by Jessica Christy1
Plaintiffs who allege injury from a defective product warning face unique hurdles in proving causation. While other types of product defect claims focus on the functioning of the product in question, failure-to-warn claims raise questions about a plaintiff’s own, hypothetical conduct: how would the plaintiff have used the product if an adequate warning had been provided? The plaintiff’s traditional burdens of production and persuasion would bar recovery in many failure-to-warn cases, since the only such evidence available is often the plaintiff’s own “speculative and self-serving” testimony that they would have read and acted in accordance with an adequate warning.2
To deal with this issue, many courts have recognized the “heeding presumption,” a rebuttable presumption of causation in strict liability warning defect claims. Under this presumption, if the plaintiff can prove that the warning was defective, then the burden shifts to the defendant to prove that the defect did not cause the plaintiff’s injury. Proponents argue that the presumption is necessary to enforce manufacturers’ duty to warn because it provides for the compensation of injured plaintiffs and incentivizes manufacturers to present reasonable safety information.3 On the other hand, some courts and scholars argue that the presumption has harmful unintended consequences because it 1) undermines consumer safety by permitting manufacturers to provide excessive warnings in lieu of safe product design;4 and 2) imposes excessive liability on manufacturers.5 Both sides of the debate have compelling policy reasons for either embracing or rejecting the presumption of causation.
This Contribution argues that the heeding presumption is an appropriate means for courts to enforce strict liability duty to warn. The presumption creates proper incentives for manufacturers to reasonably warn and protect consumers, and the policy rationales for rejecting it incorrectly conflate unrelated elements of a strict products liability case.
The heeding presumption has its origins in the Restatement (Second) of Torts section 402A comment j: “Where warning is given, the seller may reasonably assume that it will be read and heeded; and a product bearing such a warning, which is safe for use if it is followed, is not in defective condition, nor is it unreasonably dangerous.”6 A significant number of courts have derived from this comment the corollary rebuttable presumption that where a warning is not given, the factfinder may reasonably assume that it would have been read and heeded had it been provided.7 Courts give two related rationales for the presumption’s application. First, the presumption makes manufacturers’ duty to warn enforceable by relieving plaintiffs of their potentially “impossible burden” of convincing jurors that an adequate warning would have prevented their injury.8 Second, it counteracts manufacturers’ economic incentives to conceal the potential hazards of their products.9 For courts that recognize the presumption, these policy concerns combined with the language of comment j justify shifting the traditional burdens of production and persuasion on the element of causation.
Other courts and scholars fear that the presumption will undermine consumer welfare by permitting manufacturers to substitute warnings for safe product design,10 and by forcing responsible product users to subsidize careless ones by imposing near-absolute liability on manufacturers, who would then pass on the increased costs to consumers as a whole.11 However, both of these concerns misconstrue the Restatement (Second) of Torts. The former concern confuses the duty to design reasonably safe products with the duty to provide reasonable warnings, while the latter concern confuses the question of defect with the question of causation. The proper response to these fears is not to do away with the heeding presumption, but to treat each element of a warning defect claim with appropriate analytical rigor.
When a legislature chooses to recognize a duty, it intends for that duty to be enforced. However, under a plaintiff’s standard burden of proving causation, individuals who were injured by products with legally defective warnings may be unable to recover, since the only proof they can offer may be their own testimony about how they would have responded to a reasonable warning. This kind of testimony is inherently unreliable: any injured consumer suing a manufacturer for failure to warn will naturally claim that they would have read applicable warnings and heeded them had they been provided, and so it is exceedingly difficult for jurors to determine when such testimony should be trusted.12 If jurors faithfully apply the preponderance of the evidence standard, then they are unlikely to conclude that the injury would not have occurred if the warning had been in place in all but the most egregious of cases. This will leave injured plaintiffs severely undercompensated, in contravention of a legislature’s apparent will when it decided to impose a strict liability duty to warn on manufacturers. The heeding presumption neatly solves this issue by directing that the impossibility of proving a hypothetical should weigh in the plaintiff’s favor. Unless the defendant can present evidence that the plaintiff would not have read or followed an adequate warning, the court presumes that the warning defect caused the improper use behind the injury.13
Holding manufacturers liable for placing products with defective warnings on the market does more than just compensate injured plaintiffs; it also helps protect the public as a whole. Without an enforceable duty to warn, sellers may be tempted to under-inform buyers about the dangers their goods pose, since consumers are likely to find obviously hazardous products less desirable.14 Making manufacturers liable for absent or ineffective warnings ensures that they will fully instruct the public about their products’ risks. This, in turn, enables consumers to make informed safety decisions about if and how they use potentially hazardous goods.15 Admittedly, a significant number of consumers fail to read and/or follow safety information16—but that is in itself a safety decision. Consumers cannot make the choice to either heed or ignore a product’s warnings if no warnings are presented.
In spite of the above consumer safety rationales, some courts and scholars have rejected the heeding presumption on the grounds that it has inadvertent consequences that leave consumers less safe. For example, Professor Howard Latin argues that comment j and any doctrines derived from it should be disavowed because its plain language arguably allows manufacturers to substitute cheap additional warnings for expensive product redesigns.17 If, as per the comment, “a product bearing … a warning, which is safe for use if followed, is not in defective condition, nor is it unreasonably dangerous,” then it is conceivable that courts would permit manufacturers to “paper over” design defects by alerting customers to the condition with a warning label.18 However, there is virtually no evidence that this concern is justified. The three primary cases that Professor Latin cites for the proposition did not set off a wave of courts allowing manufacturers to choose warnings over reasonably safe design—and one of those cases had already been overruled entirely by the time his article was published.19 Since then, courts have broadly rejected the notion that a manufacturer can “merely slap a warning onto its dangerous product, and absolve itself of any obligation to do more.”20 The duty to warn of a product’s inherent dangers and instruct users in its safe use in no way undermines manufacturers’ responsibility to place reasonably safe products into the stream of commerce; the two duties operate independently of each other.21
Moreover, Professor Latin’s concern ignores the fact that some products cannot be reasonably redesigned, in which case consumer protection depends on a robust duty to warn. For instance, the Nevada Supreme Court echoed Latin’s argument in its rejection of the heeding presumption in Rivera v. Philip Morris, Inc.22 When asked whether Nevada law demanded the recognition of the heeding presumption, the court rejected the presumption in large part because the justices considered it “better public policy not to encourage a reliance on warnings because this will help ensure that manufacturers continue to strive to make safe products.”23 However, the very product at issue in that case undermined the court’s rationale: Rivera’s wrongful death claim pertained to a failure to warn of the inherent dangers of cigarettes.24 There is no known way to manufacture a non-carcinogenic tobacco product, and so it seems improbable that manufacturers like Philip Morris would take the heeding presumption as license to substitute warnings for safe design. It is the unavoidable danger of the product that underscores the need for robust enforcement of warnings law.25 A life-saving medicine may carry an unavoidable risk of harmful side effects, a powerful machine may be deadly if used incorrectly, and no form of bleach can be safely mixed with an ammonia-based cleaning product. Such products are not more dangerous than consumers expect and are therefore not defective, yet they still carry enormous risks. Consumers have the right to be instructed about their potential hazards and proper operations. Such warnings save limbs and lives.
Others have argued that the heeding presumption threatens to subject manufacturers to absolute liability because it allows plaintiffs to recover for any perceived shortfall in a product’s warning.26 This would be bad for consumers as a class because it would be expensive and unjust to force responsible product users to pay for every injury suffered by careless users. However, this concern is best addressed on the element of defect, not the element of causation. Before a plaintiff can benefit from the heeding presumption, they must first prove that the warning in question was legally defective.27 Reasonableness is the touchstone of the defect analysis.28 As such, a plaintiff’s mere allegation that a product should have contained a more comprehensive, more prominent, or more forceful warning is insufficient to establish defect. Injured plaintiffs may try to claim that it was unreasonable for a manufacturer to withhold virtually any proposed warning, since the benefit of disclosure would have vastly outweighed the cost of the ink and paper to print it.29 However, the full cost of a warning is far more than the expense of a physical label. The true measure of reasonableness comes from information costs: each warning that is presented risks diluting the effect of others. Unnecessary warnings make consumers less safe, not more.30 It may be reasonable for manufacturers to withhold information about certain risks, since seemingly absurd warnings about unlikely or insultingly obvious dangers may lead customers to ignore warnings about more salient risks because they believe that all of the product’s safety information is just “for the lawyers.”31
Additionally, manufacturers can make reasoned decisions about how to present information. Not only the content, but also the design and conspicuousness of a warning are relevant to its adequacy. It may be entirely in the interest of customer safety for a seller to, for instance, draw special attention to serious hazards while printing information about more remote risks in a users’ guide; if everything is emphasized, then effectively nothing is emphasized.32 Accordingly, courts can protect against the dangers of absolute liability by properly instructing juries on the reasonableness test for warning defects. Absolute liability for warning-related injuries would hurt consumers, and the proliferation of useless warnings as manufacturers try to protect themselves from this absolute liability is a valid safety concern. However, these issues are properly addressed through a rigorous defect analysis—not by making recovery prohibitively difficult on the issue of causation.
The policy reasons for rejecting the heeding presumption ultimately ring hollow, for they all stem from doctrinal confusion unrelated to the question of causation. If a state legislature enacted strict liability for failure to warn, then the duty, and the consumer safety concerns which motivate that duty, are best served by ensuring that plaintiffs will not be forced to prove the impossible. The heeding presumption ensures that manufacturers’ duty to warn will be enforced, plaintiffs with valid claims will be compensated, and the public will be properly informed.
1. Jessica Christy is a J.D. Candidate (2021) at New York University School of Law. This piece is a commentary on the 2020 August A. Rendigs, Jr. National Products Liability Moot Court Competition. The issue in the problem dealt with whether a state’s strict products liability failure-to-warn law based on the Restatement (Second) of Torts demanded that the state recognize the heeding presumption.
2. Reyes v. Wyeth Lab’ys, 498 F.2d 1264, 1281 (5th Cir. 1974).
3. See, e.g., Kevin J. O’Connor, Comment: New Jersey’s Heeding Presumption in Failure to Warn Product Liability Actions: Coffman v. Keene Corp. and Theer v. Philip Carey Co., 47 Rutgers L. Rev. 343, 371–74 (1994) (reviewing New Jersey courts’ rationales for adopting the presumption).
4. See Howard Latin, “Good” Warnings, Bad Products, and Cognitive Limitations, 41 UCLA. L. Rev. 1193, 1293 (1994) (“The comment j presumption encourages producers to substitute warnings for safer product designs and, in conjunction with failure-to-warn doctrines, the presumption also induces manufacturers to provide … often excessive disclosure at the cost of sacrificing clarity .…”).
5. See James A. Henderson, Jr. & Aaron D. Twerski, Doctrinal Collapse in Products Liability: The Empty Shell of Failure to Warn, 65 N.Y.U. L. Rev. 265, (1990) (arguing that the presumption makes it “excessively easy” for a plaintiff to win a warning defect case, in large part because it is so difficult to rebut).
6. Restatement (Second) of Torts § 402A cmt. j (Am. Law Inst. 1965).
7. See, e.g., Tech. Chem. Co. v. Jacobs, 480 S.W.2d 602, 606 (Tex. 1972). State courts in approximately nineteen states have recognized this rebuttable presumption of causation in warning defect claims, although in four of those states, the legislature subsequently abolished or limited the presumption. Federal courts in an additional eleven states have predicted that state courts would apply the presumption if presented with the issue. James M. Beck, Who Heeds the Heeding Presumption, Drug & Device L. (Nov. 7, 2014), https://www.druganddevicelawblog.com/2014/11/who-heeds-heeding-presumption.html.
8. Payne v. Soft Sheen Prods., 486 A.2d 712, 725 (D.C. Cir. 1985).
9. See Nissen Trampoline Co. v. Terre Haute First Nat’l Bank, 332 N.E.2d 820, 827 (Ind. Ct. App. 1975) (commenting that adequate warnings are likely to “impair the marketability of the product”), rev’d on other grounds, 358 N.E.2d 974 (Ind. 1976).
10. Latin, supra note 4, at 1293.
11. See Aaron D. Twerski & Neil B. Cohen, Resolving the Dilemma of Nonjusticiable Causation in Failure-to-Warn Litigation, 84 S. Cal. L. Rev. 125, 139 (2010) (warning that the heeding presumption results in “overcompensation and overdeterrence”).
12. See Coffman v. Keene Corp., 628 A.2d 710, 719 (N.J. 1993) (reasoning that, when jurors are faced with “unreliable or self-serving” testimony, they will need to turn to “extraneous, speculative considerations” to evaluate it). Such extraneous considerations will likely include inappropriate factors such as the plaintiff’s education, socio-economic class, race, and/or country of origin. See, e.g., Sheri Lynn Johnson, The Color of Truth: Race and the Assessment of Credibility, 1 Mich. J. Race & L. 261, 316–17 (1996) (discussing how race affects jurors’ credibility determinations). Of course, juries must always make credibility determinations, but when assumptions about a plaintiff’s credibility and personal responsibility are virtually all that the jury can consider when evaluating the element of causation, concerns about jurors’ innate biases are particularly pressing.
13. See Seley v. G.D. Searle & Co., 423 N.E.2d 831, 838 (Ohio 1981) (stating that a plaintiff’s causal burden is satisfied if defendant does not produce rebutting evidence).
14. See Nissen Trampoline Co., supra note 9.
15. See Michael A. Pittenger, Note, Reformulating the Strict Liability Failure to Warn, 49 Wash. & Lee L. Rev. 1509, 1539 (1992) (framing the duty to warn as a matter of market honesty and informed consumer choice).
16. See Twerski & Cohen supra note 11, at 135 (compiling studies that indicate that approximately half of consumers fail to read warnings, while perhaps a third fail to comply with them).
17. Latin supra note 4, at 1195–97 (1994) (citing Skyhook Corp. v. Jasper, 560 P.2d 934, 936 (N.M. 1977), overruled by Klopp v. Wackenhut Corp., 824 P.2d 297 (N.M. 1992); Sherk v. Daisy-Heddon, 450 A.2d 615, 618–19 (Pa. 1982); Dugan v. Sears, Roebuck & Co., 454 N.E.2d 64, 65–67 (Ill. App. Ct. 1983)).
18. Restatement (Second) of Torts § 402A cmt. j.
19. See id. at 1195–96 nn.3–6 and accompanying text.
20. Rogers v. Ingersoll-Rand Co., 144 F.3d 841, 844 (D.C. Cir. 1998).
21. See, e.g., Emery v. Federated Foods, 863 P.2d 426, 431–32 (Mont. 1993) (holding that, while marshmallows are not inherently defective products, it is appropriate to let a jury decide if they require a warning about the choking hazard they pose to young children).
22. 209 P.3d 271, 276–77 (Nev. 2009).
23. Id. at 277.
24. Id. at 273.
25. See Kenneth Ian Weissman, A “Comment J” Parry to Howard Latin’s “Good” Warnings, Bad Products, and Cognitive Limitations, 70 St. John’s L. Rev. 629, 629–30 (1996) (“[S]ome socially beneficial products are inherently unsafe or cannot be made accident-proof without imposing prohibitive costs.”).
26. See Twerski & Cohen, supra note 11, at 139 (“The warning presumption, though formally denominated as rebuttable, was for all practical purposes absolute. Thus, by the seemingly dry procedural mechanism of burden shifting, the court transformed traditional tort principles, which, as we have demonstrated, may result in underenforcement and underdeterrence, into an alternative regime with opposite results. Just as a regime of underenforcement and underdeterrence is not acceptable, a regime that results in overcompensation and overdeterrence is similarly unjust.”).
27. See, e.g., Seley, supra note 13 at 836 (describing the question of whether a warning was adequate as the “initial” issue in a failure-to-warn case).
28. See, e.g., Jarrell v. Monsanto Co., 528 N.E.2d 1158, 1161 (Ind. Ct. App. 1988) (“In Indiana, the issue of the adequacy of warnings in a strict liability case is governed by the same concepts as in negligence.” (citation omitted)).
29. See, e.g., Moran v. Faberge, Inc., 332 A.2d 11, 15 (Md. 1975) (observing that the minimal cost of adding a label to a product “will almost always weigh in favor of an obligation to warn”).
30. See, e.g., Cotton v. Buckeye Gas Prods., 840 F.2d 935, 938 (D.C. Cir. 1988) (“The primary cost is, in fact, the increase in time and effort required for the user to grasp the message. The inclusion of each extra item dilutes the punch of every other item.”).
31. See Latin, supra note 4, at 1247 (“As warnings proliferate in number and length, consumers may come to believe that some (or many) are included more to protect manufacturers against potential liability than to inform users of significant dangers.”). However, it is widely accepted doctrine that manufacturers do not need to warn for risks that are commonly known in the community, that are obvious, or that are unforeseeable according to the industry’s best available knowledge. See, e.g., Phelps v. Sherwood Med. Indus., 836 F.2d 296, 303 (7th Cir. 1987) (citation omitted) (stating that there is no need to warn for obvious or unforeseeable dangers); Restatement (Second) of Torts § 402A cmt. j (disclaiming a duty to warn of generally recognized dangers, such as the risk of getting intoxicated from alcohol). But see Campos v. Firestone Tire & Rubber Co., 485 A.2d 305, 309 (N.J. 1984) (criticizing “obvious danger rule” as bar to recovery); Beshada v. Johns-Manville Prods. Corp., 447 A.2d 539, 544 (N.J. 1982) (holding asbestos manufacturers liable for failure to warn of risks that were scientifically unknown at the time of sale).
32. See, e.g., Scott v. Black & Decker, Inc., 717 F.2d 251, 254 (5th Cir. 1983) (rejecting plaintiff’s contention that a specific warning should have been printed directly on an electric saw because “[t]o require that one explicit warning be placed on the saw would be to require all twenty”).