Summary
In Barneck, this court interpreted "competing provisions" of the UGIA—one provision that waived governmental immunity for "any injury caused by... a defective, unsafe, or dangerous condition of any highway [or]... culvert," and another provision creating an exception to such waiver (i.e., reinstating immunity) where "the injury arises out of, in connection with, or results from... the management of flood waters" or the "repair, or operation of [a] flood or storm system[]."
Summary of this case from Scott v. Wingate Wilderness Therapy, LLCOpinion
No. 20130429.
06-12-2015
Michael BARNECK, Linda Barneck, individually and as heirs of Justine Barneck, deceased, and Heidi Paulson, Appellants, v. UTAH DEPARTMENT OF TRANSPORTATION, and the State of Utah, Appellee.
David M. Bennion, Scott S. Bell, Alan S. Mouritsen, Salt Lake City, for appellants. Sean D. Reyes, Att'y Gen., Peggy Stone, Reed Stringham, Ass't Att'y Gen., Salt Lake City, for appellee.
David M. Bennion, Scott S. Bell, Alan S. Mouritsen, Salt Lake City, for appellants.
Sean D. Reyes, Att'y Gen., Peggy Stone, Reed Stringham, Ass't Att'y Gen., Salt Lake City, for appellee.
Associate Chief Justice LEE authored the opinion of the Court, in which Chief Justice DURRANT, Justice DURHAM, Justice PARRISH, and Judge TOOMEY joined. Justice NEHRING did not participate herein due to his retirement; Court of Appeals Judge KATE A. TOOMEY sat. Justice HIMONAS became a member of the Court on February 13, 2015, after oral argument in this matter, and accordingly did not participate.
Opinion
Associate Chief Justice LEE, opinion of the Court:
¶ 1 During a brief but significant rainstorm, a culvert under SR–35 in Duchesne County became obstructed, causing some fifteen feet of water to back up on the north side of the road. After Utah Department of Transportation workers removed debris from the road and tried unsuccessfully to unclog the culvert, they left the scene. The standing water sat next to the road for several hours and eventually caused it to collapse, leaving a gaping hole in the middle of the road. Later that night, plaintiffs' vehicles careened into this chasm. Plaintiffs subsequently filed this suit against UDOT for negligence and wrongful death. These claims were dismissed on summary judgment in the district court, on the basis of the determination that UDOT was entitled to immunity under the Governmental Immunity Act.
¶ 2 In reviewing this decision, we are asked to interpret and apply competing provisions of the Governmental Immunity Act—provisions that waive immunity for “any injury caused by ... a defective, unsafe, or dangerous condition of any highway [or] ... culvert,” Utah Code § 63G–7–301(3)(a)(i), while also providing an exception to such waiver where “the injury arises out of, in connection with, or results from ... the management of flood waters” or the “repair, or operation of [a] flood or storm system[ ],” id. § 63G–7–301(5)(p), (q). We reverse and remand. In so doing, we interpret the statutory references to “dangerous condition” of a “culvert,” the “management of flood waters,” and the “operation of a flood or storm system.” We also clarify the relationship between the statutory waivers of immunity and exceptions therefrom, in a manner repudiating the but-for standard of causation articulated in some of our prior cases and adopting instead a standard of proximate causation.
I
¶ 3 On a July afternoon in 2011, over an inch of rain fell over the course of about an hour in the area near mile marker 46 on SR–35 in Duchesne County. Such rain would usually collect in and run down a natural gully and through a culvert passing under SR–35. For reasons yet unknown, however, the culvert on this particular day had become obstructed. And the water quickly backed up, pooling at a depth of around fifteen feet. The water—and debris it carried—then ran across the surface of SR–35. In response to this and other occurrences at different locations along SR–35, UDOT dispatched a team of workers to clear the debris.
The briefs identify this area a bit differently. They refer to it as “near mile marker 46.5.” We are unsure of what to make of that formulation, as we suppose that a “mile marker” is in fact a mile marker and not a half-mile marker, and see no indication in the record or elsewhere that UDOT uses half-mile markers. But we note this discrepancy anyway. We do so in case there literally is a “mile marker 46.5” on SR–35. Cf. J.K. Rowling, Harry Potter and the Sorcerer' Stone89–90 (1998) (noting Uncle Vernon and Aunt Petunia's disbelief in the notion of a Platform 9¾ at King's Cross Station); John Ingold, Colorado Hopes a Mile 419.99 Sign on Interstate 70 Thwarts Stoners, Denver Post(Jan. 10, 2014), http://www. denverpost.com/news/ci_24889289/colorado–hopes–mile–419–99–sign–interstate–70 (noting the replacement of mile marker 420 on I–70 with marker 419.99; noting that the number 420 is “[i]n sports terms,” the “ ‘Roll Tide’ of weed,” and explaining that marker 420 repeatedly had been stolen by “marijuana enthusiasts”) (Only in Colorado. Or so we assume.).
The facts set forth here are largely undisputed, but presented in a light most favorable to the nonmoving party given that this case comes before us on an appeal from summary judgment.
¶ 4 Upon arriving on the scene, UDOT workers cleared the debris on the road. They also attempted to clear the obstruction in the culvert using a backhoe. UDOT's attempts to unblock the culvert were ultimately unsuccessful, and the workers left for the day at about 4:00 p.m. The decision to leave was based on the determination that there was no change in driving conditions and nothing obstructing the roadway for motorists. A short time later, a single UDOT worker returned to inspect the site one last time. He saw some water flowing on the south side of SR–35. But he could not see if the water was flowing from the culvert or if it was leaching through the embankment under the road. The worker then left for the day.
¶ 5 The pooled water sat for several hours and apparently resulted in “hydraulic piping”—a phenomenon in which water seeps through and displaces the road base. This process of hydraulic piping continued through the afternoon and night, and eventually the road collapsed. The result was a chasm in SR–35 that was twenty feet deep and thirty feet across. There were no signs to alert oncoming traffic.
¶ 6 Plaintiff Heidi Paulson was traveling eastbound on SR–35 later that night when suddenly, and without warning, her car slammed into the east wall of the chasm, falling nose-first to the bottom. Paulson was severely injured. Only a short time later, plaintiff Michael Barneck and his fifteen-year-old daughter Justine were traveling westbound when they violently crashed into the chasm, killing Justine and injuring Michael.
¶ 7 Plaintiffs sued UDOT on the basis of its alleged negligent maintenance of the road and the clogged culvert. After discovery, UDOT moved for summary judgment. UDOT claimed that it was immune under the Governmental Immunity Act because the plaintiffs' injuries arose out of the “management of flood waters” and the “operation” of a “flood or storm system.” The district court granted summary judgment in favor of UDOT on both theories. Plaintiffs now appeal.
¶ 8 We review the district court's decision granting summary judgment de novo, affording it no deference. Torian v. Craig, 2012 UT 63, ¶ 13, 289 P.3d 479. In so doing, we determine whether UDOT has established that there are no genuine issues of material fact and that it is entitled to judgment as a matter of law. Utah R. Civ. P. 56(c).
II
¶ 9 Utah's Governmental Immunity Act, Utah Code §§ 63G–7–101 to –904, at once waives sovereign immunity and carves out express exceptions to those waivers. The waiver of relevance to this case is in the provision waiving immunity for “any injury caused by ... a defective, unsafe, or dangerous condition of any highway [or] ... culvert.” Id. § 63G–7–301(3)(a)(i). And the relevant exception, in turn, is in the provision retaining immunity where “the injury arises out of, in connection with, or results from ... the management of flood waters” or the “repair, or operation of [a] flood or storm system[ ].” Id. § 63G–7–301(5)(p), (q).
¶ 10 In advancing its motion for summary judgment, UDOT relies on the above-cited exception provision. It claims that the negligence alleged by the plaintiffs concerned UDOT's “management of flood waters” or its “repair[ ] or operation of [a] flood or storm system[ ].” Plaintiffs see the case differently. They claim that their injuries were caused by “a defective, unsafe, or dangerous condition of [a] highway [or] ... culvert,” and that the question of UDOT's immunity is thus controlled by the above-cited waiver provision. And in any event, to the extent there are disagreements about the cause of plaintiffs' injuries, plaintiffs point to those disagreements as an alternative ground (of genuine issues of material fact) for denying UDOT's motion for summary judgment.
¶ 11 To resolve this dispute, we must first interpret the terms of the operative waiver and exception provisions of the Governmental Immunity Act. We must also clarify the relationship between these provisions by articulating the governing standard of causation in a case (like this one) in which the plaintiffs' injuries might fairly be described as arising both out of conduct that is covered by a waiver provision (defective or dangerous culvert) and out of conduct that is described in an exception provision (management of flood waters or operation of a storm system).
¶ 12 We proceed in that manner in the paragraphs that follow. We (A) interpret the terms of the waiver of immunity for injuries caused by “a defective, unsafe, or dangerous condition of any highway [or] ... culvert”; (B) set forth our understanding of the exception for the “management of flood waters”; (C) construe the exception for the “operation of a flood or storm system”; and (D) establish the causation standard that applies in a case in which an injury can be understood to arise both out of conduct that is described in a waiver provision and in an exception.
A. Dangerous Condition of a Culvert
¶ 13 By statute, governmental immunity is waived for “any injury caused by ... a defective, unsafe, or dangerous condition of any highway [or] ... culvert.” Utah Code § 63G–7–301(3)(a)(i). The threshold question presented concerns the meaning of the statutory reference to a “defective, unsafe, or dangerous condition.”
¶ 14 The operative terms of this waiver provision are imported from tort law. Specifically, the reference to “defective, unsafe, or dangerous condition[s]” is an unmistakable transplant from the law of premises liability. In that field, this court has long held that landowners are liable in tort for defective or dangerous conditions on their land causing injuries to invitees. Similar standards are embedded in settled tort law in other jurisdictions. ¶ 15 These principles of premises liability have also long formed the basis of an exception to the common law doctrine of sovereign immunity. It is accordingly unsurprising to see such principles incorporated in the statute that displaced the common law in this field—the Governmental Immunity Act. At various points in that Act, the legislature speaks the language of premises liability in codifying waivers of the government's immunity and exceptions thereto. Utah Code § 63G–7–301(3)(a)(i) (waiving immunity for injuries caused by “a defective, unsafe, or dangerous condition” of various public properties); id. § 63G–7–301(3)(b)(i) (retaining immunity for “latent dangerous or latent defective condition[s]” of those same public properties); id. § 63G–7–301(1)(c) (retaining immunity for injuries caused by the Division of Water Resources' failure to deliver water when it is due to a “natural condition”); id. § 63G–7–301(5)(k) (retaining immunity for injuries caused by “any natural condition on publicly owned or controlled lands”).
See Glaittli v. State, 2014 UT 30, ¶¶ 26–27, 332 P.3d 953 (Lee, J., concurring in the judgment) (noting that the Governmental Immunity Act's waiver provisions incorporate “classic terms of art from premises liability in the law of tort” such as “dangerous condition” and “latent condition” and “natural condition” (internal quotation marks omitted)).
See, e.g., Burt v. Utah Light & Power Co., 26 Utah 157, 72 P. 497, 497–98 (1903) (affirming jury verdict in a case in which the defendant maintained a “tunnel conduit, 6 feet in diameter” near a county highway, “a break or rift in the conduit permitted about one-fourth of the water flowing ... to escape and flow to the county road” below, causing the death of plaintiff's fifteen-year-old-son; noting that “the road in question was in the same dangerous condition for several days,” and affirming the admissibility of evidence offered to show the “time and opportunity to the defendant to discover and remedy the dangerous condition”); Scoville v. Salt Lake City, 11 Utah 60, 39 P. 481, 482 (Utah Terr.1895) (holding that even if ice that had accumulated on a sidewalk was not from a “natural cause” but an “artificial cause,” it still constituted a “dangerous condition” and a “defect” in the sidewalk, resulting in the municipality's liability).
See Restatement (Second) of Torts§ 343 (1965) (setting forth the terms of the law under which “[a] possessor of land is subject to liability for physical harm caused to his invitees by a condition on the land”); 2 Dan B. Dobbs et al., The Law of Torts§ 276, at 84 (2d ed.2011) (“The landowner owes to the invitee a nondelegable duty of care to make conditions on the land reasonably safe....”); W. Page Keeton et al., Prosser and Keaton on Torts§ 61, at 419–32 (5th ed.1984) (discussing the premises liability notion of a dangerous condition as applied to injuries to invitees).
See Black's Law Dictionary335 (9th ed.2009) (noting that the existence of a dangerous condition may “result[ ] in [a] waiver of sovereign immunity”); see also Davis v. Provo City Corp., 1 Utah 2d 244, 265 P.2d 415, 419 (1953) (Crockett, J., concurring) (“Respected authorities have recognized that where a static condition of extreme danger is knowingly permitted to persist, that the city may be held liable [even if the city is engaged in a governmental function].”).
See also Glaittli, 2014 UT 30, ¶¶ 26–27, 332 P.3d 953 (Lee, J., concurring in the judgment) (citing these and other provisions).
¶ 16 We accordingly construe the language of the statutory waiver of immunity for a “defective, unsafe, or dangerous condition of any highway [or] ... culvert” to incorporate the term-of-art sense of these terms from premises liability in tort. “It can be no accident that the relevant, operative terms of the Governmental Immunity Act—those addressed to the government's immunity as regards its role as possessor of land or other property—coincide with the key terms that have long been used to define the scope of premises liability in tort.” Glaittli v. State, 2014 UT 30, ¶ 30, 332 P.3d 953 (Lee, J., concurring in the judgment). Thus, we read the statute's waiver of immunity for “defective, unsafe, or dangerous condition[s]” of highways and culverts “as a transplant from premises liability in tort law.” Id. ¶ 31. And we therefore “interpret that term in a manner incorporating the ‘old soil’ that it has long carried at common law.” Id.
See Maxfield v. Herbert, 2012 UT 44, ¶ 31, 284 P.3d 647 (“[W]hen a word or phrase is ‘transplanted from another legal source, whether the common law or other legislation, it brings the old soil with it.’ ”) (quoting Felix Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev.527, 537 (1947)); Hansen v. Hansen, 2012 UT 9, ¶ 19, 270 P.3d 531 (“Because the Utah support statute adopts a legal term of art ... with a settled meaning in the law, we interpret the statute to embrace the meaning of the term as it is understood in that context.”).
¶ 17 Under this term-of-art understanding, a dangerous condition is “[a] property defect creating a substantial risk of injury when the property is used in a reasonably foreseeable manner.” Black's Law Dictionary 335 (9th ed.2009). In other words, the statutory waiver for “defective, unsafe, or dangerous condition[s]” applies to injuries caused by a “defect[ ] or dangerous condition[ ] which [the government defendant] created, or of which [it] was aware, and which [it] should reasonably foresee would expose others to an unreasonable risk of harm.” Stephenson v. Warner, 581 P.2d 567, 568 (Utah 1978).
Accord Akins v. Cnty. of Sonoma, 67 Cal.2d 185, 60 Cal.Rptr. 499, 430 P.2d 57, 63 (1967) (noting statutory definition of “dangerous condition” as “a condition of property that creates a substantial ... risk of injury when such property ... is used with due care in a manner in which it is reasonably foreseeable that it will be used.” (internal quotation marks omitted)); Restatement (Second) of Torts§ 343 (“A possessor of land is subject to liability for physical harm caused to his invitees by a condition on the land if, but only if, he (a) knows or by the exercise of reasonable care would discover the condition, and should realize that it involves an unreasonable risk of harm to such invitees, and (b) should expect that they will not discover or realize the danger, or will fail to protect themselves against it, and (c) fails to exercise reasonable care to protect them against the danger.”).
B. Management of Flood Waters
¶ 18 The statutory waiver of immunity for defective or dangerous conditions of highways and culverts is subject to exceptions. One exception relevant here is as to injuries resulting from the “management of flood waters.” Utah Code § 63G–7–301(5)(p). UDOT seeks refuge in this provision. It advocates a broad understanding of “management of flood waters” that would encompass the broad range of conduct by UDOT that forms the basis of plaintiffs' allegations of negligence.
¶ 19 Plaintiffs offer a more limited view of this provision. They would have us limit the concept of “flood waters” to “major events that rarely occur”—a standard that in their view was not satisfied here because the rain storm in question was not so rare. Alternatively, plaintiffs insist that in any event UDOT did not engage in any effective “management” given that its efforts were minimal and unsuccessful.
¶ 20 We adopt a middle position. We read the statutory reference to “flood waters” to embrace a term-of-art understanding from tort law—an understanding that forecloses plaintiffs' principle limiting this provision to “major” or “rare” events. And we interpret “management” in a manner that is inconsistent with plaintiffs' theory requiring successful management.
1. Flood waters
¶ 21 “Flood waters” is not a statutorily defined term. As with the phrase “dangerous condition,” however, the term “flood waters” is a legal term of art incorporated from tort law and other fields.
¶ 22 In trespass liability in tort, “flood waters” is an established term of art. It is best understood in relation to other, related terms—“watercourse” and “standing water.” As the Restatement (Second) of Torts commentary indicates:
[I]n times of flood a stream may be swollen far beyond its normal size. The additional volume of water is called flood water. As long as this flood water remains a part of the stream and continues to flow with it either in the main channel or in a separate flood channel ..., it is part of the watercourse. But when flood water departs from the flowing stream and ceases to flow as part of the stream or becomes stagnant, it is no longer part of the watercourse.
¶ 23 Restatement (Second) of Torts § 841 cmt.k (1965). Thus, “flood water” is water that flows beyond a regular watercourse. And a “watercourse,” in turn, is “a stream of water ... flowing constantly or recurrently on the surface of the earth in a reasonably definite natural channel.” Id. § 841(1) ; see alsoBlack's Law Dictionary,supra at 1729 (defining “watercourse” as “[a] body of water, usu. of natural origin, flowing in a reasonably definite channel with bed and banks”).
This is not to say that all abnormal volumes of water within a watercourse are “flood water.” The Restatement goes on to note that “[s]ome streams have overflow or flood channels that carry the excess water of the stream during high water or flood periods.” Restatement (Second) of Torts§ 841 cmt.f. So long as these channels are “reasonably defined,” “they are part of the water-course.” Id. And additional volumes of water in those channels are not “flood waters.” Id.; see also Fitzpatrick v. Okanogan Cnty., 169 Wash.2d 598, 238 P.3d 1129, 1134 (2010) (en banc) (noting that water within the “flood channel” of a stream is still a part of “[a] natural watercourse,” and is not “flood water” or “surface water” for purposes of the common enemy doctrine (internal quotation marks omitted)); Mogle v. Moore, 16 Cal.2d 1, 104 P.2d 785, 789 (1940) (“The term ‘flood waters' is used to indicate waters which escape from a water course ... and flow over adjoining lands in no regular channel ....” (emphasis added)).
¶ 24 A “watercourse” includes the channel or bed itself and does not require that the bed carry running water year-round. So a watercourse may be dry “in time[s] of drought” and running with water at other times of the year. Restatement (Second) of Torts § 841 cmts. a, c. “Surface water” on the other hand, is “water from rain, melting snow, springs or seepage, or detached from subsiding floods, that lies or flows on the surface of the earth but does not form a part of a watercourse....” Id. § 846.
¶ 25 In accordance with these principles, rain water falling and accumulating outside of a “watercourse”—even in significant quantities—is not “flood water” but “surface water.” See, e.g., S. Pac. Co. v. Proebstel, 61 Ariz. 412, 150 P.2d 81, 83 (1944) ; Horton v. Goodenough, 184 Cal. 451, 194 P. 34, 37 (1920). Thus, it is only after that water joins a watercourse (whether the watercourse is natural or artificial) and then spills out over its banks (whether the cause is natural or artificial) that it becomes “flood water.” See, e.g., Maricopa Cnty. Mun. Water Conservation Dist. No. 1 v. Warford, 69 Ariz. 1, 206 P.2d 1168, 1175 (1949) ; Williams v. Carbon Cnty. Bd. of Educ., 780 P.2d 816, 818 (Utah 1989) (rejecting a school district's claim of immunity—where water collected on a parking lot the district had resurfaced damaged neighboring lands—on the ground that the damages resulted “from the runoff surface waters ” and not the “management of flood waters ” (emphasis added) (internal quotation marks omitted)).
¶ 26 Accordingly, the status of water as “flood water” does not depend, as plaintiffs suggest, on the rarity of the rainstorm that produced it. We don't doubt, as plaintiffs note, that the highly unusual flooding that occurred in downtown Salt Lake City in 1983 animated the legislature's attention to this matter. But, as we have stated before, it is “an erroneous premise” to assume that “statutory provisions are addressed only to the specific problems giving rise to their adoption.” Graves v. Ne. Servs., Inc., 2015 UT 28, ¶ 68, 345 P.3d 619 ; see also Hooban v. Unicity Int'l, Inc., 2012 UT 40, ¶ 17, 285 P.3d 766 (“[W]e cannot presume that the legislature meant only to deal with [one] particular problem, as legislative bodies often start with one problem in mind but then reach more broadly in their ultimate enactment.”). Thus, the fact that unusual rainfall animated the passage of the flood waters provision does not limit the scope of the “flood waters” provision to such events. The status of water as “flood water” depends on its connection and location relative to a watercourse, not the water's unusualness. “Flood waters” are those that have joined a watercourse and spilled over its banks. Extending these principles to the Governmental Immunity Act, UDOT retains immunity for injuries caused by its management of flood waters—of waters that have escaped a watercourse, by exceeding its bounds and flowing out over adjacent property. To the extent UDOT's activities are directed at a defective culvert as the source of the flood waters, on the other hand, the statute's immunity does not attach. See Utah Code § 63G–7–301(3)(a)(i).
2. Management
¶ 27 As with “flood waters,” the term “management” is not defined in the Governmental Immunity Act. And the parties again offer competing definitions of the term. UDOT advances a broad definition—encompassing any management-level decision, including a decision to leave flood water as is (i.e., omissions). Plaintiffs, on the other hand, contend for a narrower construction. They insist that management encompasses only active —and successful —attempts to direct flood water. We reject plaintiff's argument as untenable and adopt UDOT's position in part.
¶ 28 The term management “is not expressly defined in the Act, and does not appear to be a technical term of art.” Hi–County Prop. Rights Grp. v. Emmer, 2013 UT 33, ¶ 18, 304 P.3d 851. We accordingly “construe it to partake of the ordinary meaning the word would have to a reasonable person familiar with the usage and context of the language in question.” Id. (internal quotation marks omitted). A starting point for assessing ordinary meaning is the dictionary. We start there because the dictionary attests to a range of senses that a given term has been given over time. See id. ¶ 19.
¶ 29 The attested senses of “management” and of its root verb “manage” leave room for elements of the meanings advanced by both parties. As UDOT indicates, sometimes “management” is simply “the act or art of managing,” as in “the conducting or supervising of something,” especially “the executive function of planning, organizing, coordinating, directing, controlling, and supervising.” Webster's Third New International Dictionary 1372 (2002). Yet the plaintiffs' notion of “management” also finds tenable support in the dictionary. An alternative notion of “manage” is “to control and direct.” Id. And sometimes “manage” even conveys success —as in the notion of “manage” as “[t]o succeed in accomplishing, achieving, or producing.” American Heritage Dictionary of the English Language 1065 (5th ed.2011). But seeWebster's Third New International Dictionary 1372 (defining “manage” as to “handle either well or ill”).
See also American Heritage Dictionary of the English Language1064–65 (5th ed.2011) (“management”: “[t]he act, manner, or practice of managing; handling, supervision, or control”; “manage”: “[t]o have charge of; direct or administer”); Random House Dictionary of the English Language1166 (2d ed.1987) (“management”: “the act or manner of managing; handling, direction, or control”; “manage”: “to take charge or care of”); American Heritage Dictionary of the English Language792 (2d ed.1981) (“management”: “[t]he act, manner, or practice of managing, handling, or controlling something”; “manage”: “to direct or administer”).
See also American Heritage Dictionary of the English Language1065 (5th ed.) (providing one definition of “management” as “[s]kill in managing; executive ability”); Random House Dictionary of the English Language1166 (same).
¶ 30 Thus, the dictionary itself cannot resolve the contest of meanings of management put forward by the parties. But the structure and context of the Governmental Immunity Act serve to do so. In context, it makes no sense to read “manage” to be limited to the notion of successfully accomplishing or achieving something. A provision preserving the government's immunity from liability for injuries caused by the “management” of “flood waters” necessarily presumes that such management may not be successful. Otherwise there would be no need for immunity from suit. For that reason we cannot agree with plaintiffs that “making a futile effort that is doomed to failure ... does not qualify as management.”
¶ 31 We likewise conclude that “management” cannot be limited to the physical function of actively “control[ling] and direct[ing]” the flood waters themselves. Webster's Third New International Dictionary 1372. That notion of management makes no sense in the context of a provision preserving immunity for injuries arising out of “the management of flood waters, earthquakes, or natural disasters. ” Utah Code § 63G–7–301(5)(p) (emphasis added). One cannot control or direct an earthquake or a tornado. The only way that government can manage those phenomena is in the broad sense advanced by UDOT—by “conducting or supervising” the government's efforts to deal with them, by “planning, organizing, coordinating, and supervising.” Webster'S Third New International Dictionary 1372. So that sense of management must suffice to trigger the flood waters immunity, since under the canon of consistent usage “management” cannot properly mean one thing as applied to two of the objects in a series (earthquakes and natural disasters) but something else as applied to the other object in the same series (flood waters).
Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts173 (2012) (describing this canon and explaining that “the more connection [one use of a word in a statute] has with the [use of the word] under consideration, the more plausible the argument becomes” that they share the same meaning).
This is not to say that active efforts at directing or controlling flood waters would not qualify as “management.” Such efforts would undoubtedly count as one of the more common ways for the government to deal with flood waters, and would thus qualify for immunity. But the statute renders plaintiffs' limited view of “management” untenable. That term cannot be limited to active attempts to direct or control for reasons noted above.
¶ 32 Thus, we interpret the statute's immunity for injuries arising out of the “management” of flood waters to refer to executive efforts at planning, organizing, coordinating, or supervising the government's response to such waters. Such efforts could certainly encompass a studied decision to do nothing. On that basis we reject the plaintiffs' assertion that “omissions” could not count as “management.”
¶ 33 Yet we also stop short of a whole-hearted endorsement of UDOT's (and the district court's) position—that any and all “inaction” counts as “management.” Certainly there are some decisions not to act (like a decision to allocate flood mitigation resources to one affected area and not another) that easily qualify as managerial omissions. But if government literally does nothing—making no studied assessment of flood waters and rendering no decision as to how to deal with the problem—then there would appear to be no management at all, and thus no sense in which an injury could arise out of such management.
See, e.g., American Heritage Dictionary of the English Language1064–65 (5th ed.) (“management”: “[t]he act, manner, or practice of managing; handling, supervision, or control ” (emphasis added)).
C. Operation of a Flood or Storm System
¶ 34 A second basis for immunity under the Act is for injuries arising out of “the construction, repair, or operation of flood or storm systems.” Utah Code § 63G–7–301(5)(q). UDOT contends that the culvert in question is a “storm system.” Its principal argument for that conclusion is the notion that it groups the culvert in question together with other culverts in the same general area for certain reporting purposes. The district court agreed. It concluded that the culvert in question was part of a “storm system” in the sense of “an organized or established procedure or method or the set of materials or appliances used to carry it out.” Webster's Third New International Dictionary 2322; see also id. (“an aggregation ... of objects joined in regular interaction or interdependence”).
¶ 35 We disagree. UDOT has not asserted that the culvert is a storm system; at most it has suggested that it is part of one. That cannot be enough to sustain the applicability of the statutory exception. If that were sufficient, the “storm system[ ]” exception would swallow the “culvert” rule. In a statute that waives immunity for injuries caused by a defective “culvert” but reinstates immunity for injuries arising out of the operation of “storm systems,” a culvert cannot logically be equated with a storm system.
D. Proximate Cause
¶ 36 The Immunity Act's waivers extend to injuries “proximately caused” by a governmental entity's negligence, Utah Code § 63G–7–301(4), or “caused by” a “defective, unsafe, or dangerous condition” of a “culvert,” id. § 63G–7–301(3)(a)(i). In the exception provisions, however, immunity is reinstated for injuries that “arise[ ] out of, in connection with, or result[ ] from” a latent defective condition of a culvert, id. § 63G–7–301(b)(i) ; from the “management of flood waters,” id. § 63G–7–301(5)(p) ; or from the “construction, repair, or operation of flood or storm systems,”id. § 63G–7–301(5)(q).
¶ 37 The parties offer competing constructions of these causation standards. In UDOT's view, immunity attaches if there is any causal connection between the management of flood waters or the operation of a storm system and the plaintiffs' injuries. The plaintiffs disagree. They insist that “UDOT should not be immune from any damages ... for time immemorial,” and that there must be some “sufficient causal nexus”—based on the “actions taken,” the time between the government action and the injury, and “the degree of risk at issue.”
¶ 38 We concede that UDOT's approach finds support in our precedents. But we reject it on the ground that it yields an expansive notion of the Immunity Act's exceptions that effectively swallows the antecedent waivers. With that in mind, and in an attempt to reconcile the results of our prior decisions (if not the analysis of our prior opinions) with the operative terms of the statute, we repudiate the language of some of our prior opinions and adopt a new standard. Under the new standard (explained further below), we hold that an immunity-invoking condition (such as the management of flood waters or operation of a storm system) must be a proximate cause of the plaintiff's injuries in order to sustain the reinstatement of immunity.
¶ 39 In a number of prior opinions, we have concluded that a but-for causal connection is sufficient to trigger a statutory reinstatement of immunity under an exception provision of the Act. See Hoyer v. State, 2009 UT 38, ¶ 32, 212 P.3d 547 ; Blackner v. Dep't of Transp., 2002 UT 44, ¶ 15, 48 P.3d 949 ; Taylor ex rel. Taylor v. Ogden City Sch. Dist., 927 P.2d 159, 163 (Utah 1996). This test has some arguable basis in the statutory text. The exception provisions do not speak explicitly in terms of proximate cause. They reinstate immunity for an injury that arises out of, in connection with, or results from “one of the exceptions. Utah Code § 63G–7–301(5). And the results from” formulation may properly be understood as the invocation of a but-for test, see Burrage v. United States, ––– U.S. ––––, 134 S.Ct. 881, 887–89, 187 L.Ed.2d 715 (2014) (collecting cases holding that similar phrases indicate but-for causation), particularly if read in contrast to the express proximate cause standard in the waiver of immunity for negligence, see Utah Code § 63G–7–301(4).
¶ 40 In the abstract, the Immunity Act's different causation formulations—of “proximate[ ] cause[ ]” in the negligence waiver but “results from” in connection with the statutory exceptions—might imply two different standards of causation. This could arguably follow from the premise that where a statute “has used one term in one place, and a materially different term in another, the presumption is that the different term denotes a different idea.” Scalia & Garner,supra note 13, at 170. But this presumption is a rather weak one. Id. at 171 (referring to the canon as “often disregarded” and “particularly defeasible by context”). It can easily be rebutted by context. And in any event it begs the question—of whether the second term is ultimately a “materially different” one (or instead merely a synonym). In this instance we deem the reference to “arises out of” as synonymous with “proximately caused,” for reasons set forth below.
¶ 41 First, the “results from” formulation in the exception provisions is not obviously “materially different” from the proximate cause standard in the waiver provision. Sometimes “results from” is understood to convey the principle of proximate cause. For years this court and others interpreted the Federal Employers' Liability Act—which recognizes a tort cause of action for federal railroad employees for “injury or death resulting” from a rail-road's negligence—to require proof of proximate cause. Raab v. Utah Ry. Co., 2009 UT 61, ¶¶ 12, 41–42, 221 P.3d 219. And although the U.S. Supreme Court reached a different conclusion in CSX Transp., Inc. v. McBride, –––U.S. ––––, 131 S.Ct. 2630, 180 L.Ed.2d 637 (2011), that court can hardly be understood to have ruled that this terminology always conveys a but-for standard of causation. Instead it has recognized that courts “read phrases like ‘results from’ to require but-for causality” only “[w]here there is no textual or contextual indication to the contrary.” Burrage, 134 S.Ct. at 888 (interpreting the phrase “results from” in the federal Controlled Substances Act to require but-for causation, rather than a “substantial” factor causation standard, before imposing mandatory 20–year sentencing enhancement, without reaching question of whether it also requires a showing of proximate causation).
See CSX Transp., Inc. v. McBride, ––– U.S. ––––, 131 S.Ct. 2630, 2644–45, 180 L.Ed.2d 637 (2011) (Roberts, J., dissenting) (noting that the Supreme Court's “own cases, for 50 years after the passage of FELA, repeatedly recognized that proximate cause was required for recovery under that statute”); see also id. at 2646 (noting that the Court has “applied the standard requirement of proximate cause to actions under federal statutes where the text did not expressly provide for it” in everything from securities fraud, to RICO, to Antitrust, to environmental cases under NEPA).
¶ 42 Second, in this instance there are strong “textual” and “contextual” indications that the “results from” terminology of the exception provisions is in line with the “proximately caused” standard in the waiver provision. Most important is the fact that a “but-for” reading of “arises out of” would allow the statutory exceptions to nullify the immunity waivers in a number of common circumstances. It is easy to imagine circumstances in which the government's waiver of immunity for negligence would be completely erased by a but-for connection to wide-ranging activity covered by an exception—like the performance of a “discretionary function,” Utah Code § 63G–7–301(5)(a), “the collection of and assessment of taxes”, id. § 63G–7–301(5)(h), or the “issuance, denial, suspension, or revocation of, or by the failure or refusal to issue, deny, suspend, or revoke, any permit, license, certificate, approval, order, or similar authorization,” id. § 63G–7–301(5)(c).
¶ 43 If a but-for connection to these activities were enough to override the government's waiver of immunity for negligence, the statutory waiver would be overridden in a wide range of cases. That makes little sense in the context of a statute aimed at waiving governmental immunity for negligence and other governmental activity specified in the waiver provisions. The problem with the but-for standard is evident in the context of this case. Any case of a defective culvert would also (quite inevitably) encompass a but-for connection to discretionary functions, to taxes, and to permits or licenses. Yet if such a but-for connection were sufficient to reinstate immunity, the waiver for defective culverts would be ified. We reject the but-for standard on that basis.
See Kerr v. City of Salt Lake, 2013 UT 75, ¶¶ 16, 21–23, 322 P.3d 669 (rejecting a city's reliance on the “discretionary function” exception in a case of injury caused by the defective condition of a sidewalk; explaining that “[b]ecause all cities must decide how to allocate scarce public funds to maintain sidewalks, Salt Lake City's interpretation of the discretionary function exception would completely negate the explicit waiver of governmental immunity for defective or dangerous sidewalks”); Thayer v. Wash. Cnty. Sch. Dist., 2012 UT 31, ¶ 57, 285 P.3d 1142 (Lee, J., dissenting) (rejecting a but-for causation standard in the context of the licensing exception; noting that nearly all negligent driving decisions by government employees could be shielded by “plausibly trac[ing] the injury back to the issuance of the employee's driver's license”).
¶ 44 If we are to give effect to both sets of provisions (as we must), we cannot properly reinstate immunity on a mere showing of “some causal relationship,” e.g., Taylor, 927 P.2d at 163, between the excepted immune act and the plaintiff's injury. Our analysis must vindicate—and mediate—both sets of provisions (waivers and exceptions). We therefore hold that the exception provisions (reinstating immunity) are properly invoked only where a plaintiff's injury is proximately caused by immune conduct.
¶ 45 In so holding, we need not and do not overrule the results of our prior decisions, as most of them can be recast in terms consistent with the standard we adopt today. As noted above, three of our prior cases adopt the but-for standard—Taylor, Blackner , and Hoyer. All three cases arguably would have come out the same way under the proximate cause standard.
¶ 46 Taylor arose out of a fight between students in a middle school restroom. Id. One of the students pushed the plaintiff, whose hand then went through a glass window. Id. The student, through his parents, sued the school district, claiming negligence in not having installed safety-glass in the bathroom. Id. And the district then invoked immunity under an exception for injuries arising “out of assault.” Id. at 160. Adopting and applying a but-for test, we held that “the nerve and tendon damage to [the plaintiff's] hand was the result of [the plaintiff's] being shoved into the window by a fellow student.” Id. at 163. In Taylor, however, it was beyond question that the student pushing the plaintiff into the window was also a proximate cause of the injury. So the invocation of the but-for test was immaterial. The case would have come out the same way under the standard we adopt today.
¶ 47 In Blackner the plaintiff was stopped on a canyon road waiting for a UDOT front-end loader to clear snow from an avalanche that had occurred earlier that morning. 2002 UT 44, ¶ 4, 48 P.3d 949. A UDOT employee was inspecting avalanche-prone areas nearby and noticed that many cars, including the plaintiff's, were parked directly beneath a known avalanche zone. Id. ¶ 5. The employee cautioned others about the problem, but a determination was made that the loader could continue clearing the road. Id. Only moments after removing the last bit of snow, a second avalanche hit, injuring the plaintiff. Id. ¶¶ 6–7. UDOT invoked the “natural condition” exception under the Act, and we affirmed on the ground that the avalanches were natural conditions and both but-for causes of the plaintiff's injuries. Id. ¶¶ 15–16. Again, however, the avalanches were also quite apparently the proximate cause of the plaintiff's injuries—the impact on the plaintiff was direct and quite foreseeable. So the Blackner result would be correct under our new standard despite the fact that UDOT's alleged negligence was likely also a proximate cause.
See Fluehr v. City of Cape May, 159 N.J. 532, 732 A.2d 1035, 1041 (1999) (holding in statutory tort sovereign immunity case that “the natural conditions of the ocean,” rather than lifeguards' negligence, “were the legal cause ” of a surfer's broken neck (emphasis added)).
That is the net effect of a determination that a particular injury is a proximate result of both an act for which the government has waived immunity (such as negligence) and an act for which the statute reinstates it (such as a natural condition): To give effect to both the waiver and the exception, immunity is first waived and then reinstated.
¶ 48 The Hoyer case is more difficult to reconcile with our new standard, but even that case may arguably have come out the same way. In Hoyer the plaintiff sued the Department of Wildlife Resources for negligently failing to care for snakes it had seized from the plaintiff during the execution of a search warrant. 2009 UT 38, ¶¶ 2–3, 212 P.3d 547. The plaintiff offered to have an expert come in and care for the snakes while DWR held them, but DWR refused. Id. ¶ 3. All but eight of the snakes died as a result. Id. DWR claimed immunity under the “judicial or administrative proceeding” exception under the Act. Id. ¶ 24. We held that DWR was immune because “the snakes would not have died” “[b]ut for these judicial proceedings.” Id. ¶ 32. Yet again this same result arguably could still have obtained under a proximate cause inquiry—if, for example, it could be shown that the seizure of a fickle breed of boa constrictors to instigate a criminal proceeding heightened the scope of the risk of the snakes dying, making such a result foreseeable under the circumstances.
See, e.g., B.R. ex rel. Jeffs v. West, 2012 UT 11, ¶ 28, 275 P.3d 228 (explaining that “proximate cause” requires an inquiry into “whether the precise causal mechanism of a plaintiff's injuries was a foreseeable result of” the defendant's allegedly tortious actions); Chylinski v. Wal–Mart Stores, Inc., 150 F.3d 214, 217 (2d Cir.1998) (“In order to determine whether a defendant's conduct is the proximate cause of an injury, it is ... necessary to determine whether the harm caused is within the foreseeable scope of the risk created by the defendant's conduct.”).
¶ 49 The point is not to say that all of our prior cases are consistent with the new standard we adopt today. We adopt this standard because we deem it required by the structure and text of the Immunity Act. And we repudiate our prior decisions to the extent they are irreconcilable with our new standard.
III
¶ 50 The above legal background sets the stage for our analysis of the motion for summary judgment before us on appeal. In assessing UDOT's motion, we consider whether UDOT has established that “there is no genuine issue as to any material fact” and that it is “entitled to a judgment as a matter of law.” Utah R. Civ. P. 56(c). We conclude that UDOT has failed to carry that burden under the legal standards set forth above, and accordingly reverse.
¶ 51 As an initial matter, it seems apparent that plaintiffs have properly invoked the waiver of immunity for injuries caused by a defective culvert. The culvert contained a “defect” (the blockage) that created a “substantial risk of injury” (hydraulic piping and the eventual chasm). And the plaintiffs' injuries are (at least arguably) proximately connected to the defect in the culvert—in that the blockage heightened the scope of the risk in a manner leading to injuries that were easily foreseeable.
¶ 52 It seems equally as clear that at least some of the water involved in this incident was “flood water.” During the downpour, water began running in a watercourse—the gully and culvert. As a result of a defect in the culvert, the water backed up on the north side of SR–35, swelling beyond its bounds and flowing out “in no regular channel” over the surface of SR–35. Whether the water that remained pooled on the north side of the road was “flood water” remains to be seen, as neither of the parties presented any evidence on whether that water had gone beyond the bounds of the gully other than having spilled over the road at some point before receding. Ultimately, however, at least some of the water involved in this incident—the amount that went over the road—was “flood water” within the meaning of the Act.
¶ 53 That said, UDOT has not established that the government activity forming the basis of the plaintiffs' claims amounted to “management” of any such flood waters. The claim on review on summary judgment is rooted in the allegation that the “legal and proximate cause” of plaintiffs' injuries was UDOT's negligence in its “fail[ure] to eliminate the blockage” in the culvert, its failure to “disburse the water” that had collected in the gully, and its failure “to warn of or protect travelers from the dangerous conditions on SR–35.” Those allegations are focused on UDOT's actions in dealing with the culvert and with the water pooled next to the embankment (water that has not been shown to be “flood water”).
¶ 54 Plaintiffs' evidence on summary judgment was along the same lines. In the briefing on summary judgment in the district court, plaintiffs presented evidence focused on UDOT's alleged negligence in its attempts to unclog the culvert. In their statement of undisputed facts (which were eventually agreed to by UDOT for purposes of summary judgment), plaintiffs pointed to UDOT's knowledge of the dangers of pooled water next to an embankment, its failure to inspect the south side of the gully to see if the water was leaching through the embankment or flowing from the culvert, its failure to leave warning signs, and its minimal efforts in trying to unblock the culvert. And although plaintiffs referenced the water that spilled over SR–35, at no time did they allege or was any evidence put forward to show that their injuries were causally related to UDOT's management of that water. The only water UDOT was alleged to have managed in a manner causing injuries to plaintiffs was water that apparently was still inside a watercourse (the water pooled on the north side of the road in the gully).
The parties did not develop a factual record as to the character of the pooled water or the natural bounds of the gully on the north side of the road. We assume at this point that it was simply water still inside the watercourse (the gully). We do not foreclose the possibility that this water was “flood water,” though we do not endorse that view either. We reserve the matter for further inquiry on remand.
--------
¶ 55 UDOT's argument on appeal accepts this premise, at least in part. In its briefing to this court, UDOT asserts that “the culvert became blocked; the water in the adjacent wash escaped from the wash and flowed over normally dry round; and UDOT made an effort to eliminate the blockage.” “Those facts,” UDOT concludes, “establish that UDOT managed the flood water.” (Emphasis added). But that argument only serves to emphasize the nature of the plaintiffs' claims as set forth above. UDOT's “effort to eliminate the blockage” was an attempt to remedy a defect in a culvert, not to manage flood waters. See supra ¶ 25. UDOT has accordingly failed to establish that there are no genuine issues of fact and that it is entitled to judgment as a matter of law.
¶ 56 UDOT also failed to establish a right to judgment as a matter of law under the “storm system” exception. This argument fails as a matter of law. For reasons noted above, a single culvert might qualify as a component part of a “storm system,” but it is not a storm system in itself. So the plaintiffs' injuries cannot be shown to be proximately connected to the operation of such a system, as the only allegation here is in connection with UDOT's efforts in connection with this culvert.
IV
¶ 57 For the above reasons we hold that UDOT failed to carry its burden of establishing that there is no genuine issue of material fact and that it is entitled to judgment as a matter of law. In reversing the entry of summary judgment, however, we do not deem UDOT categorically ineligible for immunity. We simply hold that it has failed to carry its burden on the record before us.
¶ 58 In remanding, we leave open the possibility that UDOT may yet advance evidence that plaintiffs' injuries were proximately caused by its “management” of “flood waters” as those terms are defined above. To the extent it can do so, moreover, we note that it may ultimately qualify for immunity to the extent the plaintiffs' injuries are proximately connected to that activity.