But-For Cause (a.k.a. Cause in Fact or Actual Cause
Substantial Factor (Multiple Redundant Causes)
The classic case applying alternative liability was Summers v. Tice. In that case two defendants negligently shot in the plaintiff's direction, but the evidence was insufficient to determine which defendant's shot caused the damage to the plaintiff. The court shifted the burden of proof to the defendants to establish which of them was not the responsible party. Courts apply this doctrine of alternative liability where (1) each of the defendants negligently created a risk of harm toward the plaintiff; (2) all of the potential defendants are present in the court room; and (3) the plaintiff's harm probably resulted from one of the defendant's conduct. The effect of the application of this doctrine is to shift the burden of proof to the defendants; if they do not carry the burden of proof, all will be held jointly and severally liable. If one or more defendants persuades the jury that he was not responsible, the liability will shift to the remaining defendants.
Although the more probable than not standard for "but-for" causation applies in most cases, courts sometimes permit exceptions where justice requires it. These exceptions apply in three distinct types of cases: (1) where the actor's identity is unknown; (2) where the plaintiff would probably have been injured anyway, but lost a chance to avoid the injury; and (3) where an accumulation of causes resulted in injury, and a strict application of the but-for test would exonerate all defendants.
But-For Cause (a.k.a. Cause in Fact or Actual Cause)
Tort law ordinarily does not permit a plaintiff to recover damages unless the defendant's conduct was a cause of the plaintiff's injury. The first part of that test is that the injury would not have occurred but for the defendant's negligence. A defendant's behavior (driving while intoxicated, for example) may be quite dangerous and greatly increase the chance of accidents in general. However, unless the plaintiff's injury would have been prevented by someone who was not behaving negligently (e.g., by driving sober), the negligence cannot be said to have caused the injury.
Where two or more defendants act in concert to create a risk of harm toward the plaintiff, each will be liable for the plaintiff's injury even if it cannot be determined which of the defendants actually caused the injurious consequence. For example, if two teenagers engage in a drag race and one of the vehicles strikes a pedestrian, both may be held liable on the theory that each is an agent of the other. Where the defendants do act in concert, they become jointly and severally liable for the injuries that each inflicts upon the plaintiff.
Occasionally the but-for test would work an injustice where more than one defendant has injured the plaintiff, and the injuries accumulate in such a way that collectively they cause serious injury, but no one injury by itself made a significant different. For example, in the asbestos cases an asbestos worker typically inhaled asbestos from as many as twenty different suppliers. It can not be said more probably than not that but for the conduct of any one of the them the injury would be avoided. However, if the but-for test were applied individually, all the defendants would escape liability. Therefore, the courts carve out a different rule: the substantial factor test.
Enterprise liability can be applied where the members of an industry or enterprise (such as the blasting manufacturers in Hall v. DuPont) control the standards setting process which could have mandated (but negligently failed to mandate) safety measures to prevent the injury to the plaintiff. This differs from concert of action in that none of the defendants is an agent of the other, and it differs from alternative liability in that not all of the defendants need be present in the court room. However, it does require a showing that the defendants had it within their power as a collective entity to take the safety steps that the plaintiff claims they negligently failed to do. Note also that enterprise liability is a means of establishing causation; it does not eliminate the need for the plaintiff to show that the defendants acted negligently in failing to adopt safety measures.
In addition proving that the defendant's negligence was a but-for cause of his injury, the plaintiff must also show that it was a legal cause. Whereas the but-for test revolves around the question of probabilities, the legal cause test revolves around questions of policy--whether it is fair in a given circumstance to require the defendant to pay for the plaintiff's injury. A defendant's negligence is not a legal cause of the plaintiff's injury if any of three conditions is present: (1) the defendant did not increase the chance of the plaintiff being injured; (2) the defendant's negligence was too remote from the injury because of a superseding cause; or (3) the chain of events resulting in the plaintiff's injury made the plaintiff an unforeseeable victim of the defendant's negligence.
In the ordinary case the plaintiff is required to show that more probably than not the injury would not have occurred but for the defendant's negligence. In some cases the defendant clearly acted negligently, but the evidence may indicate that the plaintiff more probably than not would have been injured anyway. Under a traditional application of the burden of proof, the plaintiff would lose such a case. However, in a limited number of circumstances an exception may be made based upon "loss of a chance."
If the jurisdiction recognizes the loss of a chance theory, it will permit a plaintiff to recover the value of a chance that is lost when a defendant acts negligently, the plaintiff is injured, but the evidence shows that the plaintiff more probably than not would have been injured anyway. This theory is used most often in medical malpractice cases, on the theory that, in the absence of such a rule, defendants would be immunized from liability for cases where the plaintiff's chance of recovery was less than 50% when the procedure began.
This is essentially a limited version of alternative liability. Where the plaintiff is injured by a generic product manufactured by several of the defendants, and the specific injury-causing agent cannot be identified, a plaintiff may be able to recover from the defendants on the basis of their market share of the market for the generic product. This still requires a showing that the defendant acted negligently in marketing the product, and limits the defendant's liability to the percentage of the plaintiff's damages corresponding to the defendant's percentage share of the product market.
The standard used for "but-for" causation is whether or not the accident more probably than not would have been prevented through the exercise of reasonable care. This standard does not require certainty that the accident would have been avoided in the absence of negligence, but rather a standard similar to the burden of proof used elsewhere in tort law: a preponderance of the evidence. A plaintiff must produce evidence (rather than mere conjecture) that the accident would not have occurred but for the defendant's negligence; however, all analyses of but-for causation inevitably involve comparing a hypothetical world in which the defendant did not act negligently with the real world in which the plaintiff was injured. Unless the jury finds by a preponderance of the evidence that in that hypothetical world the plaintiff would not have been injured, the court ordinarily will deny recovery. However, note the exceptions to this rule below.
In order to recover, the plaintiff must establish that the defendant's negligence (or other breach of duty) was a proximate cause of the plaintiff's injury. This means that both but-for cause AND legal cause can be established.
Substantial Factor (Multiple Redundant Causes)
Occasionally the but-for test would work an injustice where more than one defendant has injured the plaintiff, and the injuries accumulate in such a way that collectively they cause serious injury, but no one injury by itself made a significant different. For example, in the asbestos cases an asbestos worker typically inhaled asbestos from as many as twenty different suppliers. It cannot be said more probably than not that but for the conduct of any one of the them the injury would be avoided. However, if the but-for test were applied individually, all the defendants would escape liability. Therefore, the courts carve out a different rule: the substantial factor test.
In the kinds of cases where the defendants' collective behavior results in the plaintiff's injury, but no one can be said to be a but-for cause of the injury, the courts will ask in place of the but-for test whether the defendant's conduct was a "substantial factor" in causing the injury.
In order to be a legal cause of the plaintiff's injury, the defendant's negligence must be connected to the injury by a "direct and unbroken sequence." While the direct sequence may include intervening causes, the causal chain is broken if the intervening force is so different from the defendant's original act of negligence so as to characterized as a superseding cause. Courts have difficulty providing reliable criteria by which to distinguish "merely" intervening causes (which will not destroy the causal chain) from superseding causes (which do break the causal chain). Like other questions of legal cause, this goes to the issue of whether it is fair to hold a defendant liable for consequences that may have been largely due to the intentional, even criminal, acts of a third party. Here’s a visualization of the concept:
Sometimes a plaintiff is injured by a negligent party, but the evidence does not permit a determination of which of several parties caused the injury. Courts have used four different doctrines to deal with cases of known negligence but unknown identity of the responsible defendant.
In the classic case of Mrs. Palsgraf, Justice Cardozo held that a railroad could not be found liable for injuries to a person who was so far removed from the defendant's negligent act that an injury to her was unforeseeable. Cardozo held that a defendant's duty to reasonable care extends only so far as a reasonable person would expect the consequences of his negligent act to reach. To people beyond that "zone of danger" the defendant could anticipate no injury, and therefore would not be under an obligation to use reasonable care. Andrews, dissenting in that case, thought the issue was less a matter of whether the defendant owed a duty of care, and more a question of whether the defendant's negligent act was a proximate cause of the injury. Andrews thought that the question of proximate cause was less a matter of legal doctrine and more a pragmatic weighing of policy concerns and issues of fairness. Most jurisdictions have adopted an approach closer to Andrews' reasoning than Cardozo's. Here’s a visualization of the concept:
