Dooley, J.
Defendant Navistar International Transportation Corporation (Navistar) appeals from a jury verdict that held defendant liable to plaintiffs Bruce and Martha Webb on a theory of strict products liability for a design defect and/or failure to warn plaintiffs of dangers inherent in the design of a tractor. A majority of the Court agrees to reverse and remand this case.
[122]*122Justice Morse, Justice Peck and I hold that principles of comparative causation apply in this products liability action. We do not agree, however, to a general rule on when comparative principles apply in strict products liability actions, nor on how to implement these principles when they do apply. I believe this case must be reversed and remanded for a new trial because the trial judge failed to instruct the jury on comparative causation. Justice Morse concludes that Bruce Webb is more than fifty percent responsible, as a matter of law, and therefore, under 12 V.S.A. § 1036 (comparative negligence), he would enter judgment for defendant. Justice Peck agrees with Justice Morse that judgment should be entered for defendant but on the ground that the tractor is not defective as a matter of law. Justice Johnson and Justice Gibson would affirm the judgment for plaintiffs; they would hold that comparative principles are not applicable in products liability actions.
The positions of the Justices produce no majority on the mandate. A majority of the Court agrees, however, that the judgment entered on the jury verdict cannot be affirmed. Thus, although Justices Morse and Peck would prefer a mandate of “reversed,” they have joined in a mandate of “reversed and remanded” because it accurately represents the center of gravity of the Court. See Cell v. Drake, 100 P.2d 949, 951 (Idaho 1940). Thus, the mandate requires a new trial based on principles of comparative causation.
The dissent criticizes the majority for creating a rat’s maze from which there is no exit. We strongly disagree with the characterization of the cause or consequence of the Court’s voting. If the dissent would accept that comparative causation has now become the rule in products liability cases in Vermont, and participate in an implementation design to guide future cases, the trial judge in this case might know exactly what to do on remand. If a majority could not agree on an implementation design, the trial judge and parties would at least know the full range of options and votes in support of each on this Court. I share Justice Peck’s view that we should do all in our power . to avoid stalemate, if possible.
I.
On November 13, 1985, at approximately 9:30 p.m., Bruce Webb learned that some of his cows might be out of the pasture. He and his father got out their tractor, a 1978 Model 464 farm tractor manufactured by Navistar, and they proceeded down Route 207 with Bruce [123]*123Webb standing on the draw bar and his father driving. En route, the tractor was struck in the rear by a car driven by an allegedly intoxicated operator. As a result of the accident, Bruce Webb suffered serious injuries to his legs.
Plaintiffs filed suit against Navistar, the driver of the car, and others. The complaint alleged negligence, breach of warranty, and strict products liability. Claims against all defendants other than Navistar were ultimately dismissed, and the case proceeded to trial against Navistar solely on the products liability claim. Plaintiffs argued that the tractor was defectively designed because (1) it allowed operation of a white field light at highway speeds without provision for separate red tail lights, and (2) it failed to provide a safe passenger location so that Bruce. Webb could have ridden on the tractor without exposure to injury They contended further that defendant failed to provide adequate warnings of these dangers.
The ease was tried, and the trial court directed a verdict in defendant’s favor on both claims. On appeal, we affirmed the directed verdict regarding defendant’s failure to provide a safe passenger location, but reversed as to whether the design of the field light was defective and whether the manufacturer’s warning on its use was inadequate. See Webb v. Navistar Int’l Transp., No. 91-384 (Vt. July 1, 1992) (Webb T) (unpub. mem.).
The second trial focused on the lighting system of the tractor. The Model 464 tractor has a red taillight, two amber lights with road flashers, two red rear reflectors, a reflective slow-moving-vehicle triangle and a white field light mounted on the left rear bumper. A cautionary decal on the left front fender directs operators to use the flashing amber lights at all times when on public roads. The light system is designed so that when the flashing amber lights are in use, the red taillight activates and the white field light does not work. At the time of the accident, the flashing amber lights1 and the taillight did not work, and the reflectors were missing. In addition, by riding on the draw bar, Webb blocked the view of the reflective triangle. The cautionary decal also warned against riding the tractor unless a seat or platform is provided and instructed the operator to “[k]eep others off.”
The owner’s manual for the tractor also provides warnings and instructions. On pages 3 and 4, the manual sets forth rules for safe [124]*124operation of the tractor. Here, the manual warns: “No riders allowed.” It also contains an instruction not to use the white field light on the highway on page 55, under the heading CAUTION!
Webb testified that while travelling on the highway he employed both the headlights and the rear field light on the rationale that more light was better than less light. He indicated that it had not occurred to him that operating the tractor on the highway at night with the rear field light on was a hazard. The operator of the automobile that collided with the tractor testified that he believed the white field light mounted on the left rear bumper was the headlight of an approaching “one-eyed” car.
Plaintiffs tried the case on two theories: (1) that the lighting system was defective because it allowed the tractor to be operated on highways with the field light illuminated, and (2) that defendant failed to adequately warn consumers of the known risk of using the field light while operating the tractor on the highway. The jury returned a verdict in favor of plaintiffs on liability, and the parties stipulated to damages. Defendant appeals, arguing that the evidence was insufficient to support the verdict and that the court erred by failing to instruct the jury that it may apportion liability between the parties. We have the benefit of briefs of amicus parties on both sides of the comparative liability issue.2
II.
Defendant argues that the evidence was insufficient for the jury to find that its tractor was defective, that its warnings were inadequate, and that either the defective tractor or the inadequate warnings proximately caused Webb’s injuries. These arguments all [125]*125war with our decision in Webb I. On this point, I understand that four of us agree. Except for those eliminated by our decision in Webb I, the parties’ claims and defenses remained essentially the same at the second trial. If Webb I remains the law, the evidence was sufficient for the jury to reach a judgment against defendant. See McGee Constr. Co. v. Neshobe Dev., Inc., 156 Vt. 550, 556, 594 A.2d 415, 418 (1991) (verdict will be sustained if, considering evidence in light most favorable to verdict and excluding effect of modifying evidence, there is evidence fairly and reasonably tending to support it).
The jury could reasonably conclude that the danger of operating the tractor on a highway at night with the field light illuminated was not a danger obvious to the ordinary consumer, and plaintiffs presented evidence of a safety device that could have been installed by defendant to prevent such use. Despite the evidence that the automobile operator was intoxicated and changed his story between the time of the accident and the trial, the jury could reasonably conclude that the lighting system on the tractor was defective and was the cause of the accident. Moreover, the question of whether a manufacturer provided adequate warnings about foreseeable dangers is a question of fact properly left to the jury. See McCulloch v. H.B. Fuller Co., 981 F.2d 656, 658 (2d Cir. 1992) (applying Vermont law).
III.
I do not believe, however, that the judgment in this case can be affirmed. I agree with defendant that comparative liability principles are applicable in strict products liability actions and should have been charged to the jury in this case. Because the split in the Court reserves the details of implementing comparative principles for another day, I state only the reasons we adopt a comparative causation rule.
The doctrine of strict products liability was first developed by the California Supreme Court in Greenman v. Yuba Power Products, Inc., 377 F.2d 897, 901 (Cal. 1962), and then set forth in the Restatement (Second) of Torts § 402A (1965). This doctrine was created in response to the limitations of traditional negligence and warranty actions for injuries caused by defective consumer goods. Butaud v. Suburban Marine & Sporting Goods, Inc., 555 F.2d 42, 44 (Alaska 1976). In negligence actions, plaintiffs were unable to isolate the negligence of the manufacturer as manufacturing processes became more complex. Id. In actions based on warranty theories, [126]*126plaintiffs confronted defenses of disclaimer, notice of breach and lack of privity, also problematic in complex distribution systems. Id.-, see also O’Brien v. Comstock Foods, Inc., 125 Vt. 158, 162, 212 A.2d 69, 72 (1965) (abolishing privity requirement to prove liability of food producer). Strict products liability removed the difficulties plaintiffs faced in proving warranty or negligence claims against mass producers and distributors by imposing liability upon them without regard to fault or privity of contract. See W Keeton, et al., Prosser and Keeton on the Law of Torts § 98, at 692 (5th ed. 1984). The purpose of this judicially created doctrine is to lessen the burden of proof for plaintiffs injured by defective products. Daly v. General Motors Corp., 575 F.2d 1162, 1168 (Cal. 1978); see also Zaleskie v. Joyce, 133 Vt. 150, 154-55, 333 A.2d 110, 113 (1975) (one reason to adopt strict products liability is to respond to problems of availability of proof).
Justifications for reducing plaintiffs’ burden rest upon two public policies. First, strict liability protects the consumer, see Restatement (Second) of Torts § 402A cmt. c (consumer entitled to maximum protection), by creating an incentive for manufacturers to produce safe products, see, e.g., Whitehead v. Toyota Motor Corp., 897 S.W.2d 684, 693 (Tenn. 1995) (strict liability encourages greater care in manufacture of products), or as other courts have stated, a deterrence to producing unreasonably dangerous products. See, e.g., Kimco Dev. Corp. v. Michael D’s Carpet Outlets, 637 A.2d 603, 607 (Fa. 1993) (strict products liability imposes deterrent to production of dangerous products). Second, strict products liability is justified on the ground that manufacturers are in the best position to spread the cost of injury resulting from defective products by passing it on to consumers as a cost of doing business. Zaleskie, 133 Vt. at 154-55, 333 A.2d at 113; Greenman, 377 F.2d at 901; see also Restatement (Second) of Torts § 402A cmt. c (public policy demands that burden of accidental injuries caused by products be placed on those who market them who may treat as cost of production against which liability insurance may be obtained).
In 1975, we adopted the doctrine of strict products liability set forth in the Restatement (Second) of Torts § 402A. See Zaleskie, 133 Vt. at 155, 333 A.2d at 114 (judicially adopting doctrine of strict products liability as set forth in the Restatement (Second)). Under that doctrine, a manufacturer is strictly liable for physical harm or property damages resulting from a defective product that reaches a user without undergoing substantial change. Kinney v. Goodyear [127]*127Tire & Rubber Co., 184 Vt. 571, 574, 367 A.2d 677, 679 (1976). A defective product “must be 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) of Torts § 402A cmt. i. The plaintiff bears the burden of proving that the product is defective, and that its defect was the proximate cause of the harm, see Gilman v. Towmotor Corp., 160 Vt. 116, 119, 621 A.2d 1260, 1261 (1992) (proximate cause is one element of strict liability), but is relieved of showing that the defendant was negligent. Kinney, 134 Vt. at 574, 367 A.2d at 679.
A manufacturer also has a duty to warn users and consumers when it knows or has reason to know of dangers inherent in the product at the time the product is sold, Restatement (Second) of Torts § 402A cmt. k, or when the product is dangerous to an extent beyond that which would be contemplated by an ordinary consumer. Menard v. Newhall, 135 Vt. 53, 55, 373 A.2d 505, 507 (1977). To establish strict liability for an inadequate warning, a plaintiff must prove that the inadequate warning made the product unreasonably dangerous and was the proximate cause of the injury. Id. at 54, 373 A.2d at 506.
Under the Restatement (Second) formulation of products liability, defenses are limited. Assumption of risk is a complete bar to recovery. “If the user or consumer discovers the defect and is aware of the danger, and nevertheless proceeds unreasonably to make use of the product and is injured by it, [the user or consumer] is barred from recovery.” Restatement (Second) of Torts § 402A cmt. n. Product misuse has traditionally been a bar to recovery as well. Id. § 402A cmt. h; see, e.g., Kennedy v. City of Sawyer, 618 P.2d 788, 796 (Kan. 1980) (assumption of risk and product misuse traditionally barred all recovery for strict products liability claims); Smith v. Smith, 278 N.W.2d 155, 161 (S.D. 1979) (recovery barred where consumer assumes risk or misuses product). On the other hand, the Restatement (Second) provides that negligence that “consists merely in a failure to discover the defect in the product, or to guard against the possibility of its existence” is not a defense at all. Restatement (Second) of Torts § 402A cmt. n.
The Restatement (Second) does not address the issue of shared responsibility and does not address the effect of the user’s negligence beyond the limited circumstances described in comment n to § 402A. If we view these omissions as intentional, and we choose to follow [128]*128§ 402A, we are left with the harsh “all-or-nothing” approach of negligence actions prior to the adoption of our comparative negligence statute. Compare Langevin v. Gilman, 121 Vt. 440, 446, 159 A.2d 340, 344 (1960) (there can be no recovery for plaintiff if plaintiff negligence contributed “in the least degree” to accident) with 12 V.S.A. § 1036 (contributory negligence shall not bar recovery by plaintiff in negligence action, but plaintiff negligence may proportionately reduce damages).
The overwhelming majority of states have rejected the “all or nothing” rule, either by rejecting the limits of § 402A or by supplementing its provisions, and have applied principles of comparative liability in strict products liability actions. Whitehead, 897 S.W.2d at 691; see, e.g., Kaneko v. Hilo Coast Processing, 654 P.2d 343, 354 (Haw. 1982) (“[W]e eliminate the harshness of the ‘all or nothing’ bar to recovery that results if a plaintiff is found to have misused the product or to have assumed the risk of using the product.”); Suter v. San Angelo Foundry & Machine Co., 406 A.2d 140, 147 (N.J. 1979) (applying comparative negligence statute to strict liability action to relieve inequities incurred by plaintiffs and defendants as result of all-or-nothing approach to recovery); see also Butaud, 555 P.2d at 45-46 (adopting comparative principles in strict products liability actions); Daly, 575 P.2d at 1172 (same); Kennedy, 618 P.2d at 796-97 (same); Austin v. Raybestos-Manhattan, Inc., 471 A.2d 280, 288 (Me. 1984) (same); Zahrte v. Sturm, Ruger & Co., 661 P.2d 17, 18-19 (Mont. 1983) (same); Thibault v. Sears, Roebuck & Co., 395 A.2d 843, 848 (N.H. 1978) (same); Day v. General Motors Corp., 345 N.W.2d 349, 357 (N.D. 1984) (same); Sandford v. Chevrolet Div. of Gen. Motors, 642 P.2d 624, 628 (Or. 1982) (same); Fiske v. MacGregor, Div. of Brunswick, 464 A.2d 719, 727 (R.I. 1983) (same); Duncan v. Cessna Aircraft Co., 665 S.W.2d 414, 425 (Tex. 1984) (same); Star Furniture Co. v. Pulaski Furniture Co., 297 S.E.2d 854, 863 (W Va. 1982) (same); Dippel v. Sciano, 155 N.W.2d 55, 64-65 (Wis. 1967) (same).3 The United States District Court for the District of Vermont has endorsed this approach and predicted we will do so also. See Smith v. Goodyear Tire & Rubber Co., 600 F. Supp. 1561, 1568 (D. Vt. 1985).
[129]*129In addition, the tentative draft of the Restatement (Third) of Torts provides for apportioning liability between the plaintiff and the manufacturer or seller. See Restatement (Third) of Torts: Products Liability § 7 (Tentative Draft No. 1, 1994). Similarly, the Uniform Comparative Fault Act § 1 provides that a claimant’s contributory fault proportionately reduces compensatory damages in strict products liability actions. 12 U.L.A. 127 (1996). Many commentators maintain that adopting comparative liability principles in strict products liability actions is the fairest approach. See, e.g., Keeton, et al., supra, § 102, at 712 (comparative fault system fairest way to allocate costs of accidents); D. Noel, Defective Products: Abnormal Use, Contributory Negligence, and Assumption of Risk, 25 Yand. L. Rev. 93, 117-18 (1972) (contributory negligence should diminish plaintiff’s damages); V. Schwartz, Strict Liability and Comparative Negligence, 42 Tenn. L. Rev. 171,179-81 (1974) (comparative principles should apply in strict products liability); J. Wade, On the Nature of Strict Tort Liability for Products, 44 Miss. L. J. 825, 850 (1973) (same).
The primary reason that courts adopt comparative liability principles in strict products liability actions is “because it is fair to do so.” Daly, 575 P.2d at 1172. Adopting comparative liability principles “will accomplish a fairer and more equitable result” because the plaintiff’s award is reduced by an amount equal to the degree to which the plaintiff is responsible for the accident. Kaneko, 654 P.2d at 352. Most courts reject the framework that places the burden of loss on one party where two parties contributed to causing the injury. Daly, 575 P.2d at 1172. Comparative liability principles also further fairness by preventing a negligent plaintiff from recovering as much as a plaintiff who has taken all reasonable precautions. Smith v. Goodyear Tire & Rubber Co., 600 F. Supp. at 1568.
Moreover, there is no reason to impose the cost of a plaintiff’s negligence upon the manufacturer to spread among other consumers of the product. Daly, 575 P.2d at 1168-69; see also Restatement (Third) of Torts: Eroducts Liability § 7 cmt. a (Tentative Draft No. 1, 1994) (unfair to impose costs of substandard plaintiff conduct on manufacturers, who will be impelled to pass on costs to all consumers, including those who use and consume product safely). The instant case is illustrative. Here, plaintiff stood on the draw bar of the tractor while it traveled down a public road. Although he understood the importance of the warning against such action, he chose to disregard the warning. As a result, he blocked the view of the reflective triangle [130]*130and the single amber flashing light that may have been operable. Moreover, he failed to maintain the reflectors and the other flashing light. If the jury may reduce plaintiff’s recovery to the extent that his injuries were caused by his negligence, defendant is not held liable for the cost of injuries attributed to plaintiff’s negligence and does not pass this cost on to those farmers who heed the warnings posted on their tractors. Strict products liability was intended to spread the cost of injuries resulting from defective products; it was never intended to spread the cost of injuries resulting from user negligence. Daly, 575 P.2d at 1166.
Apportioning liability more effectively spreads recoveries from manufacturers for selling defective products than the “all or nothing” framework. Under the “all or nothing” framework, some plaintiffs receive windfalls because they collect damages for injuries caused by their own negligence in addition to damages for injuries caused by the product defect. On the other hand, some plaintiffs receive nothing because the court or jury has determined that their negligence constitutes misuse, assumption of risk or an intervening cause, concepts often difficult to distinguish. See Sunday v. Stratton Corp., 136 Vt. 293, 300, 390 A.2d 398, 402 (1978). Applying principles of comparative liability will reduce the total damages awarded to some plaintiffs but will also extend recoveries to some plaintiffs formerly barred from any recovery; thus, recoveries will be more equitably distributed among plaintiffs.
A minority of courts have rejected comparative liability principles in the context of strict products liability actions and continue to impose the “all or nothing” framework set forth in the Restatement (Second) of Torts § 402A. See, e.g., Bowling v. Heil Co., 511 N.E.2d 373, 380 (Ohio 1987) (finding no rationale to persuade it that comparative fault principles should apply to products liability actions); Kimco Dev. Corp., 637 A.2d at 606 (declining to extend negligence concepts to strict products liability area); Smith v. Smith, 278 N.W.2d at 160-61 (negligence of either party is irrelevant in strict liability).
We draw two reasons from those decisions for retaining the “all or nothing” rule. First, several courts have suggested that it is too confusing to inject negligence concepts into strict liability actions, see, e.g., Kimco Dev. Corp., 637 A.2d at 606 (conceptual confusion would ensue should negligence and strict liability concepts be commingled), and that juries will be unable to compare a defective product with a plaintiff’s negligent conduct to apportion liability. See Smith v. Smith, 278 N.W.2d at 161 n.7 (adoption of comparative fault would present unworkable problems for juries).
[131]*131Most courts have rejected this concern as semantic and theoretical. “We are convinced that in merging the two principles what may be lost in symmetry is more than gained in fundamental fairness,” Daly, 575 P.2d at 1172, and “fairness and equity are more important than conceptual and semantic consistency.” Kaneko, 654 P.2d at 352. Further, apportioning liability will be less difficult for juries than the current framework, which requires juries to distinguish between defenses that courts and scholars are often unable to differentiate. As the Supreme Court of Texas noted, assumed risk and unforeseeable misuse are nothing more than extreme variants of contributory negligence. See Duncan, 665 S.W.2d at 423. And the line between contributory negligence — resulting in total recovery — and assumed risk or misuse — resulting in no recovery — is difficult to draw. See Thibault, 395 A.2d at 848 (distinction between assumption of risk and contributory negligence merely semantics); Noel, supra, at 128 (distinctions between misuse, contributory negligence and assumption of risk are not clear); cf. Perkins v. Windsor Hosp. Corp., 142 Vt. 305, 310, 455 A.2d 810, 813-14 (1982) (error to introduce assumption of risk language into charge on comparative negligence). There is no need to draw shadowy lines between misuse, assumption of risk and contributory negligence, however, if all defenses may constitute a basis for apportioning liability. Schwartz, supra, at 175.
Second, the “all or nothing” courts maintain that comparative principles would undermine the purposes of imposing strict liability on manufacturers because this approach reduces the incentive to produce safe products and fails to allocate the risk for loss from injury to manufacturers who are in a better position to absorb it. See Kimco Den Corp., 637 A.2d at 606-07. On the contrary, applying principles of comparative liability in strict products liability actions is completely consistent with the purposes of imposing strict liability on manufacturers. Indeed, it will have no effect on the principal purpose of adopting this doctrine; the plaintiff is still relieved from proving negligence of the manufacturer or privity of contract with it. See Daly, 575 P.2d at 1168 (plaintiffs continue to be relieved of proving negligence). Manufacturers remain strictly liable for injuries resulting from their defective products. Kaneko, 654 P.2d at 353.
Nor is it clear that adopting comparative principles will significantly reduce the incentive to produce safe products. Cf. Swett v. Haig’s, Inc., 164 Vt. 1, 7 n.3, 663 A.2d 930, 933 n.3 (1995) (in dram shop action, dram shop’s incentive to avoid serving intoxicated persons is not reduced by availability of contribution from intoxicated motorist). [132]*132Recoveries may be reduced in some cases, but more plaintiffs will recover if assumption of the risk and product misuse are no longer total bars to recovery. Overall, the cost of a defect may be the same under either approach.
Courts rejecting comparative liability assume that the primary purpose in strict products liability actions is to spread the cost of injury. Because manufacturers are in a better position than plaintiffs to spread this cost, they reason that it is inconsistent with strict products liability to reduce recoveries in proportion to plaintiff negligence. See, e.g., Bowling, 511 N.E.2d at 380; Kimco Dev. Corp., 637 A.2d at 606. If spreading the cost of all injuries were the goal, then apportioning liability between the parties would be adverse to the goal.4 We note, however, that the purpose has been to spread the cost of injuries resulting from defective products. The issue here is whether to spread the cost of injuries resulting from user negligence in addition to that resulting from a defect. No rationale to support such risk allocation has been presented. Strict liability is not absolute liability; manufacturers are not insurers of user safety. Daly, 575 P.2d at 1166; Kaneko, 654 P.2d at 353; Mulherin v. Ingersoll-Rand Co., 628 P.2d 1301, 1302 (Utah 1981).
On balance the reasons to adopt comparative principles greatly outweigh the reasons to reject this approach. The comparative approach is fairer to all parties, and properly implemented, will not reduce the incentive to produce safe products.
IV.
Unlike Justice Morse, I reach this conclusion as part of the development of the common law of products liability in this state and not because of the Vermont comparative negligence statute, 12 V.S.A. § 1036. The statute applies only to “an action... to recover damages for negligence.” (Emphasis added.) We must presume that the Legislature intended the plain meaning of the statutory language. See Bisson v. Ward, 160 Vt. 343, 348, 628 A.2d 1256, 1260 (1993); Armstrong v. Cione, 738 P.2d 79, 82 (Haw. 1987) (plain meaning of words shows Legislature did not intend to include actions based on strict liability within coverage of comparative negligence statute). [133]*133The wording covers actions based on negligence, but not on strict liability. The majority of courts confronting this question have reached the same conclusion. See, e.g., Armstrong, 738 P.2d at 82; Thibault, 395 A.2d at 848; Day, 345 N.W.2d at 354; Duncan, 665 S.W.2d at 426; Mulherin, 628 P.2d at 1303.
Nor can I agree that the statute reaches some kinds of products liability actions and not others. Nothing in our case law suggests that some products liability cases are based on negligence principles and others are not. Section 402A(2)(a) of the Restatement (Second) provides that a seller of a defective product is liable although “the seller has exercised all possible care in the preparation and sale of his product.” Relying on § 402A, we have squarely held that “[ljack of negligence. . . does not bar the liability.” Kinney, 134 Vt. at 574, 367 A.2d at 679.
Even though the comparative negligence statute does not apply, we could construct a comparable causation rule that would mirror its terms. In this case, the main significance of such a rule is that plaintiffs could not recover if the causal effect of the negligence of Bruce Webb was greater than the causal effect of the liability of defendant. Using this test, Justice Morse would hold that, as a matter of law, a majority of plaintiffs’ damages were caused by the negligence of Bruce Webb so that plaintiffs cannot recover at all.
I do not subscribe to the “half-or-nothing” framework of § 1036 for products liability cases. The rule is inconsistent with the policy of ensuring that manufacturers bear the cost of casting defective products into the market. The manufacturer must remain responsible for damages resulting from the defect, regardless of the extent to which other factors contributed to the injuries. See Day, 345 N.W.2d at 357 (“half or nothing” policy of comparative negligence statute is incompatible with basic concepts of strict liability); see also Butaud, 555 P.2d at 45-46 (pure comparative liability is fair method to apportion liability in strict products liability cases); Armstrong, 738 P.2d at 82-83 (pure comparative principles necessary to prevent all-or-nothing decisions and to create economic incentives for safer products); Duncan, 665 S.W.2d at 429 (policies underlying strict liability best served by pure comparison); Schwartz, supra, at 179 (pure comparative principles should apply to further policy of risk distribution). I would allow plaintiffs to recover irrespective of the extent to which Bruce Webb’s negligence contributed to his injuries; of course, the amount of recovery would be reduced to reflect the contribution of this negligence.
[134]*134The dissent characterizes the adoption of comparative causation as a major step toward abolishing the doctrine of strict products liability. I find this conclusion to be greatly exaggerated. I doubt that a balanced and properly designed rule on comparative causation will significantly reduce the incentive for manufacturers to produce safe products; indeed, it may increase the incentive. The posture of this case prevents us from fashioning a fair and balanced rule that eliminates or reduces the adverse side-effects claimed by the dissent. The first question before us is whether comparative causation can ever apply in a products liability case, and we have divided on that question. I emphasize the narrowness of the question because much of the criticism of comparative causation rules applies to rules I would not adopt.
If comparative principles ever apply in a strict liability case, they should apply here. The jury could find that a number of Bruce Webb’s actions or omissions reflected lack of due care for his safety. Some of these actions or omissions do not involve the condition of the tractor and are not related to plaintiffs’ liability theory For example, irrespective of what lighting was available or in use, the jury could find that Bruce Webb was negligent in riding on the draw bar and covering up a reflector and an amber light while the tractor was being operated on a highway On remand, I would allow at least that determination.
Reversed and remanded.