The opinion
In the
United States Court of Appeals
For the Seventh Circuit
____________________
Nos. 24-3060, 24-3061, 24-3062 & 24-3063
CALEB BARNETT, et al.,
Plaintiffs-Appellees,
v.
KWAME RAOUL, Attorney General of
the State of Illinois, et al.,
Defendants-Appellants.
____________________
Appeals from the United States District Court for the
Southern District of Illinois.
Nos. 3:23-cv-209, 3:23-cv-141, 3:23-cv-192 & 3:23-cv-215 —
Stephen P. McGlynn, Judge.
____________________
ARGUED SEPTEMBER 22, 2025 — DECIDED JULY 9, 2026
____________________
Before BRENNAN, Chief Judge, and EASTERBROOK and ST.
EVE, Circuit Judges.
ST. EVE, Circuit Judge. In 2023, six months after a mass
shooting at a Chicago suburb’s Independence Day parade left
seven dead and dozens more wounded, Illinois enacted the
Protect Illinois Communities Act. Among other things, the
2 Nos. 24-3060 et al.
Act criminalizes the manufacture, sale, delivery, purchase,
and possession of assault weapons and large-capacity maga-
zines. A grandfather clause permits preexisting lawful own-
ers of the regulated items to continue possessing them.
Plaintiffs across Illinois swiftly challenged the Act, suing
state and local officials for declaratory and injunctive relief
protecting their right to keep and bear arms. One federal court
granted a preliminary injunction, two did not, and all three
losing parties appealed. We consolidated their appeals and
held in Bevis v. City of Naperville, 85 F.4th 1175 (7th Cir. 2023),
that the plaintiffs were unlikely to succeed on the merits of
their challenges.
One of the consolidated cases in Bevis was Barnett v. Raoul,
which itself was the lead case for a set of four similar chal-
lenges pending in the Southern District of Illinois. After build-
ing out the record following Bevis, the Barnett parties pro-
ceeded to a bench trial. In the end, the district court held that
much of the Act violated the Second Amendment and that the
offending provisions were not severable. The court therefore
enjoined enforcement of the Act in its entirety. The defend-
ants appealed.
For reasons that follow, we focus on the Act’s application
to AR-15s and thirty-round rifle magazines. The Act’s re-
strictions on these items, we hold, are consistent with the prin-
ciples that underpin our Nation’s tradition of firearm regula-
tion. Whether to adopt them is thus a decision reposed in our
elected representatives, and we reverse.
Nos. 24-3060 et al. 3
I. Background
A. The Protect Illinois Communities Act
The Protect Illinois Communities Act, see Pub. Act 102-
1116 (2023), 2022 Ill. Laws 8833, covers a broad array of mat-
ters, both substantive and administrative, but its restrictions
on weapons, parts, accessories, and the like are the subject of
the disputes before us.
Beginning with the restrictions on firearms, the Act makes
it unlawful to knowingly carry, possess, manufacture, sell, de-
liver, import, or purchase any assault weapon. 720 ILCS 5/24-
1.9(b)–(c); see 720 ILCS 5/24-1(a)(15)–(16), (b). Given the range
of conduct proscribed, the Act effectively amounts to a ban.
That ban, however, does not apply to qualified law enforce-
ment officers, members of the military performing official du-
ties, and other similar groups. 720 ILCS 5/24-1.9(e).
The Act defines four types of weapons as assault weapons.
First are semiautomatic rifles that can accept a detachable
magazine and have at least one of the following features: a
pistol grip or thumbhole stock; a protruding grip that can be
held by the non-trigger hand; a folding, telescoping,
thumbhole, or detachable stock; a flash suppressor; a grenade
launcher; or a barrel shroud. Id. 5/24-1.9(a)(1)(A). Second and
third, the Act bans semiautomatic pistols and shotguns that
can accept detachable magazines and have at least one feature
from enumerated lists, the details of which are not relevant
here. Id. 5/24-1.9(a)(1)(C), (F). Finally, the Act bans certain fire-
arms based on the type of ammunition feeding device they
utilize—for example, semiautomatic weapons that can accept
a belt ammunition feeding device. Id. 5/24-1.9(a)(1)(B), (D)–
(E), (G). As with the expired federal assault weapon ban from
4 Nos. 24-3060 et al.
which Illinois borrowed, the Act defines assault weapons not
only by reference to these features but also through a list of
prohibited models, including the AR-15. Id. 5/24-1.9(a)(1)(J)–
(L). The Act also bans “assault weapon attachments”—that is,
“any device capable of being attached to a firearm that is spe-
cifically designed for making or converting a firearm into” an
assault weapon. Id. 5/24-1.9(a)(3), (b)–(c).
The Act next criminalizes the knowing manufacture, de-
livery, sale, purchase, or possession of “large capacity ammu-
nition feeding devices.” Id. 5/24-1.10(b)–(c), (g). That term is
defined to include “a magazine, belt, drum, feed strip, or sim-
ilar device that has a capacity of” more than ten rounds for
rifles and shotguns, and more than fifteen rounds for hand-
guns. Id. 5/24-1.10(a)(1). We will focus, as the parties have, on
magazines, the most common type of ammunition feeding de-
vice. And we will refer to the restricted ones as “large-capac-
ity magazines.”
Finally, the Act bans certain .50 caliber rifles and their car-
tridges, as well as any device or accessory—like a bump
stock—“that is designed to and functions to increase the rate
of fire of a semiautomatic firearm above the standard rate of
fire for semiautomatic firearms.” Id. 5/24-1.9(a)(6), 5/24-
1(a)(14).
The Act’s grandfather clauses provide significant excep-
tions to these restrictions. As to assault weapons, assault
weapon attachments, .50 caliber rifles, or .50 caliber car-
tridges, preexisting lawful owners could continue possessing
them if they provided the Illinois State Police an “endorse-
ment affidavit” containing certain information by January 1,
2024. Id. 5/24-1.9(d). The required information consisted of the
owner’s firearm license number, an affirmation that he pos-
Nos. 24-3060 et al. 5
sessed the item before the Act went into effect, and the make,
model, caliber, and serial number of the item. Id. 5/24-
1.9(d)(1)–(3). This registration process, which was free, “cre-
ate[d] a rebuttable presumption that the person [was] entitled
to possess and transport the” restricted item. Id. 5/24-1.9(d)(3).
As to large capacity ammunition feeding devices, preexisting
lawful owners could continue possessing them (subject to lo-
cational restrictions) with no registration required. Id. 5/24-
1.10(d). Those who move to Illinois today and wish to possess
the restricted items may avail themselves of the grandfather
clauses by applying for an Illinois firearm license and (for eve-
rything but the ammunition feeding devices) completing an
endorsement affidavit within sixty days of moving. Id. 5/24-
1.9(d), 5/24-1.10(d).
B. Procedural History
Four related cases, which the district court consolidated,
are now before us. See Langley v. Kelly, No. 3:23-cv-192 (S.D.
Ill.); Harrel v. Raoul, No. 3:23-cv-141 (S.D. Ill.); Barnett v. Raoul,
3:23-cv-209 (S.D. Ill.); Federal Firearms Licensees of Ill. v. Pritzker,
No. 3:23-cv-215 (S.D. Ill.). The plaintiffs consist of individuals,
participants in the commercial firearm market, and organiza-
tions that advocate for Second Amendment rights. The de-
fendants are a wide range of state and local officials charged
with enforcing the Act in one way or another.
Each of the plaintiffs, relying on 42 U.S.C. § 1983, chal-
lenged the Act’s ban on assault weapons, large-capacity mag-
azines, and assault weapon attachments under the Second
Amendment. The Langley and Federal Firearms Licensees plain-
tiffs further challenged the Act’s ban on .50 caliber rifles and
6 Nos. 24-3060 et al.
cartridges. Only the Langley plaintiffs attacked the Act’s reg-
istration requirement. 1
After each set of plaintiffs moved for a preliminary injunc-
tion, the district court consolidated the four actions, designat-
ing Barnett as the lead case. It then granted the plaintiffs pre-
liminary relief. See generally Barnett v. Raoul, 671 F. Supp. 3d
928 (S.D. Ill. 2023). But in functionally identical Second
Amendment challenges, two district judges in the Northern
District of Illinois came out the other way. See generally Bevis
v. City of Naperville, 657 F. Supp. 3d 1052 (N.D. Ill. 2023); Her-
rera v. Raoul, 670 F. Supp. 3d 665 (N.D. Ill. 2023). So when all
three losing parties appealed, we consolidated the three cases
to unify the Act’s status across Illinois and to guide the district
courts and parties in litigating the cases to final judgment. See
generally Bevis, 85 F.4th 1175.
When we decided Bevis, no circuit court had applied the
Supreme Court’s seminal decision in New York State Rifle &
Pistol Ass’n v. Bruen, 597 U.S. 1 (2022), to assault weapons and
large-capacity magazines. In Bruen, the Supreme Court estab-
lished a new test governing Second Amendment claims:
“When the Second Amendment’s plain text covers an individ-
ual’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.” Id. at 24.
Bevis sought to flesh out the Bruen inquiry, particularly its
first step. Based on an analysis of District of Columbia v. Heller,
1 The Langley plaintiffs also asserted a due process challenge below,
but the district court granted summary judgment to the defendants on that
claim. The Langley plaintiffs did not cross-appeal that decision.
Nos. 24-3060 et al. 7
554 U.S. 570 (2008), Bevis concluded that Bruen’s first step re-
quired “the plaintiffs in each of the cases before us [to show]
that the weapons addressed in the pertinent legislation are
Arms that ordinary people would keep at home for purposes
of self-defense, not weapons that are exclusively or predomi-
nantly useful in military service, or weapons that are not pos-
sessed for lawful purposes.” Bevis, 85 F.4th at 1194. If a regu-
lated weapon meets this definition, it is an “Arm” covered by
the Second Amendment’s plain text, satisfying Bruen’s first
step; if not, the regulation is constitutional without need to
evaluate our historical tradition of firearm regulation.
Applying this test, Bevis concluded the plaintiffs were not
likely to succeed on the merits of their Second Amendment
challenges because AR-15s (which the court used as repre-
sentative of the banned weapons) and large-capacity maga-
zines “are much more like machineguns and military-grade
weaponry than they are like the many different types of fire-
arms that are used for individual self-defense (or so the legis-
lature was entitled to conclude).” Id. at 1195. AR-15s, Bevis
concluded, are not “materially different” from M16s—which
Heller confirmed may be banned. Id. at 1195–97; see Heller, 554
U.S. at 627. To be sure, Bevis recognized that AR-15s are lim-
ited to semiautomatic fire, whereas M16s are capable of both
semiautomatic and automatic fire. See Bevis, 85 F.4th at 1195–
96. And Bevis recognized this difference in firing modes trans-
lates to a difference in firing rates. Id. at 1196. But Bevis ulti-
mately concluded that the distinctions between AR-15s and
M16s paled in comparison to their similarities, placing AR-15s
beyond the Second Amendment’s protections.
After assuming that the regulated items were “Arms” un-
der the Second Amendment, Bevis went on to hold that the
8 Nos. 24-3060 et al.
Act also passed muster under Bruen’s second step. Our Na-
tion’s history of firearm regulation, Bevis concluded, revealed
a tradition of reserving especially dangerous weapons for mil-
itary use while leaving many other weapons available for ci-
vilians. See id. at 1197–1202. And the Act “respects and relies
on” that tradition. Id. at 1202. We therefore vacated the pre-
liminary injunctions that the district court had entered in Bar-
nett, Harrel, Langley, and Federal Firearms Licensees. See id. at
1203. 2
Following Bevis’s direction, the plaintiffs and defendants
in Barnett and the consolidated cases developed the record.
The parties stipulated to presenting much of the evidence on
the papers, but the court also held a four-day bench trial. At
the conclusion, the district court held in a thorough opinion
that the Act’s ban on assault weapons, assault weapon attach-
ments, and large-capacity magazines violated the Second
Amendment, as did the registration requirement.
Tackling Bevis’s “Arms” inquiry, the district court first con-
cluded that ordinary citizens choose assault weapons, large-
capacity magazines, and assault weapon attachments for self-
defense. With respect to the firearms themselves, the court re-
lied on both statistical evidence of these items’ wide circula-
tion and testimony from self-defense experts, firearm instruc-
tors, and a gun store owner. As for the large-capacity maga-
zines, the court noted that “every round matters in a self-de-
fense scenario.” And the attachments, the district court rea-
soned, are “well-suited for self-defense,” especially for “an in-
dividual who is infirm, small-statured, or has limited firearms
2 The plaintiffs filed a petition for certiorari, which the Supreme Court
denied. See Harrel v. Raoul, 144 S. Ct. 2491 (2024).
Nos. 24-3060 et al. 9
training.” The district court concluded, by contrast, that law-
abiding citizens would not choose .50 caliber rifles, their am-
munition, .50 caliber pistols, belt-fed weapons, and grenade
launchers for self-defense, so these items failed to qualify for
presumptive constitutional protection under Bevis.
Moving to whether the regulated items are exclusively or
predominantly useful in military service, the district court
concluded that AR-15s are materially distinct from the M16
rifles that the U.S. military issues. The court first stressed that
no military has ever issued AR-15s to its troops, largely be-
cause of their different firing modes: AR-15s can fire only sem-
iautomatically, whereas M16s are capable of semiautomatic,
automatic, and burst fire. 3 The court stressed, too, that AR-15s
are not “subject to exact standards of military specificity and
rigorous quality-insurance [sic] inspections.” The court fur-
ther held that large-capacity magazines and assault weapon
attachments are not predominantly useful in military service
because the military does not issue them for use in combat.
And finally, the court concluded that people do not possess
AR-15–style semiautomatic rifles, large-capacity magazines
holding up to thirty rounds, and assault weapon attachments
(other than grenade launchers) for unlawful purposes be-
cause most of the weapons are not used in illegal activity.
3 Though Bevis had identified the difference in firing rates between
AR-15s and M16s as important, the district court did not expressly make
a finding on this matter. It came close, however, in a footnote—located in
the court’s summary of circuit caselaw—where it cited a U.S. Army man-
ual stating that an M16 in automatic mode has a maximum effective firing
rate of 150–200 rounds per minute, whereas in semiautomatic mode it has
a maximum effective firing rate of 45–65 rounds per minute.
10 Nos. 24-3060 et al.
Having concluded that the items the Act regulates qualify
as “Arms” for Second Amendment purposes under Bevis, the
court proceeded to Bruen’s second step, where it held that the
defendants failed to establish that the Act is consistent with
our Nation’s history of firearm regulation. At this stage, the
court did not make or rely on factual findings undercutting
the applicability of Bevis’s historical assessment. Instead,
looking at essentially the same record as existed at the prelim-
inary injunction stage, the district court found Bevis’s dissent-
ing opinion more persuasive than its majority, holding that
the historical analogues on which Bevis relied were insuffi-
cient to sustain the Act.
Because the district court concluded that the offending
provisions of the Act were not severable, it enjoined the Act
in its entirety. The court’s permanent injunctions (one for each
of the four cases), entered under Federal Rule of Civil Proce-
dure 65, enjoined “the State of Illinois” from enforcing the Act
against anyone. 4 The defendants appealed, and we granted
their request to stay the district court’s injunctions pending
appeal.
II. Discussion
A. Standard of Review
We begin with our standard of review. We review the dis-
trict court’s decision to grant a permanent injunction for
abuse of discretion. See eBay Inc. v. MercExchange, L.L.C., 547
4 Enjoining enforcement of the Act against anyone—i.e., entering a uni-
versal injunction—became problematic after the Supreme Court’s deci-
sion in Trump v. CASA, Inc., 606 U.S. 831, 837 (2025), but our conclusion in
these appeals renders that flaw moot.
Nos. 24-3060 et al. 11
U.S. 388, 391 (2006); Republic Techs. (NA), LLC v. BBK Tobacco
& Foods, LLP, 135 F.4th 572, 587 (7th Cir. 2025). On the merits,
however, our standard of review is somewhat murkier.
Though it is clear (and undisputed) that we review “the un-
derlying question of constitutional law de novo,” Schoenthal v.
Raoul, 150 F.4th 889, 901 (7th Cir. 2025), it is less clear (and is
disputed) how much deference, if any, we owe the district
court’s findings of fact.
As a general matter, Federal Rule of Civil Procedure 52(a)
prescribes that on appellate review of a bench trial,
“[f]indings of fact … must not be set aside unless clearly erro-
neous.” That standard is satisfied only if we are “left with the
definite and firm conviction that a mistake has been commit-
ted.” Anderson v. City of Bessemer City, 470 U.S. 564, 573 (1985).
But the Supreme Court has recognized that appellate review
may be more searching as to legislative facts than adjudicative
facts. See Lockhart v. McCree, 476 U.S. 162, 168 n.3 (1986) (“We
are far from persuaded … that the ‘clearly erroneous’ stand-
ard of Rule 52(a) applies to the kind of ‘legislative’ facts at is-
sue here.”); see also Doe v. Prosecutor, Marion Cnty., 705 F.3d
694, 697 n.4 (7th Cir. 2013). Though the line between the two
can be blurry, roughly speaking, “[l]egislative facts are those
general considerations that move a lawmaking or rulemaking
body to adopt a rule, as distinct from the facts which deter-
mine whether the rule was correctly applied.” Menora v. Ill.
High Sch. Ass’n, 683 F.2d 1030, 1036 (7th Cir. 1982); see also
Frank v. Walker, 773 F.3d 783, 795 (7th Cir. 2014) (Posner, J.,
dissenting) (“The concept of a legislative fact comes into its
own when there is no reason to believe that certain facts per-
tinent to a case vary from locality to locality, or from person
to person; a typical definition of legislative facts is broad, gen-
eral facts that are not unique to a particular case and provide
12 Nos. 24-3060 et al.
therefore an appropriate basis for legislation of general appli-
cation.”). And even for adjudicative facts, an exception to
Rule 52(a) for so-called constitutional facts—ultimate facts in
constitutional cases—might create yet another obstacle to
clear-error review. See Bose Corp. v. Consumers Union of U.S.,
Inc., 466 U.S. 485, 510–14 (1984) (making an independent de-
termination of “actual malice” for First Amendment pur-
poses); A Woman’s Choice-E. Side Women’s Clinic v. Newman,
305 F.3d 684, 689 (7th Cir. 2002) (explaining that de novo re-
view of constitutional facts “prevent[s] the idiosyncrasies of a
single judge or jury from having far-reaching legal effects”).
We need not wade into these murky waters, however. In-
stead, “[b]ecause we do not ultimately base our decision to-
day on the invalidity of the lower court[’s] factual findings,
we need not decide the standard of review issue.” McCree, 476
U.S. at 168 n.3 (citation modified). We therefore assume that
the district court’s factual findings are shielded by clear-error
review. 5
B. The Plaintiffs’ Facial Challenges
Each of the plaintiffs presses a facial challenge against the
Act’s operative provisions in 720 ILCS 5/24-1.9(b)–(c) and
5/24-1.10(b)–(c). The former provisions make it unlawful to
5 The defendants separately contend that the district court did not
make any factual findings under Rule 52(a)(1)—which requires “the
court” to “find the facts specially and state its conclusions of law sepa-
rately”—because its opinion contained only introduction, background,
and analysis sections. But substance, not form, determines compliance
with Rule 52(a)(1). See Bartsh v. Nw. Airlines, Inc., 831 F.2d 1297, 1304 (7th
Cir. 1987); see also Valsamis v. Gonzalez-Romero, 748 F.3d 61, 63 (1st Cir.
2014). The district court made factual findings, even if it did not specifi-
cally delineate them.
Nos. 24-3060 et al. 13
“knowingly manufacture, deliver, sell, import, or purchase …
an assault weapon, assault weapon attachment, .50 caliber ri-
fle, or .50 caliber cartridge,” 720 ILCS 5/24-1.9(b), or “to know-
ingly possess” these items, id. 5/24-1.9(c). The latter provisions
make it unlawful “to knowingly manufacture, deliver, sell,
[or] purchase … a large capacity ammunition feeding de-
vice,” id. 5/24-1.10(b), or “to knowingly possess” one, id. 5/24-
1.10(c). 6
The plaintiffs’ decisions to pursue facial claims “come[] at
a cost,” as the Supreme Court has “made facial challenges
hard to win.” Moody v. NetChoice, LLC, 603 U.S. 707, 723 (2024);
see Wash. State Grange v. Wash. State Republican Party, 552 U.S.
442, 450–51 (2008) (explaining that facial challenges are “dis-
favored” because they “often rest on speculation,” “short cir-
cuit the democratic process,” and “run contrary to the funda-
mental principle of judicial restraint”). To prevail on a claim
of facial invalidity, the challenger must “establish that no set
of circumstances exists under which the [challenged provi-
sions] would be valid.” United States v. Rahimi, 602 U.S. 680,
693 (2024) (quoting United States v. Salerno, 481 U.S. 739, 745
(1987)). In the context of these appeals, that standard requires
us to side with the defendants as to 720 ILCS 5/24-1.9(b) and
(c) if the Act’s restriction of any “assault weapon, assault
weapon attachment, .50 caliber rifle, or .50 caliber cartridge”
is constitutional. The same goes with respect to 720 ILCS 5/24-
1.10(b) and (c) if the Act’s restriction on any “large capacity
ammunition feeding device” is valid. See, e.g., Bianchi v.
6 The Federal Firearms Licensees plaintiffs also challenge the endorse-
ment affidavit requirement (part of the Act’s grandfather clause), which
we address below.
14 Nos. 24-3060 et al.
Brown, 111 F.4th 438, 452–54 (4th Cir. 2024) (en banc) (exem-
plifying this approach); Capen v. Campbell, 134 F.4th 660, 668–
69 (1st Cir. 2025) (same).
The plaintiffs cannot satisfy this demanding standard. In-
deed, they do not even advance arguments as to some of the
items that 720 ILCS 5/24-1.9 and 5/24-1.10 regulate, all of
which must be constitutionally protected for their facial
claims to be viable. The Act, for example, restricts semiauto-
matic rifles that have grenade launchers attached, 720 ILCS
5/24-1.9(a)(1)(A)(v), yet the district court unsurprisingly held
that the Act may ban these weapons because they are not in
common use for lawful purposes, and no plaintiff argues oth-
erwise on appeal. Likewise, the plaintiffs focus on magazines
holding up to thirty rounds, but 720 ILCS 5/24-1.10’s re-
striction of “large capacity ammunition feeding device[s]”
also includes, among other things, belts and ammunition
feeding devices holding more than thirty rounds. Notwith-
standing their failure to substantiate their attack on every item
that the challenged provisions regulate, the plaintiffs ask that
we enjoin the Act in its entirety. But because the plaintiffs
have not made the showing that Supreme Court precedent re-
quires, “the broad relief their facial challenge seeks is not ours
to grant.” Bianchi, 111 F.4th at 453; see also, e.g., United States v.
Charles, 159 F.4th 545, 547 (8th Cir. 2025) (rejecting facial chal-
lenge to federal machine gun ban because some machine
guns, such as the M230 machine gun mounted on military hel-
icopters, are not bearable and thus lie beyond the Second
Amendment’s scope).
Following the Fourth Circuit’s model under similar cir-
cumstances, we nonetheless proceed to evaluate the Act’s re-
strictions on assault weapons and large-capacity magazines
Nos. 24-3060 et al. 15
based on the circumstances on which the parties have princi-
pally focused: an AR-15 (as representative of the banned ri-
fles) and a thirty-round rifle magazine. 7 See Bianchi, 111 F.4th
at 452–54. As in Bianchi, “the parties thoroughly briefed the
issue of whether the Second Amendment protects a citizen’s
ability to purchase and possess” these items, and moreover,
the parties developed a record focused primarily on them. Id.
at 453–54. “Not to address [them] would be to bypass the very
heart of the dispute in this proceeding.” Id. at 454. It would
also leave the two other cases Bevis consolidated, which re-
main pending in the Northern District of Illinois, in limbo.
C. The Second Amendment
The Second Amendment provides: “A well regulated Mi-
litia, being necessary to the security of a free State, the right of
the people to keep and bear Arms, shall not be infringed.” The
Supreme Court recognized in Heller that these words secure
an individual right to keep and bear arms for self-defense, see
554 U.S. at 576–600—a right that Justice Story called “the pal-
ladium of the liberties of a republic,” 2 Joseph Story, Commen-
taries on the Constitution of the United States 620 (4th ed. Boston,
Little, Brown & Co. 1873). Two years after Heller, the Court
confirmed that the states must respect that right by virtue of
the Fourteenth Amendment. See McDonald v. City of Chicago,
561 U.S. 742, 750 (2010). All Americans, therefore, may enjoy
7 Thus, we do not resolve the constitutionality of the Act’s application
to the pistols and shotguns that the Act defines as assault weapons. Nor
do we address pistol or shotgun magazines that qualify as large-capacity
magazines under the Act. These restrictions, and any others not addressed
in this opinion, are better left for another day and remain open to chal-
lenge on an as-applied basis.
16 Nos. 24-3060 et al.
the “means of self-defense” that “the right secures.” Rahimi,
602 U.S. at 690.
Fundamental as it is, “the right secured by the Second
Amendment is not unlimited.” Heller, 554 U.S. at 626. In this
respect, the right to keep and bear arms is just like other con-
stitutional rights. The Free Speech Clause does not enshrine
an absolute right to express oneself however one sees fit. The
Free Exercise Clause does not protect believers from any and
all burdens on their religious practice. And the Second
Amendment does not secure “a right to keep and carry any
weapon whatsoever in any manner whatsoever and for what-
ever purpose.” Id.; see Wolford v. Lopez, 609 U.S. ----, No. 24-
1046, 2026 WL 1825723, at *5 (June 25, 2026) (“[W]hile the
founding generation cherished the Second Amendment right,
they did not think it was absolute.”).
What does distinguish the Second Amendment is the na-
ture of the standard governing challenges brought under it.
Rejecting means-end scrutiny in favor of history and tradi-
tion, the Court in Bruen set out a two-step framework for ad-
judicating challenges to firearm regulations. Courts first ask
whether “the Second Amendment’s plain text covers an indi-
vidual’s conduct”; if it does, “the Constitution presumptively
protects that conduct.” 597 U.S. at 17. At that point, the burden
shifts to the government to “demonstrate that the regulation
is consistent with this Nation’s historical tradition of firearm
regulation.” Id. Only if the government carries that burden
“may a court conclude that the individual’s conduct falls out-
side the Second Amendment’s ‘unqualified command.’” Id.
(quoting Konigsberg v. State Bar of Cal., 366 U.S. 36, 50 n.10
(1961)).
Nos. 24-3060 et al. 17
In assessing whether a state’s proffered historical ana-
logues satisfy its step-two burden, courts should consider
“the number of jurisdictions in which they were adopted,”
“the extent to which they were well-accepted,” and whether
they are “‘relevantly similar’ to the modern law.” Wolford,
2026 WL 1825723, at *6. Ascertaining whether historical ana-
logues are “relevantly similar” to a modern law in turn “re-
quires consideration of ‘how’ the analogue restricted the
keeping or bearing of arms” and “‘why’ the analogue re-
stricted the keeping or bearing of arms.” Id.
Especially because “states were not bound by the Second
Amendment until the Fourteenth Amendment was ratified in
1868,” we may rely on nineteenth-century statutes to shed
light on the prevailing understanding of the right to keep and
bear arms. Schoenthal, 150 F.4th at 913; see United States v. He-
mani, 608 U.S. ----, No. 24-1234, 2026 WL 1751710, at *6 n.3
(June 18, 2026) (reserving the question “‘whether courts
should primarily rely on the prevailing understanding of an
individual right when the Fourteenth Amendment was rati-
fied in 1868’ or when the Bill of Rights was ratified in 1791”
(quoting Bruen, 597 U.S. at 37–38)); Wolford, 2026 WL 1825723,
at *13 (indicating that the adoption of the Fourteenth Amend-
ment may be a relevant time period for purposes of ascertain-
ing the Second Amendment’s meaning); see also Nat’l Ass’n for
Gun Rts. v. Lamont, 153 F.4th 213, 235 (2d Cir. 2025); Nat’l Rifle
Ass’n v. Bondi, 133 F.4th 1108, 1121 (11th Cir. 2025).
D. AR-15s and Large-Capacity Magazines
We turn now to whether the Act’s restrictions on AR-15s
and thirty-round rifle magazines pass muster. We will assume
for purposes of today’s ruling that the regulated items are
“Arms” under Bevis, such that their possession is presump-
18 Nos. 24-3060 et al.
tively entitled to constitutional protection, and move to
Bruen’s second step. We hold that the Act is consistent with
the principles that underpin our regulatory tradition. In short,
legislatures have long imposed restrictions on particularly
dangerous weapons, and the Act is but another chapter in that
story.
As an initial matter, little has changed since we held in
Bevis that the Act satisfies Bruen’s historical inquiry. See 85
F.4th at 1197–1202. To be sure, the parties developed the rec-
ord in the years since Bevis. But that development—and the
district court’s factual findings—went almost exclusively to
step one. By contrast, the historical laws in the record, and the
parties’ arguments about them, remain essentially the same
as they stood at the preliminary injunction stage. Indeed, no-
table intervening developments favor the defendants. First,
the Court decided Rahimi, which “clarified that the Bruen
standard should not be misunderstood to mean that modern
firearm regulations require close founding-era comparators.”
United States v. Reyna, 165 F.4th 1056, 1062 (7th Cir. 2026). And
second, every circuit to have confronted the issue has agreed
with our conclusion that legislatures may ban AR-15s and
large-capacity magazines. 8 Though we do not rest on this
point, we note that creating a conflict under these circum-
8 See Capen, 134 F.4th at 668–77 (assault weapons and large-capacity
magazines); Ocean State Tactical, LLC v. Rhode Island, 95 F.4th 38, 43–52 (1st
Cir. 2024) (large-capacity magazines); Lamont, 153 F.4th at 235–47 (assault
weapons and large-capacity magazines); Bianchi, 111 F.4th at 446–72 (as-
sault weapons); Duncan v. Bonta, 133 F.4th 852, 865–84 (9th Cir. 2025) (en
banc) (large-capacity magazines); Hanson v. District of Columbia, 120 F.4th
223, 234–43 (D.C. Cir. 2024) (large-capacity magazines).
Nos. 24-3060 et al. 19
stances would be imprudent. See United States v. Tuggle, 4
F.4th 505, 522 (7th Cir. 2021).
In any event, we remain persuaded by the unanimous cir-
cuit consensus. Using one or another label—“dangerous and
unusual,” “unusually dangerous,” “especially dangerous,”
“particularly capable of unprecedented lethality”—these
courts have coalesced around a largely overlapping set of his-
torical regulations imposing targeted restrictions on weapons
whose danger and lethality stand out. These courts have fur-
ther concluded that those regulations justify restrictions on
AR-15s and large-capacity magazines equivalent to those the
Act imposes. See Ocean State Tactical, 95 F.4th at 44–52; Capen,
134 F.4th at 669–73; Lamont, 153 F.4th at 240–47; Bianchi, 111
F.4th at 464–72; Duncan, 133 F.4th at 874–84; Hanson, 120 F.4th
at 234–40. Given the ground already covered, we think it un-
necessary to reiterate the courts’ analyses of each of the stat-
utes comprising this tradition, which spans from the pre-
Founding going-armed laws through today’s machine gun
ban. 9 Although we adopt the courts’ well-reasoned analyses
of these statutes, we focus here instead on a leading example
of this tradition: regulations of the Bowie knife 10—or, as one
Reconstruction-era court called it, the “instrument of almost
certain death.” Cockrum v. State, 24 Tex. 394, 402 (1859).
9 To be clear, we recognize that twentieth century machine gun bans
are “insufficient to support a tradition of regulating [arms] in and of them-
selves.” Hanson, 120 F.4th at 239. They “fit nicely,” however, “into the tra-
dition of regulating weapons particularly capable of unprecedented le-
thality.” Id.
10 The Second Amendment’s plain text is not limited to firearms. See
Caetano v. Massachusetts, 577 U.S. 411, 411–12 (2016) (per curiam).
20 Nos. 24-3060 et al.
Popularized in the 1830s, Bowie knives came in a variety
of forms, but “in its purest form” it was “a large knife with a
clipped point” making the tip more piercing. Harold L. Peter-
son, American Knives: The First History and Collectors’ Guide 26
(1958). These features, combined with the technological short-
comings plaguing the era’s prominent firearms—which re-
quired the user to carefully reload after every shot, such that
one missed shot could leave the user defenseless—made
Bowie knives a popular choice for fights and duels.
Contrary to the dissenting opinion’s efforts (often without
citation) to conclusively link Bowie knives to criminality,
however, Bowie knives were both widespread and used for
lawful purposes. One expert in the history of arms in Amer-
ica, for example, explained that in the nineteenth century,
“European visitors who ventured beyond the Appalachians
found [the Bowie knife] such an integral part of the American
way of life that they felt compelled to comment on it at length
in accounts of their adventures…. In many communities, no
man, whether hunter, gambler, tradesman or political leader
felt himself fully clothed without one.” Peterson, supra, at 25.
Similarly, the historian who (literally) wrote the book on
Bowie knives noted they were “widely carried by Americans
of all stripes”; “served everyone equally, upstanding citizens
and villains” alike; were “wide[ly] popular[], in the North and
South” during the Civil War; and were “common[]”—“a
weapon carried by men of all walks of life.” Norm Flayder-
man, The Bowie Knife: Unsheathing an American Legend 20, 125,
130 (2004). Other scholars agree. See David B. Kopel, Clayton
E. Cramer & Joseph Edward Olson, Knives and the Second
Amendment, 47 U. Mich. J.L. Reform 167, 184 (2013) (“During
the nineteenth century, Bowie knives were commonly present
in many areas of the United States. Contemporary sources
Nos. 24-3060 et al. 21
leave no question that Bowie knives, Arkansas Toothpicks,
and similar knives were a common part of American life until
well after the Civil War….”).
Bowie knives were also “particularly suitable for self-de-
fense” and “typically possessed for self-defense.” Id. at 180,
185; see Peterson, supra, at 26 (“[T]he original knife made for
James Bowie was a large heavy knife suitable for both self-
defense and general utility in the woods.”); Cockrum, 24 Tex.
at 402 (noting that “[t]he gun or pistol may miss its aim, and
when discharged, its dangerous character is lost, or dimin-
ished at least,” whereas the Bowie knife is “difficult to defend
against”). Hunting served as another lawful purpose for
which people owned Bowie knives. See Joseph E. Worcester,
Dictionary of the English Language 165 (1860) (defining the
term as “[a] large knife or dagger, used as a weapon, and car-
ried by hunters in the South-western part of the United
States”); see also Wolford, 2026 WL 1825723, at *4 n.4 (“[M]ost
Americans in the late 18th century also ‘undoubtedly
thought’ that the codified Second Amendment was ‘im-
portant for hunting,’ which, particularly for those moving
west, was an important source of sustenance.” (quoting Heller,
554 U.S. at 599)). But their “large blades … wreaked particu-
larly bloody and gruesome injuries,” Lamont, 153 F.4th at 243
(citation modified), and no doubt criminals misused them.
Legislatures responded accordingly, enacting criminal
prohibitions that often authorized imprisonment. Some juris-
dictions banned the carry of Bowie knives, with narrow ex-
ceptions or none at all—a broad proscription. See 1871 Tex.
Laws 1st Sess. 25; 1881 Ark. Acts 191; 1889 Ariz. Sess. Laws 30.
Others prohibited their concealed carry. See, e.g., 1820 Ind.
Acts 39; 1838 Va. Acts 76; 1838 Tenn. Pub. Acts 200; 1839 Ala.
22 Nos. 24-3060 et al.
Acts 67; 1878 Miss. Laws 175; 1879 N.C. Sess. Laws 231; 1880
S.C. Acts 448. While these restrictions targeted what one did
in public, others burdened one’s ability to own Bowie knives
for use in private too, such as for defense of the home, by ban-
ning the sale of Bowie knives or imposing burdensome, and
sometimes prohibitive, taxes. See 1838 Tenn. Pub. Acts 200
(banning sale); 1881 Ark. Acts 192 (same); 1837 Ala. Laws 7
($100 tax per sale, which is equivalent to several thousand
dollars today); 1838 Fla. Terr. Laws 36 ($200 per year tax on
Bowie-knife sellers and $10 per year on Bowie-knife carriers).
The dissenting opinion seeks to undermine our reliance on
the Texas, Arkansas, and Arizona carry bans, but it misappre-
hends these statutes. The dissenting opinion first claims that
the Texas and Arizona statutes permitted carry for self-de-
fense, but their exceptions were significantly narrower. The
Arizona statute, for example, excepted only “one who has rea-
sonable ground for fearing an unlawful attack upon his per-
son, and the danger is so imminent and threatening as not to
admit of the arrest of the party about to make such attack
upon legal process.” 1889 Ariz. Sess. Laws 30. The Texas stat-
ute employed a similar standard. See 1871 Tex. Laws 1st Sess.
25. In other words, neither statute permitted carry for self-de-
fense generally; they allowed it only for those with a height-
ened and urgent need. Bruen’s holding—striking down a New
York regime under which concealed carry licenses were avail-
able to only those with “a special need for self-protection dis-
tinguishable from that of the general community,” 597 U.S. at
12—underscores that notwithstanding Arizona’s and Texas’s
narrow exceptions, their statutes meaningfully burdened the
right to armed self-defense.
Nos. 24-3060 et al. 23
The dissenting opinion next claims that the Arkansas stat-
ute did not apply to public carry because of its exception for
those “carrying any weapon when upon a journey.” 1881 Ark.
Acts 191. The dissenting opinion ignores the meaning of
“journey” in this statute. If “journey” were synonymous with
“public carry,” the statutory prohibition on “wear[ing] or
carry[ing]” Bowie knives “in any manner whatever” would
be sapped of all meaning. Instead, as the Supreme Court of
Arkansas explained, “journey” meant something far more
narrow. See Carr v. State, 34 Ark. 448, 449 (1879) (“The excep-
tion [for journeys] in [a statute proscribing concealed carry of
a pistol] is to enable travelers to protect themselves on the
highways, or in transit through populous places—not to al-
low them the privilege of mixing with the people in ordinary
intercourse, about the streets, armed in a manner which, upon
a sudden fit of passion, might endanger the lives of others.
Travelers do not need weapons, whilst stopping in towns, any
more than citizens do. They should lay them aside, unless the
delay be slight, and the journey soon resumed.”); Hathcote v.
State, 17 S.W. 721, 722 (Ark. 1891) (“The prohibition was de-
signed to stop the carrying of weapons on the streets, in soci-
ety, and among one’s habitual associates. The exception [for
journeys] was designed to permit it when necessary to defend
against perils of the highway to which strangers are exposed,
and that are not supposed to exist among one’s own neigh-
bors.”).
Contemporaneous court decisions rejected Second
Amendment challenges (or those predicated on state-consti-
tutional analogues) to statutes like these. In Aymette v. State,
for example, the Supreme Court of Tennessee sustained, over
the defendant’s constitutional objection, a conviction for con-
cealed carry of a Bowie knife. See 21 Tenn. 154, 156–62 (1840).
24 Nos. 24-3060 et al.
The Supreme Court of Tennessee later upheld another convic-
tion for concealed carry of a similar knife; though this defend-
ant’s challenge was on statutory, not constitutional, grounds,
the court’s explanation of why the legislature regulated such
knives is instructive: “The design of the statute was to prohibit
the wearing of bowie-knives, and others of a similar descrip-
tion, which the experience of the country had proven to be
extremely dangerous and destructive to human life; the car-
rying of which by truculent and evil-disposed persons but too
often ended in assassination.” Haynes v. State, 24 Tenn. 120,
122 (1844). The Supreme Court of Texas weighed in too, up-
holding the application of a penalty enhancement for homi-
cides committed with a Bowie knife. Cockrum, 24 Tex. at 402.
In doing so, the court stressed the knife’s great danger, calling
it “an exceeding[ly] destructive weapon,” “the instrument of
almost certain death,” and “the most deadly of all weapons in
common use,” such that carrying the knife makes one “more
dangerous to the rights of others … than if [one] carried a less
dangerous weapon.” See id. at 402–03; see also Wolford, 2026
WL 1825723, at *18 (Barrett, J., concurring) (favorably citing
Cockrum as an example of a state court decision that “upheld
[a] gun regulation[] … because the State[] [was] pursuing spe-
cific regulatory ends, not because [it was] hostile to gun
rights”).
Contrary evidence is limited. The plaintiffs highlight the
Supreme Court of Georgia’s decision in Nunn v. State, 1 Ga.
243 (1846), which upheld an ambiguous statute to the extent
it prohibited concealed carry but struck it down to the extent
it prohibited open carry. Id. at 251. Most fundamentally, Nunn
is only one case; just as in Bruen the Court refused to “give
disproportionate weight to a single statute and a pair of state-
court decisions,” we refuse to place on Nunn more weight
Nos. 24-3060 et al. 25
than it can bear. Bruen, 597 U.S. at 65; cf. Wolford, 2026
WL 1825723, at *6. Furthermore, it is hard to discern what, if
anything, we can draw from Nunn, which conceived of the
Second Amendment as protecting the right “to keep and bear
arms of every description,” 1 Ga. at 251 (emphasis added)—a
view plainly inconsistent with the Court’s pronouncements in
Heller that machine guns and sawed-off shotguns are not pro-
tected, see 554 U.S. at 625, 627; see also Bianchi, 111 F.4th at
513 n.47 (Richardson, J., dissenting) (explaining that, for this
reason, Nunn is an “outlier of little value in discerning the na-
ture of ‘dangerous and unusual’ weapons in the Anglo-Amer-
ican legal tradition”). 11
In evaluating whether a modern law coheres with the
principles underpinning our regulatory tradition, “[a] court
must ascertain whether the new law is ‘relevantly similar’ to
laws that our tradition is understood to permit.” Rahimi, 602
U.S. at 692. Doing so requires comparing the modern and his-
torical regulations across at least two metrics: “how and why
the regulations burden a law-abiding citizen’s right to armed
self-defense.” Bruen, 597 U.S. at 29; see Hemani, 2026
WL 1751710, at *5 (“The more closely a contemporary law
11 Seeking to bolster Nunn, the dissenting opinion states that Heller
concluded Nunn “‘perfectly captured’ the meaning of the Second Amend-
ment.” More precisely, however, Heller said that Nunn “perfectly captured
the way in which the operative clause of the Second Amendment furthers the pur-
pose announced in the prefatory clause.” 554 U.S. at 612 (emphasis added).
Heller’s endorsement of Nunn’s recognition that the Second Amendment
protects more than just a militia-centered right does not imply wholesale
endorsement of Nunn’s every word—especially when, as relevant here
and as explained above, Nunn’s conception of the scope of protected arms
conflicts with Heller’s.
26 Nos. 24-3060 et al.
mirrors a well-established historical analogue in purpose and
operation, the more likely it is to be upheld.”). The “how”
component of the Bruen inquiry in turn encompasses at least
four considerations: how the Act operates mechanically, how
much it burdens the right of armed self-defense, the re-
striction’s duration, and the penalty associated with violating
it. See Rahimi, 602 U.S. at 698–99 (noting that 18 U.S.C.
§ 922(g)(8), like historical antecedents, disarms individuals
only temporarily and only after an individualized judicial de-
termination of dangerousness, and that its penalty is con-
sistent with the regulatory tradition); Bruen, 597 U.S. at 29
(asking “whether modern and historical regulations impose a
comparable burden on the right of armed self-defense”). The
“why” inquiry asks whether the burdens imposed by modern
and historical regulations are “comparably justified.” Bruen,
597 U.S. at 29.
While the Act “is by no means identical to” the historical
regimes, “it does not need to be.” Rahimi, 602 U.S. at 698. In
both how and why it burdens the right to keep and bear arms,
the Act is “‘relevantly similar’ to laws that our tradition is un-
derstood to permit.” Id. at 692 (quoting Bruen, 597 U.S. at 29).
Start with “how.” Like historical regulations, the Act op-
erates by imposing targeted restrictions on particular weap-
ons, leaving untouched a wide variety of other arms—includ-
ing many handguns, which Heller called “the quintessential
self-defense weapon.” 554 U.S. at 629. It also applies categor-
ically to most of the population while carving out exceptions
for special groups like law enforcement, a dividing line that
Bevis noted our regulatory tradition establishes. See 85 F.4th
at 1201–02. The Act also “fits within the regulatory tradition,”
Rahimi, 602 U.S. at 699, because it applies indefinitely and au-
Nos. 24-3060 et al. 27
thorizes imprisonment for violations, like the Bowie knife
statutes.
Moreover, the Act’s “burden on the right of armed self-
defense,” Bruen, 597 U.S. at 29, is minimal. To be sure, the dis-
trict court found that AR-15s offer self-defense utility and that
“every round matters in a self-defense scenario”; we accept
those findings and therefore accept that the Act imposes some
burden. We also of course agree, as the dissenting opinion em-
phasizes, that self-defense has long been a cherished and
closely guarded right. But the burden the Act imposes on that
right is mitigated by what the record indicates about how fre-
quently individuals actually use AR-15s and more than ten
rounds in self-defense—a matter on which the district court
made no factual findings and on which the only evidence
came from the defendants’ expert. 12 See Ocean State Tactical, 95
F.4th at 50 (“Depriving citizens of a device that is virtually
never used in self-defense imposes less of a burden on [the
right of armed self-defense] than does banning a weapon that
is, in fact, traditionally used in self-defense.”). According to
her report, using more than ten rounds in self-defense is “ex-
tremely rare”: She identified only two such incidents in a sam-
ple of roughly five thousand instances of armed self-defense.
Indeed, those engaged in self-defense fire only 2.2 rounds on
average. It is also quite rare for individuals to use rifles in self-
defense: The same expert’s analysis of an online database con-
taining over one thousand defensive gun uses reveals that
12 The dissenting opinion claims that “many Americans in fact use
[AR-15s] for self-defense,” but the evidence it relies on—consumer sur-
veys about why Americans buy AR-15s and expert testimony about AR-
15s’ properties—does not support that assertion.
28 Nos. 24-3060 et al.
whereas 90% of defenders employed a handgun, a mere 4%
turned to a rifle. 13 These figures remained the same after ex-
cluding states with assault weapon bans, meaning they reflect
self-defenders’ choices when unfettered by regulation. 14
While in the district court the plaintiffs cited a smattering
of online articles reporting on the self-defensive use of items
that the Act regulates, the plaintiffs submitted no evidence on
the frequency of such occurrences. The articles that the plain-
tiffs cited, moreover, are entirely consistent with the defend-
ants’ evidence; the defense expert did not conclude that the
regulated items are never used for self-defense—only that
such use is exceedingly rare. 15 The Act therefore imposes a
limited burden on the right of armed self-defense, one com-
parable to historical antecedents considering Bowie knives’
13 These are the figures for incidents in which the gun type was
known.
14 Perhaps Heller’s reasoning for why handguns are “the quintessen-
tial self-defense weapon” explains this phenomenon. See 554 U.S. at 629
(“There are many reasons that a citizen may prefer a handgun for home
defense: It is easier to store in a location that is readily accessible in an
emergency; it cannot easily be redirected or wrestled away by an attacker;
it is easier to use for those without the upper-body strength to lift and aim
a long gun; it can be pointed at a burglar with one hand while the other
hand dials the police.”).
15 To be clear, we are not, as the dissenting opinion suggests, shifting
the burden to the plaintiffs. No doubt, the defendants bear the burden at
Bruen’s second step, and we are relying on the evidence they submitted—
the expert report referenced above—to evaluate whether they carried that
burden. We reference the plaintiffs’ evidentiary showing only to make
clear that we are not improperly resolving a factual dispute about how
frequently AR-15s and more than ten rounds are actually used in self-de-
fense.
Nos. 24-3060 et al. 29
utility and use for self-defense. That is especially so given that
the Act’s grandfather clause permitted citizens who already
owned AR-15s and large-capacity magazines to continue pos-
sessing them.
The plaintiffs object that the Act’s “how” differs too
greatly from those of historical regulations. Many of the
Bowie knife regulations, the plaintiffs stress, imposed less
burdensome restrictions than does the Act—mostly because
they regulated only carry or only concealed carry, but also be-
cause some only imposed penalty enhancements for crimes
committed with the knives or severely taxed them. The Act
therefore incomparably burdens the right of armed self-de-
fense, the plaintiffs contend.
We accept the premise that the Act lacks a “historical
twin,” Rahimi, 602 U.S. at 701, but that is not the touchstone,
so we reject the conclusion that the historical antecedents on
which the defendants rely are insufficient. First, the Supreme
Court itself took just the leap that the plaintiffs argue is im-
permissible. In Heller, the Court invoked “the historical tradi-
tion of prohibiting the carrying of ‘dangerous and unusual
weapons’” in explaining that M16 rifles “may be banned.” 554
U.S. at 627 (emphasis added). And while Heller was not ap-
plying Bruen’s inquiry per se, Bruen made clear that its ap-
proach was consistent with—even more, stemmed from—Hel-
ler. See, e.g., 597 U.S. at 26 (“The test that we set forth in Heller
and apply today requires courts to assess whether modern
firearms regulations are consistent with the Second Amend-
ment's text and historical understanding.”).
Second, Rahimi, which “refine[d]” Bruen, “confirmed that
a closer match to a historical precursor is not necessary” by
rejecting distinctions no smaller than those here. Reyna, 165
30 Nos. 24-3060 et al.
F.4th at 1064. Rahimi considered the constitutionality of 18
U.S.C. § 922(g)(8), which “prohibits an individual subject to a
domestic violence restraining order from possessing a fire-
arm if that order includes a finding that he ‘represents a cred-
ible threat to the physical safety of [an] intimate partner,’ or a
child of the partner or individual.” Rahimi, 602 U.S. at 684–85
(alteration in original) (quoting 18 U.S.C. § 922(g)(8)). In re-
jecting a facial challenge to that provision, the Court relied on
two historical regimes: surety laws and going-armed laws.
Neither regime, however, imposed anything particularly
close to the restriction of § 922(g)(8).
Under the surety laws, a judicial officer could require one
proven likely to misuse firearms to post a bond that would
forfeit if one ultimately misused them. See id. at 695–96. But
that was all—it was, “in a nutshell, a fine on certain behavior.”
Id. at 753 (Thomas, J., dissenting). “After providing sureties, a
person kept possession of all his firearms; could purchase ad-
ditional firearms; and could carry firearms in public and pri-
vate. Even if he breached the peace, the only penalty was that
he and his sureties had to pay a sum of money.” Id. at 764. As
for the going-armed laws, those prohibited “carrying certain
weapons … in a particular manner … and in particular
places.” Id. at 770. In stark contrast to both these regimes,
§ 922(g)(8) completely denies an individual the ability to keep
and bear arms for the duration of his restraining order; re-
gardless of how much he is willing to put up as bond, he may
not possess any firearm in any place or in any manner. No
matter: Without denying the existence of the gaps the dissent
emphasized, the Court’s majority had “no trouble” conclud-
ing, with an emphasis on the principles underlying the histor-
ical regulations, that § 922(g)(8) was facially constitutional. Id.
at 700 (majority opinion).
Nos. 24-3060 et al. 31
The same goes here. True, the Act is “by no means identi-
cal” to its antecedents, but neither was § 922(g)(8). Id. at 698.
If anything, the distinctions between § 922(g)(8) and its ante-
cedents seem of a greater magnitude than those between the
Act and its antecedents. But most important is that the princi-
ples underlying each tradition justify the modern regulation:
For § 922(g)(8), disarming those found by a court to pose a
credible threat to the physical safety of another, and for the
Act, imposing a targeted restriction on particularly dangerous
weapons while leaving a host of others available for self-de-
fense.
Finally, and most generally, the Supreme Court recently
stressed that the primacy of history and tradition does not cre-
ate “a law trapped in amber.” Id. at 691. Just as the Second
Amendment protects more than just “muskets and sabers,” so
too it “permits more than just those regulations identical to
ones that could be found in 1791” or 1868. Id. at 692. The ques-
tion confronting courts, then, is not whether there is a “dead
ringer” or “historical twin” for the challenged regulation, but
instead “whether the challenged regulation is consistent with
the principles that underpin our regulatory tradition.” Id. (em-
phasis added); see id. at 740 (Barrett, J., concurring) (“Histori-
cal regulations reveal a principle, not a mold.”). Among other
things, an excessively exacting approach would risk mistak-
ing a historical legislature’s decision not to maximally regu-
late as marking the outer limits of the Second Amendment’s
original meaning, when legislative restraint or raw political
compromise may in fact have motivated the choice. See id. at
739–40 (Barrett, J., concurring) (explaining that “imposing a
test that demands overly specific analogues” inappropriately
“assumes that founding-era legislatures maximally exercised
32 Nos. 24-3060 et al.
their power to regulate, thereby adopting a ‘use it or lose it’
view of legislative authority”).
As for the Act’s justification—the “why”—the defendants
seem to offer both “protect[ing] Illinois communities” and
“protect[ing] the public from weapons that pose a special
danger to its safety.” We cannot accept the former. Because
every firearm regulation is at some level about public safety,
that justification operates “at such a high level of generality
that it waters down the right.” Id. at 740. But the latter is more
specific, appropriately so, and better coheres with the statu-
tory text of the challenged provisions, which focuses not on
safety generally but on especially dangerous weapons specif-
ically. And that justification is consistent with the rationales
underlying the historical regulations canvassed above, as our
sister circuits have concluded. See, e.g., Hanson, 120 F.4th at
240; Lamont, 153 F.4th at 246.
The record confirms that AR-15s equipped with thirty-
round magazines are indeed particularly dangerous—far
more like the bannable M16 than “the quintessential self-de-
fense weapon” that we know may not be banned for the entire
population, the handgun. Heller, 554 U.S. at 629. There is no
dispute that when chambered with identical bullets, the AR-
15 and the M16 fire rounds at approximately the same veloc-
ity—about 3,100 feet, or ten football fields, per second. That is
about three times the velocity of a standard handgun bullet.
There is also no dispute that AR-15s and M16s firing the same
ammunition have the same maximum effective range of about
500 yards, compared to just 50 yards for a 9mm handgun.
Likewise, the parties agree that AR-15s and M16s firing the
same ammunition will have consistent wounding capabilities,
as measured by the kinetic (or muzzle) energy of the rounds
Nos. 24-3060 et al. 33
fired. Moreover, the plaintiffs do not dispute the state’s evi-
dence that the 5.56 NATO bullets commonly fired by AR-15s
and M16s are more likely than slower, heavier handgun bul-
lets to rotate within the body and potentially fragment, which
causes additional damage to tissue and organs adjacent to the
bullet’s entry point. AR-15s, in other words, are dangerous
not only in the sense that all firearms are dangerous but also
relative to the semiautomatic handguns that Heller confirmed
are protected. And large-capacity magazines amplify each
and every one of the AR-15’s dangerous characteristics by al-
lowing a shooter to fire more of these lethal rounds without
breaking to reload.
We recite these uncontested facts not to suggest that “the
gravity of the harms the legislation was designed to avert and
the appropriateness of the mechanism they adopt” justify the
Act or to engage in interest balancing. Bevis, 85 F.4th at 1200.
Bruen clearly rejected means-end scrutiny. We recite these
facts, instead, to demonstrate that the Act and its antecedents
are “comparably justified.” Bruen, 597 U.S. at 29; see Schoen-
thal, 150 F.4th at 910 (recognizing that “the fact that similar
points can be made under different tests is a familiar aspect of
the law”). The justification for the Act is akin to that seen in
our history and tradition. As the Reconstruction-era judicial
decisions cited earlier demonstrate, our ancestors regulated
particularly dangerous weapons for sufficiently specific simi-
lar reasons as Illinois is regulating AR-15s and large-capacity
magazines. See, e.g., Haynes, 24 Tenn. at 122 (“The design of
the statute was to prohibit the wearing of bowie-knives, and
others of a similar description, which the experience of the
country had proven to be extremely dangerous and destruc-
tive to human life….”); Cockrum, 24 Tex. at 402; see also Wol-
ford, 2026 WL 1825723, at *17 (Barrett, J., concurring) (criticiz-
34 Nos. 24-3060 et al.
ing the challenged regulation because “[r]ather than identify-
ing a specific threat to public peace and safety,” the state “is
responding to the general danger associated with the pres-
ence of firearms”).
We must address one final point. The district court, plain-
tiffs, and dissenting opinion have made much of the fact that
AR-15s and large-capacity magazines are popular—i.e., there
are many in civilian hands—and therefore (they argue) “in
common use” as Heller and its progeny have used that term.
See Heller, 554 U.S. at 627. We do not deny that AR-15s are
“both widely legal and bought by many ordinary consum-
ers.” Smith & Wesson Brands, Inc. v. Estados Unidos Mexicanos,
605 U.S. 280, 297 (2025). As in Bevis, however, “we decline to
base our assessment of the constitutionality of [the Act] on
numbers alone.” 85 F.4th at 1198–99. We need not belabor this
point, which the Supreme Court’s post-Bevis precedents do
not address and which our precedents in Bevis and Friedman
v. City of Highland Park have forcefully made. See 85 F.4th at
1198–99; 784 F.3d 406, 408–09 (7th Cir. 2015). We add only that
Bruen cuts against the conclusion that a weapon’s “common
use” leaves it immune from regulation. After confirming that
“handguns are weapons in common use today for self-de-
fense,” Bruen, 597 U.S. at 32 (citation modified), the Court pro-
ceeded to undertake an “extended analysis of the Govern-
ment’s proposed historical analogues, hardly an obiter dic-
tum,” Hanson, 120 F.4th at 234.
* * *
Before moving on, we must address a methodological di-
vergence between our opinion and the dissenting opinion.
The dissenting opinion asserts that we “break fresh ground”
Nos. 24-3060 et al. 35
by not applying its understanding of “[t]he only relevant his-
torical tradition the Court has recognized”—“the regulation
of ‘dangerous and unusual’ weapons.” But the Court has not
set out a comprehensive framework through which to evalu-
ate challenges to restrictions on particular weapons, as its re-
cent grant of certiorari in cases similar to this one indicates.
See Viramontes v. Cook County, --- S. Ct. ----, 2026 WL 1871322
(mem.) (June 30, 2026); Grant v. Higgins, --- S. Ct. ----, 2026
WL 1871312 (mem.) (June 30, 2026). Accordingly, far from
breaking fresh ground, we have applied the general method-
ological principles established in the Court’s recent Second
Amendment jurisprudence by identifying relevantly similar
historical analogues.
E. A More Nuanced Approach
Bruen’s recognition that “cases implicating unprecedented
societal concerns or dramatic technological changes may re-
quire a more nuanced approach” to its historical inquiry for-
tifies, but is not essential to, our conclusion. 16 597 U.S. at 27.
This consideration—the flipside of the more “straightfor-
ward” inquiry applicable where the challenged regulation ad-
dresses an age-old problem—reflects that the farther today’s
regulatory challenges get from “those that preoccupied the
Founders in 1791 or the Reconstruction generation in 1868,”
the greater the need to “reason[] by analogy.” Id. at 28; see
Wolford, 2026 WL 1825723, at *6–7 (cases involving “distinc-
tively modern” conduct, as when “the modern law addresses
a situation that could not have arisen when the Second or
16 In other words, we do not—as the dissenting opinion claims—con-
clude that the Constitution would protect AR-15s but for dramatic techno-
logical changes and unprecedented societal concerns.
36 Nos. 24-3060 et al.
Fourteenth Amendment was adopted,” “call[] for a more dif-
ficult exercise of judgment”).
A more nuanced approach reinforces our conclusion here,
as our sister circuits have concluded. See, e.g., Lamont, 153
F.4th at 236–40; Hanson, 120 F.4th at 240–42; Bianchi, 111 F.4th
at 463–64. Indeed, this case implicates both of the grounds
Bruen identified in an interdependent way: The dramatic tech-
nological change embodied in AR-15s equipped with large-
capacity magazines has enabled the unprecedented societal
concern of mass killings speedily carried out by lone shooters.
See Bianchi, 111 F.4th at 464 (“These are not our forebears’
arms, and these are not our forebears’ calamities.”).
From 1791 through the mid-nineteenth century, the ubiq-
uitous firearm was a single-shot, muzzle-loaded firearm. In
other words, these guns could not fire consecutive rounds;
someone seeking to fire two shots would have needed to re-
load the gun in between rounds by meticulously loading the
projectile and a propellant into the muzzle—a process that
could take half a minute. As such, they “did not have the ca-
pacity to occasion a societal concern with mass shootings….”
Hanson, 120 F.4th at 240.
AR-15s equipped with large-capacity magazines are a far
cry from these antecedents. With this combination, a shooter
can discharge thirty rounds—each of which travels five foot-
ball fields in half-a-second and releases ten times the energy
of a musket ball upon impact—as quickly as he can pull the
trigger. And when that magazine is empty, he can replace it
in just a few seconds, reloading thirty rounds in a fraction of
the time it would have taken to reload one in 1791 or 1868.
Taking all of this together, a modern shooter armed with an
AR-15 and thirty-round magazines can fire almost sixty ex-
Nos. 24-3060 et al. 37
traordinarily lethal rounds—assuming about one pull of the
trigger per second, with a few seconds to change out maga-
zines—in the time it took to fire a pair of far less damaging
shots at the Founding. In other words, Illinois faces a “regu-
latory challenge[]” unlike “those that preoccupied the Found-
ers in 1791 or the Reconstruction generation in 1868.” Bruen,
597 U.S. at 27. 17
True, as the plaintiffs and dissenting opinion note, repeat-
ing arms (i.e., those capable of firing multiple rounds without
reloading) existed during this era. But through most of the
nineteenth century, these weapons remained technologically
flawed curios with limited practical utility. The 1779 Giran-
doni air rifle, for example, was fragile, complex, and imprac-
tical, while the 1821 Jennings flintlock rifle “was beset by tech-
nical challenges,” Hanson, 120 F.4th at 249 (citation modified),
such as its use of “superposed loads” (stacking multiple loads
in the barrel at once) that “were prone to explode if the se-
quencing between rounds was off,” Lamont, 153 F.4th at 237
(citation modified). These firearms, moreover, were exceed-
ingly rare and never penetrated the commercial or military
17 Relying on our identification of these technological advancements,
the dissenting opinion incorrectly imputes to us the position that the Sec-
ond Amendment protects only those arms known to the Founders. That is
certainly not our stance. Just as the First Amendment protects speech
made on the Internet and the Fourth Amendment protects cell-phone lo-
cation information, the Second Amendment “is not limited only to those
arms that were in existence at the founding.” Rahimi, 602 U.S. at 691. Still,
the Supreme Court has recognized that “cases implicating … dramatic
technological changes may require a more nuanced approach.” Bruen, 597
U.S. at 27. Or, put another way, that cases involving “distinctively mod-
ern” conduct “call[] for a more difficult exercise of judgment.” Wolford,
2026 WL 1825723, at *7. This is such a case.
38 Nos. 24-3060 et al.
markets—so it is hard to see how either could have posed a
“regulatory challenge[]” for historical legislatures, let alone
one that “preoccupied the Founders in 1791 or the Recon-
struction generation in 1868.” Bruen, 597 U.S. at 27. And the
Winchester repeating rifle, introduced in 1866 and refined in
1873, was not a true semiautomatic firearm—the shooter was
required to manipulate a lever forward and then backward
between each shot—and was reloaded one round at a time. 18
In other words, AR-15s with large-capacity magazines repre-
sent a dramatic technological change over these arms, too.
“Their advanced military-like features enable them to inflict
catastrophic injuries that bear no similarity to those injuries
caused by the comparatively primitive firearms that were
widely available in the founding and reconstruction eras.” La-
mont, 153 F.4th at 237.
The plaintiffs’ argument that the capabilities and ubiquity
of repeating rifles grew gradually as America grew—and not
as a “dramatic” technological change—overlooks the extent
to which incremental innovation, compounded over more
than a century, can yield a dramatic technological change
over the long run. There need not be a singular, transforma-
tive moment in the history of firearm innovation for AR-15s
and large-capacity magazines to be “different in form and in
kind from arms in common use during the Founding and Re-
construction eras, the relevant periods for assessing the origi-
18 The Act does not ban such firearms. See 720 ILCS 5/24-1.9(a)(2) (ex-
cluding from the definition of “assault weapon” any “firearm that is man-
ually operated by … lever … action, unless the firearm is a shotgun with
a revolving cylinder”).
Nos. 24-3060 et al. 39
nal understanding of the Second and the Fourteenth Amend-
ments, respectively.” Hanson, 120 F.4th at 242.
This dramatic technological change, moreover, has given
rise to the unprecedented societal concern of mass shootings,
especially those with high fatality counts carried out by lone
shooters. Whereas the technological limitations of our fore-
bears’ arms “prevented a single gunman from using them to
unleash a massacre in a matter of seconds,” Lamont, 153 F.4th
at 237, the same assuredly cannot be said for AR-15s and
large-capacity magazines. Indeed, the first known mass
shooting resulting in at least ten deaths did not occur until
1949. And the frequency of such events has only increased
over time. We need not belabor the all too familiar prevalence
of these episodes.
Neither of the plaintiffs’ counterarguments persuades us
otherwise. They emphasize (echoed by the dissenting opin-
ion) the district court’s conclusion that “[m]ass murder has
been a fact of life in the United States since the mid-nineteenth
century,” accusing the defendants of overlooking atrocities
committed against groups like Native Americans and slaves.
But this asserted problem of “mass murder” is framed at too
high a level of generality, obscuring distinctions with im-
portant consequences for how a legislature might respond.
When the Second and Fourteenth Amendments were ratified,
mass killings were a “group activity,” Bianchi, 111 F.4th at 463,
relying on “primitive firearms and melee weapons” because
individuals lacked “the means to inflict mass casualties on
their own,” Lamont, 153 F.4th at 239. A regulatory response to
that problem was not to impose a targeted restriction on cer-
tain arms, no one of which was principally responsible for the
lawlessness; rather, it was to target armed groups. And that is
40 Nos. 24-3060 et al.
just what Founding- and Reconstruction-era legislatures did.
See Mark Anthony Frassetto, Mass Violence and the Second
Amendment, 76 Ala. L. Rev. 43, 54–58 (2024) (collecting statutes
prohibiting armed groups). But whereas the firepower
needed to carry out a mass killing used to come from an ag-
gregation of people, now it comes from an AR-15 and large-
capacity magazines—a problem for which the historical solu-
tion is self-evidently unfit. The difference between the shame-
ful history on which the plaintiffs rely and the problem facing
modern legislatures therefore weakens the inference we can
draw from the lack of historical regulations that are extremely
similar to the Act. See Bruen, 597 U.S. at 27 (recognizing that
nuance is required where “[t]he regulatory challenges posed
by firearms today are” different from “those that preoccupied
the Founders in 1791 or the Reconstruction generation in
1868”); cf. Wolford, 2026 WL 1825723, at *17 (Barrett, J., concur-
ring) (because “the right to bear arms is misused in … ways
that were unknown to our forebears but pose an equivalent
risk to persons or property,” historical antipoaching laws can
justify more than merely modern antipoaching laws—they
“support the principle that when a State identifies specific
places that are prone to particular abuses of the right, it can
respond with focused regulations to address the threat” (cita-
tion modified)).
The dissenting opinion, moreover, relies almost exclu-
sively on twentieth-century mass shootings to undermine our
claim of an unprecedented societal concern. Putting aside that
the scale of today’s mass-shooting problem is unprecedented
even compared to a century ago, twentieth-century incidents
cannot undermine the propriety of adopting a more nuanced
approach. The conceptual underpinning of the more nuanced
approach is straightforward: The strength of the inference a
Nos. 24-3060 et al. 41
court can draw from a lack of precise historical precursors is
mitigated where “[t]he regulatory challenges posed by fire-
arms today are not … the same as those that preoccupied the
Founders in 1791 or the Reconstruction generation in 1868.”
Bruen, 597 U.S. at 27. That concept remains coherent only if
courts measure changes in technology or society against the
time period from which courts may draw historical ana-
logues. But we agree with the dissenting opinion that twenti-
eth-century analogues cannot sustain a modern firearm regu-
lation, so it must also be the case that twentieth-century regu-
latory challenges do not undermine the “distinctively mod-
ern” nature of the conduct here. Wolford, 2026 WL 1825723, at
*7. Put another way, because legislative responses to those
twentieth-century incidents could not sustain the Act, the in-
cidents themselves cannot undercut the unprecedented (in
the relevant sense) nature of the societal problem the Act ad-
dresses.
The plaintiffs also argue that, even if there is an unprece-
dented societal concern of mass shootings, the arms the Act
bans are not intrinsic to that problem. The undisputed record
evidence undercuts that claim, showing that the presence of
assault weapons and large-capacity magazines is strongly
correlated with the severity of the societal problem. In a sam-
ple of mass shootings with at least four fatalities, 24% in-
volved assault weapons and 63% involved large-capacity
magazines. Raise the fatality count to six, and assault-weapon
and large-capacity-magazine usage rates jump to 53% and
80%, respectively, between 2019 and 2022. Raise it further, to
twenty fatalities, and the respective figures climb to 75% and
88% for all incidents between 1991 and 2022. And since Sep-
tember 11, 2001, the seven deadliest individual acts of inten-
tional criminal violence in the United States have been mass
42 Nos. 24-3060 et al.
shootings, six of which used assault weapons and large-ca-
pacity magazines that the Act regulates.19 In short, whatever
else may be contributing to America’s mass-shooting epi-
demic, the record makes one thing clear: The more people
killed, the more likely it is that the killer used an assault
weapon and large-capacity magazines.
Overall, then, it is appropriate to take a more nuanced ap-
proach to the historical inquiry. Though we do not see this
conclusion as necessary to upholding the Act, it nevertheless
reinforces our holding that the Act is consistent with our reg-
ulatory tradition. The distinctions between the Act and its an-
tecedents are yet less material in light of the dramatic techno-
logical changes and unprecedented societal concerns that
have taken root since the Founding and Reconstruction eras.
* * *
In sum, the plaintiffs’ request that we find the Act facially
unconstitutional and enjoin it entirely falls short twice over.
First, and most simply, the plaintiffs did not even address
every application of the Act’s operative provisions (5/24-
1.9(b)–(c) and 5/24-1.10(b)–(c)) and thus cannot show, as Su-
preme Court precedent requires, that “no set of circumstances
19 The dissenting opinion looks past this record evidence to an online
database stating that more mass shooters use handguns than rifles. That
database, however, includes all incidents with multiple victims, whether
injured or deceased—such as an incident in which a shooter wounded two
victims and killed none. While such incidents are undoubtedly serious,
the fact that handguns are used more often than rifles for mass shootings
thus defined does not undermine our point that AR-15s and large-capacity
magazines are more likely to be used as the societal concern becomes more
grave—as, in other words, mass shootings cause more casualties.
Nos. 24-3060 et al. 43
exists under which the [challenged provisions] would be
valid.” Rahimi, 602 U.S. at 693 (quoting Salerno, 481 U.S. at
745). That is enough to deny their claim of facial invalidity
and leave the Act’s more contested applications for as-applied
challenges. Second, though, we also hold that the Act’s appli-
cation to AR-15s and thirty-round rifle magazines is constitu-
tional, supplying a second, independent basis for reversal.
F. Endorsement Affidavit Requirement
As noted above, the Federal Firearms Licensees plaintiffs
also challenge the Act’s registration requirement. Recall that
to effectuate the Act’s grandfather clause, the Act required
those wishing to continue possessing certain restricted items
to provide the Illinois State Police an “endorsement affidavit”
containing certain information. 720 ILCS 5/24-1.9(d). Registra-
tion was free, simple, and not subject to official discretion;
providing basic information about one’s firearm license and
the firearms in question automatically entitled the registrant
to keep them.
We easily (and unanimously) upheld the constitutionality
of the registration requirement in Bevis, and the record casts
no doubt on that conclusion. See 85 F.4th at 1202. Registration
is free of charge, subject to well-defined standards, and leaves
no room for government discretion. It is therefore akin to the
sort of “shall-issue” licensing regime that Bruen approved in
dicta. See 597 U.S. at 38 n.9; id. at 80 (Kavanaugh, J., concur-
ring) (“[S]hall-issue licensing regimes are constitutionally
permissible, subject of course to an as-applied challenge if a
shall-issue licensing regime does not operate in that manner
in practice.”); see also Maryland Shall Issue v. Moore, 116 F.4th
211, 222–23 (4th Cir. 2024) (holding that “shall-issue” licens-
44 Nos. 24-3060 et al.
ing regimes are presumptively constitutional under Bruen);
McRorey v. Garland, 99 F.4th 831, 837 (5th Cir. 2024) (same).
III. Conclusion
We REVERSE the judgments of the district court, and we
REMAND the cases with directions to enter judgments for the
defendants.
Nos. 24-3060 et al. 45
BRENNAN, Chief Judge, dissenting. Illinois has banned the
best-selling rifle in America and its standard magazine. That
ban—part of an extraordinarily broad and strict set of laws—
was preliminarily upheld as constitutional in Bevis v. City of
Naperville, 85 F.4th 1175 (7th Cir. 2023).
Now, with perhaps the most comprehensive trial record
in any Second Amendment case to date, this court repeats its
error. Our Nation’s enduring traditions forbid governments
from prohibiting firearms commonly owned for self-defense.
Because the people have overwhelmingly chosen the AR-15 ri-
fle and its magazine as their weapon of choice, they are pro-
tected by the Second Amendment.
Instead, the majority opinion holds that a state may pro-
hibit the possession and sale of AR-15s and 30-round maga-
zines. That conclusion cannot be reconciled with the “com-
mon use” test as applied in District of Columbia v. Heller, 554
U.S. 570 (2008), and is not supported by the historical record.
The district court’s permanent injunction should be affirmed.
I therefore respectfully dissent.
I. Background
A. The Act
At the center of these consolidated appeals is the Protect
Illinois Communities Act, Pub. Act 102-1116 (2023) (“the
Act”). The Act makes it a crime to manufacture, deliver, sell,
import, or purchase an “assault weapon, assault weapon at-
tachment, .50 caliber rifle, or .50 caliber cartridge.” 720 ILCS
5/24-1.9(b). With a few exceptions, the Act also makes it un-
lawful for any person to knowingly “possess an assault
weapon.” Id. § 1.9(c)–(d).
46 Nos. 24-3060 et al.
In all, the Act covers nearly 1,000 different firearms. This
includes “assault weapon[s],” which encompasses certain
semiautomatic rifles and pistols. Id. § 1.9(a)(1). A semiauto-
matic rifle is covered if it “has the capacity to accept a detach-
able magazine or … may be readily modified to accept a
detachable magazine” and has one or more of the following
features: a pistol grip or thumbhole stock; any feature capable
of functioning as a protruding grip that can be held by the
non-trigger hand; a folding, telescoping, thumbhole, or de-
tachable stock; a flash suppressor; a grenade launcher; or a
barrel shroud. Id. § 1.9(a)(1)(A)(i)–(vi). A semiautomatic rifle
with a fixed magazine capacity of greater than 10 rounds is
also covered. Id. § 1.9(a)(1)(B). As a result, Illinois’s ban cap-
tures more than 20% of all the types of firearms sold in the
United States in 2020.
The Act also outlaws particular models of “assault weap-
ons.” Id. § 1.9(a)(1)(J). Among those are “all AR types,” in-
cluding AR-15s. Id. § 1.9(a)(1)(J)(ii). The AR-15 is one of the
most widely owned firearms in the Nation and is the most
popular rifle in the country. Smith & Wesson Brands, Inc. v.
Estados Unidos Mexicanos, 605 U.S. 280, 297 (2025) (“AR-15
rifles, AK-47 rifles, and .50 caliber sniper rifles … are both
widely legal and bought by many ordinary consumers.”).
Millions are in circulation in the United States. Their popular-
ity stems from their self-defense properties. For example, they
are easier to aim and control than shotguns and machine
guns. And their grips make them safer and easier to aim, par-
ticularly for those who are infirm, small-statured, or have lim-
ited training. So, it is no surprise that “AR-15s are legal in 41
of the 50 States.” Snope v. Brown, 145 S. Ct. 1534, 1534 (2025)
(Kavanaugh, J., statement respecting the denial of cert.).
Nos. 24-3060 et al. 47
Certain pistols and shotguns are also banned under the
Act. That includes all semiautomatic pistols with “the capac-
ity to accept a detachable magazine” that have “a threaded
barrel,” “a second pistol grip,” a protruding grip, a barrel
shroud, or other features. 720 ILCS 5/24-1.9(a)(1)(C)(i)–(vi).
Further, any semiautomatic pistol with a “fixed magazine
with the capacity to accept more than 15 rounds” is banned.
Id. § 1.9(a)(1)(D). All semiautomatic shotguns with a fixed
magazine that holds more than five rounds, have a pistol grip,
or can accept a detachable magazine, among other features,
are also covered. Id. § 1.9(a)(1)(F)(i)–(vi).
The Act also prohibits types of magazines. Section 5/24-
1.10 covers large-capacity magazines, or magazines with a ca-
pacity of “more than 10 rounds of ammunition for long guns
and more than 15 rounds of ammunition for handguns.” 720
ILCS 5/24-1.10(a)(1). These large-capacity magazines cannot
be sold, delivered, manufactured, or purchased. Id. § 1.10(b).
And their possession is outlawed too, save certain exceptions.
Id. § 1.10(c)–(d). Illinois prohibited these magazines despite
their wide ownership—“easily more than 100 million” exist.
Duncan v. Bonta, 133 F.4th 852, 902 (9th Cir. 2025) (en banc)
(Bumatay, J., dissenting). As the D.C. Circuit recognized over
a decade ago, “[t]here may well be some capacity above which
magazines are not in common use,” but “that capacity surely
is not ten.” Heller v. District of Columbia, 670 F.3d 1244, 1261
(D.C. Cir. 2011).
B. Proceedings
In 2023 several groups of plaintiffs challenged the Act as
violating their Second Amendment rights. The appeals from
these cases were consolidated and heard together in Bevis v.
City of Naperville, 85 F.4th at 1181–84. Over my dissent, this
48 Nos. 24-3060 et al.
court concluded that the Act does not violate the Second
Amendment. Id. at 1197.
The majority opinion resolved the case on Bruen’s first
step. 1 See New York State Rifle & Pistol Ass’n v. Bruen, 597 U.S.
1, 24 (2022). “We begin by looking at the ‘plain text’ of the Sec-
ond Amendment to see whether the assault weapons and
large-capacity magazines … fall within the scope of the
‘Arms’ that individual persons are entitled to keep and bear.”
Bevis, 85 F.4th at 1192. “Arms,” the court concluded, are not
weapons “exclusively or predominantly useful in military
service.” Id. at 1194. AR-15s, and the other weapons covered
in the Act, were “much more like machineguns and military-
grade weaponry.” Id. at 1195. “Indeed, the AR-15 is almost the
same gun as the M16 machinegun,” with the only meaningful
difference being the AR-15 has semiautomatic fire only. Id.
My dissent disagreed with the Bevis majority’s arbitrary
definition of “Arms.” “My colleagues read the passages
in [District of Columbia v. Heller, 554 U.S. 570 (2008)] discussing
weapons with military capabilities too broadly, however,
placing controlling weight on supporting or explanatory lan-
guage in that decision.” Id. at 1222 (Brennan, J., dissenting).
Others agreed. “[T]he Seventh Circuit reached its conclusion
by conducting a lengthy (and questionable) exegesis of some
dicta in Heller about ‘dangerous and unusual’ weapons.” Wil-
liam Baude & Robert Leider, The General-Law Right to Bear
Arms, 99 NOTRE DAME L. REV. 1467, 1502 (2024).
1 The court moved to Bruen’s second step “for the sake of complete-
ness.” Id. at 1197. That treatment is dicta because that appeal was resolved
on step one. Id. (“[W]e are satisfied that these appeals can be resolved at
the first step of the Bruen framework.”).
Nos. 24-3060 et al. 49
The Supreme Court denied certiorari, but Justice Thomas
disagreed with the Bevis majority. Bevis, 85 F.4th at 1175, cert.
denied sub nom. Harrel v. Raoul, 144 S. Ct. 2491, 2492 (2024)
(Thomas, J., statement). “By contorting what little guidance
our precedents provide, the Seventh Circuit concluded that
the Second Amendment does not protect ‘militaristic’ weap-
ons.” Id. But, despite its “tautological[]” and “unmoored” rea-
soning, the majority opinion was preliminary. Id. at 2492–93.
If, he cautioned, “the Seventh Circuit ultimately allows Illi-
nois to ban America's most common civilian rifle, we can—
and should—review that decision once the cases reach a final
judgment.” Id.
On remand, the parties developed an expansive and com-
prehensive trial record. Plaintiffs proffered nearly 200 pieces
of evidence. This included live and written testimony from ci-
vilians, firearm design and self-defense experts, firearm in-
structors, weapon experts, and a firearm retailer. For its part,
the State submitted expert reports from a firearms expert, a
public safety expert, a doctor, and professors of history and
political science. In addition, two military experts testified on
behalf of the State: Col. Craig Tucker and Lt. Col. (ret.) Jason
Dempsey. Dempsey himself owned an AR-15 for self-defense
and disapproved of the State’s efforts to “outright” ban com-
mon firearms. In all, the parties offered initial and rebuttal re-
ports from 24 experts.
The district court heard four days of live testimony. Four
experts testified for plaintiffs; two testified for the State. The
parties then filed over 400 pages of fact-intensive post-trial
briefing. The district court wrote an insightful and compre-
hensive 160-page opinion, which detailed the history of the
50 Nos. 24-3060 et al.
relevant Second Amendment jurisprudence and evaluated a
mountain of technical evidence.
The district court concluded that the Act violates the Sec-
ond Amendment and entered a permanent injunction. First,
the court made numerous findings of fact, distinguishing AR-
15s from M16s in several ways:
• “[T]he M16 and M4 are designed to fulfill a specific
niche; their semiautomatic fire feature permits pre-
cise target shooting while their ability to fire in a
fully automatic capacity is designed to provide
suppression fire in a situation where members of a
squad are moving to or from an objective. Fully au-
tomatic fire is incredibly inaccurate and impracti-
cal, even in a military situation.”
• “A machine gun is … extremely difficult, if not im-
possible,” for the ordinary person “to control … or
to fix … on a discrete target.”
• M16 and M4 have distinct barrels to prevent over-
heating and receive quality inspections. By con-
trast, an “AR-15 is, frankly, not at all the same
weapon as the M16 rifle or M4 carbine used by the
United States military.”
• “A semi-automatic rifle does not suffer from the
lack of control as is inherent to machineguns and
sawed-off shotguns.”
• The civilian AR-15 has not been used by any mili-
tary force.
• Millions of Americans own AR-15s.
Nos. 24-3060 et al. 51
The court also found that the 30-round magazines prohib-
ited by the Act have legitimate self-defense purposes. “Every
round matters in a self-defense scenario”: the more rounds
fired, the less time spent reloading, meaning “the difference
between life and death for a person defending him or herself
in the home.” And these are widely owned because they are
commonly sold with semiautomatic rifles. 2
The district court then applied Heller’s “dangerous and
unusual” test. The “semiautomatic rifles, shotguns, and some
of the large-capacity magazines and attachments proscribed
by PICA are in common use.” In addition, “the AR-15 (and
similar copycat weapons) are ideally suited for self-defense in
the home.” Such weapons can be controlled and aimed effec-
tively, unlike machine guns and sawed-off shotguns. Thirty-
round magazines, too, could not be lawfully banned. They are
in common use and serve legitimate self-defense purposes.
After concluding that such firearms and magazines are
protected, the court moved to Bruen’s second step: whether
the government can identify a history and tradition of restrict-
ing these firearms or relevantly similar weapons. See Bruen,
597 U.S. at 24. Acts like a 1746 Massachusetts Bay Colony ban
on discharging a firearm or bans on the sale of Bowie knives
did not evince a history and tradition of banning weapons
commonly owned for self-defense. What is more, historical
“prohibitions on the carrying of certain weapons do not
amount to a categorical ban of whole classes of firearms.” For
that reason, the district court concluded that the Act violated
the Second Amendment. The State timely appealed.
2 But the district court found .50 caliber rifles and ammunition and
grenade launcher attachments served little self-defense utility.
52 Nos. 24-3060 et al.
Months later, the Supreme Court denied certiorari in a
similar case. Snope, 145 S. Ct. at 1534. Three Justices would
have granted the petition for a writ of certiorari. Justice Ka-
vanaugh wrote a statement respecting the denial of certiorari.
Id. at 1534 (Kavanaugh, J., statement respecting the denial of
cert.). In his view, Maryland’s ban on AR-15s was likely un-
constitutional because AR-15 bans are outliers among the
states. Id. In addition, AR-15s are “analytically difficult to dis-
tinguish” from the handguns held protected in Heller. Id. Alt-
hough the Court denied certiorari, Justice Kavanaugh called
for the Courts of Appeals to assist the Supreme Court’s “ulti-
mate decisionmaking on the AR-15 issue.” Id.
In its opinion permanently enjoining the Act, the district
court made findings of fact and conclusions of law. I agree
with my colleagues that the court’s factual findings here are
reviewed for clear error. See United States v. Griffin, 806 F.3d
890, 892 (7th Cir. 2015). The clear error standard is “highly
deferential” and allows this court to deviate from the district
court only when we have a “definite and firm conviction that
a mistake has been made.” Id. (citations omitted). By contrast,
all conclusions of law are reviewed de novo. 3M v. Pribyl, 259
F.3d 587, 597 (7th Cir. 2001).
II. Bruen Step One
The majority opinion correctly declines to recommit to the
analysis in Bevis on Bruen step one—whether the plain text of
the Second Amendment covers the AR-15. But for the sake of
completeness, I explain why an AR-15 qualifies as an “Arm”
covered by the Second Amendment.
Nos. 24-3060 et al. 53
A. Bevis Does Not Control
The holding in Bevis that AR-15s are not “arms” lurks in
the background of these appeals, but it does not bind us. “The
Supreme Court has held that legal and factual rulings made
as part of a preliminary-injunction analysis are not binding
upon panels when they later consider the matter on the mer-
its.” Tully v. Okeson, 78 F.4th 377, 381 (7th Cir. 2023) (citing
Univ. of Tex. v. Camenisch, 451 U.S. 390, 395 (1981)); see also
Lackey v. Stinnie, 604 U.S. 192, 201 (2025) (“Likelihood of suc-
cess” is not “success.”). The Bevis majority recognized this, re-
peatedly emphasizing the preliminary status of its holding.
Bevis, 85 F.4th at 1197. “There thus will be more to come, and
we do not rule out the possibility that the plaintiffs will find
other evidence that shows a sharper distinction between AR-
15s and M16s (and each one’s relatives) than the present rec-
ord reveals.” Id.
The conclusions offered in support of the holding in Bevis
have not aged well. For example, it stated, “[t]he M16 has an
automatic firing rate of 700 rounds per minute, while the AR-
15 has a semiautomatic rate of ‘only’ 300 rounds per minute”
without citation. Id. at 1196. But “record sources … give good
reason to doubt that figure.” Id. at 1223 (Brennan, J., dissent-
ing). According to the U.S. Army marksmanship manual, the
M16A1 can shoot 45 to 65 rounds per minute in semiauto-
matic mode and 150 to 200 rounds per minute in automatic
mode. And in Bevis, “Arm” was defined based on the Su-
preme Court’s “comments about the role of the militia.” Id. at
1193. That contradicts the holding of Heller: the right to keep
and bear arms is an individual right. McDonald v. City of Chi-
cago, 561 U.S. 742, 767 (2010). The scope of the Second Amend-
ment “is defined not by what the militia needs, but by what
54 Nos. 24-3060 et al.
private citizens commonly possess.” Friedman v. City of High-
land Park, 136 S. Ct. 447, 449 (2015) (Thomas, J., dissenting
from the denial of cert.). The Bevis majority opinion also
barely scratched the historical surface. Baude, supra, at 1501–
02 (criticizing the Bevis majority for ignoring caselaw, trea-
tises, legislative precedents and drawing a conclusion not de-
rived from history). We should move past that opinion and
follow Bruen and Heller in the following way.
B. The Text of the Second Amendment
Bruen’s first step instructs courts to decide whether the
Second Amendment’s text covers an individual’s conduct. 597
U.S. at 24. In other words, are the firearms at issue “Arms?”
The answer is yes.
The Act prohibits the possession of AR-15s sold after the
law takes effect. That limits the constitutional right to “keep
… Arms,” as Heller defined that right to mean to “have weap-
ons.” 554 U.S. at 582. Cf. Wolford v. Lopez, No. 24–1046, 2026
WL 1825723, at *9 (2026) (“The plain text of the Second
Amendment protects what petitioners want to do: carry
handguns for self-defense.” (citation modified)). The Act also
forbids buying AR-15s. Purchasing and acquiring firearms
is a prerequisite to exercising Second Amendment rights.
Ortega v. Grishman, 148 F.4th 1134, 1143 (10th Cir. 2025)
(“[T]he right to bear arms requires a right to acquire
arms.”).
The AR-15 rifle is an “Arm.” Heller, relying on historical
sources, defined “Arm” to mean “any thing that a man wears
for his defence, or takes into his hands, or useth in wrath
to cast at or strike another.” 554 U.S. at 581. A person fires
an AR-15 to protect himself and strike assailants. It is there-
Nos. 24-3060 et al. 55
fore an “Arm.” Harrel, 144 S. Ct. at 2492–93 (Thomas, J.,
statement respecting denial of cert.) (AR-15 is an “Arm”);
Snope, 145 S. Ct. at 1535 (Thomas, J., dissenting from denial
of cert.) (same); cf. United States v. Reyna, 165 F.4th 1056,
1062 (7th Cir. 2026) (deserialized firearms are “Arms”).
Following Heller, another circuit concluded that a ma-
chine gun is an “Arm.” See, e.g., United States v. Bridges, 150
F.4th 517, 524 (6th Cir. 2025). Others, however, have applied
the atextual “common use” test at step one. Bevis, 85 F.4th at
1193; Bianchi v. Brown, 111 F.4th 438, 460 (4th Cir. 2024) (en
banc).
The “common use” test does not belong in Bruen’s first
step. A firearm is still a firearm, regardless of its popularity.
Bevis, 85 F.4th at 1209 (Brennan, J., dissenting) (“The nature
of an object does not change based on its popularity.”). In
addition, the “common use” test derives from the historical
tradition of regulating “dangerous and unusual” weapons.
Id. at 1209–10; Snope, 145 S. Ct. at 1534 (Kavanaugh, J., state-
ment respecting denial of cert.) (“The Court’s later Second
Amendment decisions in Bruen and Rahimi did not disturb
the historically based ‘common use’ test.” (emphasis
added)). Accordingly, the common use test more properly
belongs in Bruen’s second step, where courts examine the
Nation’s “historical tradition of firearm regulation.” Bruen,
597 U.S. at 24.
Plaintiffs have proved that the Second Amendment’s plain
text covers their conduct, so it is presumptively constitution-
ally protected.
56 Nos. 24-3060 et al.
III. Bruen Step Two
At Bruen’s second step, Illinois has the burden to “justify
its regulation by demonstrating that it is consistent with the
Nation’s historical tradition of firearm regulation.” Id. The
contemporary restriction must be “relevantly similar” to that
historical tradition. Id. at 29. When making that determina-
tion, it is imperative for a court to analogize at the proper level
of generality. If a court takes too narrow a view, the Second
Amendment is “trapped in amber” and cannot respond to
technological developments. United States v. Rahimi, 602 U.S.
680, 691 (2024). But at too broad a generality, any historical
regulation can look like a contemporary restriction. For exam-
ple, both a colonial era ban on storing gunpowder and the
District of Columbia’s 2000s handgun ban were enacted to
keep citizens safe. But see Heller, 554 U.S. at 632 (“Nothing
about those fire-safety laws undermines our analysis; they do
not remotely burden the right of self-defense as much as an
absolute ban on handguns.”). Thus, I tread carefully not to
read historical principles “at such a high level of generality”
to “water[] down the right.” Rahimi, 602 U.S. at 740 (Barrett,
J., concurring).
To ensure fidelity to Bruen and the Second Amendment, a
court looks to the “[w]hy and how” of historical regulations.
Id. at 692. “[I]f laws at the founding regulated firearm use to
address particular problems, that will be a strong indicator
that contemporary laws imposing similar restrictions for sim-
ilar reasons fall within a permissible category of regulations.”
Id. “[T]he principles underlying the Second Amendment”
guide Bruen’s historical tradition step. Id.
Although the majority opinion here thinks it unnecessary
to review the applicable historical traditions, I do so because
Nos. 24-3060 et al. 57
they inform the “dangerous and unusual” test as understood
today. Historical analysis reveals three enduring traditions.
First, and most importantly, “dangerous and unusual” weap-
ons—those suited for or overwhelmingly used by criminals—
could be prohibited. Second, the people of England and the
United States closely guarded the right of self-defense. Third,
the weapons chosen by the people for self-defense were not
outlawed.
A. The History of the “Common Use” Test
“From Blackstone through the 19th-century cases, com-
mentators and courts routinely explained that the right was
not a right to keep and carry any weapon whatsoever in any
manner whatsoever and for whatever purpose.” Heller, 554
U.S. at 626. Today, we know this as the historical tradition of
prohibiting “dangerous and unusual weapons.” Id. at 627.
This “dangerous and unusual” test “roughly tracks” Bruen’s
“why” and “how” inquiry. Bridges, 150 F.4th at 534 (Nal-
bandian, J., concurring in part) (citation modified).
1. Pre-Founding English Regulations
The Second Amendment protects a preexisting right “in-
herited from our English ancestors.” Heller, 554 U.S. at 599
(citation omitted). Restrictions on the right to own arms that
predate the Second Amendment, then, inform how we under-
stand the original meaning of the Amendment. But those his-
torical restrictions do little to “illuminate the scope of the
right” if legal conventions changed in the following years.
Bruen, 597 U.S. at 34. Thus, courts should give weight only to
a “long, unbroken line of common-law precedent” rather than
a “short-lived, 14th-century English practice.” Id. at 35.
58 Nos. 24-3060 et al.
Until the late 1600s, England had no standing army or po-
lice force. Instead, an Englishman defended himself and his
family with his arms. Bianchi, 111 F.4th at 484 (Richardson, J.,
dissenting) (citing JOYCE LEE MALCOLM, TO KEEP AND BEAR
ARMS: THE ORIGINS OF AN ANGLO-AMERICAN RIGHT 2 (1994)
(“MALCOLM”)). Militia service was also mandatory to respond
to domestic and foreign threats. MALCOLM, at 2–3.
At the same time, sovereigns restricted the use of arms for
public safety. For example, squires, footmen, and spectators
sometimes rioted at jousting and melee tournaments. R.W.
HUNT, ET. AL., STUDIES IN MEDIEVAL HISTORY PRESENTED TO
FREDERICK MAURICE POWICKE 257, 261–62. (1948). So, in 1260,
the Statuta Armorum forbade tournament attendees from
bearing lethal weapons, such as “sword, or dagger, or staff, or
mace, or stone.” Statuta Armorum (The Statutes of Arms), c.
1260. Another problem was roving gangs attacking jurors
traveling to jury duty, preventing the King’s courts from func-
tioning. David Kopel, English Legal History and the Right to
Carry Arms, VOLOKH CONSPIRACY (Oct. 31, 2015), https://rea-
son.com/volokh/2015/10/31/english-legal-history-and-the/. In
response, a 1351 royal proclamation forbade “any one” from
going “armed with haketon, or with plate, or with habergeon
[or with sword], or with long dagger, or with any other man-
ner of arms suspected, within the City of London, or within
the suburbs.” 25 Edw. 3 (1351). The proclamation noted that
the “riots, and disputes” “impeded” the “business of our Lord
the King” and presented an “alarmed thereat [sic]” to the
“great people.” Id. Such “going armed” laws were passed to
“bar[] people from misusing weapons to harm or menace oth-
ers.” Rahimi, 602 U.S. at 693.
Nos. 24-3060 et al. 59
A similar example is the 1328 Statute of Northampton. The
law provided that Englishmen could not “come before the
King’s Justices, or other of the King’s Ministers doing their of-
fice, with force and arms, nor bring no force in affray of the
peace, nor to go nor ride armed by night nor by day, in fairs,
markets, nor in the presence of the justices or other ministers,
nor in no part elsewhere, upon pain to forfeit their armour to
the King, and their bodies to prison at the King’s pleasure.” 2
Edw. 3 c. 3 (1328); Bruen, 597 U.S. at 40. But the statute “pro-
hibited people from carrying [weapons] with such ill intent.”
Bianchi, 111 F.4th at 505 (Richardson, J., dissenting). As Black-
stone observed, “the Statute codified the preexisting, com-
mon-law crime” of “riding or going armed, with dangerous or
unusual weapons, which would terrify the good people of the
land.” Id. (citation modified).
Of these “dangerous or unusual” weapons, one example
was a launcegay, a medieval lance. The lance was “generally
worn or carried only when one intended to engage in lawful
combat or—as most early violations of the Statute show—to
breach the peace.” Bruen, 597 U.S. at 41. This is why a “going
armed” statute prohibited its public carry. Id. But at the same
time, daggers and knives, often carried for self-defense, were
not covered by the Statute. Id. at 41–42.
Up to this point, there were no blanket prohibitions on
owning classes of weapons. Rather, sovereigns outlawed the
public or concealed carry of weapons particularly useful for
crime or breaching the peace, like launcegays. Or they prohib-
ited the carrying of certain weapons in common places, like in
cities and tournaments. The law focused on who was carrying
arms, where they were carried, and how they were being
used.
60 Nos. 24-3060 et al.
Two hundred years and a Welsh rebellion later, King
Henry VIII sought to incorporate Wales into the realm of Eng-
land. Among several acts passed in 1534, any resident or per-
son living in Wales was forbidden from carrying “any bill,
long-bow, cross-bow, hand-gun, sword, staff, dagger, halbert,
morespike, spear or any other manner of weapon, privy coat
or armour defensive” in certain public areas. 26 Hen. 8, c. 6
(1534). Thus, certain groups believed to be dangerous—like
the rebellious Welsh—could have their weapons seized.
Around the same time, small handguns and crossbows
were banned. 33 Hen. 8 c. 6 (1542); Bruen, 597 U.S. at 42. Two
aspects of these laws are notable. First, as Bruen recognized,
Henry VIII’s prohibition stemmed from a concern that
“handguns threatened Englishmen’s proficiency with the
longbow—a weapon many believed was crucial to English
military victories.” 597 U.S. at 42. Second, the Act stated that
crossbows and handguns were being used “to the great [peril]
and contynuall feare and daunger of the Kinge most [loving]
subjecte.” 33 Hen. 8 c. 6 (1542). These weapons engendered
public concern, and were thus banned, because they were
“concealable” and “frequently employed in crime.”
MALCOLM, at 9. That is why the Act banned handguns with
barrels less than one yard in length. Id. at 10. Thus, like the
laws prohibiting the public carry of launcegays, Henry’s ban
proscribed arms overwhelmingly used by or designed for
criminals.
Similarly, a 1579 Proclamation outlawed the carry of
“Dagges, Handgunnes, Harquebuzes, Calliuers and Cotes of De-
fence (Harquebuzes were early rifles; Dagges were small
handguns). 21 Eliz. (1579). That Proclamation noted that such
weapons were carried by the “evil disposed, who … do so
Nos. 24-3060 et al. 61
commonly carry such offensive weapons, being in time of
peace only [meant] for thieves, robbers & murderers.” Id. (ci-
tation modified). Because these weapons, from what I can tell,
were used by thieves and robbers, the weapons, including
handguns, were banned. But by the 1600s, “the public carry
of handguns was no longer widely proscribed.” Bruen, 597
U.S. at 42. A similar 1616 proclamation banning dagges and
pistols was disregarded. Bianchi, 111 F.4th at 504 (Richardson,
J., dissenting). So, once the arm was widely adopted for law-
ful purposes, it was not banned.
This takes us to the late 1600s, during which a firearm re-
striction would have enormous consequences on our under-
standing of the right to bear arms. For our purposes, the most
revealing pre-Founding law was the Game Act of 1671,
passed by Parliament but supported by Charles II. 22 & 23
Car. 2, ch. 25 (1671). The Game Act confiscated firearms from
wide swaths of the English population. MALCOLM, at 74, 76.
“Few acts affected daily life in the countryside as profoundly.
It wounded the pride of the more than 90 percent of the pop-
ulation who were forbidden not only to kill a rabbit on their
own land but to own a gun for their personal protection.” Id.
at 76. As Heller hints, this law was likely pretextual; the Stuart
kings intended to disarm Protestants. 554 U.S. at 593. Today,
the Game Act of 1671 remains the closest historical analogy to
the Illinois Act’s blanket ban on possessing certain weapons
used for self-defense.
The Game Act was largely reviled. That Act, like the Mili-
tia Act of 1662, “caused Englishmen to be extremely wary of
concentrated military forces run by the state and to be jealous
of their arms.” Id. at 593; MALCOLM, at 76. This led to assur-
ances from William and Mary in the Declaration of Rights (the
62 Nos. 24-3060 et al.
English Bill of Rights) about bearing weapons for self-defense.
Heller, 554 U.S. at 593. “Subjects which are Protestants, may
have Arms for their Defence suitable to their Conditions, and
as allowed by Law.” Id. As Blackstone explained, this en-
shrined “the natural right of resistance and self-preservation”
and “the right of having and using arms for self-preservation
and defence.” Id. at 594 (quoting 1 William Blackstone, Com-
mentaries *144). This individual right codified in the English
Bill of Rights is “the predecessor to our Second Amendment.”
Id. at 593.
This historical record yields several conclusions. History
and tradition allow the government to prohibit the carry or
possession of dangerous and unusual weapons. But at the
same time, the right of self-defense was recognized and
guarded closely—indeed, it was necessary to maintain peace
throughout England. So, when the sovereign attempted to
outlaw weapons used for self-defense and disarm a signifi-
cant portion of the populace in the Game Act of 1671, the peo-
ple pushed back, and they were ensured the right to keep
arms for self-defense. That promise is the predecessor to our
Second Amendment.
2. The Founding
Across the Atlantic, these traditions endured. Regulations
on Arms in the colonies mirrored the purposes and limits of
their predecessors. Those weapons used overwhelmingly by
criminals—”dangerous and unusual weapons”—could be
prohibited. Yet the right of self-defense was rigorously safe-
guarded against sovereign intrusions, and the weapons cho-
sen by the people were not proscribed.
Nos. 24-3060 et al. 63
“Many colonial charters expressly guaranteed the right to
have arms.” Bianchi, 111 F.4th at 488 n.9 (Richardson, J., dis-
senting). But restrictions did exist. A 1686 East New Jersey
statute, for example, banned the private wearing of “pocket
pistol[s] … daggers or dirks.” These arms were banned be-
cause “there hath been great complaint by the inhabitants of
this Province, that several persons wearing swords, daggers,
pistols, dirks … or any other unusual or unlawful weapons”
caused “several persons in this Province, [to] receive great
abuses. An Act Against Wearing Swords, &c., ch. 9, in AARON
LEAMING AND JACOB SPICER, GRANTS, CONCESSIONS, AND
ORIGINAL CONSTITUTIONS OF THE PROVINCE OF NEW JERSEY 290
(2d ed. 1881) (GRANTS AND CONCESSIONS). This prohibition on
the concealed carry of pocket pistols “presumably did not by
its terms touch the open carry of larger, presumably more
common pistols.” Bruen, 597 U.S. at 48. Two identical statutes
from colonial Massachusetts and New Hampshire authorized
the arrest of those who “shall ride or go armed Offen-
sively … by Night or by Day, in Fear of Affray of Their Maj-
esties Liege People.” Id. at 46 (quoting statutes). Considered
together, these show “colonial legislatures sometimes prohib-
ited the carrying of ‘dangerous and unusual weapons,’” as
England had. Id. at 47.
One aspect of the historical record is especially revealing:
like others, I have not located a Founding-era statute proscrib-
ing the possession of an entire class of weapons. See, e.g., Bian-
chi, 111 F.4th at 506 (Richardson, J., dissenting) (“I am una-
ware of any laws before the American Founding that de-
prived citizens of the right to possess certain weapons.”);
Bridges, 150 F.4th at 534 (Nalbandian, J., concurring in part)
(“From what we can tell, no American jurisdiction prohibited
any kind of weapon in the eighteenth century.”); David B. Ko-
64 Nos. 24-3060 et al.
pel & Joseph G.S. Greenlee, The History of Bans on Types of
Arms Before 1900, 50 J. LEGIS. 223, 261 (2024) (“Before, during,
and after the Revolution, no state banned any type of arm,
ammunition, or accessory. Nor did the Continental Congress,
the Articles of Confederation Congress, or the federal govern-
ment created by the U.S. Constitution in 1787.”). 3
Some, including my colleagues, maintain that AR-15s
would be unrecognizable to the Founders. See Bianchi, 111
F.4th at 452 (“[AR-15s] are not the modern equivalents of
weapons that were commonly possessed and employed for
self-preservation by your shopkeeper, or your butcher, or
your blacksmith up the road in colonial America.”). That sug-
gests the Second Amendment protects only those pistols and
muskets known to our Founders. This argument “border[s]
on the frivolous.” Heller, 554 U.S. at 582. And repeating rifles
did exist during the Founding. The Belton repeating rifle,
circa 1777, fired several rounds before reloading. Or. Firearms
Fed. v. Kotek, 682 F. Supp. 3d 874, 901–02 (D. Or. 2023). The
Girandoni air rifle had a twenty-two-shot magazine capacity.
David Kopel, The History of Firearm Magazines and Magazine
Prohibitions, 78 ALB. L. REV. 849, 853 (2015). Meriwether Lewis
took one on his expedition. Bevis, 85 F.4th at 1225 (Brennan, J.,
dissenting). So, “multi-shot guns predate Colonel Colt by
over two centuries.” Kopel, supra, at 849.
On this point, the majority opinion states that at the
Founding, certain knives were more dangerous than that era’s
firearms, as the latter required users to “carefully reload after
3 At oral argument, the State pointed to a Founding-era New Jersey
law banning trap guns. Oral Argument at 22:20–28. I explain below why
this is not a categorical ban on a class of weapons.
Nos. 24-3060 et al. 65
every shot.” To my colleagues, this is evidence that modern
semiautomatic weapons present a societal problem unimagi-
nable at the Founding. Yet the majority opinion then acknowl-
edges that repeating firearms were available; for example,
that the Girandoni air rifle was “fragile, complex, and imprac-
tical.” That proto-semiautomatic weapons were available and
not banned is evidence that AR-15s may not be prohibited
consistent with our historical regulatory tradition. The wide
availability of revolvers and semiautomatic pistols is also
overlooked by the majority opinion. See Kopel, supra at 856.
And it accords no weight to the absence of categorical bans on
any classes of weapons—knives or firearms—when the Sec-
ond Amendment was enacted.
As tensions flared between the Colonists and British, the
right of self-defense became even more important to the citi-
zens of the burgeoning republic. The British would impose
stringent gun control measures. David Kopel, How the British
Gun Control Program Precipitated the American Revolution, 6
CHARLESTON L. REV. 283, 285 (2012). General Gage confiscated
gunpowder and destroyed public firearms. Id. at 313. By dis-
arming Americans, “the British were attempting to make the
practical exercise of the right of personal self-defense much
more difficult.” Id. at 301. These attempts “provoked polemi-
cal reactions by Americans invoking their rights as English-
men to keep arms.” Heller, 554 U.S. at 594. As a New York
newspaper stated, “[i]t is a natural right which the people
have reserved to themselves, confirmed by the Bill of Rights,
to keep arms for their own defence.” Id. Reverend John With-
erspoon, early President of Princeton and teacher to James
Madison, pronounced that the people of England “mean to
force us to be absolute slaves,” so the colonial legislatures
should ensure “all Americans [] provide themselves with
66 Nos. 24-3060 et al.
arms … in case they should be reduced to hard necessity of
defending themselves from murder and assassination.” John
Witherspoon, Thoughts on American Liberty, (August 1774)
(transcript available at the N.J. state library).
Other dangers at the time required firearms for self-de-
fense. “There were no police forces, and most Americans lived
in rural areas, often in homes located miles from the closest
neighbor.” Wolford, 2026 WL 1825723, at *4. Access to firearms
was thus imperative to ward off hostile Indians, criminals, or
wild beasts. Given these threats, “the right to keep and bear
arms was included among the other treasured liberties pro-
tected by the Bill of Rights.” Id.
After the Revolutionary War, “[f]our States adopted ana-
logues to the Federal Second Amendment in the period be-
tween independence and the ratification of the Bill of Rights.”
Heller, 554 U.S. at 601. “[A]ll four of these pre-Second Amend-
ment state constitutional provisions … secured an individual
right to bear arms for defensive purposes.” Id. at 602. This in-
cluded Pennsylvania and Vermont, which “clearly adopted
individual rights unconnected to militia service.” Id. at 601.
This time period—which best illuminates the meaning of
the Second Amendment 4—highlights that the colonies and
states banned the carrying of “dangerous and unusual” weap-
ons. Id. at 47–48. But no Founding-era law prohibited the pos-
session and ownership of an entire class of weapons. At the
same time, the right to own firearms used for self-defense was
respected and enshrined in the Constitution for the new Re-
public.
4 Bruen, 597 U.S. at 82–83 (Barrett, J., concurring).
Nos. 24-3060 et al. 67
3. Post-Founding/Nineteenth-Century Regulations
Courts must not “give[] postenactment history more
weight than it can rightly bear.” Bruen, 597 U.S. at 35. Still,
evidence of how the Second Amendment was interpreted
during this era can be a “critical tool,” particularly if a regular
course of practice has “liquidate[d],” or settled, the meaning
of the right. Id. (quoting Heller, 554 U.S. at 605). The three his-
torical trends described above continue in this time period: (1)
states regulated “dangerous and unusual” weapons, (2) there
were few to no bans on weapons in common use for lawful
ends, and (3) self-defense was a cherished, closely guarded
right.
States regulated the carry of “dangerous and unusual”
weapons. One example is the Bowie knife, which has a long,
fixed blade with cross guards to protect the attacker’s hand
and was often used in fights and duels. The knife was named
after Jim Bowie, who used one to kill several men during the
infamous Sandbar fight. See Bianchi, 111 F.4th at 465. Crimi-
nals using Bowie knives led many states to regulate them in
some way. From what I could find, Georgia was the only pre-
Civil War state to ban their possession. 5 But the ban was
promptly held unconstitutional. Nunn v. State, 1 Ga. 243 (Ga.
1846). Instead, most states banned the concealed carry of
Bowie knives. Bevis, 85 F.4th at 1216 (Brennan, J., dissenting).
The majority opinion cites three statutes—two from states
and one from a territory—of broad Bowie knife bans. These
statutes cannot bear the weight placed on them. They were
enacted after 1871, much later than the Second Amendment
5 Act of Dec. 25, 1837, § 1, 1837 Ga. Laws 90, 90 (possession or sale of
Bowie knives, pistols, dirks, sword canes, or spears).
68 Nos. 24-3060 et al.
and after the ratification of the Fourteenth Amendment. So,
their import for constitutional meaning is limited. Even more,
it is doubtful that three statutes establish a tradition. Cf. Bruen,
597 U.S. at 46 (“[W]e doubt that three colonial regulations
could suffice to show a tradition.”).
Most importantly, these are not the “broad proscrip-
tion[s]” my colleagues believe them to be. Both the 1871 Texas
and the 1889 Arizona (territory) bans allowed the carrying of
pistols and knives for self-defense. See 1871 Tex. Laws 1st Sess.
25; 1889 Ariz. Sess. Laws 30. That leaves Arkansas’s 1881 ban.
But the majority opinion cites only one section of that Act. See
1881 Ark. Acts 191, § 1. (“That any person who shall wear or
carry, in any manner whatever … any pistol of any kind …
shall be guilty of a misdemeanor.”). As far as I can tell, a dif-
ferent part of the Arkansas Act states it does not apply to pos-
session or public carry: “Provided, further, That nothing in this
act be so construed as to prohibit any person from carrying
any weapon when upon a journey, or upon his own prem-
ises.” Id. An Act to Preserve the Public Peace and Prevent
Crime, no. 96, §§ 1–8, 1881 Ark. Acts 191, 191–92 (Mitchell &
Bettis). Accordingly, these Bowie knife “bans”—rare and nar-
row—are far from analogous to Illinois’s extraordinarily
broad prohibitions. Cf. Wolford, 2026 WL 1825723, at *17 (Bar-
rett, J., concurring) (Hawaii cannot broadly restrict the carry
of firearms in public when no historical laws did so).
States did not enact laws to limit the possession or carry of
Bowie knives because they were extremely dangerous. Ra-
ther, they did so because they were commonly concealed and
Nos. 24-3060 et al. 69
used by criminals. 6 This is evident from several cases discuss-
ing these bans under the right to bear arms. For example, in
1840, William Aymette was convicted for wearing a concealed
Bowie knife, violating a Tennessee statute. Aymette v. State, 21
Tenn. 154, 155 (Tenn. 1840). The Tennessee Supreme Court
drew the boundaries of the right to self-defense. “The Legis-
lature, therefore, have a right to prohibit the wearing or keep-
ing weapons dangerous to the peace and safety of the citizens,
and which … would not contribute to the common defence.”
Id. at 159. But those arms “usually employed in private broils,
and which are efficient only in the hands of the robber and the
assassin,” may be banned. Id. at 158. Because Bowie knives
were not used for lawful but criminal purposes, the Supreme
Court upheld the conviction. The Louisiana Supreme Court
reached the same conclusions when analyzing that state’s ban
on the concealed carry of Bowie knives and like weapons.
Such weapons are “not even designed for any purpose of pub-
lic defence, and used most frequently by evil-disposed men
who seek an advantage over their antagonists, in the disturb-
ances and breaches of the peace which they are prone to pro-
voke.” State v. Smith, 11 La. Ann. 633, 633 (1856).
For support, the majority opinion cites Cockrum v. State, 24
Tex. 394 (1859). There, as my colleagues point out, the Su-
preme Court of Texas upheld a penalty enhancement for a
homicide committed with a Bowie knife. Id. at 402. But the
court emphasized that the legislature could not prohibit the
carry of Bowie knives wholesale, as it would “deter the citizen
from its lawful exercise.” Id. at 402–03. Instead, as the court
6 “The antebellum right to bear arms was fought primarily over the
early concealed weapons statutes.” Robert Leider, Our Non-Originalist
Right to Bear Arms, 89 IND. L. J. 1857, 1601 (2014).
70 Nos. 24-3060 et al.
stated, its holding is confined to the context of manslaughter
and murder: “The legislature has the power to put all cases of
manslaughter, committed with deadly weapons, on the same
footing with murder.” Id. at 403.
Next, consider bans on using trap guns. These are loaded
guns “intended to go off or discharge itself, or be discharged
by any String, Rope or other Contrivance” to protect property.
1771 N.J. Laws 346, An Act for the Preservation of Deer and
Other Game, and to Prevent Trespassing with Guns, ch. 539,
§ 10. New Jersey prohibited the “set[ting]” of a trap gun. Id.
At the outset, it is questionable whether trap guns are even
protectable “Arms.” They are not “bearable.” After all, “the
whole design of the trap gun was to allow a firearm to be dis-
charged without a person needing to ‘keep’ or ‘bear’ it.” Dun-
can, 133 F.4th at 908 (Bumatay, J., dissenting). And even if they
are “Arms,” they are unhelpful for personal self-defense. In-
deed, they fire indiscriminately, possibly even injuring the
owner. Bevis, 85 F.4th at 1218 (Brennan, J., dissenting).
In sum, prohibitions on Bowie knives and trap guns illus-
trate how states understood the “dangerous and unusual”
test. A weapon could be banned if it was particularly suited
for criminality and it was not used for self-defense.
Caselaw from the high courts of Texas, Arkansas, Louisi-
ana, and Georgia also informs our understanding of the “dan-
gerous and unusual” test.
In State v. Duke, 42 Tex. 455 (1875), the Supreme Court of
Texas evaluated the State’s ban on the “keeping and bearing
of deadly weapons.” Id. at 456. That court spoke to which
arms received constitutional protection: “The arms which
every person is secured the right to keep and bear … must be
Nos. 24-3060 et al. 71
such arms as are commonly kept, according to the customs of
the people, and are appropriate for open and manly use in
self-defense.” Id. at 458. Because the law excepted carry for
self-defense and in the home, the court concluded the statute
did not infringe the right to bear arms. Id. at 459.
The Supreme Court of Arkansas, when faced with a de-
fendant’s criminal indictment for carrying a pistol, empha-
sized that sovereigns should punish wrongdoers, not strip the
populace of the right to self-defense. Wilson v. State, 33 Ark.
557, 559 (1878). “If cowardly and dishonorable men some-
times shoot unarmed men with army pistols or guns, the evil
must be prevented by the penitentiary and gallows, and not
by a general deprivation of a constitutional privilege.” Id. at
560. The judgment therefore was reversed and the case re-
manded for a new trial on self-defense. So too, the Louisiana
Supreme Court concluded that citizens had a right to carry
arms because “[t]his is the right guaranteed by the Constitu-
tion of the United States, and which is calculated to incite men
to a manly and noble defence of themselves,” State v. Chandler,
5 La. Ann. 489, 490 (La. 1850).
Important for this evaluation is Nunn v. State, 1 Ga. 243
(1846). Georgia made it a crime “to sell, or offer to sell, or to
keep[,] or to have about their persons” Bowie knives, pistols,
dirks, sword-canes, spears—not too different from Illinois’s
Act here. Id. at 246. The Supreme Court of Georgia enjoined
the ban on carrying pistols openly. Id. at 251. That court
stated, “[t]he right of the whole people … to keep and
bear arms of every description, and not such merely as are
used by the militia, shall not be infringed.” Id. The majority
opinion disclaims reliance on Nunn, despite the Supreme
Court in Heller concluding that Nunn “perfectly captured” the
72 Nos. 24-3060 et al.
meaning of the Second Amendment. 554 U.S. at 612. During
this period, state courts continued to recognize the im-
portance of self-defense. “In Nunn … the Georgia Supreme
Court construed the Second Amendment as protecting the
‘natural right of self-defence.’” Id.
The most vulnerable members of the public understood
better than anyone the importance of self-defense. “Blacks
were routinely disarmed by Southern States after the Civil
War. Those who opposed these injustices frequently stated
that they infringed blacks’ constitutional right to keep and
bear arms.” Heller, 554 U.S. at 614. Horrific injustices were
committed against black Americans at the time. For example,
“members of a white militia … brutally murdered as many as
165 black Louisianians congregating outside a courthouse.”
McDonald, 561 U.S. at 808–09 (Thomas, J., concurring in part)
(discussing United States v. Cruikshank, 92 U.S. 542 (1876)).
Republicans in Congress sought to suppress armed gangs of
ex-Confederates from terrorizing the newly freed blacks by
passing the Freedmen’s Bureau Act of 1866 and the Civil
Rights Act of 1866. Both guaranteed blacks the right to self-
defense. Id. at 771–74. And the co-founder of the NAACP, Ida
Wells, wrote, “a Winchester rifle should have a place
of honor in every black home, and it should be used for that
protection which the law refuses to give.” Ida B. Wells-Bar-
nett, Southern Horrors and Other Writings: The Anti-Lynching
Campaign of Ida B. Wells, 1892–1900 70 (Jacqueline Jones
Royster ed., 1997).
Twentieth-century bans do not help inform the “danger-
ous and unusual” test. In 1934 and 1968, Congress severely
restricted the possession of machine guns and similar fire-
arms. Bevis, 85 F.4th at 1202. Then, in 1986, machine guns
Nos. 24-3060 et al. 73
manufactured after that year were banned. Id. But those en-
actments happened too recently to be considered an enduring
national tradition. See Bruen, 597 U.S. at 83 (Barrett, J., concur-
ring) (cautioning against a “freewheeling reliance on histori-
cal practice from the mid-to-late 19th century to establish the
original meaning of the Bill of Rights”).
To conclude, this history yields three traditions. First,
from England until the late 19th century, self-defense was a
natural right, imperative to the functioning of democracy and
a safe society. Second, the weapons that could be banned were
overwhelmingly used to commit crimes. In particular, Eng-
land and the early United States focused on easily concealable
arms. Such arms are “dangerous and unusual.” Third, arms
commonly used by the people for self-defense were not
banned. And the few times sovereigns attempted to do so, in
the 1671 Game Act or by the State of Georgia in 1837, the peo-
ple and courts pushed back, reaffirming the right to own
weapons for self-defense.
B. The “Dangerous and Unusual” Test Today
As discussed, a “dangerous and unusual” weapon is one
particularly suited for and used by criminals and not widely
owned for lawful self-defense. 7 This follows Heller’s instruc-
tion that “the Second Amendment does not protect those
weapons not typically possessed by law-abiding citizens for
lawful purposes … . That accords with the historical under-
standing of the scope of the right.” 554 U.S. at 625. Thus, the
“dangerous and unusual” test and the “common use” test are
7 I do not discuss whether other lawful purposes could protect a fire-
arm, such as hunting or community self-defense, but I do not doubt that
is true.
74 Nos. 24-3060 et al.
two sides of the same coin: a “dangerous” gun, particularly
suited for and used by criminals, is not used “for lawful pur-
poses.” And an “unusual” arm is not “in common use.” So,
the critical inquiry becomes (1) how many of the firearms are
in private hands and (2) whether they are overwhelmingly
used for self-defense.
That test properly focuses on the people. The Second
Amendment right is an individual right, as Heller held. The
Court rejected a test in which judges decide what weapons are
necessary for self-defense. “There are many reasons that a cit-
izen may prefer a handgun for home defense: It is easier to
store in a location that is readily accessible in an emergency;
it cannot easily be redirected or wrestled away by an attacker;
it is easier to use for those without the upper-body strength to
lift and aim a long gun; it can be pointed at a burglar with one
hand while the other hand dials the police.” Id. at 629.
All that mattered was that handguns were in common use
for a lawful purpose: they “are the most popular weapon cho-
sen by Americans for self-defense in the home.” Id. So the ban
was unconstitutional. See also Joel Alicea, Bruen Was Right, 174
U. PA. L. REV. 13, 39 (2025) (“Bruen describes the rejected form
of means-end scrutiny … that involves ‘the evolving product
of federal judges,’ and it contrasts this form of scrutiny with
its own approach, in which the balance of means and ends is
the ‘product of an interest balancing by the people.’”).
Bowie knife bans illustrate the application of the “danger-
ous and unusual” test. Suppose a court is deciding for the first
time whether such knives can be prohibited. First, Bowie
knives were overwhelmingly used by thieves and murderers
to terrorize the public. See Aymette, 21 Tenn. at 158. People did
not choose them “for [the] open and manly use in self-de-
Nos. 24-3060 et al. 75
fense.” Duke, 42 Tex. at 458. The knives are therefore “danger-
ous” (or “not used for lawful purposes”). Second, Bowie
knives are not “commonly kept, according to the customs of
the people.” Id. They are therefore “unusual” (or not in “com-
mon use”). As a result, they can be prohibited.
Or count the number of states that prohibit the firearm at
issue. In Bruen, two justices concurred, emphasizing that New
York’s licensing regime was an “outlier” among the states. 597
U.S. at 79 (Kavanaugh, J., joined by Roberts, C.J., concurring)
(noting that only six states had such a regime). The same is
true with AR-15 bans. “AR-15s are legal in 41 of the 50 States,
meaning that the States such as Maryland that prohibit AR-
15s are something of an outlier.” Snope, 145 S. Ct. at 1534 (Ka-
vanaugh, J., statement respecting denial of cert.). So, a court
can account for how many of the firearms are owned and how
many states ban their possession to determine whether a fire-
arm is in common use.
The State contends the common-use test is circular. See also
Friedman v. City of Highland Park, 784 F.3d 406, 409 (7th Cir.
2015). Take the AR-15 as an example. The Federal Assault
Weapons Ban of 1994 prohibited the sale and manufacture of
AR-15s. That ban expired in 2004. Then sales of AR-15s sky-
rocketed. A test that considers only how many AR-15s are in
existence would conclude they are in common use. But had
the federal ban not lapsed, AR-15s may not be widely owned,
meaning they would not be in “common use,” and Illinois
could lawfully ban them. In other words, what can be banned
depends on what has been banned.
Counting the number of state bans has the same flaw. If
on Monday two states banned AR-15s, then on Tuesday 30
states banned them, then a state ban passed on Wednesday
76 Nos. 24-3060 et al.
would be constitutional—an AR-15 ban would no longer be
an outlier among the states. Yet that same ban passed on Mon-
day would be unconstitutional.
Two counterarguments spring to mind. First, the Heller
majority did not find this circularity argument persuasive. In
dissent, Justice Breyer observed that if the ban on machine
guns were lifted, and they became ubiquitous, “the Court will
have to reverse course and find that the Second Amend-
ment does, in fact, protect the individual self-defense-related
right to possess a machinegun” despite the majority holding
that the Second Amendment does not protect machine guns.
Heller, 554 U.S. at 721 (Breyer, J., dissenting). “There is no
basis for believing that the Framers intended such circular
reasoning.” Id. The majority in Friedman embraced this argu-
ment, 784 F.3d at 409, as did the majority in Bevis, 85 F.4th at
1190. But it is not for us to inject the circularity argument back
into Second Amendment law. If a majority of the Supreme
Court did not adopt that argument in Heller, neither should
we.
Second, the circularity concern is not as problematic as it
appears. Its proponents cite the machine gun as paradigmatic
of the problems with a weapons count test. Because machine
guns were banned, they have never been in common use and
thus can now be banned. Friedman, 784 F.3d at 409. But ma-
chine guns were available for sale to the public. United States
v. Alsenat, 174 F.4th 45, 47 (11th Cir. 2026).; David B. Kopel,
Machine Gun History and Bibliography, 25 WYO. L. REV. 45, 89
(2025) (describing how machine guns were advertised as “a
machine gun for the home” and to ranchers to fend off “des-
peradoes”). “For commercial sales to law-abiding citizens, the
Thompson submachine gun was a flop, but gangsters loved
Nos. 24-3060 et al. 77
it.” Kopel, supra at 89 (citation modified). And after some
time, it became obvious that the American people were not
choosing machine guns for self-defense; rather, they were
commonly used by gangsters in shootings like the Saint Val-
entine’s Day Massacre. 8 The same happened with Bowie
knives. After their invention, their use became widespread in
unlawful duels and knife fights. In short, the Constitution
looks to the people to decide which weapons are used for self-
defense. Those weapons overwhelmingly chosen or particu-
larly suited for criminality are considered “dangerous and
unusual.”
Another criticism of the “dangerous and unusual” test is
that it creates an under-protection problem. If a new firearm
is released, a government can act quickly to ban it, and that
firearm will never be in “common use” and thus protected.
But the “dangerous and unusual” test would not allow for
that. When a new firearm is introduced to the market, it will
almost always be too early to tell whether it is “dangerous and
unusual.” To immediately regulate this weapon, the govern-
ment must proffer evidence that it is being used by criminals,
not law-abiding citizens for lawful self-defense. At a mini-
mum, the government must “wait and see” whether and how
the public adopts and uses the firearm before it can be
banned. That accords with the history of weapons bans, such
8 See Staples v. United States, 511 U.S. 600, 626 n.4 (1994) (Stevens, J.,
dissenting) (“The late 1920s and early 1930s brought … a growing percep-
tion of crime both as a major problem and as a national one … Criminal
gangs found the submachinegun (a fully automatic, shoulder-fired
weapon utilizing automatic pistol cartridges) and sawed-off shotgun
deadly for close-range fighting.” (citation modified)).
78 Nos. 24-3060 et al.
as Bowie knives and machine guns, and respects the right of
the people to choose which firearms to use for self-defense. 9
It may well be that the “common use” test has an aspect of
circularity, but that is far from an anomaly in constitutional
law. For example, the Fourth Amendment’s reasonable expec-
tation of privacy test is circular. See Richard A. Posner, The
Uncertain Protection of Privacy by the Supreme Court, 1979 S. CT.
REV. 173, 188. Circularity, then, is an insufficient reason to jet-
tison a constitutional test, particularly one backed by im-
mense historical support.
The majority opinion states: “It is also quite rare for indi-
viduals to use [AR-15] rifles in self-defense,” and the plain-
tiffs’ “smattering of online articles” are insufficient to show
any real burden on the right to self-defense. But this lacks sup-
port for many reasons.
Foremost, it incorrectly shifts the burden of proof that the
firearms are used for self-defense onto the plaintiffs. After
they satisfy their burden on Bruen step one, a firearm is “pre-
sumptively” protected. Bruen, 597 U.S. at 24. So, Illinois must
justify its ban on the most popular rifle in America and its
magazine. To do so, the State must proffer evidence that AR-
15s are “dangerous and unusual.” That means Illinois must
do more than find historical statutes showing that legislatures
had the power to ban such weapons. The State must also show
that AR-15s are not in common use for lawful purposes. Illi-
nois must supply the evidence showing that AR-15s are pre-
9 Circularity and new firearms may also be a theoretical problem, not
so much a practical problem. Rarely will a weapon enter the market that
is entirely new; most are variants of an already legal firearm. Friedman, 784
F.3d at 416 n.5 (Manion, J., dissenting).
Nos. 24-3060 et al. 79
dominantly used by criminals for illegal purposes. But no
such evidence exists in the record.
Even if the burden rested on the plaintiffs—which it does
not—at trial they offered extensive evidence that AR-15s are
chosen for self-defense. They cited four consumer studies
showing that a principal reason consumers purchase AR-15s
is for self-defense. 10 And experts explained to the district
court that the AR-15 is chosen for its self-defense properties.11
So, the district court’s conclusions that AR-15s are “ideally
suited” for self-defense, and that many Americans in fact use
them for self-defense, are not clearly erroneous.
This point also reveals a broader concern that the majority
opinion cannot be squared with Heller. There, the Supreme
Court held D.C.’s handgun ban unconstitutional, but it did
not require “evidence on the frequency” of handguns actually
being used for self-defense, as under the reasoning of the ma-
jority opinion. Instead, Heller concluded that the Second
Amendment protects handguns because they “are the most
popular weapon chosen by Americans for self-defense in the
home, and a complete prohibition of their use is invalid.” 554
U.S. at 629. That is, because handguns are owned by millions
of people, under Heller they are in common use for self-de-
fense. The same is true here. That millions of AR-15s are
owned and have obvious self-defense properties means the
firearm is in common use for self-defense.
My colleagues also reason that Illinois’s Act leaves other
weapons available for self-defense. Yet Heller rejected this ar-
10 ECF No. 253 at 22, ¶ 87.
11 Id. at 22, ¶¶ 86–87.
80 Nos. 24-3060 et al.
gument. “It is no answer to say, as petitioners do, that it is
permissible to ban the possession of handguns so long as the
possession of other firearms (i.e., long guns) is allowed.” Hel-
ler, 554 U.S. at 629. Likewise, AR-15s are owned in the mil-
lions. So it is “no answer” to say they can be banned because
handguns are available.
C. AR-15s Pass the “Common Use” Test
In applying the “dangerous and unusual” test, courts may
start with the “unusual” prong (which, as shown, is the oppo-
site of “common use”). For a firearm to be “unusual,” it must
be highly unusual in society at large. Bruen, 597 U.S. at 47.
“Society” refers to today, not at the adoption of the Second
Amendment. Id. at 48.
To say AR-15s are not in “common use” does not pass the
“red face” test. The district court correctly found that “mil-
lions of Americans own AR-15s at this very moment.” 12 Per
the State’s political scientist Professor Louis Klarevas, that
number is conservative. 13 Even accounting for the AR-15s
owned by firearms dealers (which the government argued
was around 37.1% of all AR-15s), “there are still millions of
weapons in circulation in the United States (and, it follows, in
Illinois) that the Illinois Government has rendered illegal.”14
The record evidence here well supports these conclusions.
Plaintiffs collected several articles discussing how AR-15s are
exceedingly popular. 15 Professor Klarevas estimated the
12 ECF No. 253 at 24, ¶ 98; ECF No. 258 at 101.
13 Id.
14 Id.
15 ECF No. 253 at 21–22, ¶ 85.
Nos. 24-3060 et al. 81
number of Americans who own an AR-15 to be around 14.1 to
18.2 million. 16 Indeed, one of the State’s witnesses, retired
Lieutenant Colonel Dempsey, keeps an AR-15 for self-de-
fense. 17 The district court’s finding is well supported in the
record and certainly not clearly erroneous.
The Supreme Court agrees that AR-15s are widely owned.
“AR-15 rifles, AK-47 rifles, and .50 caliber sniper rifles … are
both widely legal and bought by many ordinary consumers.”
Smith & Wesson Brands, Inc. v. Estados Unidos Mexicanos, 605
U.S. 280, 297 (2025); Garland v. Cargill, 602 U.S. 406, 430
(2024) (Sotomayor, J., dissenting) (the shooter in the 2017
Las Vegas mass shooting used a “commonly available,
semiautomatic rifle[]” with a bump stock attached). Justice
Thomas and Justice Kavanaugh recently agreed AR-15s are
widely owned and thus in common use. Snope, 145 S. Ct. at
1534 (Kavanaugh, J., statement respecting the denial of cert.)
(“Given that millions of Americans own AR-15s and that a sig-
nificant majority of the States allow possession of those rifles,
petitioners have a strong argument that AR-15s are in ‘com-
mon use.’”); id. at 1538 (Thomas, J., dissenting from denial of
cert.). And if tasers are protected because “[h]undreds of
thousands of Tasers and stun guns have been sold to private
citizens,” AR-15s are certainly protected because millions
have been sold to consumers. Caetano v. Massachusetts, 577
U.S. 411, 420 (2016) (Alito, J., concurring).
Another measure of “common use” is to examine the
number of state prohibitions on an arm. Bruen, 597 U.S. at 79
(Kavanaugh, J., concurring) (observing that New York’s
16 Id. at 24 ¶ 98.
17 Id. at 25 ¶ 100.
82 Nos. 24-3060 et al.
“may-issue” licensing regime was an “outlier” because only
six states used such a regime). By that metric, AR-15s are also
in “common use” because only nine or ten states prohibit
them. 18 “AR-15s are legal in 41 of the 50 States, meaning that
the States such as Maryland that prohibit AR-15s are some-
thing of an outlier.” Snope, 145 S. Ct. at 1534 (Kavanaugh, J.,
statement respecting denial of cert.).
The majority opinion, however, does not apply the “com-
mon use” test at all. It cites a purported historical tradition of
regulating “particularly dangerous weapons” or “weapons
whose danger and lethality stand out.” To my colleagues,
such weapons can be regulated regardless of whether they are
in common use for lawful purposes or not.
But the Supreme Court has not adopted or endorsed such
a test. The only relevant historical tradition the Court has rec-
ognized is the regulation of “dangerous and unusual” weap-
ons—which requires proof that the weapons are both “dan-
gerous” and “unusual.” See Caetano, 577 U.S. at 411–12 (per
curiam); id. at 417 (Alito, J., concurring in the judgment). So,
my colleagues break fresh ground by defining a new “tradi-
tion” that has not been recognized by the Supreme Court.
The next inquiry is whether AR-15s are “dangerous”—the
focal point of the majority’s analysis. Though all guns are in
some sense “dangerous,” recall that this is a term of art. His-
torically, “dangerous” weapons were those particularly
suited for criminality or overwhelmingly used by criminals.
18 Colorado enacted S.B. 25-03, which requires that state’s residents to
obtain a license and receive firearms training before purchasing certain
semiautomatic firearms. See COLO. REV. STAT. § 18-12-116 (2026).
Nos. 24-3060 et al. 83
Sovereigns were particularly concerned with weapons that
could be concealed.
AR-15s are widely owned for self-defense. After hearing
from firearm instructors as well as self-defense and military
experts, the district court concluded that AR-15s are “ideally
suited for self-defense in the home.” Both “experts … and fact
witnesses attest to the fact that law-abiding citizens choose
[AR-15s] for self-defense.” The court found that AR-15s are
light, short, and have less recoil, making them easier to handle
and fire. The district court also heard from self-defense ex-
perts who all reported “recoil[,] … lighter weight, shorter bar-
rel, and ergonomic stock and grip” make “AR platform rifles”
well suited for self-defense, which is why they are commonly
used in popular “defensive carbine course[s].” 19
Evidence offered by plaintiffs supports the district court’s
findings. One reason AR-15s are easy to shoot is the manner
in which the firearm expels gas, which reduces felt recoil. 20
Reduced recoil also helps keep the muzzle on target, increas-
ing accuracy. 21 And the AR-15’s expulsion design removes the
need for the shooter to manually reload, meaning the firearm
“runs itself.” 22 These features are all useful for self-defense.
To my colleagues, AR-15s are more dangerous than hand-
guns. The rifle’s ammunition has a higher velocity, and AR-
15s are more accurate at longer ranges. The State, for its part,
19 ECF No. 253 at 26–27, ¶¶ 107–09. The FBI Ballistic Research Facility
confirmed the same. See id. at 26 ¶ 107.
20 Id. at 14, ¶¶ 38–39.
21 Id. at 15, ¶ 41.
22 Id. at 14, ¶ 39.
84 Nos. 24-3060 et al.
focuses on range, penetration, and rate of fire to argue that
AR-15s are poor self-defense weapons.
But a citizen may prefer an AR-15 over a handgun for self-
defense for many reasons. An AR-15 is more accurate than a
pistol and easier to aim. David B. Kopel, Rational Basis Analy-
sis of “Assault Weapon” Prohibition, 20 J. CONTEMP. L. 381, 403
(1994). And “[i]t would be rather irrational to ban a firearm
because it was particularly accurate and, hence, posed a
smaller danger of stray shots.” Id. Those who chose a firearm
that does not fire wildly, possibly endangering bystanders,
should not be punished. These self-defense benefits rebut the
assertion that an AR-15 is unusually dangerous and might
suggest the State simply dislikes the weapon. But distaste for
a particular firearm cannot justify proscribing it. Cf. Wolford,
2026 WL 1825723, at *11 (The Second Amendment does not
yield to Hawaii’s “spirit of Aloha … any more than it can
yield to the spirit of … the Windy City.”).
Whether a firearm is useful for self-defense is not a deci-
sion for judges. The people choose which weapons to own for
self-defense. As Heller tells us, “There are many reasons that
a citizen may prefer a handgun for home defense,” but
“[w]hatever the reason, handguns are the most popular
weapon chosen by Americans for self-defense.” Heller, 554
U.S. at 629. “Our Constitution allows the American peo-
ple—not the government—to decide which weapons are
useful for self-defense.” Snope, 145 S. Ct. at 1537 (Thomas,
J., dissenting from denial of cert.). Judges are to examine
what weapons Americans do own for self-defense. And the
people have spoken: AR-15s are the most popular rifle in
America, so a prohibition on the possession of that firearm
and its magazine is unconstitutional.
Nos. 24-3060 et al. 85
Judges do not define the scope of other constitutional
amendments, as would follow from my colleagues’ posi-
tion. A judge could not label a particular type of speech as
“dangerous” or “lacking value” and thus conclude it is not
protected by the First Amendment. Quite the opposite.
“The First Amendment protects an individual’s right to
speak his mind regardless of whether the government con-
siders his speech sensible and well intentioned or deeply
misguided.” 303 Creative LLC v. Elenis, 600 U.S. 570, 586
(2023) (citation modified). The same applies for the Second
Amendment. AR-15s are overwhelmingly purchased and
possessed for self-defense. See Heller, 554 U.S. at 629. Courts
do not evaluate the prudence of that choice, just as they do
not assess the value of particular speech.
Finally, the focus of the majority opinion on “particularly
dangerous weapons” invites difficult line-drawing problems.
AR-15s are very similar to the semiautomatic pistols Heller
held were protected. “The semiautomatic mechanism in an
AR-15 rifle is, in all material respects, the same as in a semi-
automatic handgun.” Bevis, 85 F.4th at 1215 (Brennan, J., dis-
senting). “That mechanism is gas powered, and the impact of
the pin firing the bullet pushes back the lock mechanism,
ejects the old shell, and loads the new round from the maga-
zine.” Id. As a result, “it can be analytically difficult to distin-
guish the AR-15s at issue here from the handguns at issue in
Heller.” Snope, 145 S. Ct. at 1534 (Kavanaugh, J., statement re-
specting denial of cert.). Variations between AR-15s and
handguns in the rate of fire and penetration matter little in a
gunfight.
AR-15s are no more suited for criminal activities than a
handgun. If anything, “a handgun could be viewed as more
86 Nos. 24-3060 et al.
dangerous than an AR-15 rifle because the handgun is less ac-
curate and more concealable.” Bevis, 85 F.4th at 1215 (Bren-
nan, J., dissenting). By contrast, short-barreled rifles and
shotguns are easy to conceal and add “little—if any—func-
tionality to the firearm for lawful use.” United States v.
Rush, 130 F.4th 633, 637 (7th Cir. 2025); see also Duke, 42 Tex.
at 458 (The right to keep and bear arms “does not in-
clude … such pistols at least as are not adapted to being
carried concealed.”); MALCOLM, at 9 (discussing Henry VIII’s
ban on concealable firearms “frequently employed in crime”).
Under the reasoning in the majority opinion, both Bowie
knives and AR-15s would qualify as “particularly danger-
ous”—but not handguns. That cannot follow as a practical
matter and as a matter of history.
D. Nuanced Approach
As an alternative, the majority opinion invokes the so-
called “nuanced approach,” which derives from one sentence
in Bruen: “cases implicating unprecedented societal concerns
or dramatic technological changes may require a more nu-
anced approach.” 597 U.S. at 27; see also Duncan, 133 F.4th at
869–70. The majority’s primary holding does not rest on this
“nuanced approach,” but I submit this method should not be
invoked at all.
The “nuanced approach” is not its own test—it is a truism.
A 19th-century law regulating Bowie knives may be com-
pared with a contemporary ban on switchblades, as the
knives are similar. That same Bowie knife law, though, cannot
be compared to a ban on rocket launchers. To do so, more nu-
ance is necessary, which “undeniably necessitates an exercise
of judgment.” Wolford, 2026 WL 1825723, at *6.
Nos. 24-3060 et al. 87
But the application of the “nuanced approach” is too free-
wheeling. My colleagues believe that the “unprecedented so-
cietal concern of mass killings speedily carried out by lone
shooters,” facilitated by “dramatic technological change[s]
embodied in AR-15s equipped with large-capacity maga-
zines,” gives legislatures more leeway to regulate arms that
otherwise would be presumptively protected under the Sec-
ond Amendment. Yet that is not how Bruen has been applied.
Although courts focus on the Second Amendment’s princi-
ples, Rahimi, 602 U.S. at 692, those principles must be suffi-
ciently narrow to permit reasoning-by-analogy. “[A] court
must 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). Ironically, the “nuanced approach” allows
courts to be even less nuanced in drawing historical compar-
ators, directly contrary to the guidance from Bruen and
Rahimi.
Even more, applying that approach here makes it more
difficult to reconcile the majority opinion and Heller. Semiau-
tomatic handguns today are far more effective than the single-
shot, muzzle-loaded firearm that was ubiquitous at the
Founding. And many more mass shooters use handguns than
rifles—a ratio of roughly 74% using at least one handgun to
33% involving at least one rifle.23 The assertion that techno-
logical changes facilitated the rise of mass shootings cannot
justify a ban on AR-15s, because the same logic would require
banning handguns. Yet the Supreme Court has told us that a
handgun ban violated the Second Amendment. And the
23 Rockefeller Institute of Government, “Mass Shooting Factsheet,”
SUNY (June 3, 2026), https://rockinst.org/gun-violence/mass-shooting-
factsheet/.
88 Nos. 24-3060 et al.
Court did so over Justice Breyer’s dissent, which would have
made new societal concerns a central component of his anal-
ysis. Heller, 554 U.S. at 681–82, 696–99.
The majority opinion asserts that lone-wolf mass shoot-
ings are “unprecedented.” It states “the first known mass
shooting resulting in at least ten deaths did not occur until
1949.” Tragically, however, gun violence in America has a
much older pedigree. Documented instances of school shoot-
ings date back to at least the 1850s. 24 Mobs with guns often
committed serious atrocities, like when members of a white
militia murdered as many as 165 black Louisiania
This text is long and has been trimmed here. Open the source document for the complete record.