RICHARDSON, Circuit Judge:

When do constitutional rights vest? At 18 or 21? 16 or 25? Why not 13 or 33? In the law, a line must sometimes be drawn. But there must be a reason why constitutional rights cannot be enjoyed until a certain age. Our nation’s most cherished constitutional rights vest no later than 18. And the Second Amendment’s right to keep and bear arms is no different.

Plaintiffs seek an injunction and a declaratory judgment that several federal laws and regulations that prevent federally licensed gun dealers from selling handguns to any

18-, 19-, or 20-year-old violate the Second Amendment. We first find that 18-year-olds

possess Second Amendment rights. They enjoy almost every other constitutional right, and they were required at the time of the Founding to serve in the militia and furnish their own weapons. We then ask, as our precedent requires, whether the government has met its burden to justify its infringement of those rights under the appropriate level of scrutiny.

To justify this restriction, Congress used disproportionate crime rates to craft over- inclusive laws that restrict the rights of overwhelmingly law-abiding citizens. And in doing so, Congress focused on purchases from licensed dealers without establishing those dealers

as the source of the guns 18- to 20-year-olds use to commit crimes. So we hold that the challenged federal laws and regulations are unconstitutional under the Second

Amendment. Despite the weighty interest in reducing crime and violence, we refuse to relegate either the Second Amendment or 18- to 20-year-olds to a second-class status.

I. Background

Prospective handgun buyers sued the Bureau of Alcohol, Tobacco, Firearms and Explosives seeking an injunction and a declaratory judgment that federal statutes prohibiting Federal Firearm Licensed Dealers from selling handguns and handgun ammunition to 18-, 19-, and 20-year-olds (and the federal regulations implementing those statutes) violate the Second Amendment. See 18 U.S.C. § 922(b)(1), (c)(1); 27 C.F.R.

§§ 478.99(b)(1), 478.96(b), 478.124(a). 1

Nineteen-year-old Natalia Marshall had good reason to seek protection. She had

been forced to obtain a protective order against her abusive ex-boyfriend who, since the issuance of the order, had been arrested for unlawful possession of a firearm and controlled substances. He was released on bail but never came to court, leading to the issuance of a capias for his arrest. Along with the threat from her ex-boyfriend, Marshall works as an equestrian trainer, often finding herself in remote rural areas where she interacts with unfamiliar people. Having grown up training with guns, she believes that a handgun’s ease of carrying, training, and use makes it the most effective tool for her protection from these and other risks. But because Marshall was 18 when she tried to buy a handgun, a federal law prevented her from buying from a licensed dealer who would perform a background check to verify that she was not a felon or other prohibited person. She preferred using a licensed dealer because they tend to have a wider supply, a good reputation, and a guarantee

Plaintiffs also contend that the laws violate the equal-protection component of the that the guns have not been used, stolen, or tampered with. She is now 19 and remains unable to buy a handgun from a federally licensed dealer for self-defense.

The other Plaintiff, Tanner Hirschfeld, tried to buy a handgun from a licensed dealer and was denied because he was only 20. But he has since turned 21 and is no longer

affected by these laws. His claims are therefore moot. See Craig v. Boren, 429 U.S. 190, 192 (1976). Marshall, however, is not yet 21, so this appeal may still proceed.

A. The challenged laws

In 1964, Congress, concerned about increasing gun violence, began a “field

investigation and public hearings.” S. Rep. No. 88-1340, at 1 (1964). Congress concluded, among other things, “that the ease with which” handguns could be acquired by “juveniles

without the knowledge or consent of their parents or guardians . . . is a significant factor in the prevalence of lawlessness and violent crime in the United States.” Omnibus Crime

Control and Safe Streets Act of 1968, Pub. L. No. 90-351, tit. IV, § 901(a)(2), 82 Stat. 197, 225; see, e.g., S. Rep. No. 90-1097, at 77 (1968). Statistics, testimony from law enforcement, and other evidence showed that a large portion of violent crime and its escalation stemmed from minors with guns. ^[2] Based on the testimony from Internal Revenue Service Commissioner Sheldon Cohen, Congress concluded that juveniles obtained these guns from licensed dealers who sell to “emotionally immature, or thrill-bent

juveniles and minors prone to criminal behavior.” Pub. L. No. 90-351, tit. IV, § 901(a)(6), 82 Stat. at 226; see Federal Firearms Act: Hearings Before the H. Comm. on Ways and Means, 89th Cong. 31 (1965) (testimony of Sheldon S. Cohen) (explaining that “importers, manufacturers, and dealers who operate under licenses issued by the Federal Government”

are responsible for putting firearms in the hands of juveniles). They also found that “the handgun is the type of firearm that is principally used in the commission of serious crime” and “the most troublesome and difficult factor in the unlawful use of firearms.” S. Rep.

No. 89-1866, at 4–7 (1966). Seeking only to curtail the “clandestine” purchase of weapons by juveniles without their parents’ knowledge, Congress sought to permit the ownership and use of firearms with parental permission. Id. at 58–59; S. Rep. No. 90-1097, at 79;

114 Cong. Rec. ^[12],279, 12,309 (1968) (statement of Sen. Thomas J. Dodd).

To address these concerns, Congress passed the Omnibus Crime Control and Safe

Streets Act of 1968, Pub. L. No. 90-351, 82 Stat. 197. The preamble declares:

[T]he ease with which any person can acquire firearms other than a rifle or shotgun (including criminals, juveniles without the knowledge or consent of their parents or

testified that “[t]he greatest growth of crime today is in the area of young people, juveniles and young adults” and that “[t]he easy availability of weapons makes their tendency toward wild, and sometimes irrational behavior that much more violent, that much more deadly.” Federal Firearms Act: Hearings Before the Subcomm. to Investigate Juvenile Delinquency of the S. Comm. on the Judiciary, 90th Cong. ^[57] (1967) (testimony of Sheldon S. Cohen). Law enforcement officers from New York City, Los Angeles, St. Louis, Chicago, Philadelphia, and Atlanta provided Congress with “statistics documenting the misuse of firearms by juveniles and minors,” which took on “added significance” because the “lawful acquisition of concealable firearms by these persons was prohibited by statute” in each of those cities. S. Rep. No. 89-1866, at 58–60.

guardians, narcotics addicts, mental defectives, armed groups who would supplant the functions of duly constituted public authorities, and others whose possession of such weapons is similarly contrary to the public interest) is a significant factor in the prevalence of lawlessness and violent crime in the United States.

Id. § 901(a)(2), 82 Stat. at 225.

The Act prohibited licensed dealers from selling handguns to anyone under age 21 but permitted the sale of shotguns and rifles to those individuals. Id. § 902, 82 Stat. at 230.

Later that year, Congress amended the law to prohibit licensed dealers from selling any

firearm to those under 18 and maintained the ban on the sale of handguns for 18-, 19-, and 20-year-olds. Gun Control Act of 1968, Pub. L. No. 90-618, tit. I, § 102, 82 Stat. 1213, 1218. 3 The section, which has remained unchanged since 1968, reads:

[It is] unlawful for any licensed importer, licensed manufacturer, licensed dealer, or licensed collector to sell or deliver any firearm or ammunition to any individual who the licensee knows or has reasonable cause to believe is less than eighteen years of age, and, if the firearm, or ammunition is other than a shotgun or rifle, or ammunition for a shotgun or rifle, to any individual who the licensee knows or has reasonable cause to believe is less than twenty-one years of age.

Id. (codified at 18 U.S.C. § 922(b)(1)). The implementing regulation, 27 C.F.R.

§ 478.99(b)(1), contains similar language. The law did not eliminate any age group’s ability to own, possess, or use a gun. Id. 4

Another part of the law requires licensed dealers to obtain a sworn statement of age before transferring a firearm to a remote purchaser. ^[18] U.S.C. § 922(c) (providing that a

licensed dealer “may sell a firearm to a person who does not appear in person at the licensee’s business premises . . . only if the transferee submits to the transferor a sworn

statement” affirming “that, in the case of any firearm other than a shotgun or a rifle, [the buyer is] twenty-one years or more of age”). And regulations promulgated by the Bureau of Alcohol, Tobacco, Firearms and Explosives require licensed dealers to keep a firearms transaction record, Form 4473, which records, among other things, each transferee’s date

and place of birth. See 27 C.F.R. § 478.124(a) (mandating that “[a] licensed importer, licensed manufacturer, or licensed dealer shall not sell or otherwise dispose, temporarily

or permanently, of any firearm to any person, other than another licensee, unless the licensee records the transaction on a firearms transaction record, Form 4473”); id.

§ 478.124(c)(1), (f) (describing the Form 4473 requirements).

The upshot of these statutes and regulations is that 18-, 19-, and 20-year-olds may not buy handguns or handgun ammunition from a licensed dealer. This is what Marshall challenges.

B. Procedural history

Plaintiffs’ complaint challenged the facial validity of the federal laws and regulations restricting their ability to buy a handgun from a licensed dealer. For relief, they sought a broad declaration that the challenged provisions violate the Second Amendment

and an injunction preventing their enforcement. The government moved to dismiss for failure to state a claim, and Plaintiffs moved for summary judgment.

The district court held that the laws were facially valid because they “are among the ‘longstanding prohibitions’ and ‘conditions and qualifications on the commercial sale of arms,’ which the Supreme Court in Heller did not ‘cast doubt’ on.” Hirschfeld v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 417 F. Supp. 3d 747, 756 (W.D. Va. 2019).

The court also found that the laws fell outside the scope of the Second Amendment because

similar regulations were historically common among the states. Id. In the alternative, the court relied on the legislative record and some of the amici’s evidence to find that the laws

passed intermediate scrutiny. Id. at 758. So the court denied Plaintiffs’ motion for summary judgment and dismissed the complaint. Plaintiffs timely appealed.

We have subject-matter jurisdiction over this appeal under 28 U.S.C. § 1291. The denial of summary judgment and the grant of a motion to dismiss are reviewed de novo.

Garnett v. Remedi Seniorcare of Va., LLC, 892 F.3d 140, 142 (4th Cir. 2018); Betton v. Belue, 942 F.3d 184, 190 (4th Cir. 2019).

II. Discussion

The Second Amendment provides: “A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.” U.S. Const. amend. II. This “inherent,” “fundamental,” “pre-existing”

individual right is not limited to militia service. Dist. of Columbia v. Heller, 554 U.S. 570, 592–95, 628 (2008). Its core is “the right of law-abiding, responsible citizens to use arms in defense of hearth and home.” Id. at 635. So in Heller, the Court struck down a handgun

ban and a law requiring guns to be unloaded and locked within the home as violating the core protection of the Second Amendment. Id. at 635–36. But in doing so, the Court made clear that the rights protected by the Second Amendment are not “unlimited.” Id. at 626.

It did not “cast doubt on” “presumptively lawful regulatory measures,” such as

“longstanding prohibitions on the possession of firearms by felons and the mentally ill, or laws forbidding the carrying of firearms in sensitive places such as schools and government buildings, or laws imposing conditions and qualifications on the commercial sale of arms.”

Id. at 626–27, 627 n.26.

Several years later, the Court struck down a similar handgun ban in Chicago, holding that the Second Amendment was incorporated against the states. McDonald v. City of Chi., 561 U.S. 742, 750 (2010). It explained that while the original understanding of the Second Amendment focused on opposing a tyrannical federal government with a standing army, that fear eventually dissipated. Id. at 770–78; see also Heller, 554 U.S. at

599. Instead, around the time of the Civil War, Congress repeatedly voiced concern that southern states were disarming abolitionists and freed slaves to harass and harm them.

McDonald, 561 U.S. at 772–77. These concerns permeated the drafting and ratification of the Fourteenth Amendment, which reinforced the existing civil rights laws. Id. at 775.

Reviewing this history, the Court held that “it is clear that the Framers and ratifiers of the Fourteenth Amendment counted the right to keep and bear arms among those fundamental rights necessary to our system of ordered liberty.” Id. at 778.

But the Court has cautioned that it has not “undertake[n] an exhaustive historical analysis [] of the full scope of the Second Amendment.” Heller, 554 U.S. at 626. As a result, many questions remain, including what framework courts should use to analyze

Second Amendment claims. To fill that gap, most circuits, including our own, seem to follow some variation of a two-part inquiry. United States v. Chester, 628 F.3d 673, 680

(4th Cir. 2010). 5 And so we apply that inquiry here. See United States v. Hosford, 843

F.3d 161, 165 (4th Cir. 2016) (holding that courts “generally engage in the . . . two-pronged

[Chester] analysis for facial Second Amendment challenges”).

The first step asks “whether the challenged law imposes a burden on conduct falling within the scope of the Second Amendment’s guarantee.” Chester, 628 F.3d at 680. This

“historical inquiry seeks to determine whether the conduct at issue was understood to be

within the scope of the right at the time of ratification.” Id. The government bears the burden of showing that the conduct was not covered by the Second Amendment. See id. at 681–82; see also Ezell v. City of Chi., 651 F.3d 684, 702–04 (7th Cir. 2011); Tyler v.

Hillsdale Cnty. Sheriff’s Dep’t, 837 F.3d 678, 688 (6th Cir. 2016). If the historical record is silent or ambiguous, we assume the conduct is protected. Chester, 628 F.3d at 680–82.

If, however, the challenged law does not impose a burden on conduct falling within the scope of the Second Amendment’s historically understood guarantee, that is the end of the inquiry. Id. at 680.

But if the challenged law burdens Second Amendment rights, “we move to the second step of applying an appropriate form of means-end scrutiny.” Id. Without articulating the applicable standard for Second Amendment challenges, the Supreme Court found that the handgun ban in Heller would fail any level of scrutiny that applies to

enumerated constitutional rights. 554 U.S. at 628–29. It did offer some guidance, however, in rejecting a rational-basis test. Id. at 628 n.27. So at least some form of heightened scrutiny that places the burden on the government must be employed. Chester, 628 F.3d at 680.

To determine which form of heightened scrutiny applies to Second Amendment challenges, the Fourth Circuit has looked to the First Amendment. Kolbe v. Hogan, 849

F.3d 114, 133 (4th Cir. 2017) (en banc). Just as the First Amendment employs strict scrutiny for content-based restrictions but intermediate scrutiny for time, place, and manner regulations, the scrutiny in this context “depends on the nature of the conduct being regulated and the degree to which the challenged law burdens the right.” Chester, 628 F.3d at 682.

The Second Amendment’s “core protection” is “the right of law-abiding, responsible citizens to use arms in defense of hearth and home.” Kolbe, 849 F.3d at 131 (quoting Heller, 554 U.S. at 634–35). When this right is “severely burden[ed],” we apply

the “demanding” strict scrutiny test, which requires the government to show that the challenged law is “‘narrowly tailored to achieve a compelling governmental interest.’” Id. at 133, 145 (quoting Abrams v. Johnson, 521 U.S. ^[74], 82 (1997)). But if the right is

burdened only on the margins, intermediate scrutiny applies. See United States v. Masciandaro, 638 F.3d 458, 470–71 (4th Cir. 2011). And intermediate scrutiny in the Second Amendment context has been described as requiring “a ‘reasonable fit’ between the challenged regulation and a ‘substantial’ government objective.” Chester, 628 F.3d at

683 (quoting Bd. of Trs. of State Univ. of N.Y. v. Fox, 492 U.S. 469, 480 (1989)).

Our inquiry into the challenged provisions’ facial validity changes, however, if we find them to be presumptively valid based on Heller’s list of “presumptively lawful regulatory measure[s].” United States v. Moore, 666 F.3d 313, 318 (4th Cir. 2012); see

also Hosford, 843 F.3d at 165–67. 6 That list includes “longstanding prohibitions on the possession of firearms by felons and the mentally ill, [] laws forbidding the carrying of firearms in sensitive places such as schools and government buildings, [and] laws imposing conditions and qualifications on the commercial sale of arms.” Heller, 554 U.S. at 626–

27, 627 n.26; see Christopher M. Johnson, Second-Class: Heller, Age, and the Prodigal

Amendment, 117 Colum. L. Rev. 1585, 1600–03 (2017) (describing measures that seemingly fall within Heller’s list). If a law falls within this list, we conduct a more streamlined analysis. Chester, 628 F.3d at 679–80; Moore, 666 F.3d at 318; Hamilton v. Pallozzi, 848 F.3d 614, 624 (4th Cir. 2017). This often effectively ends the inquiry and upholds the law as facially valid. But not always. Hamilton, 848 F.3d at 624–25; United

States v. Pruess, 703 F.3d 242, 245–46 (4th Cir. 2012). So before turning to the full two- part test our circuit has adopted, we begin with the presumptively lawful inquiry.

A. The challenged laws are not presumptively lawful

The government and amici argue that the challenged laws fall into two categories on Heller’s presumptively valid list: conditions on commercial sales and longstanding regulations. Both contentions are incorrect.

First, these laws are not “conditions and qualifications on the commercial sale of arms.” Heller, 554 U.S. at 627. A condition or qualification on the sale of arms is a hoop someone must jump through to sell a gun, such as obtaining a license, establishing a lawful

premise, or maintaining transfer records. See Hosford, 843 F.3d at 166. Here, though, the restrictions operate as a total ban on buying a gun from a licensed dealer that has met the required conditions and qualifications to sell arms. There is nothing a law-abiding 18- to

20-year-old can do to buy a handgun from a licensed dealer except wait until she turns 21.

The amici compare this case to one of our precedents, Hosford, 843 F.3d at 165–67.

There, we upheld a regulation requiring those who regularly sell firearms to obtain a

license. Id. at 167. That law regulated only those who wanted to sell guns commercially, but it placed no restrictions on those seeking to buy guns. Id. at 166. We described the licensing law as a “mere condition” on commercial sales that “affects only those who regularly sell firearms, not owned for personal use, in the course of trade or business for the principal purpose of profit.” Id. In doing so, we found that requiring regular sellers of firearms to obtain a license was a limited condition on commercial sales. Id. (noting the requirements for obtaining a license).

Hosford recognized that a law’s substance, not its form, determines whether it qualifies as a condition on commercial sales. Id. (explaining that though the regulation was “framed as a prohibition against unlicensed firearm dealing,” it was “in fact a requirement that those who engage in the commercial sale of firearms obtain a license”).

We cited a district court case as an example of a commercial law that would be “so prohibitive as to turn [a] condition or qualification into a functional prohibition”: a

Chicago ordinance that allowed firearm transfers only outside city limits. Id. (citing Ill.

Ass’n of Firearms Retailers v. City of Chi., 961 F. Supp. 2d 928, 938–939, 947 (N.D. Ill.

2014)). Similarly, in Ezell, the Seventh Circuit struck down a ban on firing ranges within city limits because the ban was “not merely regulatory” but “a serious encroachment” preventing “the ‘law-abiding, responsible citizens’ of Chicago from engaging in target practice in the controlled environment of a firing range.” 651 F.3d at 708–09; see also

Ezell v. City of Chi., 846 F.3d 888, 894 (7th Cir. 2017) (Ezell II) (finding that a commercial zoning and distancing law worked in tandem to functionally preclude any gun ranges, thus

“severely” restricting Second Amendment rights).

The nature of these functional bans is highlighted by Heller, where the Court struck

down a handgun possession ban and a law requiring legally held guns to be unloaded and locked within the home. 554 U.S. at 635–36. As to the handgun possession ban, the Court explained that “the American people have considered the handgun to be the quintessential self-defense weapon” and concluded that the ban “amount[ed] to a prohibition on an entire class of ‘arms.’” Id. at 628–29. Similarly, the Court discussed how the law requiring legally held guns to be unloaded and locked within the home amounted to meaningless

firearm possession because it “ma[de] it impossible for citizens to use [the firearms] for the core lawful purpose of self-defense.” Id. at 630. While it was still theoretically possible to have a firearm, Heller recognized that the reality was a functional ban.

Here, the substance of the challenged laws dictates that they are a functional prohibition on buyers, not a mere condition or qualification on sellers. They ban an entire

group of adult, law-abiding citizens from purchasing handguns from licensed dealers, substantially burdening the group’s rights. The category of conditions and qualifications on commercial sales referenced in Heller is limited, as Hosford highlights, to laws that place conditions and qualifications on the seller, like licensing. It does not encompass a ban on a class of people seeking to purchase handguns from a qualified seller.

The dissent argues that any restriction on a buyer is also a restriction on a seller and vice versa. No doubt a restriction on buyers indirectly affects sellers by reducing demand.

But that cannot be enough. If that was enough, then a ban on possession by 18-to 20-year-

olds could qualify for this “exception” since that restriction on buyers also impacts the commercial sale of firearms. Cf. United States v. Deeb, 175 F.3d 1163, 1167–68 & n.8

(9th Cir. 1999) (“It is clear to us that the word ‘sale’ contained in all of these statutes and regulations refers only to sellers, not buyers.”). In fact, the ban in Heller could be justified by blurring this divide. By the dissent’s reading, this “exception” functionally swallows the Second Amendment. Cf. Marzzarella, 614 F.3d at 92 n.8.

Moreover, this restriction falls asymmetrically on buyers rather than sellers. While

the licensing regime in Hosford placed burdens on sellers with an indirect effect on buyers, the restrictions here fall on the buyers and operate as a functional ban. Under these laws, sellers lose a subset of customers but may still sell to anyone above 20. Those in the restricted age range, however, can do nothing to purchase a handgun from a licensed dealer. ^[7] And other options are not always readily available to many individuals. Not all young adults have friends or family members who are able or willing to gift them a gun.

And secondary markets are not always available to everyone or easy to navigate safely.

On the other hand, licensed dealers come with assurances of quality, safety, legality, and more. These laws push young law-abiding adults to a less safe, less regulated market to defend themselves. Hosford, 843 F.3d at 169 (highlighting the benefits of pushing customers to licensed dealers for safety and regulatory purposes). Not to mention that these restrictions apply to ammunition as well. So the severity and difference in burdens shows why the laws at issue operate as a functional ban on 18- to 20-year-olds rather than a condition on the sale of firearms.

Second, some courts have found laws to be presumptively valid because similar regulations could be described as “longstanding.” See, e.g., Heller II, 670 F.3d at 1253–

54; Drake v. Filko, 724 F.3d 426, 431–34 (3d Cir. 2013). The government and amici argue that age restrictions for purchasing guns are longstanding because many states had similar laws in the late nineteenth and early twentieth centuries. See Nat’l Rifle Ass’n, 700 F.3d at

196–97, 203–04. We do not, however, read the word “longstanding” in Heller as a standalone exception to the Second Amendment. Consistent with the text of Heller, this

Court has never employed that novel view. In our only case discussing the “longstanding” requirement, we correctly did not treat it as its own exception. Hosford, 843 F.3d at 166.

We instead treated it as a potential limit on commercial conditions, requiring the law to be both a commercial condition and longstanding to be presumptively valid. Id. As Hosford

recognized, the sentence in Heller makes clear that “longstanding” serves as a modifier of part or all of the sentence, not as a freestanding category. ^[8] Heller’s historical, textual, and structural analysis counsels against creating a freestanding category of laws exempt from

Second Amendment scrutiny based solely on how long similar laws have existed.

The laws are therefore not presumptively lawful either as commercial conditions or longstanding laws. ^[9] B. Step one: 18- to 20-year-olds are protected by the Second Amendment

At step one of the Chester inquiry, we ask “whether the conduct at issue was understood to be within the scope of the right at the time of ratification.” Chester, 628

F.3d at 680 (emphasis added) (citing Heller, 554 U.S. at 625). The government bears the burden to show that the regulation clearly falls outside the scope of the Second

Amendment. See id. at 681–82; Ezell, 651 F.3d at 702–03; Tyler, 837 F.3d at 688. And in the face of historical silence or ambiguity, we assume the conduct is protected. Chester, 628 F.3d at 680–82. At the very least, this inquiry requires us to consider text, structure, history, and practice to reveal the original public meaning of the Second Amendment. See

Heller, 554 U.S. at 576–628. The relevant question is: What did the right to keep and bear arms mean to the public at the time of ratification?

When evaluating the original understanding of the Second Amendment, 1791—the year of ratification—is “the critical year for determining the amendment’s historical meaning.” Moore v. Madigan, 702 F.3d 933, 935 (7th Cir. 2012) (citing McDonald, 561

U.S. at 765 & n.14). Sources near the time of ratification help us understand the original public meaning of the Second Amendment. Those sources shed light on how language was understood and how similar laws were read at the time. But because language, laws, cases, treatises, practices, and understandings change over time, the light provided dims as we

move further from ratification. Heller, 554 U.S. at 601–03, 614; Nat’l Rifle Ass’n of Am., Inc. v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 714 F.3d 334, 337, 339 (5th

Cir. 2013) (Jones, J., dissenting from denial of rehearing en banc).

One relevant source is state law, as various state statutes and constitutions illuminate

the pre-existing right embodied in the language of the Second Amendment. See Heller, 554 U.S. at 584–86, 600–03 (relying on state constitutions and statutes to clarify the linguistic meaning of the Second Amendment). Indeed, the definitions and interpretive tools that would have been used to understand the language of state laws would have also been used by the public to understand the Constitution. John O. McGinnis & Michael B.

Rappaport, Original Methods Originalism: A New Theory of Interpretation and the Case

Against Construction, 103 Nw. U. L. Rev. 751, 769–70 (2009). State laws prove especially

helpful in determining who the public would have considered to be included in the “militia.” This is especially true prior to ratification, as organized militias were creatures of state law. ^[10] In 1788, however, the Constitution granted the federal government the power to call

“forth the Militia to execute the Laws of the Union, suppress Insurrections and repel

Invasions” and “provide for organizing, arming, and disciplining, the Militia.” See U.S.

Const. art. I, § 8, cl. ^[15], 16. And the Supreme Court has held that state militia laws were preempted to the extent they conflicted with federal militia laws. Houston v. Moore, 18

U.S. (5 Wheat.) 1, 24 (1820); J. Norman Heath, Exposing the Second Amendment: Federal

Preemption of State Militia Legislation, 79 U. Det. Mercy L. Rev. ^[39], 43–47, 67 (2001).

So after 1788 federal law became more relevant to understanding the linguistic meaning of the term “militia” as used in the Bill of Rights.

This is not to suggest that state militia laws became irrelevant, as the Constitution still recognized important roles for the states in governing and defining their own militias. ^[11]

Indeed, besides appointing officers, the states could pass laws governing their militias to the extent they did not conflict with federal law and could freely control their militias when they were not called on by the federal government, which was almost always the case. ^[12]

And the public would be most familiar with the composition of their own state militia. So both federal and state laws inform the linguistic meaning of “militia” near the time of ratification.

But state laws are less illuminating the further one moves from ratification. This is, in part, because the states initially were not restricted by the Bill of Rights. Barron v. City of Balt., 32 U.S. (7 Pet.) 243, 247–48 (1833). And as time passes, states may diverge from a common understanding and experiment, and language is more likely to evolve. New

State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting) (praising states as laboratories of democracy); Thomas R. Lee et al., Corpus Linguistics & Original

Public Meaning: A New Tool to Make Originalism More Empirical, 126 Yale L. J. Forum

20, 22 (2016). The Heller Court recognized this, as it looked to state constitutions and laws at the time of ratification to determine the Second Amendment’s public meaning but noted that post-Civil War sources “do not provide as much insight into its original meaning

as earlier sources.” 554 U.S. at 614; Nat’l Rifle Ass’n, 714 F.3d at 337, 339 & n.5 (Jones, J., dissenting) (discounting late nineteenth- and early twentieth-century sources). Indeed, the Supreme Court has made clear that Heller relied on nineteenth-century sources “as mere confirmation of what the Court thought had already been established” by Founding- era sources. Gamble v. United States, 139 S. Ct. 1960, 1975–76 (2019). To the extent later sources conflict with evidence closer to ratification, their relevance is limited.

Evidence about the ratification of the Fourteenth Amendment also highlights the waning usefulness of state laws. While the Founders sought to oppose a tyrannical federal government, the self-defense of freed Blacks was central to the passage of the Fourteenth

Amendment. Akhil Reed Amar, Heller, HLR, and Holistic Legal Reasoning, 122 Harv. L.

Rev. 145, 176 (2008). And McDonald reviewed pre- and post-Civil-War-era laws to show that a right to self-defense was part of the Second Amendment that was intended to be incorporated. 561 U.S. at 770–78. But in doing so, the Supreme Court’s review focused on federal laws, such as the Freedmen’s Bureau Act of 1866, and debates in Congress. Id.

at 773–77. Many of those debates focused on legislators’ alarm about southern laws and militias being employed to disarm recently freed slaves. Id. at 772, 775–76. The ineffectiveness of federal law at preventing southern states from disarming free people motivated Congress to enact the Fourteenth Amendment to stop states from violating the right to keep and bear arms. Id. So we should hesitate to look to Civil-War-era state laws that restricted the right to bear arms as evidence of the scope of the Second Amendment right at the time of the Founding.

Looking through this historical lens to the text and structure of the Constitution reveals that 18- to 20-year-olds have Second Amendment rights. Virtually every other constitutional right applies whatever the age. And the Second Amendment is no different.

The militia laws in force at the time of ratification uniformly required those 18 and older

to join the militia and bring their own arms. While some historical restrictions existed, none support finding that 18-year-olds lack rights under the Second Amendment.

1. The text and structure of the Constitution

As with any matter of constitutional interpretation, “our inquiry begins with the text of the Constitution.” Altman v. City of High Point, 330 F.3d 194, 200 (4th Cir. 2003). Both

the text and structure of the Second Amendment, along with its place within the Constitution as a whole, reveal that it protects 18- to 20-year-olds. First, nothing in the text of the Second Amendment limits its application by age. Second, the most analogous rights to the Second Amendment, those in the First and Fourth Amendments, similarly

contain no age limits. Third, most other constitutional rights are not age limited. And fourth, the few rights that may not apply to those under 18 or that change by age are not analogous to the Second Amendment, and most of those rights become applicable at age

18, not 21.

As to the first point, while various parts of the Constitution include age

requirements, the Second Amendment does not. The Founders set age requirements for Congress and the Presidency, but they did not limit any rights protected by the Bill of Rights to those of a certain age. See U.S. Const. art. I, § 2 (age 25 for the House); id. art.

I, § 3 (age 30 for the Senate); id. art. II, § 1 (age 35 for the President); cf. id. amend. XXVI

(setting voting age at 18). In other words, the Founders considered age and knew how to set age requirements but placed no such restrictions on rights, including those protected by the Second Amendment.

Turning to the second point, the two rights most like the Second Amendment apply to all people regardless of age. The Second Amendment refers to the “right of the people,” which is a phrase also used in the First and Fourth Amendments to denote an individual right. Heller, 554 U.S. at 579. Like the Second Amendment, both the First and Fourth

Amendments codify a pre-existing, fundamental, inalienable individual right. Id. at 592.

The Supreme Court and the Fourth Circuit have used First Amendment law as a guidepost for interpreting the Second Amendment. Heller, 554 U.S. at 591, 595, 626;

Kolbe, 849 F.3d at 133; Chester, 628 F.3d at 678, 682. And the First Amendment applies to all persons, even those under 18. Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S.

503, 506 (1969) (free speech); see also W. Va. State Bd. of Educ. v. Barnette, 319 U.S. 624, 642 (1943) (free exercise). To be sure, minors’ First Amendment rights are qualified to

some degree, see Tinker, 393 U.S. at 506, 513, but those qualifiers do not eliminate the rights altogether, for time, place, and manner regulations are a part of everyone’s First

Amendment rights regardless of age, see Ward v. Rock Against Racism, 491 U.S. 781, 791 (1989). Instead, Tinker presumes minors have similar rights as adults because they do not

“shed their constitutional rights to freedom of speech or expression at the schoolhouse gate.” 393 U.S. at 506; accord Brown v. Ent. Merchs. Ass’n, 564 U.S. 786, 805 (2011)

(enjoining a California law prohibiting the sale of violent video games to those under 18 without a guardian’s permission as a violation of the First Amendment).

The logic of Tinker has also been applied in the Fourth Amendment context. It is

“indisputable” that the Fourth Amendment protects the rights of all persons, including

minors. New Jersey v. T.L.O., 469 U.S. 325, 334 (1985). But like the First Amendment, the context matters in determining the scope of the right. See id. at 337–38 (explaining that minors’ expectations of privacy differ based on the item searched and whether the search occurs on or off school property). But the existence of the right is not limited by age.

To be clear, in considering analogous rights from the First and Fourth Amendments and their application to minors, we are not suggesting that the protections of the Second

Amendment necessarily extend in full force to those under 18. 13 Nor do we suggest that minors have an unfettered right to purchase firearms. We merely emphasize that the First and Fourth Amendments’ protection of “the people,” including those under 18, confirms that “the people” protected by the Second Amendment includes at least those 18 and older.

As for the third point, less-analogous rights, such as the enumerated rights to due process and equal protection, also apply to all persons, including minors. See, e.g., Brown v. Bd. of Educ. of Topeka, 347 U.S. 483, 493 (1954) (Do separate-but-equal schools

“deprive the children of the minority group of equal educational opportunities? We believe that [they] do[].”); Goss v. Lopez, 419 U.S. 565, 574 (1975) (due process). Newer

unenumerated rights do as well. See, e.g., Planned Parenthood of Central Mo. v. Danforth, 428 U.S. ^[52], 74 (1976) (enjoining a law requiring a minor to obtain parental consent to get an abortion, declaring that “[c]onstitutional rights do not mature and come into being magically only when one attains the state-defined age of majority”).

And while the Eighth Amendment applies differently based on age, it does so by providing greater constitutional protection to those under 18. In Roper v. Simmons, 543

U.S. 551, 574 (2005), the Supreme Court held that the Eighth Amendment’s application to minors prohibited the use of the death penalty for those who were under 18 at the time they committed the offense. The Court drew the line at that age and has held to it. For “[t]he age of 18 is the point where society draws the line for many purposes between childhood and adulthood. It is . . . the age at which the line for death eligibility ought to rest.” Id.

Since Roper, no federal court has expanded this logic to provide greater protection to those over 18. Similarly, no federal court has ever held that one loses protection under the Eighth

Amendment—or somehow escapes its applicability altogether—by reaching the age of majority.

Finally, there are some rights that do not apply to those under 18. But those rights also suggest that 18-year-olds are protected by the Second Amendment because the Court has drawn the line for their application at age 18, not 21. The jury trial right does not apply to juvenile proceedings, McKeiver v. Pennsylvania, 403 U.S. 528, 545, 550–51 (1971), but many state laws end the jurisdiction of juvenile courts at age 17, with a few at age 16 and just recently one at age 18, Anne Teigen, Juvenile Age of Jurisdiction and Transfer to Adult

Court Laws, Nat’l Conf. St. Legis. (April 8, 2021), (saved as ECF opinion attachment).

And many view the protections of juvenile proceedings for minors as better and want to increase the age at which they apply. See McKeiver, 403 U.S. at 545, 550. The other relevant right is voting, which the Constitution explicitly limits to those 18 and older. U.S.

Const. amend. XXVI.

There are only two unenumerated rights states may restrict for those under a certain age: marriage and sex. See Lawrence v. Texas, 539 U.S. 558, 578 (2003) (finding a right for sexual autonomy but differentiating minors); Obergefell v. Hodges, 576 U.S. 644, 675

(2015) (holding that the right to marry is fundamental). But even still, the legal age of marriage in all but two states is 18, and even in those two states it appears that one can marry at 18 with parental consent. See Miss. Code Ann. § 93-1-5; Neb. Rev. Stat. §§ 42-

105, 43-2101. For sex, the oldest state-imposed age of consent (the age at which an individual can legally consent to sexual intercourse) is 18. 14 So even these rights apply to some degree to 18-, 19-, and 20-year-olds.

But whatever their import, the treatment of these rights cannot overcome the strong evidence that the Second Amendment right vests at least at age 18. Indeed, it would be odd to treat the Second Amendment like marriage and sex rather than contemporaneously ratified rights such as the First and Fourth Amendments that have been described as fundamental pre-existing rights analogous to the Second Amendment. Heller, 554 U.S. at

592.

There are many things that minors and even those under 21 cannot do. See Ent.

Merchs. Ass’n, 564 U.S. at 836–37 (Thomas, J., dissenting) (explaining that minors cannot drive for hire or drive a school bus, buy tobacco, play bingo for money, or execute a will).

But none of those restrictions implicate constitutional rights, so states have great leeway to regulate those activities under their general police powers. And while the Court has

“recognized that the State has somewhat broader authority to regulate the activities of children than of adults,” that does not mean that children necessarily have different rights

than adults. Danforth, 428 U.S. at 74. Often they have the same rights as adults, but the states’ interests are stronger with regard to minors so restrictions may more easily pass constitutional scrutiny. Ent. Merchs. Ass’n, 564 U.S. at 794–95. So it is hard to conclude that 18- to 20-year-olds have no Second Amendment rights when almost every other constitutional right affords them protection. This conclusion becomes inescapable when we consider the history.

2. Founding-era militia laws

The historical evidence from the Founding shows that 18- to 20-year-olds are protected by the Second Amendment. Founding-era militia laws provide powerful historical evidence. Near the time of ratification, the federal government and every state required 18-year-old men to be part of the militia and bring their own arms. See Heller, 554 U.S. at 596, 627. This evidences the Founding public’s understanding that the “militia,” and thus the Second Amendment, encompassed 18-year-olds.

These Founding-era militia laws illuminate the broader individual right enshrined in the Second Amendment. Id. at 600–03. The phrase “[a] well regulated Militia, being

necessary to the security of a free State” augments the individual “right of the people” and helps us understand its scope and resolve ambiguities. Id. at 577–78, 595–99.

The civic purpose of preserving our nation’s security sits at the Second

Amendment’s core. Id. at 597–98. Being part of the militia, having a firearm, and being prepared for action were vital to protecting the new Republic. This civic purpose addressed internal threats because there was no police force. So members of the militia “served various law enforcement functions,” including protection “against private thugs, such as pirates and renegade Indians, as well as against more organized threats.” Amar, supra, at

164. For this reason, many states required citizens to be armed and keep firearms at home even outside militia service. ^[15] Indeed, these practices have deep roots in the common-law doctrines of hue and cry and the posse comitatus, where privately armed citizens would be called to respond to crimes or other public emergencies. ^[16]

Militias were also important to repel invasions while reducing the need for the standing army feared at the time of the Founding. Heller, 554 U.S. at 597–98; The Federalist No. 29, at 182–87 (Alexander Hamilton) (J. Cooke ed., 1961); The Federalist

No. 46, at 321 (James Madison). 17 The third purpose of the militia is closely related: to act as a check on a tyrannical government. ^[18] Many feared that a standing army would be used to disarm and oppress the people. 19 The militia democratized self-defense and gave the people an important check on and voice in government action. ^[20] In order to serve these purposes, a wide swath of the public had to be armed.

As George Mason observed, “[T]o disarm the people; that [is] the best and most effectual way to enslave them.” 3 Jonathan Elliot, The Debates in the Several State Conventions on the Adoption of the Federal Constitution 380 (1863). Hamilton noted “[i]f the representatives of the people betray their constituents, there is then no resource left but in the exertion of that original right of self-defence, which is paramount to all positive

forms of government.” The Federalist No. 28, at 178 (Alexander Hamilton) (J. Cooke ed., 1961); see also 3 Joseph Story, Commentaries on the Constitution of the United States 746

(1833) (calling the right to keep and bear arms “a strong moral check against the usurpation and arbitrary power of rulers”). Thus, as Patrick Henry concluded, the people must

“[g]uard with jealous attention the public liberty. Suspect every one who approaches that jewel. Unfortunately, nothing will preserve it but downright force. Whenever you give up that force, you are inevitably ruined.” Elliot, supra, at 45. So “[t]he great object is, that every man be armed . . . . Every one who is able may have a gun.” Id. at 386.

Madison and other Founders often explained how the militia was key to safety and defense broadly. To serve those functions, and to avoid becoming a standing army, individuals needed to be privately armed. Indeed, one of the “advantage[s]” that Madison

thought “Americans possess over the people of almost every other nation” was the “advantage of being armed.” The Federalist No. 46, at 321 (James Madison).

In public discourse, Madison contrasted this American advantage of an armed citizenry with Europe, where governments were “afraid to trust the people with arms.” Id. at 322. For European governments recognized an armed citizenry could not be conquered by the largest standing army that could be supported. Id. Indeed, the American Revolution proved this very point, as American militias could not be conquered by British regulars.

Id. at 321. If European powers permitted an armed citizenry and local governments that appointed militia officers, then “the throne of every tyranny in Europe would be speedily overturned, in spite of the legions which surround it.” Id. at 322. For Madison, this

combination in America—the “advantage of being armed” alongside “the existence of subordinate governments, to which the people are attached and by which the militia officers are appointed”—formed “a barrier” against tyranny. Id. at 321–22. And the Constitution protected both advantages to check arbitrary power: Its structural design protected local governments separate from the federal government and the Second Amendment protected private citizens’ right to privately keep and bear arms.

This civic purpose reflected the central belief that people voluntarily created the state to protect their own rights from others. 21 And one of those rights was self- preservation, which citizens could assert individually in self-defense. ^[22] The privately

armed populace furthered these interconnected rights and duties at the core of the state and 21 1 WILLIAM BLACKSTONE, COMMENTARIES 123–25, 130 (St. G. Tucker ed., 1803); M. de Montesquieu, The Spirit of the Laws bk I, pp. 4–6, bk XXVI, pp. 211 & 231, in THE COMPLETE WORKS OF M. DE MONTESQUIEU (London: T. Evans 1777); John Locke, TWO TREATISES OF GOVERNMENT 129–30 (Haffner Pub. 1947) (1694); Thomas Hobbes, LEVIATHAN ch. XIII–XIV (London, George Routledge & Sons ed., 1894) (1651); see also Kopel, supra note 16, at 826; Don B. Kates Jr., The Second Amendment and the Ideology of Self-Protection, 265 CONST. COMMENT. 87, 89–92, 101–04 (1992).

the Second Amendment: defense of oneself and one’s community. See Heller, 554 U.S. at 577–78. 23

So while the individual right of self-defense was “the central component” of the Second Amendment, the civic purpose “was codified” based on the fear that a tyrannical government would eliminate the civically minded militia. Id. at 597–99. Both reflected

the pre-existing right of self-defense that the Founders valued as the core purpose of the state. “In the Founders’ world, individual self-protection and community defense were not wholly separate spheres.” Amar, supra, at 164. 24

The relationship between the narrower civic purpose and the broader individual right provides us one way to understand the scope of the individual right. The individual right belongs to “the people,” which “unambiguously refers to all members of the political

community.” Heller, 554 U.S. at 580. The “militia,” on the other hand, is “a subset of ‘the people’—those who were male, able bodied, and within a certain age range.” Id. Those excluded from the militia were not excluded from the Second Amendment, but those included in the militia were surely covered by the Second Amendment. So because the militia is a subset of “the people,” those in the militia share the same rights as “the people.”

With this in mind, we look to militia laws in effect at the time of ratification to determine how the public at that time would have understood the word “militia.” Those who were obligated to serve in the militia and bring their own arms fall unambiguously within the original public meaning of “the people” who had a right “to keep and bear

Arms.” a. Militia laws at the time of the Founding show that 18-year- olds have Second Amendment rights

“Men trained to Arms from their Infancy, and animated by the Love of Liberty, will afford neither a cheap or easy Conquest.” 25

Every militia law near the time of ratification required 18-year-olds to be part of the militia and bring their own arms. Around the time the Second Amendment was ratified in 1791, Congress began debating invoking its power under the Militia Clauses to better

organize the militias for federal use in emergencies. U.S. Const. art. I, § 8, cls. 15–16. The effort was pushed by Secretary of War Henry Knox, who argued to Congress that while the “military age has generally commenced at sixteen,” the age for the federal select militia

should be set at 18 because “the youth of sixteen do not commonly attain such a degree of robust strength as to enable them to sustain without injury the hardships incident to the field.” 2 Annals of Cong. 2088, 2092, 2099 (1790). One of the first rejected proposals

25 2 JOURNALS OF THE CONTINENTAL CONGRESS, 1774–1789, at 169 (1905) (Continental Congress’s address to the inhabitants of Great Britain in 1775).

suggested that the federal government would arm everyone, but this drew sharp rebukes, such as Representative Wadsworth arguing “their parents would rather give them guns of their own, than let them take others from the U.S. which were liable to be taken away at the very moment they were most wanted.” 14 Documentary History of the First Federal

Congress: Debates in the House of Representatives, Third Session: December 1790–March

1791, at 62 (1995). This proposal raised the concern that any check on tyranny could be

undermined by federal control of the militia’s arms. In the end, Congress settled on the first federal Militia Act in 1792. That Act required “each and every free able-bodied white

male citizen of the respective states, resident therein, who is or shall be of the age of eighteen years, and under the age of forty-five years (except as is herein after excepted)

shall severally and respectively be enrolled in the militia.” Act of May 8, 1792, ch. 33, § 1, 1 Stat. 271, 271 (emphasis added). The Act also required that “every citizen so enrolled and notified, shall, within six months thereafter, provide himself” with a firearm and ammunition. Id. 26 State laws also help us understand the original public meaning of “militia” because, as the Constitution recognizes, many militias were defined by state laws. See U.S. Const. art. I, § 8, cl. ^[16]. And those state laws only further prove the point.

Before ratification, when militias were solely defined by state law, most colonies and states set the age for militia enlistment at 16. 27 Right after ratification, when federal

law set the age at 18, every state also set their militia age at 18. See Appendix 2. Like the federal Militia Act, these state militia laws required enrolled citizens, including 18-year- olds, to bring their own firearms and ammunition either explicitly or as a general background principle. ^[28] Generally firearms were owned privately and kept by the members

27 J.A. 303–05; Cramer, supra note 15, at 2–10. While not every state set its militia enlistment age at 16 before ratification and some fluctuated, no state appears to have set its age higher than 18 in decades preceding ratification. See Appendix 1.

of the militia in their homes as the government did not provide or keep the guns unless the citizen was too poor to afford one. ^[29] That 18-year-olds had to be part of the militia and bring their own arms establishes that 18-year-olds were included among “the people” who enjoyed the Second Amendment right to keep and bear arms. See Heller, 554 U.S. at 580.

Many of those in the militia were quite young. As the Founder Tench Coxe wrote, “the powers of the sword are in the hands of the yeomanry of America from sixteen to sixty . . . . Their swords . . . are the birthright of an American.” Tench Coxe, A

Pennsylvanian, No. 3, Pa. Gazette, Feb. ^[20], 1788, at 22. Because the individual right is broader than the Second Amendment’s civic purpose, those required to serve in the militia

and bring arms would most assuredly have been among “the people” who possessed the right. See Heller, 554 U.S. at 579–81. And the indisputable historical evidence establishes that 18-year-olds had to join the militia and bring their own firearms. So those 18 and older would have been understood to be covered by the Second Amendment at the Founding. b. The militia laws cited by the government do not undermine the conclusion that 18-year-olds are protected

Despite the overwhelming evidence that federal and state militia laws around the Founding required those 18 and older to serve in the militia with their own firearm, the government suggests that more specific militia exclusions and requirements undermine the general laws. But none of the laws it cites undermine the Founding-era evidence that those older than 18 have Second Amendment rights.

In the decades after ratification, several states enacted laws that at times raised the age for mandatory militia service to 21. But even if 18-year-olds were not part of the compulsory militia, they were not necessarily excluded either from service in the militia or the Second Amendment’s protection. For, once again, the militia is but a subset of those with Second Amendment rights. Heller, 554 U.S. at 580. For example, most militia laws did not require those over a certain age to serve in the militia, but we would not suggest that those over that age lack any Second Amendment rights. Militia laws are helpful because they provide a baseline for determining the relevant political community that enjoyed Second Amendment rights. They support the affirmative conclusion that 18-year- olds are protected by the Second Amendment. But even if history were less clear, 18-year- olds would not necessarily be excluded from the Second Amendment’s protections. ^[30]

Further, the substance and timeline of those laws fail to aid the government. The law closest to ratification cited by the government was enacted by Pennsylvania in 1793.

That law, like all the militia laws at the Founding, required 18-year-olds to join the militia. 31 It only exempted those under 21 from active militia service during times of peace.

The next closest law to ratification comes from Delaware in 1807. It also required 18-year- olds to be enrolled in the militia, only exempting those under 21 from duty in peacetime. 32

The remaining laws are decades newer, and even some of those only provide peacetime exemptions. J.A. 287–89; see also Nat’l Rifle Ass’n, 714 F.3d at 342 (Jones, J., dissenting)

(making a similar argument that a 1843 Ohio law fails to show that 18-year-olds lack

Second Amendment rights because even though they were exempt from service during peacetime, they were still required to be in the militia). 33 These peacetime-exemption laws do not exclude those under 21 from enrolling in the militia. In fact, they often still required

18-year-olds to join the militia. They just exempted them from certain militia requirements during peacetime, such as bringing a firearm. So these laws do not undermine, but instead reinforce, the conclusion that 18-year-olds were included in the public’s conception of the militia.

And, as noted previously, because states were not subject to the Bill of Rights, the further one moves from ratification, the more state variation will occur and the less probative the laws will be in interpreting the Second Amendment. ^[34] While we should not completely discount evidence closer to the Civil War, these later militia laws are less instructive. In the decades after the Founding, the militia had become much less important while individual self-defense became more meaningful. ^[35] As a result, fewer people needed to be in the militia, and militias were less tied to gun ownership. Robert Leider, Our Non-

Originalist Right to Bear Arms, 89 Ind. L.J. 1587, 1634 (2014). So militia laws enacted decades after ratification, especially when the militias were in decline, provide limited

evidence of the original public meaning of the Second Amendment. See Nat’l Rifle Ass’n, 714 F.3d at 339, 342 (Jones, J., dissenting).

The government also relies on a 1778 New Jersey law that set the militia age at 21 for a single call to arms. That New Jersey law called for 1,000 men ages 21 and older to enlist to defend the frontiers. See generally An Act to Embody, for a Limited Time, One

Thousand of the Militia of this State, for the Defence of the Frontiers Thereof §§ 1, 3

(1778), in Acts of the General Assembly of the State of New Jersey. At a Session Begun at Trenton on the 27th Day of October, 1778 and Continued by Adjournments 58, 58–59 (1778). The government’s reliance on this law reveals the sinking sand on which it stands.

This law was a one-time call to arms for 1,000 people for a single purpose. Id. It permits officers to enlist those ages 16 to 21 but only calls upon those 21 and older for this particular

service. Id. § 3. Moreover, New Jersey still generally required 16-year-olds to enroll in the militia, both before and after this temporary call to arms. ^[36] The mandatory enrollment age increased to 18 around ratification 37 and remained there through 1829 when the state enacted a peacetime-exemption law for those under 21. 38 Taken together, none of these

New Jersey laws reflect that 18-year-olds were excluded from militia service.

The government also points to a series of Virginia laws that set varying ages for militia service. But at the time of ratification, Virginia required militia service for all men

ages 18 and older. An Act for Regulating the Militia of this Commonwealth § 12 (1792), in A Collection of All Such Acts of the General Assembly of Virginia 282, 284 (1792). In 1705, Virginia set the age of service at 16. An Act for Settling the Militia (1705), in 3

William Waller Hening, The Statutes at Large; Being a Collection of All the Laws of Virginia, from the First Session of the Legislature, in the Year 1619, at 335, 335 (1823).

In 1723, Virginia raised the mandatory age for enlistment to 21 but made it clear that 16- year-olds could be admitted as substitutes. An Act for the Settling and Better Regulation of the Militia § 2, in 4 William Waller Hening, The Statutes at Large; Being a Collection of All the Laws of Virginia, from the First Session of the Legislature, in the Year 1619, at

118, 118 (1820). Virginia kept the minimum age for enlistment at 21 in 1738 and used that age of service for a special call to arms for the French and Indian War in 1754. 39 The next

38 J.A. 288; An Act to Exempt Minors from Militia Duty in Time of Peace § 1 (1829), in ACTS OF THE FIFTY-FOURTH GENERAL ASSEMBLY OF THE STATE OF NEW JERSEY, AT A SESSION BEGUN AT TRENTON, ON THE TWENTY-SEVENTH DAY OF OCTOBER, ONE THOUSAND EIGHT HUNDRED AND TWENTY-NINE 3, 3 (1829).

year, however, Virginia adopted 18 as the age of service, then reduced the age to 16 during the Revolutionary War, and returned to age 18 near the time of ratification. ^[40]

Pennsylvania follows a similar story. In 1755, at the height of the French and Indian

War, Benjamin Franklin drafted and got passed a militia law that required the enrollment of men over 21. 41 But in 1777 Pennsylvania lowered the age to 18 where it remained through ratification. ^[42]

So every state required enrollment at least by age 18 during the decades around ratification, which is the critical time for determining the historical meaning of the Second

Amendment. See Moore, 702 F.3d at 935 (citing McDonald, 561 U.S. at 765 & n.14);

Chester, 628 F.3d at 680 (citing Heller, 554 U.S. at 625).

Next, the government points to three state laws requiring parental consent to enlist in the militia. J.A. 291. But two of those states were not states at the Founding and their laws were not passed until 1835 and 1863 respectively. ^[43] The earliest of these laws was passed by New York in 1818, but it is a stretch to claim that it required parental consent for anyone under 21, as the law required 18-year-olds to enroll in the militia and only required parental consent to join particular companies such as cavalry and artillery. ^[44]

Before ratification New York set the enlistment age at 16 and right after ratification it was raised to 18 in line with the federal Militia Act. ^[45] And even true parental-consent laws exempted 18-year-olds from mandatory service in the militia but still permitted them to join with their parents’ permission.

The government also points to a handful of laws requiring parents to furnish arms and pay fees for their enlisted children. J.A. 298–301. But those laws do little to suggest that those under 21 were not required to keep and bear arms. Most of those laws require

18-year-olds to enlist but do not set an age for parental liability, just requiring guardians

to be liable for the equipment and food of those “who shall be under their care.” Id. And the existence of these laws is unsurprising given that some states required those older than 16 to enroll in the militia. ^[46] Regardless, “the point remains that those minors were in the militia and, as such, they were required to own their own weapons,” even if their parents had to buy those weapons or consent to them joining. Nat’l Rifle Ass’n, 714 F.3d at 342

(Jones, J., dissenting).

So contrary to the government’s claims, state and federal militia laws show that 18- year-olds had a right to keep and bear arms. Near ratification, every state and the federal

government required 18-year-old men to be in the militia and have their own weapons, meaning that those people were part of “the people” protected by the Second

Amendment. ^[47] Nothing in the few and distant militia laws the government identifies undermines this conclusion.

3. The government’s arguments do not point to a different conclusion

Founding-era militia laws provide strong evidence that 18-year-olds were covered by the Second Amendment. So to undermine this evidence, the government turns to other historical arguments. First, the government contends that the Second Amendment is tied to virtue and minors were not considered virtuous. The government thus posits that because the age of majority at the Founding was 21, those under 21 were not covered by the Second Amendment. Second, it points to state laws in the late nineteenth and early twentieth centuries that restricted the sale of handguns based on age. We find neither argument persuasive. a. The age of full majority

Both parties rely on the age of majority to make their case, but they look to the age of majority from different time periods. Neither is relevant.

Plaintiffs argue that because the age of majority is now 18, 18-year-olds have

Second Amendment rights. In response to a similar argument, the Fifth Circuit correctly explained that “‘majority or minority is a status’” that “lack[s] content without reference

to the right at issue.” Nat’l Rifle Ass’n, 700 F.3d at 204 n.17 (quoting Jeffrey F. Ghent, Annotation, Statutory Change of Age of Majority as Affecting Pre–Existing Status or Rights, 75 A.L.R.3d 228 § 3 (1977)). Yet inexplicably, the Fifth Circuit then concluded that 18-year-olds do not have Second Amendment rights because the age of majority was

21 at the Founding. Id. at 201–03. The government likewise relies heavily on the idea that the age of majority was 21 at the Founding. See 1 William Blackstone, Commentaries 463 (St. G. Tucker ed., 1803); Black’s Law Dictionary 847 (9th ed. 2009) (suggesting that the age of majority was 21 at the Founding).

But the age of majority—even at the Founding—lacks meaning without reference

to a particular right. As Blackstone’s Commentaries makes clear, the relevant age of majority depended on the capacity or activity. ^[1] William Blackstone, Commentaries 463–

64; see also id. at 465 (noting the “different capacities which [individuals] assume at different ages”). For example, a man could take an oath at age 12, be capitally punished in a criminal case at age 14, and serve as an executor at age 17. Id. at 463–64. And a woman could consent to marriage at age 12, choose a guardian at age 14, and serve as an executrix at age 17. Id. at 463. Both sexes had to wait until age 21 to dispose of their lands. Id. So while the full age of majority was 21, that only mattered for specific

activities. Id. And even so, constitutional rights were not generally tied to an age of majority, as the First and Fourth Amendments applied to minors at the Founding as they do today. So the age of majority Blackstone identifies for different activities tells us little about the scope of the Second Amendment’s protections.

To square the circle, some have attempted to connect the age of majority to the Second Amendment by arguing that the right to keep and bear arms depends on “civic virtue,” and those under 21—the Founding-era age of majority—were not considered virtuous. Nat’l Rifle Ass’n, 700 F.3d at 201. But even if a group could be categorically restricted from possessing Second Amendment rights because of the group’s perceived virtue or dangerousness, we find the fact that those 18 and older had to serve in the militia at the time of ratification means that they are not such a group. Even so, any determination of a group’s dangerousness fits more appropriately at step two of this analysis.

But this age-of-majority argument has been proffered at step one to contend that those historically considered without virtue are outside the scope of the Second

Amendment’s protection. See, e.g., Nat’l Rifle Ass’n, 700 F.3d at 200–01; United States v. Rene E., 583 F.3d 8, 15–16 (1st Cir. 2009). There is an ongoing debate about the relevance of virtue, 48 but whether the Second Amendment hinges on notions of virtue is irrelevant for our purposes. As the Sixth Circuit said, “that the Founders understood the Amendment to protect the virtuous does not explain who was counted among that class.” Tyler, 837

F.3d at 688–90 (rejecting the claim that the mentally ill were categorically unvirtuous).

Even if virtue matters, there is no Founding-era evidence that 18- to 20-year-olds were considered unvirtuous, as only a few of the cited sources refer to minors and the ones that do fail to rely on primary sources. ^[49] That is not historical evidence.

The cited evidence to support the contention that minors could be prevented from

buying firearms comes from a passing reference in a footnote in Thomas Cooley’s treatise, which was published nearly a century after ratification. Thomas M. Cooley, Treatise on

Constitutional Limitations 740 n.4 (5th ed. 1883) (“[t]hat the State may prohibit the sale of arms to minors” (citing State v. Callicutt, 69 Tenn. 714 (1878), which held a Tennessee

state law, 1856 Tenn. Pub. Acts 92, 92, that made it illegal for someone to sell, loan, or give a minor a pistol except for hunting was valid)). Putting aside the limited weight that should be given to this passing statement in a footnote, Cooley’s lone source is a state case applying a state law from decades after ratification. Id. Neither the law nor the case carries

much value in determining original public meaning. Moreover, the case cited, State v. Callicutt, relies on Aymette v. Tennessee, 21 Tenn. (2 Hum.) 154, 158 (1840), for the proposition that the Second Amendment only protects the right “to keep and bear arms for their common defense” in a military context, not an individual right. Callicutt, 69 Tenn.

proposition); Don B. Kates & Clayton E. Cramer, Second Amendment Limitations and Criminological Considerations, 60 HASTINGS L.J. 1339, 1360 (2009) (cited by the government but only providing sources for felon restrictions, not restrictions on minors); Don B. Kates, Jr., Second Amendment, in 4 ENCYCLOPEDIA OF THE AMERICAN REVOLUTION 1640 (Leonard W. Levy et al. eds., 1986) (claiming that minors could be prevented from buying firearms but providing no sources); Cornell, supra note 48, at 679 (discussing civic virtue but not discussing minors); Glenn Harlan Reynolds, A Critical Guide to the Second Amendment, 62 TENN. L. REV. 461, 480 (1995) (implying minors are not virtuous based on Kates’s scholarship); Yassky, supra note 11, at 588, 626 (cursorily discussing virtue and not in the context of minors). These sources merely cite each other without offering primary sources supporting the idea that minors were considered unvirtuous. See Folajtar, 980 F.3d at 915–18 (Bibas, J., dissenting); see also Eugene Volokh, Implementing the Right to Keep and Bear Arms for Self-Defense: An Analytical Framework and a Research Agenda, 56 UCLA L. REV. 1443, 1511–12 (2009) (arguing that a danger justification could apply to those under age 18 but the justification was not persuasive considering the other rights enjoyed by those 18 and older).

at 716 (also citing Page v. State, 50 Tenn. (3 Heisk.) 198 (1871)). Not only did Heller explicitly reject this limited military and collective view of the Second Amendment, but it also rejected Aymette as offering an “odd reading of the right . . . [and] not the one we

adopt.” 554 U.S. at 579–80, 613. So the only primary source the government can muster, that comes long after ratification, does nothing to support its case.

And even so, in Cooley’s separate discussion of the Second Amendment as an

individual right he states that “[t]he meaning of the provision undoubtedly is, that the people, from whom the militia must be taken, shall have the right to keep and bear arms;

and they need no permission or regulation of law for the purpose.” Thomas M. Cooley, The General Principles of Constitutional Law in the United States of America 271 (1880).

So if anything, Cooley’s perspective on the Second Amendment reinforces our conclusion that those who were in the militia at the time of ratification had Second Amendment rights.

The idea that 18- to 20-year-olds are not virtuous in the same manner as felons is a strange contention. Why would the federal government and every state require unvirtuous

people to have guns and be in a militia? Especially if being unvirtuous means that you can, and probably should, be disarmed? It is also odd considering that this group has most other

rights, while at least some of those rights are circumscribed for felons. As a result, the government’s arguments fail to persuade us that 18- to 20-year-olds were considered unvirtuous and thus outside the Second Amendment’s protections. b. Historical gun regulations

Failing to find purchase in Founding-era militia laws or the age of majority, the government turns to historical gun regulations. See Nat’l Rifle Ass’n, 700 F.3d at 200–01.

While some gun regulations existed at the Founding, there were no regulations restricting minors’ ability to possess or purchase weapons until two states adopted such laws in 1856.

See id. at 202 n.14. The Fifth Circuit relied on the prevalence of gun-safety regulations around the Founding to conclude that “gun use and gun control have been inextricably intertwined” and that restrictions on particular groups were also allowed. Id. at 200–01.

Ironically, the very laws the Fifth Circuit relied on to permit general gun regulation were

rejected in Heller as justifications for broader regulations. Id. at 337–38 (Jones, J., dissenting) (rejecting the majority’s argument based on Heller). Compare id. at 200–01

(majority opinion), with Heller, 554 U.S. at 683–86 (Breyer, J., dissenting) (using similar laws as the panel majority in Nat’l Rifle Ass’n), and id. at 631–34 (majority opinion)

(rejecting those laws as not probative). Perhaps for that reason, the government does not force this argument but turns instead to laws that deal with age.

The government offers a long list of state laws that set various ages for purchasing or possessing guns. J.A. 51–71. At least at step one of our analysis, however, these laws carry little weight. Instead, the list reveals that near the time of ratification there were no laws restricting the sale of firearms to 18-year-olds. The earliest laws cited were passed over 60 years after ratification, and most were enacted after the Civil War.

According to the government, by the end of the nineteenth century, 18 states and D.C. in some way restricted gun ownership by those under age 21. Id. This was less than half of the 45 states admitted by 1899. Only 5 states and D.C. explicitly restricted gun ownership for those under 21; 50 the rest, the government argues, restricted gun ownership by “minors” and those under 21 were considered minors. ^[51] But the nineteenth-century cases that are typically cited to support such restrictions either do not deal with this issue, apply only to those under 18, do not address constitutional concerns, or rely on logic rejected by Heller. ^[52]

By the early 1920s, another state explicitly restricted gun ownership by those under

age 21 while two others did so by implication, bringing the grand total to 21 states and D.C. ^[53] So even if we consider laws from more than a century after the Founding era, a majority of states, 27, seem not to have restricted 18-year-olds from purchasing or possessing handguns; they instead restricted younger people. ^[54] In fact, the cited 1868

Oregon law affirmatively protects the gun rights of those older than 16. 55 As explained earlier, we focus the historical inquiry in step one on the time of ratification in 1791. Chester, 628 F.3d at 680 (citing Heller, 554 U.S. at 625); Moore, 702

F.3d at 935 (citing McDonald, 561 U.S. at 765 & n.14). At the time of ratification, there were no laws restricting minors’ possession or purchase of firearms. Most laws affecting minors post-date the Civil War, and the only two states to pass such laws before the Civil

War did so immediately before the war (over 60 years after ratification). See J.A. 51–71;

1856 Ala. Acts 17, 17; 1856 Tenn. Pub. Acts 92, 92. These laws from later time periods provide little insight into the Amendment’s original public meaning. Heller, 554 U.S. at

614; Nat’l Rifle Ass’n, 714 F.3d at 337, 339 (Jones, J., dissenting) (discounting late nineteenth- and early twentieth-century sources). And these laws were passed by states that were not bound by the Second Amendment. See Barron, 32 U.S. (7 Pet.) at 247–48;

New State Ice Co., 285 U.S. at 311; Lee, supra, at 22; Nelson Lund, The Second

Amendment, Heller, and Originalist Jurisprudence, 56 UCLA L. REV. 1343, 1354, 1359, 1368–69 (2009).

It would also be strange to rely on two southern laws restricting gun rights that were enacted before the Civil War given Congress’s grave concerns about southern states disarming freed Blacks during this period. Amar, supra, at 176; McDonald, 561 U.S. at

770–78. State laws passed decades after the ratification restricting gun ownership—at a

time when state laws were used to disarm disfavored groups—is weak evidence of the original scope of the Second Amendment.

* * *

A review of the Constitution’s text, structure, and history reveals that 18-year-olds

are covered by the Second Amendment. See Nat’l Rifle Ass’n, 714 F.3d at 337 (Jones, J., dissenting). Like most other constitutional rights, the Second Amendment has no explicit

age limit, and the most analogous constitutional rights apply equally to everyone. And when a constitutional right applies differently to minors, the age cutoff has consistently been set at 18, not 21.

Founding-era history confirms that the public understood the Second Amendment to cover 18-year-olds. At the time of ratification, every state and the federal government required 18-year-old men to enroll in the militia. Those in the militia, as Heller reminds us, were a subset of the political community known as “the people” that enjoyed Second

Amendment rights. 554 U.S. at 580. While most of the militia laws the government identifies still allowed 18-year-olds to join the militia, those that required enlistees to be

21 are few in number and distant in time from ratification. So the historical basis for those

18 and older having Second Amendment rights rests on firm ground.

Not only has the government failed to carry its burden to show that the regulated conduct was outside the Second Amendment’s scope, see Chester, 628 F.3d at 680–81;

Ezell, 651 F.3d at 702–03; Tyler, 837 F.3d at 688, but the Constitution’s text, structure, and history affirmatively prove that 18-year-olds are covered by the Second Amendment.

C. Step two: the laws do not pass intermediate scrutiny

Having found that 18-year-olds are protected by the Second Amendment, our precedent requires that we apply “an appropriate form of means-end scrutiny.” Chester, 628 F.3d at 680. The Second Amendment’s “core protection” is “the right of law-abiding, responsible citizens to use arms in defense of hearth and home.” Kolbe, 849 F.3d at 131

(quoting Heller, 554 U.S. at 634–35). When that right is “severely burden[ed],” we apply

the “‘demanding’” strict scrutiny test. Id. at 133, 138 (quoting City of Boerne v. Flores, 521 U.S. 507, 534 (1997)). But if the right is burdened only on the margins, intermediate scrutiny applies. Masciandaro, 638 F.3d at 470–71. Because we find that the laws do not even pass intermediate scrutiny, we need not decide how close to the core of the Second

Amendment these laws strike.

Intermediate scrutiny requires the government to show “a reasonable fit between the challenged regulation and a substantial government objective.” Chester, 628 F.3d at 683

(internal quotations marks omitted). 56 Various evidence can be mustered to meet this standard but “reference to legislative findings, academic studies, or other empirical data is necessary.” Tyler, 837 F.3d at 694; see also Ezell, 651 F.3d at 709 (lack of empirics

doomed the range ban). And while “case law, and even common sense” may be relied on, the government may not “rely upon mere ‘anecdote and supposition.’” Tyler, 837 F.3d at

694 (quoting United States v. Playboy Ent. Grp., Inc., 529 U.S. 803, 822 (2000)). The government must therefore offer more than speculation to connect a ban on 18- to 20-year- olds purchasing handguns from licensed dealers to a substantial interest. See 18 U.S.C.

§ 922(b)(1).

To begin, the government’s interests in preventing crime, enhancing public safety, and reducing gun violence are “not only substantial, but compelling.” Kolbe, 849 F.3d at

139; see Schall v. Martin, 467 U.S. 253, 264 (1984) (“The ‘legitimate and compelling state interest’ in protecting the community from crime cannot be doubted.”); Schenck v. Pro–

Choice Network of W. N.Y., 519 U.S. 357, 376 (1997) (referring to the “significant governmental interest in public safety”); see also Wrenn v. Dist. of Columbia, 864 F.3d

650, 655 (D.C. Cir. 2017) (“[G]un laws almost always aim at the most compelling goal—

saving lives—while evidence of their effects is almost always deeply contested.”). Thus, we look to whether the regulations are a reasonable fit to that interest.

At the summary judgment stage, the government only proffered evidence from the Congressional Record to support finding a reasonable fit between its compelling interests and the challenged laws. Before the district court, amici offered more contemporary evidence to address the intermediate-scrutiny analysis. The government, however, neither adopted nor relied on that evidence. So it is unclear whether we may rely on amici’s evidence on appeal. ^[57] But even if we consider amici’s evidence, it does not change our analysis, for amici merely offer contemporary evidence to reinforce the congressional record, and that evidence does not alter or remedy its flaws.

The basic logic of the argument is: (1) the brains of those under age 25 are not fully developed so those individuals are prone to reckless and emotional behavior; 58 (2) because of this underdevelopment, youth, especially those ages 18 to 20, commit a disproportionate

But unlike in Hobby Lobby, the amici’s arguments here were raised and adopted by the district court, and some of their evidence is in the record. J.A. 341–401; Hirschfeld, 417 F. Supp. 3d at 758–59 (depending on the Congressional Record and Fifth Circuit’s reasoning but noting that amici’s evidence supported the conclusion). As in Hobby Lobby, however, amici offer an “intensely empirical argument,” and it is not clear that the government shares their views.

amount of violence; 59 (3) handguns are the main tool used to commit that violence; 60 and (4) many of those handguns are bought from licensed dealers, so restricting that avenue of purchase for the relevant age group will reduce crime. ^[61] Amici then try to tie the bow by arguing that the passage of state gun laws has coincided with decreases in youth suicides and unintentional deaths. 62

Whatever credence these contentions may have, we find that this chain of logic suffers from two debilitating flaws: (1) a showing of disproportionate bad conduct by a group cannot justify categorical restrictions on rights when the percentage of the group engaged in the unwanted conduct is minuscule, and (2) the evidence does not sufficiently link purchases from licensed dealers to crimes committed by youth.

The Supreme Court case Craig v. Boren, 429 U.S. 190 (1976), guides much of our analysis. In the first case to announce the intermediate-scrutiny standard, the Court

enjoined an Oklahoma law that prohibited licensed sellers from “sell[ing], barter[ing] or giv[ing]” low-alcohol beer (3.2% alcohol) to males under the age of 21 and females under the age of 18 because the state could not show that the sex classification was “substantially related” to road safety. Id. at 191 n.1, 204 (quoting Okla. Stat. tit. ^[37], § 241 (1953); Okla.

Stat. tit. ^[35], § 245 (1972)). 63 The Court reviewed studies that showed that 18-, 19-, and 20-year-old males were disproportionately arrested for, killed by, and injured by drunk driving. Id. at 200–01. The Court accepted that only 0.18% of females compared to 2% of males had been arrested for drunk driving (over 10 times as many). Id. at 201–02. But it rejected this evidence as sufficient to satisfy intermediate scrutiny: “[I]f maleness is to serve as a proxy for drinking and driving, a correlation of 2% must be considered an unduly

tenuous ‘fit.’” Id. The Court also found it significant that the state did not “measure the use and dangerousness of 3.2% beer as opposed to alcohol generally” to tie the challenged restriction to its interest in reducing drunk driving. Id. at 203. So in total, the state’s

evidence did “not satisfy [the Court] that sex represents a legitimate, accurate proxy for the regulation of drinking and driving.” Id. at 204.

The same flaws plague the evidence here. ^[64] First, while 18- to 20-year-olds commit

a disproportionate amount of crime compared to other age groups, very few members of that group commit crime, making the correlation between age and gun crimes “an unduly

tenuous ‘fit.’” Id. at 201–02. Second, like the connection between low-alcohol beer and drunk driving, the connection between handgun purchases from licensed dealers—as opposed to the many other ways to obtain handguns—and crimes committed by 18- to 20-

year-olds has not been made. So just as sex was a poor proxy for drunk driving in Craig, age is a poor proxy for gun violence in this case.

1. Disproportionate rates cannot justify vastly over-inclusive group restrictions

The laws at issue here rest on the premise that young adults’ Second Amendment rights can be limited because they are responsible for a disproportionate amount of violent crime. But these laws over-inclusively restrict the rights of a large group of law-abiding citizens to target a tiny portion of them. In 1967, Congress was concerned about a

disproportionate increase in crimes committed by youth. But the disproportionate rate of crime committed by this group tells us nothing about the percentage of the group that

commits gun violence. Amici’s contemporary sources purport to show that 18-, 19-, and 20-year-olds make up around 4% of the population but account for around 9.4% of violent crime arrests (about 37,000 of the 400,000 arrests each year). 65 So while 18- to 20-year- olds may commit a disproportionate share of violent crime, an exceedingly small percentage, around 0.3% and definitely less than 1%, of the 13 million young adults in this group commit those crimes. ^[66] This alone cannot justify restricting the entire group’s rights.

What’s more, those calculations do not account for repeat offenders. In fact, a year after being released, over 40% of criminals are arrested again, and that number approaches

80% in the 6 years after release. ^[67] This means that an even smaller portion of the group is likely responsible for its disproportionate commission of violent crime.

Beyond its overbreadth, consider the implications of this argument. Men commit disproportionally more violent crime than women. ^[68] Does that mean that the Second

Amendment rights of men can be categorically restricted? See Volokh, supra note 49, at

1512–13. Treating the sexes differently would also receive intermediate scrutiny. So why

not? A similar argument could be made for any number of groups, such as those living in a specific geographic location. See Heller, 554 U.S. at 681–82 (Breyer, J., dissenting)

by the group’s total population (13,256,712) multiplied by the percentage of the total population that the FBI statistics covered (0.775). See Petition for Rehearing En Banc at 12 n.1, Nat’l Rifle Ass’n, 714 F.3d 334 (No. 11-10959) (citing FBI and census data). Plaintiffs’ initial brief in Nat’l Rifle Ass’n uses the same method to find that 0.64% of young adults were arrested for committing a violent crime in 2009. See Opening Br. at 55 n.17, Nat’l Rifle Ass’n, 700 F.3d 185 (No. 11-10959) (citing FBI and census data). These similar numbers confirm our general point that a tiny fraction of individuals in this age group commit violent crimes. Bureau of Justice Statistics, 2018 Update on Prisoner Recidivism: A 9-Year (arguing that because crime is higher in urban centers more restrictions are allowed); id. at

634 (majority opinion) (rejecting the dissent’s “judge-empowering ‘interest-balancing

inquiry’” that permits greater regulation in urban areas). And amici’s own evidence and arguments would support restricting 21- to 25-year-olds’ rights, if not older adults, as much as it supports drawing the line at age 21. See supra notes 58, 59. Is that acceptable, and if so, where is the line?

Craig explains why such laws would fail when it declared that “a correlation of 2% must be considered an unduly tenuous ‘fit.’” Craig, 429 U.S. at 201–02. The question is

not about the disparity between groups or the total crime each group commits, but the percentage of the group that uses a gun in an impermissible manner. That statistic gives us some insight into the likelihood that any one person in the group will use a gun

improperly and pose a danger to the public. Whatever the precise answer to that question, here it falls well below the unduly tenuous threshold from Craig.

The rights of more than 99% of a group cannot be restricted because a fraction of 1% commit a disproportionate amount of violent crime. And it is already illegal for felons, fugitives, drug users, and aliens unlawfully in the United States, among others, to buy firearms from licensed dealers. So the laws at issue by their nature prevent a more law- abiding, less dangerous subset of 18- to 20-year-olds from purchasing from a more regulated market. For reducing gun violence and crime, restricting a whole group that is almost entirely law-abiding is the definition of an unduly tenuous fit. And, because those who are young are also disproportionately the victims of crime, 69 preventing them from purchasing handguns implicates the self-defense core of the Second Amendment the most. ^[70]

The irony does not escape us that, under the government’s reasoning, the same 18- to 20-year-old men and women we depend on to protect us in the armed forces and who have since our Founding been trusted with the most sophisticated weaponry should

nonetheless be prevented from purchasing a handgun from a federally licensed dealer for their own protection at home. We refuse to accept this conclusion. These men and women

who, historically, have served either voluntarily or by conscription may not be read out of “the people” in the Second Amendment.

2. Handgun purchases from licensed dealers have not been connected to gun violence

The second major flaw with the laws at issue is that the licensed dealers whose sales

to 18- to 20-year-olds are the subject of the challenged regulations are not connected to the incidence of crime Congress sought to control. It is not enough to target guns generally and argue that less access to guns means less crime, as this would justify almost any restriction and eviscerate the Second Amendment. In Craig, the government could not

Bureau of Justice Statistics, Criminal Victimization, 2019, at 21, 45 (2019), (saved 69 as ECF opinion attachment).

merely contend that access to alcohol generally led to drunk driving by men to justify a sex-based restriction on low-alcohol beer purchases. 429 U.S. at 203. Instead, the Court was concerned that no objective evidence purported to “measure the use and dangerousness

of 3.2% beer as opposed to alcohol generally.” Id. The government needed to connect the low-alcohol beer to drunk driving to justify the disparate treatment. Id. at 202–03. So too here. Access to guns through licensed dealers—rather than access to handguns generally— must be sufficiently connected to 18- to 20-year olds’ use of those firearms to commit violence. Yet neither the Congressional Record nor amici’s evidence can draw this connection, further undermining any claim to a reasonable fit.

In its congressional findings, Congress concluded that concealable firearms had

been “widely sold by federally licensed importers and dealers to emotionally immature, or thrill-bent juveniles and minors prone to criminal behavior,” which caused an increase in crime. Pub. L. No. 90–351, tit. IV, § 901(a)(6), 82 Stat. at 226; see also id. § 901(a)(3), 82

Stat. at 225. 71 The basis for that conclusion appears to be the hearing testimony of IRS Commissioner Sheldon Cohen. During one hearing, Commissioner Cohen stated that

“[t]he vast majority, in fact, almost all of these firearms, are put into the hands of juveniles by importers, manufacturers, and dealers who operate under licenses issued by the Federal

Government.” Federal Firearms Act: Hearings Before the H. Comm. on Ways and Means, 89th Cong. 31 (1965) (testimony of Sheldon S. Cohen); see also Federal Firearms Act:

Hearings Before the Subcomm. to Investigate Juvenile Delinquency of the S. Comm. on the Judiciary, 90th Cong. ^[57] (1967) (testimony of Sheldon S. Cohen). The government and amici rely on this statement to connect guns bought from licensed dealers to crimes committed by young adults. But there are several problems with this line of reasoning.

First, Commissioner Cohen’s testimony notes that “almost all” firearms in the hands of minors—not that “almost all” guns used by minors to commit violent crime—came from a licensed dealer. Cohen only discusses access to firearms generally, which tells us little about the connection between the guns minors bought from licensed dealers and the guns minors use in crime.

Second, his testimony could be read to make the point that “almost all” of the firearms have, at some point, come from federally licensed “importers, manufacturers, and dealers,” even if indirectly. That fact, while true, does little to show the regulations are a reasonable fit to the goal of reducing gun violence. Most guns, at some point, come from a manufacturer whether they are later sold on the black market or by a licensed dealer. So read, this statement provides no support for a link between handgun purchases directly

from licensed dealers and gun violence committed by young adults. So this reading of Cohen’s statement is far too broad to tell us anything meaningful.

The only way Commissioner Cohen’s statements could support this causal link is if he meant that “almost all” firearms that (1) are used by minors to commit crimes (2) were

directly sold to those minors by “importers, manufacturers, and dealers.” But nothing in the Congressional Record says that, and Commissioner Cohen’s statements fall considerably short of supporting such a contention.

To begin, while the IRS regulated licensed dealers, this information seems beyond

Commissioner Cohen’s experience. He was not in law enforcement, did not know from experience where guns used in crimes come from, and presented no studies or expertise from law enforcement to support his contention. Federal Firearms Act: Hearings Before the H. Comm. on Ways and Means, 89th Cong. 31 (1965) (testimony of Sheldon S. Cohen).

Interestingly, Commissioner Cohen did cite studies and the experiences of law enforcement to support many of his other statements, but he was brief and conclusory in his contention

that almost all guns in the hands of minors came from licensed dealers. Id. While the studies sought in Craig may not always be necessary, one person’s unsupported, inexpert, and conclusory testimony is not enough to establish a link between licensed dealers and gun crimes.

It is true that we sometimes give weight to Congress’s “predictive judgments” under

Turner Broadcasting System Inc. v. FCC, 512 U.S. 622, 665–66 (1994). But this does not mean that we should blindly abdicate our obligation to review Congress’s actions under heightened scrutiny. Indeed, as we have made clear “[o]ur obligation is simply ‘to assure that, in formulating its judgments, [the legislature] has drawn reasonable inferences based on substantial evidence.’” Kolbe, 849 F.3d at 140 (emphasis added) (quoting Turner, 512

U.S. at 666). As Turner explained, “deference afforded to legislative findings does ‘not foreclose our independent judgment of the facts bearing on an issue of constitutional law.’”

512 U.S. at 666 (quoting Sable Commc’ns of Cal., Inc. v. FCC, 492 U.S. 115, 129 (1989));

see also Moore, 702 F.3d at 942 (noting that the “legislative facts” used to determine the constitutionality of statutes do “not present factual questions for determination in a trial”

but are instead questions of law that judges resolve outside the normal strictures of summary judgment or a motion to dismiss).

Congress’s conclusion depends on Commissioner Cohen’s testimony, which itself is conclusory and rests on no evidence. But the government “must supply actual, reliable evidence,” such as empirical studies, not mere “speculation” even if facially reasonable.

Ezell, 651 F.3d at 709; see also Tyler, 837 F.3d at 694. Yet all the government offers to connect licensed dealers to young-adult crime is a conclusory finding by Congress based on a conclusory statement from someone who is not an expert on this issue. That is not

“substantial evidence,” and it is not due deference. Turner, 512 U.S. at 666. Congress may not pass intermediate scrutiny by relying on unsupported conclusory testimony to justify its desired outcome. As the statements are merely conclusory, they are owed no special deference, lest we edge perilously close to rational-basis review, a standard Heller prohibits.

Furthermore, one of the contemporary studies provided by amici affirmatively undermines the connection between handguns bought from licensed dealers and violent crime. ^[72] That study reflects self-reporting from 253 inmates convicted of offenses

committed with any firearm. Vittes et al., supra note 61, at 26–31. Of those inmates, 13.4% claimed to have bought the firearm used to commit the crime from a gun store or pawn shop. Id. at 30. And only 11.8% of inmates in the study obtained their gun from that

source legally. Id. 73 But unlike another study conducted by the Bureau of Justice Statistics, this study does not appear to account for whether the reported gun store or pawn shop was

licensed or whether the purchase was made under the inmate’s own name. A Bureau of Justice Statistics study suggests that 1.9% of all prisoners possessing a firearm bought it from a retailer, and 7–8% of inmates who possessed a firearm during their offense bought the gun under their own name from a licensed firearm dealer while 1–2% used another name at the licensed dealer and 1–2% bought the gun from an unlicensed seller at a retail source. ^[74] As a result, very few inmates of any age bought the firearm used at licensed dealers generally, much less legally.

Both studies substantially undermine Commissioner Cohen’s claim that “[t]he vast majority, in fact, almost all” firearms in the hands of minors come from licensed dealers.

Federal Firearms Act: Hearings Before the H. Comm. on Ways and Means, 89th Cong. 31

(1965) (testimony of Sheldon S. Cohen). Instead, these studies show that few guns used in crimes come from licensed dealers and those guns are rarely transferred from the dealer

directly to the criminal. The majority come from friends and family, the black market, the street, or some other unlicensed source. ^[75] Commissioner Cohen’s statement was already insufficient to support the connection between licensed dealers selling firearms to minors and minors using those firearms to commit crimes, and these studies show that his contentions do not have legs to stand on.

This defect raises an under-inclusivity problem. While under-inclusivity matters much less than over-inclusivity, as Congress does not have to combat every facet of a problem at once, a law that so under-inclusively addresses the government’s interest by targeting one of the least problematic avenues of purchasing a handgun raises serious

questions about whether the law actually furthers that interest. Compare Buckley v. Valeo, 424 U.S. ^[1], 105 (1976), with Nat’l Inst. of Fam. & Life Advoc. v. Becerra, 138 S. Ct. 2361, 2376–77 (2018). For example, in Sorrell v. IMS Health Inc., 564 U.S. 552, 572–73 (2011), the Supreme Court found that a restriction on sharing prescriber information was too

narrow to further the interest of physician confidentiality because it allowed the information to be given to anyone, just not for marketing purposes. The Court noted that a broader “more coherent” policy might withstand scrutiny, but the narrowness of the law as it stood burdened rights without meaningfully furthering the government’s stated interests. Id.

The same is true here. At most, amici’s study suggests that 11.8% of inmates who used guns to commit crimes bought them legally from licensed dealers. But even closing

off access to licensed dealers does not mean that the 11.8% would no longer obtain guns, as they would likely move to unregulated markets, like most other criminals, without a background check or other safety measures. In fact, the study itself concludes that

“criminals avoid the regulated gun market of licensed sellers and prefer the largely unregulated market involving unlicensed sellers . . . . The lack of regulation of firearm sales by unlicensed sellers is likely to significantly limit the government’s ability to keep firearms from prohibited individuals.” Vittes et al., supra note 61, at 30. Indeed, only 20% of inmates who could legally purchase a weapon from a licensed dealer chose that avenue rather than another more clandestine route. Id. Even aside from Commissioner Cohen’s testimony, these studies undermine any argument that there is a connection between licensed dealers and firearms used in crime. This is especially true for otherwise legally purchased firearms from licensed dealers, which is exactly what this law bans for 18- to

20-year-olds.

We previously relied on amici’s study to conclude that “very few offenders

purchased a weapon from a federal firearms dealer, in large part because of the background-check requirement.” Hosford, 843 F.3d at 169. There, we upheld a licensing regime in part because the government established that pushing buyers into a regulated

market reduced crime because the regulated market requires background checks, tracking, oversight, and more. Id. Yet strangely, the government here wants to push 18- to 20-year- olds away from licensed dealers and into the unregulated markets where most guns used for criminal purposes are obtained and where government oversight is more difficult.

Congress justified these restrictions to increase parental oversight of firearm purchases in order to reduce youth criminality, but Congress and the government have no evidence that parental oversight is effective, much less preferable, to the protections provided by licensed dealers. See Pub. L. No. 90-351, tit. IV, § 901(a)(3), 82 Stat. at 225;

S. Rep. No. 90-1097, at 79; 114 Cong. Rec. ^[12],309 (statement of Sen. Dodd). In fact, Congress went to pains to assure the country that minors could still have various guns and possess handguns through other means. See 114 Cong. Rec. ^[12],309 (statement of Sen.

Dodd); S. Rep. No. 90-1097, at 79; S. Rep. No. 89-1866, at 58; Pub. L. No. 90–351, tit.

IV, § 901(b), 82 Stat. at 226. And the ban “perversely assures that when such young adults obtain handguns, they do not do so through licensed firearms dealers, where background checks are required, see 18 U.S.C. § 922(t), but they go to the unregulated market.” Nat’l Rifle Ass’n, 714 F.3d at 346 (Jones, J., dissenting). These laws leave youth with three

options: (1) receive their gun as a gift from their parents or others (not an option for many), (2) buy it privately (this law does not provide parental oversight of those sales), or (3) obtain it illegally. This incentivizes young adults to get guns in less regulated ways, all without assuring parental input. So these laws seem to do little to increase parental guidance while pushing youth to unlicensed markets without background checks.

Finally, it is unclear whether these laws have been effective at all. While we

recognize that it would be difficult to determine the effectiveness of these laws now, Congress could have considered the effectiveness of similar state laws before passing these laws. In fact, one study found that these laws did not influence youth crime rates.^[76]

Amici’s evidence fares no better. Their primary study shows a decrease in youth suicides and accidental deaths after the massive 1994 Crime Bill that, in part, banned firearm possession for those under 18. 77 But that study has no nexus to 18- to 20-year-old adults.

It is hard to draw any conclusions on the effect of a handgun-purchase ban from this narrow study of a broader ban. And the study admits that state laws that prohibit handgun possession for those under a certain age have been ineffective. Gius, supra note 62, at 169; see also Daniel W. Webster et al., The Case for Gun Policy Reforms in America 8 (2012)

(noting that evaluations of laws setting a minimum legal age for handgun purchases or possession “have failed to find any beneficial effects of these laws on either juvenile

homicide victimization or youth suicide”). The study also tells us nothing about crime and very little about the cause of the decrease in suicides, as it admits that there were many other gun-control regulations that could have been contributing causes. Gius, supra note

622, at 173. At best, amici’s study shows a potential correlation between one part of a much broader law and fewer suicides and accidental deaths decades after the law was passed. And “the share of violent crime arrests among the 18- to 20-year age group has increased, and the use of guns by that group is still disproportionately high.” Nat’l Rifle

Ass’n, 714 F.3d at 346 (Jones, J., dissenting). Neither the government nor amici can show that the burden the challenged laws impose on 18- to 20-year-olds’ rights has led to any meaningful or measurable positive effects. This highlights the lack of a reasonable connection between licensed dealers and gun crimes.

These laws are both over- and under-inclusive in contravention of the principles espoused in Craig. As a result, the laws lack the reasonable fit required to pass intermediate scrutiny.

* * *

Those of good faith on both sides of the gun debate wish to protect lives and uphold

individual rights. We appreciate the seriousness of gun violence in this country and applaud Congress’s laudable desire to curb senseless violence. But we also recognize that the Second Amendment embodies a fundamental, pre-existing right that enables “the people” to preserve their own life, liberty, and property. Striking a balance between those interests is a difficult exercise that draws intense passions on both sides. And that is for good reason.

But while Congress—or judges—may have struck a different balance long after ratification, that role is foreclosed to us by the balance that the Founders chose. We cannot now second-guess or undermine their choice. History makes clear that 18- to 20-year-olds were understood to fall under the Second Amendment’s protections. Those over 18 were universally required to be part of the militia near the ratification, proving that they were considered part of “the people” who enjoyed Second Amendment rights, and most other

constitutional rights apply to this age group. And Congress may not restrict the rights of an entire group of law-abiding adults because a minuscule portion of that group commits a disproportionate amount of gun violence. Congress’s failure to connect handgun purchases from licensed dealers to youth gun violence only serves to highlight the law’s “unduly tenuous ‘fit’” with the government’s substantial interests. Craig, 429 U.S. at 202.

Eighteen- to twenty-year-olds have Second Amendment rights, and the challenged

laws impermissibly burden those rights. As a result, we vacate the district court’s grant of the motion to dismiss, reverse the denial of summary judgment, and remand for further proceedings.

VACATED, REVERSED, AND REMANDED Appendix 1: Pre-Ratification Militia Laws Colony/State Age Citations An Act for Forming, Regulating, and Conducting the Military Connecticut 16 Force of this State (1786), in ACTS AND LAWS OF THE STATE OF CONNECTICUT, IN AMERICA 144, 144, 150 (1786). An Act for Establishing a Militia Within this State §§ 2, 5 Delaware 18 (1778). An Act for Regulating the Militia of the State, and for Georgia 16 Repealing the Several Laws Heretofore Made for that Purpose (1786). Maryland 16 An Act to Regulate the Militia § 2 (1777). An Act for Regulating and Governing the Militia of the Commonwealth of Massachusetts, and for Repealing All Laws Heretofore Made for that Purpose § 2 (1785), in THE Massachusetts 16 PERPETUAL LAWS OF THE COMMONWEALTH OF MASSACHUSETTS, FROM THE COMMENCEMENT OF THE CONSTITUTION, IN OCTOBER 1780, TO THE LAST WEDNESDAY IN MAY, 1789, at 338, 339 (1789). An Act for Forming and Regulating the Militia Within this State, and for Repealing All the Laws Heretofore Made for that Purpose (1786), in THE LAWS OF THE STATE OF NEW- HAMPSHIRE, TOGETHER WITH THE DECLARATION OF New Hampshire 16 INDEPENDENCE: THE DEFINITIVE TREATY OF PEACE BETWEEN THE UNITED STATES OF AMERICA AND HIS BRITANNIC MAJESTY: THE CONSTITUTION OF NEW-HAMPSHIRE, AND THE CONSTITUTION OF THE UNITED STATES, WITH ITS PROPOSED AMENDMENTS 356, 357 (1792). An Act for the Regulating, Training, and Arraying of the Militia, and for Providing More Effectually for the Defence and Security of the State § 10 (1781), in ACTS OF THE FIFTH GENERAL ASSEMBLY OF THE STATE OF NEW-JERSEY, AT A SESSION BEGUN AT TRENTON ON THE 24TH DAY OF OCTOBER, New Jersey 16 1780, AND CONTINUED BY ADJOURNMENTS 39, 42 (1781); An Act for the Better Regulating the Militia § 1 (1777), in ACTS OF THE GENERAL ASSEMBLY OF THE STATE OF NEW-JERSEY, AT A SESSION BEGUN AT PRINCETON ON THE 27TH DAY OF AUGUST 1776, AND CONTINUED BY ADJOURNMENTS 26, 26 (1777). An Act to Regulate the Militia (1786), in 1 LAWS OF THE STATE OF NEW YORK, COMPRISING THE CONSTITUTION, AND THE New York 16 ACTS OF THE LEGISLATURE, SINCE THE REVOLUTION, FROM THE FIRST TO THE FIFTEENTH SESSION, INCLUSIVE 227, 227 (1792).

North Carolina 16 An Act to Establish a Militia in this State § 2, (1777), in ACTS OF ASSEMBLY OF THE STATE OF NORTH CAROLINA 1, 1 (1777). An Act to Regulate the Militia of the Commonwealth of Pennsylvania § 2 (1777), in 9 THE STATUTES AT LARGE OF Pennsylvania 18 PENNSYLVANIA FROM 1682 TO 1801, at 75, 77 (1903); An Act for the Regulation of the Militia of the Commonwealth of Pennsylvania § 3 (1780), in 10 THE STATUTES AT LARGE OF PENNSYLVANIA FROM 1682 TO 1801, at 144, 146 (1904). The Act for Better Forming, Regulating and Conducting the Military Force of this State (1779), in AT THE GENERAL ASSEMBLY OF THE GOVERNOR AND COMPANY OF THE STATE OF RHODE-ISLAND, AND PROVIDENCE-PLANTATIONS, BEGUN Rhode Island 16 AND HOLDEN AT SOUTH-KINGSTOWN, WITHIN AND FOR THE STATE AFORESAID, ON THE LAST MONDAY IN OCTOBER, IN THE YEAR OF OUR LORD ONE THOUSAND SEVEN HUNDRED AND SEVENTY-NINE, AND IN THE FOURTH YEAR OF INDEPENDENCE 29, 29. An Act for the Regulation of the Militia of this State (1782), in ACTS PASSED AT A GENERAL ASSEMBLY, BEGUN AND HOLDEN South Carolina 16 AT JACKSONSBURGH, IN THE STATE OF SOUTH-CAROLINA 20, 20. An Act Regulating the Militia of the State of Vermont (1787), Vermont 16 in STATUTES OF THE STATE OF VERMONT, PASSED BY THE LEGISLATURE IN FEBRUARY AND MARCH 1787, at 94, 94 (1787). An Ordinance for Raising and Embodying a Sufficient Force, for the Defence and Protection of this Colony (1775), in 9 WILLIAM WALLER HENING, THE STATUTES AT LARGE; BEING A COLLECTION OF ALL THE LAWS OF VIRGINIA, FROM THE FIRST SESSION OF THE LEGISLATURE, IN THE YEAR 1619, at 9, 16 (1775); 16–17 (1821); An Act to Amend and Reduce into One Act, the Virginia 18 (1785) Several Laws for Regulating and Disciplining the Militia, and Guarding Against Invasions and Insurrections § 3 (1785), in 12 WILLIAM WALLER HENING, THE STATUTES AT LARGE; BEING A COLLECTION OF ALL THE LAWS OF VIRGINIA, FROM THE FIRST SESSION OF THE LEGISLATURE, IN THE YEAR 1619, at 9, 10 (1823).

Appendix 2: Post-Ratification Militia Laws Colony/State Age Citations An Act for Forming and Conducting the Military Force of this State, Conformable to the Act of Congress, Passed the Eighth Day of May, A.D. 1792, Which Is as Follows:—“An Act More Connecticut 18 Effectually to Provide for the National Defence, by Establishing an Uniform Militia Throughout the United States” § 1 (1792), in ACTS AND LAWS OF THE STATE OF CONNECTICUT IN AMERICA 298, 298–99 (1796). An Act for Establishing the Militia in this State § 1 (1793), in 2 LAWS OF THE STATE OF DELAWARE FROM THE FOURTEENTH Delaware 18 DAY OF OCTOBER, ONE THOUSAND SEVEN HUNDRED, TO THE EIGHTEENTH DAY OF AUGUST, ONE THOUSAND SEVEN HUNDRED AND NINETY-SEVEN 1134, 1134 (1797). An Act to Revise and Amend the Militia Law of this State, and to Adapt the Same to the Act of the Congress of the United States, Passed the Eighth Day of May, One Thousand Seven Hundred and Ninety–Two, Entitled “An Act More Effectually Georgia 18 to Provide for the National Defence, by Establishing and Uniform Militia Throughout the United States” § 9 (1792), in DIGEST OF THE LAWS OF THE STATE OF GEORGIA 348, 350 (1802). An Act to Regulate and Discipline the Militia of this State Maryland 18 pmbl. (1793), in LAWS OF MARYLAND, NOVEMBER SESSION 1793 (1793). An Act for Regulating and Governing the Militia of the Commonwealth of Massachusetts, and for Repealing All Laws Heretofore Made for that Purpose; Excepting an Act Intitled “An Act for Establishing Rules and Articles for Governing the Troops Stationed in Forts and Garrisons, Within this Massachusetts 18 Commonwealth, and also the Militia, When Called into Actual Service” § 2 (1793), in ACTS AND LAWS, PASSED BY THE GENERAL COURT OF MASSACHUSETTS, BEGUN AND HELD AT BOSTON, IN THE COUNTY OF SUFFOLK, ON WEDNESDAY THE TWENTY-NINTH DAY OF MAY, ANNO DOMINI, 1793, at 289, 290 (1793). An Act for Forming and Regulating the Militia Within This State, and for Repealing All the Laws Heretofore Made for that Purpose (1792), in THE LAWS OF THE STATE OF NEW New Hampshire 18 HAMPSHIRE, PASSED AT A SESSION OF THE HONORABLE GENERAL-COURT, BEGUN AND HOLDEN AT EXETER, NOVEMBER 1972, at 441, 441 (1793).

An Act for Organizing and Training the Militia of this State § 4 New Jersey 18 (1792), in ACTS OF THE SEVENTEENTH GENERAL ASSEMBLY OF THE STATE OF NEW JERSEY 824, 825 (1792). An Act to Organize the Militia of this State (1793), in LAWS OF New York 18 THE STATE OF NEW YORK, PASSED AT THE SIXTEENTH SESSION OF THE LEGISLATURE 440, 440. An Act for Establishing a Militia in this State § 1 (1786), in THE LAWS OF NORTH-CAROLINA 18, 18 (amended by An Act to Carry into Effect an Act of Congress, Entitled, “An Act More Effectually to Provide for the National Defence, by North Carolina 18 Establishing an Uniform Militia Throughout the United States,” Also to Amend an Act, Passed at Fayetteville, in the Year One Thousand Seven Hundred and Eighty Six, Entitled, “An Act for Establishing the Militia in this State” (1793)). An Act for the Regulation of the Militia of the Commonwealth Pennsylvania 18 of Pennsylvania § 1 (1793), in THE STATUTES AT LARGE OF PENNSYLVANIA FROM 1682 TO 1801, at 454, 455 (1909). Rhode Island 18 * An Act to Organize the Militia Throughout the State of South Carolina, in Conformity with the Act of Congress (1794), in ACTS AND RESOLUTIONS OF THE GENERAL ASSEMBLY, OF THE STATE OF SOUTH-CAROLINA, PASSED IN APRIL, 1794, at 1, 2 South Carolina 18 (1794) (evidencing a shift from the former militia age of 16 mandated in An Act for the Regulation of the Militia of this State (1782), in ACTS PASSED AT A GENERAL ASSEMBLY OF THE STATE OF SOUTH-CAROLINA, &C, 20, 20). An Act, for Regulating and Governing the Militia of this State § 1 (1797), in 2 THE LAWS OF THE STATE OF VERMONT, Vermont 18 DIGESTED AND COMPILED INCLUDING THE DECLARATION OF INDEPENDENCE, THE CONSTITUTION OF THE UNITED STATES, AND OF THIS STATE 122, 122 (1808). Virginia 18 **

* The most relevant Founding-era law we could find from Rhode Island set the militia age at 18 in 1794. An Act to Organize the Militia of this State (1794), in AT THE GENERAL ASSEMBLY OF THE GOVERNOR AND COMPANY OF THE STATE OF RHODE-ISLAND AND PROVIDENCE-PLANTATIONS, BEGUN AND HOLDEN BY ADJOURNMENT AT EAST- GREENWICH, WITHIN AND FOR THE STATE AFORESAID, ON THE LAST MONDAY IN MARCH, IN THE YEAR OF OUR LORD ONE THOUSAND SEVEN HUNDRED AND NINETY-FOUR, AND OF INDEPENDENCE THE EIGHTEENTH 14, 14–15 (1794) (reprinting the federal Militia Act and organizing the militia in line with federal law setting the age at 18). Prior laws had set the militia age at 16. The Act for Better Forming, Regulating and Conducting the Military Force of this State (1779), in AT THE GENERAL ASSEMBLY OF THE GOVERNOR AND COMPANY OF THE STATE OF RHODE-ISLAND, AND PROVIDENCE-PLANTATIONS, BEGUN AND HOLDEN AT SOUTH-KINGSTOWN, WITHIN AND FOR THE STATE AFORESAID, ON THE LAST MONDAY IN OCTOBER, IN THE YEAR OF OUR LORD ONE THOUSAND SEVEN HUNDRED AND SEVENTY-NINE, AND IN THE FOURTH YEAR OF INDEPENDENCE 29, 29; An Act, Regulating the Militia in this Colony, in THE CHARTER, GRANTED BY HIS MAJESTY, KING CHARLES II. TO THE GOVERNOR AND COMPANY OF THE ENGLISH COLONY OF RHODE-ISLAND AND PROVIDENCE-PLANTATIONS, IN NEW-ENGLAND, IN AMERICA 179, 179 (1767).

** Preceding ratification, Virginia required 18-year-olds to join the militia and bring their own arms. An Act to Amend and Reduce into One Act, the Several Laws for Regulating and Disciplining the Militia, and Guarding Against Invasions and Insurrections § 3 (1785), in 12 WILLIAM WALLER HENING, THE STATUTES AT LARGE; BEING A COLLECTION OF ALL THE LAWS OF VIRGINIA, FROM THE FIRST SESSION OF THE LEGISLATURE, IN THE YEAR 1619, at 9, 10, 12 (1823). This law also created a separate company for 18- to 25-year- olds that trained more often than the rest of the militia. Id. § 3, in HENING, supra, at 14– 15. Following ratification, Virginia’s militia law did not mention age or equipment, focusing more on the organization by county. An Act for Regulating the Militia of this Commonwealth (1792), in A COLLECTION OF ALL SUCH ACTS OF THE GENERAL ASSEMBLY OF VIRGINIA, OF A PUBLIC AND PERMANENT NATURE, AS ARE NOW IN FORCE 282, 282–90 (1803). But the act did not alter the age requirements set in 1785 and kept the light company of 18- to 25-year-olds. Like many other statutes at the time, however, Virginia’s law said that it was helping to “carry the [federal Militia Act] into effect.” Id. § 1, in A COLLECTION OF ALL SUCH ACTS, supra, at 282. And the federal Militia Act required 18-year-olds to enlist and bring their own arms.

WYNN, Circuit Judge, dissenting:

Today, my good colleagues in the majority break new ground by invalidating a modest and long-established effort to control gun violence. The majority holds that

Congress may not enact a law making 21 the minimum age to purchase handguns from federally licensed gun dealers.

But the majority’s decision to grant the gun lobby a victory in a fight it lost on

Capitol Hill more than fifty years ago is not compelled by law. Nor is it consistent with the proper role of the federal judiciary in our democratic system. Cf. Kolbe v. Hogan, 849 F.3d

114, 149–150 (4th Cir. 2017) (en banc) (Wilkinson, J., concurring) (arguing that because

“[n]o one really knows what the right answer is with respect to the regulation of firearms,” we federal judges ought not “[d]isenfranchis[e] the American people on this life and death

subject” by arrogating to ourselves “decisions that have been historically assigned to other, more democratic, actors”).

To be sure, the Second Amendment’s right to keep and bear arms is an exceptional right, just not in the way the majority imagines. According to my colleagues, even though all individual constitutional rights are subject to limitations, the Second Amendment risks being relegated to a disfavored “second-class status.” Maj. Op. at 3.

While they are not alone in this concern, see, e.g., Silvester v. Becerra, 138 S. Ct.

945, 945 (2018) (Thomas, J., dissenting from denial of certiorari) (arguing that the Second

Amendment protects a “disfavored right” that lower courts have treated “cavalierly”), I do not share it. Indeed, in a country that boasts a Congress, bench, bar, academy, and electorate that are all attentive to the prerogatives of gun owners, where many may conceal their weapons, 1 carry them openly, 2 or “stand their ground,” 3 and where civilian gun ownership rates are second to none, 4 the majority’s second-class status concern is simply surreal. ^[5]

No, the Second Amendment is exceptional not because it is uniquely oppressed or imperiled, but rather because it is singularly capable of causing harm. As other courts have recognized, while there are dangers inherent in other constitutionally protected rights—

1 At present, “[e]very state—as well as the District of Columbia—allows the carrying of concealed weapons in some form.” Giffords Law Center to Prevent Gun Violence (“Giffords”), Concealed Carry, https://giffords.org/lawcenter/gun-laws/policy- areas/guns-in-public/concealed-carry/. 2 Giffords, Open Carry, https://giffords.org/lawcenter/gun-laws/policy-areas/guns- in-public/open-carry/ (noting that “many states now place few or no restrictions on open carry”). 3 Giffords, Stand Your Ground, https://giffords.org/lawcenter/gun-laws/policy- areas/guns-in-public/stand-your-ground-laws/, (“A majority of states (30) have now enacted Stand Your Ground laws applicable in all public places.”). 4 Aaron Karp, Estimating Global Civilian-Held Firearms Numbers: Briefing Paper, Small Arms Survey, June 2018 at 4, http://www.smallarmssurvey.org/fileadmin/docs/T- Briefing-Papers/SAS-BP-Civilian-Firearms-Numbers.pdf (establishing that the United States has the highest rates of civilian possession of firearms, both in terms of absolute numbers and in terms of the number of firearms per 100 residents, and that our nearest peers on both fronts—India and Yemen, respectively—each come in a distant second). 5 Cf. Linda Greenhouse, Uncomfortable Timing for a Supreme Court Gun Fight, N.Y. Times (Apr. ^[22], 2021), https://www.nytimes.com/2021/04/22/opinion/supreme- court-gun-rights-second-amendment.html (“I’ve always been puzzled by the notion that any limitation on the Second Amendment converts it into some sort of second-class right. I can’t think of a constitutional right that the Supreme Court has interpreted as absolutely unlimited.”).

like the rights to speak and assemble—the Second Amendment alone protects a direct and lethal right to endanger oneself and others. ^[6]

We are fortunate, then, that our courts have hitherto been careful to construe the Second Amendment in a manner that gives the people latitude to try to control gun violence through the democratic process. In the Fourth Circuit, as in nearly every other circuit, we protect the individual right to keep and bear arms while honoring the public’s “right not to be shot,” Lowy, supra, at 204, by applying a two-step analysis. ^[7]

6 E.g., Bonidy v. U.S. Postal Serv., 790 F.3d 1121, 1126 (10th Cir. 2015) (“The risk inherent in firearms and other weapons distinguishes the Second Amendment right from other fundamental rights that have been held to be evaluated under a strict scrutiny test, such as the right to marry and the right to be free from viewpoint discrimination, which can be exercised without creating a direct risk to others.”); Heller v. Dist. of Columbia, 801 F.3d 264, 283 (D.C. Cir. 2015) (“Heller III”) (Henderson, J., concurring in part and dissenting in part) (“Although our Second Amendment precedent draws on First Amendment and voting-rights cases, the right to bear arms is meaningfully different from the rights to speak and vote. . . . [T]he reality of gun violence means our constitutional analysis should incorporate deference to the legislature.” (internal citations omitted)); Piszczatoski v. Filko, 840 F. Supp. 2d 813, 816 (D.N.J. 2012) (“[The Second Amendment] privilege is unique among all other constitutional rights to the individual because it permits the user of a firearm to cause serious personal injury—including the ultimate injury, death—to other individuals, rightly or wrongly.”), aff’d sub nom. Drake v. Filko, 724 F.3d 426 (3d Cir. 2013); see also Jonathan Lowy & Kelly Sampson, The Right Not to Be Shot: Public Safety, Private Guns, and the Constellation of Constitutional Liberties, 14 Geo. J.L. & Pub. Pol’y 187, 194 (2016) (“No other right exposes the public to such grave risks of lethal, imminent harm.”). 7 On a public right to live—or “right not to be shot,” see Lowy, supra, at 204–05. On the ubiquity of analyzing Second Amendment challenges through a two-step framework, see Maj. Op. at 11 n.5 (collecting cases). The Eighth Circuit, which has not expressly addressed the issue, appears to be the lone holdout in not adopting the Second Amendment two-step inquiry. But see United States v. Adams, 914 F.3d 602, 607–08 (8th Cir. 2019) (Kelly, J., concurring in the judgment) (faulting the majority for not applying the “sensible, two-pronged approach” employed by the other circuits for analyzing gun rights claims).

Under this approach, we first ask “whether the challenged regulation ‘burdens or regulates conduct that comes within the scope of the Second Amendment.’” Harley v. Wilkinson, 988 F.3d 766, 769 (4th Cir. 2021) (quoting United States v. Chester, 628 F.3d

673, 680 (4th Cir. 2010)). Then, “[i]f the challenged regulation satisfies this first prong, or if we assume without deciding that the regulation meets this requirement, we turn to perform under our second prong a ‘means-end’ review, in which we consider the regulation under the appropriate level of constitutional scrutiny.” Id. (quoting Chester, 628 F.3d at

680).

We have also applied a “simplifie[d]” version of this analysis for “prohibitions

deemed ‘presumptively lawful’” by the Supreme Court in District of Columbia v. Heller, 554 U.S. 570 (2008). United States v. Hosford, 843 F.3d 161, 165 (4th Cir. 2016). As

discussed below, Heller deemed longstanding conditions and qualifications on the commercial sale of firearms to be presumptively lawful, and our circuit precedent in United

States v. Hosford establishes that facial challenges against such regulations must fail.

Purporting to apply our two-step approach, the majority breaks new ground and creates a circuit split by striking down a modest effort at gun control passed more than fifty years ago that does not prevent young adults from purchasing, possessing, or using handguns.

Conversely, I would hold that the challenged provisions are facially valid

longstanding conditions and qualifications on the commercial sale of arms. In the alternative, I would assume on the first prong of our two-part approach that the challenged provisions burden conduct protected by the Second Amendment. I would then join the Fifth Circuit in holding on prong two that the challenged provisions easily survive intermediate scrutiny.

Because the majority holds otherwise, I respectfully dissent. In doing so, I consign neither “the Second Amendment [n]or 18- to 20-year-olds to a second-class status.” Maj.

Op. at 3.

I.

In Heller, the Supreme Court recognized the “right of law-abiding, responsible citizens to use arms in defense of hearth and home.” 554 U.S. at 635. However, Heller itself emphasized that it was “not meant ‘to clarify the entire field’ of Second Amendment jurisprudence,” Hosford, 843 F.3d at 165 (quoting Heller, 554 U.S. at 635), and that it should not “be taken to cast doubt on longstanding prohibitions on the possession of firearms by felons and the mentally ill, or laws forbidding the carrying of firearms in sensitive places such as schools and government buildings, or laws imposing conditions and qualifications on the commercial sale of arms,” Heller, 554 U.S. at 626–27. Heller dubbed these sorts of firearm restrictions “presumptively lawful” and explained that it identified these measures “only as examples” and that its “list d[id] not purport to be exhaustive.” Id. at 627 n.26. This was no idle prattle. Cf. Carlton F.W. Larson, Four Exceptions in Search of a Theory: District of Columbia v. Heller and Judicial Ipse Dixit, 60 Hastings L.J. 1371, 1372 (2009) (criticizing Heller’s discussion of presumptively lawful regulatory measures as “ad hoc” and “patchy” but nonetheless acknowledging that to the extent the discussion was dicta, it was “dicta of the strongest sort”). In fact, in its only significant post-Heller Second

Amendment case, McDonald v. City of Chicago, 561 U.S. 742 (2010), the Supreme Court

“repeat[ed] [its] assurances” that its holding in Heller “did not cast doubt on such longstanding regulatory measures as . . . [‘]laws imposing conditions and qualifications on the commercial sale of arms.’” 561 U.S. at 786 (plurality opinion) (quoting Heller, 554

U.S. at 626–27).

As the majority recognizes, we have fashioned “a more streamlined analysis” for Second Amendment challenges to regulatory measures deemed presumptively lawful in Heller. Maj. Op. at 13–14. Specifically, we have held that facial challenges to such

measures fail. ^[8] Hosford, 843 F.3d at 166 (“Hosford’s facial challenge fails if the prohibition against unlicensed firearm dealing is the type of regulation deemed

‘presumptively lawful’ in Heller.”). Because Plaintiff Natalia Marshall 9 brought a facial, 8 In dicta, the majority suggests that a facial challenge to a presumptively lawful measure will not always fail. See Maj. Op. at 13–14 (“If a law falls within [Heller’s list of presumptively lawful measures], we conduct a more streamlined analysis. This often effectively ends the inquiry and upholds the law as facially valid. But not always.” (internal citations omitted)). In support of that statement, the majority cites to our precedent in Hamilton v. Pallozzi, 848 F.3d 614, 624–25 (4th Cir. 2017), and United States v. Pruess, 703 F.3d 242, 245–46 (4th Cir. 2012). But it is unclear how these cases, both of which involved as-applied challenges to felon-in-possession laws, support the majority’s position. To the extent that the majority believes a facial, rather than as-applied, challenge to a presumptively lawful measure can succeed, that belief squarely contradicts Hosford. See Hosford, 843 F.3d at 167 (“[T]he prohibition against unlicensed firearm dealing is a longstanding condition or qualification on the commercial sale of firearms. As a result, Hosford’s facial Second Amendment challenge fails.”). To be sure, plaintiffs can still make as-applied challenges to presumptively lawful measures, id., but no such challenge was brought in this case. ^[9] As the majority finds, this case is moot as to Plaintiff Tanner Hirschfeld.

and not an as-applied challenge, we can therefore resolve this case quickly if we determine that the challenged provisions are presumptively lawful as (1) longstanding (2) conditions and qualifications (3) on the commercial sale of arms. ^[10] Id. at 166–67.

Here, the challenged provisions indisputably cover “only the commercial sale of firearms.” Id. at 166. Young adults aged 18 to 20 may still possess and use handguns, may still receive handguns as gifts, and can even purchase handguns through unlicensed, but

otherwise legal, private sales. ^[11] Nat’l Rifle Ass’n of Am., Inc. v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 700 F.3d 185, 189–90 (5th Cir. 2012) (“NRA”). In other

words, in enacting the challenged provisions, Congress was careful not to burden use, possession, or non-commercial sales. ^[12]

The challenged provisions are also longstanding. Importantly, “a regulation can be deemed ‘longstanding’ even if it cannot boast a precise founding-era analogue.” NRA, 700

F.3d at 196–97, 202–03 (collecting authorities and concluding that the same provisions at

10 While the district court conducted both a facial and an as-applied analysis, “[w]e look to the scope of the relief requested to determine whether a challenge is facial or as- applied in nature.” AFSCME Council 79 v. Scott, 717 F.3d 851, 862 (11th Cir. 2013) (citing Doe v. Reed, 561 U.S. 186, 194 (2010)). Here, Marshall requested a broad declaration that the challenged provisions were unconstitutional and requested that they be enjoined generally, not simply as to her. Her claim was thus facial in nature. The majority agrees. ^[11] And, of course, they can still purchase and use long guns. See 18 U.S.C. § 922(b)(1). 12 While every sale is “commercial” in the sense that an exchange of value occurs, we distinguish between commercial sales—that is, sales made by those who “regularly sell firearms, not owned for personal use, in the course of trade or business for the principal purpose of profit”—and non-commercial sales, that is, infrequent, non-professional sales as may occur, for example, between friends. Hosford, 843 F.3d at 166.

issue in this case are “consistent with [the] longstanding, historical tradition” of “age-based restrictions on the purchase of firearms”). Indeed, much more recent regulations can qualify. See Hosford, 843 F.3d at 166–67 (finding a licensing requirement for firearm dealers to be “longstanding” when it had been required by the federal government since

1938); Drake, 724 F.3d at 431–34 (relying on late nineteenth- and early twentieth-century authorities to conclude that a New Jersey law requiring applicants to demonstrate a

“justifiable need” to obtain a permit to publicly carry a handgun was a “longstanding”

regulation); Silvester v. Harris, 843 F.3d 816, 818 (9th Cir. 2016); id. at 831 (Thomas, C.J., concurring) (contending that California’s ten-day waiting period to buy a gun was

“longstanding” because it had been on the books in some form since 1923).

This rule follows from Heller, which listed “prohibitions on the possession of firearms by felons” as examples of the type of “longstanding” regulations that it did not

“cast doubt on.” Heller, 554 U.S. at 626–27. This is an illustrative example because, as the D.C. Circuit observed in Heller II, “states did not start to enact [prohibitions on the possession of firearms by felons] until the early 20th century.” Heller v. Dist. of Columbia, 670 F.3d 1244, 1253 (D.C. Cir. 2011) (citing C. Kevin Marshall, Why Can’t Martha

Stewart Have a Gun?, 32 Harv. J.L. & Pub. Pol’y 695, 708 (2009)). Furthermore, “Heller

considered firearm possession bans on . . . the mentally ill to be longstanding” despite the fact that the “current version[] of th[is] ban[] [is] of mid-20th century vintage,” dating back to 1968. NRA, 700 F.3d at 196–97 (citing United States v. Skoien, 614 F.3d 638, 640–41

(7th Cir. 2010) (en banc) (explaining that the federal ban on possession by the mentally ill was passed in 1968)).

Here, the challenged provisions are longstanding because they, just like the federal ban on possession of firearms by the mentally ill, were passed in 1968—over fifty years ago—and because similar provisions making 21 the minimum age to purchase or use certain firearms have been commonplace for more than a century. As the Fifth Circuit noted in NRA, “by the end of the 19th century, nineteen States and the District of Columbia had enacted laws expressly restricting the ability of persons under 21 to purchase or use particular firearms, or restricting the ability of ‘minors’ to purchase or use particular firearms while the state age of majority was set at age 21.” Id. at 202. Furthermore, by

1923, “twenty-two States and the District of Columbia had made 21 the minimum age for the purchase or use of particular firearms.” Id. Considering this pedigree, the challenged provisions are sufficiently longstanding. ^[13]

The question of whether the challenged provisions constitute “conditions and qualifications” is somewhat trickier because the Supreme Court “did not explain what precisely it meant by the phrase.” Silvester v. Harris, 41 F. Supp. 3d 927, 962 (E.D. Cal.

13 The majority contends that the Fourteenth Amendment was enacted, in part, to protect the rights of recently emancipated slaves to bear arms in the face of southern efforts to disarm them and that we should therefore be cautious about using “Civil-War-era state laws that restricted the right to bear arms as evidence of the scope of the Second Amendment right at the time of the Founding.” Maj. Op. at 24. However, here, I merely seek to understand whether laws restricting the ability of young adults to buy guns are “longstanding.” The existence of antebellum and Reconstruction-era laws regulating the use of firearms by young adults is undoubtedly relevant to this inquiry. Furthermore, while I leave for another day the question of what—if any—role America’s racial history ought to play in our Second Amendment jurisprudence, I am not convinced that the Second Amendment is best understood as a guarantor of the rights and liberties of freedmen and their descendants. See generally Carol Anderson, The Second: Race and Guns in a Fatally Unequal America (2021).

2014), rev’d on other grounds, 843 F.3d 816. Fortunately, the terms “conditions” and “qualifications” have uncontroversial common meanings. Silvester, 843 F.3d at 830

(Thomas, C.J., concurring). A condition “is [s]omething established or agreed upon as a requisite to the doing or taking effect of something else” and a qualification is a “condition precedent that must be complied with for the attainment of a status,” such as “for admission to an office.” Id. (alteration in original) (quoting dictionary definitions). Thus, a condition and qualification on the commercial sale of arms is simply a prerequisite or requirement that a purchaser or seller must comply with before engaging in a commercial gun transaction.

In Hosford, we further refined this definition by distinguishing such conditions and qualifications—which, if they are also longstanding, are presumptively lawful—from regulations “so prohibitive as to . . . functional[ly] prohibit[]” the sale of firearms. 843 F.3d at 166. The challenged provision at issue in Hosford was a licensing scheme that required

a prospective firearm dealer to obtain a license by submitting an application, paying a fee, establishing lawful premises for selling firearms, and being at least 21 years old. Id. In upholding the scheme, we distinguished it from the functional prohibitions on the sale of weapons discussed in two other cases. Id. (citing Illinois Ass’n of Firearms Retailers v. City of Chicago, 961 F. Supp. 2d 928 (N.D. Ill. 2014); Teixeira v. Cnty. of Alameda, 822

F.3d 1047 (9th Cir. 2016), rev’d on other grounds, 873 F.3d 670, 673 (9th Cir. 2017) (en banc)).

The first, Illinois Association of Firearms Retailers, invalidated a Chicago law that totally banned the sale and transfer of all firearms except by inheritance. 961 F. Supp. 2d at 930–931. In the second, a divided panel of the Ninth Circuit reasoned that a zoning ordinance that flatly banned the sale of guns would be subject to stricter scrutiny than an ordinance that merely regulated the location of gun stores. Teixeira, 822 F.3d at 1059–60.

However, both the majority and the partially dissenting judge agreed that the zoning ordinance at issue—which prohibited establishing a gun store within 500 feet of a

residentially zoned district—merely placed conditions and qualifications on the commercial sale of arms. ^[14] Id. at 1050, 1058; id. at 1064 (Silverman, J., concurring in part and dissenting in part).

In contrast, the challenged provisions here impose no functional prohibition on the sale, or, to flip to the other side of the coin, purchase, of firearms. Rather, they merely impose a limited condition and qualification on whom a subset of sellers (federally licensed dealers) may sell a subset of firearms—handguns—to (not those aged 18 to 20). Put another

way, while the law at issue in Heller flatly banned handgun possession, and while the Chicago ordinance at issue in Illinois Association of Firearms Retailers banned all commercial sales (and even gifts) of guns generally, the challenged provisions in this case permit federally licensed dealers to sell handguns to most adults, and young adults aged 18 to 20 can still purchase handguns from other sources. Moreover, unlike regulations that

amount to functional prohibitions on the sale of arms, this condition and qualification is of a temporary duration as to potential purchasers—it evaporates once the would-be purchaser turns 21.

14 The panel majority nevertheless concluded that the law was not “longstanding” and therefore was not presumptively lawful under Heller. Teixeira, 822 F.3d at 1058.

I would therefore hold that the challenged provisions are presumptively lawful as longstanding conditions and qualifications on the commercial sale of arms and thus are facially constitutional.

The majority disagrees. While it does not contest the fact that the challenged provisions are longstanding and regulate only the commercial sale of arms, it contends that

Heller’s presumptively lawful category of conditions and qualifications is limited “to laws that place conditions and qualifications on the seller, like licensing,” and “does not encompass a ban on a class of people seeking to purchase handguns from a qualified seller.”

Maj. Op. at 16.

There are two problems with this claim. First, as explained above, the challenged provisions are neither a ban on selling handguns nor even a ban on young adults buying handguns. Second, and more fundamentally, the challenged provisions cannot be excluded from the ranks of presumptively lawful conditions and qualifications on the grounds that they place a greater burden on buyers than sellers because every restriction on a seller is also a restriction on a buyer, and vice versa.

Furthermore, the majority asserts that the challenged provisions cannot be conditions and qualifications on the commercial sale of arms because “[t]here is nothing a law-abiding 18- to 20-year-old can do to buy a handgun from a licensed dealer except wait until she turns 21.” Id. at 14. I take the majority to mean that a condition and qualification must be something the burdened actor can overcome through her own initiative, rather than simply through aging. That is wrong under our precedent. As noted, one of the conditions and qualifications to obtain a firearm license in Hosford was that the prospective dealer

“be at least twenty-one years old.” 843 F.3d at 166.

The challenged provisions are therefore longstanding conditions and qualifications on the commercial sale of arms. As such, Marshall’s facial challenge fails. Case closed.

II.

Even if Marshall’s claim did not fail as a facial challenge to a longstanding condition and qualification on the commercial sale of arms, it would still fail under our normal two- step analysis of Second Amendment challenges.

A.

Once again, at the first step of our two-step analysis, we ask “whether the challenged regulation ‘burdens or regulates conduct that comes within the scope of the Second

Amendment.’” Harley, 988 F.3d at 769 (quoting Chester, 628 F.3d at 680). “This historical

inquiry seeks to determine whether the conduct at issue was understood to be within the scope of the right at the time of ratification.” Chester, 628 F.3d at 680. “If it was not, then the challenged law is valid.” Id.

Here, the majority has done careful historical research and has assembled persuasive

evidence that young adults aged 18 to 20 had Second Amendment rights at the time the amendment was ratified. In light of this evidence, and given this Circuit’s “prudent” practice of not “impart[ing] a definitive ruling” at the first step of our two-step inquiry, I assume for purposes of this opinion that the challenged provisions burden or regulate conduct within the scope of the Second Amendment. Woollard v. Gallagher, 712 F.3d 865, 875 (4th Cir. 2013); see also, e.g., Harley, 988 F.3d at 769. I therefore proceed to the second step of our two-step inquiry.

B.

Having assumed that the challenged provisions burden conduct within the scope of the Second Amendment, I move to our second step: considering the challenged provisions

“under the appropriate level of constitutional scrutiny.” Harley, 988 F.3d at 769.

Because Heller ruled out the possibility of applying rational-basis review to Second

Amendment challenges, 554 U.S. at 628 n.27, we must “select between strict scrutiny and intermediate scrutiny,” Chester, 628 F.3d at 682. In making this selection, courts consider

“the nature of the conduct being regulated and the degree to which the challenged law burdens the right.” Id.

“[I]ntermediate scrutiny is the appropriate standard [where the challenged

provision] does not severely burden the core protection of the Second Amendment, i.e., the right of law-abiding, responsible citizens to use arms for self-defense in the home.” Kolbe, 849 F.3d at 138. A law severely burdens this core right if it “effectively disarm[s] individuals or substantially affect[s] their ability to defend themselves.” Id. at 139 (holding that a law that did not have this disarming effect was subject to intermediate scrutiny)

(quoting N.Y. State Rifle & Pistol Ass’n, Inc. v. Cuomo, 804 F.3d 242, 260 (2d Cir. 2015)).

Extremely few laws are so burdensome. In fact, “there has been near unanimity in the post-

Heller case law that, when considering regulations that fall within the scope of the Second

Amendment, intermediate scrutiny is appropriate.” Mai v. United States, 952 F.3d 1106, 1115 (9th Cir. 2020) (quoting United States v. Torres, 911 F.3d 1253, 1262 (9th Cir.

2019)).

Here, the majority assumes that intermediate scrutiny applies. This is a safe assumption. As the Fifth Circuit held in NRA, the challenged provisions should be subject to intermediate scrutiny because “they do not prevent 18-to-20-year-olds from possessing and using handguns in defense of hearth and home,” and “regulate commercial sales through an age qualification with temporary effect.” 700 F.3d at 206–07 (internal quotation marks omitted); accord Nat’l Rifle Ass’n of Am., Inc. v. McGraw, 719 F.3d 338, 342, 348

(5th Cir. 2013) (applying intermediate scrutiny to Texas law preventing adults under 21 from carrying handguns in public). Simply put, a temporary restriction on which sellers that young adults can buy handguns from does not severely burden the Second

Amendment’s core right to at-home self-defense. Accordingly, intermediate scrutiny applies. ^[15]

15 Plaintiff Marshall’s arguments to the contrary are not compelling. First, she posits that we must apply strict scrutiny because the challenged provisions constitute a “universal prohibition on the . . . purchase of new or un-owned handguns and . . . ammunition,” and that this prohibition is particularly suspect given the importance of handguns to self- defense. Opening Br. at 34–36. While the Fourth Circuit has shown special solicitude for handguns in the past, see Kolbe, 849 F.3d at 138–139, Marshall provides no authority for a constitutionally protected right to purchase a new weapon for self-defense. Her attempt to jury-rig a class of arms out of what is really a subset of handguns—those purchased from a federally licensed dealer—is unpersuasive. See id. at 138 (“The initial weakness in the plaintiffs’ theory is that the banned assault weapons cannot fairly be said to be a ‘class’ like that encompassing all handguns, in that the banned assault weapons are just some of the semiautomatic rifles and shotguns in existence.”). Second, Marshall argues that we should decline to apply traditional levels of heightened scrutiny altogether and should instead focus on the “text, history, and tradition” of the Second Amendment. Opening Br. at 31–32. Binding Fourth Circuit precedent forecloses this approach. E.g., Kolbe, 849 F.3d at 133.

C.

“Under the standard of intermediate scrutiny, the government bears the burden of establishing a reasonable fit between the challenged law and a substantial governmental objective.” Harley, 988 F.3d at 769. This fit need not be perfect or even the least intrusive means of accomplishing the desired objective, United States v. Staten, 666 F.3d 154, 162

(4th Cir. 2011), and a law “may meet this standard despite being overinclusive in nature,”

Harley, 988 F.3d at 769.

“[T]he Constitution does not mandate a specific method by which the government must satisfy its burden under heightened judicial scrutiny.” United States v. Carter, 669

F.3d 411, 418 (4th Cir. 2012). Rather, as the majority recognizes, “[v]arious evidence can

be mustered” to show a reasonable fit, including legislative findings, academic studies, case law, and even common sense, though “anecdote” and “supposition” alone will not cut it. Maj. Op. at 61 (collecting authorities) (internal quotation marks omitted).

Furthermore, in deciding whether a firearm regulation is reasonably related to a substantial government interest, we owe “substantial deference to the predictive judgments of [Congress.]” Kachalsky v. Cnty. of Westchester, 701 F.3d 81, 97 (2d Cir. 2012) (quoting

Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 195 (1997)); accord Schrader v. Holder, 704 F.3d 980, 990 (D.C. Cir. 2013) (same); Drake, 724 F.3d at 436–37 (same). We owe such substantial deference “[i]n the context of firearm regulation” because “the legislature

is ‘far better equipped than the judiciary’ to make sensitive public policy judgments (within constitutional limits) concerning the dangers in carrying firearms and the manner to combat those risks.” Kachalsky, 701 F.3d at 97 (quoting Turner Broad. Sys., Inc. v. FCC, 512 U.S.

622, 665 (1994) (plurality opinion)); accord Kanter v. Barr, 919 F.3d 437, 451 (7th Cir.

2019) (same).

Finally, as the Supreme Court has explained, the “quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised,” with justifications that are

plausible and not novel requiring less empirical support. Nixon v. Shrink Mo. Gov’t PAC, 528 U.S. 377, 391 (2000); see id. at 391–95 (finding an affidavit from a state senator, citations to newspaper articles, and a statewide vote in support of campaign expenditure restrictions to be sufficient evidence to support a state law limiting campaign expenditures).

Where, as here, intermediate scrutiny applies and the justifications offered by Congress are

“both familiar and plausible,” they need only be supported by “minimal” empirical evidence. Carter, 669 F.3d at 418 (citing Satellite Broad. & Commc’ns Ass’n v. FCC, 275

F.3d 337, 355, 360 (4th Cir. 2001)). In fact, “[e]ven when applying strict scrutiny—

requiring a more taxing proof threshold than the one we apply here—the government may, in appropriate circumstances, carry its burden by relying solely on history, consensus, and simple common sense.” Id.(emphasis added) (internal quotation marks omitted) (quoting

Fla. Bar v. Went For It, Inc., 515 U.S. 618, 628 (1995)).

With this framework in place, I will now consider whether the challenged provisions can withstand intermediate scrutiny. As set forth below, they can.

*** The majority’s intermediate-scrutiny analysis gets off to a good start. As it correctly

finds, “the government’s interests in preventing crime, enhancing public safety, and reducing gun violence are ‘not only substantial, but compelling.’” Maj. Op. at 62 (quoting

Kolbe, 849 F.3d at 139). But the majority—essentially applying strict scrutiny in sheep’s clothing—errs in concluding that no reasonable fit exists between the challenged provisions and these laudable objectives. To demonstrate the reasonableness of this fit, I first consider the provisions’ text and legislative history, which establish that Congress designed a bespoke solution to a specific problem. I then consider scientific research on the immaturity of young adults, modern crime and accident statistics, and the state of state law on young adults and firearms, all of which further confirm the reasonableness of Congress’s approach. i. The challenged provisions contain their own limiting principles

The place to start a consideration of the fit between the challenged provisions and the government’s interest in public safety is with the legislative text—that is, with what the law says it prohibits, and for how long. It cannot be repeated often enough: the provisions

the majority strikes down today do not keep young adults from owning or using handguns, and they impose no limitations whatsoever on the use of long guns. See 18 U.S.C.

§ 922(b)(1), (c)(1); 27 C.F.R. §§ 478.99(b)(1), 478.96(b), 478.124(a). Little wonder, then, that the plaintiffs in this case never alleged that they have been unable to acquire handguns.

In this regard, the challenged provisions are much less restrictive than the restrictions imposed on various classes of individuals by 18 U.S.C. § 922(g), which forbids the possession of “any firearm or ammunition” by a member of a restricted class, generally for life. ^[18] U.S.C. § 922(g) (emphasis added). And yet, appellate courts—including this

Court—regularly uphold provisions of § 922(g) against Second Amendment challenges. ^[16]

My point is not to compare law-abiding young adults to stalkers, fugitives, or other

categories of persons prevented by federal criminal law from possessing firearms. Rather, I bring up those provisions to emphasize how much less restrictive the challenged provisions are than those lifelong bans. This “narrow ambit” weighs in favor of upholding the provisions. NRA, 700 F.3d at 205 (explaining that these same provisions’ narrowness militated against applying strict scrutiny).

Moreover, the limited restrictions that the challenged provisions do impose are only

imposed for a short time. All restrictions vanish when the clock strikes midnight and the young adult in question turns 21. As we explained in United States v. Carter, statutes with

“limited temporal reach” are inherently “less intrusive than other statutes that impose a permanent prohibition on the possession of firearms.” 669 F.3d at 419. This case involves time-limited provisions that “Congress tailored . . . to cover only the time period during

16 See, e.g., United States v. Moore, 666 F.3d 313, 319 (4th Cir. 2012) (§ 922(g)(1)’s ban on felons possessing firearms is facially constitutional); United States v. Carter, 750 F.3d 462, 470 (4th Cir. 2014) (§ 922(g)(3)’s ban on possession by those unlawfully using or addicted to a controlled substance is constitutional); Mai, 952 F.3d at 1109 (§ 922(g)(4)’s ban on possession by those previously committed to a mental institution is constitutional as applied to plaintiff); United States v. Carpio-Leon, 701 F.3d 974, 975 (4th Cir. 2012) (§ 922(g)(5)’s ban on possession by those unlawfully in the United States is constitutional); United States v. Jimenez, 895 F.3d 228, 231 (2d Cir. 2018) (§ 922(g)(6)’s ban on possession by those dishonorably discharged from the military withstands intermediate scrutiny as applied to plaintiff); United States v. Mahin, 668 F.3d 119, 120 (4th Cir. 2012) (§ 922(g)(8)’s ban on possession while subject to a domestic violence protective order is constitutional); Harley, 988 F.3d at 769–70 (§ 922(g)(9)’s ban on possession by individuals convicted of domestic violence misdemeanors is constitutional as applied to plaintiff (citing Staten, 666 F.3d at 168)).

which it deemed [the prohibited] persons to be dangerous.” Id. And this feature of the challenged provisions “contributes to [their] proportionality” under heightened review. Id.

Thus, under our precedent—and indeed, under common sense—a time-limited regulation of Second Amendment rights is less burdensome and better-tailored to meet a substantial government objective than is a lifelong restriction.

In sum, the challenged provisions contain their own built-in limitations of scope and duration. These limitations are powerful evidence that the provisions burden no more conduct than is necessary to meet their objectives and so are constitutional. ii. The legislative history of the challenged provisions strongly supports their constitutionality

In addition to the legislative text, a provision’s legislative history is one of the “wide

range of sources” the government can rely on to “establish the fit between a regulation and a governmental interest” under intermediate scrutiny. Id. at 418. Here, the legislative

history of the challenged provisions reinforces that they are reasonably related to the government’s compelling interest in promoting public safety by limiting the ability of young adults to purchase handguns from federally licensed firearm dealers.

During the mid-1960s, “Congress conducted a multi-year investigation that revealed a causal relationship between the easy availability of firearms to young people under 21 and the rise in crime.” NRA, 700 F.3d at 207. This investigation was led by Connecticut

Senator Thomas Dodd and culminated in the enactment of the challenged provisions as part of the Omnibus Crime Control and Safe Streets Act and Gun Control Acts of 1968.

See generally Adam Winkler, Gunfight: The Battle Over the Right to Bear Arms in America, 248–53 (2011) (describing Dodd’s leadership role in Congressional efforts to regulate guns); William J. Vizzard, The Gun Control Act of 1968, 18 St. Louis Univ. Pub.

L. Rev. 79, 79–86 (1999) (same); Gabrielle Westcott, The Fight for the Gun Control Act of 1968, Univ. of Conn. Library Archives and Special Collections Blog, Sept. ^[15], 2016

[hereinafter Westcott Blog] (same). 17

Dodd’s legislative efforts started in 1961 when he began serving as head of the Juvenile Justice Subcommittee of the Senate Judiciary Committee. Vizzard, supra, at 80.

After two years of researching gun violence and in the immediate aftermath of President

Kennedy’s assassination, Dodd introduced a bill, S. 1975, that sought to address “the ease with which juveniles and criminals could anonymously purchase mail-order guns and thus

circumvent state laws regarding the sale of firearms.” Westcott Blog; see also Winkler, supra, at 248. This bill stalled and then died in committee. Vizzard, supra, at 80; Westcott

Blog.

Following President Johnson’s election in 1964, Dodd reintroduced his gun control bill at the start of the 89th Congress in January 1965. Vizzard, supra, at 80; Westcott Blog.

But soon, at the Johnson Administration’s behest, he introduced a new measure, Senate

Bill 1592, that called for prohibiting the sale of handguns to those under 21. See Vizzard, supra, at 80, 82; Westcott Blog. Though it would take three more years to become law, this prohibition was eventually enacted in the form of the challenged provisions.

17 Westcott’s piece is available at blogs.lib.uconn.edu/archives/2016/09/15/the- fight-for-the-gun-control-act-of-1968/.

I start with this scene-setting to emphasize a basic point: the challenged provisions were not a slapdash effort. Rather, they were the careful, deliberative “culmination of five

years of legislative effort and seven years of investigation on the part of Senator Dodd and the Subcommittee on Juvenile Delinquency.” Westcott Blog. And what did Senator Dodd and his congressional colleagues find at the culmination of their years of inquiry? They

found “a causal relationship between the easy availability of firearms other than a rifle or shotgun and juvenile and youthful criminal behavior.” Pub. L. No. 90-351, tit. IV, § 901(a)(6).

This conclusion tying young adults’ ability to access handguns with threats to public safety is adequately supported by the legislative record, provided that we view our job as looking for a reasonable fit between government action and objective and not as

“weigh[ing] conflicting evidence and mak[ing] policy judgments.” Kolbe, 849 F.3d at 140

(quoting Woollard, 712 F.3d at 881). After all, the latter “is the legislature’s job, not ours.”

Id. (quoting Woollard, 712 F.3d at 881). I consider Congress’s chief conclusions step by step.

Start with Congress’s determination that Americans under 21 pose a special danger to society and to themselves. As the majority notes, Congress had before it ample evidence that individuals under 21 were involved in a substantial share of the nation’s criminal

activity. See Maj. Op. at 5–6 & n.2 (acknowledging that at the time of the enactments, minors under 21 accounted for 64 percent of all arrests for serious crimes in the United

States and “35 percent of the arrests for the serious crimes of violence including murder, rape, robbery, and aggravated assault and 21 percent of the arrests for murder” (internal citations and quotation marks omitted)). For instance, Congress knew that in New York

City in 1962, 46.3 percent of all those arrested for felonies were under 21. S. Rep. 88-1340, at 15–16 (1964). Indeed, during the 1960s, many believed the country to be experiencing an epidemic of youthful criminality. See, e.g., Maj. Op. at 5–6 n.2 (quoting federal officials’ testimony that the “greatest growth of crime [at the time was] in the area of young people, juveniles and young adults”). This fear was hardly baseless. See, e.g., 109 Cong.

Rec. 13946 (1963) (statement of Sen. Dodd) (“[A]ccording to the FBI, during the 5 years between 1955 and 1960, when the juvenile population increased by 25 percent, the number of youngsters arrested for carrying guns went up nearly twice as fast.”); 114 Cong. Rec.

13322 (1968) (statement of Sen. Dodd) (“[S]ince 1960, arrests of juveniles [for robbery] have increased 55 percent.”). 18

In addition to worrying about youthful crime and gun violence, Congress was also concerned about self-harm and accidents involving firearms. Back in 1963, Senator Dodd lamented that “[o]f the more than 18,000 annual suicides [in the country], fully half [were] committed with the use of firearms.” 109 Cong. Rec. 13946. He raised the same concerns

18 While I recognize “the inherent danger of drawing causal connections between rates of arrest and incidences of criminal activity,” Powell v. Tompkins, 926 F. Supp. 2d 367, 393 (D. Mass. 2013), other statistics help confirm that America did indeed experience an increase in violent crime during the 1960s, lessening any concern about relying on arrest statistics, see, e.g., Federal Firearms Legislation: Hearings Before the Subcomm. to Investigate Juvenile Delinquency of the Sen. Comm. on the Judiciary, 90th Cong. ^[1] (1968) (statement of Senator Dodd) (citing FBI statistics establishing that between 1964 and 1967 murder by gun “increased 51 percent; aggravated assault by gun increased an incredible 84 percent; and armed robbery increased 57 percent”); Lauren-Brooke Eisen & Oliver Roeder, America’s Faulty Perception of Crime Rates, Brennan Ctr. for Just. (Mar. ^[16], 2015), https://www.brennancenter.org/our-work/analysis-opinion/americas-faulty-perception- crime-rates (“[T]he violent crime rate increased by 126 percent between 1960 and 1970.”).

during the final push to pass the Omnibus Crime Control and Safe Streets Act of 1968, stating on May 8, 1968 that “[t]he surest and easiest way to kill one’s self is with a firearm—especially when the firearm is immediately at hand. And when it comes to suicides by firearms, the discrepancy between the rate in this country and the rate in other civilized countries is equally astounding.” 114 Cong. Rec. 12305. He was joined in this

sentiment by Attorney General Nicholas Katzenbach, who argued that “[i]n a country in which half the 20,000 suicides of 1963 were committed by firearms, congressional action is called for now.” Proposed Amendments to The National Firearms Act and The Federal

Firearms Act: Hearings Before the Comm. on Ways and Means, 89th Cong. ^[73] (1965)

[hereinafter Ways and Means: 1965 Hearing] (statement of Attorney General Katzenbach).

Beyond suicide, Congress was also concerned with simple gun accidents and the toll they took on young people. For instance, Senator Dodd’s Juvenile Delinquency

Subcommittee heard testimony from a Pittsburg police officer who testified regarding a

“typical case” of gun injury in his city in which a 20-year-old bought a handgun, loaded it, and then “attempted to expel the cartridge by striking it with a hatchet which resulted in his being wounded in the right cheekbone.” S. Rep. 88-1340, at 17 (1964); see also id. at

27 (concluding that the “indiscriminate sale of [mail-order] firearms ha[d] resulted in accidental tragedies including loss of life and serious injury”).

Though the majority ignores this evidence, prominent supporters of the challenged provisions also highlighted the depressing pervasiveness of gun accidents while advocating for passage of the Crime Control and Safe Streets Act. See 114 Cong. Rec. 12299 (1968)

(statement of Sen. Fong) (estimating that in 1966, firearms were involved in 10,000 suicides and 2,600 accidental deaths and arguing that this “staggering toll” helped

demonstrate “the urgent need for a comprehensive [gun] law”); id. at 12303 (statement of Sen. Dodd) (arguing that Title IV, which housed the challenged provisions, “represent[ed] a long-overdue legislative reaction to the terrible toll of life and limb extracted each year in our country by firearms,” including an estimated “270,000 suicides, and 145,000 deaths

by accident” between 1900 and 1966); cf. Federal Firearms Act: Hearings Before the Subcomm. to Investigate Juvenile Delinquency of the Sen. Comm. on the Judiciary, 90th

Cong. ^[29], 31 (1967) (speech of Sen. Kennedy of Mass.) (arguing that Senate Bill 1—a legislative predecessor to the challenged provisions that also prevented those under 21 from buying handguns from federally licensed dealers—could help “avoid one murder, one suicide, one accident, or ten, or a hundred, or a thousand” and was “the best measure anyone ha[d] devised” to prevent crime, suicides, and gun accidents).

Next, in contemplating its legislative options for responding to the significant harm caused by people armed with guns in the form of crime, suicides, and accidents, Congress heard from law-enforcement professionals who believed that further regulating access to firearms would promote public safety. For example, FBI Director J. Edgar Hoover concluded that “those who claim that the availability of firearms is not a factor in murders

in this country are not facing reality.” Ways and Means: 1965 Hearing, at 73 (statement of Attorney General Katzenbach (quoting Hoover’s 1963 Law Enforcement Bulletin)).

Other law-enforcement professionals agreed that regulating youth access to firearms, in particular, would reduce gun injuries and deaths. Senator Dodd noted that “law enforcement experts virtually to a man agree that we can curb these serious increases in crimes of violence by young people, by restricting the availability of guns to them.” 114

Cong. Rec. 13643–44 (1968). In fact, broad swaths of the nation’s law-enforcement establishment favored stricter gun control measures than those Congress ultimately

enacted. See S. Rep. 90-1097, at 191 (1968) (statement of Sen. Tydings) (“The President, the Attorney General, the Director of the FBI, the President’s Commission on Law

Enforcement and the Administration of Justice, the International Association of Chiefs of Police, the American Bar Association, and state and local law enforcement officials all across the country have recommended federal firearms control legislation much more stringent than Title IV[, the provisions of the Crime Control and Safe Streets Act that barred those under 21 from purchasing handguns from federally licensed dealers,] provides.”).

Further, as the majority notes, “[l]aw enforcement officers from New York City, Los Angeles, St. Louis, Chicago, Philadelphia, and Atlanta provided Congress with statistics documenting the misuse of firearms by juveniles and minors.” Maj. Op. at 5–6

n.2 (internal quotation marks omitted). For instance, Herbert T. Jenkins, the chief of Atlanta’s Police Department, stated that of 404 recent aggravated assaults involving firearms in his city, about 93 percent were committed by individuals under the age of 21.

S. Rep. 89-1866, at 59 (1966) (from Individual Views of Senators Dodd, Bayh, Kennedy of Massachusetts, Tydings, Fong, Javits, Smathers, and Long of Missouri). And Chief

Jenkins further concluded that “[m]ost of these cases would have ended as a simple altercation, except for the ready availability of the gun.” Id. (emphasis added).

Supporters of the enactments believed that limiting youth access to firearms would be a particularly effective way to limit violent deaths because keeping guns out of the hands of young people could keep them from turning to crime or, if they did, could keep them from killing anyone while engaging in criminal activity. Notably, contemporary statistics

showed that 21% of people who were assaulted with a gun died because of the assault, compared with just 3% of individuals assaulted with another weapon. 109 Cong. Rec.

13946 (statement of Sen. Dodd). They also showed that in 1962, “more than 3,200 persons met violent deaths by a gun in the hand of a noncriminal.” Id. (emphasis added).

IRS Commissioner Sheldon Cohen and Attorney General Katzenbach also explained the link they saw between firearms, youth, and violence. According to Cohen— who, as IRS Commissioner, was the executive branch official then responsible for firearm regulation 19—the “easy availability of weapons” exacerbated the “tendency [of young

people, juveniles, and young adults] toward wild, and sometimes irrational . . . violent” and “deadly” behavior. Federal Firearms Act: Hearings Before the Subcomm. to Investigate

Juvenile Delinquency of the Sen. Comm. on the Judiciary, 90th Cong. ^[57] (1967) (testimony

of Commissioner Cohen). Likewise, Attorney General Katzenbach believed that while the 19 See Federal Firearms Act: Hearings Before the Subcomm. to Investigate Juvenile Delinquency of the Sen. Comm. on the Judiciary, 89th Cong. 29–30 (1965) [hereinafter Firearms Act: 1965 Hearings] (statement of Hon. Henry H. Fowler, Secretary of the Treasury) (explaining that Cohen, as IRS commissioner, was “responsible for the administration of the Federal Firearms Act” as the head of the IRS’s Alcohol and Tobacco Tax Division, the forerunner to the modern Bureau of Alcohol, Tobacco, Firearms and Explosives). Given Cohen’s subject-matter expertise in firearm regulation, the majority’s dismissal of his testimony for lack of personal law-enforcement experience cannot hold water. We must defer to Congress’s decision to trust someone with such clearly relevant credentials instead of conducting a kind of retroactive Daubert analysis. While courts of course have an obligation to serve as gatekeepers when it comes to expert testimony cited by the parties before us in the course of normal litigation, we also have an obligation to show a much greater level of deference when it is a coequal branch who has opted to rely on a particular expert when performing its constitutionally assigned duties.

gun control measures might only make it “a little harder” for “major criminals” to get guns, another purpose of the proposed measures was “to take guns away from people who might not otherwise become criminals,” and to ensure that if they did, they would “at least” not

“use guns to kill somebody in the process.” Ways and Means: 1965 Hearing, at 85

(emphasis added).

Senator Dodd echoed the Commissioner and Attorney General’s beliefs that keeping guns out of the hands of young adults would increase public safety. He argued:

Some of the opponents of the bill . . . have also argued that it is not guns that kill people, but the criminals behind the guns. And they have further argued that, criminals being what they are, they will succeed in obtaining weapons despite any laws which we may enact. The basic fallacy in this argument is [that] it lumps all gun criminals together in a single criminal category. It may be true that the hardened criminal . . . will succeed in obtaining weapons no matter what laws we may pass. But the great majority of those involved in gun crimes are not professional criminals. They are . . . amateur criminals, some of them one-time offenders, some of them “hopped- up” juveniles involved in their first breach of the law. In the case of these amateur criminals, it is frequently the weapon that kills rather than the criminal himself. The professional criminal—and I have known some, I have spent some part of my life in criminal work—will not attempt to hold up an old man for the paltry few dollars he may have in his pocket, and then, if he encounters any resistance, panic and fire. He does not operate that way. But every year there are hundreds of murders perpetrated in precisely this way by amateur criminals—by kids, youngsters, and juveniles. . . . The amateur criminal [is frequently] a person of inferior intelligence and inferior initiative who would find it very difficult to obtain a gun for himself if the law placed a few obstacles in his way.

114 Cong. Rec. 12307–08 (1968). 20

20 Senator Dodd had intimate, first-hand experience with gun crime. As a special agent for the FBI during the 1930s, he partook in a shootout between the G-Men and John Dillinger’s gang. Winkler, supra, at 248–49.

Agree or disagree with Senator Dodd’s assessment, there is no denying that it had

support at the time the challenged provisions were enacted. For instance, Curtis Brostron, St. Louis’s Chief of Police, testified that he supported “mak[ing] the acquisition of firearms more difficult,” in part because he believed young people would be less tempted to acquire weapons and to commit crime if obtaining a firearm were harder. Firearms Act: 1965

Hearings at 580–81 (statement of Chief Brostron). In short, Congress had ample evidence

before it that, to truly enhance public safety, it would need to do more than target “major criminals” and repeat offenders.

Having identified a specific problem—death and injuries from guns in the hands of young people—and a general solution—regulating youth access to firearms—Congress then began tailoring a narrow fix.

First, it focused only on handguns, which it termed “the most formidable and most frequently used tool of the criminal.” S. Rep. 89-1866, at 4 (1966). Congress reached this conclusion that handguns were uniquely the friend of criminals by relying on two sources of evidence: (1) “the existence in many States of laws controlling [handguns]” and (2)

“statistics showing its dominance as the weapon used in unlawful activities.” Id. First, as

Congress noted, many states and cities regulated handguns to a “much greater extent” than

long guns, suggesting a greater fear of their misuse. Id. Second, regarding crime statistics, Congress had before it FBI statistics and commentary from Director Hoover that

“handguns were used in 70 percent of . . . murders committed with firearms” and in most armed robberies involving a gun. Id. at 5. Thus, Congress concluded “[f]rom these statistics [and state and local regulations that] it [was] quite clear that the principal offender in the unlawful use of firearms is the handgun.” Id.

Congress then further connected the broader trends of youth violence and handgun violence together by studying the very specific trend of youth violence perpetrated with handguns. For instance, in 1964, Senator Dodd’s Senate Subcommittee on Juvenile

Delinquency learned that from 1960 to 1962, firearms were involved in approximately

seventy percent of all arrests in New York City of individuals under 21 for four kinds of violent felonies and that these firearms were mostly handguns. S. Rep. 88-1340, at 16

(1964).

Finally, having concluded that public safety could plausibly be addressed by regulations targeting youth access to handguns, Congress focused yet more narrowly on the specific problem of handguns purchased from federally licensed firearms dealers. See

NRA, 700 F.3d at 208 (“The legislative record . . . demonstrates that Congress was particularly concerned with the [role federally licensed firearm dealers played] in the crime problem.”). To this end, Congress heard from IRS Commissioner Cohen, who testified that the “misuse of firearms by juveniles” had “reached alarming proportions”; that “[t]he vast majority, in fact almost all of these firearms, are put into the hands of juveniles by . . . dealers who operate under licenses issued by the Federal Government”; and that “[t]he way to end this dangerous practice [was] to stop these Federal licensees from selling firearms to juveniles.” 21 Firearms Act: 1965 Hearings at 67 (statement of Commissioner Cohen).

Thus, in context, Commissioner Cohen’s remarks demonstrate his conclusion that the ability of young adults to buy firearms from federally licensed dealers was contributing to the nation’s crime problem. Congress agreed, and, combining this conclusion with its finding that handguns were the primary weapon of concern, concluded that handguns had

been “widely sold by federally licensed importers and dealers to emotionally immature, or thrill-bent juveniles and minors prone to criminal behavior,” leading to “youthful criminal behavior.” Pub. L. No. 90–351, tit. IV, § 901(a)(6).

To summarize, as the Fifth Circuit found in NRA, “Congress was focused on a

particular problem: young persons under 21, who are immature and prone to violence, easily accessing handguns, which facilitate violent crime, primarily by way of [federally licensed dealers].” 700 F.3d at 208. “Accordingly, Congress restricted the ability of young

persons under 21 to purchase handguns from [federally licensed dealers].” Id. Nothing in the majority’s opinion contests that Congress studied this exact issue and agreed on a solution designed to solve it. Applying intermediate scrutiny, and not strict scrutiny masquerading as such, this ought to be enough.

Instead, in targeting the chain of conclusions that led Congress to enact the challenged provisions, the majority trains its fire on the weakest link: the specific

21 While Commissioner Cohen did not clarify the exact age group he meant when he referred to “juveniles,” these remarks came during his broader endorsement of Senate Bill 1592, which, as discussed above, endeavored to prevent individuals under twenty-one from purchasing handguns.

connection between federally licensed dealers and handgun crime. To be sure, Commissioner Cohen’s testimony could have been more robust, and the legislative record does lack a study establishing that most guns used by young adults to commit crimes were sold to them directly by federally licensed dealers. But, for several reasons, I would not find this shortcoming in the record to be fatal to the challenged provisions.

First, the majority’s exclusive focus on the link between guns sold by federally licensed dealers to young adults and crime is too narrow. The legislative history establishes

that Congress was concerned not just with crime, but also with accidents and suicides. And common sense tells us that fewer law-abiding young adults will harm themselves with handguns if they cannot purchase them from federally licensed dealers. ^[22]

22 The commonsense idea that reducing access to lethal weapons will reduce firearm accidents and suicides is also supported by the scientific literature. See, e.g., Matthew Miller et al., Firearm Availability and Unintentional Firearm Deaths, 33 Accident Analysis & Prevention 477, 479–480 (2001) (finding that “children and adults, men and women, and members of all racial groups were significantly more likely to die from unintentional firearm injuries if they lived in states with more rather than fewer guns,” and that this “robust, positive and statistically significant association” between gun availability and fatal gun accidents persisted “even after controlling for state level poverty, urbanization and regional location”); David M. Studdert et al., Handgun Ownership and Suicide in California, 382 New Eng. J. of Med., 2220, 2228 (2020) (“Our study bolsters and extends the message from previous research: ready access to firearms, particularly handguns, is a major risk factor for suicide.”); Matthew Miller et al., Suicide Mortality in The United States: The Importance of Attending to Method in Understanding Population- Level Disparities in the Burden of Suicide, 33 Ann. Rev. of Pub. Health 393, 397, 401–03 (2012) (providing figures that establish that in 2001, approximately 84.9% of suicide-by- firearm attempts in America were fatal compared with just 4.1% of all non-firearm attempts and discussing the “potential for reducing the population-level suicide burden by reducing population-level access to lethal means”). A recent study following more than 26 million Californians over the age of 21 for up to 12 years dramatically illustrates this relationship between access to firearms and suicide. It found that despite having “lower rates of all-cause mortality than nonowners,” Second, the challenged provisions aimed to prevent crime not just by keeping handguns out of the hands of hardened criminals, but also by making it somewhat harder

for law-abiding young adults to obtain handguns and to then turn, gun in hand, to crime in the first place. The majority’s statistics on the percentages of gun crimes committed with legally purchased weapons do not address these potential crimes that may have been prevented by the challenged provisions. Such potential crimes are easy to imagine when

we consider Congress’s findings on the impulsivity and hotheadedness of youth. Thus, even assuming that relatively few young adults commit a gun crime with a handgun they purchased directly from a federally licensed dealer, it was reasonable for Congress to believe that the challenged provisions could help reduce gun crime.

Third, it is important to remember what we are supposed to be doing here:

“weighing a legislative judgment, not evidence in a criminal trial” or even an agency’s record in an administrative proceeding. Pena v. Lindley, 898 F.3d 969, 979 (9th Cir. 2018)

(explaining the proper application of intermediate scrutiny to a Second Amendment

challenge). When Congress enacts a statute, it “is not obligated . . . to make a record of the handgun owners committed suicide at much higher rates than nonowners. Studdert, supra, at 2220, 2224. Specifically, from 2004 to 2016, male handgun owners committed suicide at a rate three times higher than male nonowners, and female handgun owners took their lives at a rate seven times higher than female nonowners. Id. at 2224. These differential rates were “attributable to much higher rates of suicide by firearm,” not to “higher rates of suicide by other methods.” Id. at 2224, 2226. While these precise numbers could theoretically prove unique to California, or to those over age 21, we also know that suicide attempts “are often impulsive acts, driven by transient life crises,” and that “most people who attempt suicide do not go on to die in a future suicide.” Id. at 2221. It therefore stands to reason that reducing access to firearms—which are extremely effective as a means of death—will reduce suicides. See id.

type that an administrative agency or court does to accommodate judicial review,” even as to constitutional issues. Satellite Broad. & Commc’ns Ass’n, 275 F.3d at 358 (quoting

Turner, 512 U.S. at 666 (plurality opinion)) (First Amendment case). Thus, when considering legislative judgments to determine if an enactment survives intermediate scrutiny, not only do we defer to the policy judgments of the legislature, we also “allow

[the government] to rely on any material ‘reasonably believed to be relevant’ to substantiate its interests.” Mai, 952 F.3d at 1118 (alteration in original) (quoting Pena, 898 F.3d at 979)

(discussing judicial review of congressionally enacted gun control measure).

Commissioner Cohen’s testimony regarding federally licensed dealers surely meets that standard.

Plus, as noted, the “quantum of empirical evidence needed to satisfy heightened

judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised.” Nixon, 528 U.S. at 391. And where the justifications offered by Congress are “both familiar and plausible,” they need only be supported by

“minimal” empirical evidence. Carter, 669 F.3d at 418 (quoting Satellite Broad. & Commc’ns Ass’n, 275 F.3d at 360).

Here, it is undeniable that the justifications Congress offered for enacting the challenged provisions were both familiar and plausible to the legislators who passed the bills and to the voters that elected them. That’s why proponents of the measures made repeated appeals to common sense in pushing for their passage 23 and why the enactments had such broad-based public support. ^[24] As such, the provisions need only be supported by minimal empirical evidence, a bar they easily clear.

Finally, even if I thought that Congress did not sufficiently show its work back in 1968 by justifying the challenged provisions through detailed studies of the public-safety impacts of preventing young adults from buying handguns from federally licensed dealers—which I do not—I still would not find the absence of such studies to cut decisively against the government in our intermediate-scrutiny inquiry. To do so would be profoundly anachronistic. As the Third Circuit noted in Drake, legislatures contemplating gun control measures pre-Heller can hardly have been expected to “muster . . . reports, statistical information, and other studies” to fortify their bills against Second Amendment challenges

23 See, e.g., Firearms Act: 1965 Hearings at 32 (statement of Secretary Fowler) (“I think it is self-evident that minors [under age 21] should not have access to pistols [or] other concealable firearms and weapons of vast destructive power . . . .”); Ways and Means: 1965 Hearing at 28 (statement of Hon. Joseph W. Barr, Undersecretary of the Treasury) (making the same statement); 114 Cong. Rec. 13643 (1968) (statement of Sen. Dodd) (“[I]t is simply common[ ]sense not to allow federally licensed dealers to dispense weapons of death to the young and immature. Unless they are properly supervised, firearms in the hands of juveniles can result in great tragedy. There are scores of thousands of tragedies in the police files of our country to attest to this.”). 24 Polling from the spring of 1968 showed that Americans favored “strict control over the sale of firearms by 71 to 23 percent” and that gun owners favored such controls by 65 to 31 percent. S. Rep. 90-1097, at 184 (individual views of Senator Fong) (discussing a Harris Poll from April 22, 1968 and contending that it represented Congress’s “clear mandate from the people” to expand Title IV to include long guns in addition to handguns). This popular support translated into endorsements for the enacted measures from all walks of American life. See, e.g., 114 Cong. Rec. at 12314–16 (1968) (statement of Sen. Dodd) (documenting support for the challenged provisions from, among others, the American Bar Association, the AFL-CIO, civil rights leader Roy Wilkins, and the General Federation of Women’s Clubs, “the parent organization of some 15,000 women’s clubs”).

when the Supreme Court had not yet taught us that the Second Amendment protected an

individual right to bear arms. 724 F.3d at 437–38; see also Dearth v. Lynch, 791 F.3d 32, 49 (D.C. Cir. 2015) (Henderson, J., dissenting) (“The plaintiffs also fault the Government for submitting what is admittedly a sparse evidentiary record demonstrating the Congress’s rationale for applying the ban on receipt of firearms to non-resident citizens. It is hardly surprising that the Government is not able to offer more because the provisions at issue here were enacted in 1968 and 1994—long before the individual-right view of the Second

Amendment became the law of the land.” (internal citation omitted)). So too here.

In conclusion, the legislative history of the challenged provisions illustrates that

Congress studied a very specific problem—threats to public safety caused by youth access to handguns purchased from federally licensed dealers—and aimed to address that exact problem through the challenged provisions. This evidence strongly supports holding that the challenged provisions are a reasonable fit for the government’s compelling interest in public safety. iii. The challenged provisions are supported by modern scientific research

Beyond the evidence cited in the legislative history and common sense, Congress’s decision to restrict the ability of young adults aged 18 to 20 to purchase handguns from federally licensed dealers is also supported by more recent scientific research showing that the youngest adults, as a group, are decidedly less mature than other adults—even those just a few years older. 25 See Carter, 669 F.3d at 418 (explaining that courts can consider

“empirical evidence” in assessing the fit between a government’s interest and its enactment).

Importantly, the majority does not question the accuracy of this body of research.

This is a wise decision given that courts around the country have looked to similar scientific evidence to uphold age-based restrictions on Second Amendment rights. See NRA, 700

F.3d at 210 n.21 (concluding that “Congress’s finding that minors under 21 are prone to violent crime, especially with guns-in-hand,” the very finding at issue here, was entitled to

deference and was otherwise supported by modern scientific research on the impulsivity of young adults); Horsley v. Trame, 808 F.3d 1126, 1127, 1133 (7th Cir. 2015) (finding that

“scholarly research on development through early adulthood” supported the constitutionality of an Illinois measure that prevented those under 21 from possessing a

firearm without the approval of their parents or the Director of State Police); Mitchell v. Atkins, 483 F. Supp. 3d 985, 990, 995–96 (W.D. Wa. 2020) (relying in part on research

“show[ing] that 18- to 20-year-olds are developmentally immature compared with older

25 While the government has not put forward all the scientific evidence and crime and accident statistics I discuss, the amici in this case introduced similar evidence before the district court, and—like other courts—the district court explicitly relied on that evidence. See Maj. Op. at 62–63 n.57 (“[T]he amici’s arguments here were raised and adopted by the district court, and some of their evidence is in the record.”); Hirschfeld v. BATFE, 417 F. Supp. 3d 747, 759 (W.D. Va. 2019) (district court below finding that “the Amici parties highlight substantial evidence supporting Congress’s decision to draw the line at age 21”); NRA, 700 F.3d at 210 n.21 (relying on sources cited by amici in a separate case to uphold the same provisions at issue here). These facts mitigate any concern about considering arguments put forward on appeal by the amici.

adults” to uphold the constitutionality of a Washington law barring adults under 21 from purchasing semiautomatic rifles).

As surely as “any parent knows” that children are less mature and responsible than adults, Roper v. Simmons, 543 U.S. 551, 569 (2005), so too does any older adult know that late adolescence or early adulthood is a distinct and often more chaotic season of life than the ages that follow, cf. Jeffrey Arnett, Reckless Behavior in Adolescence: A

Developmental Perspective, 12 Developmental Rev. 339, 339–40 (1992) (defining

American adolescence as “extending from puberty to the early 20’s” and explaining that

“[a]dolescence bears a heightened potential for recklessness compared to other developmental periods in every culture and in every time”).

But I need not rest on the wisdom of age to substantiate this difference between the youngest adults and older adults, for it is well-established in the scientific literature. We now know, for example, that while “a 19-year-old might possess a brain that looks ‘adult-

like’ and that supports mature cognitive performance under calm or ‘neutral’ conditions, that same brain tends to look much more like that of a younger kid when evocative emotions are triggered, resulting in significantly weaker cognitive performance.” Jason

Chein, Adolescent Brain Immaturity Makes Pending Execution Inappropriate, Bloomberg

Law (Sept. ^[17], 2020 4:00 AM), https://www.bloomberglaw.com/bloomberglawnews/us- law-week/XBBCKGKK000000; see also, e.g., Adam Ortiz, Juvenile Justice Ctr., Am. Bar

Ass’n, Cruel and Unusual Punishment: The Juvenile Death Penalty; Adolescence, Brain Development and Legal Culpability, 2 (Jan. 2004) (“The evidence now is strong that the brain does not cease to mature until the early 20s in those relevant parts that govern impulsivity, judgment, planning for the future, foresight of consequences, and other characteristics that make people morally culpable . . . . Indeed, age 21 or 22 would be closer

to the ‘biological’ age of maturity.” (quoting declaration of Dr. Ruben Gur of the University of Pennsylvania Medical Center)).

Furthermore, modern research also instructs that the prefrontal cortex, the part of the brain that makes it possible to “exercise good judgment when presented with difficult life situations,” does not finish maturing until age 25. J.A. 346, 352 (reprinting Mariam

Arain et al., Maturation of the Adolescent Brain, 9 Neuropsychiatric Disease & Treatment

449, 453, 459 (2013)); see also United States v. Ramsay, ---F. Supp. 3d---, 2021 WL

1877963, at *9 (S.D.N.Y. May 11, 2021) (Rakoff, J.) (surveying the literature on

“psychosocial maturity—the capacity to exercise self-restraint, especially in emotionally- arousing contexts,” and noting that this kind of maturity “continues to mature throughout the teens and into the twenties,” perhaps even to age 30 (cleaned up)).

The upshot of these findings is that in precisely the situations where violence is most likely—that is, emotionally evocative situations or situations requiring the intense weighing of the exigencies of the moment against future consequences—young adults think more like children than they do like older adults. See Daniel W. Webster et al., John

Hopkins Ctr. for Gun Policy and Research, Firearms on College Campuses: Research

Evidence and Policy Implications (Oct. ^[15], 2016), at 19 (explaining that “[g]uns may be called on in the very situations in which adolescents are most developmentally vulnerable”). 26

In short, as the Fifth Circuit has already concluded in its study of this exact issue, “modern scientific research supports the commonsense notion that 18-to-20-year-olds tend to be [less mature] than young adults aged 21 and over.” NRA, 700 F.3d at 210 n.21. I would therefore join the courts around the country that have held that this science buttresses reasonable restrictions on the gun rights of minors and young adults. iv. The challenged provisions are supported by crime and accident statistics

Based on the most up-to-date scientific evidence, therefore, we should expect that young adults are more likely to be involved in violent crime than older adults, given their biological immaturity. It is thus completely unsurprising that abundant modern evidence shows that young adults in fact perpetrate more violent crime generally and more

26 The phenomenon of young adults under twenty-one thinking and behaving more like children than they do like older adults is also well-established in legal scholarship, much of which has focused on questioning whether the death penalty is appropriate for those whose crimes were committed as children or young adults. See, e.g., John H. Blume et al., Essay, Death By Numbers: Why Evolving Standards Compel Extending Roper’s Categorical Ban Against Executing Juveniles from Eighteen to Twenty-One, 98 Tex. L. Rev. 921, 930–35 (2020); Andrew Michaels, A Decent Proposal: Exempting Eighteen-to- Twenty-Year-Olds From the Death Penalty, 40 N.Y.U. Rev. L. & Soc. Change, 139, 161– 67 (2016); Melissa S. Caulum, Comment, Postadolescent Brain Development: A Disconnect Between Neuroscience, Emerging Adults, and the Corrections System, 2007 Wis. L. Rev. 729, 731–33 (2007); Tirza A. Mullin, Note, Eighteen Is Not A Magic Number: Why The Eighth Amendment Requires Protection for Youth Aged Eighteen to Twenty-Five, 53 Mich. J.L. Reform, 807, 812–16 (2020); Zoe Jordan, Note, The Roper Extension: A California Perspective, 71 Hastings L.J. 197, 206–210 (2019).

firearm crime specifically than do older adults. 27 In fact, according to an analysis of FBI supplementary homicide reports conducted by amicus Everytown for Gun Safety, from

2013 to 2017, young adults aged 18 to 20 “commit[ted] gun homicides at a rate nearly four times higher than adults 21 and older.” Everytown for Gun Safety Analysis, Uniform Crime

Reporting Program: Supplementary Homicide Reports (SHR) 2013–2017 (2018)

(emphasis added), https://everytownresearch.org/stat/eighteen-to-20-year-olds-commit- gun-homicides-at-a-rate-nearly-four-times-higher-than-adults-21-and-older/; see also Br. for Amicus Curiae Brady at 11 (quoting Everytown for Gun Safety, Concealed Carry

Reciprocity: Overriding State Public Safety Laws (June 25, 2018), https://everytownresearch.org/report/concealed-carry-reciprocity-overriding-state-public- safety-laws/ (same)).

27 See, e.g., Off. of Juvenile Just. & Delinquency Prevention, U.S. Dep’t of Justice, Arrest Rates by Offense and Age Group 2019, https://www.ojjdp.gov/ojstatbb/crime/ucr.asp?table_in=1&selYrs=2019&rdoGroups=1&r doData=r (documenting that in 2019, Americans aged 18 to 20 committed 320.8 violent crimes per 100,000 individuals, compared with rates of 167.3 per 100,000 for all individuals 18 and older and 145.2 per 100,000 for all individuals 25 and older); NRA, 700 F.3d at 210 (“In 2009, 18-to-20 year-olds accounted for over 19% of all murder and non- negligent manslaughter arrests, 14% of all arrests for forcible rape, almost 24% of all robbery arrests, and 12% of all aggravated assault arrests . . . even though they comprised only about 4.3% of the population.”); Horsley, 808 F.3d at 1133 (“An FBI analysis of crime in 2014 reflects that 18-to-20-year-olds were responsible for more than 15.8% of all charges issued for murder and nonnegligent manslaughter,” despite comprising just “5.4% of the population over the age of 14.”); Mitchell, 483 F. Supp. 3d at 996 (noting that young adults age 18 to 20 account “for approximately one-quarter of firearm homicides committed where an offender was identified” and that “[a]rrest rates for murder, robbery, and other violent crimes peak around ages 17 to 20”); Powell, 926 F. Supp. 2d at 392–93 (citing a 2011 survey conducted by the FBI and the DOJ finding that “eighteen- to twenty- year-olds accounted for approximately 16.6% of all arrests made for weapons offenses” despite comprising “only about 4.3% of the national population”).

Nor is the connection between age and firearm injuries limited to intentional violence. It also appears that young adults are much more likely to die in a gun accident than older adults are. See Sarah J. Solnick and David Hemenway, Unintentional Firearm

Deaths in the United States 2005–2015, Inj. Epidemiology, 6:42 at 3–4 (2019) (finding in a survey of 16 states that “children and teens, ages 10 to 19” and “young adults, aged 20 to

29” are at an elevated risk of dying in a gun accident with “[a]ll other age groups hav[ing] a markedly lower rate” of accidental deaths).

These statistics “give credence to the notions that young adults’ access to firearms is an issue of significant governmental concern and that there is a close fit between this concern and the [challenged provisions].” Powell, 926 F. Supp. 2d at 393 (finding that crime statistics supported the constitutionality of an age-based restriction on carrying firearms). Thus, I would treat these crime and accident statistics, like the modern scientific research on the biological immaturity of young adults, as yet more evidence confirming the reasonableness of the challenged provisions. v. The challenged provisions are supported by state enactments

Finally, when considering whether the challenged provisions are a reasonable fit for Congress’s interest in promoting public safety, I find it highly relevant that “all fifty states in the Union, as well as the District of Columbia, feature laws imposing age-based

restrictions [on] firearms, and [that] many of these laws set the minimum age for the possession and use of firearms at twenty-one.” Id. at 387. Specifically, at least twenty-one states and the District of Columbia regulate, and often prohibit altogether, the sale of handguns to or the purchase of such guns by individuals under twenty-one. ^[28]

While state legislatures cannot determine the parameters of constitutional rights, the fact that many states currently agree with Congress on the need to regulate young adults’ access to handguns speaks to the ongoing reasonableness of the actions Congress took back

in 1968. In other words, courts may look to the ubiquity of state-level restrictions in determining the reasonableness of a challenged provision’s fit, and thus its constitutionality. See, e.g., United States v. Yancey, 621 F.3d 681, 683–84 (7th Cir. 2010)

(per curiam) (finding that state-level restrictions on the ability of habitual drug users to possess firearms supported the constitutionality of 18 U.S.C. § 922(g)(3)’s federal ban on firearm possession by drug addicts, as such restrictions demonstrated that “Congress was not alone in concluding that habitual drug abusers are unfit to possess firearms”).

***

To summarize: Congress had a compelling interest in promoting public safety and the challenged provisions are a reasonable fit for that objective. The reasonableness of the fit is abundantly illustrated by the legislative text and history, scientific research, crime and 28 See Cal. Penal Code § 27505; Conn. Gen. Stat. § 29-34(b); D.C. Code § 22-4507; Del. Code tit. ^[24], §§ 901, 903; Fla. Stat. § 790.065(13); Haw. Rev. Stat. § 134-2(d); 430 Ill. Comp. Stat. 65/3(a), 65/4(a)(2)(i-5); Iowa Code § 724.22(2); Md. Code, Pub. Safety § 5-134(b)(1); Mass. Gen. Laws ch. 140, § 130; Mich. Comp. Laws § 28.422(1), (3); Mo. Rev. Stat. § 571.080; Neb. Rev. Stat. §§ 69-2403, 69-2404; N.J. Stat. § 2C:58-3(c)(4).; N.Y. Penal Law 400.00(1); Ohio Rev. Code Ann. § 2923.21; 11 R.I. Gen. Laws § 11-47- 37; Vt. Stat. Ann. tit. ^[13] § 4020; Wash. Rev. Code § 9.41.240(1); W. Va. Code § 61-7- 10(c); Wyo. Stat. Ann. § 6-8-404(d).

accident statistics, and contemporary state laws, all of which show that Congress reasonably believed it could reduce gun harm by enacting the challenged provisions.

D.

The majority, of course, disagrees. Its disagreement is rooted in three arguments.

First, as discussed above, the majority believes that Congress did not adequately document

the link between handguns sold by federally licensed dealers and crime. Second, the majority contends that the challenged provisions are impermissibly overinclusive—that “a showing of disproportionate bad conduct by a group cannot justify categorical restrictions on rights when the percentage of the group engaged in the unwanted conduct is minuscule.”

Maj. Op. at 65. Third, the majority contends that the challenged provisions are impermissible because “it is unclear whether these laws have been effective at all.” Id. at

81.

I believe these arguments are misguided and do not justify concluding that the challenged provisions fail intermediate scrutiny. Having addressed the majority’s first argument above, I will now address its remaining two. i. The challenged provisions are not impermissibly overinclusive

The majority’s over-inclusiveness analysis is guided by the Supreme Court’s decision in Craig v. Boren, 429 U.S. 190 (1976). According to the majority, “just as sex was a poor proxy for drunk driving in Craig,” so too is age “a poor proxy for gun violence in this case.” Maj. Op. at 67. While Craig undeniably supports the unremarkable idea that overinclusive enactments may lack the reasonable fit we demand under intermediate scrutiny, the majority’s comparison of this matter to Craig suffers from at least four problems that substantially reduce its relevance to this case.

First and most fundamentally, the majority attempts to calcify the “flexible” world of intermediate scrutiny into a rigid statistical framework. Heller III, 801 F.3d at 282–83

(Henderson, J., concurring in part and dissenting in part) (“Intermediate scrutiny is a flexible framework that allows for different perspectives and a range of approaches to firearms regulation.”); see also Kolbe, 849 F.3d at 133 (explaining that intermediate scrutiny is “less onerous” than strict scrutiny, requiring only a reasonable fit with a substantial governmental interest). According to the majority, because less than one percent of all young adults commit a violent crime in any given year, the challenged provisions fail the “unduly tenuous threshold” of 2 percent supposedly established by Craig. Maj. Op. at

71.

But nothing in Craig purported to establish a roving two-percent threshold applicable to all contexts in which courts might apply intermediate scrutiny, and I am unaware of a case in the Fourth Circuit or anywhere else treating it as having done so.

Rather, our circuit precedent on the Second Amendment rejects such efforts to establish

hard and fast statistical rules for determining whether an enactment is a reasonable fit for furthering an important government interest. See Carter, 750 F.3d at 468–69 & n.15

(explaining that Congress is free to regulate based on correlation evidence and that while

“[s]cientists may insist on p-values of 0.05,” Congress “is not so constrained”).

Second, any statistical comparison between Craig and the instant matter is muddied by the fact that violent crime is nearly 100 hundred times less common than drinking and driving. ^[29] The majority believes that “[t]he question is . . . the percentage of the group that uses a gun in an impermissible manner.” Maj. Op. at 71. According to the majority, “[t]hat statistic gives us some insight into the likelihood that any one person in the group will use a gun improperly.” Id. So, the majority reasons, because the percentage of young adults that will misuse a gun in any given year is less than two percent—which, remember, is just

the percentage of young men arrested for drunk driving in Oklahoma at one point in the 1970s—the connection between youth and gun violence is “unduly tenuous [under the] threshold from Craig.” Id.

This approach might make more sense if drinking and driving and gun violence were equally common. But in reality, the percentage of young adults that commit a gun crime in any given year is below two percent because gun violence, blessedly, is much

rarer than drinking and driving. That alone tells us nothing about the reasonableness of regulating youth access to firearms.

To illustrate this point, consider two extreme examples. Suppose some high

percentage of all people, say 80%, will engage in some infraction in a given year. Well, then, a statistic that only 2% of a specific subset of people will commit that infraction means not only that they rarely engage in that behavior, but also that they are far less likely

29 There are approximately “111 million self-reported episodes of alcohol-impaired driving among U.S. adults each year.” Impaired Driving: Get the Facts, Ctrs. for Disease Control & Prevention, https://www.cdc.gov/transportationsafety/impaired_driving/ impaired-drv_factsheet.html. In contrast, there were approximately 1.2 million violent crimes committed in the United States in 2019. Federal Bureau of Investigation, 2019: Crime in the United States, https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.- 2019/tables/table-1.

than the average person to do so. Now suppose that only 0.00000001% of all people will engage in the infraction in a given year. Under this new hypothetical, the fact that 2% of a specific subset of people engages in that behavior takes on a very different meaning: it might still be rare among that subset, but they are far more likely than the average person to engage in the behavior. Governmental regulation of the subset is exponentially more

justifiable in the second case than in the first, even though the 2% figure remained constant, because of the difference in frequency of the infraction in the overall population.

In other words, by plucking the two-percent statistic from a specific factual context and plopping it down in a very different one, the majority erroneously justifies striking down a long-established Congressional enactment. Craig mandates no such judicial immodesty.

Third, Craig and the instant matter involve two very different kinds of classification—one is based on gender, the other on age. We treat classifications based on

gender with suspicion because gender “generally provides no sensible ground for differential treatment,” City of Cleburne v. Cleburne Living Ctr., 473 U.S. 432, 440 (1985), and because such classifications are generally rooted “in our Nation’s ‘long and unfortunate history of sex discrimination,’” Wilcox v. Lyons, 970 F.3d 452, 459 (4th Cir.

2020) (quoting Frontiero v. Richardson, 411 U.S. 677, 684 (1973) (plurality opinion)).

Meanwhile, age is not a suspect classification at all. Kimel v. Fla. Bd. of Regents, 528 U.S.

62, 83 (2000); see also Am. Entertainers, L.L.C. v. City of Rocky Mount, 888 F.3d 707, 711–12, 722–23 (4th Cir. 2018) (applying rational-basis review to equal protection challenge to city law barring 18- to 21-year-olds from owning “sexually oriented businesses”). The restriction at issue in Craig was therefore repugnant to the nation’s commitment to equal protection in a way the challenged provisions in this case are not. See

Craig, 429 U.S. at 204, 208 n.22 (arguing that “proving broad sociological propositions by statistics” was “in tension with the normative philosophy that underlies the Equal

Protection Clause”). 30

Finally, even assuming that Craig is relevant here, the majority’s calculations still have a numerator problem. The majority estimates that approximately 0.3% of young adults aged 18 to 20 commit a violent crime each year. Because that figure is below the so- called Craig threshold, the majority concludes that the challenged provisions are over-

30 The majority wonders whether “the Second Amendment rights of men can be categorically restricted” because they commit more crimes than women. Maj. Op. at 70. To state the obvious, any such enactment would face major Second Amendment and Equal Protection Clause problems not faced by the challenged provisions. For a start, sex is generally immutable, meaning that any restriction placed on male access to firearms would be permanent for most people and therefore a much greater infringement on gun rights than the temporary provisions at issue here. Any such restriction would also run straight into an equal-protection problem not raised by classifying on the basis of the non-protected status of age. This hypothetical is hardly a checkmate. The majority also wonders about restrictions for “those living in . . . specific geographic location[s]” and restrictions on those under age 25. Maj. Op. at 70–71. Our analysis of any such regulation would turn on the specifics of the enactment. Any such assessment would be highly fact-bound and is therefore not suited for abstract and speculative discussion. With that said, municipalities can enact their own firearm regulations, and some commentators believe that local conditions should be permitted to play a role in assessing the constitutionality of those provisions. See, e.g., Joseph Blocher, Firearm Localism, 123 Yale L.J. 82, 85 (2013) (arguing that Second Amendment jurisprudence “should incorporate the longstanding and sensible differences regarding guns and gun control in rural and urban areas, giving more protection to gun rights in rural areas and more leeway to gun regulation in cities”). As for any aged-based regulation applying to individuals under twenty-five, the same forces that make armed young adults under twenty-one unique threats to the safety of themselves and others may indeed extend later into those young adults’ twenties. But we have no cause to address that matter today.

inclusive. But this exclusive focus on the percentage of young adults that commit a violent crime each year is too narrow.

As discussed above, while the challenged provisions were aimed primarily at

reducing violent crime, they also aspired to promote public safety by targeting deaths and injuries from suicides and accidents. While exact numbers are difficult to come by, it appears that each year, hundreds, if not thousands, of Americans aged 18 to 20 are accidentally injured in handgun accidents or intentionally harm themselves with a handgun. 31 These figures for accidents and suicides, while not large enough to get above the Craig “threshold,” should still be included in any over-inclusivity analysis as Congress aimed to promote public safety by reducing both gun crime and other forms of gun harm.

Similarly, any proper comparison of this case to Craig would have to factor in the number

31 It can be estimated that about 900 Americans aged 18 to 20 are injured by handguns each year, either unintentionally or through self-harm. I derive this estimate from figures provided by Victor Lee et al., Emergency Department Visits for Firearm-Related Injuries Among Youth in the United States, 2006–2015, 48 J. L., Med. & Ethics 67, 70 (2020). The study found that in the 18 to 20 cohort between 2006 and 2015, 36,780 individuals reported to a hospital emergency room for unintentional firearm injury and 2,399 reported for firearm injury caused by a suicide attempt or self-harm. Id. The study further reported that handguns were involved in 23 percent of all injuries this age cohort sustained over this period, with the categories “other” and “unspecified” accounting for 70.9 percent of injuries, and shotguns, hunting, and military firearms making up the remaining 6.2 percent of injuries. Id. I then assumed that because handguns were used in 23 percent of all injuries for this cohort, they were used in 23 percent of injuries caused unintentionally or through self-harm. I then calculated ((36,780 + 2,399) x 0.23) / 10 (the years of the study) for a total of 901.1 handgun injuries per year from accidents and self- harm. While it could be that handguns are involved in less than 23 percent of such injuries to this cohort, which would inflate my estimate, there are deflationary pressures on my estimate as well—namely, the classification of a huge majority of guns as “other” or “unknown” and the fact that not all individuals injured by a firearm go to the emergency room.

of individuals who, as a result of the challenged provisions being in effect, never purchased handguns and thus never harmed themselves or others with a handgun. This number is also missing from the majority’s calculations. 32

In light of all these issues with comparing this case to Craig, I cannot agree that

Craig renders the challenged provisions unconstitutionally overbroad in violation of the Second Amendment. ii. The purported ineffectiveness of the challenged provisions does not render them unconstitutional

The majority also argues that the challenged provisions “lack the reasonable fit required to pass intermediate scrutiny” because they are ineffective at reducing youth crime

rates. Maj. Op. 78–82. Relatedly, the majority argues that the law is underinclusive, meaning that it is ineffective or at least not as effective as it could be. Thus, according to the majority, Congress did not do enough to combat harm caused by guns, and what it did do was too ineffective—and maybe even counterproductive—to be allowed. This approach is strongly misguided for at least four reasons.

First, it forgets that the Second Amendment is an area in which we owe substantial deference to the judgments of Congress. Congress’s judgment that enacting the challenged

32 While there is evidence to suggest that the challenged provisions reduce gun harm in the form of crime, accidents, and suicides, Craig did not establish that the Oklahoma laws at issue reduced drunk driving. See 429 U.S. at 202 n.14 (noting the apparent “futility of controlling driving behavior by the 3.2% beer statute” and that the Court “obviously ha[d] no means of estimating how many individuals, if any, actually were prevented from drinking by these laws”).

provisions would promote public safety is “precisely the type of judgment that legislatures are allowed to make without second-guessing by a court.” Kolbe, 849 F.3d at 140.

Second, it unfairly punishes Congress for not passing a more restrictive measure and thus violates the standard principle that “a statute is not invalid under the Constitution

because it might have gone farther than it did.” NRA, 700 F.3d at 211 (quoting Buckley v. Valeo, 424 U.S. ^[1], 105 (1976) (per curiam)); see also N.Y. State Rifle & Pistol Ass’n, 804

F.3d at 263 (“[G]un-control legislation ‘need not strike at all evils at the same time’ to be

constitutional.” (quoting NRA, 700 F.3d at 211)); Mance v. Sessions, 896 F.3d 699, 701, 704, 708–09 (5th Cir. 2018) (per curiam) (holding that federal laws preventing federally licensed dealers from directly selling handguns to out-of-state buyers survived strict scrutiny despite their underinclusivity, as governments “need not address all aspects of a problem in one fell swoop” (quoting Williams-Yulee v. Fla. Bar, 575 U.S. 433, 449

(2015))).

Third, as the majority recognizes, it “[is] difficult to determine the effectiveness of these laws now.” Maj. Op. at 81. Given this conceded uncertainty, I would not have this

Court stray far beyond its area of expertise to strike down a long-established law because of its perceived ineffectiveness. Cf. United States v. Masciandaro, 638 F.3d 458, 475–76

(4th Cir. 2011) (“We do not wish to be even minutely responsible for some unspeakably tragic act of mayhem because in the peace of our judicial chambers we miscalculated as to Second Amendment rights. . . . If ever there was an occasion for restraint, this would seem to be it.” (emphasis added)).

Fourth, courts should be especially hesitant to hold gun control measures unconstitutional due to their perceived infectiveness, because doing so will place the nation and its lawmakers in a formidable catch-22: pass too onerous a regulation and see it struck

down for violating the Second Amendment; pass too permissive a measure and suffer the same result. This heads-I-win, tails-you-lose approach is a recipe for national inaction on gun violence. While the people are free to choose that course, it ought not be imposed from on high by us. ^[33]

33 This would be a particularly perverse and heartbreaking result here given the widespread belief back in 1968 that far stronger legislative action was needed to curb gun violence. The challenged provisions are a very modest effort at gun control. See Winkler, supra, at 253 (“In an era that would become known for the aggressive, bold federal legislation enacted to combat a variety of social ills . . . the gun control laws of 1968 were notable for their timidity.”). Indeed, they have been recognized as cautious measures since before their birth. While the provisions were still being debated, Senator Joseph Tydings of Maryland called Title IV of the Crime Control and Safe Streets Act “a very limited, stripped down, bare minimum gun traffic control bill.” 114 Cong. Rec. 13639 (1968) (statement of Sen. Tydings). Senator Hiram Fong of Hawaii agreed, stating that the challenged provisions and the rest of Title IV were “entirely inadequate” and fell “far short of the strong and effective firearms control legislation so urgently required to control crime.” S. Rep. 90-1097, at 184 (1968) (individual views of Sen. Fong); accord 114 Cong. Rec. 12298–99 (1968) (statement of Sen. Fong) (“[The provision barring those under 21 from purchasing handguns from federally licensed dealers] represents the least Congress can do to meet the pressing public interest to protect against unrestricted gun traffic . . . . But it is not nearly enough.”). President Johnson echoed the same theme in his signing statement, calling the measures a “halfway step toward the protection of our families and homes.” 114 Cong. Rec. 18091 (1968) (reprint of statement of President Johnson upon signing the Omnibus Crime Control and Safe Streets Act of 1968). In short, many of the individuals who helped to enact the challenged provisions wanted to go further but believed the provisions were better than nothing. Now, more than fifty years later, the majority becomes the first court in the country to strike these measures down, leaving the American people with nothing in their place—because the provisions did not go further. The irony is bitter.

I therefore would not usurp the legislative role by invalidating these measures based on a judicial assessment that they have not worked. ^[34]

III.

The words of our colleague, Judge Wilkinson, bear repeating as a fitting admonition for the majority’s decision: “No one really knows what the right answer is with respect to

the regulation of firearms,” thus, we federal judges ought not “[d]isenfranchis[e] the American people on this life and death subject” by arrogating to ourselves “decisions that have been historically assigned to other, more democratic, actors.” Kolbe, 849 F.3d at 149–

150 (Wilkinson, J., concurring).

Today, by failing to heed that admonition, the majority fundamentally errs as a matter of law. I must, respectfully, dissent.

34 Having concluded that the challenged provisions violate the Second Amendment, the majority does not resolve whether the provisions also discriminate against Marshall based on her age in violation of the equal protection guarantee of the Fifth Amendment. Because “age is not a suspect classification,” Kimel, 528 U.S. at 83, the government “may ‘discriminate on the basis of age without offending’ the constitutional guarantee of equal protection ‘if the age classification in question is rationally related to a legitimate state interest,’” NRA, 700 F.3d at 212 (quoting Kimel, 528 U.S. at 83) (concluding that the challenged provisions survived rational-basis review). Because I would hold that the challenged provisions survive intermediate scrutiny, I would also, of course, conclude that they survive rational-basis review. Therefore, I would hold that Marshall’s Fifth Amendment claim fails.