Concept
Judge and Jury Functions
Concurrent causes and the substantial factor test
Judge and Jury Functions
Restatement (Second) of Torts § 434 splits two questions, causation and apportionment, between judge and jury, and it draws the line with one phrase: whether the jury may reasonably differ. When the evidence leaves room for reasonable people to come out either way, the jury decides whether the defendant's conduct was a substantial factor in the harm and how the harm divides among two or more causes. When the evidence leaves no such room, the court decides both questions itself. The two subsections of § 434 are mirror images and should be read together.1
The test is a judgment about the evidence, not about the verdict the judge would prefer. A judge who asks whether reasonable people could differ stays inside the gate. A judge who asks who ought to win has stepped into the jury box. The court acts first, so its role comes first below, and the jury's role follows.
The Court's Gatekeeping Role
Section 434(1) gives the court three tasks. It decides whether the evidence makes an issue on which the jury may reasonably differ about whether the defendant's conduct was a substantial factor in causing the harm. It decides whether the harm is capable of apportionment among two or more causes. And it decides causation and apportionment itself in any case in which the jury may not reasonably differ.2 The first two tasks are screens that run before the jury deliberates. The third is what the court does when a screen shows that only one answer is open.
Before the trier of fact passes on liability, the court must determine whether a jury would be justified in finding a negligent act and a causal connection between that act and the injury. It does so by asking whether the facts in evidence support any reasonable inference of a causal relation between the negligence and the injury, in whole or in part.3 Two verbs mark the division of labor. It is for the judge to say whether the jury can reasonably find the causal relation, and, if the judge decides that it can, for the jurors to say whether they do. The judge asks whether the causal relation may be reasonably inferred from the evidence, and the jurors ask whether it is inferred.4
The screen has this particular shape for two reasons. It exists because the plaintiff must prove causation: this power of the judge, properly exercised, materially lessens the danger of an unjust result as to causation.5 It is limited because causation is a fact the jury finds. A judge who asked whether the evidence persuades the judge personally would be doing the jury's work, so the judge asks only whether a reasonable finding is possible. Where witnesses contradict one another, or the evidence supports reasonable inferences in either direction, a reasonable finding is possible, and the choice among those inferences belongs to the jury.
The inquiry runs injury by injury. The extent of an injury is reached only if there is a causal connection, in whole or in part, between the negligence and the injury, so the court asks the causation question about each injury, or part of an injury, for which the plaintiff seeks damages.6 The second screen carries the same logic into damages. Whether the harm is capable of apportionment among causes is the question that separates divisible from indivisible harm, which the preceding section explains.7
Perkins v. Texas & New Orleans Railroad Co. shows why a finding of negligence does not settle causation. The railroad's own safety rules set a speed limit of 25 miles per hour in town, the plaintiff conceded that this was a safe speed at the crossing, and the train was traveling at 37.8 The railroad's violation of its own safety rule was evidence of negligence, and the court found the trainmen negligent. Even so, it described the prime issue as whether the excessive speed was a cause in fact of the fatal collision.9 Breach was established, and cause in fact remained a separate question that the record had to answer.
The warehouse and the railroad crossing both involve a negligence claim and a causation problem, but they sit at opposite settings of the gate. In the warehouse, each side's version, if believed, supports a definite conclusion about causation, and the jury chooses between them. A record like the one Perkins describes, where the speed violation is established but nothing yet ties the extra speed to the collision, calls for the third task. If the evidence supports no reasonable inference that the negligence made the difference, the jury may not reasonably differ, and the court decides causation itself.12 That task cuts both ways, since the court may also decide causation for the plaintiff when the evidence admits only that answer. On any causation dispute, ask whether a reasonable finding is possible either way. If it is, the case goes to the jury. If it is not, the court resolves the question.
The Jury's Role
Section 434(2) gives the jury two tasks and conditions both on the phrase that anchored the court's role. In any case in which the jury may reasonably differ on the issue, the jury decides whether the defendant's conduct has been a substantial factor in causing the harm to the plaintiff, and it decides the apportionment of the harm to two or more causes.14 The jury's role is the other half of the gate the court operates. The court asks whether reasonable minds can differ, and where they can, the jury answers the questions the court has left it.
The jury receives the substantial-factor question because it is a question of degree. The factfinder must make an intuitive judgment both about what degree of causation is substantial and about whether the defendant's negligence reaches that level.15 An illustration quoted in the cases shows the judgment at work. When a child drowns in a swimming pool, no one can say with absolute certainty that a lifeguard would have saved the child. Yet the common experience of the community permits the conclusion that the guard would more probably than not have done so, which means the guard's absence played a substantial part in the death. Such questions, the passage concludes, are normally for the jury, and the court may seldom rule on them as matters of law.16
Once the evidence clears the threshold, submission is mandatory. Where the issue is not so plain that reasonable people could not find the causal relation and award damages for the injury, or for part of it, the court must leave to the jury the question whether the defendant's conduct was a substantial factor in producing the injury or part of the injury.17 A judge who finds the plaintiff's causation evidence thin, or the defense witnesses more convincing, may not withdraw the question while a reasonable finding remains possible. Choosing among reasonable inferences is the jury's work.
The missing guard is the same in the hotel and in the quoted drowning illustration, so the negligence cannot be what separates them. What separates them is whether the evidence leaves a reasonable finding open. With the undisputed cardiac finding, the only reasonable answer is that a guard would have made no difference. Without it, and with experts reasonably disagreeing about the timing, the question is the one the illustration describes, and it belongs to the jury.
Apportionment, the second jury task, follows the same logic. Whether a harm is capable of apportionment among causes is first a gate question for the court under § 434(1)(b).19 If the harm can be divided and the evidence leaves room for reasonable disagreement about the division, the jury decides how the harm is apportioned among the causes.20 If the court determines that the harm is not capable of apportionment, there is nothing for the jury to divide, and the preceding section's treatment of indivisible harm governs.