specially concurring in part; dissenting in part.
For the reasons stated in my dissent in Gattman v. Favro, 86 Or App 227, 739 P2d 572 (1987), I dissent from the *500majority’s holding that plaintiff has an actionable statutory tort claim. I reluctantly agree with the majority that plaintiff does state claims for negligence and negligence per se, because Fazzolari v. Portland School Dist. No. 1J, 303 Or 1, 734 P2d 1326 (1987), precludes us from holding that the shooting of the decedent was unforeseeable to defendants as a matter of law. However, I think it is unreasonable for defendants to be chargeable with foresight of what occurred here.
The anomaly which follows from the majority’s decision is that defendants are potentially liable for the fatal injuries which their customers inflicted on the cab driver because they did exactly what they should have done by calling a cab rather than have the customers drive while intoxicated.1 I recognize that drunk driving is not the only socially undesirable behavior in which intoxicated people are more likely to engage than sober persons and that aggressive tendencies are sometimes exacerbated by alcohol. However, that generalization does not alter the fact that it is far more likely that a drunk patron will drive a car after leaving a serving establishment than that he will kill the driver if someone calls a cab for him.
The message today’s opinion conveys to owners and servers is that, if they take responsible measures to prevent a likely and identifiable risk of harm, that preventive action itself will create potential liability for every event, however improbable and unpredictable, which may ensue. I hope that message will not deter the owners and employes of serving establishments from recognizing that the likelier risk of harm carries the greater risk of liability with it; it is equally possible, however, that the majority’s decision will discourage the salutary practice of sending intoxicated customers home in cabs. New cab drivers will be saved as a result. To belabor the obvious, relatively few intoxicated riders present a homicidal menace, but every drunk driver does.
*501Whatever merit there may be to the general proposition that those who serve alcohol should be held to anticipate everything that their intoxicated customers may do, it is a far more basic proposition that the rules of law which courts develop should not be in conflict with each other and should have some capability of realistic application to a world of real events. The majority’s conclusion on the statutory tort issue and its compelled conclusions on the common law and negligence per se issues fall well short of those objectives.
In my view, the question here should not be whether defendants ought to have foreseen both that their customers might drive while drunk and that they might kill the cab driver, but whether, as a matter of law, they ought to have prevented the very foreseeable and likely risk of drunk driving by calling a cab. By any reasonable estimation of probabilities, the harm that befell the driver was very unlikely, and that is not altered by the fact that he was harmed. I would conclude, if we were free to do so, that policy considerations should foreclose an alcohol server’s liability for an arguably foreseeable but very remote risk when that risk is incurred to avoid the highly foreseeable and dangerous risk of an intoxicated customer driving a car.21 reiterate that our decision today places alcohol servers at risk of liability for doing exactly what they should do.
I do not know whether defendants’ motives for calling the cab included the prevention of drunk driving by the customers. For purposes of this opinion, it is unnecessary to decide whether that was their motive or whether their motive is significant. It is beside the point I am making here whether these defendents were so motivated. The majority’s opinion will reach other owners and servers as well as the defendants in this case.
Although the Supreme Court’s opinions in Fazzolari v. Portland School Dist. No. 1J, supra, and related cases, do not leave us free to decide this case as I have described, the appeal may provide an occasion for the Supreme Court to give further definition to its suggestion in Donaca v. Curry Co., 303 Or 30, 38, 734 P2d 1339 (1987), that there are “extreme cases” in which foreseeability is a question for the court rather than the factfinder. See Stewart v. Jefferson Plywood Co., 255 Or 603, 469 P2d 783 (1970).