¶34 (concurring) — I concur in the majority’s conclusion that the principles of strict liability set out in Ulmer v. Ford Motor Co., 75 Wn.2d 522, 452 P.2d 729 (1969); Seattle-First Nat'l Bank v. Tabert, 86 Wn.2d 145, 542 P.2d 774 (1975), and their progeny, apply retroactively in this case.
¶35 However, I do not agree that the court’s discretion should be curtailed by strict application of the rules respecting retroactivity set out in Robinson v. City of Seattle, 119 Wn.2d 34, 830 P.2d 318 (1992). In particular, I disagree with the majority’s unwise edict that the only exception to the general rule of retroactivity is pure prospectivity which can be determined only in the case in which the new rule is announced. We have not, in the years since Robinson was decided, followed such a rigid approach, and for good reason.
¶36 In fact, in In re Detention of Audett, 158 Wn.2d 712, 719-23, 147 P.3d 982 (2006), we explicitly and deliberately applied the Chevron Oil factors to determine whether a rule regarding mental evaluations of alleged sexually violent predators announced in a prior case should be given prospective application or selective prospectivity rather than retroactive application. See Chevron Oil Co. v. Huson, 404 U.S. 97, 92 S. Ct. 349, 30 L. Ed. 2d 296 (1971), overruled in part by Harper v. Va. Dep’t of Taxation, 509 U.S. 86, 113 S. Ct. 2510, 125 L. Ed. 2d 74 (1993).
¶37 The majority cannot reconcile Audett with Robinson, and so it says instead that the discussion of Chevron Oil *287was “unnecessary to reach the holding” in Audett and that the result was “consistent with Robinson.” Majority at 281. The majority says the same is true of State v. Atsbeha, 142 Wn.2d 904, 16 P.3d 626 (2001). Majority at 281.
¶38 Regardless of the majority’s after-the-fact recharacterization, our analysis in Audett was not mere window-dressing. It was deliberate and deliberative. Audett plainly directs that in a case following the case in which the rule at issue is announced, the issue of retroactivity may be considered with prospective application remaining a possibility even though the rule was applied in the case in which it was announced.21
¶39 The majority allows that the Chevron Oil factors have a place in determining the question of pure prospectivity, which the majority says must be determined in the very same case in which the rule is announced, but they cannot be used to determine prospectivity in any succeeding case. Audett is completely to the contrary. Clearly abandoning the absolutes of Robinson, we recognized in Audett that fairness concerns may demand that we exercise our discretion and apply a prior decision prospectively.
¶40 In addition, the issue of retroactivity-prospectivity is often not addressed or even mentioned in the parties’ briefing in the case in which a judicially determined rule is first set out and it is often not addressed by the court in that case. This was exactly what happened (or did not happen) in In re Detention of Williams, 147 Wn.2d 476, 55 P.3d 597 (2002), the case announcing the rule that was at the center of the retroactivity-prospectivity question in Audett. Frequently, the issue of retroactivity or prospectivity first comes to the court’s attention in a subsequent case. At that point *288forceful arguments might be made showing unacceptable unfairness in applying the rule retroactively. Yet under the majority’s decision overruling Audett, our hands are now tied. We cannot do justice.
¶41 It is our responsibility, when developing the common law, “to endeavor to administer justice according to the promptings of reason and common sense, which are the cardinal principles of the common law.” Sayward v. Carlson, 1 Wash. 29, 41, 23 P. 830 (1890). If we reason that solely because the new rule has once been applied it must always be applied, we do not carry out this responsibility. There is nothing reasonable about retroactively applying a rule of law, no matter the reliance, surprise, hardship, or unfairness involved in retroactive application, merely because it has once been applied, and it is particularly unjust to do so if there has never been a considered decision on the issue of its retroactivity or prospectivity.
¶42 It is true that in Audett the Chevron Oil analysis did not lead us to the conclusion that retroactivity was fundamentally unfair. But another case, with another set of facts, and another new rule of law could lead us to an entirely different conclusion.
¶43 It must be remembered that the reason the court adopted the rule of retroactivity and abrogated selective prospectivity in Robinson was because we perceived that the United States Supreme Court had “recently limited the Chevron Oil . . . rule regarding retroactive application” in James B. Beam Distilling Co. v. Georgia, 501 U.S. 529, 111 S. Ct. 2439, 115 L. Ed. 2d 481 (1991), a split decision. Robinson, 119 Wn.2d at 73. It is obvious that the court found great significance in the fact that the United States Supreme Court altered its own retroactivity analysis — the analysis that we had been applying as well. Our court ultimately concluded that the reasoning in Beam Distilling was sound and accepted the premise that similarly situated litigants must always be treated equally. Id. at 77.
*289¶44 Then, a year after Robinson was decided, the Court explicitly held in Harper, 509 U.S. at 97, that when it applied “a rule of federal law to the parties before it, that rule is the controlling interpretation of federal law and must be given full retroactive effect in all cases still open on direct review and as to all events, regardless of whether such events predate or postdate” the court’s announcement of the rule. However, the Court also explicitly distinguished between rules of federal law and rules of state law. While a state court must follow Harper with regard to rules of federal law, state courts retain freedom to limit retroactive application of their interpretations of state law. Great N. Ry. v. Sunburst Oil & Ref. Co., 287 U.S. 358, 364-66, 53 S. Ct. 145, 77 L. Ed. 2d 360 (1932); see Harper, 509 U.S. at 100.
¶45 As the Montana Supreme Court explained, many state courts responded negatively to Harper’s retro-activity analysis, and of these a number continued to apply the Chevron Oil analysis or a similar analysis for determining whether a decision should apply prospectively. Dempsey v. Allstate Ins. Co., 2004 MT 391, 325 Mont. 207, 215, 104 P.3d 483. The Montana court also explained its own history in this area, which included adoption of Chevron Oil in 1978, its subsequent application of Harper, and then its reversion to Chevron Oil without reference to the line of cases following Harper. Id. at 210-11.
¶46 The Montana court’s history and ours are similar, in that this court adopted Chevron Oil’s analysis, then purportedly adopted Beam Distilling (Harper made Beam Distilling’s split decision explicitly the law), and then in Audett and other cases reverted to Chevron Oil. Montana ultimately decided to apply retroactivity as the presumptive rule but retained Chevron Oil’s prospectivity analysis as an exception when all of its factors favor prospectivity.
¶47 In Beavers v. Johnson Controls World Services, Inc., 118 N.M. 391, 881 P.2d 1376 (1994), cited in Audett, 158 Wn.2d at 722, the New Mexico court acknowledged the “compelling force of the desirability of treating similarly situated parties alike” and accordingly adopted a “presump*290tion of retroactivity for a new rule imposed by a judicial decision in a civil case.” Beavers, 118 N.M. at 398. However, the court retained the Chevron Oil analysis because it did not find this “reason so powerful that it requires a rule of blanket retroactivity.” Id. at 397. Rather, the court reasoned that in some cases the Chevron Oil factors, “particularly the factor or subfactor of the parties’ reliance on the old rule — will argue so strongly for nonretroactivity that the factor ... of similar treatment of similarly situated parties will simply be outweighed.” Id.
¶48 Significantly, and in marked contrast to the majority’s harsh analysis here, the New Mexico court “decline [d] to follow the Supreme Court’s lead” and pointedly concluded that it could apply a rule prospectively “even though (as in this case) the decision announcing the new rule has already been applied retroactively to the conduct of the litigants in the case in which the rule was announced.” Id.
¶49 Like Montana, the Ohio Supreme Court recently surveyed case law respecting states’ analyses for retroactive or prospective application of rules announced in judicial decisions, observing that Harper overruled Chevron Oil only insofar as it applied to federal law. DiCenzo v. A-Best Prods. Co., 120 Ohio St. 3d 149, 2008-Ohio-5327, 897 N.E.2d 132. The court stated that in Ohio the general rule is that a decision applies retrospectively unless a party has contract or vested rights under the prior decision. Id. at 156. However, an Ohio court “has discretion to apply its decision” prospectively under the Chevron Oil factors and under exceptional circumstances prospective application is justified. Id. at 156-57.
¶50 Like the New Mexico State Supreme Court, the Ohio court rejected the argument that if the case announcing the rule does not contain language imposing only prospective application, the rule was and continues to be retroactive. Id. at 156. The court did not agree that “the passage of time and appellate cases that have applied [the new rule] retrospectively preclude” a court from applying the rule prospectively. Id. The court said that “[t]he mere passage of time, *291without more, does not diminish our authority to impose a prospective-only application of a court decision.” Id. at 157.
¶51 Thus, the Ohio Court held that whenever the issue of retroactivity-prospectivity is first addressed, the court may exercise discretion and apply a rule prospectively if the Chevron Oil factors show this is appropriate. The court refused to give up its authority and discretion to decide that a decision may be prospective where the issue of prospectivity-retroactivity had not previously been determined.
¶52 Like these courts, in Audett we clearly recognized that retroactivity is the general rule. However, we also recognized that this general rule must yield in the face of compelling reasons favoring prospectivity, regardless of the fact that the new rule of law was applied in the announcing case. In accord with the views expressed by the New Mexico and Ohio courts, in Audett we considered whether the rule at issue should be applied in Audett or instead should be applied prospectively, even though the rule had been applied in the case in which it was announced (Williams, 147 Wn.2d 476).
¶53 Unlike the inflexible analysis of Robinson, which was, as noted, founded on changes to federal retroactivity law, our decision in Audett respects the importance of treating similarly situated litigants alike while retaining the court’s discretion to apply a state rule prospectively if the injustice of retroactive application outweighs the interest in similar treatment.
¶54 I believe Audett can be fairly read to mean only one thing: Even if a state rule is applied in the case in which it is announced, i.e., it is applied “retroactively” in that case, the court may consider in a subsequent case whether under the Chevron Oil factors the rule should nevertheless be given prospective effect. Because it fails to follow this analysis, the majority decision fails to follow our precedent — for Audett is precedent just as Robinson was, and it is Audett that is the later case. We did, in fact, sub silentio overrule Robinson insofar as it was intended to abro*292gate the possibility of any selective or modified rule of prospectivity.22
¶55 I believe the better rule is that there should be a presumption that a new rule applies retroactively, but this presumption can be overcome if an analysis under the Chevron Oil factors favors prospectivity. Prospectivity does not have to be determined in the same case that announces the new rule, but may be determined in a subsequent case.
Conclusion
¶56 The majority decides that we must surrender our discretion to apply a judicially based state rule of law prospectively even if would be inequitable and unjust to apply it retroactively. I believe the majority fails to carry out our responsibility to administer justice with the reason and common sense necessary to development of the common law. I would follow Audett and retain the court’s discretion to decide whether a judicially determined rule of law should be prospectively applied, regardless of whether the rule was applied in the case in which it was announced.
Alexander, C.J., and J.M. Johnson, J., concur with Madsen, J.Under the Chevron Oil standard, a court considers whether the rule should be given prospective or selectively prospective application by (1) considering whether the rule at issue is a new principle of law, because it either overruled clear past precedent upon which litigants relied or decided an issue of first impression and the decision was not clearly foreshadowed; (2) considering the prior history of the rule, its purpose and effect, and whether its operation would be furthered or retarded by retroactive application; and (3) weighing any inequity involved in retroactive application. Chevron Oil, 404 U.S. at 106-07.
In a strange statement about this court’s power, the majority says, “Because we have not overruled Robinson and decline to do so now, selectively prospective application of strict product liability is not an option.” Majority at 283. We have overruled Robinson, in part, albeit sub silentio. But even if we had not, there is no bar to our doing so now.