dissenting.
I respectfully dissent because I feel the majority opinion misses the mark in its application of equal-protection law by attempting to distinguish asbestos litigation from other strict-liability warning cases and in other particulars which are detailed below.
At the outset, I am hesitant to find a denial of equal protection arising out of the rulings in Beshada v. Johns-Manville, 90 N.J. 191, 447 A.2d 539 (1982), Feldman v. Lederle Laboratories, 97 N.J. 429, 479 A.2d 374 (1984), and In the Matter of Asbestos Litigation Venued in Middlesex County, 99 N.J. 201, 491 A.2d 700 (1984). Vance v. Bradley, 440 U.S. 93, 99 S.Ct. 939, 59 L.Ed.2d 171 (1979), counsels deference to the government where its actions are challenged on equal-protection grounds since “even improvident decisions will eventually be rectified by the democratic process.” Id. at 97, 99 S.Ct. at 942. In the unusual situation before us, however, where we are called upon to review a judge-made rule of law, the efficacy of the checks and balances inherent in the “democratic process” is substantially reduced. In my view, therefore, our deference to the state court should be tempered and our application of the rationality test sharpened accordingly.
Furthermore, it must be remembered throughout that this is exclusively a fourteenth amendment controversy, therefore disagreement relating purely to underlying policy considerations has no rightful place in the discussion, nor can it be cloaked in the guise of equal-protection doctrine. Only sound constitutional arguments can be heard in objection to the majority’s conclusion.
As a final preliminary note, it should be stated that the issue before the court poses a genuinely troubling dilemma. The debate hinges on whether a real and substantial difference separates defendants in asbestos-related, failure-to-warn products-liability actions from those in all other failure-to-warn cases. Reed v. Reed, 404 U.S. 71, 76, 92 S.Ct. 251, 254, 30 L.Ed.2d 225 (1971). We can go two routes. If such a difference exists, asbestos firms are denied a potentially exculpatory defense which the New Jersey Supreme Court has itself expressly held available in other strict-liability contexts. Feldman, 97 N.J. at 451-52, 479 A.2d 374; O’Brien v. Muskin Corp., 94 N.J. 169, 463 A.2d 298 (1983) (state-of-the-art evidence is relevant in design-defect cases). If, on the other hand, the difference is nonexistant, we are inexorably drawn to the conclusion that the New Jersey Supreme Court has blinked on the doctrine of equal protection. Neither result is appealing. Nevertheless, reason, experience, common sense, and the Constitution dictate that we choose the second, and perhaps less-travelled, route, regrettable though it may be to disagree with the state’s highest court.
There can be no argument with the constitutional framework employed by the majority. Under traditional analysis, Beshada/Feldman denies equal protection only if there is no legitimate state purpose rationally promoted by the challenged classification. Western & Southern Life Insurance Co. v. California State Board of Equalization, 451 U.S. 648, 668, 101 S.Ct. 2070, 2083, 68 L.Ed.2d 514 (1981); San Antonio Ind. School Dist. v. Rodriguez, 411 U.S. 1, 17, 93 S.Ct. 1278, 1288, 36 L.Ed.2d 16, rehearing denied, 411 U.S. 959, 93 S.Ct. 1919, 36 L.Ed.2d 418 (1973).
My opposition originates in the soft and uncritical application of rule to fact. For satisfaction of the “legitimate purposes” requirement, the majority relies expressly on nothing more than the elimination of *781jury confusion and the expedition of trials. These are certainly significant objectives in any setting, but here those reasons simply are not enough to rebut a challenge on constitutional grounds where the underlying tort liability at stake is particularly high. In this regard, the opinion conspicuously lacks citation to case law for the critical proposition that such relatively slight goals have been, or can be, recognized as sufficient “legitimate state purposes” in an equal-protection setting. In addition, the reference to “legitimate concerns of case management, economics, and social welfare and policy,” at page 11 is undeveloped. It does little to truly differentiate asbestos from other difficult tort litigation or to rationalize the severe outgrowth of the proffered distinction.
It is also mere conjecture that elimination of state-of-the-art evidence will streamline litigation and reduce jury confusion. There has been no data to support the majority’s assertions in this regard. Indeed, in Wolf v. Procter & Gamble, 555 F.Supp. 613 (D.N.J.1982), it was held that state-of-the-art is a valid defense to negligence claims. Such evidence is also relevant to the consideration of punitive damages. Thus, state-of-the-art evidence may well find its way into asbestos trials regardless of Beshada/Feldman. Confusion would then be compounded, not reduced, since jurors would be required first to distinguish strict liability from negligence and then to discretely apply the state-of-the-art evidence only to the latter. Severance might avoid these results, but at the same time defeat the original aim of streamlining trial of the issues. In sum, accomodation of juries, though praiseworthy, carries little weight in the equation.1
The equal-protection test’s second element, too, is not sufficiently treated by the majority. In the single paragraph — indeed, the single sentence — devoted to this subject at page 13, the majority seems to assume rather than to identify and explain a rational relationship between the legitimate state purposes and- the segregation of asbestos defendants. Aside from expedience, a term employed by the majority at page 779 and one not known for its positive connotations, I am unable to discern from the opinion or to imagine of my own the necessary connection bridging means and ends.
Accordingly, in the absence of a stronger argument to satisfy both elements of the equal-protection formula, I am constrained to conclude that by prohibiting asbestos defendants from introducing state-of-the-art evidence the Supreme Court of New Jersey violates the fourteenth amendment.
Further considerations also compel my view of this matter. I disagree, for one, with the majority’s assessment that asbestos actions are “self-evidently” and substantially different from other products cases. Looks are deceiving, in litigation as in life; thus, no surer trap can be set than for a court to judge a suit by the code number2 it bears. I refer specifically to three areas in which the majority itself, and through its incorporation of decisions by Judge Gerry3 and Judge Keefe,4 seeks support for the distinctive treatment of asbestos: (1) volume of cases, (2) numerosity of parties, and (3) complexity of issues.
The volume of asbestos cases on court dockets5 is absolutely irrelevant to consid*782eration of the present equal-protection challenge. “The constitutional imperatives of the Equal Protection Clause cannot be satisfied by mere conjecture as to administrative convenience.” Frazier v. Manson, 703 F.2d 30, 35 (2d Cir.), cert. denied, 464 U.S. 934, 104 S.Ct. 339, 78 L.Ed.2d 308 (1983). More to the point, however, is the realization that if, as the majority premises, the elimination of juror confusion and resultant prejudice is at the core for precluding state-of-the-art evidence, what difference does it make then whether the court’s docket lists one or a million and one asbestos actions? One jury will, in most instances, decide but one case. And it matters not one iota to Monday’s jury that a million asbestos cases will still remain on Tuesday, nor does such knowledge contribute one measure of confusion to that panel’s deliberation. As Owens-Illinois, Inc. astutely observes in its brief6 on these issues, it would appear that there are more routine auto accident cases than asbestos cases in the courts, yet the many motorists in litigation are not denied the right to prove, for example, comparative negligence on the scant basis of their prolific numbers.
As for the majority's point concerning the abundance of plaintiffs and defendants in the asbestos cases, it must be noted that not every action is a multi-party behemoth. To be sure, there are cases where the parties are almost innumerable. But there are also instances where the named plaintiffs and defendants can each be counted on one hand. E.g., Miller v. Johns-Manville, Civil Action No. 85-389 (husband and wife v. five named defendants); Gregory v. General Motors Corp., Civil Action No. 82-4031 (administratrix and son v. one named defendant). Where in these smaller cases is the danger in permitting a jury to weigh the evidence of the actual and constructive knowledge of the handful of defendants who comprise the chain of asbestos distributors? I see none. Thus, the majority’s overly broad numerosity argument is misplaced and the New Jersey Supreme Court’s preclusion of state-of-the-art evidence constitutes needless and prejudicial overkill.
Now of course the majority is concerned lest the cumulative effect of complicated legal theories, multiple parties, and prolonged latency of injury overwhelm our juries. But even at its hypothetical worst, the danger is not so extreme as to warrant the pre-emptive strike against the asbestos litigants’ defensive arsenal. Fundamental fairness must prevail over cures that are worse than the disease. The proper remedy for projected jury confusion, prejudice, or delay is not arbitrary and radical excision of potentially dispositive evidence, but rather time-tested methods including thoughfully prepared jury instructions and charges, well-organized presentation of all evidence, expert testimony, and other ingredients which make up a well-tried case.
Even if it were agreed that juror confusion and facilitation satisfy the rationality test, I am not entirely convinced that it is proper for us to ascribe these purposes to the New Jersey Supreme Court. Admittedly, Delaware River Basin Commission v. Bucks County Water & Sewer Authority, 641 F.2d 1087 (3d Cir.1981), permits us to hypothesize as to the purposes behind a classification. But, in that very same opinion, the circuit court also recognized that the “Supreme Court has sometimes indicated that rationality analysis must consider the actual purposes of the legislature, rather than post hoc justifications offered by government attorneys or hypothesized by the court itself.” 641 F.2d at 1094. The United States Supreme Court has often looked for an actual articulation of the state’s purposes where the subject matter at hand was unusually sensitive or troublesome. See, e.g., San Antonio School District, 411 U.S. at 17, 44-52, 93 S.Ct. at 1288, 1302-06; McGinnis v. Royster, 410 U.S. 263, 270-71, 93 S.Ct. 1055, 1059-60, 35 *783L.Ed.2d 282 (1973). Thus, where a decision (1) emanates from a court and is therefore not readily susceptible to the aforementioned correctives of the democratic process, and (2) has generated substantial critical commentary,7 I do not feel it improper to impose an articulation requirement. Since my reading of the cases reveals no express or implied explanation for the discriminatory fence raised around asbestos defendants, the failure of the New Jersey Supreme Court to disclose its rationale is, in and of itself, fatal to the distinction. There is simply no way to escape the reality of this state of affairs and the majority compounds and perpetuates an unfortunate error.
Judge Gerry has also commented on this issue in an earlier opinion, noted above, from which the majority quotes extensively. An important distinction must be drawn regarding those excerpted comments. In looking generally at tort law, Judge Gerry observed by way of example, that this court recognizes numerous distinctions made between classes of litigants under the New Jersey Tort Claims Act and Workers Compensation Act and in medical malpractice litigation. Gogol at 974-75. In none of these contexts though has the classification effectively stripped litigants of a substantive right without at the very least providing some procedural quid pro quo. Here, the New Jersey Supreme Court has withheld a theory, which, if proved, could conceivably relieve numerous defendants of liability, but offered no substitute or counter-measure. Had the court somehow “re-evened” the score between asbestos plaintiffs and defendants or made the latter’s isolation less glaringly conspicuous, perhaps the Beshada/Feldman rule would be more palatable.
Furthermore, the majority has created a potential Pandora’s Box at pages 11 and 12 with its usage of the phrase “atypical products liability cases,” by no means a synonym for “asbestos cases.” Whatever Beshada/Feldman stands for, the damage should be kept to a minimum at all costs. Yet the imprecision that is brought to the opinion by the term “atypical” — however qualified it may be in context — unnecessarily opens the door for an extension of the challenged rule to an unforeseeable maze of applications.
Finally, the only point left for discussion is to surmise what in fact the New Jersey Supreme Court might have meant by the controversial language that restricted Beshada “to the circumstances giving rise to its holding.” Feldman, 97 N.J. at 455, 479 A.2d 374. As stated above, I find judicial speculation in this area to be inappropriate; therefore, I will hazard no guess as to what this phrase intends. I note only that had the supreme court actually wished the result reached by the majority, it could easily have said “state-of-the-art evidence is relevant in failure-to-warn cases but not where asbestos is involved,” or “but not where the case is big and clumsy.” The silence is indeed clanging.
In sum then, the New Jersey Supreme Court’s classification of asbestos defendants should fall for a multitude of reasons, for in the supposed hope of reducing that prejudice generated by confusion, the court and today’s majority achieve the very result they seek to avoid. Asbestos defendants are entitled to the same opportunity to be heard as manufacturers and distributors of all other products. The deprivation of this opportunity has the effect of inappropriately classifying these defendants and subjecting them to a different and harsher standard of liability by denying the chance to present relevant and probitive evidence.
The difficulty of the facts in a case is no ground for ignoring those facts. That is precisely what the majority proposes we do in our asbestos cases. Expedience is thereby exalted over fairness and justice is su*784bordinated to simplicity. It is inconceivable that these results were intended by the New Jersey Supreme Court.
. Not to be overlooked in this discussion of unburdening the jury is the distinct possibility that asbestos cases will be tried to the bench. Taking for granted the competence and acuity of the state and federal bench in New Jersey, the argument regarding simplification of the issues for the trier of fact therefore goes largely for naught.
. All asbestos cases commenced in the District of New Jersey are marked by the court clerk with the distinctive computer code number 368.
. Gogol v. Johns-Manville, 595 F.Supp. 971 (1984).
. In the Matter of Asbestos Litigation Venued in Middlesex County, Superior Court of New Jersey, Law Division, Docket No. L-52237-81, transcript of proceedings of Sept. 9, 1984.
. The district court clerk reports that approximately 133 asbestos-related cases were pending in the District of New Jersey as of January 1, 1986. The court's full docket consisted of 5,570 civil actions, thus 2.3% were asbestos related.
. See Memorandum of Law in Support of Motion of Defendant Owens-Illinois, Inc. to Admit Evidence Relating to its Actual or Constructive Knowledge of the Potential Hazards of its Product submitted to Judge Sarokin in Valori v. Johns-Manville Sales Corp., Civil Action No. 82-2686.
. See Schwartz, The Post-Sale Duty to Warn: Two Unfortunate Forks in the Road to a Reasonable Doctrine, 58 N.Y.U.L.Rev. 892, 901-04 (1983); Comment, Requiring Omniscience: The Duty to Warn of Scientifically Undiscoverable Product Defects, 71 Geo.L.J. 1635, 1644-53 (1983); Note, Products Liability — Strict Liability in Tort — State-of-the-Art Defense Inapplicable in Design Defect Cases, 13 Seton Hall L.Rev. 625, 636-42 (1983).