Reset A A Font size: Print

Supreme Court of Oregon,En Banc.

Sharon DUNN, Respondent on Review, v. CITY OF MILWAUKIE, an Oregon municipality, Petitioner on Review.

(CV07040247; CA A139386; SC S059316).

Decided: May 08, 2014

Michael A. Lehner, Lehner & Rodrigues PC, Portland, argued the cause and filed the brief for petitioner on review. Kenneth P. Dobson, Chenoweth Law Group, PC, Portland, argued the cause and filed the brief for respondent on review and filed the response to the League of Oregon Cities' amicus curie brief. Terence L. Thatcher, Senior Deputy City Attorney, Portland City Attorney's Office, Portland, filed a brief on behalf of amicus curiae League of Oregon Cities.

The City of Milwaukie (the city) used highly pressurized water to clean the sewer lines adjacent to plaintiff's house, causing sewage to back up through toilets and bathroom fixtures. Plaintiff brought this action against the city seeking compensation for the damage to her home on two theories—negligence and inverse condemnation. The trial court dismissed the negligence claim before trial as barred by the statute of limitations. See ORS 30.275 (requiring notice of tort claim against public body to be given within 180 days of injury). The inverse condemnation claim went to trial before a jury. At the close of plaintiff's case, the city moved for a directed verdict, arguing that the evidence did not establish a compensable taking of property under Article I, section 18, of the Oregon Constitution.1 The trial court denied the city's motion, and the jury found for plaintiff, awarding $58,333 in damages. On appeal, the Court of Appeals affirmed. Dunn v. City of Milwaukie, 241 Or.App. 95, 102, 250 P3d 7 (2011). We allowed the city's petition for review to consider whether a backup of sewage water into a house and the resulting damage, as happened here in the course of maintenance and cleaning of the city's sewage system, amounts to a compensable taking of property. As we will explain, on the facts before us, we conclude that the city's actions did not give rise to a compensable taking. We therefore reverse the decision of the Court of Appeals.


On review, we state the facts in the light most favorable to plaintiff.2 As part of its maintenance of the city sewer system, the city regularly uses highly pressured water—a process called “hydrocleaning”—to clean the city's sewer lines. The hydrocleaning process permits the city, after the cleaning, to use a video camera to inspect the lines for breaks and repairs needed as part of preventative maintenance. Hydrocleaning can cause water in the sewer lines to backflow through the lateral lines that run from the main sewer line into adjacent houses and then possibly backup through the toilets or other fixtures in the house. If a backup into a house occurs, the city designates the area involved on a map as a “low-pressure” area. That designation lets workers know to use reduced water pressure for future hydrocleaning in that area to prevent another water invasion. Plaintiff's house was not in an area marked for low pressure cleaning as of August 2005, when this backup occurred (although the area has since been so designated). As a result, the two city employees cleaning the sewer in the vicinity of plaintiff's house used the higher pressure that is routine for the hydrocleaning process.

Plaintiff first became aware of a backup when she heard a “loud roar,” felt her house shake, and then saw “brown and gray gunky sewer water that stunk” come out of her toilets and bathroom fixtures. Soon afterwards, water was dripping from her the hallway and into her living room. Plaintiff went out and approached a woman standing near a city truck outside her house. The worker seemed “shocked” that sewer water was coming into plaintiff's house and offered her towels. Plaintiff cleaned the house herself with towels and a wet/dry vacuum. As far as the record shows, plaintiff did not vacate the house or otherwise have to interrupt her occupation of it.

About six or seven months after the sewage water backup, plaintiff began to notice that her wood floor felt clammy (and eventually began to buckle), her wallpaper began to peel, and at times she smelled a sewage-like scent. Over the next two months, plaintiff hired several people to inspect her home for water and microbial damage and to provide repair estimates. About 10 months after the incident, plaintiff filed a formal complaint with the city. More inspections and repair estimates followed, and, ultimately, about 20 months after the incident, plaintiff brought this action against the city for negligence and inverse condemnation.

At trial, no witness could explain why the sewage backed up into plaintiff's house when it did. Plaintiff's house had been built in 1976, and plaintiff had lived in it since 1984. There is no evidence that, before this incident, sewage had backed up into plaintiff's house or into any other house in the near vicinity as a result of the city's sewer maintenance activities. The two city workers who were hydrocleaning the sewers near plaintiff's house when the backup occurred explained that they did everything “by the book” and that their equipment was “operating properly” that day. They did not know why the sewer water backed up into plaintiff's house when it did. They could say only that backups into a house can occur for reasons relating to the vents in that house's plumbing and the design of the lateral lines running from the main sewer line to the house.3 A plumber who inspected plaintiff's house at some point after the backup found nothing out of the ordinary in the plumbing system itself, however. According to him, backups sometimes occur if the hydrocleaning cannot clear a blockage in the main line, which can cause sewage water in the main line to backup into a lateral line running to a nearby house. But he could not say why one area, and not another, might be prone to such backups.

By all accounts, however, sewage backups into adjacent homes during city hydrocleaning are rare. One of the two workers hydrocleaning near plaintiff's house when the backup occurred had worked for the city for about seven years. The backup into plaintiff's house was the only one that he had personally experienced, although a few years later he heard of one other. The plumber who inspected plaintiff's house was generally aware of other sewer backups such as the one into plaintiff's house, and he characterized them as “uncommon.”

The sewage water that backed up into plaintiff's house caused damage to the wood floors, crawl space, and furnace. The estimated cost to repair that damage was $57,905.83, plus another “couple thousand” to replace the vapor barrier and insulation in the crawlspace, and $8,000 for ruined carpets.4

As earlier noted, at the close of plaintiff's case, the city moved for a directed verdict, arguing that plaintiff's evidence did not establish a compensable taking of property for purposes of Article I, section 18, of the Oregon Constitution. The city argued in particular that, to prove a compensable taking, plaintiff had to demonstrate that the city had acted intentionally, which required evidence that the city took the actions that it did knowing that the invasion of plaintiff's property was “substantially certain” to occur or was otherwise a “normal consequence” of the city's actions. Plaintiff, in response, argued that she had to show only that the backup into her house was the “natural and ordinary consequence” of the city's actions and that where, as here, there was no evidence of any other causes, the evidence was sufficient to go to the jury.5 The trial court denied the motion for directed verdict. The jury found in favor of plaintiff and awarded plaintiff $58,333 in compensation.

The city appealed, challenging, among other issues, the denial of its motion for directed verdict. In support of their respective positions, the parties effectively renewed the arguments that they had advanced to the trial court. Relying on this court's decision in Vokoun v. City of Lake Oswego, 335 Or 19, 56 P3d 396 (2002), the Court of Appeals affirmed. It agreed with the city that intent was a necessary element of plaintiff's claim, but reasoned that the jury reasonably could infer the city's intent to cause the sewer backup if the backup was “a natural and ordinary consequence” of the city's sewer cleaning:

“The question is not whether the harmful result occurs frequently; it is whether the result is a natural and ordinary consequence of the government's action at the time and place where that action occurred. * * * [A] consequence that was the last link in a chain of events that began with the governmental action and proceeded, without unnatural or extraordinary intervening causes, to produce the damage.”

Dunn, 241 Or.App. at 102 (emphasis omitted). The court concluded that the jury could infer the city's intent from evidence that the city had carried out the cleaning according to normal procedures and that doing so in some areas of the city can cause sewer backups in private houses. Id.

We allowed the city's petition for review. Mindful that the Court of Appeals was applying what it understood to be the “natural and ordinary consequences” test reaffirmed by this court in Vokoun, the city, along with amicus curiae League of Oregon Cities, argues for a modification of—or at least, a clarification of—that test. Specifically, the city asserts that intent should be inferable only from evidence that an invasion of plaintiff's property interests was “substantially certain” to occur as a result of the government conduct. The city also argues that, regardless of whether this court modifies the test from Vokoun, proof of intent requires some evidence that the intended result was expected. The city concludes that plaintiff offered insufficient evidence of its intent to survive a motion for directed verdict because there was no testimony or other evidence at trial that “the [c]ity expected water to enter plaintiff's home.”

Plaintiff disagrees that the “natural and ordinary consequences” test requires modification or refinement.6 In plaintiff's view, that test effectively serves what plaintiff sees as its primary purpose: to distinguish between governmental negligence and intentional takings. Plaintiff asserts that she sufficiently proved the city's intent by showing that the flooding of her house was the direct result of the city's purposeful act of hydrocleaning. It does not matter, plaintiff urges, that sewers regularly are hydrocleaned without causing water to back up into nearby homes. Plaintiff emphasizes that there was no evidence that the city acted negligently (i.e., failed to exercise reasonable care). According to plaintiff, “[i]n the absence of any unnatural or extraordinary intervening events, such as negligence on the [c]ity's part [ ], the jury had ample evidence that the flood of sewer water was the natural and ordinary consequence of blasting high pressure water in this particular area.”

So framed, this case calls on this court to examine what “intent” means in the context of a takings claim and, concomitantly, the nature and quality of evidence that will support an inference that the government acted with the requisite intent. To resolve that issue, we begin with the principles that guide our analysis of takings claims generally. We then turn to the “intent” element of plaintiff's claim and what was required of plaintiff's proof to satisfy that element.


A “taking” of property is a shorthand description for an exercise of the government's power of eminent domain, which is the power of the sovereign to take property for “public use” without the property owner's consent. Coast Range Conifers v. Board of Forestry, 339 Or 136, 142–43, 117 P3d 990 (2005) (discussing the term “taking”); 1 Nichols on Eminent Domain § 1.11, 1–7 (Julius L. Sackman ed., 3d ed 2013) (describing power of eminent domain generally). The power of eminent domain requires no grant of authority for its exercise, but instead is an inherent attribute of sovereignty. Tomasek v. Oregon Highway Com'n, 196 Or 120, 142, 248 P.2d 703 (1952); 1 Nichols on Eminent Domain § 1.11 at 1–7. Thus, Article I, section 18, is not the source of the state's eminent domain power. Tomasek, 196 Or at 142–43. Instead, by declaring that “[p]rivate property shall not be taken for public use * * * without just compensation[,]” it states a familiar limitation on the state's power of eminent domain-that, when the state takes property, it must pay for it. See id. at 143 (Article I, section 18, is protection for property owner rather than source of eminent domain authority).

Typically, government exercises its eminent domain power by initiating a condemnation proceeding and, through that proceeding, compensating a property owner before appropriating property for a public purpose. See Cereghino et al v. State Highway Com., 230 Or 439, 443–44, 370 P.2d 694 (1962) (so stating). But the power of eminent domain can be exercised de facto as well as well as de jure, which occurs when the government takes property interests through its actions without first initiating condemnation proceedings. When that happens, the property owner can bring an inverse condemnation action to obtain the just compensation that Article I, section 18, guarantees. Id. at 444.

Consistently with the idea that the takings clause is not the source of the state's power to take property, but instead requires compensation for property taken, Article I, section 18, itself does little to inform the understanding of when a government action constitutes a compensable taking. The most that can be said is that there must be an “appropriation of private property” for a public purpose that is characteristic of an exercise of eminent domain authority. See generally Coast Range Conifers, 339 Or at 143 (“take” implies that governmental acts resulting in “the appropriation of private property for public use” will constitute a compensable taking). In other words, the compensable taking is not limited to real property; it includes personal property as well, at least when that property is permanently taken. Hawkins v. City of La Grande, 315 Or 57, 69–70, 843 P.2d 400 (1992). Nor is the concept limited to the physical or tangible property itself; it encompasses as well the owner's fundamental legal interests in property, such as the right to possess, use, and dispose of property. See Cereghino, 230 Or at 445 (citing authorities). But whether the invasion is to real property or personal property, and to the physical property or the intangible but essential rights of ownership in it, the one principle that holds true is that the government's conduct must be “tantamount to a public appropriation” of property, both in nature and in degree. See generally Coast Range Conifers, 339 Or at 147 (discussing various tests and observing that, under them, government act must be “tantamount to a public appropriation of private property”).

The vexing problem over time has been distinguishing between intrusions that amount to a taking and those that do not. In our state constitutional jurisprudence, no single or uniform legal test has emerged. Id. at 146–47. Instead, this court has distinguished among takings claims depending on the nature of the governmental action that gives rise to the claim. Id. at 146. Thus, for example, the court has consistently found a taking when government has intentionally authorized a physical occupation of private property that substantially has interfered with the owner's rights of exclusive possession and use. See, e.g., Vokoun, 335 Or at 31 (applying that test to claim based on governmental diversion of storm water onto private property); Morrison v. Clackamas County, 141 Or 564, 568–69, 18 P.2d 814 (1933) (same). On the other hand, when governmental regulation, rather than physical occupation, restricts a property owner's right of possession, enjoyment, and use, the test is whether the property retains some economically viable or substantial beneficial use. See, e.g., Boise Cascade Corp. v. Board of Forestry, 325 Or 185, 198, 935 P.2d 411 (1997) (applying some economically viable use test to regulatory taking claim); Dodd v. Hood River County, 317 Or 172, 182, 855 P.2d 608 (1993) (applying some substantial beneficial use test to regulatory taking claim).

There likewise has developed no uniform or single test under the federal takings clause,7 on which Article I, section 18, was modeled. Coast Range Conifers, 339 Or at 144 (discussing federal roots of Article I, section 18). The United States Supreme Court has recognized that there is a nearly infinite variety of ways that government action or regulation can affect property interests. Arkansas Game and Fish Com'n v. U.S., ––– U.S. ––––, 133 S Ct 511, 518, 184 L.Ed.2d 417 (2012). Because of that, no “magic formula” has been identified to enable federal courts to determine, in every variety of case, whether a given interference with property is a taking. Id. Only a few bright lines have developed, for purposes of the federal takings analysis. One is that a permanent physical occupation of property authorized by government for a public purpose is a taking. Id. (citing Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419, 426, 102 S Ct 3164, 73 L.Ed.2d 868 (1982)). Likewise, a regulation that permanently divests a property owner of all economically beneficial use of land is a taking. Id. (citing Lucas v. South Carolina Coastal Council, 505 U.S. 1003, 1019, 112 S Ct 2886, 120 L.Ed.2d (1992)). For claims that do not fall within those bright line categories, however, the federal analysis depends on “situation-specific” factual inquiries.


With those general principles as a backdrop, we turn to the issue that this case presents: the role of intent in analyzing whether government action has de facto given rise to a compensable takings claim. As noted, under our state case law, the intentional physical occupation or invasion of property by government for a public purpose generally amounts to a taking, if there is a substantial interference with the property owner's protected interests. Vokoun, 335 Or at 26; Morrison, 141 Or at 568–69.8

The parties agree in this case that, to prevail on her inverse condemnation claim, plaintiff must prove that the government acted intentionally. The parties further agree that intent can be inferred from the circumstances, as our cases also hold. Vokoun, 335 Or at 29. The dispute in this case centers on what intent entails in this context, and the nature and quality of the evidence that will thus suffice to prove intent. Although our past cases have not explored that issue directly, they provide significant guidance, as do the general principles underlying Article I, section 18.

This court first articulated intent as an element of a takings claim in Gearin v. Marion County, 110 Or 390, 402, 223 P 929 (1924). There, according to the plaintiff's complaint, an unusual flood caused the Willamette River to overflow its banks and cover a large portion of the plaintiff's land to a depth of about eight feet. The same flooding conditions caused logs, trees, stumps, and other debris to collect around the piers on a bridge over the river, which threatened to wash out the bridge. County employees cleared the debris and, in doing so, moved it to a place where it floated across the plaintiff's property, instead of into the main channel of the river. The debris, in turn, caused damage and destruction to buildings, and to the contents of the buildings, on the plaintiff's land. Id. at 392–93. Without extensive analysis, this court rejected the plaintiff's takings claim because the county's actions in clearing the debris “were done without any intention to exercise the power of eminent domain or to take [the] plaintiff's property or any part thereof for a public use[.]” Id. at 402. Instead, the claim sounded “purely in tort.” Id. at 401.

Nine years later, in Morrison, this court adhered to the idea that, to be a taking, a physical occupation of private land by the government had to be intentional, but unlike in Gearin, found the pleadings sufficient to allege the required intent. In Morrison, Clackamas County had built a jetty that directed a portion of the Sandy River toward the plaintiff's property. During the next annual high water, the jetty diverted the river's flow over the plaintiff's property, destroying the surface of the land as well as the improvements on it. Morrison, 141 Or at 566–67. The court rejected the idea that the intent required for a taking was “specific intention on the part of [the] defendant to appropriate [the] plaintiff's property[.]” Id. at 569. Instead, the court declared, a governmental actor can “be held to have intended to do those things which are the natural and ordinary consequences of [its] act.” Id. Morrison thus was the source of the “natural and ordinary consequences” test at issue in this case.

Morrison did not discuss the contours of that test explicitly. But the sources from which the court drew, as well as the court's application of the test to the case before it, suggest that that test expressed a distinctive concept: Natural and ordinary consequences were those that were the necessary or certain result of the government's authorized acts. Morrison described the complaint, for example, as having alleged that “the natural and necessary effect” of the construction of the jetty was to alter the flow of the river in such a way as to force its full flow “immediately toward and against [the] plaintiff's land.” Id. at 566 (emphasis added). “Natural and ordinary” consequences, as the court used those terms, did not mean simply effects or consequences that could possibly or plausibly follow; they were ones that ordinarily would follow, as a matter of course and with an element of certainty. Immediately after declaring that the government could be held to intend the natural and ordinary consequences of its acts, the court captured that idea again by observing of the alleged facts before it: “Doubtless the defendant county intended to construct the jetty. The natural consequence, of course, followed.” Id. at 569 (emphasis added). Two cases cited by the court in support of its approach similarly emphasized that, where a government action results in a physical invasion of property, a taking arises only if the injurious invasion was the necessary, inevitable, or otherwise certain consequence of the government's intentional act. See Miller v. City of Morristown, 47 NJ Eq 62, 66–67, 20 A 61, 63 (1890) (for purposes of takings claim, government can be held to have intended the natural “and inevitable” consequences of its acts); Great Northern Ry. Co. v. State, 102 Wash 348, 356, 173 P 40, 43 (1918) (taking arises where damage is “necessary” result of government activity and cannot be avoided).

The intent element of a takings claim did not surface again as a significant issue in any of this court's cases again until Vokoun, which this court decided nearly 80 years after Morrison first articulated the “natural and ordinary consequences” test.9 In Vokoun, the City of Lake Oswego had constructed a storm drain that ran underground near the plaintiffs' property and released water into a ravine at the base of the hillside on which the plaintiffs' house sat. Vokoun, 335 Or at 21–22. During the 25 years that followed, the storm drain created a drainage course in the ravine and undermined the hillside, ultimately causing a landslide that destroyed much of the plaintiffs' property. Id. at 30. The trial court denied the city's motion for directed verdict, and the Court of Appeals reversed, concluding that plaintiffs' inverse condemnation action was predicated on the city's negligence, which was insufficient to establish a takings claim. Id. at 24–25. On review to this court, the plaintiffs argued that the City of Lake Oswego should be liable for the damage to their property, regardless of whether the consequences of constructing the storm drain were expected or intended. Id. at 26.

This court reaffirmed that negligence alone will not support a claim for inverse condemnation and that intent to take is an element of such a claim. Id. at 27. In doing so, the court reiterated Morrison 's observation that specific intent is not required and that the government can be held to have intended the natural and ordinary consequences of its act. Id. at 28. The parties' arguments did not directly dispute the nature and quality of what a plaintiff should be required to prove under that test. As a result, the court did not explore that question. But the court implicitly recognized, as Morrison 's broader discussion had as well, that the concept encompassed more than an indirect causal connection between the government's acts and the physical invasion of and damage to a plaintiff's property. The court began by discussing Morrison and its ordinary and natural consequences test, observing that in Morrison the court had found the complaint sufficient to state a claim because the plaintiff had pleaded that the county in that case had “intended to construct the jetty in a manner that necessarily caused the flooding that destroyed the plaintiff's property[.]” Id. The court then cited Levene et ux v. City of Salem, 191 Or 182, 196–97, 229 P.2d 255 (1951), noting that it was in accord with Morrison, for the proposition that a taking arises when government diverts the flow of a stream or constructs a drain or sewer in such a way that flooding of private property is “a necessary result” of the construction. Vokoun, 335 Or at 28. And, in concluding that the plaintiffs' evidence was sufficient to create a jury question on intent, the court described the evidence as permitting the reasonable inference that the landslide caused by water discharging from the storm pipe “was the natural and ordinary (even inevitable ) consequence” of the manner in which the city had intentionally constructed the storm drain. Id. at 30 (emphasis added).10

As that description of the case law conveys, where compensation is sought for injuries caused by physical invasions or occupations of property, the intent element of a takings claim is fundamental in distinguishing between those actions that are the equivalent of an exercise of eminent domain and those that are actionable as ordinary torts. The power of eminent domain is affirmative in nature. It is a power exercised for a particular purpose—the public's benefit—and intentionally. The idea that the sovereign's power of eminent domain could be exercised through error, accident, or inadvertence, is at odds with the nature of the power itself. Inadvertent and unintended acts give rise to liability, if at all, as ordinary torts, not takings. As the Seventh Circuit Court of Appeals has pointedly put it:

“So when does error ‘take’ property? Suppose agents of the FBI, while chasing a kidnapper, demolish someone's car, or suppose a postal van runs over a child's tricycle. Do these accidents ‘take’ the car and tricycle? Certainly they are casualties of the operation of government. * * * Accidental, unintended injuries inflicted by governmental actors are treated as torts, not takings. And torts are compensable only to the extent the Federal Tort Claims Act permits. The Court has never treated limitations on liability in tort as mere pleading obstacles, to be surmounted by shifting ground to [a takings claim brought under] the Tucker Act.” Chicago, Milwaukee, St. Paul and Pacific R. Co. v. U.S., 799 F.2d 317, 325–26 (7th Cir1986), cert den, 481 U.S. 1068 (1987). Intent, as an element of a takings claim based on a physical intrusion or occupation of property, serves the important function of helping to distinguish between acts that are tantamount to appropriations of private property for a public purpose and those for which a remedy lies only in tort. 11


If the natural and ordinary consequences test is understood as Morrison and later cases appear to have meant it—that is, as permitting an inference of the requisite intent to take when the consequences of governmental action are necessary, inevitable, or substantially certain to result—the test serves well. But the test has the potential to be misunderstood as allowing intent to be inferred based on simple causation alone. The Court of Appeals, in fact, appears to have applied the test in that way in its opinion in this case. The court began by phrasing the test as one that looked to “a” natural and ordinary consequence, rather than “the” natural and ordinary consequence, of the government's action. Dunn, 241 Or.App. at 102. The shift, although subtle, conveys a different idea. Something that is “the” natural and ordinary consequence implies that it will follow with a degree of certainty. Something that is “a” natural and ordinary consequence implies that it is one of many possible consequences, so that it might or might not follow from the government's actions. That rephrasing by the Court of Appeals led to an analysis that appears to allow an inference of intent when the government's act is simply a but for cause of the invasion or damage to property:

“The question * * * is whether the result is a natural and ordinary consequence of the government's action at the time and place where that action occurred. * * * a consequence that was the last link in a chain of events that began with the governmental action and proceeded, without unnatural or extraordinary intervening causes, to produce the damage.”

Id. (emphasis added). In other words, as long as the government's actions were a cause-in-fact of the injury to plaintiff's property, that was enough to render the conduct 19 “intentional.”12

That articulation and application of the natural and ordinary consequences test is not faithful to what the test was designed to convey. The phrase itself—natural and ordinary consequences—imports a stronger relationship between the government's act and the result that follows. In particular, it conveys that, in the ordinary course of events, a certain act will naturally have a certain consequence.13 That consequence, because it is the natural one that will ordinarily follow, is the necessary or inevitable result of undertaking a particular act, unless some other force or event comes into play to alter what will otherwise occur. That is how Morrison understood and applied the phrase, as we have explained. 141 Or at 566 (complaint adequately alleged that natural and “necessary” effect of jetty construction was to alter flow of river across plaintiff's land during seasonal flooding). It is also consistent with the relationship between the government's act and the resulting injury to property required by the cases that Morrison 13 cited with approval, which looked to whether the resulting injury was the inevitable and necessary result of the government's act in the sense that it was sure to follow. Miller, 47 NJ Eq at 66–67, 20 A at 63; Great Northern Ry. Co., 102 Wash at 356, 173 P at 43. And, finally, a test that looks to the inevitability or certainty with which particular results will follow from particular government action appears consistent with the way that the natural and ordinary consequences has been understood by courts in general, and federal courts in particular.14

We decline the city's and amicus League of Oregon Cities' invitation to modify the test for intent in this context by adopting the Restatement (Second) of Torts (1965) definition. Under the Restatement, a person acts intentionally when “the actor desires to cause consequences of his act, or * * * he believes that the consequences are substantially certain to result from it.” § 8 A. As the Restatement formulates the test, intent requires either specific intent or a state of mind that serves as a surrogate for specific intent (a person's subjective knowledge that particular consequences are “substantially certain” to result from the person's act). Morrison rejected specific intent as a requirement for a taking. 141 Or at 569. Vokoun reaffirmed that aspect of Morrison. 335 Or at 28. Neither the city nor the amicus articulate a persuasive reason for us to impose a specific intent requirement—or something close to it—where we have not before.15

But, contrary to plaintiff's argument, the natural and ordinary consequences test can benefit from clarification. If, as happened in this case, it is understood as permitting an inference of intent from “but for” causation, then the test eliminates the requirement of intent altogether. So transformed, the natural and ordinary consequences test does not adequately distinguish between governmental negligence and intentional takings and does not serve the constitutional principle at work.

The natural and ordinary consequences test, as originally embraced in Morrison, conveys a sound concept, and we adhere to it as clarified by our analysis in this case. The test examines whether the government intentionally undertook to act “in a manner that necessarily caused” the injurious invasion of the plaintiff's property. Vokoun, 335 Or at 28 (citing Morrison, 141 Or at 569). A factfinder is entitled to impute the requisite intent to take property if the invasion to the property owner's interests was the necessary, substantially certain, or inevitable consequence of the government's intentional acts. In other words, a plaintiff need not prove that the governmental actor subjectively intended the consequential invasion of property interests or undertook action knowing (even if not desiring) that the consequences would follow. Evidence of specific intent, although it will suffice, is not required. And although a plaintiff's burden is less than specific intent would make it, it is still exacting. A plaintiff still must show that the government intentionally undertook its actions and that the inevitable result of those actions, in the ordinary course of events, was the invasion of the plaintiff's property that is the basis for the plaintiff's inverse condemnation claim. Thus, if a plaintiff's best evidence is that the invasion was a less than certain consequence—such as a conceivable, possible, or plausible outcome, or one that otherwise might or might not occur—that is not enough for a factfinder to infer that the invasion was intentional.


The remaining question is whether plaintiff's evidence in this case meets the natural and ordinary consequences test as we have clarified it. Plaintiff argues that the jury could infer the city's intent to cause the sewage backup in plaintiff's house from evidence that the backup “was the direct consequence of the [c]ity's purposeful act of blasting high-pressure water * * * into the sewer line adjacent to [p]laintiff's house.” Such an inference is particularly appropriate, plaintiff argues, because she never argued to the jury or presented any testimony or evidence that the city acted negligently by failing to exercise reasonable care in cleaning the sewer lines to and around plaintiff's house. According to plaintiff, there was no evidence of “unnatural or extraordinary intervening events” that caused the damage to plaintiff's property. On the contrary, asserts plaintiff, “everything was done ‘by the book.” ’ Plaintiff concludes that, because there was no evidence that the city acted negligently, “there was sufficient evidence that the invasion of sewer water was the natural and ordinary consequence of the [c]ity's actions and intent could therefore be inferred.”16

The fact that conduct is not negligent does not establish, however, that it is intentional. As classically conceived in the law, intent is a state of mind. Prosser and Keeton on the Law of Torts § 8, 34 (W. Page Keeton ed., 5th ed 1984). Negligence, on the other hand, is conduct and not a state of mind. Id. at § 31, 169. The “essence” of negligence is “behavior which should be recognized as involving unreasonable danger to others.” Id. As Prosser explains the tort concepts of negligence and intent:

“In negligence, the actor does not desire to bring about the consequences which follow, nor does he know that they are substantially certain to occur, or believe that they will. There is merely a risk of such consequences, sufficiently great to lead a reasonable person in his position to anticipate them, and to guard against them.”

Id. Thus, negligence and intent are not flip sides of the same coin; they are different coins. For a person to act not negligently does not establish how he did act or with what mental state.17 The person could have acted with intent, but a range of other possibilities exists as well, including the possibility that what happened was purely accidental or 11 inadvertent, and not due either to negligence or intent.18

But more to the point, under the natural and ordinary consequences test as we have clarified it, the issue in this case turns on the certainty or inevitability that the city's act of hydrocleaning the sewer would cause the sewage backup into plaintiff's home. Here, there is no dispute that the city's manner of hydrocleaning the sewer, using high-pressure water, was intentional. The disputed issue at trial was whether anything more was required to establish that the alleged taking was intentional and, if so, what. Here, as we have explained, to establish intent, plaintiff also had to show that the backup into her home was the necessary result of the city's intentional actions.

As a matter of law, plaintiff's proof, viewed in the light most favorable to her, was not sufficient to meet that test. The record establishes that the city regularly cleans its sewers using the hydrocleaning process. Despite that fact, backups of sewage into adjacent houses due to the city's hydrocleaning are rare and uncommon occurrences. Indeed, they are so uncommon that one city worker, who had been cleaning city sewers for seven years, personally had experienced only the backup at issue in this case. And, by the time of trial almost two years after this back-up, he had heard of only one other. No one could explain why this backup into plaintiff's house occurred while, day in and 17 day out, the city hydrocleans sewers without similar backups occurring. To be sure, on this record, a factfinder could find that the city's hydrocleaning was a “but for” cause of the backup. But some other factor, one not identified on this record, had to be at work as well.

The conclusion most favorable to plaintiff on this record is that the intrusion of sewage water into one or more nearby houses was a known risk of hydrocleaning generally, but one that rarely came to pass. Under the natural and ordinary consequences test, for the city to be found to have intended the invasion of plaintiff's property, and not just the acts that, in some causal way, led to or contributed to that invasion, the evidence had to establish the likelihood of that invasion with greater certainty. Without any evidence that the sewage backup into plaintiff's house was the necessary, certain, predictable, or inevitable result of the city's intentional manner of hydrocleaning the adjacent sewer, the evidence was insufficient to support plaintiff's inverse condemnation claim. The trial court should have directed a verdict for the city on that claim, and the Court of Appeals erred in concluding otherwise.19

We emphasize that plaintiff and other property owners who suffer property damage under circumstances of this kind are not necessarily without any remedy. Oregon has abrogated its traditional sovereign immunity; both the state and other governmental units can be sued on common-law tort theories. A property owner in Oregon therefore has the same recourse against the government as against a private tortfeasor, subject to the requirements of the Tort Claims Act (ORS 30.260–30.302).20 Plaintiff in fact attempted to pursue a tort claim in this case, but the trial court dismissed it for lack of timely notice. Because the Court of Appeals affirmed the trial court's denial of the city's motion for directed verdict, it declined to address plaintiff's cross-assignments of error regarding the trial court's dismissal of her tort claim. Dunn, 241 Or.App. at 97 n 2. Plaintiff's arguments in that regard remain to be resolved on remand to the Court of Appeals.

The decision of the Court of Appeals is reversed, and the case is remanded to that court for further proceedings.


1.  Article I, section 18, provides, in part: “Private property shall not be taken for public use * * * without just compensation[.]”

2.  See Brown v. J.C. Penney Co., 297 Or 695, 705, 688 P.2d 811 (1984) (after denial of motion for directed verdict, reviewing court considers evidence and all inferences it supports in light most favorable to prevailing party).

3.  City workers explained that, for example, if the plumbing vents in a house are plugged, air that builds up as the lines are cleaned may have nowhere to escape other than through the toilets or other fixtures of that house. The design of the lateral lines running to a home from the main sewer also can be a factor. If the lines are particularly short, air pushed through the line from the cleaning process may not dissipate, which can cause a backup through a house's toilet. Similarly, a backup can result if the lateral line is lower than the main line, which makes it easier for the water to backflow into the house.

4.  The city vigorously disputed plaintiff's evidence of what damage could be traced to the sewage water backup in August 2005 and the value of certain items (such as the carpets). The city also vigorously disputed whether, as plaintiff's expert appraiser testified, the sewer water backup created a stigma that decreased the market value of the property by $100,000. The damages awarded by the jury tracked closely with the repair and replacement damages that plaintiff presented, except for the value that plaintiff placed on her carpets; the jury did not award plaintiff damages in an amount comparable to plaintiff's claimed loss of market value. Consistently with our standard of review, therefore, we describe the evidence in the light most favorable to plaintiff's favorable jury verdict, but that description, in this case, does not include market value loss.

5.  The city also argued that the damage to plaintiff's house was not a sufficiently substantial interference with her property rights to constitute a “taking.” The trial court concluded that the evidence presented a jury question on that point, as did the Court of Appeals. Dunn, 241 Or.App. at 103. The city continues to make that argument on review. Because our conclusion on the city's intent argument is dispositive, we do not reach that aspect of the city's argument.

6.  Plaintiff argues fleetingly that the city did not preserve its argument because it did not urge the trial court or the Court of Appeals to modify or refine the “natural and ordinary consequences” test that Vokoun reaffirmed. The city, however, made the same arguments that it makes now about what Vokoun's test, correctly understood, requires of a plaintiff's evidence. To preserve its position, the city did not have to make a futile request that the trial court or the Court of Appeals modify or refine the test from Vokoun to avoid its misapplication or misinterpretation, as the city urges happened in this case. That argument is more appropriately directed to this court.

7.  The Fifth Amendment to the United States Constitution, as relevant here, provides, “[N]or shall private property be taken for public use, without just compensation.”

8.  We emphasize, as Coast Range Conifers held, that there is no unitary test for takings claims, and the test varies with the nature of the claim. 339 Or at 146–47. We concern ourselves in this case only with a physical invasion of property, one that results in damage to that property. As Coast Range Conifers observed, this court consistently has held intent to be an element of such a takings claim. Id. at 146. Our analysis therefore is confined to that context and does not necessarily extend to other forms of takings claims, such as those based on regulation or planning (e.g ., Dodd, 317 Or at 181; Fifth Avenue Corp. v. Washington Co., 282 Or 591, 608–614 (1978)); nuisance (e.g., Thornburg v. Port of Portland, 233 Or 178, 180, 376 P.2d 100 (1963); Lincoln Loan v. State Hwy. Comm., 274 Or 49, 55, 545 P.2d 105 (1976)); or cases involving physical occupations of a kind that amount to an ongoing easement (e.g., Cereghino, 230 Or at 446).

9.  The principal cases touching on the intent element of a takings claim between the time that Morrison and Vokoun were decided were Tomasek, 196 Or 120, and Hawkins, 315 Or 57. Neither provides any particular guidance here. In Tomasek, the principal issue was whether the state was immune from suit on an inverse condemnation theory. 196 Or at 140. The court recited and quoted at length from prior cases involving takings claims, including Morrison and its statements on the element of intent. Id. at 142–51. The court did so, however, only in the course of analyzing whether the state was amenable to suit on an inverse condemnation theory. Although the defendant disputed the sufficiency of the evidence to support a takings claim, the defendant's challenge focused on whether natural causes, as opposed to the government's construction of a bridge and relocation and grading of a highway, were the cause-in-fact of the injury to the plaintiff's property. Id. at 139–40. The adequacy of the plaintiff's proof of intent was not an issue. In Hawkins, it was undisputed that the city-defendant had released sewage “intentionally” when storm conditions overran the capacity of its sewage treatment facility. 315 Or at 60. Although Hawkins implicitly treated intent as an element of the takings claim in that case, neither the nature and quality of the evidence required to prove intent nor the legal adequacy of the evidence of intent were issues on appeal.

10.  Elsewhere in the opinion, Vokoun's analysis may have been less exacting. In describing the sufficiency of the evidence in that case, the court explained that there was no dispute that the storm drain had caused erosion in the drainage channel. And although the parties disputed the cause of the landslide, the evidence was that the hillside had been stable before the storm drain was built; there had been no drainage course in the ravine before then; the resulting drainage was “consistent with the way that the drain had been designed and built;” and there had been no intervening causes. Vokoun, 335 Or at 29–30. From that evidence, the court concluded, a jury could infer that the landslide was “the natural and ordinary (even inevitable) consequence of the city's construction of the storm drain in that manner.” Id. at 30. To the extent that Vokoun can be read to permit causation-in-fact alone to suffice, we disavow that understanding for the reasons we explain in this opinion. If, however, Vokoun relied on the fact that the landslide was the necessary or inevitable result of the manner in which the drain had been constructed, then it is consistent with Morrison and other previously decided cases, as well as with the test that we adopt in this case.

11.  Some courts have expanded “takings” law to reach government negligence, often “to circumvent the traditional immunity of governmental bodies from liability for tort[.]” A.W. Gans, Damage to Private Property Caused By Negligence of Governmental Agents as “Taking,” “Damage,” or “Use” for Public Purposes, in Constitutional Sense, 2 ALR 2d 677, 680 (1948), see generally id . at 681–87 (canvassing case law). Other jurisdictions have a more expansive test for government takings, because the takings clauses of their constitutional provisions expressly require compensation for government actions that “damage” private property as well as those that “take” private property. See Moeller et ux v. Multnomah County, 218 Or 413, 425–27, 345 P.2d 813 (1959) (contrasting Oregon Constitution's takings clause with those of states that include “damage” provisions, which has led those states to extend compensation to a broader range of cases); see generally Patterson v. Horsefly Irrigation Dist., 157 Or 1, 18, 69 P.2d 282 (1937) (emphasizing that “unnecessary damage,” as often occurs from negligence, is at odds with exercise of eminent domain power, which arises from necessity of taking private property for public purpose). Oregon, however, has abrogated its sovereign immunity for torts and has no damage provision in its takings clause. Both considerations are additional reasons to adhere to requiring intent as an element of a takings claim in this context.

12.  We are not alone in reading the Court of Appeals opinion to have so transformed the test. Others have described the Court of Appeals' analysis in this case as “a tort-like proximate cause test to determine whether there is a sufficient causative link between the government action and the plaintiff's harm.” Jan G. Laitos and Teresa Helms Abel, The Role of Causation When Determining the Proper Defendant in a Takings Lawsuit, 20 Wm & Mary Bill Rts J 1181, 1184 (2012).

13.  Certainty or inevitability, in this context, does not require—at least not, necessarily—regularity or frequency. If, in the ordinary course of events, particular circumstances are substantially certain to occur on a seasonal or other intermittent basis (such as flooding), and if the government's actions will necessarily result in an invasion of a plaintiff's property when those circumstances arise (as with a bridge that will divert water onto a plaintiff's land when flood waters rise sufficiently), such a result can be found to be certain and inevitable for purposes of a takings claim. Compare Ridge Line, Inc. v. U.S., 346 F3d 1346, 1356 (Fed Cir2003) (repeated increased rain runoff caused by government development, even though intermittent, an intentional taking if runoff was the “direct, natural, or probable result” of development) with Thune v. U.S., 41 Fed Cl 49, 52–53 (1998) (no intent to take could be inferred where deliberately set forest service burn went out of control due either to negligence or unexpected and unforecast wind change, damaging private property).

14.  Morrison's understanding of the certainty required of the test that it embraced is consistent with how that test appears to have been used in other cases during that same time period, particularly federal cases. Although the test was sometimes termed the “natural and probable consequences” test, as well as the “natural and ordinary consequences test,” both looked to the certainty with which the consequences would follow government action. See, e.g., Jed Michael Silversmith, Takings, Torts & Turmoil: Reviewing the Authority Requirement of the Just Compensation Clause, 19 UCLA J Envtl L & Pol'y 359, 379–83 (2001) (natural and probable consequences test, as used in Fifth Amendment cases for more than 100 years, addresses certainty; destruction of a plaintiff's property could be ascertained to a certainty before the government engaged in its authorized conduct).

15.  Other jurisdictions with similar constitutional provisions have adopted either the Restatement intent test or one akin to it in the inverse condemnation context. See City of Dallas v. Jennings, 142 SW3d 310, 314 (Tex 2004) (government is liable for inverse condemnation if it “knows that the specific property damage is substantially certain to result from an authorized government action”); Electro–Jet Tool Mfg. Co., Inc. v. City of Albuquerque, 114 NM 676, 683, 845 P.2d 770, 777 (1992) (“acting with knowledge that the damage was substantially certain to result from [government] conduct” gives rise to a takings claim); Robinson v. City of Ashdown, 301 Ark 226, 231–32, 783 S.W.2d 53, 56 (1990) (government's knowledge that “an invasion of another's interest in the use and enjoyment of land is substantially certain to result” from its conduct results in an intentional act). But in general, state tests for what constitutes a compensable taking are all over the board, in part because state jurisprudence developed without any common-law antecedents and independently of federal law, and because a state action has traditionally been a prerequisite to pursuing a claim under federal law. See generally Jadd F. Masso, Mind the Gap: Expansion of Texas Governmental Immunity Between Takings and Tort, 36 St. Mary's L J 265, 270–72 (2005) (discussing development of takings law in United States). States, in particular, vary significantly in whether intent is required to prove a compensable taking and, if so, whether specific intent or some other test of intent is required. Id. at 277–84 (discussing state and federal case law); see also James S. Burling and Luke A. Wake, Takings and Torts: The Role of Intention and Foreseeability in Assessing Takings Damages, in Condemnation 101: Making the Complex Simple in Eminent Domain 449–51 (ALI–ABA Committee on Continuing Professional Education eds., 2011) (discussing state cases).

16.  As earlier noted, plaintiff initially advanced a negligence claim as well, but that claim was dismissed before trial. That procedural posture of the case leaves plaintiff in the unusual position of having advanced a negligence claim while simultaneously relying on the lack of evidence that the city acted negligently.

17.  Of course, on this record, there is no evidence or jury determination that the city was “not negligent.” There instead is simply a void left by the dismissal of plaintiff's negligence claim. Even if, however, this record established that the city was “not negligent,” as plaintiff argues, the argument fails, as we have explained. Classic logic provides a further explanation for why it fails. Plaintiff's argument suffers from the “fallacy of negative premises.” Judge Ruggero J. Aldisert discusses the fallacy in his book Logic for Lawyers: A Guide to Clear Legal Thinking 156 (3d ed 1997), explaining that, when two premises of a syllogism are negative, “we cannot determine anything regarding their relationship to one another.” By way of example, Judge Aldisert points out that, “[f]rom the premises, James is not a lawyer; lawyers are not steelworkers, we cannot conclude that James is or is not a steelworker.” Id. Likewise, from the premises that the city workers in this case did not act negligently and negligent acts are not intentional acts, we cannot conclude that city workers did or did not engage in intentional conduct. As Judge Aldisert emphasizes, “Not knowing that something exists is simply not knowing.” Id.

18.  As Prosser further explains, early common law imposed strict liability for trespasses that resulted in injury to person or property, with the result that purely accidental injuries were actionable; all that was required was a voluntary act. Prosser and Keeton on the Law of Torts, at § 29, 163. But the rule now is that liability generally does not attach for an “unavoidable accident,” which is an occurrence that was not intended and that, under all the circumstances, could not be foreseen or prevented by the exercise of reasonable precautions. Id. at § 29, 162. To be sure, as Prosser emphasizes, no accident “is entirely inevitable, so long as it results from a voluntary human act[,]” because the harm might have been avoided had the human act not been undertaken. Id. But liability now does not arise in the absence of “some wrongful intent or negligence.” Id. at § 29, 163.

19.  Our conclusion on these facts accords with those of courts in other jurisdictions that have resolved takings claims based on sewage overflow into private houses. See, e.g., City of Dallas, 142 SW3d at 315 (evidence established that city efforts to unclog sewer lines did not ordinarily cause residential flooding; no taking was shown given lack of evidence that damage was substantially certain to occur and that city took actions with knowledge of that fact); Edwards v. Hallsdale–Powell Utility Dist. Knox County, Tenn ., 115 SW3d 461, 467 (Tenn 2003) (no showing that backup of sewer was caused by purposeful or intentional act in maintenance of sewer lines, as opposed to clog from natural causes or negligence of utility district). Some courts have reached the same conclusion even though their state constitutions more broadly guarantee compensation for damage to property, as well as takings of property. See, e.g., Henderson v. City of Columbus, 285 Neb 482, 496, 827 NW2d 486, 496–97 (2013) (no taking or compensable damage under state constitution for single incident of sewage flooding where evidence could not support finding that city knew damage would occur from its actions in responding to malfunction of sewer system); see generally Moeller, 218 Or at 425–27 (contrasting Oregon Constitution's taking clause with those of states that include “damage,” which has led those states to extend compensation to a broader range of cases than Oregon's clause reaches).Cases in which property owners have succeeded in sewage invasion takings claims typically have involved repeated or chronic sewage invasions that permitted a finding that the government, in failing to correct the source of the overflow, acted intentionally or maintained an intentional nuisance. See, e.g., Robinson, 301 Ark 226, 228–29, 232, 783 S.W.2d 53, 54, 56 (1990) (recurrent sewage invasion of plaintiff's home over nine-year period, caused by chronically malfunctioning lift station pump, which continued despite plaintiff's pleas to the city for relief, was a compensable taking; city knew that invasion was substantially certain to result from its failure to remedy problem; city appropriated use of plaintiff's property for the public purpose of serving as overflow dump for sewage and appropriately should have to purchase the property so taken); see also DeKalb County v. Orwig, 261 Ga 137, 138–39, 402 S.E.2d 513, 514–15 (1991) (whether city's action was taking was factual issue for jury where city failed to take remedial action after first sewage backup, and second sewage backup occurred).

20.  To the extent that the conduct in question is not a tort, but instead is purely accidental or otherwise nonactionable, the protection for a property owner may lie in the purchase of private insurance to cover such events, whether they involve governmental or private actors.