This is a products liability action arising from a single-vehicle car accident involving a 1958 Volkswagen van designed and distributed by appellees Volkswagenwerk Aktiengesellschaft and Volkswagen of America, Inc. (collectively, "VW"). Appellants, the Roy family, were seriously injured when their 1958 VW van, driven by Doug Roy, Jr., rolled and caught fire. The Roys filed suit in the United States District Court against VW alleging a design defect in the van. The district court had jurisdiction under 28 U.S.C. Sec. 1332.

A jury trial was held after the district court denied the Roys' motion to bifurcate the trial of liability and damages. The jury returned a unanimous verdict for the Roys in the amount of $3 million.1 The jury assessed 60% of the fault to VW and 40% to Mr. Roy, finding that there was a defect in the design of the van and that the defect was a proximate cause of the injuries. The jury considered and refused an award of punitive damages.

Subsequently, the district court granted VW's motion to set aside the verdict and ordered a new trial. Roy v. Volkswagenwerk Aktiengesellschaft, 600 F. Supp. 653 (C.D.Cal.1985). The Roys appealed this order, but we dismissed the appeal because the order was interlocutory and, thus, not appealable. Roy v. Volkswagenwerk Aktiengesellschaft, 781 F.2d 670 (9th Cir.1985).

On return to the district court,2 the trial was bifurcated. The trial on liability resulted in a unanimous jury verdict for VW and judgment was entered accordingly. The Roys appeal the decisions in both trials, asserting that the jury verdict at the first trial was wrongly set aside and that the second trial was infected by various errors. Concluding that the district court erred in granting the new trial, we reverse and reinstate the original jury verdict for the Roys.

On August 14, 1982, the Roys embarked in their 1958 VW van on a family trip from their home in California to Texas. During the trip, the van experienced periodic mechanical problems related to the fuel line that caused the van to run roughly. Mr. Roy, who was experienced in repairing automobiles, replaced various parts hoping to remedy the problem. Though each repair initially seemed to make the van run well, the problem soon recurred.

At nightfall on August 16, 1982, the van again began to display symptoms of fuel deprivation and was losing speed. After examining the engine, Mr. Roy determined that the flow of fuel from the fuel line was restricted. Mr. Roy decided it best to make a temporary repair to the vehicle to allow the family to proceed to their destination where he could obtain the needed parts. He removed the cap from the fuel fill pipe, plugged the opening in the fuel line and connected a make-shift fuel line by running a hose from the fuel pump into the fuel tank through the filler neck. Then, he wrapped a rag tightly around the hose in the filler neck to create a seal. After Mr. Roy's temporary repair, the van ran smoothly until the time of the accident.

What happened next is somewhat unclear. At approximately 8:00 a.m. on August 17, 1982, Mr. Roy was driving east on Interstate 10 just outside of Sonora, Texas. At that location, Interstate 10 is a level road with two lanes going in each direction. There is a paved shoulder which drops two to three inches to an unpaved coliche surface. Mr. Roy recalls driving in the slow lane at approximately 50-55 miles per hour. Suddenly, for reasons hotly disputed by the parties, the van overturned and began to roll. During the roll, the van caught fire. Each of the Roys, especially Doug Roy III, suffered significant injuries and trauma as a result of the accident.

The judgment of the district court in the second trial is a final judgment and is appealable to this court under 28 U.S.C. Sec. 1291. The interlocutory order granting the second trial is now subject to review. See Evers v. Equifax, Inc., 650 F.2d 793, 796 (5th Cir.1981).

"The trial court may grant a new trial, even though the verdict is supported by substantial evidence, if 'the verdict is contrary to the clear weight of the evidence, or is based upon evidence which is false, or to prevent, in the sound discretion of the trial court, a miscarriage of justice.' " Hanson v. Shell Oil Company, 541 F.2d 1352, 1359 (9th Cir.1976), cert. denied, 429 U.S. 1074, 97 S. Ct. 813, 50 L. Ed. 2d 792 (1977) (citing Moist Cold Refrigerator Co. v. Lou Johnson Co., 249 F.2d 246, 256 (9th Cir.1957). While the trial court may weigh the evidence and credibility of the witnesses, the court is not justified in granting a new trial "merely because it might have come to a different result from that reached by the jury." Wilhelm v. Associated Container Transportation (Australia) Ltd., 648 F.2d 1197, 1198 (9th Cir.1981). We review an order granting or denying a motion for new trial for an abuse of discretion. Air-Sea Forwarders, Inc. v. Air Asia Company Ltd., 880 F.2d 176, 190 (9th Cir.1989). Thus, if the jury's verdict is not clearly against the weight of the evidence, the trial court abuses its discretion in ordering a new trial. McGhee v. Arabian Am. Oil Co., 871 F.2d 1412, 1420 (9th Cir.1989).

B. The District Court's Grant of a New Trial

In granting a new trial, the district court ruled that the jury's verdict finding a design defect in the 1958 VW van was against the great weight of the evidence. See Roy v. Volkswagenwerk Aktiengesellschaft, 600 F. Supp. 653, 655 (C.D.Cal.1985). We disagree.

The district court correctly instructed the jury at the first trial that:

A product is defective in design if it fails to perform as safely as an ordinary consumer would expect when used in an intended and reasonably foreseeable manner, or if there is a risk of danger inherent in the design which outweighs the benefits of that design.

Barker v. Lull Engineering Co., 20 Cal. 3d 413, 429-30, 432, 143 Cal. Rptr. 225, 235-38, 573 P.2d 443, 453-56 (1978). After hearing the evidence, the jury determined that the Barker test had been satisfied with respect to the VW van. The trial judge rejected this finding, relying heavily on the long use and popularity of the VW van. Though relevant, consumer acceptance alone does not insulate VW from a finding that its van is defectively designed. See Culpepper v. Volkswagen of America, Inc., 33 Cal. App. 3d 510, 518, 109 Cal. Rptr. 110, 115 (1973) (suspension system of the VW "bug", which enjoyed wide consumer acceptance, found defectively designed where car had propensity to roll over). Long use and popularity of the VW van therefore do not alone justify the granting of the new trial. We turn, then, to the evidence presented at the first trial concerning the design of the VW van and the circumstances surrounding the accident.

The expert evidence presented at the first trial concerning the design of the VW van was extensive and contradictory. The Roys contended that the swing-axle suspension system employed in the 1958 VW van was unstable and defective. William P. Thomas, who directed the race car programs for the Chevrolet Division of General Motors from 1956 to 1970, testified that when the van is turned, the VW suspension caused the back of the van to rise up and the wheels to drop. This reaction is known as "jacking." The jacking exacerbated the high center of gravity and narrow wheel base that makes a van inherently more unstable than other vehicles. Dr. Igor Paul, a tenured associate professor of mechanical engineering at the Massachusetts Institute of Technology, also testified that the van would jack when turned. Dr. Paul suggested that jacking made the van susceptible to rolling over under circumstances that would not have caused similar American vans to roll.

The Roy's also presented to the jury the results of a "bunji cord" test developed by Paul O'Shea to monitor the reaction of vehicles when turned sharply.3 Mr. O'Shea's test showed that the VW van would roll when turned sharply at between 18.2 and 18.4 miles per hour, whereas a 1967 Chevrolet van would not roll under the same conditions.

Dr. Paul and Mr. O'Shea further testified that the 1958 VW van had a tendency to change abruptly from "understeer" to "oversteer." During an evasive maneuver, this tendency could cause the van to turn more sharply than the driver intended or expected, compounding the instability of the van.

VW's experts challenged the credibility and conclusions of the Roys' experts. Some testified that jacking and the change from understeer to oversteer did not make the van less stable. In fact, Edward Heitzman, an automotive test engineer, testified that several top sports cars are designed to change from understeer to oversteer at high G's and that this characteristic was helpful in an emergency. VW also showed the jury a videotape of tests it performed on its own van and comparable American vans. These tests indicated that the VW van was no less stable than other vans on the market. In short, each side presented persuasive evidence in support of its position.

2. Was the Design Defect a Proximate Cause of the Accident?

Not only was the existence of a design defect at issue in this trial; the parties also disputed whether the alleged defect caused the accident. Again, the evidence was contradictory. Appellants contended that the driver, Mr. Roy, was traveling at approximately 55 miles per hour when he turned sharply to the left to avoid a dog, causing the van to roll. They claimed that the van never left the pavement and that any vehicle that would roll on a smooth pavement under those circumstances did so as the result of a design defect. Mr. Roy and two eye witnesses, Wade Fitzgerald and Jerry Jardee, all testified that the van never left the pavement.

VW painted a different picture. It contended that Mr. Roy fell asleep while driving and drifted onto the unimproved portion of the highway. When he awoke, he turned abruptly to the left to return to the road. The combination of the ruts created by the wheels digging into the dirt and the edge of the pavement tripped4 the van. Both parties agree that the fact that a vehicle rolls after being tripped is not indicative of a design defect.

VW introduced evidence to impugn the testimony of the Roys' witnesses concerning where the roll began. They cited testimony by Mr. Roy indicating that he did not know whether the vehicle ever left the paved surface. Further, Carr Thomson, a professional surveyor and licensed civil engineer in the State of Texas, testified that, from their distant position in an oncoming lane, Mr. Fitzgerald and Mr. Jardee could not have observed whether the van left the paved surface of the highway. He suggested that the elevation of the highway and the crown of the surface of the roadway made it impossible to observe the surface of the highway, the lane markings on the highway, the adjoining paved shoulder, or the adjoining gravel shoulder. The Roys did not dispute that the highway markings were not visible from the vantage point of the eye witnesses, but contended nevertheless that Mr. Fitzgerald and Mr. Jardee could have determined in which lane the van was traveling.

VW also presented the testimony of Dr. John Habberstad who purported to reconstruct what happened during the accident. In developing his accident reconstruction, Dr. Habberstad relied on the notes of John Ferguson, a Texas trooper who investigated the accident, and on photographs taken by Raymond Gesch, the Chief of Police for Sonora, Texas. Dr. Habberstad's reconstruction supported VW's theory of the accident.

The Roy's, in turn, challenged the validity of Dr. Habberstad's reconstruction. They cited Dr. Habberstad's testimony that he considered only that evidence supporting his theory and ignored that which did not. They further noted that Trooper Ferguson's testimony at trial was substantially different from his deposition testimony where he stated, "I believe [the van] began to overturn on the white shoulder stripe which divides the improved shoulder with the main traveled portion of the roadway." He did not change his opinion to conclude that the van rolled while on the unimproved shoulder until after conferring with Dr. Habberstad.

In addition, the parties disputed the cause of the fire. VW's experts testified that the van's fuel system was well-designed, that Mr. Roy's repairs caused the fire, and that there would have been no fire had the gas cap been in place. The Roy's experts concluded that the van's tap valve was susceptible to clogging, that Mr. Roy's actions were a good temporary repair, and that the fire would have occurred even with the gas cap in place. Again, both sides presented significant evidence to support their theories.

The district court also based its granting of a new trial on its perception that the jurors were swayed by emotion and sympathy for Doug Roy III. It pointed to no evidence, however, to support this ruling. Certainly a jury presented with a case like this one, involving horrible burn injuries to a child, will be likely to feel great sympathy. The question, however, is whether the jury was able to put aside its personal feelings during its deliberations and decide the case as the evidence and the law dictated. There is no evidence that the jury here did otherwise. The district court repeatedly instructed the jury that sympathy could play no role in their deliberations. Moreover, the damages awarded were consistent with the amount the experts estimated would be required to compensate the Roys for their injuries and treatment. Punitive damages were denied. In light of the conflicting evidence, the verdict for the Roys does not necessarily indicate that the jury's decision was born of sympathy.

VW presented weighty evidence to the jury to counter the Roys' claims that a defect in the design of the VW van was the primary cause of the Roys' injuries. Nevertheless, the Roys' also strongly supported their claims. "[I]t is the function of the jury as the traditional finder of the facts, and not the Court, to weigh conflicting evidence and inferences, and determine the credibility of witnesses." Boeing Company v. Shipman, 411 F.2d 365, 375 (5th Cir.1969) (en banc). Although under the evidence the jury could have concluded that the van was not defective, or that the defect was not a proximate cause of the accident, this possibility does not warrant granting a new trial. The evidence in this case was substantially balanced. VW's evidence did not so preponderate that it fell within the discretion of the trial judge to rule that the jury's verdict was against the great weight of the evidence.

The District Court abused its discretion in setting aside the jury verdict in the first trial and ordering a new trial. We therefore reinstate the jury verdict in the first trial in favor of the Roys. In light of our ruling, it will not be necessary to address any of the errors alleged to have occurred in the second trial.

REVERSED.