concurring in part and dissenting in part:
I agree with the majority that the language of the PLCAA’s predicate exception is ambiguous. The interpretation on which the majority ultimately settles, however, requires deciding what I consider to be a substantial constitutional question: whether, if the PLCAA requires the dismissal of Plaintiffs’ pending state causes of action, the , statute will unconstitutionally deprive them of a protected property interest.
The majority resolves this question by concluding that the PLCAA’s mandatory dismissal provision is rationally related to a legitimate government interest and that no heightened level of constitutional scrutiny is warranted. ’ The majority’s cursory discussion of the constitutional issue belies the sweeping nature of what it reads the PLCAA to do, and the difficult questions of constitutional law required to uphold that reading. Neither the Supreme Court nor this Circuit has ever made clear that rational basis review is the proper standard on which to review a federal statute that retroactively requires the dismissal of pending caubes of action for injuries cognizable at common law but does not leave any alternative means of redress. Moreover, even if we were to assume that no heightened level of scrutiny is appropriate, I am not convinced that such a statute would survive the rational basis review outlined by the Supreme Court in Duke Power Co. v. Carolina Envtl. Study Group, 438 U.S. 59, 98 S.Ct. 2620, 57 L.Ed.2d 595 (1978).
The majority tacitly breaks new ground in deciding these questions. It need not— and should not — do so. The canon of constitutional avoidance counsels that we *1147should “construe the statute to avoid [serious constitutional questions] unless such a construction is plainly contrary to the intent of Congress.” Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575, 108 S.Ct. 1392, 99 L.Ed.2d 645 (1988); see also Clark v. Martinez, 543 U.S. 371, 380-81, 125 S.Ct. 716, 160 L.Ed.2d 734 (2005); NLRB v. Catholic Bishop of Chicago, 440 U.S. 490, 507, 99 S.Ct. 1313, 59 L.Ed.2d 533 (1979). In accordance with that maxim, I would adopt a different, but in my view equally supportable, reading of the ambiguous statutory language, which would permit the Plaintiffs’ suit as against Defendants Glock and RSR (“Defendants”) to go forward. I therefore respectfully dissent from the majority’s discussion of Plaintiffs’ substantive due process challenge, as I would not decide the question, and from its holding that their lawsuit does not come within the PLCAA’s predicate exception.
I.
A.
Plaintiffs have raised a number of constitutional challenges. The one that concerns me here is their substantive due process argument. Before I explain why their challenge presents a serious constitutional question, I think it useful, as a preliminary matter, to identify the property interest on which Plaintiffs contend the PLCAA intrudes. See Wedges/Ledges of Cal., Inc. v. City of Phoenix, 24 F.3d 56, 62 (9th Cir.1994).
Generally speaking, if a plaintiff alleges that a fundamental right is burdened by a state action, the state action is subject to strict scrutiny and cannot stand unless it is “narrowly tailored to serve a compelling state interest.” Washington v. Glucksberg, 521 U.S. 702, 721, 117 S.Ct. 2258, 138 L.Ed.2d 772 (1997) (internal quotation marks omitted). If a lesser liberty or property interest is at stake, the state action is subject to rational basis review, which requires “a reasonable relation to a legitimate state interest.” Id. at 722, 117 S.Ct. 2258. Various forms of intermediate scrutiny, in between these two poles, have sometimes been found applicable. See Hutchins v. District of Columbia, 188 F.3d 531, 563 & n. 24 (D.C.Cir.1999) (Rogers, J., concurring in part) (collecting cases).
Plaintiffs here have asserted a property interest in maintaining their pending state-law causes of action. They acknowledge that the monetary award those causes of action might ultimately yield if they proceeded to a final judgment is uncertain, so they currently have no enforceable right to any particular amount of damages. But, they assert, the PLCAA also trenches on a separate and independently valuable interest: their interest in maintaining their causes of action prior to judgment and not having them dismissed.
The majority implicitly agrees with Plaintiffs’ position that this interest is cognizable and protected by the Due Process Clause. This much is clear, because the majority applies rational basis review. If Plaintiffs’ interest were wholly unprotected, no scrutiny, rational basis or otherwise, would be required.
I pause to emphasize this point, because the majority then goes on to make a potentially misleading statement: that Plaintiffs have no “vested property right in their accrued state-law causes of action.” Maj. Op. at 1141. What the majority appears to mean is that Plaintiffs have not stated a fundamental property interest deserving of heightened scrutiny. I do not understand the majority to mean that Plaintiffs have no property interest in their causes of action at all. If that were what the majority meant, it would be quite wrong. Like stocks or business operating licenses, the probable value of a plaintiffs cause of action may fluctuate over time — here, be*1148tween filing and the entry of judgment— but such fluctuations do not mean that the cause of action is without value.1 Plaintiffs have expended time and money to maintain their lawsuit, and, if they were so inclined, they could have “sold” them causes of action to defendants by settling for a sum of money reflecting the expected recovery at that point in the litigation.2 A pending cause of action, therefore, may be more or less valuable at various points during its pendency, but, even before it is reduced to a final dollar amount, it is a “species of property protected by the ... Due Process Clause,” as the Supreme Court held in Logan v. Zimmerman Brush Co., 455 U.S. 422, 428, 102 S.Ct. 1148, 71 L.Ed.2d 265 (1982) (citations omitted).
I therefore read the majority’s statement that Plaintiffs lack a “vested property right in their accrued state-law causes of action,” Maj. Op. at 1141, to indicate the majority’s view that although there is a protected property interest at stake, it is not a fundamental right, so rational basis review, rather than strict scrutiny, is sufficient. This usage, though odd, appears to be consistent with our case law. Lyon v. Agusta S.P.A., 252 F.3d 1078 (9th Cir.2001), for example, stated that “a party’s property right in any cause of action does not vest until a final unreviewable judgment is obtained,” id. at 1086 (internal quotation marks and emphasis omitted), but it also assumed that a statute depriving the party of the ability to litigate that cause of action must nevertheless withstand rational basis review. See id. (“Of course, the legislature must act in a rational manner; that almost goes without saying. Here the choice was assuredly rational.”) (internal citations omitted). Accord Duke Power, 438 U.S. at 88 n. 32, 98 S.Ct. 2620 (stating that “[a] person has no property, no vested interest, in any rule of the common law,” but applying rational basis review to federal statute precluding suit) (internal quotation marks and citation omitted); Fields v. Legacy Health Sys., 413 F.3d 943, 955-56 (9th Cir.2005) (upholding statute against substantive due process challenge on rational basis review; recognizing a distinction between vested and non-vested property rights only in the context of petitioner’s procedural due process claim).3
*1149B.
If this case were directly controlled by the case law just cited, I would agree that rational basis review is the proper level of scrutiny to apply here. And if the PLCAA were indistinguishable from the statutes at issue in Lyon, Duke Power, Fields and Austin, I would be constrained to hold that the PLCAA’s intrusion on Plaintiffs’ interest survives rational basis review.
But the PLCAA is unlike those other statutes in critical respects. We have never upheld against substantive due process attack a federal statute with precisely the PLCAA’s constellation of characteristics: (1) It completely extinguishes an individual litigant’s ability to litigate a cause of action, rather than limiting the amount of recovery or the procedure for bringing suit, and it leaves no alternative channel by which the individual may address his injury; and (2) the individual’s cause of action is for an injury that would be cognizable under state common law, and it was filed and pending at the time of the federal statute’s enactment.4 To hold such a stat*1150ute constitutional on rational basis review, despite the absence of any provision for alternative forms of redress, is to step onto new and uncertain constitutional territory.
In general, the majority is correct that Congress may limit or abrogate rights recognized at common law when enacting legislation “adjusting the burdens and benefits of economic life,” so long as the abrogation is rationally related to a permissible goal. Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15, 96 S.Ct. 2882, 49 L.Ed.2d 752 (1976). It is not clear from past case law, however, whether rational basis review is the appropriate level of scrutiny for a statute that abrogates common-law remedies without providing or leaving open a substitute remedial scheme.
In fact, the Supreme Court as a whole and individual Justices of the Court have repeatedly recognized that “[qjuite serious constitutional questions might be raised if a legislature attempted to abolish certain categories of common-law rights in some general way.” PruneYard Shopping Ctr. v. Robins, 447 U.S. 74, 93-94, 100 S.Ct. 2035, 64 L.Ed.2d 741 (1980) (Marshall, J, concurring). See also New York Central R.R. Co. v. White, 243 U.S. 188, 201, 37 S.Ct. 247, 61 L.Ed. 667 (1917) (expressing uncertainty as to whether “a state might, without violence to the constitutional guaranty of due process of law,’ suddenly set aside all common-law rules respecting liability as between employer and employee, without providing a reasonably just substitute”). To date, the Supreme Court has never decided what level of constitutional scrutiny applies to a statute that abrogates a common-law cause of action and leaves no alternative remedy available. Dissenting from the dismissal of certiorari in a ease that would have presented this question squarely, Justice White noted:
Whether due process requires a legislatively enacted compensation scheme to be a quid pro quo for the common-law or state-law remedy it replaces, and if so, how adequate it must be, ... appears to be an issue unresolved by this Court, and one which is dividing the appellate and highest courts of several States.
Fein v. Permanente Med. Group, 474 U.S. 892, 894-95, 106 S.Ct. 214, 88 L.Ed.2d 215 (1985) (White, J., dissenting from dismissal of certiorari). This same question was expressly left unresolved in Duke Power, which declined to decide what constitutional test would apply to a statute that left no alternative remedies available:
Initially, it is not at all clear that the Due Process Clause in fact requires that a legislatively enacted compensation scheme either duplicate the recovery at common law or provide a reasonable substitute remedy. However, we need not resolve this question here since the Price-Anderson Act does, in our view, provide a reasonably just substitute for the common-law or state tort law remedies it replaces.
*1151438 U.S. at 88, 98 S.Ct. 2620 (footnote omitted).
Following Duke Power’s lead, when this Circuit confronted a substantive due process challenge to a portion of the Atomic Energy Act abolishing individuals’ ability to sue certain government contractors for injuries recognized at state common law in Atmospheric Testing, it upheld the Act on rational basis review only in conjunction with the observation that substitute remedies — both the Federal Tort Claims Act and veterans benefits legislation — were available to compensate the would-be plaintiffs for the unavailability of personal injury and wrongful death suits. See Atmospheric Testing, 820 F.2d at 990-91.5 Other courts of appeal have done the same. See, e.g., Hammond, 786 F.2d at 14 (noting that, in addition to the Federal Tort Claims Act’s administrative scheme, “there may be government compensation available to many of those injured by radiation, including the plaintiff here, under veterans benefits legislation, ... or the [Federal Employees Compensation Act].”); Ducharme v. Merrill-Nat’l Labs., 574 F.2d 1307, 1309 (5th Cir.1978) (upholding the Swine Flu Act’s prohibition on tort suits against private manufacturers of swine flu vaccine, but also noting that “[t]he [substitute] cause of action provided by the Swine Flu Act to an injured person against the United States is substantially the same as that afforded ... under [Louisiana law] except that under the Swine Flu Act no trial by jury is afforded and the plaintiff is required to seek first administrative review of his claim.”)
Just as the federal courts are reluctant to construe a statute as abolishing common-law rules without providing some alternative method of redress, so Congress is reluctant to pass legislation immunizing a private industry from common-law tort liability — for example, when that industry is acting in concert with a governmental program or playing a vital role in the nation’s economy or defense — without preserving state tort law as a parallel track or providing some alternative mechanism to compensate injured parties.6 Similarly, the Supreme Court has sometimes declined to read federal statutes as extinguishing the availability of state tort causes of action in preemption cases — even where no due process argument was advanced by the parties — when the statute *1152lacks a clear statement that Congress intended to do so. See, e.g., Wyeth v. Levine, — U.S.-, 129 S.Ct. 1187, 1200, 173 L.Ed.2d 51 (2009); Sprietsma v. Mercury Marine, 537 U.S. 51, 64, 123 S.Ct. 518, 154 L.Ed.2d 466 (2002).
The statute at issue here is different. As interpreted by the majority, the PLCAA retroactively extinguishes Plaintiffs’ pending state tort causes of action, and it leaves them without any remedy for the injuries they claim they have suffered due to Defendants’ unlawful acts. No controlling case law establishes that such legislation survives constitutional scrutiny. Rather, the cases canvassed above suggest that at least a modified form of rational basis review, and perhaps a more searching type of review, may be warranted for such a statute.
More specifically, if one applied to a statute abolishing pending common-law causes of action and providing no alternative remedy a test no more searching than the rational basis review applied to the Price-Anderson Act in Duke Power, it is possible that such a statute would be held to violate due process. Duke Power inquired whether “the legislature has acted in an arbitrary and irrational way,” 438 U.S. at 84, 98 S.Ct. 2620 (quoting Usery, 428 U.S. at 15, 96 S.Ct. 2882), but, as part of its “arbitrary and irrational” inquiry, engaged in a lengthy examination of the Price-Anderson Act, taking into consideration among other things that the Act provided a compensation scheme that was a “reasonably just substitute” to the common law, id. at 88, 98 S.Ct. 2620, and perhaps even an improvement on the common law.7 In other words, the availability and effectiveness of alternative remedies was a factor in determining whether Congress had acted rationally or not. Thus, Duke Poiver applied a modified rational basis test, not unlike the due process test applied in cases involving retroactive legislation. See Pension Benefit Guar. Corp. v. R.A. Gray & Co., 467 U.S. 717, 729, 104 S.Ct. 2709, 81 L.Ed.2d 601 (1984) (economic legislation that Congress plainly intends to have retroactive effect must not only be rationally related to a legitimate state interest as a general matter, but there must also be a rational basis for its retroactive application). Duke Power suggests that, at a minimum, a modified rational basis test would apply here, rendering the PLCAA constitutional only if there were rational bases both for Congress’s failure to supply any alternative remedies and for its decision to apply the dismissal provision retroactively to extinguish the remedies for common-law injuries that had already occurred and were the subject of pending litigation.8
*1153Alternatively, Duke Power leaves open the possibility that where the challenged statute leaves no alternative remedy available, the statute should be subject not to rational basis review but to a heightened form of scrutiny. See Duke Power, 438 U.S. at 88, 98 S.Ct. 2620. Concurring in PruneYard, which upheld a California state constitutional limitation on the availability of private trespass claims against peaceful leafletters, Justice Marshall agreed that states may alter the protection of the common law to a point, but he emphasized:
I do not understand the Court to suggest that rights of property are to be defined solely by state law, or that there is no federal constitutional barrier to the abrogation of common-law rights by Congress or a state government. The constitutional terms “life, liberty, and property” do not derive their meaning solely from the provisions of positive law. They have a normative dimension as well, establishing a sphere of private autonomy which government is bound to respect. Quite serious constitutional questions might be raised if a legislature attempted to abolish certain categories of common-law rights in some general way. Indeed, our cases demonstrate that there are limits on governmental authority to abolish “core” common-law rights, including rights against trespass, at least without a compelling showing of necessity or a provision for a reasonable alternative remedy.
PruneYard, 447 U.S. at 93, 100 S.Ct. 2035 (Marshall, J., concurring) (footnote omitted; emphases added). In other words, even though no one has a protected property interest in any particular rule of the common law “entitling him to insist that it shall remain unchanged for his benefit,” White, 243 U.S. at 198, 37 S.Ct. 247, an individual does have a weighty property interest in having some legal means available to redress an injury that would have been compensable at common law. Justice Marshall’s concurrence in PruneYard suggests this interest may be so weighty as to require not merely a rational relation to a legitimate governmental interest, but a “compelling showing of necessity” — a heightened form of scrutiny. PruneYard, 447 U.S. at 93, 100 S.Ct. 2035 (Marshall, J., concurring).
C.
I will not venture further into this unsettled constitutional territory. For purposes of the avoidance canon, it is sufficient to determine that a serious constitutional question exists, and the case law I have just canvassed demonstrates that this is so. Clark, 543 U.S. at 381, 125 S.Ct. 716. The majority disagrees, noting that “[t]he dissent does not, and cannot, point to a single case in which we, the Supreme Court, or any sister circuit has held that a federal statute violates substantive due process for the reasons asserted by Plaintiffs.” Maj. Op. at 1143. But that is precisely the point of the constitutional avoidance canon — to avoid open questions. As I have explained, the Supreme Court in Duke Power expressly left open the question whether “the Due Process Clause ... requires that a legislatively enacted compensation scheme either duplicate the recovery at common law or provide a reasonable substitute remedy.” Duke Power, 438 U.S. at 88, 98 S.Ct. 2620. See also PruneYard, 447 U.S. at 93, 100 S.Ct. 2035 (Marshall, J., concurring) (noting that the question whether Congress can constitutionally “abolish certain categories of corn*1154mon-law rights in some general way ... without a compelling showing of necessity or a provision for a reasonable alternative remedy” is unresolved). Nor has this Court ever upheld against constitutional challenge a statute with the same sweepingly preclusive effect on state tort remedies as the majority reads the PLCAA to have.9 Precisely because we have no precedent to follow in resolving Plaintiffs’ substantive due process challenge, we should not do so if we can avoid it.
I therefore cannot join the majority in rejecting on the merits Plaintiffs’ substantive due process challenge. The applicable case law does not compel the majority’s conclusions that, applying rational basis review, the PLCAA survives review. Nor does it foreclose the possibility that a focused form of rational basis review, taking special account of the retroactivity feature and the lack of any alternative means of redress, or, even, heightened scrutiny applies. Reading the PLCAA to extinguish Plaintiffs’ claims without providing any alternative scheme for compensation thus raises serious constitutional questions that neither we nor the Supreme Court have resolved. I do not know how I would resolve these questions if they were unavoidably before me, but I am certain that they are more serious and complex than the majority’s brief treatment suggests.
II.
Given my view of the constitutional issue in this case, I am constrained to apply the venerable maxim of statutory interpretation prescribing that where ambiguous statutory language is capable of bearing two or more interpretations, courts should adopt the interpretation that does not raise a serious constitutional question “unless such construction is plainly contrary to the intent of Congress.” DeBartolo, 485 U.S. at 575, 108 S.Ct. 1392. See also INS v. St. Cyr, 533 U.S. 289, 299-300, 121 S.Ct. 2271, 150 L.Ed.2d 347 (2001) (“[I]f an otherwise acceptable construction of a statute would raise serious constitutional problems, and where an alternative interpretation of the statute is ‘fairly possible,’ we are obligated to construe the statute to avoid such problems.”) (quoting Crowell v. Benson, 285 U.S. 22, 62, 52 S.Ct. 285, 76 L.Ed. 598 (1932)).
As I explain below, there is such an alternative interpretation of the PLCAA’s predicate exception: one that would avoid raising the substantive due process question altogether, by construing Plaintiffs’ state-law causes of action as arising under “statute[s] ... applicable to the sale and manufacture” of firearms. 15 U.S.C. § 7903(5)(A)(iii). This alternative interpretation is not only “fairly possible” in light of the statute’s text and legislative history, see St. Cyr, 533 U.S. at 300, 121 S.Ct. 2271 (internal citation and quotation marks omitted), but also preferable to the majority’s reading raising the constitutional questions just outlined.
A.
The PLCAA’s predicate exception creates a carve-out from the PLCAA’s mandatory dismissal provision for “action[s] in which a manufacturer or seller of a qualified product knowingly violated a State or Federal statute applicable to the sale or marketing of the product, and the violation was a proximate cause of the harm for *1155which relief is sought.” 15 U.S.C. § 7903(5)(A)(iii).
The ambiguity in this provision derives primarily from its use of the word “applicable.” As the majority notes, the phrase “applicable to X” can mean, broadly, “capable of being applied to X,” or, narrowly, “specifically or even exclusively relevant to X.” Construing other statutes and enactments, courts have read the word “applicable” more broadly or more narrowly, depending upon contextual clues. See, e.g., Fong v. Glover, 197 F.2d 710, 711 (9th Cir.1952); McGee v. Peake, 511 F.3d 1352 (Fed.Cir.2008); Snyder v. Buck, 75 F.Supp. 902, 907 (D.D.C.1948), vacated on other grounds, 179 F.2d 466 (D.C.Cir.1949). Like the majority, I think ease law construing what Congress meant when it used the word “applicable” in other statutes unrelated to the PLCAA does little to illuminate the word’s meaning here. See Fogerty v. Fantasy, Inc., 510 U.S. 517, 522-24, 114 S.Ct. 1023, 127 L.Ed.2d 455 (1994). What is clear, at least, is that the predicate exception cannot possibly encompass every statute that might be “capable of being applied” to the sale or manufacture of firearms; if it did, the exception would swallow the rule, and no civil lawsuits would ever be subject to dismissal under the PLCAA. I therefore agree with the majority that a limiting principle must be found, and that rather than trying to locate it in the word “applicable” itself, we must look to the predicate exception’s surrounding words.
In my view, the key to interpreting the predicate exception is Congress’s use of the word “knowingly.” 15 U.S.C. § 7903(5)(A)(iii). Generally speaking, a “knowing” violation of a given law requires “proof of [the defendant’s] knowledge of the facts that constitute the offense.” Bryan v. United States, 524 U.S. 184, 193, 118 S.Ct. 1939, 141 L.Ed.2d 197 (1998); see also Cal.Penal Code § 7, para. 5 (stating that for purposes of the California Penal Code, “[t]he word ‘knowingly’ imports only a knowledge that the facts exist which bring the act or omission within the provisions of this code. It does not require any knowledge of the unlawfulness of such act or omission.”). Knowing conduct thus stands in contrast to negligent conduct, which typically requires only that the defendant knew or should have known each of the facts that made his act or omission unlawful and/ or the harm that was likely to occur.10
Neither of the California statutes on which Plaintiffs’ suit is based requires actual knowledge as a prerequisite for liability. The negligence statute, Cal. Civ.Code § 1714(a), requires as part of the proximate cause analysis that the harm caused by a defendant’s act or omission be “reasonably foreseeable.” Lugtu v. Cal. Highway Patrol, 26 Cal.4th 703, 716, 110 Cal.Rptr.2d 528, 28 P.3d 249 (2001). A private nuisance under the nuisance statute, Cal. Civ.Code §§ 3479-80, generally requires a showing that the defendant’s act caused an “unreasonable invasion of [the plaintiff]’s interest in the free use and enjoyment of [hi]s property.” Hellman v. La Cumbre Golf & Country Club, 6 Cal.App.4th 1224, 1230, 8 Cal.Rptr.2d 293 (1992). This “unreasonableness” factor plays a similar role to the “foreseeability” factor in the negligence context, and California courts have recognized that an action for nuisance will *1156often require functionally the same showing as an action for negligence. See El Escorial Owners’ Ass’n v. DLC Plastering, Inc., 154 Cal.App.4th 1337, 1349, 65 Cal.Rptr.3d 524 (2007); Pamela W. v. Millsom, 25 Cal.App.4th 950, 954 n. 1, 30 Cal.Rptr.2d 690 (1994); Lussier v. San Lorenzo Valley Water Dist., 206 Cal.App.3d 92, 103-04, 253 Cal.Rptr. 470 (1988).
That neither Cal. Civ.Code § 1714 nor Cal. Civ.Code §§ 3479-80 requires knowing conduct is not the end of the inquiry. The PLCAA’s predicate exception does not limit its application to suits for “violations of State or Federal statutes that require knowing conduct”; rather, it applies to suits for “knowing[ ] violation[s][of] ... State or Federal statute[s].” 15 U.S.C. § 7903(5)(A)(iii). The difference is material: The PLCAA’s actual knowledge requirement can quite reasonably be read to create a mental-state overlay, a heightened requirement that a plaintiff must meet if his lawsuit is to proceed under the new PLCAA regime, regardless of whether the underlying statute requires such a mens rea.11
Plaintiffs here allege that Defendants knowingly committed a range of acts in violation of California negligence and nuisance law. Specifically, they allege that “Defendants ... knowingly participate in and facilitate the secondary market where persons who are illegal purchasers[, including Furrow,] ... obtain their firearms,” First Amended Complaint (“FAC”) ¶ 31 (emphasis added), and that “Defendants] ... select and develop distribution channels that they know regularly provide guns to criminals and underage end users ... [and, despite information from government crime trace reports,] knowingly supply a range of disreputable distributors, dealers, gun shops, pawnshops, gun shows, and telemarketers in the State of California....” Id. ¶ 32 (emphases added).12 Plaintiffs also allege that Defendants intentionally flood police departments with *1157frequent waves of upgrades, enabling Defendants to resell police departments’ retired models on the secondary market. And Plaintiffs allege that “Defendants have full knowledge that their policies and practices will and regularly do result in substantially increased levels of firearms use in crime ... in California, and that their conduct ... [unreasonably] interferes with the public safety, health or peace ....” Id. ¶ 126-29 (emphasis added).
With their allegations, Plaintiffs are not imputing vicarious liability to Defendants for Furrow’s unlawful acts. Rather, they are alleging that Defendants themselves knowingly engaged in unlawful conduct: dangerous distribution and marketing practices, and the knowing maintenance of a nuisance. And while the majority is correct that Furrow’s shooting was the last link in the causal chain that occasioned Plaintiffs’ suit, see Maj. Op. at 1130 n.1, the violations of law for which Plaintiffs seek redress as against Defendants Glock and RSR are separate from the violations of law that Furrow himself committed. In other words, Plaintiffs advance a theory of direct liability, not vicarious liability, against Defendants. Their cause of action is premised on the allegation that Defendants’ own wrongful conduct proximately caused them harm.13
One could quarrel with Plaintiffs’ theory of causation, to be sure, and perhaps they would lose on summary judgment or at trial. But this Court determined on a prior appeal that Plaintiffs have properly stated violations of California law (including satisfying the statutes’ requirement of proximate cause) for purposes of surviving a motion to dismiss, see Ileto v. Glock, 349 F.3d 1191, 1194 (9th Cir.2003), and that holding is the law of the case. Maj. Op. at 1132 & n.2. The question now before us is whether, assuming that Plaintiffs have stated violations of California law, they have also alleged sufficient facts to come within the PLCAA’s predicate exception and avoid its mandatory dismissal requirement.
At the time Plaintiffs’ complaint was originally filed, the PLCAA had not yet been enacted. So, by alleging that Defendants had actual knowledge of the impact *1158of their intentional actions rather than that the impact was reasonably foreseeable, Plaintiffs put forward more than they needed to state a claim under the California statutes.14 When Congress passed the PLCAA in 2005, though, it effectively raised the bar concerning what Plaintiffs must allege to avoid a motion to dismiss. As a result, even when the underlying state statute does not itself require “knowing” action, plaintiffs must now allege and, ultimately, prove such actions to survive PLCAA preemption.15
In short, Plaintiffs have presciently undertaken to prove that Defendants knew the impact of their actions and undertook them anyway, even though the underlying state statutes require only that Defendants should have known, and not that they actually knew, the impact of their actions. Because Plaintiffs have adequately and specifically pleaded actual knowledge, they have shown “knowing! ] violations]” of the California statutes that form the basis for their suit. 15 U.S.C. 7903(5)(A)(iii).
Understanding the phrase “knowingly violated” as imposing a heightened pleading requirement for litigants who seek to come within the predicate exception thus gives sense and structure to an otherwise ambiguous provision. 15 U.S.C. 7903(5)(A)(iii). Having located the predicate exception’s limiting factor, it makes sense to read the term “applicable” broadly, encompassing statutes that are “capable of being applied” to the sale or marketing of firearms, as Cal. Civ.Code §§ 1719 and 3479-80 certainly are.
B.
My interpretation of the predicate exception is fully consistent with surrounding provisions of the PLCAA’s text. Unlike the majority, I read the PLCAA’s text as strongly supporting the view that the Act’s purpose is to protect firearms manufacturers and sellers from liability for acts solely those of third parties.
First, according to the PLCAA’s “Purposes” section, the purpose of the Act is “[t]o prohibit causes of action against manufacturers [and sellers] ... for the harm solely caused by the criminal or unlawful misuse of firearm products or ammunition products by others.” 15 U.S.C. § 7901(b)(1) (emphasis added). The majority ignores Congress’s use of the word *1159“solely,” reading this provision to suggest that Congress’s intention was “to preempt common-law claims, such as general tort theories of liability.” Maj. Op. at 1136. I disagree. The purpose articulated at § 7901(b)(1) is, by its own terms, not to reject tort theories of liability in general, but rather to prevent claims alleging strict or vicarious liability.
Indeed, the PLCAA repeatedly describes the sorts of law-suits with which Congress was concerned as lawsuits based particularly on vicarious liability theories. See 15 U.S.C. §§ 7901(a)(3) (“harm caused by the misuse of firearms by third parties”); 7901(a)(5) (“harm caused by those who criminally or unlawfully misuse firearm products or ammunition products that function as designed and intended”); 7901(a)(6) (“harm that is solely caused by others”). Reading the predicate exception to encompass only lawsuits alleging that defendants themselves knowingly engaged in unlawful conduct is fully consistent with this purpose, as such a reading would preclude the filing of suits that allege liability arising solely out of acts taken by third parties which the defendants failed to correct or prevent.
Second, the majority notes that the PLCAA elsewhere speaks of “Federal, State, and local laws” that “heavily regulate[ ]” the manufacture and sale of firearms, 15 U.S.C. § 7901(a)(4), and it concludes that Congress likely had only this narrow subset of laws (apparently, firearm-specific laws and regulations) in mind when drafting the predicate exception as well. Maj. Op. at 1136. I draw precisely the opposite conclusion from Congress’s choice of language in § 7901(a)(4). That subsection speaks of “Federal, State, and local laws” (not “statutes”) that “heavily regulate [ ]” (rather than “apply to”) firearms. Id. § 7901(a)(4) (emphases added). If Congress had intended the predicate exception to reach only those statutes specifically regulating the sale or marketing of firearms, to which it appears § 7901(a)(4) refers, surely the more straightforward way to do so would be to mirror the language of § 7901(a)(4) more closely. See Boise Cascade Corp. v. United States Envtl. Prot. Agency, 942 F.2d 1427, 1432 (9th Cir.1991) (“[W]e presume that words used more than once in the same statute have the same meaning.”). Instead, Congress chose to use the broader phrase “State or Federal statute[s] applicable to the sale or marketing of [firearms]” in the predicate exception. Id. § 7903(5)(A)(iii). If any inference can be drawn from this difference in language, it is not that § 7901(a)(4) and § 7903(5)(A)(iii) should be read as coterminous, as the majority suggests, but that they should be read differently. See Tang v. Reno, 77 F.3d 1194, 1197 (9th Cir.1996) (“[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion.”).
Third, the majority states that “Congresses] intention to create national uniformity” — an intention to which the PLCAA adverts only indirectly, in the “Purposes” section at 15 U.S.C. § 7901(b)(4) (“[t]o prevent ... unreasonable burdens on interstate ... commerce”) — is incompatible with a reading of the predicate exception that would allow Plaintiffs’ claims to survive. Maj. Op. at 1136. I disagree here as well. Reading the predicate exception to impose an actual knowledge requirement on litigants does create a nationally uniform baseline standard of liability. At the same time, this reading also accommodates another of the PLCAA’s stated purposes — “preserving] and protecting] the important principles of federalism [and] State sovereignty,” 15 U.S.C. § 7901(b)(6) — by allowing the con*1160tinued enforcement of state laws that, like California’s, reflect the considered policy-choices of the legislature. It is worth noting, in this regard, that the California Penal Code specifically states that the unlawful possession of a firearm is a nuisance. See Cal.Penal Code § 12028(b). And several years after Defendants committed the acts alleged in Plaintiffs’ complaint, the California legislature amended the negligence statute to specify that “[t]he design, distribution, or marketing of firearms and ammunition is not exempt from the duty to use ordinary care and skill that is required by this section.” Cal. Civ.Code § 1714(a); see also A.B. 496, 2002 Cal. Legis. Serv. 906 (West) (2002).
I therefore conclude that the PLCAA’s text is fully consistent with the reading of the predicate exception I have suggested: that “statute[s] applicable to the sale or marketing of [firearms]” includes any statutes capable of being applied to the sale or marketing of firearms, but that to proceed under the exception, litigants must allege that defendants “knowingly violated” those statutes. 15 U.S.C. § 7903(5)(A)(iii). Far from “ignoring] the text and purpose of[the] statute,” see Maj. Op. at 1138 n.8 (quoting Boumediene v. Bush, — U.S. -, 128 S.Ct. 2229, 2271, 171 L.Ed.2d 41 (2008)), I think this to be an entirely fair reading of the PLCAA.
C.
DeBartolo counsels that courts employing the canon of constitutional avoidance look not only to the statutory text, but also to the legislative history, to ensure that their reading of the statute is not contrary to Congress’s clear intent. DeBartolo, 485 U.S. at 583-84, 108 S.Ct. 1392; see also Catholic Bishop, 440 U.S. at 504, 99 S.Ct. 1313. Unlike the majority, I read the PLCAA’s legislative history not to foreclose the reading I suggest here, but rather in large part to support it.
Senator Sessions, one of the bill’s leading supporters, stated that “[t]his bill is incredibly narrow. It only forbids lawsuits brought against lawful manufacturers and sellers of firearms or ammunition if the suits are based on criminal or unlawful misuse of the product by a third party.” 151 Cong. Rec. S8908 (July 26, 2005) (statement of Sen. Sessions). Reading through the legislative history, it becomes clear that like Senator Sessions, the bill’s supporters broadly understood that the PLCAA would not do away with all tort liability, but rather that it would (1) limit firearms manufacturers or sellers’ tort liability to their oum conduct, not the conduct of third parties (thus imposing a narrow view of foreseeability and proximate cause), and (2) impose an extra “knowledge” requirement on state-defined duties of care.16 The bill was viewed essentially *1161as a tort-reform measure, aimed at restraining the supposed expansion of tort liability beyond its “traditional[ ]” boundaries, 151 Cong. Rec. S8910 (July 26, 2005) (statement of Sen. Sessions), particularly by “activist judge[s]” and municipalities suing the gun industry on public nuisance and strict liability theories. Id. at S8911.17 Suits for the wrongful acts of firearms manufacturers and sellers themselves were not the focus of the dismissal provision.
I recognize, of course, that individual legislators at times suggested divergent views of what sorts of lawsuits the PLCAA would affect if it were passed into law.18 Some of those views appear perhaps implausibly narrow or implausibly broad, likely because the bill excited *1162strong emotions from both its supporters and its opponents; As courts have long cautioned, however, the statements of single lawmakers do not establish congressional intent. Thompson v. Calderon, 151 F.3d 918, 928-29 (9th Cir.1998) (“[I]ndividual senators do not make laws; majorities of the House and Senate do.”); see also Chrysler Corp. v. Brown, 441 U.S. 281, 311, 99 S.Ct. 1705, 60 L.Ed.2d 208 (1979). That is why committee reports are more persuasive indicators of “the considered and collective understanding of those-Congressmen involved in drafting and studying proposed legislation.” Garcia v. United States, 469 U.S. 70, 76, 105 S.Ct. 479, 83 L.Ed.2d 472 (1984) (internal quotation marks omitted); see also Gen. Elec. Co. v. EPA, 360 F.3d 188, 193-94 (D.C.Cir.2004) (the broad reading of an ambiguous statutory preemption provision put forward in the “floor statement by a single senator” is “hardly persuasive evidence of congressional intent” in the face of committee reports that read the provision more narrowly).
Here, the House Judiciary Committee Report confirms the picture that emerges from the legislative history as a whole— that the PLCAA was intended to preclude the imposition of strict or vicarious liability on the gun industry for the criminal actions of third parties. The House Report describes the bill as aimed at combating a trend of “[r]ecent litigation ... against the firearms industry ... based on novel claims that invite courts to dramatically break from bedrock principles of tort law,” H.R. Rep. No. 109-124, at 11 (2005), and to hold firearms manufacturers and sellers “liable for the injuries caused by the criminal action of third parties.” Id. at 6.19
Reading the predicate exception as preserving causes of action for injuries caused by gun manufacturers and sellers’ own knowingly unlawful conduct is fully consistent with that view.
The majority correctly points out that both Senator Craig and Representative Stearns listed Ileto v. Glock among the lawsuits that they expected the PLCAA would preempt. See Maj. Op. at 1137-38. But there is no indication in their descriptions of the case that these lawmakers actually understood what Plaintiffs were alleging in this case: that Defendants themselves knowingly committed unlawful acts. Quite the contrary, Senator Craig and Representative Stearns’s remarks suggest that they believed Ileto was purely a vicarious liability suit.20 Whatever effect *1163these two lawmakers thought the PLCAA would have on Plaintiffs’ suit, their apparently ill-informed projections do not amount to “clear congressional intent” to enact a law that would immunize firearms manufacturers and sellers from tort liability for even their own knowing unlawful acts. DeBartolo, 485 U.S. at 574, 108 S.Ct. 1392.
D.
Applying the PLCAA’s predicate exception as written — that is, as applying to all statutes capable of being applied to the sale or marketing of firearms, but imposing an actual knowledge requirement— would prohibit a swath of lawsuits against firearms manufacturers and sellers, including those brought by municipalities for violations of no-fault or absolute liability statutes or those brought by individuals alleging vicarious liability under state tort law for the conduct of third parties of which the gun manufacturers or sellers were not aware.
It may well be that the PLCAA’s application to these other state actions would be constitutionally problematic for the same reasons outlined in Part II above. The reading of the predicate exception’s ambiguous language I have suggested might simply delay those hard constitutional questions for another case. But that is, ultimately, what the canon of constitutional avoidance is meant to do.21 The legislation or the constitutional law could change in the meantime, or no concrete case could arise in which the constitutional issue needs to be addressed. The reading I have suggested here would resolve the case in front of us, allowing Plaintiffs’ suit to go forward and leaving the constitutional issue for another day, should that day arise.
CONCLUSION
I would hold that the PLCAA does not require the dismissal of Plaintiffs’ suit, and not decide the difficult questions of constitutional law that the statute would otherwise raise. I therefore respectfully dissent.
. Stock ownership creates a property interest, even though the value of stock is uncertain until the moment it is sold. See Western Pac. R.R. Corp. v. Western Pac. R.R. Co., 197 F.2d 994, 1008 (9th Cir.1952), rev’d on other grounds, 345 U.S. 247, 73 S.Ct. 656, 97 L.Ed. 986 (1953). Similarly, a state operating license (such as a liquor license) that has been granted and can be revoked only "for cause” creates a cognizable property interest, even though the ultimate worth of that license, in terms of how much of a profit the licensee will earn in a year by operating under it, is uncertain. See Barry v. Barchi, 443 U.S. 55, 64, 99 S.Ct. 2642, 61 L.Ed.2d 365 (1979). And pending causes of action qualify as "property of the estate” in bankruptcy under 11 U.S.C. § 541(a)(1) — including causes of action sounding in tort, such as personal injury, for which the ultimate amount of recovery is uncertain. See, e.g., In re Arnold, 252 B.R. 778 (9th Cir.BAP2000); Sierra Switchboard Co. v. Westinghouse Elec. Corp., 789 F.2d 705, 707-09 (9th Cir.1986).
. Additionally, California law designates certain civil actions as "dioses in action,” which may be assigned to third parties while they are still pending. See Cal. Civ.Code Sec. 954 (tort actions for damage to personal property are assignable). Although apparently not applicable to Plaintiffs' causes of action in this case, see Pony v. County of Los Angeles, 433 F.3d 1138, 1143 (9th Cir.2006) (tort actions for personal injury are typically not assignable under California law), the "chose in action” concept further underscores the fact that pending causes of action generally have value prior to judgment.
. The majority also cites Austin v. City of Bisbee, 855 F.2d 1429 (9th Cir.1988), as supporting its constitutional holding. Austin is confusing. It states first that "[a] cause of action is a species of property protected by *1149the Fourteenth Amendment's Due Process Clause,” id. at 1435 (internal quotation marks omitted), then that a pending cause of action is "inchoate and affords no definite or enforceable property right until reduced to final judgment,” id. at 1436 (citation omitted), and finally that "[e]ven if Austin had a cognizable property right to overtime compensation, his claim fails on due process grounds.” Id. If the middle one of these statements is read as indicating that even rational basis scrutiny is inapplicable, it would be clearly at odds with Duke Power, which identifies rational basis as the appropriate level of review for such a deprivation. It would also be inconsistent with the very case Austin cites as support: In re Consolidated U.S. Atmospheric Testing Litigation, 820 F.2d 982, 988-89 (9th Cir.1987), which adopted the First Circuit's analysis in Hammond v. United States, 786 F.2d 8 (1st Cir.1986). Both Atmospheric Testing and Hammond, despite their observation that one has no "vested” interest in a pending cause of action, apply rational basis review to the deprivation of that interest. See Atmospheric Testing, 820 F.2d at 990; Hammond, 786 F.2d at 12-13. (The First Circuit recently clarified that Hammond does not stand for the proposition that a litigant has no protected interest in a cause of action prior to the entry of judgment; if it did, it would be "squarely in tension with the Supreme Court’s recognition in Logan that a cause of action is a protected property interest.” See Dr. Jose S. Belaval, Inc. v. Perez-Perdomo, 465 F.3d 33, 37 n. 4 (1st Cir.2006).) I therefore do not read the opaque line in Austin as anything other than a factual statement about the nature of a pending cause of action.
. All the cases on which the majority relies are distinguishable on one or more of these grounds.
Lyon involved a challenge to the General Aviation Revitalization Act's statute of repose, which barred any "civil action” against a manufacturer if the accident occurred more than eighteen years after the aircraft was delivered to the purchaser. See 252 F.3d at 1081. The statute did not completely extinguish litigants’ rights to sue in tort, but only limited the window in which such causes of action could be filed. The Lyon plaintiffs had not yet filed an action at the time of the statute's passage. See id.
Austin involved a challenge to an amendment of the Fair Labor Standards Act that barred recovery of unpaid overtime wages accrued prior to 1986. See 855 F.2d at 1431. The amendment did not abrogate state tort law; rather, it curtailed a statutorily-created cause of action. See id. at 1436 ("Property rights to public benefits are defined by the statutes or customs that create the benefits. When, as here, the statute authorizing the benefits is amended or repealed, the property right disappears.”) (internal quotation marks omitted).
Fields involved a challenge to Oregon’s statute of limitations for wrongful death suits and its statute of repose for medical malpractice suits. The state provisions did not extinguish individuals’ ability to sue entirely, but only narrowed the temporal window in which suits could be filed. See 413 F.3d at 956-57. Moreover, Fields noted that the plaintiffs’ "right of action for wrongful death is purely statutory and ... in Oregon there was no right of action for wrongful death at common law.” Id. at 959. In addition, Oregon’s statutes of limitation and repose were enacted long before Fields filed his suit. See id. at *1150949; see also Or.Rev.Stat. §§ 30.020(1) (statute of limitations), 12.110(4) (statute of repose).
Duke Power involved a challenge to the Price-Anderson Act, which limited federally licensed nuclear facilities' accident liability to $560 million. See 438 U.S. at 66-67, 98 S.Ct. 2620. The Act did not abrogate state tort remedies, but only imposed a federal limit on the maximum amount of recovery plaintiffs could obtain. The Act also provided for cost-sharing among nuclear operators and the mandatory waiver of defenses in case of an accident, id. which the Court held was a “reasonably just substitute for the [status quo at] common-law[.]” Id. at 88, 98 S.Ct. 2620. The plaintiffs, who lived close to planned nuclear facilities, had not filed any cause of action at the time of the Price-Anderson Act’s passage; rather, they sought a declaratory judgment that the Act was unconstitutional, premised on the prospective due process violation they would suffer should a nuclear accident occur. Id. at 67-70, 98 S.Ct. 2620.
. Kyle Rys., Inc. v. Pacific Admin. Servs., Inc., 990 F.2d 513 (9th Cir.1993), is not to the contrary. The plaintiff in Kyle had not filed a state claim for unjust enrichment at the time of ERISA’s passage, so the property interest he was asserting was an interest in bringing a lawsuit, not in maintaining a pending one. Kyle, 990 F.2d at 518-19. Additionally, although ERISA did not furnish Kyle with a replacement federal cause of action, id. at 519, it does provide certain “quid pro quos” to individuals covered by employee benefit plans. The PLCAA, in contrast, provides nothing but a "gap in the law.” Id.
. Consider, for example, the National Childhood Vaccine Injury Act, 42 U.S.C. §§ 300aa-1 et seq., which provides a no-fault compensation scheme as an alternative to tort law, see Schafer v. American Cyanamid Co., 20 F.3d 1, 1-2 (1st Cir.1994); and the Air Transportation Safety and System Stabilization Act, 49 U.S.C. §§ 40101 et seq., which creates a compensation fund and provides "an exclusive federal cause of action for ... claims [arising out of the September 11, 2001, airplane crashes] to be brought in the Southern District of New York, and adjudicated on the basis of applicable state law.” Benzman v. Whitman, 523 F.3d 119, 126 (2d Cir.2008) (internal citations omitted). But see FISA Amendments Act of 2008, § 802, 122 Stat. 2436, Pub.L. No. 110-261 (July 10, 2008) (providing that a "civil action ... in a Federal or State court" arising out of an individual or telephone company’s cooperation with the government’s counter-terrorism wiretap program "shall be promptly dismissed” upon the Attorney General’s certification). No federal court has yet considered whether the FISA Amendments Act violates due process.
. See id. at 90-92, 98 S.Ct. 2620 ("We view the congressional assurance of a $560 million fund for recovery, accompanied by an express statutory commitment ] to take whatever action is deemed necessary and appropriate to protect the public from the consequences of a nuclear accident, to be a fair and reasonable substitute for the uncertain recovery of damages of this magnitude from a utility or component manufacturer, whose resources might well be exhausted at an early stage.... Nor are we persuaded that the mandatory waiver of defenses required by the Act is of no benefit to potential claimants.... All of these considerations belie the suggestion that the Act leaves the potential victims of a nuclear disaster in a more disadvantageous position than they would be in if left to their common-law remedies — not known in modern times for either their speed or economy.") (internal citations omitted).
. Of course, as rational basis review does not require that legislation be the least restrictive means of achieving Congress's ends, the alternative remedy would not need to be an equally good substitute for the tort remedy it displaced. See Atmospheric Testing, 820 F.2d at 991 (acknowledging that "Congress could have ... [achieved its goal of protecting independent government contractors from suit] and still provided those injured by radiation a more generous substitute compensation scheme,” but concluding that "we cannot say *1153that Congresses] choice of means was without any rational basis.”) (internal quotation marks omitted).
. The majority asserts that "scores of cases concerning very similar statutes have held that the statutes do not violate substantive due process principles,” Maj. Op. at 1144, but it cites to none. As I have shown above, supra note 4, each of the cases on which the majority relies in rejecting Plaintiffs' substantive due process argument involves a statute that is distinguishable from the PLCAA in some significant respect.
. See Restatement (Third) of Torts § 3, cmt. g (“To establish the actor's negligence, it is not enough that there be a likelihood of harm; the likelihood must be foreseeable to the actor at the time of conduct. Foreseeability often relates to practical considerations concerning the actor's ability to anticipate future events or to understand dangerous conditions that already exist. In such cases, what is foreseeable concerns what the actor 'should have known.’ ”).
. I note, in addition, that Congress’s use of the word "violation” does not necessarily suggest a distinction between common-law-based duties and legislatively-imposed duties. Cf. Riegel v. Medtronic, Inc., - U.S. -, 128 S.Ct. 999, 1008, 169 L.Ed.2d 892 (2008) ("Congress is entitled to know what meaning this Court will assign to terms regularly used in its enactments. Absent other indication, reference to a State's [']requirements[’] includes its common-law duties.... [CJommon-Iaw liability is premised on the existence of a legal duty, and a tort judgment therefore establishes that the defendant has violated a state-law obligation.") (emphasis added); Medtronic, Inc. v. Lohr, 518 U.S. 470, 486, 495, 116 S.Ct. 2240, 135 L.Ed.2d 700 (1996) (plurality opinion) (holding that § 360k of the Medical Device Amendments, which preempts "any requirement (1) which is different from, or in addition to, any requirement applicable under this chapter to the device,” does not "den[y] Florida the right to provide a traditional damages remedy for violations of common-law duties when those duties parallel federal requirements.”) (emphasis added).
. Further, Plaintiffs’ complaint alleges that "the easy availability of firearms for criminal purposes is a direct, known result of [Defendants’ marketing and distribution policies and practices.” FAC ¶ 58 (emphasis added). Although the Bureau of Alcohol, Tobacco and Firearms has "reported ... that ... about 1% [of dealers nationwide] account for over half of the successfully traced guns used in crime,” id. ¶ 49, and although ATF regularly forwards Defendants crime-trace data on particular distributors and dealers, Defendants "choose not to use the data ... to change their marketing and distribution practices to reduce the foreseeable risk that their firearms will become possessed by prohibited persons.” Id. ¶ 66 (emphasis added). Further, Plaintiffs allege, Defendant manufacturers choose not to train their dealers, id. ¶ 77, or to cut off contracts with distributors who sell to dealers with disproportionately high sale-to-crime rates. Id. ¶ 72. Defendants also allegedly market their products to appeal to prospective purchasers with criminal intent, emphasizing characteristics such as easy concealability and rapid fire capability. Id. ¶ 81-88.
. In this respect, the case law concerning suits against municipalities under 42 U.S.C. § 1983 may provide a useful analogy, as it throws the distinction between direct liability and respondeat superior — a type of vicarious liability — into sharper relief. Under Monell v. Dep't of Soc. Servs., 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611 (1978), "a municipality cannot be held liable solely because it employs a tortfeasor — or, in other words, a municipality cannot be held liable under § 1983 oh a respondeat superior theory.” Id. at 691, 98 S.Ct. 2018. But, Monell held, a municipality can be held directly liable for a violation of the Constitution or a federal law under § 1983 if its own "policy or custom ... inflicts the injury[.]” Id. at 694, 98 S.Ct. 2018. Later cases have clarified that a municipality will be liable for a policy of inadequate training or supervision of police officers only if the policy "reflects a ‘deliberate’ or ‘conscious’ choice by a municipality” not to avoid the risk of harm. City of Canton, Ohio v. Harris, 489 U.S. 378, 389, 109 S.Ct. 1197, 103 L.Ed.2d 412 (1989).
Suppose, for example, that a municipality maintains a policy of hiring police officers without running criminal background checks on them or providing firearms training, and that it knows several of its officers have committed violent crimes in the course of duty in the past. If a police officer hired and retained under this policy then shoots and kills a group of innocent bystanders, the municipality would not be liable to the victims for the shooting on a respondeat superior theory, but it would be directly liable for its own wrongful act: maintaining a policy of inadequate screening and supervision with "deliberate indifference" to the risk of harm that policy created. See Bd. of County Commissioners of Bryan County, Oklahoma v. Brown, 520 U.S. 397, 117 S.Ct. 1382, 137 L.Ed.2d 626 (1997). The violations are distinct, even though the same act — the shooting — occasions the victim’s suit.
. I note, in this respect, that California courts do recognize "an aggravated form of negligence,” sometimes called "willful misconduct,” for which the “pleading requirements are similar to negligence but stricter”: A plaintiff must show "(1) actual or constructive knowledge of the peril to be apprehended, (2) actual or constructive knowledge that injury is a probable, as opposed to a possible, result of the danger, and (3) conscious failure to act to avoid the peril.” Berkley v. Dowds, 152 Cal.App.4th 518, 526-28, 61 Cal.Rptr.3d 304 (2007) (internal alterations, quotation marks, and citations omitted). The PLCAA, on my reading, would require something slightly more onerous: a showing of actual, not constructive, knowledge of all the elements that establish a violation of law.
. When Congress is acting in an area over which it has constitutionally delegated authority, there is nothing particularly unusual about a federal statute adding to the proof burdens that would be applicable under state law. See, e.g., Gorman v. Wolpoff & Abramson, LLP, 552 F.3d 1008, 1025-27 (9th Cir.2009) (holding that the Fair Credit Reporting Act permits state defamation claims to avoid preemption, if at all, only if plaintiffs plead and prove "malice or willful intent,” a mens rea not inherent in most state defamation actions); 800 Adept, Inc. v. Murex Securities, Ltd., 539 F.3d 1354, 1369 (Fed.Cir.2008) ("State tort claims against a patent holder, including tortious interference claims, based on enforcing a patent in the marketplace, are preempted’ by federal patent laws, unless the claimant can show that the patent holder acted in 'bad faith’ in the publication or enforcement of its patent.”).
. See, e.g., 151 Cong. Rec. S9088 (July 27, 2005) (statement of Sen. Craig) ("[This bill] does not prevent [gun manufacturers and sellers] from being sued for their own misconduct. This bill only stops one extremely narrow category of lawsuits[:] lawsuits that attempt to force the gun industry to pay for the crimes of third parties over whom they have no control. We have tried to make that limitation as clear as we possibly can....”); id. (statement of Sen. Craig) ("This bill responds to a series of lawsuits filed primarily by municipalities to shift the financial burden for criminal violence onto the law-abiding business community. These suits are based on a variety of legal theories .... seeking to hold gun manufacturers and sellers liable for the cost of injuries caused by people over whom they have no control — criminals who choose to use firearms illegally.”); id. at S9089 (statement of Sen. Craig) ("This is not a gun industry immunity bill. It prohibits one kind of lawsuit[:] a suit trying to fix the blame of a third party’s criminal acts or misdeeds on the manufacturer or the seller of the firearm used in that crime.”); 151 Cong. Rec. S8908-11 (July 26, 2005) (statement of Sen. Sessions) (“Manufacturers and sellers are still responsible for their own negligent or criminal conduct ....”); id. ("It is simply wrong ... to allow those manufacturers who comply with the many mies we have set forth *1161... to be sued for intervening criminal acts .... [But][i]f they knew, if they had reason to know, if they were negligent in going through the requirements of the law or failed to do the requirements of the law, they can [still] be sued [despite the PLCAA]”); id. at S8911 (statement of Sen. Sessions) ("Plaintiffs can go to court if the gun dealers do not follow the law, if they negligently sell the gun, if they produce a product that is improper or they sell to someone they know should not be sold to or did not follow steps to determine whether the individual was [eligible] to buy[ ] a gun.”); 151 Cong. Rec. S9226 (July 28, 2005) (statement of Sen. Graham) ("What ... [this bill will prohibit are suits that seek] under a gross negligence or simple negligence standard [to] create a duty on the part of sellers and manufacturers for an event that they can't control which is the intentional misuse of a weapon to commit a crime....”).
. See 151 Cong. Rec. S9088 (July 27, 2005) (statement of Sen. Craig) ("This bill responds to a series of lawsuits filed primarily by municipalities to shift the financial burden for criminal violence onto the law-abiding business community.”); id. at S9088-89 (statement of Sen. Craig) (characterizing the "junk lawsuits” that the PLCAA would prohibit as threatening to "reverse a longstanding legal principle in this country ... that manufacturers of products are not responsible for the criminal misuse of those products”); 151 Cong. Rec. S9378 (statement of Sen. Sessions) ("We ha[ve] a group of activist, anti-gun litigators who sometimes buddy up with a city or mayor somewhere — usually a big city — and try to conjure up some way to make a legitimate manufacturer of a firearm liable for intervening acts of criminals and murderers. That has never been the principle of American law, but it is a reality that is occurring today and it threatens an industry that supplies our military with weapons.”).
Relatively little of the debates focused on tort suits brought by injured individuals, as opposed to municipalities. But see 151 Cong. Rec. S9386 (July 29, 2005) (statement of Sen. Reed) (advocating an amendment, which ultimately failed, that would have "preserve[d] the right of an individual to sue for negligence when they have been harmed and when that negligence can fairly be attributed to a gun manufacturer ... [or] dealer____[I]f we are confronted with this legislation, I propose we step back and perhaps reluctantly eliminate suits by municipalities, but for goodness sakes, we can have and maintain suits by individuals.... At a minimum, we have to allow the tort law of the various States ... to be operative ____”); id. at S9389 (July 29, 2005) (statement of Sen. Allen) (supporting the bill as written, and stating that "[t]his legislation does carefully preserve the right of individuals to have their day in court with civil liability actions for injury or danger caused by negligence on the firearms dealer or manufactureras part] or defective product. ... ”).
. Compare 151 Cong. Rec. S9226 (July 28, 2005) (statement of Sen. Kyi) (supporting the bill and opposing an amendment that would have expressly permitted suits alleging gross negligence or recklessness to go forward; arguing that the bill as written already would allow suits alleging "gross negligence or reckless conduct ... [as] the proximate cause of death or injury”), and 151 Cong. Rec. S9926 (July 25, 2005) (statement of Sen. Graham) (same), with 151 Cong. Rec. S9385 (July 29, 2005) (statement of Sen. Schumer) (opposing the bill as written, and arguing that "[e]ven when somebody is grossly negligent ... they will” be immune from suit), and id. at S9380 (statement of Sen. Kennedy) (same).
. Like the floor debates, much of the House Report focuses on the perceived need to curtail lawsuits brought by municipalities, rather than those brought by individuals. See id. at 18 (describing municipal plaintiffs as attempting to "regulate firearms whereas only the State had the power to regulate in this area”); id. at 13 ("The various public entities that have brought suit against the gun industry in recent years have raised novel claims that seek reimbursement of government expenses — including costs for police protection, emergency and medical services, and pension benefits — associated with gun-related crimes.”). (No Senate Report was published.)
. See 151 Cong. Rec. S9394 (July 29, 2005) (statement of Sen. Craig) ("Another example of a lawsuit captured by this bill is the case of Ileto v. Glock .... The United States Ninth Circuit Court of Appeals said Glock and RSR could be sued for a criminal shooting when Glock sold the pistol to a Washington State police department and the distributor RSR never owned, nor sold, nor possessed the firearm.”); 151 Cong. Rec. E2163 (Oct. 25, 2005) (statement of Rep. Stearns) (extensions of remarks) ("Another example is the case of Ileto v. Glock .... The facts, if you can believe it, are that the manufacturer, Glock, sold the pistol later criminally misused, to a Washington State police department and the distributor being sued never owned, sold, nor possessed the firearm that was criminally misused.").
. The majority asserts that the “point” of the constitutional avoidance canon "is to adopt an alternative interpretation of the statute that avoids any constitutional problem.” Maj. Op. at 1143 n.12 (emphasis added). But surely, that is not so. The point, rather, is avoid constitutional questions actually raised by a given case. Whatever constitutional problems might be raised in other cases — or even in this one at a later stage in the litigation, if Plaintiffs were unable to prove the knowing conduct they have pleaded and could show only negligence — are not before us today, and cannot guide our choice between two plausible readings of the statute when one would raise a serious constitutional question in this case.