I dissent.
As I read the majority’s opinion, it appears that the key to their holding on the res judicata issue is their determination that there was an implied finding that the damages to the Ellena lands were reasonably foreseeable. If that position is sound, it is dispositive of the case, and the trial court must be affirmed.
However, for there to be an implied finding of foreseeability, there must be some evidence in the record to support it. As I read the record, *264what happened in 1969 as a matter of fact was totally unforeseen by everyone, including the state’s engineers who designed the facility here under question. The testimony of the state’s experts affirmed again and again that what happened was not foreseeable. The witness Jackson at one point testified, “[t]he big unknown here—and I hate to keep harping on it, but it is the case when you’re dealing with drainage, when you’re crossing an alluvial fan, it’s impossible to predict future flow patterns with certainty.”
It was this inability to predict the drainage pattern over the alluvial fan that in turn made unforeseen what resulted from the runoff which accompanied the heavy rains in 1969. What were those results? Item: the bench drains failed to operate as anticipated and because of this there was a greatly increased flow of water passing through the cross drains onto the Ellena property. Next item: excluding repairs to the bench drains, repairs to the freeway facilities damaged by the runoff and resulting again from a failure of the entire design to function as anticipated amounted to $112,551.25. Surely, if the storm consequences were foreseeable, the state’s own design engineers would not have knowingly built a $112,551.25 defect into their own structure.
In its opinion, the majority points to certain evidence as supporting the implied finding of foreseeability. More particularly, they remind that the state’s experts “testified repeatedly and without equivocation, however, that it was reasonably foreseeable, even to a layman, that when the surface waters reached the freeway embankment and were collected and discharged through the four cross drains, the concentrated flow issuing from them would result in erosion.” I respectfully suggest that this testimony begged the real question. As noted, the failure of the bench drains (unforeseen) greatly increased the flow of water through the cross drains and thus the volume of such increased flow was also unforeseen. This unforeseen volume of flow was what caused the damage, and in the face of this extrinsic, uncontradicted evidence, it defied reason and logic for the state’s witnesses to have continued to protest that the extent of the erosion which did occur was foreseeable. Such generalized statements of opinion in view of the physical facts were and are inherently unbelievable.
However, there is a more fundamental defect in the treatment of the case by the majority. In their characterizing the implied findings as including a finding “that the damages that did occur were the reasonably foreseeable consequences of the construction of the improvements as *265proposed,” they treat foreseeability as a matter of fact. It is on this basic determination that I disagree with the majority, for I see that issue as one of law.
To illustrate, suppose Ellena had taken his case to trial in 1967. Suppose at that trial he had offered to show by expert testimony the amount of erosion which could later have carved through his lands IF there were to be a storm of the size of the 1969 storm, IF the flow over the freeway cut were to be of the volume which it was in 1969, and IF the bench drains were to silt up and allow this flow to pour down the side of the cut and through the four cross drains. In the face of this offer of proof, the trial judge in the imagined trial in 1967 would have been faced with a legal determination in deciding whether to admit such evidence.
It seems clear, in view of the holding in People v. Ayon, 54 Cal.2d 217 [5 Cal.Rptr. 151, 352 P.2d 519], that the trial judge in the 1967 hypothetical trial would have rejected the offer of proof. He would have considered “[s]uch possible damages . .. too speculative to be recovered] in [that] action.” (Id., at p. 229.) “Too speculative” and “not foreseeable” are shorthand forms which describe a policy determination by the courts which operates to define the perimeters of permissible recovery.
Although Ellena abandoned his claim based on negligence, that body of law provides a ready and valid analogy of the kind of determination here involved. In negligence cases “foreseeability” is a matter of law. It is a label for the policy assessment which leads to or denies the imposition of duty to pay for the consequences of a failure to behave within a factually determined standard of care.
The cases talk of proximate cause and foreseeability interchangeably, and, while whether a person is exercising reasonable care is a matter of fact, what is deemed to be the proximate result of the want of reasonable care or what is deemed to be a foreseeable risk as a consequence of that want of reasonable care is a question of law.
Turning to the hallowed Cardozo language, “[t]he risk reasonably to be perceived defines the duty to be obeyed, and risk imports relation; it is risk to another or to others within the range of apprehension.” (Palsgraf v. Long Island R. Co., 248 N.Y. 339 [162 N.E. 99, 100, 59 A.L.R. 1253].)
*266Stated otherwise, in the absence of overriding policy considerations, a “[defendant owes a duty, in the sense of a potential liability for damages, only with respect to those risks or hazards whose likelihood made the conduct unreasonably dangerous, and hence negligent, in the first instance. [Citations.]” (Dillon v. Legg, 68 Cal.2d 728, 739 [69 Cal.Rptr. 72, 441 P.2d 912, 29 A.L.R.3d 1316].)
Surely, if “foreseeability” in negligence cases is the shorthand label for the sum of those seven policy considerations which are used as aids in defining the legal duty to behave with reasonable care (Commercial Standard Ins. Co. v. Bank of America, 57 Cal.App.3d 241, 248 [129 Cal.Rptr. 91]; Derrick v. Ontario Community Hospital, 47 Cal.App.3d 145, 153 [120 Cal.Rptr. 566]), then it can be conscientiously urged that “foreseeability” as used in cases of the kind at bench is likewise a question of law. In other words, if at the hypothetical trial of 1967 the trial judge would have been faced with a legal determination of whether the prospective erosion damages were too speculative or not foreseeable, how does that legal determination become metamorphosed into a factual one when the plaintiff seeks by a latter day effort to recoup his losses as prescribed in People v. Ayon, supra, 54 Cal.2d 217.
The majority opines that this reliance on Ayon is misplaced because the facts of that case involved an attempt by the land owner to claim damages for prospective events in the nature of temporary interference with access to his property during construction. I respectfully suggest that this factual difference is not a basis for distinguishing the legal principle involved. The court in Ayon excluded the offered proof not because the damages to be shown thereby would have resulted from a temporary interference but because such damages were too speculative. That is precisely the principle that is here applicable.
This is of course a close question turning upon a legal determination of policy, and it does not meet the issue to state as does the majority that “[o]ne need not possess the expertise of a hydraulic engineer to foresee that in the event of a prolonged heavy rainstorm, surface waters intercepted by the freeway embankment and discharged on the remaining lands through the cross drains would result in erosion at the points of discharge.” I have never suggested that erosion at the discharge end of the cross drains was not foreseeable at the time that the 1967 judgment was reached by stipulation. What was not foreseeable to the layman lacking the “expertise of a hydraulic engineer” was that the bench drains *267would fail and result in a collection of water into the cross drains far in excess of what the project’s designers themselves ever contemplated.
Turning again to the language of Ayon, “. . . if the improvement is not carried out as proposed, and, if as a result, unnecessary or unreasonable interference with appellants’ rights of possession or access occurs, then they may bring an action for such damages after they have occurred.” (People v. Ayon, supra, 54 Cal.2d 217, 229; italics added.)
Continuing, “[b]ut such damages cannot be recovered before they have occurred, for the obvious reason that they may never occur.” (Id., at p. 229.) Applied to the Ellena situation, years could have passed without there having been a storm the size of the one in 1969, and should such a storm never have occurred, there would have been no way for the defect in the design of the facility to have manifested itself. However, the storm did occur, the defect in design did result in substantial damage to the lands below the freeway, and under the rule laid down in Ayon, Ellena should have been allowed to prove his damages in the subsequent action.
Thus the misapplication of the rule of res judicata by the trial court operated to foreclose the plaintiff from litigating an issue which had not and could not have been litigated before the heavy rains in late 1969. The affirmance of the judgment below countenanced the denial of Ellena’s constitutional right to just compensation and as such should not be tolerated. I would reverse the judgment and allow the case to proceed to the second phase of the bifurcated trial.
A petition for a rehearing was denied May 11, 1977. McDaniel, J., was of the opinion that the petition should be granted. Appellant’s petition for a hearing by the Supreme Court was denied July 6, 1977.