(dissenting) — I cannot agree with the majority’s application of the rule of Smith v. Rodene, 69 Wn.2d 482, 418 P.2d 741 (1966), that there can be no recovery for mental and emotional distress in cases not involving malice or wrongful intent, unless there has been an actual invasion of a plaintiff’s person or security, or a direct possibility thereof. I am convinced that the traditionally advanced justifications for retaining this limitation on recoveries for tortious conduct are no longer appropriate or justifiable, and that, contrary to the majority’s assertion, the instant case is an appropriate one for 'altering an old and, I think, outmoded rule, legal hypothesis or principle. This court should therefore reverse the action of the trial court granting summary judgment and allow this case to proceed to trial on the issue of the liability of the parents of Reed Christensen for the emotional distress suffered by Mr. and Mrs. Schurk.
The argument for retention of the old rule is usually phrased in terms of public policy considerations. I believe that an analysis of the opinions denying recovery in such cases discloses that the alleged public policy is based on one or more of the following reasons: (1) there is no precedent for such recovery; (2) allowance of such recovery will increase litigation; (3) damages from emotional injuries *659are difficult to prove and measure, and allowance of recovery for such injuries would therefore lead to fraudulent claims and actions; and (4) affowance of such recovery will lead to unlimited liability of defendants. See Throckmor-ton, Damages for Fright, 34 Harv. L. Rev. 260, 273-74 (1921); Rodrigues v. State, 52 Hawaii 156, 472 P.2d 509, 519 (1970).
The first reason given for the existence of the alleged public policy is no reason at all. “If it were, every case of first instance would be decided against the party invoking the new rule of law or the new application of an old rule.” Throckmorton, Damages for Fright, at 274. Furthermore, precedent does now exist for the allowance of recovery for emotional harm suffered by one who is not within the zone of danger of a tort. The California Supreme Court, in Dillon v. Legg, 68 Cal. 2d 728, 441 P.2d 912, 69 Cal. Rptr. 72 (1968) , allowed recovery by a mother standing outside the zone of danger who witnessed the death of her child in an automobile accident. This rule has been extended in Archibald v. Braverman, 275 Cal. App. 2d 253, 79 Cal. Rptr. 723 (1969) , to allow recovery by a mother who did not actually witness her child’s accident but came upon the scene soon after the accident had occurred.
Other jurisdictions have begun to follow the lead of the California court in expanding the right of recovery for negligently inflicted emotional distress. The Supreme Court of Hawaii, in upholding a judgment for a homeowner who suffered serious mental distress from flood damage caused hy a culvert which was negligently allowed to become blocked, held that there is a duty to refrain from negligent infliction of emotional distress. Rodrigues v. State, supra. And in Wallace v. Coca-Cola Bottling Plants, Inc., 269 A.2d 117, 121 (Me. 1970), the Supreme Court of Maine adopted a similar rule in stating that:
[W]e adopt the rule that in those cases where it is established by a fair preponderance of the evidence there is a proximate causal relationship between an act of negligence and reasonably foreseeable mental and emotional *660suffering by a reasonably foreseeable plaintiff, such proven damages are compensable even though there is no discernable trauma from external causes. The mental and emotional suffering, to be compensable, must be substantial and manifested by objective symptomatology.
Furthermore, the English courts have long allowed recovery for negligently inflicted emotional distress. Hambrook v. Stokes Bros., 1 K.B. 141 (1925). Most recently, the principle has been extended in England to allow recovery by the widow of a man who suffered severe psychoneurotic symptoms after serving, with no possibility of injury to himself, as a rescuer following a gruesome train wreck. Chadwick v. British Rys. Bd., 1 W.L.R. 912 (1967). Commentators have argued for recognition of similar rights in the United States, with the liability of the defendant being limited only by the foreseeability of the harm caused:
In other cases, however, plaintiff is outside the zone of physical risk (or there is no risk of physical impact at all), but bodily injury or sickness is brought on by emotional disturbance which in turn is caused by defendant’s conduct. Under general principles recovery should be had in such a case if defendant should foresee fright or shock severe enough to cause substantial injury in a person normally constituted. Plaintiff would then be within the zone of risk in very much the same way as are plaintiffs to whom danger is extended by acts of third persons, or forces of nature, or their own responses (where these things are foreseeable).
(Footnotes omitted.) 2 F. Harper and F. James, The Law of Torts 1035-36 (1956).
Nor can it be argued that there is no precedent for the recovery of emotional damages in other circumstances in the state of Washington. We have consistently held that damages for mental suffering are recoverable as incident to physical harm. Green v. Floe, 28 Wn.2d 620, 183 P.2d 771 (1947); Cote v. Allen, 50 Wn.2d 584, 313 P.2d 693 (1957). Furthermore, such damages are recoverable even without accompanying physical harm where theré is a direct possibility of such harm to the plaintiff. See Smith v. Rodene, 69 Wn.2d 482, 418 P.2d 741 (1966); Frazee v. Western Dairy *661Prods., 182 Wash. 578, 47 P.2d 1037 (1935). And emotional distress, if inflicted intentionally, constitutes an independent tort. Gadbury v. Bleitz, 133 Wash. 134, 233 P. 299, 44 A.L.R. 425 (1925); Browning v. Slenderella Systems, 54 Wn.2d 440, 341 P.2d 859 (1959). This court has apparently not been deterred from allowing damages for emotional harm by the argument that such damages are too difficult to measure or assess.
The second reason assigned for a public policy not allowing recovery for emotional damages is that allowing such damages will increase litigation. The answer to this reasoning is simple; the courts exist to provide a means of obtaining redress for injuries, not to restrict litigation.
[T]he frequency with which such suits are brought, after making liberal allowance for groundless claims, is the strongest evidence of the amount of injury inflicted and of the justice of the rule allowing a recovery.
Throckmorton, Damages for Fright, 34 Harv. L. Rev. at 276.
Third, it is claimed that damages from nervous shock are difficult to prove and measure, and that allowance of recovery for such injuries would therefore lead to the bringing of actions involving fraudulent claims. This reasoning was valid in the eighteenth century; it is not valid today. Even at the beginning of this century Kennedy, J., in Dulieu v. White & Sons, 2 K.B. 669, 681 (1901), wrote that:
My experience gives me no reason to suppose that a jury would really have more difficulty in weighing the medical evidence as to the effects of nervous shock through fright, than in weighing the like evidence as to the effects of nervous shock through a railway collision or a carriage accident, where, as often happens, no palpable injury, or very slight palpable injury, has been occasioned at the time.
Today we know that mental anguish and emotional distress are injuries as real and as physical in their causes and effects as those that might be produced by the common law tort of battery. Medical science has learned how to diagnose and, in large measure, how to cure, these injuries. Their existence or nonexistence is as susceptible of proof *662(or disproof) as many physical injuries for which recovery is unquestionably allowed; for instance, a whiplash injury to the neck.
Furthermore, the distinction between physical and mental injury is at bottom an empty one. On one hand it can be said that all pain, whether from “physical” or “mental” injuries, is mental. “Psychological evidence and phenomenological theory suggests that all pain is mental in that it is consciously perceived — that if it is not perceived, there is no pain.” Comment, Mental Distress in Psychological Research, 21 Baylor L. Rev. 520, 522-23 (1969). On the other hand, fear or emotional distress cannot exist without accompanying physical manifestations; all acute emotional experiences involve physical change. Specifically, emotionality involves changes in digestion, elimination, heartbeat, respiration, electrical potential of the skin, sugar level, and pilomotor responses. Id; see Selye, The Physiology and Pathology of Exposure to Stress 336-40 (1960). Given the present extent of medical knowledge, it must be conceded that:
Fear is more than a purely emotional and subjective response. There are marked physical reactions which, in certain cases, have deep and significant effects upon the entire human organism. Mental distress induced by fear is, therefore, capable of clear and convincing proof.
(Footnotes omitted.) Comment, Bystander Recovery for Mental Distress, 37 Fordham L. Rev. 429,448 (1969).
Of course it must be conceded that damages are not as precisely ascertaináble in this case as they are, for example, under a carefully drafted liquidated damages clause of a contract. However, we have consistently held that recovery of substantial damages is not to be denied merely because the extent or amount thereof cannot be ascertained with mathematical precision, provided the evidence is sufficient to afford a reasonable basis for estimating loss. E.g. Prier v. Refrigeration Eng’r Co., 74 Wn.2d 25, 442 P.2d 621 (1968); Reefer Queen Co. v. Marine Constr. & Design Co., 73 Wn.2d 774, 440 P.2d 448 (1968). Nor has the diffi*663culty of precisely ascertaining emotional damages kept this court from allowing recovery for such damages in the instances discussed above. The absurdity of flatly denying any recovery for emotional damages because of difficulty of proof is easily demonstrated by considering the case of a defendant who negligently frightens a plaintiff. S'ay that the fright causes the plaintiff to experience a nervous shock and also to fall, and that the fall results in bodily bruises. Clearly, under the doctrine of Green v. Floe, supra, and Cote v. Allen, supra, that damages for mental suffering are recoverable as incident to physical harm, the plaintiff could recover not only for the bruises but also for the nervous shock. Yet in this case the nervous shock for which recovery is allowed is no more susceptible of proof 'and no less remote from the fright which caused it than it would have been if unaccompanied by the fall and bruises.
It seems to me entirely too tenuous, perhaps even biased and irrational for serious judicial credence to be given to the argument that causes of action such as the one now before this court are particularly likely to lead to fraudulent claims. Certainly, such claims are no more likely to be false than claims for damages for intentionally inflicted emotional distress, already recognized as an independent tort in this state. Gadbury v. Bleitz, supra. In any case, we should reject the argument that we must deny recovery upon a legitimate claim because other fraudulent claims may be urged. See Throckmorton, Damages for Fright, 34 Harv. L. Rev. at 276.
[T]he fact that there may be greater opportunity for fraud or collusion in one class of cases than another does not warrant courts of law in closing the door to all cases of that class. Courts must depend upon the efficacy of the judicial processes to ferret out the meritorious from the fraudulent in particular cases.
Borst v. Borst, 41 Wn.2d 642, 653, 251 P.2d 149 (1952).
The fourth and last reason assigned for the alleged public policy against recovery in cases of this type is that the allowance of such recovery will lead to unlimited liability *664on the part of defendants. To the extent that this is not merely a restatement of the preceding reason, it can be answered simply by saying that allowance of recovery in such cases will lead only to as much liability as the courts, applying society’s notions of just compensation, allow it to lead to. We can enunciate these standards in the future, as appropriate cases come before us; I would not now decide whether in the absence or reduced weight of various of the factors present in this case, I would conclude that an emotional injury to someone who saw or learned about an accident was not reasonably foreseeable and that therefore defendant owed no duty of due care to the plaintiff. I do not pretend or imagine that these standards will be easy to set. But I do not believe that this court should let the fear of the inability to fix boundaries prevent it from giving appropriate relief in this case. This court should not deny justice to parties in the present in order to save itself from imagined legal difficulties in the future.
I believe, therefore, that the reasons which 'are given to support the alleged public policy against recovery in such cases are without merit. Accordingly, such a dubious public policy should not override the general principles of tort liability and prevent recovery in this case.
As the commentators have suggested, the problem should be solved by the application of the principles of tort, not by the creation of exceptions to them. Legal history shows that artificial islands of exceptions, created from the fear that the legal process will not work, usually do not withstand the waves of reality and, in time, descend into oblivion.
Dillon v. Legg, 68 Cal. 2d 728, 747, 441 P.2d 912, 69 Cal. Rptr. 72 (1968); see also 2 F. Harper and F. James, The Law of Torts, supra at 1039, Throckmorton, Damages for Fright, 34 Harv. L. Rev. at 277.
Not only do I believe that the alleged public policy prohibiting recovery in cases of this type is untenable and should be declared inapplicable, I am also convinced that this case presents unique circumstances which justify and *665require the granting of relief even if the existing policy against such relief is retained.
The majority opinion quotes at length from Dillon and emphasizes certain factors which the California court listed as being relevant in determining whether defendant owed plaintiff a duty of due care — the proximity of the plaintiff to the scene of the accident, and the directness of the emotional impact which caused the shock. The majority opinion then points out that these factors or tests are less well satisfied in the instant case than they were in Dillon.
I do not deny that the factors enunciated by the California court are relevant in determining the foreseeability of harm, and hence, under the rule of Rikstad v. Holmberg, 76 Wn.2d 265, 456 P.2d 355 (1969), the duty of care owed to a plaintiff, in the case of an automobile accident where the defendant had no prior contact with the plaintiff.
To begin with, however, we are not dealing with an automobile accident in this case. The case before us involves the rape of a 5-year-old girl. The terminology of the auto accident cases has no applicability here. It means nothing to consider whether the rape of the daughter actually invaded the security of the mother. It is meaningless to talk about the zone of danger of a rape; if the mother had been present or within the zone of danger, there would have been no rape.
Nor does the existence of duty of care in this case have to be predicated only upon the foreseeability of harm to the parents of the victim. The question to be asked is not whether in raping the daughter the defendant Reed Christensen could foresee that emotional harm would be suffered by her parents (though such injury is in my opinion clearly foreseeable), but whether in recommending their son as a babysitter while having knowledge of his sexual propensities, the Christensens breached a duty which they owed to Maria Argo’s parents.
It is the rule in this state that if a plaintiff is to have his case submitted to the jury he must first produce substantial evidence to demonstrate that:
*666(1) there is a statutory or common-law rule that imposes a duty upon defendant to refrain from the complained-of conduct . . (2) the defendant’s conduct violated the duty; and (3) there was a sufficiently close, actual causal connection between defendant’s conduct and the actual damage suffered by plaintiff.
Rikstad v. Holmberg, 76 Wn.2d 265, 268, 456 P.2d 355 (1969). Foreseeability is not appropriately considered as part of the causation issue, though it is useful in determining the limits of the defendant’s duty. Id; Wells v. Vancouver, 77 Wn.2d 800, 467 P.2d 292 (1970).
In the instant case, it cannot be disputed that the causal connection requirement is satisfied. To me it is equally clear that the defendants had a duty to refrain from recommending their son as a babysitter when they knew of his bizarre sexual proclivities, and that they violated that duty when they recommended him. This duty was owed to the Schurk parents, not to the daughter, since the parents were the ones who were negligently misled. In considering the liability of the defendant parents to the plaintiff parents, the tort that was committed was not the intentional rape of 5-year-old Maria Argo by Reed Christensen, but rather the negligent misrepresentation of Reed Christensen’s abilities and reliability as a babysitter by his parents. This is not an emotional-injury-to-a-bystander case. The breached duty and the resultant liability flow directly from the defendant parents to the plaintiff parents. The principles of Dillon and Smith v. Rodene, 69 Wn.2d 482, 418 P.2d 741, (1986), do not apply to the instant case.
In this case, the contact of defendant parents was with the plaintiff parents. They recommended their son as a babysitter. They thereby breached their duty to use reasonable care. It may make sense to say, as the California Supreme Court did in Dillon, that distance of time and space limits a defendant’s duty to a plaintiff as to whom he has directly breached no duty of care. But where a defendant has dealt directly with the plaintiffs and has breached a duty owed to them, such a rule has no place. It is as if the *667father of the negligent defendant driver in Dillon had told a child’s mother that she would be safe in letting her child play in a particular field where he knew that his son often drove his car, and then tried to disclaim liability for the mother’s emotional injuries from the resulting death or injury of her daughter on the grounds that the mother was not within the zone of danger of his son’s tort.
Though under the above analysis it is not necessary to do so, it is possible to view the representations of Mrs. Christensen as to her son’s competence as a sort of contractual commitment. Under this interpretation, the “zone of danger” rule which has previously been used to determine foreseeability in automobile accident cases in this state (Smith v. Rodene, supra), is irrelevant, since the requisite breach of duty is provided, not by foreseeability of harm to the plaintiff parents, but by the assurances made to them that Reed Christensen was a competent and reliable babysitter. The law has long recognized the right of witnesses or discoverers of physical harm to recover for their emotional suffering when such suffering resulted from a contractual duty owed to them by the defendant. See Gulf, C. & S. F. Ry. v. Coopwood, 96 S.W. 102 (Tex. Civ. App. 1906); Holland v. St. Paul Mercury Ins. Co., 135 So. 2d 145 (La. App. 1961).
The majority opinion takes final refuge in the assertion that “this is not the case for consideration of a change of this long established rule . . .”1 disagree. This is a very appropriate case for reexamination of the general rule against recovery for negligently caused mental anguish and emotional distress unaccompanied by physical impact or the possibility thereof. The negligent misrepresentation of their son’s babysitting capabilities constitutes a breach of duty running directly from the defendant parents to the plaintiff parents, a breach which involves clearly foreseeable and causally related harm to the plaintiff parents. As discussed above, we can find the existence of a breached duty without over-ruling the “zone of danger,” “invasion of plaintiff’s security” rules as they are applied to auto acci*668dent cases. Though I personally doubt that it would do so, closer examination could reveal that those rules adequately serve to define the limits of foreseeability of harm, and hence the duty of care, in such cases.
The case now before us presents a clear case of breached duty and causally related physically manifested emotional harm. I would not let a misapplication of precedent, a misplaced sympathy for beleaguered defendants, or an eighteenth century view of the proficiency of medical and psychiatric evidence, prevent this court from allowing the claims of the Schurk parents to proceed to trial on the merits.
I would reverse and remand for trial.
Hamilton, C.J., and Rosellini, J., concur.
Petition for rehearing denied July 11, 1972.