Concept
Unforeseeable Plaintiffs
Foreseeable and unforeseeable plaintiffs
Unforeseeable Plaintiffs
A careless defendant does not answer to everyone a careless act happens to injure. Restatement (Second) of Torts § 281(b) requires that the actor's conduct be “negligent with respect to the other, or a class of persons within which he is included.”1 The preceding section used that clause generously: a manufacturer owes due care to any foreseeable plaintiff, so a remote user within the class a defect endangers may sue without privity.2 This section uses the same clause to mark the other edge. A plaintiff outside the group the conduct put at risk cannot say the defendant was careless toward her, and the duty does not reach her, however real her injury.
The label “unforeseeable plaintiff” invites a common mistake. It describes a plaintiff who was not apparently at risk when the defendant acted, not one who was hurt in a surprising way. A plaintiff inside the endangered class who suffers an unexpected injury is still a foreseeable plaintiff; how closely the manner of harm must be foreseen is a separate question that a later section takes up. The subsections below follow the unforeseeable plaintiff from Cardozo's majority opinion in Palsgraf, through Andrews's rival approach and the rule that class membership suffices, to the bystander who suffers emotional distress.
Palsgraf's Majority Rule
Palsgraf v. Long Island Railroad Co. supplies the standard statement of the unforeseeable-plaintiff rule. The New York Court of Appeals held in 1928 that a defendant's duty in negligence reaches only plaintiffs within the apparent zone of danger, the plaintiffs to whom harm could reasonably be foreseen.3 A plaintiff outside that zone cannot recover for carelessness that endangered someone else.4
The facts are brief. A railroad employee, assisting a passenger to board a train, dislodged a package the passenger carried.5 The package was small, about fifteen inches long, and covered by a newspaper. It held fireworks, but nothing in its appearance gave notice of that. When the package fell, the fireworks exploded, and the shock threw down some scales at the other end of the platform, many feet away. The scales struck Mrs. Palsgraf.6 A judgment for her was reversed, 4–3.7
Cardozo, writing for the majority, saw the issue as a duty problem. Negligence toward someone is not enough; the plaintiff must show a breach of duty toward her, and there is no duty unless harm to her was foreseeable. Because the package gave no notice of its contents, the railroad's employees had no reason to foresee harm to Mrs. Palsgraf from their conduct, and the case should not have gone to the jury.8 Whatever the employees owed the passenger, the railroad was not negligent in relation to her.9
Cardozo rested that conclusion on what negligence is. Negligence is not actionable unless it invades a legally protected interest, and “Proof of negligence in the air, so to speak, will not do.” Standing on the platform, Mrs. Palsgraf could claim protection against intentional invasion of her bodily security, and no one charged that. She could also claim protection against unintentional invasion by conduct that reasonable people would see as creating an unreasonable hazard that such an invasion would follow. Those were the bounds of her immunity.10 The employees' conduct created no hazard of that kind toward her, so she had to “sue in her own right for a wrong personal to her, and not as the vicarious beneficiary of a breach of duty to another.”11 Courts compress the idea into Cardozo's phrase: the risk reasonably to be perceived defines the duty to be obeyed.12
Later courts apply the same filter across far longer chains of events. In Stutts v. De Dietrich Group, plaintiffs injured by chemical weapons sued banks that had issued letters of credit to the suppliers of the raw materials. The court dismissed, invoking the risk-defines-the-duty phrase. Given the procession of events between the letters of credit and the injuries, it found “nothing to suggest to the most cautious mind that a letter of credit would cause the harm the plaintiffs allege,” and to attribute such foresight to the banks would be to attribute “a prescience that is beyond human ken.”13
The filter admits plaintiffs as readily as it excludes them when the defendant's task was performed for people in the plaintiff's position. In re September 11 Litigation, the court spoke of the expectation that the duty of screening was performed for the benefit of passengers and of those on the ground, including those present in the Twin Towers on the morning of September 11, 2001.14 Those victims fell within the class the duty protected. What separates the two cases is whether harm to the plaintiff's class was a risk the defendant's conduct made apparent, not how grave the injuries were.
Cardozo took Palsgraf from the jury because he did not think reasonable people could disagree about foreseeability: he thought injury to Mrs. Palsgraf was not even remotely foreseeable.15 Her original complaint, before amendment, had alleged a different mechanism, that a stampede of frightened passengers knocked the scale over.16 The rule therefore does its full work only for a plaintiff the defendant had no reason to anticipate at all. If injury to someone in her position was even remotely foreseeable, the question may belong to the jury.17 The usual exam error is to cite Palsgraf against a plaintiff who stood in the evident path of the danger.
Courts and commentators label the holding differently. The railroad clearly owed Mrs. Palsgraf a duty of some sort, a duty to use reasonable care to avoid foreseeable risks of harm to a person on its platform, so Cardozo can be read to hold that the railroad did not breach it, because harm to her was not a foreseeable risk of what the defendant had done.18 Whatever the label, the result is the same: she was an unforeseeable plaintiff as a matter of law.19 Most courts and the Restatement now state the limit as the risk principle, which confines negligence liability for physical harm to foreseeable consequences within the scope of the risk and to foreseeable plaintiffs.20 Judge Andrews, in dissent, would have drawn the line elsewhere.
Cardozo Versus Andrews
Judge Andrews rejected Cardozo's breach-of-duty analysis in dissent. Once the railroad was negligent in some way, he argued, the question was whether its negligence was a proximate cause of the plaintiff's injury, and on these facts that question was for the jury.22 For Andrews, proximate cause marks the point beyond which the law declines to trace a series of events, and he fixed that point by “convenience, of public policy, of a rough sense of justice.” Foreseeability may enter the judgment, but only as one factor and never as the determining one, and the jury would set the limits.23
His factors reached beyond foreseeability. One was how close the plaintiff stood in time and space to the defendant's negligence, and he saw foreseeability itself as a more malleable concept than Cardozo did.24 His idea of proximate cause therefore closely approximates the “direct causation” test,25 under which a consequence is proximate if it follows in an unbroken natural sequence from the defendant's act on existing conditions and forces already in operation, regardless of whether anyone could have foreseen it.26
The majority view runs the other way. Under the risk principle, the test for whether negligence proximately caused a result is essentially the test for whether the conduct was negligent at all, so far as foreseeability limits go.27 Courts cite Palsgraf for almost every scope-of-liability problem because it so effectively contrasts limits based on the scope of the risk and foreseeability of harm with limits based on the directness of the consequences.28
The dispute had older roots. In Ryan v. New York Central R.R. Co., decided by the New York Court of Appeals in 1866, Justice Hunt asked what principle lets A recover and Z fail when sparks thrown by a defendant's negligence start a fire. He put his opinion on the ground that destruction of the building the sparks reached was to have been anticipated the moment the fire was communicated to it, the ordinary and natural result of its being fired.29
In Palsgraf the two opinions disagreed on two points: whether the issue was breach of duty or proximate cause, and whether the issue, whichever it was, should go to the jury.30 The first matters less than the judges supposed. Cardozo said the defendant must have breached a duty of reasonable care toward the plaintiff before proximate cause could even be reached, and that there was no breach here as a matter of law; Andrews said the issue was proximate cause. Under either explanation the basis for the majority's result is the same: Mrs. Palsgraf was an unforeseeable plaintiff as a matter of law.31 The second point mattered more for the case itself.32 Cardozo did not think reasonable people could disagree, because he did not think injury to Mrs. Palsgraf was even remotely foreseeable. Andrews thought it was remotely foreseeable and thought other factors, such as her closeness to the negligence in time and space, were relevant, so he regarded proximate cause as a question for the jury.33
Andrews's closeness factors survive in modern law as evidence of foreseeability, not as free-standing tests. Other things being equal, the closer the defendant's negligence is in time and space to the plaintiff's harm, and the more direct the connection, the more foreseeable the harm was.34 Most courts no longer accept remoteness and directness as independent tests, because harm can be remote and indirect yet highly foreseeable: a defendant that negligently manufactures fireworks in New York and ships them across the country in January can foresee harm in California on July 4th.35 Courts deciding whether to send proximate cause to the jury therefore ask how foreseeable a result a jury could find, treating closeness in time and space and the directness of the harm as relevant but not determinative.36
That is why the plaintiffs forty feet and three blocks from the demolition wall land on opposite sides of the line. Distance matters to the extent it changes what a juror could find about foreseeability. Most courts, and the Restatement, limit negligence liability for physical harm to foreseeable consequences within the scope of the risk and to foreseeable plaintiffs,37 while a minority treat a consequence as proximate if it occurs as a direct result of the tortious act, foreseeable or not.38 Once some harm to the plaintiff is even remotely foreseeable, the analysis shifts to how closely the manner of harm must have been foreseen, which a later section takes up.
Class Membership Suffices
Section 281(b) is satisfied when the actor's conduct is negligent toward the plaintiff or toward “a class of persons within which he is included.”40 The second alternative keeps the foreseeable-plaintiff limit from undoing MacPherson v. Buick Motor Co., which established that a manufacturer owes due care to any foreseeable plaintiff, users, consumers, and bystanders, not just the immediate purchaser, so that the absence of privity is no defense.41 The question is therefore about membership in a group. A plaintiff qualifies by belonging to the set of people the defendant's carelessness put at risk, and the defendant's ability to identify her in advance is irrelevant.
A rule demanding foresight of the individual plaintiff would reinstate the privity barrier under another name. A manufacturer sells through dealers and typically does not learn who will use a particular unit. If the plaintiff had to be identifiable in advance, nearly every remote user would be unforeseeable, and anonymity would do the work the missing contract once did.
Cardozo made the point in MacPherson itself. The contractor who builds a scaffold invites the owner's workmen to use it, and the manufacturer who sells an automobile to a retail dealer invites the dealer's customers to use it. “The invitation is addressed in the one case to determinate persons and in the other to an indeterminate class, but in each case it is equally plain, and in each its consequences must be the same.”42 Whether the persons are named or unnamed does not matter to the duty; what matters is that the sale invites a class of people to use the product.
Two later cases show how courts handle the argument that a plaintiff was outside the intended class. In Hall v. E.I. du Pont De Nemours & Co., the defendants argued that children injured by blasting caps were neither purchasers nor intended users and so were unforeseeable as a matter of law and outside the duty of reasonable care. The court treated that argument as raising a question of duty: although cast in terms of foreseeability, “the issue is finally one of duty — a question of law and public policy,” with foreseeability one of the vectors bearing on it.43 Dillon v. Legg adds that foreseeability does not turn on whether the particular plaintiff would actually have foreseen the exact accident and loss; courts decide case by case what the ordinary person under the circumstances should reasonably have foreseen.44 A driver who strikes a child can expect that the child's mother is somewhere nearby, though the driver knows nothing about her, and the bystander subsection below builds on that move.45
Anonymity cannot separate MacPherson from Palsgraf, because the defendant could not have named either plaintiff in advance. Class membership does. A manufacturer that sells a car to a dealer invites the dealer's customers to use it, so those customers are the class the sale endangers.46 Mrs. Palsgraf stood outside any such class. The package gave no notice of its contents,47 so the employees' conduct put no one at the far end of the platform at apparent risk, and negligence toward the holder of the package was not negligence toward her.48
Exam answers go wrong in two opposite directions. One denies a duty because the defendant could not have identified the plaintiff, an argument that would have decided MacPherson for the manufacturer. The other reads Palsgraf as a rule against anonymous plaintiffs, when she lost because nothing the employees did placed anyone in her position at apparent risk. The working question is whether a reasonable actor, looking at the conduct and the circumstances, would see people like this plaintiff among those the carelessness could hurt.
Other duty questions turn on the same attention to the risk to a particular person. One dissenting justice agreed that a duty to warn should arise when a psychiatrist, in terminating a patient, increases the risk of violence toward a particular individual, but found no support for such a duty based solely on the psychiatrist-patient relationship, and other courts have agreed that the relationship alone does not ground a duty to warn.49
Bystanders and Emotional Distress
Emotional harm complicates the picture of the endangered class. A parent who watches from the curb as a car hits her child is in no danger of being hit, yet the shock can make her physically ill. Does such an onlooker belong to the class the defendant endangered? The full law of negligent infliction of emotional distress has its own treatment; the cases here show only how foreseeability sorts plaintiffs when the harm is emotional.
The respondents in Dillon v. Legg would have characterized the injured child's mother as an “unforeseeable” victim of the driver's negligence, outside the zone of any apparent danger and therefore excluded by Palsgraf, which denied liability against a railroad whose agent, assisting a passenger to board a train, dislodged a package of fireworks that exploded and brought scales many feet away down on the plaintiff.51 The argument has an internal logic. If duty runs only to people in the zone of apparent danger, and danger means the risk of physical impact, a bystander with no risk of impact is a stranger to the defendant's carelessness. The court in Amaya v. Home Ice, Fuel & Supply Co. had denied recovery in part because of the difficulty it anticipated in defining limits on such claims.52
Dillon abandoned the zone-of-danger limit as an arbitrary restriction on recovery.53 It rejected the difficulty of drawing limits as a basis for denying recovery and made foreseeability the primary consideration in finding duty.54 The opinion relied on a treatise to explain how a plaintiff outside the zone of physical risk can still be within the zone of risk: recovery should be had when the defendant should foresee fright or shock severe enough to cause substantial injury in a person normally constituted, and the plaintiff is then within the zone of risk “in very much the same way as are plaintiffs to whom danger is extended by acts of third persons, or forces of nature, or their own responses (where these things are foreseeable).”55 Applied to the mother, the reasoning is that a driver who negligently causes the death of a young child may reasonably expect that the mother will not be far distant and will suffer emotional trauma on witnessing the accident; in Prosser's words, “when a child is endangered, it is not beyond contemplation that its mother will be somewhere in the vicinity, and will suffer serious shock.”56
The court confined its ruling to a plaintiff whose shock resulted in physical injury, and it named three factors that courts will take into account:
- Nearness: whether the plaintiff was located near the scene of the accident, as opposed to a distance away from it.
- Direct perception: whether the shock resulted from sensory and contemporaneous observance of the accident, as opposed to learning of it from others after its occurrence.
- Relationship: whether the plaintiff and the victim were closely related, as opposed to having no relationship or only a distant one.
The nearness and relationship factors track Prosser's observation about a child in danger, and the second separates a witness from someone who hears the news later. The court did not decide what follows when some factors are absent or reduced in weight, and it said later courts would draw lines on facts more subtle than those before it.57 Reasonable foreseeability remains an objective judgment: it does not turn on whether this plaintiff would actually have foreseen the exact accident and loss, and courts decide case by case what an ordinary person in the circumstances should have foreseen, excluding the remote and unexpected.58
The father's claim matches Dillon's own facts and succeeds on every factor. The commuter is far from the scene and has no relationship to the child, and the aunt is related but was neither near the scene nor present when it happened. Each of those missing factors weighs against foreseeability, because the driver has far less reason to anticipate harm to a person in that position. An English decision discussed in Dillon marks the outer edge. In Bourhill v. Young, the court held that a casual bystander's physical injury from shock at witnessing an accident was so unusual and unforeseeable as to warrant a directed verdict for the defendant, since a driver may assume that the ordinary frequenter of the streets has the fortitude to endure the noise of a collision and the sight of injuries to others.59 The aunt's claim is weaker without being necessarily lost, because Dillon left open the effect of absent or reduced factors.
Relying on foreseeability carried a cost, which the California Supreme Court weighed in Thing v. La Chusa. Dillon had made foreseeability the primary consideration while recognizing that policy required restrictions narrowing the class of potential plaintiffs, and the resulting test for limiting liability was itself amorphous.60 Post-Dillon decisions relaxed the guidelines as arbitrary and gave little consideration to the limitless exposure that a pure foreseeability test of duty would create.61 The difficulty lies in how far foreseeability travels. One negligent act that injures a child can foreseeably distress parents, siblings, grandparents, friends, and every onlooker who sees or hears of it, and a test that asks only whether distress was foreseeable gives each of them an argument and gives courts no principled place to stop.
A bystander problem therefore runs in two steps. Palsgraf's idea that the risk defines the duty shows why a plaintiff outside the zone of physical danger can still be a foreseeable plaintiff. Nearness, contemporaneous perception, and closeness of relationship then measure how foreseeable this plaintiff was, and the policy need to keep liability finite explains why courts insist on measuring it.