim’s death. “But-for” causation is a necessary part of proving culpability, but that is just the start.20 There must also be a sufficient relationship between the nature of the risk the defendant created and the way that the victim’s death transpired for the defendant to be blamed for it. This principle is codified in section 263 of the Delaware Criminal Code, which provides that if the actual result of a defendant’s conduct was “outside the risk of which . . . the defendant should be aware” and was “too remote or accidental in its occurrence to have a bearing on [her] liability or on the gravity of the offense,” then the defendant is not criminally responsible for what transpired.21
Both of these requirements—criminal negligence and causation—must have been proved beyond a reasonable doubt for Tracy to have been adjudged delinquent of criminally negligent homicide.
The Family Court found that both of these requirements were met. On the first requirement—criminal negligence—the court found that Tracy’s attack posed a sufficiently obvious risk of death to blame her for not recognizing it, linking that risk not so much to what Tracy did as to where she did it:
The attack on [Alcee] had known potential consequences, risks. . . . A physical assault on another carries with it intended consequences that is to injure the victim. . . . While the extent of that injury may not specifically be intended or contemplated, the risk nevertheless exists. The attack carried out by [Tracy] in the close confines of the school bathroom stall posed risk of potential catastrophic physical harm including death by virtue of the tile floor, walls and fixtures. . . . It is within the realm of contemplation that a physical assault intended to cause injury could in fact result in death; that is death was clearly a risk of [Tracy’s] conduct.22
As for the second requirement—whether there was a sufficiently close link between the risk that Tracy created (death by hard bathroom surfaces) and the way that Alcee died (from a rare cardiac condition) to justify blaming Tracy’s negligence for her death—the court concluded that there was:
For there to be criminally negligent homicide, the death must be the natural and probable consequence of the unlawful act and not the result of intervening causes. The natural and probable consequence of a physical attack on an individual is physical and emotional trauma which may, depending on the particular circumstances of the victim, result in varying degrees of physical harm, up to in its most severe degree, the death of the victim. . . . In this case, the evidence establishes beyond a reasonable doubt that the death of [Alcee] was caused by the action of [Tracy].23
Tracy challenges both of the Family Court’s findings. First, she believes that when the Family Court found that she had acted with criminal negligence, the court conflated the question of whether her attack posed some conceivable risk of death with whether that risk was so obvious that it was grossly abnormal of her to have not perceived it. She also stresses that she was only 16 at the time and that it is doubtful that a reasonable 16-year-old would realize that an attack like hers could have deadly consequences—let alone that it would be a gross deviation for a 16-year-old not to perceive that risk.
Second, she argues that even if the location of the attack, in the close confines of the bathroom, did pose such an obvious risk of death, the Family Court gave short shrift to the fact that this risk had nothing to do with Alcee’s death. Tracy points out, correctly, that although she raised it below, the Family Court never cited section 263, the part of the Delaware Criminal Code that requires an examination into whether the actual result of a defendant’s negligent conduct was “outside the risk of which . . . [she] should be aware” and “too remote or accidental in its occurrence” to warrant a finding of criminal culpability. She believes that if the court had looked at the events through the lens of section 263, it would have realized that the risk the court admonished her for creating was too far removed from the way Alcee died to blame her for Alcee’s death.
We will take up these two contentions in turn.
Both sides vigorously debate whether it was reasonable for the Family Court to find that Tracy acted with criminal negligence. Much of their disagreement centers on whether Tracy’s perception of the risks of her conduct should be judged by the standard of a reasonable adult or a reasonable person of her own age.
Tracy urged the Family Court to view the events of that day through the eyes of a “reasonable sixteen-year-old sophomore girl.”24 She pointed out that it is “well-established” in context of tort-law negligence that a juvenile’s conduct must be measured against the standard of a reasonable person of the same age, experience, maturity, and capacity,25 not a reasonable adult, and she argued that no less should be true in the context of criminal negligence, where it is not just money damages at stake, but criminal stigmatization. For support, she drew from research in the field of adolescent development, which has found that “adolescent brains are not yet fully mature in regions and systems related to higher-order executive functions such as impulse control, planning ahead, and risk avoidance.”26 That research played a central role in the United States Supreme Court’s recent trilogy of juvenile-sentencing cases—Roper v. Simmons,27 Graham v. Florida,28 and Miller v. Alabama29—all of which recognized that there are “significant gaps between juveniles and adults,” which leave juveniles prone to “transient rashness,” a “proclivity for risk,” and an “inability to assess consequences.”30 Those characteristics not only inhibit their ability to conform to an adult standard of care, but they also make them less blameworthy when they fall short.
At trial, Tracy offered testimony from an expert in adolescent development whose work informed both Roper and Miller. Echoing the Court’s observations in those cases, he explained that even juveniles in their mid- to late-teens lag behind in the “development of higher-ordered, more advanced thinking abilities and the development of self-control and self-regulation,” which makes them “less likely to even consider whether something is risky, less likely to think about the harmful consequences that could come from doing something risky, and . . . less likely to think that those consequences are going to be very negative.”31 He opined that “it would be very unlikely that a kid getting into a fistfight would think that somebody might die as a result of it,” particularly given how “common it is for kids to get into fights at school” and how rare it is for a physical altercation among high school students to end in death.
The State maintained that Tracy should be judged by an adult standard, but it was unable to muster a single example of a court that had rejected the notion that a juvenile facing a criminal negligence charge in a delinquency proceeding must be measured against the standard of a person her own age—except for cases where juveniles had engaged in certain characteristically-adult-like activities, like driving a car. Faced with that reality, the State was left to argue that when Tracy attacked Alcee, she was engaged in adult-like behavior, even though the so-called adult-activity exception tends to be confined in practice to juveniles operating some type of motorized vehicle,32 and even though lashing out physically like Tracy did is quintessentially childlike behavior. The State also argued that Tracy—at 16—was too old for the special child-negligence standard, even though examples abound (in tort law, at least) of courts applying the child-negligence standard to juveniles until they reach the legal age of majority,33 and even though it is a matter of both science and common sense that “a 16-year-old [still] customarily lacks the maturity of an adult.”34
The Family Court sided with Tracy, deeming it “altogether appropriate” to consider her age in determining whether it was grossly abnormal for her not to realize that her conduct could have deadly consequences.35 But the court nonetheless concluded that it would be grossly deviant by even a reasonable-16-year-old standard not to perceive that risk, rejecting the evidence Tracy presented about adolescent development and risk-perception with an observation that “[e]ven reasonable 16-year-olds have standards.”36
Neither side was happy with the outcome. Tracy, of course, disagrees with the Family Court’s ultimate assessment, while the State maintains that the Family Court erred by not holding her to an adult standard.
We need not decide whether the Family Court was right to measure Tracy’s conduct against a reasonable person of her own age, because in our view, no reasonable factfinder could conclude under even an adult standard that Tracy’s attack posed such a risk of death that it was a gross deviation for her not to be aware of it.
The Family Court properly recognized that to have acted with criminal negligence, Tracy must have “fail[ed] to perceive that her conduct created a risk of death” and “her failure to perceive such a risk [must be] a gross deviation from the standard of conduct that a reasonable person would observe in the situation.”37 And in line with that standard, the Family Court ultimately concluded that Tracy’s “failure to perceive the risk of death from [her] attack . . . was a gross deviation.”38
But the explanation the court gave for that finding focused almost entirely on whether there was a risk of death inherent in Tracy’s conduct—not whether that risk was so readily apparent that failing to recognize it would be grossly abnormal. The court observed, for example, that attacking Alcee in the close confines of the bathroom “posed risk of potential catastrophic harm including death by virtue of the tile floor, walls and fixtures.”39 That may be so. But what matters is the nature and degree of that risk, not just that the risk was present. The court also suggested, more generally, that “it is within the realm of contemplation that a physical assault intended to cause injury could in fact result in death.”40 But to find someone criminally negligent, it is not enough that a risk of death was “within the realm of contemplation”—it must have been “so great that the . . . failure to perceive it constitutes a gross deviation from the standard of care that a reasonable person would exercise under the circumstances.”41 Even the court’s strongest statement—that “death was clearly a risk of [Tracy’s] conduct”42—does not get us to criminal negligence. That it is possible to say, in hindsight, that death was clearly a risk of her conduct does not mean that the risk would have been clear to someone in Tracy’s position who was bound up in the moment.
The court’s reluctance to settle on any one characterization of the apparentness of the risk—describing it in turns as a risk that was “posed,” “within the realm of contemplation” and, as the court suggested at one point, a risk that was “not specifically . . . intended or contemplated” but that “nevertheless exist[ed]”43—leaves the distinct impression that the court was not truly comfortable saying (beyond a reasonable doubt, no less) that the risk was so apparent that it was grossly deviant of Tracy not to recognize it.
We think that was for good reason. Tracy’s attack—reprehensible as it was—was not so violent that it presented a readily apparent risk of death. As we said at the outset, we do not disagree with the Family Court’s finding that Tracy’s conduct was an “attack” and an “act of violence,” and “not a fight between two teenagers squaring off to settle some mutual grievance.”44 Nor do we disagree that “[i]t is within the realm of contemplation that a physical assault . . . could in fact result in death.”45 Every lawyer has heard of a case where, by some unusual confluence of events, a victim of a fairly innocuous scuffle winds up dead.46 But for criminal negligence, what matters is whether Tracy’s attack posed not just the possibility of death, but a risk so apparent that Tracy veered far off course by not realizing that her attack may have fatal consequences. The facts do not support such a conclusion here.
We are fortunate to have a video of the attack—not because it makes for easier second-guessing of the trier of fact, but because physical attacks can vary greatly in their violence, and the recording offers an unvarnished look at the circumstances of this one. And what is evident from that recording is that this was not a severely violent attack. Tracy pulled Alcee to the ground, threw a quick succession of awkward punches, pulled her by her hair, and then jumped on top of her. By the end, the two of them were on the floor, grappling and kicking at each other until other students pulled them apart. In under a minute, it was over. The impression the video imparts is that this was not an attack that posed a readily apparent risk of death—an impression that is reinforced by the results of Alcee’s autopsy, which revealed little evidence of physical harm.
The Family Court stressed that the danger here was from the surroundings—the “close confines of the school bathroom” with its “tile floor, walls and fixtures.”47 But the video reveals that the surroundings were not all that close. The attack took place in what appears to be a large area in the rear of the bathroom, and the girls do not come within a body-length of a sink or toilet. Had Tracy swung Alcee’s head toward a wall or hurled her toward one of those fixtures, that may have been enough to give a reasonable person pause—and might suggest that Tracy would have been grossly deviant to have not thought twice. But the circumstances here were little different than if the attack had unfolded in a hallway outside of class.
To be sure, whether a defendant was criminally negligent is a question of fact, firmly committed to the factfinder to decide. But when the record does not reasonably permit a finding, beyond a reasonable doubt, that the defendant acted with that culpable mental state, the finding cannot stand.
Tracy’s second argument on appeal concerns the link between the risk the Family Court said she created—the possibility of death from the close confines of the bathroom—and the way that Alcee died. As we explained at the outset, even if a defendant created a risk of death and it was so great a risk that she was criminally negligent, the State must prove that it was her negligence that caused the victim’s death.
The threshold inquiry is whether the defendant’s risky conduct was “an antecedent but for which the result in question would not have occurred”48—the traditional concept of “but-for” causation. There is no question that aspect of the causation inquiry is satisfied here. The deputy chief medical examiner concluded that the “emotional and physical stress due to physical assault” was a “contributing” cause to Alcee’s death,49 and as the Family Court recognized, had it not been for Tracy’s attack, Alcee would not have died that morning.
But the causation inquiry does not stop there. It is not enough for criminal culpability that a defendant’s negligence happened to set in motion a chain of events that led to the victim’s death. “As the law has consistently recognized, some limitation of this broad principle is necessary,”50 and under Delaware law, any time the “actual result” of a defendant’s conduct “is outside the risk of which the defendant . . . should [have] be[en] aware,” a more searching inquiry must be conducted into the relationship between the risks that the defendant created and the actual harm that resulted.51 This principle is codified at 11 Del. C. § 263, which is broken into two parts to cover the two types of scenarios where the actual result of a defendant’s conduct deviates from the scope of the risk that the defendant should have recognized:
The element of . . . negligent causation is not established if the actual result is outside the risk of which . . . the defendant should be aware unless:
(1) The actual result differs from the probable result only in the respect that a different person or different property is injured or affected or that the probable injury or harm would have been more serious or more extensive than that caused; or
(2) The actual result involves the same kind of injury or harm as the probable result and is not too remote or accidental in its occurrence to have a bearing on the actor’s liability or on the gravity of the offense.
Subsection (1) deals with the problem of the “transferred intent” cases. This is not one of those cases. But this case does implicate subsection (2), which deals with “the traditional causation cases”52—cases where “the precise injury inflicted was different or occurred in a different way” than the defendant’s risky conduct would have suggested.53 In these types of cases, the defendant should be relieved of liability if the actual result that transpired was “too remote or accidental in its occurrence,” relative to the risk that the defendant created, “to have a bearing on the actor’s liability or on the gravity of the offense.”54
Despite the fact that Alcee’s death, by cardiac arrest, had little to do with the risk that the Family Court said Tracy created (the risk of death by contact with a hard surface)—and despite the fact that Tracy’s lawyers argued to the Family Court that section 263 required a finding in her favor because of that disconnect—the Family Court did not mention this provision in its opinion. The State acknowledges that omission, but it is confident that “a review of [the court’s] decision reveals that the court did not ignore the statute, but instead analyzed the facts fully, considered the issue and accurately applied the law.”55
It is true that there are some passages in the court’s opinion that suggest that the court was aware that it needed to look beyond mere but-for causation and examine the relationship between the risks of Tracy’s conduct and the way that Alcee died. But rather than analyze the issue under section 263, the Family Court spoke in the old common law terms of proximate cause, finding Alcee’s death to be the “natural and probable consequence” of Tracy’s behavior and declining to find Alcee’s rare heart condition to be an “intervening cause[].”
The innovation of section 263 is that it did away with the “varying and sometimes inconsistent” language of proximate cause and the overall “obscurity of that concept.”56 Terms like “natural and probable consequences” and “intervening and superseding” cause have been rightfully criticized as “ritualistic formula[s]”57 and “legal mumbo-jumbo,”58 and they tend to only obscure the real question of whether it is just for a defendant to be held responsible for a series of events that may have had little to do with the risk she created.59 Section 263 is designed to shine a light directly on that question.
In our view, even if the Family Court were correct that the hard surfaces of the bathroom posed such an obvious risk of death that Tracy was grossly deviant for not recognizing it, the facts at trial did not establish a sufficient relationship between that risk and the way Alcee died to blame Tracy for her death. The Family Court erred when it failed to use the framework of section 263—which is designed for difficult causation cases like this one.
In Bullock v State,60 we addressed how section 263 modifies section 261’s rigid “but-for” causation standard. Bullock was charged with recklessly causing another motorist’s death while driving drunk at excessive speed.61 But at trial, the evidence suggested that the collision occurred because the other motorist disregarded a red light and collided with Bullock, who had entered the intersection while the light in his direction was yellow.
He was convicted by a jury of manslaughter, and he appealed. The State emphasized that but-for Bullock’s speed and impairment, he would not have been in that intersection at that moment, and the collision would not have occurred. In an argument with obvious parallels to the State’s position in the case before us now, the State contended that “[b]ecause Bullock should have foreseen that a collision with [the other motorist] could have resulted from his speeding while impaired by alcohol, . . . Bullock should be criminally liable for the consequences of his actions.62 But Bullock, relying upon section 263, contended that the trial judge erred by not instructing the jury that “reckless causation is not established if the actual result”—the other motorist’s deadly failure to heed a red light—was “outside the risk of which [Bullock] was aware”—the possibility that his intoxication could cause him to operate his vehicle unsafely.63
We held that the trial judge’s failure to give that instruction—even though Bullock had not thought to ask for it at trial—was plain error. We rejected the State’s “facile[]” approach to reckless causation because it ignored section 263’s modification of section 261’s “but-for” causation test. The essential problem, we pointed out, was that absent that instruction, the jury would not “know, under the instruction as given, that Bullock could act recklessly as the State described yet not be criminally responsible for the collision and death.”64 Bullock involved a charge of recklessness, not criminal negligence, but the essential question in both cases is the same.
The essence of section 263 is that it negates negligent causation if the actual result of a defendant’s conduct is outside the risk of which she should have been aware—unless it can be said that the actual result was “not too remote or accidental in its occurrence to have a bearing on the actor’s liability or on the gravity of the offense.”65 To properly undertake this analysis, the “risks” and “results” to which section 263 speaks must be defined in terms of their “specific manner and character of occurrence.”66 In other words, a risk or result should not be defined generally, such as “death” (as the State suggests in its briefing), or even “death following a physical altercation,” but more specifically, such as “death from an impact with a hard surface in a bathroom” or “sudden cardiac death caused primarily by unknown heart defect.”67
Under this framework, we are compelled to identify the actual result of Tracy’s conduct as Alcee’s death by sudden cardiac arrest. So what then was the “probable result” of Tracy’s conduct? The Family Court found that Tracy’s conduct “posed [a] risk of potential catastrophic physical harm including death by virtue of the tile floor, walls and fixtures.” Although the court articulated that finding in terms of “possibility” rather than probability, that was as close as the court got to identifying the “probable result,”68 and we are willing to accept it as such for the purpose of our section 263 analysis.
The next consideration under section 263 is whether “the actual result involves the same kind of injury or harm as the probable result.” Because death was the harm or injury under both, the State’s case easily clears this hurdle.
But the “ultimate criterion” by which the significance of Alcee’s death occurring in a manner different than Tracy might have anticipated “ought to be judged . . . [is] whether the actual result is too remote or accidental in its occurrence to have a . . . bearing on her liability or the gravity of the offense.”69 Professor LaFave described this fundamental principle thusly:
[T]he forbidden result which actually occurs must be enough similar to, and occur in a manner enough similar to, the manner or result which . . . [the defendant’s] negligence created a risk of happening that the defendant may fairly be held responsible for the actual result even though it differs or happens in a different way from the . . . hazarded result.
Tracy contends that the proper application of section 263 to the factual record precluded a finding of negligent causation beyond a reasonable doubt, and we agree. We acknowledge that, like the question of whether a defendant acted with criminal negligence, questions of causation generally and of remoteness specifically are to be determined by the factfinder.70 And the drafters of section 263 recognized that the question of remoteness “will require an ad hoc determination in each case.”71 But given the statistical improbability that a 16-year-old like Alcee might suffer from a rare, undiagnosed heart condition that would leave her mortally vulnerable to the type of assault that occurred in this case, we can reach only one conclusion in this case: even reviewing the evidence in the light most favorable to the prosecution, this essential element could not be proven beyond a reasonable doubt.
The State insists that Alcee’s death was not “too remote or accidental” from the risk of death-by-hard-surfaces for Tracy to be fairly blamed for it, and for support, the State invokes another vestige of the common law: the eggshell-victim doctrine. In the State’s view, Tracy had to “take her victim as she found her,” so the fact that Alcee died as a result of rare medical condition—rather than from blunt force trauma from the bathroom’s hard surfaces—should not in any way lessen Tracy’s culpability.
The eggshell-victim doctrine has long been a fixture in both the civil and criminal law. In the words of one commentator, “[t]he principle of taking one’s victim as one finds her is so well embedded in the criminal law that it might seem fatuous to ask why this should be so.”72 But whether, and to what extent, it may ever have been a part of the criminal law in Delaware,73 the adoption of section 263 does not allow for a victim’s unknown medical condition to be dismissed out of hand as irrelevant to a defendant’s culpability. In every case where the “precise injury inflicted . . . occurred in a different way” than the defendant had intended or risked, Delaware law requires consideration of whether the result played out in a way that was “too remote or accidental” to blame the defendant,74 and the statute makes no exception for cases where an unknown medical condition was at work. Quite the opposite: the commentary to section 2.03 of the Model Penal Code—on which section 263 was modeled—expressly contemplates that a victim’s “unexpected physical condition” should be taken into account in gauging the defendant’s culpability.75
It may be the case that when a defendant intends to kill her victim, it is not particularly troubling—from a culpability standpoint—that the victim’s death ended up resulting from an unforeseen medical condition rather than through whatever machination the defendant had in mind. When death was the defendant’s intent, she “got exactly what [she] wanted,” so “[w]hat right does [she] have to complain,” even if things played out differently than she had planned.76
But the opposite might “well be expected” in a criminal negligence case, where the defendant’s only misdeed was to be ignorant of a risk she created—a risk that ends up having nothing to do with the way the victim died.77 Criminal negligence pushes us out to “the fringes of criminal responsibility,”78 and we are naturally more wary of allowing an unforeseeable result to have too great an influence on a defendant’s culpability when we are uneasy about her culpability from the start. That is particularly true on this record, given just how unforeseeable it is that the stress from an attack like Tracy’s would cause a 16-year-old to suffer cardiac arrest.
That compels us to conclude that Alcee’s death, from cardiac arrest, was simply “too remote” from the hazards of Tracy’s conduct and “too accidental in its occurrence” to transform what Tracy did from a physical attack into criminally negligent homicide.
The Family Court’s adjudication of Tracy as delinquent of criminally negligent homicide is reversed. Tracy’s delinquency adjudication on the misdemeanor conspiracy charge, which she did not appeal, stands. Because the punishment that the Family Court imposed was based on both offenses, we remand this case to the Family Court to resentence Tracy solely on the offense of conspiracy in the third degree.
I agree with the Court’s conclusion that the evidence is insufficient to support a finding that Ms. Cannon acted with the criminal negligence required for criminally negligent homicide. That finding alone is enough to decide the case.
I write separately because I do not think it is necessary to engage in a § 263 analysis. In my view, stated simply, that section comes into play when the actor’s state of mind satisfies the definition of criminal negligence, but the person injured is not the probable person or the actual injury is not the probable injury. I think that where the actor does not possess a criminally negligent state of mind to begin with, § 263 need not be considered.