Horton v. Kansas City Southern Railway

JUSTICE BOYD delivered the opinion of the Court. JUSTICE BUSBY filed a concurring opinion, in which Justice Devine, Justice Blacklock, and Justice Young joined. Petitioners sued the Kansas City Southern Railway Company (KC Southern) for the wrongful death of their mother, alleging KC Southern negligently maintained a railroad crossing by raising the crossing grade over time to form a “humped crossing” and by failing to replace a missing yield sign. In response to a single broad-form negligence question, the jury found both parties negligent and equally responsible for causing the accident, and the trial court entered judgment on the jury’s verdict. The court of appeals held that the evidence supports a finding that the missing yield sign proximately caused the accident, but that federal law preempts a negligence claim based on the humped crossing. 666 S.W.3d 1, 9, 10 (Tex. App.—Dallas 2021). Because the appellate court could not determine which of the two allegations the jury relied on when it found KC Southern negligent, it reversed and remanded for a new trial. Id. at 12. We disagree with the court of appeals on both issues and hold: (1) federal law does not preempt the humped-crossing claim, and (2) no evidence supports the jury’s finding that the absence of the yield sign proximately caused the accident. Based on these holdings, however, we agree with the court of appeals that a new trial is required. We thus affirm the court of appeals’ judgment, but for different reasons. I. Background A KC Southern train collided with Ladonna Sue Rigsby’s pickup truck as she drove across a railroad track on a rural county road near her home. The track had been there for overa century, and KC Southern maintained it by lifting and adding materials under the rails and ties, incrementally raising the track over the course of many years. This created a “humped crossing,” with the mid-point rising around thirty inches above the level road thirty feet away. No signal lights, bells, or barrier gates protected the crossing, but “crossbuck” signs—white, X-shaped signs reading “Railroad Crossing”—marked the tracks from both directions. The posts holding those signs also previously included yield signs, but—for reasons no one could explain—the yield signs were missing at the time of Rigsby’s accident. According to a video of the accident, recorded by a camera installed on the train, Rigsby slowed her vehicle to around seven miles per hour as she approached the track, and then to three or three-and-a-half miles per hour as she began to ascend the hump. Rigsby, who was deaf in her left ear, continued to cross the track as if she never saw or heard the train approaching from her left. She did not survive the collision. Rigsby’s adult children, Angela Horton and Kevin Houser (together, Horton), sued KC Southern, alleging negligence based on the humped crossing and missing yield sign. KC Southern filed a summary-judgment motion, asserting, among other things, that federal law preempts Horton’s claim. The trial court denied the motion, and KC Southern filed a motion for reconsideration, arguing federal law at least preempts Horton’s claim to the extent it is based on the humped crossing. The trial court did not rule on that motion, and the case proceeded to trial. The trial court submitteda single broad-form question to the jury, asking whether the negligence of Rigsby or KC Southern proximately caused the accident. KC Southern objected to the question, arguing the court should submit two separate negligence questions—one based on Horton’s humped-crossing allegation and the other on the missing-yield-sign allegation. The trial court overruled that objection. The jury found both Rigsby and KC Southern negligently caused the accident and assigned fifty percent of the responsibility to each. The trial court entered a final judgment based on the verdict, awarding Horton $200,000 in damages. KC Southern appealed, and the court of appeals reversed, with one justice dissenting. 666 S.W.3d at 4, 25. The court concluded the evidence supports liability under the yield-sign allegation, but federal law preempts the claim to the extent it is based on the humped-crossing allegation. Id. at 14–15. Because the court could not “determine whether the jury rested its liability determination on [Horton's] preempted humped crossing theory, which should not have been submitted, or the missing yield sign theory,” it remanded the case for a new trial only on the yield-sign allegation. Id. at 18. Horton and KC Southern both filed petitions for review, which we granted. II. Preemption We begin by addressing whether federal law preempts Horton’s negligence claim based on the humped crossing.[1] The United States Constitution provides that federal law “shall be the supreme Law of the Land; and the Judges in every State shall be bound thereby.” U.S. CONST. art. VI, cl. 2.Asa result, federal statutes may preempt state laws and render them ineffective. Altria Grp., Inc. v. Good, 555 U.S. 70, 76 (2008). They may do this expressly, by declaring that intent on the face of the statute, Arizona v. United States, 567 U.S. 387, 399 (2012), or impliedly, by demonstrating an intent to “occup[y] the field” or creating an irreconcilable “conflict,” Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 516 (1992) (first quoting Pac. Gas & Elec. Co. v. State Energy Res. Conservation & Dev. Comm’n, 461 U.S. 190, 204, (1983), and then quoting Fid. Fed. Sav. & Loan Ass’n v. De la Cuesta, 458 U.S. 141, 153 (1982)). KC Southern asserts—and the court of appeals agreed—that provisions of the federal ICC Termination Act of 1995, Pub. L. No. 104-88, 109 Stat. 803 (1995) (codified at 49 U.S.C. §§ 10101–16106), expressly and impliedly preempt Horton’s humped-crossing claim. We disagree. To explain, we describe the ICCT Act and its relevant provisions, the separate Federal Rail Safety Act and its relationship to the ICCT Act, and other court decisions addressing preemption under the ICCT Act before turning to our own preemption analysis. A. The ICCT Act Congress enacted the ICCT Act “to reform economic regulation of transportation, and for other purposes.” 109 Stat. at 803. The Act amended numerous federal statutes, including Subtitle IV of Title 49 of the United States Code, which addresses and governs interstate transportation. Id. at 803–04. Part A of Subtitle IV specifically addresses rail transportation. 49 U.S.C. §§ 10101–11908. KC Southern’s contention that the Act preempts Horton’s humped-crossing claim relies on this part, which we refer to as the Rail Provisions. As part of its reform of “economic regulation of transportation,” the ICCT Act abolished the Interstate Commerce Commission and created a new federal agency called the Surface Transportation Board, granting it “jurisdiction over transportation by rail carrier that is . . . only by railroad” or, in some circumstances, “by railroad and water.” 109 Stat. at 807; see 49 U.S.C. § 10501(a)(1). More specifically, Section 10501(b) grants the Surface Transportation Board “exclusive” jurisdiction over: (1) transportation by rail carriers, and the remedies provided in [the Rail Provisions] with respect to rates, classifications, rules (including car service, interchange, and other operating rules), practices, routes, services, and facilities of such carriers; and (2) the construction, acquisition, operation, abandonment, or discontinuance of spur, industrial, team, switching, or side tracks, or facilities, even if the tracks are located, or intended to be located, entirely in one State. 49 U.S.C. § 10501(b). Immediately following this jurisdictional grant, the section includes a preemption clause: “Except as otherwise provided in [the Rail Provisions], the remedies provided under [the Rail Provisions] with respect to regulation of rail transportation are exclusive and preempt the remedies provided under Federal or State law.” Id. (emphasis added). The Act defines the term “rail carrier” to mean “a person providing common carrier railroad transportation for compensation.” Id. § 10102(5). And it defines the term “transportation” to include a “facility, instrumentality, or equipment of any kind related to the movement of passengers or property, or both, by rail,” and “services related to that movement.” Id. § 10102(9). The parties here do not dispute that KC Southern is a “rail carrier” and that its tracks and crossings qualify as “facilities” or “equipment,” and thus “transportation.”[2] B. The Federal Rail Safety Act The ICCT Act is not the only federal statute that addresses rail transportation. The Federal Rail Safety Act (the Safety Act) created a separate statute contained within Subtitle V of Title 49 of the United States Code. See id. §§ 20101–21311. The Safety Act directly addresses rail-safety concerns, as its express purpose “is to promote safety in every area of railroad operations and reduce railroad-related accidents and incidents.” Id. § 20101. The Safety Act delegates rail-safety regulation to the federal Secretary of Transportation, not to the Surface Transportation Board. Id. § 20103. Because Horton alleges the humped crossing created safety issues, the Safety Act complicates the question of whether the ICCT Act preempts Horton’s common-law negligence claim, particularly because the Safety Act contains its own preemption provisions that explicitly address common-law claims concerning safety issues. The Safety Act’s preemption provisions permit states to adopt their own regulations governing rail safety, but only if the state regulation is “not incompatible with” federal regulations and “does not unreasonably burden interstate commerce.” Id. § 20106(a)(2); see CSX Transp., Inc. v. Easterwood, 507 U.S. 658, 664 (1993) (holding federal regulations adopted under the Safety Act preempt state-law regulations “if the federal regulations substantially subsume the subject matter of the relevant state law”). In 2002,however, Congress amended the Safety Act to “clarify” that nothing in Section 20106 “shall be construed to preempt an action under State law seeking damages for personal injury, death, or property damage” if the action alleges that a party (a) “failed to comply with the Federal standard of care established by a regulation or order issued by the Secretary of Transportation,” (b) “failed to comply with its own plan, rule, or standard that it created pursuant to a regulation or order issued by” the Secretary, or (c) “failed to comply with a State law, regulation, or order that is not incompatible with subsection (a)(2).” 49 U.S.C. §20106(b). As a result, federal rail-safety regulations can provide the “standard of care by which a defendant’s actions are judged for negligence.” Gallo v. Union Pac. R.R. Co., 372 F. Supp. 3d 470, 483 (W.D. Tex. 2019) (addressing cases). In the absence of a federal safety regulation covering the subject matter at issue, the Safety Act has no preemptive effect. Id. (citing Easterwood, 507 U.S. at 664). Addressing the relationship between the Safety Act and the ICCT Act, courts have generally agreed that “the federal statutory scheme places principal federal regulatory authority for rail safety with the Federal Railroad Administration . . . not the [Surface Transportation Board]” and thus the Safety Act “ provides the appropriate basis for analyzing whether a state law, regulation or order affecting rail safety is pre-empted by federal law.” Island Park, 559 F.3d at 107; see BNSF Ry. Co. v. Hiett, 22 F.4th 1190, 1195–96 (10th Cir.), cert. denied sub nom. City of Edmond v. BNSF Ry. Co., 142 S. Ct. 2835 (2022). As the Sixth Circuit explained, the Federal Railroad Administration’s and the Surface Transportation Board’s “complementary exercise of their statutory authority accurately reflects Congress’s intent for the [ICCT Act] and [the Safety Act] to be construed in pari materia,” granting the Federal Railroad Administration “primary authority over rail safety matters” and subjecting state laws related to rail safety to “preemption analysis under the Safety Act,” not under the ICCT Act. Tyrrell v. Norfolk S. Ry. Co., 248 F.3d 517, 523 (6th Cir. 2001). Moreover, the Safety Act specifically addresses safety issues affecting “grade crossings and railroad rights of way,” which it expressly refers to as “the railroad grade crossing problem.” Iowa, Chi. & E. R.R. Corp. v. Washington County, 384 F.3d 557, 559 (8th Cir. 2004) (quoting 49 U.S.C. § 20134(a)). As a result, the Safety Act, rather than the ICCT Act, appears to be the primary authority governing federal preemption of state-law claims addressing rail safety, and particularly the safety of railroad crossings. But the Safety Act and the ICCT Act have a “complicated” relationship, particularly “when a state action or common law claim falls at the intersection of [the ICCT Act's] realm of economic regulation and the [Safety Act]‘s realm of safety regulation.” Ezell v. Kan. City S. Ry. Co., 866 F.3d 294, 300 n.6 (5th Cir. 2017). The Surface Transportation Board has suggested that “the overwhelming weight of precedent holds that safety issues are generally governed by [the Safety Act] preemption.” Jimmy Lee Waneck & Starr Swearingen Waneck, et al.-Petition for Declaratory Order, FD 36167, 2018 WL 5723286, at *4 (S.T.B. Oct. 31, 2018). Although “there can be rare cases when both the Safety Act and [the ICCT Act] preemption may apply,” such cases do not include allegations of unsafe conditions at a railroad crossing. KC Southern does not contend the Safety Act preempts Horton’s humped-crossing negligence claim but instead asserts that the final sentence of Section 10501(b) of the ICCT Act expressly preempts Horton’s claim because it would regulate KC Southern’s construction, repair, and maintenance of the humped crossing. And in the event it does not, KC Southern contends the Rail Provisions impliedly preempt such claims. We must thus consider whether these facts provide one of the circumstances under which the ICCT Act may have a preemptive effect over a state-law rail-safety claim. C. Precedent We are not the first to consider this thorny question. Numerous courts have addressed the scope of preemption under Section 10501(b) and have reached varying conclusions. Many have addressed the question of whether the section preempts state and local legislative enactments—particularly statutes, regulations, and ordinances—as opposed to state common-law claims like the negligence claim at issue in this case. Most of these have concluded Section 10501(b) expressly or “completely”[3] preempts state and local legislation that attempts to  ”regulate,” “manage,” “govern,” or “restrain” a rail carrier’s “operations,” or at least its operations “in the economic realm.”[4] Conversely, others have held Section 10501(b) does not expressly preempt state and local legislation that does not attempt to “regulate” or “interfere with” a rail carrier’s operations.[5] Several courts have also addressed Section 10501(b)’s preemptive effect on common-law claims, including claims for negligence,[6] trespass, nuisance, and even inverse condemnation. Some have held that Section 10501(b) expressly preempted such common-law claims—which complained of noise and vibrations from a rail carrier’s operations or of a train’s speed, length, scheduling, use of side tracks, and extended blocking of crossings—because such claims (like the state and local legislation discussed above) sought to regulate, manage, or govern a rail carrier’s operations or rail transportation.[7] Other courts, however, have held that Section 10501(b) does not expressly preempt common-law claims, for various reasons. Some concluded the rail carrier’s allegedly negligent conduct was not the type of “transportation”-related conduct the Rail Provisions address and for which they provide remedies that carry preemptive power under Section 10501(b).[8] Others focused on the idea that an adverse judgment on a common-law claim would not necessarily require the carrier to alter its operations, see Elam, 635F.3dat 813–14, or observed that a common-law claim seeking only compensatory damages does not attempt to “manage” or “govern” operations, even in the “economic realm,” id. at 813.[9] And several concluded that common-law claims, by their nature, do not “directly” address rail-carrier operations and instead have only an “incidental” effect on rail transportation. See, e.g., id. (citing Franks, 593 F.3d at 411). The Fifth Circuit focused on this direct-versus-incidental distinction in Franks, holding that Section 10501(b) did not expressly preempt a landowner’s claim seeking to enforce a covenant in an easement deed to prevent a rail carrier from removing a track crossing on the landowner’s land, explaining that the claim was “governed by Louisiana property laws and rules of civil procedure that have nothing to do with railroad crossings. Railroads are only affected when the [easement] happens to cross a railroad.” 593 F.3d at 411. The same court later applied that distinction to conclude the ICCT Act did not expressly preempt a negligent-failure-to-warn claim, explaining that, “[l]ike state property laws and rules of civil procedure that generally ‘have nothing to do with railroad crossings,’ the effects of state negligence law on rail operations are merely incidental.” Elam, 635 F.3d at 813 (quoting Franks, 593 F.3d at 411). As a Texas federal district court explained when applying this Fifth Circuit approach, although tort claims may “affect the management or governance of railroads if the railroad company is the tortfeasor, these claims arise under state common law and are not intended to regulate railroad transportation even if they may incidentally affect it.” Gallo, 372 F. Supp. 3d at 480.[10] Two other federal district courts within the Fifth Circuit, however, have addressed “humped crossing” negligence claims like the one at issue here, and both concluded that Section 10501(b) expressly  preempted such claims. Addressing claims by those injured and killed when a train struck a tour bus that got stuck on a humped crossing, a federal district court in Mississippi reasoned that the plaintiffs’ negligent-maintenance claims were “tantamount to a claim regarding the design and construction of the crossing” and thus “directly attempt to manage or govern a railroad’s decisions in the economic realm such as the construction and operation of tracks.” Waneck v. CSX Corp., No. 1:17CV106-HSO-JCG, 2018 WL 1546373, at *5 (S.D. Miss. Mar. 29, 2018). Addressing similar claims arising from a similar train–bus collision, a Texas federal district court reasoned that an adverse judgment would require “changes in design and construction of railroad tracks and crossings” and thus “have the effect of managing or governing rail transportation in the economic realm.” Voight v. CSX Transp., Inc., No. 3:17-CV-01018-N, slip op. at 7, 9 (N.D. Tex. June 19, 2017). A federal district court in Kentucky, however, expressly rejected the reasoning in Voight and Waneck, concluding that both “cases are inconsistent with Sixth Circuit and Supreme Court law on complete preemption.” Minton v. Paducah & Louisville Ry., Inc., 423 F. Supp. 3d 375, 383 (W.D. Ky. 2019). And the Surface Transportation Board itself rejected their reasoning in response to requests for administrative rulings from the parties involved in Waneck. See Waneck, 2018 WL 5723286, at *4. According to the Board, the question of federal preemption of these types of humped-crossing claims “should be governed by the preemption provisions of the Federal Railway Safety Act . . . and not by 49 U.S.C. § 10501(b).” Id. at *1. Expressly rejecting the courts’ holdings in Voight and Waneck, the Board opined that the plaintiffs’ negligent-maintenance claims based on humped crossings “appear to be focused on purely safety-related issues” and thus “are not in direct conflict with the Board’s exclusive jurisdiction over transportation that is part of the interstate rail network.” Id. at *7. Having described the ICCT Act’s relevant provisions and its complex relationship with the Safety Act, as well as the broad array of precedent considering this question, we now turn to the question of whether Section 10501(b) expressly or impliedly preempts Horton’s humped-crossing claim. We agree with KC Southern that federal law can expressly preempt common-law negligence claims in some circumstances,[11] but we hold that the ICCT Act does not expressly preempt Horton’s humped-crossing claim. We reach this conclusion based on the ICCT Act’s clear language, which much of the precedent has underemphasized. D. Express Preemption It is clear that Section 10501(b) expressly “preempts” something, but we must focus on the statute’s language to determine the scope of that preemption. “Where, as in this case, Congress has superseded state legislation by statute, our task is to ‘identify the domain expressly pre-empted.’ To do so, we focus first on the statutory language, ‘which necessarily contains the best evidence of Congress’ pre-emptive intent.’” Dan’s City Used Cars, Inc. v. Pelkey, 569 U.S. 251, 260 (2013) (first quoting Lorillard Tobacco Co. v. Reilly, 533 U.S. 525, 541 (2001), and then quoting Easterwood, 507 U.S. at 664). As noted, Section 10501(b) states that “the remedies provided” under the Rail Provisions “with respect to regulation of rail transportation” are “exclusive and preempt” the “remedies” provided under state law. 49 U.S.C. § 10501(b). In a statement other courts have often quoted, one federal district court suggested that it “is difficult to imagine a broader statement of Congress’s intent to preempt state regulatory authority over railroad operations.” CSX Transp., Inc. v. Ga. Pub. Serv. Comm’n, 944 F. Supp. 1573, 1581 (N.D. Ga. 1996). We agree with the Eleventh Circuit, however, that “[a]lthough this subsection on its surface seems to provide for broad pre-emption, the text contains limitations on the reach of pre-emption.” West Palm Beach, 266 F.3d at 1330. Specifically, the section grants preemptive power only to the “remedies” provided in the Rail Provisions “with respect to regulation of rail transportation.” 49 U.S.C. § 10501(b). To determine the scope of preemption, we must explore the section’s references to “remedies,” “with respect to,” and “regulation” of rail transportation. 1. “Remedies” Section 10501(b)’s preemption clause uses “remedies” twice—first to describe what has preemptive power (remedies “provided under [the Rail Provisions] with respect to regulation of rail transportation”), and then to describe what is preempted (remedies “provided under Federal or State law”). Id. Thus, per the clause’s explicit text, the “remedies” that preempt state law are those provided in the Rail Provisions. The Rail Provisions provide “remedies” in Sections 11701 through 11708, including the recovery of “damages sustained by a person as a result of an act or omission of” a rail carrier “in violation of” the Rail Provisions. Id. § 11704(b). A person who sustains such damages as a result of a violation may assert a claim for those damages before the Surface Transportation Board and through a civil action to enforce the Board’s order. Id. § 11704(c)(1)–(2), (d)(1)–(2). Relying primarily on the ICCT Act’s purpose and legislative history, some courts have concluded that the Rail Provisions’ remedies address only economic regulation of rail transportation. See, e.g., Elam, 635 F.3d at 805 (relying on legislative history to “observe Congress was particularly concerned about state economic regulation of railroads when it enacted the [ICCT Act]“). Based on this conclusion, they have construed Section 10501(b) to grant preemptive power to (and against) only “laws (and remedies based on such laws) that directly attempt to manage or govern a railroad’s decisions in the economic realm.” Id. at 807 (emphasis added).[12] Other courts have disagreed, see, e.g., N.Y. Susquehanna & W. Ry. Corp. v. Jackson, 500 F.3d 238, 252 (3d Cir. 2007) (“[The ICCT Act] does not preempt only explicit economic regulation.”), as has the Surface Transportation Board, see CSX Transp., Inc.—Petition for Declaratory Order, Fed. Carr. Cas. (CCH) 37186, 2005 WL 584026, at *7−8 (S.T.B. Mar. 14, 2005) (concluding that Section 10501′s preemption scope “is broader than just direct economic regulation of railroads” and that states and municipalities “cannot take an action that would have the effect of foreclosing or unduly restricting a railroad’s ability to conduct its operations”). Based on the ICCT Act’s text, we also disagree. The Act’s introduction states that its purpose is “to reform economic regulation of transportation, and for other purposes,” 109 Stat. at 803 (emphasis added), and we find nothing in its text that strictly limits its remedies to address only violations of “economic” regulations. Yet we note that, as even the Board has suggested, characterizing the ICCT Act as regulating “economic” operations can provide a helpful label for distinguishing between the ICCT Act and the Safety Act, which generally governs rail-safety issues, along with their respective preemption provisions. See Waneck, 2018 WL 5723286, at *4 (agreeing that “safety issues are generally governed by [the Safety Act] preemption” but concluding “there can be rare cases when both [the Safety Act] and [the ICCT Act] preemption may apply”). In other words, whether we label[13] the ICCT Act’s regulatory scope as “economic” or merely “non-safety operational-related,” as the Board has labeled it,[14] the important point is that the two Acts generally address different regulatory scopes, although they may overlap. These observations about the remedies the Rail Provisions provide (in contrast to those the Safety Act provides) matter here because the types of state-law “remedies” Section 10501(b) preempts are the same types of federal-law remedies the Rail Provisions provide. Although the sentence does not expressly limit its reference to state-provided remedies to those “with respect to regulation of rail transportation,” the sentence clearly imposes that limitation. See, e.g., Franks, 593 F.3d at 410 (explaining that the preempted state-law “remedies receive their meaning from the earlier part of the sentence”). We thus conclude that Section 10501(b) grants preemptive power only to remedies provided in the Rail Provisions “with respect to regulation of rail transportation,” and those remedies only preempt state-law remedies “with respect to regulation of rail transportation.” We thus turn to the meaning of that limitation. 2. “With respect to regulation” Consistent with the jurisdiction the ICCT Act grants exclusively to the Surface Transportation Board, the Rail Provisions address a rail carrier’s “rates, classifications, rules . . . practices, routes, services, and facilities” and the “construction, acquisition, operation, abandonment, or discontinuance of” its tracks and facilities. 49 U.S.C. § 10501(b); see Franks, 593 F.3d at 409 (addressing the Act’s exclusive remedies); Jackson, 500 F.3d at 252 (same). Section 10501(b), however, does not grant preemptive power to or against all laws that merely affect these types of activities, but instead grants such power only to and against remedies provided “with respect to regulation” of such activities. Rather than encompassing any law that might indirectly touch on the relevant subject matter, the phrase “with respect to” limits the clause’s preemptive effect so that it includes only those remedies that directly “concern” or “involve” the matter the clause describes. Dan’s City, 569 U.S. at 261–62.[15] And the matter the clause describes— “regulation of rail transportation”—further narrows its preemptive scope. Section 10501(b) does not preempt “all state laws”[16] or even all state-law remedies “with respect to rail transportation.” Instead, it preempts state-law remedies with respect to “regulation of” rail transportation. 49 U.S.C. § 10501(b). The inclusion of the word “regulation” within the description of the section’s preemptive scope “necessarily means something qualitatively different from remedies ‘with respect to rail transportation.’” West Palm Beach, 266F.3dat 1331. Both the Supreme Court and this Court have recognized the narrow scope of a statutory reference to laws that “regulate” a subject matter. Concluding that an insured’s common-law claims against an insurance carrier for breach of contract, breach of fiduciary duty, and fraud did not “regulate insurance,” the Supreme Court explained that a “common-sense view of the word ‘regulates’ would lead to the conclusion that in order to regulate insurance, a law must not just have an impact on the insurance industry, but must be specifically directed toward that industry.” Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 50 (1987) (emphasis added).[17] And applying that same “common-sense view,” we held that Texas laws that empower state-agency executives to discipline licensees for violations of other laws that “regulate” abortions are not themselves laws that regulate abortions because such laws are not “specifically directed toward” that subject. Whole Woman’s Health v. Jackson, 642 S.W.3d 569, 578 (Tex. 2022). Similarly, as discussed above, the Fifth Circuit has held that Section 10501(b) expressly preempts state “laws that have the effect of managing or governing rail transportation.” Franks, 593 F.3d at 410 (emphasis added). Under this “common-sense view,” a state statute that restricts the amount of time a train may block a crossing “regulates” rail transportation because the statute “homes in on ‘railroad compan[ies]‘” and “has no application” at all “except with respect to the operation of railroads at rail crossings.” Elam, 635 F.3d at 807 (quoting MISS. CODE § 77–9–235). But general laws that are not specifically directed toward rail transportation, such as general state property laws, “are not meant to regulate railroad transportation, though at times they may have an incidental effect on railroad transportation.” Franks, 593 F.3d at 411. The same is true of “standard building, fire, and electrical codes,” which do not specifically “target[] the railroad industry,” Jackson, 500 F.3d at 254, and even common-law contract laws, which enforce “[v]oluntary agreements between parties” and “are not presumptively regulatory acts,” PCS Phosphate Co. v. Norfolk S. Corp., 559 F.3d 212, 218 (4th Cir. 2009). In the same way, a common-law negligence claim does not ordinarily “regulate” rail transportation because it is not specifically directed toward rail transportation and only incidentally affects rail transportation when the alleged tortfeasor happens to be a rail carrier. As we have previously recognized, although “the term ‘law’ can include both common law and statutory law” and “jury awards can have an effect akin to regulation,” generally, such a “regulatory effect is not as direct as that of positive enactments,” and thus a federal law that preempted state “laws and regulations” did not preempt state common-law claims. Moore v. Brunswick Bowling & Billiards Corp., 889 S.W.2d 246, 247, 249–50 (Tex. 1994) (holding the Federal Boat Safety Act did not preempt a “state law tort claim that a boat was defective because it lacked a propeller guard”).[18] The combination of these phrases, with their respective histories, is determinative, especially when considered within the context of the chapter that also includes the Safety Act. Section 10501(b) does not expressly preempt this common-law negligence claim. It may be, as other courts have held, that a common-law claim could so directly seek to control, manage, or govern the core operational functions of a rail carrier that it could only be said to seek a “remedy with respect to regulation of rail transportation.”[19] But in our view, negligence claims based on railroad-crossing safety will rarely meet that standard. Considering Section 10501(b)’s language limiting its preemptive effect to “remedies” provided by state law “with respect to regulation” of rail transportation, the Safety Act’s primary role in addressing rail-safety issues, and its clause expressly allowing certain common-law claims, we conclude that Section 10501(b) does not expressly preempt Horton’s common-law claim that KC Southern negligently maintained the crossing resulting in an allegedly distracting and dangerous hump. E. Implied Preemption We now turn to the issue of whether Section 10501(b) impliedly preempts Horton’s humped-crossing claim. As stated above, a federal law may impliedly preempt state law if Congress intended federal law to occupy the field exclusively or if the state law is in actual conflict with and creates an obstacle for the federal law. See Cipollone, 505 U.S. at 545 (quoting Hines v. Davidowitz, 312 U.S. 52, 67 (1941)). Within the context of the ICCT Act, courts have relied on the “conflict” prong of implied preemption to hold that Section 10501(b) may impliedly preempt a generally applicable state-law remedy if, as applied to a particular case, that remedy has the effect of “unreasonably burdening or interfering with rail transportation.” Elam, 635 F.3d at 805 (quoting Franks, 593 F.3d at 410); see CSX Transp., Inc. v. City of Sebree, 924 F.3d 276, 284 (6th Cir. 2019); Tubbs, 812 F.3d at 1145–46; PCS Phosphate Co., 559 F.3d at 220–21; Jackson, 500 F.3d at 254; Emerson, 503 F.3d at 1133. We might question whether an unreasonable burden or interference, standing alone, is sufficient to create the kind of legal “conflict” required to establish implied preemption, but the parties here agree that this is the proper test in this context. So, for purposes of this case, and for the sake of consistency with courts across the country, we will apply this standard here, without suggesting that it provides the proper test for evaluating conflict preemption in other contexts. A party arguing for implied preemption has the burden on that issue. Mo. Pac. R.R. v. Limmer, 299 S.W.3d 78, 84 (Tex. 2009). In the context of Section 10501(b) and a claim involving a rail crossing, this means the rail carrier must provide specific evidence regarding the crossing at issue, rather than rely on assertions about the effect of grade crossings on rail transportation in general. See Elam, 635 F.3d at 813; Franks, 593 F.3d at 415. This requirement is consistent with the requirements of an “as-applied” preemption analysis, which considers the degree to which a specific scenario conflicts with requirements and objectives of the federal law at issue. Thus, for example, evidence that private crossings can affect drainage, increase track maintenance costs, and cause trains to move at slower speeds, without evidence that the particular private crossing at issue would have those effects, is insufficient to establish the kind of unreasonable burden or interference necessary to trigger implied preemption. Franks, 593 F.3d at 415; see Emerson, 503 F.3d at 1133. Here, Horton’s negligence claim does not seek a court order requiring KC Southern to alter its operations. Horton sought only economic damages, not any sort of injunctive requirement that could prevent KC Southern from maintaining the lines, crossings, services, rates, or other operations in which it was engaged at the time of Rigsby’s death. Consistent with the typical purpose of a negligence claim, Horton sought only compensation for damages resulting from the accident. See Akin, Gump, Strauss, Hauer & Feld, L.L.P. v. Nat’l Dev. & Rsch. Corp., 299 S.W.3d 106, 122 (Tex. 2009) (“A negligence claim . . . is about compensating an injured party.”). As KC Southern notes, however, the effect of a successful negligence claim can far exceed the payment of compensatory damages to a particular claimant. KC Southern contends that a successful claim in this case would effectively require KC Southern to alter all of its humped crossings, or at least this particular line and crossing, and that requirement would unreasonably burden KC Southern or interfere with its operations. The trial court made no factual findings on this assertion, so we must presume the court resolved any factual disputes against preemption. See Pharo v. Chambers County, 922 S.W.2d 945, 948 (Tex. 1996). Thus, KC Southern must conclusively demonstrate its contention as a matter of law. Nevertheless, even if we assume that KC Southern could have established an unreasonable burden or interference based on changes in its tracks and operations (as opposed to the damages award), we conclude KC Southern’s evidence is insufficient here. A lack of definitive evidence regarding costs and operational methods will render evidence insufficient to establish an unreasonable burden or interference as a matter of law. In Gallo, for example, the federal district court found the evidence of unreasonable burden was insufficient when the parties offered competing visions for how a railroad could remedy a potential drainage issue, and different cost estimates for those changes, without any clear evidence that the defendant’s more burdensome proposed change was required. See 372 F. Supp. 3d at 481. On the other hand, in Union Pacific Railroad Co. v. Taylor Truck Line, Inc., No. 15-0074, 2018 WL 1750516 (W.D. La. Apr. 10, 2018), the railroad offered evidence that lowering the specific crossing at issue would require extensive studies and redesign of the drainage, signal circuits, and nearby crossings; rehabilitation of the switches, adjacent tracks, and drainage culverts; and extensive construction work at the crossing and three-quarters of a mile of track on both sides, at a cost of approximately $2,000,000. Id. at *7–8. It also provided evidence that the work would require closure of the mainline track through the area for at least three to four days, impacting five road crossings. Id. The court found this evidence, in the absence of any sufficient opposing evidence, established that the change would pose an unreasonable burden and result in regulation of the railroad. Id. at *8–9. Here, KC Southern failed to provide such definitive evidence. At trial, Horton’s railroad-maintenance expert, Allen Blackwell, testified without opposition that KC Southern could address the humped crossing either by lowering the track to the level of the county road or by raising the county road to create a gradual incline up to the track. KC Southern did not present its own expert or other specific evidence of the likely cost and burden of either option. Instead, it relied on Blackwell’s deposition testimony that lowering the track might cost up to $300,000. But at trial, Blackwell testified that number was inaccurate and that the project would more likely cost between $50,000 to $150,000. Horton also offered evidence that the cost of removing the hump would be comparable to KC Southern’s routine-maintenance process, which, incidentally, created the hump in the first place, saving KC Southern about $100,000 that could be deducted from the cost of removing the hump. Even if this cost were unreasonable, courts have generally concluded that increased costs alone cannot create the type of “unreasonable” burden or interference necessary to trigger implied preemption. See Barrois, 533 F.3d at 335; Adrian & Blissfield R.R. v. Village of Blissfield, 550 F.3d 533, 541 (6th Cir. 2008); City of Sebree, 924 F.3d at 284–85. KC Southern therefore also relied on Horton’s expert to describe the type of construction required to lower the track to the road level. Blackwell agreed that lowering the track by thirty-two inches—the amount required for KC Southern to comply with its own adopted safety guidelines—would require “major revision to the alignment, elevation, or profile of the track.” According to Blackwell, KC Southern would have to remove asphalt up to six feet away from each side of the track rails, cut and remove the rails and track panel at the crossing, resurface the crossing area with new ballast, pour new asphalt, and then install new crossing panels. The parties also agreed that KC Southern would need to extend the renovation out at least 661 feet on each side to maintain required evenness of the track. KC Southern’s corporate representative testified that a culvert present under the track could make it impossible to lower the track by the full thirty-two inches. Thus, KC Southern argued this project would far exceed the scope and cost of a typical crossing rehabilitation. On the other hand, Horton provided evidence that, despite the seemingly broad scope of work, it would take relatively little time. Blackwell testified that undercutting the track would take the same amount of time as a routine resurfacing (six to eight hours). He also specified that, to the extent work could not be completed in a day, KC Southern could run trains in the evening at a reduced speed, with subsequent speed and tonnage restrictions lasting no more than forty-eight hours. KC Southern did not contest this evidence. All of this evidence, however, addressed only one possible means of eliminating the hump. Other testimony established that KC Southern could address the safety issue by raising the county road on each side of the crossing. Blackwell testified that, while KC Southern would not have a sufficient right-of-way to complete that project alone, it would not be unusual for the railroad to coordinate with county and state authorities to complete the work. KC Southern argued it could not alter the road alone but failed to dispute that it could do so in coordination with the county. More importantly, it offered no evidence regarding the probable costs and burdens of such a project. We conclude that KC Southern did not meet its burden to establish that Horton’s negligence claim complaining of the humped crossing would pose the “unreasonable burden or interference with rail transportation” required to trigger implied preemption under Section 10501(b). Even assuming we should consider more than the burden of the compensatory damages alone, KC Southern failed to provide definitive evidence of the cost of any of the possible solutions needed to eliminate the hazard at this specific crossing. And even if we assume, as KC Southern argues, that Horton’s claim could require it to lower all the humps on all of its crossings, it provided no evidence of those costs and burdens either. The most specific evidence KC Southern presented to establish an “unreasonable” burden involved testimony that the curvature of this section of track, which was in a hilly area, would require undercutting a longer section of track than normal and would be further complicated by the presence of a flood-control culvert in the area. KC Southern’s witness testified that this would not be “a very feasible function for us to perform.” But even accepting KC Southern’s assertions that lowering the track would exceed the scope of a typical crossing rehabilitation, KC Southern failed to contest Horton’s evidence that lowering the track would close the track for less than a day and would only require weight and speed restrictions for forty-eight hours. As to the less burdensome option of raising the road, KC Southern provided no evidence of possible delays associated with such a solution, nor did it successfully refute Horton’s evidence that partnering with the local road authority for such a project would be routine. We thus hold that KC Southern failed to meet its burden to establish that Section 10501(b) impliedly preempts Horton’s humped-crossing negligence claim. And having also concluded that the section does not expressly preempt the claim, we reverse the court of appeals’ judgment to the extent it was based on the trial court’s contrary conclusion. III. Yield Sign In addition to the complaint about the humped crossing, Horton also argued and put on evidence that KC Southern negligently caused the accident by failing to ensure that yield signs remained in place on the posts containing the crossbuck signs at the crossing. KC Southern argues that no evidence could support a finding that the lack of a yield sign proximately caused the accident. We agree. In reviewing the legal sufficiency of evidence to support a jury verdict, we honor the rule that the jurors “are the sole judges of the credibility of the witnesses and the weight to give their testimony,” and it is their role to resolve any conflicts in the evidence. City of Keller v. Wilson, 168 S.W.3d 802, 819–21 (Tex. 2005) (citations omitted). We “credit favorable evidence if reasonable jurors could, and disregard contrary evidence unless reasonable jurors could not.” Id. at 827. We must consider the evidence “in the light most favorable to the verdict, and indulge every reasonable inference that would support it,” and we must credit any evidence that “allows of only one inference.” Id. at 822 (citations omitted). We cannot substitute our judgment for the jury’s. Id. Applying these standards, we conclude that the following evidence, presented through expert testimony and supporting studies and documents, supports Horton’s yield-sign theory: – Like a yield sign, “[t]he purpose of a crossbuck sign is to tell a motorist they need to yield.” In a sense, it sends the same message as a yield sign. But it is “amore specific yield sign,” specifically telling drivers “there’s a railroad crossing” and they need to “yield for [a] train.” – “[S]tudies have shown that [a] crossbuck [sign] alone does not give the kind of warning . . . that reminds people that there is a potential train coming.” – This is because the same type of crossbuck sign is used at both active rail crossings (those with lights, bells, or protective gates that automatically warn a driver that a train is approaching) and passive crossings (those lacking such automatic signals). Asa result, drivers “tend to regard the crossbuck sign as marking there’s a railroad track here” and then rely on “active signals as being the devices that control whether they should stop or go through the crossing.” So, at passive crossings, drivers “sometimes think the crossbuck sign merely marks the location of the grade crossing when, in fact, it . . . also needs to [warn drivers to] yield to trains.” – Because “research . . . showed road users do not fully comprehend the message being communicated by the crossbuck” sign, and because “there is an advantage for awareness improvement with the use of a yield sign,” in 2009, federal law began requiring yield signs be posted along with crossbuck signs at passive crossings. – Studies “indicate that [adding a] yield sign . . . conveys the message that the driver has the responsibility to look for and yield to an oncoming train better than . . . the crossbuck alone.” – Adding the yield sign “adds awareness and it makes people recognize yield more than they recognize a crossbuck [sign]. So the combination together works well.” – This is particularly true for drivers who are not as familiar with the crossing. “[D]rivers who are crossing the tracks for the first time or very infrequently would be more prone to respond by slowing somewhat and more conscientious about searching for oncoming traffic.” But drivers who cross the tracks “on a frequent basis” are likely to “eventually . . . revert back to [their] behavior before the yield sign was installed.” Warnings “tend to be less effective” for those who “use the same crossing over and over.” We agree with Horton that this evidence would support a finding that adding a yield sign to an already existing crossbuck sign would help alert drivers and cause them to look for an oncoming train before actually crossing an otherwise unprotected track. And so conversely, we agree that the evidence would support a finding that the absence of the yield sign made it less likely that Rigsby would have looked and stopped for the train that struck her pickup. But it is not sufficient to support the finding Horton had to obtain to prevail on the yield-sign claim: that, more likely than not, the absence of the yield sign proximately caused Rigsby to proceed into the train’s path. Other evidence, which a reasonable juror could not have ignored, established that no studies or empirical data confirm that adding a yield sign helps reduce crashes. The literature on which the experts relied provides “little empirical basis regarding the change in crash rates at crossings with either a yield or a stop sign,” and “no study has been conducted on crash effect when yield signs are used.” As one report explained, “[i]t is expected that this knowledge [a yield sign provides] should increase advanced searching [for oncoming trains], but how this apparent effectiveness carries over to actual locations, especially if most passive crossings were to have a yield sign, is a matter of conjecture.” A jury finding of proximate causation cannot be based on such “conjecture.” Horton’s expert conceded that he could not point to any study or evidence that the absence of the yield sign, “more likely than not, would have been a difference” for this accident. And he could not say that “this accident more likely than not would not have happened if the yield sign was present.” Nor does any other evidence support such a finding. Horton’s expert testified that, in his opinion, the addition of a yield sign more likely than not provides a more effective warning, but he based that opinion merely on the fact that the federal government recommended the addition of yield signs in 2000 and required them in 2009. This is not the type of basis that could support such an opinion, and the opinion itself was insufficient to support a finding that the absence of the yield sign more likely than not caused this particular accident. As to this accident, the video recording confirms that Rigsby, who lived near the crossing and regularly crossed it for years up until she was hospitalized several weeks before the accident in fact did slow down as she approached the crossing, and then slowed even more, to three or three-and-a-half miles per hour, before reaching the rails. As Horton’s expert confirmed, Rigsby “was cautious. She showed approach. It looked like she was intent. The purpose of those kind of signs and a warning sign is to alert that.” We must conclude that no evidence supports a finding that, more likely than not, Rigsby would have approached the crossing any more cautiously or intently had the yield sign been present, or that the absence of the yield sign more likely than not caused Rigsby to drive into the train’s path. IV. Harmful Error Having concluded that the ICCT Act does not preempt Horton’s negligence claim based on the humped crossing, but no evidence supports the negligence claim based on the missing yield sign, we must finally determine whether the trial court’s submission of both negligence theories through a single broad-form negligence question constituted harmful error. KC Southern objected to the broad-form question on the ground that it would permit the jury to find negligence on an invalid liability theory, and it offered a proposed charge that included two blanks for the jury to separately determine the parties’ negligence liability for the humped crossing and the missing yield sign. The trial court overruled the objection and refused the proposed question, believing it improperly granulated a single negligence cause of action. The court of appeals—after agreeing with KC Southern that federal law preempts the humped-crossing claim and agreeing with Horton that the evidence supports the yield-sign claim—concluded that the trial court erred by submitting both theories in one question and that the error was harmful under Crown Life Insurance Co. v. Casteel, 22 S.W.3d 378 (Tex. 2000), because the court could not determine whether the jury found negligence on a valid or invalid theory. See 666 S.W.3d at 19. We have also concluded that only one of Horton’s two allegations can support the jury’s negligence finding, albeit the opposite one. Thus, we too must determine whether submission of the broad-form question over KC Southern’s objection resulted in harmful error. KC Southern argues it did because the question “commingle[d] valid and invalid liability theories . . . and a proper reason for the verdict cannot be ascertained from the record.” Horton argues it did not because the question properly submitted Horton’s single claim for negligence, leaving it to the jury to determine what acts would support that claim. We agree with KC Southern. We held in Casteel that “when a trial court submits a single broad-form liability question incorporating multiple theories of liability, the error is harmful, and a new trial is required when the appellate court cannot determine whether the jury based its verdict on an improperly submitted invalid theory.” 22 S.W.3d at 388. Horton argues Casteel does not apply here, however, because that case involved multiple, distinct “theories of liability,” some of which were valid and others of which were not. But here, Horton contends, the court submitted just one liability theory—negligence. According to Horton, Casteel does not apply because “[f]ailing to maintain tracks and failing to post a yield sign are different negligent acts, not separate theories of liability.” We disagree. It is true that Casteel involved a single broad-form liability question with instructions addressing “thirteen independent grounds for liability,” four of which we concluded were invalid. Id. at 387. But in holding that the error was harmful because appellate courts could not determine whether the jury based its verdict on an invalid theory, we relied on and reaffirmed our prior decision in Lancaster v. Fitch, 246 S.W. 1015 (Tex. 1923), in which “the trial court submitted a single general negligence issue with instructions regarding three distinct theories of negligence liability.” Casteel, 22 S.W.3d at 389. Lancaster applied the same rule in a case in which the plaintiff “pleaded three separate acts of negligence as the proximate cause of his injury” and the trial court submitted a single negligence question. 246 S.W. at 1015–16. We held that the submission of one invalid negligence theory along with two valid theories, where it was impossible to tell which theory the jury relied on, was harmful error. Id. at 1015–17. We have since applied Casteel‘s harmful-error rule in cases involving a variety of circumstances that created the same problem for the appellate courts, including the broad-form submission of multiple elements of damages, Harris County v. Smith, 96 S.W.3d 230, 231 (Tex. 2002), the inclusion of two theories within a single apportionment-of-responsibility question, Romero v. KPH Consol., Inc., 166 S.W.3d 212, 215 (Tex. 2005), and a trial court’s refusal to submit necessary instructions with a broad-form question, Columbia Rio Grande Healthcare, L.P. v. Hawley, 284 S.W.3d 851, 865 (Tex. 2009). Most recently, and most importantly, we specifically rejected Horton’s argument in Benge v. Williams, 548 S.W.3d 466 (Tex. 2018). At the trial in Benge, the plaintiff “argued and offered evidence that her physician was negligent both in using an inexperienced resident to assist with performing her surgery and in not disclosing the resident’s level of involvement.” Id. at 467–68. But the plaintiff only claimed a right to recover based on the physician’s negligent use of the inexperienced resident and did “not claim a right to recover for the nondisclosure.” Id. at 468. The trial court submitted a broad-form negligence question, after refusing the physician’s request for an instruction that the jury should disregard the plaintiff’s arguments and evidence regarding the nondisclosure, and the jury found in the plaintiff’s favor. Id. at 470. We held that the court erred in refusing to submit the requested instruction, and we found that error harmful under Casteel because the jury could have found negligence based on the nondisclosure even though the plaintiff “does not assert that claim.”Id.at474.We did so even though the question the trial court submitted, “unlike the one in Casteel, did not include multiple theories, some valid and some invalid. It inquired about a single theory: negligence.” Id. at 475. To the extent Horton contends that the Casteel rule applies only when a broad-form question permits a liability finding based on a theory or ground that is legally “invalid” as opposed to, as here, a ground lacking sufficient evidence, our precedent has also rejected that argument. We have applied the rule not only when the question permits a finding based on a legally “invalid” theory, but when it permits an erroneous finding based on a ground the evidence does not support, Harris County, 96 S.W.3d at 231; Romero, 166 S.W.3d at 227–28, a ground that is “jurisdictionally barred,” Tex. Comm’n on Hum. Rts. v. Morrison, 381 S.W.3d 533, 535 (Tex. 2012), and, in Benge, a claim the plaintiff simply “does not assert,” 548 S.W.3d at 474. As we explained in Hawley, “[s]ubmission of an invalid theory” simply “involves ‘[a] trial court’s error in instructing a jury to consider erroneous matters.’” 284 S.W.3d at 865 (quoting Harris County, 96 S.W.3d at 233). Horton relies on our opinion in Dillard v. Texas Electric Cooperative, where we stated that, under broad-form submission rules, “jurors need not agree on every detail of what occurred so long as they agree on the legally relevant result. Thus, jurors may agree that a defendant failed to follow approved safety practices without deciding each reason that the defendant may have failed to do so.” 157 S.W.3d 429, 434 (Tex. 2005) (citing Burk Royalty Co. v. Walls, 616 S.W.2d 911, 924 (Tex. 1981)). Indeed, we went on to say in Dillard that jurors “could have unanimously found [the defendant] negligent, even if half believed the negligent act was overloading his truck and half believed it was failing to warn oncoming traffic—acts that preceded two different collisions.” Id. We read Horton’s reliance on Dillard to argue that the trial court’s broad-form submission of his two negligence allegations was not error at all, much less harmful under Casteel. But Dillard involved a completely different issue: “whether the trial court abused its discretion in refusing to submit one of two different instructions on the defendants’ inferential rebuttal defenses.” Id. at 430. We held that the court sufficiently instructed the jury on those defenses and thus committed no error, and we made the statements on which Horton relies to explain why additional instructions would have been duplicative and unnecessary. Id. Our discussion addressed only the defendants’ defenses, not the plaintiff’s claims, and the statements on which Horton relies presumed that each of the acts supporting a negligence finding were themselves valid and supported by the evidence. Here, by contrast, we have concluded that the evidence does not support one of the acts on which Horton relied for a negligence finding. Jurors finding negligence may not all have to agree on the same valid and supported grounds to find negligence, but they cannot rely on invalid or unsupported grounds. Finally, Horton argues that application of the Casteel rule in this case would undermine our strong preference for broad-form submissions, as set forth in Texas Rule of Civil Procedure 277. We think this argument goes too far. Although Rule 277 is “intended to simplify jury charges for the benefit of the jury, the parties, and the trial court,” it “was certainly never intended to permit, and therefore encourage, more error in a jury charge.” Romero, 166 S.W.3d at 230. As we explained when addressing this argument in Romero, Rule 277 requires that issues be submitted to a jury in broad form “whenever feasible.” Id. (quoting TEX. R. CIV. P. 277). We adhere to that rule today, but “Rule 277 is not absolute,” and “[s]ubmitting alternative liability standards when the governing law is unsettled might very well be a situation where broad-form submission is not feasible.” Casteel, 22 S.W.3d at 390 (quoting Westgate, Ltd. v. State, 843 S.W.2d 448, 455 n.6 (Tex. 1992)). Our holding does not overhaul the general preference for broad-form submission. Rather, it emphasizes that, despite our rules’ preference for broad-form jury questions, “broad-form submission cannot be used to broaden the harmless error rule to deny a party the correct charge to which it would otherwise be entitled.” Romero, 166 S.W.3d at 230. Where, as here, true doubt exists as to the validity of one underlying theory and the trial court must resolve a close call[20] as to whether sufficient evidence supports a separate act of negligence, submitting either separate questions or separate blanks within the same question may be helpful. Separate jury questions are not the only means to avoid a Casteel problem. In some cases, rephrasing the question or giving an instruction not to consider theories that are unpled, invalid, or lacking in evidentiary support will be sufficient.[21] And that alternative is preferable to separate questions when it is feasible. Again, we emphasize that this holding does not undermine the general preference for broad-form submission, but rather provides additional guidance as to how courts should approach instances where broad-form submission is not feasible. Because the trial court submitted Horton’s negligence claim as a broad-form question subsuming both his humped-crossing theory and his missing-yield-sign theory, we are unable to discern whether the jury found KC Southern negligent based on the yield-sign theory, which the evidence does not support. Because the question “allows a finding of liability based on evidence that cannot support recovery,” Casteel‘s “presumption-of-harm rule must be applied.” Benge, 548 S.W.3d at 475. V. Conclusion We hold that the ICCT Act does not expressly preempt Horton’s humped-crossing negligence claim, that KC Southern failed to bear its burden of proving that the Act impliedly preempts that claim, that the evidence does not support liability based on Horton’s missing-yield-sign negligence claim, and that the trial court’s submission of both negligence theories using a broad-form question constituted harmful error in this case. We thus affirm the court of appeals’ judgment reversing the trial court’s judgment and remanding the case to that court for a new trial. Jeffrey S. Boyd Justice OPINION DELIVERED: June 30, 2023