OPINION OF THE COURT BY LUM, C.J.

In this products liability case, Defendants-Appellants General Motors Corporation (GM) and Servco Pacific, Inc. (Servco) appeal a trial court judgment in favor of Plaintiffs-Appellees Steven Masaki, and his parents Frank and Sumiye Masaki. Steven was severely injured when he was run over by a GM van which Appellees claimed "self-shifted" into reverse gear due to a product defect. A jury found that the van was defective and that GM was negligent, and awarded Masaki compensatory and punitive damages. Appellants allege numerous trial court errors including erroneous jury instructions on punitive damages and liability, insufficient evidence to support liability and punitive damages, and lack of a legal basis for Masaki's parents' recovery of damages for loss of filial consortium and emotional distress.

Because we conclude that the jury should have been instructed to award punitive damages only if the supporting evidence is clear and convincing, we reverse the punitive damages award and remand for a new trial on the issue of punitive damages. We affirm the judgment of the trial court in all other respects.

I.

This case arises from a tragic accident which occurred while Steven Masaki, a 28-year old auto mechanic, was working on a 1976 Chevrolet van. On August 15, 1984, Masaki's employer sent him to jump start a van belonging to International Inflight Catering Co. (IICC) and bring it back to the shop for maintenance. The van had been manufactured by General Motors and sold to IICC by Servco. Masaki was accompanied by co-worker Sandra Reyes.

Masaki initially attempted to start the van by turning the key in the ignition. The engine turned over several times but did not catch. Masaki then decided to use a "remote starter," a device which bypasses the vehicle's ignition switch. He attached one of the two five-foot leads of the remote starter to the solenoid located underneath the van, and the other to the battery terminal under the hood. Masaki did not set the parking brake or block the wheels before going under the van, and could not recall whether the gear shift indicator ("PRNDL") was pointing to park.

Masaki pushed the starter button on the remote starter, causing the van to start. While the engine was running, he began to crawl under the van to disconnect the remote starter. The van made a sound like it was being put into gear, and immediately lurched backward. The van hit the back of Masaki's head, breaking his neck.

Sandra Reyes, who was sitting in a truck parked next to the van, facing in the opposite direction, heard the engine start and run for approximately twenty seconds. She then heard the van change gears and saw it jerk backwards. She discovered Masaki pinned underneath the van, turned off the engine, and went for help. Two witnesses observed that the gear shift lever was pointing to reverse immediately following the accident and that the parking brake was not set. Masaki's injuries rendered him a quadriplegic.

The Masakis filed a complaint against GM and Servco alleging negligence, breach of warranty, and strict products liability. Masaki claimed that the van's transmission was defectively designed and manufactured, and that he had not been warned of the defect. His parents sued for emotional distress and loss of Steven's consortium, services and support.

A jury trial lasting approximately two months commenced on December 7, 1987. On February 19, 1988, the jury returned a verdict finding that the van was defective and that GM was negligent, and that both the defect and GM's negligence were legal causes of Steven Masaki's injuries. The jury also found Masaki forty per cent contributorily negligent for his own injuries. The jury awarded Masaki $6,776,115 in compensatory damages and $11,250,000 in punitive damages. His parents were awarded $560,000 each for loss of consortium and $460,000 each for emotional distress. The court reduced all compensatory awards to Steven and his parents by forty per cent. Judgment was entered on March 15, 1988.

On March 24, 1988, Appellants moved for judgment notwithstanding the verdict, or in the alternative, a new trial, or a remittitur of the damages. The motion was denied on April 5, 1988, and this appeal followed.

Appellants raise numerous points of error. We will consider only those points which merit discussion, seriatim.

II. PUNITIVE DAMAGES

We turn first to the issue of punitive damages. GM urges reversal of the $11.250 million punitive damages award, claiming several grounds of trial court error: (1) erroneous jury instructions; (2) insufficient evidence to support the award; (3) failure to bifurcate the issue of punitive damages; (4) failure to set aside or remit the award; and (5) the unconstitutionality of punitive damages. Because we conclude that the jury was improperly instructed as to the standard of proof required in order to award punitive damages, we reverse the award and remand for a new trial on the issue of punitive damages. Therefore, we find it unnecessary to address the remaining grounds of error.

A.

Before addressing the question of the standard of proof for punitive damages, we find it useful to briefly review the doctrine of punitive damages, its underlying purpose and rationale, its compatibility with a product liability action, and its development in the law of Hawaii.

1.

Punitive or exemplary damages are generally defined as those damages assessed in addition to compensatory damages for the purpose of punishing the defendant for aggravated or outrageous misconduct and to deter the defendant and others from similar conduct in the future. D. Dobbs, Handbook on the Law of Remedies § 3.9, at 204 (1973); Restatement (Second) of Torts § 908 (1979). Thus, the practice of awarding punitive damages is an exception to the general rule that damages are aimed at compensating the victim for his injuries. C. McCormick, Handbook on the Law of Damages § 77, at 275 (1935).

Since the purpose of punitive damages is not compensation of the plaintiff but rather punishment and deterrence, such damages are awarded only when the egregious nature of the defendant's conduct makes such a remedy appropriate. Thus, "[w]here the defendant's wrongdoing has been intentional and deliberate, and has the character of outrage frequently associated with crime, all but a few courts have permitted the jury to award . . . [punitive damages.]" W.P. Keeton, Prosser Keeton on the Law of Torts § 2, at 9 (5th ed. 1984); Restatement (Second) of Torts § 908 comment b. While the concepts of punishment or deterrence usually do not enter into tort law, in this "one rather anomalous respect . . . the ideas underlying the criminal law have invaded the field of torts." Prosser Keeton, supra, at 9.

In determining whether an award of punitive damages is appropriate, the inquiry focuses primarily upon the defendant's mental state, and to a lesser degree, the nature of his conduct. Dobbs, supra, at 205. In the case of most torts, ill will, evil motive, or consciousness of wrongdoing on the part of the tortfeasor are not necessary to render his conduct actionable. In a negligence action, for example, the defendant may be required to make compensation if it is shown that he failed to comply with the standard of care which would be exercised by an ordinary prudent person, no matter how innocent of desire to harm. McCormick, supra, at 280. In contrast, to justify an award of punitive damages, "a positive element of conscious wrongdoing is always required." Id. Thus, punitive damages are not awarded for mere inadvertence, mistake, or errors of judgment. Restatement (Second) of Torts § 908 comment b; Prosser, supra, at 10. "Something more than the mere commission of a tort is always required for punitive damages." Prosser, at 9.

The practice of allowing punitive damages is centuries old. It first received explicit recognition by the English common law in the eighteenth century. The doctrine was rapidly transported to America and gained substantial acceptance in this country by the middle of the nineteenth century. Owen, Punitive Damages in Products Liability Litigation, 74 Mich. L. Rev. 1258, 1262-63 (1976). The concept of "exemplary" damages originated as a means of justifying awards of damages in excess of the plaintiff's tangible harm where the defendant's conduct was particularly egregious. Thus, the doctrine originally had a compensatory as well as punitive purpose. As the concept of actual damages broadened to include intangible harm, such as mental suffering, the compensatory function of exemplary damages came to be filled by actual damages. Thus, courts today generally speak of exemplary damages in terms of punishment and deterrence. Note, Exemplary Damages in the Law of Torts, 70 Harv. L. Rev. 517, 518 (1957). See generally Ghiardi Kircher, Punitive Damages Law and Practice, Ch. 1 (1985).

While the practice of allowing punitive damages is followed in an overwhelming majority of jurisdictions today, it has been the subject of substantial criticism and has been exhaustively analyzed by commentators for over one hundred years. See, e.g., Tuttle v. Raymond, 494 A.2d 1353, 1355-60 (Me. 1985) (discussing and rejecting various arguments raised against the doctrine of punitive damages); Ghiardi Kircher, supra, Ch. 2.

Despite its critics, the punitive damages doctrine has remained firmly established in the common law. "[T]he doctrine of punitive damages survives because it continues to serve the useful purposes of expressing society's disapproval of intolerable conduct and deterring such conduct where no other remedy would suffice." Mallor Roberts, Punitive Damages: Toward a Principled Approach, 31 Hastings L.J. 639, 641 (1980). While "[a]n award of compensatory damages may be sufficient when injury has resulted from well-intentioned, but poorly advised behavior[,] when the defendant's conduct can be characterized as malicious, oppressive, or otherwise outrageous, a stronger sanction is needed." Id. at 648. Imposing punitive damages "effectively expresses to the defendant that such conduct will not be tolerated." Id. In such circumstances, utilizing "the civil law to shape social behavior is both logical and desirable." Tuttle, supra, 494 A.2d at 1356.

2.

Although the imposition of punitive damages is well-established in traditional tort actions, the application of punitive damages in product liability cases has stirred much controversy. Those opposed to its use argue that punitive damages are incompatible with a products liability action based on a strict liability theory. See Ghiardi Kircher, supra, § 6.01. These critics maintain that, in strict products liability, a defendant's liability arises not from any finding of fault, but rather from a finding that the product is defective. By contrast, fault is a critical element in the recovery of punitive damages. Indeed, the imposition of a punitive damage award requires a finding of aggravated fault, and the conduct of the defendant is the determinant of liability. Id. § 6.04; Comment, The Imposition of Punitive Damages in Product Liability Actions in Pennsylvania, 57 Temp. L.Q. 203, 205 (1984).

Although commentators have debated the wisdom of permitting punitive damages in product liability suits, a majority of courts that have addressed the issue have permitted such awards. Comment, supra, at 210; Ghiardi Kircher, supra, § 6.14. Although a defendant's state of mind is irrelevant in determining whether compensatory damages are appropriate in a strict product liability action, courts have found it "a simple matter to allow the plaintiff to make a supplementary showing of aggravating conduct for the purpose of proving entitlement to punitive damages." Drake v. Wham-O Mfg. Co., 373 F. Supp. 608, 611 (E.D. Wis. 1974).

In rejecting the argument that exemplary damages are incompatible with the underlying theories of negligence and strict liability, the Wisconsin Supreme Court reasoned:

This court has rested its analysis of punitive damages not on the classification of the underlying tort justifying compensatory damages but on the nature of the wrongdoer's conduct. . . .

Punitive damages rest on allegations which, if proved, demonstrate a particular kind of conduct on the part of the wrongdoer, which has variously been characterized in our cases as malicious conduct or willful or wanton conduct in reckless disregard of rights or interests.

Wangen v. Ford Motor Co., 97 Wis. 260, 266-67, 294 N.W.2d 437, 442 (1980). So long as the plaintiff established the requisite "outrageous" conduct, the court concluded, there was no reason to disallow an award of punitive damages in a negligence or strict liability action. Id. at 443. The Alaska Supreme Court reached a similar conclusion in Sturm, Ruger Co. v. Day, 594 P.2d 38, 46-47 (Alaska 1979).

This court has long recognized that punitive damages are recoverable in tort action based on negligence. In Bright v. Quinn, 20 Haw. 504, 511 (1911), we declared that "[w]hile the propriety of the doctrine has been questioned, it is now too well established to admit of argument that in actions of tort punitive damages may, under certain circumstances, be awarded in addition to such sum as the plaintiff may be found entitled to purely by way of compensation for his injuries and suffering." We went on to describe the aggravated conduct on the part of the defendant which must be established in order to justify an award of punitive damages:

Such damages may be awarded in cases where the defendant "has acted wantonly or oppressively or with such malice as implies a spirit of mischief or criminal indifference to civil obligations"; or where there has been "some wilful misconduct or that entire want of care which would raise the presumption of a conscious indifference to consequences."

Id. at 512 (citations omitted).

We see no reason why punitive damages may not also be properly awarded in a products liability action based on the underlying theory of strict liability where the plaintiff proves the requisite aggravating conduct on the part of the defendant. Although strict liability dispenses with the need to prove fault in order to find the defendant liable, it does not preclude consideration of the defendant's aggravating conduct for the purpose of assessing punitive damages. See Owen, Punitive Damages in Products Liability Litigation, 74 Mich. L. Rev. 1258, 1268-71 (1976). Thus, we find no logical or conceptual difficulty in allowing a claim for punitive damages in a products liability action based on strict liability.

3.

Our decisions dating back to the nineteenth century recognized that "vindictive" damages may be awarded in tort cases "if circumstances of aggravation exist," Bernard v. Loo Ngawk, 6 Haw. 214 (1877); however, Bright v. Quinn, supra, was the first case to clearly set forth the nature of the defendant's conduct and state of mind which the plaintiff must prove in order to recover an award of punitive damages.

Our decisions since Bright have reaffirmed this standard. See, e.g., Chin Kee v. Kaeleku, 29 Haw. 524, 532-33 (1926); Howell v. Associated Hotels, Ltd., 40 Haw. 492, 499 (1954); Goo v. Continental Casualty Co., 52 Haw. 235, 239, 473 P.2d 563, 566 (1970) ("willful, malicious, wanton or aggravated wrongs where a defendant has acted with a reckless indifference to the rights of another"); Quedding v. Arisumi Bros., Inc., 66 Haw. 335, 340, 661 P.2d 706, 710 (1983) ("conduct amounting to wanton, oppressive, malicious, or reckless behavior").

We described the nature and purpose of punitive damages in Kang v. Harrington, 59 Haw. 652, 660, 587 P.2d 285, 291 (1978), a case involving fraud, as follows:

Punitive damages are in no way compensatory and are not available as a matter of right. An award of punitive damages is purely incidental to the cause of action. They may be awarded by the grace and gratuity of the law. They also act as a means of punishment to the wrongdoer and as an example and deterrent to others. (Citations omitted).

Similarly, in Lauer v. Young Men's Christian Association of Honolulu, 57 Haw. 390, 402, 557 P.2d 1334, 1342 (1976), we recognized that "[t]he award of punitive or exemplary damages constitutes an exception to the purely compensatory aspect of the damages concept as a means to right a wrong[,]" and that "[t]he deterrent or retributive effect of punitive damages must be placed squarely on the shoulders of the wrongdoer."

Thus, our decisions clearly reflect the dual purposes of punitive damages as punishing the defendant for aggravated misconduct and deterring the defendant and others from engaging in like conduct in the future. Our decisions further demonstrate that "something more" than mere commission of a tort is required to justify the imposition of punitive damages, as we have repeatedly emphasized that "[p]unitive damages may be awarded only in cases where the wrongdoer `has acted wantonly or oppressively or with such malice as implies a spirit of mischief or criminal indifference to civil obligations'; or where there has been `some wilful misconduct or that entire want of care which would raise the presumption of a conscious indifference to consequences.'" Kang v. Harrington, 59 Haw. at 660-61, 587 P.2d at 291.

B.

It is with these principles in mind that we now turn to the issues at hand. We first address the Appellants' contention that the trial court erred in instructing the jury that they may award punitive damages upon only "a preponderance of the evidence." Appellants urge us to adopt the more stringent requirement of "clear and convincing" evidence for the imposition of an exemplary award. While this court has never specifically declared the standard of proof required to justify an award of punitive damages, we have recognized that a stricter degree of proof than a mere preponderance of the evidence is required in certain classes of civil cases. For example, "clear and convincing" evidence is required to support a finding of fraud, Dobison v. Bank of Hawaii, 60 Haw. 225, 226, 587 P.2d 1234, 1235 (1978), and to prove the element of malice in a defamation action, Beamer v. Nishiki, 66 Haw. 572, 578, 670 P.2d 1264, 1270 (1983). For the reasons stated below, we conclude that such a higher standard of proof is also appropriate for a claim for punitive damages.

The purpose of fixing a particular standard of proof is to "instruct the factfinder concerning the degree of confidence our society thinks he should have in the correctness of factual conclusions for a particular type of adjudication." In re Winship, 397 U.S. 358, 370, 90 S.Ct. 1068, 1076, 25 L.Ed.2d 368, 379 (1970) (Harlan, J., concurring). The choice of standard of proof also "serves to allocate the risk of error between the litigants and to indicate the relative importance attached to the ultimate decision." Addington v. Texas, 441 U.S. 418, 423, 99 S.Ct. 1804, 1808, 60 L.Ed.2d 323, 329 (1979).

The law has evolved three standards of levels of proof for different types of cases. In most civil proceedings, such as a case involving a monetary dispute between private parties, the plaintiff must show by a "preponderance of the evidence" that his or her claim is valid. Under the preponderance standard, the parties share the risks of an erroneous verdict in roughly equal fashion. Id. at 423, 99 S.Ct. at 1808, 60 L.Ed.2d at 329. The preponderance standard directs the factfinder to decide whether "the existence of the contested fact is more probable than its nonexistence." E. Cleary, McCormick on Evidence § 339, at 957 (3d ed. 1984). As one commentator points out, to prevail, "[a] plaintiff need only offer evidence sufficient to tip the scale slightly in his or her favor, and a defendant can succeed by merely keeping the scale evenly balanced." Comment, The Imposition of Punitive Damages in Product Liability Actions in Pennsylvania, 57 Temp. L.Q. 203, 224 (1984).

At the other end of the spectrum, in criminal proceedings, the government is required to prove its case "beyond a reasonable doubt." Society has judged that it is significantly worse for an innocent man to be found guilty of a crime than for a guilty man to go free. McCormick, supra, § 341, at 962. Therefore, as stated by the Supreme Court, "[w]here one party has at stake an interest of transcending value — as a criminal defendant his liberty — this margin of error is reduced as to him by the process of placing on the other party the burden . . . of persuading the factfinder at the conclusion of the trial of his guilt beyond a reasonable doubt." Speiser v. Randall, 357 U.S. 513, 525-26; 78 S.Ct. 1332, 1341, 2 L.Ed.2d 1460, 1472 (1958); see also Addington, 441 U.S. at 423-24, 99 S.Ct. at 1808, 60 L.Ed.2d at 329.

The level of proof between these two extremes is that of "clear and convincing" evidence. This more exacting standard has been applied to a wide variety of civil cases where for policy reasons the courts require a higher than ordinary degree of certitude before making factual findings. It is typically used

in civil cases involving allegations of fraud or some other quasi-criminal wrongdoing by the defendant. The interests at stake in those cases are deemed to be more substantial than mere loss of money and some jurisdictions accordingly reduce the risk to the defendant of having his reputation tarnished erroneously by increasing the plaintiff's burden of proof.

Addington, 441 U.S. at 424, 99 S.Ct. at 1808, 60 L.Ed.2d at 330.

Thus, "clear and convincing" evidence may be defined as an intermediate standard of proof greater than a preponderance of the evidence, but less than proof beyond a reasonable doubt required in criminal cases. It is that degree of proof which will produce in the mind of the trier of fact a firm belief or conviction as to the allegations sought to be established, and requires the existence of a fact be highly probable. See Welton v. Gallagher, 2 Haw. App. 242, 245-46, 630 P.2d 1077, 1081 (1981); Bud Wolf Chevrolet, Inc. v. Robertson, 519 N.E.2d 135, 138 (Ind. 1988); McCormick, supra, § 340, at 959-60.

As one court explained:

[C]lear and convincing proof is a standard frequently imposed in civil cases where the wisdom of experience has demonstrated the need for greater certainty, and where this high standard is required to sustain claims which have serious social consequences or harsh or far reaching effects on individuals to prove willful, wrongful and unlawful acts to justify an exceptional judicial remedy. . . .

So, in a number of cases where an adverse presumption is to be overcome, or on grounds of public policy and in view of peculiar facilities for perpetrating injustice by fraud or perjury, the degree of proof required is expressed in such terms as . . . `clear and convincing' . . . and the phrase `preponderance of the evidence' has been expressly disapproved as an insufficient measure of the proof required[.]

Travelers Indem. Co. v. Armstrong, 442 N.E.2d 349, 360 (Ind. 1982).

We have repeatedly said that the fundamental purpose underlying an award of exemplary or punitive damages is to punish the wrongdoer and to deter him and others from committing similar wrongs and offenses in the future. Kang v. Harrington, supra; Howell v. Associated Hotels, Ltd., 40 Haw. 492, 499 (1954). Thus, punitive damages are a form of punishment and can stigmatize the defendant in much the same way as a criminal conviction. It is because of the penal character of punitive damages that a standard of proof more akin to that required in criminal trials is appropriate, rather than the preponderance of the evidence standard generally employed in trials of civil actions. Orkin Exterminating Co. v. Traina, 486 N.E.2d 1019, 1022 (Ind. 1986). A more stringent standard of proof will assure that punitive damages are properly awarded. For "although punitive damages serve an important function in our legal system, they can be onerous when loosely assessed." Tuttle v. Raymond, 494 A.2d 1353, 1363 (Me. 1985); Linthicum v. Nationwide Life Ins. Co., 150 Ariz. 326, 723 P.2d 675 (1986).

Accordingly, for all punitive damage claims we adopt the clear and convincing standard of proof. The plaintiff must prove by clear and convincing evidence that the defendant has acted wantonly or oppressively or with such malice as implies a spirit of mischief or criminal indifference to civil obligations, or where there has been some wilful misconduct or that entire want of care which would raise the presumption of a conscious indifference to consequences. Bright, supra.

We hold that the trial court erred in instructing the jury that they may award punitive damages by a preponderance of the evidence. Therefore, we reverse the punitive damage award and remand for a new trial on the issue of punitive damages only.

III. EMOTIONAL DISTRESS

We now turn to Appellants' claim that the trial court erred in awarding damages to Steven Masaki's parents for emotional distress.

In Rodrigues v. State, 52 Haw. 156, 173-74, 472 P.2d 509, 520 (1970), we recognized the independent tort of negligent infliction of mental distress and enunciated the rule that "[s]erious mental distress may be found where a reasonable man, normally constituted, would be unable to adequately cope with the mental stress engendered by the circumstances of the case." We rejected the requirement that the plaintiff must exhibit physical symptoms of mental distress in order to recover. See also Leong v. Takasaki, 55 Haw. 398, 403, 520 P.2d 758, 764-65 (1974). Nor have we found it necessary that the plaintiff actually witness the tortious event in order to recover for emotional distress. See Rodrigues, supra; Campbell v. Animal Quarantine Station, 63 Haw. 557, 632 P.2d 1066 (1981).

Nevertheless, in their eighteenth and nineteenth points of error, Appellants claim that the Masakis are not entitled to recover damages for emotional distress because they were not present at the scene of the accident and did not suffer any physical manifestations of emotional distress. In support of their contention, Appellants maintain that our decision in Kelley v. Kokua Sales Supply, Inc., 56 Haw. 204, 532 P.2d 673 (1975), is dispositive of this case. We disagree.

In Kelley, we denied recovery to a father, who resided in California, for emotional distress upon learning of the deaths of his daughter and granddaughter which occurred in Hawaii. We concluded that because the father's location from the scene of the accident was too remote for defendants to have reasonably foreseen the consequences of their conduct, the defendants owed no duty to him. Id. at 209, 532 P.2d at 676. In Campbell, however, while recognizing the geographical limitation imposed in Kelley, we allowed the plaintiffs to recover since both they and their dog "were located within Honolulu." 63 Haw. at 562, 632 P.2d at 1069.

We find the facts of this case analogous to those in Campbell and Rodrigues. While Frank and Sumiye Masaki were not present at the scene of their son's accident, they resided on the same island, and witnessed the consequences of the accident. Upon learning of Steven's injuries, the Masakis went immediately to the hospital and were told that their son would never walk. We conclude that it was reasonably foreseeable for Appellants to have anticipated the Masakis' emotional distress. The fact that the Masakis did not witness the accident is not a bar to recovery, but rather is a factor in determining the degree of mental stress suffered. Whether the degree of stress engendered by the circumstances of this case was beyond that with which a reasonable man can be expected to cope is a question for the jury. Leong v. Takasaki, 55 Haw. at 410, 520 P.2d at 766.

We therefore conclude that the trial court did not err in awarding damages for emotional distress to the Masakis.

IV. LOSS OF CONSORTIUM

In their twentieth point of error, Appellants present a question of first impression in this state, whether there is a cause of action available to parents for the loss of consortium of an adult child who has been severely and permanently injured due to the defendant's negligence. We conclude that such a cause of action ought to be available and therefore affirm the trial court judgment awarding Frank and Sumiye Masaki damages for the lost consortium of their son Steven.

Loss of filial consortium is a recognized cause of action in Hawaii under our wrongful death statute, Hawaii Revised Statutes (HRS) § 663-3. To date, however, neither the legislature nor this court has recognized a cause of action for loss of filial consortium resulting from injury short of death caused by the wrongful conduct of another.

We find ourselves in agreement with those jurisdictions which have recognized that severe injury may have just as deleterious an impact on filial consortium as death. These jurisdictions recognize that, in the case of a severely injured child, the quality of the parent-child relationship, as well as the parents' expectations of a normal family life, can be seriously impaired.

We agree with the Arizona Supreme Court that "no meaningful distinction can be drawn between death and severe injury where the effect on consortium is concerned. Often death is separated from severe injury by mere fortuity[.]" Howard Frank, M.D., P.C. v. Superior Court, 150 Ariz. 228, 230, 722 P.2d 955, 957-58 (1986). The court reasoned as follows:

"Perhaps the loss of companionship and society experienced by the parents of a child permanently and severely injured . . . is in some ways even greater than that suffered by parents of a deceased child. Not only has the normal family relationship been destroyed, as when a child dies, but the parent also is confronted with his loss each time he is with his child and experiences again the child's diminished capacity to give comfort, society, and companionship." Note, The Parental Claim for Loss of Society and Companionship Resulting From the Negligent Injury of a Child: A Proposal for Arizona, 1980 Ariz. St. L.J. 909, 923.

Id. at 231, 722 P.2d at 958. Thus, it would be anomalous to take the position that, if a child is injured, but does not die, the parents may not recover.

Appellants maintain however, that the Masakis are not entitled to recover for loss of filial consortium because Steven was twenty-eight years old at the time of his injury. We realize that a number of courts which recognize the parents' cause of action for loss of consortium of their injured children restrict the action to minor children. This rule is generally premised on the rationale that upon emancipation, parents are no longer entitled to the services and earnings of their children. We find such reasoning outmoded and illogical. At common law, the child, like the wife, was relegated to the role of a servant and considered an economic asset to the family. W.P. Keeton, Prosser and Keeton on the Law of Torts § 125, at 931 (5th ed. 1984). In the modern family, however, children have become less of an economic asset and more of a financial burden to their parents. Today children are valued for their society and companionship. See Shockley v. Prier, 66 Wis.2d 394, 225 N.W.2d 495, 498 (1975). Thus, services have become only one element of the consortium action while the intangible elements of love, comfort, companionship, and society have emerged as the predominant focus of consortium actions. Howard Frank, 150 Ariz. at 232, 722 P.2d at 959.

It is irrelevant that parents are not entitled to the services of their adult children; they continue to enjoy a legitimate and protectible expectation of consortium beyond majority arising from the very bonds of the family relationship. Surely nature recoils from the suggestion that the society, companionship and love which compose filial consortium automatically fade upon emancipation; while common sense and experience teach that the elements of consortium can never be commanded against a child's will at any age. The filial relationship, admittedly intangible, is ill-defined by reference to the ages of the parties and ill-served by arbitrary age distinctions. Some filial relationships will be blessed with mutual caring and love from infancy through death while others will always be bereft of those qualities. Therefore, to suggest as a matter of law that compensable consortium begins at birth and ends at age eighteen is illogical and inconsistent with common sense and experience. Human relationships cannot and should not be so neatly boxed.

Id. at 233, 722 P.2d at 960. See generally Love, Tortious Interference with the Parent-Child Relationship: Loss of an Injured Person's Society and Companionship, 51 Ind. L.J. 590, 622-23 (1976).

Moreover, we note that in wrongful death actions pursuant to HRS § 663-3, no arbitrary age limit is placed upon recovery for loss of filial consortium. HRS § 663-3 makes no distinction between minor and adult children. We see no reason why in cases of severe injury the result should be any different. Howard Frank, 150 Ariz. at 233, 722 P.2d at 960.

We hold that a parent may recover damages for the loss of filial consortium of an injured adult child. We further hold that the parents' cause of action shall be joined whenever feasible with that of the child for the child's personal injuries. See Shockley v. Prier, 66 Wis.2d at 404, 225 N.W.2d at 501; Howard Frank, 150 Ariz. at 234, 722 P.2d at 961.

We therefore affirm the lower court's award of damages for the loss of consortium to Frank and Sumiye Masaki.

V. JURY INSTRUCTIONS ON LIABILITY

We turn next to Appellants' seventh through tenth points of error which are addressed to the Court's Instruction No. 47 relating to products defect liability. Appellants first maintain that the instruction failed to inform the jury that they must find the product "dangerously" defective.

In Ontai v. Straub Clinic Hospital, Inc., 66 Haw. 237, 241, 659 P.2d 734, 739 (1983), we stated the following rule for strict product liability:

The rule, as thus adopted for this jurisdiction, provides that where a seller or lessor, who is engaged in the business of selling or leasing a product, sells or leases a defective product which is dangerous to the user or consumer, and injury results from its use or consumption, the seller or lessor will be held strictly liable in tort for the injury. (Emphasis added).

Similarly, in Johnson v. Raybestos-Manhattan, Inc., 69 Haw. 287, 288, 740 P.2d 548, 549 (1987), we stated that "the plaintiff need only show that the seller is engaged in the business of selling the product, that the product contains a defect dangerous to the user or consumer, and that the defect is the cause of the injury." (Emphasis added).

The jury was instructed that plaintiffs must "establish that [the product] failed to perform as safely as an ordinary user of the product would expect when used in an intended or reasonably foreseeable manner, including reasonable foreseeable misuses." In Ontai, we defined "dangerous" in language almost identical to that used in the jury instruction:

[T]he plaintiff need not show that the article was dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases or uses it. It is enough that the plaintiff demonstrates that because of its manufacture or design, the product does not meet the reasonable expectations of the ordinary consumer or user as to its safety. (Emphasis added.)

Ontai, 66 Haw. at 241, 659 P.2d at 739 (footnote and citations omitted). Thus, if the jury found the van defective under the first test listed in the instruction, it would necessarily also find that the van was "dangerously" defective within the meaning of Ontai.

Appellants next complain that Instruction No. 47 erroneously shifted the burden to the defendant to prove that the design was not defective by stating: "The second test is that the product is defective in design if Plaintiffs prove that the product's design was a legal cause of the injuries and Defendant fails to prove that the benefits of the design outweigh the risk of danger inherent in the design." (Emphasis added.) We disagree. In Ontai, we expressly adopted the following tests for finding a product defective in design:

First, a product may be found defective in design if the plaintiff establishes that the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner. Second, a product may alternatively be found defective in design if the plaintiff demonstrates that the product's design proximately caused his injury and the defendant fails to establish, in light of the relevant factors, that, on balance, the benefits of the challenged design outweigh the risk of danger inherent in such design.

66 Haw. at 242, 659 P.2d at 739-40 (quoting Barke v. Lull Engineering Co., 20 Cal.3d 413, 432, 143 Cal.Rptr. 225, 237-38, 573 P.2d 443, 455-56 (1978)). The jury instruction is clearly consistent with the test we adopted in Ontai.

Appellants further argue that it was error not to instruct the jury that the "feasibility and beneficial effect" of including a warning is a factor in determining whether the lack of a warning makes a product defective. Instruction No. 47 states in part: "The third test is that the product is defective in design . . . if the use of the product . . . involves a substantial danger that would not be readily recognized by the ordinary user of the product and the manufacturer fails to give adequate warnings of the danger." Appellants argue that the "feasibility and beneficial effect" of a warning should have been included as a factor in determining whether a danger is "substantial." We disagree. While we have recognized that a "manufacturer must give appropriate warning of any known dangers which the user of its product would not ordinarily discover," including "warning as to dangers inherent in improper use," Ontai, 66 Haw. at 248, 659 P.2d at 743, we conclude that it was not error to fail to specifically instruct the jury that they must consider the "feasibility and beneficial effect" of a warning.

In their ninth point of error, Appellants contend that, because there was no evidence that a warning would have prevented the accident, the jury should not have been instructed that failure to warn of a substantial danger could make a product defective. Appellants also argue, on the same ground, that the issue of negligent failure to warn should not have been given to the jury. We conclude that there was sufficient evidence on the question of causation to submit both theories of liability to the jury.

To support a jury instruction, there must be sufficient evidence presented on that issue of fact. Goo v. Continental Casualty Co., 52 Haw. 235, 239, 473 P.2d 563, 566 (1970). The evidence presented at trial indicates that an active warning system could have been installed at a low cost. GM expert engineer, Robert Lange, testified that warning systems were technologically and economically feasible, and could be installed at a cost of between $20 and $50. He described a system that causes a horn to go off and a light to flash if the driver begins to get out of the car when the gear shift lever is not latched in park. Thus, a driver who misshifts the vehicle, leaving it between park and reverse, would hear a loud noise and see a bright red light flashing on the shift indicator while still sitting in the driver's seat. Lange went on, however, to opine that such a warning system would not work because the driver would not understand what it meant, would not pay attention, and the alarm would go off at inappropriate times.

GM expert engineer, Roger McCarthy, testified that a person's reaction to the warning system for the first time would be one of surprise and shock and that it would cause the person to investigate. He further opined, however, that such a system would lose its effectiveness because of false alarms, and should not be installed because it would not work.

We have recognized that "[i]nasmuch as causal relation is one of fact, `[i]t is [a question] for the jury [to decide], except when the facts are such that they will support only one reasonable inference.'" (Emphasis in original). Knodle v. Waikiki Gateway Hotel, Inc., 69 Haw. 376, 389, 742 P.2d 377, 385 (1987). The evidence indicated that when Steven Masaki exited the van on the driver's side, the gear shift lever was not latched in park. The evidence also indicated that GM could have installed a warning device that would cause an alarm to go off in the event of a misshift. It was a question for the jury to decide whether a blaring horn and flashing light would have caused Masaki to discover the danger and thereby have prevented the accident. Although the Appellants presented testimony that such a system would not be effective, the jury, as the ultimate arbiter of credibility, could have rejected such testimony. We therefore conclude that the jury was properly instructed on the theories of negligent failure to warn and product defect.

Finally, in their twelfth point of error, Appellants argue that the trial court erred in instructing the jury as to a "manufacturing" defect when there was no substantial evidence to prove one. Instruction No. 80 provided: "A defect in the manufacture of a product exists if the product differs from the manufacturer's intended result or differs from apparently identical products from the same manufacturer."

The basis for Masaki's theory of a manufacturing defect was a "stickiness" in the van's gear shift which would make an incomplete shift into park more likely. This theory was supported by the testimony of expert engineer, Melvin Richardson, who opined that the fit of the parts in the steering column prevented the system from operating as freely as it should, and that this friction was present at the time of manufacture. In addition, two employees of IICC, the owner of the van, testified they had driven the van and found it difficult to shift into park. Thus, there was sufficient evidence that the van "differ[ed] from the manufacturer's intended result or differ[ed] from apparently identical products from the same manufacturer" to submit this theory of liability to the jury, and we find no error in the instruction given.

VI. SUFFICIENCY OF EVIDENCE ON LIABILITY

Appellants, in their thirteenth point of error, argue that the jury's verdicts of liability were not supported by substantial evidence. We disagree. Upon a careful review of the record, we conclude that the verdicts of negligence and product liability were supported by substantial evidence. See Adair v. Hustace, 64 Haw. 314, 325, 640 P.2d 294, 302 (1982).

VII. EVIDENTIARY ERRORS

Finally, Appellants' fourteenth through seventeenth points of error are addressed to certain evidentiary rulings of the court below. In their opening brief, Appellants acknowledged that they would "necessarily surrender" these alleged errors if they were not permitted to attach to their opening brief an appendix consisting of over 500 pages of testimony and record references. We denied Appellants' motion to attach the supplemental material to their brief. Accordingly, we hold that Appellants have waived points of error fourteen through seventeen on appeal.

VIII. CONCLUSION

Accordingly, we vacate the lower court's award of punitive damages and remand for a new trial on the issue of punitive damages only. In all other respects, we affirm the judgment of the trial court.