(dissenting)—I would affirm the trial court's judgment n.o.v. because, inter alia, I believe that, as a matter of law, a manufacturer has no duty to warn of a risk as unlikely as that involved here.
Beyond cavil, the risk in this case was exceedingly remote: from 1932 to the time of the accident in 1985, Johnson & Johnson had sold over 500 million bottles of *298baby oil without a single report of aspiration. To hold that Johnson & Johnson should have warned of a known, but exceedingly remote, danger not only is unfair and unwise, it simply ignores the language of RCW 7.72.030(l)(b), which reads as follows:
A product is not reasonably safe because adequate warnings or instructions were not provided with the product, if, at the time of manufacture, the likelihood that the product would cause the claimant's harm or similar harms, and the seriousness of those harms, rendered the warnings or instructions of the manufacturer inadequate and the manufacturer could have provided the warnings or instructions which the claimant alleges would have been adequate.
Statute's Balancing Test
The lack of a warning must be analyzed both in terms of the balancing required by RCW 7.72.030(l)(b), and the consumer expectations test of RCW 7.72.030(3). Falk v. Keene Corp., 113 Wn.2d 645, 655, 782 P.2d 974 (1989). Although it is clear, as the majority notes, that the seriousness of the potential injury is extremely great, the test nevertheless requires balancing the likelihood and seriousness of the harm against the burden of preventing the harm through effective warnings. RCW 7.72.030(l)(b); see also Falk, 113 Wn.2d at 654.
While the majority pays lip service to the balancing test, it fails completely to consider likelihood when it states at page 295:
Although it may be unlikely that the product will cause harm of the gravity experienced here, nevertheless the seriousness of the risk is extremely great considering what mineral oil can do when aspirated.
(Italics mine.)
Having accepted the "unlikelihood" of aspiration, the majority then purports to balance the seriousness of the harm against the burden of providing an effective warning. The majority states that the cost of printing and affixing a label would be small, but would it? The warning would have to be lengthy and complicated, warning the potential user not only that there is a risk of harm, but that children *299who have aspirated baby oil may show no outward signs and that immediate medical attention is necessary to determine if aspiration occurred. Must aspiration also be explained?
The unexamined premise that warnings are not costly in risk-utility balancing is, in our considered opinion, highly questionable. Warnings, in order to be effective, must be selective. They must call the consumer's attention to a danger that has a real probability of occurring and whose impact will be significant. One must warn with discrimination since the consumer is being asked to discriminate and to react accordingly. . . . Those who argue for warning as the judicial solution to latent defect cases labor under a naive belief that one can warn against all significant risks. The truth is that such a marketing scheme is not feasible. The warning process, in order to have impact, will have to select carefully the items which are to become part of the consumer's mental apparatus while using the product. Making the consumer account mentally for trivia or guard against risks that are not likely to occur imposes a very real societal cost.
(Footnote omitted.) Twerski, Weinstein, Donaher, Piehler, The Use and Abuse of Warnings in Products Liability— Design Defect Litigation Comes of Age, 61 Cornell L. Rev. 514-15 (1976); see also Cotton v. Buckeye Gas Prods. Co., 840 F.2d 935, 938 (D.C. Cir. 1988).
We must, then, consider the possibility that individuals will overreact to a warning; that they will assume that the risk is much higher than truly it is; and that they will stop using a beneficial product because of a very remote risk. More importantly, we also must consider the possibility that consumers will pay less attention to all risks and all warnings because of the overload of information. In either event, the consuming public loses.
The Consumer Expectations Test
The majority's analysis of consumer expectations also ignores important considerations. The court should consider the cost of the product, the seriousness of the potential harm, and the cost and feasibility of minimizing the risk. "The nature of the product, or the nature of the claimed defect may make other factors relevant to the *300issue." (Italics mine.) Seattle-First Nat'l Bank v. Tabert, 86 Wn.2d 145, 154, 542 R.2d 774 (1975).
The majority, without explanation, chooses to discuss only the nature of the product and the nature of the defect and then concludes that the jury could have found that the product was not reasonably safe as judged by consumer expectations. The analysis is as follows:
Here, the product was composed of an oil that has the potential for great harm if it gets into the lungs, but is nevertheless promoted for use on and around babies. The warning was not merely deficient, it was nonexistent.
(Italics mine.) Majority, at 296. But this begs the question by assuming that a nonexistent warning is deficient. The question is whether there was a need for any warning. To answer the question, we ask if the product is "unsafe to an extent beyond that which would be reasonably contemplated by the ordinary consumer." Tabert, 86 Wn.2d at 154; RCW 7.72.030(3); Falk, 113 Wn.2d at 655.
The ordinary consumer evaluates a product in terms of safety, recognizing that virtually no product is or can be made absolutely safe. . . .
... It must be borne in mind that we are dealing with a relative, not an absolute concept.
Tabert, 86 Wn.2d at 154. Could a reasonable consumer believe that defendant's baby oil is not reasonably safe after weighing factors such as the gravity of the potential harm, the cost and feasibility of eliminating or minimizing the risk and the nature of the product? I would answer, "no".
If one factors in the extreme unlikelihood of aspiration and the burdens of providing an effective warning, as discussed above, the scales tip even farther against the reasonableness of requiring a warning here. Therefore, I would hold as a matter of law that the baby oil without the warning was not unsafe beyond the reasonable expectations of the ordinary consumer.
Any other result is unfair because it punishes Johnson & Johnson for having a deep pocket, not for producing a product that presents an unreasonable risk of harm. Any *301other result also is unwise, because it requires a manufacturer to warn against every conceivable danger, thus diluting the importance of any warning given.
Need for Specific Warning
The majority also holds that the plaintiff need not offer the exact wording of a warning. The rationale is that because liability for failure to warn and for defective design both are created by RCW 7.72.030(1) and stand on the same footing, and because plaintiffs in defective design cases are not required to show the existence of alternative safe designs, Couch v. Mine Safety Appliances Co., 107 Wn.2d 232, 239, 728 P.2d 585, 78 A.L.R.4th 139 (1986), it follows that the plaintiffs in a failure to warn case are not required to prove the specific warning that should have been given. I disagree.
However, RCW 7.72.030(1) (b) expressly provides that a product is not reasonably safe if, inter alia, "the manufacturer could have provided the warnings or instructions which the claimant alleges would have been adequate." The majority completely ignores this language. A jury cannot possibly decide if the manufacturer could have provided the warnings unless it is told what those warnings ought to have been. If the majority's approach is adopted, the manufacturer must prove a negative: that there is no possible alternative warning.
Without knowing the content of the alternative warning the trier of fact cannot balance the cost of providing this warning against the likelihood and severity of the danger. Understandably the Legislature has placed upon the plaintiff the burden of proposing alternative warnings; this court should not shift that burden in the guise of judicial construction of the statute.8
*302In conclusion, the problem may be approached as either a question of a duty to warn or a question of legal causation.9 The concepts, though not interchangeable, are closely related and bear consideration in tandem. Hartley v. State, 103 Wn.2d 768, 779, 698 P.2d 77 (1985); see also 1 American Law of Products Liability 3d § 4:4 (1987).
If the question is framed in terms of duty, it may be put as follows: Was the likelihood of aspiration together with the gravity of the harm sufficient to impose a duty on the defendant to guard against that danger? See Berg v. General Motors Corp., 87 Wn.2d 584, 592, 555 P.2d 818 (1976); see also Rose v. Nevitt, 56 Wn.2d 882, 885, 355 P.2d 776 (1960). I would hold that the risk of harm in this case was insufficient to impose a duty upon Johnson & Johnson.
If the question is framed in terms of legal causation, it may be phrased as follows: Assuming that the plaintiffs proved the necessary elements of strict liability, should this court, as a matter of policy, determine that liability should attach? See Hartley, 103 Wn.2d at 779. I would answer that, as a matter of policy, the failure to warn should not be considered the legal cause of the injury.
In this regard, I must say that the majority's "foreseeability" discussion, pages 294-96, focusing on the product's "use on babies", the description "pure and gentle", and the "suggestions" that it "be used on baby's scalp, which, of course, is near both the mouth and nose" is a red herring— plain and simple. This tragedy did not happen because the product was being used in any of the ways intended or described. Rather, aspiration occurred because a teenaged sister broke house rules and (1) removed a tiny-apertured bottle from its safe place; (2) transferred the oil into a *303widemouthed receptacle; (3) placed the unmarked container in her purse; (4) left the purse on the floor within reach of the baby and (5) left her door open. At this point, the mother, observing the child in the act of drinking from the bottle, cried out, producing a startled reaction of aspiration. How can it be said that but for defendant's failure to give some unspecified warning, this unfortunate accident would not have happened?
In my view, and for all the reasons outlined above, I would affirm the trial judge's granting of judgment n.o.v.
Review granted at 116 Wn.2d 1001 (1991).
It is conceivable that each individual juror had a different conception of the defendant's duty and of the particular type of warning that should have been given. This juror's view may or may not have agreed with that of any fellow juror. Thus, no rule of law can be extrapolated from the verdict.
The question has been treated as one of foreseeability although the analysis remains a matter of "whether the event was likely enough 'to induce action to avoid it on the part of a reasonable mind.'" Hentschel v. Baby Bathinette Corp., 215 F.2d 102, 106 (2d Cir. 1954), cert. denied, 349 U.S. 923 (1955), quoted in L. Frumer & M. Friedman, Products Liability § 2.22[3] (1989); see also L. Frumer & M. Friedman § 2.25[2].