TEXAS COURT OF APPEALS, THIRD DISTRICT, AT AUSTIN
NO. 03-98-00123-CV
Doyle Wilson Homebuilder, Inc., Appellant
v.
David Pickens, Barbara Pickens and Terry Maxwell Electric, Inc., Appellees
FROM THE DISTRICT COURT OF TRAVIS COUNTY, 126TH JUDICIAL DISTRICT
NO. 96-07260, HONORABLE JOSEPH H. HART, JUDGE PRESIDING
After a fire destroyed their home, appellees David and Barbara Pickens sued
appellant Doyle Wilson Homebuilder, Inc. ("Doyle Wilson")
(1) and appellee Terry Maxwell
Electric, Inc., for various causes of action related to the construction and sale of the house. A jury
determined that neither defendant acted negligently, but found that Doyle Wilson breached two
warranties and failed to comply with the home sale agreement. Based on the jury's verdict, the
trial court awarded the Pickenses $299,399.88 in actual damages, plus prejudgment interest and
attorney's fees. In six points of error, Doyle Wilson challenges the legal and factual sufficiency
of the evidence supporting the trial court's judgment. We will reverse the judgment and remand
the cause for a new trial.
FACTUAL AND PROCEDURAL BACKGROUND
In 1992, David and Barbara Pickens signed a contract (the "home sales agreement")
with Doyle Wilson for the construction of a house in Austin. In this written agreement, Doyle
Wilson warranted that the construction would substantially comply with (1) the plans and
specifications, (2) construction standards required under a warranty it offered, and (3) applicable
building codes. Doyle Wilson representatives also told the Pickenses that Doyle Wilson would
provide quality workmanship and materials, and that the house would be something they could live
in for a long time. Doyle Wilson entered into a contract with Terry Maxwell Electric (the
"contractor base agreement") to install the electrical wiring in the Pickenses' house. The
contractor base agreement warranted that all materials furnished by Terry Maxwell would be of
good quality and free of defects. The Pickenses moved into their new home in November 1992
and lived there until June 1994, when a fire broke out while the Pickenses were away and
destroyed their home.
The Pickenses' theory was that the fire resulted from problems with electric wiring
in a truss area between the first and second floors, caused by either improper installation or
defective materials. They sued Doyle Wilson and Terry Maxwell Electric, alleging negligence,
breach of implied warranties of habitability and good and workmanlike performance of services,
and violations of the Texas Deceptive Trade Practices Act. See Tex. Bus. & Com. Code Ann. §§
17.41-.63 (West 1987 & Supp. 1999) ("DTPA"). They also brought a strict liability claim against
Terry Maxwell Electric, alleging that the company installed defective wiring. The Pickenses
claimed actual damages for the loss of their house, out-of-pocket expenses, mental anguish, and
emotional distress; they also sought attorney's fees, court costs, and interest.
The Plaintiffs' Witnesses
David and Barbara Pickens both testified at trial. Their testimony established that
there was no history of electrical problems with the house, and that they had never had any
electrical repairs performed prior to the fire. While the house was under construction, they saw
nothing in the work done by employees of Terry Maxwell Electric that gave them any cause for
concern. During the eighteen months they lived there, the Pickenses felt that the house had been
constructed in a good and workmanlike manner and they made no warranty claims. Nor did they
experience any "tripping" of the circuit breakers due to electrical problems during that time.
Raymond Reynolds and Philip Wagner were the Pickenses' expert witnesses; they
investigated the fire scene, co-authored a report on the fire that was offered in evidence, and
testified at trial. Reynolds, a cause-and-origin expert with forty-two years of experience in fire
protection, examined the house three days after the fire. He determined that the fire started in a
truss assembly above the southeast corner of the family room. (2) Since the electric wiring running
through the truss area was the only heat energy source capable of starting a fire in that location,
Reynolds concluded that the fire was caused by a heat buildup in the wiring. He then relied on
Philip Wagner, an electrical engineer, to examine the wiring further.
Reynolds and Wagner identified numerous electrical "faults" (3) in the wiring running
through that area. In their report, they noted two defects that made them question the overall fire
safety of the Pickenses' home. First, they observed some wiring tightly wrapped around a metal
brace in the truss assembly; Reynolds testified that the sharp angle of the metal brace could wear
through the wire insulation over time and eventually cause a fire. Second, they saw a staple nailed
through a metal brace into the truss assembly. (4) However, both of these defects were located at
least five or six feet from the fire's point of origin, and Reynolds and Wagner both acknowledged
that the defects did not cause the fire. Wagner agreed that neither defect constituted a violation
of the National Electric Code; moreover, he stated in the report that the "overall electrical
workmanship was considered to be good." Finally, the circuit breaker box was intact after the
fire with all but four of the twenty breaker switches "tripped," indicating that the breakers had
functioned properly.
Wagner inspected electrical wiring that was removed from the fire scene and detailed his
findings in a section of the report. He found evidence of arcing, which he described as a
manifestation of electrical energy "jumping" across a gap between two conducting wires.
However, he was unable to identify any specific faulted wire as having caused the fire. Wagner's
engineering analysis in the report concluded:
It is our professional judgement that the fire was initiated by an electrical
fault that occurred in the bundle of electrical cables located in the ceiling of the
family room. There were multiple cables faulted in the same area and no definitive
initiating event could be determined.
Since these cables were faulted in an area with no obvious interference or
obstruction, and no electrical storm or power surge event had occurred, it is the
opinion of the engineer that the faults were caused by either a construction related
problem with, or by an insect or rodent attack on, the electrical insulation. The
owner reported he had no[t] experienced any problems with rodents or insects.
Our examination found no indication of insect or rodent infestation. Therefore, the
insulation was probably faulty or damaged during the cable installation process.
(Emphasis added.)
At trial, Wagner testified about several samples of wiring that were introduced in
evidence. Wagner noted the large size of a fault bead (5) on one of the conductor wires, indicating
that the wire had short-circuited. On the same section of wires, he pointed out that three of the
four conducting wires had short-circuited but the other had not. Wagner described this as
"interesting" since usually all four wires will short-circuit and weld together, or else none will;
however, he did not relate this condition to the fire in any way. On a second sample, he testified
that the presence of a copper bead on the wire signified that "the conductors were energized and
there was electrical activity occurring." Wagner also testified that wiring can be damaged during
installation if it is pulled tight against a sharp surface, which removes the wiring insulation "just
like a potato peeler."
At trial, Wagner testified: "With regard to this fire we eliminated everything but
an installation problem, and we noted examples of poor installation practices, and, therefore,
concluded that the faulting was caused by a poor installation practice." However, under cross-examination he acknowledged that he could not say exactly what caused the fire. Wagner also
testified on direct examination that he eliminated manufacturing defects in the wiring insulation; (6)
on cross-examination, however, he acknowledged the conclusion in his report that a manufacturing
defect and installation defect were the two possible conditions that could have caused the fire, and
that the report did not specify if one was a more likely explanation than the other. Wagner
admitted that a manufacturing defect remained a possible explanation, but said he did not detect
any manufacturing defects in the wiring he examined. Although Wagner testified that there are
some manufacturing defects that an electrician would not likely detect during installation, he stated
that a defect in the wiring resulting in a lack of insulation would be "very obvious to the installer."
The Defendants' Witnesses
The first defense witness called at trial was Ed Kirkham, a former fire chief and an
investigator with more than forty years of experience; he testified for the defense regarding the
origin and spread of the fire. Kirkham concluded that the fire probably started on the second floor
of the house. It was his opinion that any problems with the installation of the wiring would have
manifested themselves in some way during the eighteen months between the date Pickenses moved
in and the date of the fire. Since the Pickenses had not reported any electrical problems during
that time, and based on his review of the physical evidence of wiring from the fire, Kirkham
concluded that the fire was not caused by anything Terry Maxwell Electric did in installing the
wiring. Finally, Kirkham testified that structural wiring is responsible for only seven percent of
residential fires, and that in all reasonable probability electrical problems did not cause the fire in
the Pickenses' home.
Terry Maxwell testified about the electrical wiring his company installed in the
Pickenses' house. Before a home buyer moves in, the wiring work is inspected by one of Terry
Maxwell Electric's master electricians, by Doyle Wilson's inspector, and by an inspector with the
City of Austin (the "City"). Maxwell personally performed a rough inspection of the Pickenses'
house in 1992 and was satisfied with the work, and he testified that the house passed rough and
final inspections by the City as well. He also noted that neither the City nor Doyle Wilson placed
"red tag" correction notices on any of the work his company performed. Maxwell testified that
his company's electricians are trained to look for defects in wiring insulation and improper wiring
installation, and testified that if a wire was touching a metal plate it would be "red tagged" by his
inspectors, Doyle Wilson's inspectors, and the City's inspectors. He claimed that his company's
work was passed by the City on its first inspection over ninety-five percent of the time.
Maxwell also testified that there were a number of tradespersons hired by Doyle
Wilson to construct the Pickenses' house, such as plumbers, carpenters, and air conditioning
installers. After the electrical installation was performed, Maxwell had no control over who was
allowed to work around the electrical lines. He also agreed that the type of wiring insulation used
in the Pickenses' house can be damaged by any number of construction activities, including
stepping or dropping items on it. However, there was no evidence that any tradespersons actually
damaged the electrical wiring.
Art Driskill, an electrical inspector for the City, testified by videotape deposition.
Driskill testified that he conducted a final inspection of the Pickenses' house in October 1992 and
that it passed both local and national standards. Driskill noted that he only inspected the light
fixtures, plug switches, air conditioning connections, and the appliances that were installed at that
time; however, he testified that those items had all been installed with reasonable care. Driskill
testified that the electrical installation performed by Terry Maxwell Electric met the requirements
of the National Electric Code, the State Rules Standard Book, and the City Code Amendments.
John Golding and John Kiracofe are respectively the chief and assistant chief of the
Jollyville Fire Department; portions of their deposition testimony were read to the jury. Golding
testified that he and Kiracofe responded to the Pickenses' fire and performed an unofficial
investigation at the request of the Hudson Bend Fire Department. Golding and Kiracofe concluded
that the fire originated in an exterior wall area near the junction between the first and second floors
directly under a storage closet located in the sitting room of the second floor. Although they ruled
out arson, they were unable to determine an exact cause.
Michael Owen is an electrical training and consulting expert and holds a master
electrician's license; he testified as a defense expert by deposition. Owen testified that for a fire
to occur as the Pickenses theorized, there would first have to be an electrical fault, such as a short
circuit, and then a failure of the circuit breaker protecting the wiring circuit. From the statement
in Reynolds's and Wagner's report that after the fire most of the circuit breakers were found to
have "tripped," Owen concluded that the breakers had functioned properly. Owen further testified
that for the fire to have been caused by damaged wiring insulation, (1) the outer jacket would have
to be torn; (2) at the same spot the insulation on at least two wires would have to be torn; (3) there
would have to be physical contact between the two wires; and (4) the circuit breaker would have
to fail to function properly. Based on this sequence of necessary events, Owen concluded that it
was not likely that the fire resulted from damage to the wiring insulation. Owen also pointed out
a faulty cause-and-effect assumption in Reynolds's and Wagner's report regarding faulting. Owen
testified that the evidence of faults after the fire does not itself indicate that faulting caused the
fire; on the contrary, it was more likely that the fire burned away the insulation around the wiring,
creating a fault that tripped the circuit breakers.
The jury failed to find that either defendant's negligence proximately caused the
fire. However, it found that Doyle Wilson breached (1) a warranty to perform services in a good
and workmanlike manner, and (2) a warranty of habitability. It further found that Doyle Wilson
failed to comply with the home sales agreement, and engaged in "false, misleading or deceptive"
acts in violation of the DTPA. However, the jury did not find that Terry Maxwell Electric failed
to comply with the contractor base agreement. The jury awarded David and Barbara Pickens
$20,000 each for mental anguish. Doyle Wilson moved for judgment non obstante veredicto
("n.o.v."); however, the trial court rendered a final judgment against Doyle Wilson for
$299,399.88, (7) plus $51,458.38 in prejudgment interest and $40,000 in attorney's fees. The trial
court ordered that the Pickenses take nothing by their claims against Terry Maxwell Electric.
Doyle Wilson then moved for a new trial, which motion the court overruled.
Doyle Wilson raises six points of error challenging the trial court's judgment. In
its first four points, appellant alleges that there is legally or factually insufficient evidence that it:
(1) breached a warranty to build the house in a good and workmanlike manner; (2) breached a
warranty of habitability; (3) failed to comply with the home sales agreement; and (4) engaged in
DTPA violations. In its fifth point of error, Doyle Wilson argues that if we determine that it is
liable to the Pickenses because of a wiring defect, then the jury's failure to find that Terry
Maxwell Electric breached the contractor base agreement is contrary to the overwhelming weight
and preponderance of the evidence. If the fifth point of error is sustained, Doyle Wilson asserts
as a conditional sixth point of error that there is either no evidence or insufficient evidence to
support the jury's finding that Doyle Wilson and Terry Maxwell Electric entered into the
contractor base agreement with the intention that the agreement "directly and primarily" benefit
the Pickenses.
DISCUSSION
In reviewing a no-evidence challenge, we consider all the evidence in the light most
favorable to the prevailing party, indulging every reasonable inference in that party's favor. See
Associated Indem. Corp. v. CAT Contracting, 964 S.W.2d 276, 285-86 (Tex. 1998). We will
uphold the finding if more than a scintilla of evidence supports it. See Burroughs Wellcome Co.
v. Crye, 907 S.W.2d 497, 499 (Tex. 1995); Seideneck v. Cal Bayreuther Assocs., 451 S.W.2d
752, 755 (Tex. 1970); In re King's Estate, 244 S.W.2d 660, 661 (Tex. 1951). The evidence
supporting a finding amounts to more than a scintilla if reasonable minds could arrive at the
finding given the facts proved in the particular case. See Crye, 907 S.W.2d at 499; Transportation
Ins. Co. v. Moriel, 879 S.W.2d 10, 25 (Tex. 1994); see also William Powers, Jr. & Jack Ratliff,
Another Look at "No Evidence" and "Insufficient Evidence," 69 Tex. L. Rev. 515, 522 (1991).
In reviewing a jury verdict to determine the factual sufficiency of the evidence, we consider and
weigh all the evidence and set aside the judgment only if the evidence is factually so weak, or the
verdict so contrary to the overwhelming weight of the evidence, as to make the judgment clearly
wrong and unjust. See Cain v. Bain, 709 S.W.2d 175, 176 (Tex. 1986); Garza v. Alviar, 395
S.W.2d 821, 823 (Tex. 1965); King's Estate, 244 S.W.2d at 661; see also Pool v. Ford Motor
Co., 715 S.W.2d 629 (Tex. 1986); see generally Powers & Ratliff, 69 Tex. L. Rev. 515.
Doyle Wilson's first point of error alleges that there is either no evidence, or, in
the alternative, insufficient evidence to support the jury's finding that Doyle Wilson breached a
warranty to build the house in a good and workmanlike manner. As part of this argument, Doyle
Wilson urges that Wagner's testimony was not scientifically reliable under E.I. du Pont de
Nemours v. Robinson, 923 S.W.2d 549 (Tex. 1995), and Sorensen v. Shaklee Corp., 31 F.3d 638
(8th Cir. 1994). We will address this "unreliability" portion of appellant's no-evidence argument
first.
Appellant's Challenge to the Scientific Reliability of Wagner's Testimony
Appellees claim that Doyle Wilson has "waived its right to appeal based on no
evidence and insufficient evidence as no objection was made to the reliability of the scientific
evidence or admission of expert testimony at trial." In support of this argument, appellees cite
Maritime Overseas Corp. v. Ellis, 971 S.W.2d 402 (Tex.), cert. denied, 119 S. Ct. 541 (1998),
in which the supreme court stated: "[T]o prevent trial or appeal by ambush, we hold that the
complaining party must object to the reliability of scientific evidence before trial or when the
evidence is offered." Id. at 409-10. Although appellees claim that Doyle Wilson did not request
a Daubert-Robinson (8) hearing, the record indicates that at a pretrial hearing appellant's counsel
moved to strike the testimony of Philip Wagner, the Pickenses' electrical expert, on the basis that
his testimony was not scientifically reliable (9) under the Robinson standard. The trial court
overruled appellant's motion. We conclude that Doyle Wilson has preserved its no-evidence
points claiming unreliable methodology, and we will address them on appeal.
In Robinson, the Texas Supreme Court adopted the United States Supreme Court's
"reliability" and "relevancy" requirements for scientific expert testimony from Daubert v. Merrell
Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), and used them to decide whether the trial court
had properly excluded testimony from the plaintiffs' expert that the defendant's contaminated
Benlate product had damaged the plaintiffs' pecan trees. In affirming the trial court's exclusion,
the Robinson court noted that the expert had no proof that the plaintiffs' Benlate was contaminated,
and had no knowledge as to what amount or concentration of herbicides would damage pecan
trees. See Robinson, 923 S.W.2d at 559. The supreme court reiterated the Eighth Circuit's
criticism leveled at the testimony offered in Sorensen: "Instead of reasoning from known facts
to reach a conclusion, the experts here reasoned from an end result in order to hypothesize what
was needed to be known but what was not." Id. (quoting Sorensen, 31 F.3d at 649) (emphasis in
Robinson omitted).
In the present case, Wagner testified that he eliminated other possible causes for
the fire and explained his reasons for doing so. He then concluded that an electrical problem in
the wiring had started the fire. We find Wagner's testimony distinguishable from the testimony
held properly excluded in Robinson. While Doyle Wilson labels Wagner's testimony "rank
speculation," Wagner did not simply guess at the fire's cause; instead, he testified that he deduced
it by eliminating other possible causes, such as gnawing rodents or a surge from an electrical
storm. In making his pretrial objection to Wagner's testimony, appellant's counsel did not develop
any argument beyond his assertion that Wagner was improperly reaching a conclusion first and
then hypothesizing facts to support that conclusion. Nor did counsel attempt to impeach Wagner
at trial with any evidence that an electrical storm or gnawing animals actually caused the fire. The
trial court did not abuse its discretion by allowing Wagner to testify. We therefore reject the
Daubert-Robinson "unreliability" argument appellant raises in points of error one through four.
Appellant's General Claim of Factual and Legal Insufficiency
Our holding that the trial court did not abuse its discretion in refusing to exclude
Wagner's testimony does not end our inquiry. Doyle Wilson's "unreliability" argument
constitutes only a portion of its evidentiary sufficiency challenge; the remainder of appellant's
argument is a broader assertion that there is no evidence, or insufficient evidence, to support the
jury's finding that Doyle Wilson breached a warranty to perform services in a good and
workmanlike manner. Points of error two through four make similar challenges to the jury's
findings that Doyle Wilson breached a warranty of habitability, failed to comply with the home
sales agreement, and committed DTPA violations. Because all four points of error are premised
on the same underlying argument--that there is insufficient evidence linking any act or omission
on Doyle Wilson's part to the cause of the fire--we will address these points together.
The Pickenses' theory at trial and on appeal is that the fire was caused by electrical
wiring that was either (1) defective or (2) damaged when it was installed. The jury failed to find
negligence on the part of either Doyle Wilson or Terry Maxwell Electric; it also did not find that
Terry Maxwell Electric breached the contractor base agreement with Doyle Wilson. Therefore,
to support the jury's answers that Doyle Wilson breached a warranty to perform services in a good
and workmanlike manner, breached a warranty of habitability, and committed DTPA violations,
there must be some evidence that the wiring was either (1) defective, or (2) damaged after
installation by someone not under the control of Terry Maxwell Electric. Appellant cites a number
of fire cases, which we will discuss.
In Systech Financial Corp. v. Vaughn, 558 S.W.2d 105 (Tex. Civ. App.--Tyler
1977, no writ), a corporate landlord sued its tenant after a fire damaged the tenant's apartment;
the suit alleged that tenant negligently used or disposed of smoking materials. The evidence
adduced at trial was that the tenant smoked and that the fire started in a sofa in the tenant's
apartment. An investigator for the Dallas Fire Department also testified that in his opinion the fire
was caused by careless smoking. At the close of the plaintiff's case, the trial court granted the
tenant's motion for directed verdict. See id. at 105.
The sole issue on appeal was whether the trial court erred by instructing a verdict
for the defendant because the plaintiff presented sufficient probative evidence of the tenant's
negligence to create a fact issue for the jury. In affirming, the court of appeals explained:
A presumption of fact cannot rest upon a fact presumed, or in other words, one
presumption cannot be based upon another presumption nor an inference of fact
upon other inferences . . . . Facts may be established circumstantially, but the
circumstances themselves must be shown by direct evidence, and cannot be inferred
from other circumstances . . . .
Id. at 105-06 (quoting Alamo Cas. Co. v. Stephens, 259 S.W.2d 729, 735 (Tex. Civ.
App.--Austin 1953, writ ref'd n.r.e.)). The court noted that to find that the tenant was smoking
on the sofa, the jury would have to infer that fact from testimony about his smoking habits; after
"indulging in this inference," the jury would then have to take another step and infer that the
tenant was smoking carelessly. Id. at 106. Finally, the jury would have to make a third inference
that careless smoking was the cause of the fire. Since there was no direct testimony allowing the
court to pyramid those inferences, the court held that the testimony at best could arouse a mere
surmise or suspicion of the tenant's negligence: "Evidence that it is possible that the [tenant]
caused the fire here involved cannot be accepted as evidence he did so. It must be proved." Id.
Appellant also cites John R. Francis Building Co. v. Bob Meador Co., 517 S.W.2d
693 (Tex. Civ. App.--Houston [14th Dist.] 1974, no writ). Francis was a homebuilder; his
company sued Meador, an air conditioning installation subcontractor, after a fire damaged a home
Francis was building. At the time of the fire, the house was completed except for the installation
of an air conditioning compressor. Francis testified that on the morning of the fire, he drove by
the house and recognized an employee of the defendant's unloading an air conditioning compressor
from one of the defendant's trucks. Francis knew that the air conditioning work was the only
work scheduled to be done on the house that day. When he returned several hours later, the house
was on fire. See id. at 694.
The plaintiff's theory was that the defendant was negligent in using an acetylene
torch to solder tubes from the compressor to tubes protruding from the side of the house. Francis
took photographs showing three- to four-inch tubes coming out of the exterior wall, and testified
that it was customary for such tubes to extend two feet. He testified that if shorter tubes are used,
flames from the torch could easily travel through a tube and ignite insulating materials on the
interior wall of the house, a problem that would be exacerbated by the layer of air between the
exterior brick of the house and the inside wall. The trial court granted an instructed verdict for
the defendant at the close of the plaintiff's testimony. See id. at 694-95.
The court of appeals affirmed, stating:
Negligence and causation may be established by circumstantial evidence. But
neither negligence nor causation will be presumed. Nor can one presumption be
based upon another presumption nor an inference of fact upon other
inferences. . . . The mere happening of damage by fire is no evidence of a defect
in the welding machine or that it was used so as to permit sparks to enter the home.
Id. at 695 (citations omitted). From having seen the defendant's employee unloading an air
conditioning compressor, together with the fact that no other work was scheduled for the day, the
plaintiff inferred that the employee's negligence caused the fire. However, the court noted that
the plaintiff failed to exclude the possibility that other persons were in the house during his
absence or that the fire was started in some way other than the means alleged. The court held that
the plaintiff failed to discharge its burden of proof that the defendant's employee was negligent and
that his negligence proximately caused the fire.
The Pickenses argue that John R. Francis Building Co. and Systech Financial are
distinguishable from the instant case because both were appeals from directed verdicts against the
plaintiffs, and therefore the plaintiffs had the burden of proof on appeal. We do not find this
distinction significant in the context of an evidentiary sufficiency review. Both cases are
instructive on evidentiary sufficiency in fire cases and generally support appellant's argument;
however, they address only the sufficiency of the evidence of negligence. Although negligence
is not an issue in the present appeal because the jury failed to find either defendant negligent, these
cases are still relevant because the Pickenses must link the fire's cause to Doyle Wilson to prevail.
Appellant also cites Leatherwood Drilling Co. v. TXL Oil Corp., 379 S.W.2d 693
(Tex. Civ. App.--Dallas 1964, writ ref'd n.r.e.), to support its argument. In that case,
Leatherwood had finished drilling an oil well and the rig was being serviced when a leak
developed, resulting in a spill of crude oil on the ground. Several days later, the well caught fire
and the rig was damaged. Leatherwood filed suit against TXL alleging negligence. Because there
was no direct evidence as to the cause of the fire, Leatherwood relied on the following evidence:
electric lights were left burning; valves on the casing head assembly were left open and electric
motors were left running; and no flow line was attached to the casing head to enable the leaking
oil to be carried away from the rig. The jury found that the fire was the result of unavoidable
accident. See id. at 695-96.
One of the issues on appeal was whether the trial court erred in refusing to submit
jury questions on whether the failure to turn off the motors was negligent. The court noted:
There is evidence that sparks could have come from the motors and could have
caused the fire. But this is nothing more than speculation. There was also evidence
that the fire could have been caused by a lighted match carelessly dropped, or a
burning cigarette thrown to the ground. Evidence that a happening is possible
cannot be accepted as evidence that it did happen.
Id. at 697. Because Leatherwood had offered no evidence that defendant's negligence caused the
fire, the court affirmed the trial court's refusal to submit negligence issues to the jury. See id.
Appellees also claim that Leatherwood is inapposite because in that case, "there was
evidence that the fire could have been caused by both a lighted match carelessly dropped, or a
burning cigarette thrown to the ground. Conversely, the Pickens' experts ruled out every other
conceivable cause of the fire and concluded that the fire was caused by the wiring." The problem
with this argument is that appellees equate proof that the fire started in the wiring with proof of
Doyle Wilson's liability. As John R. Francis Building Co. holds, causation may not be presumed.
See 517 S.W.2d at 695. The plaintiff must offer some evidence of what the defendant actually did
or failed to do to cause the fire--as opposed to mere conjecture about what might have
happened--to survive a challenge to the sufficiency of the evidence. See Bass v. General Motors
Corp., 447 S.W.2d 443, 447 (Tex. Civ. App.--Fort Worth 1968, writ ref'd n.r.e.). Although not
submitted on that theory, the Pickenses' case bears strong resemblance to res ipsa loquitur: the
fire appears to have started in an isolated area containing electrical wiring, therefore either the
builder or the electrician must be at fault. (10) The case law does not support this approach.
In Bass, the plaintiff parked his car and returned six hours later to find the car on
fire. Bass sued General Motors under several theories, including negligence and breach of
warranty, alleging that the vehicle's defective wiring caused the fire. See id. at 444. A fire chief
who had been at the scene testified that the fire was an electrical fire that resulted from a short
circuit. Bass testified that he pulled the electric wires loose from the car's horn on the day
preceding the fire, and had not reconnected the horn or done anything to safeguard the car from
the loosened live wires. Bass had never complained to the dealer about the electrical system, and
testified that the car operated normally after he disconnected the horn. The trial court granted the
defendant's motion for instructed verdict. See id. at 444-45.
The court of appeals observed:
In our opinion plaintiff proved nothing more than the fact that the automobile was
damaged as a result of an electrical fire. . . . There is no evidence in the record
that the wiring was defective, that it was defective when it left General Motors, that
the defect, if any, rendered the automobile unreasonably dangerous, or that any
defect was the proximate cause of the fire, nor is there any evidence of faulty
design. Proximate cause is not to be presumed. It must be proved by evidence.
Id. at 447. Later, the court added:
All the evidence shows is that a fire occurred. Plaintiff concedes the fact
that there is no evidence that the fire occurred because of any alleged defects, but
argues that the evidence shows in all probability the fire resulted from alleged
defects. Neither court nor jury could infer that the fire that did the damage was
caused by a defect in the electrical system because the proof failed to show any
defects existed in the electrical system.
Id. The court concluded that there was no evidence in the record to sustain the alleged breach of
an express warranty, and affirmed the instructed verdict for the defendant.
Bass is facially distinguishable from the present case on at least two grounds. First,
in Bass the plaintiff may have caused the fire by disconnecting the horn, whereas in the present
case, there is no evidence that the Pickenses were at fault in any way. However, this fact would
not appear to change the Bass court's conclusion that Bass "proved nothing more than the fact that
the automobile was damaged as a result of an electrical fire." Bass, 447 S.W.2d at 447. Second,
in Bass there was no expert testimony that the car's defective electrical wiring caused the fire; in
contrast, the Pickenses' experts concluded that a faulty installation caused the home fire, although
they conceded that a manufacturing defect remained a possibility. As we will discuss, however,
the jury's implicit rejection of both theories makes this distinction less meaningful.
Appellant also cites Leroy v. Texas Gulf Sulpher Co., 309 S.W.2d 550 (Tex. Civ.
App.--Galveston 1957, writ ref'd n.r.e.). In Leroy, the plaintiffs' decedents had been passengers
in a motorboat operated by the defendant's employee Daunie. Upon leaving the dock, the boat
was experiencing engine trouble; the boat was later found abandoned and burned to the waterline,
and the bodies of Daunie and the plaintiffs' decedents were recovered from the river. There was
no direct evidence as to what caused the fire. See id. at 552.
At trial, the plaintiffs' expert opined that the boat burned after an explosion from
a gasoline fire. Testimony from other witnesses established that Daunie's body was more badly
burned than those of the plaintiffs' decedents. From this evidence the plaintiffs' expert concluded
that Daunie had been working near the engine and created a spark that caused the explosion. The
court of appeals affirmed an instructed verdict for the defendant, noting that the expert's testimony
"is manifestly merely surmise or suspicion on the part of the witness and not fact." The only fact
in evidence, the court stated, was that the motorboat on which the plaintiffs' decedents had been
riding had burned; any opinion as to the cause of the fire rested "purely upon speculation and
surmise or upon inference." Id. at 554.
Appellees argue that Leroy is distinguishable from the present case because the
expert in Leroy admitted that the boat fire could have been caused by other means, and that he did
not know what had taken place. We disagree. In this case, Reynolds and Wagner similarly
acknowledged that they could not state precisely what caused the fire; they arrived at their
conclusion by theorizing various possible causes and using a process of elimination. Only minimal
physical evidence supports their conclusion. And contrary to appellees' claim, they failed to
eliminate all causes other than defective installation. We find Leroy more analogous than
distinguishable.
While numerous witnesses testified in the present case, the record reveals a lack of
evidence of the fire's cause. The Pickenses felt that the house had been constructed in a good and
workmanlike manner; they experienced no electrical problems and made no warranty claims
during the eighteen months they lived there. Their experts Reynolds and Wagner admitted that
they found no physical evidence of the fire's cause. Wagner stated that the overall electrical
workmanship was good. While they noted two installation defects in their report, it is undisputed
that neither imperfection caused the fire, and Wagner testified that neither constituted a violation
of the National Electric Code. In sum, the plaintiffs' evidence presents at best a weak inference
of causation that can be linked to Doyle Wilson.
Testimony from the defense witnesses tends to negate any suggestion of liability on
Doyle Wilson's behalf. Defense expert Ed Kirkham testified that any problems with the wiring
caused by the installation would likely have manifested themselves sooner than eighteen months;
that structural wiring is responsible for only seven percent of residential fires; and that in all
reasonable probability electrical problems did not cause the fire in the Pickenses' home. The
investigating firefighters John Golding and John Kiracofe were unable to determine an exact cause
of the fire. Terry Maxwell Electric's work on the Pickenses' house passed inspections by Doyle
Wilson and by City inspectors, among others. All the testimony indicated that the work was of
good quality. While Terry Maxwell testified that there were other tradespersons working around
the electrical wiring that he did not control, no witness testified that any of the tradespersons
actually damaged the wiring or were at fault in any way. No physical evidence suggests that one
of them caused the fire. Defense expert Michael Owen testified that it was not likely that the fire
resulted from damage to the wiring insulation. He also challenged Wagner's conclusions by
stating that it was more likely that the fire caused faulting in the electrical wiring, rather than the
other way around.
The jury's negative answer to the question of whether Terry Maxwell Electric
breached the contractor base agreement with Doyle Wilson is also revealing. In the contractor
base agreement, Terry Maxwell Electric warranted that all materials would be "of good quality"
and "free of defects." (12) The jury's failure to find a breach of this agreement was an implicit
finding that the electrical wiring used by Terry Maxwell Electric was of good quality and free of
defects, since any materials not in conformity were considered defective under the contract. While
not conclusive due to the conflicting nature of the verdict, the jury's answer to this question
indicates that the jury did not believe that either a manufacturing defect or improper installation
by Terry Maxwell Electric caused the fire.
While the process-of-elimination methodology used by the Pickenses' experts makes
the present case distinguishable from those cited above, the distinction is greater in form than in
substance. (13) The experts' testimony cannot be given much weight, since the jury rejected their
conclusion that the fire was caused by a "poor installation process" by failing to find that Terry
Maxwell Electric was negligent or breached the contractor base agreement. Moreover, the
plaintiffs' theory involves exactly the type of piling inferences on inferences or presumption upon
presumption that appellate courts have consistently rejected. One would have to infer from the
two sample defects that a sharp edge or staple damaged the wiring insulation, and then infer that
a fire resulted from this condition. "The circumstance that one of these surmises may correspond
with a similar surmise made by a jury cannot raise the surmise to anything more than a conjecture
which will not support a jury finding." Texas Sling Co. v. Emanuel, 431 S.W.2d 538, 541 (Tex.
1968). Looking at all the evidence in the record as whole, we find the evidence that Doyle Wilson
breached a warranty to perform services in a good and workmanlike manner, breached a warranty
of habitability, failed to comply with the home sales agreement, and committed DTPA violations
so factually weak as to be manifestly wrong and unjust. We sustain the factual-sufficiency
challenge appellant raises in points of error one through four.
On the other hand, we cannot say that there is no evidence of causation, i.e., legally
insufficient evidence. Looking at the evidence in the light most favorable to the jury's verdict,
we must accept Reynolds's and Wagner's testimony that they arrived at their conclusions by
scientifically eliminating all other possible causes. Wagner's testimony, taken in conjunction with
the faulted wiring samples entered in evidence and the two examples of wiring defects, can be
viewed as providing more than a scintilla of support for the plaintiffs' theory that the wiring was
defective or improperly installed. We believe that reasonable minds could differ on the issue of
Doyle Wilson's liability under the theories raised in points of error one through four. Doyle
Wilson's no-evidence arguments are therefore overruled.
Because we have sustained appellant's factual sufficiency argument, we do not reach
Doyle Wilson's conditional points of error five and six.
CONCLUSION
The Pickenses had the burden of proof in this case. When their house was
destroyed by fire, most of the physical evidence that might have indicated the cause of the fire was
destroyed as well. The plaintiffs attempted to overcome this obstacle through the testimony of two
expert witnesses, who used a process-of-elimination method to arrive at the conclusion that the fire
resulted from poorly installed wiring, although a manufacturing defect remained a possibility.
While there may be more than a scintilla of evidence against Doyle Wilson on the liability theories
advanced, after carefully weighing all the evidence we believe that it is factually insufficient to
support the jury's findings of liability. It would be manifestly unjust to allow the verdict against
Doyle Wilson to stand. We therefore reverse the trial-court judgment and remand this cause for
a new trial.
Bea Ann Smith, Justice
Before Justices Jones, B. A. Smith and Yeakel
Reversed and Remanded
Filed: May 27, 1999
Publish
1. We will use "Doyle Wilson" to refer only to the appellant corporation, not the individual
with that name.
2. The truss area was a concealed space between the first and second floors of the house about
eighteen inches high through which were laid heating and air conditioning vent pipes, as well as
electrical wiring circuits to and from the circuit breaker panel box.
3. In the electrical context, a "fault" is defined as "a defective point in an electric circuit due
to a crossing of wires, a ground, a break in the circuit, [or] a failure of insulation." Webster's
Third New International Dictionary 829 (Philip B. Gove ed., 1986). Our examination of the
record did not reveal that either expert provided the jury with any explanation for the term.
4. The report also mentions that a wire inside a breaker box had an inch-long slit in its
insulation, which Wagner described at trial as being "nicked." However, Reynolds and Wagner
both referred only to two defects in their report and generally throughout their respective
testimony at trial.
5. Wagner explained to the jury that a "fault bead" results when copper wire melts due to an
electrical condition (such as a short circuit) and resolidifies into a bead shape.
6. Reynolds also testified that neither a manufacturing defect nor an installation defect could
be ruled out; however, several times he disavowed having any expertise in electrical wiring, and
suggested that Wagner would be better equipped to answer questions pertaining to how the fire
might have started in the electrical wiring.
7. Question number ten, which asked the jury to determine the amount of the Pickenses' mental
anguish damages, was the only damages question submitted. Appellant did not object to evidence
that the Pickenses incurred $299,399.88 in actual damages, and that amount was apparently
stipulated. In any event, the amount of damages is not directly at issue on appeal.
8. See Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993); E.I. du Pont de Nemours
& Co. v. Robinson, 923 S.W.2d 549 (Tex. 1995).
9. The parties apparently agree that Wagner's testimony is scientific in nature, and not
"technical" or "other specialized knowledge" under Rule 702 of the Texas Rules of Evidence.
We therefore assume without deciding that this label is appropriate. However, even if we were
to view Wagner's testimony as non-scientific in nature, the Daubert-Robinson requirements of
reliability and relevance would still apply. See Gammill v. Jack Williams Chevrolet, Inc., 972
S.W.2d 713, 726 (Tex. 1998); see also Kumho Tire Co., Ltd. v. Carmichael, 119 S. Ct. 1167
(1999) (same result under Federal Rule 702).
10. For example, plaintiffs' counsel asked David Pickens:
Q: Mr. Pickens, is Defendant's 8 [a photograph showing the gutted remains of the
house] consistent with the representations made to you by Doyle Wilson
representatives about the workmanship and the habitability of your home?
A: Clearly not.
Plaintiffs' counsel echoed this theme in closing argument when he told the jury:
"[The defendants] had an obligation to put it [the house] in a good and workmanlike
fashion. They didn't. They had an obligation to put it in a habitable condition. That
fire gutted house is not habitable." (Emphasis added.) (11)
11. At trial, Wagner testified as follows regarding the procedure he and Reynolds used:
[The Pickens's] home was new, it was well kept, it wasn't cluttered, it wasn't
dirty, there were no nests--no birds' nests, insects, squirrels or anything like
that in the area, and we asked the owners before we left, you know, Have you
had any problems, and they said, No. Why are you asking this question? But I
mean, it was just a process of addressing potential causes so that--you know, so
it wouldn't appear that we were just moving in on one thing.
(Emphasis added.) We share some of the concerns about the reliability of the experts'
methodology voiced by appellant's counsel at the pretrial hearing. However, the trial court
found Wagner's methodology sufficiently reliable and relevant to be admissible under
Robinson, and on the limited record before us, we cannot say this was an abuse of discretion.
"
'
' "
12. The contractor base agreement specified that all materials used in the work "will be new
unless otherwise specified in the contract documents, of good quality, free of defects, and in
conformity with the contract documents. It is understood between the parties hereto that all
equipment and materials not so in conformity are defective."
13. At trial, Wagner testified as follows regarding the procedure he and Reynolds used:
[The Pickenses'] home was new, it was well kept, it wasn't cluttered, it wasn't dirty,
there were no nests--no birds' nests, insects, squirrels or anything like that in the
area, and we asked the owners before we left, you know, Have you had any
problems, and they said, No. Why are you asking this question? But I mean, it was
just a process of addressing potential causes so that--you know, so it wouldn't
appear that we were just moving in on one thing.
(Emphasis added.) We share some of the concerns about the reliability of the experts'
methodology voiced by appellant's counsel at the pretrial hearing. However, the trial court found
Wagner's methodology sufficiently reliable and relevant to be admissible under Robinson, and on
the limited record before us, we are satisfied that the court's ruling was not erroneous.
be ruled out; however, several times he disavowed having any expertise in electrical wiring, and
suggested that Wagner would be better equipped to answer questions pertaining to how the fire
might have started in the electrical wiring.
7. Question number ten, which asked the jury to determine the amount of the Pickenses' mental
anguish damages, was the only damages question submitted. Appellant did not object to evidence
that the Pickenses incurred $299,399.88 in actual damages, and that amount was apparently
stipulated. In any event, the amount of damages is not directly at issue on appeal.
8. See Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993); E.I. du Pont de Nemours
& Co. v. Robinson, 923 S.W.2d 549 (Tex. 1995).
9. The parties apparently agree that Wagner's testimony is scientific in nature, and not
"technical" or "other specialized knowledge" under Rule 702 of the Texas Rules of Evidence.
We therefore assume without deciding that this label is appropriate. However, even if we were
to view Wagner's testimony as non-scientific in nature, the Daubert-Robinson requirements of
reliability and relevance would still apply. See Gammill v. Jack Williams Chevrolet, Inc., 972
S.W.2d 713, 726 (Tex. 1998); see also Kumho Tire Co., Ltd. v. Carmichael, 119 S. Ct. 1167
(1999) (same result under Federal Rule 702).
10. For example, plaintiffs' counsel asked David Pickens:
Q: Mr. Pickens, is Defendant's 8 [a photograph showing the gutted remains of the
house] consistent with the representations made to you by Doyle Wilson
representatives about the workmanship and the habitability of your home?
Plaintiffs' counsel echoed this theme in closing argument when he told the jury:
"[The defendants] had an obligation to put it [the house] in a good and workmanlike
fashion. They didn't. They had an obligation to put it in a habitable condition. That
fire gutted house is not habitable." (Emphasis added.) (11)
11. At trial, Wagner testified as follows regarding the procedure he and Reynolds used:
[The Pickens's] home was new, it was well kept, it wasn't cluttered, it wasn't
dirty, there were no nests--no birds' nests, insects, squirrels or anything like
that in the area, and we asked the owners before we left, you know, Have you
had any problems, and they said, No. Why are you asking this question? But I
mean, it was just a process of addressing potential causes so that--you know, so
it wouldn't appear that we were just moving in on one thing.
(Emphasis added.) We share some of the concerns about the reliability of the experts