Opinion

When a flood control levee fails to retain waters within its design capacity, are plaintiff property owners who suffer damage from the resultant flooding entitled to recover on a theory of inverse condemnation without showing that the damage was caused by unreasonable conduct on the part of the defendant public entities? We conclude that in the case of an unintended breach of a flood control improvement, plaintiffs are not entitled to recover absent proof of such unreasonable conduct.

Procedural Background

On the morning of February 21, 1980, a flood control levee on the left bank of the San Jacinto River located just downstream from its confluence with Bautista Creek gave way, freeing the river to flood parts of the City of San Jacinto. Several months later, 17 property owners damaged by the flood joined in filing a complaint in inverse condemnation against the Riverside County Flood Control District (District), which owned and operated the levee, as well as the State of California (State). The case was later consolidated with eight other like actions. Five of the individual plaintiffs’ cases were selected for trial to the court, the parties stipulating that the judgment in these cases would be binding in all the other actions. Trial was bifurcated, with the issues of inverse condemnation and liability to be tried first, followed if necessary by trial of the remaining issues of damages and compensation.

The cases proceeded to trial. At the close of plaintiffs’ case, the District moved for judgment pursuant to Code of Civil Procedure section 631.8 (the State elected not to move for judgment under § 631.8). The trial court granted the motion but reserved its statement of decision until completion of the State’s case-in-chief. The State put on its defense. The trial court thereafter issued statements of decision and entered judgment in favor of both the State and the District. Plaintiffs appealed.

The Court of Appeal affirmed the judgment on two separate grounds: 1) that there was no evidence the levee was the “proximate cause” of plaintiffs’ damages, and 2) that defendants had not “substantially participated” in the design or construction of the levee so as to be subject to inverse condemnation liability. We granted plaintiffs’ petition for review to determine the propriety of each facet of the Court of Appeal’s decision. We subsequently granted leave to six separate groups representing over one hundred public agencies to file amicus curiae briefs on behalf of defendants.

Factual Background

The evidence adduced at trial as to the factual causes of the flood was relatively straightforward and undisputed. The levee failure followed a series of heavy storms over a period of several days. Expert testimony and a United States Army Corps of Engineers study undertaken after the flood revealed that the breach was caused by the undermining of the levee foundation or “toe” through a process of “scouring,” i.e., the removal by water action of the foundation which underlay the levee. The scouring caused the rock-faced revetment on the riverward side of the levee to collapse, exposing the levee’s sand foundation to the eroding force of the floodwaters. A breach appeared and water began to escape. Within a relatively short period, the original gap of 20 feet widened to approximately 1,500 feet, or a quarter of a mile, in length.

The scouring process was caused in part by the presence of two other levees in the area, a “Ring” levee on the right bank of Bautista Creek where it meets the San Jacinto River and a Bureau of Indian Affairs levee on the right bank of the San Jacinto River across from the District levee. The configuration of the three levees created a “flow impingement” which forced the channel waters to flow against the District levee at a 25-degree angle. The flow impingement caused deep scouring which undermined the levee toe and resulted in the failure. Both the Ring levee and the Bureau of Indian Affairs levee were constructed prior to the District levee. Neither was owned or operated by the District.

Plaintiffs never adduced any substantial evidence that defendants had acted unreasonably in the design, construction, operation or maintenance of the levee, or that plaintiffs’ damages were caused by any act or failure to act on the part of defendants. Indeed, plaintiffs maintained that they were not required to prove the levee failure was the result of any act or omission of defendants. Plaintiffs asserted that inverse condemnation liability required proof only of the fact that the levee had failed to function within its design capacity, and that this failure caused their damages. In this regard, plaintiffs’ evidence showed that the levee was designed and constructed to contain a “standard project storm” or flood discharge of 86,000 cubic feet per second (cfs). The maximum discharge at the time of the breach was approximately 25,000 cfs, well within the levee’s design capacity.

Though the trial court issued separate statements of decision as to the State and the District, its findings on the issue of inverse condemnation liability were essentially identical. The trial court found, in pertinent part, as follows: “1. Plaintiffs incurred damage to their property, which was caused by an unintended breach in the project levee on February 21, 1980. [Í1] 2. The design capacity of the project levee was a water flow of 86,000 cubic feet per second (CFS). At the time of the breach, the flow in the river channel was approximately 25,000 CFS and the project levee was operating as designed and constructed. [fl] 3. Plaintiffs’ damages did not result from an overflow of channel waters in the San Jacinto River. Rather, the losses resulted from a failure in a portion of the project levee, by reason of a breach at a particular point in the levee which allowed the channel waters to escape the river channel and flow onto plaintiffs’ property. The channel waters did not overtope [szc] the levee.”

The trial court’s findings continued as follows: “fl[] 4. Before construction of the project levee in 1962, the land where plaintiffs’ properties are situated were [szc] subject to periodic flooding by waters of the San Jacinto River. Such land is located within the area which is to be protected by the project levee from San Jacinto flood waters flowing at a maximum volume per second, namely 86,000 CFS. [fl] 5. Plaintiffs or their predecessors built businesses, structures, and/or conducted businesses and/or built homes and/or purchased homes and resided on such lands after 1962 and prior to February 21, 1980. They and their predecessors reasonably relied on the project levee containing the San Jacinto River waters within the river channel, at least to a volume capacity of 86,000 CFS. During such period they reasonably relied on the project levee not failing at any particular point so as to allow river waters to escape and flood their properties. In acting on such reliance, they made substantial expenditures of moneys and other material resources, [fl] 6. The project levee did not increase the risk of damage or impose any easement, servitude, or other burden on plaintiffs’ property.”

Notwithstanding the trial court’s finding that the channel waters had not exceeded the levee’s design capacity, the court concluded that the District was “neither a substantial concurring, nor a sole, independent cause of the . . . breach in the project levee . . .” and further concluded that the breach “was not a result of any act, omission or failure to act by the District in carrying out its operation and maintenance responsibilities . . . .” As to the State, the trial court concluded that “no State conduct . . . was a substantial contributing cause of [plaintiffs’] damages and losses.”

Based on additional findings relating to defendants’ respective participation in the planning, design, funding, construction, operation and maintenance of the levee, the trial court also concluded as a separate basis of decision that neither the District nor the State was a substantial participant in the levee project so as to be subject to inverse condemnation liability for plaintiffs’ damages. (See Chatman v. County of Alameda Flood Control etc. Dist. (1986) 183 Cal.App.3d 424, 431-432 [ 228 Cal.Rptr. 257 ]; Sheffet v. County of Los Angeles (1970) 3 Cal.App.3d 720, 734-735 [ 84 Cal.Rptr. 11 ].)

In affirming the judgment in favor of defendants, the Court of Appeal focused on findings (4) and (6) above. The court reasoned as follows: Because plaintiffs’ property had been subject to periodic flooding prior to construction of the District levee (finding No. 4), and because the levee had not increased the risk or the extent of the flooding (finding No. 6), then it followed that the levee had not “proximately caused the plaintiffs’ damages, i.e., had [not] caused more flooding on plaintiffs’ property than there would have been without the levee . . . .”

The Court of Appeal also affirmed the trial court’s holding that neither the District nor the State was a substantial participant in the design or construction of the levee so as to be subject to inverse condemnation liability.

As will appear, we conclude that the judgment in favor of defendants was correct, although not for the reasons stated by the Court of Appeal. Accordingly, we shall affirm the judgment.

Discussion

A. Inverse Condemnation Liability and Proximate Cause

Article I, section 19 (formerly § 14) of the California Constitution provides: “Private property may be taken or damaged for public use only when just compensation . . . has first been paid . . . .’’In the landmark case of Albers v. County of Los Angeles (1965) 62 Cal.2d 250, 263-264 [ 42 Cal.Rptr. 89 , 398 P.2d 129 ], we construed this provision to mean that, with two exceptions (to be discussed below), “any actual physical injury to real property proximately caused by [a public] improvement as deliberately designed and constructed is compensable under article I, section 14, of our Constitution whether foreseeable or not.” As we subsequently explained in Holtz v. Superior Court (1970) 3 Cal.3d 296 [ 90 Cal.Rptr. 345 , 475 P.2d 441 ], Albers fully recognized the fears that an open-ended, absolute liability rule of inverse condemnation could inhibit the construction of beneficial public improvements: “Thus we limited our holding of inverse condemnation liability, absent fault, to ‘physical injuries of real property’ that were ‘proximately caused’ by the improvement as deliberately constructed and planned.” (Id. at p. 304.) In Holtz , we also underscored the fundamental “policy” basis of the constitutional requirement of just compensation: “‘The decisive consideration is whether the owner of the damaged property if uncompensated would contribute more than his proper share to the public undertaking.’ In other words, the underlying purpose of our constitutional provision in inverse—as well as ordinary—condemnation is ‘to distribute throughout the community the loss inflicted upon the individual .. . .’ ” ( 3 Cal.3d at p. 303 .)

Plaintiffs here contend that the evidence adduced at trial satisfied the Albers criteria. The undisputed evidence clearly established that plaintiffs incurred actual physical injury to real property. In addition, the trial court found that the breach was “not caused by volume, speeds and heights of the water beyond the design capacity of the levees,” i.e., the water flow at the time of the breach was within the range the levee was designed to handle.

However, based upon the trial court’s additional findings that plaintiffs’ property was subject to flooding prior to construction of the levee and that the levee had not increased the risk of flooding, the Court of Appeal concluded that the element of proximate causation was missing. In so holding the Court of Appeal was mistaken.

After our decision in Albers , Professor Arvo Van Alstyne authored a seminal article on inverse condemnation liability in which he observed that Albers’s proximate cause requirement “involves a troublesome conceptual premise.” (Van Alstyne, Inverse Condemnation: Unintended Physical Damage (1969) 20 Hastings L.J. 431 , 435-438.) Our decision in Albers, supra, 62 Cal.2d 250, contained the seeds of confusion through its combination of “proximate cause” terminology with the elimination of foreseeability as an element of inverse condemnation. Noting this paradox, Professor Van Alstyne suggested that the true measure of proximate cause should be stated in terms of a “ ‘substantial’ cause-and-effect relationship which excludes the probability that other forces alone produced the injury.” (Van Alstyne, supra, 20 Hastings L.J. at p. 436, italics added.) Although the issue was not raised, we subsequently acknowledged in Holtz “the greater precision, as Professor Van Alstyne contends (id. at p. 436) in restating this element of the Albers test in terms of ‘substantial’ causation.” ( 3 Cal.3d at p. 304, fn. 9 .)

Subsequent to Holtz , and in reliance thereon, several Court of Appeal decisions held that inverse condemnation liability may be established where the public improvement constitutes a substantial cause of the damage, albeit only one of several concurrent causes. (See Souza v. Silver Development Co. (1985) 164 Cal.App.3d 165, 171 [ 210 Cal.Rptr. 146 ]; Ingram v. City of Redondo Beach (1975) 45 Cal.App.3d 628, 633-634 [ 119 Cal.Rptr. 688 ]; Blau v. City of Los Angeles (1973) 32 Cal.App.3d 77, 84-85 [ 107 Cal.Rptr. 727 ].) In Ingram , for example, plaintiffs sued the defendant city for damages incurred when heavy rains caused the earthen wall of a drainage sump to collapse, allowing flood waters to inundate plaintiffs’ homes. (45 Cal.App.3d at pp. 630-631.) The trial court rendered judgment for the city, finding that “no damages to plaintiff occurred as a proximate result of any public improvement of defendant . . . .” (Id. at p. 630.)

The Court of Appeal reversed. Relying on our endorsement of the “substantial cause” test in Holtz , the court reasoned as follows: “There is no controversy over the fact that the wall of the sump gave way, releasing a flood onto plaintiffs’ properties. This clearly could be a substantial factor in causing the claimed damages .... If, however, the damage is caused, solely, by an unforeseen and supervening cause, then liability will not follow.” ( 45 Cal.App.3d at p. 633 , italics added.)

Thus, in order to establish a causal connection between the public improvement and the plaintiff’s damages, there must be a showing of “ ‘a substantial cause-and-effect relationship excluding the probability that other forces alone produced the injury.’ [Citations.]” (Souza v. Silver Development Co., supra, 164 Cal.App.3d at p. 171 , fn. omitted.) Where independently generated forces not induced by the public flood control improvement—such as a rainstorm—contribute to the injury, proximate cause is established where the public improvement constitutes a substantial concurring cause of the injury, i.e., where the injury occurred in substantial part because the improvement failed to function as it was intended. The public improvement would cease to be a substantial contributing factor, however, where it could be shown that the damage would have occurred even if the project had operated perfectly, i.e., where the storm exceeded the project’s design capacity. In conventional terminology, such an extraordinary storm would constitute an intervening cause which supersedes the public improvement in the chain of causation.

Turning to the case at bar, the record amply supports plaintiffs’ claim that the levee was a substantial concurring cause of the damages to their property. Plaintiffs’ evidence established, and the trial court expressly found, that the design capacity of the levee was 86,000 cfs, that the maximum flow in the channel at the time of the breach was approximately 25.000 cfs, and that the breach was not caused by volume, speeds or heights of water beyond the design capacity of the levee. Thus, notwithstanding the heavy storms which preceded the breach, plaintiffs demonstrated that the levee’s failure to function as intended constituted a substantial concurring— or proximate—cause of the damages.

Nevertheless, defendants contend, as the Court of Appeal expressly concluded, that the District levee was not and could not have been the proximate cause of the damages, because plaintiffs’ property was subject to flooding prior to construction of the levee, and because the property would have flooded even in the absence of the levee. There is no merit to this reasoning.

The fact that plaintiffs’ property was subject to flooding prior to construction of the levee is not determinative of the question of whether the failure of the levee in this case caused damage to the plaintiffs. As the trial court here found, plaintiffs “reasonably relied on the project levee containing the River waters within the River channel, at least to a volume capacity of 86.000 CFS,” and “acting on such reliance . . . made substantial expenditures of moneys and other material resources.” By inducing plaintiffs to make substantial improvements in reliance on its providing protection to a flow of 86,000 CFS, and then failing to provide such protection, the levee plainly constituted a “substantial cause” of plaintiffs’ damages. (Holtz v. Superior Court, supra, 3 Cal.3d at p. 304, fn. 9 .)

Nor is there merit to defendants’ contention that proximate cause was not established because plaintiffs failed to demonstrate the improvement increased the risk of flooding, or diverted floodwaters that would not otherwise have crossed plaintiffs’ property. Inverse condemnation liability for failure of flood control projects is not predicated upon proof that the public improvement made a preexisting hazard worse. In this respect, defendants’ reliance on two Court of Appeal decisions, Tri-Chem, Inc. v. Los Angeles County Flood Control Dist. (1976) 60 Cal.App.3d 306 [ 132 Cal.Rptr. 142 ] and Shaeffer v. State of California (1972) 22 Cal.App.3d 1017 [ 99 Cal.Rptr. 861 ], to argue to the contrary is misplaced.

In Tri-Chem, Inc. v. Los Angeles County Flood Control Dist., supra, 60 Cal.App.3d 306 , the plaintiffs sought damages on a theory of inverse condemnation after a drainage system operated by the district overflowed and flooded their property. The jury found in favor of the plaintiffs. On appeal, the District argued that the levee failure “was not a proximate cause of plaintiffs’ harm” (id. at p. 310, italics added) since the drainage system had a design capacity of only 550 to 700 cfs, whereas runoff during the storm which resulted in the flood reached approximately 1,096 cfs. (Id. at p. 309.) The Court of Appeal agreed, stating: “All that the record here shows is that a more extensive flood control system might have diverted more water from its natural flow across plaintiffs’ properties. There is no evidence that the system did not work as it was designed to function . . . .” (Id. at pp. 311-312, italics added.)

Shaeffer v. State of California, supra, 22 Cal.App.3d 1017 , presents a factually distinct but analytically analogous scenario to Tri-Chem. Plaintiffs sought recovery in inverse condemnation for damages caused by a flood along the Feather River. The evidence showed that the State operated a flood control project along the river which consisted of a levee system and the Oroville Dam. At the time of the flood, however, the dam was only partially completed. Thus, certain property owners were completely protected by the partially completed dam, while others like plaintiffs would have received greater protection had the dam been completed. The trial court concluded that the plaintiffs had sustained no damages attributable to the flood control project of the State.

In affirming the judgment, the Court of Appeal noted that “a public entity is not liable for damages merely because its flood control improvements do not provide the same degree of protection to all property owners in the area.” ( 22 Cal.App.3d at p. 1021 .) There was no evidence, the court noted, that the flood control improvements had failed to operate as intended. The mere fact that “other properties receive[d] more benefit than those of plaintiffs because of the phase and stages of the flood control project” was not a sufficient basis of inverse condemnation liability. (Ibid.)

Thus, it is readily apparent that the results in both Tri-Chem and Shaeffer are perfectly consistent with the proximate cause analysis we have outlined. In each case the court concluded that proximate cause was absent—not because the risk of floods preexisted the flood control projects, nor because the projects had failed to increase the risk or the extent of the floods—but rather because there was no evidence that the projects failed to function as intended-, in both cases the flooding occurred in spite of the flood control improvements, not because of them.

Here, in contrast, plaintiffs adduced substantial evidence that the District levee failed to function within its design capacity and thus constituted a substantial concurring—or proximate—cause of the damages. The Court of Appeal’s holding to the contrary was erroneous.

Having thus established that the levee failed within its design capacity, and that such failure constituted a substantial cause of their damages, plaintiffs insist that they are entitled to recover as a matter of law on a theory of absolute liability under Albers, supra, 62 Cal.2d 250 , without any further showing that the flooding was the result of any unreasonable act or omission attributable to defendants. As explained below, we reject this contention.

B. The Requirement of Unreasonable Conduct in Flood Control Cases

Prior to our decision in Albers, supra, 62 Cal.2d 250 , it was standard judicial practice to analyze inverse condemnation liability by analogy to tort and property law principles. (See Mandelker, Inverse Condemnation: The Constitutional Limits of Public Responsibility (1966) Wis.L.Rev. 3, 6-9; Van Alstyne, supra, 20 Hastings LJ. at pp. 440-442.) This approach was in part predicated on the general understanding that inverse condemnation liability was limited to cases in which a private party would be held liable under like circumstances. (See, e.g., Archer v. City of Los Angeles (1941) 19 Cal.2d 19, 24 [ 119 P.2d 1 ] [“If the property owner would have no cause of action were a private person to inflict the damage, he can have no claim for compensation from the state.”]; Clement v. State Reclamation Board (1950) 35 Cal.2d 628, 637 [ 220 P.2d 897 ] [“[Pjlaintiffs have no right to compensation under article I, section 14, if the injury is one that a private party would have the right to inflict without incurring liability.”]; accord Youngblood v. Los Angeles County Flood Control Dist. (1961) 56 Cal.2d 603, 608 [ 15 Cal.Rptr. 904 , 364 P.2d 840 ]; Bauer v. County of Ventura (1955) 45 Cal.2d 276, 282-283 [ 289 P.2d 1 ]; Granone v. County of Los Angeles (1965) 231 Cal.App.2d 629, 645 [ 42 Cal.Rptr. 34 ].)

As earlier noted, however, Albers shifted the focus in inverse condemnation cases from the common law to the Constitution. As we subsequently explained in Holtz v. Superior Court, supra, 3 Cal.3d at page 302 : “In Albers v. County of Los Angeles . . . this court explicitly rejected the notion that there need be a congruence between public and private liability in inverse condemnation actions.” The critical issue, we held, was not whether the plaintiff would have a cause of action under tort or property law if the damage were inflicted by a private person, but rather whether the plaintiff should recover “as a matter of interpretation and policy [under] article I, section 14, of the Constitution . . . .” (Albers v. County of Los Angeles, supra, 62 Cal. 2d at p. 262 ; Holtz v. Superior Court, supra, 3 Cal.3d at p. 303 .) As Holtz explained, “In [inverse condemnation cases] the purposes of the constitutional clause, rather than the limits established by a rule of statutory or common law allocating rights and responsibilities between private parties, must fix the extent of a public entity’s responsibility.” ( 3 Cal.3d at p. 302 .)

Balancing the constitutional purposes of the taking clause with the “competing considerations which caution against an open-ended, ‘absolute liability’ rule of inverse condemnation,” we held that “any actual physical injury to real property proximately caused by the improvement as deliberately designed and constructed is compensable under article I, section 14, of our Constitution, whether foreseeable or not.” (Albers v. County of Los Angeles, supra, 62 Cal.2d at pp. 261-262; Holtz v. Superior Court, supra, 3 Cal.3d at pp. 302-304.)

In so holding, however, we also explicitly “identified two strains of decisions in which the urgency or particular importance of the governmental conduct involved was so overriding that considerations of public policy inveighed against a rule rendering the acting public entity liable absent fault.” (Holtz v. Superior Court, supra, 3 Cal.3d at pp. 304-305.) Of the two doctrinal categories expressly exempted from Albers’s generalized strict liability rule, the second, or so-called Archer exception (Archer v. City of Los Angeles, supra, 19 Cal.2d 19 ), “encompassed those cases in which the state at common law ‘had the right to inflict the damage.’ ” (Holtz v. Superior Court, supra, 3 Cal.3d at p. 305 , quoting from Albers v. County of Los Angeles, supra, 62 Cal.2d at p. 262.) (Italics in original.)

As we later explained in Holtz , the “doctrine of the common law ‘right to inflict damage,’ emanating from the complex and unique province of water law, has been employed in only a few restricted situations, generally for the purpose of permitting a landowner to take reasonable action to protect his own property from external hazards such as floodwaters.” ( 3 Cal.3d at p. 306 , italics added.) Frequently referred to as the “common enemy” doctrine, the notion is that the owner of land subject to flooding has the right to erect defensive barriers and that any injury caused thereby to lower landowners as the result of the increased discharge or velocity of water is considered damnum absque injuria. (See, e.g., Clement v. State Reclamation Board, supra, 35 Cal.2d at pp. 635-636; The Weinberg Co. v. Bixby (1921) 185 Cal. 87, 95 [ 196 P. 25 ]; McDaniel v. Cummings (1890) 83 Cal. 515, 519-521 [ 23 P. 795 ].) The unique legal privilege which cloaks such protective measures undoubtedly reflects the overriding interest, in a developing economy, of making land freely available for settlement and improvement. (Holtz v. Superior Court, supra, 3 Cal.3d at p. 307, fn. 11 .)

Although, as noted, we expressly excepted the Archer line of decisions from Albers’s rule of liability without fault, we also cautioned against the “facile” assumption that an activity which is “privileged” when performed by a private party is equally privileged when undertaken by a public entity. (Holtz v. Superior Court, supra, 3 Cal.3d at pp. 307-308, fn. 13.) Different policy considerations, we noted, inform the public and the private spheres. While “certain socially beneficial conduct may appropriately be designated ‘privileged’ for private individuals in order that they will not be deterred from undertaking the activity, the public entity may continue to engage in this same ‘privileged’ activity even if it must bear the loss of resulting damages.” (Ibid., italics added.)

Thus, while we recognized in Albers that strict inverse condemnation liability may not be appropriate in the case of flood control improvements, we emphasized in Holtz that such improvements should not be cloaked with the same immunity as private flood control measures. The question, therefore, is what standard applies in such cases. We draw the answer from prior case law, public policy and common sense.

On the one hand, a public agency that undertakes to construct or operate a flood control project clearly must not be made the absolute insurer of those lands provided protection. On the other hand, the damage potential of a defective public flood control project is clearly enormous. Therefore, as we observed in Holtz , the courts have consistently held that “even when a public agency is engaged in such ‘privileged activity’ as the construction of barriers to protect against floodwaters, it must [at least\ act reasonably and non-negligently. [Citations.]” (Holtz v. Superior Court, supra, 3 Cal.3d at p. 307, fn. 12 , italics added; see also Shaeffer v. State of California, supra, 22 Cal.App.3d at p. 1021 .) Contrary to plaintiffs’ position, the fact that a dam bursts or a levee fails is not sufficient, standing alone, to impose liability. However, where the public agency’s design, construction or maintenance of a flood control project is shown to have posed an unreasonable risk of harm to the plaintiffs, and such unreasonable design, construction or maintenance constituted a substantial cause of the damages, plaintiffs may recover regardless of the fact that the project’s purpose is to contain the “common enemy” of floodwaters. (See Bauer v. Ventura County, supra, 45 Cal.2d at pp. 285-286; House v. Los Angeles Flood Control Dist., supra, 25 Cal.2d at pp. 395-396; San Gabriel V.C. Club v. County of Los Angeles, supra, 182 Cal. at pp. 399-400; Granone v. County of Los Angeles, supra, 231 Cal.App.2d at p. 647 .)

Permitting recovery where the public entity’s unreasonable conduct constitutes a substantial cause of damage to property owners negates the apprehension commonly associated with a rule of absolute liability—the discouragement of beneficial flood control improvements—yet properly compensates for losses unfairly incurred. As Professor Van Alstyne observed: “Mindful of the enormous damage-producing potential of defective public flood control projects, the courts have insisted that public agencies must act reasonably in the development of construction and operational plans so as to avoid unnecessary damage to private property. Reasonableness, in this context, is not entirely a matter of negligence, but represents a balancing of public need against the gravity of private harm.” (Van Alstyne, supra, 20 Hastings L.J. at p. 455, fns. omitted.)

The reasonableness of the public agency’s conduct must be determined on the facts of each individual case, taking into consideration the public benefit and the private damages in each instance. (See Keys v. Romley (1966) 64 Cal.2d 396, 409-410 [ 50 Cal.Rptr. 273 , 412 P.2d 529 ].) Inverse condemnation liability ultimately rests on the notion that the private individual should not be required to bear a disproportionate share of the costs of a public improvement. (Holtz v. Superior Court, supra, 3 Cal.3d at p. 303 .) Thus, compensation for damages incurred as the result of a flood control agency’s unreasonable conduct, measured in light of this balancing test, constitutes no more than a reimbursement to the damaged property owners of their contribution of more than their “proper share [to] the public undertaking.” (Ib id.)

Plaintiffs contend that recent Court of Appeal decisions appear, nevertheless, to have applied an absolute liability standard to cases involving flood damages. A review of these decisions reveals that in each case cited the public improvement resulted in a diversion of surface or flood waters from their natural channel or drainage. (See Yee v. City of Sausalito (1983) 141 Cal.App.3d 917 [ 190 Cal.Rptr. 595 ] [storm drainage system which diverted surface waters across plaintiff’s property ruptured, allowing water to seep into soil adjacent to plaintiff’s property, resulting in massive soil subsidence]; Marin v. City of San Rafael (1980) 111 Cal.App.3d 591 [ 168 Cal.Rptr. 750 ] [underground drainage pipe which diverted surface waters under plaintiffs’ property burst, causing property damage]; Imperial Cattle Co. v. Imperial Irrigation Dist. (1985) 167 Cal.App.3d 263 [ 213 Cal.Rptr. 622 ] [irrigation system overflowed, diverting surface waters onto plaintiff’s property]; McMahan’s of Santa Monica v. City of Santa Monica (1983) 146 Cal.App.3d 683 [ 194 Cal.Rptr. 582 ] [water main which diverted waters under plaintiff’s building ruptured, damaging the building].)

As earlier noted, the “common enemy” doctrine did not confer the right to divert or obstruct waters from their natural channels or drainages, and several pre- as well as post -Albers decisions, relying on this principle, appear to have endorsed a rule of inverse liability without fault where such diversions were present. (See, e.g., Youngblood v. Los Angeles County Flood Control Dist., supra, 56 Cal.2d at p. 607 [“[W]hen waters are diverted by a public improvement from a natural watercourse onto adjoining lands the agency is liable for the damage . . . even though no negligence could be attributed to the installation of the improvement.”]; see also Yee v. City of Sausalito, supra, 141 Cal.App.3d at pp. 920-923.)

We need not examine the validity of these decisions here, for there was no evidence presented that the District levee affirmatively diverted or burdened plaintiffs’ property with floodwaters in excess of those which would have escaped in the absence of the levee. It is doubtful, however, whether evidence of an unintended “diversion”—an elusive concept to begin with (see Van Alstyne, supra, 20 Hastings L.J. at pp. 460-461)—would elevate the test of inverse condemnation liability to absolute liability, rather than a reasonableness standard. As earlier discussed, the purposes of the Constitution, rather than the rules “emanating from the complex and unique province of water law,” must fix the extent of a public entity’s responsibility. (Holtz v. Superior Court, supra, 3 Cal.3d at p. 306 .) It is sufficient for our purposes here to hold that when a public flood control improvement fails to function as intended, and properties historically subject to flooding are damaged as a proximate result thereof, plaintiffs’ recovery in inverse condemnation requires proof that the failure was attributable to some unreasonable conduct on the part of the defendant public entities.

Defendants contend that inverse condemnation damages should not be available even where the public entity’s conduct is unreasonable unless plaintiffs can establish that the improvement affirmatively diverted floodwaters onto property that would not otherwise have been affected. There is no merit to this contention. As noted above, Albers explicitly rejected the notion that public and private liability principles are necessarily coextensive. (Albers v. County of Los Angeles, supra, 62 Cal.2d at pp. 260-262.) Those historical policy considerations which conditioned private water law actions on an affirmative diversion or obstruction of stream, surface or floodwaters do not necessarily apply to claims for “just compensation” against the government. (Holtz v. Superior Court, supra, 3 Cal.3d at pp. 306-307.) When a public agency erects improvements to protect against floodwaters, it must, at a minimum, act reasonably; its liability for unreasonable acts or omissions does not require proof that the public agency affirmatively diverted waters where they would not otherwise have flowed.

Conclusion

There is perhaps no rule of review more firmly established than the principle that a ruling or decision correct in law will not be disturbed on appeal merely because it was given for the wrong reason. If correct upon any theory of law applicable to the case, the judgment will be sustained regardless of the considerations that moved the lower court to its conclusion. (Green v. Superior Court (1985) 40 Cal.3d 126, 138 [ 219 Cal.Rptr. 186 , 707 P.2d 248 ]; D’Amico v. Board of Medical Examiners (1974) 11 Cal.3d 1, 19 [ 112 Cal.Rptr. 786 , 520 P.2d 10 ].) Accordingly, notwithstanding the Court of Appeal’s error in concluding that plaintiffs failed to establish proximate cause, the decision must be upheld because plaintiffs did not proceed on a negligence theory, and with respect to their inverse condemnation claim adduced no substantial evidence that the flooding was the result of any unreasonable act or omission attributable to defendants. The judgment in favor of defendants was correct as a matter of law.

The judgment of the Court of Appeal is affirmed.

Lucas, C. J., Broussard, J., Panelli, J., Arguelles, J., and Eagleson, J., concurred.

I agree with the majority that plaintiffs in their inverse condemnation claim established they suffered actual physical damage to their real property that was directly or substantially caused by flood control improvements operating as deliberately planned and built. But I disagree that plaintiffs nonetheless failed to prove their case by omitting to allege and establish “unreasonable conduct.”

I.

As the majority correctly observe, the interpretation of article I, section 14, of the California Constitution long was guided by principles of private law. (Van Alstyne, Inverse Condemnation: Unintended Physical Damage (1969) 20 Hastings L.J. 431 (hereafter Van Alstyne).) Early decisions are legion that viewed the just compensation clause exclusively as a vehicle to surmount difficulties of sovereign immunity; the provision thus was understood “not to create new causes of action but only to give the private property owner a remedy he would not otherwise have against the state . . . .” (Bauer v. County of Ventura (1955) 45 Cal.2d 276, 282 [ 289 P.2d 1 ]; see also Van Alstyne, p. 441.) So construed, the clause articulated no independent legal relationship between the state and its citizenry.

Two decades ago we decisively rejected this reasoning and embraced a competing conception of article I, section 14. In the seminal case of Albers v. County of Los Angeles (1965) 62 Cal.2d 250, 263 [ 42 Cal.Rptr. 89 , 398 P.2d 129 ] (hereafter Albers), this court reaffirmed earlier precedent declaring that the constitutional provision expresses an independent policy “ ‘to distribute throughout the community the loss inflicted upon the individual by the making of the public improvements,’” albeit within parameters imposed by the countervailing risk of “ ‘impeding] . . . beneficial public improvements because of the greatly increased cost.’ ” (Quoting Bacich v. Board of Control (1943) 23 Cal.2d 343, 350 [ 144 P.2d 818 ]; see, e.g., Reardon v. San Francisco (1885) 66 Cal. 492, 505 [ 6 P. 317 ].) As we subsequently explained in Holtz v. Superior Court (1970) 3 Cal.3d 296 [ 90 Cal.Rptr. 345 , 475 P.2d 441 ], this cost-spreading objective, “rather than the limits established by a rule of statutory or common law allocating rights and responsibilities between private parties, must fix the extent of a public entity’s responsibility.” (Id. at p. 302.)

After considering the various interests implicated by this objective, we concluded in Albers that with two exceptions, “any actual physical injury to real property proximately caused by the improvement as deliberately designed and constructed is compensable under article I, section 14, of our Constitution whether foreseeable or not.” (Albers, pp. 263-264.) The tort concept of fault—contingent on the foreseeability of injury—thus was omitted from the Albers standard for inverse condemnation liability. As the late Professor Van Alstyne observed in a related context, there was good reason for this omission: “Consistent with the intent of the framers of the just compensation clause to protect property interests against even the best intentioned exercises of public power, it avoids as well a fruitless search for the somewhat artificial moral elements inherent in the tort concepts of negligence and intentional wrongs.” (Van Alstyne, p. 495, fn. omitted.)

The Albers court nevertheless recognized two limited circumstances “in which the urgency or particular importance of the governmental conduct involved was so overriding that considerations of public policy inveighed against a rule rendering the acting public entity liable absent fault. The first exception . . . involved noncompensable damages ‘inflicted in the proper exercise of the police power’; the second exception, the Archer exception (Archer v. City of Los Angeles (1941) 19 Cal.2d 19, 24 [ 119 P.2d 1 ] [hereafter Archer]), encompassed those cases in which the state at common law ‘had the right to inflict the damage.’ ” (Holtz v. Superior Court, supra, 3 Cal.3d at pp. 304-305, italics in original, citation omitted.) The latter exception, “emanating from the complex and unique province of water law, has been employed in only a few restricted situations, generally for the purpose of permitting a landowner to take reasonable action to protect his own property from external hazards such as floodwaters.” (Id. at p. 306.) The majority embrace this exception, as modified, as the ratio decidendi of the case, reasoning that a public entity’s conduct with regard to flood control improvements is privileged if “reasonable” and in that case is beyond the reach of the Albers rule. They proceed to hold that “unreasonable conduct” must be proved to recover in cases falling within the modified Archer exception, and conclude that plaintiffs failed to satisfy the required showing.

II.

Unlike the majority, I would not attempt to save the Archer “right to inflict damage” exception, grounded as it is in anachronistic tort and property rules for determining liability for defective flood control improvements. The Albers court inexplicably endorsed the exception and its vestigial reasoning while elsewhere assailing the very principles upon which the exception rests. (See Albers, pp. 262-263.) Although I concur that flood control is a sufficiently unique and compelling public endeavor to fall outside the strict liability standard of Albers , the alternative rule of liability for such undertakings should nevertheless derive from the objectives of the just compensation clause itself rather than inapposite analogies to private law. (Van Alstyne, pp. 498-499.)

The Archer exception traces its origins to the “common enemy” doctrine of water law, which recognizes a private landowner’s right to protect his land by erecting defensive barriers against the “common enemy” of flood-waters. (Cl ement v. State Reclamation Board (1950) 35 Cal.2d 628, 635-636 [ 220 P.2d 897 ].) Such conduct is privileged, and any resulting injury inflicted on downstream property owners is not compensable. (Id. at p. 636.) The initial rationale for the rule was to foster private land development: “Not to permit an upper land owner to protect his land against the stream would be in many instances to destroy the possibility of making the land available for improvement or settlement and condemn it to sterility and vacancy. Such a rule would seriously interfere with the development of the country.” (San Gabriel V. C. Club v. Los Angeles (1920) 182 Cal. 392, 401 [ 188 P. 554 , 9 A.L.R. 1200 ].) The privilege, however, has never applied to obstructions or diversions of a natural channel, or to the creation of new channels “by which the natural stream waters of the river are carried onto the lands of another that would have been protected therefrom but for the creation of the artificial channel. . . .” (Clement v. State Reclamation Board, supra, 35 Cal.2d at p. 636 .)

Relying on the now-discredited principle that the just compensation clause waived sovereign immunity but created no new causes of action, the court in Archer reasoned that the private law “common enemy” privilege should similarly protect the state in its construction of flood control improvements. (Archer, pp. 24-25.) However, the Archer exception “is not applied as an unlimited rule of privileged self-help. Mindful of the enormous damage-producing potential of defective public flood control projects, the courts have insisted that public agencies must act reasonably in the development of construction and operational plans so as to avoid unnecessary damage to private property.” (Van Alstyne, p. 455, fn. omitted; see Holtz v. Superior Court, supra, 3 Cal. 3d at p. 307, fn. 12 .)

While declining to “examine the continued validity of . . . these long-lived doctrines, [or] question the applicability of the policies supporting the ‘privileged’ status of a particular private activity in the context of public improvements,” the court in Holtz registered its uneasiness with the Archer exception by noting that “significant policy differences . . . exist between the private and the public spheres. Although certain socially beneficial conduct may appropriately be designated ‘privileged’ for private individuals in order that they will not be deterred from undertaking the activity, the public entity may continue to engage in this same ‘privileged’ activity even if it must bear the loss of resulting damages. In such a case, there may well be no reason to depart from the general constitutional guarantee of just compensation.” (Holtz v. Superior Court, supra, 3 Cal.3d at pp. 307, 308, fn. 13.)

As Van Alstyne illustrates, the concerns voiced by the Holtz court were well-founded: “The rationale of the ‘common enemy’ rule . . . is of dubious validity when considered in the context of governmentally administered flood control projects developed for the collective protection of entire regions. ... [H] Piecemeal construction, often an inescapable feature of such major flood control projects, creates the possibility of interim damage to some lands left exposed to flood waters while others are within the protection of newly erected works. Indeed, the partially completed works, by preventing escape of waters that previously were uncontrolled, actually may increase the volume and velocity of flooding .... The prevailing private law doctrine embodied in the ‘common enemy’ rule, however, imposes no duty upon the public entity to provide complete protection against flood waters .... Increased or even ruinous damage incurred by the temporarily unprotected owners, due to the inability of the improvements to provide adequate protection to all, therefore, is not a basis of inverse liability. The constitutional promise of just compensation for property damage for public use thus yields to the overriding supremacy of an anomalous rule of private law.” (Van Alstyne, pp. 500-501, fns. omitted.)

The Archer exception does not merely bar compensation for certain deserving plaintiffs, however, but elsewhere generates unreasonably expansive public liability. Tied to the prerequisites of the common law privilege, Archer offers no protection for flood control improvements which “divert” stream waters from their natural drainage. (Clement v. State Reclamation Board, supra, 35 Cal.2d at pp. 637-638.) Thus any damage resulting from the diversion of water pursuant to a flood control program remains compensable under the strict liability standard of Albers. (Ibid.; see Tyler v. Tehama County (1895) 109 Cal. 618 [ 42 P. 240 ].) While sensible in the private context with respect to the interests of adjacent landowners, the distinction becomes much less reasonable in the broader context of public flood control systems: “Stream diversions . . . may be integral features of coordinated flood control . . . projects undertaken on a large scale by public entities organized for that very purpose. Where this is so, the community may suffer more by general fiscal deterrents resulting from indiscriminately imposed strict liabilities than by specifically limited liabilities determined by the reasonableness of the risk assumptions underlying each diversion.” (Van Alstyne, p. 502, fn. omitted.)

The distinction between “common enemy” and “diversion” cases not only fails to meaningfully address the interests implicated by inverse condemnation, but furthermore turns on the relatively arbitrary characterization of the particular facts of each case: “If the improvements are regarded as causing an alteration in the direction of force of the normal current within the channel, they may readily be thought of as having ‘diverted’ the stream. This approach supports a holding of inverse liability without fault for resulting downstream erosion of the banks. By describing the channel improvements as measures to fight off the common enemy of flood waters, however, attention is focused upon the issue of fault and the alleged defective nature of the improvement plan. The result is to make liability vel non turn ostensibly upon the unarticulated premises that control the classification process, rather than upon a conscientious appraisal of the relativity of public advantage and private harm in the particular factual situation.” (Van Alstyne, p. 462, fn. omitted.) In the case at bar, for example, the failed levee was positioned to effectively block oncoming water channeled by two preexisting levees on the opposing bank; the water “diverted” by the new levee eventually scoured its foundation and caused the collapse. Accordingly, one could plausibly conclude that this dispute is actually a diversion case, insofar as the damage resulted from interference with the preexisting flow of the river, and thus reverse the assessment of liability. Surely the inquiry should turn on more substantial distinctions than these.

One might perhaps argue that the negligence standard should be employed to allocate liability for damages caused by flood control improvements within the exception’s reach. Motivated by an understandable concern that an excessively permissive standard for compensation would deter the construction of beneficial projects, one might turn to the familiar tort principle to delimit the state’s potential exposure. Fault, however, is an exceedingly blunt and inapt device to accomplish the desired effect.

As previously noted, the touchstone of inverse condemnation liability is “whether the owner of the damaged property if uncompensated would contribute more than his proper share to the public undertaking.” (Clement v. State Reclamation Board, supra, 35 Cal.2d at p. 642 .) Balanced against this cost-spreading objective is the reality that boundless liability will thwart the development of beneficial public improvements. (Albers, p. 263.) This constraint on compensation for damages inflicted by the state, however, is entirely unrelated to the blameworthiness of the damaging conduct; the framers of the just compensation clause sought to protect private property without regard to the decency of the state’s intentions. (Van Alstyne, Statutory Modification of Inverse Condemnation: The Scope of Legislative Power (1967) 19 Stan.L.Rev. 727, 771-776; Van Alstyne, supra, 20 Hastings L.J. at p. 495.) Accordingly, the “fruitless search for the somewhat artificial moral elements inherent in the tort concepts of negligence and intentional wrongs” (ibid.) should be recognized as irrelevant to the determination of inverse liability. (See Kratovil & Harrison, Eminent Domain—Policy and Concept (1954) 42 Cal.L.Rev. 596, 621.) Contemporary appellate decisions uniformly agree. (McMahan’s of Santa Monica v. City of Santa Monica (1983) 146 Cal.App.3d 683, 697 [ 194 Cal.Rptr. 582 ]; Yee v. City of Sausalito (1983) 141 Cal.App.3d 917, 921 [190 Cal.Rptr. 595]; Marin v. City of San Rafael (1980) 111 Cal.App.3d 591, 595 [ 168 Cal.Rptr. 750 ]; see also Imperi al Cattle Co. v. Imperial Irrigation Dist. (1985) 167 Cal.App.3d 263, 274 , fns. 4, 5 [ 213 Cal.Rptr. 622 ].)

Consistent with the purposes of the just compensation clause, nonnegligently inflicted damages should be compensable if the aggrieved citizen bears more than his proper share of the unanticipated additional cost of the flood control improvement, and if the imposition of such compensation is not reasonably likely to deter future beneficial public works. As Van Alstyne observed, “practically every governmental decision to construct a public improvement involves, however remotely, at least some unforeseeable risks that physical damage to property may result. In the presumably rare instance where substantial damage does in fact eventuate ‘directly’ from the project, and is capable of more equitable absorption by the beneficiaries of the project (ordinarily either taxpayers or consumers of service paid for by fees or charges) than by the injured owner, absence of fault may be treated as simply an insufficient justification for shifting the unforeseeable loss from the project that caused it to . . . the equally innocent owners.” (Van Alstyne, pp. 493-495, fns. omitted.)

III.

In view of the foregoing considerations, I believe it is erroneous to base inverse liability in the flood control context on the multiple irrelevancies of the modified Archer exception; compensation should instead rest directly on an explicit balancing of the interests implicated by the constitutional provision. Accordingly, I would assess such liability in the following manner: once a landowner establishes that he has suffered actual physical damages, whether foreseeable or not, that were directly or substantially caused by flood control improvements operating as deliberately planned and built, he should receive compensation for the loss unless the public entity can demonstrate that his interest in compensation is outweighed by the risk that such compensation will unreasonably inflate the cost of the project in a manner likely to deter the future construction of beneficial flood control improvements. The greater the owner’s interest in compensation, the more compelling must be the public entity’s showing of unreasonableness.

The landowner’s compensation interest should be understood to denote the extent to which he has contributed more than his proper share to the public undertaking, and should be measured on the basis of such factors as the extent of the damage, “the degree to which the loss is offset by reciprocal benefits,” “the extent to which damage of the kind sustained is generally regarded as a normal risk of land ownership,” and “the degree to which like damage is distributed at large over the beneficiaries of the project or is peculiar to the claimant.” (Van Alstyne, p. 502.) The countervailing disincentive effect, expressed in terms of the reasonableness of the compensation costs to be imposed on the public entity, should be measured according to such factors as “the availability to the public entity of feasible preventive measures or of adequate alternatives with lower risk potential” and “the severity of damage in relation to risk-bearing capabilities.” (Ibid.; see generally id. at pp. 491-493.)

This liability standard addresses the fundamental cost-spreading objective of the just compensation clause while guarding against excessively expansive compensation which might impede the construction of public improvements. Utilization of the rule in place of the Albers standard recognizes that flood control improvements pose the unavoidable risk of extraordinarily vast liabilities; compensation for flood damages thus requires individualized consideration of disincentive effects which in the context of other public works are sufficiently “remote” to justify strict liability. (Albers, p. 263.) Nevertheless, this standard and the Albers rule both follow directly from the policies underlying inverse condemnation, which simply yield different measures of liability for disparate factual circumstances.

I turn now to the case at bar. As stated above, plaintiffs established actual physical damage directly or substantially caused by flood control improvements operating as deliberately planned and built. Defendants, however, were not required to show, and did not show, that their interest in compensation was outweighed by the risk that such compensation would unreasonably inflate the cost of the project in a manner likely to deter the future construction of beneficial flood control improvements. Because defendants did not make that showing, the superior court erred by entering judgment in their favor and the Court of Appeal erred by affirming that judgment.

For the reasons stated, I would reverse the judgment of the Court of Appeal with directions to reverse the judgment of the superior court for further proceedings in accordance with this opinion.

Appellants’ petition for a rehearing was denied February 2, 1989. Mosk, J., was of the opinion that the petition should be granted.

Prior to trial, plaintiffs expressly waived all causes of action against the defendant public entities except inverse condemnation. In this connection, the trial court had earlier granted summary adjudication as to the following issue: “That the breaks in the levee were not caused by volume, speeds and heights of the water beyond the design capacity of the levees.” The other exception recognized in Albers , the so-called Gray exception (Gray v. Reclamation District No. 1500 (1917) 174 Cal. 622 [ 163 P. 1024 ]) holds certain physical injuries to property to be noncompensable if inflicted pursuant to the proper exercise of the government’s “police power.” This doctrine of noncompensable damages applies “only under ‘emergency’ conditions; i.e., when damage to private property is inflicted by government ‘under the pressure of public necessity and to avert impending peril.’ ” (Holtz v. Superior Court, supra, 3 Cal.3d at p. 305 , quoting House v. L.A. County Flood Control Dist. (1944) 25 Cal.2d 384, 391 [ 153 P.2d 950 ]; see Albers v. County of Los Angeles, supra, 62 Cal.2d at pp. 256-261.) “Instances of this character are the demolition of all or parts of buildings to prevent the spread of conflagration, or the destruction of diseased animals, of rotten fruit, or infected trees where life or health is jeopardized.” (House v. L.A. County Flood Control Dist., supra, 25 Cal.2d at p. 391 ; Holtz v. Superior Court, supra, 3 Cal.3d at p. 305, fn. 10 .) The common law privilege, however, does not permit a property owner to obstruct or divert a stream from its natural channel (Clement v. State Reclamation Board, supra, 35 Cal.2d at pp. 635-636; The Weinberg Co. v. Bixby, supra, 185 Cal. at p. 95 ), or to collect and discharge surface waters so as to increase the natural servitude. (Shaw v. Sebastopol (1911) 159 Cal. 623, 624 [ 115 P. 213 ]; Granone v. County of Los Angeles, supra, 231 Cal.App.2d at p. 646 ; Farrell v. City of Ontario (1919) 39 Cal.App. 351, 358-395 [ 178 P. 740 ].) One scholar has suggested that the development of this exception to the “common enemy” doctrine reflects the countervailing interest of property owners who develop land “in justifiable reliance upon the continuance of existing watercourses as means of natural drainage.” (Van Alstyne, supra, 20 Hastings L. J. at p. 458.) See, e.g., San Gabriel V.C. Club v. Los Angeles (1920) 182 Cal. 392, 401 [ 188 P. 554 , 9 A.L.R. 1200 ] (“Not to permit an upper land owner to protect his land against the stream would be in many instances to destroy the possibility of making the land available for improvement or settlement and condemn it to sterility and vacancy. Such a rule would seriously interfere with the development of the country.”). The dissenting opinion in this matter purports to share our concern that strict inverse condemnation liability in these particular circumstances may “deter future public works,” but nevertheless would require compensation for injuries “whether foreseeable or not . . . unless the public entity can demonstrate that [the] interest in compensation is outweighed by the risk that such compensation will unreasonably inflate the cost of the project in a manner likely to deter . . . future construction . . . .” (Dis. opn. at p. 574.) This amorphous hindsight test of liability appears to us unworkable. In addition, it is inconsistent with the purpose of avoiding deterrence of beneficial projects, because the decision to build or not build a project is necessarily determined on the basis of foresight, not hindsight. See footnote 4, ante, at page 564. There appears to be no merit to the dissenting opinion’s suggestion that this could “plausibly” be characterized as a diversion case. The trial court expressly found, as the evidence amply demonstrated, that plaintiffs were subject to periodic flooding before the levee was constructed, and that the levee did not “increase the risk of damage or impose any easement, servitude, or other burdens on plaintiffs’ property.” Indeed, it is not even accurate to state that the “diversion” principle applies any longer in private water-law actions. In Keys v. Romley, supra, 64 Cal.2d 396 , this court reviewed the historical development of the law governing surface waters, noting that the rule had been that an upper landowner was entitled to discharge surface waters from his land as the water natu rally flows, but was liable for damage caused by any disturbance of the natural drainage. (Id. at pp. 405-406.) In Keys , however, we modified this rule by inserting a requirement of reasonableness in every action by adjacent landowners: “No party, whether an upper or a lower landowner, may act arbitrarily and unreasonably in his relations with other landowners and still be immunized from all liability. [j|] It is therefore incumbent upon every person to take reasonable care in using his property to avoid injury to adjacent property through the flow of surface waters.” (Id. at p. 409.) Relying on Keys , several recent Court of Appeal decisions have made similar modifications to the rules pertaining to flood and stream waters in actions between private landowners. (See Linvill v. Perello (1987) 189 Cal.App.3d 195 [ 234 Cal.Rptr. 392 ], holding that the rule of reasonableness was applicable even where defendant’s flood control levee had not interfered with the natural channel of a stream or diverted the natural flow of the waters in such channel; Ektelon v. City of San Diego (1988) 200 Cal.App.3d 804, 810 [ 246 Cal.Rptr. 483 ], holding that an upstream landowner does not have an absolute right to protect his land from floodwaters by erecting structures which merely increase the downstream flow of water in its natural watercourse, but rather is governed by ordinary principles of negligence.) Our holding renders it unnecessary to reach the second issue raised in plaintiffs’ petition relating to whether the State or the District was a “substantial participant” in the levee project so as to be subject to inverse condemnation liability. The majority intimate without holding that future diversion cases may be treated within the modified Archer exception by expanding its reach to encompass all “flood control” cases. (Maj. opn., ante, at pp. 566-567.) They suggest that the common law “diversion” rule can be set aside in this context in light of the principle that “the purposes of the constitutional clause, rather than the rules ‘emanating from the complex and unique province of water law,’ must fix the extent of a public entity’s responsibility.’ ” (Id. at p. 567, quoting Holtz v. Superior Court, supra, 3 Cal.3d at p. 306 .) Having embraced this analysis, however, the majority relinquish any rational basis for their continuing adherence to the Archer exception even as modified. It would be far easier to acknowledge the fundamental incompatibility of Archer's private law underpinnings with the purposes of inverse condemnation, and set it aside in its entirety.