Robert Bevis v. City of Naperville

                               In the

     United States Court of Appeals
                 For the Seventh Circuit
                        ____________________
No. 23-1353
ROBERT BEVIS, et al.,
                                                Plaintiffs-Appellants,
                                 v.

CITY OF NAPERVILLE, ILLINOIS and JASON ARRES,
                                       Defendants-Appellees,

                                and

THE STATE OF ILLINOIS,
                                                Intervening Appellee.
                        ____________________

         Appeal from the United States District Court for the
           Northern District of Illinois, Eastern Division.
          No. 1:22-cv-04775 — Virginia M. Kendall, Judge.
                        ___________________

No. 23-1793
JAVIER HERRERA,
                                                  Plaintiff-Appellant,

                                 v.

KWAME RAOUL, et al.,
                                               Defendants-Appellees.
2                                                        Nos. 23-1353 et al.

                         ____________________

           Appeal from the United States District Court for the
             Northern District of Illinois, Eastern Division.
            No. 1:23-cv-00532 — Lindsay C. Jenkins, Judge.
                          ___________________

No. 23-1825 1
CALEB BARNETT, et al.,
                                                        Plaintiffs-Appellees,

                                       v.

KWAME RAOUL and BRENDAN F. KELLY,
                                                    Defendants-Appellants.
                         ____________________

           Appeals from the United States District Court for the
                       Southern District of Illinois.
          No. 3:23-cv-00209-SPM — Stephen P. McGlynn, Judge.
                         ____________________

      ARGUED JUNE 29, 2023 — DECIDED NOVEMBER 3, 2023
                   ____________________

    Before EASTERBROOK, WOOD, and BRENNAN, Circuit Judges.
    WOOD, Circuit Judge. The Second Amendment to the Con-
stitution recognizes an individual right to “keep and bear
Arms.” Of that there can be no doubt, in the wake of the

    1 Consolidated with No. 23-1826, Harrel v. Raoul (S.D. Ill. No. 3:23-cv-

00141-SPM); No. 23-1827, Langley v. Kelly (S.D. Ill. No. 3:23-cv-00192-SPM);
and No. 23-1828, Federal Firearms Licensees of Illinois, et al. v. Pritzker (S.D.
Ill. No. 3:23-cv-00215-SPM).
Nos. 23-1353 et al.                                                           3

Supreme Court’s decisions in District of Columbia v. Heller, 

 (2008); McDonald v. City of Chicago, 

(2010); Caetano v. Massachusetts, 

 (2016) (per cu-
riam); and New York State Rifle & Pistol Ass’n v. Bruen, 

 (2022). But as we know from long experience with
other fundamental rights, such as the right to free speech, the
right peaceably to assemble, the right to vote, and the right to
free exercise of religion, even the most important personal
freedoms have their limits. Government may punish a delib-
erately false fire alarm; it may condition free assembly on the
issuance of a permit; it may require voters to present a valid
identification card; and it may punish child abuse even if it is
done in the name of religion. The right enshrined in the Sec-
ond Amendment is no different.
    The present cases, which we have consolidated for dispo-
sition, relate to the types of “Arms” that are covered by the
Second Amendment. 2 This presents a line-drawing problem.
Everyone can agree that a personal handgun, used for self-
defense, is one of those Arms that law-abiding citizens must
be free to “keep and bear.” Everyone can also agree, we hope,
that a nuclear weapon such as the now-retired M388 Davy
Crockett system, with its 51-pound W54 warhead, can be re-
served for the military, even though it is light enough for one
person to carry.3 Many weapons, however, lie between these



    2 For ease of exposition, we will use the term Arms to refer to those

weapons that fall within the scope of the Second Amendment.
    3 See Matthew Seelinger, The M28/M29 Davy Crockett Nuclear Weapon

System, THE ARMY HISTORICAL FOUNDATION, https://armyhistory.org/the-
m28m29-davy-crockett-nuclear-weapon-system/; see also Jeff Schogol,
The Story of the ‘Davy Crockett,’ a Nuclear Recoilless Rifle Once Fielded by the
4                                               Nos. 23-1353 et al.

extremes. The State of Illinois, in the legislation that lies at the
heart of these cases, has decided to regulate assault weapons
and high-capacity magazines—a decision that is valid only if
the regulated weapons lie on the military side of that line and
thus are not within the class of Arms protected by the Second
Amendment. Several municipalities have done the same. The
plaintiffs in these cases challenge that conclusion. Using the
tools of history and tradition to which the Supreme Court di-
rected us in Heller and Bruen, we conclude that the state and
the affected subdivisions have a strong likelihood of success
in the pending litigation. We therefore affirm the decisions of
the district courts in appeals No. 23-1353 and 23-1793 refusing
to enjoin these laws, and we vacate the injunction issued by
the district court in appeals No. 23-1825, 23-1826, 23-1827, and
23-1828.
                       I.     Background
                            A. The Act
    At the center of these appeals lies a new statute in Illinois
that took effect on January 10, 2023—a measure called the Pro-
tect Illinois Communities Act, Pub. Act 102-1116 (2023) (“the
Act”). Some of the consolidated cases also implicate three mu-
nicipal laws that cover much of the same ground, though the
details vary: Cook County Ordinances No. 54-210 to 54-215;
City of Chicago Municipal Ordinances 8-20-010 to 8-20-100;
and City of Naperville Ordinances No. 3-19-1 to 3-19-3. We
make note of the municipal laws only when their specific pro-
visions affect our analysis. For the interested reader, the chart



US Army, TASK & PURPOSE (Sept. 19, 2022), https://taskandpur-
pose.com/history/army-davy-crockett-tactical-nuclear-weapon/.
Nos. 23-1353 et al.                                             5

in the Appendix to this opinion summarizes the relevant dif-
ferences among these enactments.
   The Act is a sprawling piece of legislation made up of 99
sections that cover a vast array of regulatory and record-keep-
ing matters, along with the provisions of interest here. The
Act’s wide scope led to a challenge in Illinois’s courts for fail-
ing to comply with state-law requirements such as the single-
subject rule, the three-readings requirement, and the ban on
special legislation. See Caulkins v. Pritzker, 

(Aug. 11, 2023). The state supreme court upheld the Act
against those contentions, and it also ruled that the Act did
not violate the state constitution’s equal protection clause. It
did not reach any argument about the Second Amendment,
because it found that the plaintiffs had waived any reliance
on that theory. The plaintiffs in these cases have not argued
that the Act is invalid under state law.
    The critical part of the Act for our purposes is its treatment
of so-called assault weapons and large-capacity magazines.
Those sections institute something close to a ban on “assault
weapons,” through the Act’s general prohibitions of the sale,
possession, and use of a defined set of weapons. The Act also
bans large-capacity magazines. The plaintiffs have not speci-
fied exactly which provisions of the Act they believe are un-
constitutional under the Second Amendment, but we assume
that their principal targets are 720 ILCS 5/24-1.9 and 5/24-1.10.
Section 5/24-1.9 addresses the “[m]anufacture, possession, de-
livery, sale, and purchase of assault weapons, .50 caliber rifles,
and .50 caliber cartridges,” and section 5/24-1.10 deals with
“[m]anufacture, delivery, sale, and possession of large capac-
ity ammunition feeding devices.”
6                                                  Nos. 23-1353 et al.

    The Act defines “assault weapon” using language that is
largely borrowed from the expired Federal Assault Weapons
Ban, which was a subsection of the Violent Crime Control and
Law Enforcement Act of 1994, 

 Stat.
1796. 4 The Illinois Act bans certain semiautomatic rifles and
pistols. A semiautomatic rifle falls under the Act’s proscrip-
tions if it has the capacity to accept a detachable magazine and
one or more of the following features: a pistol grip or
thumbhole stock; any feature capable of functioning as a pro-
truding grip for the non-trigger hand; a folding, telescoping,
thumbhole, or detachable stock or a stock that otherwise en-
hances the concealability of the weapon; a flash suppressor; a
grenade launcher; or a barrel shroud. 720 ILCS 5/24-
1.9(a)(1)(A). The definition also includes a semiautomatic rifle
with a fixed magazine capacity of greater than 10 rounds, ex-
cept those that accept only .22 caliber rimfire ammunition. 

5/24-1.9(a)(1)(B). Finally, there is a lengthy list of particular
models that fall within the scope of the statute. See 5/24-
1.9(a)(1)(J). Subpart (i) of that section covers all AK weapons,
and subpart (ii) covers all AR types. In the remainder of this
opinion, we will refer often to the AR-15 as a paradigmatic
example of the kind of weapon the statute covers. We use it
only illustratively, however; our analysis covers everything
mentioned in the Act.
    The Act makes it unlawful for any person within Illinois
knowingly to “manufacture, deliver, sell, import, or pur-
chase … an assault weapon, assault weapon attachment, .50
caliber rifle, or .50 caliber cartridge.” 

 5/24-1.9(b). (Unless
the context requires otherwise, from this point we use the

    4 The more formal name of the relevant part of the law was the Public

Safety and Recreational Firearms Use Protection Act.
Nos. 23-1353 et al.                                            7

term “assault weapon” to cover all four covered items, in the
interest of readability.) With some exceptions, the Act also
makes it unlawful as of January 1, 2024, for any person within
the state knowingly to “possess an assault weapon.” 

5/24-1.9(c).
    There are two significant exceptions to these prohibitions.
Using the terminology the Supreme Court of Illinois adopted
in Caulkins, the first is for “trained professionals” and the sec-
ond is for “grandfathered individuals.” 

.
The list of trained professionals, set forth in 5/24-1.9(e), in-
cludes peace officers; qualified active and retired law-enforce-
ment officers; prison wardens and “keepers”; members of the
Armed Services, Reserves, or Illinois National Guard; nuclear
facility guards; and licensed private security personnel. 

5/24-1.9(e)(1)–(7). The “grandfather” provision can be found
at 5/24-1.9(d). It states that the Act’s prohibitions do “not ap-
ply to a person’s possession of an assault weapon … if the
person lawfully possessed” that weapon as of the effective
date of the law and then the person “provide[s] in an endorse-
ment affidavit, prior to January 1, 2024, under oath or affirma-
tion” certain specified information to the Illinois State Police.

 5/24-1.9(d)(1)–(3). A completed endorsement affidavit
“creates a rebuttable presumption that the person is entitled
to possess and transport the assault weapon.” 

 5/24-1.9(d),
at ¶ 2. The Act restricts the places where authorized persons
may possess their weapons to the following: (1) private prop-
erty owned or controlled by the person; (2) other private
property, with the express permission of the owner or con-
troller; (3) premises of a licensed firearms dealer or gunsmith
for lawful repairs; (4) licensed firing ranges or sport shooting
competition venues; and (5) in transit to or from any of those
locations, if the weapon is unloaded and in a container. 

8                                            Nos. 23-1353 et al.

5/24-1.9(d), at ¶ 3(1)–(5). The parties have not focused on
these locational restrictions, and so neither will we.
    Section 5/24-1.10 sets out the rules for large-capacity am-
munition feeding devices. They are defined as a magazine (or
similar mechanism) that can accept “more than 10 rounds of
ammunition for long guns and more than 15 rounds of am-
munition for handguns.” 

 5/24-1.10(a), at ¶ 3(1). This provi-
sion also grandfathers in those who lawfully possessed a
large-capacity magazine before the effective date of the Act,
so long as the device is used in a permitted place. 

 5/24-
1.10(d). It has an analogous set of exceptions for trained pro-
fessionals. 

 5/24-10(d), at ¶ 1.
    Broadly speaking, violations of the assault-weapon ban
are classified as felonies when the violation involves guns or
gun parts, and as misdemeanors when the violation involves
.50 caliber cartridges. 

 5/24-1(b).
                       B. The Lawsuits
   The ink was barely dry on the pages of the Act when liti-
gation began. Before us now are six related cases, in which 26
plaintiffs have challenged the Act and the three municipal or-
dinances we mentioned earlier. All of the challengers contend
that the legislation in question violates their Second Amend-
ment right to keep and bear Arms. A brief review of the indi-
vidual cases should help keep the issues straight.
          1. Bevis v. City of Naperville (No. 23-2353)
    This case, filed in the Northern District of Illinois, was
brought by three parties: (1) Robert Bevis, a Naperville resi-
dent and owner of Law Weapons, Inc.; (2) Law Weapons, Inc.,
a commercial firearms store in Naperville; and (3) the Na-
tional Association for Gun Rights. We refer to them
Nos. 23-1353 et al.                                              9

collectively as Bevis. Once the suit was filed and landed in
Judge Kendall’s court, Bevis’s first step was to seek a prelimi-
nary injunction against both the Naperville ordinance and the
Act. They were unsuccessful. Applying the standard four-
part test for preliminary injunctions established in Winter v.
Natural Resources Defense Council, Inc., 

 (2008),
Judge Kendall decided that the plaintiffs were unlikely to suc-
ceed on the merits. This would have been an easy conclusion
under our decision in Friedman v. City of Highland Park, 

 (7th Cir. 2015), but the judge was concerned that
Friedman’s methodology may have been undermined by
Bruen, and so she undertook a fresh analysis of the merits us-
ing only Bruen. (We address Friedman’s continuing vitality be-
low.)
    Judge Kendall’s efforts convinced her that “[t]he history of
firearm regulation … establishes that governments enjoy the
ability to regulate highly dangerous arms (and related dan-
gerous accessories).” Bevis v. City of Naperville, No. 22 C 4775,

, at *14 (N.D. Ill. Feb. 17, 2023). She took par-
ticular note of longstanding regulations on Bowie knives and
other “melee weapons.” 

 at *10–11. Next, she found that as-
sault weapons fit within this tradition because they pose “an
exceptional danger” compared with “standard self-defense
weapons such as handguns.” Id. at *14. Critically for our pur-
poses, after citing statistics about the lethality and injury rates
of assault weapons, id., she highlighted the fact that “[a]ssault
rifles can … be easily converted to … mimic military-grade
machine guns,” id. at *15. Quoting from the Fourth Circuit,
she observed that
       the very features that qualify a firearm as a
       banned assault weapon—such as flash
10                                            Nos. 23-1353 et al.

       suppressors, barrel shrouds, folding and tele-
       scoping stocks, pistol grips, grenade launchers,
       night sights, and the ability to accept bayonets
       and large-capacity magazines—serve specific,
       combat-functional ends.
Id. (quoting Kolbe v. Hogan, 

 (4th Cir. 2017) (en
banc), abrogated on other grounds by Bruen, 142 S. Ct. at 2126–27)
(cleaned up). Finally, the judge noted that the high-capacity
magazines exhibited similar dangers. 

    This was enough, in her view, to show that the plaintiffs
were not likely to succeed on the merits. Quickly looking at
the other three criteria for a preliminary injunction, she also
found that without a presumption of irreparable harm related
to the alleged Second Amendment violation, plaintiffs could
not satisfy that factor. Bevis had not shown that the gun shop
would lose substantial sales because of the two laws, and the
organizational members retained other effective weapons for
self-defense. Id. at *16. Finally, Judge Kendall concluded that
neither the balance of equities nor the public interest favored
plaintiffs sufficiently to overcome the inadequate showing on
the other issues. Id. at *17.
               2. Herrera v. Raoul (No. 23-1793)
   The plaintiff in our next case, Javier Herrera, is a Chicago
emergency room doctor who owns several assault weapons
and large-capacity magazines. After the Act was passed, he
filed a suit seeking both a temporary restraining order and a
preliminary injunction against the Act, the Chicago ordi-
nance, and the Cook County ordinance. Unlike Bevis, he also
challenged the Act’s registration requirements (through
which the grandfathering provisions are administered). This
Nos. 23-1353 et al.                                           11

case was assigned to Judge Jenkins, who largely agreed with
the reasoning in Bevis. See Herrera v. Raoul, No. 23 CV 532,

 (N.D. Ill. Apr. 25, 2023). She rejected
Hererra’s attempt to distinguish Bevis on the ground that his
suit focused on the defense of his home, rather than on the
public-carry right. Although she recognized that the analo-
gies to Bowie knives and melee weapons were not perfect, she
noted that Bruen did not demand a “dead ringer” or a “histor-
ical twin,” especially if there are “‘dramatic technological
changes’ or ‘unprecedented societal concerns’ [that] may re-
quire a ‘more nuanced approach.’” Id. at *7, *9 (quoting Bruen,
142 S. Ct. at 2133, 2132).
    With respect to the need to register a covered weapon in
order to take advantage of the Act’s grandfathering provision,
Judge Jenkins first assured herself that the question was ripe
even though Herrera had not yet taken steps to register his
guns. Id. at *8. Herrera made clear that he intended to disobey
that law, that his intended conduct “[ran] afoul of a criminal
statute,” and that the effective date of the registration require-
ment was “sufficiently imminent.” Id. (quotations omitted).
On the merits, however, she concluded that Herrera was un-
likely to succeed because historical evidence showed that the
“colonies required gun registration in a variety of ways,” such
as colonial “muster” requirements and a variety of tax re-
quirements, “which in essence required that firearms be iden-
tified and disclosed to the government.” Id. at *9. She also took
note of several 19th- and 20th-century laws as evidence of a
“continuing tradition of state and national registration re-
quirements.” Id. She found support for her ruling in the Bruen
Court’s comment that “nothing in our analysis should be in-
terpreted to suggest the unconstitutionality of existing ‘shall-
12                                            Nos. 23-1353 et al.

issue’ licensing laws.” Id. at *10 (quoting Bruen, 142 S. Ct. at
2138 n.9 (cleaned up)).
    Although lack of likely success on the merits pointed
strongly toward denial of preliminary injunctive relief, Judge
Jenkins also looked briefly at the other three factors and found
that they pointed in the same direction. She rejected the argu-
ment that there is an established presumption of irreparable
harm for all Second Amendment challenges. Id. at *11. She
was also unpersuaded by Herrera’s argument that the laws
prevented him from protecting himself in his home and at-
tending his monthly SWAT training (because of the commute
time to retrieve his assault weapons from an out-of-county lo-
cation). Herrera owned other compliant guns suitable for self-
defense, and he had managed the commute since 2018. Id. at
*12. Lastly, she found that neither the public interest nor the
equities pushed the needle far enough to justify an injunction.
Id. at *13.
               3. Barnett v. Raoul (No. 23-1825)
    The perspective reflected in the third case, which arose in
the Southern District of Illinois, is quite different from the first
two. In Barnett and the three other cases that were consoli-
dated with it, the plaintiffs included individual gun owners,
commercial firearms dealers, and various organizations de-
voted to protecting and enhancing Second Amendment
rights. Like their counterparts in the Northern District, these
plaintiffs sought a preliminary injunction against the Act. Un-
like the others, they succeeded. Judge McGlynn concluded
that because the plaintiffs had brought a facial challenge to
the Act, “the entirety of [the Act] as codified will be enjoined.”
Barnett v. Raoul, No. 3:23-cv-00209-SPM (Lead Case), 

, at *2 (S.D. Ill. Apr. 28, 2023). (We put to one side the
Nos. 23-1353 et al.                                           13

fact that there are many provisions of the Act that have noth-
ing to do with gun ownership or regulation. See generally
Pub. Act 102-1116 (2023). Presumably the judge did not mean
to enjoin them, but if that is so, then the injunction does not
comply with Federal Rule of Civil Procedure 65. That rule re-
quires an injunction to indicate clearly what is forbidden or
mandated—a rule necessitated by the fact that injunctions are
enforceable by contempt. We need not explore this further,
given our ultimate conclusion in these appeals.)
    With obvious reference to the two sections of the Act that
address assault weapons and high-capacity magazines, Judge
McGlynn chose to start with the issue of irreparable injury,
rather than likelihood of success on the merits. He found that
there is a presumption of irreparable harm when plaintiffs
mount a facial challenge under the Second Amendment, and
even if there were not, these plaintiffs had shown irreparable
injury because the restrictions on their ability to buy or sell
the weapons and accessories covered by the Act limited their
right to armed self-defense. 

, at *4–5.
    The judge then moved on to likelihood of success on the
merits. He rejected the defendants’ arguments that many of
the Act’s provisions regulated only accessories (such as
threaded barrels and pistol grips), which in themselves were
not the Arms protected by the Second Amendment. Those
items were “important corollar[ies] to the meaningful exercise
of the core right to possess firearms for self-defense.” 

 at *8
(quoting Ezell v. City of Chicago, 

 (7th Cir.
2011)). He then moved on to consider whether the Act was
“consistent with this Nation’s historical tradition of firearm
regulation.” Id. at *9. For this purpose, he assigned to the de-
fendants the burden of “(1) demonstrat[ing] that the ‘arms’ in
14                                            Nos. 23-1353 et al.

[the Act] are not in ‘common use;’ and (2) ‘identify[ing] a well-
established and representative historical analogue’ to [the
Act].” Id. (quoting Bruen, 142 S. Ct. at 2128, 2133). He rejected
the defendants’ argument that the weapons had to be in com-
mon use for self-defense. The defendants failed to carry their
burden, he held, because they “focused almost entirely on
AR-15 rifles and their commonality or lack thereof” instead of
the many other weapons and accessories covered by the Act.
Id. at *10. Accepting an argument of the plaintiffs in the cases
now before us (as well as their amici curiae), the judge held
that AR-15s and large-capacity magazines are “in common
use” because a large number of people own them. Id.
    Wrapping up, the judge characterized the defendants’
proposed historical analogues as inapt, because they were
simply concealed-carry regulations, not outright bans on pos-
session. Id. at *11. The balance of harms, in his view, decidedly
favored the plaintiffs, as (in his words) “there can be no harm
to a government agency when it is prevented from enforcing
an unconstitutional statute,” id. (cleaned up and quotation
omitted), and he saw no evidence in the record indicating
how the Act would help Illinois communities. He noted that
the Act “was purportedly enacted in response to the Highland
Park [mass] shooting,” id. at *12, but that fact was not enough
to overcome the injury it inflicted.
                    II.    Governing Law
             A. Preliminary Injunction Standard
    As our account of the proceedings in the district courts
shows, we are not here today to rule definitively on the con-
stitutionality of the Act or any of the municipal ordinances.
The only issue before us concerns preliminary injunctive
Nos. 23-1353 et al.                                           15

relief. The Bevis and Herrera courts denied motions for such
an injunction, which would have suspended the operation of
720 ILCS 5/24-1.9 and 5/24-1.10 (and the corresponding Na-
perville, Chicago, and Cook County ordinances), and the Bar-
nett court granted the injunction (ostensibly against the entire
Act, as we mentioned). We entered a stay of the Barnett injunc-
tion pending the resolution of these interlocutory appeals,
which are authorized by 

(a)(1); the order stip-
ulated that the stay would remain in effect “until these ap-
peals have been resolved and the court’s mandate has is-
sued.”
    As we mentioned earlier, the leading Supreme Court de-
cision establishing the standard for granting preliminary in-
junctive relief is Winter v. Natural Resources Defense Council,
Inc., 

 (2008). The Court summarized the pertinent
requirements as follows:
           A plaintiff seeking a preliminary injunction
       must establish that he is likely to succeed on the
       merits, that he is likely to suffer irreparable
       harm in the absence of preliminary relief, that
       the balance of equities tips in his favor, and that
       an injunction is in the public interest.

. It elaborated on these factors in a later case dealing
with the criteria for staying a court decision, Nken v. Holder,

 (2009), noting there that “[t]here is substantial
overlap between [the criteria for a stay] and the factors gov-
erning preliminary injunctions.” 

 at 434 (citing Winter, 

). The two most important considerations are likeli-
hood of success on the merits and irreparable harm. 

 With
respect to the former, the Court said that “[i]t is not enough
that the chance of success on the merits be ‘better than
16                                              Nos. 23-1353 et al.

negligible.’” 

 (quoting and disapproving Sofinet v. INS, 

 (7th Cir. 1999)). Nor is a mere possibility enough.

 As we put it in Illinois Republican Party v. Pritzker, 

, 763 (7th Cir. 2020), although the party seeking the injunc-
tion need not demonstrate likelihood of success by a prepon-
derance of the evidence, that party must nevertheless make a
“strong” showing that reveals how it proposes to prove its
case. Similarly, a mere possibility of irreparable harm will not
suffice. See Nken, 556 U.S. at 434–35; Winter, 

.
    Decisions such as Winter and Nken reflect the fact that “[a]
preliminary injunction is an extraordinary remedy never
awarded as of right.” Winter, 

. The party seeking
the injunction bears the burden of showing that this type of
relief is warranted. Nken, 556 U.S. at 433–34. We must also
bear in mind, when a party is seeking to enjoin a statute, that
legislative enactments are entitled to a presumption of consti-
tutionality. See Flemming v. Nestor, 

 (1960)
(quoting Fletcher v. Peck, 

 (1810)).
Though we carefully evaluate any claim that a statute violates
the Constitution, we assume that the legislative body—
whether Congress or a state legislature—was aware of consti-
tutional limitations and endeavored to follow them.
    Finally, we note that a hybrid standard of review applies
to interlocutory review of a preliminary injunction: “we re-
view the district court’s findings of fact for clear error, its legal
conclusions de novo, and its balancing of the factors for a pre-
liminary injunction for abuse of discretion.” Doe v. University
of Southern Indiana, 

, 791 (7th Cir. 2022) (brackets
and quotation omitted).
Nos. 23-1353 et al.                                             17

                  B. The Second Amendment
    The basic contours of the second article of the Bill of Rights
have become familiar, and so we will only summarize them
here. In a crisp, if not enigmatic, way, it says this: “A well reg-
ulated Militia, being necessary to the security of a free State,
the right of the people to keep and bear Arms, shall not be
infringed.” U.S. CONST. amend. II. For many years, both the
Supreme Court and scholars thought that there was a relation
between the prefatory clause, which refers to the Militia, and
the operative clause, which refers to the right to keep and bear
Arms. See, e.g., ROBERT H. BORK, SLOUCHING TOWARDS
GOMORRAH: MODERN LIBERALISM AND AMERICAN DECLINE 166
(rev. ed. 2003). But in Heller the Supreme Court severed that
connection. Undertaking its own examination of the events
that led up to the Amendment’s inclusion in the Constitution,
it concluded that the Amendment recognized an individual
right to keep and bear Arms.
   At the same time, Heller held that “[l]ike most rights, the
right secured by the Second Amendment is not unlimited.”

. It continued as follows:
       From Blackstone through the 19th-century
       cases, commentators and courts routinely ex-
       plained that the right was not a right to keep
       and carry any weapon whatsoever in any man-
       ner whatsoever and for whatever purpose.

 This opened up new frontiers of litigation: Which weapons
are covered? What manner of “keeping and bearing” is pro-
tected? What purpose must or may the user have? Which peo-
ple hold this right? The Heller Court recognized that there was
much left to be resolved. It did give some hints, however. One
18                                            Nos. 23-1353 et al.

important tea leaf for present purposes was its refusal to en-
dorse the idea that the Amendment protects “only those
weapons useful in warfare.” 

. It called this a “star-
tling reading,” since that would have implied that ma-
chineguns—quintessential weapons used exclusively by the
military, not private citizens—could not be regulated, in the
face of the National Firearms Act’s restrictions on those weap-
ons. Id.; see also 

 Stat. 1236 (1934).
    Perhaps the most important expansion of Heller occurred
in McDonald, in which the Supreme Court confirmed that the
Second Amendment, like the First, Fourth, Fifth, Sixth, and
Eighth Amendments, applies to the states through incorpora-
tion pursuant to the Fourteenth Amendment. See 

. The late date of the McDonald decision—2010—explains
why there are so few cases exploring the Second Amendment
implications of state laws regulating weapons from the time
the Amendment became part of the Constitution (1791) to the
present. Under the view that prevailed until McDonald, the
states were free to regulate weapons in any way compatible
with their own constitutions. See generally Jeffrey S. Sutton,
51 Imperfect Solutions: States and the Making of American Consti-
tutional Law (2018). And they did so in countless ways—a
point of some significance when we come to consider the his-
tory and tradition of regulation in this area.
    After McDonald, most courts of appeals adopted a two-
step test for legality under the Second Amendment. See, e.g.,
Ezell, 651 F.3d at 702–03. Step one asked whether the “chal-
lenged firearms law regulates activity falling outside the
scope of the Second Amendment right as it was understood
[historically].” 

 If the regulated activity was unprotected,
then the law in question was not subject to further Second
Nos. 23-1353 et al.                                           19

Amendment review. If, however, history showed that the ac-
tivity was protected, or the evidence was inconclusive, step
two called on the court to balance the public benefit the gov-
ernment was seeking to achieve against the regulatory means
it selected, using a form of heightened scrutiny. Id. at 703.
    Some courts, including our own, steered clear of that two-
step approach. That explains the path we chose in Friedman,
which dealt with exactly the same issue we face now: a ban
on assault weapons and large-capacity magazines. Although
the district court in Bevis thought that the reasoning in Fried-
man might not have survived Bruen, we see Friedman as basi-
cally compatible with Bruen, insofar as Friedman anticipated
the need to rest the analysis on history, not on a free-form bal-
ancing test.
    After briefly reviewing the holdings in Heller and McDon-
ald, Friedman turned to the question of the scope of the indi-
vidual right to keep and bear Arms. It began by summarizing
the Court’s own historical analysis in Heller:
       [The Court] cautioned against interpreting the
       [Heller] decision to cast doubt on “longstanding
       prohibitions,” including the “historical tradi-
       tion of prohibiting the carrying of ‘dangerous
       and unusual weapons.’” [554 U.S.] at 623, 627. It
       observed that state militias, when called to ser-
       vice, often had asked members to come armed
       with the sort of weapons that were “in common
       use at the time”, id. at 624, and it thought these
       kinds of weapons (which have changed over the
       years) are protected by the Second Amendment
       in private hands, while military-grade weapons
       (the sort that would be in a militia’s armory),
20                                               Nos. 23-1353 et al.

       such as machine guns, and weapons especially
       attractive to criminals, such as short-barreled
       shotguns, are not. Id. at 624–25.
784 F.3d at 407–08. The plaintiffs in Friedman had contended
that “there is no ‘historical tradition’ of banning possession of
semi-automatic guns and large-capacity magazines.” Id. at
408. But, we observed, “this argument proves too much: its
logic extends to bans on machine guns, … [but] Heller deemed
a ban on private possession of machine guns to be obviously
valid.” Id. (citing Heller, 

). That was so even
though states “didn’t begin to regulate private use of machine
guns until 1927,” and the federal government did not do so
until 1934. 

   The critical question of “[h]ow weapons are sorted be-
tween private and military uses,” we noted, “has changed
over time.” 

 Anticipating Bruen, we rejected a historical fo-
cus on the 1920s, when these bans started to come into exist-
ence, and turned instead to the time of the Second Amend-
ment’s adoption. 

 With respect to the common ownership
and use question, we cautioned against circular reasoning:
       Machine guns aren’t commonly owned for law-
       ful purposes today because they are illegal;
       semi-automatic weapons with large-capacity
       magazines are owned more commonly because,
       until recently (in some jurisdictions), they have
       been legal. Yet it would be absurd to say that the
       reason why a particular weapon can be banned
       is that there is a statute banning it, so that it isn’t
       commonly owned. A law’s existence can’t be the
       source of its own constitutional validity.
Nos. 23-1353 et al.                                                      21

Id. at 409. 5 We were not persuaded by the plaintiffs’ efforts to
put semiautomatic weapons on the “private” or “mixed” side
of the line between private or mixed private/military weap-
ons, on the one hand, and weapons exclusively for military
use, on the other. We were reluctant to place semiautomatic
weapons in the former category for the simple reason that the
Heller Court had not done so. Instead, in distinguishing United
States v. Miller, 

 (1939), we reaffirmed “the rule
that the Second Amendment does not authorize private per-
sons to possess weapons such as machine guns and sawed-off
shotguns that the government would not expect (or allow) cit-
izens to bring with them when the militia is called to service.”

.
    Conspicuously absent from our Friedman analysis is any
hint of the two-part test that Bruen disapproved. We looked
instead to the type of Arms that the Second Amendment has
always protected for private use and contrasted them with
weapons reserved for military use. We expressly declined to
subject Highland Park’s law to means-end scrutiny. 

.
Instead, we said, “we think it better to ask whether a regula-
tion bans weapons that were common at the time of ratifica-
tion or those that have ‘some reasonable relationship to the
preservation or efficiency of a well regulated militia,’ … and
whether law-abiding citizens retain adequate means of self-
defense.” 

 (quoting Heller, 

 (quoting Miller,

)). This approach, we believe, is consistent with
the methodology approved in Bruen.


    5 The dissent embraces the reasoning we rejected in Friedman; it asserts

that circularity concerns are more hypothetical than actual. See post at 62
n.4 (citing Friedman, 

 n.5 (Manion, J., dissenting)).
22                                              Nos. 23-1353 et al.

    Pointing to Wilson v. Cook County, 

 (7th Cir.
2019) (per curiam), the dissent sees Friedman differently. It
notes that one can find language in Wilson that characterizes
Friedman as “evaluat[ing] the importance of the reasons for
the [assault weapons ban] to determine whether they justified
the ban’s intrusion on Second Amendment rights.” 937 F.3d
at 1036. But this language is pure dicta. It may represent the
Wilson panel’s attempt to put a gloss on Friedman, but it did
not change the actual legal test that Friedman applied. The is-
sue in Wilson, recall, was whether Friedman could be recon-
ciled with Ezell, which struck down Chicago’s ban on firing
ranges within city limits. See id. at 1035. On that issue, Wilson
found that “Friedman fits comfortably under the umbrella of
Ezell” and that it “represents the application and extension of
its principles to the specific context of a ban on assault weap-
ons and large-capacity magazines.” Id. at 1036. Indeed, Wilson
is notable for what it did not say: it never said that Friedman
had used intermediate scrutiny or means-end balancing; and
it did not depict Friedman as evaluating only the importance
of the reasons behind the ordinance at issue there. The fleeting
reference to the city’s reasons for adopting the ordinance, in
short, was not part of the panel’s reasoning, and so, while cer-
tainly disapproved in Bruen, does not undermine the central
analysis in the case.
    We have now referred many times to Bruen, and finally, it
takes center stage. Rejecting the two-part test adopted by the
courts of appeals (which it derided as having “one step too
many,” 142 S. Ct. at 2127), the Bruen Court elaborated on the
test that Heller requires. See 142 S. Ct. at 2129–30. First, it said,
the trial court must decide whether “the Second Amend-
ment’s plain text covers an individual’s conduct.” Id. If so,
then “the Constitution presumptively protects that conduct.”
Nos. 23-1353 et al.                                             23

Id. at 2130. The analysis then moves to the second step, which
calls on the “government [to] justify its regulation by demon-
strating that it is consistent with the Nation’s historical tradi-
tion of firearm regulation.” Id. The Court predicted that this
second step would be relatively easy in some instances, when
historical analogues are easy to find. But in other instances, it
recognized that the task would be challenging. It singled out
“cases implicating unprecedented societal concerns or dra-
matic technological changes,” which “may require a more nu-
anced approach.” Id. at 2132.
    Bruen also confirmed some additional points that inform
our analysis. First, the Court said (not for the first time) that
the Arms protected by the Second Amendment are not limited
to those that were in existence at the time of its ratification,
1791, or at the time the Fourteenth Amendment took effect,
1868. Id. Second, the search is for a historical regulation that is
relevantly similar, not identical. Bearing in mind that “the cen-
tral component” of the Second Amendment right is individual
self-defense, id. at 2133 (quoting McDonald, 

(emphasis in original)), the question is whether the modern
and historical regulations “impose a comparable burden on
the right of armed self-defense and whether that burden is
comparably justified,” 

 And the Court made it clear that this
search was a meaningful one, not just a subterfuge for either
upholding or striking down all modern laws:
       [A]nalogical reasoning under the Second
       Amendment is neither a regulatory straight-
       jacket nor a regulatory blank check. On the one
       hand, courts should not uphold every modern
       law that remotely resembles a historical ana-
       logue, because doing so risk[s] endorsing
24                                            Nos. 23-1353 et al.

       outliers that our ancestors would never have ac-
       cepted. On the other hand, analogical reasoning
       requires only that the government identify a
       well-established and representative historical
       analogue, not a historical twin. So even if a mod-
       ern-day regulation is not a dead ringer for his-
       torical precursors, it still may be analogous
       enough to pass constitutional muster.

 (quotation and citation omitted, and second alteration and
emphases in original). Finally, the Court’s decision in Bruen
builds on, rather than disturbs, Heller and McDonald. See id. at
2157 (Alito, J., concurring); id. at 2161 (Kavanaugh, J., concur-
ring). Justice Alito in particular took care to make this point
when he wrote “[n]or does [Bruen] decide anything about the
kinds of weapons that people may possess.” Id. at 2157 (Alito,
J., concurring). Bruen simply “made the constitutional stand-
ard endorsed in Heller more explicit” and applied it to the
handgun regulation at issue. Id. at 2134.
   Our task is to apply Bruen’s methodology to the four laws
before us. We begin by assessing whether the assault weapons
and large-capacity magazines described in those laws are
Arms for purposes of the Second Amendment. If not, then the
Second Amendment has nothing to say about these laws:
units of government are free to permit them, or not to permit
them, depending on the outcome of the democratic process. If
they are properly characterized as Arms, then we must pro-
ceed to Bruen’s second step, at which the governments bear
the burden of proof, and determine whether these laws pass
muster.
Nos. 23-1353 et al.                                             25

               III.    Application to the Cases
            A. Are the Covered Weapons “Arms”?
    We begin by looking at the “plain text” of the Second
Amendment to see whether the assault weapons and large-
capacity magazines (terms that we, like the parties, continue
to use as short-hand for the many items covered by these
laws) fall within the scope of the “Arms” that individual per-
sons are entitled to keep and bear. Both Supreme Court deci-
sions and historical sources indicate that the Arms the Second
Amendment is talking about are weapons in common use for
self-defense. That is not to say that there are no other lawful
uses for weapons—sporting uses, collection, and competi-
tions come to mind as examples. But the constitutional pro-
tection exists to protect the individual right to self-defense,
and so that will be our focus.
   Our starting point is, once again, Heller. It began by inter-
preting the object of the Second Amendment right: Arms. See

. It is worth a close look at this part of the opin-
ion:
       The 18th-century meaning is no different from
       the meaning today. The 1773 edition of Samuel
       Johnson’s dictionary defined “arms” as
       “[w]eapons of offence, or armour of defence.” 1
       Dictionary of the English Language 106 (4th ed.)
       (reprinted 1978). Timothy Cunningham’s im-
       portant 1771 legal dictionary defined “arms” as
       “any thing that a man wears for his defence, or
       takes into his hands, or useth in wrath to cast at
       or strike another.” 1 A New and Complete Law
       Dictionary; see also N. Webster, American
26                                            Nos. 23-1353 et al.

       Dictionary of the English Language (1828) (re-
       printed 1989) (similar).
           The term was applied, then as now, to weap-
       ons that were not specifically designed for military
       use and were not employed in a military capacity.
       For instance, Cunningham’s legal dictionary
       gave as an example of usage: “Servants and la-
       bourers shall use bows and arrows on Sundays,
       &c. and not bear other arms.” … Although one
       founding-era thesaurus limited “arms” (as op-
       posed to “weapons”) to “instruments of offence
       generally made use of in war,” even that source
       stated that all firearms constituted “arms.” 1 J.
       Trusler, The Distinction Between Words Es-
       teemed Synonymous in the English Language
       37 (3d ed. 1794) (emphasis added).
554 U.S. at 581–82 (first emphasis and ellipsis added, and
“hereinafter” parentheticals omitted). Summarizing, the
Court said that “the Second Amendment extends, prima facie,
to all instruments that constitute bearable arms.” 

.
    But what exactly falls within the scope of “bearable”
Arms? Not machineguns, the Court said, because they can be
dedicated exclusively to military use. See 

. Yet a nor-
mal person can certainly pick up and carry a machinegun, or
for that matter the portable nuclear weapons we mentioned at
the outset. “Bearable” thus must mean more than “transport-
able” or “capable of being held.” See 

 (discussing
“weapons that are most useful in military service—M16 rifles
and the like,” which “may be banned”).
Nos. 23-1353 et al.                                             27

    The Court’s comments about the role of the militia shed
light on the scope of the term “Arms.” It explained that “[t]he
traditional militia was formed from a pool of men bringing
arms ‘in common use at the time’ for lawful purposes like self-
defense.” 

. It then concluded that “the Second
Amendment does not protect those weapons not typically
possessed by law-abiding citizens for lawful purposes, such
as short-barreled shotguns. That accords with the historical
understanding of the scope of the right.” 

 (emphasis
added). We take from this that the definition of “bearable
Arms” extends only to weapons in common use for a lawful
purpose. That lawful purpose, as we have said several times,
is at its core the right to individual self-defense.
    This approach is consistent with the historical antecedents
on which the Second Amendment was based. Chief among
those was the 1689 English Bill of Rights, which is a key pre-
cursor to the bills of rights in the U.S. state and federal consti-
tutions. The 1689 Bill of Rights “explicitly protected a right to
keep arms for self-defense.” McDonald, 

. Simi-
larly, Blackstone explained that at the root of the right to bear
arms, there is a “natural right of resistance and self-preserva-
tion,” and “the right of having and using arms for self-preser-
vation and defence.” Heller, 

 (quoting
1 WILLIAM BLACKSTONE, COMMENTARIES *139, *140). State con-
stitutional protections from the Founding Era confirm this un-
derstanding. As Heller observed, “nine state constitutional
provisions written in the 18th century or the first two decades
of the 19th … enshrined a right of citizens to bear arms in de-
fense of themselves and the state or bear arms in defense of
himself and the state.” 554 U.S. at 584–85, 585 n.8 (citing the
state constitutions of Pennsylvania, Vermont, Kentucky, Ohio,
28                                            Nos. 23-1353 et al.

Indiana, Mississippi, Connecticut, Alabama, and Missouri)
(quotations omitted).
    In order to show a likelihood of success on the merits, the
plaintiffs in each of the cases before us thus have the burden
of showing that the weapons addressed in the pertinent leg-
islation are Arms that ordinary people would keep at home
for purposes of self-defense, not weapons that are exclusively
or predominantly useful in military service, or weapons that
are not possessed for lawful purposes. This search for the cor-
rect meaning of “Arms” for the Second Amendment is con-
sistent with our approach to its companions in the Bill of
Rights. When interpreting the text of a constitutional provi-
sion or a statute, we often resort to contemporaneous diction-
aries or other sources of context to ensure that we are under-
standing the word in the way its drafters intended. In Fourth
Amendment cases, we ask whether the place or item searched
falls within the Amendment’s scope. See, e.g., California v. Ci-
raolo, 

, 213–14 (1986) (aerial view of backyard). For
purposes of the Sixth Amendment, before we apply the Con-
frontation Clause we must ensure that a particular statement
was testimonial. See, e.g., Ohio v. Clark, 

, 243–44,
247 (2015) (child’s responses to questions from a teacher). The
famous Fifth Amendment right against compulsory self-in-
crimination attaches only if the person is in custody, despite
no mention of custody in the “plain text” of the Amendment.
See, e.g., New York v. Quarles, 

 (1984).
   We find substantial support for the proposition that the
Arms protected by the Second Amendment do not include
weapons that may be reserved for military use. We already
have pointed to language in the Supreme Court’s opinions to
Nos. 23-1353 et al.                                                        29

this effect. 6 The dissent, relying heavily on Staples v. United
States, 

 (1994), contends that the Court has already
decided that the AR-15 is in common use, and thus that the
weapon is presumptively immune from regulation. See post at
67. We see no such holding in Staples. That case had nothing
to do with the Second Amendment, which is mentioned no-
where in the opinion. The Court handed down the Staples de-
cision five months before Congress enacted the Federal Assault
Weapons Ban, when as a matter of federal law it was lawful
to own an AR-15. (We assume that this statute is of little rele-
vance to our historical inquiry, given the Supreme Court’s in-
sistence that the relevant time to consult is 1791, or maybe
1868, not the late 20th century.) The status of the AR-15 at the
time Staples was decided provides a ready explanation for
why the Court asserted (with no empirical support) that the
AR-15 is among the weapons that have been “widely accepted
as lawful possessions.” 

. Interestingly, the Sta-
ples Court contrasted the AR-15s with grenades, the posses-
sion of which it said “is not an innocent act.” 

 (quo-
tation omitted). It said the same about “machineguns, sawed-
off shotguns, and artillery pieces.” 

. Overall, we see


    6 We note, too, that this court was not the first to observe the line that

Heller recognized, and which was applied to the states in McDonald. For
example, over a decade ago, and three years before Friedman, one scholar
of the Second Amendment wrote that “Heller and McDonald … focused on
the right of a law-abiding person to have a handgun in his or her home for
self-protection,” but “[n]either case foreclosed reasonable gun regula-
tions,” including “bans on military weapons wholly unnecessary for ordi-
nary self-defense,” “limits on the size of gun clips,” and “registration and
permit requirements.” See Akhil Reed Amar, Gun Control After Newtown
(Dec. 26, 2012), reprinted in THE CONSTITUTION TODAY: TIMELESS LESSONS
FOR THE ISSUES OF OUR ERA 230, 231 (2016).
30                                            Nos. 23-1353 et al.

nothing in Staples that decides whether the Second Amend-
ment protects AR-15s, though we do find much in the opinion
that reinforces the line we discern from Heller, and which is
confirmed by history.
    When we compare the AR-15s and other semiautomatic
weapons covered by the Act and its counterparts, we come to
the same conclusion. Indeed, we asked the plaintiffs at oral
argument to explain what distinguishes AR-15s from M16s,
the military’s counterpart that is capable of both fully auto-
matic operation and semiautomatic operation. The question
is important precisely because Heller itself stated that M16s
are not among the Arms covered by the Second Amendment;
they are instead a military weapon. See 

.
    The plaintiffs’ responses to our question were unconvinc-
ing. They argued, for instance, that civilians do not regard
machineguns as useful for self-defense, but that is because
they cannot purchase machineguns. It is not too much of a
stretch to think that some people might like the fully auto-
matic feature of a machinegun, if they were hoping to defend
their families, their property, and themselves from invaders.
The plaintiffs also noted that machineguns are more expen-
sive than semiautomatic weapons, but we cannot believe that
an item’s entitlement to constitutional protection depends on
its price. Finally, with a nod to the “lawful use” criterion, the
plaintiffs said that when machineguns were available to civil-
ians (early in the 20th century), they were primarily used by
criminals. But this tells us nothing about how use of those
Nos. 23-1353 et al.                                                      31

guns would have evolved, had they remained legal and read-
ily available. 7
    Coming directly to the question whether the weapons and
feeding devices covered by the challenged legislation enjoy
Second Amendment protection, at the first step of the Bruen
analysis, we conclude that the answer is no. We come to this
conclusion because these assault weapons and high-capacity
magazines are much more like machineguns and military-
grade weaponry than they are like the many different types
of firearms that are used for individual self-defense (or so the
legislature was entitled to conclude). 8 Indeed, the AR-15 is al-
most the same gun as the M16 machinegun. The only mean-
ingful distinction, as we already have noted, is that the AR-15
has only semiautomatic capability (unless the user takes ad-
vantage of some simple modifications that essentially make it
fully automatic), while the M16 operates both ways. Both




    7 It appears that there is a large and growing demand for guns in gen-

eral. Since 1986, the number of guns manufactured each year has almost
quadrupled, from around 3 million in 1986 to almost 11 million in 2013.
See Scott Horsley, Guns in America, by the Numbers, NPR (Jan. 5, 2016),
https://www.npr.org/2016/01/05/462017461/guns-in-america-by-the-
numbers. There is no reason to think that machineguns would not have
followed the same pattern, had they been lawful in civilian hands.
    8 Obviously, many weapons are “dual use”: private parties have a

constitutionally protected right to “keep and bear” them and the military
provides them to its forces. In this sense, there is a thumb on the scale in
favor of Second Amendment protection. When we refer to “military”
weapons here, we mean weapons that may be essentially reserved to the
military.
32                                                    Nos. 23-1353 et al.

weapons share the same core design, and both rely on the
same patented operating system. 9
    The similarity between the AR-15 and the M16 only in-
creases when we take into account how easy it is to modify
the AR-15 by adding a “bump stock” (as the shooter in the
2017 Las Vegas event had done) or auto-sear to it, thereby
making it, in essence, a fully automatic weapon. In a decision
addressing a ban on bump stocks enacted by the Maryland
legislature, another federal court found that bump-stock de-
vices enable “rates of fire between 400 to 800 rounds per mi-
nute.” Maryland Shall Issue v. Hogan, 

(D. Md. Nov. 16, 2018) (quotation omitted). To the same effect,
the Fourth Circuit noted that “[t]he difference between the
fully automatic and semiautomatic versions of [the AR-15 and
AK-47] is slight. That is, the automatic firing of all the ammu-
nition in a large-capacity thirty-round magazine takes about
two seconds, whereas a semiautomatic rifle can empty the
same magazine in as little as five seconds.” Kolbe, 

. The District of Columbia Circuit also noted that “semi-
automatics … fire almost as rapidly as automatics.” Heller v.
District of Columbia, 

 (D.C. Cir. 2011), on re-
mand from Heller, 

; see also ATF Ruling 2006-2, at
2 (Dec. 13, 2006) (discussing a device (apparently the “Akins
Accelerator,” an early bump-stock device) that “is advertised
to fire approximately 650 rounds per minute”).



     9 See ARMALITE, INC., Technical Note 54: Direct Impingement Versus Pis-

ton Drive (July 3, 2010), available at https://wayback.archive-
it.org/all/20120905024032/http://www.armalite.com/im-
ages/Tech%20Notes%5CTech%20Note%2054,%20Gas%20vs%20Op%20R
od%20Drive,%20020815.pdf.
Nos. 23-1353 et al.                                            33

    There are a few other differences between the AR-15 and
the M16, but none that is relevant. The 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—unless,
as we have just noted, it is modified with, for example, a
bump stock or a “binary” trigger, which can double the rate
at which semiautomatic weapons can be fired. Both models
use the same ammunition, deliver the same kinetic energy
(1220–1350 foot-pounds), the same muzzle velocity (2800–
3100 feet per second), and the same effective range (602–875
yards). And these comments apply with equal force to the
high-capacity handguns that are restricted by these laws. The
latter are almost indistinguishable from the 17- or 21-round
M17 and M18 pistols that are standard-issue in the military.
    But what about the possibility that the AR-15 (and its
many cousins covered by the Act) as sold is an Arm, even
though simple modifications can transform it into a military
weapon? On the one hand, this might support an argument
against the Act, which focuses initially on the product as sold.
On the other hand, there is a serious question whether the leg-
islature sought to prevent users from deconstructing weapons
into (or assembling weapons from) their constituent parts in
order to evade the core regulation. If the AR-15 by itself is not
a machinegun because it fires “only” at the rate of 300 rounds
per minute, and the auto-sear is also not a machinegun be-
cause it is just a component that holds a hammer in the cocked
position, that would be a road map for assembling ma-
chineguns and avoiding legitimate regulations of their private
use and carry. A question of this nature is raised in VanDerStok
v. Garland, No. 4:22-cv-00691-O, 

 (N.D. Tex.
June 30, 2023), appeal docketed, No. 23-10718, 

(5th Cir. July 24, 2023), and stay pending appeal granted sub nom.
34                                            Nos. 23-1353 et al.

Garland v. Vanderstok, No. 23A82, 

 (U.S. Aug.
8, 2023), where the Supreme Court has issued a stay of a dis-
trict court’s order vacating a federal “ghost gun” regulation,

 (Apr. 26, 2022). See also Garland v. Black-
hawk Mfg. Grp., Inc., No. 23A302, 

 (U.S. Oct.
16, 2023) (vacating a second injunction limited to the parties).
    Neither the parties nor the evidence before us addressed
these points, but the district courts may explore them as the
cases move forward. Better data on firing rates might change
the analysis of whether the AR-15 and comparable weapons
fall on the military or civilian side of the line. We note in this
connection that it is one thing to say that the AR-15 is capable
of firing at a rate of 300 rounds per minute and the compara-
ble rate for the M16 is 700 rounds per minute, but quite an-
other to address actual firing capacity, which accounts for the
need to change magazines. No one here has suggested that
the M16 comes with a 700-round magazine, or for that matter
that the AR-15 comes with a 300-round magazine. Either one
must be reloaded multiple times to fire so many rounds. Fac-
toring in the reloading time, the record may show that the two
weapons differ more—or less—than it appears here.
    Turning now to large-capacity magazines, we conclude
that they also can lawfully be reserved for military use. Recall
that these are defined by the Act as feeding devices that have
in excess of 10 rounds for a rifle and 15 rounds for a handgun.
Anyone who wants greater firepower is free under these laws
to purchase several magazines of the permitted size. Thus, the
person who might have preferred buying a magazine that
loads 30 rounds can buy three 10-round magazines instead.
    Based on the record before us, we are not persuaded that
the AR-15 is materially different from the M16. Heller informs
Nos. 23-1353 et al.                                           35

us that the latter weapon is not protected by the Second
Amendment, and therefore may be regulated or banned. Be-
cause it is indistinguishable from that machinegun, the AR-15
may be treated in the same manner without offending the Sec-
ond Amendment.
    We conclude this portion of the opinion by stressing again
that this is just a preliminary look at the subject. That assess-
ment persuades us, as it did Judges Kendall and Jenkins, that
the plaintiffs have not shown a strong likelihood of success on
the merits. But, as we previously have recognized, Second
Amendment challenges to gun regulations often require more
evidence than is presented in the early phases of litigation. See
Atkinson v. Garland, 

, 1023–25 (7th Cir. 2023) (va-
cating the district court’s order dismissing a Second Amend-
ment challenge to a federal statute and remanding with a list
of specific questions to consider as the case proceeded). There
thus will be more to come, and we do not rule out the possi-
bility 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 record reveals.
                      B. Historical Tradition
    Although we are satisfied that these appeals can be re-
solved at the first step of the Bruen framework—are the weap-
ons among the Arms protected by the Second Amendment—
for the sake of completeness we now turn to the question
whether, if the weapons covered by the statutes before us
ought to be considered bearable “Arms,” the laws nonetheless
pass muster under Bruen’s second step. In short, are these
laws consistent with the history and tradition of firearms reg-
ulation? Here, too, at the preliminary injunction stage, we
36                                           Nos. 23-1353 et al.

conclude that the plaintiffs have not shown the necessary like-
lihood of success on the merits.
    In discussing whether these assault weapons and large-ca-
pacity magazines are Arms protected by the Second Amend-
ment, we have (as instructed by Bruen) confined ourselves to
textual considerations. There is another aspect of the Bruen
framework, which is whether the regulated weapons are “in
common use.” There is no consensus on whether the com-
mon-use issue belongs at Bruen step one or Bruen step two.
The plaintiffs argue that it belongs at the second step. We will
assume (without deciding the question) that this is a step two
inquiry, where the state bears the burden of proof. Even with
that leeway, we do not find this factor to be very helpful.
    In this respect, we find the analysis in Friedman to be par-
ticularly useful, and unlike the district courts, we do not be-
lieve that the relevant portion was undermined by Bruen. We
recognized in Friedman that “common use” is a slippery con-
cept. Suppose, for example, a new type of handgun is intro-
duced to the market on January 1, 2024. As of that day, zero
guns of that type have been sold. Yet if its characteristics are
analogous to those of the many other types of handguns avail-
able for consumers, no one would say that this new handgun
was not within the class of Arms protected by the Second
Amendment. At the other end of the spectrum, consider the
actual case of machineguns, which for a time were available
for civilian purchase, but which were eventually withdrawn
from that market. However popular machineguns might have
been, either in organized crime circles or more generally, be-
cause their characteristics were military in nature, the deci-
sion to reserve them to military use was within the power of
the legislature.
Nos. 23-1353 et al.                                          37

    The dissent repeatedly makes the point that the assault
weapons covered by the challenged legislation are obviously
in common use, because there are so many in private hands.
Indeed, the dissent’s argument boils down to two proposi-
tions: first, it contends that the fact that many people own as-
sault weapons insulates them from regulation; and second, it
makes the surprising assertion that assault weapons are not
particularly dangerous. The latter proposition finds no empir-
ical support in the record, and the former, as we will explain,
does not carry the day.
    The plaintiffs present basically the same argument. One
brief asserts that at least 20 million AR-15s and similar rifles
are owned by some 16 million citizens (though they do not
specify how many of these owners would fall within the large
carveout created by the grandfather and the trained profes-
sional exceptions to the Act). The plaintiffs also assert that at
least 150 million magazines with a capacity greater than 10
rounds have been bought for private use. (The state criticizes
these numbers for being based, it says, on “an unpublished,
non-peer-reviewed paper recounting an online survey that
does not disclose its funding or measurement tools.” We have
no need for present purposes to resolve that dispute.) Cook
County offers a different perspective, noting that of all the
firearms in the country, only 5.3% are assault weapons, and
that percentage includes those held by law-enforcement agen-
cies. One is reminded of Mark Twain’s apocryphal remark,
“There are three kinds of lies: Lies, Damned Lies, and Statis-
tics.”
    For the reasons set forth in more detail in Friedman, we de-
cline to base our assessment of the constitutionality of these
laws on numbers alone. Such an analysis would have
38                                                    Nos. 23-1353 et al.

anomalous consequences. The problem with this approach
can be seen in the case of the AR-15. When, in 1994, the Fed-
eral Assault Weapons Ban made civilian possession of AR-15s
(among other assault weapons) unlawful, see Pub. L. No. 103-
322, § 110102, 

, 1996, few civilians owned AR-
15s. But in 2004, after the legislation was allowed to expire
pursuant to its sunset provision, 

 § 110105(2), 108 Stat. at
2000, these weapons began to occupy a more significant share
of the market. Indeed, most of the AR-15s now in use were
manufactured in the past two decades. 10 Thus, if we looked to
numbers alone, the federal ban would have been constitu-
tional before 2004, but unconstitutional thereafter. This con-
clusion is essential to the plaintiffs’ position, yet it lacks both
textual and historical provenance.
    As this example illustrates, the idea of “common use” can-
not be severed from the historical scope of the common-law
right that the Second Amendment was designed to protect
against encroachment. In other words, the relevant question
is what are the modern analogues to the weapons people used
for individual self-defense in 1791, and perhaps as late as
1868. This would exclude the weapons used exclusively by
the military—and every Framer of the Second Amendment
was well aware by 1791 that the King of England had an im-
pressive standing army, and that such weapons existed. The
weapons used for self-defense are the ones that Heller,
McDonald, Caetano, and Bruen had in mind—not a militaristic



     10 See Aaron O’Neill, Annual Share of AR-15 Assault Rifles in the Total

Number of Firearms Manufactured in the United States from 1990 to 2020,
STATISTA (June 2, 2023), https://www.statista.com/statistics/1388010/share-
ar-15-united-states-firearm-production-historical/.
Nos. 23-1353 et al.                                            39

weapon such as the AR-15, which is capable of inflicting the
grisly damage described in some of the briefs.
    Bruen recognized that even Arms (i.e., non-militaristic
weapons) may be regulated, as long as the regulation is “part
of an enduring American tradition of state regulation.” 142 S.
Ct. at 2155. A regulation is a part of this tradition if one can
provide answers to two questions: (1) how, and (2) why, does
a given regulation “burden a law-abiding citizen’s right to
armed self-defense”? Id. at 2133. With respect to the “how”
question, judges are instructed to consider “whether modern
and historical regulations impose a comparable burden” on
that right. Id. For all its disclaiming of balancing approaches,
Bruen appears to call for just that: a broader restriction bur-
dens the Second Amendment right more, and thus requires a
closer analogical fit between the modern regulation and tra-
ditional ones; a narrower restriction with less impact on the
constitutional right might survive with a looser fit. It is at this
stage that many courts, as well as the state parties here, point
to the long-standing tradition of regulating the especially
dangerous weapons of the time, whether they were firearms,
explosives, Bowie knives, or other like devices. (The regula-
tions we list below are representative of this tradition.) The
dissent cannot deny that regulation existed; it relies only on
the fact that the particulars of those regulations varied from
place to place, and that some were more absolute than others.
But the same is true in our case. The laws before us have one
huge carve-out: people who presently own the listed firearms
or ammunition are entitled to keep them, subject only to a reg-
istration requirement that is no more onerous than many
found in history. In addition, as we noted at the outset, the
laws do not purport to regulate many other special uses. This
40                                            Nos. 23-1353 et al.

is enough, in our view, to satisfy the “how” question Bruen
identified.
    The “why” question is another one that at first blush
seems hard to distinguish from the discredited means/end
analysis. But we will do our best. Bruen makes clear that the
question whether a burden is “comparably justified” cannot
be answered by pointing to the gravity of the harms the legis-
lation was designed to avert and the appropriateness of the
mechanism they adopt. See id. at 2133, 2129. The dissent
chooses to take a purposive approach to this question: what
were the reasons motivating the historical regulations, and do
they map well onto the reasons behind the modern law? We
confess to some skepticism about any test that requires the
court to divine legislative purpose from anything but the
words that wound up in the statute. Legislator A may have
had one goal; Legislator B may have had another; and Legis-
lator C might have agreed to vote for one bill in exchange for
a reciprocal vote for Legislator D’s pet project later. That is
why, as the author of Heller reminded us, “The text is the law,
and it is the text that must be observed.” ANTONIN SCALIA, A
MATTER OF INTERPRETATION 22 (1997).
   The best one can say is that if the text of the legislation
evinces its purpose (perhaps in an introductory Statement of
Purpose, which many bills contain, or in some other prefatory
provision), that is a valid source to consult in answering the
“why” question. When we consult the text of the Act, we find
the best indication of its purpose in its name: “Protect Illinois
Communities Act.” See Pub. Act. 102-1116, at § 1 (2023). Cf.
Johnson v. Robison, 

 (1974) (noting that the
name of a statute can emphasize its purpose). Historical reg-
ulations show that at least since the Founding there has been
Nos. 23-1353 et al.                                                         41

an unbroken tradition of regulating weapons to advance sim-
ilar purposes.
    Once again, the dissent cannot dispute the existence of this
enduring American tradition. It tries to escape it, asserting
that “stop[ping] a mass casualty event,” or perhaps “stopping
escalating gun violence,” is the purpose of the statute, post at
71, 74, yet it points to nothing in the Act that supports either
of these specific characterizations. To be sure, the dissent
notes that the bill enacted by the City of Naperville recites a
few of the many mass shootings that have occurred during
the last decade. See post at 71 n.13. 11 But the bill also expressly
states that the purpose of the ordinance is to protect public
health, safety, and welfare. See City of Naperville, Ill., Ordi-
nance No. 22-099, at 4 (Aug. 16, 2022). The mass-shooting de-
tails appear to be nothing more than particular examples il-
lustrating that broader purpose. The state’s attorney also in-
forms us that the legislation was enacted after the Highland
Park July 4 massacre. But we have not rested our opinion on
this point, because in our view it comes too close to the
means/end scrutiny that Bruen rejected. In any event we do
not think it is appropriate to rely on extratextual considera-
tions to answer the “why” question. The issue, whether we
separate out “how” and “why” or we consider them a unified
test, is whether the tools the legislature used were limited to
those that the Second Amendment left for it, after (as the
Court said in Heller, 

, and Bruen, 142 S. Ct. at
2133 n.7) the Second Amendment itself performed the

    11 Indeed, the dissent relies solely on the municipal bill’s recitations as

proof of the state statute’s purpose. It is quite the puzzle to try to square
this interpretive method with the dissent’s lengthy criticism of our brief
invocation of the name of the Act. See post at 63-65.
42                                                   Nos. 23-1353 et al.

necessary means/end balancing. As we have explained, we
think that the legislatures involved here did stay within those
boundaries.
    Harking back to our examination of covered Arms, we
find the distinction between military and civilian weaponry
to be useful for Bruen’s second step, too. Both the states and
the federal government have long contemplated that the mil-
itary and law enforcement may have access to especially dan-
gerous weapons, and that civilian ownership of those weap-
ons may be restricted. 12 Many other weapons remain that are
more universally available. That is enough to assure us that
we are not creating some unbounded “military veto” over the
types of Arms that can be regulated. History and tradition
leave no doubt that certain weaponry is for the state only:
weapons such as the grenades, the machineguns, the artillery
pieces, and the like mentioned in Staples. See 511 U.S. at 611–
12. (And recall that the laws before us carve out not only the
military, but police and security forces too, from their cover-
age.) And, as we now show, the distinction between the two
uses is one well rooted in our history.
     The following examples suffice to make the point:



     12 We realize that all guns are dangerous when used as intended: a

gunshot wound may be fatal or life-threatening. The Centers for Disease
Control and Prevention estimate that 48,830 people died as a result of a
firearm in 2021. See CENTERS FOR DISEASE CONTROL AND PREVENTION, Na-
tional Center for Health Statistics: All Injuries (Sept. 13, 2023),
https://www.cdc.gov/nchs/fastats/injury.htm. But the record indicates
that there are important differences in the lethality of the military-grade
weapons, as compared with guns that are commonly owned and used for
self-defense and other lawful purposes.
Nos. 23-1353 et al.                                          43

       •   In 1746, Boston outlawed the discharging of
           any cannon, gun, or pistol within city limits,
           but it explained that soldiers were still per-
           mitted to discharge weaponry on their train-
           ing days. See Chapter 11—An Act to Prevent
           the Firing of Guns Charged with Shot[t] or
           Ball in the Town of Boston, §§ 1–3, in 3 THE
           ACTS AND RESOLVES OF THE PROVINCE OF THE
           MASSACHUSETTS BAY 1742-1756, at 309 (1878).
       •   Other cities, such as Cleveland, Ohio, imple-
           mented similar ordinances throughout the
           19th century, again exempting military com-
           panies during drills. See Chapter 33—Fire
           Arms, §§ 417–423, in ORDINANCES OF THE
           CITY OF CLEVELAND 136–37 (H.L. Vail & L.M.
           Snyder, eds., 1890).
       •   There are dozens of examples of Bowie knife
           regulations, forbidding or limiting the use of
           these dangerous weapons. Several of those
           featured military exceptions. In 1884, for ex-
           ample, Arkansas outlawed the sale of all
           dirks, Bowie knives, cane-swords, metal
           knuckles, and pistols, except as for use in the
           army or navy of the United States. See Chap-
           ter 53—Carrying Weapons, §§ 1907–1909, in
           A DIGEST OF THE STATUTES OF ARKANSAS 490
           (W.W. Mansfield, ed., 1884).
       •   Several city ordinances in the late 1800s fol-
           lowed suit, restricting the carry of a wide ar-
           ray of dangerous and concealable weapons
           (slingshots, metal knuckles, Bowie knives,
44                                           Nos. 23-1353 et al.

         daggers, pistols, and clubs), but exempting
         “peace officers” and “conservator[s] of the
         peace.” See Chapter 6—Offenses Against the
         Peace of the City, § 182, in THE REVISED
         ORDINANCES OF PROVO CITY 106–07 (1877);
         Chapter 534—Ordinances of Baltimore,
         § 742A, in THE BALTIMORE CITY CODE 297–98
         (John Prentiss Poe, ed., 1893).
     •   The federal government continued this tra-
         dition when it began passing gun control
         laws. The National Firearms Act of 1934 im-
         posed taxation and registration require-
         ments on all guns, but it exempted transfers
         to the U.S. government, states, territories,
         political subdivisions, and peace officers.
         See 

 1-12, § 13, 

, 1236-40, 1240 (1934).
     •   Federal restrictions expanded in 1968, when
         sale and delivery of destructive devices (de-
         fined as an “explosive, incendiary, or poison
         gas bomb, grenade, mine, rocket, missile, or
         similar device”) and machineguns were se-
         verely restricted. See Omnibus Crime Con-
         trol and Safe Streets Act of 1968, 

(a)(4), § 922(b), 

, 227,
         230 (1968). Once again, these provisions did
         not apply to items sold to the United States
         or to any individual state. 

 § 925(a), 82
         Stat. at 233.
     •   Machineguns were banned by the Firearm
         Owners’ Protection Act of 1986. Since then,
Nos. 23-1353 et al.                                            45

          civilian ownership has been capped at pre-
          1986 levels and only military and law en-
          forcement have access to these weapons. See
          

(9), 

, 453
          (1986).
   In short, there is a long tradition, unchanged from the time
when the Second Amendment was added to the Constitution,
supporting a distinction between weapons and accessories
designed for military or law-enforcement use, and weapons
designed for personal use. The legislation now before us re-
spects and relies on that distinction.
               IV.    Concluding Observations
    We conclude with a few remarks about several additional
issues in some of these cases that do not require immediate
attention, and a reminder about the limits on our ruling.
    First, we briefly comment on Herrera’s challenge to the
constitutionality of the registration requirement that imple-
ments the grandfather exemption. He regards it as a burden
on his Second Amendment rights, and he worries that it may
in the future lead to confiscatory acts on the part of the state.
If we are correct in our prediction that the state will prevail in
its defense of the Act against the Second Amendment argu-
ments, then the registration requirement will be valid as long
as it can withstand rational basis review. At this juncture, we
see nothing particularly onerous about it, though as with eve-
rything we have said, this is a preliminary assessment. Her-
rera has until the end of 2023 to file the necessary forms, and
if he does so, he may retain all of the covered weapons he al-
ready owns; the Act will prohibit only his acquisition of addi-
tional assault weapons or high-capacity feeding devices. For
46                                                  Nos. 23-1353 et al.

its own reasons, the dissent agrees with us that the registra-
tion requirement should not be enjoined. See post at 76.
    Second, in this court none of the parties has developed any
coherent argument that would distinguish restrictions on
possession, on the one hand, from restrictions on sale or man-
ufacture, on the other. One of the parties in Bevis is a gun
store, but the implications of that have yet to be addressed.
We thus have no comment on it.
   Finally, we have no need to decide whether an alleged Sec-
ond Amendment violation gives rise to a presumption of ir-
reparable harm, and if so, whether any such presumption is
rebuttable or ironclad. Given our decision that the plaintiffs
have not shown that they have a strong likelihood of success
on the merits, we think it best to save this point for another
day. We also have no comment on the other two parts of the
Winter inquiry: where the balance of equities lies, and what
the public interest dictates. 13
    We close with an important reminder. Nothing that we
have said here indicates that any state or municipality must
enact restrictions on the ownership of assault weapons or
high-capacity magazines. Unless preemptive federal legisla-
tion requires otherwise, this is an issue for the political pro-
cess in each jurisdiction. The people of some states may find

     13 The governmental parties devoted considerable attention in their

briefs to the horrors of the mass shootings that have occurred with dis-
tressing regularity throughout the country. Illinois reports that the mass
shooting in the town of Highland Park on July 4, 2022, in which seven
people were killed and another 48 were injured, inspired the Act. We have
not relied on this point, however, because, as we have mentioned, it ap-
pears to depend on the type of means/end analysis that Bruen disap-
proved.
Nos. 23-1353 et al.                                           47

the arguments in favor of a lack of restrictions to be persua-
sive; the people of other states may prefer tighter restrictions.
As long as those restrictions do not infringe on the constitu-
tionally protected right to keep and bear the Arms covered by
the Second Amendment, either choice is permissible. In the
cases now before us, however, the plaintiffs have not shown a
likelihood of success on the merits, based on the fact that mil-
itary weapons lie outside the class of Arms to which the indi-
vidual right applies.
    In Nos. 23-1353 and 23-1793, we AFFIRM the district
courts’ orders denying preliminary injunctive relief. In Nos.
23-1825, 23-1826, 23-1827, and 23-1828, we VACATE the dis-
trict court’s order granting preliminary injunctive relief. We
also confirm that the stay we issued in these appeals will re-
main in effect until our mandate issues.
                                                SO ORDERED.
48                                                                                                 Nos. 23-1353 et al.
                                                 APPENDIX


     Jurisdiction         Actions Prohibited                 Firearms Covered                 Magazines Covered

                                                       •   Semiautomatic rifles that
                                                           have one or more assault
                    •   Manufacture                        weapon-like modifications
                    •   Delivery                       •   Semiautomatic pistols that
                    •   Sale                               have one or more assault
                                                           weapon-like modifications      •   Rifles—over 10 rounds
                    •   Purchase
       Illinois                                        •   Semiautomatic shotguns        •   Handguns—over 15
                    •   Possession (pre-existing           with a revolving cylinder         rounds
                        possession and private             or that have one or more
                        post-Act use are permissi-         assault weapon-like modi-
                        ble if registration require-       fications
                        ments are met)
                                                       •   Various assault weapons
                                                           listed by name

                                                       •   Semiautomatic rifles that
                                                           can accept more than 10
                                                           rounds or that have one or
                                                           more assault weapon-like
                                                           modifications
     Naperville     •   Commercial sale                                                              N/A
                                                       •   Devices that accelerate the
                                                           rate of fire of a semiauto-
                                                           matic rifle
                                                       •   Various assault weapons
                                                           listed by name
Nos. 23-1353 et al.                                                                                               49




                                                      •   Semiautomatic rifles that
                                                          have one or more assault
                                                          weapon-like modifications
                       •   Manufacture
                                                      •   Semiautomatic pistols that
                       •   Sale
                                                          have one or more assault
                       •   Offer or display for sale       weapon-like modifications

                       •   Give                       •   Semiautomatic shotguns
                                                          with a revolving cylinder
         Cook County   •   Lend                                                        •   Magazines over 10 rounds
                                                          or that have or one more
                       •   Transfer of ownership          assault rifle-like modifica-
                                                          tions
                       •   Acquire
                                                      •   Combination kits from
                       •   Carry
                                                          which an assault weapon
                       •   Possession                     can be assembled
                                                      •   Various assault weapons
                                                          listed by name
50                                                                                    Nos. 23-1353 et al.



                                           •   Semiautomatic rifles with
                                               the ability to accept a de-
                                               tachable magazine and that
                                               have one or more assault
                                               weapon-like modifications
                                           •   Semiautomatic rifles that
                                               have a fixed magazine with
                                               the capacity to accept more
               •   Import                      than 10 rounds
               •   Sale                    •   Devices that accelerate the
                                               rate of fire of a semiauto-    •   Handguns—over 15
     Chicago   •   Manufacture
                                               matic rifle                        rounds
               •   Transfer of ownership
                                           •   Shotguns that have one or
               •   Possession                  more assault weapon-like
                                               modifications
                                           •   Semiautomatic handguns
                                               that have one or more as-
                                               sault weapon-like modifi-
                                               cations
                                           •   Various assault weapons
                                               listed by name
Nos. 23-1353 et al.                                             51

    BRENNAN, Circuit Judge, dissenting. The Second Amend-
ment “right of the people to keep and bear Arms” is not a sec-
ond-class right. Yet the State of Illinois and several Illinois mu-
nicipalities have categorically banned law-abiding citizens
from keeping and bearing a sweeping range of firearms and
magazines. In a remarkable conclusion, the majority opinion
decides that these firearms are not “Arms” under the Second
Amendment. Because the banned firearms and magazines
warrant constitutional protection, and the government parties
have failed to meet their burden to show that their bans are
part of the history and tradition of firearms regulation, pre-
liminary injunctions are justified against enforcement of the
challenged laws. I respectfully dissent.
                                I
   The Protect Illinois Communities Act, Pub. Act 102-1116,
challenged in each case before us, dramatically redefines the
legality of firearms and magazines in Illinois. It goes far be-
yond the prohibition of “assault rifles.” The Act eliminates the
ownership, possession, and use for self-defense of many of
the most commonly-owned semiautomatic handguns, shot-
guns, rifles, and magazines. Exceptions to the Act are narrow.
    Specifically, the Act covers firearms, magazines, and an
endorsement process for registration. The Act bans the man-
ufacture, delivery, sale, import, and purchase of a vast array
of weapons, 720 ILL. COMP. STAT. §§ 5/24-1(a)(16), 75/24-1.9(a),
5/24-1.10(a)–(b), prohibiting them by their features, by their
functions, and by name. The Act bans semiautomatic rifles
with detachable magazines and one additional qualifying at-
tachment, such as a pistol grip or a flash suppressor. Id. § 5/24-
1.9(a)(1)(A). “[A]ll AR type[]” rifles are banned, including 43
named variants, such as the AR-15. The Act further prohibits
52                                              Nos. 23-1353 et al.

“copies, duplicates, variants, or altered facsimiles with the ca-
pability of any such weapon.” Id. § 5/24-1.9(a)(1)(J)(ii). It also
bans almost 100 more rifles by name. Id. § 5/24-1.9(a)(1)(J).
    The Act restricts various other firearms as well. For exam-
ple, a law-abiding citizen in Illinois can no longer purchase
semiautomatic pistols that have “a fixed magazine with the
capacity to accept more than 15 rounds,” regardless of any
attachments. Id. § 5/24-1.9(a)(1)(D). The same goes for a semi-
automatic shotgun with a fixed magazine holding more than
five shells. Id. § 5/24-1.9(a)(1)(F)(v). The list of restricted weap-
ons includes nearly all detachable magazines holding more
than 10 rounds of ammunition for long guns and 15 rounds
of ammunition for handguns. Id. § 5/24-1.10(a)(1)–(2). Many
handguns, the “quintessential self-defense weapon” for the
American people, District of Columbia v. Heller, 

 (2008), come standard with magazines carrying more than
15 rounds. As with semiautomatic rifles, after banning pistols
by their features, the Act bans “[a]ll AR type[] pistols” and
approximately 40 semiautomatic pistols by name. 720 ILL.
COMP. STAT. § 5/24-1.9(a)(1)(K).
   Three municipal laws are also challenged, which are as or
more restrictive than the Act. The City of Naperville ordi-
nance is similar to the Act in most respects; both are
challenged in Bevis. The Cook County and City of Chicago or-
dinances, challenged along with the Act in Herrera, are even
broader. Cook County bans possession of “assault weapons,”
COOK COUNTY, ILL. CODE § 54-211 and § 54-212, which in-
cludes semiautomatic pistols with the capacity to accept a de-
tachable magazine and contain a qualifying attachment (such
Nos. 23-1353 et al.                                                        53

as a muzzle brake). The City of Chicago ordinance is similar.
See CHI. MUN. CODE §§ 8-20-010, 8-20-075, 8-20-085. 1
                                     II
   The parties dispute whether the state, county, and city
bans respect the constitutional right to keep and bear arms. In
New York State Rifle & Pistol Ass’n, Inc. v. Bruen, 

(2022), the Supreme Court set forth the framework for ad-
dressing those disputes. Rejecting means-end scrutiny, the
Court held: “When the Second Amendment’s plain text

    1 The majority opinion uses the phrase “assault weapon” to simplify

the covered arms. The appendix to the majority opinion uses a variety of
terms to summarize the types of arms the four challenged laws categori-
cally ban.
     Still, the description in the appendix of the Act’s ban is underinclusive
in some ways. The Act bans semiautomatic rifles with fixed magazines
over 10 rounds (unless it fires .22 rimfire and is loaded with a tubular
mechanism). ILL. COMP. STAT. § 5/24-1.9(a)(1)(B). The appendix uses the
phrase “[s]emiautomatic pistols that have one or more assault weapon-
like modifications,” most likely a reference to ILL. COMP. STAT. § 5/24-
1.9(a)(1)(C). More precisely, the Act also bans semiautomatic pistols with
fixed magazines over 15 rounds. Id. § 5/24-1.9(a)(1)(D). Not included in
the appendix are bump stocks and binary triggers (a device enabling the
firing of two-rounds per trigger pull), which are both prohibited by the
Act. Id. § 5/24-1(a)(14).
     The Cook County ordinance mirrors the Act’s prohibitions, although
it is stricter than the Act in that it bans semi-automatic handguns with
fixed magazines over 10 rounds (as opposed to 15 rounds under the Act).
COOK COUNTY, ILL. CODE § 54-211(2).
    The City of Chicago ordinance is underinclusive in its description of
the magazines covered. The ordinance prohibits any magazine holding
greater than 15 rounds, encompassing magazines for all types of firearms
(except for attached devices that only accept and operate with .22 rimfire
ammunition), not just handguns. CHI. MUN. CODE § 8-20-010.
54                                            Nos. 23-1353 et al.

covers an individual’s conduct, the Constitution presump-
tively protects that conduct. The government must then jus-
tify its regulation by demonstrating that it is consistent with
the Nation’s historical tradition of firearm regulation.” Bruen,
142 S. Ct. at 2129–30.
    The Second Amendment states in part, “the right of the
people to keep and bear Arms, shall not be infringed.” U.S.
CONST. amend. II. The amendment presents several condi-
tions for plain text coverage, which raise questions including:
      •   Is the regulated population a covered “peo-
          ple?” See, e.g., Range v. Att’y Gen. United
          States, 

, 101–03 (3d Cir. 2023) (en
          banc); United States v. Sitladeen, 

,
          983 (8th Cir. 2023); and
      •   Is the conduct regulated “keep[ing]” or
          “bear[ing]” arms? See, e.g., Heller, 554 U.S. at
          582–92.
We consider another question: Are the instruments regu-
lated “Arms”?
    “Arms” in the Second Amendment is a broad term that
“covers modern instruments that facilitate armed self-de-
fense.” Bruen, 142 S. Ct. at 2132. The term “extends, prima fa-
cie, to all instruments that constitute bearable arms, even
those that were not in existence at the time of the founding.”
Heller, 

. When the plain text of the Second
Amendment covers an individual’s conduct, then the Consti-
tution presumptively protects the conduct. Bruen, 142 S. Ct. at
2129–30. That presumptive protection is of all bearable instru-
ments that facilitate armed self-defense, even those not in ex-
istence at the time of the Founding. Id. at 2132, 2143 (citing
Nos. 23-1353 et al.                                                          55

Caetano v. Massachusetts, 

, 411–412 (2016) (per cu-
riam), and Heller, 

). 2
    As an initial matter, magazines—ammunition feeding de-
vices without which semiautomatic firearms cannot operate
as intended—are “Arms.” Such devices are required as part of
the firing process. This court has recognized that corollaries
to firearms fall within Second Amendment protection. See
Wilson v. Cook County, 

, 1032 (7th Cir. 2019)
(quoting Ezell v. City of Chicago, 

 (7th Cir.
2011)). Further, the Act’s ban on magazines holding more
than ten rounds for rifles and more than fifteen rounds for
handguns effectively bans firearms that come standard with
magazines over the limit.
    As for the broader definition of “Arms,” that term should
be read as “Arms”—not “Arms in common use at the time.”
In Heller, the Supreme Court recognized a “historical tradition
of prohibiting the carrying of ‘dangerous and unusual weap-
ons,’” 

, which may be regulated—a point it re-
peated in Bruen, 142 S. Ct. at 2143.
   The Court “did not say that dangerous and unusual weap-
ons are not arms.” Teter v. Lopez, 

, 950 (9th Cir.
2023) (emphasis in original) (ruling that Hawaii statute ban-
ning butterfly knives violated Second Amendment). To be

    2 When the Supreme Court issued Bruen, it vacated several federal ap-

pellate decisions upholding gun controls laws, remanding them for recon-
sideration. Two of them—Duncan v. Bonta, 

 (9th Cir. 2021),
and Ass’n of N.J. Rifle & Pistol Clubs Inc. v. Att’y Gen. N.J., 

 (3d
Cir. 2020)—concerned magazine limits of 10 rounds, and Bianchi v. Frosh,

 (4th Cir. 2021) (per curiam) (unpublished), upheld Mar-
yland’s “assault weapons” ban.
56                                                       Nos. 23-1353 et al.

sure, this does not mean that the Second Amendment bars
governments from regulating weapons long held improper
for civilian use. This reading of Bruen permits the govern-
ment, for example, to preclude civilian ownership of military
weaponry when the history and tradition of weapons regula-
tion so dictates. As other examples, the government may pro-
hibit sawn-off rifles and shotguns, which properly qualify as
dangerous and unusual firearms as they are not ordinarily
used by law-abiding citizens. See Heller, 

 (“[T]he
Second Amendment does not protect those weapons not typ-
ically possessed by law-abiding citizens for lawful purposes,
such as short-barreled shotguns.”). But that distinction does
not determine whether a weapon is an “Arm.”
    The government parties limit the Second Amendment
right by importing the phrase “in common use” to assess
whether firearms are “Arms.” But their reading improperly
restricts the constitutional right. The banned firearms propel
bullets by explosive force from gunpowder, yet the govern-
ment parties ask us to conclude that these rifles and pistols
are not “Arms.” As one amicus curiae submitted, “in common
use” is a sufficient condition for finding arms protected under
the history and tradition test in Bruen, not a necessary condi-
tion to find them “Arms.” 3 The nature of an object does not
change based on its popularity, but the regulation of that ob-
ject can.
    The government parties also incorrectly attempt to place a
burden on the plaintiffs to show that the plain text of “Arms”
includes the banned firearms. Bruen does not say that. In-
stead, Bruen states that when the Second Amendment’s text


     3 See D.E. 99, Brief for Amici Curiae Idaho, et al., at 6.
Nos. 23-1353 et al.                                              57

covers an individual’s conduct, the Constitution presump-
tively protects it. 142 S. Ct. at 2126, 2129–30. It is undisputed
that the government then bears the burden of proof under
Bruen’s history and tradition framework.
    Whether a firearm is “in common use” is asked as part of
the history and tradition analysis. At least two reasons sup-
port this reading. First, the “in common use” test in Bruen is
drawn from the “historical tradition” of restrictions on “dan-
gerous and unusual weapons.” Id. at 2143. The test is not
drawn from a historical understanding of what an “Arm” is.
Id. at 2132. Second, if a weapon is an “Arm,” it is only prima
facie protected by the Second Amendment. Bruen, 142 S. Ct.
at 2132 (quoting Heller, 

); see Teter, 76 F.4th at
949–50 (placing “in common use” test in history and tradition
test of Bruen).
     The limitation of the Second Amendment right is ad-
dressed in Bruen’s history and tradition test. This requires the
government to identify well-established and representative
historical analogues to show that the modern regulation is
consistent with a historical tradition of firearms regulation.
Bruen, 142 S. Ct. at 2133. In performing this analogical inquiry,
it is critical to fly at the right level of generality. Id. (“[A]na-
logical reasoning under the Second Amendment is neither a
regulatory straightjacket nor a regulatory blank check.”); see
J. HARVIE WILKINSON, COSMIC CONSTITUTIONAL THEORY 44
(2012). Fly too high, and we risk any historical firearms regu-
lation becoming an analogue. Under Bruen, courts must not
“uphold every modern law that remotely resembles a histor-
ical analogue.” Bruen, 142 S. Ct. at 2133. (quoting Drummond
v. Robinson, 

, 226 (3d Cir. 2021)). Fly too low, and
we risk myopia—nitpicking differences because a historical
58                                             Nos. 23-1353 et al.

regulation is not a “dead ringer.” 

 We are looking for “a
well-established and representative historical analogue, not a
historical twin.” 

    Before reviewing the approach to decide whether a regu-
lation is an analogical fit, it helps to address what history and
tradition refer to here. “History” means that analogous laws
must be “longstanding” and from the relevant “timeframe.”
Id. at 2131, 2133 (citing Heller, 

). “History” helps
establish the public meaning of the Constitution as “under-
stood ... when the people adopted” it. 

 (citing Heller, 554
U.S. at 634–35). The Court tells us that only two historical
timeframes are relevant to the public understanding of the
Second Amendment—the adoption of the Second Amend-
ment in 1791 and the ratification of the Fourteenth Amend-
ment in 1868. Id. at 2136. Laws enacted after the “end of the
19th century” must be given little weight. Id. at 2136–37
(cleaned up). “Tradition” means that the comparison must be
to laws with wide acceptance in American society. Id. at 2136.
Laws that enjoyed “widespread” and “unchallenged” sup-
port form part of our tradition. Id. at 2137.
    In Bruen, the Court reaffirmed that “individual self-de-
fense is ‘the central component’ of the Second Amendment
right,” id. at 2133 (citing McDonald, 

 (quoting
Heller, 

)), and expressly identified two ques-
tions to assess the analogical strength of a historical regula-
tion: “how and why the regulations burden a law-abiding cit-
izen’s right to armed self-defense.” Id. at 2133. Put another
way, how does the regulation limit the Second Amendment
right, and why does it do so?
   How. How a historical regulation addressed a particular
problem, or whether it did at all, matters. “[I]f earlier
Nos. 23-1353 et al.                                             59

generations addressed the societal problem, but did so
through materially different means, that also could be evi-
dence that a modern regulation is unconstitutional.” Id. at
2131. Whether a given regulation was ever enforced, and to
what extent, can be relevant here as well. Id. at 2149.
   Courts must also evaluate how historical “regulations bur-
den a law-abiding citizen’s right to armed self-defense.” Id. at
2133. Modern regulations that impose a “comparable burden
on the right of armed self-defense” are more likely to be up-
held. Id.
    In assessing these comparable burdens, we consider the
breadth of the ban and the weapon banned. For the breadth
of the ban, the more expansive the limitation, the greater the
burden on the Second Amendment right, which necessarily
requires a close analogical fit. For the weapon banned, the
burden on the right to keep and bear arms necessarily corre-
lates with whether the prohibited weapon is “in common use
at the time” of regulation. Id. at 2128, 2134, 2143. So, it is nat-
ural that categorical bans of weapons in common use will
require an even stronger analogical fit with historical regula-
tions. See id. at 2143–44 (rejecting the analogical value of
alleged colonial era categorical bans on “dangerous and unu-
sual” weapons because handguns are “unquestionably in
common use today”).
    Why. Why a historical regulation addressed a particular
problem, or whether it did at all, is also key to evaluating its
analogical value. In considering whether a historical regula-
tion is an analogical fit, courts are to address whether the
modern regulation and proposed historical analogue have
comparable justifications for burdening the right to bear
arms. Id. at 2133. If the reasons motivating the historical and
60                                            Nos. 23-1353 et al.

modern regulations differ, there is no analogue. See id. at 2140,
2144. Beyond doubt, this inquiry should not allow a return to
interest balancing. See id. at 2131 (explaining that the Second
Amendment itself “is the very product of an interest balancing
by the people” (quoting Heller, 

)). Rather, the
state’s current rationale for arms regulation only matters in-
sofar as a historical regulation was motivated by similar rea-
sons. If not, the analogy fails. See id. at 2144 (discussing the
context of the colonial New Jersey restrictions, in which land
disputes between planters and the colony’s proprietors
caused planters to carry pistols).
     The government can only defend a regulation by proving
it is consistent with this country’s history and tradition. See
Atkinson v. Garland, 

, 1020–21 (7th Cir. 2023).
Whether that history and tradition allows regulating firearms
in sensitive places, for the mentally ill, and for felons, is cur-
rently under debate. See, e.g., United States v. Rahimi, 

, 460–61 (5th Cir. 2023) (ruling that federal statute prohib-
iting possession of firearm by individual subject to domestic
violence restraining order violates Second Amendment as in-
consistent with historical tradition), cert. granted 

.
    This understanding of the Bruen framework is different
from that of my colleagues. First, the majority opinion
acknowledges Bruen’s “in common use” language but criti-
cizes it as spawning unworkable circularity issues: If the Sec-
ond Amendment protects firearms in common use, then that
right would turn on how quickly a state enacts regulations. If
a firearm is outlawed quickly following its introduction to the
market, then it has no chance of gaining common use and en-
joys only limited or no Second Amendment protection. This
Nos. 23-1353 et al.                                           61

cannot be how the Second Amendment functions, the argu-
ment goes, as the speed of regulation should not bear on an
arm’s constitutionality.
    This circularity concern is far less pressing when the “in
common use” language is properly situated. Because that
consideration plays into the history and tradition analysis—
and not the scope of the Second Amendment’s text—it is not
an “on-off” switch for constitutional protection. Just because
a weapon is not in common use does not mean it falls outside
the text of the Second Amendment; and just because a
weapon is in common use does not necessarily mean a gov-
ernment is barred from regulating it. Proper inquiry requires
full examination of the government’s evidence and historical
analogues, keeping in mind that bans of weapons “in com-
mon use” are constitutionally suspect.
    The Supreme Court certainly was not worried about cir-
cularity. In Bruen, the Court explicitly linked the Second
Amendment analysis to “in common use.” See 142 S. Ct. at
2128 (quoting Heller, 

) (explaining that “the Sec-
ond Amendment protects only the carrying of weapons that
are those ‘in common use at the time,’ as opposed to those that
‘are highly unusual in society at large’”). The Court reasoned
that even if handguns were once “dangerous and unusual,”
such firearms “are unquestionably in common use today”
and therefore receive robust Second Amendment protection.
Id. at 2143. In Caetano, the Court addressed Second Amend-
ment protections for a new electronic weapon. So many were
in circulation (200,000 stun guns, far fewer than the approxi-
mately 25 million AR rifles) that the electronic weapon was
deemed “commonly possessed by law-abiding citizens for
lawful purposes … .” 577 U.S. at 420. We are not free to ignore
62                                                     Nos. 23-1353 et al.

the Court’s instruction as to the role of “in common use” in
the Second Amendment analysis. 4
    Next, my colleagues disagree with my approach to Bruen’s
“why” question, raising the specter of purposivism. The ma-
jority opinion urges respect for the text of a statute alone,
which I share. Indeed, a fair reading of a statute always “re-
quires an ability to comprehend the purpose of the text, which
is a vital part of its context.” See ANTONIN SCALIA & BRYAN A.
GARNER, READING LAW: THE INTERPRETATION OF LEGAL TEXTS
33 (2012); see also John O. McGinnis, The Contextual Textualism
of Justice Alito, 14 HARV. J. L. & PUB. POL’Y PER CURIAM, at 2
(2023) (describing Justice Alito’s use of context in interpreta-
tion). This is certainly a different task than interpreting a stat-
ute by reference to the intent of its drafters, which I agree is
an inappropriate job for judges.
    Still, Bruen requires us to consider the historical context
giving rise to the statute (the “why”). Bruen looks at history
and tradition to determine “the content of the preexisting le-
gal right to bear arms.” Randy E. Barnett & Lawrence B. So-
lum, Originalism After Dobbs, Bruen, and Kennedy, 118 NW.
U. L. REV. 433, 469 (2023). And Bruen’s history and tradition
approach is a different endeavor than statutory interpreta-
tion.
    Often a statute takes center stage for a purpose other than
to discern the scope of its legal rule, even when determining
whether it violates a constitutional right. See Vill. of Arlington


     4 The circularity argument also is not new. See Friedman v. City of High-

land Park, 

 (7th Cir. 2015). As Judge Manion explained in dis-
sent, circularity concerns deal in the hypothetical more than the actual. 

at 416 n.5.
Nos. 23-1353 et al.                                            63

Heights v. Metro. Hous. Dev. Corp., 

 (1977)
(considering whether a discriminatory purpose was a moti-
vating factor in a city’s zoning rules). For example, in Bruen
the Court considered Henry VIII’s “displeasure with hand-
guns” due to his concern that they would “threaten[] English-
men’s proficiency with the longbow,” which led to
Parliament’s passage of handgun restrictions. 142 S. Ct. at
2140. East New Jersey prohibited the concealed carry of
pocket pistols in response to “‘strife and excitement’ between
planters and the Colony’s proprietors ‘respecting titles to the
soil.’” Id. at 2143–44. And Heller discusses the “public-safety
reasons” behind several Colonial-era individual-arms-bear-
ing statutes. Heller, 

.
     When looking to the text in its “why” analysis, the major-
ity opinion relies on the Act’s title, Protect Illinois Communi-
ties Act. Set aside for the moment that “for interpretive pur-
poses,” courts should only rely on titles to “shed light on
some ambiguous word or phrase” in the text. See Trainmen v.
Baltimore & Ohio R. Co., 

 (1947). Titles and sec-
tion headings have a short history in the Anglo-American in-
terpretive tradition—legislatures did not always include the
title while debating the act. See SCALIA & GARNER at 221. If
there is serious doubt as to whether those titles and headings
received a fair shake in the legislative process, relying on
them would make little sense. One influential treatise im-
plores judges to check a state’s constitution for provisions that
vouchsafe interpretive usefulness on a statutory title. See id. at
224.
   As it turns out, the title of the Protect Illinois Communities
Act has little interpretive utility. The Illinois Constitution has
a provision grounding the title in the legislative process, but
64                                              Nos. 23-1353 et al.

there is serious doubt whether the legislature obeyed it here.
The so-called three-readings clause states: “A bill shall be read
by title on three different days in each house.” ILL. CONST. art.
IV, § 8(d). Reading rules exist precisely to ensure “that each
House knows what it is passing and passes what it wants.”
Schwegmann Bros. v. Calvert Distillers Corp., 

(1951) (Jackson, J., concurring) (explaining that the federal
three-readings rule helps draw a line where debate ends and
drafting begins).
     Consider the procedural path of the Act, during which the
Illinois legislature may very well have ignored the three-read-
ings rule. See Caulkins v. Pritzker, No. 129453, 

, at *17 (Ill. Aug. 11, 2023) (White, J., dissenting). A
group of firearms owners challenged the Act in Illinois state
court, arguing it violated several provisions of the Illinois
Constitution. Id. at *1. The three-readings clause is one of
these provisions, and the Supreme Court of Illinois rejected
that claim only because the plaintiffs failed to cross-appeal it,
a jurisdictional error warranting dismissal. Id. This legislation
began in the Illinois House with the title, “an Act concerning
regulation,” and its synopsis described changes to the state’s
insurance code. Id. at *17 (White, J., dissenting). The House
read it three times by this title, then sent it to the Illinois Sen-
ate. Id. The Senate read it twice before the Senate adopted an
amendment that “completely stripped the insurance provi-
sions[,] … replaced them with the ‘Protect Illinois Communi-
ties Act[,]’” and added the new bill’s popular title. Id. The day
the legislation became the “Protect Illinois Communities Act,”
the Senate read it for the first time under the new title and
passed it. Id. The Act was returned to the House the day after
Nos. 23-1353 et al.                                                      65

that and passed without a reading. Id. The Illinois Governor
signed it later that day. Id. 5
    Though the Act’s possible three-readings problem bears
on neither the Second Amendment question nor the Act’s le-
gitimacy, it remains a good reason to be skeptical of the inter-
pretive value of language extrinsic to the operative text. In-
stead, I focus on permissible indicators of meaning.
                                    III
     Turning to this interlocutory appeal, the plaintiffs make a
facial challenge to the Act and ordinances at the preliminary
injunction stage. According to the Supreme Court in Nken v.
Holder, 

 (2009), the two most important con-
siderations at this stage are likelihood of success on the merits
and irreparable harm. For the reasons explained below, plain-
tiffs have satisfied both considerations.
                                    A
    As for likelihood of success on the merits, the firearms and
magazines banned by the Act and ordinances are “Arms” un-
der the plain text of the Second Amendment. These firearms
and magazines are therefore presumptively protected. 6 The
government parties embrace a contrasting, very narrow view
of the scope of the Second Amendment. They would limit this
constitutional right to the facts in Heller and Bruen. Yet, as

    5 The Illinois Supreme Court decided that the Act does not violate cer-

tain provisions of that state’s constitution. Caulkins, 

, at
*4–6. The court also ruled that a challenge based on the federal Second
Amendment had been waived. Id. at *6.
    6 Debates about grenades or rocket launchers are off subject. Some

military weaponry is covered by federal statute, see 18 U.S.C. ch. 44, which
is not challenged here.
66                                                Nos. 23-1353 et al.

examples, the First and Fourth Amendments would surely
not be read in such a cramped manner.
    Under Bruen’s history and tradition test, the government
parties bear the burden to show that the banned arms are not
in common use—or in other words, are not dangerous and
unusual—and to identify historical analogues. As described
above, Bruen reviewed Heller and set forth its test to deter-
mine if regulations satisfied the “how” and “why” test. Bruen,
142 S. Ct. at 2128 (citing Heller, 554 U.S. at 626–34).
    The Act and ordinances here do not fall within a “histori-
cal tradition of prohibiting the carrying of ‘dangerous and un-
usual weapons.’” Heller, 

; see Bruen, 142 S. Ct. at
2143. The banned arms are “in common use,” including for
self-defense, hunting, and sporting pursuits. Each side
chooses its metric—regulators divide the banned guns by the
total number of firearms, and gun owners use gross numbers
of the banned guns and magazines. Under either measure, the
banned weapons and magazines meet the definition of “com-
mon”: “the quality of being public or generally used.” BRYAN
GARNER, GARNER’S DICTIONARY OF LEGAL USAGE 179 (Oxford,
3d ed. 2011). More than 24 million AR rifles are estimated to
be in circulation in this country. 7 Magazines number far more:
in 2020 it was estimated that approximately 160 million pistol
and rifle magazines with a capacity of 11 rounds or more were
in U.S. consumer possession from 1990–2018. 8

     7 Commonly Owned: NSSF Announces over 24 Million MSRs in Circula-

tion,   NAT’L    SHOOTING       SPORTS     FOUND.    (July 20, 2022),
https://www.nssf.org/articles/commonly-owned-nssf-announces-over-
24-million-msrs-in-circulation/ [https://perma.cc/2LX6-UN3B].
     8 Firearm Production in the United States, NAT’L SHOOTING SPORTS

FOUND. 7 (2020), https://www.nssf.org/wp-content/uploads/2020/11/IIR-
Nos. 23-1353 et al.                                            67

    Federal courts have recognized that the AR-15 rifle is com-
mon. In Staples v. United States, 

 (1994), the Su-
preme Court offered comments in dicta stating how common
AR-15s were at that time in this country. That case, which did
not address the Second Amendment, turned on the question
of mens rea, and the Court decided that to convict a person of
possession of an unregistered machinegun, the government
must prove the defendant knew that it would fire automati-
cally. 

. In Staples, the Court contrasted the semiauto-
matic AR-15 with the automatic M16. 

 at 602 n.1, 603.
Acknowledging “a long tradition of widespread lawful gun
ownership by private individuals in this country,” the Court
stated, “[e]ven dangerous items can, in some cases, be so com-
monplace and generally available that we would not consider
them to alert individuals to the likelihood of strict regula-
tion.” 

 at 610–11. Staples contrasted ordinary firearms such
as the AR-15 in that case with “machineguns, sawed-off shot-
guns, and artillery pieces,” stating “guns falling outside those
categories traditionally have been widely accepted as lawful
possessions.” 

.
    Albeit pre-Bruen, two federal appellate courts also con-
cluded that AR platform rifles are common. N.Y. State Rifle &
Pistol Ass’n, Inc. v. Cuomo, 

 (2d Cir. 2015)
(“Even accepting the most conservative estimates cited by the
parties and by amici, the assault weapons and large-capacity
magazines at issue are ‘in common use’ as that term was used
in Heller.”); Heller v. District of Columbia, 

(D.C. Cir. 2011) (“We think it clear enough in the record that
semi-automatic rifles and magazines holding more than ten

2020-Firearms-Production-v14.pdf [https://perma.cc/3WK8-TVAV] (sum
of pistol and rifle magazines with 11 or more rounds).
68                                                   Nos. 23-1353 et al.

rounds are indeed in ‘common use,’ … .”). The firearms
banned by the Act and ordinances here have achieved com-
mon use in the United States. They are not unusual.
     As for magazines, Heller recognizes that ammunition feed-
ing devices may store rounds in a way that the ammunition
can be used immediately. 

. The Act and ordi-
nances limit the number of rounds a magazine may contain to
10 and 15. Nothing in the record supports these arbitrary lim-
its. “Large”- or “high”-capacity magazine is a relative term,
as pistols may ship with magazine sizes ranging from 5 to 20
rounds, and common self-loading rifles have a standard mag-
azine capacity of between 20 and 30 rounds. 9 The numbers
chosen in the Act and ordinances do not track the gun market
and are not “in common use.”
    Even if AR platform rifles were unusual, they are not more
dangerous than handguns. (Recall the test is “dangerous and
unusual.” (emphasis added). See 

; Bruen, 142 S. Ct. at
2143.) The semiautomatic mechanism in an AR-15 rifle is, in
all material respects, the same as in a semiautomatic handgun.
That mechanism is gas powered, and the impact of the pin
firing the bullet pushes back the lock mechanism, ejects the

     9 David B. Kopel, The History of Firearm Magazines and Magazine Prohi-

bitions, 78 ALB. L. REV. 849, 874 (2015) (“It is indisputable in the modern
United States that magazines of up to thirty rounds for rifles and up to
twenty rounds for handguns are standard equipment for many popular
firearms.”); id. at 859 (“The most popular rifle in American history is the
AR-15 platform, a semiautomatic rifle with standard magazines of twenty
or thirty rounds.”). Springfield, for example, ships a small handgun with
a 5-round magazine. See XD-S Mod.2 OSP 3.3” Single Stack .45 ACP Hand-
gun,     SPRINGFIELD      ARMORY     (2023),     https://www.springfield-ar-
mory.com/xd-series-handguns/xd-s-mod-2-osp-handguns/xd-s-mod-2-
osp-3-3-single-stack-45-acp-handgun [https://perma.cc/64NQ-KRWM].
Nos. 23-1353 et al.                                                          69

old shell, and loads the new round from the magazine. If
Bruen and Heller provide that semiautomatic handguns do not
fail under the “dangerous” prong, the mechanism in the AR-
15 must survive scrutiny. Indeed, a handgun could be viewed
as more dangerous than an AR-15 rifle because the handgun
is less accurate and more concealable. 10
    AR-15s are not more dangerous because of the projectile
used. The regulations challenged here do not speak to the
type of round employed, but to the capacity of the magazines
and the rate of fire. In this respect, an AR-15 and a semiauto-
matic handgun are very similar. Controlling for the same cal-
iber of round, the difference between a Glock semiautomatic
pistol and an AR-15 is just the stock and barrel length. Their
rate of fire depends on how fast a trigger can be pulled. On
that metric, an AR-15 is closer to a semiautomatic handgun




    10 One pre-Bruen analysis offered a test for “Arms” consistent with the

elements Heller pointed to: common use, unusualness, dangerousness,
and use by law-abiding citizens for lawful purposes. Eugene Volokh, Im-
plementing the Right to Keep and Bear Arms for Self Defense: An Analytical
Framework and a Research Agenda, 56 UCLA L. REV. 1443, 1481–82 (2009).
Volokh suggested that “Arms” with the same level of practical danger-
ousness as those in common use are protected. Id. Machineguns fail this
test due to their rapid rate of fire and the difficulty of firing them in a dis-
criminating way. The same with short-barreled shotguns, which combine
the lethality of a shotgun at the short distance characteristic of a criminal
attack, and the concealability of a handgun. Id. at 1482.
     The weapons banned by the Act and the ordinances here have the
same practical dangerousness as those in common use among law-abiding
citizens. See id. at 1485.
70                                                  Nos. 23-1353 et al.

(protected in Bruen and Heller) than an automatic rifle such as
the M16. 11
     Though dangerousness can be measured by many metrics,
it is best to focus on what we know. The traditional demarca-
tion for regulation has been between automatic and semiau-
tomatic weapons. Fully automatic weapons have long been
heavily regulated, and lawfully owned, fully automatic fire-
arms are very rare and expensive. 12 The Act and ordinances
violate that tradition.
    The banned arms are “in common use.” They are com-
monly possessed by law-abiding citizens for lawful purposes,
including self-defense. They may be “dangerous”—as are all
firearms—but they are not “unusual,” and thus would not be
within the history and tradition recognized in Heller of pro-
hibiting “dangerous and unusual” weapons.
    The Act and ordinances burden the rights of hundreds of
thousands of law-abiding citizens to keep and bear the types
of weapons and magazines that have long been deemed
appropriate for self-defense. This leaves one option for the
government parties—they must identify analogous weapons
regulations from at or near the time of the Founding. These
are the “how” and “why” questions of Bruen’s history and tra-
dition test—“how” did the regulation burden the Second


     11 See STEPHEN P. HALBROOK, AMERICA’S RIFLE: THE CASE FOR THE AR-

15, at 9 (2022) (“The features that make an otherwise legal semiautomatic
firearm an ‘assault weapon’ under various laws do nothing to affect the
firearm’s functional operation and, if anything, promote safe and accurate
use.”).
     12 See GARY KLECK, TARGETING GUNS: FIREARMS AND THEIR CONTROL

108–10 (1997).
Nos. 23-1353 et al.                                                  71

Amendment right, and “why” was this regulation adopted?
The government parties offer a variety of historical regula-
tions on weapons. These regulations show, they argue, that
the Act and ordinances are consistent with the Nation’s his-
tory and tradition. But the governments’ examples are not rel-
evantly similar—their “how” and “why” set them apart from
the Act and ordinances here.
    The government parties first point to regulations limiting
the public carry of certain weapons, such as pistols, dirks (a
long-bladed dagger), Bowie knives, and clubs. See, e.g., 

 (restricting concealed carry of weapons like
pocket pistols, dirks, or swords in a cane, unless the individ-
ual was “travelling on a journey”); 1813 La. Acts. 172; 

. But those regulations are limited only to the pub-
lic carry of certain weapons. The Act and ordinances here do
more, prohibiting the sale and eventually the possession of
the banned firearms. The “how” of the current regulations is
more burdensome than historical regulations limiting public
carry of weapons.
    The Bowie knife example offered by the government par-
ties and relied on by the district court in Bevis falls short as a
historical analogue under the “how” and “why” questions.
The Bowie knife was not categorically banned, just burdened
in certain ways. The “how” is different, as it was taxed, or it
could not be carried. The “why” for the Bowie knife was also
different. The knife was regulated because it was used in du-
els, not to stop a mass casualty event—the “why” proffered
here. 13 Laws banning Bowie knives are also a poor analogue


   13 For example, the Naperville ordinance states its bans are a direct

response to mass shootings over the last decade. See NAPERVILLE, ILL.
72                                                  Nos. 23-1353 et al.

because of what they ban. Guns and knives present different
dangers. Bodily harm is inflicted up-close and personal with
a knife, and from a distance with a gun. These differences cau-
tion that the “how” and “why” behind historical Bowie knife
regulations are not so comparable to justify the bans here.
    Elsewhere, the government parties note historical bans on
the sale, possession, and carry of pocket pistols, revolvers,
and other kinds of weapons. Such regulations appear to have
been uncommon. One example is an 1837 Georgia statute stat-
ing, “it shall not be lawful for any merchant … or any person
or persons whatsoever, to sell, or offer to sell, or to keep, or to
have about their person or elsewhere, any of the hereinafter
described weapons, to wit: Bowie, or any other kinds of
knives, manufactured and sold for the purpose of weapon, or
carrying the same as arms of offence or defense, pistols, dirks,
sword canes, spears … save such pistols as are known and
used as horseman’s pistols … .” 1837 Ga. Acts 90, § 1; see also
1879 Tenn. Pub. Acts 135–36, An Act to Prevent the Sale of
Pistols, chap. 96 § 1; 

, An Act to Preserve
the Public Peace and Prevent Crime, ch. XCVI, § 3.
   These regulations also tended to restrict only unusual
kinds of pistols, preserving the right to continue carrying
army or navy pistols. Even more, Heller, McDonald, and Bruen
have solidified the constitutional right to own and carry
handguns, so it is unclear what insights to draw from these
defunct regulations. The “how” of regulations like the Geor-
gia statute are thus distinguishable. The current regulations
do far more than limit small, uncommon handguns or other


MUN. CODE tit. 3, ch.19 (reciting list of mass shootings and incorporating
them into text of the ordinance).
Nos. 23-1353 et al.                                           73

outlier weapons. They limit access to many of the most popu-
lar models of semiautomatic rifles, handguns, shotguns, and
magazines. The Act and ordinances therefore impose a far
greater burden on the right to keep and bear arms. If all that
is not enough, the Supreme Court of Georgia declared the
1837 statute unconstitutional to the extent it limited one’s con-
stitutional right to carry arms openly. See Nunn v. State, 

 (1846); Bruen, 142 S. Ct. at 2147 (discussing Nunn and
the 1837 Act).
    Cook County contends that historical regulations on gun-
powder support their current ordinance. The County argues
that the “why” of those regulations is comparable to the
“why” of the Act and the county’s ordinance—preventing
mass casualty events. But the County’s argument “flies too
high.” The “why” of the gunpowder regulations was to stop
fires resulting from the combustion of stored flammable ma-
terials. Moreover, while gunpowder storage was regulated,
purchasing and possessing gunpowder was not prohibited.
Fire-safety laws do not create a comparable burden to an ab-
solute ban on arms. See Heller, 

 (“Nothing about
those fire-safety laws undermines our analysis; they do not
remotely burden the right of self-defense as an absolute ban
on handguns.”). Even more, the Court rejected this gunpow-
der analogy in Heller. 

 (“Justice Breyer cites ... gunpowder-
storage laws that he concedes did not clearly prohibit loaded
weapons, but required only that excess gunpowder be kept in
a special container or on the top floor of the home. 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.”).
74                                            Nos. 23-1353 et al.

    Various government parties also offer as historical ana-
logues regulations on trap or spring guns, fully automatic ma-
chineguns, and short-barreled rifles and shotguns. See, e.g., 

(a)(4) (short-barreled shotguns and rifles); 

§ 922(o) (machineguns); 1763–1775 N.J. Laws 346, An Act for
the Preservation of Deer and Other Game, and to Prevent
Trespassing with Guns, ch. 539, § 10 (trap guns). But the
“how” and “why” of those restrictions are materially differ-
ent as well. Trap or spring guns—rigged to fire when a string
or other device is triggered by contact—do not provide a his-
torical analogue. They fire indiscriminately, and the “why” of
banning them—the imbalance of using lethal force to protect
property versus human life—is different than the “why” the
Act and ordinances seek to address of stopping escalating gun
violence. Just so, machineguns can expend hundreds more
rounds per second than even the fastest semiautomatic fire-
arm, disqualifying such a law as an analogue.
    The majority opinion also relies on anti-carry laws as ana-
logues. But the challenged Act and ordinances ban possession
of arms. The distinction between anti-carry and anti-posses-
sion laws is critical: the first limits only the way a person may
use a firearm in public; the second categorically denies pos-
session of a firearm for any purpose. To elide this difference
between anti-carry and anti-possession laws ignores Heller
and Bruen. Bruen states that the “central” consideration in as-
sessing historical analogues is “whether modern and histori-
cal regulations impose a comparable burden on the right of
armed self-defense and whether that burden is comparably
justified.” 142 S. Ct. at 2133.
  This leaves only those regulations restricting semiauto-
matic firearms and ammunition feeding devices, but those
Nos. 23-1353 et al.                                          75

regulations all come from the twentieth century. Even if valid
for other reasons, Bruen states that regulations so far from the
time of the Founding cannot meaningfully inform the history
and tradition analysis. 142 S. Ct. at 2136–37 (“[W]e must also
guard against giving postenactment history more weight than
it can rightly bear.”).
    Even if the government parties had identified a historical
analogue that satisfied the “how” and “why” inquiries of
Bruen’s history and tradition test, a single such regulation was
not enough in that case. 142 S. Ct. at 2153. In fact, three ana-
logues were not enough in Bruen. Id. One can ask if there is
any “why” in support of the Act and ordinances that did not
also apply to the ownership and public carry of handguns in
Bruen. If the “how” and “why” of handguns did not satisfy
Bruen, what about these regulations supply a different
“why”? This question was not adequately answered at oral
argument. 14
    Because the Act and ordinances fail the “how” and “why”
questions of Bruen, the government parties have not met their
burden that these regulations are “relevantly similar” to a his-
torical law. Some hypothetical laws might satisfy the history
and tradition test—say, a law that banned carbine rifles that
hold more than six rounds, or possession of a pistol that need
not be reloaded. Magazines fall within the category of
“Arms,” so banning them must also satisfy the history and
tradition test. For example, if there had been a historical ana-
logue of “25 or fewer bullets is the number of shots a gun shall
fire,” the government parties might rely on that. But no such
laws have been cited for firearms or magazines. The


   14 Oral Arg. at 15:20.
76                                              Nos. 23-1353 et al.

government parties have failed to show that the Act and
ordinances are consistent with the Nation’s history and tradi-
tion of firearm regulation. History and tradition do not sup-
port banning firearms and magazines so many citizens own,
possess, and use for lawful purposes.
   To finish up likelihood of success on the merits, I agree
with my colleagues that on this record, the registration re-
quirement does not appear to be unconstitutional.
                                 B
    On the second consideration for a preliminary injunction,
an alleged constitutional violation often constitutes irrepara-
ble harm. See Int’l Ass’n of Fire Fighters, Loc. 365 v. City of East
Chicago, 

, 450 (7th Cir. 2022); 11A CHARLES ALAN
WRIGHT & ARTHUR R. MILLER, FEDERAL PRACTICE AND
PROCEDURE § 2948.1 (3d ed. 2022) (“When an alleged depriva-
tion of a constitutional right is involved ... most courts hold
that no further showing of irreparable injury is necessary.”).
For some constitutional violations, particularly First Amend-
ment violations, irreparable harm is presumed. Christian Legal
Soc’y v. Walker, 

, 867 (7th Cir. 2006). Although the
Supreme Court has not recognized a presumption of irrepa-
rable harm for Second Amendment violations, it has empha-
sized that the constitutional right to bear arms for self-defense
is not “a second-class right, subject to an entirely different
body of rules than the other Bill of Rights guarantees.” Bruen,
142 S. Ct. at 2156 (citing McDonald v. City of Chicago, 

 (2010) (plurality opinion)).
    This court has held that when a law is facially challenged
under the Second Amendment, “the form of the claim and the
substance of the Second Amendment right” create a “harm
[that] is properly regarded as irreparable and having no
Nos. 23-1353 et al.                                        77

adequate remedy at law.” Ezell, 651 F.3d at 699–700. In Ezell,
the court likened the plaintiff’s alleged Second Amendment
harm to a First Amendment challenge, implying a presump-
tion of irreparable harm. 

 In accord, the Ninth Circuit has
held that there is a presumption of irreparable harm where a
Second Amendment right is violated. See Baird v. Bonta, 

, 1046 (9th Cir. 2023) (“[W]e presume that a consti-
tutional violation causes a preliminary injunction movant ir-
reparable harm and that preventing a constitutional violation
is in the public interest.”) Pre-Bruen, the D.C. Circuit con-
cluded the same. See Wrenn v. District of Columbia, 

, 667–68 (D.C. Cir. 2017).
   Accordingly, a violation of the Second Amendment right
presumptively causes irreparable harm. The Act and other or-
dinances challenged here violate the Second Amendment,
and thus, irreparable harm has occurred. The majority opin-
ion does not speak to irreparable harm.
    Neither of the final two preliminary injunction factors—
balance of the equities and what the public interest dictates—
cuts against the plaintiffs. Gunshot victims and gun owners
each claim harms, and what is in the public interest on ques-
tions of gun possession and ownership is constantly under
public debate. So, I would rule that preliminary injunctions
are justified against enforcement of the challenged laws.
                             IV
    In reaching the opposite result, the majority opinion ap-
plies precedent and reasoning that Bruen abrogated.
                              A
   Notwithstanding Bruen, the majority opinion relies on rea-
soning from this court’s decision in Friedman v. City of
78                                            Nos. 23-1353 et al.

Highland Park, 

 (7th Cir. 2015). See also Wilson v.
Cook County, 

, 1035 (7th Cir. 2019) (relying on
Friedman to dismiss a Second Amendment challenge to the
Cook County ordinance banning assault weapons and large-
capacity magazines). It is true that the Act regulates firearms
and magazines in substantially the same way as the ordi-
nances in Friedman (Highland Park) and in Wilson (Cook
County), which were upheld. Compare 720 ILL. COMP. STAT. §§
5/24-1.9(a)(1), 1.10(a) with Friedman, 

 and Wilson,
937 F.3d at 1029–30. As noted in I., the City of Chicago and
City of Naperville ordinances are functionally similar to the
Act and the Cook County ordinance.
    In Friedman, this court announced a unique test for Second
Amendment questions: “whether a regulation bans weapons
that were common at the time of ratification or those that have
‘some reasonable relationship to the preservation or efficiency
of a well regulated militia,’ … and whether law-abiding citi-
zens retain adequate means of self-defense.” 

.
The government parties assert Friedman focused on the con-
siderations identified by Heller and Bruen, specifically,
historical evidence and the impact of the regulation on an in-
dividual’s meaningful opportunities for self-defense. Id.; Wil-
son, 937 F.3d at 1033. Friedman is therefore compatible with
the constitutional analysis endorsed by Bruen, the govern-
ment parties submit, and Friedman remains good law and
should control the outcome here.
    But after Bruen, Friedman’s test is no longer viable, and
much of Friedman is inconsistent with it. The Second Amend-
ment’s “reference to ‘arms’ does not apply only to those arms
in existence in the 18th century.” Bruen, 142 S. Ct. at 2132
(cleaned up). That amendment’s operative clause “does not
Nos. 23-1353 et al.                                            79

depend on service in the militia.” Id. at 2127. Indeed, the dis-
sent in Bruen admitted that under the majority opinion’s hold-
ing the scope of the right to bear arms has “nothing whatever
to do with service in a militia.” Id. at 2177–78 (Breyer, J. dis-
senting). And “the right to bear other weapons is ‘no answer’
to a ban on the possession of protected arms.” Caetano, 577
U.S. at 421 (quoting Heller, 

).
    This court in Friedman based its decision in substantial part
on its view of the benefits of the ordinance, including that the
arms ban reduced “perceived risk” and “makes the public feel
safer.” 784 F.3d at 411–12. But Bruen emphatically rejected this
sort of interest-balancing. 142 S. Ct. at 2127. Friedman also held
that categorical bans may be proper even if the limits do not
“mirror restrictions that were on the books in 1791.” 

. The Bruen decision superseded that, concluding that a re-
striction on Second Amendment rights will survive scrutiny
only if “the government identif[ies] a well-established and
representative historical analogue” to the regulation. 

.
    Friedman looked to history when it held that a court must
ask whether the arms were common at the time of ratification.

. But in Bruen, the Court was clear that “the
Second Amendment’s definition of ‘arms’ … covers modern
instruments that facilitate armed self-defense,” “‘even those
that were not in existence at the time of the founding.’” 142 S.
Ct. at 2132 (quoting Heller, 

); see Caetano, 577
U.S. at 411–12 (holding lower court’s decision that arms were
unprotected because they were not in common use at the time
of ratification was “inconsistent with Heller”).
   In Wilson, this court described Friedman as “evaluat[ing]
the importance of the reasons for the [ban] to determine
80                                           Nos. 23-1353 et al.

whether they justified the ban’s intrusion on Second Amend-
ment rights,” such as the “‘substantial’ interest[]” in “making
the public feel safer” and “overall dangerousness.” Wilson,
937 F.3d at 1036. But Bruen rejected that interest-balancing ap-
proach as “inconsistent with Heller’s historical approach.”
Bruen, 142 S. Ct. at 2129. Governments may no longer “simply
posit that the regulation promotes an important interest,” id.
at 2126, or advances a “substantial benefit,” Friedman, 

. Wilson described Friedman’s application of an interest-
balancing test as “intermediate scrutiny,” Wilson, 937 F.3d at
1036, the approach Bruen expressly left behind.
    Recently, in Atkinson v. Garland, 

 (7th Cir.
2023), this court considered the new world Bruen presented
for Second Amendment jurisprudence, in the context of pos-
session of a firearm as a felon. Id. at 1022. There, we declined
to avoid a Bruen analysis by relying on Heller and instead
stated, “[w]e must undertake the text-and-history inquiry the
Court so plainly announced and expounded upon at great
length.” Id. Neither the majority nor the dissent in Atkinson
discussed or even cited Friedman, although those opinions re-
lied on other pre-Bruen precedents from our court.
    In sum, Bruen effectively abrogated Friedman and Wilson.
The “history and tradition” methodology of Bruen is not the
framework applied in either of those cases. “Stare decisis can-
not justify adherence to an approach that Supreme Court
precedent forecloses.” Fed. Trade Comm’n v. Credit Bureau Ctr.,
LLC, 

, 767 (7th Cir. 2019). “When an intervening
Supreme Court decision unsettles [this court’s] precedent, it
is the ruling of the [Supreme] Court … that must carry the
day.” United States v. Wahi, 

 (7th Cir. 2017).
That happened here, and the district court in Bevis correctly
Nos. 23-1353 et al.                                           81

concluded that Friedman cannot be reconciled with Bruen.
This court should review the challenged laws under Bruen’s
framework, distinct from any interest-balancing approach,
and separate from the reasoning employed in Friedman and
Wilson.
    The majority opinion is correct that Friedman’s test differs
from the two-step interest balancing test of other courts that
Bruen replaced. Nevertheless, Friedman cites to history to com-
pare the arms the regulation bans, rather than the regulations
themselves. 

. Friedman discusses the features of
the weapons, including whether they are in common use for
militia or police functions. 

 It also examines the gun’s char-
acteristics—such as its weight, caliber, and magazine capac-
ity—as determinative of its value to self-defense. 

.
Representative of that analysis, the majority opinion engages
in a matching exercise between the AR-15 and the M16, as-
sessing the similarity and differences of the characteristics of
the two firearms.
    In stark contrast, in Bruen the Court did not say “Arms”
are defined by using the history and tradition of military ver-
sus civilian weaponry, such as the line drawn in the majority
opinion. Rather, the Court looked to common usage to define
the term “Arms.” Even more, the assessment in Bruen is
whether a firearm regulation has a historical analogue, 142 S.
Ct. at 2133, not whether a weapon does. Under Bruen’s frame-
work, courts can entertain the parties’ arguments as to
whether a regulation is a historical analogue. Per Bruen,
whether firearm regulations were historically grounded in a
military versus civilian distinction is to be performed as part
of the history and tradition analysis, not in the plain text re-
view, as the majority opinion does.
82                                                   Nos. 23-1353 et al.

                                     B
   The majority opinion’s reasoning departs from Bruen in
other ways, which I examine next.
     1. A weapon’s military counterpart does not determine
     whether it is an “Arm.”
    The AR-15 is a civilian, not military, weapon. No army in
the world uses a service rifle that is only semiautomatic.15
Even so, the majority opinion uses a civilian firearm’s military
counterpart to determine whether it is an “Arm.” But neither
Heller nor Bruen draw a military/civilian line for the Second
Amendment. Similarity between the AR-15 and the M16
should not be the basis on which to conclude that the AR-15
is not a weapon used in self-defense.
    The majority opinion concludes that Heller limits the scope
of “Arms” in the amendment to those not “dedicated to mili-
tary use” and those possessed for a lawful purpose. Citing to
“historical support” that “the Arms protected by the Second
Amendment do not include weapons for the military,” the
majority opinion focuses on Heller’s comment about the M16
rifle. 

. The AR-15 and the M16 are similar weap-
ons, my colleagues conclude, which means the AR-15 is be-
yond protection under the Second Amendment.
   My colleagues read the passages in Heller discussing
weapons with military capabilities too broadly, however,
placing controlling weight on supporting or explanatory lan-
guage in that decision. For example, Heller did not limit the
scope of “Arms” to those without an analogous military


     15 E. Gregory Wallace, “Assault Weapon” Myths, 43 S. ILL. U. L.J. 193,

205–06 (2018).
Nos. 23-1353 et al.                                             83

capacity. 554 U.S. at 581–82. The majority opinion emphasizes
the statement in Heller that “Arms” are “weapons that were
not specifically designed for military use and were not em-
ployed in a military capacity.” Maj. Op. at 26 (emphasis omit-
ted). But this passage most naturally means that the public
understanding of “Arms” encompassed more than weapons
designed for or employed in a military capacity. At that sec-
tion of Heller, the Court was refuting the argument that the
Second Amendment only protected a military right to keep
and bear arms. Instead, “Arms” was broad enough to include
“any thing that a man wears for his defence, or takes into his
hands, or useth in wrath to coast at or strike another.” Heller,

. That passage in Heller does not support a read-
ing that weapons the military uses are not “Arms.”
    Relying on Heller’s discussion of United States v. Miller—
the Supreme Court’s 1939 decision upholding a conviction
under the National Firearms Act against a Second Amend-
ment challenge—the majority opinion points out that milita-
ristic weapons are not “bearable” and thus not “arms” at all.
Justice Stevens in dissent in Heller viewed Miller as endorsing
a military-only view of the Second Amendment. To him, Mil-
ler says regulating “the nonmilitary use and ownership of
weapons” is fine—so the Amendment protects only the “right
to keep and bear arms for certain military purposes.” 

 at
637–38 (Stevens, J., dissenting).
   But according to Heller, Miller does not say that the Second
Amendment protects machineguns as part of ordinary mili-
tary equipment. Rather, Miller explains that a short-barreled
shotgun, the weapon at issue, is not “‘any part of the ordinary
military equipment’” nor “‘could contribute to the common
defense.’” 

 at 622 (quoting Miller, 

). In Heller,
84                                             Nos. 23-1353 et al.

the Court explained, “we therefore read Miller to say only that
the Second Amendment does not protect those weapons not
typically possessed by law-abiding citizens for lawful pur-
poses, such as short-barreled shotguns.” Id. at 625.
    The majority opinion here quotes this passage and re-
frames it as a military-analogue test. It introduces the passage
with Heller’s observation that an M16 is “most useful in mili-
tary service.” Id. at 627. But after Heller, we know Miller does
not address a weapon’s military use. Because the National
Firearms Act of 1934 targeted the firearms most commonly
used by criminals and gangs, Miller’s “lawful use” language
relates to criminal use, not military use.
    One example of this military-analogue test falling short is
when the majority opinion compares the rates of fire of the
AR-15 and the M16. My colleagues credit the AR-15’s rate of
fire as “‘only’ 300 rounds per minute,” which they do not see
as a relevant difference from the M16’s 700 rounds per mi-
nute. Maj. Op. at 33. The two record sources they point to do
not support a 300-rounds-per-minute rate; in fact, those
sources give good reasons to doubt that figure.
    The first is the district court’s opinion in Bevis, which ex-
plains: “[A] shooter using a semiautomatic weapon can
launch thirty rounds in as little as six seconds, with an effec-
tive rate of about a bullet per second for each minute of firing,
meeting the U.S. Army definition for ‘rapid fire.’” Set to the
side the district court’s concession that the effective rate is ac-
tually only sixty rounds per minute. For the 300-rounds-per-
minute figure, the district court cited a law journal article that
spends nine pages discussing the dubious origins of the 300-
Nos. 23-1353 et al.                                           85

rounds-per-minute claim. 16 Wallace agrees that 30 shots in six
seconds is possible—if you are an expert at operating firearms
and you neglect aiming and reloading.17
    The second source that might be referenced for the figure
is a government witness’s report in Herrera. James Yurgealitis
included a chart listing weapons, an ammunition type, and
the “semiautomatic cyclic rate” of each. Each rifle, including
“M-16/AR-15 Rifle,” has a cyclic rate of exactly 300 rounds per
minute, and the three pistols have a rate of “300–400 rounds
per minute.” Yurgealitis offers no source for his calculations.
He does not describe the firing conditions or how the shooter
timed the shots.
    Yurgealitis describes the rate as “cyclic,” a type of fire
where “the gunner holds the trigger to the rear while the as-
sistant gunner feeds ammunition into the weapon.” DEP’T OF
THE ARMY, ARMY TRAINING PUBLICATION: INFANTRY PLATOON
AND SQUAD, ATP 3-21.8, at Appendix F. The cyclic rate “pro-
duces the highest volume of fire the machine gun can fire”
and is a drastic step, as it “can permanently damage the
machine gun and barrel and should be used only in case of
emergency.” Id. It is difficult to see how a gunner could fire
an AR-15 cyclically. Because it is a semiautomatic firearm, if
the trigger were held to the rear, the cyclic rate would be one
round per minute. Yurgealitis does not explain how this can
be done.
   The effective rate of fire, rather than the cyclic rate, would
be a better comparison. There, Yurgealitis helps. He includes
in his report a table from an Army field manual on rifle

   16 See Wallace, supra note 15 at 214–22.

   17 See id. at 217–18.
86                                              Nos. 23-1353 et al.

marksmanship listing the M16’s maximum semiautomatic ef-
fective rate at 45 rounds per minute—more than four times
slower than its maximum automatic effective rate.
    Heller does not draw a line between firearms that are mil-
itary counterparts and those that are not. That demarcation
should not decide whether firearms and magazines are pro-
tected under the Second Amendment.
     2. A “military weapon” is defined too broadly.
   Even if Heller drew such a line, the majority opinion’s
standard for what constitutes a “military weapon” renders
the “military” category substantially overbroad.
    The majority opinion draws a line between “private” or
“mixed private/military” weapons on one side (also charac-
terized as “dual use” weapons) and “military weapons” on
the other side. Military weapons are defined as “weapons that
may be essentially reserved to the military,” Maj. Op. at 31
n.8—meaning that a military weapon is one not made availa-
ble for public use. The only “characteristic” that matters is
that the government decided to ban it. “Dual use” weapons
are those “private parties have a constitutionally protected
right to ‘keep and bear’” and “the military provides [] to its
forces.” Id. “In this sense, there is a thumb on the scale in favor
of Second Amendment protection.” Id. Under the majority
opinion’s definition, “dual use” weapons are on the side of
the line protected by the Second Amendment.
    Applying their framework, my colleagues find the AR-15
“more like” the M16 by comparing the firearms’ characteris-
tics. Id. To my colleagues, the firearms look the same (“same
core design”), operate the same (“same patented operating
system”), and have similar specifications (same ammunition,
Nos. 23-1353 et al.                                              87

kinetic energy, muzzle velocity, and effective range), identi-
fying “the only meaningful distinction” as an M16’s auto-
matic-fire capability. Id. at 31–32. But because the AR-15 is not
“essentially reserved to the military” and shares characteris-
tics with “private” weapons, such as being semiautomatic, the
AR-15 is at most a “dual use” weapon. So under the majority
opinion’s categories, the AR-15 should warrant Second
Amendment protection.
     In any event, because the majority opinion defines a mili-
tary weapon as any that “may be essentially reserved to the
military,” a weapon’s characteristics are not relevant to how
it is categorized. Thus, any combat weapon would be a mili-
tary weapon. This effectively allows the U.S. Armed Forces to
decide what “Arms” are protected under the Second Amend-
ment. Such a “military veto” is mistaken for at least three rea-
sons.
    First, the military has historically selected for commission
firearms already publicly available and thus on the “dual use”
side of the line. Privately available repeating and semiauto-
matic rifles, and the arms the American military selected for
wartime use, overlapped substantially at least until the 1930’s.
    When the Second Amendment was ratified, repeaters—
firearms capable of repeated firing before they required man-
ual reloading—were useful for military purposes and were
widely available for civilian purchase. The Girandoni air rifle,
for example, was invented for the Austrian army.18 The
“state-of-the-art repeater” at the time, the Girandoni was


   18 NICHOLAS J. JOHNSON, ET AL., FIREARMS LAW AND THE SECOND

AMENDMENT: REGULATION, RIGHTS, AND POLICY 2206 (3d ed., 2021).
88                                                      Nos. 23-1353 et al.

useful for hunting as well—Meriwether Lewis took one on his
expedition. 19 In 1828, the military awarded a contract to a
gunsmith to produce the Jennings repeater for military use.20
But the military only “considered the guns promising” after
seven years of “private use,” as the repeater had been circu-
lating at least since 1821. 21 Another repeater, the Henry, won
a military contract after a Union captain used it to defend his
home against seven Confederates who ambushed him while
eating dinner with his family. 22
    In 1900, the military began considering semiautomatic ri-
fles but, after years of searching, decided to stick with the
.30’06 Springfield bolt-action rifle. 23 Even though “semi-auto-
matic rifles for the civilian market were abundant,” the mili-
tary declined to select one because they were too complicated
and brittle for field use. 24 In the 1930s, the military’s desire to
issue semiautomatic rifles caused it to “encourage[] … private
experimentation” in development and testing. 25 A military
veto contravenes the robust history of “dual use” weapons
beyond the private sector.




     19 See id.

     20 See id. at 2221.

     21 Id.

     22 HORACE WILLIAM SHALER CLEVELAND, HINTS TO RIFLEMEN 180–81

(1864). See also id. at 179 (reproducing letter from a private citizen testify-
ing to the exceptional quality of the weapon).
    23 See JOHNSON at 2233–34.

     24 Id. at 2233.

     25 Id. at 2234.
Nos. 23-1353 et al.                                                    89

    Second, the military has historically commissioned pistols,
a firearm that is an “Arm” under Heller. Pistols have always
been standard-issue military firearms. Under the majority
opinion’s approach, Heller would have been mistaken.
    Major Pitcairn began the American Revolution with a shot
from his pistol. 26 General George Washington carried pistols
into battle at Valley Forge, Monmouth, and Yorktown. 27 In
1811, a brigade major in the Massachusetts militia described
the pistol as a standard weapon for an infantryman in a com-
prehensive guide to the day’s military science. 28
    The military has not stopped issuing pistols. In 1911, after
lengthy trials and revisions with Colt and gun designer John
Browning, the military selected for its troops the Colt Model
1911. 29 It is unclear whether that model was available for ci-
vilian purchase after the military contract in 1911. But like
more common civilian handguns, the M1911 was semiauto-
matic and had an eight-round magazine. 30 Indeed, the Civil-
ian Marksmanship Program, a federally chartered 501(c)(3)




    26 See CHARLES WINTHROP SAWYER, 1 FIREARMS IN AMERICAN HISTORY:

1600 TO 1800, at 72 (1910).
    27 See Evan Brune, Arms of Independence: The Guns of the American Rev-

olution, AM. RIFLEMAN (July 2, 2021), https://www.americanrifle-
man.org/content/arms-of-independence-the-guns-of-the-american-revo-
lution [https://perma.cc/9S69-T56Y].
     28 See E. HOYT, PRACTICAL INSTRUCTIONS FOR MILITARY OFFICERS 111

(1811).
     29 See JOHNSON at 2232.

    30 See id.
90                                                   Nos. 23-1353 et al.

entity responsible for arranging sales of decommissioned mil-
itary service weapons to the public, sells Colt M1911s today.31
    In the 1980s, the military switched to the Beretta M9, a
handgun with a counterpart available for purchase today on
Beretta’s website. In fact, the M9 was designed and available
to civilians a decade before the military selected it as the
Beretta 92. 32 The only differences between the military-issue
M9 and the one for public sale are the markings, the dots on
the sights, and the screw heads. 33 Under the majority opinion,
the military’s decision to award Beretta a military contract for
the Beretta 92 would take the firearm out of the “Arms” pro-
tected by the Second Amendment.
   Third, the military’s decommissioning and sale of its sur-
plus weapons would mean that the Second Amendment right
might spring into and out of life. The military sometimes de-
commissions service weapons and sells them to the public
through the Civilian Marksmanship Program, as mentioned

     31      See    About,     CIV.     MARKSMANSHIP       PROG.     (2023),
https://thecmp.org/about/ [https://perma.cc/L7T5-6T5D];1911 Information,
CIV. MARKSMANSHIP PROG. (2023), https://thecmp.org/sales-and-ser-
vice/1911-information/ [https://perma.cc/7HQW-G3VJ].
      32 See American Service Pistols & Civilian Counterparts, KEYSTONE

SHOOTING CTR. (2023), https://keystoneshootingcenter.com/blog/ameri-
can-service-pistols-civilian-counterparts [https://perma.cc/UG45-V46Q].
      33 See Christopher Bartocci, Beretta Government vs Commercial M9 Iden-

tification, SMALL ARMS SOLUTIONS LLC (May 28, 2018), https://smal-
larmssolutions.com/home/beretta-government-vs-commercial-m8-identi-
fication [https://perma.cc/EDT4-JEXT]; Bob Campbell, Range Report:
Beretta’s M9 Civilian Version, CHEAPER THAN DIRT: THE SHOOTER’S LOG
(Feb. 22, 2016), https://blog.cheaperthandirt.com/berettas-m9-civilian-ver-
sion [https://perma.cc/VL7T-ZXQA] (“The M9 is a variant that’s as close
to the military M9 as possible. The sights are marked in a different man-
ner, and the finish differs from the standard M92.”).
Nos. 23-1353 et al.                                          91

above. As with the M16, the military also decides not to renew
contracts for weapons it deems no longer fit for military use.
The majority opinion does not explain the status of a weapon
like this, including whether the right to possess it springs to
life, or if its analogues become “Arms.”
   3. The examples given are not historical analogues.
    The majority opinion sets forth “the relevant question [a]s
what are the modern analogues to the weapons people used
for their personal self-defense in 1791, and perhaps as late as
1868.” Maj. Op. at 38. But when declaring its holding in Bruen,
the Court discussed historical analogues with reference not to
weapons, but to regulations. Following Heller, Bruen consid-
ered “whether ‘historical precedent’ from before, during, and
even after the founding evinces a comparable tradition of reg-
ulation.” Bruen, 

, 2131–32. “Only if a firearm
regulation is consistent with this Nation’s historical tradition
may a court conclude that the individual’s conduct falls out-
side the Second Amendment’s ‘unqualified command.’” Id. at
2126.
    The seven historical examples the majority opinion offers
as comparators are laws or ordinances which it says support
“a distinction between weapons and accessories designed for
military or law-enforcement use, and weapons designed for
personal use.” Id. at 45. For my colleagues, the challenged Act
and ordinances carry forward this same distinction. Under
Bruen, though, these examples do not satisfy the “how” and
“why” questions in the history and tradition test, and thus are
not comparators for the challenged Act or ordinances.
92                                                      Nos. 23-1353 et al.

    The first example is a 1746 Boston ordinance outlawing the
discharge of a cannon, gun or pistol within city limits. 34 The
second is an allusion to similar ordinances in Cleveland in the
nineteenth century. The fourth refers to late nineteenth cen-
tury ordinances restricting the carry of various weapons. ex-
cept for peace officers. Such prohibitions differ, however,
from a categorical ban of a class of weapons from private
ownership which burden the right of armed self-defense.
Regulations against the discharge of weapons compare better
to modern criminal statutes prohibiting, for example, the
reckless discharge of a firearm. See 720 ILL. COMP. STAT. § 5/24-
1.5(a). And prohibitions on the carrying of certain weapons
do not amount to a categorical ban of whole classes of fire-
arms. These examples thus fail the “how” question in Bruen.
    The fifth, sixth, and seventh examples are the National
Firearms Act of 1934 and two amendments to it: the Omnibus
Crime control and Safe Streets Act of 1968, and Firearm Own-
ers’ Protection Act of 1986. Yet these examples do not provide
insight into the public understanding of the Second Amend-
ment right in 1791 (or in 1868). They are too far removed from
the ratification of the Constitution (or of the Fourteenth
Amendment) to qualify as historical analogues under Bruen.
They therefore fail the “why” question in Bruen.
   The remaining third example cites dozens of Bowie knife
regulations which forbid or limit their use, specifically citing
an 1884 Arkansas statute outlawing “the sale of all dirks,
Bowie knives, cane-swords, metal knuckles, and pistols, ex-
cept as for use in the army or navy of the United States.” This


       34 Heller rejected this regulation as a historical analogue. 

.
Nos. 23-1353 et al.                                               93

law was passed after ratification of the Fourteenth Amend-
ment and banned the sale of these knives. It did not categori-
cally ban their possession. This example fails the “how” and
the “why” test of Bruen for the reasons given previously.
    Attempting to show that the “how” test has been correctly
applied, my colleagues point to what they consider a “huge
carve-out” in the Act. Maj. Op. at 39. To the contrary, excep-
tions to the categorical ban in the Act are narrow. The Act out-
right forbids the manufacture, delivery, sale, importation, and
purchasing of the covered arms within the state of Illinois. On
January 1, 2024, a total ban on possession of the covered arms
takes effect. 720 ILL. COMP. STAT. § 5/24-1.9(c). Though an ex-
ception exists for those who submit a compliant “endorse-
ment affidavit” to the Illinois State Police, id. § 5/24-1.9(d), the
majority opinion mistakes its scope. The exception is limited
to the sale or transfer of a covered arm: (1) to seven specially
excepted classes of authorized persons; (2) to the United
States; or (3) in another state or for export. Id. § 5/24-1.9(e).
And the only people who can take advantage of this exception
are current in-state residents who possess a covered arm prior
to January 1, 2024, and future in-state residents who move
into Illinois already in possession of a covered arm. Id. 35 Such
a narrow exception cannot legitimize a broad categorical ban
on the ownership, possession, purchase, and sale of a vast
swath of arms.
    For my colleagues, it is sufficient that the seven regula-
tions deemed similar “are representative of [the] tradition” of

   35 The municipal ordinances are even more limiting, excepting from

their reach only military and law enforcement personnel. NAPERVILLE,
ILL., MUN. CODE tit. 3 ch. 19 § 2; CHI. MUN. CODE § 8-20-075(b); COOK
COUNTY, ILL. CODE § 54-212(a)(1).
94                                            Nos. 23-1353 et al.

“regulating the especially dangerous weapons of the time.”
Yet, Bruen requires more. The particulars of the historical an-
alogues are critical; they illustrate whether the Act and the
municipal ordinances place comparable burdens on the Sec-
ond Amendment right when considered against historical an-
alogues. Bruen itself gave weight to the differences between
the particulars of regulations. 142 S. Ct. at 2148–49 (rejecting
nineteenth century surety statutes as sufficiently analogous to
restrictions on public carry because these laws did not consti-
tute a “ban[] on public carry,” indicating their “burden” on
public carry was “likely too insignificant.”). The examples the
majority opinion cites may illustrate weapons regulation gen-
erally. But none of them is a categorical ban on an entire class
of arms.
                               V
   Since Bruen, this is the first federal appellate court to up-
hold a categorical ban on semiautomatic weapons and certain
magazines.
    The decision in Barnett was correct. The district court
properly rejected the notion that the Second Amendment pro-
tects only the possession and use of weapons for self-defense.
The banned magazines are “Arms,” as are other appurte-
nances such as a pistol grip and a flash suppressor. The court
correctly read Heller and Bruen to locate “in common use” in
Bruen’s history and tradition and applied the “how” and
“why” test to conclude that concealed carry regulation differs
from a ban on possession and does not pass as a historical an-
alog. This led the court to correctly issue an injunction against
the Act.
   The district court in Bevis correctly found standing, noted
that unlike other constitutional amendments the Second
Nos. 23-1353 et al.                                           95

Amendment protects a tangible item, and concluded that
Friedman did not survive Bruen. I disagree, however, with the
court’s decisions in Bevis to limit “Arms” to those weapons
that are not “particularly dangerous,” and its justification of
the Act and the Naperville ordinance under the historical test
without mentioning Bruen’s “how” and “why” test. As noted
above, the court’s Bowie knife analogue misses the mark. In
Herrera the district court relied heavily on the memorandum
opinion and order in Bevis, incorporating large parts of that
decision.
    I would affirm the decision in Barnett and reverse the de-
cisions in Bevis and Herrera and lift our court’s stay on the in-
junction against the Act. I would vacate the decisions in Bevis
and Herrera and remand for the district court to reconsider the
denial of the injunction against the challenged municipal or-
dinances.
   For these reasons, I respectfully dissent.