OPINION OF THE COURT

VANASKIE, Circuit Judge.

These asbestos cases involve the availability of the "bare-metal defense" under maritime law. The defense's basic idea is that a manufacturer who delivers a product "bare metal" — that is without the insulation or other material that must be added for the product's proper operation — is not generally liable for injuries caused by asbestos in later-added materials. A classic scenario would be if an engine manufacturer ships an engine without a gasket, the buyer adds a gasket containing asbestos, and the asbestos causes injury to a worker. May the manufacturer be held liable? Some courts say no — never. Others rely on a more fact-specific standard and ask whether the facts of the case made it foreseeable that hazardous asbestos materials would be used. Neither this Court nor the Supreme Court has confronted the issue.

In that void, we survey bedrock principles of maritime law and conclude that they permit a manufacturer of even a bare-metal product to be held liable for asbestos-related injuries when circumstances indicate the injury was a reasonably foreseeable result of the manufacturer's actions — at least in the context of a negligence claim. The District Court had instead applied the bright line rule approach and entered summary judgment against the plaintiffs. We will vacate the entry of summary judgment on the plaintiffs' negligence claims, affirm the entry of summary judgment on the plaintiffs' product liability claims (which we conclude were abandoned on appeal), and will remand, for further proceedings.

I.

Appellants Roberta G. Devries and Shirley McAfee are the widows of deceased husbands who served in the United States Navy. Each couple filed a Complaint in Pennsylvania state court alleging that the husband contracted cancer caused by exposure to asbestos. Devries alleges that on the U.S.S. Turner from 1957-60, her husband was exposed to asbestos-containing insulation and components that were added onto the ship's engines, pumps, boilers, blowers, generators, switchboards, steam traps, and other devices. McAfee alleges her husband was similarly exposed through his service on two ships and in the Philadelphia Naval Shipyard.

Devries and McAfee named a number of defendants, of which Appellee manufacturers ("Manufacturers") are a subset.1 The Manufacturers each made their products "bare metal," in that if they manufactured an engine, they shipped it without any asbestos-containing insulation materials that would later be added.

Devries and McAfee's Complaints each allege claims of negligence and strict liability. The Manufacturers removed to the Eastern District of Pennsylvania and invoked the bare-metal defense in support of their respective summary judgment motions, arguing that because they shipped their products bare metal, they could not be held liable for the sailors' injuries. The District Court agreed and granted the Manufacturers summary judgment motions.

Devries and McAfee each appealed separately, raising an issue as to whether the District Court's decision addressed their negligence claims. We summarily remanded with instructions that the District Court address the negligence issue and also consider a split in authority as to whether a bright-line rule or a fact-specific standard governed the bare-metal defense's availability. In re Asbestos Prods. Liab. Litig., No. 15-2667, Order (3d Cir. May 12, 2017) (McAfee); In re Asbestos Prods. Liab. Litig., No. 15-1278, Order (3d. Cir. Feb. 5, 2016) (Devries).

On remand, the District Court applied the bright-line-rule version of the bare-metal defense, and clarified that summary judgment had been entered in favor of the Manufacturers on both the strict liability and negligence claims. The Court reasoned that the rule approach was best because, according to the Court's view of the precedents, maritime law favors uniformity and the rule approach was the majority view.

Devries and McAfee appealed for a second time. We consolidated their appeals and ordered coordinated briefing.

II.

The District Court had federal-officer jurisdiction under 28 U.S.C. § 1442(a)(1), and maritime jurisdiction under 28 U.S.C. § 1333(1). We have jurisdiction under 28 U.S.C. § 1291. We review the District Court's grant of summary judgment de novo. Faush v. Tues. Morning, Inc., 808 F.3d 208, 215 (3d Cir. 2015).

III.

The key question in this case is the bare-metal defense's availability: When, if ever, should a manufacturer of a product that does not contain asbestos be held liable for an asbestos-related injury most directly caused by parts added on to the manufacturer's product? Neither the Third Circuit nor the Supreme Court has addressed the question, and the courts from other jurisdictions that have are split. Some courts apply a bright-line rule, holding that a manufacturer of a bare-metal product is never liable for injuries caused by later-added asbestos-containing materials. See, e.g., Lindstrom v. A-C Prod. Liab. Tr., 424 F.3d 488, 492, 494-97 (6th Cir. 2005); Cabasug v. Crane Co., 989 F.Supp.2d 1027, 1038-43 (D. Haw. 2013). Others apply a more fact-specific standard, stating, for example, that a bare-metal manufacturer may be held liable if the plaintiff's injury was a reasonably foreseeable result of the manufacturer's conduct. See, e.g., Quirin v. Lorillard Tobacco Co., 17 F.Supp.3d 760, 768-70 (N.D. Ill. 2014) (determining whether the addition of asbestos material was "foreseeable" by asking whether addition of asbestos-containing materials was "inevitable," and whether those added materials were "necess[ary]" or "essential" to the manufacturer's product); Chicano v. Gen. Elec. Co., 2004 WL 2250990, at *6 (E.D. Pa. Oct. 5, 2004) (asking if the addition of asbestos-containing materials was "foreseeable").2

In addressing this question, we (1) examine the doctrinal roots of the bare-metal defense, and (2) address how it should be applied in Devries and McAfee's negligence actions.

A.

The doctrinal root of the bare-metal defense has proved to be a particularly vexing question. Some courts have rooted the defense in causation: When if ever can it be said that a bare-metal manufacturer causes an asbestos-related injury? See, e.g., Thurmon v. Ga. Pac., LLC, 650 Fed. Appx. 752, 756 (11th Cir. 2016) ("the `bare metal defense' is, essentially, a causation argument"). Others locate the defense in duty: Can a manufacturer's duty to act with reasonable care with respect to reasonably foreseeable risks and plaintiffs, be said to extend to asbestos-related injuries? See, e.g., Quirin, 17 F.Supp.3d at 767-70 (reviewing the issue as one of "legal duty"). The question is more than academic. If the elemental root is duty, the defense should be expected to operate differently in strict liability as compared to negligence, because a defendant's duty of course differs between the two types of actions. See Chesher v. 3M Co., 234 F.Supp.3d 693, 700-03 (D.S.C. 2017) (holding that the defense should apply in a weaker fashion in a negligence action, as compared to strict liability). The opposite might be true too — the defense should operate in similar fashion in both negligence and strict liability if it is rooted in causation, because the proximate cause inquiry cuts across the two types of actions. See, e.g., Lindstrom, 424 F.3d at 492 (suggesting the defense applies similarly under "both negligence and strict liability theories").

We find that both approaches are correct: the defense is rooted in both duty and cause because its keystone is the concept of foreseeability. When parties debate the bare-metal defense, they debate when and whether a manufacturer could reasonably foresee that its actions or omissions would cause the plaintiff's asbestos-related injuries. The bright-line rule approach says it is never reasonably foreseeable, and the fact-specific standard approach says it sometimes is. This debate over foreseeability sounds in both duty and cause, because foreseeability is a concept embedded in each element. See Gibbs v. Ernst, 538 Pa. 193, 647 A.2d 882, 891 (1994) (highlighting "the common law notion of foreseeability as found in the concepts of duty and proximate cause"). In the duty element in a negligence action, foreseeability limits a defendant's liability to only the risks and plaintiffs that are reasonably foreseeable. See Restatement (Third) of Torts: Phys. & Emot. Harm § 7, cmt. j (2010 Am. Law Inst.) (acknowledging "widespread use" of foreseeability as an aspect of the duty of reasonable care, despite the Restatement's disagreement with such an approach). And in proximate cause, foreseeability limits a defendant's liability to only the injuries that are a reasonably foreseeable result of the defendant's actions. Id. § 29, cmt. j (discussing foreseeability as an aspect of proximate cause in both negligence and strict-liability actions).3 Thus, the bare-metal defense is nothing more than the concept of foreseeability, as embedded in the duty of reasonable care in a negligence action and the proximate cause standard in a negligence or strict-liability action, as applied to the facts of a certain subset of asbestos cases.

This dual-elemental home for the defense does not, however, totally explain when or whether the defense's application should differ from strict-liability to negligence. It might be that the defense could apply the same in both types of actions, because of the shared proximate-cause element. Or the differences in the two actions' duty elements might mean the defense is more forceful in one action than the other. See, e.g., Chesher, 234 F.Supp.3d at 700-03 (holding that the defense is weaker in negligence and stronger in strict liability, because in strict liability the manufacturer's duty is limited to the product, but with negligence the duty extends further); Bell v. Foster Wheeler Energy Corp., No. 15-6394, 2016 WL 5780104, at *5-7 (E.D. La. Oct. 4, 2016) (same). And of course the facts of a given case could be the most important variable.

We need not settle these doctrinal distinctions today, because Devries and McAfee waived their strict liability claim in this appeal. As a general matter, an appellant waives an argument in support of reversal if it is not raised in the opening brief. McCray v. Fidelity Nat'l Title Ins. Co., 682 F.3d 229, 241 (3d Cir. 2012). Here, in this appeal Devries and McAfee focused the entirety of their briefing on their negligence claims, yet attempted to also incorporate their strict-liability claim through a footnote: "By concentrating on [negligence] issues in this brief, Appellants do not waive any issues argued in their original briefs as to Defendants' liability under [the strict liability claims]." (Appellants' Br. at 2 n. 1). This attempt to shoehorn in an argument outside the briefs is insufficient to raise an issue on appeal. See John Wyeth & Brother Ltd. v. CIGNA Int'l Corp., 119 F.3d 1070, 1076 n. 6 (3d Cir. 1997) (stating that "arguments raised in passing (such as, in a footnote), but not squarely argued, are considered waived"); see also Skretvedt v. E.I. DuPont De Nemours, 372 F.3d 193, 202-04 (3d Cir. 2004) (declining to consider arguments not properly raised and therefore waived). In particular, it fails to give fair notice of the claims being contested on appeal. Thus, Devries's and McAfee's waiver of their strict-liability arguments means that we will affirm the District Court's decision to that extent, and need not fully explore the precise contours of the defense's distinctions in strict liability and negligence, beyond the unifying principle of foreseeability.

B.

For the negligence claims, rooting the bare-metal defense in foreseeability does not on its own resolve the issue, because the split in authority can be characterized as a debate over what a bare-metal manufacturer could reasonably foresee — no asbestos-related injuries, see, e.g., Lindstrom, 424 F.3d at 492, 494-97, or some, see, e.g., Quirin, 17 F.Supp.3d at 769-70.

These two choices raise familiar tradeoffs between rules and standards.4 A rule is a legal directive that attempts to capture a background principle into an easy-to-apply form that is predictable and efficient. A speed limit is a good example: its goal is road safety, but because liability turns on speed rather than the amorphous definition of "safety" itself, it is easier for drivers, police, and insurers to shape their conduct accordingly. Rules have downsides though too, in that they necessarily result in errors of over- and under-inclusion. In the case of the speed limit, it furthers the policy of road safety, but does so imperfectly: speedy drivers get punished even if they speed safely, and slow drivers go free even if they amble along haphazardly.

A standard, on the other hand, collapses the background principle into the actual legal directive, resulting in better accuracy and "fit" with the underlying purpose, and fewer errors of over- and under-inclusion. Another road-safety example would be a reckless-driving prohibition that simply prohibits driving that is "reckless." Such a prohibition is less predictable and efficient than the speed limit, in that it is harder to predict what a decisionmaker will find to be "reckless" than whether he or she will agree that 76 miles per hour exceeds a 70 m.p.h. speed limit. But liability better tracks the actual goal of road safety, because almost all "reckless" drivers are unsafe.

The point is there are tradeoffs, and courts face those tradeoffs in choosing an approach to the bare-metal defense. The rule-based approach is efficient and predictable — bare-metal manufacturers are simply not liable — but the downside is some deserving sailor-plaintiffs will not receive their due. On the other hand, the standard-based approach is bound to be less predictable and less efficient, because the standard's fact-centered nature will push more cases into discovery, see, e.g., Quirin, 17 F.Supp.3d at 771-72 (denying defendant's motion to dismiss after applying the standard), but the most-deserving sailor-plaintiffs are less likely to be denied compensation.

Thankfully, we do not weigh these tradeoffs in a vacuum. Maritime law is undergirded by established principles, at least four of which are implicated here. First and perhaps foremost, maritime law is deeply concerned with the protection of sailors, due to a historic and "special solicitude for the welfare of those men who undertook to venture upon hazardous and unpredictable sea voyages." Moragne v. States Marine Lines, Inc., 398 U.S. 375, 387, 90 S.Ct. 1772, 26 L.Ed.2d 339 (1970). This "special solicitude" developed "unknown to the common law," and so maritime law is at times more lenient toward a sailor than a state's common law may be to a similarly-situated plaintiff. Id. This divergence is acceptable if not appropriate because the "humane and liberal character of" maritime law counsels that it is better "to give than to withhold the remedy" wherever "established and inflexible rules" do not require otherwise. Id. (quoting The Sea Gull, 21 F. Cas. 909, 910 (C.C. Md. 1865)). For example, in Moragne v. States Marine Lines, the Supreme Court made it permissible for maritime plaintiffs to bring wrongful death actions even though the common law disapproved of such actions. 398 U.S. at 381-88, 408-09, 90 S.Ct. 1772. In arriving at that holding, the Court explicitly referenced and discussed maritime law's special solicitude for sailor safety and how that solicitude permitted maritime law to have more sailor-friendly rules than the common law. Id. at 386-88, 90 S.Ct. 1772.5

Here, maritime law's special solicitude for sailors' safety similarly favors the adoption of the standard-like approach to the bare-metal defense. A standard will permit a greater number of deserving sailors to receive compensation, and compensation that is closer to what they deserve. Given that results for sailor-victims will differ under a rule as compared to a standard, and since no "established" or "inflexible" rule prohibits the more forgiving standard, the "humane and liberal character" of maritime law counsels that we follow the standard. Even if certain states' common laws would call for a more stringent rule, maritime law's more liberal attitude permits us to diverge from that path.

Second, maritime law is built on "traditions of simplicity and practicality," Kermarec v. Compagnie Generale Transatlantique, 358 U.S. 625, 631, 79 S.Ct. 406, 3 L.Ed.2d 550 (1959), but that principle cuts in both directions and does not provide much guidance. On one hand, "simplicity" might be seen as favoring the rule-based approach, because simplicity is related to predictability, and it is easier to predict how a rule will apply than a standard. On the other hand, "simplicity" could also be seen as favoring a foreseeability-based standard, because simplicity is related to familiarity, and foreseeability is such a familiar and key part of tort law. See id. at 631-32, 79 S.Ct. 406 (choosing to adopt a familiar standard over a "foreign" and "alien" rule while invoking maritime law's "traditions of simplicity and practicality").

The third and fourth principles implicated in this case are also not particularly helpful. Maritime law has a "fundamental interest" in "the protection of maritime commerce," Exxon Corp. v. Cent. Gulf Lines, Inc., 500 U.S. 603, 608, 111 S.Ct. 2071, 114 L.Ed.2d 649 (1991) (quoting Sisson v. Ruby, 497 U.S. 358, 367, 110 S.Ct. 2892, 111 L.Ed.2d 292 (1990)), and seeks out "uniform rules to govern conduct and liability," Foremost Ins. Co. v. Richardson, 457 U.S. 668, 674-75, 102 S.Ct. 2654, 73 L.Ed.2d 300 (1982). Here, the parties all argue these two principles encourage the Court to side with whatever side is winning in the split in authority. The idea is that the sooner one side wins out over the other, the sooner the split in authority is ended and the goals of seamless commerce and uniformity of rules will be achieved. The rub, however, is determining which view is the majority. The bright-line rule could be said to be in the lead because it has on its side the Sixth Circuit, the only court of appeals to weigh in. Lindstrom, 424 F.3d at 492, 494-97. The standard could similarly be said to be the majority view because the courts that have confronted the question most recently have generally favored the standard, and have done so after a much more thorough analysis than that found in the Sixth Circuit's opinion in Lindstrom, which was decided much earlier in the debate over the bare-metal defense. Compare Chesher, 234 F.Supp.3d at 696-712 (analyzing in painstaking detail the split in authority and adopting a version of the standard); Bell, 2016 WL 5780104, at *3-7 (same), with Lindstrom, 424 F.3d at 494-97 (not mentioning the split in authority). We need not decide which approach is winning in terms of wins and losses — it is enough that the score is too close for us to say that the goals of seamless commerce and rule-uniformity push in one way or the other.

In sum, the special solicitude for the safety and protection of sailors is dispositive, because it counsels us to follow the standard-based approach, and none of the other principles weigh heavily in either direction. The standard-based approach is the one we will therefore follow: foreseeability is the touchstone of the bare-metal defense; a manufacturer of a bare-metal product may be held liable for a plaintiff's injuries suffered from later-added asbestos-containing materials if the facts show the plaintiff's injuries were a reasonably foreseeable result of the manufacturer's failure to provide a reasonable and adequate warning; and although cases will necessarily be fact-specific, already-decided precedents show, for example, that a bare-metal manufacturer may be subject to liability if it reasonably could have known, at the time it placed its product into the stream of commerce, that

These may or may not be the only facts on which liability can arise. The finer contours of the defense, and how it should be applied to various sets of facts, must be decided on a case-by-case basis.

IV.

Finally, the Manufacturers advanced two alternative arguments in support of an affirmance on the negligence claims. They argued (1) insufficient evidence had been presented as to causation and was fatal to Devries and McAfee's claims, and (2) the government-contractor defense should insulate the Manufacturers from liability. These arguments were also presented below, but the District Court declined to rule on them because its bare-metal-defense holding was sufficient to enter summary judgment in favor of the Manufacturers. The Manufacturers urge us to address them now, on the grounds that we may affirm a judgment for any reason supported by the record. Brightwell v. Lehman, 637 F.3d 187, 191 (3d Cir. 2011). Addressing alternative grounds for affirmance, however, is a matter left to our discretion. See Gov't of the V.I. v. Walker, 261 F.3d 370, 376-77 (3d Cir. 2001) (declining to reach arguments raised before but not decided by the lower court, and instead remanding). Given that we are without the benefit of the District Court's well-regarded expertise, and the parties' briefing and oral argument was appropriately focused on the bare-metal defense, we will leave the insufficient-evidence and contractor-defense arguments to be dealt with on remand.

V.

In conclusion, maritime law's special solicitude for the safety and protection of sailors counsels us to adopt a standard-based approach to the bare-metal defense that permits a plaintiff to recover, at least in negligence, from a manufacturer of a bare-metal product when the facts show the plaintiff's injuries were a reasonably foreseeable result of the manufacturer's conduct. We will affirm the decision of the District Court with respect to Devries and McAfee's strict liability claims, and remand for further proceedings on their negligence claims consistent with this Opinion.