SCHAUER, J.
Plaintiff Lillian Amaya appeals from a judgment of dismissal entered upon an order sustaining de[298]*298fendants’ general demurrer to her complaint1 after she had declined an opportunity to amend.
The sole issue is whether liability may be predicated on fright or nervous shock (with consequent bodily illness) induced solely by the plaintiff’s apprehension of negligently caused danger or injury to a third person. After a comprehensive review of the authorities and the several considerations underlying decision on this issue, we have concluded that the complaint does not state facts sufficient to constitute a cause of action and that the judgment should therefore be affirmed.
In the subject complaint plaintiff-appellant alleges that she is the mother of James Amaya; that at the time of the accident she was seven months pregnant, and James was 17 months of age; that on that day she “was standing near her said infant son, watching over him” and “observed the negligent conduct of the defendants ... as the said defendants’ truck was bearing down upon” James; that she shouted a warning to defendants, but they failed to stop the truck and ran over the boy; and that she “was compelled to stand helpless and watch her infant son be struck and run over by the defendants’ truck.” She then alleges that as a “direct and proximate result” of the defendants’ negligent operation of their truck she “suffered an emotional shock and great mental disturbance . . . and became violently ill and nauseous [st'c] and was hurt and injured in her health, strength and activity, sustaining injury to her body and shock and injury to her nervous system and person. ...” She further alleges on information and belief that such injuries “will result in some permanent disability,” and prays for general damages in the amount of $50,000, medical expenses, and other relief.
Plaintiff states in her opening brief that “The Court offered plaintiff’s counsel the opportunity to amend and state that the fright and shock suffered by the plaintiff was for the fear of her own safety. Plaintiff’s counsel declined[,] stating to the Court that the plaintiff suffered fright and shock as a result of being compelled to watch her infant child crushed beneath the wheels of an ice truck, and that all the fright and shock she suffered was as a result of her fear for the safety of her child, and not out of fear for her own safe[299]*299ty.” Defendants assert that they ‘1 accept this volunteer statement as a stipulation by [plaintiff],” and we treat it therefore as an amendment to the complaint.
This court has not yet ruled specifically on whether allegations such as those of the subject complaint are sufficient to constitute a cause of action.2
At the outset it is necessary to determine whether or not the “impact rule” is in force in California: i.e., in an action for personal injuries resulting from the internal operation of negligently induced fright or shock, need the plaintiff show that there was some contemporaneous physical impact upon her person? (See generally 64 A.L.R.2d 100 et seq.) If the “impact rule” prevails it is at once evident that for such reason alone the complaint in the case at bench fails to state a cause of action. Relying on Sloane v. Southern Cal. Ry. Co. (1896) 111 Cal. 668, 680 [44 P. 320]—which dealt, however, with the analytically different question of whether “nervous disturbance” caused by fright or shock “was a suffering of the body or of the mind”—it has apparently been held by the District Court of Appeal that the “impact rule” is not the law of this state. (Cook v. Maier (1939) 33 Cal.App.2d 581, 584 [5]-585 [6] [92 P.2d 434].) No California decision has been found declaring such a requirement, and we are not disposed to introduce it into our law now. We hold, accordingly, that plaintiff’s failure to allege a contemporaneous physical impact upon her person is not, of itself, fatal to her attempt to state a cause of action.
We therefore face this question: May tort liability be predicated on fright or nervous shock (with consequent bodily illness) induced solely by the plaintiff’s apprehension of negligently caused danger or injury to a third person? Prior to today that question had been raised in three eases before this court, and in each the court expressly declined to resolve it. [300]*300(Easton v. United Trade School Contr. Co. (1916) 173 Cal. 199, 202 [159 P. 597, L.RA. 1916A 394] ; Lindley v. Knowlton (1918) 179 Cal. 298, 301-302 [176 P. 440] ; Well v. Francis. J. Lewald Goal Co. (1931) 214 Cal. 182, 184 [1] [4 P.2d 532, 77 A.L.R 675].) In Easton the court found it unnecessary to rule on the matter in view of its determination that the plaintiff's fright was, at least in part, a direct consequence of physical injuries inflicted on her in a collision caused by the defendant’s negligence; hence recovery could be justified under the principle (id. at p. 203 of 173 Cal.) that “mental anguish as a direct reasonable outcome of the illegal physical injuries is always an element of damage.” In Lindley and Well the court avoided the issue by holding that in each instance the case fell within the rule—to which we adhere— that liability may be predicated upon fright and consequent illness induced by the plaintiff’s reasonable fear for her own safety, even when the plaintiff may also have feared for the safety of her children (Lindley) or of a stranger (Well).
The first California case to rule on the question now presented appears to have been Clough v. Steen (1934) 3 Cal.App.2d 392 [39 P.2d 889]. There the plaintiff and her husband and minor child were involved in an accident with the defendant’s automobile, in which the plaintiff was injured and her son was killed. In her complaint she set forth the extent of her own injuries “and then alleged that the [301]*301knowledge of the tragic death of her son ‘immediately threw her into a state of profound shock’ which caused mental and physical disorders for which she sought damages.” (Id. at p. 393.) The District Court of Appeal reversed a judgment for the plaintiff, distinguishing Lindley and holding that “no ease or rule of law has been brought to our attention which would support a recovery by plaintiff for the shock and grief, or injury consequent thereto, growing out of the knowledge of the death of her child. In the absence of such a right at common law or by statute, the plaintiff’s recovery cannot be upheld.” (Id. at p. 394 [2].)
Two federal decisions applying California law reached similar conclusions. In Minkus v. Coca-Cola Bottling Co. of California (N.D. Cal. 1942) 44 F. Supp. 10, a minor sued for personal injuries suffered by finding a decomposed mouse in a soft drink that he had partly consumed; in separate counts his parents sought damages “for nervous shock alleged to have been caused by the happening mentioned.” The court granted a motion to dismiss the latter counts, citing Clough v. Steen (1934), supra, 3 Cal.App.2d 392, and holding the matter to be controlled by “the general rule that one may not recover damages for fright or mental shock from injuries received by another when the claimant himself sustained no physical injury.”
In Maury v. United States (N.D. Cal. 1956) 139 F. Supp. 532, an action by parents for the wrongful death of their child, the plaintiffs sought to add a second cause of action based on allegations that “during the fire which destroyed the home of plaintiffs [allegedly caused by the defendants’ negligence], plaintiffs were present; that plaintiffs with full knowledge that their child was in the said burning house, suffered extreme fright, shock and mental anguish” causing the plaintiff mother to have a nervous breakdown and be confined in a hospital. The motion to add the second count was denied. The court characterized the proposed amendment as follows (id. at p. 533 [1]) : “In essence this is a claim for damages produced by shock from fear of danger or harm to another, where the danger or harm was caused by ordinary negligence, and where the claimant was not himself endangered or harmed.” The court then observed that the general rule was contrary to the plaintiffs’ theory, and held that the California cases (citing, e.g., Clough v. Steen) were in accord with that rule. In so holding the court properly distinguished (id. at p. 534 [1]) our decisions dealing with [302]*302intentional infliction of mental suffering (State Rubbish etc. Assn. v. Siliznoff (1952) 38 Cal.2d 330, 336-339 [1-4] [240 P.2d 282]) and intentional invasion of a protectible interest of an occupant of real property resulting in “mental suffering occasioned by fear for the safety of himself and his family” (Acadia, California, Ltd. v. Herbert (1960) 54 Cal.2d 328, 337 [10] [5 Cal.Rptr. 686, 353 P.2d 294], and cases there cited).
Finally, the precise question now before us was presented in Reed v. Moore (1957), supra, 156 Cal.App.2d 43, 47 [4], In that case the plaintiff, seated in front of her place of abode, was an eyewitness to a collision approximately 130 feet away, between an automobile in which her husband was riding and one driven by the defendant. In her complaint she alleged that as the “direct and proximate result” of seeing the collision caused by the defendant’s negligence she suffered mental shock, fright, and the customary miscarriage. It was stipulated that at the time of the accident the plaintiff was in fear solely for her husband’s safety and could not herself be considered in the zone of danger. A general demurrer was sustained without leave to amend. The District Court of Appeal correctly distinguished Lindley v. Knowlton (1918), supra, 179 Cal. 298, on the ground that “The case, therefore, is authority to sustain the rule that physical injury due to fright or shock as a result of fear for one’s own safety is compensable. It is not, however, authority to sustain an action for damages produced by an apprehended danger or peril to a third person.” (Reed v. Moore (1957), supra, 156 Cal.App.2d 43, 45 [1].) Relying on the general rule of nonliability set forth in the encyclopedias, on Clough v. Steen (1934), supra, 3 Cal.App.2d 392, and on other eases discussed hereinabove, the court in affirming the judgment of dismissal specifically held (p. 47 [4] of 156 Cal.App.2d) that “since the detriment to the plaintiff ensued not by reason of any fear for her own safety but by fear of an act which impinged upon the husband alone, she may not recover for her alleged injury.”
Under the existing law of California, accordingly, the trial court in the ease at bench did not err in sustaining defendants’ demurrer and entering judgment thereon.
Turning now to the general rule of nonliability relied upon in Reed v. Moore (1957), supra, it appears from extensive research that the eases, both early and recent, fully support the following statement of that rule: “As a gen[303]*303eral rule, no recovery is permitted for a mental or emotional disturbance, or for a bodily injury or illness resulting therefrom, in the absence of a contemporaneous bodily contact or independent cause of action, or an element of wilfulness, wantonness, or maliciousness, in eases in which there is no injury other than one to a third person, even though recovery would have been permitted had the wrong been directed against the plaintiff. The rule is frequently applied to mental or emotional disturbances caused by another’s danger, or sympathy for another’s suffering. It has been regarded as applicable to a mental or emotional disturbance resulting from an injury not only to a stranger, but also to a relative of the plaintiff, such as a child, sister, father, or spouse.” (52 Am. Jur., Torts, § 70, p. 417; footnotes omitted.)4
[304]*304In every jurisdiction that has ruled on this precise question—some 18 in addition to California—the decisions appear to be uniform in upholding the rule of nonliability. Most of the leading cases prior to 1950 are collected in the annotation published in 18 American Law Reports 2d 220 et seq.
. Of these the most well known is Waube v. Warrington (1935) 216 Wis. 603 [258 N.W. 497, 98 A.L.R. 394], a wrongful death action in which a husband sought damages for the death of his wife allegedly caused by nervous shock suffered when the latter, looking out of the window of her house, saw her daughter who was crossing the highway struck and killed by an automobile negligently operated by the defendants. The Wisconsin Supreme Court unanimously held that liability for such negligence does not extend to “any recovery for physical injuries sustained by one out of the range of ordinary physical peril as a result of the shock of witnessing another’s danger.” (Id. at p. 501 [2-4] of 258 N.W.) The court accordingly directed the sustaining of the defendants’ general demurrer.
Since 1950 at least a dozen more cases have so held, and we have discovered none that has held to the contrary of the general rule. (See, e.g., Berg v. Baum (Misc. 1962) 224 N.Y.S.2d 974, 976; Lahann v. Cravotta (Misc. 1962) 228 N.Y.S.2d 371, 372-373 [1-5] ; Strazza v. McKittrick (1959) 146 Conn. 714 [156 A.2d 149, 152 [3, 4] ; Lessard v. Tarca (1957) 20 Conn. Super. 295 [133 A.2d 625, 627-628 [2-4] ] ; Klassa v. Milwaukee Gas Light Co. (1956) 273 Wis. 347 [77 N.W.2d 397, 403-404 [4]]; King v. Phillips, supra, [1953] 1 Q.B. 429; Resavage v. Davies (1952), supra, 199 Md. 479 [86 A.2d 879].)5
[305]*305The modern unanimity of view upon the issue now before us is reflected in the evolution of the relevant section of the Restatement of Torts. As it originally read in the first Restatement (1934), section -3136 was qualified by a caveat, to wit: “The Institute expresses no opinion as to whether an actor whose conduct is negligent as involving an unreasonable risk of causing bodily harm to a child or spouse is liable for an illness or other bodily harm caused to the parent or spouse who witnesses the peril or harm of the child or spouse and thereby suffers anxiety or shock which is the legal cause of the parent’s or spouse’s illness or other bodily harm.” In "the current revision of the Restatement (Tent. Draft No. 5 (1960) p. 9) the just quoted caveat is stricken and the general rule of nonliability is substituted, as follows: “(2) The rule stated in Subsection (1) [i.e., former § 313, ante, fn. 6] has no application to illness or bodily harm of another, caused by emotional distress arising solely from harm or peril to a third person, unless the negligence of the actor has otherwise created an unreasonable risk of bodily harm to the other.” The reason for the change is given (ibid.) in a Note to the Institute: “The Caveat is stricken, and Subsection (2) is substituted, because of the overwhelming weight of the case law.”7 And directly in point in the case at bench is comment d to the new subsection (2), which reads: “Thus, where the actor negligently runs down and kills a child in the street, and its mother, in the immediate vicinity, witnesses the event and suffers severe emotional distress, resulting in a [306]*306heart attack or other bodily harm to her, she cannot recover for such bodily harm unless she was herself in the path of the vehicle, or was in some other manner threatened with bodily harm, otherwise than through the emotional distress at the peril to her child.” (Rest. 2d Torts, Tent. Draft-No. 5 (1960) p. 12.)
The law of California as declared in Reed v. Moore (1957), supra, 156 Cal.App.2d 43, is in complete accord with these authorities. A majority of this court voted to deny a hearing in Beed, and “Its judgment stands, therefore, as a decision of a court of last resort in this state, until and unless disapproved by this court or until change of the law by legislative action.” (Cole v. Rush (1955) 45 Cal.2d 345, 351 [4] [289 P.2d 450, 54 A.L.R.2d 1137].) “Although this court’s denial of a hearing is not to be regarded as expressing approval of the propositions of law set forth in an opinion of the District Court of Appeal or as having the same authoritative effect as an earlier decision of this court [citations], it does not follow that such a denial is without significance as to our views [citations].” (DiGenova v. State Board of Education (1962) 57 Cal.2d 167, 178 [12] [18 Cal.Rptr. 369, 367 P.2d 865].) In the five years that have elapsed since Beed—indeed, in the nearly three decades since Clough v. Steen (1934), supra, 3 Cal.App.2d 392—the Legislature has not seen fit to act in this connection, and has adopted no statute inconsistent with the common law rule of nonliability. The remaining question is whether we should nevertheless now abrogate that rule in California by judicial decision.
No guidance is to be found in such maxims as “For every wrong there is a remedy.” (Civ. Code, § 3523.) That aphorism “can obviously have no application to any but legal wrongs or those wrongs for which the law authorizes or sanctions redress.” (Finch v. Western Nat. Bank (1914) 24 Cal.App. 331, 338 [141 P. 261].) But to say that “For every wrong for which the law authorizes or sanctions redress there is a remedy” is merely to iterate a meaningless tautology, and we are not further advanced. Nor should this fact trouble us, for “If we keep in mind that justice exists only when it is effectively administered, we may be more comfortable in admitting that it is impossible to provide a remedy for every ‘wrong.’ Efforts to so provide, as well as efforts to justify the denial of a remedy while trying to uphold the principle that a remedy exists for every wrong, [307]*307cloud the image of the law with trivial distinctions and hairsplitting that confound philosophers, let alone the public.” (Proehl, Anguish of Mind (1961) 56 Nw.U.L.Rev. 477, 495.)
If the problem is to be solved on a true axiological basis, we must consider the roots of liability for negligence. First, in ruling on a general demurrer or motion for nonsuit the dispositive issue ordinarily is—as it is here— that of duty; i.e., “the existence of a duty of care owed by the alleged wrongdoer to the person injured, or to a class of which he is a member.” (Italics added.) (Richards v. Stanley (1954) 43 Cal.2d 60, 63 [3] [271 P.2d 23], quoting from Routh v. Quinn (1942) 20 Cal.2d 488, 491 [2] [127 P.2d 1,149 A.L.R. 215]; accord, Jones v. Czapkay (1960) 182 Cal.App.2d 192, 201 [6] [6 Cal.Rptr. 182]; Rest., Torts, § 281, comments b to g.) If the plaintiff does not and cannot show such a duty owed directly to her, the action must be dismissed. In terms of the ease at bench, ‘ ‘ The problem must “ be approached at the outset from the viewpoint of the duty of defendant and the right of plaintiff, and not from the viewpoint of proximate cause. The right of the mother to recover must be based, first, upon the establishment of a duty on the part of defendant so to conduct herself with respect to the child as not to subject the mother to an unreasonable risk of shock or fright, and, second, upon the recognition of a legally protected right or interest on the part of the mother to be free from shock or fright occasioned by the peril of her child.” (Waube v. Warrington (1935), supra, 216 Wis. 603 [258 N.W. 497-498 [1]] ; accord, Resavage v. Davies (1952), supra, 199 Md. 479 [86 A.2d 879, 881 [2]] ; Cote v. Litawa (1950) 96 N.H. 174 [71 A.2d 792, 794-795 [5], 18 A.L.R.2d 216] ; Angst v. Great Northern Ry. Co. (D.C. Minn. 1955), supra, 131 F.Supp. 156, 159 [1-4]; King v. Phillips, supra, [1953] 1 Q.B. 429, 433.)
The determination of this question—the existence of a duty owed to the plaintiff by the defendant—is in the first ' instance for the court, not for the jury. The classic statement is that of Minturn, J., in Morril v. Morril (1928) 104 N.J.L. 557 [142 A. 337, 339-340 [8-9], 60 A.L.R. 102] : “Hence it becomes imperative before legal liability for conceded damages can be imposed upon a defendant, for the court in the first instance to inquire and determine the character of duty which the law under the facts imposed upon the defendant as the basis of liability; for manifestly, it cannot be conceded that the jury from their inner consciousness [308]*308may evolve in every variety of tortfeasance a legal duty as the standard of liability. ’ ’ (Accord, Rest., Torts, § 453; Prosser, Torts (2d ed. 1955) p. 281; Green, The Causal Relation Issue in Negligence Law (1962) 60 Mich. L.Rev. 543, 562-563.) Much confusion has been engendered in this connection by a misplaced reliance on the “foreseeability” formula. It is not enough to say that duty has often been defined in terms of “foreseeability,” that reasonable minds might differ as to whether the injury in the present case was “foreseeable,” and hence that the duty issue was here a question of fact that should have been submitted to the jury. As this court pointed out in Richards v. Stanley (1954), supra, 43 Cal.2d 60, 66 [8], “Necessarily involved in submitting the case to the jury, however, is a preliminary determination [i.e., to be made by the court]. that, granted a foreseeable 'risk, a duty arises. . . . [T]here are many situations involving foreseeable risks where there is no duty.” (Italics added.) We then (in addition to referring to Routh v. Quinn (1942), supra, 20 Cal.2d 488) discussed Lane v. Bing (1927) 202 Cal. 590 [262 P. 318], and Goodman v. Harris (1953) 40 Cal.2d 254 [253 P.2d 447], and concluded (at p. 67 [8] of 43 Cal.2d), “Although in both of the foregoing situations it would be difficult to say that reasonable minds could not differ as to whether or not a duty should be imposed, the question was one of law for the court, and not for the jury, to decide.” (Italics added.)
. There are sound reasons for the established rule that the determination of the existence and scope of the defendant’s duty to the plaintiff is primarily a question for the court and does not depend only on “foreseeability.” First, it will be observed that the negligence issue, i.e., the violation of duty, is submitted to the jury under instructions couched in terms of how an ordinarily prudent person would view the “foreseeability of risk” created by the defendant's conduct; yet if the issue of the existence of duty is also submitted to the jury under a “foreseeability” formula, the jury will in effect be asked to determine two distinct issues in the case by means of the same “test.” And if, as sometimes occurs, the jury are also instructed that the defendant’s conduct is not the “proximate cause” of events that he could not reasonably have anticipated, then truly “analysis has played itself false, so that a case is seemingly to be subjected thrice to the ponderous process of the ‘foreseeability’ formula” (Green, The Duty Problem in Negligence Cases I [309]*309(1928) 28 Colum. L.Rev. 1014, 1029). Confronted with repetitions of the same phrase as constituting a rule to guide them in resolving various issues, the jury may well become bewildered, take refuge in the terms of their own experience, and simply deliver a verdict according to their liking. It has been accurately said that “Foreseeability of risk . . . carries only an illusion of certainty in defining the consequences for which the defendant will be liable.” (Prosser, Palsgraf Revisited (1953) 52 Mich.L.Rev. 1, 19.) For meaningful assistance the jurors need a definition, so far as relevant, of the ultimate factual conclusions upon which, if found, the law either gives the plaintiff a right to recover or the defendant a right of exoneration.
This leads to consideration' of the second reason for the above mentioned established rule governing determination of the existence and scope of the defendant’s duty to the plaintiff. This second reason touches on the fundamental responsibility of the court to declare the law. There is a legal duty on any given set of facts only if the court or the Legislature says there is a duty. “Duty is only a word with which we state our conclusion that there is or is not to be liability; it necessarily begs the essential question [italics deleted]. . . . The word serves a useful purpose in directing attention to the obligation to be imposed' upon the defendant, rather than the causal sequence of events; beyond that it serves none.” (Klassa v. Milwaukee Gas Light Co. (1956), supra, 273 Wis. 347 [77 N.W.2d 397, 402 [1]], quoting from Prosser, Palsgraf Revisited (1953) 52 Mich.L.Rev. 1, 15.) The inquiry, then, should be concentrated on the various factors which are incorporated in the court’s conclusion that on the relevant set of facts there is or is not a “duty.” While an all-inclusive .enumeration is manifestly impossible, we have on occasion undertaken to list at least some of these factors. Thus, in a recent case raising the question of liability to a beneficiary for a negligently drawn will (Biakanja v. Irving (1958) 49 Cal.2d 647, 650 [1] [320 P.2d 16, 65 A.L.R.2d 1358]), we said: “The determination whether in a specific ease the defendant will be held liable to a third person not in privity is a matter of policy and involves the balancing of various factors, among which are the extent to which the transaction was intended to affect the plaintiff, the foreseeability of harm to him, the degree of certainty that the plaintiff suffered injury, the closeness of the connection between [310]*310the defendant’s conduct and the injury suffered, the moral blame attached to the defendant’s conduct, and the policy of preventing future harm.” (Accord, Lucas v. Hamm (1961) 56 Cal.2d 583, 588 [1] [15 Cal.Rptr. 821, 364 P.2d 685].)
It bears emphasis that “foreseeability of harm” is but one of the half dozen relevant factors mentioned in the just quoted passage from Biakanja. Nor is it the most important; indeed, in all save the most obvious of cases a harm is “foreseeable” only if, in the final analysis, a court or jury says that it is. As we have seen, the virtually unanimous opinion of the courts that have ruled on the question is that the type of harm allegedly suffered by plaintiff in the case at bench is not reasonably foreseeable. Some nonjudicial commentators have agreed with the courts in this respect, while others have disagreed; but the question is largely semantic, and any such debate tends to sterility. Moreover, as noted hereinabove, even “granted a foreseeable risk,” the court must still determine whether or not from the given factual base “a duty arises.” (Richards v. Stanley (1954), supra, 43 Cal.2d 60, 66 [8].)8
Turning to the factors that are to be weighed in the balance we consider, on the one hand, plaintiff’s undoubted interest in freedom from invasion of her bodily security, and on the other, such factors as the following:
The Administrative Factor. Justice, we observed above, exists only when it can be effectively administered. It was no doubt this fundamental precept that was in the mind of the court in Waube v. Warrington (1935), supra, 216 Wis. 603 [258 N.W. 497, 501 [2-4]], when it stated that, in circumstances such as those here shown, to impose liability would [311]*311“open the way to fraudulent claims, and enter a field that has no sensible or just stopping point.” It has become almost trite for legal writers to deprecate such reasoning; and we have not hesitated to reject it when we deemed that the problems of proof could be solved and the scope of liability could be delineated. (See, e.g., State Rubbish etc. Assn. v. Siliznoff (1952), supra, 38 Cal.2d 330, 338 [2-3].) But is this necessarily true of the case at bench?9 To begin with the problem of proof, it is to be observed that in Siliznoff we were dealing with the intentional infliction of fright; we reasoned (id. at p. 338 [2] of 38 Cal.2d) that “The jury is ordinarily in a better position, however, to determine whether outrageous conduct results in mental distress than whether that distress in turn results in physical injury. From their own experience jurors are aware of the extent and character of the disagreeable emotions that may result from the defendant’s conduct, but a difficult medical question is presented when it must be determined if emotional distress resulted in physical injury. (See Smith, Relation of Emotions to Injury and Disease, 30 Va.L.Rev. 193, 303-306.) ” Here that “difficult medical question ’ ’ cannot be so easily avoided. In the cited article (at pp. 281-285 of 30 Va.L.Rev.) Dr. Smith reports on his study of 301 eases involving injuries allegedly caused by psychic stimuli, and concludes (1) that “a majority of persons claiming injury from psychic causes possessed sub-normal resistance to such stimuli”; (2) that “In only 55 of the 301 J eases surveyed could we say actual causation was proved by a preponderance of substantial and credible evidence”; and (3) that hence “The skeptical courts were . . . correct in doubting whether adequate criteria of proof existed in this field to make administration of a remedy feasible. Law, in a commendable desire to be forward looking, outran scientific standards. Taking all cases decided between 1850 and 1944 [the date of publication of the subject article], the net balance of justice would have been greater had all courts denied damages for injury imputed to psychic stimuli alone.” (Id. at p. 285.)
Granted that simply because “naivete about the problem of [312]*312proof has caused injustice in times past does not necessarily settle the matter for the future” (ibid.), we are left with the question of whether in this area of inquiry where emotions play so large a role the law has now become sufficiently responsive to scientific reality to redress the “net balance of justice.” The question is a disturbing one, and cannot be answered merely by invoking the rule that a conflict of expert testimony is for the jury. It must be recognized “that ‘conflict of expert testimony’ at trial is no guarantee that the evidence is scientifically sufficient to support a verdict.” (Id. at p .284.) Yet the parade of expert medical witnesses continues, ushered in by both plaintiff and defendant, and the juries are daily called upon to “resolve” the disputes thus engendered. This may be difficult enough when traumatic injuries to the body are involved; but as doctors well know, the “resolution” of such conflicts often borders on fancy when the causation of alleged psychoneural disorders is at issue. There are indications, moreover, that in this area the law continues to cling to cherished preconceptions of the layman, with bland disregard for the development of modern medical knowledge. Much timeliness remains in Dr. Smith’s warning (id. at p. 212 of 30 Va.L.Rev.) that “eagerness to be progressive may cause extravagant credulity and injury to ' scientific standards of proof.” Extravagant credulity,- of course, means ultimate injustice.
, Another—and no less important—administrative factor to be .weighed is the problem of setting some limits to such liability for fright or shock allegedly caused by the apprehension of danger or injury not to the plaintiff but to a third person. Here, at least, the legal writers are generally agreed that the problem is a real one. Professor Prosser suggests the following limitations on liability (Prosser, Torts (2d ed. 1955) p. 182): First, “It is clear that the injury threatened or inflicted upon the third person must be a serious one, of a nature to cause severe shock to the plaintiff, and that the shock must result in actual physical harm. ’ ’ But what if the plaintiff was honestly mistaken in believing the third person to be in danger or to be' seriously injured ? And as we are dealing here with a negligently caused danger or injury to the third person, what, if the latter was contributively negligent? Or assumed the risk involved? Second, “The action'might well be confined to members of the immediate family, or perhaps to husband, wife, parent or child, to the exclusion of bystanders, and remote relatives.” But what if the third per[313]*313son was the plaintiff’s beloved niece or nephew, grandparent, fiancé, or lifelong friend, as dear to the plaintiff as her more immediate family ? Third,‘ the plaintiff must be present at the time of the accident, or at least the shock must be fairly contemporaneous with it, rather than follow at a later date.” But how soon is “fairly contemporaneous”? What is the magic in the plaintiff's being “present”? Is the shock any less immediate if the mother does not know of the accident until the injured child is brought home? And what if the plaintiff is present at the scene but is nevertheless unaware of the danger or injury to the third person until shortly after the accident has occurred (see, e.g., Kelly v. Fretz (1937) 19 Cal.App.2d 356 [65 P.2d 914]) ?
As Professor Prosser concedes, such limitations are quite arbitrary. If it were only a matter of logic we might well agree with Atkin, L.J., who found it ‘ difficult to explain why the duty was confined to the case of parent or guardian and child, and did not extend to other relations of life also involving intimate associations; and why it did not eventually extend to bystanders.” (Hambrook v. Stokes Bros., supra, [1925] 1 K.B. 141, 158-159.) But compelling moral and socio-economic reasons, hereinafter discussed, require that a / negligent defendant’s liability have some stopping point. None has yet been proposed that would be fair to all parties concerned, and the failings of the above quoted limitations suggest that the quest may be an inherently fruitless one. There is an appealing simplicity in the view that difficulties in delimiting an area of liability do not justify a refusal to enter that area. But this court does not act in a vacuum; we cannot fashion a rule for the case at bench without reflecting on the fact that there will be other such cases, other plaintiffs. When, as here, a wholly new type of liability is envisioned, our responsibility extends far beyond the particular plaintiff before us, and touches society at large.
The Socio-Economic and Moral Factors. As just observed, there must be some stopping point to the liability of the negligent defendant. “It is still unthinkable that any one shall be liable to the end of time for all of the results that follow, in endless sequence from his single act. Causation cannot be the answer; in a very real sense the consequences of an act go forward to eternity, and back to the beginning of the world.” (Prosser, Palsgraf Revisited (1953) 52 Mich.L.Rev. 1, 24.) There are two principal reasons why this is “unthinkable,”
[314]*314• First, to the extent that the law intervenes in any area of human activity and declares that for certain consequences of that activity the actor shall be held civilly liable in damages, both the individual actor and society as a whole feel the effects of the restraint—a psychological effect in the form of a lessening of incentive, and an economic effect in the form of the cost of insurance necessary to enable the activity to continue. Yet it is recognized that no activity could survive an 'unlimited progression of such effects. Accordingly, when the ¡general social utility of an activity is deemed to outweigh the particular interests with which it may clash, important policy reasons dictate that some limits be set to liability for Cits consequences. How do these considerations affect our problem? The law, both in California and in the many other jurisdictions which have passed on the question, now provides, as has been shown, that an actor who is merely negligent is not liable to one who claims injury through fright or shock induced by conduct directed not to the latter but to a third person. Thus, in cases where the defendant’s conduct involved negligent driving of a motor vehicle the courts eonyblude that to extend liability to spectators who were not themselves in danger “would, in our opinion, place an unreasonable burden upon users of highways.” (Cote v. Litawa (1950), supra, 96 N.H. 174 [71 A.2d 792, 795 [6], 18 A.L.R.2d 216] ; accord, Resavage v. Davies (1952), supra, 199 Md. 479 [86 A.2d 879, 883 [3-5] ] ; Waube v. Warrington (1935), supra, 216 Wis. 603 [258 N.W. 497, 501 [2-4]].) As the industrial society in which we live becomes still more complex and the use of the streets and highways and airways increases, a certain percentage of accidents therefrom appears to become statistically inevitable. There will be losses, and our present system of insurance attempts to compensate for them, and, of course, to spread the cost of compensation over those who do not, as well as those who do, cause such losses. But could that system—imperfect at best—adequately and fairly absorb the far-reaching extension of liability that would follow from judicial abrogation of the rule now before us? And what of the many other activities of everyday life that are either uninsurable or customarily uninsured, yet may well give rise to the type of “spectator injury” here alleged? We conclude, rather, that the social utility of such activities outweighs the somewhat speculative interest of individuals to be free from the risk of the type of injury here alleged.
The second reason for seeking a stopping point to the negli[315]*315gent defendant’s liability is a related one. As long as our system of compensation is based on the concept of fault, we must also weigh “the moral blame attached to the*defendant’s conduct” (Biakanja v. Irving (1958), supra, 49 Cal.2d 647, 650 [1]). Here is felt the difference between the social importance of conduct that negligently causes harm and conduct that is intended to do so. It is often said that in the latter case the defendant will be held liable for a broader range of consequences because, as the consequences are intended, they are the more “foreseeable.” But in many intentional tort eases the defendant has been held liable under this reasoning for consequences far beyond those which he actually intended. (See Bauer, The Degree of Moral Fault as Affecting Defendant’s Liability (1933) 81 U.Pa.L.Rev. 586, 588-592.) It follows that, once more, “foreseeability” is not the real answer. Bather, the increased liability imposed on an intentional wrongdoer appears to reflect the psychological fact that solicitude for the interests of the actor weighs less in the balance as his moral guilt increases and the social utility of his conduct diminishes.
There is good sense in the conclusion of the Wisconsin court that “the liability imposed by such a doctrine is wholly out of proportion to the culpability of the negligent tort-feasor.” (Waube v. Warrington (1953), supra, 216 Wis. 603 [258 N.W. 497, 501 [2-4]].) It begs the question .to argue that “If the loss is out of all proportion to the defendant’s fault, it can be no less out of proportion to the plaintiff’s innocence.” (Prosser, Palsgraf Revisited (1953) 52 Mich.L.Rev. 1, 17.) That obvious truism could be urged by every person who might adversely feel some lingering effect of the defendant’s conduct, and we would then be thrown back into the fantastic realm of infinite liability.
Having weighed each of the foregoing factors in the balance, we hold that the complaint fails to state facts sufficient to constitute a cause of action.
The judgment is affirmed.
Traynor, J., McComb, J., and White, J.,* concurred.
Retired Justice of the Supreme Court sitting pro tempore under assignment by the Chairman of the Judicial Council.