I move that §§ 26 and 27 of the Restatement Third of Torts: Liability for Physical Harm, be revised as follows:

As currently worded, §§ 26 and 27 would erroneously fail to treat many actual causes as factual causes. For example, A inflicts a wound on C that by itself would cause C to bleed to death by 2:30 p.m. Previously, simultaneously, or subsequently, B inflicts a wound on C that by itself would cause C to bleed to death by 2:15 p.m. The combined wounds cause B to bleed to death at 2:00 p.m. Neither A’s nor B’s act would be treated as a factual cause under § 26 as currently worded, since it is not true that "the harm [C’s death] would not have occurred absent [A’s] conduct" or that "the harm [C’s death] would not have occurred absent [B]’s conduct." Nor would either act be treated as a factual cause under § 27 as currently worded, since it is not true that either act alone (in the absence of the other act) would have been a but-for cause of C’s death "at the same time." This is true whether the ambiguous phrase "at the same time" is interpreted to mean "at the same time that the harm actually occurred" or "at the same time as the harm would have been caused by the other act alone."

Proper treatment of overdetermined harm situations like this, involving multiple duplicative factual causes, can be obtained by inserting the words "when it did" in § 26, as proposed. But-for either A’s or B’s act, C would not have died when she did (2:00 p.m.).

As an added bonus, the additional language would allow § 26 to handle properly many (but not all) instances of the types of overdetermined harm situations discussed in Comments f and i to § 27, in which either (1) one of the causal conditions was sufficient by itself to cause the harm and a second causal condition was sufficient only in conjunction with the first causal condition or (2) none of three or more causal conditions was sufficient by itself but any two in combination were sufficient. For example, in our wounding hypothetical, assume that the wound inflicted by A was sufficient by itself to cause C to bleed to death, and that the wound inflicted by B was not sufficient by itself but caused additional bleeding that caused C to die earlier than C would have died if C had been wounded only by A. Similarly, assume that three different people inflicted separate wounds, none of which was either necessary or sufficient by itself to cause C’s bleeding death, but any two of which combined were sufficient, and that the three wounds combined caused C to bleed to death quicker than would have been the case if there had been only two wounds. In each situation, §§ 26 and 27 as currently worded would incorrectly deny that any of the wounds was a factual cause, while the proposed amendment to § 26 would correctly result in treating each as factual cause. These types of situations are much more common than is usually understood.

As an even further bonus, the additional language would allow § 26 to handle properly many situations involving preemptive causes. For example, if C was killed at 2:00 p.m. by A’s act, and B subsequently at 2:15 p.m. inflicted a wound on C’s body that would have been lethal if C were alive, the amended § 26 will correctly treat A’s act but not B’s act as a but-for cause of C’s death at 2:00 p.m.

The proposed amendment to § 26 greatly narrows the overdetermined harm situations that need to be addressed in § 27, which now need only address overdetermined harm situations in which the harm would have occurred at the same time as it did as a result of different actual or potential causes operating by themselves or in some combination.

Unfortunately, § 27 as currently worded fails to address and properly resolve the basic issue in this type of situation: whether each potential cause was indeed a (duplicative) factual cause or, instead, one was a (preemptive) factual cause while the other was not a factual cause but rather was a preempted condition — a condition that would have been a factual cause in the absence of the other condition but the causal effect of which was cut off or preempted by the other (actually causal) condition. (Indeed, the phrases "multiple sufficient causes," "causal sets," and "cause" are misleadingly used throughout the Comments to §§ 26 and 27 to refer to noncausal preempted conditions as well as actual causal conditions.) For example, assume that A puts a lethal dose of poison in C’s coffee that C is about to drink, but before C drinks the coffee B lethally wounds C. C dies at 2:00 p.m. as a result of the lethal wound, but would have died at the same time as a result of drinking the poisoned coffee. Section 27 as currently worded would erroneously treat A’s putting the poison in the coffee, as well as B’s wounding C, as a factual cause of C’s death.

Moreover, § 27 as currently worded also fails to properly resolve the types of duplicative causation situations addressed in Comments f and i, which are discussed above, in which a certain condition was not (or could not be proven to be) a but-for or independently sufficient cause of the harm and did not speed up the occurrence of the harm but nevertheless contributed to the harm. For example, assume three different persons each put one-half of a lethal dose of poison in C’s coffee, or that one person put in a lethal dose and a second person put in half a lethal dose. C drank all the coffee and died, but the time of death was not speeded up (or could not be proven to have been speeded up) by the presence of the extra half dose of poison. In each situation, those who put in half-doses of the poison erroneously would not be treated as factual causes by § 27 (or § 26). These sorts of situations are very common, and include such well-known cases as the Anderson multiple-fires case and the Corey multiple-cycles case, in which the courts found actual causation without requiring proof that either fire or cycle was a but-for or independently sufficient cause.

The proposed revision of § 27 would properly resolve these types of situations.