JUSTICE BOYD delivered the opinion of the Court. It seems reasonable to think the law should presume that drivers will stop at a stop sign, at least absent some kind of emergency situation. And landowners, it seems, should be able to expect that drivers will obey a stop sign they post on their private property. The plaintiffs in this railroad-crossing case, however, urge us to accept that the law leaves room for other possibilities. Relying primarily on expert opinions, they contend that not every visible stop sign is visible enough or, even if it is, people cannot always expect that even reasonably prudent drivers will obey it. We conclude the evidence in this record is insufficient to create a fact issue on whether a railroad crossing protected by a stop sign in addition to a crossbuck sign was “extra-hazardous.” And assuming the crossing was “unreasonably dangerous,” no evidence exists that the landowner knew it was. We reverse the court of appeals’ judgment and reinstate the trial court’s summary judgment in favor of the railway company and the landowner. I. Background In 2015, railroad tracks intersected Texas roads in 14,013 locations, 4,817 of those on private land.[1] That same year, vehicle–train accidents resulted in nineteen fatalities at those crossings.[2] Of those nineteen fatalities, four occurred at private rail crossings.[3] This case concerns one of those fatal accidents. On September 12, 2015, Rolando Prado, Jr. died when his eastbound pickup was struck by a Union Pacific train at a rural crossing on private property owned by Ezra Alderman Ranches. Prado had just left work after his first day at a new jobsite a couple of miles away. It was just after 7:30 p.m., the sun was beginning to set behind him, and the skies were clear. Driving at or under the posted thirty-miles-per-hour speed limit,[4] Prado passed through an open ranch gate and followed the caliche road as it curved right about twenty-five degrees toward the railroad tracks a football field’s length away. Meanwhile, the Union Pacific train was heading north on the tracks at about fifty-eight miles per hour. The conductor saw Prado’s truck approaching the crossing and blew the train’s horn. About sixteen seconds before impact, the train’s horn blew for nearly four seconds, and it blew again from about nine seconds until almost four seconds before impact. Trees could have prevented Prado from seeing the approaching train until he was about six seconds from the crossing, and a fence line could have partially obstructed his view for another second or two: But nothing during those last six seconds, or even for some time before then, obstructed his view of a signpost planted just to the right of the road immediately before the tracks. The post displayed a standard red stop sign mounted above a slightly larger white sign that displayed black crossbucks and stated in black letters, “Private RR Crossing—No Trespassing—Right to Pass by Permission Subject to Control of Owner”: Everyone agrees that anyone who actually stops at or near the stop sign can clearly see a train coming from either direction. Prado, however, slowed to about seventeen miles per hour five seconds before the crossing, then to about nine miles per hour as he reached the stop sign, and then continued moving directly onto the tracks and into the train’s path. The conductor blew the horn again from about three seconds before impact until nearly three seconds after. Prado’s widow, children, and parents (collectively, the Prados) sued Union Pacific and Ezra Alderman Ranches (the Ranch) for negligence, negligence per se, and gross negligence.[5] Both defendants moved for summary judgment, and the trial court granted both motions. The Prados appealed, challenging only the adverse judgment on their negligence claims. The court of appeals reversed in part, holding that fact issues exist as to whether the crossing was extra-hazardous or unreasonably dangerous and whether the Ranch or Prado had actual knowledge of the crossing’s dangerous condition. 647 S.W.3d 731, 742, 746 (Tex. App.—San Antonio 2021). Union Pacific and the Ranch both filed petitions for review, which we granted. Because the liability standards differ for each defendant, we will address their arguments separately. II. Union Pacific Although the Prados originally alleged numerous ways in which Union Pacific negligently caused the accident, they contend in this Court only that the summary-judgment evidence creates a fact issue on whether the railroad crossing was extra-hazardous, thus imposing on Union Pacific a duty to use extraordinary means to warn drivers of an approaching train. We disagree. Railroad companies have a general legal duty “to give adequate warning of approaching trains, given whatever obstructions or other conditions exist.” Mo. Pac. R.R. Co. v. Limmer, 299 S.W.3d 78, 92 (Tex. 2009). Because “every railroad crossing is tinged with danger,” railroads must provide at least one warning sign that is adequate to “give notice of the proximity of the railroad and warn persons of the necessity of looking out for the cars.” Fitch v. Mo.-Kan.-Tex. Transp. Co., 441F.2d 1, 2 (5th Cir. 1971). For “an ordinary rural railroad crossing,” a crossbuck sign will typically satisfy that duty. Mo. Pac. R.R. Co. v. Cooper, 563 S.W.2d 233, 235 (Tex. 1978). If a crossing is “extra-hazardous,” however, the railroad company must “take extra safety measures to protect those using the crossing.” Fort Worth & Denver Ry. Co. v. Williams, 375 S.W.2d 279, 281 (Tex. 1964).[6] Such extraordinary warnings may include, for example, “lights or signal bells to warn persons approaching [the] crossing.” Fitch, 441 F.2d at 2. “Every railroad crossing is dangerous, but it is only crossings which are found to be extra hazardous that place the higher duty upon the railroad to use extraordinary means to warn travelers along the road.” Cooper, 563 S.W.2d at 235. The standard for categorizing a railroad crossing as extra-hazardous is high, requiring the plaintiff to show that a prudentperson exercising ordinary care cannot safely use the crossing unless extraordinary warnings or protections are provided. Id. (“A railroad crossing is extra hazardous when, because of surrounding conditions, it is so dangerous that persons using ordinary care cannot pass over it in safety without some warning other than the usual [crossbuck] sign.”).[7] A crossing may be extra-hazardous due to either permanent or temporary conditions, so its nature may vary from time to time. See Karr v. Panhandle & Santa Fe Ry. Co., 262 S.W.2d925, 930 (Tex. 1953). “The degree of danger,” therefore, “depends on the circumstances existing at the time of the accident.” Williams, 375 S.W.2d at 284. Whether a crossing is extra-hazardous at any particular time thus “depends on the facts of each case.” Id. at 283. Factors that may be relevant include visual obstructions like trees and buildings, the volume of train and vehicular traffic, the angle of the intersection, the grade and curvature of the road leading to the crossing, and any history of similar accidents at the crossing.[8] But a crossing is extra-hazardous only if such factors establish that a reasonably prudent person exercising ordinary care could not safely use it without extraordinary warning devices.[9] Union Pacific contends that, as a matter of law, the crossing at issue here was not extra-hazardous when Prado’s accident occurred. It notes that it posted not only a crossbuck sign but also a stop sign, which necessarily would cause a reasonably prudent person to come to a stop and look for approaching trains before proceeding across the tracks. A reasonably prudent person exercising ordinary care in that manner, Union Pacific argues, indisputably could have seen the train and heard its horn and thus avoided the accident altogether. As a matter of law, Union Pacific argues, a railroad crossing cannot be extra-hazardous if it’s only hazardous to those who fail to stop at a clearly visible stop sign. Relying primarily on two expert witnesses, the Prados argue this one was. The experts, however, relied on different theories to support that contention. The first—an engineer—testified that the stop sign and the crossbuck sign were inadequate to eliminate or control the hazards because (1) the crossbuck sign was not a standard crossbuck sign and was mounted below the stop sign instead of above it, as federal standards require, (2)the post on which the signs were mounted did not have a reflective strip as federal standards require, and (3) because of the slight curve in the road as it approaches the track, the signs were not visible early enough to permit a reasonable driver to timely respond to them. The Prados’ other expert witness—a human-factors expert—took a different approach. He agreed with the engineer that the curve in the road “competes for the driver’s attention” and thus could distract a driver from seeing the stop sign early enough to react. But contrary to the engineer, he testified that the stop sign was not obstructed “within any reasonable distance that we’re concerned about” and “visibility of the stop sign is not an issue” at all. He conceded that the stop sign was the same size, shape, and color as any standard stop sign around the state. And he conceded that a driver who stops at or near the stop sign can see an approaching train without any difficulty. But in his view, Union Pacific should not and could not reasonably expect drivers to stop at this stop sign because the sign “lacks credibility.” A stop sign lacks credibility, he explained, “when it’s consistently clear to drivers that there’s no need to stop.” In his opinion, this stop sign (like most stop signs placed with crossbucks at rural railroad crossings, he says) lacks credibility because (1) ninety-eight percent of the time there’s no train present, so it’s just “crying wolf” to ask people to stop when they know that stopping is practically never necessary, and (2) the stop sign is never enforced. When a sign tells drivers to stop but they know stopping is almost never necessary and will never be enforced, “you’re not going to get complete stops by most people.”In fact, observations he recorded at this crossing throughout one sample day revealed that only thirty-two percent of the drivers came to a stop at the stop sign. Both experts support their opinions with evidence that an unusually high number of trains (twenty-four per day)and vehicles (fifty to seventy-five per day) use this crossing. And they contend this crossing, which first opened in 1977, has had a high number (four) of prior accidents, including one fatality under similar circumstances and one accident just two weeks before Prado’s: · On November 21, 1988, at 9:35 a.m., a clear and sunny morning, a tractor-trailer driver heading east stopped with the trailer on the crossing. A train came while he was stopped and struck the trailer. The crossbuck signs were present, and there were no injuries. · On March 23, 2012, just after noon on a clear and sunny day, a driver heading east in a pickup failed to stop and was struck by an approaching train. The stop signs were present, and the driver was killed. · On March 28, 2012, at 4:30 a.m. on a rainy night, a driver heading east in a pickup pulling a trailer stopped with the trailer on the tracks. The crossbuck signs were present, and there were no injuries. · On August 29, 2015, at 8:00 p.m. on a clear evening at dusk, a tractor-trailer driver heading west (toward the setting sun) failed to stop and was struck by an approaching train. The stop signs and crossbuck signs were present, and there were no injuries. We conclude the Prados’ evidence is legally insufficient to support a finding that this crossing was extra-hazardous at the time of Prado’s accident. As explained, a crossing is only extra-hazardous if a reasonably prudent driver exercising ordinary care cannot traverse the crossing safely without some extraordinary warning device beyond the usual crossbuck sign. Cooper, 563 S.W.2d at 235. This crossing in fact contained a warning device in addition to a crossbuck sign—a stop sign. Under Texas law, a driver who approaches a railroad crossing marked only by a crossbuck sign must “yield” to any train that is “in hazardous proximity to the crossing” and must “stop” only if the train is close enough that stopping is “required for safety.” TEX. TRANSP. CODE § 545.251(c). But a driver who approaches a stop sign “shall stop” regardless of whether safety requires it, unless “directed to proceed by a police officer or traffic-control signal.” Id. § 544.010(a). A stop sign thus imposes the same legal requirement on a driver as the “extraordinary” flashing lights that often protect busier railroad crossings. See id. § 545.251(a)(1) (providing drivers “shall stop” if “a clearly visible railroad signal warns of the approach of a railroad train”). Relying on their experts’ testimony, however, the Prados contend the evidence creates a fact issue on whether these signs were effective to enable a reasonably prudent driver to stop. On the one hand, the engineering expert opined that the crossing was extra-hazardousin part because the crossbuck sign was not a standard crossbuck sign,[10] it was mounted below (instead of above) the stop sign, and the post on which both signs were mounted did not have a reflective strip as federal regulations require for some crossings. But the expert did not explain how any of these factors could have prevented a reasonably prudent driver approaching the crossing at the time of Prado’s accident from either seeing the signs or understanding that he was approaching a railroad crossing and must stop. The engineering expert also opined, however, that the curve in the road prevented drivers from seeing the signs early enough to be able to stop before the railroad tracks. Specifically, he testified that federal guidance requires at least 270 feet for a driver traveling thirty miles per hour to react to a warning sign, and these signs were 266 feet and ten inches away from the ranch gate, leaving insufficient distance for a driver to react. In his opinion, the defendants should have at least posted signs before the ranch gate, warning of a “Stop Ahead” or a “Railroad Crossing Ahead.” But the expert did not testify, and we have found no other evidence, that a driver could not have seen the signs or the tracks from the ranch gate or even earlier. To the contrary, the record contains a photograph, described as one taken from the crossing looking back up the road to the ranch gate, which confirms that the view from the gate to the sign by the tracks was unobstructed: To the extent the expert suggests that the signs were ineffective because of the difference between 270 feet and 266 feet and ten inches, we must reject that contention. Indeed, the Prados’ human-factors expert agreed that the stop sign was not obstructed “within any reasonable distance that we’re concerned about” and “visibility of the stop sign is not an issue” at all. We conclude the record contains no evidence that a reasonably prudent driver exercising reasonable care could not see the sign in time to obey it. The human-factors expert testified, however, that although a reasonably prudent driver could see the stop sign in plenty of time, he could not be expected to obey it. In his opinion, the sign lacked “credibility” because drivers knew there was almost never a reason to stop and no one ever enforced that requirement. But as we’ve explained, Texas law requires all drivers to stop at a stop sign regardless of whether safety requires it. See TEX.TRANSP.CODE § 544.010(a). And the law has long presumed that drivers will obey the law. See Int’l & Great N. Ry. Co. v. Gray, 65 Tex. 32, 36 (1885) (“[T]he public may have a right to conclude that the law and rules will be observed on any given occasion, and to act accordingly in their lawful use of the railway track.”).[11] Evidence that many or even most drivers would not stop at a particular stop sign does not establish that all reasonably prudent drivers would not, much less could not, stop at the sign. See Gulf, M. & O. R. Co. v. Grubbs, 260 So. 2d 837, 838–39 (Miss. 1972) (holding “trainmen had a right to assume that appellee would stop at the ‘Stop’ sign, as required bylaw” despite assertion that “[n]ot one in a thousand” drivers would). While the expert’s testimony might be sufficient to create a fact issue on whether Prado acted with reasonable prudence and exercised ordinary care when he rolled through the stop sign, it is no evidence that, objectively, a reasonably prudent driver using ordinary care could not stop at the sign or could not pass the crossing safely without some additional warning. The Prados also point to evidence—including the curvature of the road, the trees, the fence line, and the train’s speed—that they say creates a fact issue on whether Prado could see the approaching train early enough to be expected to stop for it. But because the stop sign and crossbuck sign required a reasonably prudent driver to stop before reaching the track, and because the evidence conclusively establishes that a driver could have clearly seen the approaching train once he had stopped, the question of when he could have first seen the train is irrelevant to whether the crossing was extra-hazardous. Finally, the Prados point to the evidence that this crossing had a relatively high volume of vehicular and train traffic compared to other rural crossings, as well as a comparatively high number of prior accidents. The prior accidents are relevant however, only if they occurred under “similar but not necessarily identical circumstances.” Cooper, 563 S.W.2d at 236. Here, only one of the prior accidents—the one involving an eastbound driver who failed to stop on a clear day, resulting in a fatality a few years earlier—meets that standard.[12] Evidence of one similar accident over a nearly forty-year period on a crossing with a relatively high traffic volume is simply no evidence that reasonably prudent drivers cannot safely traverse the crossing without warnings in addition to the stop sign and crossbuck sign. Because no evidence in this record supports a finding that a reasonably prudent driver exercising ordinary care could not safely traverse this railroad crossing without extraordinary warnings in addition to the stop sign and the crossbuck sign, Union Pacific established that the crossing was not extra-hazardous as a matter of law. And because the stop sign and crossbuck sign were sufficient warnings for an ordinarily dangerous crossing as a matter of law, the court of appeals erred in reversing the trial court’s summary judgment in Union Pacific’s favor. III. The Ranch The court of appeals held that Prado was a licensee, and no party contests that holding in this Court. 647 S.W.3d at 745. As a landowner, the Ranch owes a duty to “use ordinary care either to warn a licensee of, or to make reasonably safe, a dangerous condition of which the owner is aware and the licensee is not.” Sampson v. Univ. of Tex. at Austin, 500 S.W.3d 380, 385 (Tex. 2016) (quoting State Dep’t of Highways & Pub. Transp.v. Payne, 838 S.W.2d235, 237 (Tex. 1992)). The Ranch contends that no evidence exists to support a finding that the crossing was unreasonably dangerous or that the Ranch had actual knowledge that it was. The trial court agreed, but the court of appeals concluded the evidence raises a fact issue on both questions. 647 S.W.3d at 746. A condition is unreasonably dangerous if “there is a sufficient probability of a harmful event occurring that a reasonably prudent person would have foreseen it or some similar event as likely to happen.” United Supermarkets, LLC v. McIntire, 646 S.W.3d 800, 803 (Tex. 2022) (quoting Seideneck v. CalBayreuther Assocs., 451 S.W.2d 752, 754 (Tex. 1970)). In determining whether a condition is unreasonably dangerous, courts consider several factors, including (1) whether the relevant condition was clearly marked, (2) the size of the condition, (3) whether the condition previously caused injuries or generated complaints, (4) whether it substantially differed from similar conditions, (5)whether it was naturally occurring, and (6) whether it met applicable safety standards. Id.; Christ v. Tex. Dep’t of Transp., 664 S.W.3d 82, 87 (Tex. 2023). The Ranch contends it fulfilled any duty it owed as a landowner by posting speed-limit signs, clearing brush to maintain visibility of the track and signs, and ensuring the signs were posted. According to the Ranch, these actions eliminated any unreasonably dangerous condition and ensured that prudent drivers were fully aware of the risks. The Prados, in turn, contend that their experts’ testimony and the evidence regarding prior accidents constitutes some evidence that the crossing was unreasonably dangerous. Even if we assume that the Prados are correct, however, the record must also contain some evidence that would support a finding that the Ranch knew the crossing was unreasonably dangerous. Sampson, 500 S.W.3d at385. More specifically, they must provide some evidence that the Ranch actually knew that the crossing was unreasonably dangerous, not just that it should have known that it was. See State v. Tennison, 509 S.W.2d 560, 562 (Tex. 1974). “Although there is no one test for determining actual knowledge that a condition presents an unreasonable risk of harm, courts generally consider whether the premises owner has received reports of prior injuries or reports of the potential danger presented by the condition.” Sampson, 500 S.W.3d at 392 (quoting Univ. of Tex.-PanAm. v. Aguilar, 251 S.W.3d 511, 513 (Tex. 2008)). “Circumstantial evidence establishes actual knowledge only when it ‘either directly or by reasonable inference’ supports that conclusion.” City of Corsicana v. Stewart, 249 S.W.3d 412, 415 (Tex. 2008) (per curiam) (quoting State v. Gonzalez, 82 S.W.3d 322, 330 (Tex. 2002)). The Prados argue this record contains some evidence that the Ranch had actual knowledge that the crossing was unreasonably dangerous because the Ranch actually knew about the similar prior accident that resulted in a fatality. In support, it points to testimony of the Ranch’s president, Dustin Headlee, who said he was aware of a small cross that had been placed near the crossing and that his son, a Ranch employee, told him it was placed there after a previous accident. Headlee’s testimony is a little unclear: he says first that he knew of the cross but that he had “never noticed [it] very much.”He then said it was not until after Prado’s accident that his son told him “he’d heard that there was a prior accident.” These concessions by Headlee are simply not sufficient evidence to raise a fact issue on whether Headlee had actual knowledge of the prior accident. At most, they could support a finding that Headlee might have thought something had occurred at or near the crossing but that he was not actually aware of what occurred until after Prado’s accident. See Stewart,249 S.W.3dat 414–15 (holding that city did not have actual knowledge of flooded intersection despite knowing that intersection frequently flooded, police officers reported heavy rainfall, and city knew that motorists were stranded at other intersections); see also Reyes v. City of Laredo, 335 S.W.3d 605, 609 (Tex. 2010) (per curiam) (“Awareness of a potential problem is not actual knowledge of an existing danger.”). The Prados argue that even if Headlee did not have actual notice of the previous accident, the fact that his son knew is sufficient to impute the knowledge to the Ranch. Prado relies on Los Compadres Pescadores, L.L.C. v. Valdez, in which we held that “any knowledge” an agent had of a dangerous condition “must be imputed to” the principal. 622 S.W.3d 771, 787 (Tex. 2021) (citing La Sara Grain Co. v. First Nat’l Bank of Mercedes, 673 S.W.2d 558, 563 (Tex. 1984) (“[A] corporation[] is bound by the knowledge of one of its agents if that knowledge came to him in the course of the agent’s employment.”)).But even if the son’s knowledge could be imputed to the Ranch, there is little to impute. Headlee’s testimony establishes only that his son knew of a fatal accident at or near the crossing. Headlee testified only that the cross was “down a little ways from where they put a little memorial for Mr. Prado.” This testimony constitutes no evidence that Headlee’s son or anyone else employed by the Ranch knew that the previous fatality resulted from a train–vehicle collision, much less that they knew enough of the circumstances of that accident to know that the crossing was unreasonably dangerous. As Headlee testified, the Ranch, as owner of the road, is not privy to any type of reporting mechanism that provides notice of accidents at the crossing. And there is no evidence Headlee or any Ranch employee ever received or saw the accident report, which is the only evidence in the record that the prior fatality was caused by an eastbound driver failing to stop at the stop sign. The Prados also argue that other circumstantial evidence supports a reasonable inference that the Ranch “must have known” the crossing was dangerous. They point to the Ranch’s admission that it knew the crossing had a relatively heavy traffic volume, and their expert opined that the Ranch “had a duty and responsibility . . . to evaluate the existing railroad crossing warning signs . . . to determine whether the addition of automatic gates and/or flashing light signals would mitigate and control the known hazard[].” Thus, the Prados argue, the Ranch “must have known” the crossing was dangerous. We disagree for a few reasons. First, the expert did not opine that the Ranch had actual knowledge of any danger at the crossing. Second, circumstantial evidence may demonstrate actual knowledge only when the evidence makes knowledge of the dangerous condition “a virtual certainty.” Reyes, 335 S.W.3d at 609 (discussing and distinguishing City of San Antonio v. Rodriguez, 931 S.W.2d 535, 537 (Tex. 1996) (per curiam)). Prado’s expert opined only that the Ranch should have “evaluated” the crossing. Even assuming the Ranch had a duty to evaluate whether the crossing was dangerous and failed to do so, that would at most establish only that the Ranch should have known it was unreasonably dangerous, not that it actually knew it was. IV. Conclusion We hold that on this record the trial court correctly concluded the evidence establishes as a matter of law that the railroad crossing was not extra-hazardous at the time of Prado’s accident and that the Ranch did not have actual knowledge that it was unreasonably dangerous. We therefore reverse the court of appeals’ judgment and reinstate the trial court’s summary judgment in favor of Union Pacific and the Ranch. Jeffrey S. Boyd Justice OPINION DELIVERED: February 23, 2024
