Izzarelli v. R.J. Reynolds Tobacco Co.

******************************************************
The ‘‘officially released’’ date that appears near the
beginning of each opinion is the date the opinion will
be published in the Connecticut Law Journal or the
date it was released as a slip opinion. The operative
date for the beginning of all time periods for filing
postopinion motions and petitions for certification is
the ‘‘officially released’’ date appearing in the opinion.
In no event will any such motions be accepted before
the ‘‘officially released’’ date.
All opinions are subject to modification and technical
correction prior to official publication in the Connecti-
cut Reports and Connecticut Appellate Reports. In the
event of discrepancies between the electronic version
of an opinion and the print version appearing in the
Connecticut Law Journal and subsequently in the Con-
necticut Reports or Connecticut Appellate Reports, the
latest print version is to be considered authoritative.
The syllabus and procedural history accompanying
the opinion as it appears on the Commission on Official
Legal Publications Electronic Bulletin Board Service
and in the Connecticut Law Journal and bound volumes
of official reports are copyrighted by the Secretary of
the State, State of Connecticut, and may not be repro-
duced and distributed without the express written per-
mission of the Commission on Official Legal
Publications, Judicial Branch, State of Connecticut.
******************************************************
IZZARELLI v. R.J. REYNOLDS TOBACCO COMPANY—CONCURRENCE
ZARELLA, J., with whom ESPINOSA, J., joins, con-
curring. I agree with the majority’s answer to the certi-
fied question but not its analysis because I believe we
should replace the dual design defect standards
announced in Potter v. Chicago Pneumatic Tool Co.,
241 Conn. 199, 219–23, 694 A.2d 1319 (1997), with the
more modern standard for design defect claims set forth
in the Restatement (Third) of Torts, Products Liability.
This case presents our first occasion to directly con-
sider our design defect standards since Potter was
decided nearly twenty years ago. Potter formulated our
standards at a time when design defect law was in
transition. Courts had acknowledged that the ordinary
consumer expectations test, derived from comment (i)
to § 402A of the Restatement (Second) of Torts, was
ill-suited for judging product design cases because it
did not provide sufficient guidance to juries and was
often used to deny recovery to plaintiffs for product
related injuries. See, e.g., 1 D. Owen & M. Davis, Prod-
ucts Liability (4th Ed. 2014) § 8:4, pp. 714–16. In its
place, courts overwhelmingly turned to the risk-utility
test, an alternative to the ordinary consumer expecta-
tions test, which allows a jury to assess a product design
by weighing factors relating to its risks and benefits
against those of possible design alternatives. Id., §§ 8:6
through 8:7, pp. 722–26.
Sensitive to criticisms of the ordinary test, Potter
created the ‘‘modified’’ consumer expectations test by
incorporating risk-utility factors into the existing con-
sumer expectations test. Potter v. Chicago Pneumatic
Tool Co., supra, 241 Conn. 220, 222; see id., 221. In
formulating its standards, however, Potter rejected the
approach of a draft form of the Restatement (Third) of
Torts, Products Liability, which required, as an essential
part of its risk-utility test, that a plaintiff present evi-
dence of a reasonable alternative design. See id., 214–19,
221. Such evidence allows for a jury to assess the manu-
facturer’s chosen design by comparing it against the
costs and benefits of adopting a safer alternative. See
1 D. Owen & M. Davis, supra, § 8:10, p. 739. In Potter,
the court expressed concern that requiring this proof
might harm a plaintiff by placing too many evidentiary
hurdles along the path to recovery by, for example,
forcing the plaintiff to present expert testimony in every
case. See Potter v. Chicago Pneumatic Tool Co.,
supra, 217–19.
Both of Potter’s tests were ill-conceived, however,
and they remain problematic today, even with the
majority’s clarification of when each test should be
applied. The problems with Potter’s standards are not
limited to their lack of clarity. More fundamentally, its
rejection of a reasonable alternative design requirement
leaves a jury applying its standards without any objec-
tive basis against which to assess the product design
at issue.
Since Potter was decided, a consensus has emerged
among courts and commentators that, in design defect
cases, proof of some safer and reasonable alternative
design is generally necessary to provide the jury with an
objective basis for assessing whether a manufacturer’s
chosen design is defective. See 1 D. Owen & M. Davis,
supra, § 8:10, p. 739 (‘‘[C]ost-benefit analysis of an alter-
native design lies at the heart of design defectiveness.
. . . [D]esign defectiveness is usually best resolved by
risk-utility analysis, the purpose of which is to deter-
mine whether the risk of injury might have been reduced
or avoided if the manufacturer had used a feasible alter-
native design.’’ [Footnotes omitted; internal quotation
marks omitted.]). Proof of a reasonable alternative
design allows the jury to compare the manufacturer’s
design against safer alternatives to decide whether the
manufacturer could reasonably have made a safer prod-
uct. See, e.g., id.
Reflecting this consensus, the Restatement (Third)
requires proof of a reasonable alternative design. See
Restatement (Third), Torts, Products Liability § 2 (b), p.
14 (1998). Notably, however, the Restatement (Third),
which was adopted shortly after Potter was decided,
resolves Potter’s stated concerns by incorporating
appropriate exceptions to the reasonable alternative
design requirement and by making clear that expert
testimony is not required in all cases to satisfy this
obligation. See id., § 2, comment (e), pp. 21–22; id., § 3,
p. 111; id., § 4 (a), p. 120.
In light of these developments favoring the use of a
pure risk-utility balancing standard based on proof of
a reasonable alternative design, I believe that we should
take this rare opportunity to reconsider our design
defect standards rather than simply clarifying and reaf-
firming them, as the majority does today.1
On the basis of my review of the Restatement (Third),
I am persuaded that we should now adopt the approach
set forth therein as an accurate statement of our law
controlling design defect claims. The Restatement
(Third) has resolved the concerns identified in Potter
and provides a clearer and fairer method for resolving
design claims. Because the Restatement (Third) does
not rely on the standards contained in § 402A of the
Restatement (Second) of Torts, and does not provide
an absolute bar to an action against a cigarette manufac-
turer for defective design, I join in the majority’s answer
to the certified question, although not its analysis.
I
JUDGING DESIGN DEFECTS: RISK-UTILITY
BALANCING AND REASONABLE
ALTERNATIVE DESIGN
EVIDENCE
A
Restatement (Third)’s Design Defect Test
Consistent with our product liability law, the
Restatement (Third) recognizes three distinct catego-
ries of product defect claims: manufacturing defects,
design defects, and marketing defects, also called a
failure to warn. Restatement (Third), supra, § 2 (a), (b)
and (c), p. 14; see also Hurley v. Heart Physicians,
P.C., 278 Conn. 305, 315, 898 A.2d 777 (2006) (‘‘[a] prod-
uct may be defective due to a flaw in the manufacturing
process, a design defect or because of inadequate warn-
ings or instructions’’ [internal quotation marks omit-
ted]). Recognizing that each of these categories of
defects presents different circumstances, the Restate-
ment (Third) adopts separate liability standards for
each category. See Restatement (Third), supra, § 2 (a),
(b) and (c), p. 14. The present case implicates our stan-
dards for the second category, design defects. See id.,
§ 2 (b), p. 14.
For design defect claims, the Restatement (Third)
uses a risk-utility balancing test that allows a jury to
decide liability by comparing the risks and benefits of
the manufacturer’s design against the risks and benefits
of adopting a safer alternative. See id. At its core, the
risk-utility test asks ‘‘whether the safety benefits of
remedying a design danger [are] worth the costs.’’ 1 D.
Owen & M. Davis, supra, § 8:6, p. 723. It requires a
plaintiff challenging a product design to show that the
manufacturer could reasonably have designed its prod-
uct to be safer. See id.; see also T. Jankowski, ‘‘Focusing
on Quality and Risk: The Central Role of Reasonable
Alternatives in Evaluating Design and Warning Deci-
sions,’’ 36 S. Tex. L. Rev. 283, 320 (1995). The jury then
compares the risks and benefits of the manufacturer’s
design against the risks, benefits, and costs of adopting
the proposed alternative. See 1 D. Owen & M. Davis,
supra, § 8:10, pp. 739–41; see also T. Jankowski, supra,
343. Consistent with the approach of the Restatement
(Third), a ‘‘vast majority’’ of courts and commentators
agree that the risk-utility balancing test provides the
best standard for judging design defect claims. Evans
v. Lorillard Tobacco Co., 465 Mass. 411, 426, 990 N.E.2d
997 (2013); see also 1 D. Owen & M. Davis, supra, 8:6,
p. 724 (‘‘the risk-utility test appears to have become
America’s preferred test for design defectiveness’’); A.
Twerski & J. Henderson, ‘‘Manufacturers’ Liability for
Defective Product Designs: The Triumph of Risk-Util-
ity,’’ 74 Brook. L. Rev. 1061, 1067 (2009) (‘‘virtually every
major torts scholar who ha[s] looked carefully at the
issue of design defect over the past several decades
ha[s] embraced risk-utility balancing’’).
Under the risk-utility test set forth in the Restatement
(Third), a product ‘‘is defective in design when the fore-
seeable risks of harm posed by the product could have
been reduced or avoided by the adoption of a reason-
able alternative design by the seller . . . and the omis-
sion of the alternative design renders the product not
reasonably safe . . . .’’ Restatement (Third), supra, § 2
(b), p. 14. To guide its analysis, the Restatement (Third)
provides the jury with a number of factors to weigh in
determining whether, in light of these factors, adopting
a safer design was possible without greatly increasing
the product’s costs or risks or greatly diminishing its
usefulness. See id., § 2, comment (f), p. 23. These factors
include (1) the likelihood and magnitude of foreseeable
risks of harm posed by the product’s design, (2) the
instructions and warnings given with the product, (3)
consumer expectations about the product and its usage,
(4) the safety risks and benefits of alternative designs,
and (5) the feasibility of adopting an alternative design,
including effects on the product’s cost, functionality,
longevity and appearance. See id. Because the relevance
and importance of each factor will vary in each case
depending on the nature of the evidence, the plaintiff
is not required to present evidence regarding every fac-
tor to establish his case. See id. If the jury determines
that the manufacturer could reasonably have adopted
the safer alternative, the manufacturer’s design may be
deemed not reasonably safe, and thus defective. See
id., § 2 (b), p. 14.
Notably, unlike the ordinary consumer expectations
test from § 402A of the Restatement (Second) of Torts,
the risk-utility test does not treat consumer expecta-
tions as dispositive but as one factor among many for
the jury to weigh. See Potter v. Chicago Pneumatic
Tool Co., supra, 241 Conn. 213 and n.10. In making
consumer expectations a nondispositive factor, the
risk-utility test avoids many of the bars to liability asso-
ciated with the ordinary consumer expectations test—
especially those relating to open and obvious dangers
and injuries to foreseeable but unintended users and
bystanders. See, e.g., Restatement (Third), supra, § 2,
comment (g), pp. 27–28. Thus, a jury may still find a
manufacturer liable for obvious product dangers if it
finds that the risks posed by a product’s design could
be mitigated by adopting a reasonable alternative. See
1 D. Owen & M. Davis, supra, § 8:10, pp. 739–41; see
also Restatement (Third), supra, § 2, comment (d), p. 20.
Moreover, as I will discuss, the Restatement (Third)
does not require expert testimony to establish proof
of a reasonable alternative design and recognizes that
proof of an alternative design is unnecessary in some
limited circumstances. See Restatement (Third), supra,
§ 2, comment (f), pp. 23–24.
B
Need for Reasonable Alternative Design Evidence
The Restatement (Third) standard is consistent with
modern design defect jurisprudence, which recognizes
that design defect claims are best decided under a risk-
utility standard using proof of a reasonable alternative
design, subject to appropriate exceptions.
The need for proof of an alternative design to estab-
lish defectiveness in a design case arises from the
unique considerations presented by these types of
claims. In any product defect case, a jury needs an
objective basis against which to compare the product
at issue to determine whether the product was defec-
tive. See, e.g., T. Jankowski, supra, 36 S. Tex. L. Rev.
292. In manufacturing defect cases, the objective basis
for comparison is inherent in the nature of the claim:
the plaintiff alleges that the individual unit he received
was not manufactured according to its intended design
and that this deviation caused harm. See 1 D. Owen &
M. Davis, supra, § 7:1, pp. 651–52. To determine whether
the unit at issue was in fact defective, a jury need only
compare the plaintiff’s unit against the intended design
to determine whether the two are different.
A design defect case lacks a similar inherent objective
basis for comparison. In cases involving design defect
claims, the plaintiff’s challenge does not concern the
individual unit he purchased but the product’s specifica-
tions. See id., § 8:1, p. 708. In other words, a design
defect claim alleges that, although a product may have
been manufactured properly according to its design,
the intended design chosen by the manufacturer was
not reasonably safe. See id. (‘‘unlike a manufacturing
defect claim, which implicates merely a single product
unit, a design defect claim challenges the integrity of
the entire product line and so pierces to the very core
of the manufacturer’s enterprise’’). Any judgment that
a product design is defective, therefore, ‘‘condemns the
entire product line’’ and not just the unit that the plain-
tiff purchased. Id. Because a design claim calls the
design itself into question, the jury needs some objec-
tive basis other than the specifications against which
to compare the design at issue in determining whether
it was not reasonably safe and thus defective. See
Restatement (Third), supra, § 1, comment (a), p. 7
(‘‘when the product unit meets the manufacturer’s own
design specifications, it is necessary to go outside those
specifications to determine whether the product is
defective’’).
The lack of an inherent objective basis for compari-
son in design cases has made formulating a proper
standard for design defect claims a difficult task for
courts. See, e.g., 1 D. Owen & M. Davis, supra, § 8:1, p.
702 (‘‘[e]lusive as an elf, the true meaning of ‘design
defect’ largely escaped capture by court or commenta-
tor until quite recently, and the search therefor has led
inexorably to consternation and confusion’’); see also
3A American Law of Products Liability (3d Ed. 2007)
§ 28:5, p. 15 (noting that courts have struggled with
standard in design defect cases because such cases do
not lend themselves to ‘‘readily ascertainable’’ objec-
tive standard).
Following the adoption of § 402A of the Restatement
(Second) of Torts, courts attempted to apply its con-
sumer expectations standards to design defect claims.
See 1 D. Owen & M. Davis, supra, § 8:3, pp. 713–14.
This entails asking a jury whether the product’s design
met the expectations of the product’s ordinary consum-
ers. See 2 Restatement (Second), Torts § 402A, com-
ment (i), p. 352 (1965). If the product falls short of those
expectations, it may be deemed defective. See id. The
consumer expectations test was created, however, with
manufacturing defects in mind. A. Twerski & J. Hender-
son, supra, 74 Brook. L. Rev. 1063. For example, a prod-
uct unit that was made differently from its intended
design because of a mistake in the manufacturing pro-
cess can be understood to disappoint the expectations
of its consumers. See id., 1064, 1067. With respect to
manufacturing claims, the intended or expected design
of the product provides an objective basis for determin-
ing the expectations of consumers. 1 D. Owen & M.
Davis, supra, § 7:2, pp. 653–54.
As the majority observes in its opinion, however,
the consumer expectations test proved unsuitable for
resolving many types of design defect claims because
that standard was too vague to supply an objective basis
for assessing product designs. See id., § 5:16, p. 448;
id., § 8:5, pp. 720–21; see also A. Twerski & J. Henderson,
supra, 74 Brook. L. Rev. 1067 (explaining that consumer
expectations test has been widely rejected in design
cases ‘‘as unworkable and unwise’’). Consumers often
have little or no knowledge about how safe a product
design should be and whether it could be made safer.
1 D. Owen & M. Davis, supra, § 5:16, p. 448 (‘‘consumers
often have no meaningful idea how safely the product
really ought to perform in various situations’’). This is
especially true for products with complex designs and
those that fail in complex ways. See, e.g., Pruitt v.
General Motors Corp., 72 Cal. App. 4th 1480, 1483, 86
Cal. Rptr. 2d 4 (1999) (‘‘[t]he deployment of an air bag
is, quite fortunately, not part of the everyday experience
of the consuming public’’ [internal quotation marks
omitted]); R. Dickerson, ‘‘Products Liability: How Good
Does a Product Have To Be?,’’ 42 Ind. L.J. 301, 307
(1967) (‘‘What, for instance, should the law do about
tractors that overturn, surgical implants that break, and
rear-engined automobiles that tend to swerve at high
speeds?’’). Similar problems arise with new products.
See R. Dickerson, supra, 307 (‘‘[t]he most troublesome
situations are those in which consumer attitudes have
not sufficiently crystallized to define an expected stan-
dard of performance’’). Moreover, expectations often
vary between different consumers of the same product,
and consumers may have expectations about safety that
are beyond what is feasible for manufacturers to meet.
See, e.g., D. Fischer, ‘‘Products Liability—The Meaning
Of Defect,’’ 39 Mo. L. Rev. 339, 349–50 (1974) (‘‘[e]xpec-
tations as to safety will not always be in line with what
the reasonable manufacturer can achieve because the
average consumer will not have the same information
as experts in the field’’). As a result, design defect tests
based on consumer expectations often leave a jury with
little meaningful guidance when it considers whether a
product design is defective, and may lead it to condemn
entire product lines without any true understanding of
the product’s risks and benefits and whether the prod-
uct could be made safer without eliminating its utility.
See A. Twerski & J. Henderson, supra, 1066–67.2
The inherent limitations of the ordinary consumer
expectations test have led courts and commentators to
search for a different standard for design defect cases.
Many courts have abandoned the consumer expecta-
tions test entirely for design defect claims, whereas
some courts have restricted it, as the majority does
today, to a small category of cases in which the exis-
tence of a design defect is more obvious.3 See 1 D.
Owen & M. Davis, supra, § 5:17, p. 450; see also T.
Jankowski, supra, 36 S. Tex. L. Rev. 326; A. Twerski &
J. Henderson, supra, 74 Brook. L. Rev. 1101. Courts
have instead looked to the risk-utility standard as a
better standard for resolving design defect claims. 1 D.
Owen & M. Davis, supra, § 8:6, pp. 722–25 and nn. 1–6
(collecting authorities).
As more and more jurisdictions have embraced the
risk-utility test in the decades after the adoption of
§ 402A, a consensus has emerged that design defect
claims are best resolved by using risk-utility balancing
to compare the manufacturer’s chosen design against
safer alternatives to determine whether it was feasible
for the manufacturer to have created a safer product.
See, e.g., A. Twerski & J. Henderson, supra, 74 Brook.
L. Rev. 1094 (‘‘[r]easonable alternative design is the
answer to the comparative balancing process’’); see also
1 D. Owen & M. Davis, supra, § 8:10, pp. 740–41 (‘‘[a]s
modern products liability developed after the promulga-
tion of . . . § 402A [of the Restatement (Second) of
Torts], courts and commentators alike increasingly rec-
ognized the logical and practical necessity in most types
of design defect cases that plaintiffs prove that their
harm would have been prevented if the manufacturer
had adopted some alternative design’’); annot., ‘‘Burden
of Proving Feasibility of Alternative Safe Design in Prod-
uct Liability Action Based on Defective Design,’’ 78
A.L.R. 4th 154, 157 (1990) (‘‘The reasonableness of
choosing from among various alternative product
designs and adopting the safest one if it is feasible is
not only relevant in a design defect action, but is at the
very heart of the case. The essential inquiry is whether
the design chosen was a reasonable one from among
the feasible choices of which the defendant was aware
or should have been aware. This feasibility is a relative,
rather than an absolute, concept; the more scientifically
and economically feasible the alternative is, the more
likely it is that the product will be found to be defec-
tively designed.’’).
This approach of comparing the merits of a product’s
design against possible alternatives recognizes that a
jury cannot meaningfully assess whether a product
design is defective without knowing what design alter-
natives are available, and the risks, benefits, and costs
associated with adopting an alternative design. As one
commentator has explained: ‘‘At the center of a rational
process for evaluating design . . . decisions is the
requirement of a reasonable alternative proposed by
the claimant. This requirement is both eminently fair
and necessary. If manufacturer decisions based on com-
plex tradeoffs are being challenged as wrong, it is neces-
sary to understand the alternative decision proposed
[that] is being advanced as right.’’ (Footnote omitted;
internal quotation marks omitted.) T. Jankowski, supra,
36 S. Tex. L. Rev. 292. Notions of design safety are not
absolute, and no product design can ever be entirely
accident proof, and, thus, the defectiveness of a manu-
facturer’s chosen design depends largely on whether it
could have been made safer by the adoption of some
alternative design feature. See D. Owen, ‘‘Defectiveness
Restated: Exploding the ‘Strict’ Products Liability
Myth,’’ 1996 U. Ill. L. Rev. 743, 754–55. After all, it is
generally not unreasonable for a manufacturer to mar-
ket a product with adequate warnings that serves a
useful purpose and cannot feasibly be made any safer.
See 1 D. Owen & M. Davis, supra, § 8:10, p. 741 (‘‘Without
affirmative proof of a feasible design alternative, a plain-
tiff usually cannot establish that a product’s design is
defective. Put otherwise, there typically is nothing
wrong with a product that simply possesses inherent
dangers that cannot feasibly be designed away.’’); J.
Phillips, ‘‘The Standard for Determining Defectiveness
in Products Liability,’’ 46 U. Cin. L. Rev. 101, 104 n.18
(1977) (‘‘a manufacturer’s product can hardly be faulted
if safer alternatives are not feasible’’ [internal quotation
marks omitted]). Moreover, given the significant conse-
quences at stake when a design defect claim is
asserted—the condemnation of an entire line of prod-
ucts—it is only fair that some safer alternative be pro-
posed before allowing a jury to declare a product design
defective. See 1 D. Owen & M. Davis, supra, § 8:10, p.
741. When, however, it is established that the manufac-
turer could reasonably have adopted a safer design, it
is fair to hold a manufacturer responsible for failing to
adopt it. Cf. id., § 8:12, p. 754.
C
Concerns with Potter’s Modified Standard
and Its Rejection of a Reasonable
Alternative Design Requirement
In creating the modified consumer expectations stan-
dard, Potter replaced our reliance on the ordinary con-
sumer expectations standard from comment (i) to
§ 402A of the Restatement (Second) of Torts with a
similarly problematic standard. Just as with the con-
sumer expectations test, Potter’s modified standard
also fails to provide jurors with an objective basis for
judging a product’s design. Potter created the modified
consumer expectations standard by incorporating risk-
utility factors into the ordinary consumer expectations
analysis, but without any requirement of a reasonable
alternative design. See Potter v. Chicago Pneumatic
Tool Co., supra, 241 Conn. 221. As the majority discusses
in its opinion, the court in Potter declined to adopt a
draft form of the Restatement (Third) because it inter-
preted the draft to require proof of a reasonable alterna-
tive design in all cases and without exception. See id.,
214–19, 221. The court in Potter feared that adopting
such a requirement would harm plaintiffs by creating
too heavy of an evidentiary burden. Id., 217–19. As a
result, the court emphasized, when it created the modi-
fied standard, that the availability of a reasonable alter-
native design was only one factor for the jury to con-
sider rather than a requirement in every case. Id., 221.
Without this requirement, however, Potter’s modified
standard does no better than the ordinary consumer
expectations test in providing the jury with an objective
basis against which to assess a product’s design.
Standards relying on some form of risk-utility balanc-
ing without an accompanying requirement of a reason-
able alternative design have proven problematic, both
in theory and in practice. These standards are not truly
risk-utility standards. The risk-utility test and the rea-
sonable alternative design requirement go hand in hand
because a proposed alternative design is necessary to
provide an objective basis for comparison against the
manufacturer’s chosen design. The risk-utility test itself
does not supply the basis for comparison; rather, it
provides only the considerations that guide the compar-
ison. As one commentator has succinctly explained,
‘‘one simply cannot talk meaningfully about a risk-bene-
fit defect in a product design until and unless one has
identified some design alternative (including any design
omission) that can serve as the basis for a risk-benefit
analysis.’’ G. Schwartz, ‘‘Foreword: Understanding
Products Liability,’’ 67 Cal. L. Rev. 435, 468 (1979). Other
commentators agree. See, e.g., 1 D. Owen & M. Davis,
supra, § 8:10, p. 739 (‘‘cost-benefit analysis of an alterna-
tive design lies at the heart of design defectiveness’’);
T. Jankowski, supra, 36 S. Tex. L. Rev. 292 (explaining
that reasonable alternative design requirement ‘‘is a sine
qua non of the risk-utility process’’ [emphasis omitted]);
T. Jankowski, supra, 326 (‘‘the gravitational pull in
design defect cases has been toward the risk-utility
balance and its concomitant, the reasonable alternative
design’’); A. Twerski & J. Henderson, supra, 74 Brook.
L. Rev. 1094 (‘‘[w]hen one does risk-utility balancing
one must judge the product on trial and compare it
with some hypothetical design that could have been
adopted’’).
The risk-utility test, which traces its roots to the
famed Carroll Towing decision; United States v. Carroll
Towing Co., 159 F.2d 169 (2d Cir. 1947); is predicated
entirely on the notion that some alternative measure
could have been taken to avoid the plaintiff’s harm, and
the test was developed as a tool for comparing the
allegedly defectively designed product to its alterna-
tives. See, e.g., T. Jankowski, supra, 36 S. Tex. L. Rev.
319 (‘‘[t]he key observation to be made is that the risk-
utility test, in order to evaluate the appropriateness of
the [design] at issue . . . requires some standard . . .
for comparison’’); A. Twerski & J. Henderson, supra,
74 Brook. L. Rev. 1094 (‘‘When one does risk-utility
balancing one must judge the product on trial and com-
pare it with some hypothetical design that could have
been adopted. Reasonable alternative design is the
answer to the comparative balancing process; it is not
a factor in the equation as to whether the product was
reasonably designed.’’).
A risk-utility analysis without a reasonable alternative
design lacks an objective basis for comparison, leaving
the jury with only vague guidance about whether a
product design is defective. Without a proposed alterna-
tive, the jury is left to compare the product’s own risks
against its own benefits, which essentially is like asking
the jury to imagine a world with the product and without
the product, and to decide which is preferable. M.
Green, ‘‘The Schizophrenia of Risk-Benefit Analysis in
Design Defect Litigation,’’ 48 Vand. L. Rev. 609, 617 n.38
(1995). This puts the jury in the position of having to
decide not whether the product could have been made
safer, but whether a particular product should have
been sold at all—commonly referred to as absolute or
category liability, a concept courts have been hesitant
to embrace, even in strict liability cases. See A. Twer-
ski & J. Henderson, supra, 74 Brook. L. Rev. 1069
(‘‘American courts have never imposed category liabil-
ity, mainly because they intuitively . . . and correctly
. . . understand that it would constitute an abuse of
judicial power to decide which broad categories of
products should not be distributed at all’’). Imposing
liability for a product, despite the absence of reasonable
alternatives, could deprive consumers of an otherwise
useful product if the risk of adverse verdicts prompts
the manufacturer either to cease production or to signif-
icantly increase the cost of the product, rendering it
prohibitively expensive for some consumers. Moreover,
allowing juries to hold manufacturers liable even if the
product serves some useful purpose and cannot reason-
ably be made any safer risks turning manufacturers into
insurers of their products. See, e.g., T. Jankowski, supra,
36 S. Tex. L. Rev. 324 (‘‘Any logical treatment must
recognize that a manufacturer’s [design] decision can
only be ‘wrong’ in the context of ‘right’ alternatives
that were available. . . . Without this requirement, the
manufacturer becomes an insurer of the product.’’
[Footnote omitted.]). Even the court in Potter acknowl-
edged that this kind of absolute liability is antithetical
to our product liability laws. See Potter v. Chicago
Pneumatic Tool Co., supra, 241 Conn. 210 (‘‘strict tort
liability does not transform manufacturers into insur-
ers, nor does it impose absolute liability’’); see also
Metropolitan Property & Casualty Ins. Co. v. Deere &
Co., 302 Conn. 123, 137, 25 A.3d 571 (2011) (liability
standards that essentially convert manufacturers into
insurers of their products would be ‘‘contrary to the
purposes of our product liability laws’’).
Perhaps because of these theoretical shortcomings,
jurisdictions that purport to reject a reasonable alterna-
tive design requirement nevertheless appear to require
this proof as a practical matter. See, e.g., A. Twerski &
J. Henderson, supra, 74 Brook. L. Rev. 1094–95. For
example, commentators have noted that, despite Pot-
ter’s rejection of a requirement that the plaintiff estab-
lish a reasonable alternative design, courts applying the
standard established in Potter have required this proof
in practice. See id., 1068, 1102. In researching product
liability cases brought under Connecticut law, these
commentators discovered that, at least as of 2009, there
were no reported cases involving traditional design
defect claims since Potter that have been submitted to
a jury without proof of a reasonable alternative design.
See id. Ironically, even the plaintiff in Potter had pre-
sented extensive evidence of design alternatives. See
Potter v. Chicago Pneumatic Tool Co., supra, 241 Conn.
204–206. So did the plaintiff, Barbara A. Izzarelli, in the
present case.
D
Exceptions to Reasonable Alternative
Design Requirement
Courts that have rejected a reasonable alternative
design requirement typically do so out of fear of bur-
dening plaintiffs by placing too many evidentiary hur-
dles along their path to recovery. See, e.g., id., 217–19;
see also 1 D. Owen & M. Davis, supra, 8:10, p. 745.
Potter specifically noted two areas of concern in this
regard. First, the court was concerned that it would
require expert testimony in every case, including in res
ipsa-like cases in which the jury can infer the existence
of a defect from circumstantial evidence. See Potter v.
Chicago Pneumatic Tool Co., supra, 241 Conn. 217–18.
Second, the court observed that some product designs
could be considered unreasonably dangerous, even if
no reasonable alternative design existed. Id., 219.
The Restatement (Third) resolves these concerns,
however. First, with respect to the concerns about
requiring expert testimony, the comments to the
Restatement (Third) explain that expert testimony is
not required to meet the alternative design requirement
in every case. Restatement (Third), supra, § 2, comment
(f), p. 23. The Restatement (Third) does not require
plaintiffs to propose or build an entire new prototype
of the product—the plaintiff need only show that the
manufacturer could reasonably have designed a safer
alternative. Id., p. 24. In many instances, a plaintiff can
accomplish this without expert testimony. See id., p.
23. For example, no expert testimony is needed when
the plaintiff can show that competing products on the
market would be safer or when the availability of a
safer design is obvious to a layperson.4 Id.
Second, the Restatement (Third) also expressly rec-
ognizes several exceptions to its alternative design
requirement. Although, as I discussed previously, courts
are justifiably hesitant to impose liability on manufac-
turers when no safer alternative is available, the
Restatement (Third) recognizes that there are circum-
stances when some consideration other than a design
alternative provides a sufficient and fair basis for impos-
ing liability. In each of these instances, a test other than
risk-utility balancing is used to determine liability.
First, no such evidence is needed if the product
design violates a statute or a regulation. See id., § 4 (a),
p. 120. In these cases, proof that the design violates
existing law alone is a sufficient consideration to
impose liability because manufacturers should not sell
products that legislatures or regulatory authorities have
decided to ban. See id.
Second, a plaintiff need not proffer alternative design
evidence when the product design at issue is manifestly
unreasonable. See id., § 2, comment (e), pp. 21–22. The
Restatement (Third) acknowledges that, in rare and
extreme cases, a product design may be so obviously
unacceptable that a manufacturer can fairly be held
liable for harm even if no safer alternative is feasible.
Id. In these limited instances, a jury may ‘‘conclude that
liability should attach without proof of a reasonable
alternative design’’ when ‘‘the extremely high degree of
danger posed by [a product’s] use or consumption so
substantially outweighs its negligible social utility that
no rational, reasonable person, fully aware of the rele-
vant facts, would choose to use, or to allow children
to use, the product.’’5 Id., p. 22.
Finally, the Restatement (Third) also does not require
proof of design alternatives in res ipsa-like cases, in
which the very circumstances of a product’s failure
provide strong evidence that it was defective; for these
types of cases, the Restatement (Third) does not require
direct evidence of a specific defect. See id., § 3, p. 111.
Instead, it relies on the malfunction theory, which
allows a jury to infer the existence of some product
defect from the nature of the product’s failure, together
with evidence showing that its failure was not caused
by something other than a defect. See id., § 3, comment
(b), p. 112. Because a plaintiff need not identify a spe-
cific defect in the product, no alternative design evi-
dence is needed.6 See id. Consider, for example, a new
television that catches fire in a living room during nor-
mal use. The plaintiff need not prove that the manufac-
turer should have adopted a safer design. Liability is
instead predicated on the notion that, in the absence
of other possible causes, televisions do not ordinarily
catch fire during normal use in the absence of some
product defect. See, e.g., Liberty Mutual Ins. Co. v.
Sears, Roebuck & Co., 35 Conn. Supp. 687, 691, 406 A.2d
1254 (1979). As the majority notes, we have already
adopted the malfunction theory from the Restatement
(Third). See Metropolitan Property & Casualty Ins. Co.
v. Deere & Co., supra, 302 Conn. 137–39.
These exceptions address each of Potter’s stated con-
cerns about requiring alternative design evidence. Nota-
bly, the reporters’ note to the Restatement (Third)
expressly compares Potter’s concerns about the reason-
able alternative design requirement with the exceptions
adopted in the Restatement (Third): ‘‘The Connecticut
Supreme Court’s analysis in Potter is, in actuality, per-
fectly consistent with this Restatement,’’ and it is rec-
ommended that, ‘‘when the issue is next before [that]
court, [it] may find it easier to accept the Restatement
as consistent with its position as articulated in Potter.
Whatever ambiguities in the earlier draft may have mis-
led the court in this regard, those ambiguities have
since been eliminated.’’ Restatement (Third), supra, § 2,
reporters’ note to comment (d), pp. 72–73.
II
THE RESTATEMENT (THIRD)’S FUNCTIONAL
APPROACH TO DESIGN DEFECT CLAIMS
There are additional considerations that favor adop-
tion of the Restatement (Third) for design defect cases.
In adopting the risk-utility test, the Restatement (Third)
defines its standard ‘‘functionally’’ by focusing on the
unique considerations at issue in design defect cases,
rather than relying on traditional liability doctrines like
strict liability, negligence, contract, warranty, etc. Id.,
§ 2, comment (n), p. 35. This function based approach
is in keeping with the modern consensus that different
types of product defect cases—manufacturing defect,
design defect, marketing defect—each present issues
for juries to consider and thus require tests tailored to
the type of defect alleged. The older approach of defin-
ing product defect standards set forth in the
Restatement (Second) of Torts, which used a one-size-
fits-all strict liability test (the consumer expectations
standard), proved difficult to apply in many product
defect cases. Courts and commentators have since
turned to defining product liability standards based on
the type of defect alleged, without resort to traditional
tort liability doctrines. See 1 D. Owen & M. Davis, supra,
§ 8:1, p. 707 (‘‘[The consumer expectations standard of
the Restatement (Second) reflected a] quest by courts
for a general definition of ‘defectiveness,’ commonly
viewed in early products liability as embracing a single
principle applicable to any type of case. As products
liability law has matured, however, most courts and
commentators have come to understand that meaning-
ful evaluation of the acceptability of a product’s dangers
logically turns on considerations that vary contextually
depending [on] whether the problem was one of manu-
facture, design, or the absence of sufficient warnings.’’
[Footnotes omitted.]).
The Restatement (Third)’s functional approach to
design defect cases provides a number of benefits. First,
by defining its design defect standard in terms of the
unique considerations involved in design defect cases,
rather than by resorting to traditional doctrinal liability
theories, its risk-utility standard blends beneficial
aspects of strict liability and negligence theories with-
out their accompanying drawbacks. Second, relying on
a single, unified standard for design defect claims
improves clarity by avoiding the confusion and risk of
inconsistent verdicts that could result from submitting
a claim to a jury under multiple tests and theories (e.g.,
under the ordinary consumer expectations test, the
modified test, and a negligent design theory). Third,
adopting a unified standard is consistent with our Prod-
uct Liability Act, General Statutes § 52-572m et seq.,
which was intended to simplify pleadings and product
liability claims under a single cause of action. I now
address each consideration in greater detail.
A
Blending Strict Liability and Negligence
Consistent with the modern approach to design
defect claims, the Restatement (Third) recognizes that
the risk-utility test is neither a strict liability nor a negli-
gence standard, but reflects a blend of the two, and
thus displaces those theories in design defect cases. For
example, it resembles a negligence balancing standard
inasmuch as it requires a jury to balance foreseeable
risks of harm against the costs of adopting safer, alter-
native measures. See 1 D. Owen & M. Davis, supra,
§ 5:36, p. 501 (noting that risk-utility test is ‘‘based on
principles of foreseeability and balance that underlie
the law of negligence’’). At the same time, the risk-
utility test embraces strict liability principles because
a manufacturer cannot defend itself on the ground that
it used reasonable care in selecting its chosen design
or that its design is consistent with others used in the
industry; as long as the plaintiff demonstrates that the
manufacturer could reasonably have adopted a safer
alternative, a jury can find liability without regard to
the level of care that the manufacturer exercised in
selecting its design. See id., § 5:29, p. 476 (noting that
strict liability principles permit liability even if manufac-
turer used reasonable care in making product).
Some courts, including this court in Potter, have
claimed that the introduction of risk-utility factors into
design defect jurisprudence should not be construed
as a departure from strict liability principles, and that
the focus of the jury’s inquiry must remain on the prod-
uct, not on the manufacturer’s conduct. See, e.g., Potter
v. Chicago Pneumatic Tool Co., supra, 241 Conn. 221–
22. This is an artificial distinction. See 1 D. Owen & M.
Davis, supra, § 5:29, p. 480 (noting that, with respect
to design defect cases, there is no practical distinction
between strict liability and negligence tests, although
‘‘there remains a dwindling, yet stubborn, contingent
of courts that cling tenaciously to the view that the
doctrines of negligence and strict liability in tort are
and must be kept conceptually distinct’’). One court
explained the fiction as follows: ‘‘Although many courts
have insisted that the risk-utility tests they are applying
are not negligence tests because their focus is on the
product rather than the manufacturer’s conduct . . .
the distinction on closer examination appears to be
nothing more than semantic. As a common-sense mat-
ter, [under the risk-utility test] the jury weighs compet-
ing factors presented in evidence and reaches a
conclusion about the judgment or decision (i.e., con-
duct) of the manufacturer. The underlying negligence
calculus is inescapable.’’ (Citation omitted; emphasis
in original.) Prentis v. Yale Mfg. Co., 421 Mich. 670,
687–88, 365 N.W.2d 176 (1984); see also S. Birnbaum,
‘‘Unmasking the Test for Design Defect: From Negli-
gence [to Warranty] to Strict Liability to Negligence,’’
33 Vand. L. Rev. 593, 610 (1980) (‘‘When a jury decides
that the risk of harm outweighs the utility of a particular
design [such that the product is not as safe as it should
be], it is saying that in choosing the particular design
and cost trade-offs, the manufacturer exposed the con-
sumer to [a] greater risk of danger than [it] should have.
Conceptually and analytically, this approach
bespeaks negligence.’’).
Because the risk-utility analysis resembles a blend
of both strict liability and negligence principles, the
Restatement (Third) does not recognize separate negli-
gence and strict liability tests and uses only the risk-
utility test as the proper test for all design defect cases.
See Restatement (Third), supra, § 2, comment (n), p.
35. Thus, a jury should not receive both a risk-utility
and a negligence instruction; only the risk-utility test
may be submitted to a jury in cases involving a design
defect claim.
The Restatement (Third)’s functional approach com-
bines beneficial aspects of strict liability and negligence
theories without their accompanying drawbacks. For
example, under the Restatement (Second)’s strict liabil-
ity test, the ordinary consumer expectations test, a
plaintiff can be barred from recovering if his harm was
caused by a danger open and obvious to the ordinary
consumer, even if the manufacturer could have pre-
vented the danger with a reasonable design modifica-
tion. See 1 D. Owen & M. Davis, supra, § 8:5, pp. 718–19.
The Restatement (Third) eliminates this impediment
and, instead, makes the obviousness of a product’s dan-
ger only one factor for a jury to consider, thereby remov-
ing a potential bar to recovery while still allowing the
jury to consider evidence on this issue. See Restatement
(Third), supra, § 2, comment (g), pp. 27–28.
The Restatement (Third) also avoids trappings often
associated with negligence and even contract based
theories of recovery, including the requirement that a
plaintiff show duty or privity as a prerequisite to recov-
ery. Much of the purpose for moving to strict liability
in the first place was to avoid these requirements, which
were used by defendants to block recovery in some
instances. See id., § 1, comment (a), p. 6. Similar to the
strict liability test, the Restatement (Third) expressly
omits any privity or duty requirement; a plaintiff need
only show that his harm was caused by a defect in the
defendant’s product to have standing to recover, even
if the plaintiff was not a purchaser or a user of the
product. Id., § 1, p. 5; id., § 2, p. 14.7
Moreover, although the Restatement (Third)’s risk-
utility test displaces negligence tests in cases involving
design defect claims, it does not prevent plaintiffs from
introducing evidence relating to fault when that evi-
dence is relevant to the risk-utility calculus. The
Restatement (Third) explains: ‘‘In connection with a
claim under §§ 1 and 2 and related provisions of this
Restatement, the evidence that the defendant did or did
not conduct adequately reasonable research or testing
before marketing the product may be admissible (but
is not necessarily required) regardless of whether the
claim is based on negligence, strict liability, or implied
warranty of merchantability. Although a defendant is
held objectively responsible for having knowledge that
a reasonable seller would have had, the fact that the
defendant engaged in substantial research and testing
may help to support the contention that a risk was not
reasonably foreseeable. Conversely, the fact that the
defendant engaged in little or no research or testing
may, depending on the circumstances, help to support
the contention that, had reasonable research or testing
been performed, the risk could have been foreseen.
Moreover, as long as the requisites in [the risk-utility
test] . . . are met, the plaintiff may in appropriate
instances—for example, in connection with compara-
tive fault or punitive damage claims—show that the
defect resulted from reckless, [wilfully] indifferent, or
intentionally wrongful conduct of the defendant.’’ Id.,
§ 2, comment (n), p. 35.
Finally, it is also important to emphasize that the
Restatement (Third)’s risk-utility test displaces other,
traditional standards of liability only when the plaintiff
seeks recovery for harm caused by a design defect
existing at the time of sale; the risk-utility test does
not apply to design related claims involving the manu-
facturer’s conduct after the sale. See id., p. 37. Thus,
for example, although only the risk-utility test would
apply in a case alleging that an airbag design was defec-
tive when it was sold to the plaintiff, the risk-utility
test would not apply to a separate claim alleging that
the manufacturer should have issued a recall of the
airbag when it learned that its design was unreasonably
causing harm. For that type of claim, the Restatement
(Third) acknowledges that negligence could remain an
appropriate standard. See id.
B
Avoiding Inconsistent Verdicts
This simplified approach of using a single test for all
design defect claims also serves an important practical
purpose: to avoid the confusion and inconsistent ver-
dicts that could result from submitting two separate
standards to a jury to determine the existence of a
single defect. For example, suppose a court submits a
design defect case to the jury and gives both a risk-
utility and a negligent design instruction, and the jury
finds for the defendant on the risk-utility theory and
for the plaintiff on the negligence theory. The two ver-
dicts are logically inconsistent. If the jury decides that
no design defect existed at the time of sale under the
risk-utility test, then the manufacturer should not be
deemed negligent for selling a product that is not defec-
tive.8 See 1 D. Owen & M. Davis, supra, § 5:29, pp.
481–83. Courts and commentators offer varying expla-
nations for how a jury could reach such inconsistent
conclusions. Most explanations involve an acknowledg-
ment that the jury would most likely have been confused
in using two standards to decide essentially the same
question. See id., p. 483. Worse still, an inconsistent
verdict could be the result of a compromise based on
considerations other than the jury’s proper application
of the law to the facts. See id. Sound product liability
law should be structured to avoid such results. See id.
(‘‘[w]hatever the reason, such findings logically make
no sense, are offensive to sound jurisprudence, and
ordinarily should not be tolerated’’).
Mindful of this concern, the Restatement (Third)
emphasizes that courts should instruct the jury in a
design defect case only on the risk-utility test, regard-
less of the label a court applies to it. The Restatement
(Third) explains that ‘‘two or more factually identical
[defective design] claims . . . should not be submitted
to the trier of fact in the same case under different
doctrinal labels. Regardless of the doctrinal label
attached to a particular claim, design . . . claims rest
on a risk-utility assessment. To allow two or more factu-
ally identical risk-utility claims to go to a jury under
different labels, whether ‘strict liability,’ ‘negligence,’ or
‘implied warranty of merchantability,’ would generate
confusion and may well result in inconsistent verdicts.’’
Restatement (Third), supra, § 2, comment (n), pp.
35–36.9
Formulating liability tests based on the type of defect
alleged rather than trying to frame them within tradi-
tional doctrinal categories thus improves the clarity
and predictability of product liability law and thereby
reduces confusion. See 1 D. Owen & M. Davis, Products
Liability (4th Ed. Supp. 2015) § 5:38, p. 15; see also
Restatement (Third), supra, § 2, comment (n), pp.
35–36.
C
Product Liability Act
Adopting the Restatement (Third) approach would
be fully consistent with—and help to fulfill—the pur-
pose of Connecticut’s Product Liability Act (act), Gen-
eral Statutes § 52-572m et seq., which was intended to
simplify product liability actions by requiring a plaintiff
to bring all claims against product sellers for product
related harm within a single statutory cause of action.
See General Statutes §§ 52-572m (b) and 52-572n (a).
Prior to the act, product liability claims could be
brought under numerous, separate causes of action,
each invoking different theories of liability (e.g., negli-
gence, breach of contract, strict liability, and breach of
warranty). Each was subject to different statutes of
limitations and defenses. To eliminate this patchwork
of claims and various pleading requirements, the legisla-
ture created a single statutory cause of action, subject
to one set of limitations and defenses. This cause of
action encompassed all types of claims against product
sellers, irrespective of the underlying theory. See Lynn
v. Haybuster Mfg., Inc., 226 Conn. 282, 292, 627 A.2d
1288 (1993) (‘‘The intent of the legislature was to elimi-
nate the complex pleading provided at common law:
breach of warranty, strict liability and negligence. . . .
[T]he act was intended to merge various theories into
one cause of action rather than to abolish all prior
existing rights.’’ [Citations omitted.]). Thus, according
to the act, ‘‘[a] product liability claim . . . shall be in
lieu of all other claims against product sellers, including
actions of negligence, strict liability and warranty, for
harm caused by a product.’’ General Statutes § 52-572n
(a). The act defines a ‘‘product liability claim’’ to include
‘‘all actions based on the following theories: Strict liabil-
ity in tort; negligence; breach of warranty, express or
implied; breach of or failure to discharge a duty to warn
or instruct, whether negligent or innocent; misrepresen-
tation or nondisclosure, whether negligent or innocent.’’
General Statutes § 52-527m (b).
Although the legislature aggregated existing product
liability theories under a single cause of action, it did
not provide any substantive elements to decide liability,
with the exception of claims based on inadequate warn-
ings, which are not at issue in the present case. See
General Statutes § 52-572q (b). Instead, the legislature
relied on existing common law to provide those stan-
dards and left their further development to the courts.
See, e.g., Potter v. Chicago Pneumatic Tool Co., supra,
241 Conn. 229–30, 245–46 n.34 (refining design defect
standards after adoption of act). Our current law, under
Potter, allows plaintiffs to plead multiple theories of
recovery for a single alleged design defect, as long as
they do so under the heading of a single ‘‘product liabil-
ity’’ cause of action. Thus, a plaintiff seeking to recover
for a design defect can presently bring a claim premised
on many different theories, including for strict liability
under the modified consumer expectations test and the
ordinary consumer expectations test, and for negligent
design under standard principles of negligence. Using
multiple tests to address the same essential question
sows confusion.
Consistent with the act’s purpose of simplification,
adopting the Restatement (Third) standard would
streamline design defect claims. Using a single standard
tailored specifically to design defect claims would do
away with the need to plead or prove separate strict
liability and negligence theories and avoids the confus-
ing use of multiple theories to address the same underly-
ing issue—whether the manufacturer chose a reason-
ably safe product design.
D
Rejection of the Restatement (Third)
in Other Jurisdictions
I recognize that some other jurisdictions have also
considered and rejected the Restatement (Third)’s
design defect standard. See, e.g., Aubin v. Union Car-
bide Corp., 177 So. 3d 489, 510–12 (Fla. 2015); Tincher
v. Omega Flex, Inc., 628 Pa. 296, 104 A.3d 328, 399
(2014). I find the arguments in these cases unpersuasive.
Cases rejecting its approach seem concerned primarily
with abandoning the strict liability principles of § 402A
of the Restatement (Second) of Torts or imposing bur-
dens on plaintiffs. These concerns, however, appear to
me to elevate form over substance and do not reflect
the practical considerations involved in design defect
cases, which I have explored previously in this opinion.
See, e.g., M. Green, ‘‘The Unappreciated Congruity of
the Second and Third Torts Restatements on Design
Defects,’’ 74 Brook. L. Rev. 807, 808–11, 832–36 (2009);
T. Jankowski, supra, 36 S. Tex. L. Rev. 318–24; A. Twer-
ski & J. Henderson, supra, 74 Brook. L. Rev. 1106, 1108;
C. Perkins, note, ‘‘The Increasing Acceptance of the
Restatement (Third) Risk Utility Analysis in Design
Defect Claims,’’ 4 Nev. L.J. 609, 611–12 (2004).
III
WE SHOULD ADOPT THE RESTATEMENT
(THIRD) FOR DESIGN CLAIMS
In light of the foregoing, I would accept the invitation
of the reporters of the Restatement (Third) to recon-
sider the standard that this court employs in design
defect cases and to adopt the approach for resolving
design defect claims described in §§ 1, 2 and 4 of the
Restatement (Third). Doing so will bring our design
defect law in line with current product liability jurispru-
dence and eliminate our reliance on the now outdated
consumer expectations standard from the Restatement
(Second), which has proven ill-suited for design
defect claims.
Adopting the Restatement (Third) approach will not
substantially upend our current design defect law. We
have already taken a step toward the Restatement
(Third) model by adopting the malfunction theory from
§ 3 of the Restatement (Third). See Metropolitan Prop-
erty & Casualty Ins. Co. v. Deere & Co., supra, 302
Conn. 139–41. Our adoption of the malfunction theory
has already supplanted the ordinary consumer expecta-
tions standard in such cases, leaving little reason to
retain that standard, especially in light of the limited
role that the majority has given to it today. See M.
Green, ‘‘The Unappreciated Congruity of the Second
and Third Torts Restatements on Design Defects,’’
supra, 74 Brook. L. Rev. 834–35 (explaining that mal-
function theory ‘‘encompasses the kinds of cases that
were the model for [the ordinary consumer expecta-
tions test in §] 402A’’); J. Henderson & A. Twerski, ‘‘The
Products Liability Restatement in the Courts: An Initial
Assessment,’’ 27 Wm. Mitchell L. Rev. 7, 21 (2000) (dis-
cussing malfunction theory and noting that ‘‘most of
the cases cited by courts supporting a consumer expec-
tations test are of [the res ipsa] genre’’); J. Hoffman,
‘‘Res Ipsa Loquitur and Indeterminate Product Defects:
If They Speak for Themselves, What Are They Saying?,’’
36 S. Tex. L. Rev. 353, 377–78 (1995) (explaining similari-
ties between malfunction theory and ordinary con-
sumer expectations test); A. Twerski & J. Henderson,
supra, 74 Brook. L. Rev. 1101 (explaining that modern
application of ordinary consumer expectations test is
typically ‘‘confined . . . to cases that instantiate res
ipsa-like product failures’’).
In addition, Potter’s modified consumer expectations
test has already introduced risk-utility concepts into
our law. See Potter v. Chicago Pneumatic Tool Co.,
supra, 241 Conn. 221–22. Although the modified test
nominally rejects an alternative design requirement; id.,
221; our courts are already requiring this evidence as
a matter of practice. See A. Twerski & J. Henderson,
supra, 74 Brook. L. Rev. 1068, 1102. Adopting the
Restatement (Third) will thus bring our standards in
line with their actual application and thus provide more
consistent guidance to courts and juries applying our
law.
I would therefore disavow any continued reliance
on the ordinary or modified consumer expectations
standards and recognize only the risk-utility test from
§§ 1, 2 and 4 of the Restatement (Third) as the appro-
priate test for design defect claims.10 Res ipsa-like
claims would continue to be governed by the malfunc-
tion theory that we adopted in Metropolitan Property &
Casualty Ins. Co.
IV
APPLICATION OF RESTATEMENT (THIRD)
TO CERTIFIED QUESTION
Applying the risk-utility test to the present case, the
answer to the certified question is simple: comment (i)
to § 402A of the Restatement (Second) of Torts should
no longer be the law of this state for design defect
claims, and the Restatement (Third) does not contain
a similar provision. The expectations of consumers, and
even consumer awareness of open and obvious dangers,
are not dispositive considerations in the risk-utility
inquiry. The comments to the Restatement (Third)
explain that, ‘‘[e]arly in the development of products
liability law, courts held that a claim based on design
defect could not be sustained if the dangers presented
by the product were open and obvious. [The risk-utility
test] does not recognize the obviousness of a design-
related risk as precluding a finding of defectiveness.’’
Restatement (Third), supra, § 2, comment (d), p. 20.
The comments further explain that the risk-utility test
‘‘rejects conformance to consumer expectations as a
defense. The mere fact that a risk presented by a prod-
uct design is open and obvious, or generally known,
and that the product thus satisfies expectations, does
not prevent a finding that the design is defective.’’ Id.,
§ 2, comment (g), p. 28. Consumer expectations are,
instead, one factor for the jury to consider when
weighing the risks and benefits of a product design. Id.,
§ 2, comment (f), p. 23.
Consequently, I agree with the majority that we
should answer the certified question in the negative.
Because I cannot join the majority’s analysis in support
of this conclusion, however, I respectfully concur in
the result only.
1
The majority declines this opportunity principally because the parties
in the present case each relied on Potter in their arguments before the
United States Circuit Court of Appeals for the Second Circuit. See footnote
11 of the majority opinion. In my view, however, we should not limit our
analysis to clarifying and reaffirming Potter because, as I note in this opinion,
Potter’s standards were flawed when they were adopted nearly twenty years
ago and remain so today. Moreover, the certified question from the Second
question as it sees fit and add any pertinent questions of Connecticut law
that the [c]ourt chooses to answer.’’ Izzarelli v. R.J. Reynolds Tobacco Co.,
731 F.3d 164, 169 (2d Cir. 2013).
2
For a discussion of other design defect standards that have been consid-
ered and rejected, see T. Jankowski, supra, 36 S. Tex. L. Rev. 312–14 (dis-
cussing application of ‘‘ ‘deviation from the norm’ ’’ and ‘‘ ‘reasonable fitness
for intended purpose’ ’’ standards to design defect claims).
3
The consumer expectations test continues to be used in other contexts
in which consumer expectations tend to be well formed and more uniform.
See Restatement (Third), supra, § 2, comment (h), p. 28 (noting that con-
sumer expectations continue to play strong role in resolution of specialized
product defect claims involving food products and used products); see also
J. Phillips, ‘‘Consumer Expectations,’’ 53 S.C. L. Rev. 1047, 1061–63 (2002)
(discussing modern applications of consumer expectations standard).
4
The comments provide the following examples: ‘‘[W]hen a manufacturer
sells a soft stuffed toy with hard plastic buttons that are easily removable
and likely to choke and suffocate a small child who foreseeably attempts
to swallow them, the plaintiff should be able to reach the trier of fact with
a claim that buttons on such a toy should be an integral part of the toy’s
fabric itself (or otherwise be unremovable by an infant) without hiring an
expert to demonstrate the feasibility of an alternative safer design. Further-
more, other products already available on the market may serve the same
or very similar function at lower risk and at comparable cost. Such products
may serve as reasonable alternatives to the product in question.’’ Restate-
ment (Third), supra, § 2, comment (f), pp. 23–24.
5
The Restatement (Third) uses as an example a novelty item that has
little utility but potential to cause significant harm: an exploding cigar used
for pranks. It acknowledges that a jury could hold the manufacturer ‘‘liable
for the defective design of the exploding cigar even if no reasonable alterna-
tive design was available that would provide similar prank characteristics.
The utility of the exploding cigar is so low and the risk of injury is so high
as to warrant a conclusion that the cigar is defective and should not have
been marketed at all.’’ Restatement (Third), supra, § 2, illustration (5), p. 22.
6
Because malfunction theory cases do not turn on proof of a specific
manufacturing or design defect, the precise nature of the defect remains
undetermined. See Restatement (Third), supra, § 3, comment (b), pp. 111–12.
A finding of liability therefore does not condemn the entire product line,
making the consequences of liability under the malfunction theory much
less devastating to a manufacturer and thus making it fairer to impose
liability without requiring proof of a feasible alternative.
7
In Connecticut, questions of privity and duty are governed by statute.
See General Statutes § 52-572n (b) (claim may be asserted regardless of
whether claimant purchased product from or entered into contract with
product seller). So are other negligence related considerations, like compara-
tive fault. See General Statutes § 52-572o (setting forth comparative fault
standards for product liability claims).
8
Of course, a manufacturer separately may be deemed negligent for failing
to recall a product with a latent defect that was not foreseeable at the time
of sale.
9
For example, in adopting the Restatement (Third) approach to design
defect claims, the Iowa Supreme Court eliminated use of doctrinal reference
in design defect cases: ‘‘We question the need for or usefulness of any
traditional doctrinal label in design defect cases because, as comment n
points out, a court should not submit both a negligence claim and a strict
liability claim based on the same design defect since both claims rest on
an identical risk-utility evaluation. . . . Moreover, to persist in using two
names for the same claim only continues the dysfunction . . . . Therefore,
we prefer to label a claim based on a defective product design as a design
defect claim without reference to strict liability or negligence.’’ (Citation
omitted; emphasis in original.) Wright v. Brooke Group Ltd., 652 N.W.2d
159, 169 (Iowa 2002).
10
I recognize that we have adopted separate standards for resolving some
specialized types of design defect claims, namely, for prescription drugs.
See Vitanza v. Upjohn Co., 257 Conn. 365, 376, 778 A.2d 829 (2001). Because
liability for the design of those specialized products is not at issue in the
present case, I do not consider whether we should also apply the Restatement
(Third) to claims involving those products.