Twin Fires, Inextricable Tangles and Seeking Clarity: The Massachusetts Supreme Judicial Court Poised To Address Actual Causation in Potentially Significant Decision

In what may well be a significant decision either redefining or clarifying actual causation in negligence including medical malpractice actions, the Supreme Judicial Court heard argument on October 5, 2020 and took the issue and case under advisement. The case, Doull v. Foster, is a wrongful death medical malpractice action tried to a defense verdict after three weeks of trial and which involved allegations of lack of informed consent and negligence. The decedent died of an underlying clotting disorder, Chronic Thromboembolic Pulmonary Hypertension (“CTEPH”). The allegations included that the clotting disorder was caused by progesterone prescription; that there was a failure to inform of the risk of clotting; and that there was a failure to diagnose and refer for treatment in a more timely manner resulting in the failure to prevent a subsequent stroke and death. The defense was that the progesterone cream did not carry any risk of clotting; was not the cause of the disorder; that there was no failure to diagnose; and that any such earlier intervention at that time would not have prevented the subsequent death.

The jury found the two primary care health provider defendants negligent but further found that their negligence was not the cause of the stroke or death. The trial court opted not to instruct the jury with “substantial factor” opting to utilize “but for” (i.e., “without which the harm would not have occurred”). It determined that the “substantial factor” test was not applicable with the court otherwise influenced by the Third Restatement’s approach. The SJC took the case for direct appellate review on its own accord identifying its specific interest in whether Massachusetts should adopt the American Law Institute’s Third Restatement’s approach to actual causation.

Amicus curie briefing was submitted by the Massachusetts Academy of Trial Attorneys (MATA) as well as the Massachusetts Defense Lawyers Association (“MDLA”). MATA took the position that the substantial factor approach was preferable while the MDLA argued that the Third Restatement provided more clarity and should be adopted.

Twin Fires And Moral Culpability

The case is significant because causation has remained a difficult area particularly in conveying the requisite obligation to a lay jury in cases in which there are multiple defendants and/or multiple possible or sufficient causes of the harm. There remains uncertainty and a measure of confusion as to the role and use of “substantial factor,” whether it replaces or supplements the traditional “but for” test, and whether it provides helpful guidance or should not be used at all due to concerns it is too evaluative either heightening or lowering the burden of proof. See Weigand, The Wrongful Demise of “But For” Causation, 41 W. N.E. L. Rev. 75 (2019)

There is nearly universal agreement that the “but for” test for actual causation is the primary test and works well in most situations. It is the bare minimum necessary to impose legal responsibility with it observed “that to abandon the “but for” test is to abandon the element of causation.” However, over time efforts to modify or provide exception to the “but for” test arose to address certain instances where the “but for” test seemed to lead to an unjust result most notably the multiple sufficient cause (also known as “independent concurrent cause” or “overdetermined harm”) circumstance. The classic example is the twin fires scenario where two fires, each itself sufficient to have caused the harm merged and proceeded to concurrently cause the harm. See Anderson v. Minneapolis St. Paul & Sault Ste. Marie Ry. Co., 179 N.W. 45 (Minn. 1920). Since each was sufficient neither would be held responsible under a literal application of “but for.” The concern reduces to the moral culpability of actors whom the technical application of the “but for” test would absolve.

The “substantial factor” test thus emerged and was utilized as either as a supplement to the “but-for” test (to apply to the multiple sufficient, concurrent cause, or overdetermined-causation cases or as a complete substitute (i.e. cause must be substantial and without which it would not have occurred). Indeed, the Second Restatement (R2) which was published in 1965 defined causation in terms of “legal cause” and “substantial factor.” In order to be a legal cause, the conduct had to be both substantial and where the harm would not have otherwise occurred (i.e. “but for”). While this approach preserves the primacy of “but for” it was subsumed within how R2 defined “substantial factor.” R2 likewise adopted the multiple sufficient cause exception to “but for” stating that if there are multiple sufficient causes the defendant’s conduct would be a “substantial factor” if the defendant’s conduct alone would have been sufficient to cause the harm irrespective of any other sufficient cause. The result under R2 is different uses of substantial factor: one where it is defined as requiring “but for” and the other where it does not.

“The Test of Time”

In 2010, the Third Restatement (R3) arrived taking the view that the use of “substantial factor” was problematic proclaiming that “it has failed the test of time” and should not be used in determining causation and proceeded to purge the language from causation altogether. R3 takes the view that “substantial factor” improperly rendered the actual cause inquiry evaluative and that cause in fact was not one of degree. It was concerned the use of “substantial factor” has led to misuse and the potential lessening or heightening of the applicable burden of proof. As such, the Third Restatement reasserts the primacy of “but for” otherwise confirming that there can be more than one “but for” cause in any case. It also, as in the Second Restatement, recognizes the multiple sufficient cause exception providing that the operative test in such circumstances is whether the particular defendant’s conduct would have caused the harm absent the other potential causes. If so, it is a factual cause. R3 proceeds to add the notion of “sufficient causal sets” which renders a defendant’s conduct a factual cause of the harm if it is a necessary component of an otherwise sufficient causal set. Under this rule, a defendant’s conduct would be a factual cause of the harm even if the defendant’s conduct would not be sufficient alone to cause the harm but when combined with other causes is sufficient; all such causes of this set would be a factual cause of the harm.

R2 and R3: Not So Different and The Primacy of But For

Notably, the approach under R2 and R3 as to factual causation are largely the same with the exception that R3 does not define factual causation utilizing “substantial factor” and otherwise adds the necessary to a sufficient causal set concept. Under R2, factual cause exists where: (a) the conduct is a “but for” cause of the harm and is “substantial” or (b) the conduct is one of multiple causes operating concurrently and each is individually and by itself sufficient to have caused the harm. Under R3, factual cause exists where: (a) the conduct is a “but for” cause of the harm; (b) is one of two or more sufficient causes which would be a “but for” cause in the absence of any other cause; or (c) the conduct is necessary to a sufficient causal set. R3 would otherwise exclude, under the scope of liability constituent, an otherwise factual cause if “trivial.”

Massachusetts has largely followed R2’s approach and regularly uses substantial factor as the operative test. It has recognized the multiple sufficient or independent current cause exception as well as an inextricable tangle exception best exemplified in toxic tort situations like asbestos where it is usually impossible to say whether any contribution of any of the multiple defendants caused the harm but it is otherwise known that some combination resulted in the harm. See O’Conner v. Raymark Indus., Inc., 518 N.E. 2d 510 (Mass. 1988). It both instances “substantial factor” in lieu of “but for” is used.

The Confusion And Discreet Exception Swallowing the Rule

The difficulty with the substantial factor approach is that it can lead to misunderstanding and misuse particularly where “but for” is left out in cases where it applies. While R2 defines legal causation utilizing “substantial factor,” the definition includes “but for.” Further, under R2, once conduct is deemed to be a “but for” cause or a sufficient cause, it still must be found to be substantial under certain identified conditions set out in section 433 of R2 to constitute a “substantial factor.” This breeds confusion in the form of litigants or courts relying on these considerations without reference to the need for showing the conduct to be a “but for” cause. Adding to the milieu is the fact that R2 provides in its comments that conduct is a substantial factor where it is of sufficient significance in producing the harm as to lead reasonable persons to regard it as a cause. This offering again leaves out any reference that conduct must still be first a “but for” cause. Neither the comments nor section 433 were ever intended to eliminate or by-pass the need to show “but for” but only to inform as to what is substantial once either “but for” or the “sufficient” concurrent cause exception has been found. Finally, as to both R2 and R3 and the sufficient concurrent cause exception, the use of the term “sufficient” in both R2 and R3 does not mean that either provision would impose liability for conduct that is not a “but for” cause if only the conduct could have caused the injury but both require that the conduct (or the causal set in which it is part) would have caused the harm.

At bottom, the multiple sufficient cause/ independent concurrent cause as well as the inextricable tangle exceptions where “substantial factor” is used in lieu of “but for” and without any reference to “but for” are very limited and not triggered just because there are multiple defendants or causes. Nonetheless, courts and litigants have, at times, failed to appreciate the distinction and request or use “substantial factor” as the operative test without reference or use of “but for” anytime there are multiple defendants or causes. This is a misuse and a misunderstanding of the origins of substantial factor as well as the failure to appreciate that the multiple sufficient cause and inextricable tangle scenarios are discreet, limited circumstances where causation cannot be resolved using “but for.” Further, under the view of R3, substantial factor creates the potential to lessen or heighten the burden of proof and otherwise requires subjective evaluation where cause in fact is not one of degree.

In Doull, it was argued that substantial factor instead of “but for” should have been used because there were two defendants and because there were multiple possible or competing causes of the clotting disorder and/or the stroke and death (i.e. failure to diagnose and refer and the nature of the CTEPH). The defense contended that this was incorrect and that the facts did not present either the multiple concurrent cause or inextricable tangle exception and thus “but for” was required.

As to the multiple defendants, the trial court held that there were no multiple concurrent causes or independent concurrent causes triggering the substantial factor exception because the claim against one of the two defendants was supervisory. As such, there remained only one causal conduct as to the harm, i.e., the underlying provider (Nurse Practitioner). Absent her negligence and causal finding as to her, the lack of supervision could not be an independent cause. Indeed, as provided by R3 there can be more than one “but for” cause with the obligation to make clear that each defendant is assessed separately and independently and without regard to any finding as to any other defendant. A finding that one defendant is not a “but for” cause has no bearing on any other defendant whose causal liability rises or falls on their own “but for” test.

As to the alleged multiple causes of the clotting disorder (the progesterone cream or natural progression of CTEPH) or the competing cause of the stroke and death (defendants’ failure to more timely diagnose or the nature of the CTEPH disorder), the claim remained one of a failure to diagnose and refer earlier. The gravamen of the causal issue was whether such earlier intervention would have made any difference. There was significant and persuasive evidence at trial that it would not with the plaintiff’s own expert conceding he could only say that earlier referral would have “possibly” prevented the CTEPH and death. Regardless, the fundamental question was whether earlier diagnosis would have made any difference. The fact that the decedent had an underlying disease and that the causal dispute was between the effect of the alleged delay and the nature of CTEPH did not render the “but for” test inapplicable or otherwise lead to any misconception that there could not be more than one cause of the harm. The operative causal inquiry was whether earlier intervention would have prevented the death. This is neither the multiple sufficient cause/independent concurrent cause nor inextricable tangle exception where deviation from “but for” with substantial factor is believed to avoid misconception. “But for” works perfectly well with no such misconception.

“More important than the quest for certainty is the quest for clarity”

Doull squarely provides the Supreme Judicial Court with the opportunity to clarify actual causation. Whether it will opt to adopt the R3 view and thereby banish substantial factor from the factual causation element remains to be seen. Only two states have so far adopted R3’s approach (Iowa and Virginia) with only one known state (Delaware) expressly refusing to adopt R3 although that decision was as to the duty formulation and not causation. By incorporating the necessary to a sufficient causal set approach, R3 arguably expands liability as it would render insignificant and insufficient causal contributions a factual cause so long as part of a causal set (e.g., inextricable tangle-asbestos) while under R2 such conduct would not be a cause because it is insufficient to have caused the harm. R3 would proceed to exonerate such a contribution if otherwise determined to be trivial compared to the other contributing causes but does so as a matter of policy in the scope of liability constituent of causation and not a matter of factual cause.

If R3 is adopted it would mark a significant change and require acclimation to the nomenclature used to present the causation question to the jury. See Weigand, Duty, Causation and Palsgraf: Massachusetts and the Restatement (Third) of Torts, 96 Mass. L. Rev. 55 (2015). R3 arguably provides a more precise approach to causation including providing a clearer demarcation between the factual and policy constituents of causation. R3 is otherwise adamant that the use of substantial factor is concerning and leads to misuse and misunderstanding including the potential to impermissibly heighten or lessen the more probable than not burden of proof. A comparison of a jury instruction under the substantial factor approach and R3 illuminates the difference between the two approaches as to operative nomenclature:


Factual Cause

Substantial Factor

(1) If you decide that the defendant was negligent, you must then consider whether the defendant’s negligent conduct [caused/enhanced] the plaintiff’s injuries. Even if you find that the defendant was negligent, the defendant is not liable to the plaintiff unless [its/his/her] negligence caused the plaintiff’s harm. To meet [his/her] burden, the plaintiff need only show that there was greater likelihood or probability that the harm complained of was due to causes for which the defendant was responsible than from any other cause. The defendant’s conduct was a cause of the plaintiff’s harm if the harm would not have occurred absent the defendant’s negligence. In other words, if the harm would not have happened anyway, the defendant is not


(1) Tortious conduct must be a factual cause of harm for liability to be imposed. Conduct is a factual cause of harm when the harm would not have occurred absent the conduct. 16 One must ask what would have occurred if the actor had not engaged in the tortious conduct. The framework is two steps: identify the relevant harm for which recovery is sought and then identify the tortious conduct of the actor and ask whether the harm would have occurred if the actor had not acted.

Substantial Factor:

(2) (If multiple sufficient causes/inextricable tangle)



(2) (If multiple sufficient causes)

Substantial Factor:

(3) There may be more than one cause present to produce an injury, and more than one person legally responsible for an injury. The plaintiff does not have to prove that the defendant’s negligence was the only or even the predominant cause of the injury. If two or more causes operating together contributed to the plaintiff’s injury so that, in effect, the damages suffered were inseparable, then it is enough for the plaintiff to prove that the defendant’s negligence was a substantial contributing factor in causing the injury. By “substantial” I mean that the defendant’s contribution to the harmful result, i.e., the defendant’s negligence, was not an insignificant factor.

The defendant’s negligence must contribute along with other factors to the result; it must be a material and important ingredient in causing the harm. If the defendant’s negligence was a substantial factor, then it is considered a cause of the plaintiff’s injury, and the plaintiff is entitled to recover. If it was not a substantial factor, if the negligence was only slight or tangential to causing the harm, then even though you may have found the defendant negligent, [it/he/she] cannot be held liable to pay damages to the plaintiff on this claim.



(3) An actor’s tortious conduct need only be a factual cause of the other’s harm. The existence of other causes of the harm does not affect whether specified tortious conduct was a necessary condition for the harm to occur. Those other causes may be innocent or tortious, known or unknown, influenced by the tortious conduct or independent of it, but so long as the harm would not have occurred absent the tortious conduct, the tortious conduct is a factual cause. 18 If there are multiple potential causes and each alone would have been a but for cause of the harm, each act is a factual cause.

(If inextricable tangle case)

An actor’s tortious conduct need only be a factual cause of the other’s harm. The fact that an actor’s conduct requires other conduct to be sufficient to have caused another’s harm does not mean the conduct cannot be a factual cause of the harm. Even if there are other causes of the harm, a defendant’s conduct is a factual cause so long as it was a necessary part of a causal set which set was sufficient to cause the harm. Stated otherwise, even if the defendant’s conduct was not sufficient by itself to have the caused the harm it remains a factual cause so long as and only if it was a necessary part of a set of other causes which set was sufficient in the aggregate to have caused the harm. (note: if such conduct is trivial or negligible compared to the other causes in the set, there can be no liability which would be an instruction under “scope of liability”).

Even absent adoption of R3, the Supreme Judicial Court has the opportunity to reaffirm the primacy of “but for” as well as clarify that the use of substantial factor in lieu of “but for” is limited to very confined circumstances. It can be made clear that the multiple sufficient cause/independent concurrent cause and inextricable tangle exceptions are limited and discreet circumstances and that there is no justification to discard “but for” (without which it would not have occurred) anytime there are multiple causes or defendants. Causation in such circumstances would be defined as to require the conduct to be both substantial and without which it would not have occurred. Further, as the Doull case involved a claim and allegation of medical negligence and particularly a failure to more timely diagnose, the Court will have the opportunity to declare that but for (individually or subsumed within the instruction as to substantial factor) applies to such circumstances—commonplace in medical malpractice actions. At bottom, the Court is poised to provide needed clarity in an area that has long confounded and confused.