Washington and Lee Law Review

Volume 49 Issue 4 Article 16

Fall 9-1-1992

Reformulating The Strict Liability Failure To Warn

Michael A. Pittenger

Follow this and additional works at: https://scholarlycommons.law.wlu.edu/wlulr

Part of the Torts Commons

Recommended Citation
Michael A. Pittenger, Reformulating The Strict Liability Failure To Warn, 49 Wash. & Lee L. Rev.
1509 (1992).
Available at: https://scholarlycommons.law.wlu.edu/wlulr/vol49/iss4/16

This Note is brought to you for free and open access by the Washington and Lee Law Review at Washington and
Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law
Review by an authorized editor of Washington and Lee University School of Law Scholarly Commons. For more
information, please contact [email protected].
 NOTES
REFORMULATING THE STRICT LIABILITY FAILURE
TO WARN

I. INTRODUCTION

A person injured by a defective product may recover damages against

the manufacturer or intermediate seller of the product.' Courts recognize
three types of product defects: Manufacturing defects, design defects, and
inadequate warnings. 2 A manufacturing defect, or flaw, is an unintended
abnormality in the product that results from improper manufacture.3 A
design defect conforms to the manufacturer's specifications but, neverthe-
less, imposes an unnecessary risk of injury.4 Finally, a warning defect results
from the manufacturer's failure to provide adequate warnings to the con-
sumer of dangers inherent in the product or to provide adequate instructions
for safe use of the product. s

1. See 1 M. STUART MADDEN, PRODUCTS LABmrnrY § 1.1, at 2 (2d ed. 1988) (stating
general rule that persons injured by defective products may recover damages from manufacturer
or intermediate seller of product); see also Wnju m L. PROSSER, HANDBOOK OF THE LAw OF
TORTS 641 (4th ed. 1971) (defining products liability as law involving liability of sellers of
chattels to third persons with whom sellers are not in privity of contract).
2. See Feldman v. Lederle Labs., 479 A.2d 374, 385 (N.J. 1984) (noting three types of
defects) (citing O'Brien v. Muskin Corp., 463 A.2d 298, 304 (N.J. 1983)). See generally W.
Page Keeton, The Meaning of Defect in ProductsLiability Law-A Review of Basic Principles,
45 Mo. L. REv. 579, 585-88 (1980) (describing, in some detail, three types of defects); John
W. Wade, On the Nature of Strict Tort Liabilityfor Products, 44 Miss. L.J. 825, 830 (1973)
(describing defect requirement and three types of defects).
3. See Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 846 (N.H. 1978) (defining
manufacturing defect as accidental variation caused by mistake in manufacturing process
resulting in product's not conforming with others of same design); see also RICHARD A.
EpsTErN, MODERN PRODUCTS LLA.BnrIY LAW 68 (1980) (stating that construction defect results
when product differs from own specifications). A manufacturing defect may be a result of
inadequate materials, failure to inspect, or improper assembly, packaging, or transportation.
MADDEN, supra note 1, § 9.2, at 322.
4. See Thibault, 395 A.2d at 846 (defining design defect as defect that occurs when
manufacturer constructs product in conformity with intended design but design itself poses
unreasonable danger to consumer); see also EpSrTE, supra note 3, at 69 (stating that design
defects are not result of deviation from manufacturer's chosen design but rather result from
design itself being unsafe); Keeton, supra note 2, at 587-88 (stating that product may be
defective because of defective design and describing tests for determining design defect). In
contrast to manufacturing defects, design defects necessarily pervade the entire product line.
EPsTEiN, supra note 3, at 69.
5. See Payne v. Soft Sheen Prods. Inc., 486 A.2d 712, 725 (D.C. 1985) (stating that
defect may result from failure to attach adequate warning if product may in certain circum-
stances cause injury). See generally David C. Little, Products Liability-the Growing Uncer-

1509
1510 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

A number of serious problems currently plague failure to warn product.s

liability law. Many courts have conceded that strict liability6 failure to warn,
as the courts have formulated the doctrine, differs little, if at all, from
negligent7 failure to warn. 8 Attempts to distinguish the two theories have
been unsuccessful, 9 and the courts typically allow plaintiffs to plead both
theories simultaneously, which inevitably leads to the confusion of judges
and juries and the making of bad law. 10
Aside from the difficulties arising from the similarity between strict
liability and negligence, a number of other doctrinal problems currently
exist in failure to warn products liability law. First, the courts have failed
to distinguish the risk-reduction from the informed-choice function of
warnings and have applied risk-reduction principles in determining the

tainty about Warnings, 12 FORUM 995 (1977) (describing failure to warn products liability
doctrine); M. Stuart Madden, The Duty to Warn in ProductsLiability: Contours and Criticism,
89 W. VA. L. Rav. 221 (1987) (same). Although a product is of proper design and manufacture,
it may, nevertheless, be defective because of the manufacturer's failure to warn of dangers
inherent in the use of the product or to provide adequate instructions for safe use of the
product. Jarrell v. Monsanto Co., 528 N.E. 2d 1158, 1166 (Ind. Ct. App. 1988).
Although warnings and directions for use are conceptually different, they will hereinafter
be treated collectively as "warnings." Thus, the "failure to warn" also includes the failure to
provide adequate instructions. For discussion of the similarities and differences between
warnings and instructions for use, see Little, supra, at 999-1001; James B. Sales, The Duty to
Warn and Instruct for Safe Use in Strict Tort Liability, 13 ST. MARY'S L.J. 521, 553-57
(1982).
6. See infra notes 26-30 and accompanying text (discussing strict products liability
doctrine). Although plaintiffs often plead a theory of implied warranty in addition to the
negligence and strict liability theories, for the purposes of this discussion implied warranty will
be treated as a form of strict liability. Like strict products liability, implied warranty is a form
of liability without fault. PROSSER, supra note 1, § 97, at 650, 653. Unlike the strict liability
action, however, devices such as disclaimers and notice requirements may limit the implied
warranty action. See U.C.C. § 2-316 (1978) (describing requirements for disclaiming implied
warranties of merchantability and fitness for particular purpose); id. § 2-607(3) (stating
requirement that buyer must notify seller of injury within reasonable time); see also PROSSER,
supra note 1, § 97, at 655-56 (discussing disclaimer and notice requirements). Most courts
today treat failure to warn actions in implied warranty substantially the same as they treat
actions in strict liability. See Hays v. Ariens Co., 462 N.E.2d 273, 277-78 (Mass. 1984)
(applying strict products liability rule in implied warranty context).
For a general discussion of the implied warranty action in the products liability context,
see PROSSER, supra note 1, § 97. See also U.C.C. § 2-314 (1978) (stating that seller makes
implied warranty of merchantability to consumer, which guarantees goods are of merchantable
quality); id. § 2-315 (stating that seller makes implied warranty of fitness for particular purpose
if seller guarantees to consumer that product is safe for contemplated use).
7. See infra note 20 (discussing negligence doctrine).
8. See infra note 49 (citing cases that have recognized similarity between strict liability
and negligent failure to warn actions).
9. See infra notes 55-81 and accompanying text (describing courts' unsuccessful efforts
to establish satisfactory distinction between strict liability and negligent failure to warn).
10. See infra note 83 (noting problem of confusion inherent in courts' permitting plaintiffs
to plead both strict liability and negligent failure to warn); infra notes 55-81 and accompanying
text (describing unsuccessful attempts of courts to develop sound strict liability failure to warn
doctrine).
1992] FAILURE TO WARN

adequacy of all warnings." A second area of difficulty results from the

courts' applying the same standard in failure to warn actions as they apply
in design defect actions.' 2 Finally, the causation inquiry presents numerous
complications in the failure to warn context. 3 A common thread running
through all of these doctrinal problems is that courts generally have not
considered, in formulating failure to warn doctrine, the important policies
that the failure to warn doctrine should seek to accommodate.' 4 Rather,
the primary concern of the courts has been to develop a strict liability
failure to warn doctrine that in some way differs from negligence. 5
This note documents many of the problems currently plaguing strict
liability failure to warn law and proposes a solution. Part II examines the
doctrinal development of strict liability failure to warn and the inability of
the courts to distinguish satisfactorily the strict liability failure to warn
action from the negligence action.' 6 Part III discusses the policies that courts
should seek to promote in formulating failure to warn doctrine and also
addresses the other doctrinal problems currently afflicting failure to warn
law.' 7 A prevailing theme of Part III is that many of the doctrinal difficulties
can be attributed to the courts' failure to analyze a manufacturer's failure
to warn in light of the appropriate policies. Finally, Part IV proposes a7
reformulated strict liability failure to warn doctrine that is based upon the
relevant policies and that also eliminates many of the current doctrinal
problems, including the similarity between the strict liability action and
negligence."
II. THE DOCTRINAL DEVELOPMENT OF STRICT LIABILITY FAILURE To
WARN
A. Negligent Failure to Warn
The manufacturer has a common-law duty to provide adequate warnings
against risks of which the manufacturer has or should have knowledge and

11. See infra notes 128-53 and accompanying text (describing problems resulting from
failure of courts to recognize differences between risk-reduction and informed-choice functions
of warnings).
12. See infra notes 142-64 and accompanying text (describing difficulties resulting from
courts' application of risk-utility balancing in both warning and design defect cases); see also
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, 292-303 (1990) (discussing number
of problems courts face in applying risk-utility balancing in failure to warn context).
13. See infra notes 192-209 and accompanying text (describing problems relating to
causation issue in failure to warn context).
14. See infra part III (discussing policies that failure to warn doctrine should seek to
accommodate).
15. See infra note 86 and accompanying text (arguing that primary concern of courts
has been to develop strict liability failure to warn doctrine that is different from negligence).
16. See discussion infra part II (describing problematic doctrinal development of strict
liability failure to warn).
17. See discussion infra part III (discussing doctrinal problems of failure to warn law
and policy considerations relevant to strict liability failure to warn doctrine).
18. See discussion infra, part IV (describing proposed reformulation of strict liability
failure to warn doctrine).
1512 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

that may render the product unreasonably dangerous for its intended purpose
in the absence of adequate warnings. 9 Under this negligence principle, 20 the
duty to warn is triggered only when a manufacturer knows or should know
of the dangers involved in the use of the product. 21 For the duty to warn
to attach, the danger presented by the use of the product in the absence of
an adequate warning must be unreasonableJ 2 The determination of whether
the danger is unreasonable requires a balancing of the seriousness of the
harm, the probability that the harm will result, and the burden involved in
taking the necessary precautions. 2 The manufacturer has no duty to warn

19. See Dempsey v. Virginia Dare Stores, Inc., 186 S.W.2d 217, 219 (Mo. App. 1945)
(stating general rule that manufacturer has duty to warn of dangers of which manufacturer
knows or should know and of which user has no knowledge). The RESTATEMENT (SEcoND) OF
TORTS § 388"(1965) states the general rule:
One who supplies directly or through a third person a chattel for another to use is
subject to liability to those whom the supplier should expect to use the chattel with
the consent of the other or to be endangered by its probable use, for physical harm
caused by the use of the chattel in the manner for which and by a person for whose
use it is supplied, if the supplier
(a) knows or has reason to know that the chattel is or is likely to be dangerous for
the use for which it is supplied, and
(b) has no reason to believe that those for whose use the chattel is supplied will
realize its dangerous condition, and
(c) fails to exercise reasonable care to inform them of its dangerous condition or of
the facts which make it likely to be dangerous.
Id.; see also PROSSER, supra note I, § 96, at 646-47 (noting general rule that seller is under
duty to warn of unreasonable dangers involved in use of product of which the seller knows
or should know).
20. See RESTATEMENT OF TORTS § 282 (1934) (defining negligence as conduct that "falls
below the standard established by law for the protection of others against unreasonably great
risk of harm"). The negligence cause of action traditionally involves five factors: (1) duty, (2)
breach of duty, (3) cause in fact, (4) proxmate cause, and (5) damages. See PROSSER, supra
note 1, § 30. For a discussion of the cost-benefit balancing involved in determining the duty
issue in negligence, see infra note 23. For a discussion of negligence in the products liability
context, see PROSSER, supra note 1, § 96.
21. See, e.g., Rivers v. Stihl, Inc., 434 So.2d 766, 773 (Ala. 1983) (stating general rule
that manufacturer has duty to warn under negligence principles only if manufacturer knows
or should know of risk); Payne v. Soft Sheen Prods., Inc., 486 A.2d 712 (D.C. 1985) (same);
Jarrell v. Monsanto Co., 528 N.E.2d 1158, 1161-62 (Ind. Ct. App. 1988) (same); Mitchell v.
Sky Climber, Inc., 487 N.E.2d 1374, 1376 (Mass. 1986) (same); RESTATEMENT (SEcoND) OF
TORTS § 388 (1965) (same).
22. See Moran v. Faberge, Inc., 332 A.2d 11, 15 (Md. 1975) (stating that manufacturers
have duty to warn under negligence .principles when risk is unreasonable); see also PROSSER,
supra note 1, § 96, at 646-47 (stating that seller is under duty to give adequate warning of
unreasonable dangers involved in use of product); Madden, supra note 5, at 234-35 (noting
general rule in negligence requires manufacturer to warn of unreasonable risks).
23. See Moran, 332 A.2d at 15 (stating that determination of duty to warn involves
balancing of probability and seriousness of harm against cost of taking precautions to reduce
risk of harm).
In United States v. Carroll Towing, 159 F.2d 169 (2d Cir. 1947), a negligence action
involving a ship that broke away from its moorings, Judge Learned Hand stated that the
defendant's duty under negligence doctrine was a matter of balancing "(1) the probability that
she will break away; (2) the gravity of the resulting injury if she does; [and] (3) the burden
of adequate precautions." Id. at 173.
1992] FAILURE TO WARN 1513

of dangers that are obvious to the product user 24 or that result from an
unforeseeable use of the product. 2
B. Strict Liability Failure to Warn
In 1965, the American Law Institute adopted strict products liability in
section 402A of the Restatement (Second) of Torts. 26 Pursuant to section
402A, anyone who sells a product in "a defective condition unreasonably
dangerous" to the user or consumer will be liable for any physical injury
caused by that condition. 27 Section 402A applies regardless of. whether the
seller or manufacturer used all possible care in the preparation and sale of
the product. 28 Thus, analysis purportedly focuses on the condition of the
product rather than on the negligent conduct of the manufacturer or seller. 29

24. See, e.g., Dudley Sports Co. v. Schmitt, 279 N.E.2d 266, 275 (Ind. Ct. App. 1972)
(explaining general rule that manufacturer has no duty to warn of dangers that are obvious);
Dempsey v. Virginia Dare Stores, Inc., 186 S.W.2d 217, 220 (Mo. App. 1945) (same); Griebler
v. Doughboy Recreational, Inc., 449 N.W.2d 61, 64-65 (Wis. Ct. App. 1989) (same). See
generally Madden, supra note 5, at 253-59 (discussing effect of obviousness of risk on duty
to warn). Dean Prosser stated the general rule: "[a manufacturer] is not liable for dangers
that are known to the user, or are obvious to him, or are so commonly known that it can
reasonably be assumed that the user will be familiar with them." PROSSER, supra note 1,
§ 96, at 649 (footnotes omitted). For example, it is obvious that a knife will cut, that a match
will take fire, or that a hammer may smash one's finger. Id.
25. See, e.g., Payne v. Soft Sheen Prods., Inc., 486 A.2d 712, 721 (D.C. 1985) (stating
general rule that manufacturer has no duty to warn of risks that are not foreseeable); Jonescue
v. Jewel Home Shopping Serv., 306 N.E.2d 312, 316 (Ill. App. Ct. 1973) (same); Higgens v.
Paul Hardeman, Inc., 457 S.W.2d 943, 948 (Mo. Ct. App. 1970) (holding that abnormal use
relieves manufacturer of liability for failure to warn only if misuse was unforeseeable). See
generally Dix W. Noel, Products Defective Because of Inadequate Directions or Warnings, 23
Sw. L.J. 256, 274-81 (1969) (discussing and explaining general rule that manufacturer has duty
to warn only of foreseeable risks). Dean Prosser stated, "a normal foreseeable use may include
such relatively uncommon ones as standing on a chair, eating coffee, testing heat fixtures, or
wearing a cocktail robe in proximity to the kitchen stove." PROSSER, supra note 1, § 96, at
647.48 (footnotes omitted).
26. RESTATEMENT (SEcOND) OF TORTS § 402A (1965). Section 402A reads:
(1) One who sells any product in a defective condition unreasonably dangerous to
the user or consumer or to his property is subject to liability for physical harm
thereby caused to the user or consumer, or to his property, if
(a) the seller is engaged in the business of selling such a product, and
(b) it is expected to and does reach the user or consumer without substantial change
in the condition in which it is sold.
(2) The rule stated in Subsection (1) applies although
(a) the seller has exercised all possible care in the preparation and sale of his product,
and
(b) the user or consumer has not bought the product from or entered into any
contractual relation with the seller.
RESTATEMENT (SECOND) OF TORTS § 402A (1965).
27. RESTATEMENT (SECOND) OF TORTS § 402A(l) (1965).
28. RESTATEMENT (SEcoND) OF TORTS § 402A(2)(a); id. § 402A cmt. a.
29. See, e.g., Anderson v. Owens-Coming Fiberglas Corp., 810 P.2d 549, 558 (Cal.
1991) (stating that in strict liability focus is on condition of product, while in negligence focus
is on conduct of manufacturer); Woodill v. Parke Davis & Co., 402 N.E.2d 194, 198 (Ill.
1980) (same); Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1037, 1039 (Or. 1974) (same);
Falk v. Keene Corp., 767 P.2d 576 (Wash. App.) (same), aff'd, 782 P.2d 974 (Wash. 1989).
1514 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

Section 402A does not state an exclusive rule that precludes liability based
upon negligence, but rather provides an additional ground for finding
liability. 30 A number of jurisdictions adopted section 402A soon after its
introduction, 3' and today most states have either adopted the Restatement
32
version or some other version of strict liability.
Based upon several of the comments following section 402A, 33 a large
number of courts have concluded that the drafters of section 402A intended
to create a strict liability action for the failure to warn.3 4 However, comment

30. RESTATEMENT (SECOND) OF TORTS § 402A cmt. a.

31. See William L. Prosser, The Fall of the Citadel (Strict Liability to the Consumer),
50 MINN L. REv. 791, 793-98 (1966) (describing rapid pace at which states recognized strict
products liability following adoption of § 402A-by judicial decision or statute in 24 states,
by decisions indicating that change in law was imminent in 6 states, and with some limitations
on its applicability in 11 states).
32. See State Chart-Acceptance of Strict Liability, I Prod. Liab. Rep. (CCH) 4016
(Nov. 1988 & Apr. 1989) (summarizing recognition of strict products liability in states, District
of Columbia, and Puerto Rico). Thirty-seven states and the District of Columbia recognize
the Restatement's version of strict products liability. Id. Seven states and Puerto Rico recognize
other variations. Id. Delaware, Massachusetts, Michigan, North Carolina and Virginia have
not yet recognized strict products liability. Id. See generally Acceptance by State Jurisdictions,
1 Prod. Liab. Rep. (CCH) 4015 (Nov. 1988) (annotating treatment of strict products liability
doctrine by highest courts of 50 states, District of Columbia, and Puerto Rico).
33. See RESTATEMENT (SECOND) OF TORTS § 402A cmts. h, j, k (1965). Comment j
provides that in order to prevent a product from being unreasonably dangerous, the seller
may be required to warn against unreasonable dangers of which the public may be unaware.
Id. cmt. j. Comment h states that a product is in a defective condition if the seller markets
the product without an adequate warning when the seller has reason to anticipate that danger
may result from use of the product. Id. cmt. h. Comment h contains a cross reference to
comment j. Id. Finally, comment k excepts "unavoidably unsafe" products from strict liability,
but makes clear that such products must be accompanied by proper directions and warnings.
Id. cmt. k. "Unavoidably unsafe products" are those which are incapable of being made
completely safe for their intended use. Id. The marketing of unavoidably unsafe products is,
nevertheless, completely justified because such products are "apparently useful and desirable."
Id.
34. See Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1088-89, 1091 (5th Cir.
1973) (citing comments j and k in defining scope of strict liability action based upon failure
to warn), cert. denied, 419 U.S. 869 (1974); Canifax v. Hercules Powder Co., 46 Cal. Rptr.
552, 557-58 (Cal. Ct. App. 1965) (concluding, after discussing comments j and k, that
Restatement's framers intended for product to be defective under § 402A if it would be
unreasonably dangerous to market product without warning); see also Myron J. Bromberg,
The Mischief of the Strict Liability Label in the Law of Warnings, 17 SroN HALL L. REv.
526, 527, 540 (1987) (stating that courts have construed § 402A comments as having created
new strict liability failure to warn doctrine); Henderson & Twerski, supra note 12, at 271 n.21
(noting that text does not explicitly state rule applies to warning claims, but comments make
clear that drafters intended that result). The comments to § 402A have been regarded as so
integral to the black letter rule set forth in that section that they are commonly treated as
part of the rule. See Bromberg, supra, at 527.
Some commentators have argued that the framers of § 402A erred in implying a strict
liability action for failure to warn in the § 402A comments. Bromberg, supra, at 540; Henderson
& Twerski, supra note 12, at 271-72. At least one commentator has argued that the framers
of § 402A did not intend for the comments to imply an action in strict liability for failure to
warn at all. EPSTEIN, supra note 3, at 66; see also Leichtamer v. Amer. Motors. Corp., 424
1992] FAILURE TO WARN 1515

j to section 402A introduces an important limitation on strict liability for

failure to warn. According to comment j, the seller is liable for the failure
to provide an adequate warning only if the seller knows or should know
of the product-related risk.'- Considerable controversy exists as to whether
the proof-of-knowledge requirement should be a prerequisite to imposing
strict liability for the failure to warn.3 6 The majority of courts, however,
have adopted the position of comment j that the manufacturer must have
had actual or constructive knowledge of the product's dangers before the
court will impose strict liability for the manufacturer's failure to warn.37
Once a court requires proof of knowledge as a necessary element of
strict liability failure to warn, the strict liability action begins to look much
like its negligence counterpart.38 To explain why, it is necessary to describe
the means by which courts determine that a product is in a "defective
condition unreasonably dangerous." Courts most typically determine that a
product is in the requisite dangerously defective condition through the use
of a risk-utility balancing test.3 9 Under risk-utility analysis, a product is

N.E.2d 568, 578 (Ohio 1981) (refusing to recognize strict liability failure to warn and stating
that § 402A, rather than creating strict liability failure to warn, merely provides that giving
of warning is affirmative defense to strict liability claims).
35. See supra note 33 (describing comment j). Comment j limits the seller's liability for
failure to warn to situations in which "he has knowledge, or by the application of reasonable,
developed human skills and foresight should have knowledge of the presence of the ...
danger." REsTATEMENT (SEcoND) OF ToRTs § 402A cmt. j. Comment j's limitation seems to
conflict with other portions of § 402A that would make the seller subject to liability even
though he has exercised all possible care in the preparation of the product. See id. § 402A(2)(a);
id. § 402A cmt. a.
36. See, e.g., Robbins v. Farmers Union Grain Terminal Assoc., 552 F.2d 788, 794 n.15
(8th Cir. 1977) (noting split in authorities over whether strict liability failure to warn requires
actual or constructive knowledge of risk); Woodill v. Parke Davis & Co., 402 N.E.2d 194,
197 (Il1. 1980) (same); Ortho Pharmaceutical Corp. v. Chapman, 388 N.E.2d. 541, 546-47
(Ind. Ct. App. 1979) (same). For a discussion and analysis of some arguments for and against
conditioning imposition of strict liability on the seller's actual knowledge of the danger see
infra notes 68-81 and accompanying text.
37. See Charles C. Marvel, Annotation, Strict Products Liability: Liability for Failure
to Warn as Dependent on Defendant's Knowledge of Danger, 33 A.L.R.4TH 368, 370-377
(1984 & Supp.) (citing cases in which courts have required that defendant have knowledge or
reason to know of product-related risk as precondition for imposing strict liability for failure
to warn).
38. See cases cited infra note 49 (concluding that strict liability failure to warn differs
little, if at all, from negligence action); see also Henderson & Twerski, supra note 12, at 271-
73 (arguing that strict liability failure to warn and negligence are, by necessity, virtually
identical and that courts have been unsuccessful in articulating a meaningful difference); John
A. Kidwell, The Duty to Warn: A Description of the Model of Decision, 53 TEx. L. REv.
1375, 1377-79 (1975) (concluding that no relevant differences exist between strict liability and
negligent failure to warn).
39. See, e.g., Borel v. Fibreboard Paper Prods. Corp, 493 F.2d 1076, 1087 (5th Cir.
1973) (applying risk-utility test to determine whether product was "unreasonably dangerous"),
cert. denied, 419 U.S. 869 (1974); Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 846 (N.H.
1978) (applying risk-utility test to determine whether product is defective); Phillips v. Kimwood
Mach. Co., 525 P.2d 1033, 1039 (Or. 1974) (stating that risk-utility balancing is used to
1516 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

unreasonably dangerous if a reasonable person would conclude that the

magnitude of the danger outweighs the benefits of the way in which the
manufacturer designed and marketed the product. 40 This standard is nearly
identical to the cost-benefit test that courts use to determine duty in
negligence law. 41 Theoretically, however, the risk-utility test differs from
negligence because in strict liability the defendant's fault is irrelevant. 42 In

determine whether product is dangerously defective); Turner v. General Motors Corp., 584
S.W.2d 844, 847 & n.1, 851 (Tex. 1979) (approving general risk-utility approach to aid jury
in determining liability).
The Restatement suggests a consumer expectations test for determining whether a product
is in a "defective condition unreasonably dangerous." See RESTATEmENT (SECOND) OF TORTS
§ 402A cmt. g, cmt. i (1965). According to comments g and i, a product is in a "defective
condition unreasonably dangerous" if it is dangerous to an extent beyond that contemplated
by the ordinary consumer. Id. Few jurisdictions have adopted the consumer expectations test
as the sole basis for determining whether a product is defective, but some have adopted it as
an alternative to the risk-utility test. See Barker v. Lull Eng'g. Co., 573 P.2d 443, 452 (Cal.
1978) (adopting two prong test for determining defectiveness-product is defective if more
dangerous than reasonable consumer would expect or if benefits do not outweigh risks); Knitz
v. Minster Mach. Co., 432 N.E.2d 814, 818 (Ohio) (same), cert. denied, 459 U.S. 857 (1982).
For a critique of the consumer expectations test, see Keeton, supra note 2, at 588-92. For a
discussion of the benefits of the consumer expectations test, especially when used in conjunction
with the risk-utility test, see James E. Britain, Product Honesty is the Best Policy: A
Comparison of Doctors' and Manufacturers' Duty to Disclose Drug Risks and the Importance
of Consumer Expectations in Determining Product Defect, 79 Nw. L. REv. 342, 353-57, 363-
66, 408-22 (1984).
For a description and analysis of other proposed tests for determining whether or not a
product is dangerously defective see Frank J. Vandall, "Design Defect" in ProductsLiability:
Rethinking Negligence and Strict Liability, 43 OHIO ST. L.J. 61, 72-79 (1982).
40. See Borel, 493 F.2d at 1087 (stating that produict is unreasonably dangerous if, on
balance, utility of product does not outweigh magnitude of danger); Thibault, 395 A.2d at
846 (N.H. 1978) (explaining that risk-utility test involves weighing product utility against
danger). Dean Keeton, a leading proponent of the risk-utility test has formulated it as follows:
A product is defective if it is unreasonably dangerous as marketed. It is unreasonably
dangerous if a reasonable person would conclude that the magnitude of the scien-
tifically perceivable danger ... outweighed the benefits of the way the product was
designed and marketed.
W. Page Keeton, Products Liability and the Meaning of Defect, 5 ST. MARY's 1.J. 30, 37-38
(1973). Dean Keeton has made it clear that economic feasibility is a factor to consider in risk-
utility analysis. Keeton, supra note 2, at 493; see also Wade, supra note 2, at 835, 837-38
(approving risk-utility approach and listing a number of factors to be considered, including
economic feasibility).
41. See supra note 23 and accompanying text (discussing negligence cost-benefit analysis).
Compare Thibault, 395 A.2d at 846 (explaining that in weighing product utility against danger
it is necessary to consider whether risk could have been reduced without significant impact on
product effectiveness and manufacturing cost) and Knitz v. Minster Mach. Co., 432 N.E.2d
814, 818 (Ohio) (listing three factors to consider in applying strict liability risk-utility test:
Likelihood of injury, gravity of danger, and mechanical and economical feasibility of improved
design), cert. denied, 459 U.S. 857 (1982) with United States v. Carroll Towing, 159 F.2d 169,
173 (2d Cir. 1947) (holding that determination of duty under negligence law is matter of
balancing probability of risk, gravity of resulting injury, and burden of adequate precautions).
42. See supra note 28 and accompanying text (explaining that strict products liability is
liability without fault).
19921 FAILURE TO WARN 1517

order to retain the no-fault characteristic of strict liability, a number of

courts impute knowledge of the product-related risk to the defendant before
43
performing risk-utility balancing.
In strict liability failure to warn cases, courts typically utilize the
standard risk-utility test to determine whether a warning is required. 44 At
the same time, the courts require proof of the defendant's knowledge, as
suggested by comment j.41 The resulting analysis is virtually identical to the
inquiry used to fix the duty of care in negligent failure to warn. Both
theories require that the manufacturer have actual or constructive knowledge
of the danger, 46 and both theories involve some type of risk-utility or cost-
benefit balancing to determine whether a warning is needed. 47 Under either
theory, the reasonableness of the manufacturer's conduct effectively becomes
4
the determinative factor.

43. See, e.g., Helene Curtis Ind., Inc. v. Pruitt, 385 F.2d 841, 850 (5th Cir. 1967)
(stating standard as product being dangerous to extent that reasonable man would not sell
product if he knew of risk involved), cert. denied, 391 U.S. 913 (1968); Feldman v. Lederle
Labs., 479 A.2d 374, 385 (N.J. 1984) (holding that question is whether defendant acted in a
"reasonably prudent manner" in marketing product, assuming that defendant knew of prod-
uct's defect); Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1036 (Or. 1974) (holding "[t]he
test, therefore, is whether the seller would be negligent if he sold the article knowing of the
risk involved."). The courts have asserted that imputing knowledge of the risk to the defendant
before the court performs risk-utility balancing results in a characterization of the product
rather than an evaluation of the manufacturer's conduct. Phillips, 525 P.2d at 1037.
Deans Wade and Keeton have proposed that the strict liability risk-utility test is most
simply stated by first assuming that the manufacturer or seller knew of the dangerous condition
of the product and then asking whether the defendant would be negligent in marketing the
product knowing of the danger. W. Page Keeton, Products Liability-Some Observations
About Allocation of Risks, 64 MICH. L. Rav. 1329, 1335 (1966); Wade, supra note 2, at 834-
35. The supposed benefit of the Wade-Keeton approach is that it is framed in negligence
language, and negligence terminology and thought processes are already familiar to courts,
lawyers, and juries. Phillips, 525 P.2d at 1037. But cf. Cronin v. J.B.E. Olson Corp., 501
P.2d 1153, 1162-63 (Cal. 1972) (holding that phrase "unreasonably dangerous," which § 402A
introduced into definition of defective product, is not part of California law because it "rings
of negligence").
44. See, e.g., Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1087 (5th Cir.
1973) (applying risk-utility test in failure to warn context), cert. denied, 419 U.S. 869 (1974);
Anderson v. Owens-Corning Fiberglas Corp., 810 P.2d 549, 554- (Cal. 1991) (approving of use
of balancing test in warning context); Finn v. G.D. Searle & Co., 677 P.2d 1147, 1152-53
(Cal. 1984) (same); Feldman v. Lederle Labs., 479 A.2d 374, 382, 385-86 (N.J. 1984) (same);
Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1036-39 (Or. 1974) (same).
45. See Marvel, Annotation, supra note 37, at 370-377 (citing cases in which courts have
required that manufacturer have knowledge, or reason to know, of product-related risk as
precondition for imposing strict liability for failure to warn); see also RESTATEMENT (SEcoND)
oF ToRTs § 402A cmt. j (stating that knowledge or ability to know of danger is element of
strict liability failure to warn action). Some courts shift the burden of proof to the defendant
on the knowledge issue. Cases cited infra note 66.
46. See supra notes 35-37 and accompanying text (discussing knowledge requirement
under strict liability failure to warn doctrine); supra note 21 and accompanying text (discussing
knowledge requirement under negligent failure to warn doctrine).
47. See supra note 41 and accompanying text (describing similarity between strict liability
risk-utility test and negligence cost-benefit test).
48. See cases cited infra note 49 (concluding that strict liability failure to warn is virtually
1518 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

A number of courts have recognized the fact that strict liability failure
to warn differs little, if at all, from the negligent failure to warn doctrine.4 9
Many of these courts have been deeply concerned by the similarity. 0 These
courts, all of which enthusiastically adopted strict products liability in order
to escape the limitations of the negligence doctrine,5 ' have endeavored to
establish a satisfactory distinction between the two actions.5 2 Attempts to
distinguish the two theories, however, have been largely unsuccessful. Many
of the distinctions that the courts have managed to create are merely
illusory,53 and those that are not are often the product of analytical confusion
54
or ill-conceived policy.
Representative of the illusory distinctions is the argument that strict
liability failure to warn differs from negligence because in strict liability the

identical to negligence standard because reasonableness of manufacturer's conduct is determi-

native factor under both); authorities cited supra note 38 (arguing that strict liability failure
to warn is essentially the same as negligence standard). In Borel v. Fibreboard Paper Prods.
Corp., 493 F.2d 1076, 1087-88 (5th Cir. 1973), cert. denied, 419 U.S. 869 (1974), the court
stated:
...The determination that a product is unreasonably dangerous, or not reasonably
safe, means that, on balance, the utility of the product does not outweigh the
magnitude of the danger .... The fulcrum for this balancing process is the reasonable
man as consumer or as seller. Thus, a product is unreasonably dangerous only when
it is "dangerous to an extent beyond that contemplated by the ordinary consumer
who purchases it." ... In other words, for a product to be unreasonably dangerous,
"it must be so dangerous that a reasonable man would not sell the product ifhe
knew the risk involved."
Id. (emphasis added) (citations omitted). Borel makes clear that the reasonableness of the
seller's conduct is the determinative factor, just as it would be under a negligence analysis.
49. See, e.g., Russell v. G.A.F., 422 A.2d 989, 991 (D.C. 1980) (concluding that both
strict liability and negligent failure to warn impose duty of ordinary care); Bernier v. Raymark
Ind., Inc., 516 A.2d 534, 540 (Me. 1986) (explaining that strict liability failure to warn
resembles negligence because reasonableness of manufacturer's conduct is critical issue under
each); Smith v. E.R. Squibb & Sons, Inc., 273 N.W.2d 476, 480 (Mich. '1979) (holding that
when liability turns on adequacy of warning, issue is one of reasonable care regardless of
theory pleaded); Knitz v. Minster Mach. Co., 432 N.E.2d 814, 815 n.5 (Ohio) (noting that
rule imposing obligation on seller to give adequate warning is rule fixing standard of care,
regardless of whether tort is labeled strict liability) (quoting Temple v. Wean United, Inc., 364
N.E.2d 267, 273 (Ohio 1977)), cert. denied, 459 U.S. 857 (1982).
50. See cases cited infra notes 55-81 (attempting to develop distinction between strict
liability failure to warn and negligence).
51. See generally Prosser, supra note 31, at 791-800 (describing rapid development of
strict products liability and noting policies driving rapid development).
52. See infra notes 55-81 and accompanying text (describing attempts of courts to
differentiate strict liability from negligent failure to warn); see also John W. Wade, On Product
"Design Defects" and Their Actionability, 33 VAND. L. RaV. 551, 558 (1980) (noting that
number of appellate judges have become concerned that they are not differentiating strict
liability from negligence with sufficient clarity and in response have sought to devise "a
significant and well-articulated distinction").
53. See infra notes 55-61 and accompanying text (describing illusory distinctions that
courts have developed between strict liability failure to warn and negligence).
54. See infra notes 62-81 and accompanying text (describing distinctions courts have
developed between strict liability and negligent failure to warn that are result of unsound
rationale or poor policy).
19921 FAILURE TO WARN 1519

focus is on the condition of the product, whereas in negligence, the focus

is on the conduct of the manufacturer." Courts advancing this argument
typically apply the standard strict liability failure to warn test, under which
the reasonableness of the defendant's conduct is actually the determinative
factor. 6 In this sense, the product-conduct distinction is illusory; it is
difficult to see how a test that ultimately depends upon the reasonableness
of the defendant's conduct can actually focus on the condition of the
product. 7
Other examples of illusory distinctions deserve mention. Some courts
and commentators have suggested that strict liability failure to warn sets a
higher standard of care than does negligence."8 It is unlikely that strict
liability failure to warn sets a higher degree of care, however, because the
courts arguing that it does generally employ the usual strict liability failure
to warn test, which is actually little more than the "strict liability" label

55. See, e.g., Woodill v. Parke Davis & Co., 402 N.E.2d 194, 198 (Ill. 1980) (explaining
that strict liability failure to warn inquiry focuses on nature of product rather than conduct
of manufacturer); Phillips v. Kimwood Machine Co., 525 P.2d 1033, 1037 (Or. 1974) (same);
Falk v. Keene Corp., 767 P.2d 576 (Wash. App.) (explaining that strict liability test focuses
on manufacturer's defective product and is "fundamentally and irreconcilably different anal-
ysis" than negligence, which focuses on manufacturer's conduct), aff'd, 782 P.2d 974 (Wash.
1989).
56. See Woodill, 402 N.E.2d at 198 (explaining that under strict liability failure to warn,
proof of knowledge is required and proper standard is whether it would have been reasonable
for manufacturer to give adequate warning); Phillips, 525 P.2d at 1036 & n.6, 1039 (adopting
negligence with imputed knowledge test, but citing, with approval, Dean Wade who "would
impute only the knowledge existing at the time the product was sold"); see also supra notes
38-49 and accompanying text (arguing that conventional strict liability failure to warn test
actually is little more than negligence standard).
57. See Henderson & Twerski, supra note 12 at 275-76 (arguing that little, if any,
practical difference exists between product and conduct standard); William C. Powers, Jr.,
The Persistence of Fault in Products Liability, 61 TEx. L. RE. 777, 791-94, 809 (1983)
(arguing that product-conduct distinction is illusory because focus on product necessarily
involves consideration of product costs, which in turn represents an evaluation of manufac-
turer's conduct). The illusory character of the product-conduct distinction is apparent in a
statement by the court in Woodill, 402 N.E.2d at 198:
We perceive that requiring a plaintiff to plead and prove that the defendant
manufacturer knew or should have known of the danger that caused the injury, and
that the defendant manufacturer failed to warn plaintiff of that danger, is a
reasonable requirement, and one which focuses on the nature of the product and
on the adequacy of the warning, rather than on the conduct of the manufacturer.
Id. It is difficult to see how an inquiry into the manufacturer's state of mind and into the
manufacturer's failure to act can actually focus on the condition of the product. An observation
by the California Supreme Court in Anderson v. Owens-Corning Fiberglas Corp., 810 P.2d
549 (Cal. 1991) is on point: "[W]hile a manufacturing or design defect can be evaluated
without reference to the conduct of the manufacturer, the giving of a warning cannot. The
latter necessarily requires the communication bf something to someone." Id. at 558 (citations
omitted).
58. See Anderson, 810 P.2d at 558-59 (explaining that standard of care is higher in strict
liability failure to warn); Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1039 (Or. 1974)
(same); Noel, supra note 25, at 267 (arguing that a lesser degree of risk may establish duty
to warn under strict liability rules).
1520 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

applied to a negligence standard.5 9 At least one court has suggested that in

strict liability failure to warn cases, warnings must cover "all foreseeable
uses.' ' 6 How this concept would meaningfully differ from foreseeability
under a negligence theory, however, is hard to imagine. 6'
Sometimes courts have successfully articulated a "real" distinction
between strict liability failure to warn and its negligence counterpart. Illus-
trative of this line of distinctions is the argument that defenses based upon
the plaintiff's fault, such as contributory negligence, comparative fault, or
consumer misuse, are available in negligence but not in strict liability. 62 One
reason for disallowing plaintiff-fault defenses is that to allow them would
be inconsistent with the no-fault basis of strict products liability.63 However,
defenses based upon the plaintiff's conduct are not analytically inconsistent
with no-fault liability if they are viewed as presenting countervailing reasons
that outweigh the policies that support recovery rather than mitigate the

59. See Anderson, 810 P.2d at 554, 557 (approving use of risk-utility balancing test in
strict liability failure to warn context and holding that manufacturer's knowledge is component
of strict liability failure to warn action); Phillips, 525 P.2d at 1036 & n.6, 1039 (Or. 1974)
(adopting negligence with imputed knowledge test, but citing with approval Dean Wade who
"would impute only the knowledge existing at the time the product was sold"); see also supra
notes 38-49 and accompanying text (explaining that typically employed strict liability failure
to warn tests are little more than negligence standard). Of course, a jury intuitively may hold
the manufacturer to a higher degree of care under strict liability failure to warn, but only
because the court has applied the strict liability label to what is actually a negligence standard.
See Henderson & Twerski, supra note 12, at 276 (explaining that semantic "word games" of
courts may occasionally result in jury being harder on defendant when applying strict liability).
60. See Smith v. U.S. Gypsum Co., 612 P.2d 251, 254 (Okla. 1980) (holding that in
strict liability failure to warn, manufacturer must warn of dangers inherent in "all foreseeable
uses," which differs from foreseeability in negligence).
61. See Bromberg, supra note 34,. at 533 (explaining that view that requiring warnings
to cover "all foreseeable uses" somehow differs from foreseeability in negligence law fails to
account for modern meaning of foreseeability in negligence law).
62. See, e.g., Young v. Up-Right Scaffolds, Inc., 637 F.2d 810, 814 (D.C. Cir. 1980)
(stating that contributory negligence is not available in strict liability failure to warn); Payne
v. Soft Sheen Prods., Inc., 486 A.2d 712, 721 n.9 (D.C. 1985) (same) (citing Russell v. G.A.F.
Corp., 422 A.2d 989, 991 n.* (D.C. 1980)); Jarrell v. Monsanto Co., 528 N.E.2d 1158, 1167
(Ind. Ct. App. 1988) (stating that contributory negligence, or comparative fault, is not available
defense under doctrine of strict liability failure to warn). See generally 1 ROBERT D. HURSH
& HENRY J. BAILEY, AMERICAN LAW OF PRODUCTS L IArTY §§ 4.33 & 4.34 (2d ed. 1974 &
Supp. 1985) (describing general treatment by courts of issue of availability of plaintiff-fault
defenses in strict products liability and citing cases).
In addition to distinctions based upon the nonavailability of plaintiff-fault defenses in
strict liability, courts have advanced a number of other "real" distinctions. Some courts have
concluded that evidence of industry custom is not admissible in strict liability. Holloway v.
J.B. Sys. Ltd., 609 F.2d 1069, 1073 (3d Cir. 1979). Other courts have held that evidence of
warnings given in the subsequent marketing of the product is admissible in strict liability but
not in negligence. Ault v. Int'l Harvester Co., 528 P.2d 1148, 1150-52 (Cal. 1974). For a
lengthy critique of these and other distinctions courts have created between strict liability and
negligent failure to warn, see Bromberg, supra note 34.
63. See RESTATEMENT (SECOND) OF TORTS § 402A cmt. n. (1965) (explaining that because
§ 402A is not based upon seller's negligence, but rather on strict liability, contributory negligence
is not a defense).
1992] FAILURE TO WARN

plaintiff's prima facie case." In addition, if public policy demands the

conclusion that plaintiff-fault defenses should not be available in a particular
case or in failure to warn cases generally, those reasons should apply just
as well to the negligence action as they do to strict liability. 61
Another way that some courts have distinguished strict liability failure
to warn from negligence is by imputing knowledge of the risk to the
manufacturer-defendant, thus shifting the burden to the defendant of prov-
ing that the risk was unknowable at the time the product entered the
market." Although the imputed knowledge distinction does not suffer from
the same analytical problems from which the distinction based upon plain-
tiff-fault defenses suffers, it nevertheless does not withstand close scrutiny.
If sound policy reasons exist for shifting the burden of proof on the
knowledge issue, those reasons should apply just as well to negligent failure
to warn as they do to strict liability. 67

64. See Powers, supra note 57, at 799 (explaining that contributory negligence is not
inconsistent with no-fault liability if viewed as type of affirmative defense in which policies
supporting recovery are outweighed by countervailing policies). One court has articulated
another way to avoid the apparent inconsistency between contributory negligence and strict
liability. See Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 848-50 (N.H. 1978) (explaining
that plaintiff's conduct may act together with defendant's product to jointly cause injury and
that assessing damages on basis of comparative causation between plaintiff's conduct and
defendant's product alleviates inconsistency).
65. See Green v. Sterling Extruder Corp., 471 A.2d 15, 19-20 (N.J. 1984) (concluding
that if public policy requires that contributory negligence defense be unavailable in strict
products liability, then same public policy requires nonavailability of contributory negligence
in negligent products liability action); Henderson & Twerski, supra note 12, at 276-77 (arguing
that courts could just as well work within negligence framework in failure to warn case and,
remain free to resolve issue of contributory fault as they, see fit); cf. Powers, supra note 57,
at 799 n.76 (arguing that one reason courts likely have been reluctant to permit plaintiff-fault
defenses in strict products liability, is dissatisfaction with contributory negligence in general
and that, while courts feel constrained by precedent in negligence cases, they are able to avoid
aggravating problem by refusing to permit defenses in strict liability).
66. See, e.g., Bernier v. Raymark Ind., Inc., 516 A.2d 534, 538-40 (Me. 1986) (allowing
defendant to introduce state of art evidence on knowledge issue); Bilotta v. Kelley Co., 346
N.W.2d 616 (Minn. 1984) (holding that distinction between strict liability and negligence in
design defect and failure to warn cases is that in strict liability, knowledge of condition of
product and risks involved is imputed to manufacturer); Feldman v. Lederle Lab., 479 A.2d
374, 388 (N.J. 1984) (holding that as matter of policy, burden of proving scientific unknow-
ability of risk should be place on defendant).
67. See Bromberg, supra note 34, at 530 (noting that shifting burden of proof on
knowledge issue to defendant is no more appropriate to strict liability than to negligence);
David P. Griffith, Note, Products Liability-Negligence Presumed: An Evolution, 67 TEx. L.
REv. 851, 879-80 (1989) (arguing that rationale which justifies shifting the burden of proof
does not warrant separate strict liability doctrine because courts often shift burdens in negligence
context); cf. authorities cited supra note 65 (arguing that availability of contributory fault
should not necessarily depend upon whether case is styled in strict liability or negligence).
The reason courts most often state for shifting certain evidentiary burdens to the defendant
in strict liability is that the defendant is in a better position to prove certain facts. See infra
note 94 and accompanying text (discussing strict liability justification based upon need to
relieve plaintiff of onerous burden of proving negligence). However, nothing prevents courts
1522 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

Some courts have gone beyond merely shifting the burden of proof on
the knowledge issue and have held that the defendant may be liable in strict
liability for the failure to warn regardless of whether the risk was scientif-
ically knowable at the time the product entered the market. 8 The decision
of the New Jersey Supreme Court in Beshada v. Johns-Manville Products
Corp.6 9 is the most notable in this line of cases. The plaintiffs in Beshada
were asbestos workers, or survivors of asbestos workers, who claimed to
have contracted various respiratory diseases - from the use of asbestos-
containing products manufactured by the defendants.7 0 The plaintiffs alleged
that the defendants failed to warn of the dangers associated with the use
7
of asbestos1. The defendants asserted that the risks were scientifically
undiscoverable at the time the defendants placed the asbestos-containing
products on the market.7 2 The court, nevertheless, held that the manufac-
turers could be found strictly liable for the failure to warn regardless of
whether the risks were scientifically knowable when the product was mar-
keted.73

from shifting the burden of proof in the negligence context based upon the same rationale.
See Ybarra v. Spangard, 154 P.2d 687, 689-90 (Cal. 1944) (holding in medical malpractice suit
that because plaintiff was unconscious during surgery when his injury occurred, medical staff
was in better position to have pertinent evidence, which justifies res ipsa loquitur inference);
Anderson v. Somberg, 338 A.2d 1, 6-7 (N.J.) (placing burden on defendants in medical
malpractice action because defendants were in better position to identify responsible party),
cert. denied, 423 U.S. 929 (1975).
68. See Marvel, Annotation, supra note 37, at 377-80 (citing minority of cases imposing
liability upon manufacturer for failure to warn of product-related risks regardless of scientific
knowability of risks at time product entered market).
Dean Keeton is the leading proponent of holding manufacturers liable for failure to warn
of risks that were scientifically unknowable at the time the product entered the market. See
W. Page Keeton, Products Liability-Inadequacy of Information, 48 TEx. L. REv. 398, 407-
09 (1970) (arguing that fact that manufacturer was excusably unaware of the extent of danger
should be entirely irrelevant in strict liability); see also Dawn Pelletier," Note, Is there a
Distinction Between Strict Liability and Negligence in Failure to Warn Actions?, 15 SUFFoLK
U. L. REv. 983 (1981) (arguing that courts should impose liability for failure to warn of
unknowable risks in order to maintain a distinction between strict liability and negligent failure
to warn).
69. 447 A.2d 539 (N.J. 1982), limited to itsfacts, Feldman v. Lederle Labs., 479 A.2d
374 (N.J. 1984).
70. See Beshada v. Johns-Manville Prods. Corp., 447 A.2d 539, 542 (N.J. 1982), limited
to its facts, Feldman v. Lederle Labs., 479 A.2d 374 (N.J. 1984).
71. See id. at 543.
72. See id. at 542.
73. See id. at 549. Surprisingly for a case in which a court was undertaking to establish
a distinction between strict liability and negligent failure to warn, the New Jersey Supreme
Court in Beshada considered the most important inquiry to be whether the imposition of
liability for failure to warn of unknowable risks would advance the policies sought to be
achieved by strict liability. See id. at 547. The court concluded that such liability would
advance three important policies. First, the rule would allocate the costs of accidents to the
manufacturers, who could best distribute those costs through pricing of the product. See id.;
see also infra note 91 (discussing and critiquing loss-shifting policy). Second, the rule would
impose on manufacturers the costs of falling to discover product dangers, which would create
1992] FAILURE TO WARN 1523

Rules imposing liability upon manufacturers for the failure to warn of

scientifically unknowable risks have been subject to a great deal of criti-
cism.7 4 At least one commentator has argued that such a rule reduces the
incentive for manufacturers to discover product-related risks after the man-
ufacturers have placed a product on the market. 7 Other concerns are that
this type of rule defeats the important goal of economic efficiency76 or that
it could operate to deprive consurfers of potentially valuable and beneficial
products. 77 Courts and commentators have argued that imposing liability
upon a manufacturer for the failure to warn of scientifically undiscoverable
risks is tantamount to imposing absolute liability and effectively places the
manufacturer in the role of insurer. 7 s The most widespread criticism, how-

an incentive for manufacturers to invest more time and money in safety research. See Beshada,
447 A.2d. at 547-48; see also infra note 92 (discussing safety-incentive policy). Finally, the
rule would preclude the highly confusing, costly, and time consuming process that would
necessarily be involved in litigating so complex an issue as the determination of scientific
knowability. See Beshada, 447 A.2d at 548.
74. See infra notes 75-81 and accompanying text (discussing criticisms of rules imposing
liability for failure to warn of scientifically undiscoverable risks). See generally William R.
Murray, Jr., Comment, Requiring Omniscience: The Duty to Warn of Scientifically Undiscov-
erable Product Defects, 71 GEO. L.J. 1635 (1983) (criticizing rules imposing liability for failure
to warn of undiscoverable risks, with special attention to New Jersey Supreme Court's decision
in Beshada); Robert D. Casale, Comment, Beshada v. Johns-Manville Products Corp.: Adding
Uncertaintyto Injury, 35 RUTGERS L. REv. 982 (1983) (same). But see Christopher M. Placitella
& Alan M. Darnell, Beshada v. Johns-Manville Products Corp.: Evolution or Revolution in
Strict ProductsLiability?, 51 FoRDHAM L. REviEw 801 (1983) (defending result in Beshada;
written by attorneys for plaintiffs in Beshada case).
75. See Victor Schwartz, The Post-Sale Duty to Warn: Two Unfortunate Forks in the
Road to a Reasonable Doctrine, 58 N.Y.U. L. RE,. 892, 897, 904 (1983) (arguing that if
manufacturer is held liable for failure to warn at point of sale regardless of whether risk was
unknown, manufacturer has less incentive to discover and warn of danger after product has
been sold); cf. infra note 92 (arguing that manufacturer is likely to decide on same level of
safety regardless of whether manufacturer is susceptible to strict liability or negligence).
76. See Henderson & Twerski, supra note 12, at 274 (arguing that manufacturers cannot
insure against unknown risks or include costs of unknown risks in price of product. As a
result, manufacturer must charge losses against earnings or capital or go out of business);
James A. Henderson, Jr., Coping with the Time Dimension in Products Liability, 69 CALIF.
L. REv. 919, 942-48 (1981) (discussing market distorting effects of rule imposing liability for
risks unknown at time manufacturer placed product on market).
77. See Feldman v. Lederle Labs., 479 A.2d 374, 387 (N.J. 1984) (asserting that if rule
imposing liability for failure to warn of scientifically unknowable risks were deemed to hold
generally, especially to cases involving drugs vital to health, court would not agree); Sales,
supra note 5, at 546 (arguing that manufacturers' concerns about imposition of liability for
unknowable dangers could deprive marketplace of valuable products or postpone entry of
beneficial products into market because of inordinate delay and testing).
78. See Anderson v. Owens-Coming Fiberglas Corp., 810 P.2d 549, 552, 555 (Cal. 1991)
(arguing strict liability never intended to be absolute liability and that rule imposing liability
for failure to warn of unknowable dangers is absolute liability which effectively places
manufacturer in role of insurer); Woodill v. Parke Davis & Co. 402 N.E.2d 194, 199 (11.
1980) (contending that holding manufacturer liable for failure to warn of impossible-to-know
danger would make manufacturer virtual insurer of product and that strict liability is not
equivalent of absolute liability); Sales, supra note 5 (noting that imposition of liability for
failure to warn of scientifically undiscoverable risk makes manufacturer an insurer).
1524 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

ever, is based upon fairness. The critics argue that it is extremely unfair to
require manufacturers to warn of risks that are unknown and technologically
unknowable.7 9 Even the New Jersey Supreme Court was unwilling to per-
petuate the harsh consequences of the rule it had set down in Beshada and
essentially overruled Beshada two years after it had rendered the decision. s0
In fact, the vast majority of courts have refused to adopt a rule that would
hold manufacturers accountable for the failure to warn of undiscoverable
dangers and have adopted the requirement of comment j that imposes
liability only if the manufacturer knew, or should have known, of the
product-related risk.8'

C. The Current Status of Strict Liability Failure to Warn

Even disregarding the troublesome distinctions that the courts have
created between strict liability and negligent failure to warn, a serious
problem remains. Not withstanding the fact that strict liability failure to
warn, as the courts have formulated it, is virtually indistinguishable from
the negligence action, plaintiffs generally are free to plead both theories in
the same case.12 Both courts and commentators have argued that allowing

79. See Henderson & Twerski, supra note 12, at 274 (assuming that basic postulate of
legal duty is that conduct which law seeks to induce is capable of being performed and arguing
that imposition of liability for failure to warn of unknowable risks is extremely unfair); Sales,
supra note 5, at 546 (contending that it is unfair to expect manufacturer to warn of risk that
is unknown and unknowable); Schwartz, supra note 75, at 879 (arguing that imposition of
liability for failure to warn of unknowable risks is unfair); ef. Feldman v. Lederle Labs., 479
A.2d 374, 387 (N.J. 1984) (stating that "[a] warning that a product may have an unknowable
danger warns of nothing").
80. See Feldman, 479 A.2d at 388 (restricting Beshada to circumstances giving rise to its
holding). The New Jersey Supreme Court stated in Feldman:
... A warning that a product may have an unknowable danger warns of nothing.
Neither [of two earlier cases] stated that the manufacturer would be deemed to know
of the dangerous propensity of the chattel when the danger was unknowable....
If Beshada were deemed to hold generally or in all cases, particularly with respect
to a situation like the present one involving drugs vital to health, that in a warning
context knowledge of the unknowable is irrelevant in determining the applicability
of strict liability, we would not agree.
Id. at 387. The court then listed a number of articles criticizing its Beshada holding. See id.
at 388 (listing articles criticizing Beshada). The Third Circuit later upheld the constitutionality
of applying the Beshada rule, in which knowledge of the risk is irrelevant, to asbestos cases
and the Feldman rule to all other warning cases. See In re Asbestos Litigation, 829 F.2d 1233
(3d Cir. 1987) (upholding constitutionality of treating warning cases involving asbestos differ-
ently from all other warning cases), cert. denied, 485 U.S. 1029 (1988).
81. See Marvel, Annotation, supra note 37, at 370-77 (citing cases in which courts have
required that manufacturer have knowledge, or reason to know, of product-related risk as
precondition for imposing strict liability for failure to warn).
82. See, e.g., Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1088-89, 1091
(5th Cir. 1973) (noting that plaintiff pleaded theories of negligence, strict liability, and implied
warranty), cert. denied, 419 U.S. 869 (1974); Payne v. Soft Sheen Prods., Inc., 486 A.2d 712,
716, 720 (D.C. 1985) (treating plaintiff's implied warranty claim as strict liability claim in case
in which plaintiff pleaded negligence and warranty theories); Jonescue v. Jewel Home Shopping
19921 FAILURE TO WARN 1525

plaintiffs simultaneously to plead two nearly identical theories of failure to

warn, rather than being advantageous in some small way, only serves to
complicate the litigation process and confuse both judges and juries. s3 A
number of critics of strict liability failure to warn have suggested that a
possible solution to the current dilemma would be to abandon the strict
liability action altogether. 4 If, in fact, courts cannot develop a sound strict
liability failure to warn action that is meaningfully distinct from negligence,
this argument has a great deal of merit. If little or no difference exists
between the two theories, it makes little sense to complicate the litigation
process by retaining both actions.
However, the critics have failed to consider other possible formulations
of strict liability failure to warn. They have assumed that the variations of
the doctrine that the courts have so far conceived are the only possible
variations. Merely because courts have failed thus far to develop an adequate
distinction between strict liability failure to warn and negligence does not
mean that such a distinction is impossible. 5 In fact, much of the courts'
inability to formulate a well-reasoned strict liability failure to warn doctrine
based upon sound policy is attributable to the fact that careful reasoning
and policy analysis have not been the predominate concerns of the courts;

Serv., 306 N.E.2d 312, 313 (I11.App. Ct. 1973) (noting'that plaintiff's failure to warn claim
included elements of both negligence and strict liability); cf. RESTATEMENT (SECOND) OF TORTs
§ 402A cmt. a. (explaining that strict products liability rule does not preclude theory of
negligence). But see Hauenstein v. Locite Corp., 347 N.W.2d 272, 275 (Minn. 1984) (holding
that because of potential confusion, when plaintiff alleges failure to warn, plaintiff may plead
either negligence or strict liability to jury but not both).
83. See Hauenstein, 347 N.W.2d at 275 (noting confusion that may result from mixing
negligence and strict liability where only basis for liability is failure to warn); Bromberg, supra
note 34, at 540 (asserting strict liability failure to warn serves to confuse both judges and
juries because of, among other things, its similarity to negligence); Henderson & Twerski,
supra note 12, at 276, 278 (noting that confusion results from coexistence of two such closely
related theories and arguing such confusion leads to bad law); Wade, supra note 2, at 850
(arguing that although each theory has advantages and disadvantages, confusion created by
similarity between them cannot be justified). The confusion created by the similarity between
strict liability and negligent failure to warn is evident in cases in which courts have reversed
a judgement for the plaintiff because of an inconsistent jury- finding that the manufacturer
was negligent but not strictly liable. See generally, John E. Theuman, Annotation, Products
Liability: Inconsistency of Verdicts on Separate Theories of Negligence, Breach of Warranty,
or Strict Liability, 41 A.L.R.4AT 9, 16-25 (1985 & Supp.) (analyzing cases, most of which are
failure to warn cases, that have considered whether jury's finding of negligence but not strict
liability constitutes inconsistency that requires reversal).
84. See Bromberg, supra note 34, at 540 (arguing that courts should remove'strict liability
from failure to warn law); Henderson & Twerski, supra note 12, at 271-78, 289 (arguing that
courts should abandon strict liability action in failure to warn context and that failure to warn
action should lie in negligence alone); cf. John W. Wade, On the Effect in Product Liability
of Knowledge UnavailablePrior to Marketing, 58 N.Y.U. L. REv. 734, 759-60 (1983) (arguing
that if court has not already recognized applicability of strict liability in failure to warn
actions, it should consider treating failure to warn cases as negligence cases).
85. See Aaron Gershonowitz, The Strict Liability Duty to Warn, 44 WAsH. & LEE L.
REv. 71 (1987) (proposing strict liability failure to warn doctrine that substantially differs
from current formulations).
1526 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

the primary concern has been the need to formulate a doctrine separate
from negligence. 86 Ironically then, the fact that the strict liability failure to
warn doctrine has been driven largely by the need to create a doctrine that
differs in some way from negligence has rendered unsuccessful the courts'
attempts to do just that-satisfactorily distinguish the strict liability action
from negligence.

III. POLICY CONSIDERATIONS AND DOCTRINAL PROBLEMS

Negligence language, concepts, and policies continue to dominate strict
liability failure to warn. 87 If the courts are to eliminate this troublesome
state of affairs, they must phrase the strict liability doctrine in its own
language and must firmly establish the doctrine in its own policy founda-
tion. 88 Constructing such a strict liability failure to warn doctrine will
necessarily require a careful examination of the policies that strict liability
failure to warn should seek to accommodate. 89
This policy analysis should begin with an examination of the policies
that courts have advanced as establishing the foundation for strict products
liability. 90 Two of the traditionally cited policies-loss shifting9l and incentive

86. See supra notes 55-81 and accompanying text (describing efforts of courts to develop
satisfactory distinction between strict liability and negligent failure to warn). But see Beshada
v. Johns-Manville Prods. Corp., 447 A.2d 539, 547-48 (N.J. 1982) (noting that most important
inquiry in formulating strict liability failure to warn rule is whether rule promotes important
policies), limited to its facts, Feldman v. Lederle Labs., 479 A.2d 374 (N.J. 1974). Although
the Beshada court should be commended for its attempt to formulate a strict liability rule
based upon policy analysis, the Beshada decision remains an example of the problems inherent
in rules based on ill-conceived policy. See supra note 73-81 and accompanying text (discussing
criticisms of Beshadaand rules imposing liability for failure to warn of scientifically unknowable
risks).
87. See supra notes 55-81 and accompanying text (arguing that attempts to formulate
distinct strict liability failure to warn doctrine have been unsuccessful).
88. See Leon Green, Strict Liability Under Sections 402A and 402B: A Decade of
Litigation, 54 TEx L. REv. 1185, 1193 n.14 (1976) (arguing that if strict liability is to survive,
doctrine must be phrased in its own language and dressed in its own clothing); David G.
Owen, Rethinking the Policies of Strict Products Liability, 33 V.ND. L. REa. 681, 683 (1980)
(stating that strict liability cannot be expected to move forward until policy goals are set).
89. See Owen, supra note 88, at 683-84 (arguing that products liability can not be
expected to move forward in acceptable fashion until policy goals are set, and only once policy
goals are set will it be possible to construct one or more strict products liability tests).
90. See id. at 684 (stating that best place to begin critical examination of product liability
policy is with policies said to support movement to strict liability in tort).
91. See Greenman v. Yuba Power Prods., Inc., 377 P.2d 897, 901 (Cal. 1963) (stating
that "purpose of [strict] liability is to insure that the cost of injuries resulting from defective
products are born by the manufacturers that put such products on the market rather than by
the injured persons who are powerless to protect themselves"); Henningsen v. Bloomfield
Motors, 161 A.2d 69, 81 (N.J. 1960) (stating that "burden of losses consequent upon use of
defective articles is borne by those who are in a position to ... make an equitable distribution
of the losses"). Commentators have severely criticized the loss-shifting rationale on a number
of grounds: That the rationale would justify absolute liability or even a comprehensive social
insurance scheme for the victims of all accidents, that the results dictated by the rationale
1992] FAILURE TO WARN 1527

for safer products 2-do not withstand close scrutiny and should, therefore,
play little or no role in laying the groundwork for any form of strict
products liability. 93 Another policy, the ,need to relieve the plaintiff of the
onerous burden of proving negligence, justifies, at most, shifting the burden
of proof on certain issues to the defendant9 and should, therefore, play
only a limited role in establishing a strict liability failure to warn doctrine.
The remaining policies-risk reduction 95 and marketplace hones ty9-should,

would often be unfair, and that imposition of liability in accordance with loss-shifting would
be economically inefficient. See generally Richard A. Epstein, Products Liability: The Search
for the Middle Ground, 56 N.C. L. Ray. 644-45, 659-661 (1978) [hereinafter Epstein, Products
Liability Search] (discussing and criticizing loss-shifting rationale); Owen, supra note 88, at
703-07 (same). Many commentators have argued that the loss-shifting rationale should be
abandoned altogether. See, e.g., EpsTni, supra note 3, at 46 (arguing that loss-shifting
arguments should receive no weight in formulation of tort doctrine); Epstein, ProductsLiability
Search, supra, at 661 (arguing that allowing loss-shifting issues to even covertly dominate tort
law will lead to unsound results, disrepute of lawyers, courts and tort system, and will
overburden tort system to point where it can no longer effectively function); Owen, supra note
88, at 705-06 (contending that loss-shifting policy should be abandoned).
92. See Beshada v. Johns-Manville Prods. Corp., 447 A.2d 539, 548 (N.J. 1982) (con-
cluding that imposition of liability for failure to warn of risks that were scientifically
unknowable at time product entered market will motivate manufacturers to develop safer
products and invest more in safety research), limited to its facts, Feldman v. Lederle Labs.,
479 A.2d 374 (N.J. 1984). Commentators have strongly criticized the safety-incentive justifi-
cation for strict products liability. See, 'e.g., EpsTEn, supra note 3, at 40-41 (arguing that
manufacturers will likely opt for same point of safety-that at which additional precautions
are no longer cost-effective-under both strict liability and negligence theories); Epstein,
Products Liability Search, supra note 91, at 659 (arguing that liability founded upon safety
rationale tends to deal as harshly with those sensible defendants who have exercised all possible
care in making their product safe as it does with those firms that are backward and
incompetent); Owen, supra note 88, at 709-10 (arguing that safety-incentive rationale is deficient
as decision-making tool). Furthermore, the important goal of optimum product safety is more
adequately achieved by the risk-reduction policy. See infra note 105 and accompanying text.
93. See Owen, supra note 88, at 684 (stating that policies that do not survive close
scrutiny should be discarded as justification for strict products liability); see also supra notes
91-92 (noting weaknesses in loss shifting and safety incentive policies).
94. See Cronin v. J.B.E. Olson Corp., 501 P.2d 1153, 1162 (Cal. 1972) (stating "The
very purpose of our pioneering efforts in [strict liability] was to relieve the plaintiff from
problems of proof in pursuing negligence."). The need to relieve the plaintiff of the difficult
burden of proving negligence may in some circumstances justify shifting the burden of proof
on certain issues. See Barker v. Lull Eng'g Co., 573 P.2d 443, 455 (Cal. 1978) (shifting to
manufacturer burden of proving that product utility outweighs product risk because manufac-
turer has superior knowledge); Feldman v. Lederle Labs., 479 A.2d 374, 388 (N.J. 1984)
(holding that as matter of policy, burden of proving scientific unknowability of risk should
be placed on defendant because defendant is in superior position to know technological data).
The policy does not, however, justify substantive differences. See Owen, supra note 88, at
710-11 (arguing that policy often misconstrued as justifying liability based upon manufacturer's
probable negligence); Griffith, Note, supra note 67, at 879 (stating that party's ignorance alone
should not be basis for imposing liability). Some commentators have argued that shifting the
burden of proof because of the consumer's ignorance, in relation to the manufacturer, on
certain issues is as appropriate in negligence as in strict liability. See authorities cited supra
note 67 (arguing that burden shifting appropriate in negligence if public policy so demands).
95. See infra notes 100-10 and accompanying text (describing risk-reduction policy).
96. See infra notes 111-19 and accompanying text (describing marketplace-honesty policy).
1528 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

on the other hand, play a significant role in the process of reformulating

the strict liability failure to warn.
Consideration of the risk-reduction and marketplace-honesty policies,
however, is not sufficient. Doctrinal problems that transcend the trouble-
some strict liability-negligence predicament currently plague both strict lia-
bility and negligent failure to warn. 97 These problems are, in part, attributable
to the failure of courts to give careful consideration to the various goals,
values, and concerns that have special significance in the area of failure to
warn.98 These policies need to be examined thoroughly so that they may
provide guidance in the reformulation of the strict liability failure to warn
doctrine. 9

A. Risk-Reduction Policy
The risk-reduction policy, stated simply, is that courts should place
liability upon the one who could best, but did not, avoid the accident or
reduce the risk.100 The risk-reduction rationale is based largely upon the
notion that with the developing complexity of modern products and the
corresponding complexity of potential dangers inherent in those products,
manufacturers are often in a better position than the consumer to discover,
evaluate, and act upon risks inherent in the'products they make. 01' Some-

97. See infra notes 192-209 and accompanying text (describing problems in failure to
warn law relating to causation issue); infra notes 128-53 and accompanying text (explaining
problems resulting from failure of courts to consider different functions that warnings serve
in formulating failure to warn liability rules); infra notes 154-72 and accompanying text
(discussing problems pertaining to use of same standard in design and warning cases). See
generally Henderson & Twerski, supra note 12 (discussing many of the problems currently
afflicting failure to warn law).
98. See infra notes 120-209 and accompanying text (discussing policies and concerns
specifically relevant to failure to warn and problems that have resulted, or may result, from
failure of courts to adequately consider these policies).
99. See infra notes 120-209 and accompanying text (summarizing policies that should be
considered in reformulating strict liability failure to warn doctrine).
100. See Escola v. Coca Cola Bottling Co., 150 P.2d 436, 440 (Cal. 1944) (Traynor J.,
concurring) (stating that courts should fix responsibility "wherever it will most effectively
reduce the hazards of life and health inherent in defective products that reach the market");
Henningsen v. Bloomfield Motors, 161 A.2d 69, 81 (N.J. 1960) (stating that "burden of losses
consequent upon use of defective articles is borne by those who are in a position to ...
control the danger"); Britain, supra note 39, at 349-50 (noting that risk-reduction policy,
which author calls "injury prevention," places burden of loss on party that can most efficiently
and affordably prevent injury); Owen, supra note 88, at 712 (stating that risk-reduction policy,
which author calls "risk control," places loss on party in best position to avoid accident); cf.
Guido Calabresi and John T. Hirschoff, Toward a Test for Strict Liability in Torts, 81 YALE
L.J. 1055, 1060 (1972) (proposing strict liability rule that places loss on "cheapest cost
avoider"-i.e., party "in the best position to make the cost-benefit analysis between accident
costs and accident avoidance costs and to act on that decision once it is made") (emphasis
omitted).
101. See Escola, 150 P.2d at 441, 443 (Traynor J., concurring) (asserting that manufacturer
is in better position to understand its own manufacturing processes because those processes
are either inaccessible to or beyond the understanding of consumers); Feldman v. Lederle
19921 FAILURE TO WARN 1529

times, however, the consumer is better able to control the risks that manifest
themselves, 10 2 as is often the case when the manufacturer has given directions
for the safe use of a product.' 3
Placing liability upon the party who is in the best position to control
the risk seems intuitively fair'04 and also provides incentive to all parties
involved to achieve the optimum level of safety. 0 The risk-reduction policy
may also provide protection against economic waste.'06 Thus, the risk-

Labs., 479 A.2d 374, 388 (N.J. 1984) (noting that manufacturer is in superior position to
know technological data relating to manufacturer's products); Owen, supra note 88, at 711
(explaining that complex and hidden dangers inherent in modern products are normally beyond
skill and understanding of consumers).
102. See Britain, supra note 39, at 349-50 (noting that in some circumstances consumer's
capacity to prevent injury exceeds that of manufacturer); Calabresi & Hirschoff, supra note
100, at 1064, 1071-72 (explaining that misuser or third parties are often in better position than
manufacturer to evaluate risks); Epstein, Products Liability Search, supra note 91, at 658-59
(noting that in many situations incentives to maintain safety are better placed on others in
chain of distribution). Professor Owen offers the example of a consumer driving a car over a
rocky stream bed at great speed. Owen, supra note 88, at 712. In such a case, the consumer
is best informed and best able to control the resulting risks. Id.
103. See infra note 128 and accompanying text (discussing risk-reduction function of
warnings).
104. See Owen, supra note 88, at 712 (stating that liability based on risk-reduction policy
seems intuitively fair).
105. See EpsTmN, supra note 3, at 41 (noting that liability rules will achieve optimum
level of safety by not only placing incentive on manufacturer to make safer products, but also
by placing incentive on other parties in chain of distribution to maintain and use product
safely); Britain, supra note 39, at 348-50 (explaining that risk-reduction policy, which-author
calls "injury prevention," provides incentives for both manufacturer and consumer to optimize
level of safety). It is important to note that the safety incentive inherent in the risk-reduction
policy is fundamentally different than the safety-incentive policy standing alone. See supra
note 92 (discussing weaknesses of safety-incentive policy). Liability rules based upon the safety-
incentive policy are one-directional, leading inevitably to the imposition of liability upon the
manufacturer. Owen, supra note 88, at 709. Such a rule arguably provides less incentive for
maintaining safety to others in the chain of distribution. See EpsTEIN, supra note 3, at 41
(arguing that liability rules based upon imposing high incentive for safety on manufacturer
encourages others to use less care in maintaining product safety). The risk-reduction policy,
on the other hand, places the safety incentive on the party best able to control the risk. Supra
notes 100-03 and accompanying text. Furthermore, the safety incentive policy, standing alone,
would encourage the marketing of products that are "too safe," which would cause an
undesirable rise in product costs. Griffith, Note, supra note 67, at 877. Implicit in the risk-
reduction policy, on the other hand, is reduction of risks to an affordable level. Britain, supra
note 39, at 349; see Owen, supra note 88, at 710 (arguing that if deterrence to a point is
underlying notion of safety-incentive rationale, then safety rationale would simply "mask"
other policies necessary in fixing that point).
106. See Owen, supra note 88, at 712 (noting that liability based upon risk-reduction
policy may promote economic efficiency). Professor Owen notes that risk-reduction seems little
different from Professor Calabresi's "cheapest cost avoider" economic model, which is arguably
the most efficient test of liability. Id. Professor Calabresi proposes a strict liability test for
product-related injuries that "requires of [courts] only a decision as to which of the parties
to the accident is in the best position to make the cost-benefit analysis between accident costs
and accident avoidance costs and to act on that decision once it is made." Calabresi &
Hirschoff, supra note 100, at 1060. In other words, "[tihe question for the court reduces to
1530 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

reduction rationale may be very helpful in identifying who should bear the
loss; the loss should be borne by the party that was in the best position to
reduce or control the risk.'°7 Allocating the loss in accordance with the risk-
reduction policy alone, however, may occasionally conflict with allocations
of loss in accordance with other policies, such as marketplace honesty, 08
or with certain "rights" that deserve protection, such as the "right" of
consumers to make an informed choice concerning product risks. 0 9 Accord-
ingly, although the risk-reduction policy has a great deal of potential for
helping to structure products liability rules, other policies will have to play
a role as well."10

B. Marketplace-Honesty Policy
Another traditional justification for the strict products liability doctrine
is a representational or marketplace-honesty theory."' The theory is based
on the notion that by placing a product on the market and packaging and
advertising the product, the seller impliedly represents to the consumer that

a search for the cheapest cost avoider." Id. Professor Calabresi argues that the "cheapest cost
avoider" test is the most efficient liability test because, among other things, it involves lesser
administrative costs than do other tests and focuses on minimizing accident costs and accident
avoidance costs rather than on accident avoidance. Id. at 1074-76 & n.74. Professor Calabresi
does consider, however, that goals besides efficiency, such as distributional goals and fairness
concerns, may conflict with and often predominate over efficiency considerations and that
these goals, too, will likely influence liability decisions. See.id. at 1076-85 (describing relation-
ship between efficiency model and other goals).
107. See sources cited supra note 100 (noting that under risk-reduction policy, loss should
lie with party in best position to control risk).
108. See Britain, supra note 39, at 351-52 (discussing tensions between risk-reduction
policy and marketplace-honesty policy); Owen, supra note 88, at 712 (noting that in some
circumstances other policies "loom larger" than risk-reduction policy). In fact, the marketplace-
honesty policy "looms larger" in failure to warn cases than does risk reduction. See Gershon-
owitz, supra note 85, at 72 passim (arguing that marketplace honesty is dominate policy in
failure to warn context); infra notes 159-68 and accompanying text (discussing role of
marketplace-honesty policy in warning cases); see also infra notes 111-19 and accompanying
text (discussing marketplace-honesty policy).
109. See infra notes 150-53 and accompanying text (discussing consumers' "right" to
make informed choice and arguing that risk-reduction policy is incapable of adequately
protecting that right); see also Owen, supra note 88, at 712-13 (noting conflict between risk-
reduction policy and "rights" that deserve protection).
110. See Britain, supra note 39, at 408-22 (arguing that marketplace-honesty policy should
assume coequal role with risk-reduction policy in products liability law); Owen, supra note 88,
at 713 (explaining that values in addition to risk reduction will have to inform decisions).
111. See Greenman v. Yuba Power Prods., Inc., 377 P.2d 897, 901 (Cal. 1962) (justifying
imposition of strict products liability by reference to marketplace-honesty policy). For an
extensive, 279 page evaluation of the representational theory, see Marshall S. Shapo, A
Representational Theory of Consumer Protection: Doctrine, Function and Legal Liability for
ProductDisappointment, 60 VA. L. REv. 1109 (1974). For a less extensive, yet comprehensive,
exploration of the appropriate role of marketplace honesty in products liability law, see Britain,
supra note 39. See also Green, supra note 88 (proposing "communicative tort" based upon
marketing theory supported by §§ 402A and 402B of the Restatement).
1992] FAILURE TO WARN

the product is safe for use."12 These implicit representations in turn create
expectations in consumers of product safety."' Sellers should not be able
to escape liability for injuries caused by their defective products4 once they
have sought and built consumer confidence in those products."
The marketplace-honesty policy contains two goals that are worthy of
promotion in products liability law. First, because sellers usually intend to
convey an impression to potential consumers that their products are safe to
use," 5 liability rules should obligate sellers to stand behind their products
when they have made such representations." 6 Second, the marketplace-
honesty policy promotes consumer autonomy because consumers can better
order their lives as they see fit if they are assured protection, at least in
the form of compensation, against unexpected product risks.1"1 In other

112. See Greenman, 377 P.2d at 901 (asserting that implicit in product's presence on
market is representation that it will function properly); Escola v. Coca Cola Bottling Co., 150
P.2d 436, 443 (Cal. 1944) (Traynor J., concurring) (noting efforts of manufacturers to build
consumer confidence in products through advertising and marketing devices such as trade-
marks); Britain, supra note 39, at 348 (stating that marketplace-honesty policy is founded on
notion that through advertising and other marketplace inducements, manufacturers explicitly
and implicitly portray product safety); Green, supra note 88, at 1185 (describing communicative
process in which manufacturer makes implied representations of safety); Shapo, supra note
111, at 1325-34. (discussing at length different ways in which manufacturers make both
unqualified assurances and general implied representations of product safety through media
ranging from "fanfare of advertising to sober detail of instruction manuals").
113. See Greenman, 377 P.2d at 901 (observing that consumers rely on manufacturer
representations); Escola, 150 P.2d at 443 (Traynor J., concurring) (noting that consumers rely
on confidences built up by manufacturer through advertising and other marketing devices);
Henningsen v. Bloomfield Motors, Inc., 161 A.2d 69, 83 (N.J. 1960) (arguing that under
modern conditions, consumers have little choice but to rely on "the importuning of colorful
advertising" by manufacturers); RFsTATEMENT (SECOND) OF TORTS § 402A cmt. c (1965) (noting
that consumer is forced to rely on seller); Green, supra note 88, at 1190 (describing process
in which marketing techniques ripen into consumer confidences and arguing that seller cannot
escape liability once it has created such confidences).
114. See Escola, 150 P.2d at 443 (Traynor J., concurring) (arguing that marketing practices
ripen into manufacturer obligation); RESTATEMENT (SECOND) OF TORTS § 402A cmt. c (1965)
(asserting that by marketing product, manufacturer undertakes special obligation to consuming
public); EPSTEIN, supra note 3, at 93-94 (noting that under representational theory, "gist" of
consumer's case is estoppel by misrepresentation); Green, supra note 88, at 1190-91 (describing
process by which manufacturer's representations of product safety ripen into obligation to
consumers).
115. See Escola, 150 P.2d at 443 (Traynor J., concurring) (noting efforts of manufacturers
to build consumer confidence in products through advertising, use of trademarks, and increas-
ingly high standards of inspection); Owen, supra note 88, at 708 (stating that manufacturers
usually intend to create impression that products are safe for use); cf. EPSTEIN, supra note 3,
at 94 (explaining that even if manufacturer has said nothing about product, product itself, by
its very appearance, makes implied representations about its use and limitations).
116. See RESTATEMENT (SECOND) OF TORTS § 402A cmt. c (1965) (stating that public has
right to expect reputable sellers to stand behind their goods); PROSSER, supra note 1, § 97, at
650 (stajing practice of reputable manufacturers is to stand behind their goods); Britain, supra
note 39, at 348 (contending that manufacturers should be responsible when expectations they
have intentionally created are frustrated); Green, supra note 88, at 1191 (stating that law
should take sellers at their word).
117. See sources cited infra note 118.
1532 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

words, consumer autonomy is best promoted if consumers can be certain

that the seller has informed them of all product-related risks or that the
seller will compensate them for injuries caused by risks about which the
seller has not provided adequate information."" Because marketplace honesty
promotes these important goals, the policy should play a significant role in
laying the foundation for products liability law. 19 Moreover, for reasons to
be discussed below, the marketplace-honesty policy is especially important
in the context of the failure to warn.

C. Insignificant Risks and the Proliferation of Warnings

Every product has some degree of risk. Courts could formulate a test
for failure to warn that would require manufacturers to warn of the most
insignificant product risks or risks with little probability of occurring 20
Such a rule inevitably would lead to a proliferation, of warnings, which
would have extremely negative consequences. One result of such a rule is
that manufacturers would have to issue a "laundry list" of warnings with
the sale of every product.' 2 ' Such a "laundry list" could contain a large
number of warnings against relatively insignificant risks, which would tend
to dilute the effectiveness of more important warnings accompanying the
same product. 122 When faced with a "laundry list" of warnings, consumers

118. See Britain, supra note 39, at 369 (arguing that marketplace-honesty policy promotes
enhanced flow of information to consumers, which facilitates individual integrity that comes
from exercise of meaningful choice); Owen, supra note 88, at 709 (noting that implied
representations policy promotes consumer autonomy because consumers are protected, in form
of compensation, against unexpected calamity caused by defective products). For a general
discussion of the law's treatment of personal autonomy, see Joseph Goldstein, For Harold
Lasswell: Some Reflections on Human Dignity, Entrapment, Informed Consent and the Plea
Bargain, 84 YALE L.J. 683 (1975).
119. See Britain, supra note 39, at 408-22 (arguing that marketplace honesty should take
on coequal role with risk reduction in products liability law); Green, supra note 88 (arguing
that representational theory should be basis for communicative tort applicable to product
defects); Owen, supra note 88, at 709 (concluding that representational theory should be
accorded significant respect in products liability law); Shapo, supra note 111 (discussing
significance of representational theory in products liability law).
120. See Moran v. Faberge, Inc., 332 A.2d 11, 13, 20 (Md. 1975) (holding that manu-
facturer of cologne is required to warn of risk that misusing cologne by slowly dripping it
onto flaming candle in order to make candle "scented" could result in burst of fire and burns
to individual). The Moran court explained that the cost of supplying a warning is so minimal
that the cost-benefit balancing process will almost always weigh in favor of an obligation to
warn. Id. at 15; see also Ross Labs. v. Thies, 725 P.2d 1076, 1079 (Alaska 1986) (arguing
that costs of adding warning are so minimal that balance must always be struck in favor of
requiring warning).
121. See Aaron D. Twerski et al., The Use and Abuse of Warnings in ProductsLiability-
Design Defect Litigation Comes of Age, .61 CORNELL L. Rav. 495, 516 (1976) (arguing that
failure to warn rule imposing duty to warn of remote and insignificant risks would force
manufacturers to issue a "laundry list" of warnings with every product).
122. See Finn v. G.D. Searle & Co., 677 P.2d 1147, 1153 (Cal. 1984) (noting inevitable
diluting effect that would result from inundating physicians with warnings of any and every
hint of product danger); Gershonowitz, supra note 85, at 103 (noting that warnings of
19921 FAILURE TO WARN 1533

likely would disregard or overlook important warnings concerning significant

product risks.'1
A related and more troublesome possibility is that an undue proliferation
of warnings would have what has been termed a "wolf-crying" effect. 24
The constant repetition of insignificant warnings on nearly every product
could cause consumers to become "immune" to all warnings on all products,
no matter how significant the risk. 21 Under these circumstances, consumers
would have no way to distifiguish insignificant warnings from significant
26
ones and would begin to disregard product warnings altogether.
To be effective, then, warnings should call attention only to significant
dangers that have a real probability of occurring. 27 Thus, any test for strict
liability failure to warn should take into consideration the adverse effects
of a rule that would force manufacturers to warn of every trivial product-
related risk.
D. The Informed-Choice and Risk-Reduction Functions of Warnings
Warnings serve two functions. One function a warning may serve is
risk reduction. Risk-reduction warnings direct consumers to use a product
in a particular safe way or allow consumers to behave more carefully than
they would if they remained ignorant of the product-related risks. 2 For

insignificant risks tend to reduce effectiveness of potentially helpful warnings against significant
risks); Henderson & Twerski, supra note 12, at 296 (same); Twerski et al., supra note 121, at
516-17 (same).
123. See sources cited supra note 122.
124. See Twerski et al., supra note 121, at 514 (defining "wolf-crying" effect as the
negative consequences resulting from constant repetition of insignificant warnings).
125. See Henderson & Twerski, supra note 12, at 296 (claiming that consumers, if
bombarded by useless warnings, may stop heeding warnings altogether); Twerski et al., supra
note 121, at 514-15 (noting potential for all warnings to come into disrepute if manufacturers
were required to "oversell" insignificant or trivial risks).
126. Twerski et al., supra note 121, at 515.
127. See Finn v. G.D. Searle & Co., 677 P.2d 1147, 1153 (Cal. 1984) (contending that
significance of warning is relevant to issue of when warning-is required in order to avoid
diluting effect of over warning); Twerski et al., supra note 121, at 514 (noting that to be
effective, warnings must be selective and warn against risks with real probability of occurring);
cf. Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 847 (N.H. 1978) (explaining that requiring
warnings against unforeseeable use and absurd misuse may harm state's economy).
A number of commentators have observed that courts generally impose an obligation on
manufacturers to warn only of significant risks. See, e.g., Kidwell, supra note 38, at 1395
(observing that courts generally impose liability only for failure to warn of "material" or
"significant" risks); Little, supra note 5, at 997 (same); Madden, supra note 5, at 234-35
(same).
128. See, e.g., Finn, 677 P.2d at 1152 (stating that some warnings instruct consumer on
how product should be safely used); Palmer v. Avco Distributing Corp., 412 N.E.2d 959, 964
(11. 1980) (holding that warning of danger of catching leg in agitator mechanism of fertilizer
spreader must be adequate to perform risk-reduction function); Glittenberg v. Doughboy
Recreational Ind., Inc., 462 N.W.2d 348, 365 (Mich. 1990) (noting that warning not to dive
into shallow end of swimming pool could have reduced chances of injury by allowing plaintiff
to alter behavior); see also Henderson & Twerski, supra note 12, at 285-86 (defining risk-
reduction function of warnings); Twerski et al., supra note 121, at 520-21 (same).
1534 WASHINGTON AND LEE LA W REVIEW [Vol. 49:1509

example, a warning on a lawn mower may warn users to keep their feet
and hands away from the rotating blade. The second function that a warning
may serve is informed choice. Pure informed-choice warnings inform con-
sumers of risks that consumers cannot reduce by any amount of safe or
careful conduct. 29 For example, some types of prescription drugs may
inevitably injure a certain percentage of users. 30 The consumer can do
3
nothing, short of foregoing use of the drug altogether, to reduce the risk.' '
Thus, a warning would be merely informative in nature.' 3 2 This informed-
choice warning would, however, promote consumer autonomy by providing
consumers with information so that they may choose whether or not they
wish to encounter unavoidable product-related risks.' The informed-choice
warning doctrine is analogous to the doctrine of informed consent in medical
malpractice law. 3 4 Just as consumer autonomy is the underlying value of

129. See Finn, 677 P.2d at 1152 (stating that some warnings inform consumers of potential
risks or side effects that may follow foreseeable use of product); Ortho Pharmaceutical Corp.,
v. Chapman, 388 N.E.2d 541, 555 (Ind. Ct. App. 1979) (explaining that warning of drug side
effects would not reduce risk but would simply inform user of risk); see also Henderson &
Twerski, supra note 12, at 285 & n.88 (defining informed choice function of warnings); Twerski
et al., supra note 121, at 514 (same).
130. See, e.g., Davis v. Wyeth Labs., Inc., 399 F.2d 121, 123-24 (9th Cir. 1968) (discussing
use of Sabin type III polio vaccine, which involves small but definite risk of causing user to
contract polio virus); Chapman, 388 N.E.2d at 554, 555 (involving oral contraceptive which
tended to increase incidence of blood clotting in some women; warning would not have reduced
risk); Feldman v. Lederle Labs., 479 A.2d 374 (N.J. 1984) (involving tetracycline drugs, which
tend to discolor teeth of users).
131. See sources cited supra note 129 (explaining that consumers cannot alter behavior to
reduce certain types of risks).
132. See sources cited supra note 129 (explaining purely informational value of informed-
choice warnings).
133. See Borel v. Fibreboard Paper Products Corp., 493 F.2d 1076, 1089 (5th Cir. 1973)
(stating that consumers are entitled to information so that they may make own choice as to
whether product benefits justify exposure to risk), cert. denied, 419 U.S. 869 (1974); Davis,
399 F.2d at 129, 131 (asserting importance of allowing consumer opportunity to make voluntary
and informed choice); Finn v. G.D. Searle & Co., 677 P.2d 1147, 1152 (Cal. 1984) (describing
function of informed choice warnings as affording consumer opportunity to make informed
choice) (citing Frank M. McClellan, Strict Liability for Drug Induced Injuries: An Excursion
Through the Maze of Products Liability, Negligence and Absolute Liability, 25 WAYNE L.
REv. 1, 32 (1978)); Chapman, 388 N.E.2d at 555 (stating that informed choice warning informs
user so that user may choose whether risk should be incurred at all).
134. See Cunningham v. Charles Pfizer & Co., 532 P.2d 1377, 1382 (Okla. 1974) (citing
informed-consent cases as support for adoption of objective causation test in failure to warn
context); Britain, supra note 39, at 368-402 (analogizing informed-choice warning doctrine to
informed-consent doctrine); Henderson & Twerski, supra note 12, at 286-89 (same). Informed-
consent doctrine concerns a physician's duty to inform patients of risks involved in proposed
medical procedures. Physicians are generally liable for injuries resulting from medical proce-
dures if the patient was not supplied with sufficient information to effectively consent to the
procedure. See Canterbury v. Spence, 464 F.2d 772, 780-83 (D.C. Cir.), cert. denied, 409 U.S.
1064 (1972). See generally Martin R. Struder, The Doctrine of Informed Consent: Protecting
the Patient's Right to Make Informed Health Care Decisions, 48 MONT. L. REv. 85 (1987)
(tracing development of informed-consent law and explaining current state of law). Informed-
consent claims concede that the physician's conduct was adequate. Id. at 85. Consequently,
1992] FAILURE TO WARN 1535

informed-choice warnings, the predominate policy underlying the informed-

consent doctrine is individual autonomy and the patient's right to choose. 3
Warnings very often serve both the risk-reduction and informed-choice
functions. 36 For example, a certain drug may react adversely with a com-
monly consumed substance, such as milk. A proper warning would serve
the risk-reduction function by providing the consumer with necessary infor-
mation so that the consumer may avoid ingesting the drug and drinking
milk at the same time. The same warning would serve the informed-choice
function by allowing milk lovers the opportunity to weigh the cost of giving
up milk with the benefits of taking the drug.
The primary policy underlying the risk-reduction function of warnings
is, not surprisingly, the risk-reduction policy. 37 The informed-choice func-
tion, on the other hand, implicates. the marketplace-honesty policy because
the marketplace-honesty policy best promotes consumer autonomy.' 3 Some
commentators have argued that because risk-reduction warnings and in-
formed-choice warnings implicate different policies, courts should develop
a separate analysis for the two types of warnings. 39 This argument overlooks
the fact that many warnings serve both functions in varying degrees. '4
Separate analysis would require one test for informed-choice warnings,
another test for risk-reduction warnings, and some hybrid test for the large
number of cases in which the warning serves both functions. Courts have
been reluctant to establish such separate analytical frameworks.' 4'

risk reduction is not involved. This is also the case with situations involving informed-choice
warnings-a warning does noi make use of the product any more safe.
135. See Canterbury,464 F.2d at 780 (asserting importance of freedom of choice); Cobbs
v. Grant, 502 P.2d 1, 9 (Cal. 1972) (noting importance of individual integrity).
136. Henderson & Twerski, supra note 12, at 286. Professors Henderson and Twerski
offer two examples of situations in which warnings serve both the risk-reduction and informed-
choice functions. First, if the warning provides information that allows the user to take safety
precautions to reduce the risk, the user may also decide to decrease the level of usage to avoid
residual risks. Id. Second, even in the case of a "pure" informed-choice warning that informs
the user that "one out of one million people who take this drug become blind as a result,"
the consumer who decides to take the drug, notwithstanding the risk, can be on the alert for
early symptoms of blindness. Id.
137. See id. at 285-86 (explaining that risk-reduction policy underlies risk-reduction func-
tion of warnings); see also supra notes 100-10 and accompanying text (discussing risk-reduction
policy).
138. See Britain, supra note 39, at 374-402 (arguing that prescription drug cases involving
informed-choice warnings cannot be justified on risk reduction grounds but only under
marketplace-honesty policy); see also supra notes 117-18 and accompanying text (noting that
marketplace honesty promotes consumer autonomy).
139. See Twerski et al., supra note 121, at 519-21 & n.59 (arguing that different functions
require different analysis and that informed-choice doctrine may require manufacturer to warn
of less significant risks and may involve different causation inquiry).
140. See supra note 136 and accompanying text (discussing warnings that serve both risk-
reduction and informed-choice function).
141. See infra note 142 and accompanying text (noting that courts generally employ same
analysis for all types of warnings).
 1536 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

Rather than establish separate tests for the different types of warnings,
courts have most often applied an analysis firmly based upon the risk-
reduction policy to all warnings. 42 Tests based upon risk-reduction policy,
the most common of which is the risk-utility test, 43 are inappropriate for
use in informed-consent cases.' 44 For example, in the case of "unavoidably
unsafe products,"' 45 such as beneficial prescription drugs that will unavoid-
ably injure a certain percentage of users, courts often hold that the lack of
an informed-choice warning renders the product "unreasonably danger-
ous."' 46 The problem with this analysis is that no safety precautions on the
part of the consumer can make use of the product any more safe. 47 Through
faulty analysis, however, the courts somehow reach the absurd conclusion
that although the product will have precisely the same level of risk regardless
of whether the manufacturer provides a warning, it is "reasonably danger-
ous" with a warning and "unreasonably dangerous" without one. 48 Simi-
larly, the application of the risk-utility test to cases involving "unavoidably
unsafe" products should result in no liability, regardless of whether the
manufacturer provides a warning, because supplying a warning would not
result in any reduction in product risk or increase in product utility. 49

142. See Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1087-89 (5th Cir. 1973)
(applying risk-utility analysis in informed-choice context), cert. denied, 419 U.S. 869 (1974);
Davis v. Wyeth Labs., Inc., 399 F.2d 121, 129-30 (9th Cir. 1968) (applying "unreasonably
dangerous" analysis in informed-choice context); see also Finn v. G.D. Searle & Co., 677
P.2d 1147, 1152 (Cal. 1984) (noting tendency of courts to lump all warnings together under
one analysis).
143. See Britain, supra note 39, at 362, 415 (observing that risk-utility test is based upon
risk-reduction policy); Gershonowitz, supra note 85, at 72 (contending that risk-utility test
well-serves risk-reduction policy, but not marketplace-honesty policy); Keeton, supra note 2,
at 592-93 (describing risk-reduction features of risk-utility test); see also supra notes 39-43 and
accompanying text (describing risk-utility test).
144. See Britain, supra note 39, at 390-93 (criticizing courts' use of risk-reduction analysis
in informed-choice context); Twerski et al., supra note 121, at 519-20 (same).
145. See supra note 33 (describing "unavoidably unsafe products" as defined in comment
k of § 402A).
146. See Borel, 493 F.2d at 1089 (holding that failure to provide informed-choice warning
rendered product unreasonably dangerous); Davis, 399 F.2d at 129-30 (same); see also sources
cited supra note 144 (criticizing courts' use of risk-reduction analysis in informed-choice
context).
147. See supra notes 129-33 and accompanying text (noting that, by definition, pure
informed-choice warnings cannot provide consumer with information that will help to reduce
risk).
148. See cases cited supra note 146 (holding that failure to provide informed-choice
warning rendered otherwise reasonably dangerous product unreasonably dangerous); sources
cited supra note 144 (criticizing courts' use of risk-reduction analysis in informed-choice
context).
149. See sources cited supra note 144 (explaining that risk-utility analysis does not justify
liability in informed-choice context). Curiously, in the informed-choice context, the argument
for no liability would appear to be even better under a risk-utility analysis when the manu-
facturer does not supply a warning because the giving of a warning adds additional cost with
no corresponding decrease in product risk or increase in product utility.
1992] FAILURE TO WARN 1537

Consumers are entitled to the information necessary to help them make

an informed choice regarding product risks. 50 Courts, therefore, have
intuitively managed to reach the right result in informed-choice cases, but
only through the application of extremely flawed analysis.' The danger
remains that a court or jury could reach the wrong result in an informed-
choice situation by correctly applying risk-reduction principles. Because few,
if any, warnings serve the risk-reduction function alone,5 2 the "proper"
application of risk-reduction analysis would frustrate the important goal of
consumer autonomy in many, if not all, failure to warn cases.' The courts
should formulate a test that accommodates the importance of consumer
autonomy in the warning context and that applies equally to risk-reduction,
informed-choice, and hybrid warnings.

E. Different Policy Considerations Underlying Failure to Warn and

Design Defects and the Relationship Between Design and Warnings
Much of the doctrinal confusion relating to strict liability failure to
warn5 4 can be attributed to the misconception that courts should accord
equal weight to the same policies in both design defect and failure to warn

150. See, e.g., Reyes v. Wyeth Labs., 498 F.2d 1264, 1294 (5th Cir.) (referring to right
of individual to choose and control what risks individual will take), cert. denied, 419 U.S.
1096 (1974); Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1089 (5th Cir. 1973)
(asserting that consumers are entitled to make their own choice as to whether product's benefit
justifies exposure to risk), cert. denied, 419 U.S. 869 (1974); Davis v. Wyeth Labs., Inc., 399
F.2d 121, 123-24 (9th Cir. 1968) (holding that if risk qualitatively and quantitatively is such
as to call for true choice judgement, warning must be given). Justice Cardozo's famous
statement in an early informed-consent medical malpractice case is on point: "Every human
being of adult years and sound mind has a right to determine what shall be done with his
own body." Schloendorff v. Society of N.Y. Hosp., 105 N.E. 92, 93 (N.Y. 1914), overruled
by, Bing v. Thunig, 143 N.E.2d 3 (N.Y. 1957); see also Canterbury v. Spence, 464 F.2d 772,
780 (D.C. Cir.) (discussing right of patients to information necessary in making an informed
decision and quoting Justice Cardozo's statement in Schloendorfj), cert. denied, 409 U.S. 1064
(1972).
151. See Britain, supra note 39, at 389-98 (arguing that informed-choice warning decisions
cannot be justified on basis of risk-reduction analysis employed by courts, but that decisions
can be justified on basis of marketplace-honesty policy); Twerski et al., supra note 121, at
519-20 (noting that Davis court was perhaps correct in concluding that consumer was entitled
to information, but that court's analysis does not support that conclusion).
152. See supra note 136 and accompanying text (observing that many warnings serve both
risk-reduction and informed-choice functions); infra note 162 and accompanying text (arguing
that consumer autonomy is implicated, to some degree, by all warnings).
153. See supra notes 144-49 and accompanying text (noting that risk-reduction analysis
should come out in favor of no liability in informed choice context regardless of whether
warning is supplied); supra note 150 and accompanying text (contending that consumers are
entitled to information necessary in making informed choice as to whether to encounter product
risks).
154. See supra part II (discussing problematic doctrinal development of strict liability
failure to warn); supra notes 142-53 and accompanying text (describing problems inherent in
applying risk-utility analysis to informed-choice warnings); infra notes 192-204 and accompa-
nying text (explaining difficulties relating to causation issue in failure to warn context).
1538 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

cases.' 55 A number of commentators have argued that risk reduction is the

primary goal in both warning and design cases. 5 6 The courts would seem
to be in accord because the prevailing analysis in both warning and design
cases centers on risk-utility balancing, 5 7 which is a test firmly grounded in
the risk-reduction policy. 58
Contrary to the views of these commentators and the prevailing analysis
of the courts, marketplace honesty, and not risk reduction, should be the
predominate policy in the law of warnings. 5 9 As shown in the previous
section, a large number of warnings involve notions of informed choice,6
under which consumer autonomy rather than risk reduction is the primary
concern.' 6' In fact, even the purest risk-reduction warning implicates con-

155. See, e.g., Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1035 & n.2 (Or. 1974)
(treating failure to warn as subcategory of design defect) (citing Wade, supra note 2, at 830);
Powers, supra note 57, at 783 n.25 (arguing that same standards should apply in both design
defect and failure to warn contexts); Twerski et al., supra note 121, at 510, 513, 517 (arguing
that same analysis should apply to design and warning cases because design and warnings
must be blended to obtain optimum level of safety).
156. See, e.g., Henderson & Twerski, supra note 12, at 271-78, 304 (arguing that failure
to warn doctrine should be fault-based, risk-reduction tort-i.e. that failure to warn actions
should lie in negligence alone); Keeton, supra note 68, at 403-04 (contending that questions
of warnings and design are inseparable and advocating a negligence-with-imputed-knowledge
test to determine if product is "unreasonably dangerous"); Twerski et al., supra note 121, at
517, 524 (advocating that courts apply risk-reduction-based risk-utility analysis to design and
warning claims because design and warning questions are inextricably woven together); Wade,
supra note 2, at 830, 834-38 (grouping design and warning defects together and proposing 7-
factor balancing test replete with risk-reduction concepts). But see Gershonowitz, supra note
85, at 72 passim (arguing that risk reduction is primary policy in design defect context, but
marketplace honesty is policy most heavily implicated by failure to warn claims).
157. See Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1087 (5th Cir. 1973)
(applying risk-utility test in failure to warn context), cert. denied, 419 U.S. 869 (1974);
Anderson v. Owens-Corning Fiberglas Corp., 810 P.2d 549, 554 (Cal. 1991) (approving use
of risk-utility balancing test in warning context); Finn v. G.D. Searle & Co., 677 P.2d 1147,
1152-53 (Cal. 1984) (same); Feldman v. Lederle Labs., 479 A.2d 374, 382, 385-86 (N.J. 1984)
(same); Phillips v. Kimwood Mach. Co., 525 P.2d 1033, 1036-39 (Or. 1974) (same); cf. Davis
v. Wyeth Labs., Inc., 399 F.2d 121, 129-30 (9th Cir. 1968) (referring to importance of freedom
of choice but, nevertheless, using "unreasonably dangerous" balancing test to determine
liability).
158. See supra note 143 (citing authorities for proposition that risk-utility test is based
upon risk-reduction policy).
159. See Gershonowitz, supra note 85, at 72 passim (arguing that marketplace honesty is
central policy in warning context). A number of courts have noted the importance of consumer
autonomy in the warning context, but have, nevertheless, applied a risk-reduction-based test
to determine whether a warning was necessary. See, e.g., Borel, 493 F.2d at 1087, 1089
(discussing consumer entitlement to right to decide but, nevertheless, employing risk-utility
test); Davis, 399 F.2d at 128-30 (discussing importance of consumer choice but adopting
"unreasonably dangerous" test); Anderson, 810 P.2d at 554, 559 (explaining that consumers
must be given opportunity to choose after having cited use of risk-utility test in warning
context with approval).
160. See supra notes 129-36 and accompanying text (explaining that large number of
warnings involve notions of informed choice).
161. See supra note 133 and accompanying text (explaining that consumer autonomy
concerns underlie informed-choice function of warnings).
1992] FAILURE TO WARN 1539

sumer autonomy considerations. 6 2 Consumer autonomy is best promoted

63
by the marketplace-honesty policy, rather than the risk-reduction policy.
Furthermore, the risk-reduction-based tests that the courts most often em-
ploy are inappropriate in many warning cases.'"
Marketplace honesty takes on more significance in the failure to warn
context for another reason as well. One of the primary concerns underlying
the marketplace-honesty policy is the need for honesty in communication
between seller and consumer. 6 5 Because the failure to warn implicates a
failure in communication more so than does a design defect,' 66 the market-
place-honesty policy is especially appropriate for use in formulating a strict
liability failure to warn test. 6 7 Conversely, the failure to warn action is
particularly well-suited for promoting the important goal of marketplace
honesty, which requires manufacturers to stand behind the68 express and
implied representations they make relating to product safety.
Just as a test relying heavily on the risk-reduction policy is often
inappropriate in warning cases, a test relying too heavily on marketplace
honesty is ill-suited for use in devising a design defect test. 69 Without
delving too deeply into design defect law, it is sufficient to note two inherent
weaknesses in using the marketplace-honesty policy to inform the design
defect doctrine. First, a design test based on marketplace honesty would
allow a manufacturer to avoid liability simply by supplying a warning or

162. Cf. Kidwell, supra note 38, at 1384-85 & n.36 (explaining that if law was to include
preferences as to what decision consumers should make in given context, it would obscure
important and unique aspect of consumer autonomy that society attempts to preserve, namely
"freedom to make foolish choices"); infra notes 203-04 and accompanying text (explaining
that relevant causal inquiry in failure to warn context is not whether consumer would have
heeded warning had manufacturer provided one, but whether consumer was given opportunity
to choose).
163. See supra notes 117-18 and accompanying text (discussing means in which market-
place-honesty policy promotes consumer autonomy).
164. See supra notes 144-53 and accompanying test (discussing inapplicability of risk-
utility test to informed-choice warnings); see also Bromberg, supra note 34, at 535-36 (asserting
that risk-utility test is never appropriate in failure to warn context); Gershonowitz, supra note
85, at 101-03 (same).
165. See supra notes 111-16 and accompanying text (discussing foundation of marketplace-
honesty policy).
166. See Anderson v. Owens-Coming Fiberglas Corp., 810 P.2d 549, 558 (Cal. 1991)
(noting that "warning defects" relate to failure extraneous to product itself; manufacturing
and design defects can be evaluate without reference to communications of manufacturer, but
giving of warning necessarily involves communication of something to someone).
167. See Gershonowitz, supra note 85, at 72 passim (arguing that failure to warn law
should be based upon marketplace-honesty policy).
168. See supra notes 112-116 (arguing that sellers should be obligated to stand behind
express and implied representations they make regarding product safety); infra notes 229-38
and accompanying text (arguing that in some situations manufacturers may promote product
to such an extent as to vitiate effectiveness of warning).
169. See Britain, supra note 39, at 356 (stating that promoting feasible design alternative
is risk reduction concern); Gershonowitz, supra note 85, at 72, 104 (noting that risk reduction
is primary policy underlying design defect doctrine).
1540 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

making the danger obvious, even if reducing the risk by redesigning the
product would be cost effective and would result in fewer accidents. 70 Also,
a test based on marketplace honesty cannot accommodate important factors
such as the cost of adding more safety features or of redesigning the
product.' 7 ' A test based on the risk-reduction policy, such as a risk-utility
test, significantly lessens these concerns and is, therefore, preferable to a
marketplace-honesty test in design cases. 72
Although the different policy concerns implicated by warning defects
and design defects necessitate different tests for each, 7 the different tests
should interact in a manner that best promotes marketplace honesty and
risk reduction. 174 Before a manufacturer places a product on the market,
the manufacturer is nearly always in a better position than the consumer
to control product-related risks because the manufacturer has greater knowl-
edge than the consumer about the risks. 75 In many situations, the manu-
facturer can best control the risk either by designing the risk out of the
product or by shifting the status of best risk controller to the consumer by
providing an adequate warning or somehow making the risk open and
obvious. 76 Both options promote marketplace honesty as well as risk re-
duction.

170. See Britain, supra note 39, at 354-55, 420 (observing that honesty policy encourages
making danger patent rather than reducing risk when cost effective); Epstein, ProductsLiability
Search, supra note 91, at 648 n.16 (noting that recovery would be denied under consumer
expectation test regardless of whether risk can be cost effectively reduced); Keeton, supra note
2, at 589 (same).
171. See Britain, supra note 39, at 354 (noting that marketplace honesty test fails to take
costs of additional safety features or redesign into account); Keeton, supra note 2, at 592
(same).
172. See Britain, supra note 39, at 420 (noting that risk-reduction policy can account for
situations in which risk can be "designed out" at little cost); Gershonowitz, supra note 85, at
72, 104 (arguing that risk-reduction policy works well in design cases); Keeton, supra note 2,
at 592 (expressing preference for risk-utility test in design defect context).
173. See supra notes 159-72 and accompanying text (arguing that failure to warn and
design defect implicate different policies and, therefore, require separate tests).
174. See Britain, supra note 39, at 408-22 (arguing that products liability rules should
seek to accommodate risk reduction and marketplace honesty equally); supra notes 100-19 and
accompanying text (explaining important functions served by both risk-reduction and market-
place-honesty policies in products liability law).
175. See Smith v. American Motors Sales Corp., 576 N.E.2d 146, 151 (II!. App. Ct.
1991) (stating that duty to warn is based upon unequal knowledge with respect to risk and
that manufacturer who has, or should have, such knowledge has obligation to warn); Bloxom
v. Bloxom, 512 So. 2d 839, 843 (La. 1987) (stating that if risk is foreseeable, manufacturer is
in best position to avoid injuries by giving warning); Madden, supra note 5, at 233-34 passim
(stressing the importance of asymmetry in safety-related information held by buyer and seller
to failure to warn doctrine).
176. See EPSTEIN, supra ncte 3, at 94 (explaining that in some situations design change
is best way to reduce product-related risks, while in other situations warning may be appro-
priate); Twerski et al., supra note 121, at 506-17 (discussing complex trade-offs between design
and warning options necessary to obtain optimum level of safety); Wade, supra note 2, at 842
(noting that warnings may often make otherwise dangerous product safe, but warning not
always sufficient).
1992] FAILURE TO WARN

If, however, the manufacturer is able to cost-effectively design a risk

out of the product without any corresponding diminution in the product's
utility, the manufacturer retains the position of best risk controller, even if
the manufacturer provides an adequate warning to the consumer. 17 7 In this
situation, marketplace honesty is satisfied, and there should be no liability
for the failure to warn, but the risk-reduction policy remains unsatisfied.'78
Thus, courts should hold the manufacturer accountable on a design defect
theory premised on risk reduction for any injuries caused by the product
79
risk.
In another situation, it may be cost ineffective, or perhaps even im-
possible, to redesign the product so as to eliminate an inherent risk without
a corresponding reduction in the product's utility. 110 The consumer, however,
may be able to take safety precautions to reduce the risk. 8 ' This would be
the case with a product such as pure sulfuric acid, in which the property 8 2
that produces the risk is also the property that provides the benefit.
Although such a product does not have a design defect, the law should still
require the manufacturer to provide an adequate warning. 83 The risk-
reduction policy mandates a warning because the manufacturer possesses
greater knowledge than the consumer as to the product-related risk and the
consumer can take precautions to reduce the risk.'8 Marketplace honesty

177. See, e.g., Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 847 (N.H. 1978) (stating
that if danger can be eliminated without excess cost or loss of product efficiency, liability may
attach even though danger is obvious or manufacturer has provided adequate warning); EPSTEIN,
supra note 3, at 94 (explaining that when design change possible or effective, design change
is preferable to warning); Calabresi & Hirschoff, supra note 100, at 1062-63 (explaining that
analysis depends not only on whether warning was adequate, but on availability of alternatives
to product); Noel, supra note 25, at 262-63 (noting that warnings are not sufficient if redesign
is feasible).
178. See sources cited supra note 177 (observing that warning is not most efficient means
of making product safe if redesign is feasible). When faced with cases in which redesign is
preferable to warning, courts often decide on dual design defect and failure to warn grounds.
See Twerski et al., supra note 121, at 500-505 (discussing cases in which courts decide on both
design defect and failure to warn grounds and arguing that such holdings effectively constitute
an instruction to redesign product).
179. See supra notes 177-78 and accompanying text (arguing that manufacturer retains
position of best risk controller if redesign is feasible and will substantially reduce risk).
180. See Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1088-89 (5th Cir. 1973)
(identifying asbestos as product that is incapable of being made safer by redesign), cert. denied,
419 U.S. 869 (1974); Dunn v. Lederle Labs., 328 N.W.2d 576, 579 (Mich. Ct. App. 1982)
(noting that Sabin polio vaccine is incapable of being made safer by redesign).
181. See Dunn, 328 N.W.2d at 579 (noting that Sabin polio vaccine is incapable of being
made safer by redesign but that persons can avoid contracting "contact polio" by avoiding
certain types of contact with patients who have taken vaccine).
182. See EPSTEIN, supra note 3, at 94 (offering pure sulfuric acid as example of product
that cannot be redesigned to reduce risk while retaining its utility).
183. See sources cited supra notes 181-82 (noting that products that cannot feasibly be
made safer by redesign should be accompanied by warning).
184. See supra note 175 and accompanying text (discussing importance to risk-reduction
policy of asymmetry between manufacturer and consumer knowledge); supra note 128 and
accompanying text (describing risk-reduction function of warnings).
1542 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

also requires a warning. 85 Thus, a marketplace-honesty-based failure to

warn action would hold the manufacturer accountable for failing to ade-
quately warn, promoting both the risk-reduction policy and marketplace
honesty.
Occasionally, a product is incapable of being made safer either by a
change in design or by the inclusion of a warning.1 6 Nonetheless, the
manufacturer is fully justified in the marketing of the product because of
the great benefit the product provides.8 7
An example of such a product is
the Sabin polio vaccine, the use of which contains an extremely remote but
definite risk of contracting the polio virus.' The consumer can take no
precautions to avoid this risk short of foregoing the use of the vaccine
altogether. 8 9
Thus, because a warning would not help to reduce the risk,
the risk-reduction policy would not require that the manufacturer warn
consumers of the risk.190 But this is precisely the type of risk against which
marketplace honesty and consumer autonomy demand that the manufacturer
warn.' 9' A marketplace-honesty-based failure to warn test would, therefore,
require that the manufacturer provide a warning or compensate the consumer
for any injury caused by the risk.

F. The Causation Inquiry

The causation inquiry presents serious problems in the area of failure
to warn. 92 Causation in the context of failure to warn is essentially this: if
the consumer would have altered his or her behavior so as to avoid or

185. See supra notes I 11-16 and accompanying text (discussing marketplace-honesty policy
and requirement that manufacturers honestly represent product safety); supra note 150 and
accompanying text (discussing importance of informed consumer choice).
186. See supra notes 129-35 and accompanying text (describing pure informed-choice
warnings).
187. See RESTATEMENT (SECOND) OF TORTS § 402A cmt. k (1965) (explaining that marketing
of "unavoidably unsafe product" is justified because of important benefits product provides).
188. See Davis v. Wyeth Labs., Inc., 399 F.2d 121, 128 (9th Cir. 1968) (identifying Sabin
type III polio vaccine as product that's marketing is justified by public interest in eliminating
polio, even though small, unavoidable risk of contracting polio accompanies ingestion of
vaccine).
189. See id. at 128-29 (explaining that no known method exists to identify those who will
contract polio from ingestion of Sabin type III vaccine); see also supra notes 129-35 and
accompanying text (explaining pure informed-choice function of warnings).
190. See supra notes 144-49 and accompanying text (reasoning that liability in pure
informed-choice context cannot be based on risk-reduction policy).
191. See supra notes 129-35, 138, 150 and accompanying text (describing informed choice
function of warnings and consumer's "right" to information regarding product-related risks).
192. See Payne v. Soft Sheen Prods., Inc., 486 A.2d 712, 725 (D.C. 1984) (noting
difficulties associated with causation issue in failure to warn context, including "impossible"
burden on plaintiff); Technical Chem. Co. v. Jacobs, 480 S.W.2d 602, 606 (Tex. 1972)
(discussing proof problems associated with causation in failure to warn context); Green, supra
note 88, at 1199-1200 (describing number of problems relating to causation issue in failure to
warn context); Henderson & Twerski, supra note 12, at 278-79, 303-10 (same); Keeton, supra
note 68, at 413-15 (same).
1992] FAILURE TO WARN 1543

reduce the possibility of injury had the manufacturer provided an adequate

warning, the failure to adequately warn caused the injury. 93 When courts
frame the causation inquiry in this way, injured persons, in order to establish
a cause of action, must prove that they would have read, understood, and
remembered an adequate warning, had the manufacturer provided one, and
also that tiey would have heeded such a warning.'9 Defendant manufac-
turers often focus on the causation issue, contending that even if they had
supplied an adequate warning, the consumer would not have read or would
not have heeded the warning. 95 This places an extremely difficult burden
of proof on the plaintiff, who musf prove something that did not happen. 96
One way that courts have dealt with the problem is through the use of a
rebuttable presumption of causation. 97 The courts simply presume that if
the manufacturer had supplied an adequate warning, the user would have
read and heeded that warning. 9 This presumption is extremely difficult,
199 to rebut. 200
but not impossible,

193. See, e.g., Reyes v. Wyeth Labs., Inc., 498 F.2d 1264, 1279-80 (5th Cir.) (holding
that defect-i.e. failure to warn-must cause injury), cert. denied, 419 U.S. 1096 (1974);
Payne, 486 A.2d at 725 (stating that plaintiff would have had to alter conduct had manufacturer
provided warning in order for failure to warn to have caused injury); Bloxom v. Bloxom, 512
So. 2d 839, 850 (La. 1987) (same); see also Henderson & Twerski, supra note 12, at 305
(explaining that plaintiff must have read, understood, and altered conduct had manufacturer
provided warning for failure to warn to have caused injury); cf. Canterbury v. Spence, 464
P.2d 772, 790 (D.C. Cir.) (holding that causal connection exists in informed consent medical
malpractice case only when disclosure of risks would have resulted in decision against treat-
ment), cert. denied, 409 U.S. 1064 (1972).
194. See sources cited supra note 193 (describing causation inquiry in failure to warn
context).
195. See Green, supra note 88, at 1199-1200 (contending that manufacturers often use
causation issue as device to divert the court from more important issues).
196. See sources cited supra note 192 (noting difficult proof problems inherent in failure
to warn causation inquiry).
197. See, e.g., Reyes, 498 F.2d at 1281 (adopting causation presumption); Payne v. Soft
Sheen Prods., Inc., 486 A.2d 712, 725 (D.C. 1984) (same); Bloxom, 512 So. 2d at 850 (adopting
rebuttable presumption of causation); Cunningham v. Charles Pfizer & Co., 532 P.2d 1377,
1382 (Okla. 1974) (adopting rebuttable presumption and citing informed consent medical
malpractice cases for support). The justification courts most often give for the causation
presumption is that it is the mirror image of the presumption given to manufacturer's under
§ 402A comment j of the Restatement. See, e.g., Ortho Pharmaceutical Corp. v. Chapman,
388 N.E.2d 541, 555 (Ind. Ct. App. 1979) (adopting presumption of causation based on
comment j); Wooderson v. Ortho Pharmaceutical Corp., 681 P.2d 1038, 1041, 1057 (Kan.)
(same), cert. denied, 469 U.S. 965 (1984); Technical Chem. Co. v. Jacobs, 480 S.W.2d 602,
606 (Tex. 1972) (adopting rebuttable presumption based on comment j). Comment j provides
that "[w]here a warning is given, the seller may assume that it will be read and heeded."
RESTATEMENT (SECOND) OF TORTS § 402A cmt. j (1965). Professors Henderson and Twerski
have argued that the causation presumption does not logically follow from the presumption
defendants enjoy under comment j. Henderson & Twerski, supra note 12, at 278-79.
198. See sources cited supra note 197.
199. See Bloxom v. Bloxom, 512 So. 2d 839, 850-51 (La. 1987) (holding that manufacturer
presented sufficient evidence to rebut causal presumption).
200. See Henderson & Twerski, supra note 12, at 306, 325-26 (noting that tools available
1544 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

The failure to warn causation inquiry, as the courts have formulated

it, is firmly grounded in risk-reduction policy, 20' and framed in this way,
fails to account for the importance of informed choice in the warning
context. 2°2 The very purpose of the informed-choice function of warnings
is to promote consumer autonomy by allowing consumers to make an
informed decision as to whether they wish to heed a warning. 203 The
possibility that a consumer would not have heeded an adequate warning,
had the manufacturer provided one, is irrelevant to whether or2 4not the
consumer was given the opportunity to make an informed choice.
Courts could avoid many of the problems pertaining to the causation
issue by simply recognizing that the causation inquiry, as they have framed
it, is inappropriate for use in the failure to warn context. Marketplace
honesty and consumer autonomy, rather than risk reduction, should be the
most important factor in framing the causation issue. 2 5 When the causation
issue is framed in terms of marketplace honesty, the plaintiff only needs to
show (1) that the manufacturer failed to provide an adequate warning about
an inherent product risk and (2) that the plaintiff's injury resulted from the
very same risk against which the manufacturer failed to provide an adequate
warning. 2 6 The causal nexus is between the risk and the injury, rather than

to defendant to rebut causation once it has been established are "almost nonexistent" and
that rebuttal is further complicated by tendency of courts to send all causation questions to
jury).
201. See Britain, supra note 39, at 374, 398-99 (explaining that causation inquiry in failure
to warn and informed consent medical malpractice is based upon risk-reduction policy);
Henderson & Twerski, supra note 12, at 304 (arguing that risk reduction requires substantial
connection between defendant's fault-providing inadequate warning-and plaintiff's injury).
202. See Britain, supra note 39, at 374 (arguing that risk-reduction-based causal inquiry
defeats objectives of consumer autonomy); cf. Jay Katz, Informed Consent-A Fairy Tail?
Law's Vision, 39 U. Prrr. L. Rav. 137, 160-64 (explaining that altered conduct requirement
of informed consent medical malpractice causation inquiry places significant limitation on
human dignity interests and frustrates patient choice); Marjory M. Shultz, From Informed
Consent to Patient Choice: A New ProtectedInterest, 95 YALE L.J. 219, 245-51 (arguing that
causal inquiry in informed consent medical malpractice context ill-serves interest of patient
autonomy).
203. See supra notes 129-35 and accompanying text (discussing informed-choice function
of warnings).
204. See Britain, supra note 39, at 400 (arguing that failure to provide adequate warning
constitutes frustration of individual integrity that deserves redress and that inquiry into whether
injuries would have been averted by better warning is irrelevant); Twerski et al., supra note
121, at 520 n.59 (contending that when informed-choice function is at issue, relevant question
is not.whether plaintiff would have altered conduct had a warning been given); cf. Katz, supra
note 202, at 160 (arguing that in informed consent medical malpractice context when protection
of human dignity is at issue, question of whether conduct would have been altered had
information been disclosed is irrelevant); Schultz, supra note 202, at 251 (explaining that in
informed consent context, freedom of choice is protected interest, and proper inquiry is whether
right to exercise freedom has been frustrated).
205. See supra notes 159-68 and accompanying text (arguing that marketplace honesty
and consumer autonomy are more important than risk reduction in failure to warn context).
206. Cf. Britain, supra note 39, at 400 (reasoning that when interest to be protected is
1992] FAILURE TO WARN 1545

between the failure to warn and the injury. 2 7 The plaintiff need not show
that the warning, if given, would have been read, understood, remembered,
and heeded. 28 The relevant inquiry, from a consumer autonomy perspective,
is whether the manufacturer, by failing to provide an adequate warning,
deprived the plaintiff of the opportunity to make an informed decision as
to whether to confront the risk or to take the necessary precautions to
reduce the risk. 209

consumer autonomy, tort arises from manufacturer's "dishonesty," rather than from causal
connection between the nondisclosure and injury); Twerski et al., supra note 121, at 520 n.59
(arguing that if pure informed choice case is properly analyzed, issue is whether a battery has
been committed due to fact that plaintiff has not been provided with adequate information to
give informed consent to "touching").
The court in Jarrell v. Monsanto Co., 528 N.E.2d 1158 (Ind. Ct. App. 1988), formulated
a very similar two step inquiry to failure to warn causation based upon an interpretation of
the Indiana products liability statute. The court found that it is more appropriate to say the
product caused the injury than to say the lack of adequate warning caused the injury. Id. at
1167. The court then inquired into whether the manufacturer provided an adequate warning.
Id. at 1167-68. The court refused to allow the defendant to argue that the plaintiff would not
have heeded an adequate warning. Id. at 1168 & n.6. Rather than justifying its holding on
grounds of consumer autonomy, however, the court employed a pseudo-estoppel argument
based upon the presumption in comment j to section 402A. Id. at 1168 n.6. Comment j states
that if an adequate warning is provided, the manufacturer may presume that it will be followed.
RESTATEMENT (SEcoND) OF TORTS § 402A cmt. j (1965). The court reasoned that because the
comment j presumption applies if the manufacturer provides an adequate warning, the
manufacturer cannot successfully argue that regardless of what warning it put on the product
the consumer would not heed the warning. Jarrell,528 N.E.2d at 1168 n.6.
207. See Jarrell, 528 N.E.2d at 1167-68 (stating that it is more accurate to say product
caused injury than to say inadequate or missing warning was defect that caused injury); cf.
Reyes v. Wyeth Labs., Inc., 498 F.2d 1264, 1279 (5th Cir.) (recognizing need for causal nexus
between risk and injury as well as between failure to warn and injury), cert. denied, 498 U.S.
1264 (1974); Finn v. G.D. Searle & Co., 677 P.2d 1147, 1153 (Cal. 1984) (reasoning that
strength of causal link between product and injury is relevant to issues of whether warning
should be given at all and whether warning is adequate); Ortho Pharmaceutical Corp. v.
Chapman, 388 N.E.2d 541, 555 (Ind. App. 1979) (recognizing need for two causal inquiries:
first, whether the failure to warn caused plaintiff's use of product, to which rebuttable
presumption of causation applies; and second, whether product-related risk was cause of
injury). Under the risk-reduction approach to causation, the causal nexus is typically between
the failure to warn and the injury. See sources cited supra note 193.
The requirement of a causal nexus between the specific risk against which the manufacturer
failed to warn and the injury eliminates a significant potential problem inherent in the risk-
reduction approach to causation. If the only causal nexus is between the failure to warn and
the injury, as in the typical failure to warn case, the plaintiff could possibly recover in a
situation in which the manufacturer has provided an adequate warning against the risk that
caused the injury but has failed to warn of other material risks. See Gershonowitz, supra note
85, at 92. For an interesting hypothetical explaining this possibility, see id. at 90-93.
208. See supra note 194 and accompanying text (describing requirement under current law
that plaintiff show that plaintiff would have read and heeded an adequate warning had one
been given in order to establish causal connection). The marketplace-honesty test also eliminates
the need for a causation presumption. See supra notes 197-200 and accompanying text
(describing causation presumption).
209. Cf. Britain, supra note 39, at 374 (arguing that relevant causal inquiry in informed
consent medical malpractice action based on integrity should be whether nondisclosure cogni-
1546 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

IV. REFORMULATING THE STRICT LLA.BmrrY FAMURE TO WARN

A. Proposed Reformulation of the Strict Liability Failure to Warn

The primary concern that the proposed reformulation of strict liability
210
failure to warn doctrine seeks to accommodate is marketplace honesty.
This policy best promotes consumer autonomy, a principal objective of the
law of warnings,23 ' and also is the policy best able to guarantee that product
sellers stand behind their explicit and implicit representations regarding
product safety. 2 2 Consequently, both the test for the adequacy of the
warning2 31 and the causation inquiry 2 4 have their foundations in the mar-
ketplace honesty policy. Although marketplace honesty is the predominate
policy upon which the proposed reformulation is based, risk reduction
remains an important consideration. 21s The proposed test also substantially
differs from negligence. 2 6 The test is set forth in the form of the require-
ments for the plaintiff's prima facie case, followed by the defenses available
to the manufacturer:
1. The plaintiff must first prove that the product contained an
inherent material risk that caused (or manifested itself in the form
of) plaintiff's injury.
2. Once the plaintiff has shown a causal connection between the
risk and the injury, the plaintiff must show that the manufacturer
did not provide adequate safety-related information pertaining to
that risk-in other words, ithat the manufacturerdid not adequately
warn of the risk.

zably diminished patient's choice of integrity); Schultz, supra note 202, at 251 (same).
Commentators have noted that the use by courts of devices such as causal presumptions, see
supra notes 197-200 and accompanying text (describing use of causal presumptions), demon-
strates the preference of courts, in the failure to warn cbntext, for the values of consumer
autonomy and freedom of choice, even though the courts purportedly apply a risk-reduction-
based causation analysis. See Britain, supra note 39, at 399-402 (arguing that use of devices
that simplify causation issue demonstrates courts' preference for personal integrity); Kidwell,
supra note 38, at 1408 (arguing that use of causation presumption demonstrates preference
for individual sovereignty ideal).
210. See supra notes 159-68 and accompanying text (arguing that marketplace honesty
should be primary policy concern underlying strict liability failure to warn doctrine).
211. See supra notes 133-36, 162 and accompanying text (discussing role of consumer
autonomy in law of warnings).
212. See supra notes 111-19 and accompanying text (explaining marketplace honesty
policy).
213. See infra notes 226-38 and accompanying text (discussing means of determining
warning adequacy).
214. See supra notes 204-09 and accompanying text (discussing causation inquiry based
upon marketplace honesty policy).
215. See supra notes 173-91 and accompanying text (explaining role of risk-reduction
policy in failure to warn context).
216. See infra notes 266-278 (describing ways in which proposal differs from negligence
doctrine).
1992] FAILURE TO WARN 1547

3. Once the plaintiff has made out a prima facie case, the manu-
facturer may rely on a number of available defenses-
- Assumption of the risk (either because the risk was open and
obvious or because the plaintiff had actual knowledge of the
risk)
- Misuse of the product by the plaintiff
- Third party fault
- Risk not scientifically knowable.
1. Causal connection and material risk.-The causation inquiry has 2 17
been framed in terms of marketplace honesty rather than risk reduction.
The plaintiff need not show, as the plaintiff would with a risk-reduction
approach to causation, that the failure to warn caused the injury. 218 Rather,
the plaintiff must show that an inherent product-related risk caused (or
manifested itself in the form ofy 19 the injury and that the manufacturer's
failure to provide an adequate warning of that risk deprived the plaintiff
of the opportunity to make an informed decision as to whether to confront
the risk or to take the necessary precautions to reduce the risk.? 0 Use of
the marketplace-honesty approach to failure to warn causation eliminates
some of the serious problems inherent in the use of the risk-reduction
approach 22' and also helps to lay a firm marketplace-honesty foundation
for the strict liability failure to warn.
The proposed formulation requires that the risk which caused the injury
be material. A risk is material if the product-related danger has reached the
level at which it would become a material fact in the decision of a reasonably
prudent consumer as to whether or not to confront the risk or to take
precautions to reduce the risk."" The material-risk requirement serves a

217. See supra notes 201-09 and accompanying text (arguing that failure to warn causation
inquiry should be based upon marketplace honesty rather than risk reduction and describing
difference between two types of causation inquiries).
218. See supra notes 192-202 and accompanying text (describing risk-reduction approach
to failure to warn causation inquiry).
219. The "manifested itself in the form of" language should be used in place of "cause"
in situations such as that of asbestos. It is known that cancer, asbestosis and other respiratory
diseases are inherent asbestos-related risks, but it is unclear exactly which specific characteristic
of asbestos causes these conditions. In such situations it is conceptually clearer to say the risk
"manifested itself in the form of" plaintiff's injury.
220. See supra notes 204-09 and accompanying text (discussing marketplace honesty
approach to failure to warn causation inquiry).
221. See-supra notes 192-204 and accompanying text (describing problems inherent in risk-
reduction based failure to warn causation inquiry).
222. See Gershonowitz, supra note 85, at 101 (defining materiality as used in failure to
warn context); cf. Canterbury v. Spence, 464 F.2d 772, 787 (D.C. Cir.) (stating that in
informed consent context "[a] risk is material when a reasonably prudent person ... would
be likely to attach significance to the risk or cluster of risks in deciding whether or not to
forego the proposed therapy") (citing Jon R. Waltz & Thomas W. Scheuneman, Informed
Consent to Therapy, 64 Nw. L. Rnv. 628, 639-41 (1970)), cert. denied, 409 U.S. 1064 (1972);
Cobbs v. Grant, 502 P.2d 1, 11 (Cal. 1972) (defining materiality in informed consent context);
Shultz, supra note 202, at 284-85 & n.285 (arguing for materiality requirement in informed
consent law).
1548 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

number of purposes. First, it balances the need to provide adequate infor-

mation of significant product risks against the potentially costly effects of
an undue proliferation of warnings.223 Second, it promotes consumer auton-
omy by not requiring that the manufacturer burden the consumer with
information that the consumer does not require, or even desire, in making
an informed decision relating to the risk.22 Third, materiality is a familiar
concept to courts, so it does not require judges and lawyers to familiarize
themselves with an entirely new concept.21
2. Adequacy of the warning.-An exhaustive analysis of warning
adequacy is far beyond the scope of this note. A brief discussion, however,
is necessary to explain the relationship between the adequacy of a warning
and the marketplace-honesty policy. An "inadequate warning" can range
from the complete absence of any warning to a warning that is given but
is inadequate.226 Determining whether a warning is adequate generally in-
volves the consideration of factors such as the clarity of the language used,
the prominence of the warning, and the impression that the warning language
is calculated to make on users of the product. 7 Whether a warning is

223. See Gershonowitz, supra note 85, at 104 (arguing materiality requirement balances
need to provide adequate information against possibility of diluting efficacy of warnings); see
also supra notes 120-27 and accompanying text (describing potential problems relating to
proliferation of warnings).
224. See Gershonowitz, supra note 85, at 103-04 (concluding that materiality requirement
preserves freedom of choice); cf. Shultz, supra note 202, at 284-85 & n.285 (arguing that
materiality requirement is necessary in informed consent law in order to best promote informed
patient choice). Warnings of insignificant risks are simply not the type of information consumers
require, or even desire, in order to go about ordering their lives as they see fit. See Davis v.
Wyeth Labs., 399 F.2d 121, 129-30 (9th Cir. 1968) (recognizing that not all risks call for true
choice judgement); Canterbury, 464 F.2d at 786-87 (same); Gershonowitz, supra note 85, at
100 (noting that if risk is insignificant, consumers would likely choose to use product
notwithstanding the risk); Shultz, supra note 202, at 284-85 & n.285 (explaining that failure
of physician to disclose nonmaterial risks does not affect patient's ability to choose). Consumer
autonomy would be frustrated considerably, however, if consumers were inundated with
information about insignificant risks to such an extent that they would be unable to distinguish
between those risks that are significant and those that are insignificant. See supra notes 120-
27 and accompanying text (describing potential consequences of undue proliferation of warn-
ings).
225. See Gershonowitz, supra note 85, at 101 & n.160 (noting that courts and lawyers
are familiar with concept of materiality from its use in law of fraud and informed consent);
cf. supra note 43 (noting contention that courts should frame strict liability rules in negligence
language because negligence concepts are familiar to courts).
226. Ortho Pharmaceutical Corp. v. Chapman, 388 N.E.2d 541, 549 (Ind. App. 1979)
(citing Sterling Drug, Inc. v. Yarrow, 408 F.2d 978, 993 (8th Cir. 1969)).
227. See Jarrell v. Monsanto Co., 528 N.E.2d 1158, 1162-63 (Ind. Ct. App. 1988) (listing
factors to be considered, including factual content of warning, manner in which warning is
expressed, method of conveying warning, and warning intensity; noting relevance of expert
testimony as to poor location of warning, insufficiently startling colors, lack of symbols, and
lack of specificity); PROSSER, supra note 1, § 96, at 646-7 n.60 (listing factors often considered
in determining warning adequacy). Dean Prosser listed the following factors as important in
determining the adequacy of a warning: (I) the prominence of warning; (2) whether the warning
is sufficiently emphatic; (3) whether the warning is calculated to reach those likely to use the
1992] FAILURE TO WARN 1549

adequate is often a question for the jury.22s

Because the proposed formulation of the strict liability failure to warn
is based largely upon the marketplace-honesty policy, a determination of
warning adequacy that focuses only on the warning itself may be insuffi-
cient. 229 A court should consider all of the manufacturer's representations
about product safety in determining the adequacy of the warning. 230 Repre-
sentations made, by a manufacturer through advertising, packaging, or
deceptive design may all have an impact upon the way in which consumers
perceive a warning. 31
Application of the proposal to facts similar to those in Heaton v. Ford
Motor Co. 212 illustrates this point. Heaton, while driving his Ford pickup
truck, ran over a five-inch rock lying on the highway. 2 3 Approximately
thirty-five miles further down the highway, the rim of the truck's wheel
separated from the interior portion of the wheel, and the truck unexpectedly
left the road, injuring Heaton.2 4 Evidence showed that Ford had advertised
the truck as "solid," "rugged," and "built like a truck. '' 235 Heaton pursued
the case on design defect grounds, 236 and the Supreme Court of Oregon
eventually affirmed a nonsuit entered against him.23 7 Assume, however, that

product; (4) whether the warning is intelligible to the ordinary user; (5) whether the warning
sufficiently covers the particular danger; and (6) whether the warning makes the danger clear.
Id. See generally Little, supra note 5, at 1002-04 (discussing adequacy of warnings and
instructions for use); Sales, supra note 5, at 557-66 (same).
228. See, e.g., Payne v. Soft Sheen Prods., Inc., 486 A.2d 712, 723 (D.C. 1984) (holding
that warning adequacy is issue for jury); Jarrell, 528 N.E.2d at 1162 (same); Bloxom v.
Bloxom, 512 So.2d 839, 844 (La. 1987) (same).
229. See Britain, supra note 39, at 407 (arguing that representations in form of advertising,
packaging, and deceptive design may need. to be considered in determining adequacy of.
warning); Little, supra note 5, at 1005-1007 (noting that some courts have found warnings
that would be adequate standing alone inadequate in light of manufacturer's actions in
representing its product). 0
230. See, e.g., Stevens v. Park, Davis & Co., 507 P.2d 653, 661-62 (Cal. 1973) (recognizing
that warnings can be made inadequate by overpromotion "watering down" warnings); Love
v. Wolf, 38 Cal. Rptr. 183, 197 (Cal. Ct. App. 1964) (concluding that evidence justified
inference that prescription drug manufacturer had "watered down" its warnings by promoting
wider use of drug by physicians than proper medical practice warranted); Incollingo v. Ewing,
282 A.2d 206, 220 (Pa. 1971) (holding that actions designed to stimulate use of product must
be considered in testing adequacy of warning); see also RESTATEMENT (SECOND) OF ToRTs §
402B (1965) (providing that seller who makes misrepresentation of material fact concerning
character or quality of product through advertising, labeling, or otherwise is subject to liability
for physical harm to consumer caused by reliance upon misrepresentation); sources cited supra
note 229 (arguing that manufacturer's marketing practices are important in determining warning
adequacy).
231. See sources cited supra note 229 (noting that manufacturer's marketing practices
should be considered in determining warning adequacy).
232. 435 P.2d 806 (Or. 1967).
233. Heaton v. Ford Motor Co., 435 P.2d 806, 807 (Or. 1967).
234. Id. at 807.
235. Id. at 810 (O'Connell, J., dissenting).
236. Id. at 808.
237. Id. at 810.
1550 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

Heaton had relied upon the marketplace-honesty-based failure to warn

theory. Also assume that Ford had provided a warning in the owner's
manual about the potential danger of running over large rocks at highway
speeds. Even if Ford had provided such a warning, the warning may not
have been sufficient to counteract the impression of Ford's representations,
made through advertising, that the truck was "tough," "rugged," and
'2
"built like a truck." 38
3. Manufacturer'sdefenses.-Assumption of the risk defenses are en-
tirely consistent with the marketplace honesty policy. If the risk is open
and obvious239 or the consumer has actual knowledge of the risk,2 ° the
failure of the manufacturer to warn does not amount to a misrepresentation
of product safety; 241 nor does it frustrate consumer autonomy.24 2 Also, in
most situations, the assumption of the risk defense promotes the risk-
reduction policy because the consumer is generally in a better position than
the manufacturer to control the risk when the risk is open and obvious or
the consumer has actual knowledge of the risk.243 In situations where the
manufacturer is the better risk controller because it can cost effectively
reduce or eliminate the risk through redesign, a design defect action remains
open to the consumer. 2" Furthermore, warnings against obvious risks or
risks of which the consumer has actual knowledge would be redundant,
would likely lead to a proliferation of warnings and the problems resulting

238. See Britain, supra note 39, at 407 (hypothesizing that, on facts of Heaton, warning
in owner's manual may not be sufficient, under marketplace honesty theory, to counteract
explicit representations of truck's safety made through advertising); cf. Heaton v. Ford Motor
Co., 435 P.2d 806, 810-11 (Or. 1967) (O'Connell, J., dissenting) (arguing that jury question
was presented as to whether manufacturer's representations that vehicle was "solid," "rugged,"
and "built like a truck" created consumer expectation that driving over large rock would not
harm such a vehicle). 0
239. See supra note 24 and accompanying text (describing obvious risks).
240. See Garrett v. Nissen Corp., 498 P.2d 1359, 1364 (N.M. 1972) (holding that
manufacturer has no duty to warn if user has actual knowledge, regardless of whether claim
is negligence or strict liability), overruled on other grounds by, Klopp v. Wackenhut Corp.,
824 P.2d 293 (N.M. 1992); Noel, supra note 25, at 273 (explaining that manufacturer has no
duty to warn of dangers about which user has actual knowledge).
241. See Smith v. American Motors Sales Corp., 576 N.E.2d 146, 151 (111. App. Ct.
1991) (stating that because purpose of warning is to inform users of dangers of which they
are unaware, no purpose is served by providing warning of risk that is open and obvious).
242. See Kidwell, supra note 38, at 1392 (stating that if risk is obvious, warning of risk
would not result in greater informed choice).
243. See Calabresi & Hirschoff, supra note 100, at 1065-66 (explaining that assumption
of risk defense is consistent with liability rule placing loss on party in best position to decide
advantages of accident avoidance versus accident costs).
244. See supra notes 177-79 and accompanying text (noting that sometimes warning is not
sufficient to shift status of best risk controller from manufacturer to consumer and that in
these situations, design defect action is available). The majority of courts have abandoned the
"patent danger rule" in the design defect context. E.g., Auburn Mach. Works Co., Inc., v.
Jones, 366 So. 2d. 1167, 1167 (Fla. 1979); Holm v. Sponco Mfg., Inc., 324 N.W.2d 207, 213
(Minn. 1982); Micallef v. Miehle Co., 348 N.E.2d 571, 573, 576-77 (N.Y. 1976).
19921 FAILURE TO WARN 1551

from such a proliferation, 24 and would decrease economic efficiency 24 with

no corresponding benefit in terms of risk reduction or marketplace hon-
27
esty.
The consumer misuse defense is very similar to the requirement in
negligent failure to warn that the risk be foreseeable before a court will
impose liability. 2" The manufacturer should not be liable for the failure 249
to
warn of a risk that attends only an extremely unusual use of the product.
Because "unforeseeable" risks are, by definition, nonmaterial, 250 for the
sake of consistency of language, it is best to view the consumer misuse
defense as mitigating the materiality requirement.2' A liability rule requiring
manufacturers to warn against risks attending extremely unusual uses of a
product would likely lead to a proliferation of warnings. 212 Also, although
the lack of a warning may frustrate the autonomy of consumers who wish
to use a product in an unusual manner, the lack of warning would not
frustrate consumer autonomy in general because consumers do not desire
information about extremely improbable risks to make an informed decision
regarding safe product use. 2 " Nor does a misuse defense based upon an

245. See supra notes 120-27 and accompanying text (describing problems likely to result
from proliferation of warnings).
246
z See Moran v. Faberge, Inc., 332 A.2d 11, 15 (Md. 1975) (noting monetary cost,
insubstantial as it may be, of adding additional warnings); Beshada v. Johns-Manville Prods.
Corp., 447 A.2d 539, 545 (N.J. 1982) (same), limited to its facts, Feldman v. Lederle Lab.,
479 A.2d 374 (N.J. 1984).
247. See supra notes 239-44 and accompanying text (noting that assumption of risk defense
is consistent with both marketplace-honesty and risk-control policies).
248. See Payne v. Soft Sheen Prods., Inc., 486 A.2d 712, 725 (D.C. 1984) (defining
misuse as use that is not foreseeable); Higgens v. Paul Hardeman, Inc., 457 S.W.2d.943, 948
(Mo. Ct. App. 1970) (holding that abnormal use relieves defendant of liability only if misuse
unforeseeable); Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 847 (N.H. 1978) (rejecting
manufacturer's obligation to warn of unforeseeable misuse); see also supra note 25 and
accompanying text (describing foreseeability requirement in context of negligent failure to
warn).
249. See, e.g., Thibault, 395 A.2d at 847 (holding that manufacturer has no obligation
to warn of unforeseeable misuse); Owen, supra note 88, at 714 (recognizing unfairness that
would result from imposing liability upon manufacturers for injuries caused by "consumer
'madness' in putting products to unpredictably dangerous use"); Powers, supra note 57, at
802 (noting that risks imposed by unforeseeable use of product are appropriately attributed to
conduct of consumer rather than to product itself).
250. See supra notes 222-25 and accompanying text (describing materiality requirement).
A risk that is imposed by an extremely unusual use of thi product would not be a "material
fact in the decision of a reasonably prudent consumer as to whether or not to confront the
risk or to take precautions to reduce the risk." See supra text accompanying note 222 (defining
materiality).
251. Cf. Thibault, 395 A.2d at 849 (discussing need for semantic and conceptual clarity
and concluding that term "plaintiff's misconduct" should replace term "contributory negli-
gence" in jury charge in order to avoid injecting "flavor of negligence" into strict liability
claim).
252. See supra notes 120-27 and accompanying text (describing problems arising from
liability rules requiring manufacturers to warn of nonmaterial risks).
253. See supra note 224 (arguing that consumer autonomy is best promoted by providing
1552 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

extremely unusual use of the product frustrate the risk-reduction policy

because consumers are in a better position than the'manufacturer to know
4
to which unusual uses they will put the product.2
A third party fault defense is also available under the proposed test for
strict liability failure to warn. Often, all or part of an injury may be
attributable to a third party,211 as is the case when a third party incorporates
the original manufacturer's product into its own. If the manufacturer has
been completely honest in providing information about product-related risks,
the risk-reduction policy demands that the party in the best position to
avoid or reduce the risk shoulder the portion of the loss attributable to that
party. 6 If the party in the position of best risk controller happens to be a
third party, rather than the manufacturer or consumer, no reason exists for
holding the manufacturer accountable for the portion of the loss attributable
7
to the third party.25
The final defense available to the manufacturer is the scientific un-
knowability, at the time the manufacturer placed the product on the market,
of the risk against which the manufacturer failed to warn. Because this
defense is inconsistent with the marketplace honesty policy, 2 8 it is best
viewed as presenting countervailing reasons that outweigh the policies that
support rdcovery. 2 9 Both courts and commentators have severely criticized
a rule that imposes liability for the failure to warn of scientifically un-

consumers with only that information they desire in making an informed decision-i.e. those
risks with significant likelihood of manifesting themselves).
254. See Epstein, Products Liability Search, supra note 91, at 658-59 (observing that in
many situations, parties other than manufacturer are in best position to control risk); Powers,
supra note 57, at 802 (noting that risks imposed by unforeseeable use of product are
appropriately attributed to conduct of consumer); see also supra note 100 and accompanying
text (defining risk-control policy as that which imposes liability upon party in best position to
control or reduce risk).
255. See Epstein, ProductsLiability Search, supra note 91, at 658-59 (noting that in many
situations parties in chain of distribution other than manufacturer are in best position to
control or reduce risk); Owen, supra note 88, at 713 (explaining that some accidents are fault
of third persons).
256. See supra notes 100-03 and accompanying text (describing risk-reduction policy).
257. See supra note 100 and accompanying text (defining risk-reduction policy as that
which imposes liability upon party in best position to avoid or reduce risk). The loss-shifting
policy would justify imposing losses attributable to third parties on the manufacturer, but the
loss-shifting policy has been discredited. See supra note 91 (describing loss-shifting policy and
its inherent weaknesses).
258. See Shapo, supra note 111, at 1322 (reasoning that imposition of liability for failure
to warn of unknowable risks is supported by representational theory, because central issue is
not manufacturer's knowledge, but objective understanding of impressions manufacturer makes
in projecting product's image).
259. See Owen, supra note 88, at 709 (noting that liability rules premised upon represen-
tational theory need to account for goals of economic efficiency and fairness that may
sometimes cut other way from representational policy); cf. Powers, supra note 57, at 799
(arguing that contributory negligence is not inconsistent with no-fault liability if courts treat
it as defense resting on countervailing reasons that outweigh, rather than mitigate, policies
supporting recovery).
1992] FAILURE TO WARN 1553

knowable risks on the grounds that it provides a disincentive for manufac-

turers to discover risks subsequent to marketing, 260 that it inhibits economic
efficiency, 26 ' that it keeps beneficial products off the market, 262 that it
effectively places the manufacturer in the role of an insurer, 263 and, most
importantly, that it is extremely unfair. 264 For these reasons, courts have
been extremely reluctant to adopt such a rule. 2 5 Also for these reasons, the
scientific unknowability of a risk should be a defense under the proposed
strict liability failure to warn action.

B. How the Proposed Test Differs from Negligence

The proposed reformulation of strict liability failure to warn differs
markedly from negligent failure to warn. Perhaps the most significant
distinction is that the proposed action has been formulated only after careful
consideration of the relevant policies. 266 The reformulated action promotes
the important goals of marketplace honesty and consumer autonomy; neg-
ligence does not. 267 Thus, courts can discontinue use of devices such as the
product-conduct dichotomy or the imputed knowledge requirement, which
the courts traditionally have employed to distinguish the strict liability action
from negligence, and courts simply can recognize that strict liability and
26s
negligence differ with regard to the policies they promote.

260. See supra note 75 and accompanying text (presenting argument and citing authorities
for proposition that rule imposing liability for failure to warn of unknowable risks provides
disincentive to discover and warn of risks subsequent to marketing).
261. See supra note 76 and accompanying text (presenting argument and citing authorities
for proposition that rule imposing liability for failure to warn of unknowable risks inhibits
economic efficiency).
262. See supra note 77 and accompanying text (presenting argument and citing authorities
for proposition that rule imposing liability for failure to warn of unknowable risks keeps
beneficial products off market or at least delays entry of such products to market pending
inordinate delay for safety research and testing).
263. See supra note 78 and accompanying text (presenting argument and citing authorities
for proposition that rule imposing liability for failure to warn of unknowable risks essentially
places manufacturer in role of insurer of product safety).
264. See supra note 79 and accompanying text (presenting argumeni and citing authorities
for proposition that rule imposing liability for failure to warn of unknowable risks is extremely
unfair).
265. See Marvel, Annotation, supra note 37, at 370-77 (citing majority of cases, which
require that manufacturer knew, or should have known, of risk at time it placed product on
market before court will impose liability for failure to warn). But see id. at 377-80 (citing
minority of cases, which have adopted rule that imposes liability for failure to warn regardless
of whether risk was scientifically knowable at time manufacturer placed product on market).
266. See Vandall, supra note 39, at 67-69 (arguing that frontal consideration of important
social policies has never been important issue in negligence).
267. See Britain, supra note 39, at 416 (arguing that negligence does not accommodate
honesty policy).
268. See id. (contending that transparent devices for distinguishing strict products liability
from negligence are unnecessary if it is recognized that strict liability unequivocally accom-
modates honesty policy and negligence does not).
1554 WASHINGTON AND LEE LAW REVIEW [Vol. 49:1509

The fact that the proposed reformulation promotes important policies

that failure to warn law should seek to accommodate and negligence does
not is, in itself, a satisfactory distinction. 269 However, the proposed refor-
mulation differs from negligence in other respects as well. First, the refor-
mulated strict liability failure to warn action frames the causation inquiry
differently than the negligence doctrine does. 2 0 In negligence, and in the
common formulations of the strict liability action, the plaintiff must show
that the failure to warn caused the plaintiff's injury before a court will
impose liability. 27' This often leads to a difficult and needless inquiry into
whether the plaintiff would have read, understood, and heeded a warning
had the manufacturer provided one. 272 The proposed reformulation elimi-
nates this problem by framing the causation inquiry in terms of marketplace
honesty and requiring a causal nexus between the risk and the injury rather
than between the failure to warn and the injury, as is the case in negli-
27 3
gence.
Second, the determination of warning adequacy is different under the
proposed strict liability action than it is under negligence. Because the
reformulated strict liability failure to warn action is based upon the mar-
ketplace honesty policy, a court should take into account all of the manu-
facturer's representations of product safety in determining the adequacy of
the warning. 274 For example, if a manufacturer advertises a product as
"tough" and "rugged," a court should consider these representations in
determining whether a warning is adequate. 27 Although the negligence action
is quite capable of taking such factors into consideration, 276 the determi-
nation of warning adequacy in a negligence action often centers on the
277
warning, or list of warnings, alone.

269. See id. (arguing that representational nature of strict liability action would sufficiently
distinguish it from negligence).
270. See supra notes 192-209 and accompanying text (describing differences between risk-
reduction-based causation inquiry, which is used in negligence, and marketplace-honesty-based
causation inquiry).
271. See supra notes 192-202 and accompanying text (describing risk-reduction-based
causation inquiry).
272. See supra notes 194-200 and accompanying text.
273. See supra notes 203-09 and accompanying text (describing marketplace-honesty-based
failure to warn causation inquiry); supra notes 217-21 and accompanying text (describing
causation inquiry of proposed strict liability failure to warn action).
274. See supra notes 226-38 and accompanying text (explaining determination of warning
adequacy under marketplace-honesty-based failure to warn action).
275. See supra notes 232-38 and accompanying text (illustrating need to account for all
manufacturer representations of product safety in determining warning adequacy by use of
example in which manufacturer advertised vehicle as "tough," "rugged," and "built like a
truck").
276. See Love v. Wolf, 38 Cal. Rptr. 183, 197 (Cal. Ct. App. 1964) (concluding, in
negligence action, that evidence justified inference that prescription drug manufacturer had
"watered down" its warnings by promoting wider use of drug by physicians than proper
medical practice warranted).
277. See sources cited supra note 227 (describing factors generally used in determining
1992] FAILURE TO WARN

Finally, the proposal eliminates terminology such .as "unreasonable"

and "foreseeable" and avoids cost-benefit-type analysis. This difference in
terminology and mode of analysis between the proposed strict liability failure
to warn doctrine and the negligence doctrine will help to eliminate the
confusion that the similarity between the two actions, as they now stand,
278
has created.

V. CONCLUSION

Since the inception of strict products liability, courts have failed to

develop a strict liability failure to warn doctrine that meaningfully differs
from negligence and also accommodates important social policies relevant
to the law of warnings. This failure is attributable to the fact that the
courts seldom have considered the relevant policies in attempting to for-
mulate the doctrine. The primary objective of the courts has been to develop
a strict liability doctrine that in some way differs from negligence. Through
a careful examination of the policies that the failure to warn doctrine should
seek to accommodate, however, the courts could develop a strict liability
failure to warn doctrine that is based upon sound policy and that differs
significantly from negligence. At the same time, the reformulated strict
liability action would eliminate many of the doctrinal difficulties, such as
the complications inherent in the causation inquiry, that plague the current
law.

MICHAEL A. PITTENGER

warning adequacy). Courts have generally taken representations of product safety made through
marketing and advertising into consideration in determining warning adequacy only in the
prescription drug context, in which "detail men" rigorously promote the benefits of the drugs
to doctors. See cases cited supra note 230 (recognizing, in prescription drug context, theory
of determining warning adequacy in which marketing activities are taken into account).
278. See supra notes 82-84 and accompanying text (explaining that similarity between
strict liability and negligent failure to warn causes confusion for both judges and juries).