The opinion of the Court was delivered by

This case raises the issue of whether a social host who enables an adult guest at his home to become drunk is liable to the victim of an automobile accident caused by the drunken *541driving of the guest. Here the host served liquor to the guest beyond the point at which the guest was visibly intoxicated. We hold the host may be liable under the circumstances of th?s case.

At the trial level, the case was disposed of, insofar as the issue before us is concerned, by summary judgment in favor of the social host. The record on which the summary judgment was based (pleadings, depositions, and certifications) discloses that defendant Donald Gwinnell, after driving defendant Joseph Zak home, spent an hour or two at Zak’s home before leaving to return to his own home. During that time, according to.Gwinnell, Zak, and.Zak’s wife, Gwinnell consumed two or three drinks of scotch on the rocks. Zak accompanied Gwinnell outside to his car, chatted with him, and watched as Gwinnell then drove off to go home. About twenty-five minutes later Zak telephoned Gwinnell’s home to make sure Gwinnell had arrived there safely. The phone was answered by Mrs. Gwinnell, who advised Zak that Gwinnell had been involved in a head-on collision. The collision was with an automobile operated by plaintiff, Marie Kelly, who was seriously injured as a result.

After the accident Gwinnell was subjected to a blood test, which indicated a blood alcohol concentration of 0.286 percent.1 Kelly’s expert concluded from that reading that Gwinnell had consumed not two or three scotches but the equivalent of thirteen drinks; that while at Zak’s home Gwinnell must have been showing unmistakable signs of intoxication; and that in fact he was severely intoxicated while at Zak’s residence and at the time of the accident.

Kelly sued Gwinnell and his employer; those defendants sued the Zaks in a third party action; and thereafter plaintiff amend*542ed her complaint to include Mr. and Mrs. Zak as direct defendants. The Zaks moved for summary judgment, contending that as a matter of law a host is not liable for the negligence of an adult social guest who has become intoxicated while at the host’s home. The trial court granted the motion on that basis. While this disposition was interlocutory (plaintiff’s claim against Gwinnell and his employer still remaining to be disposed of), the trial court entered final judgment in favor of Zak pursuant to Rule 4:42-2 apparently in order to allow an immediate appeal. Pressler, Current N.J. Court Rules, Comment R.4:42-2. The Appellate Division affirmed, Kelly v. Gwinnell, 190 N.J.Super. 320 (1983). It noted, correctly, that New Jersey has no Dram Shop Act imposing liability on the provider of alcoholic beverages, and that while our decisional law had imposed such liability on licensees, common-law liability had been extended to a social host only where the guest was a minor. Id. at 322-23. (But see Figuly v. Knoll, 185 N.J.Super. 477 (Law Div. 1982).) It explicitly declined to expand that liability where, as here, the social guest was an adult. Id. at 325-26.

The Appellate Division's determination was based on the apparent absence of decisions in this country imposing such liability (except for those that were promptly overruled by the Legislature).2 Id. at 324-25. The absence of such determina*543tions is said to reflect a broad consensus that the imposition of liability arising from these social relations is unwise. Certainly this immunization of hosts is not the inevitable result of the law of negligence, for conventional negligence analysis points strongly in exactly the opposite direction. “Negligence is tested by whether the reasonably prudent person at the time and place- should recognize and foresee an unreasonable risk or likelihood of harm or danger to others.” Rappaport v. Nichols, 31 N.J. 188, 201 (1959); see also Butler v. Acme Mkts., Inc., 89 N.J. 270 (1982) (supermarket operator liable for failure to provide shoppers with parking lot security). When negligent conduct creates such a risk, setting off foreseeable consequences that lead to plaintiffs injury, the conduct is deemed the proximate cause of the injury. “[A] tortfeasor is generally held answerable for the injuries which result in the ordinary course of events from his negligence and it is generally sufficient if his negligent conduct was a substantial factor in bringing about the injuries.” Rappaport, supra, 31 N.J. at 203; see Ettin v. Ava Truck Leasing Inc., 53 N.J. 463, 483 (1969) (parking tractor-trailer across street is substantial factor in cause of accident when truck with failed brakes collides into trailer).

Under the facts here defendant provided his guest with liquor, knowing that thereafter the guest would have to drive in order to get home. Viewing the facts most favorably to plaintiff (as we must, since the complaint was dismissed, on a motion for summary judgment), one could reasonably conclude that the Zaks must have known that their provision of liquor was causing Gwinnell to become drunk, yet they continued to serve him even after he was visibly intoxicated. By the time he *544left, Gwinnell was in fact severely intoxicated. A reasonable person in Zak’s position could foresee quite clearly that this continued provision of alcohol to Gwinnell was making it more and more likely that Gwinnell would not be able to operate his car carefully. Zak could foresee that unless he stopped providing drinks to Gwinnell, Gwinnell was likely to injure someone as a result of the negligent operation of his car. The usual elements of a cause of action for negligence are clearly present: an action by defendant creating an unreasonable risk of harm to plaintiff, a risk that was clearly foreseeable, and a risk that resulted in an injury equally foreseeable. Under those circumstances the only question remaining is whether a duty exists to prevent such risk or, realistically, whether this Court should impose such a duty.

In most cases the justice of imposing such a duty is so clear that the cause of action in negligence is assumed to exist simply on the basis of the actor’s creation of an unreasonable risk of foreseeable harm resulting in injury. In fact, however, more is needed, “more” being the value judgment, based on an analysis of public policy, that the actor owed the injured party a duty of reasonable care. Palsgraf v. Long Island R.R. Co., 248 N.Y. 339, 162 N.E. 99 (1928). In Goldberg v. Housing Auth. of Newark, 38 N.J. 578, 583 (1962), this Court explained that “whether a duty exists is ultimately a question of fairness. The inquiry involves a weighing of the relationship of the parties, the nature of the risk, and the public interest in the proposed solution.” See also Portee v. Jaffee, 84 N.J. 88, 101 (1980) (whether liability for negligently inflicted emotional harm should be expanded depends “ultimately” on balancing of conflicting interests involved).

When the court determines that a duty exists and liability will be extended, it draws judicial lines based on fairness and policy. In a society where thousands of deaths are caused each *545year by drunken drivers,3 where the damage caused by such deaths is regarded increasingly as intolerable, where liquor licensees are prohibited from serving intoxicated adults, and where long-standing criminal sanctions against drunken driving have recently been significantly strengthened to the point where the Governor notes that they are regarded as the toughest in the nation, see Governor’s Annual Message to the N.J. State Legislature, Jan. 10, 1984, the imposition of such a duty by the judiciary seems both fair and fully in accord with the State’s policy. Unlike those cases in which the definition of desirable policy is the subject of intense controversy, here the imposition of a duty is both consistent with and supportive of a social goal — the reduction of drunken driving — that is practically unanimously accepted by society.

While the imposition of a duty here would go beyond our prior decisions, those decisions not only point clearly in that direction but do so despite the presence of social considerations similar to those involved in this case — considerations that are claimed to invest the host with immunity. In our first case on the subject, Rappaport, supra, 31 N.J. 188, we held a licensee liable for the consequences of a customer’s negligent operation of his automobile. The customer was a minor who had become intoxicated as a result of the consumption of liquor at various premises including the licensee’s. While observing that a stan*546dard of conduct was contained in the statute prohibiting licensees from serving liquor to minors and in the regulation further prohibiting service to any person actually or apparently intoxicated, our decision that the licensee owed a duty to members of the general public was based on principles of common-law negligence.4

We later made it clear that the licensee’s duty is owed to the customer as well, by holding in Soronen v. Olde Milford Inn, Inc., 46 N.J. 582 (1966), that the licensee who served liquor to an intoxicated customer was liable to that customer for the death that resulted when the customer fell in the licensed premises while leaving the bar. While the situation of a licensee differs in some respects from that of a social host, some of the same underlying considerations relied on here in disputing liability are present in both: the notion that the real fault is that of the drunk, not the licensee, especially where the drinker is an adult (as he was in Soronen); and the belief — not as strong when applied to licensed premises as when applied to one’s home — that when people get together for a friendly drink or more, the social relationships should not be intruded upon by possibilities of litigation.

The Appellate Division moved our decisional law one step further, a significant step, when it ruled in Linn v. Rand, 140 N.J.Super. 212 (1976), that a social host who serves liquor to a visibly intoxicated minor, knowing the minor will thereafter drive, may be held liable for the injuries inflicted on a third party as a result of the subsequent drunken driving of the minor. There, practically all of the considerations urged here against liability were present: it was a social setting at someone’s home, not at a tavern; the one who provided the liquor to the intoxicated minor was a host, not a licensee; and all of the notions of fault and causation pinning sole responsibility on the *547drinker were present. The only difference was that the guest was a minor — but whether obviously so or whether known to the host is not disclosed in the opinion.5

In Rappaport, we explicitly noted that the matter did not involve any claim against “persons not engaged in the liquor business.” 31 N.J. at 205. We now approve Linn with its extension of this liability to social hosts. In expanding liability, Linn followed the rationale of Rappaport that the duty involved is a common law duty, not one arising from the statute and regulation prohibiting sales of liquor to a minor, neither of which applies to a social host.6 Cf. Congini v. Portersville Valve Co., — Pa. —, —, 470 A.2d 515, 517-18 (1983) (in which the Pennsylvania Supreme Court relied exclusively on statutes criminalizing the provision of alcohol to minors as the basis for extending liability to a social host). The fair implication of Rappaport and Soronen, that the duty exists independent of the statutory prohibition, was thus made explicit in Linn. As the court there noted: “It makes little sense to say that the licensee in Rappaport is under a duty to exercise care, but give immunity to a social host who may be guilty of the same wrongful conduct merely because he is unlicensed.” 140 N.J.Super. at 217.7

The argument is made that the rule imposing liability on licensees is justified because licensees, unlike social hosts, *548derive a profit from serving liquor. We reject this analysis of the liability’s foundation and emphasize that the liability proceeds from the duty of care that accompanies control of the liquor supply. Whatever the motive behind making alcohol available to those who will subsequently drive, the provider has a duty to the public not to create foreseeable, unreasonable risks by this activity.

We therefore hold that a host who serves liquor to an adult social guest, knowing both that the guest is intoxicated and will thereafter be operating a motor vehicle, is liable for injuries inflicted on a third party as a result of the negligent operation of a motor vehicle by the adult guest when such negligence is caused by the intoxication. We impose this duty on the host to the third party because we believe that the policy considerations served by its imposition far outweigh those asserted in opposition. While we recognize the concern that our ruling will interfere with accepted standards of social behavior; will intrude on and somewhat diminish the enjoyment, relaxation, and camaraderie that accompany social gatherings at which alcohol is served; and that such gatherings and social relationships are not simply tangential benefits of a civilized society but are regarded by many as important, we believe that the added assurance of just compensation to the victims of drunken driving as well as the added deterrent effect of the rule on such driving outweigh the importance of those other values. Indeed, we believe that given society’s extreme concern about drunken driving, any change in social behavior resulting from the rule will be regarded ultimately as neutral at the very least, and not as a change for the worse; but that in any event if there be a loss, it is well worth the gain.8

*549The liability we impose here is analogous to that traditionally imposed on owners of vehicles who lend their cars to persons they know to be intoxicated. Knight v. Gosselin, 124 Cal.App. 290, 12 P.2d 454 (Dist.Ct.App. 1932); Harris v. Smith, 119 Ga.App. 306, 167 S.E.2d 198 (Ct.App. 1969); Pennington v. Davis-Child Motor Co., 143 Kan. 753, 57 P.2d 428 (1936); Deck v. Sherlock, 162 Neb. 86, 75 N.W.2d 99 (1956); Mitchell v. Churches, 119 Wash. 547, 206 P. 6 (1922). If, by lending a car to a drunk, a host becomes liable to third parties injured by the drunken driver’s negligence, the same liability should extend to a host who furnishes liquor to a visibly drunken guest who he knows will thereafter drive away.

Some fear has been expressed that the extent of the potential liability may be disproportionate to the fault of the host. A social judgment is therein implied to the effect that society does not regard as particularly serious the host’s actions in causing his guests to become drunk, even though he knows they will thereafter be driving their cars. We seriously question that value judgment; indeed, we do not believe that the liability is disproportionate when the host’s actions, so relatively easily *550corrected, may result in serious injury or death. The other aspect of this argument is that the host’s insurance protection will be insufficient. While acknowledging that homeowners’ insurance will cover such liability,9 this argument notes the risk that both the host and spouse will be jointly liable. The point made is not that the level of insurance will be lower in relation to the injuries than in the case of other torts, but rather that the joint liability of the spouses may result in the loss of their home and other property to the extent that the policy limits are inadequate.10 If only one spouse were liable, then even though the policy limits did not cover the liability, the couple need not lose their home because the creditor might not reach the interest of the spouse who was not liable. Newman v. Chase, 70 N.J. 254, 266 (1976); King v. Greene, 30 N.J. 395 (1959); ESB, Inc. v. Fisher, 185 N.J.Super. 373 (Ch.Div. 1982). We observe, however, that it is common for both spouses to be liable in automobile accident cases. It may be that some special form of insurance could be designed to protect the spouses’ equity in their homes in cases such as this one. In any event, it is not clear that the loss of a home by spouses who, by *551definition, have negligently caused the injury, is disproportionate to the loss of life of one who is totally innocent of any wrongdoing.

Given the lack of precedent anywhere else in the country, however, we believe it would be unfair to impose this liability retroactively. Merenoff v. Merenoff, 76 N.J. 535 (1978); Darrow v. Hanover Twp., 58 N.J. 410 (1971); Willis v. Department of Conservation & Economic Dev., 55 N.J. 534 (1970). Homeowners who are social hosts may desire to increase their policy limits; apartment dwellers may want to obtain liability insurance of this kind where perhaps they now have none. The imposition of retroactive liability could be considered unexpected and its imposition unfair. We therefore have determined that the liability imposed by this case on social hosts shall be prospective, applicable only to events that occur after the date of this decision. We will, however, apply the doctrine to the parties before us on the usual theory that to do otherwise would not only deprive the plaintiff of any benefit resulting from her own efforts but would also make it less likely that, in the future, individuals will be willing to claim rights, not yet established, that they believe are just.

The goal we seek to achieve here is the fair compensation of victims who are injured as a result of drunken driving. The imposition of the duty certainly will make such fair compensation more likely. While the rule in this case will tend also to deter drunken driving, there is no assurance that it will have any significant effect. The lack of such assurance has not prevented us in the past from imposing liability on licensees. Indeed, it has been only recently that the sanction of the criminal law was credited with having some significant impact on drunken driving.11 We need not, however, condition the *552imposition of a duty on scientific proof that it will result in the behavior that is one of its goals. No one has suggested that the common-law duty to drive carefully should be abolished because it has apparently not diminished the mayhem that occurs regularly on our highways. We believe the rule will make it more likely that hosts will take greater care in serving alcoholic beverages at social gatherings so as to avoid not only the moral responsibility but the economic liability that would occur if the guest were to injure someone as a result of his drunken driving.

We do not agree that the issue addressed in this case is appropriate only for legislative resolution. Determinations of the scope of duty in negligence cases has traditionally been a function of the judiciary. The history of the cases cited above evidences a continuing judicial involvement in these matters. Without the benefit of any Dram Shop Act imposing liability on licensees, legislation that is quite common in other states, this Court determined that such liability nevertheless existed.12 We *553did so in 1959 and have continued to expand that concept since then. We know of no legislative activity during that entire period from 1959 to date suggesting that our involvement in these matters was deemed inappropriate; even after the judiciary expanded this liability to include social hosts in its decision in Linn, there , was no adverse reaction on the part of the Legislature. In fact, the Legislature’s passage of S. 1054, imposing criminal liability on anyone who purposely or knowingly serves alcoholic beverages to underage persons, indicates that body’s approval of the position taken eight years earlier in Linn. The subject matter is not abstruse, and it can safely be assumed that the Legislature is in fact aware of our decisions in this area. Absent such adverse reaction, we assume that our decisions are found to be consonant with the strong legislative policy against drunkén driving.

The dissent relies on two related grounds in concluding this matter should be resolved by legislation: the superior knowledge of the legislature obtained through hearings and other means enabling it better to balance the interests involved and to devise an appropriate remedy, and the ruling’s potential “extraordinary effects on the average citizen.” Post at 561. Many of the cases cited in support of this view, however, are from jurisdictions in which a Dram Shop Act was in effect and are therefore clearly distinguishable. Kowal v. Hofher, 181 Conn. 355, 436 A.2d 1 (1980); Miller v. Moran, 96 Ill.App.3d 596, 52 Ill.Dec. 183, 421 N.E.2d 1046 (1981); Behnke v. Pierson, 21 Mich.App. 219, 175 N.W.2d 303 (Ct.App. 1970); Cole v. City of Spring Lake Park, 314 N.W. 2d 836 (Minn. 1982); Schirmer v. Yost, 60 A.D.2d 789, 400 N.Y.S.2d 655 (App.Div. 1977); Edgar v. Kajet, 84 Msc.2d 100, 375 N Y.S.2d 548 (Sup.Ct. 1975), aff’d, 55 A.D.2d 597, 389 N.Y.S.2d 631 (App.Div. 1976).

*554Whether mentioned or not in these opinions, the very existence of a Dram Shop Act constitutes a substantial argument against expansion of the legislatively-mandated liability. Very simply, when the Legislature has spoken so specifically on the subject and has chosen to make only licensees liable, arguably the Legislature did not intend to impose the same liability on hosts. See, e.g., Edgar, 375 N.Y.S.2d at 551-52. We note, furthermore, that in several of the cited cases, post at 561-563, the courts’ denial of relief was not solely the result of deference to the legislature; the courts also treated the question (of whether the host was liable) as one appropriate for judicial determination. Kowal v. Hofher, 181 Conn. at 357, 436 A.2d at 2; Behnke v. Pierson, 21 Mich.App. at 221, 175 N.W.2d at 304; Schirmer v. Yost, 60 A.D.2d at 789, 400 N.Y.S.2d at 656.13

In only four of the jurisdictions cited in the dissent did the courts rule, despite the absence of a Dram Shop Act, that a host should not be liable. Cartwright v. Hyatt Corp., 460 F.Supp. 80, 82 (D.D.C. 1978); Runge v. Watts, 180 Mont. 91, 589 P.2d 145 (1979); Klein v. Raysinger, — Pa. —, 470 A.2d 507 (1983); Manning v. Andy, 454 Pa. 237, 310 A.2d 75 (1973); Halvorson v. Birchfield Boiler, Inc., 76 Wash.2d 759, 458 P.2d 897 (1969).

Whether our ruling will have such an “extraordinary” impact on “the average citizen” in his or her social and business relations (presumably the premise for the conclusion that judicial action is inappropriate) depends to some extent on an initial evaluation of the matter. We suspect some of the extraordinary change is already taking place, that it is not unusual today for hosts to monitor their guests’ drinking to some extent. *555Furthermore, the characterization of the change as one demanding prior legislative study and warranting action only after such, implies that its effects on balance may be seriously adverse. Given our firm belief that insurance is available, that compensation of innocent victims is desirable, and that the added deterrence against drunken driving is salutary, we do not perceive the potential revision of cocktail-party customs as constituting a sufficient threat to social well-being to warrant staying our hand. Obviously the Legislature may disagree.

This Court has decided many significant issues without any prior legislative study. In any event, if the Legislature differs with us on issues of this kind, it has a clear remedy. See, e.g., Van Horn v. Blanchard Co., 88 N.J. 91 (1981) (holding that under Comparative Negligence Act, a plaintiff could recover only from those defendants that were more negligent than was the plaintiff); N.J.S.A. 2A:15-5.1 as amended by L. 1982, c. 191 § 1 eff. Dec. 6, 1982 (under which a plaintiff may recover from all defendants if plaintiffs negligence is less than or equal to the combined negligence of all defendants); Willis v. Department of Conservation and Economic Dev., 55 N.J. 534 (1970) (abolishing the State’s sovereign immunity from tort claims), N.J.S.A. 59:1-1 et seq., L. 1972, c. 45 (reestablishing and defining immunity for all New Jersey governmental bodies); Dalton v. St. Luke’s Catholic Church, 27 N.J. 22 (1958), Collopy v. Newark Eye & Ear Infirmary, 27 N.J. 29 (1958), Benton v. Y.M.C.A., 27 N.J. 67 (1958) (abolishing charitable immunity), N.J.S.A. 2A:53A-7, L. 1959, c. 90 (reestablishing charitable immunity); cf. Immer v. Risko, 56 N.J. 482 (1970) (abolishing interspousal immunity in automobile negligence cases); France v. A.P.A. Transport Corp., 56 N.J. 500 (1970) (abolishing parent-child immunity in automobile negligence cases) (no subsequent legislative action on issue of familial immunity).

We are satisfied that our decision today is well within the competence of the judiciary. Defining the scope of tort liability has traditionally been accepted as the responsibility of the courts. Indeed, given the courts’ prior involvement in these *556matters, our decision today is hardly the radical change implied by the dissent but, while significant, is rather a fairly predictable expansion of liability in this area.14

It should be noted that the difficulties posited by the dissent as to the likely consequence of this decision are purely hypothetical. Given the facts before us, we decide only that where the social host directly serves the guest and continues to do so even after the guest is visibly intoxicated, knowing that the guest will soon be driving home, the social host may be liable for the consequences of the resulting drunken driving. We are not faced with a party where many guests congregate, nor with guests serving each other, nor with a host busily occupied with other responsibilities and therefore unable to attend to the matter of serving liquor, nor with a drunken host. Post at 566-567. We will face those situations when and if they come before us, we hope with sufficient reason and perception so as to balance, if necessary and if legitimate, the societal interests alleged to be inconsistent with the public policy considerations that are at the heart of today’s decision. The fears expressed by the dissent concerning the vast impact of the decision on the *557“average citizen’s” life are reminiscent of those asserted in opposition to our decisions abolishing husband-wife, parent-child, and generally family immunity in France v. A.P.A. Transport Corp., 56 N.J. at 500, and Immer v. Risko, 56 N.J. at 482. In Immer, proponents of interspousal immunity claimed that abandoning it would disrupt domestic harmony and encourage possible fraud and collusion against insurance companies. 56 N.J. at 488. In France, it was predicted that refusal to apply the parent-child immunity would lead to depletion of the family exchequer and interfere with parental care, discipline and control. 56 N.J. at 504. As we noted there, “[w]e cannot decide today any more than what is before us, and the question of what other claims should be entertained by our. courts must be left to future decisions.” Immer, 56 N.J. at 495. Some fifteen years have gone by and, as far as we can tell, nothing but good has come as a result of those decisions.

We recognize, however, that the point of view expressed by the dissent conforms, at least insofar as the result is concerned, with the view, whether legislatively or judicially expressed, of practically every other jurisdiction that has been faced with this question. It seems to us that by now it ought to be clear to all that the concerns on which that point of view is based are minor compared to the devastating consequences of drunken driving. This is a problem that society is just beginning to face squarely, and perhaps we in New Jersey are doing so sooner than others.

For instance, the dissent’s emphasis on the financial impact of an insurance premium increase on the homeowner or the tenant should be measured against the monumental financial losses suffered by society as a result of drunken driving. By our decision we not only spread some of that loss so that it need not be borne completely by the victims of this widespread affliction, but, to some extent, reduce the likelihood that the loss will occur in the first place. Even if the dissent’s view of the scope of our decision were correct, the adjustments in social behavior at parties, the burden put on the host to reasonably oversee the serving of liquor, the burden on the guests to make *558sure if one is drinking that another is driving, and the burden on all to take those reasonable steps even if, on some occasion, some guest may become belligerent: those social dislocations, their importance, must be measured against the misery, death, and destruction caused by the drunken driver. Does our society morally approve of the decision to continue to allow the charm of unrestrained social drinking when the cost is the lives of others, sometimes of the guests themselves?

If we but step back and observe ourselves objectively, we will see a phenomenon not of merriment but of cruelty, causing misery to innocent people, tolerated for years despite our knowledge that without fail, out of our extraordinarily high number of deaths caused by automobiles, nearly half have regularly been attributable to drunken driving. See supra, at 545 n. 3. Should wé be so concerned about disturbing the customs of those who knowingly supply that which causes the offense, so worried about their costs, so worried about their inconvenience, as if they were the victims rather than the cause of the carnage? And while the dissent is certainly correct that we could learn more through an investigation, to characterize our knowledge as “scant” or insufficient is to ignore what is obvious, and that is that drunken drivers are causing substantial personal and financial destruction in this state and that a goodly number of them have been drinking in homes as well as taverns. Does a court really need to know more? Is our rule vulnerable because we do not know — nor will the Legislature— how much injury will be avoided or how many lives saved by this rule? Or because we do not know how many times the victim will require compensation from the host in order to be made whole?

This Court senses that there may be a substantial change occurring in social attitudes and customs concerning drinking, whether at home or in taverns. We believe that this change may be taking place right now in New Jersey and perhaps elsewhere. It is the upheaval of prior norms by a society that has finally recognized that it must change its habits and do *559whatever is required, whether it means but a small change or a significant one, in order to stop the senseless loss inflicted by drunken drivers. We did not cause that movement, but we believe this decision is in step with it.

We are well aware of the many possible implications and contentions that may arise from our decision. We express no opinion whatsoever on any of these matters but confine ourselves strictly to the facts before us. We hold only that where a host provides liquor directly to a social guest and continues to do so even beyond the point at which the host knows the guest is intoxicated, and does this knowing that the guest will shortly thereafter be operating a motor vehicle, that host is liable for the foreseeable consequences to third parties that result from the guest’s drunken driving. We hold further that the host and guest are liable to the third party as joint tortfeasors, Malone v. Jersey Central Power & Light Co., 18 N.J. 163, 171 (1955); Ristan v. Frantzen, 14 N.J. 455, 460 (1954); Matthews v. Delaware, L. & W. R.R., 56 N.J.L. 34 (Sup.Ct. 1893), without implying anything about the rights of the one to contribution or indemnification from the other. See supra at 549 n. 8.

Our ruling today will not cause a deluge of lawsuits or spawn an abundance of fraudulent and frivolous claims. Not only do we limit our holding to the situation in which a host directly serves a guest, but we impose liability solely for injuries resulting from the guest’s drunken driving. Cf. Immer, 56 N.J. at 482 (interspousal immunity abandoned only in actions arising out of negligent operation of automobiles). Automobile accidents are thoroughly investigated by law enforcement officers; careful inquiries are routinely made as to whether the drivers and occupants are intoxicated. The availability of clear objective evidence establishing intoxication will act to weed out baseless claims and to prevent this cause of action from being used as a tool for harassment.

*560We therefore reverse the judgment in favor of the defendants Zak and remand the case to the Law Division for proceedings consistent with this opinion.

dissenting.

Today, this Court holds that a social host who knowingly enables an adult guest to become intoxicated knowing that the guest will operate a motor vehicle is liable for damages to a third party caused by the intoxicated guest. The imposition of this liability on a social host places upon every citizen of New Jersey who pours a drink for a friend a heavy burden to monitor and regulate guests. It subjects the host to substantial potential financial liability that may be far beyond the host’s resources.

My position as a strong advocate of legal measures to combat drunk driving is established. See In re Kallen, 92 N.J. 14 (1983). The majority need not parade the horrors that have been caused by drunk drivers to convince me that there is always room for stricter measurers against intoxicated drivers. I too am concerned for the injured victim of a drunken driver. However, the almost limitless implications of the majority’s decision lead me to conclude that the Legislature is better equipped to effectuate the goals of reducing injuries from drunken driving and protecting the interests of the injured party, without placing such a grave burden on the average citizen of this state.

I

Prior to today’s decision, this Court had imposed liability only on those providers of alcoholic beverages who were licensed by the State. See Rappaport v. Nichols, 31 N.J. 188 (1959). The Appellate Division also had expanded the liability to a social host who served liquor to a minor. Lind v. Rand, 140 N.J.Su*561per. 212 (App.Div. 1976).1 Although both of these cases were based on common-law negligence, the courts deemed the regulations restricting the service of alcohol to minors significant enough evidence of legislative policy to impart knowledge of foreseeable risk on the provider of the alcohol and to fashion a civil remedy for negligently creating that risk.

Many other states have considered the problem before us today but no judicial decision establishing a cause of action against a social host for serving liquor to an adult social guest is currently in force. Any prior judicial attempts to establish such a cause of action have been abrogated or restricted by subsequent legislative action. See, e.g., Cal.Civ.Code § 1714 (as amended Stats.1978, ch. 929, § 2, p. 2904); Or.Rev.Stat. § 30.955 (1979).

State courts have found that imposition of this new form of liability on social hosts is such a radical departure from prior law, with such extraordinary effects on the average citizen, that the issue is best left to a legislative determination. See Kowal v. Hofher, 181 Conn. 355, 436 A.2d 1 (1980); Miller v. Moran, 96 Ill.App.3d 596, 52 Ill.Dec. 183, 421 N.E.2d 1046 (1981); Behnke v. Pierson, 21 Mich.App. 219, 175 N.W.2d 303 *562(1970); Cole v. City of Spring Lake Park, 314 N.W.2d 836 (Minn. 1982); Runge v. Watts, 180 Mont. 91, 589 P.2d 145 (1979); Hamm v. Carson City Nugget, Inc., 85 Nev. 99, 450 P.2d 358 (1969); Schirmer v. Yost, 60 A.D.2d 789, 400 N.Y.S.2d 655 (1977); Edgar v. Kajet, 84 Misc.2d 100, 375 N.Y.S.2d 548 (Sup.Ct. 1975), aff’d, 55 A.D.2d 597, 389 N.Y.S.2d 631 (1976); Klein v. Raysinger, supra, — Pa. —, 470 A.2d 507; Halvorson v. Birchfield Boiler, Inc., 76 Wash. 2d 759, 458 P.2d 897 (1969).

I agree with the holdings of our sister states and with their misgivings about the judicial imposition of the duty that the majority today places on social hosts. Their conclusions, and the unanimity with which they express them, are instructive.

In Olsen v. Copeland, 90 Wis.2d 483, 491, 280 N.W.2d 178, 181 (1979), the Supreme Court of Wisconsin said:

A change in the law which has the power to so deeply affect social and business relationships should only be made after a thorough analysis of all the relevant considerations. * * * The type of analysis required is best conducted by the legislature using all of the methods it has available to it to invite public participation.

Similarly, in Edgar v. Kajet, supra, 84 Misc.2d at 103, 375 N.Y.S.2d at 552, the New York Supreme Court, Appellate Division, stated:

The implications of imposing civil liability on Avis herein are vast and far-reaching. Extending liability to non-sellers would open a virtual Pandora’s box to a wide range of numerous potential defendants when the Court does not believe that the legislature ever intended to enact a law that makes social drinking of alcoholic beverages and the giving of drinks of intoxicating liquors at social events actionable. Just a recitation of a few of the considerations involved herein impels this Court to conclude that any extension of liability should be a legislative act. For example, how is a host at a social gathering to know when the tolerance of one of his guests has been reached? To what extent should a host refuse to serve drinks to those nearing the point of intoxication? Further, how is a host to supervise his guests' social activities? The implications are almost limitless as to situations that might arise when liquor is dispensed at a social gathering, holiday parties, family celebrations, outdoor barbecues and picnics, to cite a few examples. If civil liability were imposed on Avis herein, it could be similarly imposed on every host who, in a spirit of friendship, serves liquor. In the final analysis, the controlling consideration is public policy, and *563any extension of liability should be carefully considered after all the factors have been examined and weighed in our legislative process, that is, after extensive hearings, surveys and investigation.

In Cartwright v. Hyatt Corp., 460 F.Supp. 80, 82 (D.D.C. 1978), the Federal District Court for the District of Columbia stated:

Valid policy considerations exist on both sides of this issue, and the Court is not prepared to adopt for the District of Columbia a rule not judicially imposed by any other court in any other jurisdiction. If such a rule is to become a part of District of Columbia law, the decision should appropriately be made by the legislature — as it has been wherever the rule has been adopted.

In Runge v. Watts, supra, 180 Mont. at 94, 589 P.2d at 147, the Supreme Court of Montana stated:

Establishing such a civil cause of action involves considerations of public policy far beyond those presented by the circumstances of the instant case.

In Holmes v. Circo, 196 Neb. 496, 505, 244 N.W.2d 65, 70 (1976), the Supreme Court of Nebraska stated:

We agree * * * that, in the final analysis, the controlling considerations are public policy and whether the court or the Legislature should declare it. We believe that the decision should be left to the Legislature. The Legislature may hold hearings, debate the relevant policy considerations, weigh the testimony, and, in the event it determines a change in the law is necessary or desirable, it can then draft statutes which would most adequately meet the needs of the public in general, while balancing the interest of specific sectors.

In Manning v. Andy, 454 Pa. 237, 239, 310 A.2d 75, 76 (1973), the Supreme Court of Pennsylvania stated:

While appellant’s proposal may have merit, we feel that a decision of this monumental nature is best left to the legislature.

II

My reluctance to join the majority is not based on any exaggerated notion of judicial deference to the Legislature. Rather, it is based on my belief that before this Court plunges into this broad area of liability and imposes high duties of care on social hosts, it should carefully consider the ramifications of its actions. The Court acts today with seemingly scant knowledge and little care for the possible negative consequences of its decision.

*564The magnitude of the problem with which we are dealing is entirely unknown. As the Illinois Appellate Court noted in Miller v. Moran, supra, 96 Ill.App.3d at 600, 421 N.E.2d at 1049, the injured party normally has a remedy against the direct perpetrator of the injury, the intoxicated driver. The majority’s portrayal of the specter of many innocent victims with no chance of recovery against drunk drivers is specious.

The Legislature of this state has enacted a comprehensive auto insurance program to guarantee that those injured on our highways have remedies. Even in cases in which the drunk driver is insolvent and has no insurance, the victim’s own automobile insurance policy is required by law, N.J.S.A. 17:28—1.1, to include coverage for all or part of the sums that the insured “shall be legally entitled to recover as damages from owners or operators of uninsured automobiles * * Furthermore, all motorists must have uninsured motorist coverage. N.J.S.A. 39:6A-14. If the drunk driver hits an uninsured pedestrian, the pedestrian, after obtaining a judgment, and unsuccessfully attempting to satisfy it, could satisfy the judgment out of the Unsatisfied Claim and Judgment Fund. N.J.S.A. 39:6-73. Thus, only in the situation in which a victim of a drunk driver is himself an uninsured motorist at the time of the injury will the victim have no remedy under our insurance laws. There, the victim, having broken the law by driving without insurance, is not entitled to collect from the Unsatisfied Claim and Judgment Fund.

I do not know whether the Legislature’s insurance scheme provides adequate protection for victims of drunk drivers and whether further relief may be necessary. I have seen no statistics to indicate the extent of this problem. However, the Legislature can collect information indicating the number of victims of drunk drivers who have not been adequately compensated. These statistics would be significant factors in ascertaining the scope of the problem and in creating solutions that would protect the injured party without excessively burdening the average citizen.

*565As stated earlier in this dissent, this Court has, in the past, imposed civil liability on commercial licensees who serve alcoholic beverages to intoxicated patrons. Commercial licensees are subject to regulation by both the Alcoholic Beverage Commission (ABC) and the Legislature. It is reasonable to impose tort liability on licensees based on their violation of explicit statutes and regulations.

I have no quarrel with the imposition of such liability because of the peculiar position occupied by the licensee. A social host, however, is in a different position. A brief discussion of the dissimilarities between the licensee and the private social host will illustrate the many problems this Court is creating by refusing to distinguish between the two in imposing liability upon them.

A significant difference between an average citizen and a commercial licensee is the average citizen’s lack of knowledge and expertise in determining levels and degrees of intoxication. Licensed commercial providers, unlike the average citizen, deal with the alcohol-consuming public every day. This experience gives them some expertise with respect to intoxication that-social hosts lack. A social host will find it more difficult to determine levels and degrees of intoxication.

The majority holds that a host will be liable only if he serves alcohol to a guest knowing both that the guest is intoxicated and that the guest will drive. Ante at 548. Although this standard calls for a subjective determination of the extent of the host’s knowledge, a close reading of the opinion makes clear that the majority actually is relying on objective evidence. The majority takes the results of Gwinnell’s blood alcohol concentration test and concludes from that test that “the Zaks must have known that their provision of liquor was causing Gwinnell to become drunk * * *." Ante at 543.

Whether a guest is or is not intoxicated is not a simple issue. Alcohol affects everyone differently. “[T]he precise effects of a particular concentration of alcohol in the blood varies from *566person to person depending upon a host of other factors. See generally Perr, ‘Blood Alcohol Levels and “Diminished Capacity”,’ 3 (No. 4) J. Legal Med. 28-30 (April 1975).” State v. Stasio, 78 N.J. 467, 478 n. 5 (1979). One individual can consume many drinks without exhibiting any signs of intoxication. Alcohol also takes some time to get into the bloodstream and show its outward effects. Experts estimate that it takes alcohol twenty to thirty minutes to reach its highest level in the bloodstream. See American Medical Association, Alcohol and the Impaired Driver (1968). Thus, a blood alcohol concentration test demonstrating an elevated blood alcohol level after an accident may not mean that the subject was obviously intoxicated when he left the party some time earlier. “Moreover, a state of obvious intoxication is a condition that is very susceptible to after the fact interpretations, i.e., objective review of a subjective decision. These factors combine to make the determination that an individual is obviously intoxicated not so obvious after all.” Comment, “Social Host Liability for Furnishing Alcohol: A Legal Hangover?” 1978 Pac.L.J. 95, 103. Accordingly, to impose on average citizens a duty to comprehend a person’s level of intoxication and the effect another drink would ultimately have on such person is to place a very heavy burden on them.

The nature of home entertaining compounds the social host’s difficulty in determining whether a guest is obviously intoxicated before serving the next drink. In a commercial establishment, there is greater control over the liquor; a bartender or waitress must serve the patron a drink. Not so in a home when entertaining a guest. At a social gathering, for example, guests frequently serve themselves or guests may serve other guests. Normally, the host is so busy entertaining he does not have time to analyze the state of intoxication of the guests. Without constant face-to-face contact it is difficult for a social host to avoid serving alcohol to a person on the brink of intoxication. Furthermore, the commercial bartender usually does not drink on the job. The social host often drinks with the *567guest, as the Zaks did here. The more the host drinks, the less able he will be to determine when a guest is intoxicated. It would be anomalous to create a rule of liability that social hosts can deliberately avoid by becoming drunk themselves.

The majority suggests that my fears about imposition of liability on social hosts who are not in a position to monitor the alcohol consumption of their guests are “purely hypothetical” in that the present case involves a host and guest in a one-to-one situation. It is unrealistic to assume that the standards set down by the Court today will not be applied to hosts in other social situations. Today’s holding leaves the door open for all of the speculative and subjective impositions of liability that I fear.

A more pressing distinction between the social host and commercial licensees is the host’s inability to fulfill the duty the majority has imposed even if the host knows that a particular guest is intoxicated. It is easy to say that a social host can just refuse to serve the intoxicated person. However, due to a desire to avoid confrontation in a social environment, this may become a very difficult task. It is much easier in a detached business relationship for a bartender to flag a patron and either refuse to serve him or ask him to leave. We should not ignore the social pressures of requiring a social host to tell a boss, client, friend, neighbor, or family member that he is not going to serve him another drink. Moreover, a social host does not have a bouncer or other enforcer to prevent difficulties that may arise when requesting a drunk to stop drinking or not to drive home. We have all heard of belligerent drunks.

Further, it is not clear from the Court’s opinion to what lengths a social host must go to avoid liability. Is the host obligated to use physical force to restrain an intoxicated guest from drinking and then from driving? Or is the host limited to delay and subterfuge tactics short of physical force? What is the result when the host tries to restrain the guest but fails? *568Is the host still liable? The majority opinion is silent on the extent to which we must police our guests.

Ill

The most significant difference between a social host and a commercial licensee, however, is the social host’s inability to spread the cost of liability. The commercial establishment spreads the cost of insurance against liability among its customers. The social host must bear the entire cost alone. While the majority briefly discusses this issue, noting that it may result in a catastrophic loss of a home to a husband and wife, it apparently does not consider this much of a problem to the average New Jersey citizen. It assumes that such liability is now covered or will be covered under the homeowner’s insurance policy.

The majority cites no authority for its belief that actions against social hosts will be covered under homeowner’s insurance. This new cause of action will be common and may result in large awards to third parties. Even if it is assumed that homeowner’s insurance will cover this cause of action, it is unrealistic to believe that insurance companies will not raise their premiums in response to it.

Furthermore, many homeowners and apartment renters may not even have homeowner’s insurance and probably cannot afford it. Other homeowners may not have sufficient insurance to cover the limitless liability that the Court seeks to impose. These people may lose everything they own if they are found liable as negligent social hosts under the Court’s scheme. The individual economic cost to every New Jersey citizen should be weighed before today’s result is reached.

The majority cites several cases in which this Court took action without prior legislative study or approval and was subsequently reversed by the Legislature. Ante at 555-556. I take no solace in the fact that the Legislature may reverse today’s decision if it disagrees with the Court’s action.

*569There are obviously instances in which this Court must take action in the absence of legislative guidance. Burlington Cty. NAACP v. Mount Laurel Tp., 92 N.J. 158, 212-13 (1983) (Mount Laurel II), is a perfect example of this kind of decision. In Mount Laurel II this Court felt compelled to act to protect the constitutional rights of citizens against discriminatory zoning where the Legislature had failed to act.

Here the Legislature has not, by refusing to act, forced this Court to fashion a remedy. Although it is true, as the majority points out, that New Jersey has no Dram Shop Statute, our Legislature and the ABC have been particularly active and diligent in creating duties and remedies to protect the public from drunk drivers.

Recently, our Legislature has enacted laws making New Jersey the unchallenged leader in the national crackdown on drunken driving. Evidence that the Legislature is still vitally interested in the area of drunken driving is Senate Bill S-1054, recently passed by the Senate and Assembly. It provides a criminal penalty for a social host who serves alcohol to a minor. The absence of any similar imposition of criminal liability on social hosts who serve adult guests should be instructive as to the Legislature’s intent on the matter before the Court.

IV

In conclusion, in trivializing these objections as “cocktail party customs”, ante at 555 and “inconvenience”, ante at 559, the majority misses the point. I believe that an indepth review of this problem by the Legislature will result in a solution that will further the goals of reducing injuries related to drunk driving and adequately compensating the injured party, while imposing a more limited liability on the social host. Imaginative legislative drafting could include: funding a remedy for the injured party by contributions from the parties most responsible for the harm caused, the intoxicated motorists; making the social host secondarily liable by requiring a judg*570ment against the drunken driver as a prerequisite to suit against the host; limiting the amount that could be recovered from a social host; and requiring a finding of wanton and reckless conduct before holding the social host liable.

I do not propose to fashion a legislative solution. That is for the Legislature. I merely wish to point out that the Legislature has a variety of alternatives to this Court’s imposition of unlimited liability on every New Jersey adult. Perhaps, after investigating all the options, the Legislature will determine that the most effective course is to impose the same civil liability on social hosts that the majority has imposed today. I would have no qualms about that legislative decision so long as it was reached after a thorough investigation of its impact on average citizens of New Jersey.

For reversal and remandment—Chief Justice WILENTZ, and Justices CLIFFORD, SCHREIBER, HANDLER, POLLOCK and O’HERN—6.

Opposed—Justice GARIBALDI—1.