Matthew JONES; Thomas Furrh; PWGG, L.P., dba Poway Weapons and Gear and PWG Range; North County Shootingining Center, Inc.; Beebe Family Arms and Munitions LLC, dba BFAM and Beebe Family Arms and Munitions; Firearms Policy Coalition, Inc.; Firearms Policy Foundation; The Calguns Foundation; Second Amendment Foundation; Kyle Yamamoto, Plaintiffs-Appellants, v. Rob BONTA,* in his official capacity as Attorney General of the State of California; Martin Horan, in his official capacity as Director of the Department of Justice Bureau of Firearms; Does, 1–20, Defendants-Appellees.
America would not exist without the heroism of the young adults who fought and died in our revolutionary army. Today we reaffirm that our Constitution still protects the right that enabled their sacrifice: the right of young adults to keep and bear arms.
California has restricted the sale of most firearms to anyone under 21. Plaintiffs challenged the bans on long guns and semiautomatic centerfire rifles under the Second Amendment. The district court declined to issue a preliminary injunction.
We hold that the district court did not abuse its discretion in declining to enjoin the requirement that young adults obtain a hunting license to purchase a long gun. But the district court erred in not enjoining an almost total ban on semiautomatic centerfire rifles. First, the Second Amendment protects the right of young adults to keep and bear arms, which includes the right to purchase them. The district court reasoned otherwise and held that the laws did not burden Second Amendment rights at all: that was legal error. Second, the district court properly applied intermediate scrutiny to the long gun hunting license regulation and did not abuse its discretion in finding it likely to survive. But third, the district court erred by applying intermediate scrutiny, rather than strict scrutiny, to the semiautomatic centerfire rifle ban. And even under intermediate scrutiny, this ban likely violates the Second Amendment because it fails the “reasonable fit” test. Finally, the district court also abused its discretion in finding that Plaintiffs would not likely be irreparably harmed. We thus affirm the district court's denial of an injunction as to the long gun regulation, reverse its denial of an injunction as to the semiautomatic centerfire rifle ban, and remand for further proceedings consistent with this opinion.
California regulates the acquisition, possession, and ownership of firearms with a multifaceted scheme. Peruta v. County of San Diego, 824 F.3d 919, 925 (9th Cir. 2016) (en banc). To start, some general requirements apply to everyone, not just young adults. 1 First, except for some intrafamily transfers and loans, 2 the state requires that all transfers of firearms happen at a licensed firearms dealer. Cal. Penal Code §§ 27545, 28050. And second, the purchaser must have a valid firearm safety certificate (“FSC”). Id. §§ 31615, 27540(e). Exempt from the FSC requirement are people with hunting licenses, active and reserve peace officers, federal officers or law enforcement agents, and active or honorably retired members of the armed forces. Id. § 31700(a)–(c).
California also regulates young adults' commerce in firearms. Specifically, after first banning only the sale of handguns, California then prohibited the sale to young adults of almost any kind of firearm. The only exception was for sales of long guns to young adults who (1) have a state hunting license, (2) are peace officers, active federal officers, or active federal law enforcement agents and are allowed to carry firearms for their work, or (3) are active or honorably discharged members of the military. 2017 California Senate Bill No. 1100, California 2017–2018 Regular Session. 3
Several young adults, gun shops, and advocacy groups sued, asking the district court to enjoin the long gun regulation under the Second and Fourteenth Amendments. Then, while the suit was pending, California again amended the law, banning sales of semiautomatic centerfire rifles 4 to young adults, and excepting only law enforcement officers and active-duty military, but not hunting license holders. In response, Plaintiffs withdrew their motion for a preliminary injunction, amended their complaint to challenge the new ban, and again sought a preliminary injunction, now of both the long gun regulation and the semiautomatic rifle ban.
The district court declined to preliminarily enjoin the laws, holding that Plaintiffs had not shown that they were likely to succeed on the merits, both because the laws did not burden Second Amendment rights and would likely survive intermediate scrutiny. The district court also held that Plaintiffs had not shown irreparable harm and that the balance of interests did not favor enjoining the laws.
First, the district court observed that other courts had held that similar laws do not burden Second Amendment rights at all. Jones v. Becerra, 498 F. Supp. 3d 1317, 1326–27 (S.D. Cal. 2020). 5 The district court noted that these courts found that similar laws were “longstanding, do not burden the Second Amendment, and are therefore presumptively constitutional.” Id. at 1327. The district court then reasoned that “[i]ndividuals under the age of 21 were considered minors or ‘infants’ for most of our country's history without the rights afforded adults” and therefore they are among those “believed unfit of responsible firearm possession and use.” Id. at 1327. It did address the tradition of militia members who were under 21 years old, but reasoned this tradition actually supported the constitutionality of the laws. Id. In the district court's view, “[m]ilitia members were required to possess their own firearms if they complied with accountability and maintenance regulations” and thus the “strict rules surrounding militia duty” show that the “right to firearm possession came with obligations to ensure public safety.” Id.
Because of other courts' holdings, the longstanding history of similar regulations, and its militia analysis, the district court reasoned that California's laws “do[ ] not burden the Second Amendment.” Id. The district court thus held that Plaintiffs were not likely to succeed on the merits.
Second, because it found no burden on Second Amendment rights, the district court did not need to apply any tier of scrutiny. Still, “in an abundance of caution,” the district court also determined that intermediate scrutiny applied and that the laws likely survived it. Id.
In determining whether to apply strict or intermediate scrutiny, the district court reasoned that the laws neither implicated the core Second Amendment right nor severely burdened that right. Id. at 1328 (citing Pena v. Lindley, 898 F.3d 969, 977 (9th Cir. 2018)). The district court bolstered its conclusion by noting that young adults could “receive otherwise prohibited firearms via transfer from immediate family.” Id. Because the laws were not, in the district court's view, complete bans, it held that only intermediate scrutiny would be required. Id.
The district court then held that the laws likely would satisfy intermediate scrutiny. The court noted first that California's goal of improving public safety was a significant objective. Id. The court then held that the laws “provide[ ] a reasonable fit” to those goals because “it remains commonly understood that Young Adults may require additional safeguards to ensure proper training and maintenance of firearms.” Id. at 1330. Thus, Plaintiffs still were not likely to succeed on the merits, even under intermediate scrutiny.
Third, the district court held that Plaintiffs failed to show irreparable harm. Id. at 1330–32. The district court observed that, after filing their amended complaint, Plaintiffs waited two months before moving for a preliminary injunction. It reasoned that this delay undermined finding irreparable harm. Id. at 1331. “More importantly,” young adults could still get firearms, either under an exception, through a transfer from family, or by using them at shooting ranges. Id.
Finally, the district court also held that the balance of interests weighed against enjoining the laws, reasoning that “[t]he potential harm of enjoining a duly-enacted law designed to protect public safety outweighs Young Adults' inability to secure the firearm of their choice without proper training.” Id. at 1332.
The district court had jurisdiction under 28 U.S.C. § 1331. We have jurisdiction to review the district court's interlocutory order declining to issue an injunction under 28 U.S.C. § 1292(a)(1).
Because we “have an independent obligation to ensure that [we] do not exceed the scope of [our] jurisdiction,” Henderson ex rel. Henderson v. Shinseki, 562 U.S. 428, 131 S.Ct. 1197, 179 L.Ed.2d 159 (2011), we must ensure that the case is not moot, even if the parties do not dispute it.
A case is moot when there is “no actual or live controversy.” Bishop Paiute Tribe v. Inyo County, 863 F.3d 1144, 1155 (9th Cir. 2017) (citing Foster v. Carson, 347 F.3d 742, 745 (9th Cir. 2003)). “If there is no longer a possibility that an appellant can obtain relief for his claim, that claim is moot and must be dismissed for lack of jurisdiction.” Id. (internal quotation marks omitted).
Defendants argue that the individual plaintiffs' claims are moot because they have turned 21, but they concede that we have jurisdiction anyway. We agree that we have jurisdiction. We need not reach whether the individual plaintiffs' claims are moot because the organizational plaintiffs' claims are not. Cf. Laub v. U.S. Dep't of Interior, 342 F.3d 1080, 1086 (9th Cir. 2003) (where “one plaintiff ha[s] standing to bring the suit, the court need not consider the standing of the other plaintiffs”). The advocacy groups sued on behalf of their young adult members, and some of those members are still under 21. Similarly, the firearm dealer plaintiffs sued because they have had to forego selling firearms to young adults and offering firearm classes for them. The case is not moot as to them either because they still cannot sell firearms to young adults or admit them to their classes.
Although we review the denial “of a preliminary injunction for an abuse of discretion” and “factual findings for clear error,” we also review “the underlying legal conclusions de novo.” Washington v. U.S. Dep't of State, 996 F.3d 552, 560 (9th Cir. 2021). If “the district court relied on an erroneous legal premise,” then it abused its discretion. Fyock v. Sunnyvale, 779 F.3d 991, 995 (9th Cir. 2015).
The district court's choice of a tier of scrutiny is a legal question that we review de novo. See Joelner v. Vill. of Wash. Park, Ill., 508 F.3d 427, 431 (7th Cir. 2007), as amended on denial of reh'g (Apr. 3, 2008). Its application of that tier of scrutiny is reviewed for abuse of discretion. Fyock, 779 F.3d at 998.
At the preliminary injunction stage, our “review of the district court's findings” is “restricted to the limited record available to the district court when it granted or denied the motion.” Sports Form, Inc. v. United Press Int'l, Inc., 686 F.2d 750, 753 (9th Cir. 1982). Ultimately, because denying a preliminary injunction lies within a district court's discretion, we may reverse only when it abused its discretion by relying on an erroneous legal premise or clearly erroneous finding of fact. See, e.g., Calvary Chapel Dayton Valley v. Sisolak, 982 F.3d 1228, 1231 (9th Cir. 2020).
“A plaintiff seeking a preliminary injunction must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.” Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008). When the government is a party, the balance of equities factor and the public interest factor merge. Drakes Bay Oyster Co. v. Jewell, 747 F.3d 1073, 1092 (9th Cir. 2014). “Under our ‘sliding scale’ approach, a stronger showing of one element may offset a weaker showing of another, as long as plaintiffs ‘establish that irreparable harm is likely.’ ” Doe v. Kelly, 878 F.3d 710, 719 (9th Cir. 2017) (quoting All. for the Wild Rockies v. Cottrell, 632 F.3d 1127, 1131 (9th Cir. 2011)).
“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. The Second Amendment “protects a personal right to keep and bear arms for lawful purposes, most notably for self-defense within the home.” McDonald v. City of Chicago, 561 U.S. 742, 780, 130 S.Ct. 3020, 177 L.Ed.2d 894 (2010). This right is “applicable to the States” through the Due Process Clause of the Fourteenth Amendment. Id. at 750, 130 S.Ct. 3020.
The “Second Amendment right is exercised individually and belongs to all Americans.” District of Columbia v. Heller, 554 U.S. 570, 581, 128 S.Ct. 2783, 171 L.Ed.2d 637 (2008). The “people” protected by the Second Amendment “refers to a class of persons who are part of a national community.” Id. (quoting United States v. Verdugo-Urquidez, 494 U.S. 259, 265, 110 S.Ct. 1056, 108 L.Ed.2d 222 (1990)).
On the merits, for challenges to firearm laws under the Second and Fourteenth Amendments, we apply a “two-step framework.” Young v. Hawaii, 992 F.3d 765, 783 (9th Cir. 2021) (en banc). First, we ask “whether the challenged law burdens conduct protected by the Second Amendment.” Fyock, 779 F.3d at 996 (internal quotations omitted). In this step, we “explore the amendment's reach based on a historical understanding of the scope of the Second Amendment right.” Mai v. United States, 952 F.3d 1106, 1114 (9th Cir. 2020) (quoting United States v. Torres, 911 F.3d 1253, 1258 (9th Cir. 2019)). As we conduct this historical analysis, we must remain “well aware that we are jurists and not historians.” Young, 992 F.3d at 785. 6 Still, if the challenged law regulates conduct historically outside the scope of the Second Amendment, then it does not burden Second Amendment rights. Mai, 952 F.3d at 1114. But if the challenged law “falls within the historical scope of the Second Amendment, we must then proceed to the second step of the Second Amendment inquiry to determine the appropriate level of scrutiny.” Jackson v. City & County of San Francisco, 746 F.3d 953, 960 (9th Cir. 2014).
In our historical analysis, the Framers' understanding of the Second Amendment at and around the time of ratification has special significance. Laws from that time are particularly important because they are “contemporaneous legislative exposition[s] of the Constitution” that took place “when the founders of our government and framers of our Constitution were actively participating in public affairs.” Myers v. United States, 272 U.S. 52, 175, 47 S.Ct. 21, 71 L.Ed. 160 (1926). If they were also “acquiesced in for a long term of years,” these legislative expositions “fix[ ] the construction” that we must give to the Constitution's parameters. Id. Because the militias originated in the states, see Heller, 554 U.S. at 596, 128 S.Ct. 2783, we also consider colonial and state laws. Since the Second Amendment was incorporated against the states through the Fourteenth Amendment, our historical analysis also must consider how the right to keep and bear arms was understood in 1868, when that amendment was ratified. See McDonald, 561 U.S. 742, 770–78, 130 S.Ct. 3020 (analyzing Reconstruction-era history).
After the historical analysis, if we conclude that the law at issue burdens Second Amendment rights, then we proceed to the second step. In this step, we determine which level of scrutiny to apply and must decide both “how close [each] law comes to the core of the Second Amendment right” and “the severity of [each] law's burden on that right.” Mai, 952 F.3d at 1115. “Strict scrutiny applies only to laws that both implicate a core Second Amendment right and place a substantial burden on that right.” Id. (citing Torres, 911 F.3d at 1262). And “[i]n weighing the severity of the burden, we are guided by a longstanding distinction between laws that regulate the manner in which individuals may exercise their Second Amendment right, and laws that amount to a total prohibition of the right.” Pena, 898 F.3d at 977 (citing United States v. Chovan, 735 F.3d 1127, 1138 (9th Cir. 2013)). Laws that regulate how individuals can exercise the right are less severe; laws that amount to a total prohibition of the right are more severe.
To withstand intermediate scrutiny, first, “the government's stated objective [must] be significant, substantial, or important,” and second, there must be “a reasonable fit between the challenged regulation and the asserted objective.” Chovan, 735 F.3d at 1139. But to satisfy strict scrutiny, the law must be “justified by a compelling government interest and [be] narrowly drawn to serve that interest.” Brown v. Ent. Merchs. Ass'n, 564 U.S. 786, 799, 131 S.Ct. 2729, 180 L.Ed.2d 708 (2011).
We analyze the laws against this legal backdrop. First, because the Second Amendment historically protected the right of young adults to possess firearms, the district court abused its discretion in finding no burden on Second Amendment rights. As to the long gun regulation, the district court properly applied intermediate scrutiny, and did not abuse its discretion in finding the law likely to survive. But semiautomatic rifles are nearly totally banned. Thus, the district court erred in applying intermediate scrutiny, rather than strict scrutiny. And even under intermediate scrutiny, the district court erred in finding the law likely to survive. Finally, the district court also abused its discretion in finding that there was no irreparable harm and that the public interest favored declining to issue an injunction.
Before engaging with the historical record, we first establish the parameters of our analysis. California regulates young adults' commerce in firearms, not their possession. And we have avoided defining “the contours of the commercial sales category because [we have] assumed the Second Amendment applied and upheld the restriction under the appropriate level of constitutional scrutiny.” Pena, 898 F.3d at 976 (collecting cases). Still, even though this is a commercial regulation, the district court's historical analysis focused not on the history of commercial regulations specifically but on the history of young adults' right to keep and bear arms generally. See Jones, 498 F. Supp. 3d at 1325–29. The district court was asking the right question.
“Commerce in firearms is a necessary prerequisite to keeping and possessing arms for self-defense.” Teixeira v. County of Alameda, 873 F.3d 670, 682 (9th Cir. 2017). We have assumed without deciding that the “right to possess a firearm includes the right to purchase one.” Bauer v. Becerra, 858 F.3d 1216, 1222 (9th Cir. 2017). And we have already applied a similar concept to other facets of the Second Amendment. For example, “[t]he Second Amendment protects ‘arms,’ ‘weapons,’ and ‘firearms’; it does not explicitly protect ammunition.” Jackson, 746 F.3d at 967. Still, because “without bullets, the right to bear arms would be meaningless,” we held that “the right to possess firearms for protection implies a corresponding right” to obtain the bullets necessary to use them. Id. (citing Ezell v. City of Chicago, 651 F.3d 684, 704 (7th Cir. 2011)).
Similarly, without the right to obtain arms, the right to keep and bear arms would be meaningless. Cf. Jackson, 746 F.3d at 967 (right to obtain bullets). “There comes a point ․ at which the regulation of action intimately and unavoidably connected with [a right] is a regulation of [the right] itself.” Luis v. United States, 578 U.S. 5, 136 S. Ct. 1083, 1097, 194 L.Ed.2d 256 (2016) (Thomas, J., concurring in the judgment) (quoting Hill v. Colorado, 530 U.S. 703, 745, 120 S.Ct. 2480, 147 L.Ed.2d 597 (2000) (Scalia, J., dissenting)). 7 For this reason, the right to keep and bear arms includes the right to purchase them. And thus laws that burden the ability to purchase arms burden Second Amendment rights.
Finally, before we dive into the history, we pause to clear up two last points. First, because the long gun regulation and the semiautomatic rifle ban regulate different categories of guns and have different exceptions, we analyze them separately. And second, the Second Amendment does not protect the right to carry “dangerous and unusual weapons.” Heller, 554 U.S. at 627, 128 S.Ct. 2783. But that doesn't mean that weapons can be banned just because they're dangerous. Rather, “dangerous and unusual weapons” is a kind of historical term of art: Heller contrasted those arms with weapons “in common use at the time.” Id. Thus “the relative dangerousness of a weapon is irrelevant when the weapon belongs to a class of arms commonly used for lawful purposes.” Caetano v. Massachusetts, 577 U.S. 411, 418, 136 S.Ct. 1027, 194 L.Ed.2d 99 (2016) (Alito, J., concurring). Here, the district court held that “[b]oth long-guns and semi-automatic centerfire rifles are commonly used by law abiding citizens for lawful purposes such as hunting, target practice, and self-defense,” and thus that they are not “dangerous and unusual weapons” under Heller, 554 U.S. at 627, 128 S.Ct. 2783. Jones, 498 F. Supp. 3d at 1325. Similarly, semiautomatic weapons “traditionally have been widely accepted as lawful possessions.” Staples v. United States, 511 U.S. 600, 612, 114 S.Ct. 1793, 128 L.Ed.2d 608 (1994). We agree: long guns and semiautomatic rifles are not dangerous and unusual weapons.
Having cleared these last preliminary hurdles, the question now is, “based on a historical understanding of the scope of the Second Amendment right,” whether the right of young adults to bear arms is “conduct [that is] protected by the Second Amendment.” Mai, 952 F.3d at 1114 (internal citation omitted).
Our analysis of the historical record reveals several points which inform our exploration of the amendment's reach. First, the tradition of young adults keeping and bearing arms is deep-rooted in English law and custom. Going back many centuries, able-bodied English men at least fifteen years old were compelled to possess personal arms and had to take part in both the militia and other institutions that required them to keep and bear personal arms. Second, the American colonists brought that tradition across the Atlantic: the colonial militias almost always included all men 18 and older, and other institutions involving keeping and bearing arms made it to our shores, too. Third, at the time of the founding, all states required young adults to serve in the militia, and all states required young adults to acquire and possess their own firearms. Just after the founding, Congress established a federal militia, which included young adults, and required them to acquire and possess their own weapons. Fourth, both at the founding and later, different states had different ages of majority, and the age of majority also varied depending on the conduct at issue. And finally, turning to the Reconstruction era, some states passed laws that regulated minors' access to firearms, but most of them only regulated handguns, and only a few banned all sales of firearms to minors. We explore each of these points in turn.
The tradition of young adults keeping and bearing arms is deep-rooted in English law and custom. As far back as medieval times, able-bodied men aged fifteen and older were compelled to possess personal arms and had a duty, when asked, to use those personal arms to maintain the king's peace and protect their communities and property. 8 “[T]he militia from its obscure origin in Saxon times has been composed of all subjects and citizens capable of bearing arms, regardless of age or parental authority.” 9
And the militia was not the only institution imposing an obligation to acquire and possess arms: “[u]nder English law originating long before the Norman Conquest of 1066, all able-bodied men were obliged to join in the hutesium et clamor (hue and cry) to pursue fleeing criminals.” 10 More generally, sheriffs, coroners, and magistrates could “summon all able-bodied males to assist in keeping the peace,” 11 and the traditional minimum age for these law-enforcement duties was typically 15 or 16 years old. 12 For example, at common law, the sheriff could command citizens—already armed—to help suppress riots, arrest criminals, and otherwise enforce civil processes. 13
This deep-rooted tradition was brought across the Atlantic by the American colonists. Heller confirmed that the “militia” in colonial America consisted of “a subset of ‘the people’—those who were male, able bodied, and within a certain age range.” 554 U.S. at 580, 128 S.Ct. 2783. Before ratification, when militias were solely defined by state law, most colonies and states set the age for militia enlistment at 16. See Appendix 1. Every colony passed, at some point, laws identifying 18-year-olds as persons required to possess arms. Id. Throughout the colonial period, the minimum age fluctuated both below and above 18, and some colonies passed laws temporarily increasing the minimum age requirements for militia service to not include 18- to 20-year-olds. Id.
Militia members had to show up for militia duty with their own arms. 14 When militia members were “called for service th[ey] ․ were expected to appear bearing arms supplied by themselves and of the kind in common use at the time.” United States v. Miller, 307 U.S. 174, 179, 59 S.Ct. 816, 83 L.Ed. 1206 (1939). Colonial governments even supplied arms to citizens too poor to purchase them, requiring them, for example, to pay back the government or work off their debt. 15
Militia membership also included some of what we might now call regulation: “members of the militia were required to meet regularly for weapons inspection and registration.” Jones, 498 F. Supp. 3d at 1327 (citing Saul Cornell & Nathan DeDino, A Well Regulated Right: The Early America Origins of Gun Control, 73 Fordham L. Rev. 487, 509–11 (2004)).
Along with the militia, the colonists also brought over the practice of posse comitatus, which again required citizens to have their own arms. 16 “Prior to the advent of centralized police forces,” posse comitatus allowed “sheriffs and others [to] compel[ ] citizens to serve in the name of the state to execute arrests, level public nuisances, and keep the peace, upon pain of fine and imprisonment.” 17 And in fact, the colonists didn't just continue the practice: posse comitatus was “a pillar of local self-governance” and “central to the broader project of protecting the public good.” 18 Colonial governments even punished citizens who would not join the posse. 19
The Second Amendment was ratified just a few months before Congress passed the Militia Act of 1792. The Militia Act required that young adults serve in the militia and acquire and possess their own weapons. The Act “purported to establish ‘an Uniform Militia throughout the United States.’ ” Perpich v. Dep't of Def., 496 U.S. 334, 341, 110 S.Ct. 2418, 110 L.Ed.2d 312 (1990) (internal citation omitted). The Act stated: “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, 1 Stat. 271. The Act also required each militia member to “provide himself with a good musket or firelock ․ or with a good rifle.” Id. The Militia Act thus “command[ed] that every able-bodied male citizen between the ages of 18 and 45 be enrolled [in the militia] and equip himself with appropriate weaponry.” Perpich, 496 U.S. at 341, 110 S.Ct. 2418.
Thus, “[a]t the time of the Second Amendment's passage, or shortly thereafter, the minimum age for militia service in every state became eighteen.” Nat'l Rifle Ass'n of Am., Inc. v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 714 F.3d 334, 340–44 (5th Cir. 2013) (Jones, J., dissenting from denial of rehearing en banc) (“NRA II”). 20 Several states adopted the exact language from the Federal Militia Act—obligating male persons 18 years old or older to acquire and provide their own firearms. See Appendix 2. Either at the same time as or right after the Act's passage, every state's militia law obliged young adults to acquire and possess firearms. Id. “[A]ny argument that 18- to 20-year olds were not considered, at the time of the founding, to have full rights regarding firearms” is “inconceivable.” NRA II, 714 F.3d at 342 (Jones, J., dissenting from denial of rehearing en banc).
Turning now to the age of majority, the common law age of majority at the time of the founding was 21 years old. “[I]t was not until the 1970s that States enacted legislation to lower the age of majority to 18.” NRA I, 700 F.3d at 201. But the relevant age of majority also depended on the capacity or activity. William Blackstone, Commentaries 463–64, 465 (1765). In other words, “the age of majority—even at the Founding—lacks meaning without reference to a particular right,” because, “[f]or 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.
Finally, we turn to the Reconstruction era. “By the 1850's, the perceived threat that had prompted the inclusion of the Second Amendment in the Bill of Rights—the fear that the National Government would disarm the universal militia—had largely faded as a popular concern, but the right to keep and bear arms was highly valued for purposes of self-defense.” McDonald, 561 U.S. at 770, 130 S.Ct. 3020. And even once the Fourteenth Amendment was ratified in 1868, it would of course still be many years before the Supreme Court incorporated the Second Amendment against the states. McDonald, 561 U.S. at 791, 130 S.Ct. 3020. So, like in the colonial and founding eras, state laws were made against the backdrop of “Second Amendment analogues in their respective [state] constitutions.” NRA I, 700 F.3d at 202 n.16.
Within a few decades of Reconstruction, some states had enacted laws regulating access to firearms by minors. Id. at 202. We identify twenty-eight such state laws passed between 1856 and 1897. See Appendix 3. Of these laws, nineteen banned sales of only pistols to minors, and several had exceptions for hunting or parental consent. Of the non-pistol bans, three only applied to minors under fifteen years old, only required parental consent, or both. Eight states banned the sale of all firearms or deadly or dangerous weapons to minors. Four of these statutes were passed between 1881 and 1885.
There were also other Reconstruction-era restrictions on the right to acquire and bear arms. In particular, some statutes were designed to disarm formerly enslaved people and members of Native American tribes. See Drummond v. Robinson Twp., 9 F.4th 217, 228 (3d Cir. 2021). Kentucky, for example, restricted firearm access by African Americans. 1860 Ky. Acts 245 § 23.
For the most part, cases from this time did not address the constitutionality of laws that regulated firearm ownership by young adults. Two cases touch on related issues, but neither addresses our question. One of them, Coleman v. State, 32 Ala. 581 (1858), summarily affirmed a lower court's application of a state statute that prohibited selling or lending a pistol to a minor. But the court did not address the constitutionality of the law or say how old the minor was. In the second case, State v. Callicutt, 69 Tenn. 714 (1878), on top of not saying how old the minor was, that court also addressed concealed carry of dangerous weapons, not the right to keep and bear arms more generally.
Professor Cooley's famous treatise from 1868, relied on by the Fifth Circuit panel in NRA I, also does not address the question: its sole reference to the issue, citing Callicutt, comes in a discussion of the states' police powers, not of the right to keep and bear arms. NRA I, 700 F.3d at 202–03 (citing Thomas M. Cooley, Treatise on Constitutional Limitations 740 n.4 (5th ed. 1883)).
We must decide what these historical facts tell us about the reach of the Second Amendment. Fyock, 779 F.3d at 996. According to Plaintiffs, these facts show that the Second Amendment protects young adults' right to bear arms, because young adults were expected to bear arms at the time of the founding.
Defendants have two main responses, both of which the district court adopted. First, it argues that the protected historical right is not a full right to bear arms, but rather only a right to bear arms that comes with some obligations of militia service, at the very least the inspection requirement. Jones, 498 F. Supp. 3d at 1327. In the district court's reading, because militia service came with some regulation, the Second Amendment does not protect the right to keep and bear arms, absent that regulation. Id.
Second, Defendants argue that the militia laws don't show anything about young adults' right to bear arms, because states in the 19th and 20th centuries also criminalized transferring firearms to young people, and because the age of majority during much of this country's history was 21 years old, not 18. Id. at 1326–27.
We agree with Plaintiffs: the historical record shows that the Second Amendment protects young adults' right to keep and bear arms. We address Plaintiffs' argument and then each of Defendants' counterarguments in turn.
“Sixteen was the minimum age for colonial militias almost exclusively for 150 years before the Constitution” and “[a]t the time of the Second Amendment's passage, or shortly thereafter, the minimum age for militia service in every state became eighteen.” NRA II, 714 F.3d at 340 (Jones, J., dissenting from denial of rehearing en banc). This historical militia tradition supports Plaintiffs' reading. Indeed, the historical evidence is so strong that even the dissenting judge in the vacated Hirschfeld opinion found it “persuasive,” did not dispute it, and simply assumed that the law did burden Second Amendment rights, disagreeing only at step two. 5 F.4th 407, 463 (4th Cir. 2021) (Wynn, J., dissenting), vacated as moot, 14 F.4th 322 (4th Cir. 2021). The Second Amendment refers to the militia, and young adults had to be in the militia and bring their own firearms. This reference implies at least that young adults needed to have their own firearms.
Defendants' first argument to the contrary is unpersuasive. Defendants agree that young adults needed to have firearms for the militia and that the Second Amendment refers to the militia. Even so, Defendants argue that the Second Amendment only protects older adults' right to keep and bear arms, and not that of young adults. In other words, young adults could keep and bear arms and had to serve in the militia, but their ability to keep and bear arms was not protected by the Second Amendment and could have been abridged at any time without posing any burden on the right. Because it strays from the most obvious historical interpretation, this reading would need to be supported by powerful evidence. It is not.
To begin, the district court's main premise has already been rejected. “[T]he Second Amendment conferred an individual right to keep and bear arms.” Heller, 554 U.S. at 595, 128 S.Ct. 2783. The right is not conditioned on militia service. Id. at 599–600, 128 S.Ct. 2783. Indeed, that was the position of the dissenters in Heller, and the Court rejected it. Id.
The district court's position here is a variation on that same, already-rejected argument. Rather than argue that all citizens' right to bear arms is conditioned entirely on militia service, as the dissenters did in Heller, the district court held that some citizens' right to bear arms is conditioned on some aspects of militia service. Jones, 498 F. Supp. 3d at 1327. And there is another problem with the district court's analysis. At the first step, we just ask whether the regulations burden Second Amendment rights at all. Few, if any, of our constitutional rights are absolute, and asking if a right is burdened is different from asking if a particular burden is constitutional. That there were some firearm regulations associated with militia membership could show that some restrictions can be constitutional. But the regulations themselves cannot dispositively show that there is no burden.
The historical analysis controls the first step of the inquiry but not the second. In applying a tier of scrutiny in the second step, we focus not on the historical record (i.e., what kinds of regulations were present at the founding), but on the gravity of the state's interest (compelling/significant/ legitimate) and the degree of tailoring between the regulation and that interest (narrow tailoring/reasonable fit/rational relation). In finding no burden on Second Amendment rights, the district court improperly relied on founding-era regulations.
We now turn to Defendants' second argument, which relies on laws passed in the 19th and 20th centuries.
First, Defendants fail to adequately address the founding-era militia tradition: “19th-century sources may be relevant to the extent they illuminate the Second Amendment's original meaning, but they cannot be used to construe the Second Amendment in a way that is inconsistent with that meaning.” NRA II, 714 F.3d at 339 n.5 (Jones, J., dissenting from denial of rehearing en banc). Defendants argue, citing NRA I, that “a regulation can be deemed ‘longstanding’ even if it cannot boast a precise founding era analog.” 700 F.3d at 196. But even if we were to agree, that would not save the argument. Here, there is not just a vacuum at the founding era: instead, the founding-era evidence of militia membership undermines Defendants' interpretation.
Even putting that aside, the Reconstruction-era laws themselves are not convincing. On top of the deeply offensive nature of many of them, nineteen out of twenty-eight banned only the sale of handguns, and California's handgun ban is not at issue. The Reconstruction-era laws show that long guns were far less regulated than handguns. Ruling out other state laws that are similarly inapplicable (laws only requiring parental consent, only banning dangerous and deadly weapons, and only applying to children under fifteen years old), we are left with only five complete bans on sales of firearms to minors. Of these five laws, three were passed in states without a Second Amendment analog in their state constitution. 21 So only two states—Kentucky and Michigan—banned the sale of firearms to minors, see 1873 Ky. Acts 359, 1883 Mich. Pub. Acts 144, and had a Second Amendment analog, see Ky. Const. of 1850, Art. 13, § 25; Mich. Const. of 1850, Art. 17, § 7. These two laws—both passed over a decade after the ratification of the Fourteenth Amendment—cannot contravene the Second Amendment's original public meaning.
As to Defendants' argument relying on the age of majority being 21, rather than 18, we agree with the Fifth Circuit and the Fourth Circuit's vacated opinion in Hirschfeld that “majority or minority is a status that lacks content without reference to the right at issue.” 5 F.4th at 435; NRA I, 700 F.3d at 204 n.17. 22 “As Blackstone's Commentaries makes clear, the relevant age of majority depended on the capacity or activity.” Hirschfeld, 5 F.4th at 435 (citing 1 William Blackstone, Commentaries at 463–65). We also agree that “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” and that “the age of majority Blackstone identifies for different activities tells us little about the scope of the Second Amendment's protections.” Id.
Finally, Defendants argue that California's laws are just “conditions and qualifications on the commercial sale of arms,” or “longstanding prohibitions on the possession of firearms” by certain groups. Because of the hunting license exception, the long gun regulation is more naturally considered a “condition or qualification,” while the semiautomatic rifle ban is more aptly categorized as a “prohibition.” Heller itself called such measures “presumptively lawful,” 554 U.S. at 626–27 n.26, 128 S.Ct. 2783, so Defendants argue that California's laws pose no burden on Second Amendment rights. We disagree. These laws burden Second Amendment rights, notwithstanding this observation from Heller.
First, the “longstanding prohibitions” referred to in Heller were “prohibitions on the possession of firearms by felons and the mentally ill,” id. at 626, 128 S.Ct. 2783, not prohibitions on a broader set of groups. Young adults are neither felons nor mentally ill. The semiautomatic rifle law does not fall within the Supreme Court's enumerated categories.
Second, as to the long gun law, there is a more fundamental problem. In Heller, the Supreme Court noted just that “nothing in [its] opinion should be taken to cast doubt on” laws such as “conditions and qualifications on the commercial sale of arms,” and that such laws were “presumptively lawful.” Id. at 627, n.26, 128 S.Ct. 2783. But this does not mean that all such laws pose no burden on Second Amendment rights at all. “On the one hand, this language could be read to suggest the identified restrictions are presumptively lawful because they regulate conduct outside the scope of the Second Amendment. On the other hand, it may suggest the restrictions are presumptively lawful because they pass muster under any standard of scrutiny.” Pena, 898 F.3d at 976 (citing United States v. Marzzarella, 614 F.3d 85, 91 (3d Cir. 2010)). The answer need not be the same for every regulation. Some presumptively lawful measures might burden conduct unprotected by the Second Amendment, while others might presumptively pass the applicable level of scrutiny.
Here, our historical analysis leads us to conclude that young adults have a Second Amendment right to keep and bear arms. Because that right includes the right to purchase arms, both California laws burden conduct within the scope of the Second Amendment. The long gun law is a “condition[ ] ․ on the commercial sale of arms,” Heller, 554 U.S. at 627, 128 S.Ct. 2783, but it still burdens Second Amendment rights. The Supreme Court's observation in Heller is no obstacle to this holding.
* * *
Ultimately, the Second Amendment protects the right of the people to keep and bear arms and refers to the militia. Young adults were part of the militia and were expected to have their own arms. Thus, young adults have Second Amendment protections as “persons who are a part of a national community.” Id. at 580, 128 S.Ct. 2783 (citing Verdugo–Urquidez, 494 U.S. at 265, 110 S.Ct. 1056). Defendants point to contemporaneous regulations, arguing that some states banned young adults from having firearms later on, and that the age of majority was 21, not 18. But these observations do not prove their point: permissible regulations can still burden the right, later laws cannot contravene the original public meaning, and the age of majority depends on the conduct. The California laws burden Second Amendment rights and the district court erred in concluding otherwise.
Having concluded that the laws burden Second Amendment rights, we now consider the district court's choice of a tier of scrutiny and its application of that tier. Fyock, 779 F.3d at 996.
The district court properly applied intermediate scrutiny to the long gun regulation but should have applied strict scrutiny to the semiautomatic rifle ban. Similarly, the district court did not abuse its discretion in holding that the long gun regulation was likely to survive. But even under intermediate scrutiny, the district court still abused its discretion in holding that the semiautomatic rifle ban was likely to survive.
First, we must determine the appropriate level of scrutiny. Reviewing de novo, see Joelner, 508 F.3d at 431, we consider both “how close [each] law comes to the core of the Second Amendment right” and “the severity of [each] law's burden on [that] right,” Mai, 952 F.3d at 1115 (citing Chovan, 735 F.3d at 1138). Laws that regulate how individuals can exercise the right are less severe; laws that amount to a total prohibition of the right are more severe. See Pena, 898 F.3d at 977 (citing Chovan, 735 F.3d at 1138). Similarly, “firearm regulations which leave open alternative channels for self-defense are less likely to place a severe burden on the Second Amendment right than those which do not.” Jackson, 746 F.3d at 961 (citation omitted). The district court properly applied intermediate scrutiny to the long gun regulation, but improperly applied it to the semiautomatic rifle ban.
As to the long gun regulation, the district court properly applied intermediate scrutiny. The burden on Second Amendment rights posed by this rule on its face is not severe. This rule facially more aptly regulates “the manner in which persons may exercise their Second Amendment right.” Id.
The long gun regulation allows a young adult to buy a long gun if he gets a hunting license. This requirement does not prevent young adults from having any firearms or from using them in any particular way. Because this regulation does not impose a significant burden on the Second Amendment right to keep and bear arms, the district court properly applied intermediate scrutiny.
As to the semiautomatic rifle ban, we part company with the district court. Strict scrutiny applies. The main difference between this ban and the long gun regulation is the exceptions. The long gun regulation has a readily available exception, at least on its face—young adults can get hunting licenses.
The semiautomatic rifle ban has no such exception: the only young adults who can buy semiautomatic rifles are some law enforcement officers and active-duty military servicemembers.
It's one thing to say that young adults must take a course and purchase a hunting license before obtaining certain firearms. But to say that they must become police officers or join the military? For most young adults, that is no exception at all. 23 In effect, this isn't an exception that young adults can avail themselves of by joining the police force or military; it is a blanket ban for everyone except police officers and servicemembers.
We have never held that intermediate scrutiny applied to a rule that banned the purchase of a major category of firearm. To the contrary, our cases applying intermediate scrutiny have dealt with two kinds of laws. First, we have applied intermediate scrutiny to laws that govern conduct outside the core of the Second Amendment because the actors are not “law-abiding, responsible citizens” under Heller. See Torres, 911 F.3d at 1262–63; United States v. Singh, 979 F.3d 697, 725 (9th Cir. 2020) (citing Torres, 911 F.3d at 1262–63); Mai, 952 F.3d at 1115. This rule does not apply here. And second, we have applied intermediate scrutiny to laws that regulate either the way people can obtain or use firearms, or auxiliary features of those firearms. See Pena, 898 F.3d at 977–78 (requiring specific safety features and microstamped serial numbers); Bauer, 858 F.3d at 1221–22 (using firearm purchase fees to fund law enforcement programs); Silvester v. Harris, 843 F.3d 816, 827 (9th Cir. 2016) (ten day waiting period to purchase a gun); Fyock, 779 F.3d at 998–99 (ban of “large capacity” magazines); Jackson, 746 F.3d at 963–65 (regulation of sale of bullets); Duncan v. Bonta, 19 F.4th 1087, 1104 (9th Cir. 2021) (“large capacity” magazines). We have held that intermediate scrutiny applies to laws like these because they regulate the way people can exercise their Second Amendment rights. Indeed, in Duncan, we were careful to avoid approving of applying only intermediate scrutiny to laws banning certain firearms entirely. Id. at 1104. We noted that the law in that case “ha[d] no effect whatsoever on which firearms may be owned” and that, as far as that law was concerned, “anyone may own any firearm at all.” Id. The opposite is true here: this law bans almost all young adults from having semiautomatic rifles. We have never applied intermediate scrutiny to a ban like this.
We have often observed that there is “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, 952 F.3d at 1115 (citing Torres, 911 F.3d at 1262). But this observation makes no difference here for a simple reason: the level of scrutiny depends on the law at issue. That states and localities at one point had passed laws that demanded only intermediate scrutiny analysis says little about what kinds of laws they may have passed later. Indeed, if states pass increasingly strict gun laws, those laws may demand higher scrutiny, especially considering their cumulative effect.
Handguns are the quintessential self-defense weapon, see Heller, 554 U.S. at 629, 128 S.Ct. 2783, but young adults already cannot purchase them, Cal. Penal Code § 27505, 18 U.S.C. § 922(b)(1). And under this ban, they also cannot purchase semiautomatic centerfire rifles. That leaves non-semiautomatic centerfire rifles, rimfire rifles, and shotguns. Non-semiautomatic rifles are not effective as self-defense weapons because they must be manually cycled between shots, a process which becomes infinitely more difficult in a life or death situation. Rimfire rifles generally aren't good for self-defense either, because rimfire ammunition has “poor stopping power” and are mostly used for things like hunting small game. David Steier, Guns 101, 13 (2011). So for self-defense in the home, young adults are left with shotguns.
Even acknowledging that shotguns are effective weapons for self-defense in the home, shotguns are outmatched by semiautomatic rifles in some situations. 24 Semiautomatic rifles are able to defeat modern body armor, have a much longer range than shotguns and are more effective in protecting roaming kids on large homesteads, are much more precise and capable at preventing collateral damage, and are typically easier for small young adults to use and handle.
Thus, we hold that California's ban is a severe burden on the core Second Amendment right of self-defense in the home. Young adults already cannot buy the quintessential self-defense weapon, Heller, 554 U.S. at 629, 128 S.Ct. 2783, and this ban now stops them from buying semiautomatic rifles, leaving only shotguns. So handguns aside, this law takes away one of the two remaining practical options for self-defense in the home, and leaves young adults with a self-defense weapon which is not ideal or even usable in many scenarios. That is a severe burden.
In arguing that the burden is not severe, the dissent points first to the intrafamily transfer and loan provisions. Dissent at 755. We disagree that these provisions sufficiently alleviate the burden. To start, young adults remain severely restricted in getting firearms through family transfers: Gifts from parents and grandparents are allowed but strawman purchases are not. See Cal. Penal Code §§ 27875 (family transfers), 27515 (strawman purchases). Moreover, allowing family transfers but not purchases makes young adults' Second Amendment rights conditional on the rights of others. The family transfer provision is unavailable to young adults whose parents or grandparents have passed away, do not have a gun to transfer, or are unable or unwilling to participate in a transfer. The first loan provision, which permits loans of up to thirty days from a slightly broader subset of family members, suffers from similar problems, and is temporally limited. Cal. Penal Code § 27880. And the remaining loan provisions are only available in even more limited circumstances: for only three days and only if the firearm is used in the presence of the loaner, Cal. Penal Code § 27885; if the firearm stays only at the loaner's residence, Cal. Penal Code § 27881; or if the loan is only for the hunting season, which is only part of the year, Cal. Penal Code § 27950. These provisions do not alleviate the sales ban's severe burden on the right of self-defense in the home.
The dissent's second rationale is that California's ban does not impose a severe burden because young adults can just wait to buy semiautomatic rifles until they are 21. Dissent at 755. It's true that we've applied intermediate scrutiny to a ten-day waiting period. Silvester, 843 F.3d at 827. But telling young adults to wait up to three years is a much more severe burden than having to wait a week and a half. We are not aware of any precedent that has adopted the dissent's rationale. Indeed, telling an 18-year-old that he can vote when he turns 21 would hardly minimize the existing constitutional deprivation.
Finally, the dissent argues that our reasoning is circular because any subset of guns can be considered a category. Dissent at 755–56; see Worman v. Healey, 922 F.3d 26, 32 n.2 (1st Cir. 2019). But we do not hold that a ban of any kind or subset of gun must necessarily receive strict scrutiny. We hold just that this ban of semiautomatic rifles requires strict scrutiny, because handguns are already banned, and semiautomatic rifles are now effectively banned. That means two of the three types of effective self-defense firearms are banned, leaving young adults with limited or ineffective alternatives in many self-defense scenarios, and severely burdens their Second Amendment rights.
Having determined that intermediate scrutiny applies to the long gun regulation, we now review for abuse of discretion the district court's application of that test. In finding that the long gun regulation was likely to survive intermediate scrutiny, Jones, 498 F. Supp. 3d at 1329–30, the district court did not abuse its discretion.
California's objective is “to increase public safety through sensible firearm control and limit access to certain firearms for some Young Adults with proper safety training.” Id. at 1330. In its brief, Defendants referred to the objective more broadly as promoting public safety and reducing gun violence and crime.
Though public safety is important, firearms were also dangerous in 1791, when the Second Amendment was ratified, and the government then also had an interest in promoting public safety. This is not a standalone government interest separate from the Second Amendment: The Second Amendment itself, “[l]ike the First, ․ is the very product of an interest balancing by the people.” Heller, 554 U.S. at 635, 128 S.Ct. 2783; see also Heller v. District of Columbia, 670 F.3d 1244, 1277 (D.C. Cir. 2011) (Kavanaugh, J., dissenting). “[T]he Second Amendment has already made the basic policy choice for us.” Ass'n of N.J. Rifle & Pistol Clubs v. Att'y Gen. of N.J., 910 F.3d 106, 129 (3d Cir. 2018) (Bibas, J., dissenting) (citing Heller, 554 U.S. at 634–36, 128 S.Ct. 2783). Thus, in the reasonable fit part of the analysis, the importance of the interest has no effect: once the interest is shown to be important, the question becomes whether the law is a reasonable fit. The importance of the interest cannot override Second Amendment rights.
Defendants will likely be able to show that California's long gun regulation is a reasonable fit for the stated objectives. The main effect of the rule is to require young adults to take a hunter education class before they can get long guns. So whether the rule is a reasonable fit depends on what the law requires to happen in these hunter education classes.
Some context for the hunter education classes is helpful. Generally, before purchasing a gun, Californians must get an FSC. Cal. Penal Code §§ 31615, 27540(e). Getting the certificate requires passing a multiple-choice test and a safe handling demonstration, both of which can happen at the point of sale. Id. In enacting the regulation at issue, however, California changed the requirements for young adults. Rather than having to get an FSC, a young adult must instead get a hunting license, which requires them to first take and pass a hunter education class. California offers in-person and hybrid class options. The course takes approximately ten hours and costs less than $30. After passing the course, a young adult may purchase a hunting license for $54.
The class covers “firearm safety information” that is “more extensive” than what is covered by the FSC test and demonstration. The class also discusses other aspects of hunting that are less relevant to non-hunting uses of long guns (e.g., conservation). Cal. Fish & Game Code § 3051(a).
So, overall, California wants to “increase public safety through sensible firearm control.” Jones, 498 F. Supp. 3d at 1330. We agree with the district court that sensible firearm control includes things like “proper training and maintenance of firearms.” Id. California has pursued that end by requiring young adults to take a class which teaches them, among other things, “firearm safety information.” Because the hunting classes include other, unrelated information, the requirement is not a perfect fit. In other words, this requirement likely is neither narrowly tailored nor the least restrictive means for achieving California's goal. But it doesn't have to be: it only has to be a reasonable fit. And it likely is.
Before moving on to the semiautomatic rifle ban, we pause to make one last point. In their complaint, Plaintiffs have challenged the long gun regulation facially and as applied. But they appeal the denial of the preliminary injunction only on the basis that the law is facially invalid. And in evaluating a facial challenge, we consider only the text of the law—we judge the law on its face, not in its application. See Calvary Chapel Bible Fellowship v. County of Riverside, 948 F.3d 1172, 1176 (9th Cir. 2020). Nothing we have said forecloses the possibility that the regulation might still be unconstitutional as applied. For example, if the hunter education courses were prohibitively expensive or were only offered on a limited basis, then California might be applying the regulation unconstitutionally. Still, as to the facial challenge at issue, the district court did not abuse its discretion in finding that the regulation would survive intermediate scrutiny.
As to the semiautomatic rifle ban, because we have held that strict scrutiny applies, we reverse on that basis. Even so, we also hold in the alternative that, even if intermediate scrutiny were to apply, the district court still abused its discretion in finding that the ban was likely to survive, and reverse on this alternative basis as well. (Because we hold that the ban is unlikely to survive intermediate scrutiny, we also by implication hold that it is even less likely to survive strict scrutiny.) We first clarify the nature of the intermediate scrutiny test, and then discuss its application here.
“[A]ll forms of the [intermediate scrutiny] standard require (1) the government's stated objective to be significant, substantial, or important; and (2) a reasonable fit between the challenged regulation and the asserted objective.” Jackson, 746 F.3d at 965 (citing Chovan, 735 F.3d at 1139). Unfortunately, despite regularly acknowledging that a reasonable fit is required, we have increasingly dispensed with the fit requirement, relying instead on a cherry-picked formulation of the rule that requires only that the regulation “promote a substantial government interest that would be achieved less effectively absent the regulation.” See, e.g., Mai, 952 F.3d at 1116. Still, we have not silently transformed the test: intermediate scrutiny continues to require an analysis of whether the regulation is a reasonable fit for the government's objective, not just an assessment of whether it does anything at all.
We transported intermediate scrutiny into the Second Amendment context from First Amendment cases. See Fyock, 779 F.3d at 1000 (citing Colacurcio v. City of Kent, 163 F.3d 545, 553 (9th Cir. 1998)). To satisfy intermediate scrutiny in the First Amendment context, the regulation must first “promote[ ] a substantial government interest that would be achieved less effectively absent the regulation.” United States v. Albertini, 472 U.S. 675, 689, 105 S.Ct. 2897, 86 L.Ed.2d 536 (1985). In other words, the regulation must accomplish something. But that's not all: “[t]o be sure, this standard does not mean that a time, place, or manner regulation may burden substantially more speech than is necessary to further the government's legitimate interests.” Ward v. Rock Against Racism, 491 U.S. 781, 799, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989) (citing Frisby v. Schultz, 487 U.S. 474, 485, 108 S.Ct. 2495, 101 L.Ed.2d 420 (1988)). The government still “may not regulate expression in such a manner that a substantial portion of the burden on speech does not serve to advance its goals.” Id. This is the essence of the intermediate scrutiny test: the regulation must be a reasonable fit for the government's stated objective, which means not just that it accomplishes something, but also that it does not burden far more speech than is necessary.
Unfortunately, in our Second Amendment cases, we have sometimes omitted one-half of the inquiry. When we transplanted intermediate scrutiny from the First Amendment to the Second, we continued to say that intermediate scrutiny requires “a reasonable fit between the challenged regulation and the asserted objective,” but we did not bring the does-not-burden-more-conduct-than-necessary part. Fyock, 779 F.3d at 1000 (citing Colacurcio, 163 F.3d at 553) (but leaving off the second half of the test); Peruta, 824 F.3d at 942 (en banc); id. at 946 (Graber, J., concurring) (citing Fyock, 779 F.3d at 1000); Silvester, 843 F.3d at 829 (citing Fyock, 779 F.3d at 1000); Mahoney v. Sessions, 871 F.3d 873, 882 (9th Cir. 2017) (citing Fyock, 779 F.3d at 1000); Torres, 911 F.3d at 1264 (citing Fyock, 779 F.3d at 1000); Singh, 979 F.3d at 725 (citing Fyock, 779 F.3d at 1000); Mai, 952 F.3d at 1116 (citing Torres, 911 F.3d at 1263). And bringing only the first half of the test is “incomplete” because “[i]ntermediate scrutiny also requires that a law not burden substantially more protected activity than is necessary to further the government's interest.” Silvester v. Becerra, ––– U.S. ––––, 138 S. Ct. 945, 950, 200 L.Ed.2d 293 (2018) (Thomas, J., dissenting from denial of certiorari).
When we omit the second part of the inquiry, we neglect to consider fit at all. Something fits with something else if it is “well adapted or suited to the conditions or circumstances of the case” or if it is “proper or appropriate.” “Fit, adj.,” Oxford Dictionary of English (3d ed.) (2010). So a law is a good fit for a goal if it regulates only when it helps achieve that goal, and not in other instances. The more innocent conduct that is regulated, the less good a fit the law is. And conversely, sweeping in less innocent conduct makes for a better fit.
Asking only if the regulation accomplishes something does not address “fit” at all: A straw and a two-foot pipe both transport fluids, but only one of them is a reasonable fit for drinking a soda. Intermediate scrutiny requires us to ask whether a regulation is a reasonable fit for the government's stated objective. And that means that we have to consider fit.
The dissent offers a rationale for why intermediate scrutiny should be different in the Second Amendment context, relative to the First Amendment, and suggests that our failure to bring over the second part of the test was purposeful. Dissent at 756–57. But this is an after-the-fact rationalization, because in the series of cases in which we used a weaker version of the rule, we simply left off the second part of the test without explaining why. See, e.g., Fyock, 779 F.3d at 1000 (citing Colacurcio, 163 F.3d at 553). And contrary to the dissent's rationale, a majority of judges in a recent en banc panel also recently reaffirmed that there is “no merit to the suggestion that the Ninth Circuit's application of intermediate scrutiny in Second Amendment cases is somehow less exacting than its application of the standard in other kinds of cases.” Duncan, 19 F.4th at 1138 (Berzon, J., concurring). In any case, we have continued to acknowledge that “all forms of the [intermediate scrutiny] standard require ․ a reasonable fit between the challenged regulation and the asserted objective.” Jackson, 746 F.3d at 965 (citing Chovan, 735 F.3d at 1139). But the dissent's version of the rule does not contain an analysis of fit.
California's stated objective for the semiautomatic rifle ban is the same as for the long gun regulation: to promote public safety and reduce gun violence and crime. Jones, 498 F. Supp. 3d at 1330. The question is whether the ban—prohibiting commerce in semiautomatic rifles for all young adults except those in the police or military—is a reasonable fit for that aim.
We agree with Defendants that the fit need only be reasonable, not perfect. Jackson, 746 F.3d at 969. But the fit here is likely not even reasonable. The district court abused its discretion in finding that Defendants could likely show a reasonable fit.
In Craig v. Boren, the Supreme Court considered a law that was a much better fit than this law and still found the fit unreasonable. 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976). The law in Craig v. Boren banned the sale of some beer to men between 18 and 21, but not to women in the same age range. Id. at 191–92, 97 S.Ct. 451. Intermediate scrutiny applied, and the objective of the law was to enhance traffic safety. Id. at 199, 97 S.Ct. 451. The state argued that its law was a reasonable fit for that objective because young men were more than ten times more likely to be arrested for driving under the influence than young women. Id. at 199–201, 97 S.Ct. 451. But the plaintiff argued that the law was overbroad: only 2% of young men were arrested for drunk driving, but the law regulated all young men. In other words, the law regulated fifty times more men than was ideal: it regulated 100% of them, even though only 2% would drive drunk.
The Supreme Court struck down the law. “While such a disparity is not trivial in a statistical sense, it hardly can form the basis for employment of a gender line as a classifying device.” Id. at 201, 97 S.Ct. 451. In other words, a ten times increase in risk cannot justify regulating fifty times more people than is ideal: “if maleness is to serve as a proxy for drinking and driving, a correlation of 2% must be considered an unduly tenuous ‘fit.’ ” Id.
The fit here is far more tenuous than that. In adopting the ban at issue, the California legislature considered various statistics. In particular, it knew that young adults were less than 5% of the population but accounted for more than 15% of homicide and manslaughter arrests. In other words, young adults are more than three times more likely to be arrested for homicide and manslaughter than other adults. But as Plaintiffs point out, only 0.25% of young adults are arrested for violent crimes. In other words, California's law sweeps in 400 times (100% divided by 0.25%) more young adults than would be ideal. 25 Because it regulates so much more conduct than necessary to achieve its goal, the law is unlikely to be a reasonable fit for California's objectives.
On this point, the dissent argues that we have transformed intermediate scrutiny into an impermissible “rigid statistical framework.” Dissent at 758 (citing Hirschfeld, 5 F.4th at 479 (Wynn, J., dissenting)). But we do not hold that Craig v. Boren established a rigid, bright-line, statistical rule for reasonable fit. Instead, we simply note that Craig v. Boren is an example of a law that was not a reasonable fit. We establish no rigid statistical framework; we use a few numbers only to compare Craig v. Boren with this case, and to illustrate that the fit here is substantially more tenuous. The dissent also argues that we compare “apples to oranges” because Craig v. Boren concerned gender discrimination, for which intermediate scrutiny applies, whereas this case is about age discrimination, for which, under the Fourteenth Amendment, we only review for rational basis. Dissent at 759; see Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 83, 120 S.Ct. 631, 145 L.Ed.2d 522 (2000). But the dissent agrees that intermediate scrutiny applies here, not rational basis. It is the Second Amendment we are applying after all.
We pause here for an observation. The Second Amendment “does not demand ‘an individualized hearing’ to assess Plaintiff's own personal level of risk.” Mai, 952 F.3d at 1119 (citing Tyler v. Hillsdale Cnty. Sheriff's Dep't, 837 F.3d 678, 698 n.18 (6th Cir. 2016) (en banc)). But still, one way that states can improve regulations' fit is by having exceptions or more individualized assessment. See, e.g., Singh, 979 F.3d at 725 (reasonable fit because statute “carves out exceptions”); Horsley v. Trame, 808 F.3d 1126, 1132 (7th Cir. 2015) (reasonable fit because “a person for whom a parent's signature is not available can appeal to the Director of the Illinois State Police”). There are only limited exceptions here, and no individualized assessment of any sort.
This result tracks our prior applications of intermediate scrutiny in the Second Amendment context. In Chovan, we found a reasonable fit with a law that banned convicted domestic criminals from having guns. 735 F.3d at 1139–42. In finding a reasonable fit there, we relied on the fact that “a high rate of domestic violence recidivism exists,” and cited studies “estimating a rate of domestic violence recidivism between 35% and 80%.” Id. What's more, the law only regulated convicted domestic criminals, not anyone else. Similarly, in Mai, we relied on statistics showing that persons who had been involuntarily confined were 39 times more likely to commit suicide, as well as figures showing that even years later, the risk remained much higher than normal. 952 F.3d at 1118. And again, the law only regulated people who had been involuntarily confined, not anyone else. Both laws are a far cry from the situation here: only three times increased risk and more than 400 times overregulation.
This result also fits with the deference that we owe to the California legislature. “[W]e must accord substantial deference to [the California legislature's] predictive judgments.” Mai, 952 F.3d at 1118 (citing Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 665, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) (Turner I)). And we have. We defer to California's assessment of the harm, to its statistics about young adults, and to its assessment that banning the sale of semiautomatic rifles would promote public safety. 26 But in Turner II, the Supreme Court did not defer when assessing the fit itself. See Turner Broad. Sys., Inc. v. FCC, 520 U.S. 180, 213, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997). We also defer to the legislature's judgment only on the effect of a law, and not on the law's fit. See, e.g., Mai, 952 F.3d at 1117–21.
Ultimately, in applying intermediate scrutiny, the district court had to do two things: identify the proper legal test for “reasonable fit,” and measure the semiautomatic rifle ban against that test. Properly identifying the legal standard is a question of law that we review de novo; applying it is a mixed question that we review for abuse of discretion. The district court used the wrong legal rule. Because the district court misapprehended the intermediate scrutiny test, it abused its discretion by getting the law wrong.
The district court also erred in its analysis of the irreparable harm preliminary injunction factor. “[T]he deprivation of constitutional rights unquestionably constitutes irreparable injury.” Melendres v. Arpaio, 695 F.3d 990, 1002 (9th Cir. 2012) (citing Elrod v. Burns, 427 U.S. 347, 373, 96 S.Ct. 2673, 49 L.Ed.2d 547 (1976)) (internal quotation marks omitted). The district court offered three rationales for why Plaintiffs would not be irreparably injured. They were all error.
First, the district court erred in holding that there was no irreparable harm because young adults could still obtain firearms under an exception. Jones, 498 F. Supp. 3d at 1331. As we discussed above, the exceptions do not alleviate the ban's severe burden on Second Amendment rights. They do not allow Plaintiffs to avoid irreparable harm for the same reasons. The main exception—for young adults with hunting licenses—only applies to the long gun regulation, not the semiautomatic rifle ban. So as to the semiautomatic rifle ban, this exception makes no difference. And as we discussed above, the exception for law enforcement officers and active-duty military members does not apply for most young adults. That leaves the family transfer and loan provisions. But the loan provisions are very limited, and it is not clear that young adults really can get firearms through family transfers, because gifts from family members are allowed but strawman purchases are not. See Cal. Penal Code §§ 27875 (family transfers), 27515 (strawman purchases).
Second, the district court observed that young adults could still get firearms by using them “at shooting ranges under certain circumstances.” Jones, 498 F. Supp. 3d at 1331. But using a firearm at a shooting range does not allow young adults to exercise their core Second Amendment right of self-defense in the home. See Heller, 554 U.S. at 630, 128 S.Ct. 2783. Young adults' ability to go to shooting ranges does not affect whether the harm here is irreparable.
And third, the district court relied on the fact that “[P]laintiffs may still access firearms ․ when they turn 21.” Jones, 498 F. Supp. 3d at 1331. But a constitutional violation is not reparable just because it is definite in duration: a harm need not last indefinitely to be irreparable. In other words, we would not tell a plaintiff suing over voting right restrictions on young adults that her harm was not irreparable because she could still vote when she turned 21.
Moreover, even putting aside these errors on irreparable harm, “the district court's likelihood-of-success determination [also] tainted its evaluation of the remaining three Winter elements.” Pom Wonderful LLC v. Hubbard, 775 F.3d 1118, 1133 (9th Cir. 2014). Our determination that Plaintiffs are likely to succeed on the merits “fundamentally changes the district court's calculus.” BOKF, NA v. Estes, 923 F.3d 558, 565 (9th Cir. 2019). Still, “a preliminary injunction is an extraordinary remedy never awarded as of right.” Benisek v. Lamone, ––– U.S. ––––, 138 S. Ct. 1942, 1943, 201 L.Ed.2d 398 (2018) (citing Winter, 555 U.S. at 24, 129 S.Ct. 365). And “the grant of a preliminary injunction is a matter committed to the discretion of the trial judge.” Epona v. County of Ventura, 876 F.3d 1214, 1227 (9th Cir. 2017) (citing Sierra On-Line, Inc. v. Phoenix Software, Inc., 739 F.2d 1415, 1421 (9th Cir. 1984)). This makes sense here because reconsidering the remaining factors may be “at least in part, fact-dependent.” BOKF, 923 F.3d at 565. Thus, despite Plaintiffs' “overwhelming likelihood of success on the merits, we remand this case to the district court to consider the remaining [three] Winter factors consistent with this opinion.” Goldman, Sachs & Co. v. City of Reno, 747 F.3d 733, 747 (9th Cir. 2014) (citing Evans v. Shoshone-Bannock Land Use Pol'y Comm'n, 736 F.3d 1298, 1307 (9th Cir. 2013)).
We note for the district court's reconsideration that “the government suffers no harm from an injunction that merely ends unconstitutional practices.” Kelly, 878 F.3d at 718 (cleaned up).
In conclusion, the district court erred by holding that the California laws did not burden Second Amendment rights. It properly applied intermediate scrutiny to the long gun regulation and did not abuse its discretion in finding it likely to survive. But it erred in applying intermediate scrutiny to the semiautomatic rifle ban. And even if intermediate scrutiny applied, the district court abused its discretion in finding the ban likely to survive. Finally, the district court erred in its application of the irreparable harm factor. Thus, as to the long gun regulation, the district court's order is AFFIRMED. And as to the semiautomatic centerfire rifle ban, the district court's order is REVERSED. We REMAND the case to the district court for further proceedings consistent with this opinion.
* 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. § 1, 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.
As explained in Judge Nelson's excellent opinion, California's law effectively banning the sale or transfer of semiautomatic firearms to young adults conflicts with the text, tradition, and history of the Second Amendment. I join the opinion in full but write separately to highlight how California's legal position has no logical stopping point and would ultimately erode fundamental rights enumerated in our Constitution. Simply put, we cannot jettison our constitutional rights, even if the goal behind a law is laudable.
California justifies its law by citing statistics showing that young adults constitute less than 5% of the population but represent more than 15% of homicide and manslaughter arrests. The state argues that intermediate scrutiny should apply and that it survives that test because the law is a “reasonable fit” for the state's important public safety goal. But even assuming intermediate scrutiny applies 1 , the state's assertion of a “reasonable fit” reduces that requirement to a malleable and meaningless limit on the government's power to restrict constitutional rights. As the majority opinion capably points out, only 0.25% of young adults commit violent crimes. So California limits the rights of 99.75% of young adults based on the bad acts of an incredibly small sliver of the young adult population. That is not a “reasonable fit.”
If we accept the state's argument, it redefines intermediate scrutiny as a rational basis review with a small sprinkle of skepticism in Second Amendment cases. And that would allow the government to trample over constitutional rights just by relying on anecdotal evidence and questionable statistics that loosely relate to a worthwhile government goal. If California can deny the Second Amendment right to young adults based on their group's disproportionate involvement in violent crimes, then the government can deny that right—as well as other rights—to other groups. For example, California arguably has a more compelling case if it enacts a similar gun-control law that targets males of all ages instead of young adults. Statistics—and science—show that men almost exclusively commit violent crimes. Take mass shootings for instance. Men have been involved in 99% of all mass shootings in America since 1966, according to a database maintained by the Violence Project.2 California can thus theoretically claim that if men cannot own firearms, it will eliminate 99% of mass shootings.
But as tempting as that solution may sound to some, such a law almost certainly would not pass constitutional muster. And the reason is obvious: its scope would not be remotely, let alone reasonably, tailored to the praiseworthy goal of curbing gun violence. 3 Cf. Bd. of Trustees of State Univ. of New York v. Fox, 492 U.S. 469, 480, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989) (requiring the “governmental goal to be substantial, and the cost to be carefully calculated” under intermediate scrutiny). While men constitute almost all mass shooters, 99.9999% 4 of men are not mass shooters. In other words, such a hypothetical law would strip all men of their Second Amendment rights based on the actions of 0.000001% of the male population.
The Supreme Court rejected such tenuous logic in Craig v. Boren when it struck down a state law banning the sale of some beer to young men, who overwhelmingly are much likelier than young women to drive under the influence and cause car accident deaths. 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976). Applying intermediate scrutiny for gender-based classifications, the Court acknowledged the statistical disparity but held that the state cannot use “a gender line as a classifying device.” Id. at 201, 97 S.Ct. 451. Even though that state law would have likely saved thousands of lives—almost certainly more so than California's law—the Court invalidated it because good intentions alone cannot salvage a law.
So, too, here. To accept the state's argument would mean allowing the government to restrict individuals' enumerated constitutional rights based solely on their group membership. Unlike other gun-control laws that target a person's specific and individual characteristics or actions (e.g., commission of felony, mental illness), California's law strips individuals of their fundamental constitutional rights based solely on what other people in their group may have committed in the past. That is antithetical to the very nature of individual rights and leads us down a dark path. Cf. Stanley v. Illinois, 405 U.S. 645, 656, 92 S.Ct. 1208, 31 L.Ed.2d 551 (1972) (the Bill of Rights protects the “citizenry from overbearing concern for efficiency and efficacy that may characterize praiseworthy governmental officials no less, and perhaps more, than mediocre ones.”).
We also do not typically limit constitutional rights based on the age of adults. Young adults have the same constitutional rights as the middle-aged or the elderly—even if some of them may not necessarily have the wisdom or judgment that age and experience can bring—for the same reason that we do not limit fundamental rights based on supposed intelligence, maturity, or other characteristics. We thus allow 18-year-olds to join the military and lay down their lives in defense of our freedoms. We even allow minors to take actions that their parents may strongly oppose: the Supreme Court has held that parents and the government must yield to the wishes of, say, a 14 or 15-year-old who wants an abortion. Bellotti v. Baird, 428 U.S. 132, 96 S.Ct. 2857, 49 L.Ed.2d 844 (1976).
None of this is to downplay the tragedy of gun violence. Although we must remain impartial as judges, we are citizens, too. And whenever we hear of gun violence, our stomachs sink and our hearts break for those who have lost families or friends in these terrible and tragic events. But only a tiny number of people abuse their rights and wield guns for unlawful violence. Such cold numbers admittedly offer little solace to those who have lost loved ones because of gun violence, but it does provide a perspective on whether we should restrict a constitutional right for the larger population based on a minuscule percentage of the populace who abuses that right.
Our Constitution provides a guarantee of our rights and freedoms. For the most part, people exercise their rights in responsible and productive ways. A tiny percentage, however, does not. But we should not sanction restricting a constitutional right by solely focusing on the few who abuse it.
As judges and lawyers, we revere the First Amendment as a core fundamental right. And rightfully so: It has allowed Americans to protest unjust wars abroad as well as racism and other injustices on our soil, changing this country for the better. But in our paeans to the First Amendment, we sometimes forget that the right also allows the people to do horrendous things. The First Amendment thus empowers Nazis to march down Main Street in the predominantly Jewish suburb of Skokie. See National Socialist Party v. Village of Skokie, 432 U.S. 43, 43–44, 97 S.Ct. 2205, 53 L.Ed.2d 96, (1977). It also allows amoral and perhaps immoral businesspeople to invoke the majesty of our Constitution to market despicable videogames to minors, even though they depict people being “dismembered, decapitated, disemboweled, set on fire, and chopped into little pieces,” and encourage players to engage in “ ‘ethnic cleansing’ [of] ․ African-Americans, Latinos, or Jews” and to “rape a mother and her daughters” in the videogames. Brown v. Entertainment Merchants Ass'n, 564 U.S. 786, 789–804, 131 S.Ct. 2729, 180 L.Ed.2d 708 (2011) (invalidating law restricting violent videogames).
But we do not impinge on the First Amendment based on the outlier actions of a few who may abuse that right. Nor should we with the Second Amendment. Cf. Jackson v. City and Cty. of San Francisco, 746 F.3d 953, 960 (9th Cir. 2014) (the Second Amendment “inquiry bears strong analogies to the Supreme Court's free-speech caselaw”); Ezell v. City of Chicago, 651 F.3d 684, 706–07 (7th Cir. 2011) (“Heller and McDonald suggest that First Amendment analogues are more appropriate, and ․ have already begun to adapt First Amendment doctrine to the Second Amendment context”).
In sum, we cannot allow good intentions to trump an enumerated and “fundamental right” deeply rooted in the history and tradition of this country. See McDonald, 561 U.S. at 778, 130 S.Ct. 3020.
While the majority was correct to apply intermediate scrutiny to the long gun regulation enumerated in Senate Bill 1100 and to affirm the district court's denial of the preliminary injunction, it erred in applying strict scrutiny to and reversing the district court with respect to Senate Bill 61's semiautomatic centerfire rifle 1 regulation. On that basis, I concur with the majority's holding and reasoning with respect to the long gun regulation and dissent from its holding and reasoning with respect to the semiautomatic rifle regulation. Accordingly, this dissent deals solely with the majority's treatment of the semiautomatic rifle regulation.
Although the question of “whether the challenged law burdens conduct protected by the Second Amendment” – the first step in the proper two-step framework – is debatable, the most significant flaw in the majority's analysis arises under the second step, i.e., the appropriate tier of scrutiny. See Fyock v. Sunnyvale, 779 F.3d 991, 996 (9th Cir. 2015). Neglecting consideration of either the disproportionate perpetration of violent crime by, or the relatively immature and variable cognitive development among, adults under age 21, the majority opinion fails to conduct a legal analysis that comports with the corpus of precedent within this Circuit and elsewhere. Not only in my view is it error for the majority to apply strict scrutiny to the semiautomatic rifle regulation, but its alternative holding that the regulation fails under intermediate scrutiny suffers from a faulty assessment of whether the regulation is a “reasonable fit” for California's public policy objectives. I shall attempt to elucidate these conclusions.
In 2018, California legislators amended California Penal Code Section 27510, which regulates the sale of firearms to persons aged 18 through 20. 2 The amendment, Senate Bill 1100 (“the long gun regulation”), introduced age limitations on the sale or transfer of long guns. 2017 California Senate Bill No. 1100, California 2017–2018 Regular Session. Specifically, the long gun regulation prohibits federally licensed firearms dealers (“FFLs”) from selling or transferring long guns to young adults. However, the long gun regulation contains exceptions, removing the regulation's applicability to young adults who have a hunting license; are peace officers, active federal officers and law enforcement officers; and are active or retired members of the military. Id.
Then, on April 27, 2019, a 19-year-old opened fire with a semiautomatic rifle, a subset of long gun, killing one and injuring three others at a synagogue in Poway, California. John Wilkens, Kristina Davis, and Teri Figueroa, One Dead, Three Injured in Poway Synagogue Shooting, San Diego Union-Trib. (April 27, 2019), https://www.sandiegouniontribune.com/news/public-safety/story/2019-04-27/reports-of-several-people-shot-at-poway-synagogue; Cheri Mossburg, Poway Synagogue Shooter Sentenced to Second Life Sentence, CNN (Dec. 28, 2021), https://www.cnn.com/2021/12/28/us/poway-synagogue-shooter-sentenced/index.html. In response, California's legislature passed Senate Bill 61 (“the semiautomatic rifle regulation”), which amended Section 27510 further to remove the hunting license exception for young adults to purchase semiautomatic centerfire rifles. Senate Bill No. 61, California 2019–2020 Regular Session.
Aside from the explicit exceptions contained in section 27510, California has preserved several avenues for young adults to possess and use long guns, including semiautomatic rifles. Contrary to plaintiffs' contentions, section 27510 does not regulate possession or use; rather, it merely regulates the purchase of firearms through FFLs. California emphasizes that, as long as young adults follow otherwise applicable California laws, they may use long guns, including semiautomatic rifles for self-defense in the home or elsewhere and for a number of other lawful purposes.
Indeed, the challenged regulations permit acquisition and loan of long guns, including semiautomatic rifles, in several ways. For instance, young adults may receive long guns from immediate family “by gift, bequest, intestate succession, or other means from one individual to another[.]” Cal. Penal Code §§ 16720, 27505, 27585. Young adults may also be loaned firearms, including handguns, from a wide range of people for varying periods of time, see Cal. Penal Code §§ 27880, 27885, or for the entirety of a hunting season if they are licensed hunters. Cal. Penal Code § 27950. California provides examples of other forms of acquisition that are untouched by the challenged regulations in its briefings. In sum, neither of the regulations we consider here are categorical bans on young adults' possession or acquisition of long guns, including semiautomatic rifles.
“[I]ndividual self-defense is ‘the central component’ of the Second Amendment right,” and the need for self-defense “ ‘is most acute’ in the home.” McDonald v. City of Chicago, 561 U.S. 742, 767, 130 S.Ct. 3020, 177 L.Ed.2d 894 (2010) (quoting District of Columbia v. Heller, 554 U.S. 570, 599, 128 S.Ct. 2783, 171 L.Ed.2d 637 (2008)). However, the Second Amendment does not grant the right to “carry any weapon whatsoever in any manner whatsoever and for whatever purpose.” Heller, 554 U.S. at 626, 128 S.Ct. 2783. In the same vein, it is clear that “the right secured by the Second Amendment is not unlimited.” Id. In Heller, the Supreme Court struck down Washington, D.C.'s complete ban on the possession or use of handguns, holding that a categorical prohibition of use in the home of the most-favored type of firearm, the handgun, was inconsistent with the Second Amendment. Id. at 635, 128 S.Ct. 2783. However, the Supreme Court made clear that “nothing ․ should be taken to cast doubt on” a variety of laws affecting the right to bear arms, including “laws imposing conditions and qualifications on the commercial sale of arms.” Id. at 626–27, 128 S.Ct. 2783. The Court also clarified that its list of “presumptively lawful regulatory measures ․ does not purport to be exhaustive.” Id. at 627 n. 26, 128 S.Ct. 2783.
To discern the finite limits on the Second Amendment right, this Circuit has developed a two-step framework. Young v. Hawaii, 992 F.3d 765, 783 (9th Cir. 2021). First, we must ask “whether the challenged law burdens conduct protected by the Second Amendment.” Fyock, 779 F.3d at 996. To answer this question, we must assess “historical understanding of the scope of the right.” Silvester v. Harris, 843 F.3d 816, 821 (9th Cir. 2016). If the restriction “can be traced to the founding era” or is the “subject of longstanding, accepted regulation,” it may be upheld without proceeding to the second step of the framework. Id.; Fyock, 779 F.3d at 997.
If this question is answered affirmatively, the analysis proceeds to the second step, which entails selecting the appropriate tier of scrutiny. To so determine, we must assess “how close the law comes to the core of the Second Amendment right,” and “the severity of the law's burden on that right.” Mai v. United States, 952 F.3d 1106, 1115 (9th Cir. 2020). If a law “implicates the core of the Second Amendment right and severely burdens that right,” we must apply strict scrutiny; otherwise, we apply intermediate scrutiny. Silvester, 843 F.3d at 821. The test for intermediate scrutiny is as follows: 1) the government must have a “significant, substantial, or important” objective, and 2) there must be “a reasonable fit between the challenged regulation and the asserted objective.” United States v. Chovan, 735 F.3d 1127, 1139 (9th Cir. 2013).
First, we must “determine whether the right [of young adults to purchase or receive transfer of semiautomatic rifles from FFLs] is protected by the Second Amendment.” Young, 992 F.3d at 784. As the majority recognizes, “California regulates young adults' commerce in firearms, not their possession.” Majority at 715. It is from this baseline that the majority conducts its historical analysis, considering “the history of young adults' right to keep and bear arms generally.” Id. And, from there, its review of the historical record produces the conclusion that “the Second Amendment protects young adults' right to keep and bear arms.” Id. at 720.
The majority contends that, by restricting young adults' commerce through the semiautomatic rifle regulation's prohibition on sales to young adults by FFLs, young adults cannot “obtain arms,” and thus, their “right to keep and bear arms [is] meaningless.” Majority at 716. However, by its terms, the semiautomatic rifle regulation is not a ban on young adults' ability to obtain semiautomatic rifles. This is true even though it prohibits FFLs from selling or transferring semiautomatic rifles to young adults. 3 As the district court reasoned, “[t]he only complete ban is for any FFL to sell, deliver, or supply a handgun to a [y]oung [a]dult.” Jones v. Becerra, 498 F. Supp. 3d 1317, 1328 (S.D. Cal. 2020). The semiautomatic rifle regulation allows for family gifts and a variety of other modes of possession through acquisition or loan; for example, the regulation permits a parent to purchase a semiautomatic rifle and transfer it to their child under age 21 through gift. See id. It also allows “individuals between the ages of 18 and 20 ․ [to] possess semi-automatic rifles if they are members of law enforcement, active duty members of the Armed Forces ․ or active reserve components of the United States.” Id. To be sure, the semiautomatic rifle regulation does indeed restrict the ability of young adults to purchase semiautomatic rifles from FFLs. But to classify it as a ban without qualification is a patent misreading of the statutory text.
Given the statute's several qualifications and exemptions, I disagree with the majority's conclusion that the semiautomatic rifle regulation constitutes a ban on commerce and conclude instead that the regulation is “consistent with a longstanding tradition of targeting select groups' ability to access and to use arms for the sake of public safety.” Nat'l Rifle Ass'n of Am., Inc. v. Bureau of Alcohol, Tobacco, Firearms, & Explosives, 700 F.3d 185, 203 (5th Cir. 2012) (“NRA”), cert. denied, 571 U.S. 1196, 134 S.Ct. 1364, 188 L.Ed.2d 296 (2014). California highlights the Fifth Circuit's review of founding-era attitudes in NRA, pointing out that the age of majority at common law was 21 years old. Id. at 201; see also Infancy, Black's Law Dictionary (10th ed. 2014) (“Every person is, at the common law, considered an infant, or minor, until he has reached the age of twenty-one years ․”) (quoting Lewis Hochheimer, A Treatise on Law Relating to the Custody of Infants 1 (2d ed. 1891)). Moreover, the historical record is replete with laws restricting the possession by and sale of firearms to minors. See Possession by, Use of, and Sales to Minors and Others Deemed Irresponsible, Repository of Historical Gun Laws, Duke Ctr. for Firearms Law (last viewed May 4, 2022), https://tinyurl.com/ycknv84y (describing 92 historical restrictions on minor firearm possession and commerce in 45 states).
I also take issue with the majority's repeated reference to the historic age of militia service—around 16 years—to support the notion that young adults have a Second Amendment right to bear arms. Majority at 717. Despite its insistence that “[t]he right is not conditioned on militia service,” id. at 721 (emphasis removed), the majority makes the following syllogism: “[t]he Second Amendment refers to the militia, and young adults had to be in the militia and bring their own firearms. This reference implies at least that young adults needed to have their own firearms.” Id. at 721. In drawing this conclusion, the majority makes the same mistake as plaintiffs in confusing “the age for military service with the separate question of the age at which society can draw a line at the sale of firearms to minors.” Regardless, the district court reminds us that even “[m]ilitias were well regulated by each state in the Founding Era.” Jones, 498 F. Supp. 3d at 1327. For example, “members of the militia were required to meet regularly for weapons inspection and registration, and members who did not show up with the required equipment could be fined.” Id. The lower court was correct to deduce that the regulations on militia duty “demonstrate that as far back as the Founding Era, firearm regulations were considered necessary and an individual's right to firearm possession came with obligations to ensure public safety.” Id. If young adults were historically members of the militia, then these regulations would have applied equally to them.
Nevertheless, much of this back and forth is of limited value. In reviewing historical sources, “we are likely to fall short in some way.” Young, 992 F.3d at 785. Indeed, “the courts of appeals have spilled considerable ink in trying to navigate” the historical parameters of the Second Amendment as set forth in Heller. Pena v. Lindley, 898 F.3d 969, 976 (9th Cir. 2018). One must acknowledge that historical review in line with an originalist understanding of constitutional rights tends to produce different interpretations and conclusions depending on the level of generality from which the analysis begins. See generally Peter J. Smith, Originalism and Level of Generality, 51 GA. L. REV. 485 (2017). Moreover, “the historical record [is] mixed” and as jurists we must be careful not to “pick[ ] [our] friends and come to a fore-ordained conclusion” on the scope of the Second Amendment. See Young, 992 F.3d at 822–23. Therefore, I will “follow [the] well-trodden and ‘judicious course’ ” and “assume without deciding” that the semiautomatic rifle regulation burdens conduct protected by the Second Amendment. See Pena, 898 F.3d at 976. I now turn to the majority opinion's flawed determination of the appropriate tier of scrutiny to be applied to the semiautomatic rifle regulation.
It bears repeating that “we are guided by a longstanding distinction between laws that regulate the manner in which individuals may exercise their Second Amendment right, and laws that amount to a total prohibition of that right.” Pena, 898 F.3d at 977. On one hand, “[s]trict scrutiny applies only to laws that both implicate a core Second Amendment right and place a substantial burden on that right.” Duncan v. Bonta, 19 F.4th 1087, 1103 (9th Cir. 2021) (quoting Mai, 952 F.3d at 1115). For instance, a “total prohibition” or a law that “bar[s] firearm possession completely” would likely merit strict scrutiny. Pena, 898 F.3d at 977; Silvester, 843 F.3d at 827. On the other hand, “[i]ntermediate scrutiny is appropriate if the regulation at issue does not implicate the core Second Amendment right or does not place a substantial burden on that right.” Fyock, 779 F.3d at 998–99. Indeed, there is “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.” Silvester, 843 F.3d at 823.
The first question is whether the challenged law impacts the “core” Second Amendment right to defend oneself in the home. See Heller, 554 U.S. at 629, 128 S.Ct. 2783; Fyock, 779 F.3d at 999. The majority urges that the semiautomatic rifle regulation does indeed implicate that “core” Second Amendment right. Neither plaintiffs nor the majority cite to any sources regarding the utility, practicality, or effectiveness of semiautomatic rifles for home self-defense. Nevertheless, regardless of whether the regulation does or does not implicate the “core of the Second Amendment,” strict scrutiny is inapposite because the regulation does not impose a “substantial burden.” See Fyock, 779 F.3d at 999.
The majority disagrees on this point, contending that strict scrutiny applies to the semiautomatic rifle regulation because it lacks the hunting license exception encompassed by the long gun regulation. Majority at 724. In the majority's view, the law here amounts to a near-total prohibition on young adults' ability to obtain semiautomatic rifles, and given that this Court has “never applied intermediate scrutiny to a ban like this,” strict scrutiny must apply. Majority at 725.
However, as previously limned, the semiautomatic rifle regulation is far afield from being a total or even near-total prohibition on the right to defend oneself in the home. See infra Part II.A. The district court and California have repeatedly highlighted the numerous exemptions for FFL transfers and non-FFL avenues still available to young adults to acquire firearms, including semiautomatic rifles. See Jones, 498 F. Supp. 3d at 1328. Furthermore, the regulation “leaves open alternative channels for self-defense in the home,” primarily by allowing young adults with hunting licenses to purchase other types of long gun from FFLS. Jackson v. City and County of San Francisco, 746 F.3d 953, 964 (9th Cir. 2014).
The majority counters that allowing intrafamily transfers while prohibiting purchases “makes young adults' Second Amendment rights conditional on the rights of others.” Majority at 726. While the regulation certainly places conditions on the rights of young adults to own semiautomatic centerfire rifles, laws that place conditions on or regulate the manner in which persons may possess firearms are commonly upheld. See, e.g., Silvester, 843 F.3d at 827–29; Pena, 898 F.3d at 978, 986. We have no reason to doubt that the vast majority of young adults aged 18 to 21 can secure semiautomatic centerfire rifles through intrafamily transfer. Indeed, the challenge here is a facial one, and the likely widespread availability of intrafamily transfers to California's young adults suggests that the burden here is not a “severe” one for the purposes of selecting a tier of scrutiny.
And, importantly, “unlike regulations that amount to functional prohibitions on the sale of arms, this [regulation] is of a temporary duration as to potential purchasers—it evaporates once the would-be purchaser turns 21.” Hirschfeld v. Bureau of Alcohol, Firearms, Tobacco & Explosives, 5 F.4th 407, 462 (4th Cir. 2021) (Wynn, J., dissenting), vacated as moot, 14 F.4th 322 (4th Cir. 2021). “The narrow ambit” of a regulation such as the one we are considering “militates against strict scrutiny.” NRA, 700 F.3d at 205.
Finally, plaintiffs' classification of the semiautomatic rifle regulation as a complete prohibition on a class of firearm also suffers from logical frailties. In Worman v. Healey, which also concerned a regulation restricting the sale and transfer of “semiautomatic assault weapons,” plaintiffs made the same argument. 922 F.3d 26, 32 n.2 (1st Cir. 2019). The First Circuit rejected this argument, reasoning thus: “the plaintiff's ‘absolute prohibition’ argument is circular: essentially, it amounts to a suggestion that whatever group of weapons a regulation prohibits may be deemed a ‘class.’ By this logic ․ virtually any regulation could be considered an ‘absolute prohibition’ of a class of weapons.” Id.
Although Heller rejected the notion “that it is permissible to ban the possession of handguns so long as the possession of other firearms (i.e., long guns) is allowed,” Heller, 554 U.S. at 629, 128 S.Ct. 2783, Worman points out that the significance of the handgun as a category of guns is critical because it is “the quintessential self-defense weapon” and “the most popular weapon chosen by Americans for self-defense in the home.” Id.; see Worman, 922 F.3d at 36. This rationale cannot extend to semiautomatic rifles, for they are not handguns and therefore neither the quintessential nor the most popular self-defense weapons; nor is the regulation here anywhere near as sweeping as the handgun ban in Heller. 4 The majority nevertheless contends that strict scrutiny is appropriate because, in its view, only three classes of gun are suitable for self-defense in the home: handguns, semiautomatic centerfire rifles, and shotguns. Majority at 725–26. Because the sale of handguns to young adults is banned by another law, Cal. Penal Code § 27505, 18 U.S.C. § 922(b)(1), and the law here places significant restrictions on the sale of semiautomatic centerfire rifles to young adults, the only suitable category of firearms for self-defense available to young adults is the shotgun. According to the majority, “this law takes away one of the two [other than the handgun] remaining practical options for self-defense in the home,” and that it is thus “a severe burden.” Majority at 725–26.
While I do not dispute that the semiautomatic rifle regulation places burdens on young adults who wish to purchase or otherwise receive semiautomatic centerfire rifles from FFLs, I do not find that it is a “severe burden” on young adults' Second Amendment rights. Even if the regulation means most young adults are “unable to purchase a subset of semiautomatic weapons” from FFLs, this “does not significantly burden the right to self-defense in the home.” Pena, 898 F.3d at 978. As the majority acknowledges, young adults still have access to reasonable alternatives for self-defense in the home, including the shotgun and other forms of long gun. Moreover, the time-limited nature of the regulation and the various avenues it leaves open to young adult possession of semiautomatic centerfire rifles mitigate its severity. Because the regulation does not remove young adults' ability to acquire other forms of long gun, and simultaneously maintains several methods for acquisition and use of semiautomatic rifles, the application of strict scrutiny is inappropriate. The more appropriate tier of scrutiny to be applied here is intermediate scrutiny.
Now, I shall assess whether the semiautomatic rifle regulation survives intermediate scrutiny by analyzing the majority's determination that the regulation does not. Then, I shall apply intermediate scrutiny in a manner that, I believe, comports more properly with this Circuit's precedents.
Intermediate scrutiny has two requirements: first, that “the government's stated objective ․ be significant, substantial, or important”; and second, “a reasonable fit between the challenged regulation and the asserted objective.” Chovan, 735 F.3d at 1139. California is “not required to show that [the regulation] is the least restrictive means of achieving its interest,” but rather, that the regulation “promotes a ‘substantial government interest that would be achieved less effectively absent regulation.’ ” Fyock, 779 F.3d at 1000 (quoting Colacurcio v. City of Kent, 163 F.3d 545, 553 (9th Cir. 1998)).
California's objective of promoting public safety and reducing gun violence is a significant, important one. I disagree with the majority's conclusion that the law is not a “reasonable fit” with this objective.
The majority refers to the origins of the Second Amendment intermediate scrutiny test within the First Amendment and chides the judges of this Circuit for neglecting “one-half of the inquiry.” Majority at 729. In the First Amendment context, intermediate scrutiny “does not mean that a time, place, or manner regulation may burden substantially more speech than is necessary to further the government's legitimate interests.” Ward v. Rock Against Racism, 491 U.S. 781, 789, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989). According to the majority, “[t]his is the essence of the intermediate scrutiny test: the regulation must ․ accomplish[ ] something ․ [and] not burden far more speech than is necessary.” Majority at 729. And, in my colleagues' view, “[w]hen we transplanted intermediate scrutiny from the First Amendment to the Second, ․ we did not bring the does-not-burden-conduct-more-than-necessary part.” Id. at 729. In so doing, the majority contends, “we neglect to consider fit at all.” Id. at 729.
Although Second Amendment intermediate scrutiny is no doubt drawn from the First Amendment context, the majority's analysis makes no distinction between the cluster of rights protected by the First Amendment and those protected by the Second. It goes without saying that the First Amendment, in protecting some of our most cherished civil and political rights, demands exacting scrutiny of government regulations placing limits on free speech. The right to bear arms, by contrast, is a core right whose direct exercise can lead to bodily injury or death. Although intermediate scrutiny in the Second Amendment context is no less exacting than that of the First, the distinction between the rights protected by the First Amendment and the Second is stark. It is not an “after-the-fact rationalization”, Majority at 729, to conclude that the reasonableness inquiries under the First and Second Amendments are distinctive due to the different rights each protects, and that Second Amendment reasonableness must more often account for public safety concerns. And even despite these substantive differences in the rights protected under the First and Second Amendments, speech “directed to inciting or producing imminent lawless action” is one of the very few categories of expression not protected by the First Amendment. See Brandenburg v. Ohio, 395 U.S. 444, 447, 89 S.Ct. 1827, 23 L.Ed.2d 430 (1969). Perhaps this Court's precedents were purposeful, rather than neglectful, in tailoring a “reasonable fit” analysis that allows more room for consideration of the heightened physical public safety considerations in the Second Amendment context. “No one really knows what the right answer is with respect to the regulation of firearms,” so courts should not “[d]isenfranchis[e] the American people on this life and death subject.” Kolbe v. Hogan, 849 F.3d 114, 150 (4th Cir. 2017) (Wilkinson, J., concurring).
The majority's conflation of First Amendment intermediate scrutiny with Second Amendment intermediate scrutiny is not the only instance in which the majority makes inapt comparisons with other forms of constitutional intermediate scrutiny. Like the two judges forming the majority in Hirschfeld's divided panel, the majority here draws a curious comparison between the semiautomatic rifle regulation and the challenged law in Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976). Relying on statistical comparisons, the majority concludes that the law banning the sale of low-alcohol beer to men considered in Craig failed to pass intermediate scrutiny even though it was “a much better fit” than the law we consider here. Majority at 730. The majority's argument is as follows: because only 2% of young men were arrested for drunk driving, a law regulating 100% of them was not a “reasonable fit” for the government objective of reducing traffic fatalities in Craig. Id. at 730. Compared with the law here, where “only 0.25% of young adults are arrested for violent crimes,” a law regulating 100% of young adults who don't qualify for any exemptions is also “unlikely to be a reasonable fit for California's objectives.” Id. at 731.
There are multiple problems with comparing the law in Craig with the law at issue here. First, the majority here attempts to “calcify the flexible world of intermediate scrutiny into a rigid statistical framework.” Hirschfeld, 5 F.4th at 479 (Wynn, J., dissenting) (internal quotation marks omitted). Our precedents counsel against such an approach: “[w]hen considering California's justifications for the statute, we do not impose an ‘unnecessarily rigid burden of proof,’ and we allow California to rely on any material ‘reasonably believed to be relevant’ to substantiate its interests in gun safety and crime prevention.” Pena, 898 F.3d at 979 (quoting Mahoney v. Sessions, 871 F.3d 873, 881 (9th Cir. 2017)). Just as we are not colonial historians, we are not statisticians, and California “must be allowed a reasonable opportunity to experiment with solutions to admittedly serious problems.” Id. at 980 (quoting City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 52, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986)).
A second major problem with the comparison to Craig is, again, inherent to the majority's conflation of Second Amendment intermediate scrutiny with Fourteenth Amendment intermediate scrutiny. Under the Fourteenth Amendment, classifications based on gender are subjected to heightened, intermediate scrutiny requiring an “exceedingly persuasive justification” because “our Nation has had a long and unfortunate history of sex discrimination.” United States v. Virginia, 518 U.S. 515, 531, 116 S.Ct. 2264, 135 L.Ed.2d 735 (1996) (quoting Frontiero v. Richardson, 411 U.S. 677, 684, 93 S.Ct. 1764, 36 L.Ed.2d 583 (1973)). In contrast, a state “may discriminate on the basis of age without offending the Fourteenth Amendment if the age classification in question is rationally related to a legitimate state interest.” Kimel v. Florida Bd. of Regents, 528 U.S. 62, 83, 120 S.Ct. 631, 145 L.Ed.2d 522 (2000). Indeed, it is well-established under the Fourteenth Amendment that “age classification is presumptively rational.” Id. at 84, 120 S.Ct. 631. Therefore, the majority's comparison of the gender classification in Craig with the age-restrictive semiautomatic rifle regulation here is the proverbial comparison of apples to oranges.
Beyond its inappropriate reliance on Craig, the majority claims that the semiautomatic rifle regulation is not a “reasonable fit” because it contains “only limited exceptions” and “no individualized assessment,” i.e. appeals or hearing process for young adults deemed ineligible to purchase semiautomatic rifles from FFLs. Majority at 731. Ultimately, these factors convince the majority that it need not defer to the California legislature's enactment of the regulation. This Court has made clear, however, that “[w]e ask only whether the evidence ‘fairly supports’ [the California legislature's] ‘reasonable’ conclusions.” Mai, 952 F.3d at 1118 (quoting Pena, 898 F.3d at 979–80). “When empirical evidence is incomplete, we ‘must accord substantial deference to [California's] predictive judgments.’ ” Id. (quoting Turner Broad. Sys. v. FCC, 512 U.S. 622, 665, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994)). 5 Although the majority faults the courts of this Circuit for failing to do a proper “reasonable fit” analysis in past decisions, it is clear that the majority is holding California to an evidentiary standard that is not required under Second Amendment intermediate scrutiny.
Having concluded that the majority misapplies intermediate scrutiny to the semiautomatic rifle regulation, I will now apply that test in a manner that, I believe, hews more closely to this Court's jurisprudence. Having already determined that California's objectives of enhancing public safety and reducing gun violence are important, the remaining question is whether the law is a “reasonable fit” for those objectives. In answering this question, I intend to focus on evidence California provides and that the majority makes little to no mention of at all. The facts surrounding young adults' disproportionate commission of violent gun crime; the increasing understanding of the relative immaturity of young adults; and the exemptions and numerous channels to possess and use semiautomatic rifles open to young adults are sufficient to demonstrate that the regulation is indeed a “reasonable fit.”
To start, California highlights the fact that young adults are disproportionately more likely to commit violent crimes in general and gun violence specifically than older adults. While 18 to 20-year-olds comprise less than 5% of the U.S. population, they account for more than 15% of reported homicide and manslaughter arrests. In California alone, 18 to 19-year-olds account for roughly 12% of the state's homicide arrests. Id. According to the gun violence prevention non-profit organization, Everytown for Gun Safety, 18 to 20-year-olds commit gun homicides at a rate three times higher than adults above the age of 21. Everytown Research & Policy, Everytown for Gun Safety, https://everytownresearch.org/stat/eighteen-to-20-year-olds-commit-gun-homicides-at-a-rate-triple-the-rate-of-those-21-and-years-older/ (2018). Additional studies show that at least one in eight victims of mass shootings from 1992 to 2018 were killed by an 18 to 20-year-old; additionally, assault rifles, including semiautomatic rifles, were responsible for approximately 45% of fatalities and 62% of overall victims of mass shootings. Joshua D. Brown and Amie J. Goodin, Mass Casualty Shooting Venues, Types of Firearms, and Age of Perpetrators in the United States, 1982–2018, 108 Am. J. Pub. Health 1385, 1386 (2018).
Indeed, California had heightened motivation to enact the semiautomatic rifle regulation after a 19-year-old committed a mass shooting at a synagogue in Poway, a suburb of San Diego. By the end of July 2019, there had been 32 shootings that year in California in which four or more people were injured or killed. California also provides evidence that the vast majority of firearms used in mass shootings are purchased legally from FFLs. The state legislature manifestly was entitled to have considered the disproportionate commission of violent gun crimes by young adults, the fact that most mass shooters purchase weapons legally, and the fact that semiautomatic weapons “have been the weapons of choice in many of the deadliest shootings in recent history,” as eminently reasonable bases to curtail the ability of young adults to purchase or receive transfer of semiautomatic rifles from FFLs. Worman, 922 F.3d at 39; see also Elzerie de Jager, Eric Goralnick, Justin C. McCarty, et al., Lethality of Civilian Active Shooter Incidents With and Without Semiautomatic Rifles in the United States, 320 J. Am. Med. Ass'n 1034 (2018) (“Semiautomatic rifles are designed for easy use, can accept large magazines, and fire high-velocity bullets, enabling active shooters to wound and kill more people per incident.”).
Beyond these significant safety concerns, contemporary scientific research increasingly sheds light on the relative immaturity and incomplete cognitive development of young adults. California cites to evidence that young adults are less mature than older adults, which leads them to take more risks and behave more reactively than their elders. Young adults are thus quicker to anger than older adults and more vulnerable to intense mood swings and to making instinctive, rather than considered, decisions. This cognitive immaturity makes young adults more likely to use firearms in situations of significant emotional arousal or perceived threat, or other situations that require rapid, complex information processing. Other Circuits have credited similar evidence to uphold regulations on firearms affecting 18 to 20-year-olds. NRA, 700 F.3d at 208; Horsley v. Trame, 808 F.3d. 1126, 1133 (7th Cir. 2015). The semiautomatic rifle regulation helps to “ensure that access to these weapons is restricted to mature individuals who have successfully completed safety training,” such as members of law enforcement and the military, “furthering the public safety objectives and ensuring that the Founding Era balancing of Second Amendment rights with safety concerns continues today.” Jones, 498 F. Supp. 3d at 1328.
The regulation also leaves open multiple avenues for young adults to possess and use firearms for self-defense. As discussed previously, see infra Part II.A, the semiautomatic regulation leaves open exemptions for young adults with military and law enforcement training. California highlights the fact that, while semiautomatic rifles were popular purchases among young adults in the past, young adults have also purchased or received transfer of numerous other types of long guns during the same period.
Ultimately, California provides substantial and substantiated justifications for its enactment of the semiautomatic rifle regulation. The evidence is sufficient to show that the regulation “promotes a ‘substantial government interest that would be achieved less effectively absent regulation.’ ” Fyock, 779 F.3d at 1000 (quoting Colacurcio, 163 F.3d at 553). I am quite comfortable concluding that the semiautomatic rifle regulation satisfies constitutional muster under intermediate scrutiny.
Finally, because I find that the semiautomatic rifle regulation withstands intermediate scrutiny, I disagree with the majority's conclusions that the district court erred in finding no irreparable harm to plaintiffs, and that the district court erred in finding that a preliminary injunction would not be in the public interest. Finding no issue with the district court's legal conclusions, I review the district court's factual findings for clear error. Washington v. United States Dep't of State, 996 F.3d 552, 560 (9th Cir. 2021).
First, the majority is simply incorrect to say that the district court conflated its analysis of the two regulations challenged here when it ruled that young adults could still obtain firearms under exceptions. See Jones, 498 F. Supp. 3d at 1331. The district court was, in fact, correct that both regulations contain the following exceptions: “[y]oung [a]dults are not banned from acquiring all firearms, but may qualify under an exception, or may receive transfers from parents, grandparents, and spouses.” Id. The district court's finding of fact here is based on a clear, obvious reading of the statute before us. There is no clear error here, and I would uphold the denial of the motion for a preliminary injunction for lack of irreparable harm.
Second, the majority errs in saying that the district court conflated its analysis of the two regulations in finding that the balance of interests weighed against an injunction. The district court was, again, correct in finding that both regulations serve to “advance public safety by limiting the possession and use of deadly weapons to mature individuals who have demonstrated discipline through proper training to ensure public safety while honoring the Second Amendment rights of these individuals.” Id. at 1332. The majority contends that the “training” aspect only applies to the long gun regulation because it contains an exception for young adults with hunting licenses. However, by restricting the purchase of semiautomatic rifles to persons with military and law enforcement training, the semiautomatic rifle regulation creates an exception for persons with a higher level of safety education and training. The latter regulation's more restrictive ambit accords with the heightened dangerousness of semiautomatic rifles as compared with other types of long gun. Therefore, the district court made no conflation, and it was not clear error for it to determine that “[t]he potential harm of enjoining a duly-enacted law designed to protect public safety outweighs [y]oung [a]dults' inability to secure the firearm of their choice without proper training.” Id.
Accordingly, I would not disturb the district court's decision to deny plaintiffs' motion for a preliminary injunction, and would affirm that court's determination.
In sum, the district court was correct to hold that both the long gun regulation and the semiautomatic rifle regulation do not impermissibly burden Second Amendment rights. It correctly applied intermediate scrutiny to both laws and properly denied plaintiffs' request for a preliminary injunction. Therefore, I would affirm the district court's ruling in full and respectfully, but decidedly, dissent from the majority's holding with respect to the semiautomatic rifle regulation.
1. We use “young adults” to refer to people who are 18 years old or older but not yet 21 years old.
2. Loans and intrafamily transfers are severely restrictive. First, intrafamily transfers are only allowed from parents or grandparents. Cal. Penal Code §§ 27875, 16720. But strawman purchases are not allowed. Id. § 27515. Then limited loans are allowed: (1) loans for up to thirty days from a larger set of family members (spouses, domestic partners, parents, children, siblings, or grandparents), id. § 27880; (2) loans for up to three days, if the firearm is used in the presence of the loaner, id. § 27885; (3) loans for any amount of time, if the firearm stays only at the loaner's residence, id. § 27881; and (4) loans to licensed hunters for the hunting season, id. § 27950.
3. Transfers of handguns to young adults are also banned, except for antique handguns and intrafamily transfers. Cal. Penal Code § 27505(a).
4. A rifle is a kind of long gun. A semiautomatic rifle fires a single bullet each time the trigger is pulled and does not require the user to manually cycle between shots. And a centerfire rifle uses centerfire ammunition, in which the primer that ignites the powder is in the center of the bullet, rather than at the rim. Most rifles are centerfire rifles, and thus for ease of reference, we refer just to semiautomatic rifles.
5. The district court relied on National Rifle Association of America, Inc. v. Bureau of Alcohol and Tobacco, Firearms, and Explosives, 700 F.3d 185, 196 (5th Cir. 2012) (“NRA I”); Hirschfeld v. Bureau of Alcohol, Tobacco, Firearms and Explosives, 417 F. Supp. 3d 747, 755 (W.D. Va. 2019); and Mitchell v. Atkins, 483 F. Supp. 3d 985 (W.D. Wash. 2020). The district court's order in Hirschfeld was reversed by a divided panel, 5 F.4th 407 (4th Cir. 2021), but then the plaintiffs turned 18, the court found the case moot, and vacated the opinion, 14 F.4th 322 (4th Cir. 2021).
6. As part of this historical analysis, one tool to consider is corpus linguistics. Corpus linguistics is “an analysis of how particular combinations of words are used in a vast database of English prose.” Facebook, Inc. v. Duguid, ––– U.S. ––––, 141 S. Ct. 1163, 1174, 209 L.Ed.2d 272 (2021) (Alito, J., concurring) (citing Lee & Mouritsen, Judging Ordinary Meaning, 127 Yale L. J. 788 (2018)). It “draws on the common knowledge of the lay person by showing us the ordinary uses of words in our common language.” Wilson v. Safelite Grp., Inc., 930 F.3d 429, 439 (6th Cir. 2019) (Thapar, J., concurring in part and concurring in the judgment). Corpus linguistics “is a powerful tool for discerning how the public would have understood a statute's text at the time it was enacted,” and “[c]ourts should consider adding this tool to their belts.” Id. at 439–40.We asked the parties to file supplemental briefing addressing in part the applicability of corpus linguistics to this case. We thank the parties for their hard work. Because neither of them asks us to apply corpus linguistics here, we decline to consider it further.
7. Because the Second Amendment is not a “second-class right,” McDonald, 561 U.S. at 780, 130 S.Ct. 3020, we must treat it the same as other rights. In the context of a right to privacy, “a total prohibition against sale of contraceptives ․ would intrude upon [the right to privacy] as harshly as a direct ban.” Carey v. Population Servs., Int'l, 431 U.S. 678, 688, 97 S.Ct. 2010, 52 L.Ed.2d 675 (1977). “Indeed, in practice, a prohibition against all sales, since more easily and less offensively enforced, might have an even more devastating effect upon” the exercise of constitutional rights. Id.
8. See David B. Kopel, The Posse Comitatus and the Office of Sheriff: Armed Citizens Summoned to the Aid of Law Enforcement, 104 J. Crim. L. & Criminology 761, 788 (2014).
9. S.T. Ansell, Legal and Historical Aspects of the Militia, 26 Yale L.J. 471, 473 (1917).
10. Kopel, supra n.8, at 771–72.
11. See David B. Kopel & Joseph G.S. Greenlee, The Second Amendment Rights of Young Adults, 43 S. Ill. U. L. J. 495, 535 (2019).
12. Kopel, supra n.6, at 788, 790.
13. Id. at 792.
14. See Joyce Lee Malcolm, To Keep and Bear Arms 139 (1994).
15. See Clayton E. Cramer, Colonial Firearm Regulation, 16 J. Firearms & Pub. Pol'y 3, 24 (2004).
16. See Gautham Rao, The Federal Posse Comitatus Doctrine: Slavery, Compulsion, and Statecraft in Mid-Nineteenth-Century America, 26 Law & Hist. Rev. 1, 10 (2008) (internal citation omitted).
17. See id. at 2.
18. See id. at 10.
19. See id.
20. Of course, the Fifth Circuit panel's decision in NRA I is the law in that circuit, not Judge Jones's dissent. But as Judge Jones pointed out, the panel “d[id] not do justice to Heller's tailored approach toward historical sources.” NRA II, 714 F.3d at 336 (dissenting from denial of rehearing en banc). A bare majority of judges voted against rehearing NRA I en banc, and they did not respond to Judge Jones's dissent. Much of Judge Jones's historical analysis remains unrefuted.Also, dissents from denial from rehearing en banc, such as the one written by Judge Jones, can be persuasive judicial guideposts that “address[ ] arguments raised for the first time during the en banc process, correct[ ] misrepresentations, or highlight[ ] important facets of the case that had yet to be discussed.” Martin v. City of Boise, 920 F.3d 584, 588 (9th Cir. 2019) (Berzon, J., concurring in the denial of rehearing en banc).
21. New York passed two laws prohibiting the sale of firearms to minors. N.Y. Penal Code ch. 375 § 1 (1883); id. § 409 (1885). But its state constitution had no Second Amendment analog. (There is a Second Amendment analog in N.Y. Civ. Rights Law § 4, but it was not passed until the 20th century.) Delaware also banned the sale of firearms to minors, 16 Del. Laws 716 (1881), but it did not ratify its Second Amendment analog until 1987, Del. Const., Art. 1, § 20.
22. We find it telling that even though they came to different ultimate conclusions, the Fifth Circuit and the Hirschfeld panel agreed on this point.
23. Many young adults cannot even qualify for the “active peace officer” exception, because although individuals may enroll in the police academy at age 18, numerous political subdivisions require police officers to be 21. See, e.g., General Information and Qualifications, San Francisco Police Department, https://www.sanfranciscopolice.org/your-sfpd/careers/sworn-job-openings/general-information-and-qualifications (last visited Nov. 18, 2021).
24. Defendants argue that we may not consider Plaintiffs' facts about these categories of guns, because they were not submitted below. But these facts are legislative facts, “which have relevance to legal reasoning and the lawmaking process,” rather than adjudicative facts, which “are simply the facts of the particular case.” Advisory Comm. Note, Fed. R. Evid. 201. We have previously considered this kind of fact in a Second Amendment challenge, even over a defendant's challenge that it was not in the record below. See Chovan, 735 F.3d at 1140–41 (considering social science studies). In any case, Defendants did not contest Plaintiffs' evidence about rimfire semiautomatic rifles, and we agree with Defendants that semiautomatic shotguns likely are effective self-defense weapons.
25. The law actually sweeps even more broadly that, because the 400 times over regulation figure does not account for repeat offenders. And the denominator is also inflated because it includes all violent crimes, not just homicide and manslaughter.
26. Here, the dissent argues that we do not adequately discuss the evidence that the California legislature considered. Dissent at 758–60. But we defer to the legislature's predictive judgments, and we agree that its ban would promote public safety. Because we accept the legislature's conclusion on this point, we need not discuss its evidence further.
1. We should apply strict scrutiny under our court's two-step inquiry test, as explained in Judge Nelson's opinion. But see Rogers v. Grewal, ––– U.S. ––––, 140 S. Ct. 1865,1867, 207 L.Ed.2d 1059 (2020) (Thomas, J., dissenting from denial of certiorari) (stating that our two-step test “appears to be entirely made up” and that “its application has yielded analyses that are entirely inconsistent with Heller”).
2. The Violence Project Database, https://www.theviolenceproject.org/mass-shooter-database/ (last visited December 15, 2021). Of the 172 mass shootings since 1966, only four of them involved women. But of the four, two of them were assisting their male counterparts in the mass shooting. The organization used Congressional Research Service's definition of a mass shooting, i.e., “a multiple homicide incident in which four or more victims are murdered with firearms ․ within one event, and at least some of the murders occurred in a public location or locations in close geographical proximity ․ and the murders are not attributable to any other underlying criminal activity or commonplace circumstance.”
3. California has argued that intermediate scrutiny should apply to its gun-control laws, which is the same standard used for gender-based classifications. See, e.g., Craig v. Boren, 429 U.S. 190, 97 S.Ct. 451, 50 L.Ed.2d 397 (1976) (applying traditional intermediate scrutiny test). During oral argument, counsel for the state was non-committal on whether such a hypothetical law would survive intermediate scrutiny. If United States v. Virginia established a more heightened version of intermediate scrutiny, 518 U.S. 515, 116 S.Ct. 2264, 135 L.Ed.2d 735 (1996) (applying an “exceedingly persuasive” burden requirement), then—at the very least—that version should apply to an enumerated constitutional right “deeply rooted in this Nation's history and tradition.” McDonald v. City of Chicago, 561 U.S. 742, 778, 130 S.Ct. 3020, 177 L.Ed.2d 894 (2010).
4. Since 1966, there have been 170 mass shootings involving men. According to the U.S. Census estimate, there were 163,073,046 males as of 2020. See https://www.census.gov/quickfacts/fact/table/US/PST045219#PST045219 (last visited Dec. 15, 2021). That means that only 0.0001% of the male population committed a mass shooting. And even that miniscule percentage is still inflated because it assumes a static denominator based on the male population as of 2020 instead of all males alive since 1966.
1. I generally refer to semiautomatic centerfire rifles as “semiautomatic rifles” for ease of reference, as do my colleagues in the majority.
2. I refer to individuals who are 18 or older but not yet 21 years old as “young adults” for ease of reference, as do my colleagues in the majority.
3. The majority insists that the Second Amendment right includes the ability to purchase firearms. In doing so, it cites to only two precedents, this Court's decisions in Bauer v. Becerra, 858 F.3d 1216 (9th Cir. 2017), and Teixeira v. County of Alameda, 873 F.3d 670 (9th Cir. 2017). To be clear, neither of these precedents stands for the conclusion that the right to bear arms includes the right to purchase them. In Bauer, in response to the plaintiff's argument that the core Second Amendment right included a right to purchase firearms, this Court set aside the argument by stating, “even if we assume that the right to possess a firearm includes the right to purchase one, the burden on that right is exceedingly minimal here.” 858 F.3d at 1222. Moreover, the semiautomatic rifle regulation we consider here is a prohibition on the sale by FFLs to young adults. This Court in Teixeira held that “the Second Amendment does not independently protect a proprietor's right to sell firearms.” 873 F.3d at 690. I do not contest that the prohibition on FFLs selling semiautomatic rifles to young adults is directly tied to young adults' ability to purchase semiautomatic rifles. However, while the Second Amendment right surely protects the right to possess and use firearms, the majority's inferential leap to the assumption that it protects the right to purchase firearms goes too far.
4. The particular difference in scope between the District of Columbia's categorical firearm ban in Heller and the semiautomatic rifle regulation here is worth subjecting to explicit comparison. D.C. made it illegal to carry an unregistered handgun and simultaneously prohibited the registration of handguns; it also required residents who owned handguns lawfully to keep them unloaded and disassembled or secured by a trigger lock in the home, essentially rendering them inoperable in the event of an unforeseen emergency. See Heller, 554 U.S. at 574–76, 128 S.Ct. 2783. Here, by contrast, the semiautomatic rifle regulation prohibits FFLs from selling or transferring semiautomatic rifles to young adults who do not fall within one of its exceptions for law enforcement or military members. The law in Heller “totally ban[ned] handgun possession in the home.” Id. at 628, 128 S.Ct. 2783. The challenged law here does no such thing. As the Fifth Circuit noted, “unlike the D.C. ban in Heller, this ban does not disarm an entire community, but instead prohibits commercial [semiautomatic rifle] sales to 18-to-20-year-olds—a discrete category.” NRA, 700 F.3d at 205.
5. The majority highlights the fact that in Turner, the Supreme Court did not defer to the Government's arguments that its statutes regulating aspects of broadcast television were a “reasonable fit” for its objectives under First Amendment intermediate scrutiny. Majority at 731–32; see Turner, 512 U.S. at 666–68, 114 S.Ct. 2445. Like many of the majority's comparisons with Ninth Circuit precedent and Supreme Court caselaw, the comparison with Turner is faulty. Turner concerned the district court's grant of summary judgment in favor of the Government, and the Supreme Court remanded because genuine issues of material fact remained. Id. at 668, 114 S.Ct. 2445. The Supreme Court pointed out that the record was devoid of a great deal of evidence that could establish that no genuine issue of material fact existed, and therefore, remand was appropriate. Here, by contrast, California points to numerous statistics to substantiate its important objective specifically surrounding the fact that young adults disproportionately commit violent crime and are still developmentally maturing. Unlike in Turner, where the Supreme Court highlighted the lack of evidence supporting Congress's enactments, the majority gives here short shrift to the ample evidence California provides in support of its objectives.
R. NELSON, Circuit Judge:
Opinion by Judge R. Nelson; Concurrence by Judge Lee; Dissent by Judge Stein