In 1997, three notable events occurred which will significantly impact the future of products liability law in Connecticut. First, the American Law Institute ("ALI") altered its position with regard to the elements of proof in products liability cases when it unanimously adopted the Restatement (Third) of Torts: Products Liability ("Restatement (Third)"). Second, the Connecticut Supreme Court addressed several issues of first impression when it released its decision in Potter v. Chicago Pneumatic Tool Co., 241 Conn. 199, 694 A.2d 1319 (1997). Third, the court released its decision on several pivotal issues in Wagner v. Clark Equipment Co. Inc., 243 Conn. 168, 700 A.2d 38 (1997). Significantly, the Potter court expressly rejected what is perhaps the most controversial provision of the Restatement (Third), Section 2(b), which, in most cases, requires a plaintiff to demonstrate the existence of a reasonable alternative design in order to prevail in a design defect case. In lieu of the Restatement approach, the Potter court loosened its long standing grip on the "consumer expectation" test as the sole measure of a product defect and introduced a "modified consumer expectation" test to be used in conjunction with the traditional test. Unfortunately, the court provided only limited direction with regard to how and when the new test is to be used, leaving judges and litigants to grapple with those issues. Additionally, the court ratified the use of state-of-the-art evidence in design defect cases and delineated a burden-shifting analysis to be utilized when considering alterations or modifications made to a product. An examination of the courtÆs rulings on these issues leads to the conclusion that, while expressly rejecting the Restatement (Third) standard, the courtÆs decision is substantially aligned with the Restatement approach. This issue is discussed in detail below and will surely provide fodder for future litigation. In Wagner, the court held, for the first time in the products liability context, that two or more independent intervening forces may combine to create a superseding cause of a plaintiffÆs injury, thereby relieving a defendant of liability. The Wagner court also ruled that evidence of compliance with federal regulations is admissible to demonstrate the lack of a product defect and that evidence of postaccident modification may be used to demonstrate the existence of possible alternative designs (i.e. the presence of a product defect). Independently, each of these events is significant. In combination, the three have dramatically altered the products liability landscape in Connecticut. Not only has the substantive law changed, so too has the manner in which product liability claims must be litigated. Specifically, the evidence now needed to prosecute or defend these claims is entirely different from that which was required pre-Potter. Now more than ever, parties will be required to present evidence of "feasible alternative designs," state-of-the-art designs, along with pre-accident and postaccident modifications. The parties who identify these changes will succeed. The parties who do not will never know why they failed. I. Potter v. Chicago Pneumatic Tool Co. In Potter v. Chicago Pneumatic Tool Co., 241 Conn. 199, 694 A.2d 1319 (1997), several employees of General DynamicsÆ Electric Boat shipyard sued the Chicago Pneumatic Tool Company, Stanley Works and Dresser Industries, claiming that various tools which they used during the course of their employment as "grinders" at Electric Boat were defectively designed. According to the plaintiffs, the defective design resulted in excessive vibration of the tools causing the plaintiffs to suffer "Hand Arm Vibration Syndrome," also known as "White Finger Disease." The plaintiffs further alleged that the defendants had failed to warn them regarding the potential danger of using the tools and that this failure also resulted in harm to the plaintiffs. The trial court entered judgment in favor of the plaintiffs following the conclusion of a jury trial. The defendants subsequently appealed from the judgment claiming, among other things, that: (1) the plaintiffs had failed to offer evidence sufficient to establish a defective design; (2) the trial court improperly instructed the jury regarding their consideration of any alteration or modification of the tools; and (3) the trial court improperly instructed the jury on the use of state-of-the-art evidence in design defect cases. The Connecticut Supreme Court agreed that the trial court had improperly shifted the burden of proof to the defendants with regard to the issue of alteration or modification and ordered a new trial on that ground. In reaching its conclusion, the court made several rulings which have significantly altered the law of products liability in Connecticut. A. Introduction of the "Modified Consumer Expectation" Test 1. Proof of a Design Defect Generally Perhaps the most significant element of the Potter decision is the courtÆs introduction of the "modified consumer expectation" test as a measure of a product defect. Unfortunately, despite the courtÆs detailed analysis of this issue, its conclusions create more questions than answers. Undertaking a comprehensive review of the history of products liability law, the court recognized the modern rule as set forth in Section 402A of the Restatement (Second), which holds manufacturers "strictly liable for unreasonably dangerous products that cause injury to ultimate users." Potter, 241 Conn. at 210. Under the Restatement (Second) formulation, an "unreasonably dangerous" (and therefore defective) product is defined as a product "dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics." Restatement (Second) Section 402A, comment (I). Commonly referred to as the "consumer expectation test," the Restatement (Second) rule was, prior to Potter, well established as the only test to be employed in Connecticut products liability actions. Challenging ConnecticutÆs long history of adherence to the consumer expectation test as the measure of a productÆs allegedly defective nature, the defendants in Potter urged the court to adopt a different test. Specifically, they urged adoption of the test set forth in Section 2(b) of the draft Restatement (Third), which provides: A product . . . is defective in design when the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design by the seller or other distributor, or a predecessor in the commercial chain of distribution, and the omission of the alternative design renders the product not reasonably safe. 1 This test would require a plaintiff to prove that, at the time the defendants manufactured their products, a feasible alternative design existed which was safer than the defendantsÆ design, thus increasing the burden of proof in design defect cases. Addressing this issue, the court noted that courts in other jurisdictions have considered the existence of a feasible alternative design when determining whether a product is defective. See Barker v. Lull Engineering Co., 20 Cal. 3d 413, 573 P.2d 443 (1978). The Potter court specifically noted, however, contrary to the Restatement ReportersÆ Note to comment (c) of the draft Restatement (Third),2 "that the majority of jurisdictions do not impose upon plaintiffs an absolute requirement to prove a feasible alternative design." Potter, 241 Conn. at 216. (emphasis in original).3 The court concluded that "the feasible alternative design requirement imposes an undue burden on plaintiffs that might preclude otherwise valid claims from jury consideration." Id. at 217. Specifically, the court stated that this rule would require expert testimony even in cases where a design defect could be inferred from circumstantial evidence, an inference long recognized in Connecticut product liability cases. Id. at 217-18. See also Standard Structural Steel Co. v. Bethlehem Steel Corp., 597 F. Supp. 164, 183 (D. Conn. 1984); Living & Learning Centre, Inc. v. Griese Custom Signs, Inc., 3 Conn. App. 661, 664, 491 A.2d 433 (1985).4 Additionally, a product could be in a defective condition unreasonably dangerous to the user even in the absence of a safer alternative design. "Other products, including some for which no alternative exists, are so dangerous and of such little use that . . . a manufacturer would bear the costs of liability of harm to others." Potter, 241 Conn. at 219,5 citing OÆBrien v. Muskin Corp., 94 N.J. 169, 184, 463 A.2d 298 (1983). Accordingly, the court expressly rejected the defendantsÆ invitation to adopt the position set forth in Section 2(b) of the draft Restatement (Third) and instead continued its adherence to the traditional consumer expectation test. As a result, a plaintiff in Connecticut is still not specifically required to demonstrate the existence of a feasible alternative design in order to make out a prima facie design defect case as suggested by the Restatement (Third). 2. The "Modified Consumer Expectation Test" After concluding that Connecticut does not require proof of a feasible alternative design, the court stepped away from its traditional position by stating "that there may be instances involving complex product designs in which an ordinary consumer may not be able to form expectations of safety." Potter, 241 Conn. at 219. In such cases; i.e., where the jury is unable to determine whether the product in question did or did not meet the safety expectations of the ordinary consumer, a modified consumer expectation test is to be utilized. The court described the function of the modified test as providing the jury with evidence concerning the risks and utility of a given product and then inquiring as to whether a reasonable consumer would consider the product unreasonably dangerous based upon those factors. Id. at 220. Under this "modified consumer expectation" test, the factors a jury may consider include, but are not limited to: (1) the usefulness of the product; (2) the likelihood and severity of danger posed by the product; (3) the feasibility of an alternative design; (4) the financial cost of an improved design; (5) the ability to reduce danger without impairing its usefulness or making the product too costly; and (6) the feasibility of spreading loss by increasing the productÆs price. Id. at 221. Notably, these factors are similar to those set forth in comment (f) to Section 2 of the Restatement (Third), the same section the court previously rejected. Indeed, when the modified consumer expectation test applies, the courtÆs decision specifically indicates that the existence of a feasible alternative design is now one factor a Connecticut plaintiff may present to a jury in order to establish that the risks of a particular product outweigh its utility and that it is therefore defective. (Part III of the courtÆs opinion addressing state-of-the-art evidence suggests that, in most cases, a plaintiff must present this evidence in order to prevail even under the ordinary test. See below.) With the introduction of the modified consumer expectation test as a complement to the ordinary test, the court provides an additional tool to be used in determining whether a product is defectively designed. Unfortunately, the court offers scant guidance as to when each test is to be employed. The court creates confusion in this regard first by suggesting that the boundary between the ordinary and modified consumer expectation tests is established simply by determining whether the product design at issue is simple or complex. "There may be instances involving complex product designs in which an ordinary consumer may not be able to form expectations of safety." Potter, 241 Conn. at 219. As the courtÆs own discussion of the issues makes clear, however, the simplicity or complexity of the product is not the dispositive inquiry. It is instead a question of whether or not the average consumer knows how safe the product could be. For example, the Potter court cites with approval to the California Court of AppealsÆ decision in Akers v. Kelley Co., 173 Cal. App. 3d 633, 651 219 Cal. Rptr. 513 (1985), where the court stated "[t]here are certain kinds of accidents ù even where fairly complex machinery is involved ù [that] are so bizarre that the average juror, upon hearing the particulars, might reasonably think: æwhatever the user may have expected from that contraption, it certainly wasnÆt that.Æ" Potter, 241 Conn. at 222, citing Akers, 173 Cal. App. 3d at 651. In such a circumstance, the ordinary test would be applicable in spite of the apparent complexity of the product. Accordingly, in order to properly determine which test should be used, the trial court must consider the product at issue and the circumstances under which the harm occurred in order to determine whether under those circumstances the productÆs ordinary users could be anticipated to form safety expectations about the product. The real uncertainty over which test should apply in a given case arises not from the courtÆs formulation of the modified test, but rather from the courtÆs effort to provide direct guidance as to when each test should be used. The court states that the ordinary consumer expectation test is to be used "when the everyday experience of the particular productÆs users permits the inference that the product did not meet minimum safety expectations." Potter, 241 Conn. at 222. The modified consumer expectation test should be used "when the particular facts do not reasonably permit the inference that the product did not meet the safety expectations of the ordinary consumer." Id. These statements appear to tie the distinction between the two tests not to the ability of the ordinary consumer to form expectations of safety, but rather to the plaintiffÆs ability to establish a product defect. As a result, the distinguishing criteria utilized by the court contradicts its stated concern that there are cases "in which an ordinary consumer may not be able to form expectations of safety." Id. at 219. In an effort to explain the use to be made of the modified consumer expectation test, the Potter court cites with approval to the California Supreme CourtÆs decision in Soule v. General Motors Corp., 8 Cal. 4th 548, 882 P.2d 298 (1994). Id. The Soule court noted that ordinary consumer expectations cannot always be used as the exclusive criteria for evaluating design defects since in many cases the consumer has no expectations because he or she does not know how safe the product could be. Soule, 8 Cal. 4th at 562 (discussing Barker, 20 Cal. 3d 413). The Soule court held that the ordinary consumer expectation test is to be utilized in cases where the "everyday experience of the productÆs users permits a conclusion that the productÆs design violated minimum safety assumptions, and is thus defective regardless of expert opinion about the merits of the design." Id. at 567. Alternatively, the facts are subject to a risk-benefit analysis in which the jury weighs the benefits of the design versus the risks inherent in the design. Id. at 566-67. The Soule court concluded that "[u]nless the facts actually permit an inference that the productÆs performance did not meet the minimum safety expectations of its ordinary users, the jury must engage in the balancing of risks and benefits required by the second prong of Barker." Id. at 568. Thus, "[t]he crucial question in each individual case is whether the circumstances of the productÆs failure permit an inference that the productÆs design performed below the legitimate, commonly accepted minimum safety assumptions of its ordinary consumers." Id. at 568-69. Following Soule, the Potter court introduced the modified consumer expectation test with the implication that it would be utilized in circumstances where the ordinary consumer was not likely to have an expectation as to the safety standards applicable to a certain product. However, the court created uncertainty as to when the modified consumer expectation test should apply by also adopting the language from Soule which suggests that the application of the modified test is contingent not upon whether the ordinary consumer can form expectations of safety, but rather upon whether or not the plaintiff can prove a product defect under the ordinary consumer expectation test. A strict reading of the Potter courtÆs language suggests that the new test exists only to provide a plaintiff with a second opportunity to prove the case when he is unable to prove the existence of a defect under the ordinary consumer expectation test; a fault obviously derived from the courtÆs reliance on Soule, where the California Supreme Court made the same mistake. This subtle but significant shift in emphasis from the ability of the ordinary consumer to form expectations regarding a productÆs safety to the ability of a plaintiff to prove his case, creates a gap in the courtÆs reasoning and leaves a substantial question regarding the application of the modified consumer expectation test. The Potter court correctly noted that it is within the discretion of the trial court to determine which test is to be provided to the jury. However, the court creates further confusion by suggesting that it may be appropriate in certain circumstances to instruct the jury on both the ordinary consumer expectation test and the modified consumer expectation test. This suggestion is inconsistent with the courtÆs statements addressing the proper use of the two tests. Specifically, the Potter court stated that the ordinary test is to be applied where the facts are sufficient to permit a finding that the everyday experience of the particular productÆs users permits the inference that the product did not meet minimum safety expectations, and that the modified test should be used when the particular facts do not reasonably permit such an inference. Setting aside the problems discussed above, the trial court will be compelled to use its discretion in order to determine whether the facts of the case warrant the application of the ordinary test or the modified test. This analysis is mutually exclusive: the facts of the case either will or will not support the inference which the court requires to be drawn. If the facts support the inference, then the ordinary test should be used; if they do not, then the modified test should be used. There is no room for overlap in the application of these tests, and the courtÆs suggestion that the jury might appropriately be instructed on both tests is confusing and contradictory. Indeed, the only scenario under which the jury might be instructed on both tests is one in which the question of which test is to be applied is left to the jury. The jury could then decide whether the facts permit the required inference and apply the appropriate test. Of course, this procedure flies directly in the face of the courtÆs statement that the question of which test to apply rests within the discretion of the trial court. The Supreme Court adopted the modified test to address circumstances where "an ordinary consumer may not be able to form expectations of safety." Potter, 241 Conn. at 219. The threshold question in determining which test should apply, therefore, should be whether an ordinary consumer is able to form expectations about safety, considering the product at issue and the circumstances under which the harm occurred.6 It is classically a function of the court to decide what law is applicable to a case based on the pleadings and the evidence submitted. The jury may appropriately decide whether one or the other (or both) tests have been met, but the jury may not appropriately decide which test should apply. Accordingly, there exists no scenario under which the jury could be instructed on both tests. A trial court should not determine the applicability of either test by evaluating the sufficiency of the evidence to support a verdict for the plaintiff under the ordinary consumer expectation test. The question is whether the circumstances of the case, as framed by the pleadings and detailed in the evidence, are sufficient to have enabled an ordinary consumer to form expectations of safety. Tempting as it might be, the trial court must not delegate this function to the jury. B. State-Of-The-Art Evidence The Potter court also addressed the trial courtÆs instruction to the jury that the state- of-the-art "defense" raised by the defendants was restricted to the plaintiffsÆ failure to warn claim only. The defendants argued that not only was this "defense" equally applicable to the plaintiffsÆ design defect claim, but that it constituted an affirmative defense which, if proven, comprised a complete defense to the plaintiffsÆ claims. In reply, the plaintiffs argued that such a "defense" would be inappropriate in the context of strict products liability "because it improperly diverts the juryÆs attention from the productÆs condition to the manufacturerÆs conduct." Id. at 243. Considering this issue of first impression in Connecticut, the Potter court first addressed whether state-of-the-art evidence, defined as "the level of relevant scientific, technological and safety knowledge existing and reasonably feasible at the time of design," Id. at 247, is admissible in design defect cases. Prior decisions established that such evidence is relevant and admissible in product liability cases involving allegedly inadequate warnings, Tomer v. American Home Products Corp., 170 Conn. 681, 368 A.2d 35 (1976), but no Connecticut appellate-level court had previously considered its applicability in design defect cases. Although the court noted that other courts were split on the issue, it expressly adopted the majority view, noting that such evidence "is relevant and assists the jury in determining whether a product is defective and unreasonably dangerous, but that it is not dispositive on the issue of liability." Id. at 247-49. The court took care to note that while state-of-the-art evidence could be dispositive on the facts of a particular case, such evidence is not an affirmative defense which, if proven, automatically relieves a defendant of liability. Id. at 253. Accordingly, state-of-the-art evidence is now admissible under both the ordinary consumer expectation test and the modified consumer expectation test. With regard to the former, such evidence "supplies the jury with a relevant basis on which to determine what the ordinary consumer would expect with respect to safety features available at the time of manufacture." Id. at 252. With respect to the latter, such evidence is merely one factor to be considered as part of the risk-utility analysis. Id. The courtÆs holding on this subject is significant not only because it addresses an issue of first impression in Connecticut, but also because it significantly alters the impact of the courtÆs holding in Part I of the opinion where the court stated that a plaintiff is not required to demonstrate the existence of a feasible alternative design as part of a prima facie case. The court specifically noted that if sufficient evidence is "produced to warrant an instruction, the jury may properly consider the state of the art in determining whether the defendantsÆ [products] were defectively designed and unreasonably dangerous."7 Because a defendant now has the right to introduce state-of-the-art evidence, and because the only way to refute a showing that a product embodies "the level of relevant scientific, technological and safety knowledge existing and reasonably feasible at the time of design" is to demonstrate the converse; i.e., that a feasible alternative design existed, a plaintiff now must address the feasible alternative design issue. Moreover, since a defendant may always put on state-of-the-art evidence, regardless of whether the ordinary or modified consumer expectation test is to be used, a plaintiff who declines to introduce evidence of a feasible alternative design now does so at his peril. As a result, despite the courtÆs explicit holding that a plaintiff need not demonstrate the existence of a feasible alternative design as part of his prima facie case, its equally explicit holding allowing the introduction of state-of-the-art evidence in design defect cases creates a de facto feasible alternative design requirement in Connecticut of which counsel for both plaintiffs and defendants should be keenly aware.8 C. The "Lack of Substantial Change" Requirement and the Alteration/Modification Defense The issue which the Potter court ultimately found dispositive was whether the trial court improperly shifted the burden of proving causation to the defendants by instructing the jury that certain alterations made to the tools by Electric Boat would bar liability only if the defendants proved that the alterations had been the sole proximate cause of plaintiffsÆ injuries. While the Supreme Court agreed that the jury instruction improperly shifted the burden of proof to the defendants, it disagreed that the use of the sole proximate cause standard was incorrect as a matter of law. Under the traditional common law standard as embodied in Section 402A of the Restatement (Second), a plaintiff is required to prove as part of his case in chief that a product reached the plaintiff without a substantial change in the condition in which it was sold. "In other words, the defendant would be absolved of liability under [Section] 402A if the plaintiffÆs injury was the result of an alteration or modification to the product made by a third party." Potter, 241 Conn. at 230. In Connecticut, the common law standard has been superseded by Section 52-572p of the Connecticut General Statutes, which provides that a manufacturer is not liable for harm that would not have occurred but for a third partyÆs alteration or modification of the product. Id. at 229-30. This defense does not apply, however, if the alteration or modification was: (1) done in accordance with the manufacturerÆs instructions or specifications; or (2) done with the manufacturerÆs consent; or (3) the result of conduct that the manufacturer reasonably should have anticipated. Id. at 229; see also Connecticut General Statute Section 52-572p. Because Section 52-572p must be strictly construed to comport with the common law rule, the Potter court held that the statute does not create an affirmative defense to be raised by a defendant, but instead, merely sets forth an element of the plaintiffÆs prima facie case. Id. at 230. It is important to note that while the standard embodied in Connecticut General Statutes Section 52-572p is similar to that set forth in the Restatement (Second), it is not identical. Significantly, Section 52-572p is based on the Draft Uniform Product Liability Act ("Draft Act") and utilizes "but for" language. Such language has previously been interpreted by the court to mean "causation in fact" which in turn has been interpreted as not necessarily referring to a sole cause of injury "because there may indeed be several causes in fact . . . The test for cause in fact is, simply, would the injury have occurred were it not for the actorÆs conduct." Id. at 232 (citing Doe v. Manheimer, 212 Conn. 748, 757, 563 A.2d 699 (1989)).9 Even though the "but for" language arguably does not refer to sole proximate cause, the Potter court nevertheless concluded that the commentary to the Draft Act clarifies the standard, indicating that a manufacturerÆs liability is limited only if the subsequent alteration or modification is indeed the sole proximate cause of the plaintiffÆs injury. Id. at 232-33. Where the original defect is merely a contributing or concurrent proximate cause of the harm in conjunction with the subsequent alteration, the manufacturer remains liable. However, where the subsequent alteration "breaks the chain of causation, akin to an intervening superseding cause in negligence law," the manufacturer will be relieved of liability. Id. at 23310 Because Connecticut law provides that a plaintiff must prove that a product reached him without substantial change in condition, and because the Potter defendants argued on appeal that the trial courtÆs instruction improperly shifted to them the burden of proof on this element of the plaintiffÆs case, the court was required to discuss the partiesÆ respective burdens of proof on alteration or modification at length. Ultimately, the court adopted a burden-shifting analysis "consistent with the policies underlying strict products liability" and "adopted by the majority of courts that have considered the allocation of burden of proof in cases involving the alteration or modification defense." Potter, 241 Conn. at 237. Under this burden-shifting procedure, the plaintiff, as part of his prima facie case, must first plead and prove "that the product in question was intended to and did reach the ultimate consumer without substantial change in condition." Id. at 236. The defendant must then raise the alteration or modification defense by "specifying a modification or alteration that was the sole proximate cause, and not merely a substantial contributing factor, of the plaintiffÆs harm." Id. Failure to invoke the statute in a special defense specifically identifying an alleged alternation or modification, will deprive a defendant of the opportunity to rely on any alteration or modification of the product to counter the plaintiffÆs evidence that the product reached the consumer without substantial change in condition. The defendant also bears the burden of producing evidence of the specific alteration or modification which is alleged to have been the sole proximate cause of the plaintiffÆs harm. Once the defendant has produced this evidence, the burden shifts back to the plaintiff to rebut the claim of substantial change by proving "that the harm would have occurred notwithstanding the alteration or modification." Id. at 237. Alternatively, the plaintiff may establish that the alteration or modification falls under one of the exceptions set forth in Connecticut General Statutes Section 52-572p; as previously discussed. The court was careful to note, that "[a]t all times, . . ., the ultimate burden of persuasion (or, put another way, the risk of nonpersuasion) that the product reached the ultimate consumer without substantial change remains with the plaintiff." Id. The significance of this portion of the Potter decision is that, notwithstanding the "defense" established by Section 52-572p, a plaintiff still must prove, as part of his prima facie case, that the product in question reached him without a substantial change in condition, and defendants must be certain to specially plead and prove a specific alteration or modification that caused the plaintiffÆs injury. Further, defendants should be aware that even if the jury determines that an alteration or modification was indeed the sole proximate cause of a plaintiffÆs injuries, such a finding does not necessarily relieve a defendant of liability if the alteration or modification falls within the purview of Connecticut General Statutes Section 52-572p. Potter, 241 Conn. at 234, n. 25. Finally, both plaintiffs and defendants must consider the impact of Wagner on this issue. Although a particular alteration or modification may not constitute an intervening superseding cause, a combination of such changes or the addition of other intervening events may combine to break the chain of causation. (See discussion of the Wagner case, below). D. Potter in Brief In sum, the court in Potter: (1) has recognized the use of a risk-utility analysis in cases where a productÆs alleged failure does not fall within the ordinary consumerÆs realm of expectations regarding safety; (2) has established a de facto feasible alternative design requirement despite its explicit rejection of Section 2(b) of the Restatement (Third) by virtue of its additional holding that state-of-the-art evidence is admissible in all design defect cases; and (3) has adopted a burden-shifting analysis with regard to claims of alteration or modification to the product, which reinforces and clarifies the partiesÆ burdens with respect to the requirement that a plaintiff establish that the product in question reached him or her without substantial change in condition. II. Wagner v. Clark Equipment Co., Inc. The second new piece in ConnecticutÆs products liability puzzle is Wagner v. Clark Equipment Co, Inc., 243 Conn. 168, 700 A.2d 38 (1997). In Wagner, the Connecticut Supreme Court began to explore the reach of the Potter decision and continued its effort to gradually clarify the landscape of Connecticut Products Liability law. A. Superseding Cause of Injuries The primary issue in Wagner was whether, in the strict products liability context, two or more intervening forces may combine so as to create a superseding cause of a plaintiffÆs injuries, thereby relieving a defendant of liability. The court ruled in the affirmative. Kevin Wagner was knocked to the ground by a forklift which then ran over his foot, causing serious injuries that resulted in the amputation of his leg below the knee. He brought suit against the manufacturer and distributor of the forklift, alleging various design defects. The jury returned a verdict for the plaintiff, from which the defendants appealed, claiming, among other things, that the jury should have been allowed to consider whether the independent negligence of the plaintiff, his employer, and the forklift operator combined so as to constitute the sole proximate cause of the accident. In response, the plaintiff argued that any such intervening forces must act in concert in order to constitute a superseding cause. The court agreed with the defendants and granted them a new trial. The issue of whether two or more independent forces may combine to create a superseding cause of injury in a products liability case was one of first impression in Connecticut.11 The Wagner court began its discussion of this issue by defining proximate and superseding cause, noting that proximate cause is "[a]n actual cause that is a substantial factor in the resulting harm . . . [which] results from a sequence of events unbroken by a superseding cause, so that its causal viability continue[s] until the moment of injury or at least until the advent of the immediate injurious force." Wagner, 243 Conn. at 178 (quoting Coburn v. Lenox Homes, Inc., 186 Conn. 370, 383, 441 A.2d 620 (1982)). In contrast, "[a] superseding cause is an act of a third person or other force which by its intervention prevents the actor from being liable for harm to another which his antecedent negligence is a substantial factor in bringing about." Wagner, 243 Conn. at 179 (quoting Restatement (Second) of Torts, Section 440). If a third personÆs negligence is determined to be a superseding cause of the injury, then that negligence, rather than the negligence of the party invoking the doctrine of superseding cause, is considered to be the sole proximate cause of the injury. Id. Accordingly, where a superseding cause exists, responsibility for the harm caused may be shifted from an admittedly negligent party to another party also deemed negligent. The court relied in its analysis upon Allen v. Chance Mfg. Co., 873 F.2d 465 (1st Cir. 1989), in which the First Circuit Court of Appeals reasoned that two or more forces not acting in concert may combine to create a superseding cause and that where a defendant can show that the plaintiffÆs injury was "legally caused in its entirety by other persons or entitiesùthat is, that the sole proximate cause of the injury was elsewhere," the defendant is relieved of liability. Id. The Wagner court held that "the inquiry should properly focus on the nature of the intervening forces, and whether they were sufficient to shift the entire causation element to some entity or entities other than the defendant." Id. With regard to multiple causes, so long as the intervening force "entirely breaks the causal connection between the defendantÆs conduct and the plaintiffÆs injuries," it is irrelevant whether that intervening force was comprised of one independent act or a combination of acts. Id. The court was careful to distinguish the concept of "concurring cause" from "superseding cause," stating that "[a] concurrent cause is contemporaneous and coexistent with the defendantÆs wrongful conduct and actively cooperates with the defendantÆs conduct to bring about the injury." Id. at 183. A superseding cause "so entirely supersedes the operation of the defendantÆs negligence that it alone, without his negligence contributing thereto in any degree, produces the injury; or it must be the non-concurring culpable act of a human being who is legally responsible for such act." Id. (quoting Corey v. Phillips, 126 Conn. 246, 255, 10 A.2d 370 (1939)). Thus, it is a question for the jury whether the intervening act concurred with the defendantÆs conduct, in which case the defendant remains liable, or whether the intervening acts combined with each other so as to create a superseding cause breaking the causal connection between the defendantsÆ conduct and the harm, thereby relieving the defendant of liability. Wagner, 243 Conn. at 183. The courtÆs conclusion that two or more forces may combine so as to create a superseding cause is consistent with and clarifies the holding in Potter that a plaintiff must prove, as part of his prima facie case, that the product reached him without substantial change in condition. Under Potter, once the plaintiff has shown that the product in question reached him without a substantial change in condition, a defendant who has specially pleaded alteration or modification to the product is free to put on evidence which tends to show that such alteration or modification was either the sole proximate cause or a superseding cause of the plaintiffÆs injuries. Significantly, Wagner indicates that the alteration or modification need not stand alone as the cause of the plaintiffÆs injuries, but instead can combine with some other force or act, including the plaintiffÆs own negligence, to relieve a defendant of liability. In addition, the Wagner holding is consistent with Connecticut General Statutes Section 52-572o, which provides that in a product liability case, the plaintiffÆs negligence does not bar his recovery, but merely results in a reduction of his damages proportionate to his share of liability. Id. at 184. If liability is proven, then the plaintiffÆs damages are reduced in proportion to his own negligence, pursuant to Section 52-572o. If a superseding cause exists, however, then there is no liability as to the defendant and Section 52-572o does not come into play. Thus, Section 52-572o is irrelevant unless and until a predicate finding of liability is made as to the defendant. B. OSHA Regulations Admissible in Defective Design Claim The Wagner court confronted another issue of first impression when it decided that evidence of compliance with OSHA regulations may be considered in design defect cases. Specifically, the court addressed whether evidence that the product at issue complied with applicable OSHA regulations relating to safety was properly admitted and whether the jury should have been instructed on that issue in conjunction with the courtÆs instructions on design defect. The defendants in Wagner presented evidence that the subject forklift met, and in fact exceeded, the standards set forth in the OSHA regulation governing safety requirements for industrial forklifts. The trial court declined to give an instruction, requested by the defendants, that the jury could consider OSHA regulatory compliance when determining whether the forklift was defectively designed and whether the defendants acted with due care in their design and distribution of the forklift. The Supreme Court concluded that this was an error, agreeing with the defendants that "where the regulation at issue addresses the safety of a product, it is relevant to the issue of defective design in a product liability action." Id. at 189. The court further noted that "[e]vidence that a product complies with an OSHA regulation that addresses the safety of the product may be probative of whether the product meets consumer expectations regarding its safety." Id. at 190. Dismissing the plaintiffsÆ concerns that evidence of OSHA compliance would improperly shift the juryÆs focus from the safety of the product to the employerÆs conduct, the court reasoned that such evidence goes to whether the product complied with safety regulations and does not relate to the employerÆs or manufacturerÆs conduct. Id. In fact, the court went so far as to state that where a product complies with and exceeds OSHA standards, the jury may permissibly draw the inference that the product is not defective and that the manufacturer acted with due care in its design and distribution of the product. Id. at 190-191.12 Thus, evidence of compliance with relevant OSHA regulations, like state-of-the-art evidence, is admissible in cases utilizing either the ordinary or modified consumer expectation test. This holding is in accord with Section 4(b) of the Restatement (Third), which provides that compliance with applicable product safety statutes or administrative regulations are properly considered by the jury when weighing the risks and benefits of a particular product, but do not preclude a finding of product defect.13 Accordingly, this portion of Wagner provides a powerful tool to both plaintiffs and defendants where the product at issue is covered by OSHA regulations pertaining to safety. C. Postaccident Modifications The Wagner defendants also claimed that evidence of postaccident modification of the forklift was improperly admitted, arguing that such modification, made by a nonparty, was irrelevant to the question of whether the forklift was defectively designed and whether a feasible alternative design existed. The court disagreed and ultimately concluded that evidence of subsequent remedial measures may be highly probative because it provides a safer alternative against which to compare the allegedly defective product. Wagner, 243 Conn. at 195. This holding is consistent with Potter, where the court determined that the feasibility of a safer design was an appropriate consideration when weighing risk against utility. Potter, 241 Conn. at 220-21. The Wagner court similarly concluded that, in the context of a design defect case, postaccident modifications are relevant to the issue of whether a safer, feasible alternative design existed because such modifications may demonstrate potential alternatives to the design used by the manufacturer and therefore may be highly probative of defectiveness. Wagner, 243 Conn. at 195, 199. Chief Justice Callahan, joined by Justice McDonald, dissented to this portion of the Wagner opinion, reasoning that because the subsequent remedial measures were undertaken by a nonparty, the limited probative value of the postaccident modification evidence was outweighed by its prejudicial effect. Id. at 203 (citations omitted). He further opined that the pertinent question was whether the product was reasonably safe at the time of manufacture. The introduction of evidence relating to subsequent changes "threatened to confuse the jury by diverting its attention from whether the product was defective at the relevant time to what was done later." Id. (quoting Grenada Steel Industries, Inc. v. Alabama Oxygen Co., 695 F.2d 883, 888 (5th Cir. 1983)). While the majority rejected this argument, the door has been left open for such evidence be excluded where a defendant can demonstrate that the modifications were made for reasons unrelated to safety. The Wagner defendants also argued that even if the changes made by Electric Boat were relevant to the existence of a safer alternative design, such evidence was improperly admitted because feasibility must be controverted before evidence regarding feasibility can be introduced. Wagner, 243 Conn. at 199. To support this argument, the defendants relied on Rule 407 of the Federal Rules of Evidence, which excludes evidence of subsequent remedial measures in negligence actions except under certain circumstances. Specifically, such evidence is permitted on the issue of feasibility where feasibility has been controverted. Noting that the federal circuits are not in agreement as to whether evidence of subsequent remedial measures should be excluded in strict liability actions, and noting that Connecticut does not apply the rule excluding evidence of subsequent remedial measures in strict liability cases, the court concluded that it would not require that "feasibility be controverted before evidence regarding feasibility may be introduced." Id. This portion of Wagner, dealing with postaccident modifications, brings the discussion full circle back to Potter and the courtÆs introduction of a feasible alternative design element in design defect cases. Wagner gives plaintiffs an opportunity to demonstrate, via postaccident modifications, that a feasible alternative design existed at the time of manufacture. Evidence of postaccident modifications can provide plaintiffs with a powerful evidentiary weapon and a strong counter to a defendantÆs efforts to demonstrate that the allegedly defective product in fact embodied the state-of-the-art. D. Wagner in Brief The key concepts arising from Wagner are that: (1) the court has held that two or more intervening forces, including the plaintiffÆs negligence, may combine to create a superseding cause of injury in a strict products liability case, thus relieving the defendant of liability; (2) evidence of compliance with OSHA (or, potentially, other safety related regulations) to demonstrate a lack of defect is admissible; and (3) evidence of postaccident modifications, even where made by a third party, is admissible to demonstrate the existence of a design defect and the existence of a feasible alternative design. III. Conclusion The application of products liability law in Connecticut has been significantly altered by the Connecticut Supreme CourtÆs decisions in Potter and Wagner and by the adoption of the Restatement (Third). Both plaintiffs and defendants now have strong incentives to offer as much evidence concerning the state-of-the art and pre-accident and postaccident modification of a product as the trial court will allow. However, the court has left many questions unanswered, and litigants will be hard-pressed to deal with the new rules governing this area of the law until such time as the court revisits them and offers some clarification. 1 This section has now been adopted unchanged as Section 2(b) of the Restatement (Third).2 This Note became ReportersÆ Note to comment (d) in the final version of the Restatement (Third).6 In most cases, the court should not make this determination until after the evidence in the case is concluded. Consequently, during the course of most trials, the court must admit evidence relevant under either test.3 This Note became ReportersÆ Note to comment (d) in the final version of the Restatement (Third).4 The court initially used the term "reasonable alternative design," as used in the Restatement (Third), then switched to use of the term "feasible alternative design."5 The drafters of the Restatement (Third) agree with the Potter court that design defects can be inferred from circumstantial evidence in appropriate cases. Restatement (Third) º 2, Rptrs. n. to comment d, º IIc.6 This is a concern which the Restatement (Third) addresses in comment (e) to Section 2(b), noting that "the designs of some products are so manifestly unreasonable, in that they have low social utility and high degree of danger, that liability should attach even absent proof of a reasonable alternative design."7 Id. at 254. A possible jury charge in a case where the ordinary consumer expectation test will be used, and where sufficient evidence is adduced to warrant an instruction on the state-of-the-art, is as follows: A product is unreasonably dangerous as designed if, at the time of sale, it is dangerous to an extent beyond that which would be contemplated by the ordinary consumer or user. In determining what an ordinary consumer would reasonably expect, you may consider the ordinary knowledge common to the community as to the productÆs characteristics, which include, among other things, the productÆs intended function and/or use as well as the safety features an ordinary consumer would reasonably expect to be incorporated into the product. The manufacturer is not required to guarantee that his product is incapable of doing harm. Therefore, the mere happening of an injury does not create legal responsibility on the part of the manufacturer. The law does not require the manufacturer to produce a product which is accident-free or foolproof. A manufacturerÆs duty is to produce a product which is reasonably safe. In this case, the defendant put on evidence supporting its claim that its product embodies the state-of-the-art. "State-of-the-art" is defined as "the level of relevant scientific, technological, and safety knowledge existing and reasonably feasible at the time the product in question was designed for manufacture." In designing a product, a manufacturer cannot be held to standards which exceed the limit of scientific advances and technology that existed at the time of manufacture and therefore cannot be charged with the duty to incorporate technology or features that were not available at the time it manufactured its product. State-of-the-art evidence, however, is not a defense to the plaintiffÆs claim. Instead, this evidence is simply one factor you may consider in determining whether the product is defective and unreasonably dangerous due to a defective design. In other words, state-of-the-art evidence relates to the condition of the product and the possibility that the product could have been made safer. A manufacturer, however, has no duty to incorporate a design change if to do so would significantly interfere with the function of the product itself. Accordingly, you may consider the evidence offered by the defendant that its product incorporates relevant scientific, technological, and safety knowledge existing and reasonably feasible at the time of design to determine whether, at the time of manufacture, the product was dangerous to an extent beyond that which would be contemplated by the ordinary consumer. See Potter, 241 Conn. at 221, n. 15, 223, n. 17; Giglio v. Connecticut Light & Power Co., 180 Conn. 230, 234, 429 A.2d 486 (1980); Couch v. Mine Safety Appliances Co., 107 Wash. 2d 232, 235-36, 728 P.2d 585 (1986).8 This apparent conflict in the Potter decision has also been recognized by the Reporters of the Restatement (Third) in the now published final version. The Reporters view the adoption of the modified consumer expectation test as "inevitably steer[ing] Connecticut toward the reasonable alternative design requirement set forth in this Restatement" and opine that ambiguities in the nonfinal versions of the Restatement (Third) may have misled the court about the proof requirements a plaintiff faces under the new Restatement. Restatement (Third) Section 2, ReportersÆ Note, comment d, section II.C. The Reporters conclude that Potter is "perfectly consistent" with the Restatement (Third). Id. 9 According to the court, Connecticut is the only state to utilize "but for" language. Other jurisdictions utilize a standard of proximate or substantial cause. Because Connecticut is so out of step with other jurisdictions on this issue, Section 52-572p might appropriately be considered for legislative review.10 Note that two or more intervening forces may combine to create a superseding cause of a plaintiffÆs injuries, thereby relieving a defendant of liability. Wagner v. Clark Equipment Company, Inc., 243 Conn. 168, 700 A.2d 38 (1997). See discussion below. Thus, an alteration or modification could combine with some other intervening force to create a superseding cause of injury.11 Prior Appellate Court opinions have held that two or more forces may constitute a superseding cause, but in each case, the intervening actors were acting in concert. Wagner, 243 Conn. at 180- 81 (citing Amendola v. Geremia, 21 Conn. App. 35, 38, 571 A.2d 131, cert. denied, 215 Conn. 803, 574 A.2d 217 (1990); Burns v. Gleason Plant Security, Inc., 10 Conn. App. 480, 482-83, 523 A.2d 940 (1987)).12 The Wagner defendants presented evidence that the forklift complied with certain industry standards which were incorporated by reference into the pertinent OSHA regulation. The court noted that "compliance with a federal regulation may carry more weight with a jury than compliance with an industry standard, because a federal regulation has the imprimatur of the federal government." Id. at 191.13 RESTATEMENT (THIRD) Section 4(b). Section 4(a) goes further and provides that noncompliance with applicable product safety statutes or administrative regulations "renders the product defective with respect to the risks sought to be reduced by the statute or regulation." RESTATEMENT (THIRD) Section 4(a).