HOOD JUSTICE
Certiorari to the Colorado Court of Appeals Court of Appeals Case No. 19CA1308 Judgment
Attorneys for Petitioner: Megan A. Ring, Public Defender Meredith K. Rose, Deputy Public Defender Denver, Colorado
Attorneys for Respondent: Philip J. Weiser, Attorney General John T. Lee, First Assistant Attorney General Jessica E. Ross, Assistant Attorney General Denver, Colorado
OPINION
HOOD JUSTICE
¶1 When the defendant, Arnold Roman Martinez, stole a bicycle from C.T.'s garage, C.T. jumped in his car and gave chase. As Martinez tried to complete his getaway on the bike, C.T. first drove parallel to him and then cut him off. Martinez hit the car. Fortunately, nobody got hurt. But when criminal charges against Martinez yielded a plea agreement, the district court ordered Martinez to pay over $2,000 in restitution for the damage to C.T.'s car.
¶2 Martinez challenged this order, asserting that the district court erred in finding that Martinez proximately caused the damage. On appeal, the parties disagreed about the appropriate standard of review, as did a division of the court of appeals. The majority reviewed the district court's proximate-cause determination for an abuse of discretion, but Judge J. Jones specially concurred, arguing that the appropriate standard of review is clear error.
¶3 We hold that clear error is the appropriate standard of review for evaluating a district court's determination of proximate cause for restitution. Accordingly, we conclude that the division erred by applying abuse-of-discretion review here. But applying the clear-error standard, we nonetheless affirm the division's judgment that Martinez was obligated to pay restitution. Thus, we affirm on other grounds.
I. Facts and Procedural History
¶4 C.T. and his wife returned to their Boulder home one evening and left their garage door open. C.T.'s wife heard a noise in the garage and went to investigate. She screamed that a man, later identified as Martinez, was in the garage stealing C.T.'s bike.
¶5 As Martinez rode off on the bike, C.T. got in his car and went after him. (It was an expensive bike.) After several blocks, C.T. caught up with Martinez, drove parallel to him, and then turned in front of him, cutting him off. Martinez crashed into the front passenger-side fender of C.T.'s car. Martinez then got into another car and drove away. The bike was undamaged, but C.T.'s car was less fortunate.
¶6 The prosecution charged Martinez with second degree burglary and criminal mischief but dropped these charges as part of an agreement in which Martinez pled guilty in another case and agreed to pay restitution for the cost of repairing C.T.'s car in this case. The prosecutor sought $2,393.84. Martinez objected, claiming that he wasn't the proximate cause of the damage and therefore didn't owe restitution.
¶7 The district court granted the prosecution's restitution request. It found that C.T. pulling his car in front of Martinez wasn't an independent intervening cause that broke the chain of causation, reasoning that C.T.'s act was foreseeable and that Martinez participated in the collision. Consequently, the district court concluded that Martinez's theft was the proximate cause of the damage to C.T.'s car.
¶8 A division of the court of appeals affirmed the restitution order. People v. Martinez, 2022 COA 28, ¶ 52, 511 P.3d 739, 748. The parties disagreed on the standard of review. Martinez claimed that the division should review the district court's proximate cause determination as a sufficiency-of-the-evidence question subject to de novo review, while the prosecution claimed that the division should review for an abuse of discretion. The division majority agreed with the prosecution. Id. at ¶ 14, 511 P.3d at 742. In his special concurrence, however, Judge J. Jones stated that he would have reviewed for clear error. Id. at ¶ 60, 511 P.3d at 749 (J. Jones, J., specially concurring). Regardless of the standard applied, the division unanimously concluded that the district court didn't err in finding that Martinez proximately caused the car's damage. Id. at ¶¶ 31, 61, 511 P.3d at 745, 749-50.
¶9 We granted Martinez's petition to review the division's opinion.
We granted certiorari to review the following issues:
1. [REFRAMED] Whether an appellate court reviews a trial court's proximate cause determination for restitution purposes under an abuse of discretion standard.
2. Whether the prosecution proved this bike theft proximately caused the car damage for which the court awarded restitution.
II. Analysis
¶10 We first explain why the clear error standard should govern. In doing so, we note that abuse-of-discretion review of restitution orders is a relic from a bygone statutory era. We also reject Martinez's claim that his challenge is one based on insufficiency of the evidence, which merits mixed-question review.
¶11 Having set the bar, we then explain how the district court cleared it. Because there is record support for the court's factual finding that C.T. cutting off Martinez on the bike was not an independent intervening cause of the damage to C.T.'s car, there was no clear error here.
A. Proximate Cause for Criminal Restitution
¶12 We determine the applicable standard of review de novo. Howard-Walker v. People, 2019 CO 69, ¶ 22, 443 P.3d 1007, 1011. The standard we assign depends on the type of issue we confront. So, we must first consider how the legislature has instructed courts to calculate restitution.
¶13 A defendant convicted of a felony offense must pay restitution for any pecuniary loss he proximately caused his victim. See § 18-1.3-602(3)(a), C.R.S. (2023). Proximate cause is any "cause which in natural and probable sequence produced the claimed injury." People v. Stewart, 55 P.3d 107, 116 (Colo. 2002) (quoting CJI-Crim., 9:10, 9(3) (1983)). Thus, "[u]nlawful conduct that is broken by an independent intervening cause cannot be the proximate cause of injury to another." Id. at 121. But "[t]o qualify as an intervening cause, an event must be unforeseeable and one in which the accused does not participate." Id.
¶14 A victim's gross negligence can serve as an independent intervening cause. People v. Saavedra-Rodriguez, 971 P.2d 223, 226 (Colo. 1998). "Gross negligence is willful and wanton conduct, that is, action committed recklessly, with conscious disregard for the safety of others." Hamill v. Cheley Colo. Camps, Inc., 262 P.3d 945, 954 (Colo.App. 2011). With this framework in mind, we turn to how the division erred.
1. The Division Erred in Applying an Abuse-of-Discretion Standard
¶15 The division majority reviewed for an abuse of discretion. Martinez, ¶ 14, 511 P.3d at 742. It relied on case law that recognized courts' discretion in determining the terms and conditions of restitution orders. See, e.g., People v. Henson, 2013 COA 36, ¶¶ 9-20, 307 P.3d 1135, 1137-39; People v. Reyes, 166 P.3d 301, 302 (Colo.App. 2007); accord Pierce v. Underwood, 487 U.S. 552, 558 (1988) (identifying abuse of discretion as the appropriate standard for "matters of discretion"). But, as the special concurrence points out, these cases aren't applicable because they predate the current statutory regime for determining restitution.
¶16 The abuse-of-discretion standard that Henson, Reyes, and their predecessors applied can be traced to Cumhuriyet v. People, 615 P.2d 724, 725-26 (Colo. 1980), which interpreted former section 16-11-204, C.R.S. (1973 &1978 Repl. Vol. 8). Martinez, ¶ 55 n.1, 511 P.3d at 748 n.1 (J. Jones, J., specially concurring). That section said a court, "in its discretion," could impose "reasonably necessary" probation conditions, including restitution. § 16-11-204(1), (2)(e) (emphasis added). When this law controlled, it was appropriate for appellate courts to review decisions involving restitution for an abuse of discretion.
¶17 The General Assembly repealed section 16-11-204 in 2002, however, and replaced it with section 18-1.3-204(1)(a), C.R.S. (2023), which states that "[t]he court shall provide as [an] explicit condition[] of every sentence to probation . . . that the defendant make restitution." Ch. 318, sec. 2, § 18-1.3-204(1), 2002 Colo. Sess. Laws 1365, 1378 (emphasis added). Two years earlier, the General Assembly also passed then-section 16-18.5-103 (now section 18-1.3-603(1), C.R.S. (2023)), which provides that "[e]very order of conviction . . . shall include consideration of restitution." Ch. 232, sec. 1, § 16-18.5-103(1), 2000 Colo. Sess. Laws 1030, 1032 (emphasis added). These changes made restitution mandatory when proximate cause is established. See Walton v. People, 2019 CO 95, ¶ 13, 451 P.3d 1212, 1216 ("'Shall' is mandatory unless there is a clear indication otherwise."). Thus, it no longer makes sense to apply abuse-of-discretion review.
¶18 But if not abuse of discretion, then what? Martinez urges de novo review, claiming that the problem is one of sufficiency of the evidence. We turn to that contention now.
2. Martinez Doesn't Present a Sufficiency-of-the-Evidence Challenge
¶19 Martinez alleges that he brings a sufficiency-of-the-evidence challenge, in which appellate courts review the record de novo. Clark v. People, 232 P.3d 1287, 1291 (Colo. 2010) (examining a sufficiency challenge outside the context of restitution). Although this court hasn't weighed in on this issue in addressing restitution, multiple divisions of the court of appeals have reviewed true challenges to the sufficiency of restitution evidence de novo. See, e.g., People v. Moss, 2022 COA 92, ¶¶ 9-11, 520 P.3d 694, 696; People v. Dyson, 2021 COA 57, ¶ 15, 492 P.3d 1070, 1074; People v. Barbre, 2018 COA 123, ¶ 25, 429 P.3d 95, 99. This brings us to the crux of the issue: Does Martinez present a true sufficiency challenge?
¶20 To answer this question, we consider the challenge's substance, not its form. See Lucero v. People, 2017 CO 49, ¶ 26, 394 P.3d 1128, 1134. Sufficiency challenges address whether the quantum of evidence provided to the court is "substantial and sufficient" to support a reasonable juror's conclusion of guilt beyond a reasonable doubt, not a court's particular interpretation of the evidence before it. Clark, 232 P.3d at 1291.
¶21 Here, though Martinez labels his argument a sufficiency challenge, his point of attack suggests otherwise. His gripe is really with the district court's findings of fact related to proximate cause. As to foreseeability, for example, Martinez argues that "the district court erroneously concluded that it was . . . foreseeable that [C.T.] would attempt to prevent [Martinez] from taking property by pulling in front of [Martinez]," not that the prosecution didn't meet its burden of proof. (Emphasis added.) And as to participation, Martinez asserts that his "riding in the would-be bike lane" didn't damage the car. This is a challenge to the district court's factual conclusion that Martinez participated in the act causing the pecuniary harm. It is not a challenge to the sufficiency of the evidence the district court used to draw this conclusion.
¶22 To better understand this distinction, it's instructive to compare the case before us to another case involving a true sufficiency challenge to a restitution determination. In Barbre, for example, a woman was charged with theft and possession of a controlled substance for stealing prescription pain medications from the pharmacy where she worked. ¶ 2, 429 P.3d at 96. The district court ordered restitution for the more than $10,000 in losses Barbre had caused over more than a year of stealing pills. Id. at ¶ 8, 429 P.3d at 97. But Barbre appealed, contending that the prosecution hadn't met its burden of showing she had proximately caused the pharmacy's losses for the full year-as opposed to the seventeen-day investigation period that had led to Barbre's initial charge. Id. at ¶¶ 3, 7, 429 P.3d at 96, 97. She claimed the prosecution hadn't offered "first-hand knowledge" of the pharmacy's losses across the year-long period, id. at ¶ 29, 429 P.3d at 99, and the prosecution couldn't prove that the pills went missing on the precise days she worked at the pharmacy, id. at ¶ 35, 429 P.3d at 100. Using a de novo standard of review, the division affirmed the district court's restitution order. Id. at ¶ 25, 429 P.3d at 99. In doing so, however, it cautioned that the issue before it concerned the quantum of evidence (i.e., was there proof by a preponderance of the evidence that Barbre caused the loss for the entire year), not the interpretation of the evidence. Id. at ¶¶ 24, 26, 429 P.3d at 99.
Other divisions of the court of appeals have performed similar analyses. See Moss, ¶ 11, 520 P.3d at 696; Dyson, ¶ 15, 492 P.3d at 1074; People v. Rice, 2020 COA 143, ¶ 22, 478 P.3d 1276, 1281-82, overruled on other grounds by People v. Weeks, 2021 CO 75,498 P.3d 142; People v. Stone, 2020 COA 24, ¶ 7, 471 P.3d 1159,1162-63; People v. Jaeb, 2018 COA 179, ¶ 39, 434 P.3d 785, 792. In doing so, these divisions all applied the test outlined in People v. Bennett, 515 P.2d 466, 469 (Colo. 1973), and all that accompanies it. See, e.g., Barbre, ¶ 25, 429 P.3d at 99 (applying this test). In Bennett, the court assessed "whether the relevant evidence, both direct and circumstantial, when viewed as a whole and in the light most favorable to the prosecution, is substantial and sufficient to support a conclusion by a reasonable mind that the defendant is guilty of the charge beyond a reasonable doubt." Bennett, 515 P.2d at 469 . The prosecution doesn't ask us to overrule these cases, and we see no need to do so.
¶23 Quantitative arguments like these simply aren't the crux of Martinez's petition. We therefore decline to review his petition as a sufficiency claim.
3. The Proper Standard of Review Is Clear Error
¶24 Typically, appellate courts review a trial court's factual findings for clear error and legal conclusions de novo. C.R.C.P. 52. But the line between fact and law isn't always clear; some issues blend the two. If an issue involves both factual findings and legal conclusions, this court may treat the issue as one of fact, one of law, or as a mixed question of fact and law, in which case "the court may review the findings of fact for clear error and still look de novo at the legal conclusions that the trial court drew from those factual findings." E-470 Pub. Highway Auth. v. 455 Co., 3 P.3d 18, 22 (Colo. 2000).
¶25 Martinez argues that we should review his claim as a mixed question, deferring to the trial court as to questions of historical fact but then assessing the legal significance of the facts de novo. The prosecution argues that proximate cause is strictly an issue of fact, so appellate courts should review for clear error.
¶26 Neither this court nor the U.S. Supreme Court has established a definitive test for how to review issues-like proximate cause in criminal restitution-that may "fall[] somewhere between a pristine legal standard and a simple historical fact." Miller v. Fenton, 474 U.S. 104, 114 (1985); see also Pierce, 487 U.S. at 559. In choosing among the different standards of review, however, precedent from both courts suggests that we may consider various factors, including (1) the "nature of the inquiry itself," Miller, 474 U.S. at 115; (2) "the long history of appellate practice" regarding the challenge, Pierce, 487 U.S. at 558; (3) "the language and structure of the governing statute," id. at 559; and (4) "whether 'one judicial actor is better positioned than another to decide the issue in question,'" Matter of Wollrab, 2018 CO 64, ¶ 37, 420 P.3d 960, 968 (quoting Miller, 474 U.S. at 114). We consider these factors in turn.
¶27 First, we consider the nature of the proximate-cause inquiry. In assessing proximate cause, the factfinder first determines the historical facts. The factfinder then determines whether those facts constitute the "natural and probable sequence [that] produced the claimed injury," Stewart, 55 P.3d at 116 (quoting CJI-Crim. 9:10, 9(3) (1983)), based on "the common sense consideration of the risks created by various conditions and circumstances" and "the policy consideration of whether a defendant's responsibility should extend to the results in question," Walcott v. Total Petroleum, Inc., 964 P.2d 609, 612 (Colo.App. 1998); see also Rocky Mountain Planned Parenthood, Inc. v. Wagner, 2020 CO 51, ¶ 66, 467 P.3d 287, 299 (Hart, J., dissenting in part); Ekberg v. Greene, 588 P.2d 375, 377 (Colo. 1978) ("Rather than resting on mechanistic rules of law to determine tort liability, a court should ordinarily allow the jury to make a determination of what is reasonable in each factual setting.").
¶28 Comparing this fact-driven proximate cause inquiry with others of a more hybrid nature is helpful. In considering the nature of the voluntariness of a confession, for example, the Supreme Court concluded that the issue had "a uniquely legal dimension" because it turned on ensuring due process under the Fourteenth Amendment. Miller, 474 U.S. at 116. Such mixed questions often turn on constitutional matters, see, e.g., People v. Matheny, 46 P.3d 453, 459 (Colo. 2002) (acknowledging that whether a suspect is in custody is "a little of both" an issue of fact and of law, but concluding that a trial court's determination should be reviewed de novo because "a constitutional right is implicated"), or the application of statutory requirements, see, e.g., People v. V.K.L., 2022 CO 35, ¶ 20, 512 P.3d 132, 139 (reviewing de novo whether government actions satisfied the Indian Child Welfare Act). In short, legal questions.
¶29 But the proximate-cause inquiry is fundamentally different: it's based on individualized judgments of how facts connect a defendant to a victim, not on constitutional or statutory principles. While proximate-cause determinations undoubtedly can involve legal concepts-simple versus gross negligence, for example, as we will discuss in Part B-proximate cause is, at its core, a fact-based determination. So, this factor weighs in favor of clear-error review.
¶30 Second, we consider history. While this court hasn't considered the standard of review for proximate-cause challenges in the criminal restitution context, we typically review proximate cause challenges in tort actions as factual questions. See, e.g., Wagner, ¶ 30, 467 P.3d at 293; Westin Operator, LLC v. Groh, 2015 CO 25, ¶33 n.5, 347 P.3d 606, 614 n.5; Build It &They Will Drink, Inc. v. Strauch, 253 P.3d 302, 306 (Colo. 2011). Indeed, we only consider proximate cause as a matter of law in tort when the "facts are undisputed and reasonable minds could draw but one inference from them." Gibbons v. Ludlow, 2013 CO 49, ¶ 13, 304 P.3d 239, 244 (quoting Allen v. Martin, 203 P.3d 546, 566 (Colo.App. 2008); Smith v. State Comp. Ins. Fund, 749 P.2d 462, 464 (Colo.App. 1987)). And there's no reason that the law of proximate cause should differ between the two contexts. Proximate cause in the tort and the restitution contexts share (1) a definition, (2) a burden of proof, and (3) a purpose of ensuring that parties causing harm (and only those parties) redress that harm. See Schneider v. Midtown Motor Co., 854 P.2d 1322, 1327 (Colo. 1992) (defining proximate cause in a tort case as an event that "in the natural and probable sequence of things . . . produced the claimed injury"); People v. Clay, 74 P.3d 473, 475 (Colo.App. 2003) (using the same definition of proximate cause for restitution purposes); Branco E. Co. v. Leffler, 482 P.2d 364, 366 (Colo. 1971) (identifying preponderance of the evidence as the burden of proof for proximate cause in tort); People v. Henry, 2018 COA 48M, ¶ 15, 439 P.3d 33, 36 (same in restitution cases); N. Colo. Med. Ctr., Inc. v. Comm. on Anticompetitive Conduct, 914 P.2d 902, 908 (Colo. 1996) (identifying the purpose of proximate cause in the tort context); § 18-1.3-601(1)(a)-(b), C.R.S. (2023) (same in the restitution context). Our historical treatment of proximate cause in tort then strongly suggests that we should apply clear-error review in the restitution context, too.
¶31 Third, we consider language and structure. The legislative declaration for the criminal restitution statute states that "[a]n effective criminal justice system requires timely restitution" and that the purpose of the statute is in part to establish programs that "provide for . . . restitution for victims of crimes in the most expeditious manner." § 18-1.3-601(1)(e), (g)(I) (emphases added). This language suggests that the General Assembly wanted restitution decisions to be quick. And clear-error review, which relieves appellate courts of the obligation "to take a finetoothed comb to the factual disputes in each case," offers this efficient resolution. Carousel Farms Metro. Dist. v. Woodcrest Homes, Inc., 2019 CO 51, ¶ 18, 442 P.3d 402, 407 ("Carousel Farms"). Thus, this third factor also weighs in favor of clear-error review.
¶32 Fourth and finally, we consider institutional competence-whether "one judicial actor is better positioned than another to decide the issue in question." Wollrab, ¶ 37, 420 P.3d at 968 (quoting Miller, 474 U.S. at 114). Because proximatecause determinations involve parsing facts instead of law, trial courts are better situated to resolve them. See Carousel Farms, ¶ 19, 442 P.3d at 407 ("[A]ppellate tribunals don't (and indeed, can't) make findings of fact."). Because this and the preceding three factors all counsel in favor of applying clear-error review, we hold that a district court's determination of proximate cause in a restitution proceeding will not be disturbed unless it is clearly erroneous.
B. No Clear Error Here
¶33 On the merits, Martinez alleges that C.T. turning his car into Martinez's path constituted an independent intervening cause, which severed the causal connection between his theft of C.T.'s bike and the damage to C.T.'s car. Again, for an event to be an independent intervening cause, it must be unforeseeable (so at least grossly negligent if a victim's negligence is to be a factor), and the defendant can't have participated in the event. Stewart, 55 P.3d at 121. The district court concluded that (1) C.T.'s conduct was foreseeable, and (2) Martinez was a participant. The division affirmed, albeit under what we have now determined to be an erroneous standard of review.
¶34 Applying a clear-error standard, we must affirm the district court's findings unless they are without "support in the record." People v. Turner, 2022 CO 50, ¶ 19, 519 P.3d 353, 359 (quoting Melssen v. Auto-Owners Ins. Co., 2012 COA 102, ¶ 16, 285 P.3d 328, 333). So, we look for record support for the district court's foreseeability and participation determinations.
1. Record Support for Foreseeability
¶35 The district court found it was foreseeable that C.T. would pursue Martinez and therefore that C.T. "would attempt to prevent [Martinez] from taking his property by pulling in front of [Martinez]." It determined that C.T.'s action constituted simple negligence if anything because C.T. was "anticipating [Martinez] would stop and thus cease his theft of the bicycle." The division agreed, pointing to record evidence that showed C.T. turned "in [an] attempt to get [Martinez] to stop." Martinez, ¶ 23, 511 P.3d at 744 (alterations in original).
¶36 Martinez acknowledges that C.T.'s pursuit was foreseeable, but he draws the line at C.T. turning the car in front of him. He claims that C.T. intentionally turning into his path elevated recovery of property over safety and therefore constituted gross negligence. He argues, with some force, that "[r]easonable people understand no one should be injured or killed over a stolen bike," and that C.T. simply could have followed Martinez and provided the police with information to help apprehend him.
¶37 Even so, we conclude the district court didn't clearly err in its negligence determination. As previously stated, gross negligence requires "reckless[ness], with conscious disregard for the safety of others." Hamill, 262 P.3d at 954. And here, reasonable minds can differ as to whether C.T.'s conduct reached that threshold. Without question, a court could reasonably reach Martinez's preferred conclusion: C.T. could have left apprehension of Martinez and the bike to the police; therefore, his choice to turn instead was reckless. But the district court's conclusion that there was no gross negligence here was also reasonable. After all, common sense suggests that consciously disregarding Martinez's safety would also have meant disregarding the bike's safety (since Martinez was riding it), and we can fairly assume that C.T. pursued the bike to obtain its undamaged return. Accordingly, the record supports the notion that C.T. turned his car "in an attempt to stop [Martinez]," not in an attempt to recover the bike at all costs. In sum, there is record support for the district court's and the division's determination that C.T.'s action wasn't grossly negligent.
¶38 Martinez also argues that C.T. turning the car was unforeseeable because it exceeded the limits of Colorado's defense-of-property statute. That statute provides, "[a] person is justified in using reasonable and appropriate physical force upon another person when and to the extent that he reasonably believes it necessary to prevent what he reasonably believes to be an attempt by the other person to commit theft." § 18-1-706, C.R.S. (2023). Specifically, Martinez claims that he had already completed the bike theft when C.T. turned his car, so turning the car wasn't "necessary to prevent" Martinez's theft. Id.
¶39 This argument fails at its inception. Acts aren't inherently unforeseeable because they exceed the law's limits. See, e.g., Wagner, ¶ 38, 467 P.3d at 294 (recognizing that a mass shooting-undoubtedly outside the law's limits-could be foreseeable for proximate-cause purposes under the case's particular facts); Ekberg, 588 P.2d at 376 (same for vandalism). So, C.T.'s conduct doesn't need to be within the bounds of the defense-of-property statute to be foreseeable.
¶40 But even assuming that an act must be lawful to be foreseeable, the record supports the district court's determination that the theft wasn't complete when the collision happened, and therefore that C.T. was within the statute's bounds. A theft is completed when a thief "exercise[s] control of the property" and "move[s] it away from an area within defendant's control." People v. Oslund, 2012 COA 62, ¶ 24, 292 P.3d 1025, 1029. The court of appeals has found a theft completed when the victim needed to "find and catch" the thief. Id.; see also People v. Goedecke, 730 P.2d 900, 901 (Colo.App. 1986) (holding a defendant was outside the defense-of-property statute when he attacked a thief "[s]ome time" after the theft). But here, as the district court observed and the division credited, "[C.T.]'s wife observed the bicycle being taken from their garage," and "[C.T.]'s pursuit of [Martinez] was instantaneous." Martinez, ¶ 29, 511 P.3d at 744-45. Therefore, the district court didn't clearly err in concluding that Oslund and Goedecke were factually distinguishable.
The district court and the division majority also observed that Oslund and Goedecke were distinguishable because they arose in a different legal context-jury instructions on self-defense instead of a proximate-cause determination. Martinez, ¶ 26, 511 P.3d at 744.
2. Record Support that Martinez Was a Participant
¶41 The district court also found that Martinez "was clearly participating in the event as he was riding [C.T.]'s bicycle parallel to [C.T.]'s car while he was in the act of stealing the bicycle." The division agreed. Id. at ¶ 30, 511 P.3d at 745.
¶42 Martinez responds that the theft and the collision were separate events (because-he reiterates-the theft was completed) and that he didn't participate in the collision because the record didn't show that he could have stopped or otherwise avoided it. He likens his case to Stewart, in which we held a victim's act of allegedly jumping onto the hood of the defendant's car, even if deemed grossly negligent, wasn't an intervening cause that relieved the defendant of liability for driving over the victim after the victim fell off the car because the two acts were separate and the defendant "drove forward of his own volition." 55 P.3d at 121.
¶43 But Martinez misses Stewart's point. We reached our conclusion in Stewart as part of our foreseeability analysis. We went on to write that, despite the victim's jump and the defendant's driving being separate, the defendant did participate in the victim's jump because the victim's "alleged leap came after [the victim] brushed against [the defendant's] car and after their ensuing verbal altercation." Id. Stewart implies that a defendant can participate in an action that he provokes. It thus supports the district court's finding that Martinez participated in the collision, not Martinez's counterargument. It's less clear here than in Stewart that Martinez's theft of the bike and C.T.'s turning were separate acts. Regardless, the record shows that Martinez was involved in events leading to the collision: He was still "in the act of stealing the bicycle," and he was riding parallel to C.T. and could have stopped when C.T. turned the car to cut him off. These actions make Martinez a participant under Stewart. We therefore conclude that the district court didn't clearly err in determining that Martinez participated in the action that caused harm to C.T.'s car.
III. Conclusion
¶44 We affirm the court of appeals' judgment with respect to the finding of proximate cause, albeit on different grounds. JUSTICE GABRIEL concurred in the judgment.
JUSTICE GABRIEL, concurring in the judgment.
¶45 Relying on cases in the civil tort arena and on its view that Arnold Roman Martinez is not actually challenging the sufficiency of the evidence to support the restitution award in this case (notwithstanding the fact that Martinez asserts that he is doing just that), the majority concludes that in the context of determining a restitution award in a criminal case, the issue of proximate cause presents a factual question that an appellate court reviews for clear error. Maj. op. ¶¶ 3, 21-23, 29-30, 44. I do not agree. Rather, in my view, civil tort cases are inapposite, and the question of proximate cause in the criminal restitution context presents a mixed question of law and fact. Accordingly, in this context, I would review questions of historical fact for clear error and questions of law de novo, with the ultimate conclusion of proximate cause being a question of law that we review de novo. Such a determination is fully consistent with the fact that Martinez is, indeed, challenging the sufficiency of the evidence to support the trial court's finding of proximate cause, as well as with how we have approached similar questions in other contexts.
¶46 Reviewing, then, the trial court's proximate cause determination de novo, I reach the same ultimate conclusion as does the majority. Because I do so by employing a different legal standard, however, I concur in the judgment, only.
I. Analysis
¶47 Restitution is a matter that the trial court decides post-verdict. See § 18-1.3-603, C.R.S. (2023) (requiring a court to include consideration of restitution in most orders of conviction and detailing what the court's restitution order must include).
¶48 "Restitution" is defined as "any pecuniary loss suffered by a victim . . . proximately caused by an offender's conduct and that can be reasonably calculated and recompensed in money." § 18-1.3-602(3)(a), C.R.S. (2023). To establish a right to restitution, then, "[t]he prosecution must prove by a preponderance of the evidence that the defendant's conduct proximately caused the victim's loss and the amount of that loss." People v. Babcock, 2023 COA 49, ¶ 19, 535 P.3d 981, 987. In the context of a restitution award, proximate cause is defined as "a cause which in natural and probable sequence produced the claimed injury and without which the claimed injury would not have been sustained." People v. Rivera, 250 P.3d 1272, 1274 (Colo.App. 2010); see also People v. Stewart, 55 P.3d 107, 120-21 (Colo. 2002) ("A defendant is responsible for serious bodily injury to another if the injury is a natural and probable consequence of his misconduct.").
¶49 Unlawful conduct that is broken by an independent intervening cause, however, cannot be the proximate cause of injury to another. Stewart, 55 P.3d at 121. "An independent intervening cause 'is an act of an independent person or entity that destroys the causal connection between the defendant's act and the victim's injury and, thereby becomes the cause of the victim's injury.'" Id. (quoting People v. Saavedra-Rodriguez, 971 P.2d 223, 226 (Colo. 1998)). To constitute an intervening cause, "an event must be unforeseeable and one in which the accused does not participate." Id. As a general matter, simple negligence is deemed foreseeable and thus would not constitute an independent intervening cause. Id. Gross negligence, in contrast, is deemed unforeseeable and may constitute an independent intervening cause. Id.; see also People v. Sieck, 2014 COA 23, ¶¶ 9-11, 351 P.3d 502, 504 (noting that simple negligence is foreseeable and does not constitute an independent intervening cause while gross negligence is unforeseeable and can constitute such an independent intervening cause, and then holding that a victim's failure to wear a seatbelt was not gross negligence for restitution purposes).
¶50 Accordingly, the question of proximate cause in the context of determining restitution in a criminal case requires the trial court, not a jury, to determine whether, among other things, (1) the defendant's actions, in natural and probable sequence, caused the victim's claimed injury; (2) absent the defendant's conduct, the claimed injury would not have occurred; (3) the defendant's conduct was not broken by an independent intervening cause; (4) any intervening event was unforeseeable; and (5) any intervening event that caused the victim's injuries was the product of gross, and not simple, negligence.
¶51 In my view, these matters do not present factual questions. To the contrary, they involve the analysis of highly complex questions of law to be performed by the trial court.
¶52 For these reasons, I believe that the question of proximate cause in the context of determining restitution in a criminal case presents a mixed question of law and fact, with the ultimate determination of proximate cause being a question of law, and many courts have so determined. See, e.g., United States v. Sepulveda, 64 F.4th 700, 712-13 (5th Cir. 2023) ("[W]e generally review the legality of a restitution order de novo.... [S]ome of our cases have reviewed de novo the amount of restitution ordered where the defendant attacks the causal link between the restitution ordered and the offense. But in other cases, we have applied an abuse-of-discretion standard under similar circumstances. Recently, we stated that '[w]e review the legality of [a restitution] award de novo, . . . its amount for abuse of discretion,' and the court's '[factual] finding regarding the amount of loss . . . for clear error.'") (last five alterations in original) (quoting United States v. Williams, 993 F.3d 976, 980 (5th Cir. 2021); other citations omitted); United States v. Ruiz-Lopez, 53 F.4th 400, 404 (6th Cir. 2022) (noting that the court reviews the propriety of a restitution award de novo but the amount of the award for an abuse of discretion); United States v. Aumais, 656 F.3d 147, 154 (2d Cir. 2011) (treating a magistrate judge's determination of proximate cause in the context of a restitution award as a mixed question of law and fact, and concluding that although the magistrate judge's factual findings were supported by the record, as a matter of law, those facts did not establish the requisite causal connection); People v. Martinez-Chavez, 2020 COA 39, ¶ 20, 463 P.3d 339, 343 (construing the question of causation in the context of a restitution determination as "a mixed question of law and fact, not a purely legal question, that the prosecution bears the burden to prove"); Commonwealth v. Stoops, 290 A.3d 721, 723 (Pa. Super. Ct. 2023) (reviewing, in the context of a restitution determination, a defendant's contention that the trial court had erred in finding causation, and stating that the defendant's "claim concern[ed] the legality of his restitution sentence" and "[w]hen reviewing the legality of a sentence, we apply a plenary scope and de novo standard of review"); see also People v. Stone, 2020 COA 24, ¶ 7, 471 P.3d 1159, 1162 (noting that although a court generally reviews a court's restitution order for an abuse of discretion, when the issue is whether sufficient evidence justified the order, the court reviews the order de novo).
¶53 Treating the proximate cause determination in this way and therefore applying a de novo standard of review are consistent with the fact that challenges to restitution awards are typically in the nature of challenges to the sufficiency of the evidence to support those awards. Indeed, notwithstanding the majority's assertion to the contrary, Maj. op. ¶¶ 21-23, that is the essence of Martinez's contention here. Specifically, he contends that the evidence did not establish the requisite proximate cause to support the restitution award at issue. Challenges like this ask whether the evidence was substantial and sufficient to establish that the defendant was responsible for the sums that the defendant is being asked to pay in restitution. It is well-settled that we review challenges to the sufficiency of the evidence de novo, see, e.g., Johnson v. People, 2023 CO 7, ¶ 13, 524 P.3d 36, 40, and I would do the same here.
¶54 On this point, I find it telling that to reach the opposite conclusion, the majority both creates a new category of sufficiency of the evidence claims and necessarily recharacterizes what Martinez is arguing. The majority begins its analysis by asking whether Martinez presents "a true sufficiency challenge," creating its own distinction between "true" and, I gather, "not really" sufficiency of the evidence claims. Maj. op. ¶ 19. No party in this case has argued for such a distinction, I know of no case law supporting it, and I worry about its unintended consequences (e.g., having the People respond to future sufficiency of the evidence arguments by contending that the defendants are not asserting "true" sufficiency claims).
¶55 Having created this new category of sufficiency of the evidence claims, the majority then proceeds to conclude that although Martinez plainly states that he is challenging the sufficiency of the evidence, he is not "really" presenting such a challenge. Id. at ¶ 21.
¶56 I am not persuaded. A sufficiency of the evidence claim is a sufficiency of the evidence claim, and Martinez argues what he argues. My position honors both. The majority's position does not and simply assumes its conclusion: proximate cause is a factual question because Martinez is purportedly challenging the "findings of fact related to proximate cause." Id.
¶57 My conclusion is also consistent with how we handle similar questions in other contexts. For example, in cases involving the voluntariness of a confession, we have concluded that the question of voluntariness presents a mixed question of law and fact. See, e.g., People v. Zadran, 2013 CO 69M, ¶ 13, 314 P.3d 830, 834. Accordingly, we defer to the trial court's findings of historical fact, declining to overturn them if they are supported by competent evidence in the record, but we review the legal effect of those facts de novo. Id. Although the majority sees a distinction between such a case and the case now before us, Maj. op. ¶ 28, I do not. Specifically, contrary to the majority's suggestion that the standard of review in voluntariness cases is based on the fact that voluntariness involves a constitutional, and thus a legal, question, id., I read those cases as turning more generally on the legal nature of the question presented, regardless of the source of law at issue, see, e.g., Miller v. Fenton, 474 U.S. 104, 115-16 (1985) (noting that voluntariness is a "legal question" and that "the dispositive question of the voluntariness of a confession has always had a uniquely legal dimension"). For the reasons noted above, I see the ultimate determination of proximate cause in the context of a restitution determination likewise to constitute a legal question, and I would thus review this determination under the same standard as we review voluntariness determinations in confession cases.
¶58 Similarly, in determining whether a defendant is in custody for purposes of Miranda v. Arizona, 384 U.S. 436 (1966), the court, not the jury, is tasked with deciding whether the defendant was in custody during an interrogation, and we have concluded that this issue, too, presents a mixed question of law and fact that we review de novo. People v. Sanders, 2023 CO 62, ¶ 10, 539 P.3d 148, 151. In my view, the question of whether a defendant was in custody at the time of an interrogation is far more factual than the multi-layered question of whether the prosecution has established proximate cause in the context of a restitution determination. Accordingly, I see no reason to treat a court's custody determination in the Miranda context as a mixed question of law and fact that we review de novo while treating a court's proximate cause determination in the criminal restitution context as a purely factual one.
¶59 I am not persuaded otherwise by the majority's reliance on civil tort cases. Maj. op. ¶ 30. As every first-year law student learns, causation is an element of the tort of negligence. See Vigil v. Franklin, 103 P.3d 322, 325 (Colo. 2004). As a result, a tort plaintiff must present the issue to the factfinder, and the factfinder must decide if that element has been established.
¶60 Restitution, unlike a tort claim, is not a cause of action. Nor does it have any elements. It is an "equitable remedy [that] was intended to reimburse a person wronged by the actions of another." United States v. Christopher, 273 F.3d 294, 299 (3d Cir. 2001); accord People v. Johnson, 2020 COA 124, ¶ 10, 487 P.3d 1262, 1265, aff'd, 2021 CO 79, 499 P.3d 1045. Accordingly, in this context, I would construe the determination of proximate cause to be a legal determination that is a prerequisite to the ultimate legal conclusion of whether restitution should be awarded.
¶61 Applying this standard here, I would conclude, as does the majority, that the trial court properly awarded restitution in this case. Maj. op. ¶ 44. Specifically, reviewing the question of proximate cause de novo, I would conclude that it was foreseeable that the victim would have chased Martinez, after Martinez stole the victim's bicycle, and that the victim would have attempted to prevent Martinez's escape, resulting in injury or property damage. I would further conclude that the victim's actions here did not constitute an independent intervening cause. Although the victim's actions were perhaps unwise and should not be encouraged, I cannot say that such actions were grossly negligent, thereby breaking the causal chain. See Stewart, 55 P.3d at 121. At worst, I view such actions as simple negligence, which, as noted above, is foreseeable. Id.
II. Conclusion
¶62 For these reasons, although I would review the trial court's proximate cause determination as a mixed question of fact and law, with the ultimate causation determination to be a question of law, like the majority, I would affirm the restitution order in this case.
¶63 Accordingly, I respectfully concur in the judgment, only.
Comments