dissenting.

It is conceded that defendant was negligent. It was negligent because its employees deliberately and wrongfully released an unconscious man. For the purpose of this decision it is necessary to consider that plaintiff was as incapable of caring for himself as a blind man would be and that he was set at large without any care at all for his safety or means of travel or as to where he was going.

The case is treated as one of causation. I think it *55also involves an examination of the extent of the duty required and of policy. But on cause alone, it should be submitted to the jury. It is a difficult case made more difficult because it is the converse of the typical case. Usually the hazard is created or set at large to await or find an unknown, but predictable, victim. Here a known victim is set at large to meet an unknown but certain and predictable hazard. The cause is more elusive in the latter case because the defendant’s negligence has made plaintiff incapable of telling what happened to him. But the certainty of harm was just as great as leaving a dynamite cap exposed on a public playground. In either event the risk of harm is so great that in neither case should the court strain to imagine other concurring causes.

A more acceptable basis for deciding the court versus jury function than that adopted by the majority, leads to the conclusion that there is a permissible inference of cause.

Although the search for a test to solve the court versus jury causation problem is thus far fruitless① the “probability” test, adopted by the majority is not *56it. Professor Leon Green has best exposed the fault of the probability test:

“A so-called rule of ‘probabilities’ is not infrequently put into play as a test of causal relation. It is a highly seductive ‘decoy.’ It has the same vice as the ‘but-for’ test, namely, it takes the focus off the defendant’s conduct and goes abroad for other causes. At most it is an argument and, though a legitimate one for the advocate, can never rise to the level of a test of causal relation. This is readily demonstrated. The issue is whether defendant’s conduct contributed to the victim’s injury. To turn the issue into whether the defendant’s conduct caused the injury is a different issue, and then to require that defendant’s conduct must be found to be more probable than not the cause of the injury completely perverts the original issue.

“A victim’s hurt as the result, at least in part, of a defendant’s conduct may be highly improbable and yet admittedly true,* while on the other hand it may be highly probable and yet the result of other cause factors.* Since there are other causes in every case that also contributed to the victim’s hurt, to seek to find whether defendant’s conduct was the cause is literal nonsense, and equally so is the attempt to find whether the ‘probabilities’ are that it was the cause. The ‘probability’ test, as employed on the issue of causal relation is an ‘argumentative decoy’ not infrequently employed by the courts when it is not needed, or to reach an insupportable judgment. * * *.” Green, The Causal Relation Issue, 1962, 60 Mich L Rev 543, 556, 557, 558 and passim. (*Footnotes omitted).

In the same article he provides this more simplified formula for deciding whether or not a given case presents work for the court or the jury.

“* * * Causal relation will not be presumed from the fact of injury or from a showing defend*57ant has violated his duty.® It may he a case in which the duty and its violation did not include the risk to which the victim was subjected.* But if the details indicate that the victim’s safety was put in peril by defendant’s conduct it is for a jury to find whether that conduct did in fact contribute to the injury he suffered. The difficult question is the sufficiency of the evidence to raise the issue. So long as the evidence will support an inference that defendant’s conduct contributed to the victim’s injury, even though other inferences can be drawn that it did not, or that his injury was due to other causes, it is for the jury and not the judge to draw the inference. * * *.” 60 Mich L Rev p 560. (*Footnotes omitted).

A segment of Prosser’s analysis of cause is helpful in weighing the balance when we are concerned with the court versus jury function.

“* * ® Proof of what we call the relation of cause and effect, that of necessary antecedent and inevitable consequence, can be nothing more than ‘the projection of our habit of expecting certain consequents to follow certain antecedents merely because we had observed these sequences on previous occasions.’* If as a matter of ordinary experience a particular act or omission might be expected, under the circumstances, to produce a particular result, and that result in fact has followed, the conclusion may be permissible that the causal relation exists.

“® * * And whether the defendant’s negligence consists of the violation of some statutory safety regulation, or the breach of a plain common law duty of care, the court can scarcely overlook the fact that the injury which has in fact occurred is precisely the sort of thing that proper care on the part of the defendant would be intended to prevent, and accordingly allow a certain liberality to the *58jury in drawing its conclusion.*” Prosser, Torts (3d ed, 1964) pp 246-247. (*Footnotes omitted).②

Plaintiff’s predicament in the present case is “precisely the sort of thing that proper care on the part of the defendant would be intended to prevent.”

The majority think that defendant’s conduct in sending the unconscious plaintiff into a perilous arena will not support an inference that such negligence contributed to the harm that did befall him. This they say is true even though the admitted negligence can only be predicated on the foreseeability of harm.

I think that it is reasonable to infer that unconsciousness did make a contribution to plaintiff’s injury although other causes may also have contributed. This is not speculation. It is recognition that the foreseeable harm did occur. It does not destroy the inference to argue that unconsciousness in a protected crosswalk could not contribute to harm perhaps more directly caused by a negligent driver. A conscious pedestrian may well avoid the careless driver. More importantly, the concurring negligence does not necessarily insulate the defendant’s carelessness. Arneil v. Schnitzer et al, 1944, 173 Or 179, 144 P2d 707; Prosser, supra, p 309 et seq. In Arlington Heights Sanitarium, v. Deaderick, 272 SW 497 (Tex Civ App 1925) a causal relationship was found between a private sanitarium’s negligent act in permitting an insane inmate to escape and the death of the inmate who was later found dead on railroad tracks. The court held that the causal connection was not broken regardless of any negligence on the part of the railroad. The court likened the situation to that of permitting a small child to wander un*59attended through trafficked streets. The example is appropos here.

There is another reason why this case should not be decided in its present posture. That is the question of the burden of proof. Plaintiff mentioned the issue but it has not been seriously debated nor considered by the majority. If the case were presented to a jury on defendant’s present statement of the facts it could present a serious question as to who must carry the burden of proof. Prosser, supra, p 247, for example.

The case should be reversed.

This is an action for personal injuries. After opening statements by counsel defendant moved for an involuntary nonsuit. The court granted the motion and plaintiff appeals.①

The statement of facts most favorable to plaintiff is as follows: Plaintiff was injured in an automobile collision and taken unconscious to a hospital. The hospital . resident examined plaintiff and found no bone damage. The hospital mistakenly believed the plaintiff to be in some state of intoxication, whereas he was in a state of semi-consciousness caused by a brain concussion. The hospital negligently released plaintiff when he was unable to care for himself.

The plaintiff suffered amnesia from the time of the automobile collision until four days later. At such time he was found in bed in his house trailer. It is not known how long he was there. When he was found plaintiff had an injury to his right shoulder and frac*52tures of Ms pelvis. The cause, time or other circumstances of the injury are unknown. Nothing is known of plaintiff’s actions from the time he left the hospital until he was found in his trailer.

The problem is one of causation, — physical cause or cause in fact. Was defendant’s negligence a cause of plaintiff’s injury? There is no direct evidence upon this issue. Is there sufficient indirect evidence to warrant submitting the issue to the jury?

Plaintiff’s syllogism is as follows: The defendant hospital negligently discharged plaintiff although he was not able to care for himself. One who is not able to care for himself is likely to be injured; therefore, plaintiff’s injury probably was caused by his not being able to care for himself. Based upon this syllogism plaintiff contends “the evidence is sufficient to raise an inference that it was more probable that defendant’s negligence contributed to plaintiff’s injury than not.”

Plaintiff relies heavily upon the rationale of Eitel v. Times, Inc., 221 Or 585, 594, 352 P2d 485 (1960), particularly the statement in that opinion:

“We are of the opinion, therefore, that although defendant might suggest some probable causes which would insulate defendant’s negligence as a causal factor, there were in the present case sufficient facts to indicate that it was more probable than not that defendant’s conduct was the cause of plaintiff’s injury. Plaintiff is not required to eliminate all other possible causes in order to get her case to the jury. # * 221 Or at 596.

In that case we went on to state:

“* * * We have only our own experiences, observations and understanding of what commonly happens from which to judge whether the infer*53enees drawn from the circumstances connect up defendant’s conduct to the injury more strongly than the conduct of some other person.” 221 Or at 597.

Based upon “our own experiences, observations and understanding of what commonly happens” we conclude that under the circumstances here present a jury would have no basis upon which to decide that it is more probable that plaintiff was injured because he was not able to take care of himself than from some other cause for which defendant is not responsible.

We agree with plaintiff, and defendant does not appear to disagree, that if plaintiff were injured by some third party’s negligent or intentional act, which came about because of plaintiff’s inability to care for his own safety, defendant’s negligence could nevertheless be a cause of plaintiff’s injuries. For example, if plaintiff, unable to comprehend his danger, wandered into a busy street in front of a speeding vehicle, the defendant’s negligence could be a cause of his injuries. On the other hand, if the plaintiff were crossing a street at a controlled intersection, in the crosswalk, and in response to the “walk” signal, and was struck by a car running the red light, there would be no causal connection between defendant’s negligence and plaintiff’s injuries.

Plaintiff is correct that one unable to care for himself is more likely to be injured. This is the reason defendant’s conduct is negligent, i.e., defendant acted in a manner which created an unreasonable risk of harm to plaintiff. However, the fact that defendant’s conduct made it more likely that plaintiff would be injured does not of itself raise the inference that it is more probable than not that plaintiff was injured because he was unable to care for his own safety.

*54We know from onr experience that a substantial number of persons are injured despite their taking reasonable steps for their own safety. We do not know whether injuries occurring under such circumstances are in the majority or in the minority. The incidence of such injuries probably depends, at least partially, upon the other circumstances surrounding the infliction of the injury. Was the injured person a pedestrian, a vehicle passenger, or a vehicle driver? Did the injury occur at home, at work, or at recreation? Was the injury inflicted during daylight or darkness? We know none of the circumstances.

We conclude that a jury could not, without pure speculation, determine that it was more probable that defendant’s negligence was a cause of plaintiff’s injuries and, therefore, the issue of causation should not be submitted to the jury.②

Affirmed.