Bowman v. General Motors Corp., 427 F. Supp. 234 (E.D. Pa. 1977)
February 15, 1977
GENERAL MOTORS CORPORATION
Clyde E. RHODES, Third-Party Defendant,
Robert W. Bowman, Fourth-Party Defendant.
United States District Court, E. D. Pennsylvania.
*235 Avram G. Adler, Stanley P. Kops, Philadelphia, Pa., for plaintiff.
George J. Lavin, Jr., Philadelphia, Pa., for defendant.
OPINION AND ORDER
EDWARD R. BECKER, District Judge.
I. Preliminary Statement
This is a products liability case in which plaintiff alleges that the gas tank and surrounding structures of his 1966 Oldsmobile Toronado were defectively designed. Specifically, plaintiff contends that the Toronado was uncrashworthy because, when his car was struck in the rear by another vehicle, it burst into flames, causing serious injury to his wife. After the liability phase of a bifurcated trial lasting almost five weeks, the jury found for the defendant General Motors Corporation (GM). This opinion addresses the plaintiff's motion for a new trial and/or for a judgment n. o. v. We shall not review the facts in detail, but shall summarize only so much of the evidence and of the parties' theories as is necessary to an understanding of the issues raised by plaintiff's motion.
On May 22, 1970, at about 11:30 p.m., plaintiff Robert Bowman was driving his Toronado east on U.S. Route 30 in Downingtown, Pennsylvania, with his wife as a passenger, when his vehicle was struck in the rear by a vehicle driven by third-party defendant Clyde Rhodes. The front of the Rhodes vehicle underrode the Toronado's rear bumper and pierced its fuel tank. An explosion and fire immediately ensued and the flames invaded the passenger cabin. Although Mr. Bowman escaped without injury, Mrs. Bowman sustained extremely serious and disfiguring burns. Mrs. Bowman died prior to trial from a disease unrelated to the accident. Mr. Bowman sues as her administrator, as well as in his own right for property damage.
Plaintiff's crashworthiness allegation is that the Toronado was not designed to minimize the risk of injury to the vehicle's occupants in the event of accident. Factually, plaintiff's case rests on the following claims of defective design: (1) the fuel tank was positioned too close to the rear bumper; (2) the angle of the rear bumper invited a striking vehicle to underride it, thus exposing the fuel tank to trauma; and (3) openings in the Toronado's structure (especially the plenum drain) unnecessarily allowed flames to invade the passenger cabin.
Legally, plaintiff relied on theories both of strict products liability and of ordinary negligence liability. Plaintiff's design defect claim is founded upon Restatement of Torts 2nd, § 402(A), which has been adopted in Pennsylvania, Webb v. Zern, 422 Pa. 424, 220 A.2d 853 (1966), whose law applies to the case. Plaintiff also contended that GM was negligent in failing adequately to test the 1966 Toronado for fuel tank integrity in the event of rear end impact.
Plaintiff's theories were advanced principally through the testimony of Derwin Severy, a pioneer in the automobile crash testing field and a highly-qualified automotive design engineer. Mr. Severy testified for three days in three major segments. First, he performed a detailed accident reconstruction during which he testified that the striking (Rhodes) vehicle was not traveling *236 at an excessive rate of speed. Next, he offered testimony on defective design. Finally, he testified as to what a reasonable test program for fuel tank integrity would entail. Mr. Severy opined that GM's test program was seriously deficient, that the Toronado was defectively designed in the respects noted above, and that these factors were significant causes of the fire.
GM's trial methodology mirrored the plaintiff's. First, GM offered its own expert's accident reconstruction, which sought to establish that the Rhodes vehicle struck the Bowman vehicle at such a high rate of speed that it would have penetrated any vehicle short, perhaps, of a Sherman tank. This was meant to prove that the cause of the fire and of the resultant injuries was the negligence of the third-party defendant and not the design of the Toronado, and to support GM's position that there are extreme hazards of the road against which it is unreasonable to expect it to design. Next, through the presentation of films of its testing program and the testimony of its able chief crash test engineer, William Cichowski, GM developed evidence that its program of crash testing, which was calculated to study, inter alia, fuel tank integrity, was extensive and adequate. Finally, GM offered the testimony of its own design engineers to the effect that the Toronado's design was safe. The main thrust of the opinion of these experts was that the rupture of the Toronado's fuel tank could not have been avoided by a repositioning of the fuel tank or by a redesign of the rear bumper, and that the underride by the Ford of the Toronado bumper, and the piercing of the fuel tank followed ineluctably from design of the striking vehicle and the nature and severity of the impact. GM expert Ron Elwell also countered Mr. Severy's testimony about the plenum drain and other openings into the passenger cabin.
In rebuttal, the plaintiff offered the testimony of Alfred Baccini to refute Mr. Elwell. We found Mr. Baccini to be qualified in the field of fire safety and we permitted his testimony relative to fire safety matters including the mechanism of fire's entry into the passenger cabin, and design matters relating thereto. Although Mr. Baccini is a mechanical engineer, he lacked experience in the field of automotive design and had never inspected the Toronado; therefore, we did not find him qualified to testify about matters of automotive design in general. Alternatively, we ruled that because he was a rebuttal witness, he could not go over again the areas covered by Mr. Severy in plaintiff's case in chief. One of the two points briefed by the plaintiff in support of his motion for a new trial is our limitation on the scope of Mr. Baccini's testimony.
Because that matter can be quickly disposed of, we will address it first. We will then turn to plaintiff's other major contention, founded upon the views of Chief Justice Jones and Justice Nix in Berkebile v. Brantly Helicopter Corp., 462 Pa. 83, 337 A.2d 893 (1975), that we erred in charging the jury that, in order to find that the Toronado was defective by reason of its design when it was sold to Mr. Bowman, they had to find not only that it was defective, but also that it was unreasonably dangerous. We will devote the major portion *237 of the opinion to this point, but must preface that discussion with the following perspective.
In Bair v. American Motors Corporation, 535 F.2d 249, 250 (3d Cir. 1976) the Court of Appeals, following the incisive reasoning of our colleague Judge Daniel H. Huyett, 3rd, in Beron v. Kramer-Trenton Co., 402 F. Supp. 1268, 1277 (E.D.Pa.1975), aff'd. mem. 538 F.2d 318 (3d Cir. 1976), stated that:
the views expressed in Chief Justice Jones' opinion in Berkebile are not the law of Pennsylvania, and that it is proper to instruct a jury that it must find that a defective condition be unreasonably dangerous to the user or consumer.
This conclusion followed from: (1) the Pennsylvania rule that an opinion of the Pennsylvania Supreme Court representing the views of only two justices has no binding precedential value; and (2) the previous (binding) Pennsylvania precedent which required the unreasonably dangerous charge.
The Bair precedent is sufficient to dispose of plaintiff's objections to our charge. However, Bair was not decided until after this trial, and we could not be certain at the time of trial whether Judge Huyett's view in Beron would prevail. Plaintiff's counsel argued, not unpersuasively, that the view of two justices of the Pennsylvania Supreme Court was entitled to considerable weight on the issue of what was Pennsylvania law. At the very least it appeared to us, as we were formulating our jury charge, that Pennsylvania law was in flux, and that it was necessary to make a judgment as to what the Supreme Court of Pennsylvania would do when it definitively addressed the "unreasonably dangerous" question. The problem was exacerbated by our realization that this case might be appealed, and that the Pennsylvania Supreme Court might write on the question in the interim. Accordingly, while agreeing with Judge Huyett's Beron opinion, we extensively researched and contemplated the question whether the unreasonably dangerous concept is a necessary part of the law of strict liability in a design defect case, and formulated certain views which, inter alia, supported our jury charge.
Notwithstanding the interposition of Bair, we believe it appropriate to set forth the conclusions of our research and reflections for two reasons. First, it is appropriate that we memorialize the basis on which the charge, as it was conceived at the time, was the product of reasoned analysis. Second, we believe that the relationship between federal and state courts in diversity cases established by Erie Railroad Co. v. Tompkins, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188 (1938), which requires that a federal district judge act essentially as a state nisi prius judge, implies that where a particular trial has resulted in a survey of the law and an extensive inquiry into the factors that need to be taken into account in the continuing development of an important rule on the "cutting edge" of the law, the judge should elucidate the insight thus produced by trial so as to add to the ongoing jurisprudential dialogue, and perhaps aid the reflective processes of the state appellate courts.
*238 Our analysis of the problem has led us to this conclusion: in the context of alleged defects resulting from conscious design choices, as opposed to defects resulting from manufacturing flaws or inadvertent design errors, the § 402A notion of "defect" cannot be meaningfully discussed without the aid of the "unreasonably dangerous" concept, hence that concept is an important part of strict liability law. We shall explain this conclusion at some length below. At this point we note, however, that plaintiff's post trial motion will be denied.
II. The Limitation on Rebuttal Testimony
We consider next plaintiff's contention that we improperly limited his proffered *239 rebuttal testimony. As has been noted, plaintiff called Alfred Baccini on rebuttal to testify as an expert about his judgment as to, inter alia, the adequacy of the Toronado's design. After hearing plaintiff's proffer, we ruled that Mr. Baccini's testimony would be limited to rebutting specific points raised by GM's experts. In so ruling, we observed that to the extent that the evidence proffered was not specifically responsive to GM's case-in-chief, it was not properly offered as rebuttal, but should have been presented in plaintiff's case-in-chief. Moreover, to the extent that the evidence proffered would simply rehash plaintiff's basic theory about the defectiveness of the design, it was excludable an unnecessary cumulation.
After the voir dire on Mr. Baccini's qualifications, we further limited his testimony to crash performance and to the characteristics of fire in crash situations on the basis that he was not qualified to testify as an expert on the design of automobiles and the positioning of the bumper and gas tank. Plaintiff now asks that we reconsider those limiting rulings on the ground that they were prejudicial error. We find that the rulings were neither prejudicial nor erroneous.
Rule 403, Fed.R.Ev., authorizes a district judge to exclude relevant evidence which he finds, in his discretion, to be a "needless presentation of cumulative evidence." Rule 403 would, in itself, be a legally-sufficient ground to support the exclusion of needlessly cumulative evidence from either the case in chief or the rebuttal. The trial reached the rebuttal stage only after fourteen days of testimony. Both parties had adduced extensive expert testimony. To have permitted Mr. Baccini to present again plaintiff's general theory of the accident would have added nothing substantive. Although it might have been to plaintiff's tactical advantage to get a re-run of Mr. Severy's testimony before the jury in the wake of the lapse of time since he had testified, the trial had been reported with daily transcript and we informed plaintiff's counsel that he could take as much time as he needed in summation to review Mr. Severy's testimony which he now had verbatim. *240 Moreover, had we permitted cumulative rebuttal, we would have had to allow GM "equal time" in rejoinder, thus needlessly prolonging an already long trial.
We do not, however, rest our ruling solely on Rule 403; for there is no doubt about the power of a trial judge to regulate, in his sound discretion, the scope of rebuttal testimony. While a district judge's discretionary power to limit rebuttal evidence or, more generally, to require an orderly presentation of a case has not been questioned since the passage of the Federal Rules of Evidence and the accompanying revision of Fed.R.Civ.P. 43(a), it is unclear where the textual source of that power (if there need be any, which we doubt) lies. Until 1972, Rule 43(a) referred us generally to the rules of evidence applied in the courts of general jurisdiction of the forum state. Both plaintiff and defendant seem to have assumed that such was still the proper source of our authority, for they have cited mainly Pennsylvania cases. In fact, the new Federal Rules of Evidence neither refer us to state law on the order of presentation of evidence nor create a new federal rule on that subject.
However, it matters not at all whether we follow state or federal precedent on this issue; there is unanimous agreement that on rebuttal it is properly within the discretion of the trial judge to limit testimony to that which is precisely directed to rebutting new matter or new theories presented by the defendant's case-in-chief. See 6 Wigmore's Evidence § 1873 (3d Ed. 1940); Glen Alden Coal Co. v. Commissioners of Schuylkill County, 345 Pa. 159, 27 A.2d 239 (1942); Blair v. Turnpike Commission, 152 Pa.Super. 555, 33 A.2d 490 (1943); Feldman, Pennsylvania Trial Guide, § 9.1 (1973); French v. Hall, 119 U.S. 152, 7 S. Ct. 170, 30 L. Ed. 375 (1886); Hickok v. G.D. Searle & Co., 496 F.2d 444 (10th Cir. 1974); Hanrahan v. St. Vincent Hospital, 516 F.2d 300 (8th Cir. 1975). Conversely, the only cases in which the district court's discretion to exclude rebuttal testimony has been found to have been abused are those in which defendant's witnesses have presented an alternative theory or new facts or have otherwise created a need for a particularized response. See Weiss v. Chrysler Motors Corp., 515 F.2d 449 (2d Cir. 1975); Frankel v. Styer, 386 F.2d 151 (3d Cir. 1967); National Surety Corp. v. Heinbokel, 154 F.2d 266 (3d Cir. 1946). Such is not the case here.
We believe that we correctly applied the teaching of these authorities by requiring that plaintiff make specific offers of proof as to the points in GM's case that Mr. Baccini was prepared to rebut and by limiting the testimony to those points (concerning which Mr. Baccini was qualified to testify). Neither at trial, nor in its post trial motion, has plaintiff pointed to one area that was proper rebuttal (i. e., made necessary by defendants evidence) in which we forbade testimony. See also Fed.R.Ev. 702; United States v. Alker, 260 F.2d 135 (3d Cir. 1958), cert. denied, 359 U.S. 906, 79 S. Ct. 579, 3 L. Ed. 2d 571 (1959) ("admissibility of expert testimony . . . within the sound discretion of the trial judge"). We have also disposed of the cumulation point. Accordingly, plaintiff's "limitation of rebuttal" claim must be denied.
III. The "Unreasonably Dangerous" Concept and the Law of Strict Liability
Section 402A of the Restatement (Second) of Torts (1965) provides, in pertinent part:
One who sells any product in a defective condition unreasonably dangerous to the user or consumer . . . is subject to liability for physical harm thereby caused to the ultimate user . . ..
That provision, of course, both reflected and accelerated the development of strict liability in tort. Yet, clearly poised in the center of this strict liability provision and of the controversy about its future, stands the "unreasonably dangerous" language. For, *241 whether the "unreasonably dangerous" requirement so "rings in negligence" (see note 5, supra) as to undermine the policies which led to the birth of strict products liability is a subject of much dispute. It is not our purpose here to resolve that question by surveying the extensive literature, or by scrutinizing the words or syntax of § 402A and its comments. What we shall do is report our conclusions, forged in the trial crucible, about the difficulty of formulating a meaningful charge to guide a jury's determination of liability in conscious design choice cases where the unreasonably dangerous motion has been excised from § 402A.
We begin with the distinction between a manufacturing flaw and a design defect. The former concept relates to a product which was not manufactured in the manner that it was designed or intended to be. Guided by the excellent analysis of Professor Henderson, Judicial Review of Manufacturer's Conscious Design Choices: The Limits of Adjudication, 73 Colum.L.Rev. 1531, 1548 (1973), we parse the design defect category to distinguish inadvertent design errors and conscious design choices. Professor Henderson wrote:
At one end of the spectrum are risks of harm which originate in the inadvertent failure of the design engineer to appreciate adequately the implications of the various elements of his design, or to employ commonly understood and universally accepted engineering techniques to achieve the ends intended with regard to the product. At the other end of the spectrum are risks of harm which originate in the conscious decision of the design engineer to accept the risks associated with the intended design in exchange for increased benefits or reduced costs which the designer believes justify conscious acceptance of the risks.
We treat manufacturing flaws and inadvertent design errors in the same way. We do so not because they are both unintended, but because they are both subject to measurement against a built-in objective standard or norm of proper manufacture or design. Where, however, a conscious design choice has caused the injury, we are faced with quite a different problem; for there is no built-in objective standard by which the jury can measure the alleged defect. This result stems, at least in part, from the fact that a conscious design choice necessarily involves a trade-off among safety, utility, and cost. The trade-off may be obvious and may also be acceptable to the consumer. At the very least, it reflects the manufacturer's judgment of what would be acceptable if the terms of the trade-off were publicly known. However, the process of evaluating the trade-off, which represents the manufacturer's distillation of the forces of the marketplace, is a sophisticated one which complicates the process of products liability adjudication.
*242 When a judge instructs a jury that its task is to determine the existence vel non of a "defect," he must recognize that the term "defect," in common parlance, implies that there is a standard from which one has deviated. But our experience teaches us that, in the conscious design choice cases, where there is no other (available) standard, excision of the unreasonably dangerous concept denudes § 402A of its only vehicle for infusing into the notion of "defect" a meaningful guide to its determination. Dean Wade has written that in (conscious) design defect cases, the concept of defective condition standing alone is inappropriate, and that it has no independent meaning and is apt to prove misleading. Wade, Strict Tort Liability of Manufacturers, 19 Sw.L.J. 15 (1965). Accord, Ross v. Up-Right, Inc., 402 F.2d 943 (5th Cir. 1968). We agree. Professor Keeton believes that, in the area of design problems "defective" means unreasonably dangerous. Keeton, Product Liability and the Meaning of Defect, 5 St. Mary's L.J. 30, 32 (1973). At least to the extent of concluding that in a conscious design choice case "defect" should be defined in terms of unreasonableness of danger, we concur.
As we see it, the unreasonably dangerous concept, properly formulated, posits a risk-utility balancing test pursuant to which the jury makes a judgment as to the social acceptability of the conscious design choice trade-off. We agree with Professor Henderson that
[i]n cases involving liability for inadvertent design errors, the means employed to reach the intended ends are insufficient; in cases involving liability for conscious design choices, the intended ends themselves are out of step with prevailing social policies.
73 Colum.L.Rev. at 1548. However, we disagree with his assertion (see note 12, supra) that conscious design choice cases are beyond the competence of the Courts. We believe that a (properly instructed) jury can perform the necessary balancing test as well as any individual or agency, at least in the absence of a well-financed broadly jurisdictional federal fact-finding and standard-setting agency such as the Consumer Product Safety Commission could conceivably become.
*243 We do not pretend that our jury charge on "defect" in this case was a model of correctness or clarity. However, a recitation of relevant portions thereof will be helpful to an explication of our thesis. After distinguishing a conscious design defect from a manufacturing flaw and explaining the notion of crashworthiness (we charged the jury that GM had a duty to design its vehicles with a view to the foreseeability of accidents and to minimize their effect), we instructed the jury that a product is defective, even though made exactly as it was designed, if it is unreasonably dangerous to the users of the product. We then explained the "unreasonably dangerous" concept as follows:
In determining whether a product is unreasonably dangerous and therefore defective because of its design, there are a number of factors which you should consider. As you will see, when I have listed all these factors, you may have to balance them against each other, sort of like in a formula. In other words, some may cut one way and some the other. No one factor by itself will be all-important.
First, you should consider the likelihood that the product as thus designed will result in injury to a user. Applied to the present `crashworthiness' case, you should consider the likelihood that the '66 Toronado, as it was designed, would, in an accident situation, cause (or fail to protect an occupant from) injury. In this regard, you must in turn also consider the likelihood that a '66 Toronado would be struck in the rear on a highway: (1) at a closure speed such as you in fact find to have been present in this case; and (2) in the manner in which you find the collision to have occurred in this case. If you find that an accident of this kind was not unlikely to occur, that would be a factor to consider in the balancing formula on unreasonableness of danger. If you so find, you should also consider the likelihood that a Toronado designed in the manner of the '66 model would, in circumstances such as you find occurred, experience a fire resulting from ignition of the fuel system's gasoline vapors which would invade the passenger compartment. If you find such a likelihood of injury, you should consider that factor in deciding whether the automobile here was unreasonably dangerous.
The second factor you should consider in the unreasonably dangerous formula is the seriousness of potential injury in such circumstances.
*244 Third, you should consider the ability of the manufacturer, GM, to eliminate any unsafe characteristics which you find in the car, without impairing the usefulness of the car or significantly increasing its cost.
Fourth, you should consider whether the Toronado was dangerous to an extent beyond that which would be contemplated or expected by the ordinary user, considering the ordinary knowledge common to the community as to an automobile's characteristics. In this regard, you should consider the question of warnings, for I instruct you that a product can be rendered defective by reason of failure to give warnings as to risks or dangers in the product. As I have previously stated, the degree of warning depends upon the degree of danger since a greater degree of danger requires a greater degree of warning. However, you should also consider the ability of the manufacturer to notify or warn the user of the danger in a given situation, and the degree to which warning was meaningful.
In applying those factors to the formula you may find that some cut one way toward unreasonable danger and some the other away from that conclusion. It is for you, the jury, to evaluate them in light of the facts as you find them in the case, and to determine where the correct balance lies. You may of course find there to be other factors which appear from the evidence to shed light on whether the Toronado's design was defective, and if so you may consider them.
To recapitulate: I have told you that Mr. Bowman's second claim in this casewholly distinct from his claim of negligenceis based on the doctrine of Manufacturer's Liability for Defective Products. I have defined `defect' in terms of `unreasonableness of the danger,' because I believe that to have assigned you the task of determining whether there was a `defect' without such guidance would have been too difficult. However, I believe that it may also be helpful to give you a general definition of defect which you may also apply in your deliberations. In terms pertinent here, Webster defines `defect' to mean `an absence of something necessary to adequacy of function.' Applied to this case, the question would be: Did the design of the Toronado lack something necessary for adequate performance of its function namely, to provide a reasonably safe compartment for transportation of occupants of motor vehicles; that is, one designed not to be crashproof or to provide absolute safety against all risks of the road, but to provide reasonable safety against the foreseeable risks of the road.
In formulating the balancing test we have drawn heavily upon the works of Dean Wade.
*245 We believe, notwithstanding the Jones-Nix view in Berkebile, that our charge is not a negligence charge which focuses upon fault, but a products liability charge which focuses on the product. We concede that the ratio decidendi of a case thus submitted to the jury may be similar to that of the same case submitted on a negligence theory, and that it may be forcefully argued that some of the ingredients in the balancing formula are akin to considerations involved in negligence cases. On the other hand, let us posit a conscious design choice § 402A case where the manufacturer: (1) after exhaustive study and testing has adopted a given design which is slightly less safe than an alternative design but one which has slightly greater utility and can be sold at a slightly lower price; and (2) after considering the countervailing societal values has concluded that the public, in inflationary times, is concerned about "saving a dollar" and about a more useful product, and would prefer the slightly less safe product in return for the price and utility advantage. In a products liability lawsuit brought by someone injured by the product we believe it appropriate for a jury to apply the unreasonably dangerous risk-utility balancing formula which we have explicated to determine the presence or absence of liability. Our faith in the jury system is considerable. We have seen as many defendant's verdicts as plaintiff's verdicts *246 in products liability cases over the past several years, buttressing our conclusion that juries are as impartial as they are intelligent. In view of their broad community base they seem to us well-equipped to perform the fact-finding and judgmental tasks involved.
We conclude then that the unreasonably dangerous concept is and properly should remain a part of the law of strict liability in Pennsylvania. We believe that when the Pennsylvania Supreme Court definitively addresses the issue it will so hold. An order denying plaintiff's post trial motion follows.NOTES
 Our opinion on a discovery motion in this case is reported at 64 F.R.D. 62 (E.D.Pa.1974).
 Plaintiff does not (and could not) seriously contend that the verdict was against the weight of the evidence. The jury plainly could have gone either way based upon the superb expert testimony offered by both the plaintiff and GM. Out of the 18 points asserted in plaintiff's post trial motion, he has briefed but three; we will address only those in this opinion, one in a footnote (note 6, infra). We have carefully considered the other points and find them to be without merit. Moreover, the grounds for virtually all of our challenged rulings are amply spread upon the trial record.
 Mr. Severy testified that, at impact, the Rhodes vehicle was traveling at a speed between 42 and 50 m.p.h., resulting in a "closure speed" (considering the 15 m.p.h. speed of the Bowman vehicle which had just exited the parking lot of the Downingtown Inn where Mr. & Mrs. Bowman had dined) of 26-34 m.p.h.
 GM's accident reconstructionist, William Cichowski, testified that the Rhodes vehicle at impact was traveling at a speed of at least 65 m.p.h., with a closure velocity of at least 50 m.p.h. (assuming that the Bowman vehicle was traveling at 15 m.p.h.).
 In Berkebile, 337 A.2d at 899-900, Chief Justice Jones quoted with approval the statement of the California Supreme Court in Cronin v. J. B. E. Olson Corp., 8 Cal. 3d 121, 132-33, 104 Cal. Rptr. 433, 442, 501 P.2d 1153, 1161 (1972), that "unreasonably dangerous" was inconsistent with strict liability because it "rings in negligence." Accord: Glass v. Ford Motor Co., 123 N.J.Super. 599, 304 A.2d 562 (Superior Court, Law Division 1973). Point 17 of plaintiff's new trial motion reads:
17. The Court erred in failing to charge the jury on strict liability in accordance with Berkebile v. Brantly Helicopter Corp. . ..
 As Judge Huyett noted in Beron:
In Webb v. Zern, 422 Pa. 424, 220 A.2d 853 (1966), Pennsylvania joined the many jurisdictions that apply strict liability in tort principles to sellers of commercial products whose products cause injury to users or consumers. In so doing the Pennsylvania Supreme Court adopted the language of § 402A of the Restatement of Torts, Second, verbatim, and prior to 1975 it had never wavered in its precise adherence to the letter of that section. In the decade since § 402A was adopted the Pennsylvania Supreme Court has expressed unqualified approval for the language and the comments of § 402A, where doubts arose as to its proper interpretation.
402 F. Supp. at 1274.
 One of plaintiff's principal contentions in his brief in support of his motion for a new trial is that we should abstain from determining that very motion while plaintiff seeks a determination of Pennsylvania law from the State Supreme Court. Although we do not recall and have not been referred to any pretrial abstention motion, we do recall informal pretrial discussion of the possibility of "referring" the state law issue to that Court. (See Hart and Wechsler's The Federal Courts and the Federal System (2nd ed. 1973) at 1006 and n.3 for discussion and collection of authorities on federal court referral or certification of state law issues to state courts.) At that time, plaintiff admitted that no such procedure existed in Pennsylvania. Furthermore, plaintiff did not suggest any alternative means for obtaining a timely state decision on the issue and did not move for abstention. Plaintiff's present abstention request, which is in the nature of a suggestion sans formal motion, does nothing to obviate the problem of obtaining a state court decision on the legal issue. But even if plaintiff had formally so moved, and even if an adequate procedure existed, abstention would not be justified. Plaintiff does not distinguish the uncertainty and possibility of change in state law in this case from that which exists in garden variety diversity cases. We do not believe that mere uncertainty is sufficient to justify exercising our discretion to abstain. In addition, we find plaintiff's "suggestion," after it has lost its legal arguments before this court and after it has lost on the facts as determined by a jury, to be untimely.
The United States Supreme Court recently summarized the meaning of its prior abstention decisions in Colorado River Water Cons. Dist. v. United States, 424 U.S. 800, 96 S. Ct. 1236, 47 L. Ed. 2d 483 (1976). Speaking for the Court, Justice Brennan divided abstention into the three commonly recognized types. First, there is Pullman abstention, a development of the policy that federal constitutional issues should be decided only if they must be decided. In Railroad Comm'n of Texas v. Pullman Co., 312 U.S. 496, 61 S. Ct. 643, 85 L. Ed. 971 (1941), the Court required that a Three Judge Court abstain from determining a federal constitutional issue that could be avoided if the state courts found a state agency to have exceeded its (state) constitutional power. The second type of abstention is often associated with the case of Louisiana Power and Light Co. v. City of Thibodaux, 360 U.S. 25, 79 S. Ct. 1070, 3 L. Ed. 2d 1058 (1959). Thibodaux abstention does not involve avoidance of constitutional issues, but rather promotes comity interests where state law is unclear and special state interests are involved. Third, abstention has been required in a variety of cases in which a federal court has been asked to enjoin a state criminal proceeding. See Younger v. Harris, 401 U.S. 37, 91 S. Ct. 746, 27 L. Ed. 2d 669 (1971).
Clearly, plaintiff requests Thibodaux abstention. Just as clearly plaintiff does not demonstrate the appropriateness of his claim. We cannot but agree with plaintiff that the controlling law in diversity cases is state law. Erie R.R. v. Tompkins, 304 U.S. 64, 58 S. Ct. 817, 82 L. Ed. 1188 (1938). We do not agree, however, that a federal court, sitting in diversity, must or should abstain whenever state law is uncertain or even where the winds of change are awhirl. Were that the law, we would be required, or at least, permitted, to abstain in a large number of garden variety diversity cases. Indeed, it is common in cases which have been carried to the litigation stage, that the lawyers are able to posit, using legal creativity, some uncertainty in the underlying law, or some evidence of a change therein.
What the Supreme Court suggested in Colorado River Water Cons. Dist. is that abstention should remain the exception to be justified by special needs for comity. We find here no issue "of substantial public import whose importance transcends the result in the case then at bar." 96 S. Ct. at 1244, citing Thibodaux. Nor do we find a complex state managerial or administrative schema. See Burford v. Sun Oil Co., 319 U.S. 315, 63 S. Ct. 1098, 87 L. Ed. 1424 (1943); Alabama Public Service Comm'n v. Southern R. Co., 341 U.S. 341, 71 S. Ct. 762, 95 L. Ed. 1002 (1951). See generally 1 A.J. Moore, Federal Practice ¶ 0.203 (2d ed. 1953); Field, Abstention in Constitutional Cases: The Scope of the Pullman Abstention Doctrine, 122 U.Pa. L.Rev. 1071, 1150 n.212 (1974).
Not only do we not find special circumstances justifying abstention, but in fact we do find special circumstances demonstrating its inappropriateness in this case. Plaintiff chose the federal diversity forum at a time when state law was even more clearly set against his position than it is now. He knew that the proper forum to argue for a change in state law was the state judiciary. Having chosen this forum and lost before a jury, plaintiff now, and only now, seriously suggests abstention. To grant abstention to a party after he has argued the issue of state law to the court and lost, and after he has litigated his case before a jury and lost, would be to place an enormous financial strain on federal diversity litigants and to render the diversity jurisdiction, in such instances, a virtual farce.
 In his motion for new trial plaintiff assigns four points to this area of his claim:
11. The Court erred in sustaining defendant, General Motors' objection to the following question as not being proper rebuttal:
"Q. Mr. Baccini, do you agree with the statement of Mr. Elwell that a fuel tank system is reasonably safe if it can withstand a rear end impact of anything more than 15 miles per hour?" (N.T. 1731)
12. The Court erred in preventing the witness, A. E. Baccini, from giving an opinion in accordance with the question propounded by the plaintiff:
"Q. Based upon your experience, Mr. Baccini, do you have an opinion, taking a tank such as this into consideration, as to where it should be positioned to avoid the underride which you have just described to the jury?" (N.T. 1732, 1733)
13. The Court erred in refusing to prevent [sic] Mr. Baccini to rebut Mr. Elwell's testimony that "above 15 miles per hour you would need a tank to protect this fuel tank" and also [in ruling] that Mr. Baccini, in the light of his curriculum vitae and his vast experience in fuel-carrying containers was not sufficiently expert to testify in that area;
14. The Court erred in sustaining defendant's objection to the testimony of Mr. Baccini and the striking of his testimony of pages 1761 and 1762 on the grounds that it was not proper rebuttal.
[N.T. 1761 and 1762 contain Mr. Baccini's opinion as to the crashworthiness of the Toronado's fuel tank in the context of the design of that car. On objection, we struck his answer].
 There was nothing in plaintiff's offer of proof to suggest that Mr. Baccini's general theory of the accident and his views on the adequacy of the Toronado's design differed from that of plaintiff's prior expert, Mr. Severy. If plaintiff had so contended, then he would have had the great burden to convince us that there was adequate justification to allow a new theory, not suggested during discovery or in pretrial memos or conferences, to be presented on rebuttal.
 The trial was lengthy not only because of the extensive testimony, but also because a juror, a witness, and the judge contracted the flu, resulting in several days recess.
 In this regard we note that, had Mr. Baccini's testimony been offered in plaintiff's case-in-chief, we might have granted more leeway, despite cumulation. The reason for this difference lies in the nature of rebuttal, rather than in the nature of cumulation.
 It should be recorded here that Professor Henderson believes that conscious design decisions are "polycentric" in character and that the passing of judgment upon those decisions is beyond the competence of Courts. Indeed, Professor Henderson's even broader thesis is stated in his opening sentence: "Courts are inherently unsuited to the establishing of product safety standards in cases involving the liability of manufacturers." A polycentric problem, the type described by Professor Henderson as being beyond the limits of adjudication, is described as follows:
. . . polycentric problems are many-centered problems, in which each point for decision is related to all the others as are the strands of a spider web. If one strand is pulled, a complex pattern of readjustment will occur throughout the entire web. If another strand is pulled, the relationships among all the strands will again be readjusted. A lawyer seeking to base his argument upon established principle and required to address himself in discourse to each of a dozen strands, or issues, would find his task frustratingly impossible. As he moved from the first point of his argument to the second and then to the third, he would find his arguments regarding the earlier points shifting beneath him. Unlike most of the traditional types of cases in which litigants are able, in effect, to freeze the rest of the web as they concentrate upon each separate strand, the web here retains its natural flexibility, adjusting itself in seemingly infinite variations as each new point, or strand, in the argument is reached.
73 Colum.L.Rev. at 1536. Henderson credits Professors Lon Fuller and Michael Polanyi as creators of the concept of polycentricity. Id. at 1534.
 Several recent law review discussions of Berkebile are in accord. See generally, Comment, Elimination of "Unreasonably Dangerous" from § 402A The Price of Consumer Safety?, 14 Duquesne L.Rev. 25 (1975); Recent Cases, Products Liability, 80 Dick.L.Rev. 633 (1976); Comment, Pennsylvania Supreme Court Review, 1975, 4 Temp.L.Q. 786 (1976).
The Duquesne Commentator, supra at 51, writes:
Those courts that have eliminated "unreasonably dangerous" on the grounds that a bifurcated standard of proof is unnecessary and burdensome for the plaintiff, have discarded the only element of § 402A relevant to design problems. Without "unreasonably dangerous," the courts are left with two alternatives: (1) conclude a defect existed since a different design would have prevented injury, or (2) conclude there was no defect since the product was exactly as intended. The "unreasonably dangerous" standard, on the other hand, suggests an intermediate approach. . . .
See also Collins v. Musgrave, 28 Ill.App.3d 307, 328 N.E.2d 649 (1975) (retaining "unreasonably dangerous" in definition of defect as against Cronin & Glass challenge (see supra, note 5).
But see Recent Developments, Products Liability, 21 Vill.L.Rev. 794 (1976) (supporting the Jones-Nix view on the grounds that an "unreasonably dangerous" charge undermines strict liability policy, confuses the distinction between strict liability and negligence, and is unnecessary to make "defective" meaningful.)
Several commentators have expressed the view that the elimination of the "unreasonably dangerous" concept from § 402A may lead to absolute liability in favor of a plaintiff if he merely proves a design related injury. Case note, Products Liability "Unreasonable Danger" Eliminated from the Theory of Strict LiabilityThe Restatement Restated, 42 Fordham L.Rev. 943, 947 (1974); Temple L.Q., supra, at 790.
 We note in this regard that in Comment g to § 402A entitled "Defective Condition," defective is defined at least in part in terms of the "unreasonably dangerous" notion.
 See Consumer Product Safety Act, 15 U.S.C. § 2051 et seq. (1974).
 We informed the jury that GM had no duty to produce a crash proof automobile:
. . . General Motors does not, by way of example, have an obligation to manufacture a passenger car built like a tank so that a passenger would escape injury even if the car were hit by a fast moving locomotive. Neither, if the vehicle left the roadway and landed in a river, could it be argued that the manufacturer has a duty to produce an automobile that would float. Concomitantly, GM had no obligation to use any particular design or the safest design which could be conceived of under the state of the art at the time, i. e., what was known. However, General Motors does have an obligation to provide more than a movable platform capable of transporting occupants from one place to another along deserted streets; it has the obligation to provide the occupants with a vehicle which is not unreasonably dangerous in which to make the journey.
The correctness of this approach, and the rejection of the principles of Evans v. General Motors Corp., 359 F.2d 822 (7th Cir. 1966) now seems well established. See generally Comment. Automobile Crashworthiness: Evans Takes a Backseat, 21 Vill.L.Rev. 72 (1975-76). Huddell v. Levin, 537 F.2d 726 (3d Cir. 1976) (Aldisert, J.). Dyson v. General Motors Corp., 298 F. Supp. 1064 (E.D.Pa.1969) (Fullam, J.) Larsen v. General Motors Corp., 391 F.2d 495 (8th Cir. 1968). As Judge Aldisert noted in Huddell:
We take it as beyond peradventure that an automobile manufacturer today has some legal obligation to design and produce a reasonably crashworthy vehicle. The manufacturer is not required to design against bizarre accidents; the manufacturer is not required to produce an accident-proof vehicle. But the manufacturer is required to take reasonable steps within the limitations of cost, technology, and marketability to design and produce a vehicle that will minimize the unavoidable danger. Rephrased in the terminology of strict liability, the manufacturer must consider accidents as among the "intended" uses of its product.
537 F.2d at 735.
 We believe that the matter of warnings is among the most significant factors for the jury's consideration on the unreasonably dangerous issue (see discussion and formulations, infra). Professor Henderson agrees. 73 Colum.L.Rev. at 1563. The role of warnings was muted in this case inasmuch as the main value of warnings is to encourage consumer safety, yet no act of a driver (save perhaps driving only on rural roads or not at all) can prevent one's being struck in the rear. Of course, the warning can serve to give the consumer the choice of whether to purchase the product at all; but that factor is not helpful where the injured party is a bystander or employee.
 Dean Wade proposed a set of criteria by which the reasonableness of product danger can be measured, which reflects a
. . . realization that all products involve some degree of danger; and that what is being censured is the manufacturer's failure to strike a proper balance between design safety and social desire for utility and aesthetics. They include:
(1) The usefulness and desirability of the product its utility to the user and to the public as a whole.
(2) The safety aspects of the product the likelihood that it will cause injury, and the probable seriousness of the injury.
(3) The availability of a substitute product which would meet the same need and not be as unsafe.
(4) The manufacturer's ability to eliminate the unsafe character of the product without impairing its usefulness or making it too expensive to maintain its utility.
(5) The user's ability to avoid danger by the exercise of care in the use of the product.
(6) The user's anticipated awareness of the dangers inherent in the product and their avoidability because of general public knowledge of the obvious condition of the product, or of the existence of suitable warnings or instructions.
(7) The feasibility, on the part of the manufacturer, of spreading the loss by setting the price of the product or carrying liability insurance. [footnote omitted]
Wade, On the Nature of Strict Tort Liability-Products, 44 Miss.L.J. 825, 837-38 (1973).
Professor Fisher has also suggested some factors which could be used in a balancing test:
In deciding when to impose strict liability courts should consider, in light of the facts of the particular case, the merits of the policies underlying strict liability and balance these considerations against countervailing factors. Some of the factors that should be considered are as follows:
I. Risk Spreading
A. From the point of view of consumer.
1. Ability of consumer to bear loss.
2. Feasibility and effectiveness of self-protective measures.
a. Knowledge of risk.
b. Ability to control danger.
c. Feasibility of deciding against use of product.
B. From point of view of manufacturer.
1. Knowledge of risk.
2. Accuracy of prediction of losses.
3. Size of losses.
4. Availability of insurance.
5. Ability of manufacturer to self-insure.
6. Effect of increased prices on industry.
7. Public necessity for the product.
8. Deterrent effect on the development of new products.
II. Safety Incentive.
A. Likelihood of future produce improvement.
B. Existence of additional precautions that can presently be taken.
C. Availability of safer substitutes.
Fischer, The Meaning of Defect, 39 Missouri L.Rev. 339, 359 (1974).
In appropriate cases other factors listed by Wade and/or Fischer could be incorporated into the unreasonably dangerous charge.
 But cf. Dreisonstok v. Volkwagenwerk, A.G., 489 F.2d 1066 (4th Cir. 1974), which applied a similar balancing test to a conscious design choice case under the negligence label, and Dorsey v. Yoder, 331 F. Supp. 753 (E.D.Pa. 1971), aff'd 474 F.2d 1339 (3d Cir. 1973), in which Judge Masterson tested § 402A products liability for "defects" under the Wade formulation without attempting to distinguish "strict" liability from liability for "negligence."
It may be that the balancing test that we have posited under § 402A is quite similar to the test ordinarily applied in negligence. However, as noted in the text, the focus is different, and the charge emphasizes the particularized policies underlying modern products liability. Cf. the relationship between negligence concepts and the duties of landowners and the nascent trend to "abolish the categories." Prosser, The Law of Torts, 4th Ed. at 351-415 (1971).
 See Restatement (Second) Torts § 398 which provides:
§ 398. Chattel Made Under Dangerous Plan or Design
A manufacturer of a chattel made under a plan or design which makes it dangerous for the uses for which it is manufactured is subject to liability to others whom he should expect to use the chattel or to be endangered by its probable use for physical harm caused by his failure to exercise reasonable care in the adoption of a safe plan or design.