Justice WILLETT
delivered the opinion of the Court.
In this products liability case, Ford Motor Co. argues that the trial court reversibly erred in charging the jury by giving an incomplete definition of “manufacturing defect.” We agree. Additionally, we hold that a frequently submitted definition of “producing cause” should no longer be used. We remand the case for a new trial under a jury charge that reflects our applicable caselaw, including our decision today.
I. Background
In March 1999, Tiburicio Ledesma, Jr. purchased a new Ford F-350 Super Duty pickup truck for his construction business. The truck had four rear tires, two on each side, surrounded by fiberglass fenders extending beyond the sides of the truck.
On June 5, 1999, Ledesma turned onto a two-lane street in Austin and began to accelerate. He testified that after shifting gears the truck suddenly began to lurch, and he lost control, striking two parked cars, a Firebird and a Civic, on the side of the street. The truck then hit the street curb and came to rest. At the time of the accident, the truck’s odometer read about 4,100 miles.
Power from the truck engine is conveyed to the rear axle by the drive shaft, which connects the transmission in the front of the truck with the differential/rear axle assembly in the rear. As seen in the trial exhibit reproduced below, the rear-axle housing is attached to two sets of rear leaf springs by u-bolts, which wrap around the axle housing and are bolted to a rear spring plate that sits on top of the leaf-spring assembly. On each side of the truck, two u-bolts attach the rear-axle housing to a spring plate and set of leaf springs.
[[Image here]]
Both parties agree that the truck’s rear leaf spring and axle assembly came apart and that this separation caused the drive shaft to dislodge from the transmission. The core dispute centers on
when
and
why
this malfunction occurred and whether it
caused
the collision or
resulted
from it. That is, did a manufacturing defect trigger the right rear-axle displacement and cause Ledesma to lose control of the truck and strike the parked cars (as Ledesma claims), or did the right rear axle detach when Ledesma struck the parked cars and curb (as Ford claims)?
At trial, Ledesma claimed that he lost control of the truck when its drive shaft separated from the transmission and “pronged” on the pavement, causing him to hit the parked cars. A police officer testified that he investigated the accident scene and prepared a report based on Ledesma’s description of the accident. The report makes no mention of any other witnesses. Ledesma also presented two expert witnesses in support of his manufacturing defect claim, as discussed below.
Ford presented an expert, Dan May, in support of its theory that the axle-to-spring attachment failed, not because of a manufacturing defect, but because of the forces exerted on it when Ledesma struck the parked vehicles and curb. Among other efforts to discredit May, Ledesma emphasized to the jury that May was a longtime Ford employee and had never found a defect in a Ford product.
Ford also called the owner of the Fire-bird, Edward Plyant, who testified by deposition that he witnessed the accident from a driveway. Plyant testified that Le-desma was speeding and inattentive and struck the Firebird at a high rate of speed. Ledesma testified that Plyant did not see the accident, but came outside after hearing the ensuing commotion, and that Plyant had unsuccessfully sued Ledesma.
The jury sided 11-1 with Ledesma, finding that a manufacturing defect caused the accident and that Ledesma was not con-tributorily negligent, and awarding economic damages of $215,380. The court of
appeals affirmed.
II. Discussion
A. Admissibility of Ledesma’s Expert Testimony
Ford argues that the trial court erred in admitting the testimony of Ledesma’s two expert witnesses because their opinions were unreliable.
1. What Ledesma’s Two Experts Said
Ledesma’s expert David Hall, an accident reconstructionist, testified by deposition that he believed a mark in the road showed that the drive axle struck the pavement before the truck struck the Fire-bird, consistent with Ledesma’s theory that the truck malfunctioned and caused the collision. He reached this conclusion by reviewing a number of photographs. Based on the photographic evidence of damage to the Firebird and an engineering paper providing a method for estimating speed based on the Firebird’s body damage, Hall also estimated that Ledesma’s truck was traveling at a very slow speed when it struck the Firebird, again consistent with Ledesma’s theory of the accident and inconsistent with Ford’s theory that Ledesma was speeding and otherwise driving carelessly.
Ford raises numerous complaints about the reliability of Hall’s testimony. Ford notes that the post-accident photographs of the road were taken by Ledesma and his father with a low-quality disposable camera, that there are many spots and marks on the grainy photographs, and that the marks do not reliably indicate that the drive shaft struck the road prior to the crash with the Firebird. Ford stresses that the investigating police officer found no gouges in the pavement despite a careful search of the accident scene. Ford also raises numerous complaints about Hall’s estimate of the truck’s speed at the time of the Firebird collision. These complaints include two observations: (1) the engineering paper on which Hall relied states that it should not be used when examining side swipes such as the accident in issue; and (2) the damage to the Firebird, as confirmed by the car’s owner and a repair estimate, was far more severe than Hall assumed, rendering his estimate of a very slow impact unreliable.
Ford complains that the testimony of Ledesma’s principal expert, metallurgical and mechanical engineer Geert Aerts, was likewise unreliable. Ledesma counters that Aerts is a leading expert on truck leaf springs; Aerts has investigated about 150 leaf-spring failures. Ford does not claim that Aerts was unqualified to testify as an expert on the accident in issue but contends that his theory of the accident was unreliable for various reasons.
Aerts focused on the u-bolts holding the rear axle to the rear springs and concluded that one of them was defectively manufactured. His theory was that a rear passenger-side u-bolt was loose, causing it to vibrate. This vibration eventually fatigued and fractured a center pin holding the leaf springs to the rear spring plate, which in turn allowed the u-bolts, spring plate, and rear axle to slide backwards, pulling the drive shaft from the transmission. Aerts examined the truck’s u-bolts while they were still on the truck and found the tor
que on each bolt to be well below that required by Ford’s specifications. While he conceded that the torques might have changed due to the accident, he concluded that the u-bolt in issue was under-torqued when Ford assembled the truck.
Aerts testified extensively about how the u-bolt came to be under-torqued. One explanation was that one leg of the u-bolt was manufactured a few millimeters higher than the other, which could have caused the nut on the shorter leg to receive an insufficient torque when it and other nuts were simultaneously tightened. Aerts claimed that, according to a Ford “product engineering” sheet for the u-bolt, admitted into evidence, this slight difference in heights of the legs exceeded the tolerance in Ford’s own specifications. The engineering sheet provided diagrams and various specifications for the u-bolts.
Aerts also testified that the product engineering sheet required the bottom, curved portion of the u-bolt to be flattened to allow for a better grip on the axle, and that the flattened portion on the allegedly defective bolt was off center by about three-quarters of an inch. According to Aerts, this manufacturing defect reduced the surface area of the bolt in contact with the axle, thus reducing the bolt’s grip on the axle. This defect also allegedly allowed the bolt to dig into and deform the axle housing, causing the bolt to further loosen over time. The product engineering sheet clearly shows that the flattened portion of the u-bolt is to be centered at the bottom of the curved portion of the bolt, with flattening forty-five degrees up from either side of the bottom of the bolt, and a blend from flat to round for another forty-five degrees on both sides of the bolt.
As further evidence of a pre-accident defect, Aerts focused on the rear spring plate. The u-bolts went through this plate, which sits on top of the leaf springs. The u-bolts were tightened onto the plate with four nuts. The rear spring plate also had a center pin attaching the leaf springs to the plate. Aerts claimed that rust and marks on the spring plate near the hole for the center pin indicated that the u-bolt was loose, allowing the assembly to vibrate and the pin to scratch the spring plate over a period of time. These observations were consistent with his theory of a u-bolt that was loose before the accident, resulting in vibration that caused the center pin to suffer a fatigue fracture and precipitate the u-bolt slippage and drive shaft dislocation.
Aerts also theorized that, perhaps for a reason other than the uneven u-bolt ends and flattened portion of the bolt, Ford simply failed to tighten the u-bolts to the torque required by its own specifications, causing the assembly to vibrate and eventually fail after a few months of driving.
Ford challenged Aerts’s testimony at every turn. It argued that Aerts never confirmed the existence of denting in the axle housing and that Ford’s expert May testified that he could find no such denting. Ford contended that the u-bolt was deformed because of the tremendous forces of the accident. It further offered evidence from May, who was familiar with Ford’s manufacturing process, that the sockets on its “four spindle nut runner,” the machine used to tighten the nuts onto the u-bolts, are four inches deep, easily deep enough to properly tighten the nuts despite the slight difference in the height of the u-bolt legs. May claimed that the difference in the u-bolt lengths was within Ford’s “allowable tolerances.” May also testified that he could not see the .75-inch misalignment of the flattened portion of the u-bolt that Aerts observed.
2. Whether Their Testimony Was Sufficiently Reliable
An expert’s testimony, to be admissible, must possess a rehable foundat
ion.
“Admission of expert testimony that does not meet the reliability requirement is an abuse of discretion.”
Expert testimony is unreliable if it is based on unreliable data, or if the expert draws conclusions from his underlying data “based on flawed methodology.”
Expert testimony is also unreliable if “there is simply too great an analytical gap between the data and the opinion proffered.”
In
E.I. du Pont de Nemours & Co. v. Robinson,
923 S.W.2d 549 (Tex.1995) we set out six factors courts may consider in deciding whether expert testimony is reliable:
1. the extent to which the theory has been or can be tested;
2. the extent to which the technique relies on the subjective interpretation of the expert;
3. whether the theory has been subjected to peer review and/or publication;
4. the technique’s potential rate of error;
5. whether the underlying theory or technique generally has been accepted as valid by the relevant scientific community; and
6. the nonjudicial uses which have been made of the theory or technique.
We have recognized that the
Robinson
factors are nonexclusive
and have specifically noted that “the
Robinson
factors are not always useful in evaluating expert testimony in automobile accident cases.”
As in
Gammill v. Jack Williams Chevrolet, Inc.,
another automobile accident case, the
Robinson
factors do not readily lend themselves to a review of the expert testimony in the pending case,
but “there must be some basis for the opinion offered to show its reliability.”
And as in
Volkswagen of America, Inc. v. Ramirez,
yet another automobile accident case, “[a]n expert’s bare opinion will not suffice” and is unreliable if “based solely upon his subjective interpretation of the facts.”
We need not assess the reliability of Hall’s testimony, because we hold that Aerts’s testimony was sufficiently rehable to warrant admission and is some evidence that a manufacturing defect caused the accident.
Ford never challenged Aerts’s credentials. His testimony offered a plausible theory of how the accident occurred. Aerts based his testimony on deviations between Ford’s own specifications as to torque, alignment of the flattened portion of the u-bolt, and the tolerance for the difference in u-bolt leg lengths. He relied on observations and measurements of tangible truck components, which were documented with photographs and a videotape of his initial examination of the truck. The u-bolts and rear-axle assembly were admitted into evidence. The jury was free to examine this evidence and the Ford
specifications that were also admitted into evidence.
To a significant extent, deciding whether the truck lost its drive shaft
before
striking the Firebird, or instead the drive shaft dislodged
after
striking the parked ears and the street curb, turned on whether the jury believed Ledesma or Plyant. Ledes-ma insisted he was driving carefully when suddenly he lost control of the truck and veered into the Firebird. He claims that the truck started lurching before the impact, consistent with a truck malfunction that caused the drive shaft to drop to the street. Pylant’s testimony was inconsistent with the investigating police officer’s report and testimony, which indicated that Plyant talked to the officer but did not claim to have witnessed the accident. Le-desma likewise observed no witnesses to the accident, claiming instead that Plyant and others came outside after hearing the accident. The jury was of course entitled to weigh the credibility of these lay witnesses and apparently believed Ledesma’s account. Thé jury was also entitled to take into account that Ford’s expert May had worked for Ford for decades and had never found a defect in a Ford product.
We also conclude that Ford overstates the effectiveness of its cross-examination of Aerts. While May explained that Ford’s tightening equipment provided sockets easily deep enough to tighten the uneven legs of the u-bolt, Aerts claimed that the uneven legs might still cause the equipment to slip off the shorter leg, or under-tighten the shorter leg if the equipment lacked the flexibility to properly tighten uneven legs, although May contended that each socket was “on the same head” but “driven independently.” Aerts conceded that the u-bolt could have been stretched in the accident, perhaps accounting for the inadequate torques he measured, but he also found that the overall length of the allegedly defective u-bolt was the same length as exemplar u-bolts provided by Ford and was within one millimeter of the length of another u-bolt from the truck. He conceded that he had no photographic proof of indentation of the axle caused by the asymmetrical flat portion of the u-bolt and that “I never came back here to check that,” but he opined that the asymmetrical manufacture, regardless of indentation of the axle housing, reduced the area of contact between the u-bolt and axle. The jury could infer that Ford would not have specified flattening of the curved end of the u-bolt, which indisputably required an extra manufacturing step, unless Ford thought that procedure was needed. Aerts also testified that regardless of the shape of the u-bolt, Ford might have failed to tighten the u-bolt nuts to Ford’s own torque specification because the air pressure on the tightening equipment was incorrect or for some other reason, thus explaining the inadequate torque and vibration damage he observed. His review of the accident scene photographs further persuaded him that the accident did not cause the u-bolt slippage and to opine that “I don’t think [the u-bolt] should have come loose with what I saw happen there on that street.”
Ford presented a strong defense, but ultimately the jury rejected it. Aerts’s testimony does not present a case where “there is simply too great an analytical gap between the data and the opinion proffered,”
or where the expert’s testimony amounted to nothing more than a recitation of his credentials and a subjective opinion.
We conclude that Ford’s complaints about Aerts’s testimony go to its
weight, not its admissibility. The trial court therefore did not abuse its discretion in admitting Aerts’s testimony.
B. Jury Charge
Ford complains that, over its objection, the trial court improperly instructed the jury on the definitions of manufacturing defect and producing cause. The jury affirmatively answered Question No. 1 of the jury charge, which asked:
Was there a manufacturing defect in the 1999 Ford F-350 pickup truck at the time it left Ford’s possession that was a producing cause of the June 5, 1999 incident in question?
A “defect” means a condition of the product that renders it unreasonably dangerous. An “unreasonably dangerous” product is one that is dangerous to an extent beyond that which would be contemplated by the ordinary user of the product, with the ordinary knowledge common to the community as to the product’s characteristics.
“Producing cause” means an efficient, exciting, or contributing cause that, in a natural sequence, produces the incident in question. There may be more than one producing cause.
In defining defect, the trial court followed Texas Pattern Jury Charge (PJC) 71.3.
As specified in the comment to PJC 71.3, the trial court included in the question the definition of producing cause found in PJC 70.1. Ford objected that both PJC 71.3 and PJC 70.1 were “not accurate under the law” and failed to track this Court’s precedent. We agree. Le-desma may have
argued
a manufacturing defect to the jury, but the law requires the jury to determine specifically whether he had
proven
one. The jury here received a legally incorrect charge that omitted an indispensable element: that the product deviated, in its construction or quality, from its specifications or planned output in a manner that rendered it unreasonably dangerous.
1. The Court’s Charge on Manufacturing Defect Failed to Include the Essential Element of a Deviation from Design
The trial court submitted the pattern jury charge’s definition on manufacturing defect. We agree with Ford, however, that the model charge is erroneous, as it does not include the requirement that a manufacturing defect must deviate from its specifications or planned output in a manner that renders the product unreasonably dangerous. We established this standard in
American Tobacco Co. v.
Grinnell,
and in three other cases since
Grinnell
was issued a decade ago, we have recognized, with essentially identical statements, the “deviation from specifications or planned output” requirement.
This
requirement is separate from, and in addition to, the requirements that the product was defective when it left the manufacturer and that the defect was a producing cause of the plaintiffs injuries.
We note that the current Restatement of Torts essentially follows the same concept of a deviation from the manufacturer’s design by providing that a product “contains a manufacturing defect when the product departs from its intended design even though all possible care was exercised in the preparation' and marketing of the product....”
The requirement of a deviation from the manufacturer’s specifications or planned output serves the essential purpose of distinguishing a manufacturing defect from a design defect. PJC 71.3 refers to a “manufacturing defect” in the product “at the time it left” the manufacturer. A jury — without further guidance — may view any defect in a product at the time it leaves the manufacturer as satisfying the PJC’s reference to a “manufacturing defect,” rather than making the essential distinction between a manufacturing and design defect. As it stood, the court’s charge merely inquired whether a “condition” of the product rendered it unreasonably dangerous. That “condition” could have been a design defect or a manufacturing defect.
The distinction is material. The danger of allowing a jury to conclude that the defect was or might have been a design defect is that “[a] design defect claim requires proof and a jury finding of a safer alternative design.”
The charge did not make such an inquiry.
Moreover, requiring a deviation from specifications or planned output permits a jury to determine whether a specific defect caused the accident, rather than premising liability on a belief that a product failure, standing alone, is enough to find a product defect. Texas law does not generally recognize a product failure or malfunction, standing alone, as sufficient proof of a product defect.
Instead, we have held that “a specific defect must be identified by competent evidence and other possible causes must be ruled out.”
Our law requires more than finding an undifferentiated “condition” that renders the product unreasonably dangerous, which is all the court’s charge mandated. While a products liability claim does not of course require proof of manufacturer negligence, the deviation from design that caused the injury must be identified. Otherwise, the jury is invited to find liability based on speculation as to the cause of the incident in issue.
Requiring proof of a deviation from manufacturer specifications or planned output also comports with our recognition that expert testimony is generally encouraged if not required to establish a products liability claim.
If juries were
generally free to infer a product defect and injury causation from an accident or product failure alone, without any proof of the specific deviation from design that caused the accident, expert testimony would hardly seem essential. Yet we have repeatedly said otherwise.
For these reasons, we hold that the court’s charge was fundamentally flawed in omitting the requirement that the product deviate, in its construction or quality, from its specifications or planned output in a manner that renders it unreasonably dangerous.
2. The Erroneous Definition of Manufacturing Defect Is Reversible Error
Having determined that the court’s charge was erroneous, there remains the question of whether this charge error is reversible error.
Ledesma contends that Ford did not preserve error on this complaint. Generally, “[a] party objecting to a charge must point out distinctly the objectionable matter and the grounds of the objection.”
Preservation of error generally depends on “whether the party made the trial court aware of the complaint, timely and plainly, and obtained a ruling.”
“Failure to submit a definition or instruction shall not be deemed a ground for reversal of the judgment unless a substantially correct definition or instruction has been requested in writing and tendered by the party complaining of the judgment.”
In this case, Ford objected to the definition of manufacturing defect on grounds that it did not accurately reflect the law, and submitted in writing a proposed question that included an alternative definition of a manufacturing defect. The alternative definition provided that a manufacturing defect “is a physical departure from the product’s intended design that renders it unreasonably dangerous.” The trial court refused Ford’s alternative charge.
The proposed definition addresses the essential omission in the court’s charge described above, namely, the omission of a requirement that the product deviate from its intended design. The proposed definition is consistent with the Restatement definition of a manufacturing defect, discussed above, that includes the requirement that the “product departs from its intended design,” as well as our own case-law. In support of its proposed definition, Ford also provided the trial court with quotations from
Grmnell
and
Torrington Co. v. Stutzman,
giving their definitions of a manufacturing defect that include the requirement of a deviation from specifications or planned output.
The objection, proposed question and instruction, and supporting authorities provided the trial court with a plain objection identifying the
error in the charge that we recognize today, “with sufficient specificity to make the trial court aware of the complaint.”
Since Ford preserved error, we turn to whether the charge error is reversible. We hold that it is. “It is fundamental to our system of justice that parties have the right to be judged by a jury properly instructed in the law.”
“There can be no question that it was [the plaintiffs] burden to obtain affirmative answers to jury questions as to the necessary elements of his cause of action.”
The jury was not asked to decide an essential element of a manufacturing defect claim, namely whether the u-bolt deviated from Ford’s specifications or planned output.
If a cause of action consists of more than one element, and an element is omitted from the charge “without request or objection,” the missing element can be found by the trial court or deemed found if certain requirements are met.
But where, as here, a proper objection is made about the omission of an essential element, the failure to include it is reversible error.
Ford argues that we should reverse and render judgment in its favor because of its objected-to charge error, as we did in
State Department of Highways & Public Transportation v.
Payne,
but we disagree. In
Spencer v. Eagle Star Insurance Co. of
America,
we distinguished
Payne
and explained that where, as in the pending case, the theory of recovery was defectively submitted, as opposed to a situation where the plaintiff “refused to submit a theory of liability” after defendant’s objection, the proper remedy is to remand for a new trial.
Spencer
distinguished cases that reversed and rendered where the plaintiff “failed to submit any jury question on a controlling issue.”
We further explained, in
Spencer
and in
Stutzman,
that if the plaintiff submits a jury question on his claim that is merely “defective,” as opposed to “immaterial,” the appropriate remedy is to remand for a new trial rather than to render judgment.
In the pending case, Ledesma proffered a question on his manufacturing defect claim based on the PJC for this cause of action. He “attempted to submit a controlling issue”
on his sole cause of action, and the question submitted “was the heart of [Ledes-
ma’s] case,”
so the question cannot be characterized as immaterial, and we cannot say that Ledesma failed to submit a jury question on a controlling issue.
Further, we “may, in the interest of justice, remand the case to the trial court even if a rendition of judgment is otherwise appropriate.”
“[W]e have remanded in the interest of justice when our decisions have altered or clarified the way in which a claim should be submitted to the jury.”
In the pending case, the trial court followed the PJC. On one occasion we not only approved a PJC issue and instruction for design defect cases, we expressly disapproved of the use of any other instructions in such cases,
prompting one court of appeals to remark that “[o]ur highest court has made it abundantly clear that to deviate from the pattern jury charges in products liability cases is a perilous journey.”
Given that our trial courts routinely rely on the Pattern Jury Charges in submitting cases to juries,
and we rarely disapprove of these charges, we conclude that the interests of justice would not be served by reversing and rendering judgment in favor of Ford. The more appropriate remedy is to reverse and remand for a new trial.
3. Producing Cause
Ford separately complains that the trial court improperly instructed the jury on producing cause. The trial court, following PJC 70.1, instructed the jury: “ ‘Producing cause’ means an efficient, exciting, or contributing cause that, in a natural sequence, produces the incident in question. There may be more than one producing cause.” Ford contends that this definition is an incorrect statement of Texas law, and that a valid definition would state that producing cause “means that cause which, in a natural sequence, was a substantial factor in bringing about an event, and without which the event would not have occurred. There may be more than one producing cause.” Ford requested the trial court to use this definition.
We agree with Ledesma that the second part of the court’s definition, recognizing that there may be more than one producing cause of an event, is correct. And we have seemed to sanction the first part, employing it ourselves several times in describing producing cause.
But we have also described a producing cause as one “that is a substantial factor that brings about injury and without which the injury would not have occurred,”
the definition Ford asks us to adopt.
To say that a producing cause is “an efficient, exciting, or contributing cause that, in a natural sequence, produces the incident in question” is incomplete and, more importantly, provides little concrete guidance to the jury. Juries must ponder the meaning of “efficient” and “exciting” in this context. These adjectives are foreign to modern English language as a means to describe a cause, and offer little practical help to a jury striving to make the often difficult causation determination in a products case.
Defining producing cause as being a substantial factor in bringing about an injury, and without which the injury would not have occurred, is easily understood and conveys the essential components of producing cause that (1) the cause must be a substantial cause of the event in issue and (2) it must be a but-for cause, namely one without which the event would not have occurred. This is the definition that should be given in the jury charge.
III. Conclusion
We reverse the court of appeals’ judgment and remand this case to the trial court for further proceedings.