Kevin Bradley was a passenger in a 1986 Chevrolet C/K pickup owned by the Highway Department of the State of Wyoming when the truck collided with a bridge and spun out on a slippery road, projectiling him onto the roadway. A jury found General Motors Corporation negligent and awarded Kevin and Cheryle Bradley $4.8 million in damages for the injuries Mr. Bradley sustained. On appeal, GM contends the judgment is based upon an inconsistent verdict because the jury, having found against plaintiffs on their strict liability claim, could not conclude GM was nevertheless negligent. We agree that the verdict is irreconcilably inconsistent and reverse for a new trial.

On November 7, 1991, Mr. Bradley and Mark Wallace, Wyoming Highway Department employees, were en route to an assignment with Wallace driving. It had recently snowed, and the highway was wet and slippery. Driving at about 55 mph as he crested a hill passing another vehicle, Wallace was unable to slow the pickup which began skidding downhill onto a bridge, hitting a guardrail. The truck then traveled perpendicularly to the guardrail, the front end sliding along the right guardrail until its right fender snagged a post, spinning the truck counterclockwise. Sometime during this slide and the impact of the right rear corner against the guardrail, Mr. Bradley was ejected from the passenger seat. He suffered multiple injuries as a consequence.

The Bradleys brought this action against GM on theories of strict liability and negligence, claiming the GM manufactured seat belt buckle and passenger side door latch were defectively designed and that GM failed to properly warn of the "defective and hazardous condition." At trial, plaintiffs attempted to establish that despite the mild, "glancing blow" when the pickup hit the guardrail, traveling at not more than 11 mph, Mr. Bradley's door latch failed; the door unlocked and flew open; and he was thrown from the vehicle. Plaintiffs' experts theorized the seat belt failed because of "inertial unlatch"; that is, the sudden deceleration of the collision caused the "pawl," the part of the buckle that holds it closed, to spring open. They also maintained the detent lever which holds the door latch closed became unbalanced and inertially unlatched.

GM attempted to reconstruct the accident relying on statements of the driver, Mark Wallace, and Wyoming State Trooper, Joseph Arzy. Following the collision, both had initially stated Mr. Bradley was not wearing his seat belt. Officer Arzy ultimately vacillated on the point, however, testifying he could not be sure whether Mr. Bradley was wearing his seat belt. He also stated he saw a locking bar mark and an imprint on the belt from the patch near the end of the webbing, which he said was consistent with belt use. Although not certain, Mr. Wallace stated he "might" have said in an earlier statement that upon impact, Mr. Bradley's hand was on the door handle unlatching the door, causing him to fall out.

GM offered Highway Department testimony about the inadequacy of the tires on the pickup; Officer Arzy's opinion of the road conditions and the State's failure to send out plows; and Mr. Wallace's speed from the observations of the driver of the car he had passed. All agreed the Highway Department vehicle was equipped with hand-me-down tires from State Highway Patrol cars. The tires, representing the State's effort to save money, were too small for the truck and "played a significant role" in the accident.

GM's experts generally testified to soundness of the seat belt and latch designs, explaining that while inertial unlatch is possible, it only occurs when the absolute right forces coincide. Essentially, they argued plaintiffs' effort to demonstrate the phenomenon was a "parlor trick," which involved swinging the buckled belt around and hitting its back hard against the hip to unlatch it. To accomplish that result in this collision, GM argued the seat belt could not have been worn and the buckle could not have been fastened. Expert testimony was offered to establish -given the unlikelihood of an inertial unlatch--with the slow speed of the impact and Mr. Bradley's weight against a fastened belt, the belt was least likely to unlatch. Instead, GM contended Mr. Bradley did not wear his seat belt, as witnesses initially stated, and his hand, resting on the door handle, accidentally opened the door.

Similarly, GM sought to disprove plaintiffs' evidence the door latch was defective. Its experts testified there was no way--absent a direct blow which the parties agreed was not present here--that this impact could have caused the door to unlatch. GM's experts opined only "actuation" of the inside handle would, under these circumstances, open the door.

GM sought to introduce the settlement the Bradleys reached with the State of Wyoming and Wallace, an agreement providing, GM argued, a powerful incentive for Wallace and Arzy to tailor their testimony, permitting the State to recoup its payment of Mr. Bradley's medical expenses. GM also sought to have the jury apportion the liability of the State and Wallace, but the court refused on the ground that while these actors may have been responsible for part of the initial accident, this was a second impact case.

In response to interrogatories propounded by the court, on plaintiffs' strict liability claims the jury found neither the seat belt nor the door latch was defectively manufactured. Despite that conclusion, however, the jury also found GM was negligent "in a manner which caused Plaintiff Kevin Bradley's injuries."

As this appeal unfolds, the first question before us is not whether the evidence is sufficient to sustain the verdicts, but whether these verdicts are irreconcilably inconsistent as a matter of law. Although the parties devote considerable effort to presenting their views of the evidence, we are only concerned whether the verdict represents a proper and consistent resolution of the factual questions or a confusion of the issues in the minds of the jury.1

"A verdict that resolves separate and distinct causes of action in favor of both parties is not inconsistent on its face." Harris Mkt. Research v. Marshall Mktg. & Communications, Inc., 948 F.2d 1518, 1522 (10th Cir.1991). However, when "several causes of action are identical and defended on the same ground, a verdict for the plaintiff on one cause of action and for the defendant on another is inconsistent." Diamond Shamrock Corp. v. Zinke & Trumbo Ltd., 791 F.2d 1416, 1425 (10th Cir.1986).

In this case, the jury found GM's product--the pickup truck including its seat belt and door latch--was not defective. That finding notwithstanding, it also found GM was negligent. GM maintains these are irreconcilably inconsistent findings because if a product is found to be non-defective, building it consistent with its design cannot constitute negligence.

Plaintiffs counter by reminding if there is a way to render a jury's answers in a special verdict internally consistent, it is the duty of the court to do so. See Palmer v. City of Monticello, 31 F.3d 1499, 1505 (10th Cir.1994). They then postulate GM's position is the product of a mischaracterization of the applicable law, contending GM misinterprets McLaughlin v. Michelin, 788 P.2d 59 (Wyo.1989), the key Wyoming case. Plaintiffs take the position Michelin really involves two different kinds of product "defects." They argue Wyoming recognizes a distinction between a "defect in fact" and a "defective condition unreasonably dangerous." The distinction they draw, however, seems tenuous.

The Wyoming Supreme Court's own distinction between strict liability and negligence in the field of products liability is that under a claim of strict liability "the focus is on the product itself," whereas in a claim of negligent manufacture of design, "the conduct of the maker ... is in question." Michelin, 778 P.2d at 64 (emphasis added). The court then explained, despite the distinctions between the various theories of recovery,

"[t]he requirement of showing a defect is one element common to every products liability case, whether it is brought on a theory of negligence, breach of an express or implied warranty, strict tort liability, or a combination of theories."

Id. (quoting W. Kimble & R. Lesher, Products Liability § 53, at 69 (1979)). The court added, "[t]he burden under this rule is upon the plaintiff to demonstrate a defect in fact, i.e., that the product failed." Id. (emphasis added). Plaintiffs seize upon this last phrase to create a distinction between the defect in a product which will support a claim for strict liability and a defect that will support a claim for negligence. We can see no such distinction.

We read Michelin to say if a plaintiff cannot prove a product is defective, he cannot prove a claim of negligence. Notwithstanding the focus of a negligence claim on the conduct of the manufacturer, if a defective product does not result from that conduct, the manufacturer has no liability.

Also instructive is Garrett v. Hamilton Standard Controls, Inc., 850 F.2d 253 (5th Cir.1988), where the court held the failure to give a negligence instruction was not reversible error, the jury having found the product was not defective. The court stated,

[A] manufacturer logically cannot be held liable for failing to exercise ordinary care when producing a product that is not defective because: (1) if a product is not unreasonably dangerous because of the way it was manufactured, it was not negligent to manufacture it that way and (2) even if the manufacturer was somehow negligent in the design or production of the product, that negligence cannot have caused the plaintiff's injury because the negligence did not render the product unreasonably dangerous.

Id. at 257. In that case, the jury found the electric blankets that caught fire were not defective.

We have reached a similar disposition in Oja v. Howmedica, Inc., 111 F.3d 782 (10th Cir.1997). Construing Colorado products liability law, we stated:

[A]s with all tort claims, the plaintiff must prove the elements of causation and damages. More importantly, "[o]ne critical area of overlap is that, '[r]egardless of whether a product liability action is grounded in negligence or strict liability, a plaintiff must prove that the product was defective.' "

Id. at 791 (quoting Perlmutter v. United States Gypsum Co., 54 F.3d 659, 663 (10th Cir.1995) (quoting Mile Hi Concrete, Inc. v. Matz, 842 P.2d 198, 205 (Colo.1992)).

Consequently, because the jury here found the defendant's product without defect, it could not find it liable for negligence of any form with respect to that product. The verdict which reached that conclusion is, thus, irreconcilably inconsistent by definition. In fairness to the district court, it anticipated the possibility of just this result and suggested the plaintiffs choose which theory they wanted to pursue. They should have done so. Now, the only avenue through which this inconsistency can be resolved is by retrial.

Having reached this conclusion, we believe it is necessary to address only those remaining issues that are likely to arise on retrial. The first of these is the issue of Mr. Bradley's comparative fault. We would simply observe because comparative fault arises only in consideration of the defendant's negligence, and there is no negligence component of strict liability, a jury cannot make findings of comparative fault on such a claim. Here, if the plaintiffs had elected to proceed only on their theory of negligence, no inconsistency would exist with the findings of comparative fault. Thus, on retrial, if plaintiffs insist upon proceeding upon both theories, not only will the district court have to carefully outline the distinctions between the two claims, but also, it will have to demark the claims against which comparative negligence will be considered.

Finally, the district court found the evidence did not justify giving the jury the question of the comparative negligence of the non-parties. Given the outcome of this appeal, we will not directly respond to the issue at this time. We do, however, call the district court's attention to Cleveland v. Piper Aircraft Corp., 890 F.2d 1540, 1546-51 (10th Cir.1989).

The judgment is VACATED, and the cause is REMANDED for a new trial.