PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
JAMES L. OGLESBY,
Plaintiff-Appellant,
v. No. 98-1716
GENERAL MOTORS CORPORATION,
Defendant-Appellee.
JAMES L. OGLESBY,
Plaintiff-Appellee,
v. No. 98-1818
GENERAL MOTORS CORPORATION,
Defendant-Appellant.
Appeals from the United States District Court
for the District of South Carolina, at Charleston.
Falcon B. Hawkins, Senior District Judge.
(CA-97-403-2-11)
Argued: April 8, 1999
Decided: August 31, 1999
Before WILKINSON, Chief Judge, and NIEMEYER
and MICHAEL, Circuit Judges.
_________________________________________________________________
Affirmed by published opinion. Judge Niemeyer wrote the opinion,
in which Chief Judge Wilkinson and Judge Michael joined.
_________________________________________________________________
COUNSEL
ARGUED: Jill Waldman, GIBBS & HOLMES, Charleston, South
Carolina, for Appellant. W. Randall Bassett, KING & SPALDING,
Atlanta, Georgia, for Appellee. ON BRIEF: Allan R. Holmes,
GIBBS & HOLMES, Charleston, South Carolina, for Appellant. Chil-
ton Davis Varner, KING & SPALDING, Atlanta, Georgia; Wm.
Howell Morrison, HOLMES & THOMSON, L.L.P., Charleston,
South Carolina, for Appellee.
_________________________________________________________________
OPINION
NIEMEYER, Circuit Judge:
While James Oglesby was leaning into the engine compartment of
a seven-year old Chevrolet Silverado pickup truck to adjust a trans-
mission cable, a radiator hose detached, causing Oglesby serious burn
injuries. Contending that his injuries were caused by a defective plas-
tic hose connector between the radiator and the radiator hose, Oglesby
filed this product liability action against General Motors Corporation,
the manufacturer of the truck, alleging negligence, breach of war-
ranty, and strict liability under South Carolina law.
On General Motors' motion, the district court entered summary
judgment in General Motors' favor, rejecting under Federal Rule of
Evidence 702 the opinion of Oglesby's expert witness, a mechanical
engineer, but ruling that the principles of Daubert v. Merrell Dow
Pharmaceuticals, Inc., 509 U.S. 579 (1993), did not govern the
admissibility of that testimony. The district court also concluded that
Oglesby failed to prove that the part was defective when it left Gen-
eral Motors' manufacturing plant and that General Motors breached
its duty of care in manufacturing the product.
For the reasons that follow, we affirm.
I
In or around 1991, Blaine Nielson purchased a used 1988 Chevro-
let Silverado pickup truck. Within the first years after purchasing the
2
truck, Nielson had to overhaul the truck's engine and rebuild its trans-
mission. At the same time, he removed the radiator, cleaned and
flushed it, and replaced the radiator's hoses.
Several years later, in June 1995 when the truck had about 156,000
miles on it, the rods began "knocking," and the engine began to over-
heat. Nielson and his friend, James Oglesby, who was a certified
mechanic, completely rebuilt the engine. During this repair, not only
did they replace the thermostat in the radiator, but they also removed
and cleaned the radiator, replacing one of its hose clamps.
Within two weeks of this repair, the transmission began to shift
"too slowly," and Nielson took the truck to Oglesby's place of work
to have Oglesby adjust the transmission "detent cable," which causes
the transmission to change gears. To reach the detent cable, Oglesby
stepped on the driver's side front bumper, leaned into the engine com-
partment, and reached down toward the engine's firewall where the
cable was located. Oglesby's body was positioned on the driver's side
of the truck facing the windshield with his backside facing the radia-
tor. As Nielson described it, Oglesby "was kneeling up inside, reach-
ing to adjust the linkage. He is short, so instead of reaching over the
fender -- like I'm taller, but I don't know how to do it, so Jim was
doing that for me. But he got up inside and was leaning up over the
carburetor and stuff." As Oglesby leaned down to adjust the cable,
Nielson stated that he heard "a big `w[h]oosh' sound" and he heard
Oglesby yell. The radiator hose had detached and sprayed hot coolant
on Oglesby, severely burning his torso from the waist down and his
legs. Oglesby thereafter received extended hospital care to treat his
burns.
Following the incident, Nielson reattached the radiator hose and
continued driving the truck. Two months later, in September 1995,
Nielson and Oglesby's attorney inspected the radiator and discovered
a broken piece of the plastic inlet connector inside the radiator hose.
The inlet connector is a lipped pipe-type piece that serves as the con-
nection device between the radiator hose and the radiator.
Oglesby filed suit against General Motors, the manufacturer of the
truck, in state court in Charleston, South Carolina, alleging liability
based on negligence, breach of warranty, and strict liability. General
3
Motors removed the case to federal court, based on diversity of citi-
zenship, and filed a motion for summary judgment, contending that
the hose connector was not defective when it left control of General
Motors, that it had breached no duty of reasonable care, and that
plaintiff's expert testimony was inadmissible under both Daubert and
Federal Rule of Evidence 702.1 In answer to General Motors' motion
and in support of his claims, Oglesby presented the opinion of Doug-
las Bradbury, a former professor of mechanical engineering at Clem-
son University. Bradbury had previously consulted in an array of
cases involving the mechanical design and safety of various industrial
products, but he had no particularized experience or expertise in eval-
uating either automobile manufacturing processes or the strength of
plastic automobile component parts.
In preparing to render his expert opinion, Bradbury looked at the
broken plastic connector and the piece which broke off, took physical
measurements of the connector, and photographed the parts. These
three steps constituted the entirety of Bradbury's investigation into
the part or its manufacture. He found that "the connector [was] not
circular, but [was] a flattened oval in shape," noting that the differ-
ence between "the major and minor dimensions of the oval shape
[was] 1/10 inch." Bradbury did not know nor learn how the part was
manufactured or from what material it was manufactured. He did not
know nor learn any specifications prescribed for the part, and he did
not perform any tests or calculations to determine the strength of the
part or the stresses to which it was subjected. From his observation
and measurement, Bradbury concluded:
The out-of-roundness of the upper (inlet) hose connector
was apparently a manufacturing defect. Certainly, there
would be no reason to design it as an oval shape, consider-
ing the method of clamping. Two probable causes of the
_________________________________________________________________
1 Federal Rule of Evidence 702, entitled "Testimony by Experts," pro-
vides:
If scientific, technical, or other specialized knowledge will assist
the trier of fact to understand the evidence or to determine a fact
in issue, a witness qualified as an expert by knowledge, skill,
experience, training, or education, may testify thereto in the form
of an opinion or otherwise.
4
out-of-roundness are recognized. One is that the upper con-
nector was manufactured out-of-round. This is most
unlikely, considering that the mold was designed to produce
round, circular connectors. Also, the out-of-roundness
should have been seen during the routine on-line inspection.
It is most likely that a defect in the formulation or han-
dling of the thermo-setting plastic caused time-dependent
internal stresses which deformed the connector from round
to oval, ultimately resulting in the loosening of the upper
hose and the scalding of James Oglesby.
After Bradbury gave his opinion, a General Motors expert testified
that the part was "not a thermosetting plastic," but a "nylon compos-
ite, glass filled," that had "thermoplastic" characteristics including the
"ability to be remolten." Bradbury then submitted a supplemental affi-
davit stating, "I incorrectly stated that the inlet connector was made
of thermosetting plastic when it was really made of a glass thermo-
plastic nylon composite." Nonetheless, he concluded, still without
possessing the specific properties of the material, that the "error does
not in any way alter my opinion as to the defect."
General Motors' expert concluded that the "hose separation was . . .
due to large bending loads applied to the hose connection" from "Mr.
Oglesby leaning or pressing his torso against the radiator hose." Gen-
eral Motors also determined that Oglesby was the only individual,
past or present, to file a complaint alleging a defect in the inlet con-
nector installed in any of the 2,505,491 series C/K trucks, which
included the 1988 Silverado model, that it manufactured between
1987 and 1991.
In granting General Motors' summary judgment motion on strict
liability and breach of warranty, the district court applied South Caro-
lina product liability law and concluded that Oglesby "failed to pro-
duce competent evidence that the truck was in the same condition as
when it left the defendant's possession," explaining that the truck's
"age, mileage, and prior repairs exceed the reasonable limits of estab-
lishing the fundamental element that the product was in the same con-
dition at the time of the accident as when it left the defendant's
possession." On the negligence claim, the court ruled that "[p]laintiff
5
has shown this court nothing which would support the claim that GM
knew of a potential product defect involving this type of radiator."
While the district court concluded that Daubert did not apply to Brad-
bury's testimony because it was not scientific testimony, it neverthe-
less held that Federal Rule of Evidence 702 limited the admissibility
of testimony of an expert testifying in a field of technical or other spe-
cialized knowledge such as engineering. Applying Rule 702, the court
refused to consider Bradbury's opinion because it was mere specula-
tion:
The proposed testimony must go beyond mere speculation
or conjecture to be of assistance to the trier of fact. In this
case, while plaintiff's expert is clearly qualified as an expert
on various engineering principles, his proposed opinion in
this case lacks any probative value and would be of no assis-
tance to the trier of fact. . . . [T]his court finds that [Brad-
bury's] proposed testimony lacks the reliability, foundation
and relevance necessary for admissibility in this case [under
Federal Rule of Evidence 702]. The expert has shown no
knowledge of the manufacturing process, the radiator's
composite makeup, nor has he conducted any meaningful
testing or analysis.
On appeal, Oglesby contends that Bradbury's prior experiences in
working with industrial products qualified him to give a meaningful,
nonscientific opinion on whether the plastic inlet connector contained
a manufacturing defect. On cross-appeal, General Motors contends
that the district court erred in not subjecting Bradbury's testimony to
the rigors of Daubert, which would have provided an alternative
ground upon which to reject Bradbury's opinion.
II
After the district court entered judgment in this case but before oral
argument on appeal, the Supreme Court decided Kumho Tire Co. v.
Carmichael, 119 S. Ct. 1167 (1999), in which it held that "Daubert's
general holding -- setting forth the trial judge's general `gatekeeping'
obligation -- applies not only to testimony based on `scientific'
knowledge, but also to testimony based on `technical' and `other spe-
cialized' knowledge." Id. at 1171 (citing Fed. R. Evid. 702). In
6
Kumho Tire, the plaintiffs claimed that a manufacturing defect caused
a tire to blow out, which, in turn, resulted in numerous injuries and
one death. The tire was manufactured by the defendant and installed
on a vehicle in 1988, five years before the plaintiffs purchased the
vehicle with the tire on it. At the time of the accident, the tire had
worn bald in certain areas and had undergone prior inadequate repairs
to patch punctures. Plaintiffs' expert, who had a masters degree in
mechanical engineering and 10 years work experience at Michelin
America, Inc., as well as prior consulting experience in other tire
blowout cases, gave his opinion that a manufacturing defect or design
defect caused the plaintiffs' injuries. He based his opinion upon the
combination of his knowledge of tire failures, a personal theory of the
cause of tire failures, and his inspection of the tire at issue. His rea-
soning was syllogistic -- a properly manufactured tire carcass stays
bound to its innertread; the tread of the tire at issue did not stay so
bound; thus the tire was defectively manufactured.
The district court in Kumho excluded the expert's testimony, not
because the witness lacked the requisite general qualifications, but
because the court found his testimony unreliable. The court was par-
ticularly concerned about "the methodology employed by the expert
in analyzing the data obtained in the visual inspection, and the scien-
tific basis, if any, for such an analysis." Id. at 1173 (quoting the dis-
trict court). In reversing the Eleventh Circuit's opinion in this case,
which had held that Daubert applied only to scientific expert testi-
mony, the Supreme Court held that the district court did not abuse its
discretion in dismissing the expert's opinion as unreliable. The Court
emphasized that while the plaintiffs' expert was experienced in tire
failures, his methodology of relying on a visual inspection of the tire
was unreliable. See id. at 1176-79.
In addition to clarifying the scope of Daubert , Kumho Tire rein-
forces principles established in our cases interpreting Federal Rule of
Evidence 702. We have admonished that "a plaintiff may not prevail
in a products liability case by relying on the opinion of an expert
unsupported by any evidence such as test data or relevant literature
in the field." Alevromagiros v. Hechinger Co. , 993 F.2d 417, 422 (4th
Cir. 1993); accord Freeman v. Case Corp., 118 F.3d 1011, 1016-17
(4th Cir. 1997).
7
Accordingly, a district judge, considering a proffer of expert testi-
mony under Federal Rule of Evidence 702 -- whether based on scien-
tific, technical, or other specialized knowledge-- must, in
determining its admissibility, ensure that the evidence is "not only rel-
evant, but reliable." Daubert, 509 U.S. at 589. A reliable expert opin-
ion must be based on scientific, technical, or other specialized
knowledge and not on belief or speculation, and inferences must be
derived using scientific or other valid methods. See id. at 590, 592-93.
Reliability of specialized knowledge and methods for applying it to
various circumstances may be indicated by testing, peer review, eval-
uation of rates of error, and general acceptability. See id. at 593-94.
But at bottom, the court's evaluation is always a flexible one, and the
court's conclusions necessarily amount to an exercise of broad discre-
tion guided by the overarching criteria of relevance and reliability.
When applying these principles to the case before us, we find that
the particular analysis called for is similar to that in Kumho Tire.
Bradbury was concededly a qualified mechanical engineer who
attempted to apply general engineering principles to conclude that the
plastic inlet connector in this case was defective when it left General
Motors in 1988. His testimony was not sufficiently reliable, however,
and did not properly draw on specialized knowledge. Rather, it
depended on an imperfect syllogism constructed from unsupported
suppositions.
Bradbury observed that the inlet connector was out of round and
that, therefore, forces from the hose clamp as well as internal pressure
from the radiator contents caused the connector to crack over time. He
assumed that the manufacturer's mold was round and deduced that the
connector could become out of round only during the time that the
part was curing during "handling," presumably because it was a
"thermo-setting" plastic. He assumed that the plastic part could not
have become out of round much later because it could not be
remolded by heat.
Bradbury admitted, however, that he did not know the type or com-
position of the plastic. He did not ask the manufacturer; he did not
analyze the part; he did not test it; he did not apply any calculations.2
_________________________________________________________________
2 A sampling of Bradbury's deposition testimony on this subject fol-
lows:
8
Thus, as a matter of logic, Bradbury could not eliminate other equally
plausible causes for the plastic to have come out of round or to have
cracked, such as engine overheating; trauma to the inlet connector if
the radiator, when removed from the car, had tipped over; or stress
applied by leaning on the connector with too much weight when the
plastic was hot -- all of which would be consistent with the given his-
tory of the part in this case. Bradbury assumed that the part could
have gone out of round only after it left the manufacturer's mold but
before it was fully set, but he had no factual basis by which to reach
that conclusion.
As it turned out, the part was not manufactured out of thermo-
setting plastic but out of a nylon composite which allows it to be
_________________________________________________________________
Q. You haven't done any computer analysis of any of the
stresses or strengths of the materials we have been discuss-
ing, have you?
A. No, I have not.
Q. You have done no testing of the radiator or parts of the radi-
ator.
A. No. Just photographed it.
* * *
Q. You haven't done any analysis by way of equations or cal-
culation.
A. No.
Q. Correct?
A. That's correct. That's the reason I used the term "quantita-
tive" earlier.
Q. I understand. You haven't done any analysis of the manu-
facturing process of the radiator either by way of use of the
inspection mold process nor the assembly line process itself.
A. No.
Q. You have done no actual testing in which you would put the
radiator under pressure and apply various external forces to
it either.
A. No.
9
remolded. Despite the fact that Bradbury's manufacturing defect the-
ory appears to be premised on the inability of the part to change shape
after it was "set," Bradbury did not alter his opinion at all when con-
fronted with the error in his underlying assumption as to the composi-
tion of the part.
Under generally known engineering principles, stress applied in a
given amount in a particular direction would theoretically cause a
plastic part to reshape itself, if hot enough, or to break. To conclude,
however, that a particular stress caused a particular plastic part to
reshape itself or to break, the engineer would have to know the
amount of the particular stress and the stress resistance of the part at
various temperatures in order to ascertain that the particular stress
overcame the stress resistance of the part. Bradbury conceded, how-
ever, that he had none of the necessary data and therefore could not
make any such calculations for the part in this case. He could only
speculate as to a possibility which was no more likely than other
available possibilities. Without more information, it would appear just
as consistent with Bradbury's theory that the overheating of the
engine could have caused the part to reshape, perhaps coupled with
pressure by a mechanic leaning on it when hot.
The district court recognized this flaw, concluding that Bradbury's
opinion lacked "any probative value" because it lacked "the reliabil-
ity, foundation and relevance necessary for admissibility" under Fed-
eral Rule of Evidence 702. We cannot conclude that the district court
abused its discretion in excluding Bradbury's opinion testimony.
While the district court did not have the benefit of Kumho Tire, it
reached the same result that an analysis under Daubert would have
yielded.
III
Without Bradbury's testimony, Oglesby failed to meet his burden
of proof. He proffered no evidence demonstrating (1) that the plastic
inlet connector was in essentially the same condition at the time of the
incident in 1995 as it was when it left General Motors' plant in 1988
or (2) that General Motors failed to exercise due care.
Because Oglesby's injury in this case was sustained in South Caro-
lina and removal to the district court was based on diversity of citi-
10
zenship, see 28 U.S.C. § 1332, we apply the substantive law of South
Carolina. See Klaxon Co. v. Stentor Electric Mfg. Co., 313 U.S. 487,
496 (1941) (holding that in a diversity action the federal court applies
the state court's choice of law rules); Lister v. NationsBank, 494
S.E.2d 449, 454 (S.C. Ct. App. 1998) (holding under South Carolina
choice of law principles that "substantive law governing a tort action
is determined by the state in which the injury occurred" (citations
omitted)).
Under South Carolina law, a plaintiff in an action for strict product
liability must demonstrate "(1) that he was injured by the product; (2)
that the product, at the time of the accident, was in essentially the
same condition as when it left the hands of the defendant; and (3) that
the injury occurred because the product was in a defective condition
unreasonably dangerous to the user." Allen v. Long Mfg. NC, Inc., 505
S.E.2d 354, 356 (S.C. Ct. App. 1998) (internal quotation marks, cita-
tion, and emphasis omitted); see also S.C. Code § 15-73-10 (estab-
lishing liability of seller for defective products). Similarly, to prove
a breach of warranty claim, a plaintiff must demonstrate that the
goods were defective "at the time the sale was completed." Doty v.
Parkway Homes Co., 368 S.E.2d 670, 671 (S.C. 1988) (construing
South Carolina's Uniform Commercial Code, S.C. Code§ 36-2-314).
And when pursuing recovery under a negligence theory, a plaintiff
must prove, in addition to the three elements described in Allen,
"proof that the manufacturer breached its duty to exercise reasonable
care to adopt a safe design." Allen, 505 S.E.2d at 356 (internal quota-
tion marks and citations omitted).
The Silverado truck on which Oglesby was working was purchased
by Nielson in 1991 as a used truck and had approximately 156,000
miles on it at the time that Oglesby was injured. The record contains
no information regarding the prior owners' repair activities or treat-
ment of the truck. We do know that when Nielson owned the truck,
he had removed the radiator on at least two occasions and had
removed and repaired its hoses. During either of those repairs, the
radiator's inlet connector could have been damaged as a result of the
radiator's having been improperly removed, handled, or reinstalled.
The inlet connector could also have been damaged from engine over-
heating. Without evidence demonstrating the condition of the inlet
connector when it was sold by General Motors and left its custody,
11
Oglesby cannot carry his burden under South Carolina law for prov-
ing strict liability and breach of warranty.
Finally, Oglesby also failed to offer any evidence of negligence by
General Motors. Indeed, he was unable to demonstrate that General
Motors had knowledge of any problem with the plastic inlet connec-
tor. General Motors' statement that it had never received a complaint
about the inlet connector in connection with any of the more than
2,500,000 trucks that had the part went unchallenged.
For the foregoing reasons, the judgment of the district court is
AFFIRMED.
12