(concurring) — The facts presented do not indicate Burkhart was obviously intoxicated either at the time he drank at the Harrods' party or when he left on his motorcycle. The majority opinion's statements regarding the absence of social host liability are therefore dicta and not controlling on a future case presenting different facts. I believe there are facts under which a social host should be liable, using well settled principles of negligence law, to persons injured as a proximate result of the host serving alcohol to an obviously intoxicated guest and then failing to take reasonable measures to insure the guest does not drive on public roads. That question, however, is not before us in this case.
Considering only the relevant and admissible evidence, and viewing such evidence in a light most favorable to Burkhart, there is no evidence indicating Burkhart was obviously intoxicated either at the time the Harrods furnished him alcohol or at the time he left the Harrod residence by motorcycle.
The only relevant evidence presented regarding Burk-hart's intoxication is the way he appeared to those around him. Purchase v. Meyer, 108 Wn.2d 220, 737 P.2d 661 (1987); Wilson v. Steinbach, 98 Wn.2d 434, 439, 656 P.2d 1030 (1982). Appellants rely on a blood alcohol level of .16 and expert testimony based on the blood alcohol level to raise an inference of obvious intoxication at time of service. Given the recent decisions in Purchase and in Dickinson v. Edwards, 105 Wn.2d 457, 716 P.2d 814 (1986), evidence of blood alcohol content and expert testimony based on such test results is inadmissible on the issue of obvious intoxication.
*392Appellants contend that expert testimony based on the blood alcohol content should be admissible to establish the amount of alcohol Burkhart consumed at the party. In Dickinson, this court found evidence of the amount of alcohol consumed relevant on the issue of obvious intoxication. Dickinson, at 464-65 (plurality opinion). However, as this court observed in Purchase, at 227-28, Dickinson was factually unique because the driver admitted to the number of drinks he had consumed and the officer observed the driver just 10 minutes after leaving the drinking establishment.
Even if expert testimony could establish the amount of alcohol consumed, the expert testimony here is not competent. The experts based their opinions on Burkhart having consumed alcohol only between 2 and 6 p.m. However, appellants contend that Burkhart continued drinking after dinner ended at 6:30 p.m. Burkhart may have drunk more alcohol after stopping at a store to buy beer; a 2-quart bottle of Olde English 800 malt liquor, three-fourths full, was found at the scene. Even granting appellants every favorable inference and assuming the beer had leaked out of the bottle, the experts based their opinions on an inaccurate set of assumptions regarding when at the party Burkhart had consumed alcohol; the opinions must be considered speculative and irrelevant, and therefore inadmissible. See, e.g., Myers v. Harter, 76 Wn.2d 772, 781-82, 459 P.2d 25 (1969).
Appellants have also submitted post-service observations by a motorist who saw Burkhart driving erratically and aggressively as if he were intoxicated shortly before the accident. The key to admitting post-service firsthand observations lies in the reliability of any inference of obvious intoxication that can be drawn from such evidence. Essential to the inference's reliability is the proximity to the time Burkhart was last furnished alcohol, whether Burkhart consumed any alcohol after he left the party, and whether any time remained unaccounted for between the last furnishing and the post-service evidence. Dickinson, at 464. Dickinson's narrow facts were that the police officer's *393observations came just 10 minutes after the driver had left the drinking establishment. Purchase, 108 Wn.2d at 227. Here, the observations of Burkhart's driving came up to 1 hour after he left respondents' home and after he stopped at a store and purchased beer. These observations more closely resemble a police officer's observations "an hour or two" after service of alcohol and found inadmissible in Purchase. Appellants also submitted as evidence observations by the police officer who arrived at the scene of the accident and stated that Burkhart's body smelled of alcohol. Those observations suffer the same infirmities as the firsthand observations shortly before the accident and are inadmissible.
The only other evidence of Burkhart's obvious intoxication at the time of service was the typewritten transcript of tape-recorded interviews conducted by respondents' insurance agent. These allegedly inconsistent statements include one by the groom's sister that Burkhart "did kick one or two drinks over." When read in context this statement refers to the number of drinks consumed, and not to any obviously intoxicated behavior. In addition, the bride-to-be, when asked by the insurance investigator if she noticed Burkhart was under the influence of alcohol, answered: "That night . . . no. Because by then it was way later." This ambiguous statement loses any sting in the context of her entire statement, which made it clear that she did not observe signs of obvious intoxication at any point in the day or evening.
In sum, there is no evidence that the Harrods knew or had reason to know Burkhart was obviously intoxicated when he last drank alcohol at their party or when he left on his motorcycle. In fact, the affidavits of those observing Burkhart at the party unanimously state that Burkhart showed no signs of obvious intoxication.
Negligence Standard
Because the evidence does not indicate Burkhart was obviously intoxicated, this case does not present the issue *394of the liability of a social host with regard to obviously intoxicated guests. If such evidence were present, unlike the majority, I would find a social host has a duty to take reasonable measures to prevent a guest from driving on public roads after the host furnishes the guest with alcohol when the guest is obviously intoxicated.
I recognize that my statements in this regard, if they were in the majority opinion of this case, would be dicta. Similarly, the majority's sweeping statements immunizing social hosts from liability are dicta. Although our discussions are outside the scope of the facts presented here, the majority's statements should not be left unchallenged lest it be assumed they are unanswerable. The best way to test them is with a hypothetical scenario: A social host continues to serve alcohol to a guest who is obviously intoxicated, perhaps even encouraging the guest to continue imbibing; the host knows that the guest is driving a vehicle home but takes no steps to deter the guest; the guest then strikes a pedestrian; the guest has no resources from which to compensate the pedestrian.
We must view the issue of social host liability in the context of existing tort law. Tort law is overwhelmingly court-developed common law. Statutes influence tort law, either by creating standards of behavior or by explicitly changing substantive law rules previously developed by courts. W. Keeton, D. Dobbs, R. Keeton & D. Owen, Prosser and Keeton on Torts § 3, at 19 (5th ed. 1984). Otherwise, courts have the responsibility to develop the common law in response to social or technological changes, and to fine tune the law to be more just.6
In developing common law, we must consider not only compensation to injured victims (which the New Jersey Supreme Court considered the crucial factor in extending *395liability to social hosts in Kelly v. Gwinnell, 96 N.J. 538, 548, 551, 476 A.2d 1219 (1984)), but also public policy. One reason for imposing liability is to provide an incentive in the right place to avoid the loss altogether. W. Keeton, D. Dobbs, R. Keeton & D. Owen, Prosser and Keeton on Torts § 4, at 25 (5th ed. 1984). "There is good reason ... to make a conscious effort to direct the law along lines which will achieve a desirable social result, both for the present and for the future." W. Keeton § 3, at 16.
In the context of the hypothetical then, the question is whether the social host or the pedestrian is in a better position to avoid the loss. The answer is clear: the social host. Imposing social host liability will "make it more likely that hosts will take greater care in serving alcoholic beverages at social gatherings so as to avoid . . . economic liability". Kelly, at 552.
The second inquiry is whether preexisting principles of tort liability are consistent with social host liability. Negligent conduct consists of (1) the existence of a duty owed to the complaining party; (2) a breach thereof; and (3) a resulting injury. E.g., LaPlante v. State, 85 Wn.2d 154, 159, 531 P.2d 299 (1975). A cause of action by a party injured as a result of a social host's providing alcohol to an obviously intoxicated guest easily falls within this framework.
1. Duty
In developing the scope of tort liability for negligence actions, we have held, "The question of duty encompasses the concept of foreseeability." Maltman v. Sauer, 84 Wn.2d 975, 980, 530 P.2d 254 (1975). In other words, absent some special immunity, actors are responsible for the foreseeable consequences of their acts. "Generally, the duty to use ordinary care is bounded by the foreseeable range of danger." Wells v. Vancouver, 77 Wn.2d 800, 803, 467 P.2d 292 (1970). It is beyond question that a foreseeable result of "mixing gasoline and liquor" is, at the least, some driving error, or at the most a tragic accident. The California Supreme Court observed it is "evident that the service of *396alcoholic beverages to an obviously intoxicated person by one who knows that such intoxicated person intends to drive a motor vehicle creates a reasonably foreseeable risk of injury to those on the highway." Coulter v. Superior Court, 21 Cal. 3d 144, 152-53, 577 P.2d 669, 145 Cal. Rptr. 534 (1978) (abrogated by California Senate Bill 1645 (1978)). Our society has become increasingly aware of the dire consequences of drunk driving, partly through the admirable efforts of grass roots groups such as Mothers Against Drunk Drivers and Students Against Drunk Drivers.
2. Breach of Duty
"[Njegligence consists in the doing of an act which a reasonable man would not have done, or in the failure to do an act which a reasonable man would have done under similar circumstances." System Tank Lines, Inc. v. Dixon, 47 Wn.2d 147, 151, 286 P.2d 704 (1955). In the hypothetical, the social host's doing nothing to deter the intoxicated guest from driving was not reasonable, considering it was the host who made the guest unable to drive safely, and perhaps unable to perceive his incapacity. Naturally, it is for the finder of fact to determine if the host has an excuse for failing to deter the guest, such as fear of physical harm.
3. Resulting Injury
This court has rejected the theory that providers have not proximately caused injuries caused by those provided with alcohol. "[T]he court's analysis of proximate cause changed from 'the drinking of the liquor is the proximate cause of the injury' to 'the furnishing of the liquor is the proximate cause of the injury.'" Dickinson, at 461. In the hypothetical, the host has thus proximately caused the pedestrian's injuries. I choose not to discuss the issue of allocation of damages between hosts and guests in the absence of that issue in this case.
It requires no extension of existing tort principles to impose liability on social hosts who have put their guests in *397no position to drive. "If the actor does an act, and subsequently realizes or should realize that it has created an unreasonable risk of causing physical harm to another, he is under a duty to exercise reasonable care to prevent the risk from taking effect." Restatement (Second) of Torts § 321(1), at 132 (1965). In Robertson v. LeMaster,_W. Va. _, 301 S.E.2d 563 (1983), the West Virginia Supreme Court found an employer liable under this common law principle for encouraging an employee to work unreasonably long hours and then sending him out on the highway in an exhausted condition. As one commentator noted in comparing Robertson with the social host situation:
In both instances, the driver's condition is a natural and foreseeable result of the host's or employer's actions, which the host or employer should realize have created a dangerous situation. Although a tired or intoxicated driver always has the option not to drive, in many cases her condition may be such that she does not recognize that option. The Robertson court did not suggest that the employer should have prevented the employee from driving; it did suggest, however, that the employer should have provided the employee with an alternative.
Comment, Social Hosts and Drunken Drivers: A Duty To Intervene?, 133 U. Pa. L. Rev. 867, 890 (1985).
Because all the elements of a negligence cause of action are present in the social host hypothetical presented above, the majority's denial of compensation to victims injured as a proximate result of the host's unreasonable actions must rest on some sort of immunity.
The courts and Legislature in this state are very reluctant to grant immunity for tortious conduct, absent very compelling public interests. Some traditional common law immunities have been abolished in the interest of providing just remedies. Thus the interest of preserving "peace and tranquility of the home" was found insufficient justification to perpetuate inter-spousal immunity in Freehe v. Freehe, 81 Wn.2d 183, 187, 500 P.2d 771 (1972). The "need of financial encouragement and protection" of charitable *398institutions was found insufficient justification to perpetuate charitable immunity in Friend v. Cove Methodist Church, Inc., 65 Wn.2d 174, 176, 396 P.2d 546 (1964). Sovereign immunity was statutorily abolished. RCW 4.96.010.
Where the Legislature has enacted statutory immunities, it is generally in response to a compelling social need. Thus, providers of emergency care are immune from ordinary negligence suits under RCW 4.24.300. Reporters of child abuse are immune from suit under RCW 26.44.060. This court grants immunity to parties and counsel for allegedly libelous statements made in the course of judicial proceedings based on a public policy of granting them utmost freedom to secure justice. McNeal v. Allen, 95 Wn.2d 265, 267, 621 P.2d 1285 (1980).
The majority cites no compelling public policy, even in its dicta, for granting immunity to social hosts. The only justification the majority offers for granting immunity to social hosts where commercial providers have no such immunity is that the latter are "only a narrow slice of our population, [whereas] social host liability would touch most adults in the state on a frequent basis." Majority, at 387. This is not a proper basis for granting tort immunity. The entire population of this state is liable for the foreseeable consequences of unreasonable behavior.
Commercial hosts are liable to parties injured as a result of their serving alcohol to obviously intoxicated patrons. E.g., Dickinson. There is no logical reason to deny recovery to victims injured as a proximate result of equally unreasonable actions by social hosts. Indeed, I must now agree with Justice Durham's statement in her dissent in Dickinson, "[T]he inevitable extension of the majority rationale dictates a finding of liability for a social host." Dickinson, at 489. The consequences to the victims are the same, whether the driver's host was commercial or social. And there is no convincing reason to see the host's responsibility to the victim any differently based on whether he or she benefits commercially from serving alcohol. See Comment, *399Social Hosts and Drunken Drivers: A Duty To Intervene?, 133 U. Pa. L. Rev. 867, 869 (1985).
The majority seems concerned that hosts will worry they will be held liable for serving guests that are not obviously intoxicated. The majority thus poses a fretful list of inquiries that hosts presumably would make if confronted with liability. "Can the host determine if a guest is drinking on an empty stomach? ... Is it necessary to gauge a guest's weight and height in order to determine allowable amounts of alcohol? . . . Before guests depart, should the host . . . [ask] them to recite the alphabet or walk a straight line? . . ." Majority, at 384-85. This list, while entertaining, is irrelevant. There is no issue of liability of social hosts that serve alcohol to guests who cleverly hide their intoxication. As with the existing liability of commercial hosts, social host liability is not asserted to attach except where the guest is obviously7 intoxicated. See, e.g., Dickinson. That some hosts may have a groundless worry about whether they will be found liable when their guests are not in fact obviously intoxicated is trivial. The justice in compensating victims injured or killed as a proximate result of hosts serving truly obviously intoxicated guests alcohol and then failing to deter their driving on public roads is compelling.
Although, by definition, it will be "obvious" to social hosts when their guests are "obviously intoxicated", it will not always be clear to the hosts what steps to take in response to such guests intending to drive away from the social event. I recognize that it is not uncommon for intoxicated persons to be very assertive of their desire to drive, and some might become intimidating to their hosts. "We have all heard of belligerent drunks." Kelly, at 567 (Garibaldi, J., dissenting). On the other end of the spectrum, it might be merely embarrassing to the host to point out to a guest that the guest would be better to temporarily refrain from driving.
*400I propose the following standards to determine if a host has acted reasonably:
1. Liability attaches only where the social host has provided an obviously intoxicated guest with alcohol. The host has no liability if the guest has consumed the alcohol before arriving, or has imbibed his or her own alcohol while on the host's premises without the host's knowledge.
2. Liability attaches only where the host knows or has reason to know the guest intends to drive a motor vehicle away from the host's premises.
3. Hosts should verbally express to obviously intoxicated guests that the host does not wish the guest to drive while in that condition. Hosts should present reasonable alternatives to such guests: e.g., letting the guest stay until no longer impaired, providing alternative means of transportation, or lending money for taxi fare.
4. In no case is a host liable for failing to perform acts that might endanger his or her physical well being. See Comment, Recognizing the Liability of Social Hosts Who Knowingly Allow Intoxicated Guests To Drive: Limits to Socially Acceptable Behavior, 60 Wash. L. Rev. 389, 403 (1985).
5. Social embarrassment alone is not an excuse for failing to take reasonable steps to deter the guest from driving. Comment, 60 Wash. L. Rev., at 403.
These standards give hosts a choice. They can either refrain from providing alcohol to guests that are obviously intoxicated, or they can take reasonable steps to prevent such guests from driving. The standards certainly do not impose an onerous burden on hosts. I believe responsible hosts follow these or higher standards already.
Deference to Legislature
The majority would be correct in deferring to the Legislature if the Legislature had in fact spoken on the issue of social host liability or immunity. However, our Legislature has not determined whether to grant immunity to social hosts for injuries caused by providing alcohol to intoxicated *401guests and then failing to take reasonable measures to deter their guests from driving. Because it is the traditional role of the courts to develop the common law of tort liability, the Legislature's inaction in this area is no deterrent to this court.
We are competent to rule on this, or any other issue in which we must consider both the rights of the parties and public policy. "If each change in society occasioned by technological and other developments involving potential negligent harm to individuals requires legislation before tort liability results, justice would have been and would continue to be unnecessarily illusory in too many instances." Halvorson v. Birchfield Boiler, Inc., 76 Wn.2d 759, 768, 458 P.2d 897 (1969) (Finley, J., dissenting).
The majority attempts to infer a "legislative intent" from the absence of a statute penalizing social hosts for their furnishing liquor to persons under the influence of liquor. Majority, at 387.8 Determination of legislative intent, to the extent that the diverse members of our Legislature can be said to have a single intent, is a proper inquiry in the context of statutory interpretation. E.g., Grant v. Spellman, 99 Wn.2d 815, 818, 664 P.2d 1227 (1983) ("duty of the court is to ascertain and give effect to the intent and purpose of the Legislature, as expressed in the act"). (Italics mine.) In the absence of a statute, we do not decide cases before us by inquiring of the Legislature how it would rule.
Although "legislative intent" is irrelevant in the absence of a controlling statute, the Legislature can guide us on issues of public policy to assist in the development of common law. Moragne v. States Marine Lines, Inc., 398 U.S. 375, 390-93, 26 L. Ed. 2d 339, 90 S. Ct. 1772 (1970). "A court may look to statutes not only as mandates on issues directly addressed but also as sources of 'establishment of *402policy [that] carries significance beyond the particular scope of each of the statutes involved.'" W. Keeton, D. Dobbs, R. Keeton & D. Owen, Prosser and Keeton on Torts § 3, at 20 (5th ed. 1984) (quoting Boston Housing Auth. v. Hemingway, 363 Mass. 184, 293 N.E.2d 831, 840 (1973)). The Legislature has stated:
The legislature finds and declares:
(1) There is a need to reduce' the incidence of drivers on the highways and roads of this state who, because of their use, consumption, or possession of alcohol, pose a danger to the health and safety of other drivers;
RCW 46.20.710. Besides this unambiguous declaration, we can easily infer from the legislative acts listed by the majority and statutes such as RCW 46.20.710-.750 that the Legislature has determined public policy supports using every traditional and creative means available to deter drunk driving. See majority, at 390.
It is not only the Legislature that is concerned about the tragedies and waste associated with drunk driving. The attitudes of society have matured to the point that very few would find that social pleasures of mutual imbibing, combined with the enjoyment of driving, are more important to society than the risk to potential innocent victims caused by drunk driving. As the New Jersey Supreme Court noted when it imposed liability on social hosts:
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.
Kelly, at 545.
The "decision to defer the issue of social host liability to the Legislature", majority, at 388, is not a meaningful choice. "[W]here relevant legislation does not exist, courts must by necessity decide a controversy without legislative guidance. In doing so within a common law system in which each decision is precedent, they necessarily make law." W. Keeton, D. Dobbs, R. Keeton & D. Owen, Prosser and *403Keeton on Torts § 3, at 18 (5th ed. 1984). There is no need to reiterate that the Legislature may overturn our decision today by enacting legislation on the subject of social host liability. Given the narrow facts of this case, the majority's asserted decision to defer to the Legislature is not binding on future action of this court and is, in addition, erroneous as a matter of law and policy.
Conclusion
The summary judgment must be affirmed, but only on the fact that there is no evidence Burkhart was obviously intoxicated. The statements of the majority in dicta are unnecessary and incorrect and not controlling on future cases where the issue is before us. Compelling reasons exist not to grant social hosts immunity from liability for injuries proximately caused by their providing alcohol to obviously intoxicated guests where the hosts then fail to take reasonable steps to prevent the guests from driving.
Brachtenbach, J. (concurring) — I concur with Justice Utter only as to the first portion of his opinion, i.e., prior to that denominated "negligence standard."
When over-arching principles of justice are not accommodated by traditional theories of tort recovery, we do not hesitate to take measured steps to advance more just theories of tort recovery, as can be seen in cases such as Martin v. Abbott Labs., 102 Wn.2d 581, 689 P.2d 368 (1984).
The relevant dictionary definition of "obvious'' is: "capable of easy perception". Webster's Third New International Dictionary, at 1559.
I do not discuss the majority's reliance on the repeal of the dramshop act, as the repeal of that act has been disposed of in Halvorson v. Birchfield Boiler, Inc., 76 Wn.2d 759, 458 P.2d 897 (1969) as irrelevant to the development of the common law.