FOR PUBLICATION

UNITED STATES COURT OF APPEALS

FOR THE NINTH CIRCUIT

VIRGINIA DUNCAN; PATRICK No. 23-55805

LOVETTE; DAVID MARGUGLIO;

CHRISTOPHER WADDELL; D.C. No. CALIFORNIA RIFLE & PISTOL 3:17-cv-01017- ASSOCIATION, INC., a California BEN-JLB corporation,

Plaintiffs-Appellees, OPINION

v.

ROB BONTA, in his official capacity as Attorney General of the State of California,

Defendant-Appellant.

Appeal from the United States District Court for the Southern District of California Roger T. Benitez, District Judge, Presiding

Argued and Submitted En Banc March 19, 2024 San Francisco, California

Filed March 20, 2025 Before: Mary H. Murguia, Chief Judge, and Sidney R. Thomas, Susan P. Graber, Kim McLane Wardlaw, Richard A. Paez, Marsha S. Berzon, Sandra S. Ikuta, Andrew D. Hurwitz, Ryan D. Nelson, Patrick J. Bumatay and Lawrence VanDyke, Circuit Judges.

Opinion by Judge Graber; Concurrence by Judge Berzon; Dissent by Judge R. Nelson; Dissent by Judge Bumatay; Dissent by Judge VanDyke;

SUMMARY *

En Banc Procedures / Second Amendment

The en banc court held that a California law banning the possession of large-capacity magazines comports with the Second Amendment, reversed the district court’s contrary conclusion, and remanded with the instruction to enter judgment in favor of the Attorney General of the State of California. In 2016, the California legislature enacted Senate Bill 1446, which barred the possession of large-capacity magazines as of July 1, 2017, and imposed a fine for failing to comply. Later in 2016, California voters approved Proposition 63, which subsumed Senate Bill 1446 and added a provision that imposed a possible criminal penalty for

* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader.

DUNCAN V. BONTA 3

unlawful possession of large-capacity magazines after July 1, 2017. Cal. Penal Code § 32310(c). The en banc court affirmed its earlier rejection of Plaintiffs’ Fifth Amendment takings claim in Duncan v. Bonta (“Duncan V”), 19 F.4th 1087, 1111-13 (9th Cir. 2021) (en banc). Employing the methodology announced in New York State Rifle & Pistol Ass’n v. Bruen, 597 U.S. 1 (2022), the en banc court concluded that California’s law comported with the Second Amendment for two independent reasons. First, the text of the Second Amendment does not encompass the right to possess large-capacity magazines because large-capacity magazines are neither “arms” nor protected accessories. Second, even assuming that the text of the Second Amendment encompasses the possession of optional accessories like large-capacity magazines, California’s ban on large-capacity magazines falls within the Nation’s tradition of protecting innocent persons by prohibiting especially dangerous uses of weapons and by regulating components necessary to the firing of a firearm. Accordingly, the en banc court remanded this case to the district court with the instruction to enter judgment in favor of the California Attorney General. Concurring in full in the majority opinion, Judge Berzon, joined by Chief Judge Murguia, and Judges Hurwitz, Paez, S.R. Thomas, and Wardlaw, wrote separately to address Judge VanDyke’s dissent, which includes a link to a video that he recorded showing him handling several different handguns and explaining his understanding of their mechanics and operation. Judge Berzon pointed out two problems with Judge VanDyke’s reliance on the video: (1) The video is not part of his written dissent and it includes facts outside the record. (2) Judge VanDyke has in essence appointed himself as an expert witness in this case, providing a factual presentation with the express aim of convincing the readers of his view of the facts without complying with any of the procedural safeguards that usually apply to experts and their testimony, while simultaneously serving on the panel deciding the case. Dissenting, Judge R. Nelson agreed with Judge Bumatay’s dissent that the majority’s decision to reverse the district court on the merits flouted Bruen. The majority’s decision also spurned the statutory procedure for en banc proceedings because, as explained in his dissent from the order filed concurrently with this opinion, this en banc court lacked statutory jurisdiction to decide this new appeal. Dissenting, Judge Bumatay, joined by Judges Ikuta, R. Nelson, and VanDyke, would hold that nothing in the text, or the country’s historical understanding of the Second Amendment, warrants California’s magazine ban. Applying the Bruen framework, California’s magazine ban is presumptively unconstitutional because the plain text of the Second Amendment protects the possession of magazines capable of feeding more than ten rounds. California failed to overcome the presumption of unconstitutionality where California failed to identify a historical analogue that was relevantly similar to a ban on these magazines. Accordingly, he would hold that California’s magazine ban is unconstitutional. Dissenting, Judge VanDyke agreed with Judge Bumatay’s dissent, which demonstrated the correct approach under the Second Amendment following Bruen, but wrote separately to further highlight serious flaws in the majority’s analysis. First, in determining that large-capacity magazines

DUNCAN V. BONTA 5

are not part of the arms covered by the Second Amendment, the majority erred by taking Bruen’s guidance to mean there was an extensive first-step, arm-or-not inquiry. Whether a firearm component is an inherent and “necessary” part of the arm itself, or instead merely an “optional” and unnecessary accessory to the arm, is a hopelessly indeterminable and inadministrable distinction. Judge VanDyke included a video in his dissent to illustrate the conceptual point, made in the written portion of his dissent, that an “arm” is a broad term covering an almost limitless variety of configurations within that category. The Second Amendment cannot apply only to firearms containing just those parts that a state like California deems essential and necessary. Second, Judge VanDyke agreed with Judge Bumatay that the majority erred as a historical matter in assessing whether a ban on large- capacity magazines is consistent with history and tradition. Finally, addressing Judge Berzon’s concurrence criticizing the video portion of his dissent, Judge VanDyke responded that his dissent is clearly evidenced by a written disposition; and his criticism of the majority’s reliance on the arms-accessory definition to decide this case is fundamentally a conceptual one, not a factual one. Erin E. Murphy (argued), Paul D. Clement, and Matthew D. Rowen, Clement & Murphy PLLC, Alexandria, Virginia; Anna M. Barvir, Sean A. Brady, and C.D. Michel, Michel & Associates PC, Long Beach, California; for Plaintiffs- Appellees. Michael J. Mongan (argued), Solicitor General, California Department of Justice, San Francisco, California; Robert L. Meyerhoff, Kevin J. Kelly, John D. Echeverria, and Robert L. Meyerhoff, Deputy Attorneys General; R. Matthew Wise, Supervising Deputy Attorney General; Mica L. Moore, Deputy Solicitor General; Thomas S. Patterson, Senior Assistant Attorney General; Helen H. Hong, Principal Deputy Solicitor General; Rob Bonta, California Attorney General; California Department of Justice, Los Angeles, California; for Defendant-Appellant. Amanda Hainsworth, Assistant Attorney General; Elizabeth N. Dewar, State Solicitor; Andrea J. Campbell, Massachusetts Attorney General; Office of the Massachusetts Attorney General, Boston, Massachusetts; Angela Cai, Deputy Solicitor General; Jeremy M. Feigenbaum, Solicitor General; Matthew J. Platkin, New Jersey Attorney General; Office of the New Jersey Attorney General, Trenton, New Jersey; Kristin K. Mayes, Arizona Attorney General, Office of the Arizona Attorney General, Phoenix, Arizona; Philip J. Weiser, Colorado Attorney General, Office of the California Attorney General, Denver, Colorado; William Tong, Connecticut Attorney General, Office of the Connecticut Attorney General, Hartford, Connecticut; Kathleen Jennings, Delaware Attorney General, Office of the Delaware Attorney General, Wilmington, Delaware; Brian L. Schwalb, District of the

DUNCAN V. BONTA 7

Columbia Attorney General, Office of the District of Columbia Attorney General, Washington, D.C.; Anne E. Lopez, Hawai'i Attorney General, Office of the Hawai'i Attorney General, Honolulu, Hawai'i; Kwame Raoul, Illinois Attorney General, Office of the Illinois Attorney General, Chicago, Illinois; Aaron M. Frey, Maine Attorney General, Office of the Maine Attorney General, Augusta, Maine; Anthony G. Brown, Maryland Attorney General, Office of the Maryland Attorney General, Baltimore, Maryland; Dana Nessel, Michigan Attorney General, Office of the Michigan Attorney General, Lansing, Michigan; Keith Ellison, Minnesota Attorney General, Office of the Minnesota Attorney General, St. Paul, Minnesota; Letitia James, New York Attorney General, Office of the New York Attorney General, New York, New York; Ellen F. Rosenblum, Oregon Attorney General, Office of the Oregon Attorney General, Salem, Oregon; Michelle A. Henry, Pennsylvania Attorney General, Office of the Pennsylvania Attorney General, Harrisburg, Pennsylvania; Peter F. Neronha, Rhode Island Attorney General, Office of the Rhode Island Attorney General, Providence, Rhode Island; Charity R. Clark, Vermont Attorney General, Office of the Vermont Attorney General, Montpelier, Vermont; Robert W. Ferguson, Washington Attorney General, Office of the Washington Attorney General, Olympia, Washington; Joshua L. Kaul, Wisconsin Attorney General, Office of the Wisconsin Attorney General, Madison, Wisconsin; for Amici Curiae Massachusetts, New Jersey, Arizona, Colorado, Connecticut, Delaware, The District of Columbia, Hawai'i, Illinois, Maine, Maryland, Michigan, Minnesota, New York, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and Wisconsin. Timothy C. Hester, Daniel Weltz, and Priya Leeds, Covington & Burling LLP, Washington, D.C.; Douglas N. Letter and Shira L. Feldman, Brady Center to Prevent Gun Violence, Washington, D.C.; Esther Sanchez-Gomez, Giffords Law Center to Prevent Gun Violence, San Francisco, California; Ciara W. Malone, March for Our Lives, New York, New York; for Amici Curiae Brady Center to Prevent Gun Violence, Giffords Law Center to Prevent Gun Violence, and March for Our Lives. Priyanka G. Sen, William J. Taylor Jr., and Janet Carter, Everytown Law, New York, New York; Freya Jamison, Everytown Law, Washington, D.C.; for Amicus Curiae Everytown for Gun Safety. Joseph G.S. Greenlee and Cody J. Wisniewski, FPC Action Foundation, Las Vegas, Nevada, for Amici Curiae FPC Action Foundation, California Gun Rights Foundation, and Center for Human Liberty. Noel J. Francisco, Anthony J. Dick, and Harry S. Graver, Jones Day, Washington, D.C.; Andrew E. Lelling, Jones Day, Boston, Massachusetts; Matthew R. Modderman, Jones Day, Cleveland, Ohio; Lawrence G. Keane, The National Shooting Sports Foundation Inc., Washington, D.C.; for Amicus Curiae The National Shooting Sports Foundation Inc.. David H. Thompson and Peter A. Patterson, Cooper & Kirk PLLC, Washington, D.C., for Amicus Curiae Firearm Policy Coalition Inc.. Edward A. Paltzik, Meredith Lloyd, and Serge Krimnus, Bochner PLLC, New York, New York, for Amicus Curiae The Second Amendment Foundation.

DUNCAN V. BONTA 9

Peter M. Torstensen Jr., Deputy Solicitor General; Christian B. Corrigan, Solicitor General; Austin Knudsen, Montana Attorney General; Office of the Montana Attorney General, Helena, Montana; Joshua N. Turner, Acting Solicitor General; Raul R. Labrador, Idaho Attorney General; Steve Marshall, Alabama Attorney General; Office of the Alabama Attorney General, Montgomery, Alabama; Treg Taylor, Alaska Attorney General, Office of the Alaska Attorney General, Anchorage, Alaska; Tim Griffin, Arkansas Attorney General, Office of the Arkansas Attorney General, Little Rock, Arkansas; Ashley Moody, Florida Attorney General, Office of the Florida Attorney General, Tallahassee, Florida; Christopher M. Carr, Georgia Attorney General, Office of the Georgia Attorney General, Atlanta, Georgia; Theodore E. Rokita, Indiana Attorney General, Office of the Indiana Attorney General, Indianapolis, Indiana; Brenna Bird, Iowa Attorney General, Office of the Iowa Attorney General, Des Moines, Iowa; Kris Kobach, Kansas Attorney General, Office of the Kansas Attorney General, Topeka, Kansas; Daniel Cameron, Kentucky Attorney General, Office of the Kentucky Attorney General, Frankfort, Kentucky; Jeff Landry, Louisiana Attorney General, Office of the Louisiana Attorney General, Baton Rouge, Louisiana; Lynn Fitch, Mississippi Attorney General, Office of the Mississippi Attorney General, Jackson, Mississippi; Andrew Bailey, Missouri Attorney General, Office of the Missouri Attorney General, Kansas City, Missouri; Michael T. Hilgers, Nebraska Attorney General, Office of the Nebraska Attorney General, Lincoln, Nebraska; John M. Formella, New Hampshire Attorney General, Office of the New Hampshire Attorney General, Concord, New Hampshire; Drew H. Wrigley, North Dakota Attorney General, Office of the North Dakota Attorney General, Bismark, North Dakota; Dave Yost, Ohio Attorney General, Office of the Ohio Attorney General; Columbus, Ohio; Gentner F. Drummond, Oklahoma Attorney General, Office of the Oklahoma Attorney General, Oklahoma City, Oklahoma; Alan Wilson, South Carolina Attorney General, Office of the South Carolina Attorney General, Columbia, South Carolina; Marty J. Jackley, South Dakota Attorney General, Office of the South Dakota Attorney General, Pierre, South Dakota; Ken Paxton, Texas Attorney General, Office of the Texas Attorney General, Austin, Texas; Jason Miyares, Virginia Attorney General, Office of the Virginia Attorney General, Richmond, Virginia; Patrick Morrisey, West Virginia Attorney General, Office of the West Virginia Attorney General, Charleston, West Virginia; Bridget Hill, Wyoming Attorney General, Office of the Wyoming Attorney General, Cheyenne, Wyoming; for Amici Curiae Montana, Idaho, and 23 Other States. Jeremiah L. Morgan, William J. Olson, and Robert J. Olson; William J. Olson PC, Vienna, Virginia; John I. Harris III, Schulman, LeRoy & Bennett PC, Nashville, Tennessee; for Amici Curiae Gun Owners of America Inc., Gun Owners Foundation, Gun Owners of California Inc., Heller Foundation, Virginia Citizens Defense League, Tennessee Firearms Association, America’s Future Inc., U.S. Constitutional Rights Legal Defense Fund, and Conservative Legal Defense and Education Fund. John Cutonilli, Garrett Park, Maryland, for Amicus Curiae John Cutonilli.

DUNCAN V. BONTA 11

OPINION

GRABER, Circuit Judge:

Mass shootings are devastating events for the victims, their families, and the broader community. The first mass shooting in the United States occurred in 1949, and they have increased in frequency and in lethality, primarily because of the widespread availability of modern firearm technology: semi-automatic firearms equipped with large- capacity magazines. A large-capacity magazine is a device that, when attached to a semi-automatic firearm, allows a shooter to fire more than ten rounds without pausing. A large-capacity magazine has little function in armed self- defense, but its use by mass shooters has exacerbated the harm of those horrific events. Murderers who use large- capacity magazines need not pause between shots until they have fired 20, 30, or even 100 rounds. These pauses are crucial. Victims and law enforcement personnel take advantage of short pauses in firing to flee, take cover, and fight back. A mass shooter’s use of large-capacity magazines limits those precious opportunities. In 2016, following long traditions in our Nation of protecting innocent persons by prohibiting especially dangerous uses of weapons and by regulating components of a firearm that are necessary to the firing of a firearm, the California legislature and California’s voters banned the possession of large-capacity magazines in order to address mass shootings. Earlier, lesser measures, such as banning the sale of those magazines, had proved both ineffective and difficult to enforce. Plaintiffs challenge the constitutionality of California’s ban. In Duncan v. Bonta (“Duncan V”), 19 F.4th 1087 (9th Cir. 2021) (en banc), we upheld the law as consistent with the Second Amendment and other constitutional guarantees. After the Supreme Court introduced a new framework for deciding Second Amendment challenges in New York State Rifle & Pistol Ass’n v. Bruen, 597 U.S. 1 (2022), the Court vacated our decision and remanded for reconsideration. Duncan v. Bonta, 142 S. Ct. 2895 (2022). Employing the methodology announced in Bruen and recently applied in United States v. Rahimi, 602 U.S. 680 (2024), we again conclude that California’s law comports with the Second Amendment, for two independent reasons. First, the Founders protected the right to keep and bear “Arms,” not a right to keep and bear “Arms and Accoutrements,” a common expression at the time of the Founding. Large-capacity magazines are optional accessories to firearms, and firearms operate as intended without a large-capacity magazine. A large-capacity magazine is thus an accessory or accoutrement, not an “Arm” in itself. Possession of a large-capacity magazine therefore falls outside the text of the Second Amendment. See Bruen, 597 U.S. at 24 (instructing courts to ask whether “the Second Amendment’s plain text covers an individual’s conduct”). Second, even assuming that the text of the Second Amendment encompasses the possession of an optional accessory like a large-capacity magazine, California’s law falls neatly within the Nation’s traditions of protecting innocent persons by prohibiting especially dangerous uses of weapons and by regulating components necessary to the firing of a firearm. Plaintiffs understate the extent to which our forebears regulated firearms to promote public safety. California’s law is relevantly similar to such historical regulations in both “how” and “why” it burdens the right to

DUNCAN V. BONTA 13

armed self-defense. Like those historical laws, California’s law restricts an especially dangerous feature of semi- automatic firearms—the ability to use a large-capacity magazine—while allowing all other uses of those firearms. So far as California’s law is concerned, persons may own as many bullets, magazines, and firearms as they desire; may fire as many rounds as they like; and may carry their bullets, magazines, and firearms wherever doing so is permissible. The only effect of California’s law on armed self-defense is the limitation that a person may fire no more than ten rounds without pausing to reload, something rarely done in self- defense. The justification for California’s law—to protect innocent persons from infrequent but devastating events—is also relevantly similar to the justifications for the historical laws. California’s law is not a precise match to the historical laws, “but it does not need to be.” Rahimi, 602 U.S. at 698. By prohibiting only an especially dangerous use of a modern weapon, the law “comport[s] with the principles underlying the Second Amendment.” Id. at 692. We reverse the district court’s contrary conclusion and remand with the instruction to enter judgment in favor of Defendant Rob Bonta, Attorney General of the State of California.

FACTUAL AND PROCEDURAL BACKGROUND 1

A. Large-Capacity Magazines A magazine is a device that automatically feeds ammunition into a firearm whenever the shooter fires a

1 Applying a bedrock principle of federal appellate review, we consider only the factual record developed by the parties. Fed. R. App. P. 10. With exceptions not relevant here, such as judicial notice, “we will not consider facts outside the record developed before the district court.” United States ex rel. Robinson Rancheria Citizens Council v. Borneo, Inc., 971 F.2d 244, 248 (9th Cir. 1992). We therefore do not consider factual information introduced in Judge VanDyke’s dissenting opinion. That dissent fundamentally misunderstands that legal rule by comparing an appellate-judge-made video, which neither the district court nor any party has ever seen or had an opportunity to comment on, to our citation in Mai v. United States, 952 F.3d 1106 (9th Cir. 2020), of publicly available scientific studies. Dissent by J. VanDyke at 144. The legal issue in Mai required us to assess whether adequate scientific evidence fairly supported a legislative judgment. 952 F.3d at 1118. The plaintiff’s primary argument—before us and before the district court— was that we should agree with the Sixth Circuit’s analysis of the scientific evidence in Tyler v. Hillsdale County Sheriff’s Department, 837 F.3d 678 (6th Cir. 2016) (en banc). Both parties in Mai discussed Tyler at length. In our opinion, we cited and discussed, as had the Sixth Circuit, the primary scientific study relevant to the legal issue. Mai, 952 F.3d at 1117–18; Tyler, 837 F.3d at 695–96. Our discussion of the publicly available scientific study that underpinned the parties’ primary dispute was entirely proper in the context of the legal issue at hand. Indeed, the parties had asked us to assess the scientific evidence. Presumably for that reason, neither the parties nor a single dissent from denial of rehearing en banc asserted that we had cited facts outside the record developed before the district court. Mai v. United States, 974 F.3d 1082, 1083 (9th Cir. 2020) (order) (Collins, J., dissenting from the denial of reh’g en banc); id. at 1083–97 (Bumatay, J., dissenting from the denial of reh’g en banc); id. at 1097–1106 (VanDyke, J., dissenting from the denial of reh’g en banc). By sharp contrast to that publicly available study, the judge-made video here clearly contains facts outside the record developed before the district court. Judge VanDyke’s dissenting opinion here also cites an offhand footnote in Mai. In that footnote, we took judicial notice of the fact that evidence of a certain sort exists in other contexts, we cited a report by the American Cancer Society, and we observed that no similar evidence existed in the context relevant to the case. Mai, 952 F.3d at 1118 n.7. Judicial notice was appropriate because we took notice of the existence of evidence of a particular sort, regardless of its accuracy. Von Saher v. Norton Simon Museum of Art at Pasadena, 592 F.3d 954, 960 (9th Cir. 2010). Judicial notice plainly does not authorize the judge-made video contained in Judge VanDyke’s dissenting opinion.

DUNCAN V. BONTA 15

bullet. Although some magazines are permanently affixed to a firearm, most magazines are detachable. Ocean State Tactical, LLC v. Rhode Island, 95 F.4th 38, 42 (1st Cir. 2024), petition for cert. filed, No. 24-131 (U.S. Aug. 2, 2024). When a magazine feeds a semi-automatic firearm, the shooter may continue to fire without pause and without taking any action other than pulling the trigger to fire successive rounds. A shooter thus may fire, repeatedly and without meaningful pause, all bullets in the magazine. Many jurisdictions, including California, define “large- capacity magazine” to include any magazine or similar automatic feeding device that can hold more than ten rounds of ammunition. E.g., Cal. Penal Code § 16740; Conn. Gen. Stat. § 53-202w(a)(1); Haw. Rev. Stat. § 134-8(c). Large- capacity magazines thus allow a shooter to fire more than ten rounds without reloading. Once a magazine is empty, a shooter may reload bullets into the magazine or may attach a different magazine to the firearm. It takes anywhere from a few to ten seconds for a person to change magazines, depending on the shooter’s skill and the surrounding circumstances. For those firearms that accept magazines, manufacturers often include large-capacity detachable magazines as part of the standard package when the firearm is purchased. “Most pistols are manufactured with magazines holding ten to seventeen rounds, and many popular rifles are manufactured with magazines holding twenty or thirty rounds.” Kolbe v. Hogan, 849 F.3d 114, 129 (4th Cir. 2017) (en banc), abrogated in other part by Bruen, 597 U.S. 1. Although data are imprecise, experts estimate that approximately half of privately owned magazines hold more than ten rounds. A large-capacity magazine—which enables a shooter to fire more than ten bullets rapidly and without reloading— has almost no utility in the lawful defense of the home, but it has devastating effects in mass shootings. A shooter equipped with a large-capacity magazine may kill and injure many people in rapid succession, not only because the shooter can fire many bullets quickly but also because the shooter can fire without pausing to reload. Those pauses are crucial because they allow intended victims and law enforcement personnel to flee, take cover, and fight back. More than twice as many people have been killed or injured in mass shootings that involved a large-capacity magazine than in mass shootings that involved a smaller-capacity magazine. And in the past half-century, large-capacity magazines have been used in about three-quarters of gun massacres with ten or more deaths and in every gun massacre with twenty or more deaths. B. California’s Ban In 1994, Congress banned the possession or transfer of large-capacity magazines. Pub. L. No. 103-322, § 110103, 108 Stat. 1796, 1998–2000 (1994). Like California’s law, the federal statute applied to a magazine “that has a capacity of . . . more than 10 rounds of ammunition.” Id. The federal ban exempted magazines that were legally possessed before the date of enactment. Id. The law expired ten years later, in 2004. Id. § 110105(2). California began regulating large-capacity magazines in 2000, prohibiting their manufacture, importation, or sale in the state. Cal. Penal Code § 12020(a)(2) (2000). After the expiration of the federal ban, California strengthened its law in 2010 and again in 2013 by, among other things, prohibiting the purchase or receipt of large-capacity

DUNCAN V. BONTA 17

magazines. Cal. Penal Code § 32310(a) (2013). But possession of large-capacity magazines remained legal, and law enforcement officers reported to the California legislature that enforcement of the existing laws was “very difficult.” In 2016, the California legislature enacted Senate Bill 1446, which barred possession of large-capacity magazines as of July 1, 2017, and imposed a fine for failing to comply. 2016 Cal. Stat. ch. 58, § 1. Later in 2016, voters in California approved Proposition 63, also known as the Safety for All Act of 2016, which subsumed Senate Bill 1446 and added provisions that imposed a possible criminal penalty of imprisonment for up to a year for unlawful possession of large-capacity magazines after July 1, 2017. Cal. Penal Code § 32310(c). Proposition 63 declared that large-capacity magazines “significantly increase a shooter’s ability to kill a lot of people in a short amount of time.” Cal. Prop. 63 § 2(11). “No one except trained law enforcement should be able to possess these dangerous ammunition magazines,” and the existing law’s lack of a ban on possession constituted a “loophole.” Id. § 2(12). The law’s stated purpose is “[t]o make it illegal in California to possess the kinds of military-style ammunition magazines that enable mass killings like those at Sandy Hook Elementary School; a movie theater in Aurora, Colorado; Columbine High School; and an office building at 101 California Street in San Francisco, California.” Id. § 3(8). California law defines a “large-capacity magazine” as any ammunition-feeding device with the capacity to accept more than ten rounds, but does not include any of the following:

(a) A feeding device that has been permanently altered so that it cannot accommodate more than 10 rounds. (b) A .22 caliber tube ammunition feeding device. (c) A tubular magazine that is contained in a lever-action firearm.

Cal. Penal Code § 16740. The ban on possession of large- capacity magazines exempts persons such as active or retired law enforcement officers and security guards for armored vehicles. Id. §§ 32400–55. The law requires any current, non-exempt possessor of a large-capacity magazine to (1) remove it from the state, (2) sell it to a licensed dealer, (3) turn it in to law enforcement for destruction, or (4) permanently alter it so that it can accept no more than ten rounds. Id. §§ 16740(a), 32310(d). The District of Columbia and thirteen other states have imposed similar restrictions on large-capacity magazines. Colo. Rev. Stat. §§ 18-12-301, 302; Conn. Gen. Stat. § 53- 202w; Del. Code tit. 11, §§ 1468–69; D.C. Code § 7- 2506.01(b); Haw. Rev. Stat. § 134-8(c); 720 Ill. Comp. Stat. 5/24-1.10; Md. Code Ann., Crim. Law § 4-305(b); Mass. Gen. Laws ch. 140, §§ 121, 131(a), 131M; N.J. Stat. Ann. §§ 2C:39-1(y), 39-3(j), 39-9(h); N.Y. Penal Law §§ 265.00, 265.02(8); 2022 Oregon Ballot Measure 114, § 11; R.I. Gen. Laws §§ 11-47.1-2(2), 11-47.1-3; Vt. Stat. Ann. tit. 13, § 4021; Wash. Rev. Code §§ 9.41.010, 9.41.370.

DUNCAN V. BONTA 19

C. Procedural History Plaintiffs own, or represent those who own, large- capacity magazines, and they do not want to comply with California’s law. Plaintiffs brought this action in 2017, arguing that California’s prohibition on the possession of large-capacity magazines violates the Second Amendment, the Fifth Amendment’s Takings Clause, and the Fourteenth Amendment’s Due Process Clause. The district court preliminarily enjoined Defendant from enforcing the law, holding that Plaintiffs were likely to succeed on their Second Amendment and Takings Clause claims. Duncan v. Becerra (“Duncan I”), 265 F. Supp. 3d 1106 (S.D. Cal. 2017). On appeal, a divided panel affirmed the preliminary injunction, concluding that the district court did not abuse its discretion in holding that Plaintiffs had shown a likelihood of success on their claims. Duncan v. Becerra (“Duncan II”), 742 F. App’x 218, 221–22 (9th Cir. 2018) (unpublished); see id. at 223–26 (Wallace, J., dissenting) (voting to reverse the preliminary injunction). In 2019, the district court granted summary judgment to Plaintiffs on the Second Amendment and takings claims, and the court permanently enjoined enforcement of the law. Duncan v. Becerra (“Duncan III”), 366 F. Supp. 3d 1131 (S.D. Cal. 2019). On appeal, another divided panel affirmed the summary judgment as to the Second Amendment claim. Duncan v. Becerra (“Duncan IV”), 970 F.3d 1133 (9th Cir. 2020); see id. at 1169–76 (Lynn, D.J., dissenting) (stating that she would reject the Second Amendment claim). A majority of active judges voted to rehear the case en banc. Duncan v. Becerra, 988 F.3d 1209 (9th Cir. 2021) (order). Sitting as the en banc court, we applied the two-part test adopted by all circuit courts at the time, reversed the district court’s judgment, and remanded with the instruction to enter judgment for Defendant. Duncan V, 19 F.4th 1087. After our ruling, the Supreme Court decided Bruen, announcing a new framework for deciding Second Amendment challenges. The Supreme Court then granted certiorari, vacated the judgment, and remanded for further consideration in light of Bruen. Duncan, 142 S. Ct. 2895. We, in turn, remanded the case to the district court to consider, in the first instance, the effect of Bruen on the Second Amendment claim. Duncan v. Bonta, 49 F.4th 1228 (9th Cir. 2022) (en banc) (order). On remand, the district court declined to reopen discovery or to conduct an evidentiary hearing or trial. Instead, the court again granted summary judgment to Plaintiffs on the Second Amendment claim and permanently enjoined Defendant from enforcing the law. Duncan v. Bonta (“Duncan VI”), 695 F. Supp. 3d 1206 (S.D. Cal. 2023). Defendant timely appealed, and the en banc court chose, pursuant to Ninth Circuit General Order 3.6(b), to address the new appeal. We granted Defendant’s motion to stay the permanent injunction pending our resolution of this appeal. Duncan v. Bonta (“Duncan VII”), 83 F.4th 803 (9th Cir. 2023) (en banc) (order). After we heard oral argument, the Supreme Court decided Rahimi, 602 U.S. 680.

DISCUSSION

A. Takings Claim The Fifth Amendment provides, “nor shall private property be taken for public use, without just compensation.” U.S. Const. amend. V. Plaintiffs assert that, on its face, the statute effects a taking because it requires Plaintiffs either to

DUNCAN V. BONTA 21

(1) modify any large-capacity magazines so that they accept ten or fewer rounds, (2) sell them, (3) move them out of the state, or (4) turn them over to state officials for destruction. Cal. Penal Code §§ 16740(a), 32310(d). We rejected that claim in Duncan V, 19 F.4th at 1111–13. On remand, the district court did not consider the takings claim but, on appeal, Plaintiffs ask us to affirm on the alternative ground of a takings violation. The Supreme Court’s decision in Bruen had no effect on our takings analysis, nor have any other intervening decisions aided Plaintiffs’ position. We adopt and affirm our earlier rejection of this claim. Id. at 1111–13. Our ruling is in accord with decisions by the First and Third Circuits. See Ocean State, 95 F.4th at 52–53 (holding that the plaintiffs were unlikely to succeed on a takings challenge to Rhode Island’s ban on large-capacity magazines); Ass’n of N.J. Rifle & Pistol Clubs, Inc. v. Att’y Gen. N.J. (“ANJRPC”), 910 F.3d 106, 124–25 (3d Cir. 2018) (rejecting a takings challenge to New Jersey’s ban on large-capacity magazines), abrogated in other part by Bruen, 597 U.S. 1. B. Second Amendment Claim The Second Amendment provides: “A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.” U.S. Const. amend. II. The Amendment creates an individual right to keep and bear arms for self-defense. District of Columbia v. Heller, 554 U.S. 570, 599, 602 (2008). The right applies against States via the Fourteenth Amendment. McDonald v. City of Chicago, 561 U.S. 742, 750 (2010) (plurality opinion). The Supreme Court has instructed that, “[l]ike most rights, the right secured by the Second Amendment is not unlimited.” Heller, 554 U.S. at 626. It is “not a right to keep and carry any weapon whatsoever in any manner whatsoever and for whatever purpose.” Id. For example, the Second Amendment protects only those weapons “in common use at the time.” Id. at 627 (quoting United States v. Miller, 307

U.S. 174, 179 (1939)).

In Bruen, the Supreme Court announced the appropriate general methodology for deciding Second Amendment challenges to state laws:

When the Second Amendment’s plain text covers an individual’s conduct, the Constitution presumptively protects that conduct. The government must then justify its regulation by demonstrating that it is consistent with the Nation’s historical tradition of firearm regulation.

597 U.S. at 24. Applying that methodology to California’s ban on large- capacity magazines, we reject Plaintiffs’ Second Amendment challenge for two independent reasons. First, the plain text of the Amendment protects the right to bear “Arms,” not accessories to firearms that are neither arms themselves nor necessary to the ordinary functioning of a firearm. Because large-capacity magazines are neither weapons nor accessories that are necessary to the operation of a weapon, the Second Amendment’s plain text does not protect possession of large-capacity magazines. Second, even assuming that California’s law implicates the text of the Second Amendment, the ban on large-capacity magazines fits comfortably within our “historical tradition of firearm regulation,” Rahimi, 602 U.S. at 691 (quoting Bruen, 597

DUNCAN V. BONTA 23

U.S. at 17). California’s law fits within the traditions of protecting innocent persons by restricting a component necessary to the firing of a firearm and by restricting especially dangerous uses of weapons when those uses have proved particularly harmful. 1. Large-Capacity Magazines Are Neither “Arms” Nor Protected Accessories. We first examine “whether the plain text of the Second Amendment protects [Plaintiffs’] proposed course of conduct.” Bruen, 597 U.S. at 32. Plaintiffs assert that their proposed conduct—possessing large-capacity magazines— implicates the text of the Second Amendment because, in their view, large-capacity magazines are arms commonly chosen for the purpose of self-defense. Defendant raises several distinct arguments to the contrary: large-capacity magazines are not “Arms” within the meaning of the Second Amendment; they are not in common use for self-defense; they are most useful in military service; and they are dangerous and unusual. See, e.g., Hanson v. District of Columbia, 671 F. Supp. 3d 1, 14 (D.D.C. 2023) (holding that large-capacity magazines do not fall within the text of the Second Amendment because “they are most useful in military service”), aff’d on other grounds, 120 F.4th 223 (D.C. Cir. 2024); see also Bianchi v. Brown, 111 F.4th 438, 461 (4th Cir. 2024) (en banc) (“[T]he AR-15 is a combat rifle that is both ill-suited and disproportionate to self-defense. It thereby lies outside the scope of the Second Amendment.”), petition for cert. filed sub nom. Snope v. Brown, No. 24-203 (U.S. Aug. 21, 2024). Because we conclude that large- capacity magazines are not “Arms” within the meaning of the Second Amendment, we need not, and do not, reach Defendant’s alternative arguments pertaining to the text of the Second Amendment. 2 The text of the Second Amendment encompasses “the right of the people to keep and bear Arms.” U.S. Const. amend. II. We must look to history to understand the meaning of “Arms.” Bruen, 597 U.S. at 26. The Second Amendment’s “reference to ‘arms’ does not apply ‘only to those arms in existence in the 18th century.’” Id. at 28 (brackets omitted) (quoting Heller, 554 U.S. at 582). “[T]he Second Amendment extends, prima facie, to all instruments that constitute bearable arms, even those that were not in existence at the time of the founding.” Id. (quoting Heller, 554 U.S. at 582) (internal quotation marks omitted). “Thus, even though the Second Amendment’s definition of ‘arms’

2 “There is no consensus on whether the common-use issue” is a threshold, textual inquiry or a historical inquiry. Hanson, 120 F.4th at 232 n.3 (quoting Bevis v. City of Naperville, 85 F.4th 1175, 1198 (7th Cir. 2023), cert. denied sub nom. Harrel v. Raoul, 144 S. Ct. 2491 (2024)). In United States v. Alaniz, 69 F.4th 1124 (9th Cir. 2023), we placed this question in the initial, textual determination. Id. at 1128. Judge Bumatay’s dissenting opinion now argues, in direct contrast to his view less than eighteen months ago, that the inquiry belongs instead in the historical analysis. Compare Dissent by J. Bumatay at 92– 95 (arguing that the inquiry belongs in the historical analysis), with Duncan VII, 83 F.4th at 810 (Bumatay, J., dissenting) (placing the inquiry clearly in the textual category). Both views find some support in the text of Bruen. See Bianchi, 111 F.4th at 501–02 (Richardson, J., dissenting) (describing the support for both approaches). Because we do not reach the issue as presented by Defendant, we need not and do not address the issue here; therefore, Alaniz remains good law. To the extent that Plaintiffs’ argument about ownership statistics overlaps with, or is identical to, an “in common use today for self-defense” argument, we give Plaintiffs the benefit of the doubt and, out of an abundance of caution, resolve this question in the historical analysis, where Defendant bears the burden of proof. See Part B-2-d, below.

DUNCAN V. BONTA 25

is fixed according to its historical understanding, that general definition covers modern instruments that facilitate armed self-defense.” Id. “The 18th-century meaning [of ‘Arms’] is no different from the meaning today.” Heller, 554 U.S. at 581. Consistent with modern usage, dictionaries from the 18th century defined the term as encompassing “weapons of offence, or armour of defence” and “any thing that a man wears for his defence, or takes into his hands, or useth in wrath to cast at or strike another.” Id. (quoting, first, 1 Dictionary of the English Language 106 (4th ed.) (reprinted 1978) and, second, 1 A New and Complete Law Dictionary (1771)). The term includes commonplace weapons and is not limited to military weapons. Id. The meaning of “Arms” thus broadly includes nearly all weapons used for armed self-defense. The scope of the Second Amendment is broad in a second sense as well. As we recognized a decade ago, for the right to bear arms to have meaning, the Amendment’s text must carry an implicit, corollary right to bear the components or accessories necessary for the ordinary functioning of a firearm. See Jackson v. City & County of San Francisco, 746 F.3d 953, 967 (9th Cir. 2014) (holding that, unless understood to protect the corollary right to possess ammunition, “the right to bear arms would be meaningless”), abrogated in other part by Bruen, 597 U.S. 1; see also Fyock v. Sunnyvale, 779 F.3d 991, 998 (9th Cir. 2015) (holding that the Second Amendment encompasses a “corollary” right to possess components “necessary to render . . . firearms operable”), abrogated in other part by Bruen, 597 U.S. 1; cf. B & L Prods., Inc. v. Newsom, 104 F.4th 108, 118 (9th Cir. 2024) (reaffirming, after Bruen, that “unless the right to acquire firearms receives some Second Amendment protection, the right to keep and bear firearms would be meaningless” (emphasis omitted)); id. (“Ancillary rights are protected to the extent necessary to serve [lawful purposes such as self-defense]; otherwise, the Second Amendment is not implicated by restraints on such rights.”). A complete ban on ammunition thus would implicate the Second Amendment, as likely would a ban on, for example, firearm triggers. But the text of the Second Amendment also reveals an important limit on the scope of the right. In particular, the Amendment protects only the right to bear Arms. At the time of ratification, a clear distinction was recognized between weapons themselves, referred to as “arms,” and accessories of weaponry, referred to as “accoutrements.” Common accoutrements included flint, scabbards, holsters, and ammunition containers such as cartridge cases and cartridge boxes. “Accoutrements” were distinct from “arms.” For example, the Continental Congress promised to pay States for “[e]very horse and all arms and accoutrements, which shall be taken, by the enemy in action.” 2 Public Papers of George Clinton 828 (Wynkoop Hallenbeck Crawford Co. ed., 1900) (emphasis added). Similarly, the Duke of Wellington described the need “to collect the wounded and their arms and accoutrements” from a battlefield. 10 The Dispatches of Field Marshal the Duke of Wellington (1799-1818) 495 (Murray ed., 1838) (emphasis added). Hundreds of examples from the Founding era describe arms and accoutrements as separate, distinct items of military gear, and the phrase “arms and accoutrements” was common. By choosing to protect the right to bear “arms,” not “arms and accoutrements,” the Founders constrained the scope of the Second Amendment. The term “Arms” thus

DUNCAN V. BONTA 27

encompasses most weapons used in armed self-defense, and the Second Amendment necessarily protects the components necessary to operate those weapons. But it does not protect the right to bear accoutrements. Applying those principles here, California’s ban on large-capacity magazines does not fall within the plain text of the Second Amendment. A large-capacity magazine is a box that, by itself, is harmless. It cannot reasonably be described as an item that a person “takes into his hands, or useth in wrath to cast at or strike another.” Heller, 554 U.S. at 581. Nor can it be reasonably described, by itself, as a “weapon[] of offence, or armour of defence.” Id. Without an accompanying firearm, a large-capacity magazine is benign, useless in combat for either offense or defense. Large-capacity magazines thus fall clearly within the category of accessories, or accoutrements, rather than arms. That straightforward conclusion does not end our analysis, though. We also must consider whether the possession of large-capacity magazines falls within the corollary right to possess accessories that are necessary for the ordinary operation of a protected weapon. 3 Some (but not all) firearms require the use of a magazine in order to operate. For that reason, the Second Amendment’s text necessarily encompasses the corollary right to possess a magazine for firearms that require one, just as it protects the right to possess ammunition and triggers. Otherwise, the right to bear arms, including firearms that require the use of a magazine, would be diminished.

3 We assume for purposes of analysis that Plaintiffs intend to possess weapons that are protected by the text of the Second Amendment. But a large-capacity magazine—the only type of accessory regulated by California—is not necessary to operate any firearm. Most firearms that accept detachable magazines can be equipped with a large-capacity magazine, but the record contains no example of a firearm that requires a large-capacity magazine to function normally. To the contrary, firearms that accept magazines operate as intended when equipped with magazines containing ten or fewer rounds. Accordingly, the Second Amendment’s plain text does not encompass a right to possess large-capacity magazines. Plaintiffs point out that a magazine is attached to the firearm when the shooter fires a shot. Unlike some other accessories, then, a magazine is an integral part of the firing mechanism of some firearms. Plaintiffs contend that, for that reason, the Second Amendment’s text necessarily encompasses a right to possess a magazine for firearms that require one. We agree, for the reasons described above, that the Second Amendment’s text encompasses a right to possess a magazine in that circumstance. Plaintiffs further contend, however, that the Amendment’s text also encompasses a right to possess a large-capacity magazine because, when a shooter chooses to use a large-capacity magazine, it, too, is attached to the firearm when the shooter fires a shot. We reject that reasoning for two independent reasons. First, the function of the large-capacity magazine is completed once the magazine automatically places a new round into the chamber. The large capacity of the magazine plays no role in the firing mechanism of the firearm. In this way, a large- capacity magazine is no different than other items that hold additional ammunition, such as cartridge boxes and belts that

DUNCAN V. BONTA 29

hold bullets, yet were classified historically as accoutrements, not arms. Second, and more fundamentally, Plaintiffs’ contention misunderstands the relevant test. The proper inquiry for an item that is not an arm itself is whether the component or accessory is necessary to the ordinary operation of the weapon, not whether, when one voluntarily chooses to use an optional accessory, the accessory is attached to the weapon. Many optional accessories—such as a high- powered scope for a rifle, a gun sling, or a silencer—may be attached to a firearm without necessarily falling within the scope of the text of the Second Amendment. See, e.g., United States v. Cox, 906 F.3d 1170, 1186 (10th Cir. 2018) (“A silencer is a firearm accessory; it’s not a weapon in itself.”). A large-capacity magazine undoubtedly provides a benefit for a shooter: it allows firing an eleventh round or more without having to pause for a few seconds to reload. But the enhancement of a person’s ability to fight or to defend is a fundamental attribute of any accessory for a weapon. A sword sheathed in a scabbard, a rifle equipped with a high-powered scope, a six-shooter nestled loosely in a holster—all are superior in some way to the same weapons without the accessory. The mere fact that an accessory enhances an attribute of a weapon does not bring the accessory within the scope of the Second Amendment’s text. The Founders limited the Second Amendment’s protection to the right to bear arms, not the broader right to bear arms and accoutrements. We must respect that limitation, just as we must respect the Founders’ choice to protect the right to bear a broad range of arms. Cf. Rahimi, 602 U.S. at 691–92 (“[T]he Second Amendment permits more than just those regulations identical to ones that could be found in 1791. Holding otherwise would be as mistaken as applying the protections of the right only to muskets and sabers.”). Because the text of the Second Amendment does not encompass the right to possess large-capacity magazines, we hold that Plaintiffs’ Second Amendment claim fails. See Or. Firearms Fed’n v. Kotek, 682 F. Supp. 3d 874, 911–13 (D. Or. 2023) (explaining why large-capacity magazines are not “Arms” within the meaning of the Second Amendment), appeals filed, Nos. 23-35478, 23-35479 (9th Cir. July 17, 2023); Ocean State Tactical, LLC v. Rhode Island, 646 F. Supp. 3d 368, 384–88 (D.R.I. 2022) (same), aff’d on other grounds 95 F.4th 38; cf. Bevis, 85 F.4th at 1195 (concluding, for a different reason, that large-capacity magazines are not “Arms”). 2. California’s Ban on Large-Capacity Magazines Falls Within the Nation’s Historical Tradition of Firearm Regulation. Plaintiffs’ argument fares no better even if we assume that their proposed conduct falls within the plain text of the Second Amendment. In Bruen, the Supreme Court explained that a government must justify a regulation of firearms by demonstrating that it falls within the Nation’s tradition of regulating weapons. 597 U.S. at 24. A court’s assessment of whether a law comports with a tradition defined by historical laws “will often involve reasoning by analogy.” Id. at 28. “[D]etermining whether a historical regulation is a proper analogue for a distinctly modern firearm regulation requires a determination of whether the two regulations are ‘relevantly similar.’” Id. at 28–29. The two most important considerations are “how and why the regulations burden a

DUNCAN V. BONTA 31

law-abiding citizen’s right to armed self-defense.” Id. at 29. More specifically, we must consider whether the regulations “impose a comparable burden on the right of armed self- defense and whether that burden is comparably justified.” Id. Analogical reasoning is “neither a regulatory straightjacket nor a regulatory blank check.” Id. at 30. The government must “identify a well-established and representative historical analogue, not a historical twin.” Id. So even if a modern regulation is not a “dead ringer” for a historical analogue, “it still may be analogous enough to pass constitutional muster.” Id. The Court gave an example of the longstanding laws prohibiting the carry of firearms in “sensitive places,” such as legislative assemblies and courthouses. Id. Although few such historical laws existed, the Court “assume[d] it settled that these locations were ‘sensitive places’ where arms carrying could be prohibited consistent with the Second Amendment.” Id. “And courts can use analogies to those historical regulations of ‘sensitive places’ to determine that modern regulations prohibiting the carry of firearms in new and analogous sensitive places are constitutionally permissible.” Id. But those historical laws do not justify the conclusion that entire cities are “sensitive places” simply because people congregate there and law enforcement is present. Id. at 30–31. That analogy would be “far too broad[],” because it “would eviscerate the general right to publicly carry arms for self-defense.” Id. at 31. The Court emphasized that the historical analogies both in Bruen and in Heller were “relatively simple to draw.” Id. at 27. Heller concerned the District of Columbia’s “complete prohibition” on handguns, 554 U.S. at 629, and Bruen concerned New York’s “may issue” licensing scheme, 597 U.S. at 14. In those cases, the perceived societal problems—firearm violence in densely populated communities and the need to regulate who may possess a firearm—had existed since the Founding, and the regulations that the governments chose to impose—a ban on handguns and a “may issue” licensing scheme—were ones “that the Founders themselves could have adopted” to confront those problems. Bruen, 597 U.S. at 27. “[W]hen a challenged regulation addresses a general societal problem that has persisted since the 18th century,” then a modern regulation likely is inconsistent with the Second Amendment if the Founders either addressed the problem “through materially different means” or did not enact “a distinctly similar historical regulation.” Id. at 26. By contrast, the Court explained, “cases implicating unprecedented societal concerns or dramatic technological changes may require a more nuanced approach.” Id. at 27. “The regulatory challenges posed by firearms today are not always the same as those that preoccupied the Founders in 1791 or the Reconstruction generation in 1868.” Id. “Although its meaning is fixed according to the understandings of those who ratified it, the Constitution can, and must, apply to circumstances beyond those the Founders specifically anticipated.” Id. at 28. When considering historical sources, “not all history is created equal.” Id. at 34. Regulations enacted close to the time of ratification are the most relevant, because “[c]onstitutional rights are enshrined with the scope they were understood to have when the people adopted them.” Id. (quoting Heller, 554 U.S. at 634–35) (emphasis in Bruen). Historical evidence that long predates, or long postdates, the date of ratification is less illuminating, particularly if it

DUNCAN V. BONTA 33

contradicts the text of the Second Amendment or evidence from the time of the ratification. 4 Id. at 34–37. The Supreme Court recently applied Bruen’s framework in Rahimi, 602 U.S. 680. Eight justices joined the majority opinion, with only Justice Thomas, the author of Bruen, dissenting. Id. at 683. Rahimi concerned a challenge to 18 U.S.C. § 922(g)(8), which criminalizes the possession of a firearm by a person who is subject to a domestic-violence restraining order. Id. at 688–89. The Court had “no trouble” concluding that § 922(g)(8), as applied to Rahimi, was consistent with the Second Amendment. Id. at 700. The Court began by noting that some courts had misunderstood Bruen’s methodology and had applied it too stringently. Id. at 691. The Court emphasized that the methodology was “not meant to suggest a law trapped in amber.” Id. “[T]he Second Amendment permits more than just those regulations identical to ones that could be found in 1791.” Id. at 691–92. Rahimi summarized the methodology as follows: “[T]he appropriate analysis involves considering whether the challenged regulation is consistent with the principles that underpin our regulatory tradition.” Id. at 692. “Why and how the regulation burdens the right are central to this

4 The Court has made clear that at least one relevant date of ratification is 1791, the year in which the States ratified the Second Amendment. Bruen, 597 U.S. at 37. The Second Amendment’s protections apply to the States via the Fourteenth Amendment, which was ratified in 1868. Id. The Court twice has reserved whether, for laws enacted by States, another relevant date is 1868. Id. at 38; Rahimi, 602 U.S. at 692 n.1. We need not address that issue here, because the relevant historical traditions, discussed below, all began at the time of the Founding or earlier. inquiry.” Id. “And when a challenged regulation does not precisely match its historical precursors, ‘it still may be analogous enough to pass constitutional muster.’” Id. (quoting Bruen, 597 U.S. at 30). “The law must comport with the principles underlying the Second Amendment, but it need not be a ‘dead ringer’ or a ‘historical twin.’” Id. (quoting Bruen, 597 U.S. at 30). Applying that methodology to § 922(g)(8), Rahimi looked to “two distinct legal regimes.” Id. at 694. The first regime consisted of surety laws, laws that required a person to post a bond that would be forfeited if the person later breached the peace. Id. at 695–97. Those laws sometimes applied to persons who carried dangerous weapons publicly. Id. at 696. The second regime consisted of “going armed” laws, laws that prohibited “riding or going armed, with dangerous or unusual weapons, to terrify the good people of the land.” Id. at 697 (quoting 4 Blackstone 149) (brackets omitted).

Taken together, the surety and going armed laws confirm what common sense suggests: When an individual poses a clear threat of physical violence to another, the threatening individual may be disarmed. Section 922(g)(8) is by no means identical to these founding era regimes, but it does not need to be. See Bruen, 597 U.S. at 30. Its prohibition on the possession of firearms by those found by a court to present a threat to others fits neatly within the tradition the surety and going armed laws represent.

Id. at 698.

DUNCAN V. BONTA 35

Section 922(g)(8), the Court held, is “relevantly similar” to the Founding-era regimes in both why and how the law burdens the Second Amendment right. Id. (quoting Bruen, 597 U.S. at 29). As to “why,” both the historical regimes and the modern law “restrict[] gun use to mitigate demonstrated threats of physical violence.” Id. As to “how,” the historical and modern laws require a judicial determination of whether a defendant “likely would threaten or had threatened another with a weapon.” Id. at 699. Both the surety laws and the modern law are temporary in duration. Id. Finally, the going-armed laws provided for imprisonment, so § 922(g)(8)’s lesser restriction of disarmament is permissible. Id. In dissent, Justice Thomas pointed out that § 922(g)(8) addressed a “societal problem—the risk of interpersonal violence—‘that has persisted since the 18th century,’ yet was addressed ‘through the materially different means’ of surety laws.” Id. at 752–53 (Thomas, J., dissenting) (quoting Bruen, 597 U.S. at 26) (brackets omitted). He explained that the historical regulations were “materially different” from § 922(g)(8)’s complete ban on firearm possession arising out of private conduct. Id. at 764–71. Justice Thomas wrote that surety laws had the same “why,” but the “how” could not be more different. Id. at 764. Surety laws, he noted, had no effect whatsoever on the right to bear arms, either when posting a bond or when forfeiting that bond because of a breach of the peace: “After providing sureties, a person kept possession of all his firearms; could purchase additional firearms; and could carry firearms in public and private. Even if he breached the peace, the only penalty was that he and his sureties had to pay a sum of money.” Id. “At base, it is difficult to imagine how surety laws can be considered relevantly similar to a complete ban on firearm ownership, possession, and use.” Id. at 766. As for the going-armed laws, Justice Thomas noted that they “had a dissimilar burden and justification.” Id. at 767. Going-armed laws prohibited only distinctly public acts; they “did not prohibit carrying firearms at home or even public carry generally.” Id. at 769. And those laws “prohibited only carrying certain weapons (‘dangerous and unusual’) in a particular manner (‘terrifying the good people of the land’ without a need for self-defense) and in particular places (in public).” Id. at 770. Plus, those laws “were criminal statutes that penalized past behavior, whereas § 922(g)(8) is triggered by a civil restraining order that seeks to prevent future behavior.” Id. The majority responded to Justice Thomas’ critique by stating that it reached the opposite conclusion “[f]or the reasons we have set forth,” id. at 700 (majority opinion), and reiterating simply that “a ‘historical twin’ is not required,” id. at 701 (quoting Bruen, 597 U.S. at 30). Applying the methodology described in Bruen and Rahimi, we first conclude that (a) the “more nuanced approach” described in Bruen applies here. 597 U.S. at 27. We next consider (b) the historical regulations provided by Defendant. Considering “how” and “why” the modern and historical regulations burden the right to armed self-defense, we conclude that (c) California’s law fits within the Nation’s tradition of regulating firearms. Finally, we reject (d) Plaintiffs’ ownership-statistics argument before (e) concluding that California’s law is constitutional.

DUNCAN V. BONTA 37

a. The “More Nuanced Approach” Applies Here. In Bruen, the Supreme Court stated that “cases implicating unprecedented societal concerns or dramatic technological changes may require a more nuanced approach.” 597 U.S. at 27. The parties dispute whether this case implicates unprecedented concerns or dramatic technological changes. As an initial matter, the precise status of Bruen’s “more nuanced approach” statement is unclear following Rahimi. Rahimi addressed a societal challenge—domestic violence—that has persisted since the Founding. And nothing in the record suggested a dramatic technological change. 5 The dissent insisted that the case, like Heller and Bruen, fell neatly into the straightforward category, because “§ 922(g)(8) addresses a societal problem—the risk of interpersonal violence—‘that has persisted since the 18th century.’” Rahimi, 602 U.S. at 752 (Thomas, J., dissenting) (quoting Bruen, 597 U.S. at 26). Yet the majority did not address whether that case implicated unprecedented societal concerns or dramatic technological changes. Notwithstanding the Supreme Court’s silence on this topic in Rahimi, we will not disregard Bruen’s statement that a more nuanced approach applies to cases involving modern problems or dramatic technological changes. Those cases warrant an even more flexible approach than the Court applied in Rahimi. To conclude otherwise would be to

5 The Founders likely could not have imagined the weaponry available today, so in that sense every Second Amendment case involves dramatic technological changes. But Bruen could not have meant that type of change, because it held that Bruen itself, along with Heller, were cases that did not implicate dramatic technological changes. Bruen, 597 U.S. at 27. disregard entirely a statement by the Supreme Court. Even assuming that Bruen’s statement was dictum, we do not lightly decline to follow a clear statement by the Supreme Court. See, e.g., Fernandez-Ruiz v. Gonzales, 466 F.3d 1121, 1129 (9th Cir. 2006) (en banc) (“[A]s a lower federal court, we are advised to follow the Supreme Court’s considered dicta.” (quoting Oyebanji v. Gonzales, 418 F.3d 260, 264–65 (3d Cir. 2005))). We readily conclude that a more nuanced approach is appropriate here. This case implicates both unprecedented societal concerns and dramatic technological changes. See Hanson, 120 F.4th at 240–42 (holding that the nuanced approach applies to a challenge to a ban on large-capacity magazines); Ocean State, 95 F.4th at 44 (same); Ass’n of N.J. Rifle & Pistol Clubs, Inc. v. Platkin, No. CV 18-10507- PGS-JBD, 2024 WL 3585580, at *22 (D.N.J. July 30, 2024) (same), cross appeals filed, Nos. 24-2415, 24-2450 (3d Cir. Aug. 6 & 9, 2024); Capen v. Campbell, 708 F. Supp. 3d 65, 89–90 (D. Mass. 2023) (same), appeal filed, No. 24-1061 (1st Cir. Jan. 17, 2024); Del. State Sportsmen’s Ass’n v. Del. Dep’t of Safety & Homeland Sec. (“DSSA”), 664 F. Supp. 3d 584, 598 (D. Del. 2023) (same), aff’d on other grounds, 108 F.4th 194 (3d Cir. 2024); Hanson, 671 F. Supp. 3d at 17–20 (same); Herrera v. Raoul, 670 F. Supp. 3d 665, 675 (N.D. Ill. 2023) (same), aff’d sub nom. Bevis, 85 F.4th 1175; Kotek, 682 F. Supp. 3d at 924–27 (same); Nat’l Ass’n for Gun Rights v. Lamont, 685 F. Supp. 3d 63, 104–07 (D. Conn. 2023) (same); cf. Bianchi, 111 F.4th at 463 (concluding that the “more nuanced approach” applies to a challenge to a ban on assault weapons). Mass shootings are clearly a societal concern that arose only in the 20th century. The first known mass shooting in the United States resulting in ten or more deaths occurred in

DUNCAN V. BONTA 39

1949. Ocean State, 95 F.4th at 44. In the three decades that followed, two such shootings occurred. In 2009 alone, three mass shootings claimed ten or more lives, and many more such shootings occurred in the years that followed. See, e.g., id. at 44 n.4 (“The record suggests that mass shootings have become more frequent and more deadly.”). In other words, not a single mass shooting occurred until the middle of the 20th century, and we now experience these events regularly. See Kotek, 682 F. Supp. 3d at 897–99 (detailing the rise of mass shootings and their increasing lethality, particularly when a large-capacity magazine is used). These tragedies naturally receive significant media attention, and they have spawned legislative action and citizen initiatives. It is hard to imagine a clearer example of an “unprecedented societal concern[].” Bruen, 597 U.S. at 27; see Hanson, 120 F.4th at 241 (“There can be little doubt that mass shootings are an unprecedented societal concern.”); Bianchi, 111 F.4th at 463 (“The ripples of fear reverberating throughout our nation in the wake of the horrific mass shootings . . . stem from a crisis unheard of and likely unimaginable at the founding.”). Large-capacity magazines, when attached to a semi- automatic firearm, also represent a dramatic technological change from the weapons at the Founding. Semi-automatic firearms equipped with large-capacity magazines fire with an accuracy, speed, and capacity that differ completely from the accuracy, speed, and capacity of firearms from earlier generations. Hanson, 120 F.4th at 242, 248–51; Ocean State, 95 F.4th at 44. The single-shot, muzzle-loading firearms common at the Founding required slow, manual reloading. Repeating Henry and Winchester rifles, which became popular in the decades following the Civil War, required a shooter to pump a lever manually, a process that allowed about one shot per three seconds—much slower than the firing rate of a modern semi-automatic firearm. Hanson, 671 F. Supp. 3d at 18–19. Notably, California’s law exempts magazines designed for lever-action rifles. Cal. Penal Code § 16740(c). We have no difficulty in concluding that large- capacity magazines designed for semi-automatic firearms represent a “dramatic technological change[].” Bruen, 597 U.S. at 27; see also Bianchi, 111 F.4th at 464 (“These are not our forebears’ arms.”). For all of those reasons, we hold that a more nuanced approach applies here. At the same time, we emphasize that the result in this case does not hinge on this categorization. Because we reach the same result under Rahimi’s straightforward approach, we apply that approach below. But our conclusion is buttressed by the Supreme Court’s reservation of a more flexible analogical approach for “unprecedented societal concerns or dramatic technological changes.” Bruen, 597 U.S. at 27. Inexplicably, the principal dissenting opinion claims that we contrived a “more nuanced approach” ourselves, Dissent by J. Bumatay at 112–13, when in fact all we have done is to quote the Supreme Court’s opinion. See Bruen, 597 U.S. at 27 (“[C]ases implicating unprecedented societal concerns or dramatic technological changes may require a more nuanced approach.”). Nor is our understanding of Supreme Court precedent an outlier. As just noted, we follow essentially every circuit-court and district-court decision in holding that a more nuanced approach applies in the circumstances. The dissenting opinion also faults us for mentioning, but not utilizing, that approach. Dissent by J. Bumatay at 116. But, because the Court did not flesh out how the “more nuanced approach” operates—for instance, whether more recent analogies should be consulted—we have taken the most

DUNCAN V. BONTA 41

conservative path in our analysis by declining to apply the more nuanced approach. b. Three Historical Legal Regimes Are Relevant. Beginning before the Founding and continuing throughout the Nation’s history, legislatures have enacted laws to protect innocent persons from especially dangerous uses of weapons once those perils have become clear. We discern three distinct categories of laws relevant to our analysis. The first category of laws regulated the storage of gunpowder. Early in the Nation’s history, gunpowder was necessary to shoot a firearm. Kotek, 682 F. Supp. 3d at 903. But the storage of gunpowder increased the risk of explosions or fires, which posed an obvious threat to innocent persons. Id. To mitigate the danger to innocent lives, several colonies and states enacted laws restricting the storage of gunpowder. E.g., 1771-1772 Mass. Province Laws 167, ch. 9; 1772 N.Y. Laws 682, ch. 1549; 1782-1783 Mass. Acts 119–20, ch. 46; 1784 N.Y. Laws 627, ch. 28; 1786 N.H. Laws 383–84; 1798–1813 R.I. Pub. Laws 85, § 2; 1806 Ky. Acts 122, § 3; 1821 Me. Laws 98, ch. 25, § 1; 1825 N.H. Laws 74, ch. 61; 1832 Conn. Acts 391, ch. 25; 1836 Conn. Acts 105, ch. 1, § 20. The laws typically prohibited certain methods of storing gunpowder or restricted the amount of gunpowder that could be stored in one place. One Massachusetts law banned taking a firearm loaded with gunpowder into any house or building in Boston. 1782 Mass. Acts 119, ch. 46. Cities and towns, too, enacted laws requiring that gunpowder be stored in certain containers or on the highest story of the home. Act of Apr. 13, 1784, ch. 28, 1784 N.Y. Laws p. 627; An Act for Erecting the Town of Carlisle, in the County of Cumberland, into a Borough, ch. XIV, § XLII, 1782 Pa. Laws p. 41; An Act for Erecting the Town of Reading, in the County of Berks, into a Borough, ch. LXXVI, § XLII, 1783 Pa. Laws p. 140. In 1803, for example, Boston required that all gunpowder be stored in powder houses. 1801 Mass. Acts 507, ch. 20. In sum, legislatures in the Founding era sought to prevent a specific, infrequent type of harm to innocent persons—fires and explosions from the storage of gunpowder—by significantly restricting how and where persons could store gunpowder. The second set of laws concerns trap guns—the rigging of a firearm to discharge when a person unwittingly trips a string or wire. People typically set trap guns to defend their businesses, homes, or possessions. When trap guns became popular, some people applauded their use as a deterrent to crime. But others lamented that trap guns inevitably harmed or killed innocent persons. Legislatures responded by banning the use of firearms as trap guns. In 1771, the legislature of the Colony of New Jersey found that “a most dangerous Method of setting Guns has too much prevailed in this Province” and criminalized the setting of trap guns. 1763–1775 N.J. Laws 346, ch. 539, § 10. Other jurisdictions followed suit. Nine states banned the setting of trap guns in the 18th and 19th centuries, and nine more states enacted bans early in the 20th century. In sum, the use of firearms as trap guns posed a threat, albeit an infrequent threat, to innocent persons, and legislatures responded by prohibiting what had proved to be an especially dangerous use of firearms: the setting of trap guns. The third, and most extensive, set of laws consists of significant restrictions on weapons after their use by

DUNCAN V. BONTA 43

criminals exposed an especially dangerous use of the weapon. These laws date from long before the Founding and continue to today. The Statute of Northampton prohibited the carrying of lances, for example, because those weapons generally were appropriate only to engage in lawful combat “or—as most early violations of the Statute show—to breach the peace.” Bruen, 597 U.S. at 41. In the New World, states and colonies prohibited the concealed carry of some weapons in response to criminals’ secretly carrying those weapons to achieve illicit ends. For example, after a period of strife between planters and the colony’s proprietors, the Colony of East New Jersey prohibited the concealed carry of “pocket pistol[s], skeins, stilladers, daggers or dirks, or other unusual or unlawful weapons.” Grants, Concessions, and Original Constitutions of the Province of New Jersey, 289–90 (1881). Massachusetts enacted similar laws, both as a colony and as an early state. See 1750 Mass. Acts 544, ch. 17 § 1 (prohibiting the carry of “clubs and other weapons” in a group of twelve or more persons); 1786 Mass. Acts 87, ch. 38 (prohibiting being armed with a club or other weapon while rioting); see also Bevis v. City of Naperville, 657 F. Supp. 3d 1052, 1070 (N.D. Ill.) (describing similar restrictions), aff’d, 85 F.4th 1175 (7th Cir. 2023). As new weapons gained popularity and notoriety for their criminal use, legislatures imposed significant restrictions on them. For example, in 1827, Jim Bowie used a large, specially designed knife in a duel. “Bowie knives,” as they came to be called, proliferated in the 1830s. The knives had design features that were particularly suitable for fighting: long blades, crossguards to protect the wielder’s hands, and points designed to make it easier to harm a victim. Criminals used them widely in fights and duels. States acted quickly. By 1840, at least five States or territories had enacted laws restricting the carrying of Bowie knives or other fighting knives. By the end of the 19th century, nearly every State had enacted laws restricting Bowie knives, including by outlawing their concealed or unconcealed carry and sale, by enhancing criminal penalties for their use, or by taxing their ownership. The “slungshot” followed a similar trajectory. A slungshot is a hand-held impact weapon with a weighted object at the end of a flexible strap. Invented in the 1840s, its use by criminals and street gangs became widespread. States again reacted. In 1849, New York and Vermont prohibited the manufacture, sale, or carry of slungshots, punishable by up to five years’ imprisonment. 1849 N.Y. Laws 403, §§ 1-2, ch. 278; 1849 Vt. Acts & Resolves 26, No. 36 §§ 1-2. By the end of the century, nearly every State had enacted anti-slungshot laws. States also responded to advances in firearm technology. In the first half of the 19th century, percussion-cap pistols and revolvers allowed pistols to remain loaded for longer and to contain up to six bullets. Criminals used such pistols with increasing frequency to resolve interpersonal disputes. Several states responded by restricting the carry of concealable pistols. E.g., 1871 Tenn. Pub. Acts 81, An Act to Preserve the Peace and to Prevent Homicide, ch. 90, § 1; 1881 Ark. Acts 191, An Act to Preserve the Public Peace and Prevent Crime, chap. XCVI, § 1-2. A clear pattern emerges from a review of this third set of regulations. 6 When criminals took advantage of

6 Although we choose to stop our survey of historical laws in the 19th century, we note that, as technological advances continued to increase

DUNCAN V. BONTA 45

technological advances in weapons, legislatures acted to restrict an especially dangerous use of those weapons: Bowie knives were designed to—and did—cause significant harm in fights, with little self-defense value, so legislatures banned their carry outside the home; the slungshot proved incredibly useful to criminals but of minimal value in self- defense, so legislatures banned their carry outside the home; and pistols became easy for criminals to conceal, to the detriment of public safety, so legislatures banned their concealed carry. Legislatures sought to prevent a specific type of harm to innocent persons—criminal use of new technology—by prohibiting what had proved to be especially dangerous uses of the new weapons, primarily the ability to carry an extremely deadly weapon concealed from law enforcement and bystanders. c. California’s Ban on Large-Capacity Magazines Falls Within the Nation’s Tradition of Firearm Regulation. We discern two distinct traditions from the legal regimes described above. First, the Founding-era gunpowder- storage regulations established an early tradition of laws seeking to protect innocent persons from infrequent but devastating harm by regulating a component necessary to the firing of a firearm. Second, since the Founding era, legislatures have enacted laws to protect innocent persons from especially dangerous uses of weapons once those perils have become clear. See Hanson, 120 F.4th at 237–38

the lethality of firearms, legislatures throughout the Nation acted to restrict significantly a range of weapons, including sawed-off shotguns, “Tommy guns,” semiautomatic weapons, and automatic weapons. See, e.g., Hanson, 120 F.4th at 239 (describing some of these weapons and the resulting federal regulations); Kotek, 682 F. Supp. 3d at 909–11 (describing many of the weapons and resulting regulations by state legislatures and by Congress). (recognizing the tradition of regulating “weapons that are particularly capable of unprecedented lethality”); Ocean State, 95 F.4th at 46 (recognizing the tradition of regulating dangerous aspects of weapons “once their popularity in the hands of murderers became apparent”); Bevis, 85 F.4th at 1199 (describing “the long-standing tradition of regulating the especially dangerous weapons of the time”); see also Bianchi, 111 F.4th at 464–72 (extensively discussing “a strong tradition of regulating those weapons that were invented for offensive purposes and were ultimately proven to pose exceptional dangers to innocent civilians”). 7 Whether we view those traditions independently or together, California’s ban on large-capacity magazines falls well within them. As described below, the ban is “‘relevantly similar’” to the historical laws in both “why and how it burdens the Second Amendment right.” 8 Rahimi, 602 U.S.

7 The district court in Kotek reached the same conclusion after reviewing essentially the same regulations that we consider: Throughout this Nation’s history, new technologies have led to the creation of particularly dangerous weapons. As those weapons became more common, they became tied with violence and criminality. In response, governments passed laws that sought to address the features of those weapons that made them particularly dangerous to public safety. 682 F. Supp. 3d at 935; see also Lamont, 685 F. Supp. 3d at 71 (reaching the same conclusion because, “when a modern innovation in firearm technology results in a particular type of weapon or method of carrying being utilized for unlawful purposes to terrorize and endanger the public, the Nation has a longstanding history and tradition of regulating those aspects of the weapons or manners of carry that correlate with rising firearm violence”). 8 Our analysis below dutifully compares the “why” for the modern law with the “why” for the historical laws and compares the “how” for the

DUNCAN V. BONTA 47

at 698 (quoting Bruen, 597 U.S. at 29). California’s law thus “is consistent with the principles that underpin our regulatory tradition.” Id. at 692. We first examine the tradition set by the gunpowder- storage laws. Both those early laws and California’s modern law share the same justification for burdening the right to armed self-defense: to protect innocent persons from infrequent but devastating harm caused or exacerbated by a component necessary to the firing of a firearm. Gunpowder caused fires and explosions only infrequently, but legislatures nevertheless recognized that those infrequent events could cause devastating harm. The legislatures therefore imposed significant restrictions on how and where gunpowder could be stored, even though gunpowder was a necessary component to the firing of a firearm. The same justification underpins California’s restriction on magazine capacity: to protect innocent persons from infrequent but devastating harm. Mass shootings (at least as defined by shootings resulting in more than ten deaths) are, thankfully, not everyday occurrences. But there is no dispute that mass shootings, when they occur, cause devastating harm. Mass shootings are devastating for the entire community, and large-capacity magazines exacerbate the harm. The short pauses when a shooter must reload a firearm afford intended victims and law enforcement officers a precious opportunity to flee, take cover, and fight back. Ocean State, 95 F.4th at 47; ANJRPC, 910 F.3d at

modern law with the “how” for the historical laws. To the extent that the principal dissent reaches a different conclusion from that analysis, we respectfully disagree. But we are baffled by the dissent’s further suggestion that we somehow “compare the ‘how’ to the ‘why.’” Dissent by J. Bumatay at 119. 119–20 & n.24; Kolbe, 849 F.3d at 128; Kotek, 682 F. Supp. 3d at 898–99; Platkin, 2024 WL 3585580, at *25. Large- capacity magazines exacerbate the harm caused by mass shootings, and limiting magazine capacity thus prevents or mitigates the harm caused by mass shootings. In sum, both the Founding-era legislatures and California’s legislature and voters enacted measures that burdened the Second Amendment right in order to prevent or mitigate a known, albeit infrequent, cause of devastating harm. See Ocean State, 95 F.4th at 49 (discussing the relevance of gunpowder- storage laws). The gunpowder-storage laws and California’s law are also “relevantly similar” in how they burden the right to armed self-defense. Rahimi, 602 U.S. at 698 (quoting Bruen, 597 U.S. at 29). Both the historical laws and California’s law target the component that causes or exacerbates the devastating harm, and both affect the speed at which a person can fire a firearm. Founding-era legislatures regulated the storage of gunpowder because that component caused or exacerbated fires and explosions. Similarly, California’s legislature and voters regulated the capacity of magazines because, as discussed above, that component causes or exacerbates mass shootings. Both laws also impose a size restriction commensurate with the threat to public safety. Some gunpowder-storage laws imposed a limit on the quantity of gunpowder that could be stored in order to limit the harm caused by a fire or explosion; and California’s law limits the quantity of bullets that may be placed in a magazine in order to limit the harm caused by a mass shooting. Perhaps most pertinently to the analysis here, the laws also plainly affect the speed at which a person could fire a firearm. Laws requiring that gunpowder be stored on the

DUNCAN V. BONTA 49

highest floor of a home, for example, clearly delayed the ability of a resident to respond with a firearm to a ground- floor intruder or to an incident in the street or yard. Towns prohibiting altogether the storage of gunpowder had a much greater effect on the ability of a person to use a firearm with speed; a person wishing to fire a firearm had to retrieve gunpowder from a powder house or a location outside of town. California’s law, too, affects a person’s ability to respond with speed. California’s law permits a person to fire ten rounds without pause but, before firing the eleventh round, the shooter must pause to reload the magazine, use a second magazine, or use a second firearm. Viewed through this lens, California’s law has a significantly smaller effect on the speed of armed self-defense. Gunpowder-storage laws could impose delays of minutes before a person could fire a single shot, whereas California’s law imposes a delay of only seconds and only after a person has fired up to ten rounds. In conclusion, California’s law falls within the national historical tradition of regulating a component necessary to the firing of a firearm in order to prevent or mitigate devastating harm caused or exacerbated by that component. See Ocean State, 95 F.4th at 49 (“It requires no fancy to conclude that those same founding-era communities [that enacted gunpowder-storage laws] may well have responded to today’s unprecedented concern about [large-capacity magazine] use just as the Rhode Island General Assembly did: by limiting the number of bullets that could be held in a single magazine.”). We next turn to the national historical tradition of laws represented by trap-gun bans and restrictions on Bowie knives, slungshots, and concealable pistols. Legislatures throughout our Nation’s history have banned especially dangerous uses of weapons once the threat to innocent persons has become clear. 9 Both the historical laws and California’s law share the same justification: to protect innocent persons from harm from especially dangerous uses of weapons. Legislatures recognized the threat from the use of a firearm as a trap gun by banning that particular, especially dangerous use. Similarly, as criminals increasingly used specific weapons by carrying them outside the home or by concealing them, legislatures banned those particular, especially dangerous uses of the weapons. California’s law, too, reflects the growing threat to public safety from the use of large-capacity magazines. As described above, large-capacity magazines cause or exacerbate the harm from mass shootings, and limiting magazine capacity prevents or mitigates the harm from those events because a shooter must pause before firing an eleventh round. California’s law bans a particular, especially dangerous use of firearms—the use of a large- capacity magazine—in order to protect public health. See, e.g., Hanson, 120 F.4th at 240 (concluding that the historical laws and a ban on large-capacity magazines “share the same basic purpose: To inhibit then unprecedentedly lethal criminal activity by restricting or banning weapons that are particularly susceptible to, and were widely used for, multiple homicides and mass injuries”); DSSA, 664 F. Supp. 3d at 603 (concluding that historical regulations were

9 The principal dissent repeatedly mischaracterizes the relevant tradition as a ban on “especially dangerous weapons.” Dissent by J. Bumatay at 79, 98–99, 118, 119 (several times), 120 (twice). As we emphasize repeatedly, the relevant tradition identified here is a tradition of banning especially dangerous uses of weapons once the perils of those uses becomes clear. By mischaracterizing the relevant tradition, the principal dissent spends much effort attacking a nonexistent assertion.

DUNCAN V. BONTA 51

relevantly similar as to the “why” because those regulations, and a ban on large-capacity magazines, “were enacted in response to pressing public safety concerns regarding weapons determined to be dangerous”). We next examine “how” the regulations burden the Second Amendment right, by considering whether California’s law and the historical regulations “impose a comparable burden on the right of armed self-defense.” Bruen, 597 U.S. at 29. With respect to armed self-defense, the only effect of California’s ban on large-capacity magazines is that a person may fire a semi-automatic weapon no more than ten times without a short pause to change magazines (or reload the original magazine or fire a different weapon). In other words, the law prohibits only one very specific use of some firearms: the shooting of an eleventh (or successive) round without a brief pause. 10 The law imposes no limit whatsoever on the number of magazines a person may own, the number of bullets a person may own, or the number of firearms a person may own. The law also imposes no limit on the number of rounds a person may fire or the number of firearms a person may fire. Nor, despite Plaintiffs’ protestations to the contrary, does the law ban any weapon. A person wishing to buy any lawful

10 Like other courts, we reject the suggestion that a person “uses” a large- capacity magazine even when not firing the weapon. Ocean State, 95 F.4th at 45 n.8; Hanson, 671 F. Supp. 3d at 15. Whereas brandishing a firearm can have a deterrent effect on would-be attackers, there is no evidence in the record that others can tell that a magazine attached to an unfired weapon is large-capacity or that such magazines would provide any additional deterrent effect. See Ocean State, 95 F.4th at 45 n.8 (“[The] plaintiffs claim no plausible scenario in which a threat has proved less effective because the brandished weapon could only fire ten rounds at once without reloading.”). firearm (or other weapon) is free to do so. The owner may possess that firearm at home for self-defense and, so far as this law is concerned, may carry it in any manner, and to any place, that state law allows. The burden imposed by California’s law is comparable to the burden imposed by the historical laws. Each of those laws, like California’s law, burdened the right to armed self- defense by prohibiting a specific, particularly dangerous use of a weapon. Like California’s law, the trap-gun laws allowed persons to use firearms in every way except one way that proved, at the relevant time, particularly dangerous to innocent persons: the rigging of trap guns. Similarly, various laws throughout the Nation’s history regulated the right to armed self-defense by prohibiting specific uses of weapons that had proved particularly dangerous: we have noted the examples of concealed carry, carry more generally, carry of weapons in a group of twelve or more persons, and carry of a weapon while rioting. Those laws, too, prohibited the use of weapons in ways that had proved, in their time, especially dangerous. Legislatures throughout the Nation’s history thus have chosen, as California’s legislature and voters have chosen, to impose a confined regulation of armed self-defense by prohibiting a specific, especially dangerous use of a weapon. 11

11 When considering “how” the legislature has burdened the right to armed self-defense, we understand the Supreme Court’s focus to be about the method of burdening the right, rather than the magnitude of the burden. We therefore describe in text how California’s law and the historical laws burden the right by prohibiting a specific, particularly dangerous use of a weapon. To the extent that the magnitude of the burden is relevant, California’s law imposes only a minimal burden on the right of armed self-defense. Firing more than ten rounds occurs only rarely, if ever, in armed self-defense. The evidence in this record, and in

DUNCAN V. BONTA 53

Plaintiffs’ primary argument to the contrary is that California’s law imposes a materially different burden on armed self-defense than the historical laws because, unlike those laws, California’s law prohibits the possession of large-capacity magazines. We disagree both with Plaintiffs’ reasoning and with their premise. Most fundamentally, the burden imposed by the modern law need not be an exact match to the burden imposed by historical laws. California’s law “is by no means identical to” the relevant historical laws, “but it does not need to be.” Rahimi, 602 U.S. at 698. Its prohibition on a weapon’s component that serves the sole function of enabling a specific, and especially dangerous, use of a firearm fits

other cases, demonstrates that a person seldom shoots more than ten rounds when defending with a firearm. Hanson, 120 F.4th at 245; Ocean State, 95 F.4th at 45; Worman v. Healey, 922 F.3d 26, 37 (1st Cir. 2019), abrogated in other part by Bruen, 597 U.S. 1; Kolbe, 849 F.3d at 127; Capen, 708 F. Supp. 3d at 91; Kotek, 682 F. Supp. 3d at 896–97. And even in those extremely rare instances, the record does not disclose whether the shooter fired more than ten bullets in rapid succession, without a short pause that would have allowed reloading or switching weapons. In sum, California’s law places a limited burden on the Second Amendment’s right to armed self-defense by prohibiting only one specific and rare use of semi-automatic firearms that accept detachable magazines—the firing of more than ten rounds without a short pause after the tenth round. The magnitude of that burden is relevantly similar to the magnitude of the burden of the historical laws. The trap-gun laws undoubtedly burdened the right to armed self-defense in specific situations. A person worried about a nighttime intruder had to remain awake and alert and could not rely on a trap gun. And the historical restrictions on carrying weapons imposed an even greater burden: A person skilled at using a concealable pistol, Bowie knife, or slungshot could not depend on those weapons when leaving home. neatly within the tradition that the historical regulations represent. Rahimi is instructive. Section 922(g)(8) prohibits persons subject to a domestic-violence restraining order from possessing any firearm. Id. at 688–89. The surety laws, as Justice Thomas pointed out, imposed no burden whatsoever on armed self-defense; they required merely posting a bond and, in the case of a breach, forfeiting the bond. Id. at 764–67 (Thomas, J., dissenting). Similarly, the going-armed laws burdened armed self-defense only after a jury trial, with all its attendant protections for the accused, and only for public breaches of the peace, whereas the modern law burdened armed self-defense with few procedural protections and for the altogether different conduct of a private threat. Id. at 768–71. The Rahimi majority rejected those differences as insignificant, explaining that, “[a]s we said in Bruen, a ‘historical twin’ is not required.” Id. at 701 (quoting Bruen, 597 U.S. at 30). California’s law fits comfortably into historical tradition. Even accepting Plaintiffs’ premise—which we do not, for the reasons described below—that the burden imposed by California’s law is different in some way than the burdens imposed by the historical laws, a historical twin is not required. Like the historical laws, California’s law functionally prohibits a particular, especially dangerous use of a weapon. That law is “relevantly similar” to the historical laws. Id. at 698 (quoting Bruen, 597 U.S. at 29). Moreover, the regulation of a component necessary to the firing of a firearm is far from unprecedented. As described above, the Founding generation, through gunpowder-storage laws, imposed significant restrictions on a component absolutely necessary at the time to the firing of a firearm.

DUNCAN V. BONTA 55

To the extent that California’s law differs meaningfully—again, a premise that we reject—any difference is precisely because of the factors that Bruen mentioned. 597 U.S. at 27. The voters of California determined that modern experience had shown that laws short of bans on possession had been ineffective and nearly impossible to enforce, and mass shootings are a uniquely modern phenomenon resulting from dramatic improvements in technology. Prop. 63 §§ 2(12), 3(8). It is inconsistent with the Supreme Court’s instructions to reason that, because a ban on the carry of Bowie knives, for example, may have proved sufficient to mitigate criminal use of Bowie knives, legislatures and citizen initiatives are limited to precisely the same restrictions when addressing new technology that enables a new, more devastating type of societal harm. To the contrary, the Supreme Court has cautioned us that its “precedents were not meant to suggest a law trapped in amber.” Rahimi, 602 U.S. at 691; see also id. at 739–40 (Barrett, J., concurring) (“[I]mposing a test that demands overly specific analogues has serious problems. To name two: It forces 21st-century regulations to follow late- 18th-century policy choices, giving us ‘a law trapped in amber.’ And it assumes that Founding-era legislatures maximally exercised their power to regulate, thereby adopting a ‘use it or lose it’ view of legislative authority. Such assumptions are flawed, and originalism does not require them.” (citation omitted)). Moreover, Plaintiffs’ premise is mistaken; California’s law bans no weapon. We reiterate that a large-capacity magazine is not itself a weapon. It is an accessory whose sole function is to provide some firearms with the ability to fire more than ten rounds without a short pause. In this way, California’s law imposes less of a burden than the historical laws, which regulated weapons themselves. A person may carry a firearm freely, but Bowie knives and slungshots must be left at home. In sum, as other courts have concluded with respect to similar bans on large-capacity magazines, we conclude that California’s law is “relevantly similar” to historical regulations in “how” it burdens the right to armed self- defense. Rahimi, 602 U.S. at 698; Hanson, 120 F.4th at 237– 38 (discussing historical regulations of the Bowie knife); Ocean State, 95 F.4th at 48 (same); Bevis, 85 F.4th at 1201 (same); Platkin, 2024 WL 3585580, at *20, *24 (same); DSSA, 664 F. Supp. 3d at 600–01 (describing historical regulations of the Bowie knife, slungshot, and revolver pistol); Kotek, 682 F. Supp. 3d at 903–10, 928–33 (describing many of the same historical regulations we discuss and explaining why the burden they imposed was relevantly similar to the burden imposed by Oregon’s ban on large-capacity magazines); Bevis, 657 F. Supp. 3d at 1068– 71 (describing historical regulations of the Bowie knife, slungshot, and trap gun); see also Bianchi, 111 F.4th at 464– 68 (describing historical regulations of gunpowder, the Bowie knife, dirk, sword cane, metal knuckles, slungshot, and sand club and concluding that a ban on assault weapons is relevantly similar to those historical regulations). In conclusion, California’s law is relevantly similar to the historical laws in both why and how it burdens the right to armed self-defense, and it falls within the national historical tradition of regulating a particular, especially dangerous use of a weapon, once that use becomes a specific threat to innocent persons.

DUNCAN V. BONTA 57

d. We Reject Plaintiffs’ Ownership-Statistics Argument. In Heller, the Supreme Court overturned a District of Columbia ban on all handguns because it “amounts to a prohibition of an entire class of ‘arms’ that is overwhelmingly chosen by American society for th[e] lawful purpose” of self-defense. 554 U.S. at 628. Plaintiffs argue that California’s law falls into the same category because it, too, is “a prohibition of an entire class of ‘arms’ that is overwhelmingly chosen by American society for that lawful purpose.” Id. After all, Plaintiffs point out, many firearm owners own large-capacity magazines. According to Plaintiffs, “that should be the end of the analysis.” We reject that simplistic approach. For the reasons described in Part A, above, large- capacity magazines are not “arms.” Even assuming the contrary, large-capacity magazines remain optional, and they remain an accessory to some firearms. Accepting Plaintiffs’ argument would require concluding that the Second Amendment never permits a legislature to ban any optional accessory to any weapon, provided that enough people purchased enough of them before a legislature could act. We do not read the Supreme Court’s precedents in that rigid manner. Instead, we take at face value the instruction that a modern law “must comport with the principles underlying the Second Amendment, but it need not be a ‘dead ringer’ or a ‘historical twin.’” Rahimi, 602 U.S. at 692 (quoting Bruen, 597 U.S. at 30). Consider, for example, machine guns. Federal law prohibits their possession except in very limited circumstances. 18 U.S.C. § 922(o). The Supreme Court stated in Heller that it would be “startling” to read the Second Amendment as prohibiting a ban on machine guns, and the Court clearly signaled that one machine gun—the M-16— could be banned. 554 U.S. at 624, 627. It is estimated that civilians own more than 176,000 machine guns (and nearly one million machine guns exist in the country, when one counts those owned by law enforcement officers). DSSA, 664 F. Supp. 3d at 592. Plaintiffs do not explain why, under their ownership-statistics theory, 176,000 is insufficient while the somewhat larger, but unknown, number of large- capacity magazines suffices. 12 Moreover, if Congress chose to let the ban on machine guns expire, as Congress did with respect to large-capacity magazines, and if civilians purchased more machine guns, would a state-law ban on machine guns suddenly change from constitutional to unconstitutional? How many more would civilians have to buy before that binary change took effect? We do not read the Constitution or the Supreme Court’s precedents to hinge on the necessarily speculative answers to those questions. Instead, we must ask whether Plaintiffs’ proposed conduct falls within the text of the

12 In addition to the uncertainty about the total number of large-capacity magazines owned by civilians, we also do not know how many were truly “chosen by American society for th[e] lawful purpose [of self- defense].” Heller, 554 U.S. at 628. Among other questions, we do not know how many purchases were made simply because the large-capacity magazine comes as standard equipment with the purchase of a firearm. See, e.g., Duncan V, 19 F.4th at 1126–27 (Berzon, J., concurring) (“[A] device may become popular because of marketing decisions made by manufacturers that limit the available choices. Here, for example, large- capacity magazines come as a standard part on many models of firearms, so a consumer who wants to buy those models has no choice regarding whether the weapon will include a magazine that can fire more than ten rounds without reloading.”). Moreover, there is no constitutional right to factory settings.

DUNCAN V. BONTA 59

Second Amendment and, if so, whether the law “is consistent with the principles that underpin our regulatory tradition.” Rahimi, 602 U.S. at 692 (citing Bruen, 597 U.S. at 26–31). Heller addressed a true ban on a class of common firearms, including all uses of those weapons, and must be understood in that light. California’s law, by contrast, bans only one type of optional accessory to some firearms— functionally prohibiting only one specific use that is rarely, if ever, used in self-defense. The Supreme Court’s recent decisions in Bruen and Rahimi have described at length the approach that we must apply when assessing the constitutionality of modern firearm regulations. We reject Plaintiffs’ facile invitation to jettison that approach and hold that, any time an undefined number of people own an undefined number of any optional accessory to any weapon, no legislature may ban that accessory, no matter how rarely that accessory is used in armed self-defense. See Hanson, 120 F.4th at 233–34 (explaining why a simplistic ownership- statistics argument conflicts with Bruen); Ocean State, 95 F.4th at 50–51 (detailing at length many flaws in the same ownership-statistics argument); Bevis, 85 F.4th at 1190 (discussing a flaw in the argument); DSSA, 664 F. Supp. 3d at 592–93 (rejecting the same ownership-statistics argument); see also Bianchi, 111 F.4th at 459–61 (discussing similar flaws in the same ownership-statistics argument for assault weapons). e. California’s Law Fits the Nation’s Tradition. In sum, even assuming that Plaintiffs’ proposed conduct of possessing large-capacity magazines implicates the plain text of the Second Amendment, California’s law fits within the Nation’s tradition of regulating weapons. Accordingly, Plaintiffs’ Second Amendment challenge fails for this second, alternative reason. REVERSED AND REMANDED with the instruction to enter judgment in favor of Defendant.

BERZON, Circuit Judge, with whom MURGUIA, Chief Judge, and HURWITZ, PAEZ, SR THOMAS, and WARDLAW, Circuit Judges, join, concurring:

I concur in full in the majority opinion. Here, I address Judge VanDyke’s novel form of “dissent.” Judge VanDyke’s dissent improperly relies on factual material that is unquestionably outside of the record. See Majority Op. at 13–14 n.1. His source for these beyond-the record facts? A video that he recorded, in his own chambers, showing him handling several different handguns and explaining his understanding of their mechanics and operation. I write separately to point out two fundamental problems with Judge VanDyke’s reliance on his self-made video: First, the video is not part of his written dissent and it includes facts outside the record, so the panel is right to ignore it. Second, and more egregiously, Judge VanDyke has in essence appointed himself as an expert witness in this case, providing a factual presentation with the express aim of convincing the readers of his view of the facts without complying with any of the procedural safeguards that usually apply to experts and their testimony, while simultaneously serving on the panel deciding the case. While the facts Judge VanDyke asserts must be ignored, his wildly improper video presentation warrants additional comment, lest the genre proliferate.

DUNCAN V. BONTA 61

I. Judge VanDyke’s video is, in his words, a “visual illustration” meant to “aid [his] colleagues and the parties.” Dissent at 125. The “amateur” majority, he writes, bases its analysis on “a nonexistent reality.” Dissent at 128, 146. So Judge VanDyke (who, he suggests, does understand these matters) purports to “show[] that this reality doesn’t exist” by explaining “how guns are made, sold, used, and commonly modified”—all facts, despite Judge VanDyke’s protestations to the contrary. Dissent at 146; Lawrence VanDyke, Video, at 5:09-13 (March 20, 2025), https://www.ca9.uscourts.gov/media/23-55805/opinion. Judge VanDyke asserts that the video, to which his opinion includes a link, is “part” of his dissent, “deliver[ed] . . . orally—via video” rather than in writing. Dissent at 125. But the video is not “part” of his dissent simply because Judge VanDyke says it is. To the contrary, our circuit’s general orders require that “the determination of each appeal . . . shall be evidenced by a written disposition.” 9th Cir. Gen. Order 4.5a (emphasis added). 1 Our rules do not allow a video to operate as a “disposition,” a term that includes separate opinions. 2

1 Rules and statutes similarly require that decisions by federal district judges be memorialized in writing. For example, the Federal Rules of Civil Procedure contemplate oral decisions when district judges rule on motions or make findings and conclusions in bench trials, but the rules require that any decisions not evidenced by a written opinion or memorandum “be stated on the record” in open court. See Fed. R. Civ. P. 52(a)(1), (a)(3). And by statute, all district court sessions must “be recorded verbatim” to enable transcription. 28 U.S.C. § 753(b). 2 General Order 4.5a refers to the “disposition” and the “majority disposition” as shorthand for “written disposition,” including when it Of course, in bygone days, before computers, typewriters, cameras, or microphones, judges delivered decisions orally from the bench. Erwin C. Surrency, Law Reports in the United States, 25 Am. J. Legal Hist. 48, 55 (1981). But it’s not the 1750s anymore. We have long since moved past an “unwritten” system “retained . . . by memory and custom.” 1 William Blackstone, Commentaries *63 (1765). States began requiring courts to reduce opinions to writing as early as 1784. See, e.g., Acts and Laws of the State of Connecticut in America 267–68 (1784) (Act of May 2, 1784). In this circuit, written decisions have been part of our practice since our founding in 1891. See, e.g., United States v. Sutton, 47 F. 129 (9th Cir. 1891). So today, we ground our jurisprudence in written precedent. See Peter Tiersma, The Textualization of Precedent, 82 Notre Dame L. Rev. 1187 (2013). And we do so for good reason: Written opinions promote uniformity, predictability, accountability, and care. “[W]riting seems to have the advantage of inducing greater care,” as the Scottish judge Henry Cockburn noted; “Men don’t boggle at speaking nonsense which they would hesitate to put permanently down upon paper.” 2 Journal of Henry Cockburn 154 (Edinburgh, Edmonston & Douglas 1874). True, times have changed: New technologies might today make it easier to preserve and distribute oral opinions than in centuries past. But even in an age of online videos, written opinions are more clear, useful, and accessible, and

refers to “[a]ny separate concurring or dissenting disposition.” The General Order thus treats the term “disposition” as referring exclusively to written dispositions.

DUNCAN V. BONTA 63

there are many potential challenges with video dispositions. 3 In any event, our entire legal system has long since evolved to one built around written precedent. And in this circuit, our rules require it. Perhaps our written-disposition rule should be reevaluated in light of new technology. But we have a clearly defined process for considering such changes; that process has yet to be invoked toward that end. Judge VanDyke’s video is not, then, technically speaking, “part” of his dissent. So what is it? It is not a video that is itself part of the record. Cf., e.g., Norse v. City of Santa Cruz, 629 F.3d 966, 979 (9th Cir. 2010) (Kozinski, J., concurring). It is not a recording of an oral argument or some other court proceeding. 4 Cf., e.g., Cadena v. Customer Connexx LLC, 107 F. 4th 902, 914–15 n.11 (9th Cir. 2024). And it is not a link to a video offered as a citation in support of the judicially noticeable fact that the video exists and is available on the internet. Cf., e.g., Brown v. Ent. Merchs. Ass’n, 564 U.S. 786, 856 (2011) (Breyer, J., dissenting). Each of these reasons for linking to a video is well- established and permissible. But Judge VanDyke links to his video for a different purpose—to provide facts not in the record, by demonstration. The video itself is not part of the

3 For example, where would videos be hosted for later access? How could their contents be searched and cross-referenced? Would courts and reporters be able to ensure that the recordings remain available indefinitely? Would pro se litigants and prisoners have access to the videos online? Would all the videos be transcribed? If so, how would transcripts capture non-verbal signals like tone and facial expressions, let alone visual demonstrations like Judge VanDyke’s? 4 Judge VanDyke’s video contains a short excerpt from the oral argument in this case. I do not take issue with that portion of his recording. record, and the facts asserted therein are not subject to judicial notice. “Save in unusual circumstances, we consider only the district court record on appeal.” Lowry v. Barnhart, 329 F.3d 1019, 1024 (9th Cir. 2003); see also Fed. R. App. P. 10. There are a few narrow exceptions—for example, to correct inadvertent omissions, to take judicial notice when appropriate, or if new factual developments have rendered a case moot. Lowry, 329 F.3d at 1024. I would have thought the proposition obvious, but it apparently bears emphasis that a judge wanting to provide an unrequested “visual illustration” to help his “colleagues and the parties” understand the case, Dissent at 125, is not one of these established exceptions. Not even close. 5 The majority opinion thus considers the video no differently than it would any other factual source that is not in the record and not subject to judicial notice or another of the narrow exceptions: it ignores it.

II.

But there is another, even more troublesome problem with the recording: Judge VanDyke himself appears in the recorded presentation making factual assertions about how guns work and providing physical demonstrations to support his assertions. By doing so, Judge VanDyke casts himself in the role of an expert witness, speaking to the type of “technical” and “specialized” issues that are reserved for witnesses properly “qualified as an expert.” See Fed. R. Evid. 701, 702. He catalogues according to his own criteria

5 If an attorney (rather than a judge) submits and relies on material outside the record, we have held the conduct sanctionable. See Lowry, 329 F.3d at 1025–26.

DUNCAN V. BONTA 65

various handgun components, describes their functions, and provides a physical demonstration about how they are attached to a firearm and replaced. He opines on the relative merits of different types of takedown levers, lighter versus heavier grips, and iron sights versus red dot optics. He repeatedly avers that certain features make “the gun more dangerous when it’s misused, but also make[] the gun more effective for its intended purpose.” Lawrence VanDyke, Video, at 11:11-11:31; 10:01-10:16; 13:07-13:21 (March 20, 2025), https://www.ca9.uscourts.gov/media/23- 55805/opinion. And he speaks to his understanding as to whether certain components are factory standard or frequently used. All these topics are commonly the province of expert testimony. See, e.g., United States v. Spinner, 152 F.3d 950, 957 (D.C. Cir. 1998) (expert testimony about technical features of an AR-15 including its grip); United States v. Meadows, 91 F.3d 851, 853–54 (7th Cir. 1996) (expert testimony about the components and characteristics of a gun that was converted into a rifle); Oregon Firearms Fed’n v. Kotek, 682 F. Supp. 3d 874, 912, 922 (D. Or. 2023) (expert testimony about the mechanics of magazines and the relative merits of high-capacity magazines); Nat’l Ass’n for Gun Rts. v. Lamont, 685 F. Supp. 3d 63, 72–75 (D. Conn. 2023) (expert declaration about the mechanics of automatic and semiautomatic firearms); United States v. Hasson, No. 19- 96, 2019 WL 4573424, at *2 (D. Md. Sept. 20, 2019) (expert testimony about the features of silencers and their interaction with firearms); Barnett v. Raoul, No. 23-CV-00141, 2024 WL 4728375, at *36 (S.D. Ill. Nov. 8, 2024) (expert testimony about the merits of features like “the ability to fire semiautomatically, detachable magazines, pistol grips, forward-protruding grips, thumbhole stocks, adjustable stocks, flash suppressors, barrel shrouds, buffer tubes/braces, and threaded barrels” in facilitating self- defense). Myriad rules govern the submission and presentation of expert testimony, all of which Judge VanDyke has bypassed by introducing his factual testimony on appeal and alongside his dissent. First, expert testimony must be found to “have a reliable basis in the knowledge and experience of [the expert’s] discipline” and be “more than subjective belief or unsupported speculation.” Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 590, 592 (1993). Should a party wish to challenge the reliability of an expert’s conclusions, they may move to exclude the expert’s testimony. District courts then have an obligation to make an explicit finding as to an expert’s reliability before permitting an expert to testify. United States v. Jawara, 474 F.3d 565, 583 (9th Cir. 2007). This gatekeeping function is vital “to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 152 (1999). Second, parties may contest an expert’s admission or move to limit the scope of their testimony on the grounds that the expert is insufficiently qualified to opine on specific topics. See Fed. R. Evid. 702 (requiring that an expert witness be “qualified . . . by knowledge, skill, experience, training, or education”). Further, because expert testimony is subject to general evidentiary rules, litigants may also seek to exclude testimony on the basis that it is irrelevant, unduly prejudicial, or otherwise objectionable under the Federal Rules of Evidence. See, e.g., Fed. R. Evid. 401, 403.

DUNCAN V. BONTA 67

Third, various facets of expert testimony must be disclosed during discovery to ensure that the litigants are on notice of what evidence might be provided. These disclosures include the identity of expert witnesses, the subject of their testimony, and a summary of all of the facts or opinions they intend to provide. Fed. R. Civ. P. 26(a)(2). If the witness was specifically retained to provide testimony, they must also provide a written report disclosing, among other things, “a complete statement of all opinions the witness will express and the basis and reasons for them,” “the facts or data considered by the witness in forming them,” “any exhibits that will be used to summarize or support them,” and “the witness’s qualifications.” Fed. R. Civ. P. 26(a)(2)(B). 6 Additionally, a litigant then has the right to “depose any person who has been identified as an expert whose opinions may be presented at trial.” Fed. R. Civ. P. 26(b)(4)(A). These provisions ensure that litigants are aware of the expert evidence that might be introduced, allowing them to marshal objections, responses, or opposing expert testimony. Aside from these mechanisms of exclusion and notice, litigants have avenues for rebutting the contents of expert testimony. They may cross-examine experts at trial or during a deposition to challenge questionable aspects of their testimony, qualifications, or methodology. See Fed. R. Evid. 611(b) (scope of cross-examination); id. 705 (disclosure of expert witness’s underlying facts or data on cross- examination). They may introduce facts that undercut the expert’s credibility. See Fed. R. Evid. 607, 608. And they

6 There are additional rules governing the timeliness of such disclosures and requiring that they be updated with pertinent changes as the litigation progresses. Fed. R. Civ. P. 26(a)(2)(B), (D). can proffer competing evidence, often in the form of an opposing expert witness. None of those avenues are available when the “expert testimony” appears for the first time in an appellate opinion. The procedures governing testimony generally and expert testimony in particular are of paramount importance to our adversarial system. Basic fairness demands that parties have the opportunity to challenge the admissibility of expert testimony, cross-examine expert witnesses, and introduce countervailing evidence. Evidentiary and procedural rules allowing parties to robustly analyze and challenge expert witness testimony is critical to ensuring factfinders are able to assess the veracity and reliability of the technical evidence before them. Court evidentiary rules, including the rules governing expert testimony, have evolved over centuries to meet these purposes, adjusted over time and interpreted in judicial opinions as gaps in coverage appeared. See, e.g., Daubert, 509 U.S. 579. Permitting anyone—including a judge—to interject their observations and opinions on technical factual issues without abiding by these carefully developed rules for presentation of expert testimony defangs these procedural safeguards and severely disadvantages the litigants. Additional rules constrain the presentation of demonstratives such as the firearms and related items that Judge VanDyke features in his recording. Demonstratives may be excluded if the court finds they are not relevant to the matter at hand. See, e.g., United States v. Ortiz-Martinez, 1 F.3d 662, 669 (8th Cir. 1993); United States v. Jones, 222 F.3d 349, 351 (7th Cir. 2000). A sufficient foundation must be laid to verify that the demonstrative evidence is a fair and accurate representation of what the witness claims it to be. See, e.g., Keller v. United States, 38 F.3d 16, 32 n.10 (1st

DUNCAN V. BONTA 69

Cir. 1994); United States v. Myers, 972 F.2d 1566, 1579 (11th Cir. 1992). Such demonstratives are excluded if the court finds that their use would risk causing undue prejudice, confusion, or delay. See Fed. R. Evid. 107, 403. Barring exclusion, opposing counsel may address any inaccuracies with the demonstrative through cross-examination. See, e.g., Roland v. Langlois, 945 F.2d 956, 963 (7th Cir. 1991) (affirming use of a life-sized model where it “was admitted only on the express condition that the jury be alerted to the perceived inaccuracies”); Krause v. Cnty. of Mohave, 459 F. Supp. 3d 1258, 1272–73 (D. Ariz. 2020) (explaining that an illustrative animation’s “purported errors or inaccuracy can be sufficiently addressed through jury instruction and cross- examination”). Such demonstratives are also often subject to disclosure requirements during discovery, to ensure the parties are aware of what evidence might be presented. See Fed. R. Civ. P. 26(a)(2)(B)(iii) (requiring expert reports to include “any exhibits that will be used to summarize or support” the expert witness’s opinions); id. 26(a)(3)(A)(iii) (requiring disclosure of “each document or other exhibit, including summaries of other evidence—separately identifying those items the party expects to offer and those it may offer if the need arises”). In providing a physical illustration of how various handgun components function and how to replace them, Judge VanDyke presents demonstratives that have not been subject to the vetting procedures normally afforded such evidence. Like Judge VanDyke’s procedurally infirm factually based descriptions, commentary, and opinions, the use of these demonstratives affronts party presentation principles and flouts the rules that govern the introduction of evidence. Judge VanDyke presents his factual assertions not only in the wrong court, but also with no regard to the well- established requirements governing expert testimony and demonstrative evidence. Moreover, his presentation is doubly concerning because, had he provided his expert testimony in the district court with all of the required safeguards, he almost certainly would have needed to recuse from adjudicating this matter on appeal. This recusal issue most frequently arises in the context of district court judges, who are flatly prohibited from serving as both arbiter and witness. See Fed. R. Evid. 605 (“The presiding judge may not testify as a witness at the trial.”); see also Quercia v. United States, 289 U.S. 466, 470 (1933) (explaining that a presiding judge “may analyze and dissect the evidence, but he may not either distort it or add to it”); United States v. Berber-Tinoco, 510 F.3d 1083, 1091 (9th Cir. 2007) (“[T]he judge may not actually testify in the proceeding or interject facts. . . .”). Moreover, “[a]ny justice, judge, or magistrate judge of the United States shall disqualify himself in any proceeding in which his impartiality might reasonably be questioned.” 28 U.S.C. § 455(a). A reasonable observer would be fair to question the impartiality of an appellate judge who served as an expert witness in the proceedings before the district court. Here, Judge VanDyke attempts to have it both ways: providing a factual presentation with the express aim of convincing both the parties and the panel of the truth of his assertions, while also remaining a member of the panel adjudicating the merits of the case. That Judge VanDyke presents his factual commentary on appeal, as opposed to as an expert witness in the proper forum, exacerbates rather than cures the impropriety of the presentation. The form is

DUNCAN V. BONTA 71

no different than it would have been had Judge VanDyke sought to appear as an expert witness in district court, yet the procedural protections the litigants would have had in district court are totally absent. Contrast Judge VanDyke’s approach with that of the majority. The defendants raised the question whether high- capacity magazines may be considered an “accoutrement” rather than an “arm” before the district court. Expert opinions on that question were properly introduced before the district court, are part of the record on appeal, and are relied upon in the majority opinion. For example, in analyzing the historical distinction between “arms” and “accoutrements,” the majority draws from factual material presented by an expert on corpus linguistics. Majority Op. at 26. Likewise, the majority relies on the declaration of a firearms expert for the facts underlying its conclusion that high-capacity magazines are accoutrements rather than arms. Majority Op. at 28. The plaintiffs also introduced some pertinent evidence on this issue. One of the plaintiffs’ experts opined that magazines should not be considered accessories because many firearms cannot function as intended without a magazine during the district court proceedings. Another gave a similar opinion based on his understanding of magazine mechanics. Instead of relying on the factual material introduced by the parties in the proper forum subject to the applicable procedural rules, Judge VanDyke presents his own testimony on the matter. He attempts to justify his eleventh- hour factual interjections by asserting a need to refute the “factual fantasy” underlying the majority’s test. But, as I’ve noted, the majority’s test relied on factual submissions on the accoutrement issue in the district court, record material that Judge VanDyke ignores. Had plaintiffs wished to proffer additional expert testimony, akin to that which Judge VanDyke presents to support their argument against classifying high-capacity magazines as accessories, they had ample opportunity to do so. “In our adversarial system of adjudication, we follow the principle of party presentation.” United States v. Sineneng- Smith, 590 U.S. 371, 375 (2020). “[W]e rely on the parties to frame the issues for decision and assign to courts the role of neutral arbiter of matters the parties present.” Id. (quoting Greenlaw v. United States, 554 U.S. 237, 243 (2008)). In acting as a self-appointed expert, Judge VanDyke oversteps his role as arbiter and robs the parties of their prerogative to develop the record as they see fit, as well as the many procedural protections to which they are entitled. * * * Judge VanDyke might well be able to qualify as an expert on guns. But whatever specialized expertise we bring to the bench is irrelevant to our role as judges. Our job is not to provide the facts that support our conclusions but to apply the law to the facts as presented by the parties. Judge VanDyke’s factual presentation flips this foundational principle on its head. The majority is right to ignore the contents of Judge VanDyke’s video presentation. These outside-the-record assertions of fact, made by someone not properly admitted as an expert or subject to any of the many procedural safeguards that govern expert and demonstrative testimony, have no bearing on the proper disposition of this appeal. And while Judge VanDyke accuses the majority of “blinding itself” to the facts he presented in his video, Dissent at 143,

DUNCAN V. BONTA 73

limiting review to the facts presented by the parties is precisely what appellate courts are required to do. See Fed. R. App. P. 10. True, the prejudice to the parties here is arguably minimal because Judge VanDyke has prepared his video in support of a dissent. But if a dissent can rely on a judge’s recorded factual presentation, nothing prevents a majority opinion from doing the same thing. I therefore write separately in the hope that in the future my colleagues, whether in the majority or dissent, will do exactly and only that: write. And, although I am surprised that it is necessary to do so, I write to reemphasize that as judges, we must decide cases as they are presented to us by the parties, leaving advocacy to the attorneys and testimony to the witnesses, expert and otherwise.

R. NELSON, Circuit Judge, dissenting:

I agree with Judge Bumatay that the majority’s decision to reverse the district court on the merits flouts New York State Rifle & Pistol Association, Inc. v. Bruen, 597 U.S. 1 (2022). But the majority didn’t just butcher the Second Amendment and give a judicial middle finger to the Supreme Court. It also spurned statutory procedure for en banc proceedings. As explained in my dissent from the order filed concurrently with this opinion, this en banc court lacks statutory jurisdiction to decide this new appeal, years after it remanded the prior appeal to the district court. See R. Nelson Dissent to Order at 69–76; Moody v. Albemarle Paper Co., 417 U.S. 622, 627 (1974) (per curiam) (quoting 28 U.S.C. § 46(c)). Every other circuit applies § 46(c) to require a new en banc vote in a new appeal. As Judge Ikuta points out, the majority’s view “is not the best interpretation of this statute, in light of the history and purpose of an en banc consideration.” Ikuta Special Concurrence to Order at 56– 57. Nor is the majority’s retention of this case business as usual. See R. Nelson Dissent to Order at 59–60, 81–89. Never before has a court allowed five senior judges to control an en banc decision on behalf of the court’s active judges. We shouldn’t have been the first, let alone in such an important case. As Judge Bumatay notes, the majority’s decision was a misuse even of our own rules. See Bumatay Dissent to Order at 102. The result is a precedential Second Amendment opinion that largely reflects the views of five senior judges, not the active judges statutorily entrusted with “determin[ing] the major doctrinal trends of the future” for our court. Moody, 417 U.S. at 626 (quoting United States v. Am.-Foreign S. S. Corp., 363 U.S. 685, 690 (1960)). That does not bode well for public confidence in the courts—and our court specifically.

BUMATAY, Circuit Judge, with whom IKUTA, R. NELSON, and VANDYKE, Circuit Judges, join, dissenting:

California is at the forefront of States seeking to limit their citizens’ firearm rights. The State says its firearms laws are the strictest in the Nation. 1 From an assault-weapons ban to red-flag laws, from waiting periods to age restrictions and universal background checks, California’s gun-control

1 Official Website of the State of California, FACT SHEET: California’s strong gun safety laws continue to save lives, (Jun. 7, 2024), available at: https://perma.cc/W75M-U4GX.

DUNCAN V. BONTA 75

measures significantly reduce its citizens’ access to firearms. California believes these restrictions cut firearms-related violence. 2 As a sovereign State, California may of course do what it thinks proper to protect its citizens. But what California may not do is encroach on its citizens’ constitutional rights. We cannot ignore that California’s actions continually whittle away the Second Amendment guarantee. While California may pass laws to address gun violence, California’s choices must give way to the Constitution. And this is true no matter how well-meaning the gun control measures. It is easy to sympathize with the litany of tragedies that California cites to justify its regulatory actions. We understand California’s concern for needless gun violence. We recognize California’s desire to curtail mass shootings. And we appreciate that many will vehemently disagree on policy grounds with enforcing the constitutional limits on California’s gun-control measures. But, as federal judges, our duty is to uphold the Constitution—no matter how unpopular. After all, the right to keep and bear arms is a “fundamental right necessary to our system of ordered liberty.” McDonald v. City of Chicago, 561 U.S. 742, 778 (2010) (plurality opinion). And through incorporation under the Fourteenth Amendment, the Constitution authorizes only state regulations “consistent with the Second Amendment’s text and historical understanding.” N.Y. State Rifle & Pistol Assoc. v. Bruen, 597 U.S. 1, 26 (2022). So while California seeks to limit its citizens’ access to firearms—even with the best of intentions—it is our duty to ensure that the Second Amendment endures and that this ancient right of the people

2 Id. is given its fullest breadth. Otherwise, we abdicate our role to the whims of the political majority of the State. At issue is California’s ban on the possession of firearm magazines capable of accepting more than ten rounds of ammunition. See Cal. Penal Code §§ 32310, 16740. California prohibits the sale and purchase of these magazines, id. § 32310(a); it punishes the possession of the magazines with a fine and up to one year of imprisonment, id. § 32310(c); and it requires those who possessed the magazines before the ban to remove, sell, or surrender them, id. § 32310(d). California calls these magazines “large-capacity magazines.” That term suggests that their capacities are greater than the usual magazine. But, in truth, magazines holding more than ten rounds are the most common magazines in the country. They come standard with the most popular firearms sold nationwide. As the district court observed, “in the realm of firearms,” these magazines “are possibly the most commonly owned thing in America.” Duncan v. Bonta, 695 F. Supp. 3d 1206, 1214 (S.D. Cal. 2023) (“Duncan VIII”). By the most conservative estimates, more than a hundred million “large-capacity” magazines exist in the country today. To put it into perspective, if California’s law applied nationwide, then half of all magazines in the United States would be taken from nearly 40 million Americans. And so these magazines should be more accurately termed “standard-capacity magazines.” Simply put, the ban deprives Californians of the most popular firearm magazines for self-defense and other lawful purposes. That’s what’s at stake for California’s citizens. Whatever the moniker, California’s absolute ban on magazines with more than ten rounds of ammunition is both

DUNCAN V. BONTA 77

“unusual” and an “outlier.” See Bruen, 597 U.S. at 79 (Kavanaugh, J., concurring). They are lawfully possessed in at least 38 States 3 and under Federal law. And California’s law is novel. Aside from some Prohibition-era laws, 4 before 1990, only the District of Columbia restricted law-abiding citizens’ possession of feeding devices of any size.

3 Only twelve States and the District of Columbia ban the outright possession of magazines with more than a certain number of rounds. See California (Cal. Penal Code § 32310); Colorado (Colo. Rev. Stat. §§ 18- 12-301, 18-12-302) (limits magazine capacity to 15); Connecticut (Conn. Gen. Stat. § 53-202w); Delaware (Del. Code tit. 11, §§ 1468, 1469) (limits magazine capacity to 17); the District of Columbia (D.C. Code § 7-2506.01); Hawaii (Haw. Rev. Stat. § 134-8) (handguns only); Illinois (720 Ill. Comp. Stat. 5/24-1.10) (10 rounds for “long guns” and 15 rounds for handguns); Massachusetts (Mass. Gen. Laws Ch. 140, §§ 121, 131M); New Jersey (N.J. Stat. §§ 2C:39-1, 2C:39-3); New York (N.Y. Penal Law § 265.00, 265.02); Oregon (Or. Rev. Stat. § 166.355); Rhode Island (11 R.I. Gen. Laws §§ 11-47.1-2, 11-47.1-3); and Vermont (Vt. Stat. tit. 13, § 4021) (10 rounds for “long guns” and 15 rounds for handguns). Two other States forbid the manufacture and sale of these magazines. See Maryland (Md. Code, Crim. Law § 4-305); and Washington (Wash. Rev. Code §§ 9.41.010, 9.41.370). 4 Michigan, Rhode Island, and Minnesota enacted laws limiting magazine capacity in the 1920s and 1930s, but those laws were repealed or amended and do not serve as a ban on magazine capacity today. See Act of June 2, 1927, No. 373, § 3, 1927 Mich. Pub. Acts 887, 888 (repealed 1959); Act of Apr. 22, 1927, ch. 1052, §§ 1, 4, 1927 R.I. Acts & Resolves 256, 256–57 (amended 1959); 1933 Minn. Laws ch. 190 (amended 1963). Only the District of Columbia’s 1932 restriction of magazine capacity remains operative. See Act of July 8, 1932, Pub. L. No. 72-275, §§ 1, 8, 47 Stat. 650, 650, 652. While it appears Virginia may have defined “machine gun” to encompass guns able to fire sixteen bullets without reloading, see ch. 96, §§ 1–7, 1934 Va. Acts 137–139, Virginia has since amended the definition and the law no longer includes this language. Va. Code § 18.2-288. California’s possession ban just went into effect in 2017, and three States enacted their laws in the last three years. 5 Nothing in the historical understanding of the Second Amendment warrants California’s magazine ban. Even with some latitude in searching for historical analogues, none exist. California points to no historical laws banning the possession of firearms commonly used for self-defense. Other traditional laws—like laws against carrying certain weapons in public, laws against possessing particular weapons, gunpowder storage laws, and laws against setting trap guns—don’t remotely resemble the “how” and “why” of California’s magazine bans. Thus, neither the text of the Second Amendment nor our country’s historical tradition of firearm regulation supports California’s magazine ban. Still, the majority once again upholds California’s regulation. In doing so, the majority defies the Supreme Court. First, the majority takes the extreme position that the Second Amendment doesn’t even apply to California’s magazine ban. That’s because the majority rules that magazines holding more than ten rounds are not even “Arms” under the Second Amendment. So California’s law receives no constitutional scrutiny at all. And it bases this ruling on the flimsiest ground. Somehow, the majority believes that a “large-capacity magazine is no different than” a “belt[] that hold[s] bullets.” Maj. Op. 28–29. Such a belief displays ignorance of both firearms operations and constitutional law. Indeed, the majority bases its “Arms” analysis, in part, on a portion of a district court opinion unanimously reversed by the D.C. Circuit. See id. at 23

5 Oregon (2022), Rhode Island (2022), and Illinois (2023). See note 3.

DUNCAN V. BONTA 79

(citing Hanson v. District of Columbia, 671 F. Supp. 3d 1, 14 (D.D.C. 2023), rev’d in part, 120 F.4th 223, 232 (D.C. Cir. 2024)). So not surprisingly, no other circuit court has gone as far as the Ninth Circuit has in declaring that some magazines fall outside the scope of the Second Amendment. Second, the majority makes up a new two-test Bruen framework—the so-called “more nuanced approach” and the “straightforward,” unnuanced approach. Id. at 37–41. While giving little explanation for its invention, under the “more nuanced approach,” the majority claims that there is no need to look to historical analogues whenever we are dealing with technological or societal change because governments are simply entitled to “more flexib[ility]” in limiting the Second Amendment right. Id. at 37. And under its “straightforward” Bruen test, the majority interest- balances its way into concluding that California’s ban conforms with historical analogues. It draws the broadest generalities from California’s claimed historical analogues—fashioning a dubious tradition against “especially dangerous” weapons. This supposed tradition is no different from just requiring the State to provide a public- safety rationale. But this ignores the admonition that “court[s] must be careful not to read a principle at such a high level of generality that it waters down the right.” United States v. Rahimi, 602 U.S. 680, 740 (2024) (Barrett, J., concurring). Ignoring this limitation, the majority then finds a supposed match between that purported tradition and California’s ban by improperly minimizing its burden on self-defense. Under either of the majority’s created tests, we’ve returned to interest balancing by another name. So yet again, the majority continues to reject the Supreme Court’s Second Amendment jurisprudence. We have repeatedly sounded the alarm over the affront to the Second Amendment here. See Duncan v. Bonta, 19 F.4th 1087, 1140 (9th Cir. 2021) (en banc) (“Duncan V”) (Bumatay, J., dissenting); Duncan v. Bonta, 83 F.4th 803, 808 (9th Cir. 2023) (en banc) (“Duncan IX”) (Bumatay, J., dissenting). 6 Over the course of this litigation, the majority has taken at least three positions on how California’s novel ban should be upheld as constitutional. So this is now the third time we’ve had to warn against the majority’s violation of Supreme Court instructions. We sound the alarm yet again—but this time, it’s more dire given the extreme nature of the majority’s ruling. Its implications are vast and lead to a dangerous expansion of government power. In contrast, if our analysis here sounds familiar, it is. Our position has remained the same from the start of this litigation. Adhering to the Second Amendment’s text and historical understanding, California’s magazine ban is unconstitutional. Because the majority’s ruling again stands in opposition to the Second Amendment’s text, history, and tradition, we respectfully dissent. I.

THE SECOND AMENDMENT FRAMEWORK

The Second Amendment commands that “[a] well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.” U.S. Const. amend. II. This right flows from the inherent right of self-defense—as Blackstone said, it was central to “the natural right” to “self-preservation and

6 This case has been up and down the federal courts so often that this is the fourth decision by our en banc court in this saga.

DUNCAN V. BONTA 81

defence.” District of Columbia v. Heller, 554 U.S. 570, 593– 94 (2008) (quoting 1 William Blackstone, Commentaries on the Laws of England 139–40 (1765)). Though the Second Amendment guarantees a preexisting right, it took until Heller for the Supreme Court to recognize its rightful place as a fundamental protection of liberty. In Heller, the Supreme Court examined the Amendment’s “text and history” and concluded that it “conferred an individual right to keep and bear arms.” Id. at 595. Although “not unlimited,” at its core, the Second Amendment protects the right of “law-abiding citizens” to keep and carry arms for the “lawful purpose of self-defense.” Id. at 595, 630, 635. This guarantee is at its strongest when “arms” are used “in defense of hearth and home.” Id. at 635. And because the right is so “deeply rooted in this Nation’s history and tradition,” it is “fully applicable to the States” under the Fourteenth Amendment. McDonald, 561 U.S. at 750, 768 (simplified). Repeatedly, the Court has rejected “freestanding interest-balancing” to resolve Second Amendment questions. Heller, 554 U.S. at 634 (simplified); see also McDonald, 561 U.S. at 785 (rejecting “judicial interest balancing”). Despite the Court’s clear teachings, in Heller’s wake, many courts—including our own—created two-step, interest-balancing tests that inevitably stripped away the Second Amendment right. See, e.g., Young v. Hawaii, 992 F.3d 765, 783–84 (9th Cir. 2021) (summarizing our two-step process and collecting cases), cert. granted, judgment vacated, 142 S. Ct. 2895 (2022). As we’ve said, this two- step approach proved to be mere “window dressing for judicial policymaking.” Duncan V, 19 F.4th at 1148 (Bumatay, J., dissenting). At least in our court, no state regulation had ever been ruled unconstitutional before Bruen. See id. at 1165 (VanDyke, J., dissenting) (observing the Second Amendment’s 0-for-50 record in the Ninth Circuit). Then came Bruen. Having had enough of the balancing, Bruen established a new framework—one consistent with the text of the Constitution. Bruen expressly declared that the “Constitution demands” that we jettison the use of “interest balancing.” 597 U.S. at 26. Instead, it returned us to the “Second Amendment’s text and historical understanding.” Id. First, we look to the Second Amendment’s textual elements: Does “the Second Amendment’s plain text cover[] an individual’s conduct[?]” Id. at 17. If so, “the Constitution presumptively protects that conduct.” Id. Second, to rebut that presumption, the government bears the burden of “justify[ing] its regulation by demonstrating that it is consistent with the Nation’s historical tradition of firearm regulation.” Id. at 24. Put simply, the government “must affirmatively prove that its firearms regulation is part of the historical tradition that delimits the outer bounds of the right to keep and bear arms.” Id. at 19. If the government fails to do this, the regulation does not pass “constitutional muster.” Id. at 30. How do we know if a regulation fits in the historical tradition? Like judges do in every case, we use analogical reasoning. But rather than ask how other federal judges have conducted the interest-balancing calculus, we apply “the balance struck by the founding generation to modern circumstances.” Id. at 29 n.7. That means the government must “identify a well-established and representative historical analogue,” and our task is to decide whether the regulation and historical analogues are “relevantly similar.” Id. at 29, 30 (simplified). Of course, this “analogical reasoning . . . is neither a regulatory straightjacket nor a

DUNCAN V. BONTA 83

regulatory blank check”—some regulations will stand, some will fall. Id. at 30. But the key is finding a relevantly similar historical analogue. Analogues need not be “twin[s]” or “dead ringer[s].” Id. Instead, Bruen tells us that our “central” consideration is whether the modern regulation “impose[s] a comparable burden on the right of armed self-defense” and whether “that burden is comparably justified.” Id. at 29. In other words, we must compare the “how” and the “why” of the government’s regulation with the reported historical analogue. Id. In this way, the only means-ends scrutiny involved is the “interest balancing by the people” who ratified the Second Amendment. Heller, 554 U.S. at 635. In conducting this inquiry, we shouldn’t be overly rigid—we don’t have to look for “a law trapped in amber.” Rahimi, 602 U.S. at 691. Instead, we look to “whether the challenged regulation is consistent with the principles that underpin our regulatory tradition.” Id. at 692 (emphasis added). So what’s required is that the modern regulation “works in the same way and does so for the same reasons” as historical regulations. Id. at 711 (Gorsuch, J., concurring). Doing this faithfully may reveal guiding principles defining the scope of the right. At the same time, we must also “be careful not to read a principle at such a high level of generality that it waters down the right.” Id. at 740 (Barrett, J., concurring). After all, extracting principles too far removed from the historical record just returns us to the interest-balancing regime the Court has authoritatively repudiated. And “[h]istory, not policy, is the proper guide.” Id. at 717 (Kavanaugh, J., concurring). CALIFORNIA’S MAGAZINE BAN IS

UNCONSTITUTIONAL

With this framework in mind, we turn to California’s magazine ban. First, we examine whether the Second Amendment’s “plain text” covers the conduct that California regulates. Bruen, 597 U.S. at 17. It does. And because California’s magazine ban fits within the Amendment’s textual elements, it’s “presumptively” unconstitutional. Id. at 24. Second, we evaluate whether California has overcome the presumption of unconstitutionality. It must do so by justifying its law under our “historical tradition of firearms regulation[s].” Id. It hasn’t. Thus, California’s magazine ban is unconstitutional. A. California’s Magazine Ban is Presumptively Unconstitutional California’s ban prohibits people from owning, possessing, purchasing, or selling any magazine holding more than ten rounds, and forces people to surrender magazines already in circulation to the government. See Cal. Penal Code § 32310. As a result, California’s ban infringes on the “right of the people” to “keep” and “bear” magazines. Bruen, 597 U.S. at 32 (explaining that “bear” “naturally encompasses” “carrying handguns publicly for self-defense” and, at a minimum, “keep” means the possession of “firearms in the[] home, at the ready for self-defense”). Thus, California’s magazine ban easily fits into the Second Amendment’s “textual elements,” which makes it presumptively unconstitutional. Rather than accept this obvious conclusion, the majority contends that these

DUNCAN V. BONTA 85

magazines are not even “Arms” under the Second Amendment. That’s wrong. 1. Magazines Are “Arms” Magazines—whether they hold ten rounds, more than ten rounds, or fewer than ten rounds—are unquestionably “Arms” under the Second Amendment. As a textual matter, “Arms” include any “[w]eapons of offence, or armour of defence” or “any thing that a man wears for his defence, or takes into his hands, or useth in wrath to cast at or strike another,” that is “carr[ied] . . . for the purpose of offensive or defensive action.” Heller, 554 U.S. at 581, 584 (simplified). Arms, then, “extends, prima facie, to all instruments that constitute bearable arms, even those that were not in existence at the time of the founding.” Id. at 582. At a minimum, the meaning of Arms “covers modern instruments that facilitate armed self-defense.” Bruen, 597 U.S. at 28. California’s ban applies to instruments that are necessary to the operation of most modern firearms for self-defense. A magazine is a “container or (detachable) receptacle in a repeating rifle, machine-gun, etc., containing a supply of cartridges which are fed automatically to the breech.” Oxford English Dictionary Online (2024) (emphasis added). As that definition makes clear, the magazine is a part of the firearm. It stores and continuously feeds cartridges into the firearm’s chamber. Firearms with detachable magazines— the most commonly owned firearms—simply cannot fire as designed without a magazine. Indeed, some firearms won’t fire a single shot without an attached magazine. And so, when a person uses a firearm in self-defense, the person must operate both the firearm and magazine together. Without the magazine, the firearm would be practically useless for self- defense. Because magazines are a necessary component of using firearms in self-defense, they are integral to a bearable “instrument that facilitate[s] armed self-defense.” Bruen, 597 U.S. at 28. By its nature, the Second Amendment’s protection of “Arms” must extend to their functional components. If magazines and other components weren’t included, the Second Amendment would be a shallow right—easily infringed by basic indirect regulation. But our fundamental rights are more robust than that. Simply put, the government can’t accomplish through “indirect[] infringement” what would otherwise be a “direct interference with fundamental rights.” See Healy v. James, 408 U.S. 169, 183 (1972); see also Nat’l Rifle Ass’n of Am. v. Vullo, 602 U.S. 175, 190 (2024) (“a government official cannot do indirectly what she is barred from doing directly”). Our Constitution thus “implicitly protect[s] . . . closely related acts necessary to the exercise” of enumerated rights. Luis v. United States, 578 U.S. 5, 26 (2016) (Thomas, J., concurring). After all, “[t]here comes a point . . . at which the regulation of action intimately and unavoidably connected with [a right] is a regulation of [the right] itself.” Hill v. Colorado, 530 U.S. 703, 745 (2000) (Scalia, J., dissenting). That’s why the Second Amendment protects “necessary concomitant[s]” to the right to bear arms. N.Y. State Rifle & Pistol Ass’n, Inc. v. City of New York, 590 U.S. 336, 364 (2020) (Alito, J., dissenting). So the Second Amendment’s scope includes corollaries necessary to firearms’ use for self- defense, like the right to learn how to “keep [firearms] ready for their efficient use,” Heller, 554 U.S. at 707 (Breyer, J., dissenting) (simplified); the right to “obtain bullets necessary to use” firearms, Jackson v. City & Cnty. of San Francisco, 746 F.3d 953, 967 (9th Cir. 2014); and the “right

DUNCAN V. BONTA 87

to possess the magazines necessary to render . . . firearms operable,” Fyock v. Sunnyvale, 779 F.3d 991, 998 (9th Cir. 2015). And once properly understood as “Arms,” magazines of all stripes, including those holding more than ten rounds, are protected. After all, the government can’t limit the people to government-preferred types of magazines any more than it can limit the people to government-preferred types of firearms. See Heller, 554 U.S. at 629 (holding that the government can’t forbid handguns because long guns are available). The majority sees things differently. Although the majority begrudgingly concedes that magazines are generally protectable “Arms” and that “large-capacity” magazines “enhance[] . . . a person’s ability . . . to defend” himself, Maj. Op. 29, it nonetheless excludes the “large- capacity” magazine from the protection of the Second Amendment. And it does so using questionable reasoning. Let’s try to explain the majority’s analysis. The majority first draws a constitutional distinction between “Arms” and “accoutrements.” According to the majority, “Arms” and their necessary components are protected by the Second Amendment, but accoutrements or firearm accessories are not. Id. at 26. It then reasons that a magazine by itself isn’t an “Arm[]” protected by the Second Amendment. That’s because magazines are “harmless” when not attached to an accompanying firearm and so they can’t “reasonably” be described as a “weapon[] of offence, or armour of defence.” Id. (quoting Heller, 554 U.S. at 581). Nor is a “large- capacity” magazine a protected component of a firearm, says the majority, because no firearm “requires” a magazine holding more than ten rounds to “function normally.” Id. at 28. Stringing this altogether, the majority claims “large- capacity” magazines are not “Arms” under the Second Amendment. The majority’s analysis is flawed for at least three reasons. First, a magazine is simply not an accoutrement—it’s a necessary firearm component. No antebellum dictionary describes “accoutrement” to include firearm components. Instead, they define “accoutrement” as:

• “Attire, dress, garb, furniture.” Nathan Bailey, An Universal Etymological English Dictionary 21 (London, T. Osborne 1763). • “Dress, equipage, furniture relating to the person; trappings, ornaments.” Samuel Johnson, A Dictionary of the English Language 5 (London, W. Strahan 1773). • “[D]ress, habiliments, particularly after a warlike manner.” Thomas Dyche, A New General English Dictionary 24 (London, C. Bathurst 1777). • “In a military sense, signify habits, equipage, or furniture, of a soldier, such as belts, pouches, cartridge-boxes, saddles, bridles, &c.” William Duane, A Military Dictionary 2–3 (Philadelphia, W. Duane 1810). 7

7 In the military context, equipage “is all kinds of furniture made use of by the army,” such as “tents, kitchen furniture, saddle horses, baggage wagons, bat horses, &c.” See Duane, A Military Dictionary 139.

DUNCAN V. BONTA 89

• “[I]n a military sense, signify habits, equipage, or furniture of a soldier, such as buffs, belts, pouches, cartridge boxes, &c.” Charles James, An Universal Military Dictionary 3 (4th ed. 1816). • “Dress; equipage; trappings; ornaments.” Samuel Johnson & John Walker, Johnson & Walker’s English Dictionaries 63 (Boston, C. Ewer 1828). • “Dress; equipage; furniture for the body; appropriately, military dress and arms; equipage for military service.” Noah Webster, American Dictionary of the English Language xv (New York, S. Converse 1828). • “Equipage; trappings; ornaments.” Joseph Worcester, Primary Dictionary of the English Language 13 (Boston, Swan, Brewer, and Tileston 1861).

No modern magazine would fit within the definition of “accoutrements.” And the size of the magazine wouldn’t matter either. As a necessary component of the functioning of a firearm, it’s not part of the “habits, equipage, or furniture, of a soldier.” Duane, A Military Dictionary 2–3. Under these definitions, magazines, including those holding more than ten rounds, are not “accoutrements.” Instead, they are protected components of “Arms,” like triggers and barrels. Ignoring these facts, the majority merely asserts magazines are no different than “cartridge boxes and belts that hold bullets.” Maj. Op. 28–29. Second, the majority’s faux-Solomonic splitting of magazines based on the number of rounds they hold makes no sense. The majority concedes that magazines holding ten or fewer rounds are perfectly legal, “integral” components of “Arms”—entitled to the Second Amendment’s fullest protection. Id. at 28; see also id. at 27 (“[T]he Second Amendment’s text necessarily encompasses the corollary right to possess a magazine . . . just as it protects the right to possess ammunition and triggers.”). And common sense dictates that just because a magazine holds more than ten rounds doesn’t transform it into an accoutrement. Yet the majority seems to possess magical abilities. See id. at 27. As soon as you add one more round—poof—the magazine is no longer “integral” and it disappears from the Second Amendment’s ambit. Call this the “magic bullet” theory of the Ninth Circuit. So the majority’s “Arms” versus “accoutrements” distinction proves too much. Either a magazine is an “accoutrement,” which States may ban completely under the majority’s theory. Or it’s an “Arm[]” which affords it Second Amendment protection. The majority can’t seem to make up its mind on where magazines fit. The majority won’t—and can’t—go so far to say that magazines may be prohibited outright. But in the next breath it says some magazines aren’t protected at all. This “accoutrement” distinction is thus baffling and unhelpful. In the end, even the majority abandons the distinction as it concedes that some “accessories . . . are necessary for the ordinary operation of a . . . weapon” and fall within the Second Amendment’s protection. Id. at 27. Third, the majority misunderstands the Second Amendment inquiry. According to the majority, “[t]he proper inquiry . . . is whether the component or accessory is

DUNCAN V. BONTA 91

necessary to the ordinary operation of the weapon, not whether, when one voluntarily chooses to use an optional accessory, the accessory is attached to the weapon.” Id. at 29. But that’s wrong. The “relevant test” under the Second Amendment isn’t what’s strictly “necessary” for self- defense. Rather, the Second Amendment inquiry centers on what the people choose to “facilitate armed self-defense.” Bruen, 597 U.S. at 28. As we’ve said before, “[l]awful purpose, not necessity, is the test.” Duncan IX, 83 F.4th at 808 (Bumatay, J., dissenting). Indeed, the Supreme Court has already rejected the majority’s strictly-necessary-for-self-defense theory of the Second Amendment. In Heller, the government argued that its handgun ban was permissible because it allowed citizens to use long guns for self-defense. 554 U.S. at 629. So in the government’s view, handguns weren’t strictly necessary because citizens could defend themselves with other weapons. The Court forcefully rejected that argument,

It is no answer to say . . . that it is permissible to ban the possession of handguns so long as the possession of other firearms (i.e., long guns) is allowed. It is enough to note, as we have observed, that the American people have considered the handgun to be the quintessential self-defense weapon. . . . Whatever the reason, handguns are the most popular weapon chosen by Americans for self-defense in the home, and a complete prohibition of their use is invalid. Id. Thus, the Second Amendment grants citizens the choice of commonly owned arms to protect themselves. The government doesn’t get to decide for the people. And we would never be so parsimonious when it comes to other constitutional rights. Imagine granting only what’s strictly necessary to enjoying the free-speech or free- exercise right. No court would tolerate that. That’s like saying that, as long as the government permits speech through print or the airwaves, it may ban speech on social media platforms because they’re mere “optional accessories” for spreading information. See Meyer v. Grant, 486 U.S. 414, 424 (1988) (“That [people] remain free to employ other means to disseminate their ideas does not take [restrictions on] their speech . . . outside the bounds of First Amendment protection.”). It’s also like saying that the government can ban religious worship at home because it’s not strictly “necessary” when churches and synagogues are available. See Tandon v. Newsom, 593 U.S. 61 (2021). The examples could go on and on. But it bears repeating—the Second Amendment is no “second-class right, subject to an entirely different body of rules than the other Bill of Rights guarantees.” Bruen, 597 U.S. at 70 (simplified). Thus, the majority’s attempt to carve out magazines holding more than ten rounds from the Second Amendment’s protection of “Arms” is wrong both as a matter of firearms operations and constitutional law. 2. Common-Use Question For its part, California argues that Plaintiffs here failed to satisfy their burden of proving that magazines holding more than ten rounds are “in common use” for self-defense, which it argues is part of Bruen’s step-one textual analysis. Plaintiffs counter that the “common use” inquiry only comes into play under Bruen step two and thus it’s California’s burden to prove. Concededly, “Bruen is somewhat ambiguous on this point.” Bianchi v. Brown, 111 F.4th 438,

DUNCAN V. BONTA 93

501 (4th Cir. 2024) (Richardson, J., dissenting). It’s mentioned in both the historical and textual steps of the analysis. See Bruen, 597 U.S. at 32, 46. This question has divided panels of our court. Compare United States v. Alaniz, 69 F.4th 1124, 1128 (9th Cir. 2023) (step one) with Teter v. Lopez, 76 F.4th 938, 949–50 (9th Cir. 2023), vacated and reh’g en banc granted, 93 F.4th 1150 (9th Cir. 2024) (step two). Indeed, during the preliminary injunction stage, before the parties briefed this question, we assumed that the “in common use” inquiry was part of the Second Amendment’s “textual elements.” Duncan IX, 83 F.4th at 810 (Bumatay, J., dissenting). Besides, the question of the common usage of these magazines is undebatable— so, for this case, it didn’t matter which side carried the burden of proof. Even so, after further briefing on the matter, we agree with Plaintiffs “that the ‘common use’ inquiry best fits at Bruen’s second step.” Bianchi, 111 F.4th at 502 (Richardson, J., dissenting). We think this for three reasons. First, Bruen explained that the first step—whether the Second Amendment presumptively protects conduct— comes from determining whether the “plain text” of the Second Amendment covers the conduct at issue. 597 U.S. at 17. And, as a textual matter, nowhere in the text of the Second Amendment does “in common use” appear. See Bevis v. City of Naperville, 85 F.4th 1175, 1209 (7th Cir. 2023) (Brennan, J., dissenting) (The Second Amendment should be “read as ‘Arms’—not ‘Arms in common use at the time.’”). Nor is common usage an inherent part of the definition of “Arm,” which looks only at whether it’s a bearable “[w]eapon[] of offence, or armour of defence.” Heller, 554 U.S. at 581. Conducting the common-use inquiry at the first step “would be at odds with the fact that the common-use test is not about the semantic meaning of the Second Amendment’s plain text.” J. Joel Alicea, Bruen Was Right, 174 U. Pa. L. Rev. (forthcoming 2025) (manuscript at 12). 8 Second, under Bruen’s second step, the Second Amendment permits only firearm regulations “consistent with this Nation’s historical tradition.” 597 U.S. at 17. And whether a firearm is “dangerous and unusual” or “in common use” is borne from the “historical understanding of the Amendment.” Id. at 21. Heller itself directly tied the common-use inquiry to “the historical tradition of prohibiting the carrying of dangerous and unusual weapons.” 554 U.S. at 627 (simplified). Indeed, the “in common use” phrase originates from United States v. Miller, 307 U.S. 175, 179 (1939), in which the Court described the types of weapons colonial citizens would bring to militia service. Third, as a matter of constitutional law, it makes sense that California carries the burden of disproving “common use.” As the Court has repeatedly said, the Second Amendment is a fundamental individual right. Once a plaintiff makes “a prima facie showing of arguable [fundamental-right] infringement,” like in the First Amendment context, the burden shifts to the government to justify the regulation. See Brock v. Loc. 375, Plumbers Intern. Union of Am., AFL-CIO, 860 F.2d 346, 349–50 (9th Cir. 1988) (simplified). In the First Amendment context, the government’s burden is also “extraordinarily heavy.” ACLU of Nev. v. City of Las Vegas, 466 F.3d 784, 791 (9th Cir. 2006). So too here. Once a prima facie showing of a Second

8 Available at https://perma.cc/KV22-25HU.

DUNCAN V. BONTA 95

Amendment infringement occurs (by showing the government regulates conduct within the Amendment’s “textual elements”), everything else falls on the government. Since common usage of a firearm isn’t necessary to prove a prima facie Second Amendment infringement, it’s the government’s burden to carry at the second step. We turn there now. B. California Fails to Overcome the Presumption of Unconstitutionality Because the plain text of the Second Amendment protects the possession of magazines capable of feeding more than ten rounds, California’s ban is presumptively unconstitutional. To rebut this presumption, California must “justify its regulation” by proving that its ban fits within our “historical tradition of firearm regulation.” Rahimi, 602 U.S. at 691 (quoting Bruen, 597 U.S. at 17, 24). California doesn’t meet its burden. California supports its magazine ban based on five broad categories of historical analogues: (1) the prohibition of “dangerous and unusual” weapons; (2) laws regulating the carry of certain weapons; (3) prohibitions on possessing crossbows, slungshots, and automatic firearms; (4) bans on the setting of trap guns; and (5) regulations on the storage of gunpowder. But given that California strictly bans the ownership, possession, and use of magazines in common use today, these traditions are not “relevantly similar” to justify California’s magazine ban. 1. Prohibition of “Dangerous and Unusual” Weapons Start with prohibitions on “dangerous and unusual” weapons. From Miller to Heller to Bruen, the Supreme Court has recognized that the “Second Amendment protects only the carrying of weapons that are those ‘in common use at the time,’ as opposed to those that ‘are highly unusual in society at large.’” Bruen, 597 U.S. at 47 (simplified). As the Court has explained, this understanding is “fairly supported by the historical tradition of prohibiting the carrying of ‘dangerous and unusual weapons.’” Heller, 554 U.S. at 627. Thus, the Court has suggested that “dangerous and unusual” weapons fall outside the Second Amendment’s protection. So if California can prove that these magazines are “highly unusual” today, then that would be enough to satisfy its burden. But the magazines California bans are the furthest thing from highly unusual in modern America. In fact, firearms with magazines holding more than ten rounds are the overwhelming choice of Americans for self-defense and other lawful purposes. While estimates vary, easily more than 100 million of these magazines exist in the country. According to one estimate, these magazines account for half of all American magazines—that’s 115 million out of 230 million magazines in circulation today. See Duncan VIII, 695 F. Supp. 3d at 1217. Another estimate suggests that these magazines are even more prevalent—climbing to a staggering 542 million rifle and handgun magazines in the hands of “millions of Americans across the country.” Id. at 1216–17. Indeed, Plaintiffs assert, and California doesn’t contradict, that nearly 40 million Americans own or have owned magazines with capacity for more than ten rounds— that’s more than 10% of the Nation’s total population and about half of all American gun owners. See William English, Ph.D., 2021 National Firearms Survey: Updated Analysis Including Types of Firearms Owned 22–23 (Sept.

DUNCAN V. BONTA 97

28, 2022). Given this widespread ownership, they are necessarily used for lawful purposes. And this common usage is nothing new. As we’ve said before, a “clear picture emerges [from our history] that firearms with large-capacity capabilities were widely possessed by law-abiding citizens by the time of the Second Amendment’s incorporation.” Duncan V, 19 F.4th at 1155 (Bumatay, J., dissenting). And “large-capacity” magazines had become “common” in this country by at least “the late nineteenth century or early twentieth century.” Id. at 1130 (Berzon, J., concurring). In contrast, regulation of these magazines is unusual and new. As mentioned above, three- quarters of all States have no magazine-capacity limit like California. The federal government’s short-lived experiment with a magazine ban lapsed by 2004. See Pub. L. No. 103-322, 108 Stat. 1796 (1994) (formerly codified at 18 U.S.C. § 922(w)). Aside from D.C.’s law, most state bans were enacted after the 1990s, with many passed in just the last few years. Neither California nor the majority seriously challenge that these magazines are “in common use.” Instead, the majority buries the data. It claims that looking at “ownership[]statistics” is too “simplistic” and disregards reliance on them as too “rigid.” Maj. Op. 57. But, as lower court judges, we are not free to set aside the Supreme Court’s directions so easily. Heller and Bruen were very clear— once a firearm is “in common use,” it falls out of the historical tradition of prohibiting “dangerous and unusual” weapons and is entitled to constitutional protection unless restricted by another tradition. In Heller, the Court held that, because “handguns are the most popular weapon chosen by Americans for self-defense in the home,” they are protected by the Second Amendment and their “complete prohibition” is “invalid.” 554 U.S. at 629; see also id. at 628–29 (“[A] prohibition of an entire class of ‘arms’ that is overwhelmingly chosen by American society for [the] lawful purpose” of self-defense “fail[s] constitutional muster.”). On the other hand, Heller recognized that “the Second Amendment does not protect those weapons not typically possessed by law-abiding citizens for lawful purposes[.]” Id. at 625. Bruen put an even finer point on this question. Assuming that historical laws could show handguns were considered “dangerous and unusual” in our Nation’s past, Bruen said that this history wasn’t dispositive when handguns are “in ‘common use’ for self-defense today.” 597 U.S. at 47. So, “even if [historical] laws prohibited the carrying of handguns because they were considered ‘dangerous and unusual weapons’” sometime in the past, “they provide no justification for laws restricting the public carry of weapons that are unquestionably in common use today.” Id. So the lesson of Heller and Bruen is that the historical tradition of banning “dangerous and unusual” weapons can’t justify regulation of firearms in common use today. And because these magazines are no doubt in common use today, a tradition of banning weapons that “are highly unusual in society at large” can’t support California’s magazine ban. Next, the majority tries a different tack. Again, ignoring the Supreme Court’s instructions, the majority claims that our historical tradition included dispossessing the people of “especially dangerous” weapons—no matter how commonly they were used for self-defense. Maj. Op. 12. Again, that’s wrong. The Court has always grouped “dangerous and

DUNCAN V. BONTA 99

unusual” together. See Heller, 554 U.S. at 627; Bruen, 597 U.S. at 21, 47, 51; see also Caetano v. Massachusetts, 577 U.S. 411, 417 (2016) (Alito, J., concurring) (“[T]his is a conjunctive test: A weapon may not be banned unless it is both dangerous and unusual.”). In other words, whether a weapon is “dangerous and unusual” or “in common use” are different sides of the same coin. See Bruen, 597 U.S. at 47 (contrasting “dangerous and unusual weapons” with those that are “in ‘common use’ for self-defense today”); Heller v. District of Columbia, 670 F.3d 1244, 1272 (D.C. Cir. 2011) (Kavanaugh, J., dissenting) (observing that the Supreme Court “said that ‘dangerous and unusual weapons’ are equivalent to those weapons not ‘in common use’”) (simplified). Indeed, like the majority here, the dissent in Heller tried to justify the D.C. handgun ban by arguing that an “outright prohibition” is necessary “where a governmental body has deemed a particular type of weapon especially dangerous.” Heller, 554 U.S. at 713 (Breyer, J., dissenting). Of course, the Court rejected this view. And since it didn’t work against commonly used handguns, it won’t work against commonly used magazines. Perhaps trying to evade Heller’s clear ruling, the majority tries a new twist—recharacterizing the tradition as a ban on “especially dangerous uses of weapons once the threat to innocent persons has become clear.” See Maj. Op. 49–50. But that move fools no one. It is simply a retread of the same argument. And the bottom-line is this— neither the majority nor California has identified any historical regulation dispossessing law-abiding citizens of commonly used weapons for self-defense because they were “especially dangerous.” So this purported “tradition” is made of thin air. 2. Laws Regulating the Carry of Weapons California next relies on the tradition of regulating the public carry of certain weapons to justify its regulation. To be sure, these types of laws have been around a while. Anti- carry laws trace back to 14th-century England as well as colonial, Founding-era, and antebellum America. These historical regulations, of course, differed. Most forbade concealed carry, while some forbade open carry for certain illicit purposes—reflecting the sensibilities of the jurisdiction at the time. Only a few banned all public carry of specific weapons, while others required intent—whether the carry was meant to frighten or threaten. The throughline of all these regulations, however, is that none broadly disarmed law-abiding citizens. California begins with the Statute of Northampton of 1356. As Blackstone described the English law, the Statute prohibited the “offence of riding or going armed, with dangerous or unusual weapons,” while “terrifying the good people of the land.” 4 William Blackstone, Commentaries on the Laws of England, *148–49 (Wilfrid Prest et al. eds., 1st ed. 2016). Although it was “centrally concerned with the wearing of armor,” it’s accepted that it applied to weapons like the “launcegay.” Bruen, 597 U.S. at 41. To the Court, the Statute was notable for two reasons. First, the Statute didn’t apply to common weapons—like medieval daggers— which would be “most analogous to modern handguns.” Id. at 42. Second, the Statute applied only to those who carried weapons “with evil intent or malice.” Id. at 44. Yet, California’s ban applies to everyone—regardless of intent. All in all, the Court concluded that English law couldn’t “justif[y] restricting the right to publicly bear arms,” such as handguns, “for self-defense” today. Id. at 46.

DUNCAN V. BONTA 101

The earliest American law cited by California dates to 1686. Then, the Quaker province of East New Jersey prohibited the concealed carrying of “pocket pistol[s], skeines, stilladers, daggers or dirks, or other unusual or unlawful weapons” because they induced “great fear and quarrels.” Grants, Concessions, and Original Constitutions of the Province of New Jersey 289–290 (1881). Massachusetts followed with other carry regulations— although it only targeted armed groups. See An Act for preventing and suppressing of Riots, Routs, and unlawful Assemblies, 1750 Mass. Acts 545, ch. 17 § 1 (prohibiting being armed with “clubs or other weapons” in a group of twelve or more); An Act to Prevent Routs, Riots, and Tumultuous assemblies, and the Evil Consequences Thereof, 1786 Mass. Acts 87, ch. 38, 88 (same). California claims that other States enacted “anti-carry laws” for clubs and other blunt instruments. California next looks to historical laws limiting the carry of bowie knives, concealed weapons, and pistols. California contends that, by 1840, a handful of States implemented laws regulating the carry of bowie knives, 9 and that those restrictions eventually spread to most States by the end of

9 As examples, California points to an 1836 Tennessee statute and 1839 Alabama statute. But neither law was a sweeping ban on the carry of bowie knives. The Tennessee law only prohibited “wear[ing]” a bowie knife “under his clothes” or otherwise “keep[ing it] concealed about his person;” selling bowie knives, and “cut[ting] or stab[bing] another person” with a bowie knife. See 1837-38 Tenn. Pub. Acts 200-01, An Act to Suppress the Sale and Use of Bowie Knives and Arkansas Tooth Picks in this State, ch 137, §§ 1, 2, 4. And the Alabama law only regulated the “concealed” carry of “any bowie knife” and “any species of fire arms.” 1839 Ala. Acts 67, An Act to Suppress the Evil Practice of Carrying Weapons Secretly, § 1. the 19th century. 10 California also observes that by 1838, a handful of States had laws banning the concealed carrying of weapons, such as pistols, dirks, sword canes, and spears. 11 Finally, according to California, several States responded to an upswing in violence from the proliferation of “percussion-cap pistols” in the first half of the 19th century by restricting the carry of concealable pistols. The majority identifies two late-19th century laws banning the carry of concealable pistols. See Maj. Op. 44 (citing An Act to Preserve the Peace and to Prevent Homicide, 1871 Tenn. Pub. Acts 81, ch. 90, § 1; An Act to Preserve the Public Peace and Prevent Crime, 1881 Ark. Acts 191, chap. XCVI, § 1-2). Interestingly, courts in Tennessee and Arkansas struck down early versions of the laws because they applied to repeating or military-style revolvers. See Andrews v. State, 50 Tenn. 165, 187 (1871) (holding that if the state law applied to “the pistol known as the repeater,” “then the prohibition of the statute is too broad to be allowed to

10 California has identified no state laws banning the possession of bowie knives. An 1837 Georgia act declared that, along with prohibiting carry, a person cannot “keep, or . . . have about their person or elsewhere . . . [a] Bowie, or any other kind of knives.” An Act to Guard and Protect the Citizens of this State, Against the Unwarrantable and too Prevalent Use of Deadly Weapons, 1837 Ga. Laws 90, § 1. But in 1846, the Georgia Supreme Court held the act unconstitutional “inasmuch” as it “deprive[d] the citizen of his natural right of self-defence.” Nunn v. State, 1 Ga. 243, 251 (1846) (suggesting the ruling applied to all provisions of the act except concealed carry). 11 These States were Kentucky (1813), Louisiana (1813), Indiana (1820), Georgia (1837), Arkansas (1838), and Virginia (1838). See Clayton E. Cramer, Concealed Weapons Laws of the Early Republic: Dueling, Southern Violence, and Moral Reform 143–51 (1st ed. 1999). California also often refers to the Duke Center for Firearms Law’s Repository of Historical Gun Laws for sourcing. Available at https://perma.cc/E7FZ-

PANH.

DUNCAN V. BONTA 103

stand”); Wilson v. State, 33 Ark. 557, 559–60 (1878) (concluding that “prohibit[ing] the citizen from wearing or carrying [an army size pistol, such as are commonly used in warfare,] is an unwarranted restriction upon his constitutional right to keep and bear arms”). Regardless of a tradition to regulate the carry of certain weapons, these laws cannot justify California’s ban on the ownership and possession of magazines holding more than ten rounds. Bruen dictates this conclusion. In that case, New York relied on the same regulatory history as California to show that States may restrict the public carry of firearms. The Court agreed that “[t]he historical evidence from antebellum America . . . demonstrate[s] that the manner of public carry was subject to reasonable regulation.” Bruen, 597 U.S. at 59 (simplified). Even so, that robust history was not enough “to prevent law-abiding citizens with ordinary self-defense needs from carrying arms in public.” Id. at 60. The answer is simple then: if historical regulation of the carry of certain weapons in certain situations was not “relevantly similar” to a modern law prohibiting the carry of all weapons without a license, then those laws cannot be analogous to California’s law preventing law-abiding citizens from possessing firearms for self-defense purposes. Put differently, if targeted historical carry laws don’t justify wide-ranging restrictions on the carry of commonly owned weapons, they don’t support the outright ban of commonly owned weapons. Under Bruen, we can’t extrapolate from narrow regulations a justification for much broader regulations. While anti-carry laws “limited the intent for which one could carry arms, the manner by which one carried arms, or the exceptional circumstances under which one could not carry arms,” id. at 70, California’s law prohibits all conduct at all places at all times, even in the privacy of the home. And Heller made clear that the need for “defense of self, family, and property is most acute” in “the home.” 554 U.S. at 628; see also id. at 635 (The Second Amendment “surely elevates above all other interests the right of law-abiding, responsible citizens to use arms in defense of hearth and home.”). While historical carry restrictions may burden the Second Amendment right in one limited fashion, California’s magazine ban is different in scope—it’s a complete dispossession of a commonly owned “Arm.” Rahimi further signals the mismatch between the tradition of carry regulations and California’s magazine ban. In Rahimi, the Court determined that historical “surety laws” and “going armed laws” supported the modern regulation of disarming those under a domestic-violence protection order. See 18 U.S.C. § 922(g)(8)(C)(i). That’s because both the historical tradition and the modern regulation shared the same “how” and “why.” Both served the purpose of disarming “individuals found to threaten the physical safety of another.” Rahimi, 602 U.S. at 698. And their burdens were comparable—the temporary disarmament after a judicial determination of a “particular defendant[’s]” dangerousness. Id. at 699. In contrast, Rahimi observed that the law struck down in Bruen “broadly restrict[ed] arms use by the public generally.” Id. at 698. So while a tradition of targeted and temporary disarming of dangerous individuals may exist, that wouldn’t justify any broad-based, permanent dispossession. So too here. Limited historical laws designed to prevent confrontation, fear, and terror in public can’t justify a modern law that prevents all individuals from ever possessing a commonly owned arm for self-defense. In other words, we can’t extract any “principles” from the

DUNCAN V. BONTA 105

tradition of carry laws to fully dispossess law-abiding citizens of arms in common usage. Id. at 692. 3. Laws Banning Possession of Certain Weapons California also identifies a handful of laws throughout history that banned the outright possession of a weapon. None are a “relevantly similar” analogue to California’s magazine ban. California cites an English law from 1541 that prohibited persons with an annual income below 100 pounds from possessing a crossbow without a license. See An Act Concerning Crossbows and Handguns, 33 Hen. 8, ch. 6, § 1, (1541). This law is closer to California’s law since it appears to prohibit outright possession of a weapon—at least for certain classes of people. But by the 1700s, this law was widely considered “obsolete” and so restrictive that the Court considered the law as “not incorporated into the Second Amendment’s scope.” Bruen, 597 U.S. at 43 n.10. California points to late 19th-century state laws against slungshots. “Slungshots” refers to a wide range of hand-held weapons for striking—often with a metal or stone attached to a flexible strap or handle made of rope, leather, or other material. Between 1849 and 1890, nine jurisdictions prohibited the sales and manufacture of slungshots. 12 Only Illinois banned their outright possession, while the other

12 Vermont (1849), New York (1849), Massachusetts (1850), Kentucky (1856), Florida (1868), Dakota Territory (1877), Illinois (1881), Minnesota (1886), and Oklahoma Territory (1890). See David Kopel & Joseph Greenlee, History of Bans of Types of Arms Before 1900, 50 J. of Legis. 226, 346 (2024). jurisdictions prohibited their carry. 13 Most other states enacted laws against the carry of slungshots by the late 19th century. Even if these laws come from the historically relevant period, they don’t serve as proper analogues for California’s magazine ban. Simply, slungshots were not commonly used for self-defense. As California’s expert observed, they were “widely used by criminals and street gang members,” with no noted historical use for self-defense. Indeed, according to one source, while “[c]ourt records of the [1800s] have many cases of civilians . . . using slungshots,” “a man bringing one out after being threatened comes up rarely.” Kopel & Greenlee, 50 J. of Legis. at 345 (quoting Robert Escobar, Saps, Blackjacks, and Slungshots: A History of Forgotten Weapons 131 (2018)). So these slungshot regulations don’t evince a historical justification for regulating weapons that are used to “facilitate armed self- defense” or other lawful purposes. See Bruen, 597 U.S. at 28. Further, the burdens of slungshot regulations and California’s magazine restriction are dissimilar. Historically, most States restricted only the carry of slungshots—very different than an outright ban. Just one State banned their possession—not enough to suggest a widespread historical tradition. See Bruen, 597 U.S. at 67 (“[W]e will not stake our interpretation of the Second Amendment upon a law in effect in a single State, or a single city, ‘that contradicts the overwhelming weight of other evidence regarding the right to keep and bear arms[.]’” (simplified)).

13 Id. A Vermont law from 1849 made it a felony to possess or carry a slungshot for the “purpose of using it against another person.” Id. at 347.

DUNCAN V. BONTA 107

Finally, California raises 20th-century restrictions on automatic and semi-automatic firearms as historical analogues. According to California, from 1927 to 1934, over a dozen States restricted fully automatic and some semiautomatic firearms. We can make short work of this argument. These laws were enacted nearly 140 years after the Second Amendment’s ratification and 60 years after its incorporation. That’s simply too late to help define the meaning and scope of the Second Amendment right. The Supreme Court has made clear that our inquiry of regulations is at most cabined to the period “through the end of the 19th century.” Bruen, 597 U.S. at 35 (quoting Heller, 554 U.S. at 605). And even then, late 19th-century history may be too distant. See Bruen, 597 U.S. at 83 (Barrett, J., concurring) (“[T]oday’s decision should not be understood to endorse freewheeling reliance on historical practice from the mid-to- late 19th century to establish the original meaning of the Bill of Rights.”). So we can safely say that the 20th century is not a relevant timeframe for this inquiry. In sum, this survey of history shows no robust historical tradition of disarming law-abiding citizens of commonly owned arms used for self-defense. Indeed, the lack of outright weapons bans is telling in itself. From colonial America to Founding-era America to antebellum America, save an outlier or two, no laws banned the possession of any type of weapon. It’s even starker when looking at firearms bans. As the district court observed, between 17th-century English laws and 19th-century state laws, California “has not identified any law, anywhere, at any time . . . that prohibited simple possession of a gun or its magazine or any container of ammunition (unless the possessor was an African- American or a slave or a mulatto).” Duncan VIII, 695 F. Supp. at 1242. So if anything is “trapped in amber” here, Rahimi, 602 U.S. at 691, perhaps it’s the historical understanding that law-abiding citizens may choose any commonly owned firearm for self-defense without government interference. See Alicea, 174 U. Pa. L. Rev (manuscript at 41) (“[T]he tradition of banning dangerous and unusual weapons—and the absence of a tradition of banning weapons in common use—is very strong (if not dispositive) evidence that prohibiting arms in common use by persons protected under the Second Amendment is per se impermissible[.]”). 4. Laws Against Trap Guns California’s reliance on trap-gun regulations is even more far afield. Trap guns were contraptions using string, wire, or other contrivances to remotely discharge a firearm. Historically, they were used to thwart thieves from robbing businesses or properties and sometimes for hunting. According to California, only eleven States regulated trap guns before the 20th century. 14 And only one was from the 18th century. New Jersey’s 1771 law banned “a most dangerous Method of setting Guns.” 1763–1775 N.J. Laws 346, An Act for the Preservation of Deer and Other Game, and to Prevent Trespassing with Guns, ch. 539, § 10. The rest of these laws came into effect in the mid- to late-19th century. Even assuming these laws are temporally significant, they would not implicate the Second Amendment. Trap-gun mechanisms aren’t protectable “Arms” under the Second

14 According to California, those States are Michigan (1875), Minnesota (1873), Missouri (1891), New Jersey (1771), New York (1870), North Dakota (1891), Rhode Island (1890), South Carolina (1855), Utah (1865), Vermont (1884), and Wisconsin (1872). We assume this record is accurate.

DUNCAN V. BONTA 109

Amendment’s text. To start, they are not “bearable.” See Heller, 554 U.S. at 582 (“[T]he Second Amendment extends, prima facie, to all instruments that constitute bearable arms[.]”). On the contrary, the whole design of the trap gun was to allow a firearm to be discharged without a person needing to “keep” or “bear” it. So trap guns are not weapons to “wear, bear, or carry . . . upon the person or in the clothing or in a pocket, for the purpose . . . of being armed and ready for offensive or defensive action in a case of conflict with another person.” Id. at 584 (simplified). Nor can a person take a trap gun “into his hands, or useth in wrath to cast at or strike another.” Id. at 581. And they are not integral components of a firearm. So the majority is just simply wrong to claim that the trap guns were used for armed self- defense. Indeed, the majority mistakenly suggests that homeowners rigged trap guns in “defen[se]” of their “homes.” Maj. Op. 42. But nothing in the record supports that trap guns were used in such an improper way. Given all this, it’s hard to see how trap guns fall under the Second Amendment’s textual elements. But even if trap guns constitute “Arms,” the burdens of regulating trap-gun mechanisms are not analogous to the burdens of California’s magazine ban. First, California identifies no historical regulation that prohibited the possession of a trap-gun. Like New Jersey’s law, most States prohibited only the setting of the device. See Kopel & Greenlee, 50 J. of Legis. at 365–366. Other States only forbade their setting for hunting or for injuring another person. Id. But the broad dispossession of an arm is different from preventing individuals from rigging a firearm with a contraption to fire it automatically. The “why” of trap gun regulations is also very different. As the majority must concede, these regulations were meant to prevent the occasional tripping of trap guns by innocent persons. California didn’t ban magazines holding more than ten rounds to limit accidental firearms deaths. Rather, the purpose of the magazine ban was to respond to intentional gun violence. 5. Laws Regulating Gunpowder Storage Finally, California relies on 18th- and 19th-century gunpowder-storage laws. Concerned with the dangers of massive fires and explosions, these laws prohibited the stockpiling of large quantities of gunpowder in one place. An 18th-century law, for example, made it unlawful in New York City “to have or keep any quantity of gun powder exceeding twenty-eight pounds weight, in any one place, less than one mile to the northward of the city hall[.]” 1784 N.Y. Laws 627, An Act to Prevent the Danger Arising from the Pernicious Practice of Lodging Gun Powder in Dwelling Houses, Stores, or Other Places, ch. 28. Another 1821 Maine law regulated how much gunpowder could be possessed and stored by a person “for the prevention of damage by Fire.” 1821 Me. Laws 98–99, An Act for the Prevention of Damage by Fire, and the Safe Keeping of Gun Powder, ch. 25. These gunpowder-storage restrictions don’t establish a historical tradition supporting California’s magazine ban. First, these laws offer no comparable burden on the possession of a firearm. While California’s magazine ban prohibits using the most popular magazine for self-defense, the gunpowder laws had zero effect on self-defense. They “did not clearly prohibit loaded weapons” and “required only that excess gunpowder be kept in a special container or on the top floor of the home.” Heller, 554 U.S. at 632 (emphasis added). So they regulated the accumulation of

DUNCAN V. BONTA 111

excess explosive material by limiting where it could be stored—they didn’t prevent citizens from having ammunition at the ready for self-defense. As the Supreme Court observed when this history was used to defend a handgun ban, “[n]othing about th[e]se fire-safety laws undermines our analysis” because “they do not remotely burden the right of self-defense as much as an absolute ban on handguns.” Id. Second, the “why” is also obviously different. These laws targeted the danger of accidental explosions and widespread fire posed by improperly stored gunpowder. In contrast, California’s purpose in enacting its ban was to reduce intentional gun violence. So this tradition doesn’t support California’s magazine ban. *** In sum, the right to keep and bear firearms that use a magazine able to hold more than ten rounds is presumptively protected by the text of the Second Amendment. These magazines constitute “Arms” because they are necessary components of firearms and facilitate the firing process. Thus, California had the burden of identifying a historical analogue that is relevantly similar to a ban on these magazines—and has failed to do so. Simply, no historical analogue justifies California’s absolute ban on magazines that come standard with most firearms. While “dangerous and unusual” weapons may fall outside the Second Amendment’s scope, arms in common use today—like magazines holding more than ten rounds—clearly don’t. And though the Second Amendment seemingly permits reasonable restrictions on the public carry of some uncommon weapons, that’s not comparable to California’s outright prohibition of the most popular magazine for self- defense. And the lack of any historical regulations banning the possession of a common firearm further undermines California’s law. Finally, with no impact on armed self- defense at all, the regulations on trap guns and gunpowder storage are not remotely close to the burden and justification for California’s ban. Because California failed to meet its burden, California Penal Code § 32310 is unconstitutional.

III.

THE RETURN OF INTEREST BALANCING

The majority upholds California’s magazine ban despite Heller, McDonald, Bruen, and Rahimi. In doing so, the majority rejects the Supreme Court’s Second Amendment framework and reads the Amendment as it wants. First, the majority haphazardly establishes two Bruen tests—the so- called more nuanced and unnuanced approaches. If that sounds confusing, it is. In fact, it’s so confusing that the majority largely abandons its own creation mid-opinion. Second, rather than examine historical analogues for their similarity with California’s regulation, the majority simply cloaks interest balancing under the guise of “tradition.” So in the Ninth Circuit, we’ve returned to the old days of judicial policymaking that the Court has gone out of its way to end. A. The Majority’s Nuanced v. Unnuanced Approaches To begin, we can’t ignore the majority’s creation of alternate Bruen tests—what it dubs the “more nuanced approach” and the “straightforward,” unnuanced approach. Maj. Op. 37, 40. Plucking a few words from Bruen, the majority claims it may apply a “more nuanced approach”

DUNCAN V. BONTA 113

anytime a regulation involves “unprecedented societal concerns or dramatic technological changes.” Id. at 37 (quoting Bruen, 597 U.S. at 27). Although it’s unclear what precisely the majority means by a “more nuanced approach,” it appears to mean that we may disregard our historical tradition of firearms regulation whenever a modern regulation seeks to address modern problems or technology. Id. at 37–38. Instead, whenever that is the case, the majority calls for a more “flexible analogical approach.” Id. at 40. This flexibility apparently means that no analysis of historical analogues is necessary—all that’s needed is a determination that there’s a difference between technology or societal problems at the Founding compared to today. See id. at 39 (analyzing how the magazine ban responds to an “unprecedented societal concern” and the magazines here “represent a dramatic technological change from the weapons at the Founding” but not comparing modern regulations to any historical analogues). So the majority’s “more nuanced” approach is more accurately termed the “ahistorical” approach. There’s much to dislike about the majority’s creation. First, the Supreme Court has never endorsed a “more nuanced” versus a “straightforward,” unnuanced approach. Indeed, it would be remarkable if the Supreme Court were forming two sets of Bruen tests without telling anyone. So the majority’s test is just its own invention. Instead, there is one approach: the Second Amendment’s text and historical understanding always control. The majority claims to just be quoting Bruen. But take Bruen’s comment in context: “While the historical analogies here and in Heller are relatively simple to draw, other cases implicating unprecedented societal concerns or dramatic technological changes may require a more nuanced approach.” Bruen, 597 U.S. at 27. The Court’s note about the occasional need for a “more nuanced approach” was an unremarkable observation that making comparisons to proper historical analogies might be challenging at times. Indeed, “[h]istorical analysis can be difficult; it sometimes requires resolving threshold questions and making nuanced judgments about which evidence to consult and how to interpret it.” Id. at 25 (simplified). That’s especially the case when “unprecedented societal concerns or dramatic technological changes” may make analogical reasoning relatively tougher. Id. at 27. Bruen and Heller’s historical analogies, in contrast, were “relatively simple to draw” given the extreme nature of the bans. Id. So the Court was simply cautioning lower courts to be careful and thoughtful in scrutinizing the government’s claim of historical analogues. Simply, no approach that ignores history adheres to Bruen. But in the majority’s telling, Bruen’s discussion of a “more nuanced” approach was the Court silently embracing a whole new test in cases involving “unprecedented societal concerns or dramatic technological changes.” Maj. Op. 40 (quoting Bruen, 597 U.S. at 27). And how do we know the majority is wrong in developing its two-test regime? Just read Rahimi. That case makes no mention of the majority’s nuanced/unnuanced distinction. If this two-test method were so central to Bruen, we would expect that the Court would at least say something about it in its very first case applying Bruen. Instead, we get zilch. Undeterred, the majority plows ahead with its creation. Yet it has no explanation for Rahimi’s silence on the two-tests framework. It fails to explain whether Rahimi is a nuanced case or an unnuanced case. It skips all of this and just moves forward with its made-up analysis.

DUNCAN V. BONTA 115

But it gets worse. The majority also makes clear that its “more nuanced” test is a repackaging of this circuit’s old interest-balancing regime. The majority holds that when governments deal with modern technology or problems, like mass shootings, they need “even more flexib[ility]” and thus the governments’ regulations are constitutional. Id. at 37. All of this is just code for judicial interest balancing and policymaking. Under this balancing regime, there’s no look to the historical meaning of the Second Amendment—only a review of technological or societal change and whether that change justifies the government’s regulation. See id. at 38– 40. So, the “more nuanced” approach appears to be little more than reinvigorated judicial means-ends scrutiny in fancy dress. In the end, we have no doubt this “more nuanced” approach will resemble our old “black box” regime where “judges [simply] uphold favored laws and strike down disfavored ones.” Duncan V, 19 F.4th at 1140 (Bumatay, J., dissenting). Besides flouting the Court, this two-test approach makes no constitutional sense. The Constitution enshrines enduring principles. That means that the Constitution doesn’t so easily bow to technological or societal change. In any other context, we would scoff at the idea that the Constitution grants broad deference to the government simply based on the modernity of the problem. The free- speech right didn’t change when the internet was invented. See, e.g., Moody v. NetChoice, LLC, 603 U.S. 707 (2024). The Fourth Amendment right didn’t succumb to new technology. See, e.g., Carpenter v. United States, 585 U.S. 296 (2018). In none of these cases have courts just thrown up their hands and declared that the government needs “more flexibility.” So the majority’s creation is a grave mistake. In the end, perhaps recognizing its vast departure from the Court’s directives, the majority retreats from its two-test approach mid-opinion and purports to apply Bruen, as directed by the Court. Maj. Op. 40. Even then, the majority still gets it wrong. We turn there next. B. Interest Balancing As Analogical Reasoning By now, it should be clear what courts should do at Bruen step two when analyzing the government’s justification of its gun control laws. We look to whether the modern regulation “impose[s] a comparable burden on the right of armed self-defense and whether that burden is comparably justified.” Bruen, 597 U.S. at 29. Our task then is to see if historical analogues point to a limit on the Second Amendment’s scope that justifies the regulation. In other words, we decide whether the “how” and the “why” of the government’s regulation and purported historical analogue are “relevantly similar.” Id. Unfortunately, the majority botches this analysis. Rather than examine historical analogues for their similarity with California’s regulation, the majority cloaks interest balancing under the guise of “tradition.” But it’s easy to see the majority’s sleight-of-hand. We break it down step-by-step. Step One: The majority starts off by assessing the magnitude of California’s magazine ban’s burden on the right of self-defense. It concludes that the burden is only “minimal.” Maj. Op. 52 n.11; see also id. at 49 (“California’s law has a significantly smaller effect on the speed of armed self-defense.”). The majority asserts that the burden is small because California’s law didn’t go further. Id. at 51 (“The law imposes no limit whatsoever on the number of magazines[,] . . . bullets[,] . . . or . . . firearms a

DUNCAN V. BONTA 117

person may own. The law also imposes no limit on the number of rounds a person may fire or the number of firearms a person may fire. Nor . . . does the law ban any weapon.”). Instead, the majority says that “the law prohibits only one very specific use of some firearms: the shooting of an eleventh (or successive) round” without reloading and concludes the burden minimal. Id. at 51. It then claims that people “rarely” use magazines holding ten rounds in armed self-defense. Id. at 52 n.11. All this ends with the majority just declaring that these magazines are “no weapon” at all. Id. at 55. The majority again falls for the fallacy that “using a firearm” for self-defense equates to pulling the trigger and firing every round. But the “natural meaning” of “bear arms” is “being armed and ready for offensive or defensive action in a case of conflict with another person.” Heller, 554 U.S. at 584 (simplified). So the right of self-defense isn’t measured by how many bullets are expended. It’s determined by how law-abiding citizens choose to “ready” themselves in “case of conflict with another person.” Id.; see also Bruen, 597 U.S. at 74–75 (Alito, J., concurring) (retelling the stories of “potential victim[s who] escaped death or serious injury only” because of armed self-defense without needing to discharge a firearm or shoot the assailant). Our criminal laws don’t require a firearm to be discharged for it to be “used.” See, e.g., Smith v. United States, 508 U.S. 223, 230 (1993). Cf. Bailey v. United States, 516 U.S. 137, 143 (1996) (acknowledging that “use draws meaning from its context,” such that someone can “use a gun to protect [his] house” while “never ha[ving] to use it” (simplified)). All that matters is that California’s total ban on possession of magazines capable of holding more than ten rounds “amounts to a prohibition of an entire class of ‘arms’ that is overwhelmingly chosen by American society for th[e] lawful purpose” of self-defense. Heller, 554 U.S. at 628. Even the majority concedes that California bans magazines that “enhance[] . . . a person’s ability to fight or to defend” himself. Maj. Op. 29. And it’s hard to see how the majority’s burden analysis is any different from its step-one conclusion that these magazines are not an “Arm[]” at all. So the majority just confuses the step one and step two inquiry. In any case, we proceed with the majority’s interest balancing. Step Two: Having arrived at its stylized burden, the majority moves on to analyze historical regulations’ purported justifications. Misreading history and attempting to extract the broadest of generalities, the majority manufactures a “tradition of regulating a particular, especially dangerous use of a weapon, once that use becomes a specific threat to innocent persons.” Id. at 56, see also id. at 41 (“Beginning before the Founding and continuing throughout the Nation’s history, legislatures have enacted laws to protect innocent persons from especially dangerous uses of weapons once those perils have become clear.”). As explained above, no historical analogue supports this made- up tradition. And in reality, the majority just seeks to recognize a broad “public safety” justification for government regulation. Under the majority’s view, this justification gives States the ability to regulate any arms for public safety. Such a justification swallows the entire Second Amendment right. Step Three: The majority then makes a factual leap— finding that magazines holding more than ten rounds fit the bill of an “especially dangerous” weapon. In the majority’s

DUNCAN V. BONTA 119

view, “[l]arge-capacity magazines exacerbate the harm caused by mass shootings, and limiting magazine capacity thus prevents or mitigates the harm caused by mass shootings.” Id. at 48. Of course, the majority cites no facts to support this. Even more, it doesn’t explain how these magazines are “especially dangerous” firearms or under what baseline it makes this comparison. How are we to decide when a firearm is just “dangerous” versus “especially dangerous”? Only the majority knows. Step Four: The last step is the majority’s comparison of the burden of the magazine ban to the supposed justification for outlawing “especially dangerous” weapons. In the majority’s view, because the “same justification underpins California’s restriction on magazine capacity” as historical laws and because California’s ban imposes “less of a burden” on armed self-defense than some historical laws, it is constitutional. Id. at 47, 55–56. In other words, it takes a broad justification (the “why”) and compares it with a minimal burden (the “how”) and finds a match. So rather than compare the justifications and burdens to “relevantly similar” analogues, the majority just looks to the fit between a generalized “why’” and an in-the-ballpark “how.” But this crosses wires. We are not supposed to compare the “how” to the “why”; we’re supposed to compare the “how and why” of modern regulations to the “how and why” of analogous historical regulations. If the historical and modern regulations share a common “how and why,” then this may reveal that the regulation is outside of the scope of the Second Amendment. Put another way, we must see whether California’s magazine ban “works in the same way and does so for the same reasons” as historical regulations. Rahimi, 602 U.S. at 711 (Gorsuch, J., concurring). Instead, the majority merely points to some historical “whys” and some historical “hows” and calls it a day. Although they try to disclaim it now, make no mistake—this is what the majority has been doing and continues to do. Take the majority’s analysis of Bowie knife and slungshot regulations. The majority claims the regulation of both weapons supports its supposed tradition of regulating “especially dangerous” weapons. But it also acknowledges that the only historical burden on those weapons was to “ban[] their carry outside the home.” Maj. Op. 45. So even if the majority were right, the justification of regulating “especially dangerous” weapons only leads to the prohibition of carrying the weapons outside the home—not outright possession bans as California enacts. If comparing the “how” to the “why” of a regulation sounds familiar, it is. It’s interest balancing 101—this time masquerading as respect for the Second Amendment’s historical scope. The majority’s analysis bears all the hallmarks of judicial means-ends balancing—determining first whether California’s interest is compelling, then assessing the severity of the burden, and then evaluating whether California’s means fit its end. Look at the majority’s language pre-Bruen and post-Bruen and notice how little has changed (even after the majority attempts to mask its defiance):

DUNCAN V. BONTA 121

MAJORITY PRE-BRUEN MAJORITY POST-BRUEN

“[L]arge-capacity “Mass shootings are magazines tragically devastating for the entire exacerbate the harm community, and large- caused by mass capacity magazines shootings.” exacerbate the harm.” Duncan, 19 F.4th at 1109 Maj. Op. 47 “California’s ban on large- “California’s law imposes capacity magazines only a minimal burden on the imposes only a minimal right of armed self-defense.” burden on the exercise of Maj. Op. 52 n.11 the Second Amendment right.” Duncan, 19 F.4th at 1104 “[W]e conclude that “Its prohibition on a California’s ban is a weapon’s component that reasonable ﬁt, even if an serves the sole function of imperfect one, for its enabling a speciﬁc, and compelling goal ofespecially dangerous, use of reducing the number of a ﬁrearm ﬁts neatly within deaths and injuries causedthe tradition” of banning by mass shootings.” especially dangerous Duncan, 19 F.4th at 1110 weapons. Maj. Op. 53–54 “California’s ban on large- “California’s modern capacity magazines is a law[’s] . . . justiﬁcation for reasonable ﬁt for the burdening the right to armed self-defense” [ﬁts the need] compelling goal of to protect innocent persons reducing gun violence.” from infrequent but Duncan, 19 F.4th at 1111 devastating harm.” Maj. Op. 47

Bruen did two things: (1) it ended judicial interest balancing and (2) it provided a new framework for considering Second Amendment challenges. Despite this revolutionary change, things remain the same at the Ninth Circuit. Faithfully applying Bruen requires a course correction that the majority refuses to take. Instead, the majority just declares it knows better and charts its own path. But that disrespects the Supreme Court and the rule of law.

IV.

CONCLUSION

At each step of this case, the majority has made clear its disdain for the Supreme Court’s Heller-McDonald-Bruen- Rahimi jurisprudence. But our job is to follow even if we disagree. Because California’s magazine ban violates the Second Amendment’s text, history, and tradition, we respectfully dissent.

DUNCAN V. BONTA 123

VANDYKE, Circuit Judge, dissenting:

Three years ago, the Supreme Court vacated our court’s opinion in this very case, presumably because we tried the same tack that Bruen rejected in no uncertain terms: engaging in interest balancing after assuming that an activity falls within the scope of the Second Amendment. In other words, our court’s reliance on interest balancing (like the Second Circuit’s decision in Bruen) took “one step too many.” N.Y. State Rifle & Pistol Ass’n v. Bruen, 597 U.S. 1, 19 (2022). Because, as the Supreme Court made clear enough in District of Columbia v. Heller, the Second Amendment is itself “the very product of an interest balancing by the people,” it is not our role to “conduct [it] for them anew.” 554 U.S. 570, 635 (2008). Notwithstanding these repeated directives to stop, today’s decision doubles down on this court’s prior practice of balancing away the rights of law-abiding citizens to bear arms in self-defense—only this time under another name. Rather than balancing after concluding that the original understanding of the Second Amendment protects an individual’s conduct, the majority now merges its balancing into its determination of whether the Second Amendment protects an individual’s conduct at all. This reinforces why interest balancing must have no place in applying the Second Amendment. See Duncan v. Bonta (Duncan V), 19 F.4th 1087, 1159–60 (9th Cir. 2021) (en banc) (VanDyke, J., dissenting), cert. granted, judgment vacated, 142 S. Ct. 2895 (2022), vacated & remanded, 49 F.4th 1228 (9th Cir. 2022). I wholeheartedly agree with Judge Bumatay’s excellent dissent, which thoroughly demonstrates the correct approach under the Second Amendment following Bruen. I write separately to further highlight some serious flaws in the majority’s analysis. If some of what you read below sounds familiar, it is because, consistent with the majority’s reuse of balancing by another name, many of the flaws I identify are barely disguised retreads of those I pointed out in our last iteration of this case. Id. at 1160–70. Notwithstanding Bruen, very little has changed about the Second Amendment out here in the Ninth Circuit. I. The majority begins with Bruen’s cue to first determine whether “the Second Amendment’s plain text covers an individual’s conduct.” 597 U.S. at 17. One might expect this to be a mechanical or routine part of the analysis, especially considering the Supreme Court’s emphasis that there is only one step in assessing a Second Amendment claim. Id. at 19. But many lower courts, like ours, have instead taken Bruen’s guidance to mean there is an extensive first-step, arm-or-not inquiry. According to the majority, California is correct that a larger capacity magazine is not an “arm” within the meaning of the Second Amendment because it is “an accessory, or accoutrement, not an ‘Arm’ in itself.” As a result, the majority concludes the Constitution does not “presumptively protect[]” it. But whether a firearm component is an inherent and “necessary” part of the arm itself, or instead merely an “optional” and unnecessary accessory to the arm, is a hopelessly indeterminable and inadministrable distinction. As is true for almost any type of object—from cars to computers to sewing machines to software—there is no Platonic ideal of a firearm from which such distinctions between “inherent” and “optional” parts can be objectively derived. Yet the majority relies on its invented “arms– accessory” distinction as an on/off switch for fundamental

DUNCAN V. BONTA 125

constitutional protections. Because that distinction has no basis in reason or reality, what this new test is really cover for is the majority’s ipse dixit. A higher capacity magazine is an accessory, and thus unprotected by the Second Amendment, because the majority says so. And a lower capacity magazine (as yet undefined) is a “necessary” part of the arm, and thus protected, again because the majority says so. This is a terribly unprincipled way to analyze constitutional rights. Initially, I planned to explain my reasons for dissenting on this conceptual point through usual judicial means alone: describing in writing some real-world illustrations to explain how the majority’s supposed “arms–accessory” distinction collapses. But at argument it became clear to me that a visual illustration would greatly aid my colleagues and the parties in better grasping how this rather obvious conceptual problem specifically applies to firearms. So instead of straining to use written words to explain the many different parts of a gun and how each part could easily be deemed an “accessory” under the majority’s vacuous test, I have decided to deliver part of my dissent in this case orally—via video—under the established wisdom that showing is sometimes more effective than telling. Please click the link below and enjoy the presentation: https://www.ca9.uscourts.gov/media/23-55805/opinion As I hope the video portion of this dissent helpfully illustrates, an “arm”—just like most other categories of objects known to the human experience—is a broad conceptual term covering an almost limitless variety of configurations within that category. See Bruen, 597 U.S. at 28 (explaining that the term “extends, prima facie, to all instruments that constitute bearable arms” (emphasis added)). The majority does not and cannot dispute that. It acknowledges that “the meaning of ‘Arms’ … broadly includes nearly all weapons used for armed self-defense.” But the majority nonetheless concludes that in its view some parts of a firearm are “necessary to the operation of a weapon” and thus protected by the Second Amendment, while other parts are not necessary and therefore not protected “arms.” The majority then purports to apply its misguided new test to decide that higher capacity magazines are not arms, while lower capacity magazines are. The majority’s new constitutional test fails at the most basic level possible—the conceptual level. Its failure is not even related to anything particularly unique to firearms. The inherent indeterminability of categorizing constituent parts of a class of objects as either belonging to the class itself or, instead, merely functioning as “unnecessary accessories” to that class should be self-evident for almost everything from cars (doors?) to cellphones (cameras?) to cereal (marshmallows?). Firearms are of course no different. The only difference is that the majority’s test as applied to firearms is not just philosophically goofy, but it also has the very real and troubling result of denying Americans’ constitutional rights. It is so easy to demonstrate the conceptual failings of the majority’s new test that even a caveman with just a video recorder and a firearm could do it. For example, while the majority concedes that “triggers” are firearm components due at least some Second Amendment protection, the majority’s misguided test cannot support that conclusion. Even something as essential to the firearm as a manufacturer-issued trigger could be considered an unprotected “accessory” under the majority’s view because that particular trigger is not essential to the function of the

DUNCAN V. BONTA 127

firearm, as it could be swapped out for one with less effective, and therefore less “dangerous,” attributes. The problem with the majority’s misguided test is no different with respect to larger capacity magazines. My colleagues in the majority reason that “a magazine is an integral part” to the operation of a semi-automatic gun and therefore “that the Second Amendment’s text encompasses a right to possess a magazine.” I agree. But the majority also contends that a “large-capacity” magazine “is not necessary to operate any firearm” and is therefore not an arm or a protected component. California defines a large- capacity magazine as any magazine holding more than ten rounds. But why stop there? Under the majority’s rationale, any magazine that holds more than one round is not “necessary” for the function of the weapon. So presumably California could also ban magazines holding five rounds. Maybe even two. And why stop at magazines? According to the majority, because “firearms operate as intended without a large- capacity magazine,” and a large-capacity magazine is not “necessary to the ordinary functioning of a firearm,” large-capacity magazines are not protected under the Second Amendment. But under that logic, basically every part of a firearm is an “optional component” because each could be replaced with a less effective (aka, less “dangerous”) version of that part and the firearm would still “operate” in some sense. Nor is it at all clear what the majority means by “as intended” and “ordinary functioning.” Technically speaking, I suppose that would mean a grip or a sighting system is not a protected component of a firearm because those pieces are “optional components” not strictly necessary to make the gun fire a round. Some handguns come without any sights at all. Those guns are obviously difficult to aim accurately. So does that mean California could ban all grips and sights under the majority’s test? After all, just as a magazine is only “a box that, by itself, is harmless,” a grip could be characterized as just a piece of polymer, a barrel as just a steel tube, and a bullet as just a small hunk of metal. Each one of those pieces, just like every other individual part of a firearm, “is benign” and “useless in combat for either offense or defense” without the rest of the firearm. More basically, what do my amateur gunsmithing colleagues mean by “operate as intended?” Take a red dot optic. A firearm equipped with a red dot optic is “intended” to be operated more quickly and accurately than a firearm without one. So I suppose you could say the red dot optic is “necessary” to make the red-dot-optic-equipped firearm “operate as intended.” But of course, like many parts of modern firearms, it is not necessary at all if “operate as intended” means only the bare minimum functionality needed to send a bullet downrange. Is a red dot optic an unprotected accessory or a protected component under the majority’s test? As another example, most modern handguns have an automatic cycling mechanism that, upon firing, expels the spent cartridge, loads a new round, and resets the trigger. But plenty of firearms do not have an automatic cycling mechanism. The automatic cycling mechanism is “necessary” to make a semi-automatic firearm “operate as intended,” but it is not necessary to make, say, a revolver or a bolt-action or a single-shot break-action firearm operate. Could California ban all semi-automatic handguns by applying the majority’s logic that, because the automatic

DUNCAN V. BONTA 129

cycling mechanism is not required to make a handgun work, it’s simply not protected by the Second Amendment? 1 The majority’s test produces head-scratching results. On the one hand, the majority gives lip service to the fact that “[t]he meaning of ‘Arms’ … broadly includes nearly all weapons used for armed self-defense.” But on the other, its reasoning inevitably means that only the most dumbed-down or basic version of any component part of a gun is protected—and many parts of a gun are entirely unprotected if they aren’t strictly necessary to make a gun go bang. Similarly, the majority acknowledges that the Second Amendment “must carry an implicit, corollary right to bear the components or accessories necessary for the ordinary functioning of a firearm.” But the obvious result of the majority’s test is that almost all of the component parts of a firearm would fall completely outside the Second Amendment because, in theory, any particular part could be replaced with a dumbed-down version of the same part. In another place, the majority suggests that “any accessory” performs a specific function that “enhance[s] … a person’s

1 The majority’s historical examples fail to shed light on the bounds of its test. The majority explains that, historically, “accoutrements” like “flint, scabbards, holsters, and ammunition containers” were “distinct from ‘arms.’” But under the “necessary to the ordinary functioning” test posited by the majority, one would have assumed flint would clearly be covered as a component part of a firearm. Flint is integral to the actual firing of a flintlock firearm. Akin to how a modern-day striker or firing pin ignites the primer in a cartridge, starting the chain reaction that fires a bullet, flint creates a spark that ignites gunpowder in a flashpan that causes the gun’s discharge. Granted, flint is a material that degrades more quickly than the material comprising a modern firing pin and therefore must be replaced with some frequency, but it is absolutely a required part to make a flintlock firearm “operate as intended.” A flintlock will not work without flint. ability to fight or to defend.” So the majority seems to acknowledge that its test devolves to exclude anything that enhances a firearm’s operation—or, put more bluntly, anything that works too well. In the end, the majority’s test boils down to something like this: the Second Amendment presumptively protects only the jankiest version of a firearm and a little bit of ammunition (2.2 rounds?). Sound familiar? See Duncan V, 19 F.4th at 1173 (VanDyke, J., dissenting). The only way the majority can classify a large-capacity magazine as “an optional accessory” is if it has some idea of what is (and is not) “optional.” Put differently, the majority must have at least some idea of what a “standard” firearm is. The majority’s logic is premised on its assumption that there is some Platonic ideal of a firearm, which I guess makes sense if you think judges are the Platonic Guardians of the Second Amendment. That’s a nice job if you can get it, but it should be clear enough by now that many judges (and gun- banning governments) know next to nothing about how guns actually work, which perhaps explains why they would invent such an obviously inadministrable test for guns, but never for any other constitutional right. Ultimately, just as with televisions and sewing machines, there is no such thing as a stock-part basic firearm, unadorned and without any “accessories,” that constitutes the only “arm” protected under the Second Amendment. There are many parts that constitute the arm, most of which usually can be swapped out to emphasize and improve certain functions over others. Consider, for example, heavier grips that make the gun steadier when shooting versus lighter ones that are easier to carry and conceal. Just as the First Amendment doesn’t apply only to “necessary” or “essential” speech, the Second Amendment cannot apply only to firearms containing just those parts that a state like

DUNCAN V. BONTA 131

California deems essential and necessary. Instead, what constitutes the “arm” includes every functional component and not only the most downgraded version of a “necessary” component. Cf. Jackson v. City & Cnty. of San Francisco, 746 F.3d 953, 967–68 (9th Cir. 2014) (holding that hollow- point ammunition is covered by the Second Amendment because of the “corresponding right to obtain the bullets necessary to use firearms” (cleaned up)); Fyock v. Sunnyvale, 779 F.3d 991, 998 (9th Cir. 2015) (recognizing that this same principle applies to magazines).

II.

The majority’s analytical flaws do not end at the first step of its analysis. Despite initially (and incorrectly) determining that large-capacity magazines are not part of the arms covered by the Second Amendment at all, the majority proceeds to assess whether a ban on large-capacity magazines is consistent with history and tradition. Judge Bumatay aptly explains how the majority’s analysis errs as a historical matter. I will elaborate on just a few points. The majority explains that for “cases implicating unprecedented societal concerns or dramatic technological changes,” we should apply “a more nuanced approach.” 2 It

2 Here, the majority determines that such an approach is justified because “[l]arge capacity magazines, when attached to a semi-automatic firearm, … represent a dramatic technological change” because they “fire with an accuracy, speed, and capacity that differ completely from the accuracy, speed, and capacity of firearms from earlier generations.” The majority seems to recognize the truth that most of the “dramatic technological” jump since the Founding is attributable to the automatic cycling mechanism—not the large-capacity magazines at issue here. Presumably then, because that mechanism accounts for most of the “dramatic” change making modern firearms “especially dangerous” compared to the muzzleloaders of our forefathers, the unstated is true that the Supreme Court explained that some societal changes would be so unprecedented—and some technological changes so great—that historical regulations would not have contemplated them. Bruen, 597 U.S. at 27. In those cases, courts should “reason[] by analogy.” Id. at 28. But rather than look for close analogues that “impose a comparable burden on the right of armed self-defense” that “is comparably justified,” id. at 29, the majority warms up by opining that analogues should be “flexible.” What the majority’s new “flexible” and “nuanced” approach devolves into is that the government need only show the sloppiest of a fit between the historical example and the modern regulation. To show you what I mean, consider the majority’s use of gunpowder storage laws as an example of a historical regulation analogous to California’s ban on magazines holding more than ten rounds. Under Bruen, we must look to “why” gunpowder storage laws existed and “how” those regulations burdened the Second Amendment by requiring gunpowder to be stored in a special container or place in the home. Start with the “why.” As the majority acknowledges, those laws existed “to prevent … fires and explosions from the storage of gunpowder,” or by “prohibiting what had proved to be an especially dangerous” practice. Heller itself summarized these storage laws in a similar way. 554 U.S. at 632 (distinguishing gunpowder storage laws because they “did not clearly prohibit loaded weapons, but required only

conclusion that follows from the majority’s misguided analysis is that a state could ban semi-automatic handguns altogether.

DUNCAN V. BONTA 133

that excess gunpowder be kept in a special container or on the top floor of the home”). So far so good. But how do those laws look anything like the magazine ban at issue here? If that leap confuses you, you’re in good company. To get from gunpowder storage laws to bans on magazines with more than ten rounds, the majority must summarize the “why” as “to prevent a specific, infrequent type of harm to innocent persons,” or as they further summarize it, to stop “an especially dangerous use of firearms.” Turning to the next step, the majority characterizes the “how” as completely “prohibiting” certain methods of storage. So the majority generalizes the history-and-tradition test—first explicated in Bruen and then expanded upon in United States v. Rahimi, 602 U.S. 680 (2024)—to the level of “especially dangerous” and posits that the government can completely ban anything that fits that definition. It is hard to imagine a degree of fit much sloppier than that. 3 By taking any historical example that completely bans a specific activity and raising the level of generality for the “why” to something akin to mere “dangerous[ness],” it is hard to see how any gun regulation wouldn’t pass muster, because governments have been banning “dangerous” stuff and activities since time immemorial. In fact, I would venture to guess that even the regulations that failed in Heller or Bruen would survive such trifling scrutiny. After all, (1) lots of historical laws have prohibited dangerous things, and (2) even the jankiest firearm in the hands of the wrong person is “especially dangerous.” So what stops the

3 I use this one example to illustrate my point, but the point stands for the other historical analogues the majority suggests, as discussed in Judge Bumatay’s dissent. government from banning the possession of all firearms under the majority’s loose test? But that’s not the only way the majority overhauls the Supreme Court’s test. The majority goes on to candidly acknowledge that, in its view, “every Second Amendment case involves dramatic technological changes.” One can imagine now the reasoning our court will employ in the future to substantiate this pronouncement. Cases like Rahimi—with a focus on domestic violence—will come to stand for the proposition that all modern societal ills even conceivably related to firearms are “dramatic [societal] changes” justifying the majority’s sloppier, “more nuanced approach.” And any appreciable advance in firearms technology—whether that is the automatic cycling mechanism, higher ammunition capacity, better optics, an improved barrel, etc.—will come to represent a “dramatic technological change” that the Founders could not possibly have comprehended. The inevitable consequence—and presumably the intended effect—of such a rule is that circumstances will always be different enough to justify the “more nuanced approach.” And the “more nuanced approach” will always justify ill-fitting comparators. What is the end effect of this scheme? It essentially writes Bruen out of the United States Reports—at least out here in the Ninth Circuit. Notwithstanding Bruen’s clear command that the government will be held to its burden of showing a “distinctly similar historical regulation” to succeed in a Second Amendment case, 597 U.S. at 24, 26 (emphasis added), the judges of this circuit will always find some reason or another to excuse the government from meeting that burden. E.g., United States v. Perez-Garcia, 115 F.4th

DUNCAN V. BONTA 135

1002, 1011 (9th Cir. 2024) (VanDyke, J., dissenting from the denial of rehearing en banc). To summarize, the majority’s test boils down to whether any new gun regulation was passed to deal with situations that are “especially dangerous.” Why “especially dangerous” as opposed to just dangerous? Perhaps because the majority implicitly recognizes that merely “dangerous” would be too obviously defiant. Dangerousness itself is clearly inherent in the definition of a weapon: “An instrument used or designed to be used to injure or kill someone.” Weapon, BLACK’S LAW DICTIONARY (12th ed. 2024); see also Commonwealth v. Canjura, 240 N.E.3d 213, 221–22 (Mass. 2024) (“In the most basic sense, all weapons are ‘dangerous’ because they are designed for the purpose of bodily assault or defense.”). And because the very purpose of a weapon is to be “dangerous,” the more dangerous a weapon is—whether because of concealability, stopping power, ease of use, rate of fire, or any number of other considerations—the better it is at doing what it was designed to do. A rule that allows a ban on all “dangerous” weapons could thus effectively be a ban on all weapons. Full stop. The majority—and California—know they cannot create a test that bans any firearm that anyone could perceive as “dangerous” because every firearm has the capacity to cause serious harm if used improperly or placed in the wrong hands. So instead, California points to its conclusion that large-capacity magazines are “uniquely dangerous,” and the majority finds a historical underpinning for a ban on all weapons that are “especially dangerous.” Of course, the test for what qualifies as “especially dangerous” looks quite a lot like the majority’s interest- balancing test from the days of Duncan’s past. To begin, the majority evaluates the “how” by explaining that California’s law imposes only a “minimal burden” on a plaintiff interested in exercising his Second Amendment right, which sounds a lot like a cut-and-paste of the first step of intermediate scrutiny. Compare Duncan V, 19 F.4th at 1104, 1108 (“California’s ban on large-capacity magazines imposes only a minimal burden on the exercise of the Second Amendment right.”), with the majority here (“California’s law imposes only a minimal burden on the right of armed self-defense.”). According to the majority, the burden on such an individual is only minimal because the law “imposes no limit whatsoever on the number of magazines a person may own, the number of bullets a person may own, … the number of firearms a person may own, … the number of rounds a person may fire[,] or the number of firearms a person may fire.” Judge Graber, in both the majority opinion and her concurrence, made that point last time too. See Duncan V, 19 F.4th at 1104; id. at 1115 (Graber, J., concurring) (“[A]lternatives nevertheless remain: the shooter may carry more than one firearm, more than one magazine, or extra bullets for reloading the magazine.”). And I already pointed out how that is unrealistic. Id. at 1163 (VanDyke, J., dissenting). The majority also again emphasizes that firing more than ten rounds “occurs only rarely, if ever, in armed self-defense.” My colleagues made this same argument last time, and I explained then why it doesn’t hold water. Id. at 1167. Nevertheless, as before, the majority concludes that the large-capacity magazine ban is minimally burdensome because it only “prohibit[s] one specific and rare use of semi-automatic firearms.” Compare with id. at 1104 (majority op.) (“The ban on large-capacity magazines has the sole practical effect of requiring shooters to pause for a few seconds after firing ten bullets.”).

DUNCAN V. BONTA 137

Next, the majority goes on to evaluate the “why,” determining that California’s law is similar to its historical analogues because “[m]ass shootings are devastating for the entire community, and large-capacity magazines exacerbate the harm.” This again looks quite a lot like the discredited intermediate scrutiny test that the majority applied the last time around. Id. at 1109–11. Because California’s law seeks “to prevent or mitigate … devastating harm … to innocent persons” due to “especially dangerous uses of weapons,” California’s law must be justified. See id. at 1109 (“California’s law aims to reduce gun violence primarily by reducing the harm caused by mass shootings.”). Is it just me, or do we seem to be right back where we were before Bruen? Except somehow worse. In important ways, the majority’s lax historical balancing test is even easier for the government to satisfy than intermediate scrutiny. Here, the majority discusses the burden on an individual’s right to self-defense, just like it did in Duncan

V. But if the majority can already point to a historical

analogue of a complete ban on a weapon, why does it need to show a minimal burden at all? Really, so long as it has a matching “why”—and there always will be a matching “why” because the inherently dangerous nature of firearms will always match the majority’s permissive “especially dangerous” level of generality—any new law could similarly burden the individual’s right to self-defense in the same way as any ban at the Founding. Cf. Canjura, 240 N.E.3d at 221– 22 (recognizing that all weapons are inherently dangerous). More glaringly, when considering whether the risk of harm is justified, the majority weighs everything in favor of the government and ignores the always-corresponding burden that the ban imposes on law-abiding citizens. Like every part of a firearm, large-capacity magazines are of course “especially dangerous” in the hands of a criminal. So is having a semi-automatic instead of a single-shot handgun. And so is having good sights that help the criminal hit what he’s aiming at. On the other hand, not having good sights decreases the usefulness for law-abiding citizens. And so does having a lower capacity magazine and being forced to reload when put in a lawful self-defense situation. Take the majority’s thrice-repeated line: “The short pauses when a shooter must reload a firearm afford intended victims and law enforcement officers a precious opportunity to flee, take cover, and fight back.” But what of armed victims attempting to defend themselves and others? They too must pause to reload, and their pauses give assailants time to get off more shots. Not to mention the greater potential for malfunction with every magazine swap. And that is only if the victim happens to be carrying an extra magazine in a place where she can quickly get to it. Pauses in shooting don’t just mean a chance for victims to take cover. Pauses in shooting while trying to reload also mean a chance for victims to be overwhelmed by criminal assailants. It always works both ways. Yet the majority sees only a one-way street, claiming that the need to reload in a self-defense situation “seldom” occurs. We’ve been down this road before, too. Statistically, mass shootings almost never occur either. Duncan V, 19 F.4th at 1160 (VanDyke, J., dissenting). But for the majority, extremely rare criminal acts count against the Second Amendment while similarly rare self-defense needs are irrelevant. The majority’s assumption that the need to reload in a self-defense situation “almost never” happens in real life

DUNCAN V. BONTA 139

may not be as justified as it thinks, though, and becomes even less so in a modern society increasingly plagued by unchecked group violence. There is at least one disturbing example from right here in the Ninth Circuit—indeed, not far from where we heard oral argument in this case. In October 2021, retired police captain Ersie Joyner was pumping gas in Oakland when a group of assailants attempted to rob him at gunpoint in broad daylight. 4 The assailants pointed their guns and repeatedly told each other to shoot Joyner, even as he complied with all their demands. 5 Joyner was carrying a Glock 43, and eventually made the choice to defend himself. He fired at the assailants, and they returned fire. Joyner fired ten times, emptying his magazine. He can be seen on the gas station’s security footage having to then take cover and pretend to shoot back while the assailants continued to shoot him at close range before finally driving away. Remarkably, Joyner survived. But he was shot multiple times before the assailants fled, leaving twenty-two bullet holes in his body. Again, that is just one anecdote, and I don’t dispute that these situations are uncommon. 6 But as I explained in my

4 See KTVU Newsroom, Retired Oakland police captain wounded, 1 other killed during gas station gun battle, KTVU Fox 2 (Oct. 22, 2021, 5:30 AM), https://perma.cc/8E35-Z8SB. 5 See Lisa Fernandez & Andre Senior, Ersie Joyner ‘humbled and humanized’ after surviving 22 bullet wounds in Oakland shootout, KTVU Fox 2 (Feb. 26, 2022, 6:26 AM), https://perma.cc/54ZA-BVF7. 6 Indeed, as I’ve emphasized before, all self-defense uses of firearms are relatively uncommon. Duncan V, 19 F.4th at 1160 (VanDyke, J., dissenting) (noting “the practical infrequency of any particular person’s need to actually defend herself with a gun”). So the uncommonness of the need to use a firearm for self-defense cannot be a reason to deny the Second Amendment’s protections. previous Duncan en banc dissent, so are all instances where individuals need to actually fire a gun to defend themselves—including against mass shootings, which form the majority’s and California’s shared rationale for these bans. Id. As I stated then, the standard cannot be based on the “practical infrequency of any particular person’s need to actually defend herself with a gun,” or on the practical infrequency of her need to use multiple rounds to do so. Id. We shouldn’t be balancing the Second Amendment at all. But if the majority can’t help itself, it should at least stop loading the scales. Watering down the history-and-tradition test the way that the majority does here creates real consequences for real people exercising their Second Amendment rights.

III.

Finally, I must respond to Judge Berzon’s concurrence attacking at some length the video portion of this dissent as “wildly improper.” She levels three criticisms: (1) that our court’s rules don’t allow for part of my dissent to be presented in video format, (2) that I have “egregiously … appointed [my]self as an expert witness in this case,” and (3) that my video improperly introduces “facts outside the record.” I’ll respond to each of these accusations in turn. Demonstrating the majority’s consummate textualist bona fides, Judge Berzon’s first criticism starts and ends with the text of our court’s General Orders: “[T]he determination of each appeal … shall be evidenced by a written disposition.” 9th Cir. Gen. Ord. 4.5(a) (emphasis added). Judge Berzon emphasizes “written,” and I’m never one to dispute that words can be “a real workhorse when italicized,” particularly in Second Amendment cases. McDougall v. Cnty. of Ventura, 23 F.4th 1095, 1122 n.5 (9th

DUNCAN V. BONTA 141

Cir. 2022) (VanDyke, J., concurring). But emphasizing one word doesn’t license us to ignore the rest of the text. General Order 4.5(a) doesn’t even say that “the determination of each appeal” shall be “in writing,” much less that it shall be entirely or solely in writing. It says only that the “determination of each appeal … shall be evidenced by a written disposition.” 9th Cir. Gen. Ord. 4.5(a) (different emphasis added). It should be self-evident that if the rule requires only that “the determination … be evidenced by a written disposition” then it doesn’t require that it be “a written disposition”—just evidenced by one. In other words, our court can’t just issue an oral ruling from the bench disposing of a case that is never memorialized—i.e., “evidenced”—in writing. The administrative need for such a rule is obvious enough. My dissent clearly is “evidenced by” a written disposition. Much of the dissent is actually written, and this written portion evidences (i.e., refers to) the oral portion. And even if the rule required that the disposition itself be written, that too would be satisfied by my dissent—which again, is written in part. Indeed, only if the rule unambiguously required that the “determination of each appeal” be only in writing would Judge Berzon’s criticism have any merit. But aside from running squarely into the phrase “evidenced by,” such an extreme reading of General Order 4.5(a) would also be inconsistent with our court’s established practice. We have long included links to videos in our court’s opinions, as well as pictures, timelines, and diagrams. Nobody thought that was a problem until now, and Judge Berzon even defends that practice in her concurrence. In short, Judge Berzon’s overreading of General Order 4.5(a) is just that—an overreading. Like the majority’s invention of its facile arms–accessory test, the “textual” argument against my video dissent seems driven more by a desire for a certain result than anything in the text or reason itself. Most of Judge Berzon’s withering fire, however, is directed at the perception that I’ve made myself a factual expert in this case. First, I would be remiss if I didn’t say thank you. But as much as I may be flattered, I think the accusation misses the mark—indeed, I think my colleagues aren’t even aiming at the right target. My criticism of the majority’s reliance on the arms–accessory distinction to decide this constitutional case is fundamentally a conceptual one, not a factual one. As already noted, it has nothing to do with any unique characteristics of firearms per se, but is rather an intrinsic conceptual shortcoming with the majority’s ill-advised approach that makes a fundamental right turn on whether some object has certain “inherent” qualities or is instead an “unnecessary” add-on to the Platonic ideal of some category. Illustrating that conceptual shortcoming with the majority’s approach doesn’t necessarily require any factual “expertise” about firearms. It just requires a certain level of logical and analytical rigor combined with good judgment in not creating clearly inadministrable constitutional tests—precisely the type of legal expertise we expect in our jurists. So again, thank you. Judge Berzon’s related accusation that the video portion of my dissent introduces “facts outside the record” is misguided for the same reason. Again, the fundamental purpose of the video is to convey a conceptual point, not any particular disputable facts about guns. The same conceptual point could have been illustrated in video form using essentially any tangible object. I could, for example, have referred to the variety in foot types or bobbin styles on a sewing machine to illustrate the inherent indeterminability

DUNCAN V. BONTA 143

in making the majority’s inappropriate legal test turn on whether part of an object is an “integral part” or merely an “accessory.” Or I could have stood by a car and talked about tires and windshield wipers. The factual specifics of how any particular parts work on any particular object is not what is important—it’s the conceptual point that matters. But this is a case about guns, after all, and the Constitution protects the right to bear arms, not cars or sewing machines. So it seems appropriate to use firearms to illustrate my conceptual criticism. The majority’s odd obsession with the factual content of the video—while intentionally blinding itself to my conceptual point— appears to be a bad case of intentionally avoiding the forest by fixating on the trees. There are several strong indicators that Judge Berzon’s and the majority’s “facts outside the record” complaint about my video dissent is just a manufactured concern. First, if you have watched the video portion of my dissent and also read up to this point you are no doubt aware that the written portion of my dissent makes the same conceptual argument as the video: it talks about the same firearms parts except in written form. Yet the majority has never complained that the written portion of my dissent “includes facts outside the record.” The difference between the two formats (written and video) is not the supposed factual content, but rather that for some reason the video format is harder to ignore. So the majority has fabricated a sham procedural reason to justify ignoring it anyway. The majority’s newfound punctiliousness for scrupulously avoiding any reference to facts outside the record would perhaps ring truer if it was evenly applied in Second Amendment cases. Only a few years ago, the same judge who has authored the majority opinion in this case authored an opinion in another case denying Second Amendment rights—and relied extensively on extra-record facts in doing so. See Mai v. United States, 952 F.3d 1106, 1117 & n.6, 1118 & n.7, 1121 (9th Cir. 2020) (Graber, J.) (relying on extra-record studies, including “[i]n other contexts” like smoking, to support the majority’s conclusion). Nobody in today’s majority batted an eye. See Mai v. United States, 974 F.3d 1082, 1082–83, 1097 (9th Cir. 2020) (en banc) (containing multiple dissents from the denial of rehearing en banc without a single member of today’s majority joining). The majority doesn’t dispute (because it can’t) that the Mai panel relied on extra-record materials to directly support the outcome in that case. Instead, the majority belatedly attempts to justify that reliance because the extra-record materials existed in “publicly available scientific studies” and because “the parties had asked [the court] to assess the scientific evidence.” Okay. Whatever post-hoc rationalization the majority offers now, nothing alters that the Mai panel felt free, so long as it was rejecting a Second Amendment claim, to cite and directly rely on facts that were neither in the record nor cited by the parties. In contrast, as I’ve now explained at length, the conceptual point I’m making in both the written and oral portions of this dissent in no way relies on any specific or unique facts—“scientific” or otherwise. The video portion of my dissent doesn’t engage in any factfinding. It makes a broad conceptual point that can be easily illustrated by a literal universe of commonly known objects without being tethered to any specific facts about those objects. But if my colleagues were genuinely bothered by referencing non- record facts in Second Amendment cases, maybe they would

DUNCAN V. BONTA 145

have voted for en banc review in Mai—a case where the panel expressly relied on non-record facts to drive the outcome in that case. 7 There is also a heads-I-win, tails-you-lose quality to the majority’s crocodile tears over the supposed non-record facts in my video dissent. Remember, it is the majority, urged on by California, that has introduced a plainly conceptually flawed but supposedly fact-based constitutional test, and then purported to invent a farcical factual distinction to support its constitutional conclusion. It cannot be the case that, when judges make up such conceptually flawed constitutional tests, the further the invented test is from factual reality the more insulated it is from criticism. The majority’s real beef with my video is not that it introduces any new facts, but that it unmasks their invented constitutional test as obviously grounded in a factual fantasy. It would be one thing for me to introduce the majority’s fact-based constitutional concept for the first time in my

7 The panel’s reliance on extra-record materials in Mai was neither tangential nor “offhand.” The panel directly and repeatedly relied on the extra-record materials as factually supporting its decision. Yet the majority now attempts to partially justify that reliance as merely taking “judicial notice” because it supposedly only “took notice of the existence of evidence of a particular sort, regardless of its accuracy.” But in Mai itself, the panel never attempted to make such a claim. And for good reason—it would have been a transparent misstatement. It is obvious to anyone reading the Mai decision that the panel there was relying on the substance of the claims made in extra-record materials, not just their mere existence. See, e.g., Mai, 952 F.3d at 1117 (“The authors found that studies of persons released from involuntary commitment reported a combined ‘suicide risk 39 times that expected.’ That extraordinarily increased risk of suicide clearly justifies the congressional judgment ….” (citation and footnote omitted)). opinion, and then use my own (or someone else’s) firearms knowledge to show why that concept belongs in our Second Amendment jurisprudence. It is quite another for the majority to purportedly rely on that factual concept, introduce it into our jurisprudence, and then complain only when I show the concept is completely divorced from reality. California and the majority came up with their silly conceptual test, not me. If some extra-record facts about what is actually “integral” to firearms and what is a non- integral “accessory” have entered the picture, that is not due to my response, whether expressed via video or written word. It is because the majority invited us to analyze a nonexistent reality. Don’t shoot the messenger simply for showing that this reality doesn’t exist. Ultimately, however, any debate over my supposedly introducing facts is just a distraction. The majority is trying to manufacture a controversy over the medium I chose to make my point. But the force of my argument is the same regardless of the format. The majority’s new test was never based on some deep factual understanding gleaned from experts and well-observed reality. It was concocted based on the majority searching for some way—any way—to declare high-capacity magazines not protected by the Second Amendment. If nothing else, my colleagues are at least consistent because doing so is in line with this court’s long tradition of finding a way to neuter the Second Amendment under whatever test the Supreme Court directs us to apply. Now the majority projects onto my dissent its insecurities about the very flawed factual concept it contrived to do so. If you don’t like the video portion of my dissent, then don’t watch it. But don’t let that distract you from grasping the conceptual absurdity of the novel test the

DUNCAN V. BONTA 147

majority has concocted to, once again, justify its refusal to apply the Second Amendment. * * * To sum it up: the majority’s rationale in this case, followed to its (il)logical conclusion, means that now— perhaps even more so than before Bruen—only the jankiest guns are even facially protected by the Second Amendment. And even those can be banned outright consistent with the Second Amendment so long as the government can find a historical analogue with the flimsiest connection to the challenged law. Despite the Supreme Court’s intervention, we’re right back where we started when it comes to the Second Amendment, “trimm[ing] back that right at every opportunity.” Duncan V, 19 F.4th at 1172 (VanDyke, J., dissenting). Except worse. It sadly seems our court has somehow now established an even more government- friendly version of the very interest balancing the Supreme Court rejected in Bruen. In doing so today, this court once again improves its undefeated record against the Second Amendment, demonstrating both its misunderstanding of firearms and its disdain for the People’s constitutional right to have them in the process. And once again, I respectfully dissent.