A 14-year-old boy who was squirrel hunting with his father's 30-year-old shotgun stopped to rest, placing the gun against a bench. The gun fell and discharged despite the fact that the hammer was not cocked. He appeals an adverse verdict, contending that the district court erred by granting a partial directed verdict, in instructing the jury on Louisiana law, in denying his motions for a directed verdict or judgment notwithstanding the verdict, and in summarily terminating his cross-examination of defendant's expert witness. Finding no merit in these contentions, we affirm.

* Jeffery Miles, a fourteen year-old boy, went in January, 1987, to hunt squirrels in the woods near his home in Pineville, Louisiana. As he had many times before, he was using a Winchester Model 37 single-shot shotgun which had been given by his grandmother to his father as a Christmas present in 1957. The gun's only safety mechanism was that it would not fire unless the hammer was cocked before the trigger was pulled.

According to Jeffery's testimony, he loaded the gun when he reached the woods. Shortly thereafter he saw a squirrel, fired, and missed. He reloaded, but the squirrel was gone. He left the hammer uncocked, in the safe position. After looking for the squirrel for approximately a minute, Jeffery came upon an old, handmade bench. He laid his gun on the bench, unzipped his coveralls, and took off his flannel shirt and a T-shirt. He picked the gun up, resting its butt on the bench. It slipped. The hammer spur hit the edge of the bench. The gun discharged into Jeffery's left shoulder, inflicting a serious wound.

Robert Miles, Jeffery's father, sued Olin Corporation, a successor corporation to the gun's manufacturer, both individually and as Jeffery's guardian ad litem. Miles alleged that the shotgun was unreasonably dangerous for its intended use and that Olin had failed to adequately warn consumers that a blow to the Model 37's hammer spur could discharge the weapon.

Olin's expert, Dr. Robert Jay Block, after examining the physical evidence and conducting numerous tests, testified that he believed that the weapon was cocked and ready to fire when the hammer spur was struck, and theorized that Jeffery might have fallen forward, striking the hammer spur against the bench with his weight behind it, causing the gun to discharge. He admitted that the gun might have discharged had the accident occurred as Jeffery described it, but maintained that Jeffery's description was inconsistent with the damage to the shotgun's firing mechanism. Miles's experts testified that the gun was defectively designed, that it had discharged despite the fact that the hammer was not cocked, and that it would have been practical to design the gun to prevent such an accidental discharge.

After a five-day trial, the jury concluded, in response to special interrogatories, that the Model 37 was not unreasonably dangerous due to a design defect, and that Olin had failed to provide adequate warnings, but that its failure to warn was not a cause of Jeffery's injury.

Miles challenges the district court's judgment on several grounds, claiming that the district court erred by (1) granting a directed verdict on Miles's unreasonably-dangerous-per-se theory of liability; (2) applying the substantive provisions of the Louisiana Products Liability Act retroactively; (3) denying Miles's motions for a directed verdict and for judgment notwithstanding the verdict; and (4) summarily terminating Miles's cross-examination of one of Olin's expert witnesses.

Louisiana's law of products liability in force when Jeffery was injured was prescribed in the Louisiana Supreme Court's landmark decision in Halphen v. Johns-Manville Sales Corporation.1 Under Halphen, a plaintiff could demonstrate that a product was "unreasonably dangerous to normal use"2 by establishing that the product either (1) was "unreasonably dangerous per se" (hereinafter "unreasonable-per-se"); (2) suffered from some flaw in its construction or composition; (3) lacked adequate warnings; or (4) exhibited a "defective" design.3 A design was "defective" if (a) "the danger-in-fact, whether foreseeable or not, outweigh[ed] the utility of the product"; (b) "alternative products were available to serve the same needs or desires with less risk of harm"; or (c) "there was a feasible way to design the product with less harmful consequences."4 In his complaint, Miles asserted all of these theories of liability save the second, and asserted all three design-defect theories.

After Jeffery suffered his injury but before his case was tried, the Louisiana Legislature enacted the Louisiana Products Liability Act (LPLA).5 The LPLA was specifically intended to overrule Halphen in several respects, and to clarify many of the questions that Halphen had left unanswered.6 As the Louisiana courts have held repeatedly that the substantive provisions of the LPLA may be applied only to causes of action arising after its September 1, 1988 effective date,7 those provisions are not applicable to this case.

At the close of Miles's case, the district court granted Olin's motion for a partial directed verdict on Miles's unreasonable-per-se theory of liability. The grounds advanced by the district court to justify its ruling are ambiguous:

I'm not giving instructions on unreasonably dangerous per se. This has to do with design defects and failure to warn, and that is the method that I'm going to employ in the instructions and that has basically been the way that this game has been played since we have had the pre-trial [sic] conference.

Olin construes these remarks to be a holding by the district court that Miles waived the unreasonable-per-se theory of recovery by not presenting it as an issue in the pretrial order. We cannot agree. The district court made no reference in its remarks to the terms of the pretrial order and Olin did not advance a waiver argument in support of its motion for a directed verdict. Instead, we analyze the record to determine whether the ruling can be sustained on the ground stated in Olin's motion: that Miles failed to present enough evidence that the Model 37 shotgun was unreasonably dangerous per se to withstand the motion for directed verdict.8

A directed verdict is appropriate only if, considering all of the evidence and drawing all inferences therefrom in favor of the nonmoving party, the court is convinced that no reasonable jury could find in favor of the nonmovant.9 Halphen's dangerous-per-se theory requires the plaintiff to prove that "a reasonable person would conclude that the danger-in-fact of the product, whether foreseeable or not, outweighs the utility of the product."10

Olin presented evidence that the Model 37 and its particular firing mechanism served a number of utilities, ranging from the general utility of shotguns as hunting weapons to the specific consumer demand for a simple, inexpensive, durable shotgun. Perhaps most importantly, Olin presented evidence that the exposed safety mechanism on the Model 37 was designed to make it readily apparent to the novice hunter whether his weapon was in the safe or fire position, a feature that may very well have prevented numerous accidental injuries.

On the other hand, Miles adduced no evidence whatever that the danger-in-fact posed by the Model 37's exposed-hammer firing mechanism is substantial. Experts for both sides testified that the chances of accidentally striking the hammer spur directly and with enough force to cause the weapon to discharge are remote. Olin's expert demonstrated that this was due to the low profile of the hammer spur and the configuration of the butt stock. Indeed, Miles's expert had some difficulty in achieving a direct hit on the hammer spur even in a controlled drop test, and Olin's expert chose not to drop the gun as Jeffery had done, but clamped it in a machine and struck the hammer spur with a pendulum: that was the only way "to be sure that [he] would get a solid hit."

Furthermore, Miles introduced no evidence of any other inadvertent discharge of the type alleged in this case, whether resulting in injury or not, in spite of the fact that more than a million Model 37s--and more than six million exposed-hammer single-shot shotguns of all brands--had been produced and sold. Miles blames what his attorney terms Olin's "aggressive document destruction policy" for the fact that no records of customer complaints regarding the Model 37 were available. It is true that Olin had a records-retention policy mandating the destruction of nonessential paperwork after four years; since production of the Model 37 was discontinued 1962, next to nothing relating to it remained in Olin's files. Nevertheless, the burden of adducing some evidence that the shotgun posed a real danger-in-fact remained Miles's. He could not carry it simply by bemoaning the lack of evidence.

The only positive evidence that the Model 37 was dangerous-in-fact, other than Jeffery's accident itself, came from his expert mechanical and design engineer, John T. Butters. Even that had little probative value. Butters reiterated that the gun would indeed discharge if struck directly on the hammer spur even in the safe position, thus rendering it, in his opinion, "unreasonably hazardous ... and unsuitable for the purpose for which it was designed." He went on to explain why he thought so:

Any firearm which will discharge without the purposeful pulling of the trigger, no matter what the reason, is categorically dangerous and it is categorically defective. Any inadvertent firing of a firearm makes it categorically defective for whatever reason. It may be for design, it may be from manufacture, it may be from wear, but for whatever reason it has got to be corrected before it may be safe[l]y used.

As Olin's expert, Dr. Block, explained, however, any gun will fire accidentally under some set of circumstances. To accept Mr. Butters's opinion as sufficient evidence to submit to the jury the issue whether the Model 37 was unreasonably dangerous per se would, in effect, be to ask the jury to determine whether all firearms are unreasonably dangerous per se. That determination is properly for the Louisiana legislature, not a petit jury in a federal trial. The directed verdict was therefore warranted.

C. Jury Instructions on Miles's First and Second Design-Defect Claims

Miles next contends that the district court erred in instructing the jury on his first two design-defect claims, predicated on Halphen's risk versus utility and alternative-product theories of liability. The district court instructed the jury as follows:

You may find that the defendant's product was unreasonably dangerous in design if there was a product that was capable of preventing the damage which plaintiff sustained. If the evidence does not satisfy you that, more probably than not, such an alternative design existed, then you may not find that the defendant's product was unreasonably dangerous in design. If you conclude that the alternative design was in existence, then you must weigh the likelihood that the design utilized by the defendant would cause plaintiff's damage, and the gravity of that damage against two things: the burden on the defendant of adopting the alternative design, and the adverse effect, if any, that such an alternative design would have on the utility of the product.

Miles argues that this instruction is erroneous because it states the law as revised by the LPLA, rather than the law under Halphen. Specifically, Miles claims that the instruction erroneously (1) required proof of an alternative design to prevail under Halphen's first design-defect theory; and (2) eliminated entirely Halphen's second design-defect theory. At oral argument, Miles conceded that the instruction with respect to the alternative-design theory was essentially correct, so we do not address its propriety in that regard.

a. The First Design-Defect Theory: Risk v. Utility. The first design-defect theory articulated in Halphen is essentially identical to the unreasonable-per-se theory of liability: the plaintiff must prove that the danger-in-fact of the design outweighs the utility of the product.11 This "overlap ... makes it unnecessary to decide whether a product's defect is one of design or of another kind."12

Under Halphen, no proof that an alternative design was possible is necessary to recover under the first design-defect theory, for "the theory's purpose is to evaluate the product itself, not the manufacturer's conduct."13 The LPLA overruled Halphen in this respect by adopting a unitary design-defect theory that looks solely to whether there was a feasible alternative design available that would have prevented the plaintiff's damage.14 The district court's instruction was, in all material respects, a verbatim reading of the LPLA's change in the law on design defects. It was therefore in error.

The error, however, was harmless. Errors in instructing the jury are harmless when the evidence was so lacking that a directed verdict against the complaining party was warranted.15 That was the case here, for the same reason noted with respect to Miles's unreasonable-per-se claim: Miles failed to introduce any evidence that the danger-in-fact of the Model 37's firing mechanism outweighed the gun's utility. The district court's error thus did not "affect the substantial rights of the parties."16

b. The Second Design-Defect Theory: Alternative Products. We agree with Miles that the district court's instruction erroneously eliminated from the jury's consideration Halphen's second design-defect theory. While the instruction does mention "a product that was capable of preventing the damage which plaintiff sustained," the instruction read as a whole indicates that the district court was, in fact, speaking of alternative designs. The district court never defined for the jury the concept of an "alternative product", nor did it instruct the jury that it could find for the plaintiff if it found that such a product was or is available.

Again, however, the error was harmless. Miles adduced no evidence that alternative, safer products were or are available. He did not attempt to define, through expert testimony, proposed jury instructions, or otherwise, a universe of products that may be considered alternatives to the Model 37, nor did he attempt to prove that the exposed-hammer single-shot shotguns marketed by Olin's competitors were safer. The only evidence submitted by Miles was that other shotguns, namely Winchester's Model 21 and Model 24 double barrel shotguns, which include an automatic safety, can also be used to hunt birds and small game. But he offered no evidence that those weapons served the other utilities the Model 37 was designed to provide: the Model 37 filled a niche in the market for durable, inexpensive shotguns; it was simple to assemble and to use; and, most importantly, its exposed safety mechanism made it readily apparent to the inexperienced hunter whether his weapon was in the fire or safe position. In short, a directed verdict against Miles on the second design-defect theory would have been warranted. The district court's error does not require reversal.

D. Miles's Alternative Design Design-Defect Claim

The jury found that the Model 37 was not unreasonably dangerous because of a design defect. Miles argues that the jury's verdict was against the great weight of the evidence, and that the district court therefore erred in failing to grant his motion for a directed verdict or for judgment notwithstanding the verdict on his claim under Halphen's third design-defect theory. To prevail, Miles must demonstrate that, even considering the evidence in a light most favorable to Olin, the facts and inferences point so strongly and overwhelmingly in his favor that reasonable jurors could not have found against him.17 Miles did not meet that burden.

Miles's expert, John T. Butters, testified that there were four alternative designs available when the Model 37 was manufactured: a two-step safety, a hammer guard, a design placing the cocking lever behind the hammer, and a hammer block. On cross-examination, however, Butters admitted that the first two designs would meet with stiff resistance from hunters; that the third would not eliminate the danger of accidental discharge from a direct blow to the hammer spur; and that the fourth design was for muzzle-loaded weapons, not shotguns. Olin's expert, David Findlay, elaborated on each of these shortcomings. Butters also suggested that a transfer bar could be used to eliminate the danger, but had to admit that he had not worked out precisely how to do it, and that the company holding the patent for the transfer bar had never incorporated it into its single-shot shotguns. In short, we cannot say that the jury's decision that these designs were not feasible alternatives was unreasonable. The district court did not err in denying Miles's motion for directed verdict or JNOV.

The jury found that Olin failed to provide adequate warnings that the Model 37 would discharge if struck on the hammer spur, but that Olin's failure to warn was not a proximate cause of Jeffery's injuries. Miles makes two distinct arguments for reversal: (1) the jury's conclusion that the product was not unreasonably dangerous because of a design defect and its conclusion that Olin failed to warn are fatally inconsistent; and (2) the jury's conclusion that Olin's failure to warn was not a proximate cause of Jeffery's accident is against the great weight of the evidence. Both of Miles's arguments are based on fundamental misunderstandings of the law, and are without merit.

Miles asserts that the jury's conclusion that Olin failed to give adequate warnings necessarily includes a determination that the product was defective. That is nonsense. Under Halphen, the manufacturer has a duty to warn of "any danger "--not defect--"inherent in the normal use of its product."18 The product may be rendered unreasonably dangerous by a failure to warn, but the manufacturer's breach of its duty to warn does not mean that the product is otherwise defective. The jury concluded that the product was not rendered unreasonably dangerous by a design defect, but that it was rendered unreasonably dangerous by Olin's failure to warn. The two findings are not inconsistent.

Nor was the jury's finding that Olin's failure to warn was not a proximate cause of Jeffery's injury against the great weight of the evidence. Miles contends that, because Louisiana law presumes that a consumer will read and heed warnings if they are given,19 and because Olin offered no evidence to the contrary,20 the jury could not but conclude that the failure to warn was a proximate cause of the accident.

That argument assumes one critical fact: that the warnings, even if read and heeded, could have prevented the accident. The Louisiana Supreme Court has stated that "[a]n essential element of the plaintiff's cause of action for failure to adequately warn of a product's danger is that there be some reasonable connection between the omission of the manufacturer and the damage which the plaintiff suffered."21 Miles presented no evidence--and we are dubitante that he could--that a warning by Olin, even if Jeffery had read it and scrupulously abided by it, could have prevented Jeffery from dropping his shotgun vertically onto a bench. The jury's verdict may very well have been different had Jeffery deliberately driven the hammer spur into the bench while the shotgun was pointed at his shoulder. Halphen's failure-to-warn theory, after all, is by its very nature aimed at preventing reasonable, deliberate uses of products by consumers that unwittingly and needlessly put them in harm's way. But that is not this case. A warning has little power against purely accidental occurrences. We see no reasonable connection between the failure to warn and the accident as it occurred; therefore, we do not believe that the jury's verdict was against the great weight of the evidence.

Miles's final arguments concern the district court's termination of his cross-examination of defendant's expert, Dr. Robert Jay Block. Miles contends that the district court erred in making prejudicial comments when terminating the examination, and in refusing to admit evidence to impugn Dr. Block's credibility.

Miles complains that the following remarks by the district court, which he describes as "impatient and even condescending", prejudiced the jury against him:

I think I have endured all I can do on this cross-examination. It is over now. He has said the same thing over and over and over.... The cross-examination is over, it is redundant, repetitive and it is argumentative and it is over, so I declare the cross-examination over.

Federal Rule of Evidence 611(a) gives the district court "reasonable control over the mode and order of interrogating witnesses and presenting evidence ... to avoid the needless consumption of time."22 In exercising that control, however, the district court must maintain both objectivity and the appearance of neutrality when it intervenes to terminate the questioning of a witness.23 We review the record as a whole, not isolated comments by the trial judge, to determine whether the trial was fair and impartial.24 Because Miles did not object to the district court's comments in this case, we review only for plain error.25

We do not find the district court's comments so prejudicial or biased as to constitute error, plain or otherwise. This case is in no respects similar to either Dartez v. Fibreboard Corporation26 or Newman v. A.E. Staley Manufacturing Company,27 both of which involved extensive, highly prejudicial remarks by the trial judge favoring one party over the other. The district court's remarks in this case, however, were made only after counsel for Miles was permitted an entire afternoon--over fifty pages in the record--of cross-examination of Dr. Block, the great majority of which was indeed redundant, argumentative, and unproductive. The district court's remarks were well within the bounds of acceptable conduct by a trial judge.

Miles's final argument is that by abruptly terminating his cross-examination the district court in effect excluded testimony intended to impeach Dr. Block. During the voir dire, Miles asked Dr. Block if any court had ever refused to accept him as an expert; Dr. Block answered, "No." He had been refused, however, in a summary judgment case by the United States District Court for the Western District of Oklahoma.28 Miles claims that admission of this evidence would have undermined both Dr. Block's qualifications and his credibility.

The exclusion of the evidence was within the sound discretion granted the trial court by Evidence Rule 611(a), and was not erroneous. Counsel for Miles made a calculated strategic decision to ask the question last, apparently to maximize its dramatic effect. Unfortunately for his client, he spent a number of hours prior to asking the question belaboring the court, the jury, and the witness with redundant and argumentative questioning. The trial court was justified in concluding that enough had been said and in terminating the cross examination, whatever surprises Miles still had to spring on the witness.

Even if it were error, however, we would conclude that it was harmless. Dr. Block's credentials to testify as an expert mechanical design engineer and metallurgist were well established; his findings were specific, detailed, and well documented, and Miles was unable to cast any doubt during his extended cross-examination on the reliability of Dr. Block's methodology. The impeachment evidence would have had, at most, minimal effect on the jury's assessment of Dr. Block's qualifications or credibility. We cannot conclude that Miles's substantial rights were prejudiced.

For the foregoing reasons, the judgment of the district court is AFFIRMED.