California Law Review, Inc.

Foreword: Understanding Products Liability

Author(s): Gary T. Schwartz
Source: California Law Review, Vol. 67, No. 3, The Supreme Court of California 1977-1978.
Part I (May, 1979), pp. 435-496
Published by: California Law Review, Inc.
Stable URL: https://www.jstor.org/stable/3479949
Accessed: 27-06-2019 09:16 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

California Law Review, Inc. is collaborating with JSTOR to digitize, preserve and extend
access to California Law Review

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 California Law Review
VOL. 67 MAY 1979 No. 3

Foreword:
Liability
Gary T. Schwartzt

Unquestionably, products liability ranks as one of the most con-

spicuous legal phenomena of the last twenty years,' and the California
Supreme Court has played a leading role in the elucidation of products
liability doctrine. In Greenman v. Yuba Power Products, Inc. ,2 decided
in 1963, the court first ruled that manufacturers are subject to strict
liability for injuries resulting from defective products. In its Cronin
decision nine years later,3 the court reemphasized the "defect" prereq-
uisite for strict liability, but rejected the Restatement's "unreasonably
dangerous" gloss on "defect,"4 thereby leaving the defect concept in a
rather barren state. In Barker v. Lull Engineering Co.,' decided in Jan-
uary 1978, a unanimous court finally attempted an explication of "de-
fect." The court noted (as have others) that product defects can be
classed under three headings: "manufacturing" defects, "design" de-
fects, and defects involving inadequate warnings or instructions accom-

t Professor of Law, University of California, Los Angeles. B.A. 1962, Oberlin College;
J.D. 1966, Harvard University. A much earlier version of this Foreword was submitted as a report
to the California Legislature's Joint Committee on Tort Liability. In preparing that report and
this Foreword, I spoke to about twenty experienced products lawyers in California: to them, my
thanks. My appreciation of products liability theory has filtered through conversations with Cali-
fornia colleagues like Professors William Cohen, Robert Ellickson, and Edmund Ursin.
1. Consider, for example, three law review symposia published during 1977-78: Symposium.
1977 Products Liability Institute, 10 IND. L. REV. 755 (1977); Symposium: Product Liability, 29
MERC. L. REV. 373 (1978); Symposium on Products Liability Law. The Needfor Statutory Reform,
56 N.C.L. REV. 625 (1978). By now, the literature on products liability is voluminous; in this
Foreword, it is cited selectively.
2. 59 Cal. 2d 57, 377 P.2d 897, 27 Cal. Rptr. 697 (1963).
3. Cronin v. J.B.E. Olson Corp., 8 Cal. 3d 121, 501 P.2d 1153, 104 Cal. Rptr. 433 (1972).
4. RESTATEMENT (SECOND) OF TORTS ? 402A (1965).
5. 20 Cal. 3d 413, 573 P.2d 443, 143 Cal. Rptr. 225 (1978).

435

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 436 CALIFORNIA LAW REVIEW [Vol. 67:435

panying the product.6 For manufacturing defects,

dictum the appropriateness of a deviation-from
standard. But it is design defects that have engend
confusion, both in actual litigation and in the p
ing this confusion, Barker established a "two-pr
fying design defects. First, a product's design is d
"fails to perform as safely as an ordinary consu
used in an intended or reasonably foreseeable m
design is defective if the risks inherent in that de
the design's intrinsic benefits-whether measure
ings, improved performance, or other factors.
benefit prong, Barker also announced a seeming
of the burden of proof: once the plaintiff shows t
sign proximately caused [the plaintiffs] injury,
defective unless the manufacturer can persuade
sign's benefits exceed its associated risks.8 More
ion twice indicates that the jury is to render its ri
upon the basis of "hindsight."' And in a prov
Barker opinion raises the possibility of strict liabi
ordinary defect, for injuries caused by product
ger.'1o
This Foreword will briefly trace the historical development of the
existing products liability Tule; it will compare that rule to a genuine
strict products liability rule, which it will identify and evaluate; it will
explore the general relationship between strict products liability in its
present form and the pre-existing (and co-existing) doctrines of negli-
gence and implied warranty; it will assess each of Barker's two
"prongs;" and, finally, it will discuss Barker's references to "hindsight"
evaluations of design defectiveness and to "norm-is-danger" products.
In the course of these efforts, the Foreword will endeavor to achieve an
understanding of strict products liability and, in particular, to establish
the relationship between tort and contract principles within products
liability.

6. The absence of a warning can easily be seen as a kind of defect in design. Given this
perception, there are only two major defect types, manufacturing and design. According to a
recent federal study, 35% of all products cases involve manufacturing defects, 37% design defects,
and 18% warning defects. INTERAGENCY TASK FORCE ON PRODUCT LIABILITY, FINAL REPORT
II-54 (1977) [hereinafter referred to as FINAL REPORT].
7. 20 Cal. 3d at 426-27, 573 P.2d at 452, 143 Cal. Rptr. at 234.
8. Id. at 427, 573 P.2d at 452, 143 Cal. Rptr. at 234.
9. Id. at 431, 435, 573 P.2d at 454, 456, 143 Cal. Rptr. at 236, 239.
10. Id. at 430 n.10, 573 P.2d at 455 n.10, 143 Cal. Rptr. at 237 n.10.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 437

HISTORY: ONE STATE'S EXPERIENCE

The California court system was set up, after statehood, in 18

During the first half century of California law, not a single perso
injury action against a product manufacturer reached the appella
level, and only one such action against a product retailer did so."
1901, the supreme court, in an opinion marked by an unself-consc
blend of implied warranty and negligence law, affirmed a farme
$2,000 personal injury recovery against a company that had both m
ufactured a harvester and directly sold it to the farmer.12 During
next thirty-three years, California's appellate courts decided all of thr
actions brought by ultimate consumers against negligent manufact
ers.'3 In their opinions, the courts recognized a "privity of contrac
limitation on recovery, but also relied on a highly ambiguous "inh
ently dangerous" exception to that limitation. These ambiguities w
largely resolved in favor of product victims when the supreme co
adopted the MacPherson rule'4 in 1934.15
A year later, the court began affirming personal injury judgme
secured against product retailers under a theory of implied warrant
derived from the Uniform Sales Act, which California had enacte
back in 1931.16 These retailers were guilty of no negligence; imp
warranty was from the start a strict liability doctrine. Strict liability
of course, a constitutive feature of sales law and contract law generally
While the early warranty cases all concerned food products, the cou
opinions did not suggest that the food element was necessary to
strict liability result." In 1944, in Escola v. Coca Cola Bottling Co.

11. The retailer was held liable for willfully misrepresenting the safety of a bed purchased
a landlord for use by his tenant. Suit was brought by the tenant injured when the bed collaps
Lewis v. Terry, 111 Cal. 39, 43 P. 398 (1896).
12. Snyder v. Holt Mfg. Co., 134 Cal. 324, 66 P. 311 (1901).
13. Kolberg v. Sherwin-Williams Co., 93 Cal. App. 609, 269 P. 975 (1st Dist. 1928); Clif
California Spray Chem. Co., 83 Cal. App. 424, 257 P. 99 (3d Dist. 1927); Catlin v. Union Oil
31 Cal. App. 597, 161 P. 29 (2d Dist. 1916). See also Dahms v. General Elevator Co., 214
733, 7 P.2d 1013 (1932); Fidelity & Casualty Co. v. Parafine Paint Co., 188 Cal. 184, 204 P.
(1922).
14. MacPherson v. Buick Motor Co., 217 N.Y. 382, 111 N.E. 1050 (1916).
15. Kalash v. Los Angeles Ladder Co., 1 Cal. 2d 229, 34 P.2d 481 (1934).
16. Goetten v. Owl Drug Co., 6 Cal. 2d 683, 59 P.2d 142 (1936); Mix v. Ingersoll Candy Co.,
6 Cal. 2d 674, 59 P.2d 144 (1936); Gindraux v. Maurice Mercantile Co., 4 Cal. 2d 206, 47 P.2d 708
(1935).
17. Gindraux, the leading case, relied on both the general implied warranty doctrine and a
special food provision then in the Civil Code. By basing its warranty reasoning on Consolidated
Pipe Co. v. Gunn, 140 Cal. App. 412, 35 P.2d 350 (4th Dist. 1934), the court made clear that the
warranty doctrine extended beyond the food context. Gunn was an economic loss case involving
an unfit well-drilling casing shoe.
18. 24 Cal. 2d 453, 150 P.2d 436 (1944).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
438 CALIFORNIA LAW REVIEW [Vol. 67:435

the court liberalized the use of res ipsa loquitu

brought against manufacturers under a negligen
ing the plaintiffs burden of proving negligenc
manufacturing defects. Justice Traynor, concur
ticulated the formal strict products liability ide
Between 1944 and 1963, there were significan
ucts suits. When a manufacturing defect was i
were frequently successful under one theory or an
negligence19 and the incipient idea of an implied
rectly against the product manufacturer.20 Man
negligence in the product's design. California c
courts in certain other jurisdictions,21 were quite
able-minds-can-differ design cases to the jury.2
in dramatic anticipation of the later doctrine
able misuse,"23 even tolerated a suit against a c
injury caused by the chair's collapse when it w
this warranty and negligence litigation culminat
Justice Traynor, now speaking for the full court,
products liability rule.
Greenman couched the strict liability test in
product contained "a defect in design and man
plaintiff was not aware that made [the product]
use."25 Stimulated by Greenman, the American
a section for its Second Torts Restatement-dra
Dean Prosser-imposing strict liability on the sel
a defective condition unreasonably dangerou
sumer."26 Standing alone, "unreasonably dan
phrase that seemingly calls for a comparison of
associated with the alleged product defect.27 Bu
to section 402A, goes on to explain "unreaso
meaning dangerous "to an extent beyond that w
plated by the ordinary consumer who purchases
knowledge common to the community as to it

19. See, e.g., Burr v. Sherwin Williams Co., 42 Cal. 2d 682,

20. See Peterson v. Lamb Rubber Co., 54 Cal. 2d 339, 353 P.2d
21. See Noel, Manufacturers' Negligence of Design and Direct
YALE L.J. 816, 816 (1962).
22. See, e.g., Darling v. Caterpillar Tractor Co., 171 Cal. App.
1959); Brooks v. Allis-Chalmers Mfg. Co., 163 Cal. App. 2d 410,
23. See note 34 and accompanying text infra.
24. Phillips v. Ogle Aluminum Furniture, Inc., 106 Cal. App. 2
1951). Technically, this was a manufacturing defect case.
25. 59 Cal. 2d 57, 64, 377 P.2d 897, 901, 27 Cal. Rptr. 697, 7
26. RESTATEMENT (SECOND) OF TORTS ? 402A(1) (1965).
27. See Noel, supra note 21, at 818.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 439

provides a contract or warranty-like explanation of defect.

comment's illustrations of nondefective products-including
harms a diabetic and butter that heightens a person's c
level-are unhelpfully easy.28 In 1966, Reporter Prosser expl
the "unreasonably dangerous" standard was devised to prev
facturers from becoming "automatically responsible for all
that such things [their products] do in the world."29
In Escola and Greenman, Justice Traynor had evidently
"defect" as a rather easy concept;30 hence his belief that str
would vastly simplify the issues in products litigation.31 By th
an important speech he delivered in 1965, however, then-Ch
Traynor was beginning to appreciate various perplexities i
"defect."32 Subsequent to the adoption of the Restatem
supreme court's opinions shuttled back and forth between t
man language and the Restatement language.33 Then, i
Cronin decision,34 the court took two steps. First, it abandone
Traynor's "intended use" limitation in favor of a test maki
possible whenever the product's use is "reasonably foreseea
gardless of whether the use is otherwise unintended or imp
ond, Cronin held that the basic standard of liability should
rather than "unreasonably dangerous defect." On the latter
of the court's reasoning was that "unreasonably dangerous" i
ous, since the chief purpose of that language is to negate Repor
ser's fear of automatic liability and that purpose is fully a
"defect" standing alone. But Cronin also expressed the view
reasonably dangerous" is at least potentially nefarious. Here
was concerned with both style and substance. Stylistically, "
bly dangerous" is wrong because it "rings of negligence
"negligence complexion."35 Also, the court found the langu
of being misunderstood as subjecting the plaintiff to a two-ste
of proof, requiring him to prove not only the existence of a de

28. The comment does indicate that "good tobacco is not unreasonably dange
because the effects of smoking may be harmful."
29. Prosser, Strict Liability to the Consumer in Calfornia, 18 HASTINGS L.J. 9
30. The Escola majority opinion frequently referred to the product's "defect"
the negligence res ipsa issue, 24 Cal. 2d 453, 459, 461, 150 P.2d 436, 439, 440 (194
lihood, Justice Traynor's Escola concurrence merely accepted the majority's term
31. See, e.g., Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 463, 464, 150 P.2
442 (1944) (Traynor, J., concurring).
32. Traynor, The Ways and Meanings of Defective Products and Strict Liability,
REV. 363 (1965).
33. Compare Pike v. Frank G. Hough Co., 2 Cal. 3d 465, 476, 467 P.2d 229
Rptr. 629, 636 (1970) with Elmore v. American Motors Corp., 70 Cal. 2d 578, 451
Rptr. 652 (1969).
34. Cronin v. J.B.E. Olson Corp., 8 Cal. 3d 121, 501 P.2d 1153, 104 Cal. Rpt
35. 8 Cal. 3d at 132, 501 P.2d at 1162, 104 Cal. Rptr. at 442.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 440 CALIFORNIA LAW REVIEW [Vol. 67:435

also that the defect rendered the product unrea

dressing the substance of comment i, the cour
sumer expectations" approach, indicated an un
liability simply because a consumer's expectati
in some way-for example, by the obviousness o
The Cronin opinion, while stripping "defect"
dangerous" gloss, explicitly declined to set fort
"defect" definition.37 As a result, Cronin rece
reception in other jurisdictions38 and in the law
nia, litigation in post-Cronin cases-especially t
design-was confused, as lawyers and judges str
word "defect" standard. A few judges evidently
instructed in the language of "defect" and noth
professors, it seems fair to say that the gener
after Cronin, was that a design defect plaintiff ne
ence of some design alternative that would pro
acceptable cost. Court of appeal opinions after C
direction,41 although somewhat inconclusively.
The uncertainties engendered by Cronin prov
Barker. In Barker, the plaintiff-employee was s
he jumped off a high-lift loader which the defend
and then leased to the victim's employer. The
filling in for the loader's regular operator, who
The employee leaped from the loader after it b
being operated on a sharply sloping terrain.
launched a barrage of design defect accusation
turer.42 Although the trial in Barker was subs
confused trial judge gave the jury an "unreason

36. Cronin's companion case, Luque v. McLean, 8 Cal. 3d 136,

443 (1972), makes clear that even "patent" defects are covered by
37. See 8 Cal. 3d at 134 n.16, 501 P.2d at 1162 n.16, 104 Cal.
38. Compare Berkebile v. Brantly Helicopter Corp., 462 Pa. 83,
of Jones, C.J.) with Kirkland v. General Motors Corp., 521 P.2d
39. See, e.g., Keeton, Product Liability and the Meaning of De
(1973); Wade, On the Nature of Strict Tort Liability for Product
40. The California Trial Lawyers Association, as amicus in Bar
dorse this practice. See Barker v. Lull Eng'r Co., 20 Cal. 3d 413, 4
Rptr. 225, 234 (1978).
41. See Buccery v. General Motors Corp., 60 Cal. App. 3d 533,
14 (2d Dist. 1976); Baker v. Chrysler Corp., 55 Cal. App. 3d 710, 7
Dist. 1976).
42. See note 189 infra.
43. Telephone interview with Frank Hills, Barker's attorne
opinion gets this chronology wrong. See 20 Cal. 3d at 422-23, 57
231.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979 ] UNDERSTANDING PRODUCTS LIABILITY 441

instruction. The jury44 then returned a verdict for the def

ufacturer. The plaintiff appealed and won a reversal in
appeal on the ground that the trial court's instruction
Cronin. The manufacturer then secured a hearing in
court, arguing that in complicated cases involving produ
especially when the alleged design defect involves the o
safety device-the simple language of "defect" provid
guidance to juries; rather, for these cases the language o
bly dangerous" is sensible after all.45 The manufactu
tended for a selective return to the Restatement view.46
Rejecting this contention, the supreme court in its Ba
reaffirmed that "unreasonably dangerous" is no part of
"defect" definition. The court agreed, however, with th
turer's more general position that the word "defect," in i
adequate as a strict liability instruction in design cases. F
therefore, the court formulated the new "two-pronged" d

II

GENUINE STRICT LIABILITY

It is quite possible to conceive of a liability rule in the product

setting that is genuinely strict: this would be the rule (noted by Prosser
rendering product manufacturers automatically liable for all acciden
caused or occasioned by the use of their products.47 Thus, a car man
facturer would be liable for all injuries in which its cars are involved
power mower manufacturer would bear the liability for all pow
mower accidents, a ladder manufacturer would suffer liability for a
ladder falls, and a skateboard manufacturer would be fully liable f
all skateboarding injuries. Existing strict products liability falls dras
cally short of such a genuine strict liability rule, since it grants the vic-
tim a recovery only if the victim can demonstrate the existence o
product "defect." By stating that products liability does not cast t
manufacturer in the role of an "insurer,"48 opinions like Barker an
Cronin clearly (if somewhat awkwardly) reject the idea that the ex
isting liability rule is in any way equivalent to genuine strict liabilit
Of course, the res ipsa loquitur doctrine, given its Escola breakthroug

44. Barker's counsel describes the jury as the worst plaintiff's jury he has ever seen. Tel
phone interview with Frank Hills (Nov. 6, 1978).
45. Brief for Defendant at 1-2, 5-6.
46. The plaintiff urged the court simply to leave Cronin as is.
47. See Prosser, supra note 29, at 23. See also Wade, A Conspectus of Manufacturers'Liab
ity for Products, 10 IND. L. REV. 755, 768 (1977).
48. See Barker v. Lull Eng'r Co., 20 Cal. 3d 413, 432, 573 P.2d 443, 456, 143 Cal. Rptr. 22
238 (1978); Cronin v. J.B.E. Olson Corp., 8 Cal. 3d 121, 133, 501 P.2d 1153, 1162, 104 Cal. Rp
433, 442 (1972).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 442 CALIFORNIA LAW REVIEW [Vol. 67:435

is now seen as a way station to the Greenma

liability.49 Will today's defect doctrine be und
transitional stage in the evolution of a genuine
sibly, but this seems very doubtful. The Bar
adverting to the genuine strict liability idea,
And the keen interest in proposals like auto
liability on the product user-is out of line w
any automatic liability rule operating on pro
Yet the genuine strict liability rule calls for
rule is capable of being plausibly defended n
also by reference to those basic criteria for liab
by Justice Traynor in his historic Escola con
sively discussed in the contemporary tort lite
currence succinctly anticipated.52 Since t
provided the basis for the Greenman liability
conducive to genuine strict liability, and sinc
doctrine so discernibly hovers in the back
Barker,53 if we wish to appreciate the Greenm
ciple we would do well to consider the genuin
tive.

The analogy adducible on its behalf is, of

pensation. Workers' compensation is a model
in operation:54 the employer is held strictly liab
ing out of and in the course of the victim's emp
ers' compensation is a staple of the Ameri
usually regarded as a successful, if imperfec
recent years, it has been increasingly viewed
"precedent."

49. This was first glimpsed by Jaffe, Res Ipsa Loquitur Vind
15 (1951).
50. 20 Cal. 3d at 435, 573 P.2d at 457, 143 Cal. Rptr. at 239.
51. The leading work is G. CALABRESI, THE COSTS OF ACCIDENTS: A LEGAL AND Eco-
NOMIC ANALYSIS (1970). Also important is R. POSNER, ECONOMIC ANALYSIS OF LAW (2d ed.
1977).
52. Economic writings have frequently confused genuine strict liability with the law's ex-
isting "defect" liability rule, really discussing the former while professing to discuss the latter. See,
e.g., G. CALABRESI, supra note 51, at 161-65 (treating existing products liability as a "nearly ...
classic" example of "general deterrence," id. at 164, which is the equivalent of what is here called
genuine strict liability); Oi, The Economics of Product Safety, 4 BELL J. ECON. & MANAGEMENT
ScI. 3, 5, 14 (1973) (seemingly equating liability for "defective" products with liability for "risky"
products).
53. Thus, an Arizona court has found genuine strict liability implications in Barker's bur-
den-of-proof rule. Brady v. Melody Homes Mfr., [1978] PROD. LIAB. REP. (CCH) ? 8303 (Ariz.
App. 1978).
54. See G. CALABRESI, supra note 51, at 164-65.
55. See generally NATIONAL COMMISSION ON STATE WORKMEN'S COMPENSATION LAWS,
REPORT (1972).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 443

The policy analysis favoring genuine strict liability begin

goal of compensation through loss spreading, as first em
Justice Traynor's Escola concurrence.56 Undeniably, one
consequences of opinions like Escola and Greenman has b
ceptance of loss-spreading rhetoric in otherwise tradition
legal documents.57 Yet so long as the existing strict liabi
tinues to refuse to compensate the victims of nondefective p
alleged devotion to the loss-spreading ideal invites real s
By contrast, a genuine strict liability rule would obviousl
maximum in loss spreading for product-related injuries.
Moreover, in several ways a genuine strict liability rule m
tribute to the important objective of accident preventio
lighted by the Escola concurrence.59 To the extent that ce
injuries are inevitable, genuine strict liability would ach
prevention by allocating resources through the market m
Assuming that product purchasers are not aware (or not f
product dangers,61 genuine strict liability would raise the re
dangerous products; by so doing, it would accordingly di
sales-with a corresponding reduction in society's prod
rate.62 Insofar as product injuries are not inevitable, and
ing limited consumer knowledge, a genuine strict liabilit
create significant safety incentives. Manufacturers w
under a powerful and unremitting incentive (not requiri
vention by the jury) to adopt all appropriate safety devices63

56. 24 Cal. 2d 453, 462, 150 P.2d 436, 441 (1944) (Traynor, J., concurring).
nation of loss spreading, see G. CALABRESI, supra note 51, at 39-67.
57. See, e.g., RESTATEMENT (SECOND) OF TORTS ? 402A, Comment c (196
58. Writing in 1965, Dean Prosser regarded the loss-spreading argument for
ity as a "makeweight." Prosser, The Fall of the Citadel (Strict Liability to the Con
L. REV. 791, 800 (1965). There is not a word about loss spreading in Barke
General Motors Corp., loss spreading is discussed only in order to explain tha
should not be spread insofar as it is due to the victim's own faulty conduct
Motors Corp., 20 Cal. 3d 725, 737, 575 P.2d 1162, 1168-69, 144 Cal. Rptr. 380
59. 24 Cal. 2d 453, 462, 150 P.2d 436, 441 (1944) (Traynor, J., concurring)
60. See G. CALABRESI, supra note 51, at 68-94, on "general" or "market"
61. Even if purchasers do have knowledge of the danger, genuine strict liabil
market effects if the danger in the product runs toward third persons (for exa
whom the purchaser does not intend to protect.
62. See Michelman, Pollution as a Tort. A Non-Accidental Perspective on C
80 YALE L.J. 647, 659-60 (1971).
63. Under genuine strict liability, manufacturers would adopt all safety de
deem risk-beneficial. Under Barker "defect" liability, the manufacturer wil
safety devices which it deems risk-beneficial and which it foresees will also be f
cial by the jury. Insofar as particular jury pro-liability findings are foreseeable
turer but are nevertheless regarded by the manufacturer as incorrectly calculated
pay the resulting products liability verdicts rather than incur what it sees as th
safety device investment. See Calabresi, Torts-The Law of the Mixed Society,
519, 528, 530 (1978).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 444 CALIFORNIA LAW REVIEW [Vol. 67:435

ilar incentive to hasten the development of new

future.64

Arguments supporting a genuine strict liability rule can thus be

rationally marshalled. Whatever their force, however, there is no
avoiding the objection that many applications of the rule would fall
below minimum standards of fairness. Assume that a drunk driver
runs down a pedestrian, or that someone, allowing himself to becom
distracted, falls off an entirely well-built household ladder. M
would agree that it would be severely unfair to hold the car and lad
manufacturers liable for the resulting injuries.65 Claims of unfairn
must be reckoned with as a possible veto upon liability-rule propo
als.66

Moreover, each of the arguments inclining towards genuine strict

liability is susceptible to criticism. Take the compensation/loss-spread-
ing argument first. As an explanation for almost any products liability
rule, loss spreading is problematic. For one thing, a surprisingly large
number of the product victims who bring suit under even the existing
"defect" liability rule-for example, Gladys Escola and Ray Barker-
turn out to be employees who have suffered injuries on the job.67 The
availability of workers' compensation to reimburse these victims for
most of their out-of-pocket losses keenly embarrasses products liabil-
ity's supposed loss-spreading rationale. True, products liability does
allow recovery for elements of intangible harm like pain and suffering.
But this only attracts attention to another point: it is by no means clear
that the loss-spreading reasoning has any meaningful application to el-

64. See R. POSNER, supra note 51, at 138. Finally, a genuine strict liability rule--by remov-
ing the troublesome "defect" issue--could lower the "overhead" of the existing products liability
system. That overhead is indeed unfortunate: it is estimated that only 37.5 cents of every products
liability insurance dollar ultimately reaches product victims. See O'Connell, An Alternative to
Abandoning Tort Liability." Elective No-Fault Insurance for Many Kinds of Injuries, 60 MINN. L.
REV. 501, 511 (1976). Professor O'Connell's provocative proposal for a voluntary, limited liability
no-fault compensation program for product-caused injuries is beyond this Foreword's scope.
65. The perceived injustice of a genuine strict liability rule may lie in an intuitive sense that
the arguments supporting the rule are unconvincing and hence do not justify the heavy liability
burden. If so, then the fairness objection is not wholly independent of the criticisms of those
arguments set forth below.
66. See E. CAHN, THE SENSE OF INJUSTICE (1949); G. CALABRESI, supra note 51, at 24-26.
This analysis does not suggest any serious unfairness in workers' compensation. Workers' com-
pensation liability is genuinely strict-but it is also genuinely limited: the negligent employer es-
capes liability for full common law damages. The "quid pro quo" which was politically essential
to the original adoption of American workers' compensation also is helpful in establishing its
general fairness. Additionally, the employee's (careless) conduct that causes injury to the em-
ployee himself or to his fellow worker is conduct that is within the employer's legal control. This
fact alters the equities of the situation.
67. The federal study indicates that almost 50% of all products liability insurance "payouts"
go to employee claimants. FINAL REPORT, supra note 6, at VII-85.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 445

ements of noneconomic detriment such as pain and suf

workers' compensation-which evidently operates, at leas
loss-spreading principle--declines to take account of pain
is in this respect suggestive.69
More generally, the loss-spreading criterion, when of
tionale for any tort law rule, seems inherently unstable
basic sense promiscuous. If loss spreading is deemed th
mental purpose, a compensation right should accordingl
to the victim of every serious accident, without regard
ment in that accident of any product. Yet tort law as
"tort law" instead of a compensation program exactly b
lective-that is, because the liability rules it fashions exc
for some accident victims while permitting recovery for o
loss-spreading rhetoric is taken with sufficient seriousness
augers not the reform of tort law but rather (for bette
supersession of tort law by some general compensation
a program has now been adopted in New Zealand; and,
adoption, that country has abolished all forms of prod
strict liability, negligence, whatever.7' But so long as A
remains committed, whether wisely or not, to a tort law
loss-spreading criterion should be treated with caution.7
The accident prevention justification for a genuine st
liability rule is also questionable. That justification assu
manufacturer is the only party in any way responsible f
product-related accident. But this assumption is often
victim himself may be partly responsible, as in the ladd
scribed above. Or some other person may be partly res
the example of the drunk driver. Furthermore, the pr
second manufacturer may be partly responsible, as in th
a collision between a car and a train or, for that matte
Chevrolet car and a Ford car. For the purpose of acciden

68. See, e.g., Franklin, Tort Liabilityfor Hepatitis. An Analysis and a Pro
REV. 439, 464, 479 (1972); Franklin, Replacing the Negligence Lottery. Compen
Reimbursement, 53 VA. L. REV. 774, 800 (1967). Chief Justice Traynor, in disc
extension of products liability into a general compensation program for acciden
the withholding of any award for pain and suffering. Traynor, supra note 32,
69. See Wights v. Staff Jennings, Inc., 241 Or. 301, 310, 405 P.2d 624, 6
70. See the already classic statement of this position in R. KEETON & J. O
PROTECTION FOR THE TRAFFIC VICTIM 242 (1965).
71. See Palmer, Dangerous Products and the Consumer in New Zealand, 1
366.

72. A major objection to a general compensation program is that it could lead to terrible
results in accident prevention. See G. CALABRESI, supra note 51, at 64-67. If genuine strict prod-
ucts liability could accomplish loss spreading while also promoting accident prevention--or at
least without greatly interfering with the accident prevention effort-it would escape this objec-
tion.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 446 CALIFORNIA LAW REVIEW [Vol. 67:435

in the ladder and drunk driving examples an unfet

strict liability, by focusing solely on the acciden
cause while ignoring all its foreground behavio
uncertain merit. And for accidents involving th
product, it is not even clear what that rule would
operate.73
To be sure, a genuine strict liability rule, if adopted, would pre-
sumably be absorbed into the general fabric of California's tort law.
Under Daly v. General Motors Corp.74-decided two months after
Barker-if the victim's unreasonable conduct has contributed to a
product-related accident, the victim's recovery is diminished to
tent of his own contributory negligence;75 hence the distracte
who falls off the satisfactory ladder would probably be awarde
small amount. Also, under American Motorcycle Association v.
Court76 and Safeway Stores, Inc. v. Nest-Kart77--decided one
and four months after Barker, respectively-the auto manufa
sued by the victim of the drunk driver would be entitled to comp
indemnification from the driver himself. Given the facts as st
jury would undoubtedly assign the lion's share of the liability
driver.78 Taken in combination, then, Daly, American Motorcy
Safeway would considerably reduce the impact of a genuine st
bility rule. The Daly-Safeway sequence, however, offers little
the simple cases of the car-train and two-car accidents. He
safety benefits to be derived from genuine strict liability woul
pendent upon how liability is apportioned between the product

73. Consider the fencer injured when a sabre pierces his fencing mask. See Garcia
Vince Co., 84 Cal. App. 3d 868, 148 Cal. Rptr. 843 (2d Dist. 1978). Or consider a p
injures his back lifting a heavy bowling ball out of the deep trunk of an automobile: is
attributable to the bowling ball, the car, both, or neither? For an accident hypothetica
four products, see Blum, Book Review, 43 U. CHI. L. REV. 217 (1975). See also Ph
Standard for Determining Defectiveness in Products Liability, 46 U. CIN. L. REV. 1
(1977). In workers' compensation, the question of what injuries are legally attribut
victim's employment requires over a thousand pages of careful treatisewriting to explai
LARSON, 1 WORKMEN'S COMPENSATION chs. III-IV (1978). And workers' injuries rarely
problems of multiple employers.
74. 20 Cal. 3d 725, 575 P.2d 1162, 144 Cal. Rptr. 380 (1978).
75. For a discussion of genuine strict liability with a full contributory negligence (a
to comparative negligence) defense, see Posner, Strict Liability.: A Comment, 2 J. LE
205, 207 (1973). Attempts to utilize a contributory negligence defense to give safety inc
potential accident victims are replete with difficulties. See Schwartz, Contributory and
tive Negligence.- A Reappraisal, 87 YALE L.J. 697 (1978). Comparative negligence ma
function about as well as any other possible rule. See id. at 721, 727. This Foreword
the Daly rule as an adequate response to the phenomenon of victim causation of prod
injuries.
76. 20 Cal. 3d 578, 578 P.2d 899, 146 Cal. Rptr. 182 (1978).
77. 21 Cal. 3d 322, 579 P.2d 441, 146 Cal. Rptr. 550 (1978).
78. Safeway emphasizes submitting the apportionment issue to the good sense of the jury.
21 Cal. 3d at 332-33, 579 P.2d at 446, 146 Cal. Rptr. at 585.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 447

facturers. But if liability were imposed on the sole basis

factual involvement in an accident, one does not even
methodology to utilize in developing a "correct" app
formula.79

These doubts reveal the larger flaw in the accident-preve

sis. Almost any accident, examined closely enough, can
ing at the intersection of more than one activity or enterpr
this intersection quality, the process of attributing the accid
ticular activity or enterprise is inherently uncertain. Ind
certainties virtually compel the abandonment of the gen
allocation argument for strict liability;8" instead, about a
done is to choose among contributing activities or enter
basis of which of them may be in the best position to re
dent risk in a cost-effective way.82 But this reformulatio
dent prevention strategy brings us back at least to the ge
of the defect requirement in existing products law.83
In sum, because of the momentum provided by Justi
Escola reasoning, the idea of genuine strict liability is qui
cussing. In light of the problems that such a liability rul
ally raise, however, it is neither surprising nor obviously
that the supreme court has chosen to remain with the far m
ate rule of strict liability for product "defects" only. Bu

79. In a typical Daly case, the defendant's defective product and the plainti
conduct both entail departures from relevant safety norms, and a comparison of
two departures may be feasible. But even comparisons of this sort would not
genuine strict liability, given the absence of any requirement of subnormality.
80. See Weiler, Defamation, Enterprise Liability, and Freedom of Speech,
L.J., 278, 298 (1967):
For instance, in a level-crossing collision between automobile and train en
following "activities" (along with others) may be present: the railroad; the m
of the various components of the railroad; the various types of users of t
train (passenger, freight, etc.); the driver of an automobile; the business i
automobile was being used; the purpose for which a passenger was "hitching
care and raising of a family by the hired chauffer of the automobile; the man
this particular type of automobile; the manufacturer of the railroad; the bu
highway (province); the owner of the highway who maintains the crossing
ity). Unless most or all of these different "individuals" had engaged in vario
which coalesced in the concrete accident-situation, the harm would never ha
81. See G. CALABRESI, supra note 51, at 138: Because "judgments as to wh
the cheapest long-run cost avoider" are so very hard to make, "in many situatio
ignore the long-run judgment, as one guess is as good as another." This concessio
wiping out the resource allocation argument which Calabresi has so carefully
82. See id.: "[C]oncentrating on the short-run judgment [as to the best cos
best we can do."

83. The reduction of overhead rationale can also be reconsidered. Even if a genui
liability rule dispensed with the defect requirement, products suits would still be pla
difficult factual and legal issues-including those suggested by Daly, AMA, and Safe
the novel legal concept of accidents "caused or occasioned by the use of a product"
certain to generate legal controversies.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 448 CALIFORNIA LAW REVIEW [Vol. 67:435

opinion does leave open-in truth, it deliberatel

one limited but interesting application of the
idea. According to a Barker footnote, strict lia
nized, without proof of ordinary defect, for p
such that their "norm is danger."84 The advisab
for defect-free "norm-is-danger" products wi
section of this Foreword. Also, that section wil
foregoing analysis of genuine strict liability in
gestion of a "hindsight" approach to its risk-ben

III

EXISTING STRICT "DEFECT" LIABILITY

A. The Tort and Contract Basesfor Strict Liability

By the time Greenman introduced strict liability, negligence
already afforded the injured product victim generous compensat
rights against a negligent manufacturer, and implied warranty law
conferred on the product consumer a strict liability right runni
against the product retailer, if not the product manufacturer.86 In
Escola concurrence, Justice Traynor relied on both the negligence
the warranty precedents in formulating the strict liability prop
which he later accepted, on behalf of the entire court, in Greenman; a
the Greenman doctrine has itself been explained as an imaginative
thesis of negligence and warranty.87
If Greenman can be seen as synthesizing negligence and warran
however, Barker's two-pronged design defect test threatens to ana
away this synthesis. As the Barker opinion recognizes, its ordina
consumer expectations prong manifests the warranty heritage up
which California products liability doctrine partially rests.88 Ind
'"consumer expectations" is a plausible restatement of the criterion
liability in a common implied warranty action.89 Meanwhile, Bark

84. See 20 Cal. 3d at 431 n.10, 573 P.2d at 455 n.10, 143 Cal. Rptr. at 237 n.10.
85. It is sometimes suggested that tort warranty originally sounded in tort. What we
call express warranty has an observable tort background, through its affiliation with traditio
torts like deceit and misrepresentation. See S. WILLISTON, 1 LAW OF SALES OF GOODS ?? 195
(3d ed. 1948). Insofar as implied warranty can possibly be characterized as a tort, it is a t
intrinsically and necessarily connected to the sales transaction. Modern scholarship indicates t
the implied warranty of merchantability can best be understood as an element of contract. S
WARREN, W. HOGAN, & R. JORDAN, COMMERCIAL AND CONSUMER TRANSACTIONS 2 (2
1975); Prosser, The Implied Warranty of Merchantable Quality, 27 MINN. L. REV. 117, 125
(1943); text accompanying note 96 infra.
86. See notes 16, 17, 20 and accompanying text supra.
87. G. WHITE, THE AMERICAN JUDICIAL TRADITION 297, 299 (1976).
88. 20 Cal. 3d at 430-31, 573 P.2d at 454, 143 Cal. Rptr. at 236.
89. Barker indicates that its consumer expectations prong is "somewhat analagous to t

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 449

risk-benefit prong reveals the strong indebtedness of stric

trine to tort law's negligence principle, which itself has his
understood as calling for a risk-benefit balancing.90 Ba
tions of the warranty and negligence bases of strict liab
invite an assessment of the logic underlying the neglig
ranty theories.
Negligence law, in its product applications, seems at
no account of the point that the product manufacturer stan
of contractual relationship with the product purchaser;
the manufacturer the same obligation of reasonable care
defendants in "stranger" cases9" of the sort typified
Kendall.92 This obligation is anchored in a fairness pri
potential injurer should not egoistically rank his own we
welfare of others)93 and an accident prevention principl
should be discouraged from engaging in conduct which e
others that exceed its benefits to the actor).94 Warrant
trast, is drawn directly from the essentially contractual re
tween the consumer and the manufacturer.95 Its purp
effect to the reasonable expectations which that relation
in the parties, particularly the consumer; "the assumptio
parties themselves [to a contract of sale], had they thought
specifically have so agreed [to a merchantability guarant
The tort dimension of strict "defect" liability has bee
in recent years on grounds that juries are somewhat prone
favorable to plaintiffs in product design cases,97 and o
basis that sophisticated risk-benefit balancing is somet
process of adjudication is incapable of handling.98 Neithe

Uniform Commercial Code's warranty of fitness and merchantability." 20 C

P.2d at 454, 143 Cal. Rptr. at 236.
90. See United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir. 19
Weirum v. RKO General, Inc., 15 Cal. 3d 40, 48, 539 P.2d 36, 40, 123 Cal. Rp
As long as this indebtedness is recognized, there is no irony in finding so f
negligence law's balancing test in a strict liability opinion.
91. See Posner, A Theory of Negligence, 1 J. LEGAL STUD. 29, 36-37 (19
92. 60 Mass. (6 Cush.) 292 (1850).
93. See Schwartz, supra note 75, at 701-02.
94. See id. at 703.

95. See note 116 and accompanying text infra.

96. Hauter v. Zogarts, 14 Cal. 3d 104, 117, 534 P.2d 377, 385, 120 Cal. Rptr. 681
(quoting Ezer, The Impact of the Uniform Commercial Code on the Calfornia Law
ranties, 8 U.C.L.A. L. REV. 281, 292 (1961). See also W. WARREN, W. HOGAN, & R
supra note 85, at 26; A. Schwartz, Products Liability and Judicial Wealth Distribution
558, 575 (1976).
97. This objection seems implicit in Professor Epstein's insistence that "the plain
bear a heavy burden before he can take a case of unsafe design to the jury." Epste
Liability.: The Search for the Middle Ground, 56 N.C.L. REv. 643, 650 (1978).
98. See Henderson, JudicialReview of Manufacturers' Conscious Design Choice. T

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 450 CALIFORNIA LAW REVIEW [Vol. 67:435

jections is entirely persuasive. The complaint ab

pirical verification. A recent federal study of p
collecting data from several jurisdictions (includ
ported that at least half of all jury verdicts in
favor of the defense;99 and practicing lawyers r
juries seem puzzled, if not disturbed, by the very
ity.'" The second objection rests on the claim
cases are "polycentric"'o' in Lon Fuller's sense
for adjudication. Given Fuller's own examples of
claim seems exaggerated.103 Those who object
apparently prefer to assign product design issue
process for purposes of prospective regulation.
runs the risk of both idealizing the performance o
cies'05 and neglecting the fact that they can n
more than a limited number of product design
products liability from cases involving consciou
function to undermine the law's accident prev
that class of cases in which it can most realistically
Exactly because design decisions are conscious
level company officials, there is good reason to
decisionmaking will be influenced by those offi

Adjudication, 73 COLUM. L. REV. 1531 (1973); Henderson, Design

CORNELL L. REV. 541 (1976). In England, issues of product design a
the judicial function. See P. ATIYAH, ACCIDENTS, COMPENSATIO
1977).
99. FINAL REPORT, supra note 6, at 11-46, VI-39. See also the jury's almost incredible de-
fense verdict in Hauter v. Zogarts, 14 Cal. 3d 104, 534 P.2d 377, 120 Cal. Rptr. 681 (1975).
100. In light of this, when pretrial discovery uncovers some evidence suggesting the manufac-
turer's negligence, lawyers believe there are strong reasons for bringing it to the jury's attention.
In general, a plaintiffs' lawyer's success in products cases greatly depends upon his ability to ex-
plain the strict liability doctrine to the jury in a persuasive fashion.
101. Fuller, The Forms and Limits ofAdjudication, 92 HARV. L. REV. 353, 394-95 (1978).
102. Fuller's chief examples of "polycentric" problems are the division of a large art collec-
tion between two art museums, the promulgation of comprehensive wage and price controls, and
the establishment of lineups for a football team. Id. He makes explicit that the problem of the
design of railroad crossings, if challenged in a negligence suit, is not so polycentric as to resist
adjudication. Id. at 397-98.
103. See the criticism of the Henderson position in Twerski, Weinstein, Donaher, & Piehler,
The Use andAbuse of Warnings in Products Liabilit---Design Defect Litigation Comes ofAge, 61
CORNELL L. REV. 495, 524-34 (1976).
104. In his 1973 article, Professor Henderson relies chiefly on agency regulation as the means
for dealing with problems of "conscious design choices." Henderson, Judicial Review of Manufac-
turers' Conscious Design Choice. The Limits ofAdjudication, 73 COLUM. L. REV. 1531, 1568, 1573-
77 (1973). His 1978 article is somewhat less negative about the role of liability law. Henderson,
Manufacturers'Liabilityfor Defective Product Design. A Proposalfor Statutory Reform, 56 N.C. L.
REV. 625 (1978).
105. See Schorr, Consumer Product Safety Panel Discovers Big Hazards in Just Getting Itself
Rolling, Wall St. J., May 6, 1975, at 40, col. 1.
106. See G. CALABRESI, supra note 51, at 108-09.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 451

company's potential liability.'07

Neither of these objections, therefore, succeeds in establ
tort law should retreat from product design controversi
objections should by no means be ignored. It is undeniab
issues in design cases often become extremely difficult,
apparent that when a products case is sent to the jury under
instructions, almost anything can happen. The objections
valuable inasmuch as they suggest important criteria to r
uating the law's design-defect rules. If product design l
achieve satisfying results, the law is obliged to formula
which identify the proper variables, which are capable of
istered in the civil trial context, and which provide the jury
tive, intelligent guidance. These criteria will be returned
in the evaluation of Barker in Parts III and IV.

Historically, the chief limitation on the application of tort pr

ples in products cases was located in the rule of privity. That ru
the extent it was ever in effect in Californial09-barred negligence
against a manufacturer by anyone other than the person who had
ally purchased the product from the manufacturer. From a contem
rary vantage point, this traditional privity rule had hold o
important idea, but somehow managed to invert its significance. W
parties have entered into a contractual relationship, it is permissib
argue that the rights and liabilities running between them shou
deemed exclusively established by the provisions of their contract
press and implicit. I" By contrast, when parties have not joined in
contract, the "jurisdiction" of tort law can be most emphatically

107. For any who may be inclined to regard this hope as an ivory tower construct, l
simply quote from my morning newspaper:
In 1960 Andrea McCormack, then 3 years old, received third degree burns over mor
than 30% of her body when she accidentally tipped over a vaporizer filled with scalding
water. .

In 1967, she received a $150,000 jury verdict levied against the vaporizer's man
Hankscraft Co.
In upholding the jury award the Supreme Court of Minnesota determined . . the defect
could have been eliminated by adopting any one of several "practical and inexpensive"
alternative designs to secure the top of the jar, such as putting threads on the inside of
the plastic top. . .. [McCormack v. Hankscraft Co., 278 Minn. 322, 335, 154 N.W. 2d
488, 498 (1967)].
Today, most vaporizers have more secure tops and are designed so that the water in the
reservoir does not reach dangerous temperatures.
L.A. Times, Oct. 22, 1978, ? V, at 2, col. 2.
108. Professor Epstein, for example, would evidently immunize manufacturers from design
liability so long as their design was "in substantial use" within the industry at the time of th
product's sale. If there were two or more designs "in substantial use" at a particular time, al
would enjoy the immunity. Epstein, supra note 97, at 662 n.68.
109. See notes 11-15 and accompanying text supra.
110. See Posner, supra note 91, at 37.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 452 CALIFORNIA LAW REVIEW [Vol. 67:435

firmed."' There should be no doubt, therefore,

comes within products liability's tort-based pr
proach the same point from a different directio
ings most vividly verify that California's products
a major tort dimension, one which prevents that r
alized in a purely contractual fashion.'
No one should deny the significant contractu
in strict products liability, however, especially i
precursor. In suits brought by actual product pu
doctrine that somewhat encumbered pre-Green
can be seen as erroneous, even in strictly contr
antees of product quality implicit in the purchas
understood as running against the manufacturer ra
retailer if the purchaser expects the former to be
for product quality."' Given people's general an
ing that product defects are typically within th
trol-after all, defects both enter the product durin
process and are most easily detected there-t
properly be regarded as a participant in or acce
immediate contract with the product purchaser.
This contractual feature in products liability
commented on in recent years by prominent law
analysts have reasoned that as a matter of econo
liability rules are essentially irrelevant, given th
nism reacts to a product's safety characteristics
quiring manufacturers of more dangerous pr
prices, gives manufacturers an incentive to make t

111. See Epstein, Medical Malpractice. The Case for Contract

RESEARCH J. 87, 94; Fletcher, Fairness and Utility in Tort Theory,
546 n.38 (1972).
For an indication, however, that the manufacturer and the actu
in a kind of contractual relationship, see notes 114-16 and accompan
112. Epstein, supra note 97, at 647.
113. In affirming the bystander's products liability cause of actio
sized that products liability is "governed" by "tort rules" rather than
Elmore v. American Motors Corp., 70 Cal. 2d 578, 585, 451 P.2d
(1969).
114. See note 20 and accompanying text supra.
115. See R. POSNER, supra note 51, at 135. This assessment is implicit in all the economic
works cited in notes 117 and 122 infra.
116. See A. Schwartz, supra note 96, at 575. Alan Schwartz now advises me that he has
backed away somewhat from the conclusion he had earlier reached in applying this test. See id. at
574-78.

117. See G. CALABRESI, supra note 51, at 162-63; R. POSNER, supra note 51, at 134-36; Edit
Transcript of AALS-ABA Conference on Products Liability, 38 U. CHI. L. REV. 117, 121 (197
(Statement of G. Calabresi) (hereinafter cited as Statement of G. Calabresi).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 453

the extent this can be done at reasonable cost; informed con

purchase the degree of product safety that they individually
Most of the time, however, the economic reasoning goes on
ceive that given various practicalities concerning imperfect
information, products liability rules ultimately do acquire
meaning."1 This conclusion I entirely agree with; but the l
soning I find questionable. Knowledge is not vested in co
priori; it must always be acquired. Through day-to-day livi
people undoubtedly pick up general knowledge of certain ki
neric product hazards-for example, that matches can burn
sling-shots can injure."19 But people do not have easy access
edge of defects in particular products, whether the defects are
ufacturing or a design sort.120 Moreover, no matter how co
the consumer, the acquisition of knowledge about product
possible at all, is a burdensome, time-consuming, and unple
business.

To be sure, economists have developed a concept of "information

costs" that is hospitable to these observations-indeed, the very pur-
pose of the information-cost concept is to bring observations of this sort
into the substance of economic analysis. The concept is, however, un-
stable, for it concedes that people start out in ignorance of product de-
fects and can obtain knowledge of these defects, if at all, only by
"costly" efforts. Yet if this is true, it is far from clear why economic
theory should choose to begin its analysis with the assumption of full

118. See G. CALABRESI, supra note 51, at 163; R. POSNER, supra note 51, at 136-37; Statement
of G. Calabresi, supra note 117, at 121-22.
119. See Bojorquez v. House of Toys, Inc., 62 Cal. App. 3d 930, 133 Cal. Rptr. 483 (4th Dist.
1976) (hazard obvious to all, hence no duty to warn).
120. The best source of information may be Consumer Reports. Yet for products other than
automobiles, it is usually impossible to find an issue of that magazine that is recent enough and
specific enough (as to particular models) to satisfy one's informational needs. Anyway, Consumer
Reports generally gives no information on manufacturing defects except in the case of
automobiles, where the defects it counts (rarely of a safety nature) are evidently those detectable
during a fairly brief driving experience. The magazine does attempt to identify design defects in
the products it tests. But frequently it is unable to detect rather serious defects. See, e.g., Radial
Tires, CONSUMER REPORTS, Oct. 1973, at 604 (rating Firestone 500 radials "well above average");
Six Subcompact Cars, CONSUMER REPORTS, June 1973, at 409 (rating the Ford Pinto last among
six and criticizing the small capacity of the Pinto's gas tank, but making no mention of gas tank
location). And occasionally the magazine overstates minor design problems, thereby damaging its
credibility. Compare Chrysler's Big Mistakes, CONSUMER REPORTS, July 1978, at 376 (rating the
Omni/Horizon "not acceptable") with U.S. DEP'T OF TRANSPORTATION, PRESS RELEASE NHTSA
74-78 (July 7, 1978) (finding no real world safety problem). Of course, Consumers Union en-
counters difficulties-"costs"-in acquiring its information.
121. The consumer product most likely to be involved in accidents is the automobile. Espe-
cially given the negotiability of auto prices, many consumers find the experience of shopping for
cars so distressing that they knowingly buy quickly so as to cut short the distressful episode. Inter-
view with Martha Scallon, Secretary to Judge J. Skelly Wright, en route to Dulles Airport (Oct.
16, 1978).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
454 CALIFORNIA LAW REVIEW [Vol. 67:435

consumer information. To regard information c

of assumption-relaxing practicality is to encou
to understate their magnitude122 or to ignore the
account of simple neglect'23 or because they det
of a pure economic argument.'24 Economic mo
helpful in attempting to understand products l
will help most if it bases itself on an original
sumer knowledge of product defects, and subse
sumption to be modified only to the exten
incurring reasonable costs, can acquire some "de
implied warranty and the affiliated Barker c
prong profess to recognize that a particular pro
what the consumer might have otherwise assum
quality. By avoiding the potential for inaccuracy
tional economic analysis, these doctrines are to
Strict products liability, as noted, can be exp
of negligence and warranty law.'25 This explan
one sense, however, since even after Greenman
mains free to bring suit under either a negligence
ranty'27 theory. Each theory has not only "surv

122. Professor Oi relaxes his full information assumption in o

Oi, supra note 52, at 22. The only information problem whose ex
cedes is that a seller, knowing of a design hazard in a competitor
advertise that hazard for fear of raising the entire safety issue in
might lead them to buy a safer "substitute" product instead. R. PO
But the processes by which manufacturers and also retailers (many
products) are able to learn of the design dangers in competitors' pr
cost-free. And there is a more fundamental problem: no intelli
anything resembling face value the statements of one seller about
able characteristics) of a competitor's product.
The competitor's tort of product disparagement permits recover
and can be avoided if the disparagement is presented as an opinion.
OF THE LAW OF TORTS ? 125, at 922-24, 926 (4th ed. 1971). Mor
incur costs if it wishes to monitor what is being said about its pro
retail outlets. For whatever combination of reasons, product disp
123. See W. BLUM & H. KALVEN, PUBLIC LAW PERSPECTIVES O

58-59 (1965), later corrected in Blum & Kalven, The Empty Cabin
dents and General Deterence, 34 U. CHI. L. REV. 239, 243 n.23 (1
For strong discussions (from different perspectives) of limited
Schwartz, Private Law Treatment of Defective Products in Sales S
Shapo, A Representational Theory of Consumer Protection. Doctrin
for Product Disappointment, 60 VA. L. REV. 1111, 1127-29, 1302-
124. I suspect this is why Professor Demsetz assumes perfect
implications of alternative liability rules for workers' injuries. De
the Ownership of Rights, 1 J. LEGAL STUD. 223, 224-27 (1972).
125. See note 87 and accompanying text supra.
126. Jiminez v. Sears, Roebuck & Co., 4 Cal. 3d 379, 482 P.2d 6
127. Hauter v. Zogarts, 14 Cal. 3d 104, 534 P.2d 377, 120 Cal.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 455

of strict products liability,128 but indeed has flourished, fo

since Greenman both negligence and implied warranty ha
development. Thus, the supreme court has affirmed tha
principles apply in a meaningful general way to issues of
sign. 29 As for implied warranty, the consumer's right to su
facturer (in addition to the retailer) under an implied war
has been established,'3 as has the applicability of th
problems of product design."'3 Furthermore, in the years
man plaintiffs' lawyers have developed greater skills in
facts of a products case; these enhanced skills make it e
plaintiff to prevail, regardless of the particular products th
If strict products liability is compared to early or im
tems of warranty and negligence, the importance of the
synthesis may seem immense. The proper comparison, ho
tween a mature strict liability system on the one hand and m
ranty and negligence systems on the other-a mature negl
being one that has dispensed with privity, approved of res ip
extended itself to matters of product design, and acquire
cated plaintiffs' bar. In this comparison, the impact of strict
far less dramatic. What is it that the Greenman rule real
It is now widely appreciated that affirmative defens
and-parcel of any liability rule. Can strict liability be d
from negligence and warranty in terms of available defe
time of Barker, an overwhelming difference between strict
negligence lay in their treatment of the victim's contrib
gence. In a Greenman strict liability action, ordinary contrib

128. Thus, after Barker, in a product design case the plaintiff can allege ne
plied warranty, and the strict liability risk-benefit prong, and the strict liability
tions prong.
129. Pike v. Frank G. Hough Co., 2 Cal. 3d 465, 467 P.2d 229, 85 Cal. Rp
130. See Hauter v. Zogarts, 14 Cal. 3d 104, 534 P.2d 377, 120 Cal. Rptr. 681
sary implication).
131. See id.

132. The focus here will be on the manufacturer, who is the primary defendant
liability purposes. Suit is allowed against the retailer, Vandermark v. Ford Motor
256, 391 P.2d 168, 37 Cal. Rptr. 898 (1964), but, absent special circumstances, th
secure indemnification from the manufacturer, see Pearson Ford Co., v. Ford Moto
App. 2d 269, 78 Cal. Rptr. 279, 282 (4th Dist. 1969), under a classic implied warra
nothing else. See Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 462, 150 P.2d 4
(Traynor, J., concurring). If there is a wholesaler subject to strict liability, the analy
liability ultimately comes to rest on the manufacturer.
California law imposes strict liability on the in-the-business product lessor. See
Oil Co., 2 Cal. 3d 345, 466 P.2d 722, 85 Cal. Rptr. 178 (1970). In those (perhaps fe
in which (a) the defect cannot be traced back to the product at the time when the le
it, and (b) the lessor has been guilty of no negligence in inspecting or maintaining t
Price rule produces a new and interesting strict liability result.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 456 CALIFORNIA LAW REVIEW [Vol. 67:435

ligence had no effect upon the victim's reco

negligence claim, however, contributory neglig
plete defense until Li v. Yellow Cab Co.134 in
defense subsequent to Li. But in Daly v. Genera
cided a month after Barker, a divided supreme
contributory negligence liability-reducing rule i
liability setting. This ruling all but bridges the dr
ously separated negligence from strict liability. Th
utory/comparative negligence has of course
environment, and its status in a contract-oriented
tion has never been authoritatively clarified.'36
ple underlying comparative negligence-that a d
be required to reimburse the plaintiff for the full
which the plaintiffs own foolish conduct was
does not seem threatened by any transition from
for those who see the purpose of comparative n
potential victims with influential safety incent
hardly incompatible with the risk-assigning criter
at work in contract doctrine.139 In all, it would
wrong if, after Daly, the supreme court were to a
cape the effects of comparative negligence by
ranty.
Liability disclaimers can be regarded as a special form of affirma-
tive defense. While disclaimers of strict liability are presumptively in-
valid,140 disclaimers of negligence liability fall under a similar cloud.141
Implied warranty's contract framework may suggest that such a war-
ranty is inherently vulnerable to supersession by an express contractual
term.142 Yet the U.C.C., in setting forth implied warranty doctrine,
equivocates badly on the disclaimer issue, and at least permits the in-
ference that as a general matter implied warranty liability for personal
injuries cannot be disclaimed.143 If, however, a seller should list its

133. Luque v. McLean, 8 Cal. 2d 136, 501 P.2d 1163, 104 Cal. Rptr. 443 (1972).
134. 13 Cal. 3d 804, 532 P.2d 1226, 119 Cal. Rptr. 858 (1975).
135. 20 Cal. 3d 725, 575 P.2d 1162, 144 Cal. Rptr. 380 (1978).
136. See L. FRUMER & M. FREIDMAN, 2 PRODUCTS LIABILITY ? 16.01[3] (1978).
137. See Schwartz, supra note 75, at 721-27.
138. I am not one of them. See id. at 703-21.
139. See R. POSNER, supra note 5 1, at 74, 77-79.
140. See Seely v. White Motor Co., 63 Cal. 2d 9, 17, 403 P.2d 145, 150, 45 Cal. Rptr. 17, 22
(1965) (dictum). All that Vandermark seems to hold is that a disclaimer of contractual liabilit
does not profess to apply to a tort liability claim. Vandermark v. Ford Motor Co., 61 Cal. 2d 256
391 P.2d 168, 37 Cal. Rptr. 896 (1964).
141. See Tunkl v. Regents of Univ. of Cal., 60 Cal. 2d 92, 383 P.2d 441, 32 Cal. Rptr. 3
(1963).
142. See S. WILLISTON, supra note 85, ? 237.
143. Compare CAL. COM. CODE ? 2316 (West 1964) (establishing that implied warranty is

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 457

product at a regular price but also offer the consumer

price discount if the latter chooses to accept a liability discla
obvious contract reasons there is a good chance that
honor the disclaimer, whether it is attacked on strict lia
gence, or warranty grounds.
Lying between comparative negligence and liability d
the defense of assumption of risk. For negligence and s
purposes, Li and Daly "merge" assumption of risk into c
negligence and thereby "abolish" it as a complete defens
the extent that assumption of risk is no more than a "v
"form," of contributory negligence. 45 In cases where t
discovers the defect in his product at some time after h
purchase, I agree that contributory/comparative negligen
the exclusive issue, whatever the liability theory. But whe
is fully disclosed to the consumer prior to his product
express knowledge of that hazard seemingly becomes
purchase contract in a way that clearly weakens any later
ranty claim.'46 Consider now the purchaser who not only
product hazard prior to the purchase, but who receives
from the sales transaction that is necessarily tied to that
who chooses to accept the hazard in order to secure that ben
sent a policy of extreme paternalism, in this situation th
victim's assumption of risk should override not only an

generally disclaimable by conspicuous language) with CAL. COM. CODE ? 2

(specifying that any "limitation of consequential damages for injury to the pers
plied warranty exists is "prima facie unconscionable"). For a discussion of th
inconsistency," see Franklin, When Worlds Collide. Liability Theories and Disclai
Product Cases, 18 STAN. L. REV. 974, 995 (1966). If the U.C.C. does indeed
injury disclaimers, the U.C.C. is arguably using the implied warranty doctrine fo
purposes. See W. WARREN, W. HOGAN, & R. JORDAN, supra note 85, at 77.
MORE, THE DEATH OF CONTRACT 94 (1974).
144. See Franklin, supra note 143 at 1006, 1010-11.
145. See Li v. Yellow Cab Co., 13 Cal. 3d 804, 826, 532 P.2d 1226, 1241, 119
873 (1975); Daly v. General Motors Corp., 20 Cal. 3d 725, 739, 743, 575 P.2d 11
144 Cal. Rptr. 380, 387-388, 390 (1978).
146. See Pritchard v. Liggett & Myers Tobacco Co., 350 F.2d 479, 484-8
(dictum) (assumption of risk in its "primary" or "strict" sense a complete defen
express warranty claim). See also CAL. COM. CODE ? 2316(3)(c) (West 1964).
147. Suppose that a manufacturer offers a safety device (say, an auto airbag) a
option and that a consumer declines to purchase it because he thinks that th
benefits do not justify its price. It is hard to believe that the consumer could la
facturer on grounds that the absence of the safety device rendered the product
Even Li appreciates that in some assumption of risk situations recovery is prope
the "plaintiff is held to agree to relieve the defendant of an obligation of r
toward him. Such a situation . . involve[s] . . . a reduction of defendant's du
Yellow Cab Co., 13 Cal. 3d 804, 824-25, 532 P.2d 1226, 1240, 119 Cal. Rptr
(quoting from Grey v. Fibreboard Paper Prods. Co., 65 Cal. 2d 240, 245-46, 418
Cal. Rptr. 546, 548 (1966)).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 458 CALIFORNIA LAW REVIEW [Vol. 67:435

ranty claim but a tort claim as well, whether sou

strict liability.
The most prominent affirmative defenses thus
line-drawing among products liability theories. It i
implied warranty uniquely disadvantages produc
them to give the manufacturer notice within a "
alleged warranty violation.148 Justice Traynor's
should not be required from the victim of a trau
sonal injury led him to his Greenman conclusio
theory is necessary.149 Partly because of the no
partly because the notions of "warranty" and "
difficult to explain to the jury, lawyers prosec
while they often state implied warranty counts
plaints-rarely choose to emphasize implied w
presentations.
This accounts for the affirmative defenses available in strict liabil-
ity, negligence, and warranty. What about the more obvious question
of differences in their standards of substantive liability? The core con-
cept of Greenman strict liability is the product's "defect."'5O The core
concept of negligence liability is the manufacturer's faulty conduct.
The core concept of implied warranty liability is the product's "unmer-
chantability,"'51 its "unfitness for the ordinary purposes for which such
goods are used."'52 These concepts obviously overlap and intermingle.
The warranty literature specifies that a "defect" most clearly renders a
product unmerchantable.'53 In like manner, negligence cases have
often indicated that the presence of a "defect" raises the issue of the
manufacturer's negligence.154 And "strict" liability, given its defect re-
quirement, can plausibly be regarded as a "fault" rule after all, one that
simply shifts the focus from the "fault" of the manufacturer to the
"fault" in the product itself.'55

148. CAL. COM. CODE ? 2607(3)(a) (West 1964).

149. 59 Cal. 2d at 61-62, 377 P.2d at 900, 27 Cal. Rptr. at 700. For an argument that the
U.C.C.'s notice requirement is both sensible and modest, see Shanker, Strict Tort Theory of Prod-
ucts Liability and the Uniform Commercial Code. A Commentary on Jurisdictional Eclipses, Pi-
geonholes and Communications Barriers, 17 W. RES. L. REV. 5, 45-46 (1965).
150. "Defect" also is the pivot of the notification and recall provisions in the federal motor
vehicle safety program. See 15 U.S.C. ?? 1411, 1414(a) (1976); United States v. General Motors
Corp. (Wheels), 518 F.2d 420, 427 (1976).
151. CAL. COM. CODE ? 2314(1) (West 1964).
152. Id., ? 2314(2)(c).
153. See W. PROSSER, supra note 85, at 136-37; S. WILLISTON, supra note 85, at 642. See also
Franklin, supra note 143, at 980 (equating products-liability "defect" and warranty "unmerchant-
ability").
154. See, e.g., Escola v. Coca Cola Bottling Co., 24 Cal. 2d 453, 459, 461, 150 P.2d 436, 439,
440 (1944) (majority opinion); Snyder v. Holt Mfg. Co., 134 Cal. 324, 325, 66 P. 311 (1901).
155. See O'Connell, The Interlocking Death and Rebirth of Contract and Tort, 75 MICH. L.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 459

The Barker opinion professes to highlight the diffe

manufacturer negligence and product defect. But if it is
involves a kind of fault, one should inquire when, if eve
of negligence liability and defect liability actually lead t
sults.156 The two theories can first be tested in their appli
ufacturing defects. Typically, such defects are introdu
product by a mishap somewhere in the manufacturing
mishap, in turn, is usually due to either the negligence
of the manufacturer or the negligence of the manufact
Two vicarious liability rules-one quite ancient57 and
emerging'58 -render the manufacturer liable'59 in either
Negligence law, however, imposes on the plaintiff-vi
den of proving his case. Since the circumstances of the
hap are commonly unknown even to the manufacturer
quite beyond the plaintiffs ability to discover, the latter h
but to fall back upon res ipsa loquitur. In an early arti
ser claimed that if a victim can demonstrate a manufacturi
can secure a res ipsa negligence verdict from the jury
cases.161 Prosser subsequently backed away from this str
contented himself with the softer allegation that a negli

REV. 659, 664 n.3 (1977). The Barker opinion concedes as much: the manufactu
there is "something wrong" with its product. 20 Cal. 3d at 433, 573 P.2d at 456
238. Under Barker, the plaintiff "is no longer required to impugn the maker, b
impugn the product." 20 Cal. 3d at 436, 573 P.2d at 457, 143 Cal. Rptr. at 23
supra note 39, at 33).
156. In light of warranty law's notice requirement and trial lawyers' percep
sion concerning basic warranty concepts, the comparison here will downplay
centrate on strict liability versus negligence.
157. This is the taken-for-granted rule of respondeat superior.
158. On the vicarious liability of the manufacturer for the negligence of its
dermark v. Ford Motor Co., 61 Cal. 2d 256, 261, 391 P.2d 168, 170, 37 Cal. R
(dictum).
159. There are other possibilities. The negligence may be committed by a retailer to whom
the manufacturer delegates tasks of final assembly. If so, a vicarious liability rule probably runs
against the manufacturer. See id. (by implication). The mishap may be traceable to the malfunc-
tion of some machine on the manufacturer's assembly line. If so, one can possibly speak of the
manufacturer's vicarious liability for the torts of its machines. If this sounds anthropomorphic,
consider the common law rule rendering the farmer liable for the "trespasses" of his cattle. See
Williams v. Goodwin Dev. Corp., 41 Cal. App. 3d 496, 116 Cal. Rptr. 200 (3d Dist. 1974).
160. Similar problems arise in ascertaining the legally responsible party when medical sur-
gery produces a clearly negligent result. The supreme court dealt with this problem in Ybarra v.
Spangard, 25 Cal. 2d 486, 154 P.2d 687 (1944), by creating an essentially presumptive rule of
group negligence liability within the surgical team. The court's Ybarra opinion (in which Justice
Traynor concurred) was released only a few months after Escola, which had included Justice
Traynor's strict liability concurrence. Ybarra is explained in Clark v. Gibbons, 66 Cal. 2d 399,
411, 426 P.2d 525, 533, 58 Cal. Rptr. 125, 133 (1967).
161. Prosser, The Assault Upon the Citadel (Strict Liability to the Consumer), 69 YALE L.J.
1099, 1114-15 (1960).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 460 CALIFORNIA LAW REVIEW [Vol. 67:435

who identifies a manufacturing defect will "usu

this latter assessment is faithful to the chancin
ment. Even when the law allows resort to res
nia's "conditional" res ipsa standards'64 there is
will decline to accept the res ipsa invitation.'65
perceived as "a simple, understandable rule
dence,"'66 lawyers agree that many juries either
res ipsa theory or regard it as not entirely fair-pe
pathizing with the claim that "[s]urmise ought
strict proof when it is sought to fix a defend
ity."167

In manufacturing defect cases, therefore, the basic consequence of

the products liability rule is clear enough. The rule recognizes that
manufacturers can be held vicariously liable for any negligence in the
manufacturing process, that these defects are almost always a result of
negligence somewhere within that process, but that res ipsa can be a
cumbersome method of proving negligence in the individual case. By
imposing automatic liability on the manufacturer, the rule correctly
concludes that whatever marginal fine-tuning a mature negligence sys-
tem makes possible would not justify the corresponding complication
of the trial process, a complication that would disserve product victims
by delaying, rendering more costly, and sometimes preventing the vin-
dication of their valid negligence claims. Products liability thus serves

162. W. PROSSER, HANDBOOK OF THE LAW OF TORTS 672 (4th ed. 1971).
163. The majority opinion in Escola declined to overrule Honea v. City Dairy, Inc., 22 Cal.
2d 614, 140 P.2d 369 (1943), where the supreme court (Justice Traynor concurring) held that res
ipsa was not available in a suit brought against a milk bottler for an injury caused by an exploding
milk bottle. The distinction between carbonated and noncarbonated beverages is notable. Given
Vandermark's subsequent announcement of an assembler's vicarious liability for "upstream" neg-
ligence, see note 158 supra, the res ipsa argument in products cases like Honea (and Escola itself)
becomes much easier.

164. See CAL. EVID. CODE ? 646 (West Supp. 1978) and the accompanying Law Revis
Commission Comment. Under "conditional" res ipsa, the trial judge merely determines whe
reasonable minds could differ as to the appropriateness of res ipsa in the particular case. The
then makes up its own mind on whether res ipsa is appropriate.
165. One defense counsel has advised me that in pre- Greenman days he was consistently ab
to secure defense verdicts in those cases in which plaintiffs counsel contented himself with a
ipsa argument. And an experienced plaintiffs' lawyer reports that the lesson of his experience
that the doctrine should be renamed "res ipsa loser." Statement of Leonard Mandel, moderator
the California Continuing Education of the Bar Session on Products Liability: Pleading, Prac
and Proof, Los Angeles, September 16, 1978. But see Whitford, Strict Products Liability an
Automobile Industry.: Much Ado About Nothing, 1968 Wis. L. REV. 83, 108-09.
166. Ybarra v. Spangard, 25 Cal. 2d 486, 489, 154 P.2d 687, 689 (1944). Of course, on
multiparty facts, the res ipsa issue in Ybarra itself was anything but simple.
167. Argument of counsel in Byrne v. Boadle, 2 H. & C. 722, 724, 159 Eng. Rep. 299,
(Exch. Ch. 1863).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 461

the highly intelligent function of a "negligence shortcut."'6

these defect cases, however, negligence principles justify
insofar as they are extended by certain rules of vicarious li
should not be forgotten that vicarious liability embodies
of a limited but important sort.169
In a few remaining cases that get somewhat awkwar
under the heading of "manufacturing defect," the pro
specific product (usually a food or medical product) is or
utable not to any mishap but rather to a flaw or impuri
source which is an ingredient in the production process
of product defect, no inference of negligence immediately
ipsa or otherwise. Certain negligence claims may re
however, such as negligence in the manufacturer's choic
source,v70 negligence in the failure to separate out the
processing, and negligence in the failure to detect the im
inspection and testing.v'7 Even if the facts of the indi
some reason do not support a negligence finding, it rem
the manufacturer usually is in a good position to achieve
in the near future by identifying more sophisticated m
source selection or developing more imaginative techniq
tion and testing.
If strict liability applies in these impurity cases, its p
be to "buy off" numbers of uncertain or difficult-to-pr
claims and to give the manufacturer a strong short-run inc
novate in inspection, testing, and resource selection. Th
would resemble the incentives characteristic of genuine
but with the important difference that the incentive wo
at very specific product risks.172 Interesting issues ar

168. See Franklin, Tort Liability for Hepatitis: An Analysis and a Proposal
439, 461 (1972). See also Prosser, supra note 58, at 799-800.
169. The equivalence of strict "defect" liability and an advanced negligen
gested by Ford Motor Co. v. Mathis, 322 F.2d 267 (5th Cir. 1963) (applying
that a manufacturer is vicariously liable for the negligence of its supplier
"sound and realistic" conclusion, id. at 273, the court relied on a "public po
Justice Traynor's Escola concurring opinion, id. at 143 n.10, an opinion whic
cates strict liability.
The entire subject of vicarious liability has been greatly underdiscussed in
therefore lack a firm basis for evaluating the Mathis holding in its negligence
170. See Franklin, supra note 168, at 439, 444-45, 447-49.
171. Negligence law imposes a strong inspection requirement on m
Sheward v. Virtue, 20 Cal. 2d 410, 126 P.2d 345 (1942); Edison v. Lewis Mfg
2d 429, 336 P.2d 286 (2d Dist. 1959); MacPherson v. Buick Motor Co., 217
1050 (1916).
172. The safety incentives afforded by a genuine strict liability rule (see notes 60-64 and ac-
companying text supra) are of an inexhaustible and hence long-run character. Since in the impu-
rity cases there is a particular, limited safety problem, the assumption that the problem can be
dealt with by short-term innovations, while not valid in all cases, seems generally reasonable.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
462 CALIFORNIA LAW REVIEW [Vol. 67:435

stake here. As for the law there is a correspond

plication to report: in cases involving "natural"
able impurities, products liability has repeated
leap. From the inception of implied warranty i
temporary problems of hepatitis-infected blood
tures have been remarkably hesitant to imp
burden."' It is frankly uncertain, therefore, wh
of strict liability actually intends to step mean
gence in order to give manufacturers short-run sa
Given this uncertainty, all that can confiden
manufacturing defect cases is that strict liabili
into a sophisticated negligence system, one th
vicarious liability (and hence strict liability) ele
cases, by comparison, it is far from clear that s
even this limited function. Technically speakin
gence action should be whether the manufactur
a design decision was unreasonable. But in pract
cases the focus tends to shift from the unreaso

173. See Mix v. Ingersoll Candy Co., 6 Cal. 2d 674, 59 P.2d 144
ral" chicken bone in chicken pie). A recent decision in a similar ve
Inc., 30 N.C. App. 134, 226 S.E.2d 534 (1976). See generally Pros
174. See McDonald v. Sacramento Med. Found. Blood Bank,
Rptr. 444 (3d Dist. 1976) (declaring the obvious meaning of CAL. H
(West 1970)); Brody v. Overlook Hosp., 127 N.J. Super. 331, 317 A
66 N.J. 448, 332 A.2d 596 (1975). With the Illinois legislature ha
MacNeal Memorial Hosp., 47 Ill. 2d 443, 266 N.E.2d 897 (1970
?? 181-183 (Supp. 1978), it is unclear which jurisdictions, if any, p
for the supply of infected blood.
175. Under Barker, the question is whether the deviation-from-
for identifying liability-producing manufacturing defects is to
qualification. The Barker dictum setting forth that standard do
exceptions. 20 Cal. 3d at 429, 430, 573 P.2d at 453, 454, 143 C
Restatement's endorsement of an exception, see RESTATEMENT
Comment k (1965). The Department of Commerce's Draft Unifo
cludes a similar exception. Draft Uniform Product Liability Law
(1979) [hereinafter cited as Uniform Law].
176. The carrier test described by Professor Franklin, supra n
more sophisticated since his 1972 article. Still, the test is only able
hepatitis B infection. Hepatitis B has always been less common t
beyond the reach of existing testing and which has become, if any
years. Interview with Dr. Gary Gitnick, UCLA School of Medic
(Nov. 15, 1978). The overall blood hepatitis problem therefore is ab
been. How much more progress would have been made had court
to provide strict-liability incentives is obviously an interesting, im
The matter is complicated by the division in function between blo
that medical research is largely conducted by pharmaceutical house
extent to which blood transactions operate outside of ordinary pri

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 463

decision to the unreasonableness of the design itself."'77 And

reasonableness as a negligence notion is measured in risk-benef
the negligence issue in design cases assumes a form almost ide
the defect issue under strict liability-especially given Barke
benefit defect prong. Moreover, negligence law's willingness to b
distinction between conduct and product makes sense. Since d
cisions are made deliberately by manufacturers' design officia
reasonable decision almost necessarily leads to an unreasonable
and conversely an unreasonable design is almost always th
quence of an unreasonable design decision. Perceiving this, c
some jurisdictions have found it meaningless to apply strict li
design cases.178
In a limited number of design situations, however, the equ
of the negligence and strict liability theories can break down.
theories remain equivalent so long as the relevant risks and
either are known by the manufacturer's design officials or could
ably be known by them. What if those officials are unaware o
vant risk or benefit at the time they reach their design decision,
their lack of awareness is reasonable and nonnegligent? To t
treme examples, what if the risk that the product contains is
known and unknowable at the time the product is designed
Or what if a safety device capable of eliminating or reducing
uct's risk at acceptable cost is not invented until after the tim
product's design and sale? Clearly, negligence principles would
ford a recovery in either of these situations. To the extent th
eries are authorized by strict liability, that doctrine doe
difference. Unfortunately, however, the results that strict liabil
ally reaches in cases of this sort remain somewhat uncertain
these cases lie on the outskirts of the Greenman-Barker doctr
sideration of them helps clarify that doctrine's boundaries.179 A

177. See Noel, supra note 21, at 818, 824, 825, 834, 836, 844. The Noel article is
useful because it was written at just about the dawn of strict liability. See id. at 87
178. "[A]s to a defect in design [strict liability] has no special meaning. . . . [I]t
illogical to speak of a defective design even though the manufacturer has 'exercised
care' in the preparation of his product." Volkswagen of America, Inc. v. Young, 27
220-21, 321 A.2d 737, 747 (1974).
Pike v. Frank G. Hough Co., 2 Cal. 3d 465, 467 P.2d 229, 85 Cal. Rptr. 629 (1970
first California case to clearly apply strict liability to matters of product design. But th
was also willing to give negligence doctrine a rather full design application; and the co
gence reasoning seems almost interchangeable with its strict liability reasoning.
179. While as boundary-setters these cases have considerable analytical importance
remains that they are factually unrelated to the great bulk of product design claims.
gard, see G. FLETCHER, RETHINKING CRIMINAL LAW 143 (1978), warning against a
tion with borderline cases."

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 464 CALIFORNIA LAW REVIEW [Vol. 67:435

sion of the issues they raise will be presented in th

Foreword.

B. The Definition of Design Defect: Barker's Two Prongs

1. The Tort-Based Risk-Benefit Prong

There can be little doubt about the correctness of the risk-benefit
standard for design defect set forth in Barker.'8" That standard is a
natural outgrowth of the basic negligence principle, and can find justifi-
cation in the fairness and accident prevention ideas which support that
principle. Cronin's terseness had encouraged doubts at the trial level as
to whether a "balancing" approach was permissible once an apparently
safer design alternative had been identified. By clearly legitimating the
jury's full consideration of "trade-off" evidence, Barker clarifies the
law in a welcome manner."' Barker's holding on burden of proof is
troublesome, however. That holding specifies that once the plaintiff
shows that the product's design is the proximate cause of his injury, the
manufacturer can escape liability only by verifying that the design is
indeed risk-beneficial.'82

This holding can be seen as responsive to a certain practical prob-

lem. Prior to Barker, lawyers knew that proving a design defect was an
expensive and uncertain endeavor.183 As a result, the victim with an
injury "priced" at less than $25,000 often encountered difficulties in
finding a high-quality lawyer to take his case, even when the facts on
the liability issue were reasonably good. It has been suggested that the
implicit purpose of the Barker burden-of-proof rule is to enable com-
petent lawyers to "small budget" such a case and thereby induce them

180. Under that standard, it is clear that the manufacturer's pre-sale knowledge of some haz-
ard in the product's design by no means suffices to establish liability. Rather, the standard gives
the fact-finder the opportunity to impose liability if its risk-benefit calculation reaches an opposite
result from any favorable risk-benefit calculation the manufacturer may have originally made.
181. Barker's hostility to the language of "reasonableness" may seem surprising. In tort law,
"reasonableness" is well understood as calling for a risk-benefit balancing. See RESTATEMENT
(SECOND) OF TORTS ?? 291-293 (1965). The word carries essentially the same meaning in the law
of public product regulation. See 15 U.S.C. ? 2056(a) (1976); Aqua Slide 'N' Dive Corp. v. Con-
sumer Prod. Safety Comm'n, 569 F.2d 831 (5th Cir. 1978).
182. This burden-of-proof idea had not been mentioned (so far as I know) in any of the law
reviews. In the briefs before the court in Barker, it was no more than hinted at by one amicus.
See Brief for Amici Curiae (Siegfried Hesse & Peter Elkind), at 7; Barker v. Lull Eng'r Co., 20
Cal. 3d 413, 573 P.2d 443, 143 Cal. Rptr. 225 (1978).
183. By now there is available to plaintiffs a substantial supply of expert witnesses who are
competent to consult and testify on a wide range of product design issues. These experts charge
their fees, however. Moreover, manufacturers have a comparative advantage. While a single tire
company can retain an expert witness almost full-time, even a leading plaintiffs' lawyer is likely to
need a tire expert only once or twice a year. Employment is thus more secure on the defense side.
For the battle of expert testimony in Barker itself, see 20 Cal. 3d at 420-22, 573 P.2d at 447-49, 143
Cal. Rptr. at 229-31.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 465

to provide representation.184 Alternatively, the reduction in the

of proof can be seen as an attempt to relieve the victim of th
retain a lawyer who is already a products specialist.
Proving a product case could be rendered somewhat less d
were products law merely to recognize techniques of proof d
by negligence law during the 129 years of its California lega
Assume that a new car's brakes fail ten miles away from the show
thereby causing an injury. These facts clearly bespeak the exi
some defect in the brakes at the time the car was sold. While the
supreme court has failed to realize this,'85 it seems obvious that the
plaintiff should be allowed to prove defect by some strict-liability ver
sion of the negligence-oriented doctrine of res ipsa loquitur.'86 Now
assume that a product is lacking a safety device common in similar
products on the market. The one product's failure to comply with in
dustry custom should be deemed strong evidence of its defectiveness
Moreover, an industry custom analysis can probably address itself t
the performance of a product, as induced by its design, as well as to the
actual composition of that design: if the product's design-induced per
formance is plainly less safe than the performance of comparable prod
ucts, the burden of explanation should be imposed on the
manufacturer.187

Even if accepted in the aggregate, however, these case-proving

techniques would still leave many worthy products plaintiffs--espe-
cially those whose claims are for small amounts-with a substantia
burden. This burden may well indicate some problem for the tort sys
tem. But both the extent of this problem'88 and its actual ramifica-

184. Telephone interview with Ned Good, Member of the California State Bar (Nov. 10
1978).
185. See, e.g., Jiminez v. Sears, Roebuck & Co., 4 Cal. 3d 379, 482 P.2d 681, 93 Cal. Rptr. 769
(1971), in which the court allowed a plaintiff to plead negligence in addition to strict liability on
the idea that the negligence theory enables the plaintiff to rely on res ipsa.
186. See Henningsen v. Bloomfield Motors, Inc., 32 N.J. 358, 409-10, 161 A.2d 69, 97-98
(1960). In such a case, the plaintiff should not even be required to "declare" whether the defect is
one of manufacture or of design-a requirement that Barker seemingly imposes.
187. This variation in the custom theme is suggested by Culpepper v. Volkswagen of
America, Inc., 33 Cal. App. 3d 510, 109 Cal. Rptr. 110 (4th Dist. 1973). In Culpepper, a VW bug
rolled over when the driver turned its front wheel sharply (18 degrees) while traveling at 55 miles
per hour. The court of appeal affirmed a verdict for plaintiff mainly on the basis of expert testi-
mony that for cars driven in America, there is an "implied standard" not to roll over when being
driven in this fashion. This evidence as to performance subordinated what little evidence the
plaintiff presented concerning any of the VW's particular design features. Culpepper's direct em-
phasis on product performance is consistent with products liability's warranty basis, since the
warranty concept of "fitness for use" has a strong performance orientation. Culpepper is com-
mended in Barker, 20 Cal. 3d at 430, 573 P.2d at 454, 143 Cal. Rptr. at 236.
188. In the area of auto accident claims, minor injury claimants tend to be overcompensated
in comparison to their out-of-pocket losses, while serious injury claimants are undercompensated.
See NEW YORK INS. DEP'T, AUTOMOBILE INSURANCE ... FOR WHOSE BENEFIT? 27 (1970). This

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 466 CALIFORNIA LAW REVIEW [Vol. 67:435

tions189 require further study. In any even

problem, it does not follow that the Barker ru
gently adapted solution.
A preliminary observation is that the rule p
potential application. It is hard to conceive of a
cident which would not permit some claim that
by something in the product's design.190 Until
very paradigm of a product that is dangerous
knife that cuts.191 Yet under Barker, when a
burden evidently shifts to the manufacturer to
design is non-defective, since plainly a design f
sharp blade-has been a proximate cause of th
now it is known with near certainty that there
sign of a cigarette which is carcinogenic. There
plaintiff in a cigarette cancer case can establish
burden of proof evidently shifts to the cigarette m
that the design of the cigarette is not defective
typical automobile accident, one in which a car
miles per hour when the driver sees an obstacle
to stop in time, the car hits the obstacle and the
the windshield, suffering serious injury. These
mately caused by the product's design inasmuch
to operate at speeds as fast as 75 miles per hour, th
capable of stopping the car within 275 feet,'94 the
produces serious injuries upon impact, and so o
That some of these assessments may seem con
points to another problem with the Barker rul
enormous burden on the concept of a "product d

same pattern has been found in products cases, despite the high
entail. See INSURANCE SERVICES OFFICE, PRODUCT LIABILITY CL
NICAL ANALYSIS OF SURVEY RESULTS: HIGHLIGHTS 4 (1977). Of
does not include all "victims."

189. Because the threshold of proof is so low in auto claims-all it takes is the testimon
the victim that the motorist was speeding-insurance companies, to avoid uneconomic litig
are required to settle small "nuisance" claims for excessive amounts. A higher threshold t
not without advantage.
190. An Arizona court has rejected the Barker burden-of-proof rule for what amounts to
reason. Brady v. Melody Homes Mfr., [1978] PROD. LIAB. REP. (CCH) ? 8303 (Ariz. App. 1
191. See Traynor, supra note 32, at 367.
192. Professor Epstein agrees with this assessment of how the Barker rule works. Ep
supra note 97, at 651.
193. Melvin Belli is presently litigating a cigarette cancer case in San Francisco, to the dis
of columnist George Will. Will, Danger. A Smoky Assault on an Industry, L.A. Times, Sep
1978, ? 2, at 7, col. 4.
194. Does it matter whether this braking distance is above or below the average for o
cars? See note 187 and accompanying text supra.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 467

causes injury," a burden which the concept seems ill-eq

dle."'9 Is it or isn't it correct that the design of the car's e
or windshield proximately causes the victim's injuries?
ladders all the time, and the fact that ladders are both hig
general way unstable enables these falls to occur. Does
follow that in every case of a person's falling off a ladd
design proximately causes the fall? Also, when the desig
is satisfactory as far as it goes but is lacking a possibl
device, in what sense does the design itself proximately ca
While the Barker opinion steers clear of the word "p
both the court's articulation of its rule and the court's reference to sec-
tion 605 of the Evidence Code'97 verify that the rule does amount to a
presumption. That is, proof that the product's design proximately
caused the victim's injury creates a rebuttable presumption that the de-
sign is defective in the risk-benefit sense. So characterized, the Barker
rule sounds a bit like res ipsa loquitur-a rule which, thanks to Escola,
can be invoked in most manufacturing defect cases argued under a neg-
ligence theory. In such a case, res ipsa is understandable enough; if a
properly handled cola bottle explodes, it is reasonable to assume some
negligence on the bottler's part. By contrast, the Barker rule lacks this
reasonable, common sense quality.'98
To be sure, Barker allows the manufacturer to rebut the presump-
tion of defect.'99 A further Barker-related confusion concerns how this

195. The Oregon Supreme Court reads Barker as holding that the plaintiff can go to the jury
in a design case so long as he can show that the "product caused the injury." Wilson v. Piper
Aircraft Corp., 282 Or. 411, 579 P.2d 1287 (1978). This seems to misstate the Barker rule-but
that rule is elusive enough to encourage misstatement.
196. In Barker itself, four of the plaintiffs six design defect claims related to design omis-
sions. These six claims are listed below, the omission claims first.
1. The high-lift loader was without "outriggers"-mechanical arms extending away from
the machine-which would have provided the loader with greater stability.
2. The loader lacked seat belts or a roll bar to protect the operator in the event of a rollover.
3. The loader's transmission lacked a "park" position. (Had the loader been in "park," it
would have been more stable.)
4. The loader's "leveling" mechanism lacked an automatic lock. (The mechanism, for
whatever reason, was not "on" at the time of the accident.)
5. The leveling mechanism was so located as to make it vulnerable to inadvertent bumpings
turning it "off."
6. The loader's wheel base was too narrow and hence unstable.
20 Cal. 3d at 420-21, 573 P.2d at 447-48, 143 Cal. Rptr. at 229-30.
197. CAL. EVID. CODE ? 605 (West 1966). See 20 Cal. 3d at 432, 573 P.2d at 453, 143 Cal.
Rptr. at 237.
198. It will therefore bedevil the lower courts. In Garcia v. Joseph Vince Co., 84 Cal. App. 3d
868, 148 Cal. Rptr. 843 (2d Dist. 1978), a too-sharp-edged sabre (manufacturer unknown) wielded
by one fencer pierced the mask of another fencer, causing injury. The latter sued the manufac-
turer of the mask. The court of appeal's opinion-affirming a nonsuit--makes clear that the court
has little idea how the Barker rule works.
199. Under Barker, even when the manufacturer effects a rebuttal, the burden of persuasion

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 468 CALIFORNIA LAW REVIEW [Vol. 67:435

rebuttal process works. In the knife case, for e

knife manufacturers can easily show, relying on
jury's pooled experience, that a knife needs a sh
function as a knife. But how can one preclude t
cagelike safety device for the knife that would m
inevitable risk? If an auto accident ignites a gas
true that the location of the gas tank is a design f
has proximately caused the injury.2" To rebut t
of defect, must the car manufacturer rule out, wit
every other possible gas tank location?201 Assu
that the circumstances of the accident succeed in s
proof, how does the manufacturer discharge th
ticipate (and refute) all six of the defect allegatio
sel-unaware of the eventual Barker rule-pres
trial?202

The heart of the problem is this: one simply

fully about a risk-benefit defect in a product d
one has identified some design alternative (inclu
sion) that can serve as the basis for a risk-be
Barker rule is read literally, however, it fails to
even to point to an alternative of this sort. Within
jurisprudence, one respected canon is that the b
on the party who has control of or access to the re
this is the canon upon which Barker properly r
spectable canon is that the burden of proof shou
to require a party to prove a negative.2'" This c
violates.205

remains with the manufacturer. 20 Cal. 2d at 432-33, 573 P.2d a

doubt that this makes much of a difference in practical terms. A
evidence is probably rare; and the plaintiffs lawyer who argues to
win in the event of such a tie would be unstrategically confessin
generally Comment, The California Evidence Code.- Presumptions, 5
57 (1965).
200. See Self v. General Motors Corp., 42 Cal. App. 3d 1, 116 C
201. In general, evidence rebutting a ? 605 presumption does no
for defendant unless it is so effective that it "compels belief." See
1453. With respect to the Barker rule as stated, the directed verdic
question of "belief in what?"
202. See note 196 supra.
203. See Cleary, Presuming and Pleading.: An Essay on Juristic Im
5, 12 (1959).
204. See id. at 11. See also J. MAGUIRE, EVIDENCE: COMMON SE
(1947).
205. In particular cases this violation may be harmless enough. In Korli v. Ford Motor Co.,
84 Cal. App. 3d 895, 149 Cal. Rptr. 98 (2d Dist. 1978), the rear door of a 1965 Lincoln Continental
was "rear-hinged" rather than "front-hinged." Rear-hinging produces a certain risk (of the door
flying open if opened slightly while the car is moving rapidly), but it also provides a certain benefit

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 469

Yet while the Barker burden-of-proof rule thus see

difficulties, in practice the rule may not turn out to be
tant. The analysis above has impliedly assumed that plain
to exercise their Barker rights. However, my discussions w
attorneys make clear that in a typical case, a plaintiff w
tory facts will choose to ignore this aspect of Barker. Bark
plaintiff, having shown that the product's design caused
close his case and wait for the manufacturer to explain t
that design is sensible after all; once the manufacturer has
pro-product evidence, the plaintiff can then bring forward
dence in an effort to counter the defense's presentation
entail very poor trial strategy for a plaintiff to rely on
since it puts him on the defensive by allowing the manu
to the jury first with its explanation of the product's d
that design is a good one. If his facts are good, the plain
tactical interest in getting the jury to consider his version
issue first, before the manufacturer has a chance to tell
story. Wishing to convey to the jury the strength of his c
tiff will want to come out "with all guns blazing," presenti
est evidence as to the impropriety of the product's desig
Therefore, as sweeping as the Barker burden-of-pro
be, it may create a right which most products plaintiffs
exercise. Indeed, the pertinent question turns out to be
limited class of products cases will the plaintiff choose t
of the Barker rule? Here attention can return to the victim
torious, but small, damage claim. Will the Barker rule m
for this victim to find a competent lawyer able, by relying
to small-budget his case? Perhaps this would be the res
doubtful. If the facts on liability are favorable enough
needing to win the case in order to secure his contingent fe
incentive to make his strongest pitch to the jury; and t
well involve the waiver of his Barker rights.206

(easy ingress and egress). Given what we all know about automobiles, there
design alternative-front-hinging-with which the car's rear-hinged design c
The Korli court, which had refused to allow the plaintiff to get to the jury
Cal. App. 3d 115, 137 Cal. Rptr. 828 (2d Dist. 1978) (official reports advance she
fled opinion), persisted in this refusal after Barker, on the plainly anti-Barker
ucts liability does not allow "lay juries to simply make value judgments
desirability of one design over another." 84 Cal. App. 3d at 906, 149 Cal. Rpt
also reasoned that only the victim's opening of the car door, and not the door'
the proximate cause of the accident. This reasoning highlights the uncertaintie
mate cause concept.
206. Likewise, even under Barker the victim still must retain a lawyer wh
about products, given the victim's eventual need to cope with the manufacture
off' evidence.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 470 CALIFORNIA LAW REVIEW [Vol. 67:435

Rather, the class of cases in which the Barker b

seems to offer the clearest practical help are th
has suffered severe injuries, but in which the f
sign defect are very doubtful or weak. Given the
claim, without Barker the plaintiff might well
judgment or a directed verdict. Under Barker,
den of proof has been shifted by the plaintiffs lim
a few cases will the manufacturer be able to offer t
able evidence that overcomes the defect presum
law.207 In all remaining cases, the issue of defe
the jury, thin as the liability facts may be.
What are the implications of consigning suc
Given the possibility of jury sympathy with a s
the jury may sometimes cut corners on the evid
the victim a recovery. These cases are, by hypot
But it is hardly clear that such results would p
products defendants win at least half of all jur
sometimes resist the strict liability doctrine, not
or doubting its fairness.209 The Barker burden-
its natural logic, and it dramatizes the strict lia
queer manner. As the rule is fed to the jury, th
expected that are confused and erratic.
In sum, the Barker burden-of-proof rule eith
or its coverage is anyone's guess. Once Barker
of proof, what happens next is something of a
selective assertion of the Barker rule is likely to
results. Given these assessments, the Barker rul
is therefore led to consider a return to what m
professors' pre-Barker understanding: that the pla
burden of showing a risk-beneficial design alter
pens, this understanding does not accurately ref
tell me was sometimes happening at the trial lev
Barker. The trial process frequently seems to h
the plaintiff introduced evidence establishing t
ble212 design alternative that would have preven

207. See note 201 supra.

208. See text at note 99 supra.
209. See text at note 100 supra.
210. It has already been rejected as "not . . . persuasive" by
Wilson v. Piper Aircraft Corp., 282 Or. 411, 579 P.2d 1287 (1978)
ing).
211. This is the apparent holding of Wilson v. Piper Aircraft Corp., 282 Or. 61, 577 P.2d 1322
(1978).
212. "Feasible" here means nothing more than "it can be done."

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 471

point, whatever the law said, some juries expected the m

introduce whatever evidence it possessed showing that
design proposal would not have been a good idea-that i
been too costly, would have unduly interfered with the
formance, would have created additional safety hazards,
The jury's expectation in this regard obviously grew out of
perception of the sound access-to-information burden-o
rion. Yet since the resulting practice also spared the man
obligation of proving a negative, it equally complied w
valid criterion. Also, it avoided any overburdening of
cause concept. So long as judges are willing to enter dir
when the manufacturer's trade-off evidence is strong enou
Cronin, pre-Barker practice seems an intelligent, balanc
the burden-of-proof problem,213 one which reconciles t
Barker places in conflict. Given the intelligence of thi
Barker rule should be either modified or "interpreted"2
for its legalization. That is, once the plaintiff identifies a f
alternative which would have prevented his injury, the
should rest on the manufacturer to show why that alt
not have been risk-beneficial.

2. The Contract-Based Consumer Expectations Prong

Barker is right on one syllogistic point. Since Cronin merely he
that the failure to satisfy consumer expectations is not a necessary con-
dition for liability, the Cronin holding is not logically inconsistent with
Barker's indication that the failure to satisfy these expectations is
sufficient liability criterion. Its limited holding apart, however, t
Cronin opinion seemed almost to disdain the consumer expectation
idea.215 The revival of consumer expectations in Barker is therefo
something of a surprise. And it could be seen as a welcome surpris
since the consumer expectations standard attempts both to take adva
tage of strict liability's legitimate "warranty heritage"216 and to ta
account of limited consumer knowledge of particular produ
hazards.217

213. This assessment is in line with the Fifth Circuit's reasoning in Mitchell v. Frueh
Corp., 568 F.2d 1139, 1143-45 (5th Cir. 1978) (applying Texas law).
214. Defense lawyers are presently developing the argument that in order to show that
"product design has proximately caused" the injury, the plaintiff must prove a design alternati
which, had it been used, would have prevented the plaintif's injury. I regard this an improba
definition of the Barker concept, but insofar as it leads to the right result I am in sympathy wi
the lawyers' efforts.
215. See Cronin v. J.B.E. Olson Corp., 8 Cal. 3d 121, 132-33, 501 P.2d 1153, 1161-62, 104 C
Rptr. 433, 441-42 (1972).
216. 20 Cal. 3d at 429-30, 573 P.2d at 454, 143 Cal. Rptr. at 236.
217. See notes 118-24 and accompanying text supra.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 472 CALIFORNIA LAW REVIEW [Vol. 67:435

Examined more closely, the prong is effectiv

support for already recognized products liability
independent basis for products liability verdicts,
by doubts. These doubts can be explained by conside
nition of "consumer," and second, the nature of
nary expectations."

a. Who Is a "Consumer"?

Under the law of warranty-the wisdom of which the pr

tended to carry forward-the "consumer" is, of course, the
chaser of the product.218 Likewise, under the Restate
"consumer" whose "expectations" are legally relevant is the
purchaser, as section 402A's explicit language makes clear.2
California and in other jurisdictions, the strict products lia
has been interpreted as allowing suit by almost any victim of a
related accident, including employees who are injured by
purchased by their employers,220 passengers and pedestrians
accidents involving automobiles driven by car owners,221 an
injured by products purchased by their schools.222
How does Barker's consumer expectations prong work
brought by nonpurchasers like employees, passengers, stud
bystanders? Taking that prong in conjunction with the case's
Barker opinion can be understood as indicating that the in

A jury considers the consumer expectations question subsequent to the accident

product has performed in a certain way. Perhaps a jury will find "consumer" expect
if it-the jury-feels that a properly designed product would have performed bett
the consumer expectations prong amounts to the risk-benefit prong looked at f
loquitur point of view. Res ipsa can claim a significant role in the proof of prod
notes 185-86 and accompanying text supra. But to suggest that a res ipsa analysis is
every product design case would be plainly wrong. And even if res ipsa "applies
benefit issue, the manufacturer is entitled to rebut with trade-off evidence. Yet un
consumer expectations prong, the manufacturer has no such opportunity-a fin
sumer expectations have been denied is conclusive of liability. Given the finality un
a negative consumer expectation finding, and given that Barker expressly advances
expectations standard as an alternative to the risk-benefit standard, it is clear that
sumer expectations test cannot be regarded as merely a res ipsa approach to risk-be
218. See W. WARREN, W. HOGAN, & R. JORDAN, supra note 85, at 3, 76-79. Th
implied warranty law imposes a requirement of horizontal privity (with narrow exce
which covers members of the buyer's family). Hauter v. Zogarts, 14 Cal. 3d 104, 114
377, 383 n.8, 120 Cal. Rptr. 681, 687 n.8 (1951). The horizontal privity issue concerns
in addition to the product's actual buyer. See Morrow v. New Moon Homes, Inc.,
287 (Alas. 1976).
219. RESTATEMENT (SECOND) OF TORTS ? 402A, Comment i (1965).
220. See Barker itself (although there the product was leased to the employer
sold).
221. See Elmore v. American Motors Corp., 70 Cal. 2d 578, 451 P.2d 84, 75 Cal. Rptr. 652
(1969).
222. See Dudley Sports Co. v. Schmitt, 151 Ind. 217, 279 N.E.2d 266 (1972).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 473

ployee is a "consumer" whose expectations matter. Like

the course of discussing consumer expectations the B
drops an approving reference to the court's earlier "bys
in Elmore v. American Motors Corp.,223 the opinion per
ence that even the bystander-and a fortiori the passeng
dent-are legally relevant "consumers" as well.
If this is Barker's intended meaning, Barker seem
Strict liability's tort dimension categorically justifie
against manufacturers by these various product-acciden
in their suits these victims should be free to rely on th
risk-benefit liability standard. But Barker fails to explai
should attach decisive significance224 to these persons'
product expectations. The whole purpose of warranty's
consumer expectations is to make sure that the product
chaser acquires is of the quality he reasonably assumes i
purchaser should get the full benefit of an "honest" sales t
When the consumer-expectations prong is extended bey
to employees, passengers, students, and bystanders, it ex
dons this basic rationale. Assume that an employer b
machine knowing that it is lacking in a possibly effective b
safety device; that a person buying an automobile choo
sharp hood ornament potentially dangerous to pedestrian
high school buys unpadded football helmets because its
likes the solid "whack" that unpadded helmets prov
tact.228 Persons victimized by these products should d
lowed to proceed against the manufacturer under the to
benefit standard.229 But to allow them to sue the manufacturer on

223. 70 Cal. 2d 578, 451 P.2d 84, 75 Cal. Rptr. 652 (1969).
224. See note 217 supra.
225. See W. WARREN, W. HOGAN, & R. JORDAN, supra note 85, at 26.
226. See Professor Twerski's discussion of an "honest product." Twerski, From Defect to
Cause to Comparative Fault--Rethinking Some Product Liability Concepts, 60 MARQ. L. REV. 297,
314 (1977). The Twerski article, not cited by Barker, anticipates Barkers two-prong holding.
227. See Knippen v. Ford Motor Co., 546 F.2d 993 (D.C. Cir. 1976).
228. See Underwood, An Unfolding Tragedy, SPORTS ILLUSTRATED, Aug. 14, 1978, at 69, 73,
78, 80: "Coaches want to hear noise. They love noise. Equipment manufacturers know that
coaches, not physicians, buy helmets."
229. For that matter, the pedestrian and the student have possible negligence claims against
the auto purchaser and the school for their choice of products. See, e.g., Gerrity v. Beatty, 7 Ill. 2d
47, 373 N.E.2d 1323 (1978). And the employee injured by the product selected by the employer is
entitled to a workers' compensation recovery from the employer. To state the matter in general
terms, a contract between X and Y can impose significant costs or risks upon others. When this
happens, the law needs to consider imposing liability on the contracting parties.
As for the manufacturer's liability, the interesting question is whether its willingness to pro-
vide its purchasers with some measure of choice should have any extenuating effect upon the third
party's claim of risk-benefit defect. Is the provision of choice itself a species of "benefit"?

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 474 CALIFORNIA LAW REVIEW [Vol. 67:435

grounds that their own expectations have been

justification.
Moreover, since consumer expectations typically develop around
the sales transaction, when the victim is other than the product pur-
chaser the process of identifying the victim's product expectations be-
comes difficult, to say the least. An employee ordered by his employer
to work with a particular product has little occasion to form expecta-
tions about that product; and a person accepting a ride in another's car
ordinarily gives that car no thought. When an automobile injures a
pedestrian or the occupant of another vehicle, any inquiry about the
victim's expectations vis-A-vis that automobile is meaningless.
It is possible, however, that the Barker opinion does not intend
these various applications of its consumer expectations prong; the
Barker "consumer" may be no one other than the product pur-
chaser.230 Significantly, however, even if this is correct, it does not nec-
essarily follow that the prong has no pertinence to the claims of
nonpurchaser product victims. Every injury-inflicting product was
originally purchased by someone, and that purchaser may have in-
tended, at least implicitly, to provide some safety protection to the vic-
tim--or, more precisely, to minimize the accident risk which the
product projects upon that victim. If so, the intended third party bene-
ficiary doctrine can be invoked to give the victim the legal benefit of the
purchaser's expectations. Reasoning of this sort is not at all disingenu-
ous;231 the third party beneficiary doctrine rule is an integral part of
modem contract law.232 In pursuit of that reasoning, it can be assumed
that car buyers possess a concern for the safety of family and acquaint-
ances who at their invitation or with their permission drive the car or
accept a ride in it. The product victim in Barker was an employee. As
a general matter, employers buying factory machinery assumedly pre-
fer to minimize the likelihood of employee injuries, at least for the rea-
son that those injuries impose substantial direct (workers'
compensation) and indirect233 monetary burdens upon the employer.

230. In a different context, the opinion refers disjunctively to "consumers, users, or bystand-
ers." 20 Cal. 3d at 434, 573 P.2d at 457, 143 Cal. Rptr. at 239.
231. Indeed, it was relied on by Justice Cardozo in justifying liability in MacPherson. Mac-
Pherson v. Buick Motor Co., 217 N.Y. 382, 393, 111 N.E. 1050, 1054 (1916).
232. In the Barker context, since the purchaser's safety expectations are themselves typically
implicit, there is no overwhelming harm in the concession that the intention to benefit is implicit
also. However, this doubling-up of the implicitness in the purchaser's expectations is certainly
troublesome, and reduces one's ability to muster enthusiasm for an expectations analysis.
233. These indirect costs-including the disruption of the assembly line and the cost of
machine repair--generally exceed the costs of workers' compensation. See NATIONAL COMMIS-
SION ON STATE WORKMEN'S COMPENSATION LAWS, COMPENDIUM ON WORKMEN'S COMPENSA-
TION 6 (1973). Because the employer's intention, though explainable in employer-cost terms, is
essentially to prevent employee injuries, it is proper to regard the employee as an intended benefi-

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 475

While some form of third party beneficiary reasoning th

in order, that reasoning contains both an important special fe
important limitations. The special feature is that in the third
lawsuit, the person whose expectations are to be honored is
third party plaintiff-victim but rather the product purchaser; th
ture provides a basic guide for the resulting expectations in
The relevant limitations are present at both the particular and
eral level. As earlier illustrations have made clear,235 the facts of
ticular case can easily negate the assumptions that would oth
support an intended beneficiary claim. And at some general p
entire beneficiary reasoning attenuates. It is doubtful, for examp
product purchasers typically intend to confer meaningful safety
tion upon mere bystanders.236 Since the bystander's right to
strongest proof of product liability's pure tort element,237 it
surprising that Barker's contract-related prong encounters dif
in reaching the bystander.

b. What Are "Ordinary Expectations"?

When the victim is the product purchaser or that person's in
beneficiary, the warranty foundation of strict liability gives goo
for attending to the purchaser's product expectations. But the
remains of ascertaining, in any satisfying way, what those exp
are.

Expectations can be fostered by communications abou

uct from the manufacturer to the product purchaser. As
ufacturer has issued statements or advertisements praisin
If those statements include claims that meet minimum standards of

ciary. Given both the economic relationship between employer and employee and the fo
legal liability for the latter's injuries, the employee can almost be regarded as a creditor ben
ary.
234. This is in a way helpful news for the consumer expectations test, since an institutional
purchaser like an employer or school may have contemplated its purchase in a way that facilitates
the identification of safety expectations.
235. See notes 227-28 and accompanying text supra.
236. See Vetri, Products Liability. The Developing Framework for Analysis, 54 OR. L. REV.
293, 297 (1975). There is an obvious and interesting question here as to the incidence of purchaser
altruism.

237. There can be questions about the relationship between Barker's "ordinary consumer"
and the product's actual consumer/purchaser in the particular case. What if the particular con-
sumer has a higher expectation for the product's safety (induced, perhaps, by the comments of an
acquaintance) than the "ordinary" consumer has? It seems clear that the Barker rule would not
take this expectation into account. Conversely, what if the specific consumer, given his actual
knowledge at the time of sale, has lower expectations for the product than those of ordinary con-
sumers? If the purchaser's lower expectations result from information conveyed to him by the
product manufacturer or seller, those expectations (rather than general consumer expectations)
should be deemed controlling. But if his reduced expectations derive from his own investigatory
efforts, the question is much more difficult.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
476 CALIFORNIA LAW REVIEW [Vol. 67:435

specificity and materiality, and if the product's

these claims results in an injury, the victim has no
thing in Barker; the Restatement doctrine of fa
tion238 and the U.C.C. doctrine of express war
upon to assure a recovery.240 A consumer expec
deniably helpful in explaining the rationale for t
doctrines themselves do not justify the Barker p
recognition in pre-Barker law and the advantag
doctrinal independence. Now, what if the manu
commending its product are either too vague or
false representations or express warranties? Inso
and the U.C.C. have good reasons for delibera
these statements into account, it is very difficult
should be ruled relevant to a Barker consumer e
So much for "affirmative" communications.
tive" communications in the form of warnings a
gers? As Barker recognizes, when a product's
obvious, its obviousness usually undermines any
contravenes the purchaser's product expectation
a warning lowers the purchaser's reasonable exp
prevents him from later arguing that his expectati
the warned-of product hazard.242

238. RESTATEMENT (SECOND) OF TORTS ? 402B (1965).

239. CAL. COM. CODE ? 2313 (West 1964).
240. See Hauter v. Zogarts, 14 Cal. 3d 104, 534 P.2d 377, 120
241. To this extent I am unable to agree with the thesis expoun
marvelously rich monograph, Shapo, A Representational Theory of C
Function, and Legal Liabilityfor Product Disappointment, 60 VA.
thesis is that a manufacturer's portrayal of its product-largely t
the "initial" as well as the "principal" factor considered in determini
turer's liability. Id. at 1115. The thesis works well if the "portrayal
U.C.C. express warranty or a Restatement product representation
thesis does not easily test out. Consider, for example, the advertise
advertisements that have always praised the car for both economy
emphases. For several years, this advertising stressed economy
Saves." After a well-publicized change of advertising agencies, in
came "We Are Driven," suggesting quality and performance. Wit
ers' lots in fall 1978, its advertising shifted to "We Are Dealing," po
To my mind, these changes in advertising themes, conspicuous thou
legal rule that measures Datsun's personal injury liability to its 197
demanding than those applicable to its 1977-78 purchasers. Nor shou
any material respect from Toyota's ("If You Can Find a Better-B
242. The manufacturer's opportunity to control expectations by i
fies that it would be wrong to classify nonpurchasers as "consum
sumer expectations prong. What if, after Barker, a product pur
entirely adequate warning by the manufacturer, turns the product o
conveying the warning to that person (or, while using the produ
without warning the bystander of the product's danger)? Under Bar

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 477

Of course, given the received "defect" typology,243 war

sort are hardly at the option of the manufacturer. If a pro
a hazard that is sufficiently substantial and unusual, the law
manufacturer liable if it fails to provide the purchaser
of the hazard so that the purchaser may make an "infor
to whether to buy the product in the first place.244 Any w
by the manufacturer under the threat of liability can
might otherwise be the consumer's misimpressions. Thu
requirement in products liability law, whether it is violated
with, deals directly with most of the cases in which Bar
expectations prong has a possible application.245
Because it embodies a direct response to the problem
purchaser knowledge of product hazards that economic
quired to recognize, the warning obligation which products
poses is in general admirable. But while it is a direct res
not be a complete response. One is hesitant to conclude
ing the warning obligation causes the consumer expect
disappear. For one thing, since a warning is required onl
that pose hazards which purchasers might not expect,
quirement necessitates some identification of purch
tions.246 Moreover, where the victim is complaining ab

the unwarned victim would be entitled to sue under a consumer-expectation

warning did not reach him and hence did not reduce his expectations.
It can be argued that a manufacturer should make an effort to warn even
the hazards that its product may cause those persons to encounter. See W. P
162, at 647. But the risk-benefit standard is the right way to define the scope o
obligation. Barker cuts against this by suggesting that an unwarned third p
consumer expectation prong even when the manufacturer's efforts to warn no
unsuccessful in the individual case, were entirely reasonable. What can the ma
do to warn product bystanders?
243. See note 6 and accompanying text supra.
244. See Borel v. Fibreboard Paper Prods. Corp., 493 F.2d 1076, 1088-8
(applying Texas law); Davis v. Wyeth Labs., 399 F.2d 121, 124 (9th Cir. 19
law); RESTATEMENT (SECOND) OF TORTS ?? 388, 399, 402A, Comment k
Schwartz, supra note 96, at 10, 18.
245. Even if a warning resolves the problem of consumer expectations, th
fail the risk-benefit test. Liability could therefore follow, absent some defens
risk. The status in products liability of the defense of "reasonable" a
-assumption of risk when it is not a mere "variant" of contributory negli
quite uncertain. See notes 145-47 and accompanying text supra.
246. A warning is often an effective means for reducing a product's risk. Th
obligation to warn can therefore be derived from Barkers risk-benefit prong as
its consumer expectations prong. For risk-benefit purposes a warning should p
long as any significant minority of the product's consumers may be unawa
hazard. The consumer expectations relevant to the obligation to warn are there
as those expectations the denial of which produces liability per se.
I should add that while the idea behind the warning obligation is excellen
requirement can become perplexing. It is never sufficiently clear what hazards
what language the warning should use, and how large its letters should be.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 478 CALIFORNIA LAW REVIEW [Vol. 67:435

to provide merely informational warnings (as co

for use), the victim's claim can raise a vexing p
tion.247 Even if the warning had been given, w
bought the product anyway? But there is an ev
lem in the informational warning context. Prod
the manufacturer's instructions as to safe prod
way that is both specific in substance and clear
purchaser can read the instructions in the quiet
requirements of explicitness and clarity are adeq
contents of the instructions are in fact communicated to the consumer.
But if an informational warning is to achieve its objective, it must be
appreciated by the consumer before he purchases the product. Given
the circumstances of many retail sales (consider a hardware store on a
Saturday morning), even if such a warning is inscribed on the product's
package and is properly worded, there is no guarantee that the warning
will actually be received by the consumer, let alone absorbed by him,
before he reaches his decision to buy. "Effective" communication is
what is needed, and this may often entail an oral delivery of the warn-
ing to the would-be purchaser, presumably by the product retailer. I
leave open here the question whether the law should actually impose
liability on the manufacturer for its failure to arrange for the effective
transmission of informational warnings.249 What is clear is that only
an effective communication should be allowed to alter what would
otherwise be regarded as the consumer's expectations.
In several ways the question of ordinary consumer expectation
thus presents itself, and we need to know what those expectations
clude. Unfortunately, identifying their substance, or even developin
methodology for identifying their substance, proves to be a frustratin
task. There is a paucity of empirical studies providing any clear a
swers as to what ordinary consumers expect of the products th
buy.250 The warranty literature suggests one answer: consumers exp
products to be free of defects.251 But if one starts with this assessmen
the Barker consumer expectations standard becomes infinitely

247. Compare Haft v. Lone Palm Hotel, 3 Cal. 3d 756, 478 P.2d 465, 91 Cal. Rptr. 745 (19
with Cobbs v. Grant, 8 Cal. 3d 229, 245-46, 502 P.2d 1, 11-12, 104 Cal. Rptr. 505, 515-16 (1
248. See Anderson v. Klix Chem. Co., 256 Or. 199, 472 P.2d 806 (1970).
249. Cf Vandermark v. Ford Motor Co., 61 Cal. 2d 256, 391 P.2d 168, 37 Cal. Rptr.
(1964) (manufacturer strictly liable for retailer's failure to properly assemble final product).
250. Cf Whitford, Strict Products Liability and the Automobile Industry.: Much Ado Ab
Nothing, 1968 Wis. L. REV. 83, 148 (indicating that only one-third of new-car buyers even exp
the manufacturer to compensate them for their out-of-pocket losses if a defect in the car cau
injury; four-fifths expect compensation for the cost of car repairs).
251. See note 153 supra. This assessment would not necessarily apply, however, to
purchase of used products. See L. FRUMER & M. FREIDMAN, 2 PRODUCTS LIABIL
?? 16A[4][b][iv], 19.03[5] (1978).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 479

determinant: the product is defective if it defeats cons

tions, and it defeats consumer expectations if it is defec
this context "defect" means a manufacturing defect252
fect determined by risk-benefit balancing, then the ass
accurate enough as to consumers' implicit expectations
dependent force as a basis for liability; any verdict reach
sumer expectations approach can be independent
applying the regular manufacturing defect rule or Barker'
design defect prong.
To an increasing extent, products cases-including B
involve sophisticated products with complicated design
ucts are available for a wide range of foreseeable uses, m
"improper." If the question is what the ordinary consum
tions" are for such a product's safety performance, th
probably is that while he may have no specific expecta
product at all, he does generally trust that the product has
gently designed in light of foreseeable contingencies w
sumer's safety in mind.253 But if this is the case, the "
expectation standard collapses into the "manufacturer-o
benefit standard. As one court has put it, "[t]he two sta
same because a seller acting reasonably would be sel
product which a reasonable consumer believes he is pur
The frailty of the consumer expectations formulation
omous liability standard can be additionally clarified b
the problems that arise in attempting to render it operatio

252. Here, the problem is that under Barker the consumer expectations s
only for design defects-manufacturing defects being covered by another stan
and accompanying text supra. Barker evidently requires the plaintiff to "decla
The consumer expectations standard, therefore, offers surprisingly little help i
simple case of the new car which is destroyed when a tire explodes 10 miles aw
dealer's showroom. A res ipsa-type argument should be available to make t
notes 186-87 and accompanying text supra.
253. See Hagans v. Oliver Mach. Co., 576 F.2d 97, 100 (5th Cir. 1978) (ap
Henderson v. Ford Motor Co., 519 S.W.2d 87, 93 (Tex. 1974). See also Back v
Mass. Adv. Sh. 1874, 378 N.E.2d 964, 970 (1978) (accepting Barkers risk-ben
for implied warranty liability under the U.C.C.).
The evaluation in the text seems correct when an unsophisticated ordinary
a complicated product. When the purchaser is itself a sophisticated entity li
school, the identification of more specific expectations may prove feasible.
254. Phillips v. Kimwood Mach. Co., 269 Or. 485, 493, 525 P.2d 1033, 1
recognized that the two standards are virtually identical, the Philips court indi
for the manufacturer-oriented standard. In Brown v. Link Belt Corp., 565
Cir. 1971) (applying Oregon law), the court found no error in the giving of a c
standard, given its proximity to the proper standard. In Allen v. Rodgers M
N.W.2d 830 (Iowa 1978), the Iowa Supreme Court, having noted that both th
tions test and the manufacturer-oriented test end up in the same risk-benefit
remain with the consumer test.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 480 CALIFORNIA LAW REVIEW [Vol. 67:435

tion. Assume the product victim merely demon

some hazard in the product's design. Is the vic
more, to have his case submitted to the jury,
jurors themselves are ordinary consumers who
sumers' expectations? If the circumstances of
way complicated, this seems unsatisfactory. A
Court has stated in the context of a motor vehicle accident:

High speed collisions with large rocks are not so common... that the
average person would know from personal experience what to expect
under the circumstances. . ... The jury would therefore be un-
equipped, either by general background or by facts supplied in the rec-
ord, to decide whether this wheel failed to perform as safely as an
ordinary consumer would have expected.255
If, then, in a not-easy case the mere existence of some design dang
does not alone supply a basis for the jury's consumer-expectatio
factfinding, what evidence can the victim present that is both admi
ble and entitled to substantial weight? The victim's own after-the-f
testimony that the product failed to meet his expectations? The tes
mony of some man-on-the-street, selected by the victim and designated
by the victim as a typical consumer? Of course, under the risk-bene
standard the testimony of expert witnesses is quite appropriate, a
there are many experts competent to testify on matters of risk-benefi
product design. But one can hardly imagine what credentials a witn
must possess before he can be certified as an expert on the issue o
ordinary consumer expectations.256
In anything resembling a difficult case, therefore, the consumer ex
pectations standard is unable to stand on its own, and raises the pro
pect of haphazard, impressionistic jury decisionmaking.257 Especia
since a finding of a denial of consumer expectations is dispositive
liability-no presentation of trade-off evidence can counter or rebu
it-this prospect is disquieting. There may be a class of "easy" cases
however, in which the standard is in order. In Hauter v. Zogarts,25
mother bought, as a present for her son, a "Golfing Gizmo"-a dev
for unskilled golfers to use in improving their games. The Gizmo w
so designed, however, that if the player hit underneath the golf ball, th
ball was likely to double back and strike the player. Hitting at balls
such a way is exactly the kind of thing that inexperienced golfers-w
comprise the Gizmo's intended market-are likely to do. Fred Haute

255. Heaton v. Ford Motor Co., 248 Or. 467, 473, 435 P.2d 806, 809 (1967).
256. See id. at 478-79, 435 P.2d at 811 (O'Connell J., dissenting).
257. Its aversion to an " 'overkill' of subjectivity" in jury decisionmaking has led the Depa
ment of Commerce to eschew the consumer expectations approach in its Draft Uniform Produc
Liability Law. Uniform Law, supra note 175, at 3005.
258. 14 Cal. 3d 104, 534 P.2d 377, 120 Cal. Rptr. 681 (1975).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 481

suffered brain damage when the golf ball struck him on the
supreme court granted him a recovery as a matter of law,
part on the U.C.C. theory of implied warranty: the product
ously "not fit for the ordinary purposes for which such
used.'"259
The court's later Barker opinion, in explaining its consumer ex-
pectations prong, relies on Justice Traynor's Greenman observation
that "implicit in [a product's] presence on the market . . . [is] a repre-
sentation that it [will] safely do the jobs for which it was built."260 In
Hauter, the victim's use of the product was wholly reasonable and as
intended by the manufacturer, and there were no external forces, either
behavioral or environmental, that contributed to the accident. In cases
which comply with these restrictive conditions, it seems meaningful to
say, in line with Greenman, that what the consumer assumes261 is that
the product is flat-out "safe."262
Even here, however, a warning analysis cannot be avoided. If a
warning of the product hazard is not given to the Gizmo's purchaser,
that failure sustains liability standing alone. If, on the other hand, a
warning has been effectively communicated, the consumer expectations
argument is defeated. The warning analysis thus retains a certain ulti-
mate priority. Still, in cases like Hauter the nature of the design hazard
has an important impact. The relationship of that hazard to normal
product use generates the warning obligation in the first place. Given
that relationship, if a warning has not been issued the law should be
more than willing to resolve any cause-in-fact ambiguity in favor of the
plaintiff, since it is natural to assume that no rational purchaser, know-
ing of the hazard, would have purchased the product.263 And if a
warning has been formally given, the same assumption can strengthen
a claim that the communication of the warning was not sufficiently ef-
fective. The factfinder should be reluctant to believe that the purchaser
of such a product was meaningfully informed.

259. CAL. COM. CODE ? 2314(2)(c) (West 1964). For similar implied warranty reasoning, see
McCabe v. L.K. Liggett Drug Co., 330 Mass. 177, 112 N.E.2d 254 (1953) (coffee maker designed
so that it exploded when properly used).
260. 59 Cal. 2d at 64, 377 P.2d at 901, 27 Cal. Rptr. at 701.
261. In Hauter, these are the expectations of the mother, with her son as the intended third
party beneficiary.
262. See Brief for Amici Curiae (Siegfried Hesse & Peter Elkind), at 10-11, in Barker. As
Hauter suggests, implied warranty is also available in such a case. But one can agree with Justice
Traynor about the inaptness of implied warranty's "reasonable time" notice requirement in the
personal injury setting. See text accompanying notes 148-49 supra.
263. See Twerski, Weinstein, Donaher, & Piehler, supra note 103, at 504-05.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 482 CALIFORNIA LAW REVIEW [Vol. 67:435

IV

BOUNDARIES OF STRICT PRODUCTS LIABILITY

A. "Hindsight" Evaluation of Risks and Benefits

The previous discussion of the relationship between negligence
and strict defect liability in matters of product design left open what the
strict-liability result should be in each of two cases.264 In Case A, the
product contains a hazard which the manufacturer neither knows of
nor could reasonably know of at the time of the product's design and
sale. With knowledge, the manufacturer could easily have reduced or
eliminated the hazard, either by modifying the product or by issuing
instructions or a proper warning. In Case B, the product contains a
hazard which, though fully known, is nevertheless unpreventable by
the manufacturer at the time of the product's sale; the hazard subse-
quently becomes preventable when new safety technology is developed.
Case A raises the issue of "unknowable" dangers, which a recent
federal study, having canvassed the case law nationwide, concludes are
beyond the reach even of strict liability.265 Case B presents the issue of
"state of the art" in strict liability-"state of the art" not in the weak
sense of industry custom,266 but in the strong sense of the full range of
design options technologically available at the time of the product's
sale.267 These Cases can be placed in context by noting the kinds of
products they are likely to involve.268 Case B has special relevance for
"old" products-products sold many years ago but which are still in
use and hence still capable of causing injuries.269 Case A has special
relevance for pharmaceutical products. Here, the hazards unknowable

264. See text following note 178 supra.

265. See FINAL REPORT, supra note 6, at 11-9. For California's contribution to this case law,
see Christofferson v. Kaiser Found. Hosps., 15 Cal. App. 3d 75, 92 Cal. Rptr. 825 (1st Dist. 1971).
266. Compliance with custom clearly does not negate the possibility of defect. See FINAL
REPORT, supra note 6, at VII-34 through 37.
267. Both Cases test the maxim that strict liability dispenses with the "scienter" requirement.
See Keeton, supra note 39, at 38; Wade, supra note 39, at 834-35. Especially in Case B, that
maxim comes into conflict with the standard strict liability instruction requiring that the defect
exist in the product "when it left possession of the defendant." BAJI, CALIFORNIA JURY INSTRUC-
TIONS, CIVIL, 9.00.3, 9.00.5 (1977). For an endorsement of liability in Case B, see Phillips, supra
note 73, at 120-21. The Department of Commerce's Draft Uniform Products Liability Law is
explicitly hostile to a "hindsight" approach; it would deny liability all of the time in Case A and
most of the time in Case B. Uniform Law, supra note 175, at 2998-99, 3005, 3006-07.
268. For an example of a preliminary Case A problem followed by an interesting Case B
problem, see The Devils in the Product Liability Laws, Bus. WEEK, Feb. 12, 1979, at 72, 75: "Some
years ago . . . Black & Decker Mfg. Co. realized that metal housings on power tools sold to
consumers could present a shock hazard. The company 'changed to plastic as soon as the state of
the art got to the point where plastics could take the abuse,' [according to the statement of a
company official]."
269. The federal study's review of appellate products opinions determined that in 13% of the
cases the product allegedly causing the injury was more than 20 years old. FINAL REPORT, supra

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 483

at the time of sale may not manifest themselves until many year
or, unlike Case B products, their hazards can erupt quite promptl
the drug is brought to market.
What results does Barker suggest in these Cases? In its o
the court twice indicates that the jury's risk-benefit determ
should be rendered on the basis of "hindsight,"270 and the o
makes clear that these determinations can be adverse to the manufac-
turer even if the manufacturer "took reasonable precautions in an at-
tempt to design a safe product."271 These "hindsight" references can
easily be interpreted as mandating liability in Cases A and B. Other
passages in Barker, however, cast doubt on the propriety of this inter-
pretation.272 Thus, the opinion states that the only product "misuses"
for which a manufacturer can be held liable are those that are "reason-
ably foreseeable."273 And the opinion explicitly reserves for later deci-
sion the "state of the art" issue.274 In all, it seems prudent to regard
Cases A and B as presenting open questions. For these questions, there
are no easy answers.
At a doctrinal level, the basic issue is whether the products in
and B are "defective." Since the defect standard is to be administered
by the jury, the law's definition of defect must be one that ordinary
jurymen can respect. As a matter of ordinary language use, to cal
Product B defective seems very strange. But it is not at all difficult to
appreciate the defect in Product A-it was there all along, even if
unobserved. Of course, under products law "defect" is a proxy for "lia-
bility." Community notions of legal responsibility may well correlate
with the community's understanding of the meaning of "defect." My
sense is that those notions would incline towards liability in Case A as
an acceptable, though strict, application of the basic idea that a manu-
facturer is responsible for the products it sells. In Case B, however, the
community's sense of fairness would probably be offended by the
strongly ex post facto character of a liability imposition,275 given the

note 6, at VII-20. One Massachusetts company has recently been sued for two industrial presses
built in 1895 and 1897. See Inflation in Product Liability, Bus. WEEK, May 31, 1976, at 60.
270. 20 Cal. 3d at 431, 435, 573 P.2d at 454, 457, 143 Cal. Rptr. at 236, 239.
271. Id. at 434, 573 P.2d at 457, 143 Cal. Rptr. at 239.
272. The one time the Barker opinion formally states the risk-benefit test in a section heading,
it includes no mention of hindsight. 20 Cal. 3d at 428, 573 P.2d at 452, 143 Cal. Rptr. at 234. And
if, as the opinion indicates, the basic purpose of the risk-benefit test is to give recognition to "the
complexity of, and trade-offs implicit in, the design process," 20 Cal. 3d at 432, 573 P.2d at 456,
143 Cal. Rptr. at 238, a "hindsight" approach would seem unwarranted.
273. 20 Cal. 3d at 426-27, 573 P.2d at 452, 143 Cal. Rptr. at 234.
274. Id. at 422 n.4, 573 P.2d at 449 n.4, 143 Cal. Rptr. at 231 n.4.
275. A special ex post facto problem arises when application of strict liability is sought against
a manufacturer for a product manufactured and sold prior to the jurisdiction's adoption of the
strict liability rule. See Wansor v. George Hantscho Co., 570 F.2d 1202 (5th Cir. 1978).

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
484 CALIFORNIA LAW REVIEW [Vol. 67:435

manner in which any Case B claim is created b

vance occurring after the product's sale. Strict
ciently fair in a way that ex post facto liability
The two Cases can be further considered fro
the accident prevention goal that Barker's risk-
rates. The stipulation that the product satisfied
dards at the time of its sale indicates that th
principle implicit in negligence law does not call
however, a genuine strict liability rule could c
giving manufacturers strong incentives to accelera
developments.276 A "hindsight" approach woul
gence in the direction of genuine strict liability
manufacturer's incentives to engage in safety
product's sale. After that sale, however, a hindsi
a limiting condition sharply separating it from
While in Cases A and B knowledge of the prod
"benefit"-related technology is lacking at the time
the knowledge must become available, or the ne
developed, prior to the time of the plaintiffs la
with hindsight, no defect can be identified.
The requirement that the knowledge or tech
ing by the time of the lawsuit prevents a hindsigh
subsequent to the product's sale, the safety ince
genuine strict liability. Worse, the requirement
hindsight liability could negatively operate as a
safety discovery process. Consider the product m
of expending funds for either testing existing p
research into possible new safety devices. Under
the manufacturer's expenditure is successful th
tion it uncovers or the new safety technology it d
decisive evidence against it in later suits involvi
earlier date. Insofar as the manufacturer is able
ity consequences, it could be dissuaded from un
product safety research.277

276. See note 64 and accompanying text supra.

277. In Ault v. International Harvester Co., 13 Cal. 3d 113, 528
(1974), a divided supreme court permitted evidence to be admitted
manufacturer's post-accident change or "repair" of its product's
believe that the revised design in Aut--malleable iron in place of
gearbox--was in any way unavailable at the time of the vehicle'
does not therefore represent Case B.
Moreover, the Ault court did not really deny that allowing jur
turer's post-accident design improvements might inhibit manufact
improvements. Rather, the court's perception (probably accurate)
effectively overcome by the manufacturer's strong incentive to rede

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 485

This disincentive would adversely affect the real wo

only insofar as it is manufacturers who by their affirmati
responsible for society's product safety advances. Man
clearly in a good position to seek out information about
recognized product hazards (the Case A problem),278 an
manufacturers in this regard is hardly uncommon. To a
however, this information flows in to manufacturers w
an effort on their part.279 Also, in some instances the vict
examining the product and reconstructing the accident,
the product's hazard. Furthermore, the federal govern
launched an extensive program of product accident ana
acquire knowledge of dangers that existing products
Case B, the effective responsibility for researching new saf
ogy varies from one industrial setting to another. This
often lies with the manufacturers themselves, either ind
least as an industry. But frequently the research is underta
panies who serve (or who would like to serve) as the m
suppliers,281 or by outside entities like governmental a
versity research facilities. In both A and B, therefore, w
liability is capable of deflecting proper safety incentiv
probably happen only in a limited number of cases; giv
ground circumstances, my estimate is that the result
likely in A than in B.
Barker invokes hindsight as an aspect of the risk-b
prong. One can ask how Cases A and B should be de
Barker's other prong, concerning consumer expectat
Cases diverge in analysis and possibly in results. In Cas
typical purchaser implicitly assumes no more than that

to minimize its general liability exposure in the future. In the context of

difficult to identify any such offsetting incentives.
278. Negligence law, undoubtedly taking the manufacturer's good posi
obliges manufacturers to test their products for hidden hazards. See Dalehi
346 U.S. 15, 51-52 (1953) (Jackson, J., dissenting on other grounds); Borel
Prods. Corp., 493 F.2d 1076, 1089-90 (5th Cir. 1973) (applying Texas law); Noe
833-35.

279. Both the negligence rule of Hasson v. Ford Motor Co., 19 Cal. 3d 530, 564 P.2d 857, 138
Cal. Rptr. 705 (1977), and the recall provisions of the Federal Motor Vehicle Safety Program, 15
U.S.C. ?? 1411, 1412(b), 1414(a) (1976), seem to assume that imposing obligations on the manu-
facturers once knowledge is acquired of hazards in previously marketed products does not dis-
courage the materialization of this knowledge.
280. See 15 U.S.C. ? 2054(a) (1976), which sets up the Injury Information Clearinghouse.
281. Thus, in the power tool example described in note 268 supra, the burden of improving
the durability of plastic obviously rested with plastics suppliers rather than with the manufactuers
of power tools. I am told that there is keen competition between metals suppliers and plastics
suppliers. For an indication that pharmaceutical houses are potential suppliers of bloodbanks, see
Epstein, supra note 111, at 118.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 486 CALIFORNIA LAW REVIEW [Vol. 67:435

turer has kept abreast with contemporary technol

product does not defeat consumer expectations.
trast, if the product hazard is unknown to the ma
a fortiori unknown to product buyers. Given th
the hazard in the Case A product could well be r
consumer expectations.
It is arguable, however, that a literal applicat
sumer expectations standard in Case A is ina
may not know of the particular product's hazard,
the twentieth century he is probably aware of the
ucts (especially drugs) containing unknowable h
of consumer expectations can be expanded t
awareness, the product may not frustrate these ex
But it is hardly certain that the contract basis f
mits the expectations question to be posed at th
Even assuming that an abstract expectations inq
very least that inquiry should be comparative,
tions of the product manufacturer as well as tho
chaser. And here it is obvious that manufacturers are in a much better
position than purchasers to estimate the incidence and severity of th
hazards that their products may be later found to contain.284 In thi
way, consumer expectations reasoning, while it is consistent with th
result of no liability in Case B, buttresses the liability argument in Case
A.

Additional factors pertinent to evaluating the hindsight issue con-

cern the reliability of manufacturers' claims that particular cases do
indeed fall within A or B, and the judicial system's ability to assess
these claims. Manufacturers frequently allege that a product hazard
was unknown and not reasonably knowable at the time of the product's
sale. But often enough (though not invariably), we eventually learn
that the manufacturers did indeed possess knowledge of the hazard, or
at least had reason to know of it, from an early date.285 Moreover, in
light of the obligation negligence law imposes on manufacturers to test

282. See Bruce v. Martin-Marietta Corp., 544 F.2d 442, 447 (10th Cir. 1976) (applying
Oklahoma law).
283. See A. Schwartz, supra note 96, at 583-86. Professor Schwartz would recognize con-
sumer expectations liability only in the special case of the drug that causes the very disease it was
meant to prevent. But is the polio vaccine that induces polio really all that different from a (hypo-
thetical) polio vaccine that induces a heart attack in an ordinary person?
284. Cf R. POSNER, supra note 51, at 73-78 (justifying contract defenses of mutual mistake
and impossibility).
285. See Weinstein, Did Industry Suppress Asbestos Data?, L.A. Times, Oct. 23, 1978, at I-A,
col. 4; Harris, Story of DC-lO Defects Raises Serious Doubts About Safety Policies. The Hazard of
Cargo Door Was Known Before Crash, Emerging Details Show, Wall St. J., Mar. 3, 1975, at 1, col.
6; Kramer, Vinyl-Cloride Risks Were Known by Many Before First Deaths, Wall St. J., Oct. 2, 1974,

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 487

their products for hidden hazards,286 in many cases there w

for a strong suspicion that a reasonable testing effort woul
vealed the hazard in a timely fashion. Yet the art form of
seems poorly suited to accomplish either the confirmation o
of suspicions of this sort, given the lapse of time between the
sale and the product lawsuit, and given the elusiveness of t
have known" issue.287 Taken in combination, these practic
tions are sympathetic to liability in Case A: the differenc
knowledge-now and knowledge-then may be something that
ligent for litigation to ignore.
This assessment seems especially useful when we recog
many Case A situations illustrate that Case only in a modifi
this modification-call it Case A-1-the product contains a h
the manufacturer knows about at the time of the product's ori
but only in a vague or partial way; and the difference between
as-then-perceived and the risk-as-now-understood tips the s
risk-benefit balance.288 Here the practical arguments for c
only present knowledge in determining liability seem unusu
Once the manufacturer concedes that it had some original
of the product's risk, we may particularly doubt that its origin
edge was as limited as it now claims; we may be especially t
conclude that reasonable, nonnegligent testing would have r
hazard more fully; and we may be sharply skeptical of th
judge and jury to reconstruct the exact extent of that prior
Liability thus seems in order in Case A-1.289 Moreover, giv
ter-of-degree character, Case A-1 may be applicable to a m
number of real-world situations than Case A, in all its pur

at 1, col. 1. On very early evidence linking cigarette smoking and lung cancer, s
Liggett & Myers Tobacco Co., 295 F.2d 292, 300 (3d Cir. 1961).
286. See note 278 supra. Since negligence law, for obvious reasons of practica
tion, has not placed conspicuous obligations on manufacturers to engage in safety-d
this point applies only weakly to Case B.
287. Case B's "state of the art" issue, given its greater external objectivity, seems
ble for retroactive inquiry. Thus a 1979 court should be reasonably able to ascertain
of plastic in 1970. See note 268 supra. This and the preceding footnote indicate
liability analysis in the accompanying text applies mainly to Case A.
288. There is a similar Case B variation. In Case B-1, the product's known
prevented by the utilization of a safety alternative that is quite expensive at the tim
uct's original sale but becomes less expensive by the time of the victim's lawsu
includes not only the monetary cost of the alternative but also any way in which t
impairs product performance. The power tool example, see note 268 supra, is thus o
acter.) Since the practical problems which the shift from Case A to Case A-i exac
fully pertain to Case B in the first place, see notes 286-87 supra, the Case B-1 varia
require special discussion.
289. Case A-i can itself be extended to cover situations where the manufacturer "should have
had" partial knowledge of the hazard, although it had no actual knowledge. The practical compli-
cations that result from this extension suggest that liability is in order here as well.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 488 CALIFORNIA LAW REVIEW [Vol. 67:435

since the boundary line between A-i and A will

the appraisal that liability is called for in Case A
ment that liability is a good idea in Case A as w
In summary, several indicators, including com
tract reasoning, and the effective capacity of li
liability in Case A, while safety incentive reaso
limited cloud over the liability idea. There is a d
ever, approving of liability in Case B; and besid
incentive problem, the community's sense of f
against liability. Barker's "hindsight" languag
enough to include both Cases A and B. That
both so tersely advanced in Barker and so surr
contrary language as to call for further examin
issue. That examination suggests that on balance
is probably sensible in Case A but that it would
B.291

B. Products So Designed that Their "NWorm is Danger"

Footnote 10 of Barker raises the possibility, without deciding the

question, of strict liability for products which, while free of defects in
the risk-benefit sense, nevertheless "entail a substantial risk of harm
. even if no safer design is feasible."292 These are products whose
"norm is danger."293

290. Did the manufacturer have no knowledge (and no reason to know) of the hazard at the
time of product sale, or rather did it have some knowledge (actual or constructive) of this hazard,
although only of a limited or partial character? For observations on the undesirability of line-
drawing of this sort, see Keeton, Products Liability---Inadequacy of Information, 48 TEXAS L. REV.
398, 408-09 (1970).
291. It appears that the Oregon Supreme Court would impose liability in Case A but not in
Case B. See Wilson v. Piper Aircraft Corp., 282 Or. 61, 577 P.2d 1322, rehearing denied, 282 Or.
411, 579 P.2d 1287 (1978). A recent Indiana statute establishes state-of-the-art as a "defense" to a
products liability claim. IND. CODE ANN. ? 34-4-20A-4(b)(4) (Bums Supp. 1978). As of this writ-
ing, the California Assembly (but not the Senate) has approved a bill stipulating that compliance
with state-of-the-art creates a "rebuttable presumption" of non-defectiveness. Assembly Bill No.
382, 1979-80 Reg. Sess.
In a few instances the distinctions I see between Cases A and B can recede. Consider the
safety device that is beyond the state of the art at the time of the product's sale not because of the
limits of then-current technology but merely because nobody had yet thought of it. Consider also
the unknowable product danger that is due only to a product misuse that was wholly unforesee-
able at the time of sale or to an unforeseen change in the post-sale physical environment in which
the product is asked to perform. In the case of the unforeseeable product misuse, full liability
should probably be placed on the misuser.
292. 20 Cal. 3d at 430 n.10, 573 P.2d at 455 n.10, 143 Cal. Rptr. at 237 n.10. This footnote is
simply ignored in Garcia v. Joseph Vince Co., 84 Cal. App. 3d 868, 148 Cal. Rptr. 843 (2d Dist.
1978).
293. 20 Cal. 3d at 430 n.10, 573 P.2d at 455 n.10, 143 Cal. Rptr. at 237 n.10. The court
ventured this idea on its own; it had not been raised by counsel.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 489

The ramifications of the idea referred to in footnote 1

be understated. Assume a product whose design som
substantial risk. Either there is or is not an appropriate
tive. If there is, liability is authorized by Barker's risk-be
there is not, liability could still be imposed under the
"norm is danger" theory. If, therefore, the footnote 10
ally accepted as a supplement to the risk-benefit standa
risk" products would bear liability quite without regar
bility of feasible design alternatives. Ironically, then, the
ger" concept, while couched in terms of a risk-benefit
directly lead to the exclusion of that analysis; all "sub
products would be subject to what Part II of this Forew
of genuine strict liability.
It is difficult to believe that the Barker court would
rule. That rule would render irrelevant all issues of risk-benefit as well
as of consumer expectations-issues which the Barker court was at
pains to explore. Additionally, the Barker opinion at several junctures
indicates its wariness about genuine strict liability,294 a wariness that is
not without warrant, as Part II of this Foreword has suggested.295
Therefore, there is a need to seek a narrow exegesis of the tersely-stated
footnote 10 idea. This search can be aided by considering particular
products responsible for substantial risks. Of all the products in our
society, motor vehicles are involved in the largest number of accidents.
On the 1977 list compiled by the National Injury Information Clearing
House,296 bicycles rank first as an accident producer. Stairs hold the
number two position, with power mowers ranking sixth and
skateboards seventh (up from eighteenth the year before). Even if
these products are flawlessly designed, it can be assumed that substan-
tial risks remain.

Most of the accidents which give rise to these substantial risks,

however, are immediately caused by human inexperience or careless-
ness-the carelessness of the victim himself or some third person.
Since this carelessness is all too statistically predictable, it constitutes a
"reasonably foreseeable" product "misuse" that manufacturers of cars,
bikes, ladders, and skateboards are required to cope with for purposes
of the risk-benefit design defect standard. But assuming no such defect,
and conceding the primacy of victim or third-party carelessness as the

294. See notes 48, 50 and accompanying text supra.

295. See notes 65-83 and accompanying text supra.
296. NATIONAL INJURY INFORMATION CLEARINGHOUSE, 1977 CONSUMER PRODUCTS HAZ
ARD INDEX (1978). The Clearinghouse's rankings (which do not include motor vehicles) reflect
the absolute number of annual accidents, not the ratio of accidents to usage. For purposes of any
norm-is-danger liability rule, it is probably the ratio and not the absolute number that should be
the relevant factor.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
490 CALIFORNIA LAW REVIEW [Vol. 67:435

accident cause, the fairness and accident prev

noted in the earlier discussion of the genuine str
justify the assignment of significant liability to th
Given this analysis, the most plausible interpret
is danger" proposal would stipulate that a pro
norm only if it is responsible for a major risk e
actors-including the product designer and the
eminently reasonable care.298 This formulation
10 concept with a precedent: the rule of str
trahazardous activities,299 a rule which the supr
heartedly3?" (if only occasionally301) endorsed.
For most products that could be deemed ultr
the hazard is so conspicuous as to be apprecia
product's regular users. If the product has irregu
be aware of the hazard, the manufacturer is under
ucts liability obligations to provide those users
ing.302 Since the failure to warn provides an in
liability, for footnote 10 purposes we can assume
is knowledgeable of the product's peril.
What if the victim is not the knowledgeab
rather some other uninvolved person? Th
trahazardous activity is blasting.303 The blaster w
ously uses an explosive (and hence highly danger
is brought against the blaster rather than again
facturer. Indeed, in almost all ultrahazardous act
the entity carrying on the activity utilizes a pro

297. The person injured by a product user's carelessness has a

user. When an injury is produced by the combination of the care
negligence of some other party, fairness requires that the victim b
accident cost. See Schwartz, supra note 75, at 725-76. When the
with manufacturer conduct that is not at all negligent and a produ
the assignment of any significant liability to the manufacturer wo
298. Footnote 10's norm-is-danger idea (and language) are draw
nor's 1965 speech-article. See Traynor, supra note 32, at 367-69.
infected blood is not a norm-is-danger product-indeed, its danger e
quite safe norm. (Less than 1% of all blood units are infected.)
example of a product with a special kind of manufacturing defec
element of unpreventability links up infected blood with norm-
Traynor's discussion suggests.
299. RESTATEMENT OF TORTS ?? 519, 520(a), 524(2)(a) (1938).
300. See Luthringer v. Moore, 31 Cal. 2d 489, 190 P.2d 1 (194
301. As pointed out in Schwartz, supra note 75, at 700 n.17, th
cided a single ultrahazardous activity case in the thirty years since
302. See note 244 and accompanying text supra. See also Can
237 Cal. App. 2d 44, 46 Cal. Rptr. 552 (3d Dist. 1965) (explosive
303. See W. PROSSER, supra note 162, at 513.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 491

way or another.3' What this suggests is that a product m

with some use of the product before any ultrahazard is
sider the injuries suffered by the neighbor of the person em
plosives, by a person in the building which is being fumi
deadly gas, or by someone on the highway as a truck appr
ing a highly volatile liquid. Even if the products themse
deemed ultrahazardous, the ultrahazard is created by a par
uct use; and the law affords the victim a clear compen
against the product user who is knowingly carrying
trahazardous activity.305 Products like explosives, deadly
volatile liquids can be used in a variety of ways, and the
the risk occasioned by the product greatly depends on w
owner selects. Whether tort law's purpose is fairness or a
vention, liability should probably remain on the produc
conducting the ultrahazardous activity rather than being
to the product manufacturer.
Even when the victim of the ultrahazardous product is th
user himself, the user's knowledge of the product's risk
manufacturer liability.306 This knowledge weakens the in
could otherwise be drawn from the ultrahazardous activit
That rule rests partly on the fairness goal of preventing any
richment of the party conducting the activity at the expens
tivity's victim.307 Cigarettes can serve as an example of
danger product. Consider the cigarette manufacturer that
business, knowing of the cancer risk. For every cigarett
company sells for the sake of profit, there is a cigarette pur
consumer who also knows of the cancer risk but who is
benefits of cigarette smoking. The basic fact of this benefit
the consumer because of the consumer's choice plainly p
manufacturer's enrichment from being condemned as unj

304. See, e.g., Luthringer v. Moore, 31 Cal. 2d 489, 190 P.2d 1 (1948) (exter
with hydrocyanic acid gas); Siegler v. Kuhlman, 81 Wash. 2d 448, 502 P.2d 1181 (
ing gasoline by truck).
305. This describes the tort law rule. The employer who furnishes its emplo
preventably dangerous work tool, see, e.g., Hagans v. Oliver Mach. Co., 576 F.2d
1978), is liable under workers' compensation for any injuries that result.
306. For purposes of Barker's contract prong, this knowledge surely under
that the product defeats the consumer's expectations.
307. See Keeton, Conditional Fault in the Law of Torts, 72 HARV. L. REV. 4
308. Should this evaluation be altered in favor of liability if the consumer's
product seems less "voluntary"? Assume a patient who can be cured only by t
danger-norm drug. Escaping illness is a major benefit; indeed, it is only because
so substantial that the appearance of choice is diminished.
The lack of apparent consumer choice impressed Justice Traynor as a reason
ing liability on manufacturers of norm-is-danger products. He proposed stric
inherently dangerous drug only if "a reasonably close substitute exists," and not

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 492 CALIFORNIA LAW REVIEW [Vol. 67:435

Fairness apart, norm-is-danger liability could

in accident prevention. By raising the product
would reduce the product's sales and proport
number of accidents that the product causes. M
pothesis there is no feasible safer design at the
sale, a manufacturer's economic interest in pre
would give it a sharp incentive to conduct resear
a safer product design in the future.
This safety analysis is, of course, a restatement
vention reasoning associated with genuine strict
news is that the norm-is-danger liability prop
cases in which the product accident is at least part
dard behavior of the victim or some other pers
vulnerabilities that weaken genuine strict liabil
that the assumption that the consumer has know
danger-norm creates a new vulnerability. S
counted on to produce advantageous safety effe
that consumers are unaware of the product haza
of that hazard and hence recognize the product's
sition of liability would not diminish the prod
would not succeed in fostering safety incentives
clusion is premature. The conclusion results fr
itself assumes that the product is sold by a profit-
turer in a competitive market. These are plainly
It may well be, for example, that cigarette com
gree of monopoly power312 or that they do not
prices with standard profit maximization in min
If these are the economic realities, then c

be used despite the danger." Traynor, supra note 32, at 368-69. In a

Wade would classify unavoidably dangerous products according t
pose strict liability only on manufacturers of low utility products
that the sale of these products may be per se inappropriate. Wad
If a safe substitute drug exists, a reasonable view is that liabilit
the physician who chooses to use the more dangerous drug prod
wrong, buying and using them seems equally wrong.
309. See text accompanying notes 60-64 supra. These resource
doubtedly part of the rationale for ultrahazardous activity strict l
310. See notes 65, 73-77 and accompanying text supra.
311. See notes 61-64 and accompanying text supra. Ultrahazard
covery to the activity's "knowledgeable participant." See RESTAT
This person is the counterpart of the product user. See Gaston v. H
App. 1978).
312. On the significance of monopoly power, see G. CALABRESI, supra note 51, at 81-85.
313. For discussions of corporate pricing practices and how they deviate from classical theory,
see Pricing Strategy in an Inflation Economy, Bus. WEEK, Apr. 6, 1974, at 43; Flexible Pricing,
Bus. WEEK, Dec. 12, 1977, at 78. Additionally, some products, like drugs administered in hospi-
tals, may be distributed in a manner that departs from the market pattern.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 19791] UNDERSTANDING PRODUCTS LIABILITY 493

would not nullify the safety effects of a norm-is-danger

Even if this is so, however, the safety argument suppor
does not necessarily prevail. As the earlier discussions of
liability and the natural impurities cases314 indicated, o
products liability rule has never clearly made up its mind
it should move beyond the negligence principle in orde
future-directed safety incentives. The footnote 10 issue t
crucible, although one highly dependent on particular ec
ings, for testing the ambitiousness of products liability's
vention objective.

CONCLUSION

Strict products liability is a subtle rather than a sensational doc-

trine, as is often supposed. The law has refused even to consider a rule
of genuine strict liability, not implausible though that rule may be. The
existing moderate rule of products liability hinges on the requirement
of a "defect." Since "defect" means that there is something "wrong"
with the product, it is accurate (if ironic) to understand strict liability as
involving a "fault" system after all--one concerned with the fault in
the manufacturer's product rather than the fault in its conduct. What
becomes interesting is the attempt to identify those circumstances in
which these two kinds of fault are not equivalent.
In manufacturing defect cases, products liability essentially pro-
vides a shortcut into, and a guarantee of the results supposedly avail-
able under, a sophisticated negligence system-one that has rejected
privity, accepted res ipsa, extended itself (in a somewhat strict-liability
fashion) via various vicarious liability rules, and attracted a proficient
plaintiffs' bar. What conclusion strict liability reaches in the rare case
of the unpreventable manufacturing defect is a question that remains
intriguingly unanswered.
Barker's focus is on design defects. The design-defect problem im-
plicates the question of the relationship between tort and contract prin-
ciples within products liability. Negligence, a basic tort doctrine, and
implied warranty, a basic contract doctrine, provided the two sources
for the original strict products liability rule. Cronin, in disparaging
"consumer expectations," seemingly buried products liability's contract
element in favor ofa curtly defined (indeed, undefined) tort alternative.
Barker provides that tort element with a formal risk-benefit expression
that reaches deep into the tort tradition. But at the same time, Barker
unexpectedly revives products liability's contract tradition by endorsing
an alternative consumer expectations liability test. Barker thus con-

314. See notes 173-76 and accompanying text supra.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
494 CALIFORNIA LAW REVIEW [Vol. 67:435

firms the general jurisdiction of both tort doctr

ranty doctrine in products litigation. In earlier
doctrines had been somewhat hemmed in by pr
the privity rule in tort cases was deficient in it
counterpart rule in warranty cases made poor se
tual terms, at least as applied to the ultimate p
release from privity allowed both tort and cont
products field.
In attesting to their continued prosperity, Ba
gests that within products liability tort and co
each separate from and independent of the oth
independence is valuable in a number of ways. T
the explanation for the bystander's right to su
explain the attention paid to false product repr
likewise help make clear the relevance of produc
Barker's expectation of the separateness of tort
nerable, however, in other respects. Barker fai
negation of a contract claim can weaken an oth
claim. There has been no rush, for example, to
on the sellers of used products,315 whose purcha
expectations. And if the purchaser of a new
knowledge of a product hazard at the time of the s
a contract-related defense of assumption of risk
tort-defined liability claim.316
Barker also declines to dig into the consume
cept. The concept's contract basis suggests that o
chaser qualifies as the "consumer" whose ex
considered, although certain third party produ
garded as the purchaser's intended beneficiaries;
Barker does not discuss. Consumer expectations
manufacturer's communications about its produc
praiseworthy or cautionary. But preexisting prod
of false representation and the obligation to warn a
correct legal effect to these communications. "C
is helpful insofar as it provides a rationale-a un
these preexisting doctrines.
Communications apart, "consumer expectation
identify. There is an irony here. Barker found it
on Cronin because Cronin's simple "defect" requ

315. See note 251 supra.

316. The assumption of risk issue is most dramatic when the con
safety-device option and then in his subsequent lawsuit alleges t
rendered the product's design defective. See notes 145-47 and acc

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
 1979] UNDERSTANDING PRODUCTS LIABILITY 495

adequate guidance to factfinders in difficult design cases

sumer expectations "prong" included in the Barker elabor
from a similar inadequacy. Two lines of analysis can b
First, consumers may expect products not to cause injury
erly used in a wholly normal environment. But if a pro
does so threaten injury, the manufacturer's primary liabi
failure to warn. Second, consumers trust and expect that
they buy come equipped with intelligent designs. But w
gests is the partial convergence of the contract (consumer
standard and the tort (risk-benefit) standard, since an intell
is essentially one that is risk-beneficial. While tort law ob
person to act in risk-beneficial ways towards others, the
tionship between manufacturer and consumer does add
emphasis to that obligation. In the products context, tort
are thus able to harmonize, with the theory of the form
forced by the theory of the latter.317
Given the pertinence of both tort and contract theory, t
reason to interpret products liability in terms of either tort
contract or contract circumventing tort.318 At the ope
however, the tendency towards convergence between the
test and the consumer expectations test calls into question
ness of tort and contract that Barker posits. Other jurisdicti
this convergence, have selected one test or the other, but ha
need to opt for both.319 Since the risk-benefit test usu
better job in initiating and guiding the jury's deliberatio
only it extends to all product victims, including the bystand
nary situations it should be the test of preference. In light o
culties of identifying the particular purchaser's product expe
the accident victim, the consumer expectations test, whateve
ness in theory, is probably not worth the effort in practice.
With respect to this risk-benefit standard, Barker's prof
in the burden of proof is more gallant than wise. But t
Barker can be modified in a moderate way that make
once the plaintiff identifies a feasible design alternative
have prevented his injury, the manufacturer should be re
plain why that alternative would not have been a good i
sight" should probably be permitted in risk-benefit appra

317. Likewise, the standard of liability in medical malpractice is derived f

interweaving of tort and contract. See D. LOUISELL & H. WILLIAMS, 1 MEDIC
? 8.03, at 194-200 (1960). See generally O'Connell, supra note 155.
318. See O'Connell, supra note 64, at 533-34, for a witty narrative of produ
tory. See also G. GILMORE, THE DEATH OF CONTRACT 94 (1974) (suggesting that
ity may involve both tort and contract "swallowed up in a generalized theory of
319. See notes 253-54 and accompanying text supra.

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms
496 CALIFORNIA LAW REVIEW [Vol. 67:435

issue is the existence of a product-created risk t

known (or only partially known) by the manufa
the issue is the development after the product's sal
nology. Imposing liability on manufacturers of a
ucts whose norm is danger is perhaps defensib
such a liability rule will need to show how mark
vert the rule into a safety-research incentive op
turers. They will then need to persuade courts th
ambitious enough to step conspicuously beyond
yond contract as well, to provide this incentive.
Barker is a feast of an opinion, in all. But it
more for the fundamental products liability issu
cates than for the correctness of the rules it esp

This content downloaded from 103.231.241.102 on Thu, 27 Jun 2019 09:16:51 UTC
All use subject to https://about.jstor.org/terms