¶26 This is a vehicular homicide case-a criminal case, not a tort case. The majority holds that defendant's actions of hitting a car on the freeway and fleeing the accident scene without stopping, thereby leaving the struck disabled car on the freeway, proximately caused the death of a passerby who subsequently stopped to render aid and later died from injuries sustained in a second automobile collision at the scene. I disagree with the majority's conclusion that "as a matter of law" defendant's acts were the "legal cause" of the passerby's death. Majority at 602. For the reasons explained below, in my view, the passerby's death is simply too attenuated from defendant's hit and run to result in legal causation in the present context.
¶27 In this vehicular homicide case, Joshua Frahm, while intoxicated and driving his truck, collided with a car *604on the freeway, injuring the car's driver, Steven Klase. Frahm continued on without stopping, leaving Klase's disabled car on the freeway. Richard Irvine, seeing the disabled car on the freeway, stopped to help the disabled car's driver. Irvine pulled his vehicle onto the freeway's right shoulder, walked across the freeway, and called 911. While Irvine was talking to the dispatcher, another vehicle (a minivan) struck Klase's disabled car, propelling *603it into Irvine, who later died from his injuries. Frahm was charged with several crimes, including vehicular homicide for the death of Irvine. A jury convicted Frahm of this charge.
¶28 On appeal, Division Two of the Court of Appeals rejected Frahm's insufficient evidence challenge to his vehicular homicide conviction, determining that events occurring after Frahm fled the scene (i.e., the acts of Irvine and the van driver) were foreseeable. Thus, such intervening acts did not amount to superseding causes that would break the chain of proximate causation resulting in Frahm's liability for Irvine's death. The majority agrees; I do not. In my view, it is untenable that Frahm's rear-ending a vehicle on the freeway proximately caused the death of Irvine under the facts as above described.
¶29 Both the majority and the Court of Appeals view this case as analogous to State v. Roggenkamp, 115 Wash. App. 927, 64 P.3d 92 (2003), affd. 153 Wash.2d 614, 106 P.3d 196 (2005), a vehicular homicide and assault case. In my view, while Roggenkamp is instructive, it does not support the majority's view. In Roggenkamp, the defendant drove on a two-lane county road with a posted speed limit of 35 miles per hour. Defendant entered the oncoming traffic lane to pass another vehicle and reached a speed of approximately 70 miles per hour. While defendant was still in the oncoming lane, a vehicle turned into that lane from an intersection. Defendant collided with that vehicle, seriously injuring three passengers and killing another. The driver of the vehicle that defendant hit had a blood alcohol content over the legal limit. Id. at 931-34, 64 P.3d 92.
*605¶30 Roggenkamp held that the drunk driver's actions were not a superseding cause of the accident for two reasons. First, the drunk driver's action of turning left onto a two-lane residential roadway was reasonably foreseeable; and, second, Roggenkamp's recklessness was ongoing when the drunk driver pulled out of the intersection. Id. at 946-47, 64 P.3d 92. In the present case, Division Two concluded, "Here, as in Roggenkamp , the acts of the [van] driver and a passerby were foreseeable." State v. Frahm, 3 Wash. App. 2d 812, 821, 418 P.3d 215, review granted, 191 Wash.2d 1026, 428 P.3d 1170 (2018). And "a reasonable jury could find beyond a reasonable doubt that Frahm's rear-ending Klase's vehicle proximately caused Irvine's death." Id. at 822, 418 P.3d 215. The majority agrees that "[t]he question of foreseeability was properly given to the jury" and holds that "Frahm's acts proximately caused Irvine's death." Majority at 601-02, 602. I disagree with the majority's conclusion.
¶31 Here, the jury received Washington Pattern Jury Instruction 90.08 concerning vehicular homicide and causation as follows:
If you are satisfied beyond a reasonable doubt that the driving of the defendant was a proximate cause of the death, it is not a defense that the conduct of the deceased or another may also have been a proximate cause of the death.
However, if a proximate cause of the death was a new independent intervening act of the deceased or another which the defendant, in the exercise of ordinary care, should not reasonably have anticipated as likely to happen, the defendant's act is superseded by the intervening cause and is not a proximate cause of the death. An intervening cause is an action that actively operates to produce harm to another after the defendant's act or omission has been committed.
However, if in the exercise of ordinary care, the defendant should reasonably have anticipated the intervening cause, that cause does not supersede the defendant's original act and the defendant's act is a proximate cause. It is not necessary that the sequence of events or the particular injury be foreseeable. It is only necessary that the death fall within the general field of danger which the defendant should have reasonably anticipated.
*606Clerk's Papers at 106 (Instr. 12) (emphasis added); 11A WASHINGTON PRACTICE: WASHINGTON PATTERN JURY INSTRUCTIONS: CRIMINAL 90.08, at 278 (4th ed. 2016). Here, the general field of danger that Frahm should have reasonably anticipated is further injury to the disabled vehicle and its occupants that he left stranded on the freeway, not injury to a passerby who walked across the freeway and was struck by the disabled car, after that *604disabled car was struck by yet another vehicle on the freeway.
¶32 By contrast, in Roggenkamp, defendant's recklessly speeding vehicle struck another car causing injury and death to that car's occupants. The defendant's reckless conduct directly caused the injuries sustained and such reckless conduct was ongoing at the point in time when the injuries were sustained. That is not the case here. Frahm's conduct of striking a car with his truck and leaving the disabled car in the freeway was completed and over when the passerby (Irvine) later received injuries in the second automobile collision. Also, as discussed below, there is simply no sufficient connection between Frahm and the passerby's death to impose criminal liability. While "but for" causation is satisfied, legal causation is not.
¶33 Unlike the facts in Roggenkamp , the chain of events in this case is simply too attenuated to impose criminal liability on Frahm for the death of the passerby. This result finds support in State v. Bauer, 180 Wash.2d 929, 942, 329 P.3d 67 (2014). There, this court explained that while criminal law and tort law share the same "but for" causation, the requirement for legal causation to impose criminal liability is more demanding than in the tort context. Surveying the relevant case law and commentary on this issue, the Bauer court opined as follows:
This court, in agreement with commentators and other jurisdictions, has observed that "as to cause in fact [ (i.e., 'actual' or 'but for' causation) ], tort and criminal situations are exactly alike." [ State v. Dennison, 115 Wn.2d 609, 624 n.15, 801 P.2d 193 (1990) (citing 1 WAYNE R. LAFAVE & AUSTIN W. SCOTT, JR., SUBSTANTIVE CRIMINAL LAW § 3.12, at 397 n.31 (1986)) ].
*607Legal causation, however, is different. Commentators and courts recognize that criminal law and tort law serve different purposes and therefore have different principles of legal causation. "[W]ith crimes, where the consequences of a determination of guilt are more drastic ... it is arguable that a closer relationship between the result achieved and that intended or hazarded should be required." 1 WAYNE R. LAFAVE, SUBSTANTIVE CRIMINAL LAW § 6.4(c) at 472 (2d ed. 2003). "The wider doctrines of causation currently applied in tort law should not be extended to criminal law. ... [I]n criminal law, ... it is not normally enough merely to prove that [the] accused occasioned the harm; he must have 'caused' it in the strict sense." H.L.A. HART & TONY HONORE, CAUSATION IN THE LAW 350-51 (2d ed. 1985); see also JEROME HALL, GENERAL PRINCIPLES OF CRIMINAL LAW 254-55 (2d ed. 1960) (noting that causation in tort may be much broader than causation in criminal law).
Most states that have addressed the question agree that legal causation is defined more narrowly in criminal law than it is in tort law. The Pennsylvania Supreme Court explained, in the context of felony murder:
A closer causal connection between the felony and the killing than the proximate-cause theory normally applicable to tort cases should be required because of the extreme penalty attaching to a conviction for felony murder and the difference between the underlying rationales of criminal and tort law. The former is intended to impose punishment in appropriate cases while the latter is primarily concerned with who shall bear the burden of a loss.
Commonwealth ex rel. Smith v. Myers, 438 Pa. 218, 232, 261 A.2d 550 (1970). Cases from multiple jurisdictions, both state and federal, support this principle. E.g., United States v. Schmidt, 626 F.2d 616, 618 n.3 (8th Cir. 1980) ("[W]e believe that proof of some more direct causal connection between act and result should be required in criminal cases than would be sufficient to uphold liability in tort."); Campbell v. State, 293 Md. 438, 450-51, 444 A.2d 1034 (1982) ("[T]he tort liability concept of proximate cause is generally too broad and comprehensive to be appropriate in a criminal proceeding ....");
*608People v. Kibbe, 35 N.Y.2d 407, 413, 321 N.E.2d 773, 362 N.Y.S.2d 848 (1974) ("We subscribe to the requirement that the defendants' actions must be a sufficiently direct cause of the ensuing death before there can be any imposition of criminal liability, *605and recognize, of course, that this standard is greater than that required to serve as a basis for tort liability."); People v. Scott, 29 Mich. App. 549, 558, 185 N.W.2d 576 (1971) ("In criminal prosecutions there must be a more direct causal connection between the criminal conduct of the defendant and the homicide charged than is required by the tort liability concept of proximate cause.").
180 Wash.2d at 936-38, 329 P.3d 67 (most alterations in original) (footnote omitted). In Bauer, this court ultimately concluded that criminal liability did not attach where "the accused did not actively participate in the immediate physical impetus of harm." Id. at 940, 329 P.3d 67 (quoting with approval, Hart & Honore, supra, at 326 for the proposition that " '[W]here [the first actor's conduct] was not sufficient [to bring about the harm] without the intervention of the second actor ... most decisions relieve the first actor of responsibility' " (most alterations in original)). That is precisely the circumstance in Frahm's case. Frahm's striking of Klase's car was not sufficient to bring about Irvine's death. Irvine's death resulted from the minivan striking Klase's car, propelling it into Irvine, who sustained injuries from that impact that were ultimately fatal more than a week later. Without Irvine leaving his vehicle and crossing the freeway on foot and the minivan subsequently striking Klase's disabled car, Irvine would not have been injured.
¶34 The majority finds this case analogous to State v. Leech, 114 Wash.2d 700, 790 P.2d 160 (1990), in which this court held an arsonist criminally liable for the death of a firefighter who died while responding to the fire started by the arsonist. The majority concludes that similarly here, "Frahm's acts of hitting Klase's [car] while intoxicated and fleeing the scene were felonious and directly caused harm in and of themselves." Majority at 600. While I agree that Frahm's drunken hit-and-run "directly caused harm," such direct harm was to Klase, not to Irvine. As discussed, under *609Bauer, the legal causation required to impose criminal liability necessitates a direct causal connection that is simply not present here. In Frahm's case, while "but for" causation is satisfied, legal causation is not. This alone is a sufficient basis for reversing the Court of Appeals and overturning Frahm's vehicular homicide conviction.
¶35 Additionally, I disagree with the majority's view that "[w]hether an intervening act rises to the level of a superseding cause is an issue of actual cause [for the trier of fact]." Majority at 601 (emphasis added). The majority cites two cases as support: Maltman v. Sauer, 84 Wash.2d 975, 530 P.2d 254 (1975), and McCoy v. American Suzuki Motor Corp., 136 Wash.2d 350, 961 P.2d 952 (1998). Both cases, however, addressed application of the rescue doctrine and foreseeability in the tort context and, thus, have no application here.1
¶36 Elsewhere, Washington courts have opined that the issue of foreseeability "is not properly framed within the 'cause in fact' element of proximate cause." Bullard v. Bailey, 91 Wash. App. 750, 758, 959 P.2d 1122 (1998). "Instead, it is part of determining whether [a defendant's] actions are the 'legal cause' of [a plaintiff's] damages." Id.
[The] question of foreseeability "is in no way one of causation, and never arises until causation has been established. It is rather one of the fundamental policy of the law, as to whether the defendant's responsibility should extend to such results."
Id. (quoting WILLIAM L. PROSSER, HANDBOOK OF THE LAW OF TORTS 250 (4th ed. 1971)).
*610¶37 Further, concerning the McCoy case, the Washington Practice series notes:
In cases brought by a plaintiff who was injured in the course of trying to rescue an individual who was placed in peril by the defendant's negligence, Washington courts have added an additional inquiry as part of the "cause in fact" analysis. ...
It should be noted that intervening forces are usually considered as part of *606the question of legal cause, rather than cause in fact. However, this additional element of cause in fact may be a unique feature of rescue cases, and hence can be appropriately considered in such cases as part of analyzing both prongs of the proximate cause test.
16 DAVID K. DEWOLF & KELLER W. ALLEN, WASHINGTON PRACTICE: TORT LAW AND PRACTICE § 5:3 (4th ed. 2013) (emphasis added) (citation and footnote omitted).
¶38 In any event, whatever Maltman and McCoy say about proximate cause and foreseeability in the context of applying the rescue doctrine, those cases are limited to the tort arena that they addressed and, thus, have no application in the present criminal context. In my view, as discussed above, Bauer provides the appropriate analysis in this criminal context; and the stronger direct causal connection required to find legal causation, as explained in Bauer, is simply not present.
¶39 For these reasons, I dissent.
Gordon McCloud, J.
The rescue doctrine is a tort concept that "allows an injured rescuer to sue the party which caused the danger requiring the rescue in the first place." McCoy, 136 Wash.2d at 355, 961 P.2d 952 (citing Maltman, 84 Wash.2d at 976-77, 530 P.2d 254 ). The rescue doctrine reflects "a societal value judgment that rescuers should not be barred from bringing suit for knowingly placing themselves in danger to undertake a rescue." Id. at 356, 961 P.2d 952. But, as noted, the present case concerns imposition of criminal sanction, not compensation.