PRECEDENTIAL

      UNITED STATES COURT OF APPEALS
           FOR THE THIRD CIRCUIT
               ______________

                     No. 18-3595
                   ______________

            RAYMOND HOLLOWAY, JR.

                           v.

ATTORNEY GENERAL UNITED STATES OF AMERICA;
    DEPUTY DIRECTOR BUREAU OF ALCOHOL
      TOBACCO FIREARMS & EXPLOSIVES;
 DIRECTOR FEDERAL BUREAU OF INVESTIGATION;
         UNITED STATES OF AMERICA,
                               Appellants
               ______________

      Appeal from the United States District Court
        for the Middle District of Pennsylvania
                (D.C. No. 1-17-cv-00081)
      District Judge: Hon. Christopher C. Conner
                    ______________

                Argued October 2, 2019
                   ______________

  Before: SHWARTZ, FUENTES, and FISHER, Circuit
                    Judges.
                 (Filed: January 17, 2020)
                     ______________

                        OPINION
                     ______________



Joseph H. Hunt
Assistant Attorney General
David J. Freed
United States Attorney
Mark B. Stern
Thais-Lyn Trayer [ARGUED]
Tyce R. Walters
United States Department of Justice
950 Pennsylvania Avenue, N.W.
Civil Division, Room 7712
Washington, D.C. 20530

      Counsel for Appellants United States of America,
      Attorney General United States of America, Deputy
      Director Bureau of Alcohol Tobacco Firearms &
      Explosives, and Director Federal Bureau of
      Investigation

Adam J. Kraut
Joshua Prince [ARGUED]
Prince Law Offices
646 Lenape Road
Bechtelsville, PA 19505

      Counsel for Appellee Raymond Holloway, Jr.




                             2
Joseph G. S. Greenlee
Firearms Policy Coalition
1215 J Street, 17th Floor
Sacramento, CA 95814

      Counsel for Amici Curiae Firearms Policy Coalition
      Inc, Firearms Policy Foundation, Madison Society
      Foundation Inc, and Second Amendment Foundation
      Inc



SHWARTZ, Circuit Judge.

       Drunk driving is a dangerous and often deadly crime.
“Approximately a quarter million people are injured annually
in alcohol-related crashes,” Begay v. United States, 

, 156-57 (2008) (Alito, J., dissenting), and the number
“who are killed . . . by drunk drivers is far greater than the
number of murders committed” during many other violent
crimes, 

& n.4. “[F]rom 1982 to 2016, alcohol-related
accidents took roughly 10,000 to 20,000 lives in this Nation
every single year. In the best years, that would add up to more
than one fatality per hour.” Mitchell v. Wisconsin, 

, 2536 (2019) (emphasis omitted) (citations omitted).

       Today, we consider whether Pennsylvania’s driving
under the influence (“DUI”) law, which makes a DUI at the
highest blood alcohol content (“BAC”) a first-degree
misdemeanor that carries a maximum penalty of five years’
imprisonment, see 18 Pa. Cons. Stat. Ann. § 1104; 75 Pa. Cons.
Stat. Ann. §§ 3802(c), 3803(b)(4), constitutes a serious crime




                              3
that requires disarmament. Plaintiff Raymond Holloway, Jr.,
was convicted under this statute, and by the terms of 18 U.S.C.
§ 922(g)(1), he is prohibited from possessing a firearm.
Holloway claims this prohibition violates his Second
Amendment rights. The District Court agreed and enjoined
applying § 922(g)(1) to him. Because Holloway was convicted
of a serious crime as contemplated by Binderup v. Attorney
General United States of America, 

(3d Cir. 2016)
(en banc), applying § 922(g)(1) to him does not offend the
Second Amendment. Therefore, we will reverse the District
Court’s order and remand for the entry of judgment in favor of
the Government.

                               I

       In 2002, Holloway was convicted of a DUI at the
highest BAC, but the charge was dismissed upon his
completion of an accelerated rehabilitation program. In 2005,
Holloway was again arrested for driving under the influence
and registered a BAC of 0.192%. Holloway pled guilty to
violating 75 Pa. Cons. Stat. Ann. § 3802(c) for driving under
the influence at the highest BAC (greater than 0.16%). He
received a sentence of 60 months’ “Intermediate Punishment,”
including 90-days’ imprisonment that allowed him work
release, a $1,500 fine, and mandatory drug and alcohol
evaluation.

       In 2016, Holloway sought to purchase a firearm but was
unable to do so because of his disqualifying DUI conviction.
Holloway sued the Attorney General of the United States and
other federal officials (the “Government”) in the United States
District Court for the Middle District of Pennsylvania,
claiming that § 922(g)(1) is unconstitutional as applied to him




                              4
and seeking declaratory and permanent injunctive relief. The
parties filed cross-motions for summary judgment.

       The District Court granted Holloway’s motion for
summary judgment, awarded him a declaratory judgment, and
entered a permanent injunction barring the Government from
enforcing § 922(g)(1) against him. Holloway v. Sessions, 

, 463 (M.D. Pa. 2018). Applying Binderup, the
Court held that § 922(g)(1) is unconstitutional as applied to
Holloway because (1) Holloway’s DUI offense was a non-
serious crime that has not historically been a basis for the
denial of Second Amendment rights, 

and (2) the
Government failed to demonstrate that disarmament of
individuals like Holloway would promote the public safety,
particularly given his decade of crime-free behavior, 

62. The Government appeals.

                               II1


       1
          The District Court had jurisdiction under 28 U.S.C. §
1331. We have jurisdiction under 28 U.S.C. § 1291. Our
review of a district court’s order granting summary judgment
is plenary, Mylan Inc. v. SmithKline Beecham Corp., 

, 418 (3d Cir. 2013), and we view the facts and make all
reasonable inferences in the non-movant’s favor, Hugh v.
Butler Cty. Family YMCA, 

, 266-67 (3d Cir.
2005). Summary judgment is appropriate where “there is no
genuine dispute as to any material fact and the movant is
entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a).
The moving party is entitled to judgment as a matter of law
when the non-moving party fails to make “a sufficient showing
on an essential element of her case with respect to which she




                               5
                               A

      The sole issue on appeal is whether applying 18 U.S.C.
§ 922(g)(1)2 to Holloway, which makes it unlawful for him to
possess a firearm due to his prior conviction, violates his
Second Amendment rights.

        In District of Columbia v. Heller, the Supreme Court
held that the Second Amendment protects the right of “law-
abiding, responsible citizens to use arms in defense of hearth
and home.” 

, 635 (2008). This right, however,
“is not unlimited.” 

Indeed, the Court cautioned that
“nothing in [its] opinion should be taken to cast doubt on
longstanding prohibitions on the possession of firearms by
felons.” 

described the felon ban as just one
“example[]” of “presumptively lawful regulatory measures.”

n.26.

       Since Heller, we have been called upon to determine
whether various laws unlawfully infringe the Second
Amendment. Some of these laws regulate who can possess
firearms, see, e.g., Beers v. Att’y Gen. U.S., 

, 155-
56 (3d Cir. 2019) (ban on possession by those adjudicated


has the burden of proof.” Celotex Corp. v. Catrett, 

, 323 (1986).
        2
          Section 922(g)(1) makes it unlawful for any person
convicted of “a crime punishable by imprisonment for a term
over one year” to possess a firearm. Excluded from this
definition is any crime “classified by the laws of the State as a
misdemeanor and punishable by a term of imprisonment of two
years or less.” 18 U.S.C. § 921(a)(20)(B).




                               6
mentally defective or committed to mental institution);
Binderup, 

(ban on possession by certain
convicts). Other laws regulate the type of firearms that may be
possessed. See, e.g., Ass’n of N.J. Rifle & Pistol Clubs, Inc.
v. Att’y Gen. N.J. (N.J. Rifle), 

(3d Cir. 2018)
(large capacity magazines). In each instance, we examined the
challenged law by applying the two-part test first articulated in
United States v. Marzzarella, 

(3d Cir. 2010).
Under that test, we first “ask whether the challenged law
imposes a burden on conduct falling within the scope of the
Second Amendment’s guarantee.” 

“If it does not,
our inquiry is complete.” 

does, we move to the second
step: we evaluate the law under some form of heightened
scrutiny. See 

       After Marzzarella, we addressed a constitutional
challenge to § 922(g)(1) in United States v. Barton, 

(3d Cir. 2011). Barton recognized that § 922(g)(1) was
one of the “presumptively lawful” measures referenced in
Heller, 

but that individuals could challenge
§ 922(g)(1) on an as-applied basis, 

Barton,
however, did not expressly apply Marzzarella’s two-step
framework. 

held that a challenger could
rebut the presumption that § 922(g)(1) constitutionally applied
to him by “present[ing] facts about himself and his background
that distinguish his circumstances from those of persons
historically barred from Second Amendment protections.” 

The “historically barred” class, Barton concluded, was
individuals “likely to commit violent offenses.” 

Thus, Barton held that if an individual could show that he
posed no threat of future violence, then § 922(g)(1) could not
constitutionally apply to him. 

                               7
       We revisited Barton and as-applied challenges to
§ 922(g)(1) as an en banc Court in Binderup. Binderup
resulted in several opinions from fifteen judges: (1) an opinion
by Judge Ambro, joined in full by two judges and joined
additionally in part by four other judges; (2) an opinion by
Judge Hardiman, joined in full by four judges, and which
concurred in part with Judge Ambro and concurred in the
judgment; and (3) an opinion by Judge Fuentes, joined by six
judges (some of whom joined parts of Judge Ambro’s opinion),
which concurred in part, dissented in part, and dissented from
the judgment.

       There are no specific rules for how to identify the
holdings and legal standards from split circuit opinions. We
can, however, look to the rules we use to identify such
standards in fractured Supreme Court opinions, as set forth in
Marks v. United States, 

(1977), and its progeny.3
We need not conduct an explicit Marks analysis of the
Binderup opinions here because we already recited its
holdings, as expressed by Judge Ambro’s controlling opinion,
in 

-56;4 see also N.J. 

           Marks is often applied by judges who did not
participate in the opinion being reviewed. In this case, fourteen
of the fifteen judges who participated in Binderup remain on
our Court and know what it held and did not hold.
        4
          In Beers, we explained that at step one of Binderup, “a
challenger ‘must (1) identify the traditional justifications for
excluding from Second Amendment protections the class of
which he appears to be a member, and then (2) present facts
about himself and his background that distinguish his
circumstances from those of persons in the historically barred

(quoting 

-




                               8
130 (Bibas, J., dissenting) (describing Judge Ambro’s
Binderup opinion as the “controlling opinion”), and it binds
us.5 Mateo v. Att’y Gen. U.S., 

, 231 n.6 (3d Cir.


47). “If a challenger passes these two hurdles, ‘the burden
shifts to the Government to demonstrate that the regulation
satisfies some form of heightened scrutiny[.]’” 

836 F.3d at 347). Beers further explained that
Binderup overruled Barton in large part and “[w]here the
historical justification for disarming felons was because they
had committed serious crimes, risk of violent recidivism was
irrelevant, ‘and the seriousness of the purportedly
disqualifying offense is our sole focus throughout
Marzzarella’s first step.’” 

(quoting 

) (emphasis omitted).
        5
          Although Beers did not explicitly conduct a Marks
analysis, Beers set forth the Binderup majority holdings. In
Marks, the Supreme Court held that when “no single rationale
explaining the result enjoys the assent of five justices, the
holding of the Court may be viewed as that position taken by
those Members who concurred in the judgment on the
narrowest 

(internal quotation marks
and citation omitted). Marks expresses one way to identify a
holding from among separate opinions. The Supreme Court
has adopted other approaches for examining fractured opinions
to identify the rule or rules a majority endorsed. See, e.g.,
United States v. Jacobsen, 

, 117 n.12 (1984)
(“[T]he disagreement between the majority and the dissenters
in [a previous] case with respect to the [application of law to
fact] is less significant than the agreement on the standard to
be applied . . . .”); Moses H. Cone Mem’l Hosp. v. Mercury
Constr. Corp., 

, 17 (1983) (“[T]he Court of Appeals
correctly recognized that the four dissenting Justices and




                              9
2017) (prior panel’s precedential opinion “binding on
subsequent panels”); see also Jackson v. Danberg, 

, 165 n.10 (3d Cir. 2011) (applying a legal standard derived
from a previous panel opinion’s Marks analysis as the law of
our Circuit).

       Nevertheless, both Beers and Marks reveal the
following relevant Binderup holdings agreed to by a majority
of judges:

       (1) Marzzarella’s two-step test—and not the test
articulated in Barton—governs Second Amendment



Justice Blackmun formed a majority to require application of
the Colorado River test.”). Whatever the test, “our goal in
analyzing a fractured [opinion] is to find ‘a single legal
standard . . . [that] when properly applied, produce[s] results
with which a majority of justices in the case articulating the
standard would agree.’ . . . To that end, we have looked to the
votes of dissenting justices if they, combined with the votes
from plurality or concurring opinions, establish a majority
view on the relevant issue.” United States v. Donovan, 

, 182 (3d Cir. 2011) (quoting Planned Parenthood of
Se. Pa. v. Casey, 

, 693 (3d Cir. 1991), modified
on other grounds, 

(1992)) (first alteration added);
see also 

(deriving the rule
established in a particular case by combining one opinion that
garnered two votes with the opinion of the four dissenters);
B.H. ex rel Hawk v. Easton Area Sch. Dist., 

, 310
(3d Cir. 2013) (stating that we have “count[ed] even dissenting
justices’ votes that, by definition,” did not concur in the
judgment to identify a majority’s holding).




                              10


(Ambro, J.); 

(Fuentes,
J.);6

       (2) At Marzzarella step one for challenges to
§ 922(g)(1), we determine whether an individual has
committed a “serious” offense, and thus was an “unvirtuous
citizen[]” who was historically barred from possessing
firearms and fell out of the Second Amendment’s scope, 

(Ambro, J.); 

(Fuentes, J.);7

       6
          Chief Judge Smith and Judge Greenaway, Jr., joined
Judge Ambro’s opinion in its entirety, for a total of three
judges. Then-Chief Judge McKee and Judges Vanaskie,
Shwartz, Krause, Restrepo, and Roth joined Judge Fuentes for
a total of seven judges. Thus, any agreement between Judge
Ambro’s and Judge Fuentes’ opinions represents agreement by
ten judges.
        Judges Fuentes, Vanaskie, Krause, and Roth also
“expressly” joined the portions of Judge Ambro’s opinion
laying out the framework for as-applied challenges, for a total
of seven judges. 

n.1 (Ambro, J.);

n.72 (Fuentes, J.). Judges McKee, Shwartz, and
Restrepo did not “expressly” join Judge Ambro’s opinion
“because they reject[ed] the notion that the Marzzarella
framework can be reconciled with any aspect of Barton’s as-
applied Second Amendment analysis, which they would
overrule entirely.” 

n.1 (Ambro, J.). Thus, ten
Binderup judges rejected Barton and held that Marzzarella’s
framework governs as-applied challenges.
        7
           Although Judge Ambro, joined by two judges,
disagreed with Judge Fuentes, joined by six judges, over “how
to decide whether any particular crime is serious enough” to
warrant 

(Fuentes, J.) (emphasis




                              11
       (3) Barton’s focus on whether the challenger’s crime
was violent or whether the challenger poses a threat of violence
is overruled, 

(Ambro, J.); 

n.72 (Fuentes,
J.);

          (4) a challenger, otherwise barred from possession by
§ 922(g)(1), can make a factual showing that he falls outside
of the historically barred class, 

& n.3, 349 (Ambro,
J.); 

(Hardiman, J.);8

        (5) intermediate scrutiny applies at Marzzarella step
two, 

(Ambro, J.); 

(Fuentes, J.).9


omitted), a total of ten judges agreed that the correct test at step
one for challenges to § 922(g)(1) is whether the offense is
“serious,” not whether the offense is violent, and thus
overruled Barton’s focus on violence for this inquiry.
       8
         Judges Fisher, Chagares, Jordan, and Nygaard joined
Judge Hardiman’s opinion for a total of five judges.
       9
         Our dissenting colleague agrees that a majority in
Binderup: (1) rejected the idea that the Second Amendment
excludes only those who commit violent offenses and that,
because that majority adopted the “virtuous citizenry” theory
of serious offenses, the Second Amendment excludes “any
person who has committed a serious criminal offense, violent
or nonviolent,” Dissenting Op. at 2; 

(Ambro, J.); 

(Fuentes, J.); (2) held that we
evaluate § 922(g)(1) under intermediate scrutiny, not strict
scrutiny, Dissenting Op. at 24; 

(Ambro, J.); 

(Fuentes, J.); and (3) held that Barton
was overruled to the extent it suggested that (a) the Second
Amendment excludes only those who commit violent
offenses, 

(Ambro, J.); 

(Fuentes, J.),




                                12
        Thus, as we said in 

, Binderup
held that “the two-step Marzzarella framework controls all
Second Amendment challenges, including as-applied
challenges to § 

(Ambro, J.). At
step one, the challenger must “identify the traditional
justifications for excluding from Second Amendment
protections the class of which he appears to be a member[.]”

When the class at issue is historically excluded
convicts, as here and in Binderup, the challenger must show
that he was not previously convicted of a serious crime. 

A crime is “serious” based on circumstances related to
the offense, 

and so evidence of a challenger’s
rehabilitation or his likelihood of recidivism is not relevant, 

There are no fixed rules for determining whether an
offense is serious but various factors may be informative
including, but not limited to, whether the crime poses a danger
or risk of harm to self or others, whether the crime involves
violence or threatened violence, the classification of the
offense, the maximum penalty, the penalty imposed, and how
other jurisdictions view the crimes. See 

If a


(b) “the passage of time or evidence of rehabilitation will
restore the Second Amendment rights of people who
committed serious crimes,” 

(Ambro, J.); 

n.1, or (c) that strict scrutiny rather than intermediate scrutiny
applies at step two of the Marzzarella framework, 

, 

(Fuentes, J.); Dissenting Op. at 2, 6, 24.
        10
           In Binderup, Judge Ambro considered: (1) whether
the crime of conviction was classified as a misdemeanor or
felony, (2) whether the criminal offense involves violence or
attempted violence as an element, (3) the sentence imposed,
and (4) whether there is a cross-jurisdictional consensus as to
the seriousness of the crime. 

.




                                13
challenger makes a “strong” showing that the regulation
burdens his Second Amendment rights and that he has not
committed a “serious” crime, and thus is different from those
historically barred from possessing firearms, then “the burden
shifts to the Government to demonstrate that the regulation
satisfies” intermediate scrutiny. 

      We apply this framework to determine whether
§ 922(g)(1) as applied to Holloway violates his Second
Amendment rights.

                               B

        At the first step of the analysis, we must determine
whether the application of § 922(g)(1) burdens Holloway’s
Second Amendment rights by considering the traditional
justifications for denying certain criminals Second
Amendment rights and then examining whether Holloway’s
offense is disqualifying. We “presume the judgment of the
legislature is correct and treat any crime subject to § 922(g)(1)
as disqualifying unless there is a strong reason to do
otherwise.” 

       No majority of judges in Binderup agreed on how to
determine whether a particular offense is serious. That said,
we have viewed, albeit in a non-precedential opinion, Judge
Ambro’s factors as providing data points for determining
whether a challenger’s prior conviction was serious, King v.
Att’y Gen. U.S., 783 F. App’x 111, 113-14 (3d Cir. 2019), and
we agree with the dissent that a multifactor test should be used
to identify whether an offense is serious, at least as to
misdemeanor offenses, Dissenting Op. at 6.




                               14
                               1

       As previously stated, Heller embraced the
“longstanding prohibitions on the possession of firearms by

. Because Holloway’s DUI
misdemeanor conviction carries a maximum penalty of five
years’ imprisonment, it is deemed a disqualifying felony under
§ 922(g)(1).    Thus, the application of § 922(g)(1) is
presumptively lawful. See 

(Ambro,
J.).

                               2

       We next examine whether Holloway’s crime was
nonetheless “not serious enough to strip [him] of [his] Second
Amendment rights.” 

Under Binderup, “a person
who did not commit a serious crime retains his Second
Amendment rights,” because “a non-serious crime does not
demonstrate a lack of ‘virtue’ that disqualifies an offender from
exercising those rights.” 

      A crime that presents a potential for danger and risk of
harm to self and others is “serious.”11 See “Serious,” Black’s

       11
           The dissent asserts that our consideration of an
offense’s dangerousness steps too far from Barton. Dissenting
Op. at 16-17. Barton, however, has been overruled in nearly
all respects. Among other things, seven Binderup judges
agreed that Barton “defines too narrowly the traditional
justification for why a criminal conviction may destroy the
right to arms (i.e., it limits felon disarmament to only those
criminals likely to commit a violent crime in the future) and,
by extension, defines too broadly the class of offenders who




                               15
may bring successful as-applied Second Amendment
challenges to § 922(g)(1) (i.e., it allows people convicted of
serious crimes to regain their right to 

n.3 (Ambro, J.). Three other judges would have overruled
Barton entirely. 

n.1. Thus, ten judges rejected the
dissent’s argument that our considerations of who falls within
the historically barred class must be tied to Barton, and in
particular, “the presence of force or violence in the
challenger’s conduct.” Dissenting Op. at 16-17.
        Instead of Barton’s exclusive focus on violence,
Binderup instructs that the Founders sought to permit only the
virtuous citizen to possess a firearm. The historical record tells
us that those who present a risk of danger lack virtue and the
Founders considered danger in evaluating who had the right to
bear arms. See 

-49 (Ambro, J.); 

(Fuentes, J.).
        First, The Address and Reasons of Dissent of the
Minority of the Convention of the States of Pennsylvania to
Their Constituents (the “Address”), “a ‘highly influential’
‘precursor’ to the Second Amendment,” 

(Ambro, J.) (quoting United States v. Skoien, 

, 640 (7th Cir. 2010) (en banc) and 

),
stated “no law shall be passed for disarming the people or any
of them unless for crimes committed, or real danger of public
injury from individuals,” United States v. Bena, 

,
1184 (8th Cir. 2011) (emphasis omitted) (quoting the Address,
reprinted in Bernard Schwartz, 2 The Bill of Rights: A
Documentary History 662, 665 (1971)); see also 

(quoting same passage). While the dissent
proposes a narrow reading of the broad language “real danger
of public injury,” Dissenting Op. at 13-15, we precedentially
interpreted the Address to indicate that the legislature could




                               16
Law Dictionary (11th ed. 2019) (defining “serious” as, among
other things, “dangerous; potentially resulting in death or other
severe consequences”). “There is no question that drunk
driving is a serious and potentially deadly crime . . . . The


historically disarm those “considered dangerous to themselves
and/or to the public at large,” 

. The
dissent’s read is thus foreclosed by our precedent.
       Second, Samuel Adams’ proposed language for the
Second Amendment would have expressly limited the right to
“peaceable citizens.” 

(Hardiman,
J.) (quoting Journal of Convention: Wednesday February 6,
1788, reprinted in Debates and Proceedings in the Convention
of the Commonwealth of Massachusetts Held in the Year 1788,
at 86 (Boston, William White 1856)) (emphasis omitted). In
Adams’ time, “peaceable” meant “free from tumult;” “quiet;
undisturbed;” “[n]ot violent; not bloody;” “[n]ot quarrelsome;
not turbulent.” 1 Samuel Johnson, A Dictionary of the English
Language (5th ed. 1773). Relatedly, “[b]reaches of the peace
comprise[d] not only cases of actual violence to the person of
another, but any unlawful acts, tending to produce an actual
breach of the peace; whether the peace of the public, or an
individual, be in fact disturbed or not.” Pearce v. Atwood, 

, 332 (1816). From these sources, judges have
concluded that “founding-era legislatures categorically
disarmed groups whom they judged to be a threat to the public
safety.” Kanter v. Barr, 

, 458 (7th Cir. 2019)
(Barrett, J., dissenting).      Thus, the Pennsylvania and
Massachusetts proposals show that any right to bear arms did
not extend to those who posed a danger to the public. These
historical sources therefore support considering risk of danger
in determining whether an offense constitutes a serious crime
that deprives an offender of Second Amendment protection.




                               17
imminence of the danger posed by drunk drivers exceeds that
at issue in other types of cases.” Virginia v. Harris, 

, 979-80 (2009) (Mem.) (Roberts, C.J., dissenting from
denial of writ of certiorari); see 

(Sotomayor, J., dissenting) (“[D]runk driving poses significant
dangers that [states] must be able to curb.”); 

(“Drunk driving is an extremely dangerous crime.”).

         All three branches of the federal government have
recognized as much. The Supreme Court has described
individuals “who drive with a BAC significantly above
the . . . limit of 0.08% and recidivists” as “the most dangerous
offenders.” Birchfield v. North Dakota, 

, 2179
(2016). Congress and the Executive Branch have also
recognized the dangers posed by drunk driving. Congress
requires states to implement highway safety programs “to
reduce injuries and deaths resulting from persons driving
motor vehicles while impaired by alcohol.” 23 U.S.C.
§ 402(a)(2)(A)(iii).      The Secretary of Transportation
conditions the receipt of certain highway-related funds on
states’ implementation of programs with impaired driving
countermeasures that will “effective[ly]” “reduce driving
under the influence of alcohol.” § 405(a)(3), (d). Thus, all
branches of the federal government agree that DUIs are
dangerous, and those who present a danger may be disarmed.

       While use or the threatened use of violence is not an
element of a DUI offense, see 75 Pa. Cons. Stat. Ann.
§ 3802(c) (providing “[a]n individual may not drive, operate or
be in actual physical control of the movement of a vehicle after
imbibing a sufficient amount of alcohol such that the alcohol
concentration in the individual’s blood or breath is 0.16% or
higher”), a showing of violence is not necessary for a crime to




                              18
be deemed serious, see, e.g., 

(Ambro, J.); 

(Fuentes, J.); Medina v. Whitaker,

, 160 (D.C. Cir. 2019) (holding that fraud, by
lying on mortgage documents, is “a serious crime”). Thus, the
fact that an offense does not include the use or threatened use
of violence does not mean it is not serious.

       Moreover, though labeled as a first-degree
misdemeanor, Holloway’s DUI crime carries a three-month
mandatory minimum prison term and a five-year maximum
prison term. See 18 Pa. Cons. Stat. Ann. § 1104; 75 Pa. Cons.
Stat. Ann. § 3803(b)(4); 75 Pa. Cons. Stat. Ann. § 3804(c)(2).
While “generally the misdemeanor label . . . in the Second
Amendment context, is . . . important” and is a “powerful
expression” of the state legislature’s view, it is not dispositive.

. First, not only is the distinction
“minor and often arbitrary,” Tennessee v. Garner, 

,
14 (1985); see also Burgess v. United States, 

, 132
(2008), some states do not use the distinction at all, see, e.g.,
N.J. Stat. Ann. § 2C:1-4 (dividing offenses into “crimes,”
“disorderly persons offenses,” and “petty disorderly persons
offenses”); § 2C:43-1(a) (dividing “crimes” further into four
degrees); State v. Doyle, 

, 613 (N.J. 1964)
(“Criminal codes in New Jersey have not utilized the felony-
misdemeanor nomenclature or classification of the English
common law.”). Second, “numerous misdemeanors involve
conduct more dangerous than many felonies.” 

. Indeed, giving dispositive weight to the
felony/misdemeanor nomenclature for determining whether an
offense is serious would mean that the following offenses,
labeled under Pennsylvania law as misdemeanors and carrying
a five-year maximum penalty (the maximum Holloway faced),
18 Pa. Cons. Stat. Ann. § 1104(1), would not qualify as serious




                                19
crimes: involuntary manslaughter, § 2504(b), terrorism,
§ 2717(b)(1), assaulting a child, § 2701(b)(2), abusing a care-
dependent person, § 2713.1(b)(1), making terroristic threats,
§ 2706(d), threatening to use weapons of mass destruction,
§ 2715(b)(1), shooting a fire bomb into public transportation,
§ 2707(a), indecent assault by forcible compulsion,
§ 3126(a)(2), concealing the murder of a child, § 4303(a),
luring a child into a motor vehicle or structure, § 2910(a),
restraining a person “in circumstances exposing him to risk of
serious bodily injury,” § 2902(a)(1), and stalking,
§ 2709.1(c)(1). At bottom, Heller emphasized that the Second
Amendment right belongs to “law-abiding, responsible

, and whether labeled a felon or
misdemeanant, those who commit serious crimes are not “the
kinds of ‘law-abiding’ citizens whose rights Heller
vindicated,” 

(Fuentes, J.).

      Furthermore, the maximum penalty that may be
imposed often reveals how the legislature views an offense.12

      12
          In addition to ascribing high value to the offense’s
felony/misdemeanor label, the dissent favors focusing on the
actual penalty imposed. While the penalty imposed may
provide some insight into how a sentencing judge may have
viewed an offender, it does not necessarily reflect how the
offense itself is viewed. Binderup step one focuses on the
offense and not the offender. 

(Ambro,
J.); 

(Fuentes, J.). Because the actual sentence
imposed can be influenced by many factors, such as
cooperation, U.S.S.G. § 5K1.1, acceptance of responsibility,
U.S.S.G. § 3E1.1, and offender-related variances, 18 U.S.C.
§ 3553, the actual penalty imposed does not necessarily show
that the crime was not “serious.” Instead, the maximum




                              20
Put succinctly, “the maximum possible punishment is certainly
probative of a misdemeanor’s seriousness.” 

(Ambro,
J.).13 “[T]he category of serious crimes changes over time as
legislative judgments regarding virtue evolve,” 

and
here, the Pennsylvania legislature has demonstrated an
evolution in judgment. Pennsylvania’s DUI laws were


punishment is a more appropriate data point because it
provides insight into how a state legislature views a crime—
not how a sentencing judge views an individual. See Lewis v.
United States, 

, 325-26 (1996) (noting that an
offense’s penalty “reveals the legislature’s judgment about the
offense’s severity”); 

(noting that the maximum
punishment is an “objective indication of the seriousness with
which society regards the offense”); 

. For these reasons, it is proper to consider the
maximum penalty an offender faces, and not simply the actual
punishment imposed or whether the offense is designated as a
misdemeanor or felony, to determine whether an offense is
properly viewed as “serious.”
       13
          The dissent is mistaken to say that a majority in
Binderup rejected consideration of a maximum penalty in
favor of the felony/misdemeanor label. Judge Ambro’s
opinion for three judges reasoned that “the maximum possible
punishment is certainly probative of a misdemeanor’s
seriousness” under the first 

(Ambro,
J.). Seven judges stated that any crime which qualifies for
§ 922(g)(1) is serious. 

(Fuentes, J.). That means that
those seven judges would conclude that the penalty Holloway
faced shows his offense is serious regardless of its
misdemeanor classification. Combining the views of Judge
Ambro’s and Judge Fuentes’ opinions, a majority of the
Binderup court rejected the dissent’s view.




                               21
amended in 2003 when state legislators observed that “[t]oo
many people have been injured and killed on our highways,”
H.R. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1443 (Pa.
2003) (statement of Rep. Turzai), and unlike in other states,
which saw an eleven percent decrease in deaths caused by
drunk drivers, such deaths “continue to rise” in Pennsylvania
with a five percent increase, H.R. Legis. Journal, 187th Gen.
Assemb., Reg. Sess. 1444 (Pa. 2003) (statement of Rep.
Harper); S. Legis. Journal, 187th Gen. Assemb., Reg. Sess. 981
(Pa. 2003) (statement of Sen. Williams). At the time of the
amendment, thirteen individuals were killed every two weeks
in Pennsylvania from alcohol-related accidents. H.R. Legis.
Journal, 187th Gen. Assemb., Reg. Sess. 1445 (Pa. 2003)
(statement of Rep. Harper). “[M]ore than half of all fatal
alcohol-related accidents [were] caused by hardcore drunken
drivers, those people whose BACs are .16 or above,” H.R.
Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1444 (Pa.
2003) (statement of Rep. Harper), and “one-third of drunk
driving arrests involve[d] repeat offenders,” S. Legis. Journal,
187th Gen. Assemb., Reg. Sess. 981 (Pa. 2003) (statement of
Sen. Williams). To address this “very serious matter,” H.R.
Legis. Journal, 187th Gen. Assemb., Reg. Sess.1445 (Pa. 2003)
(statement of Rep. Harper), the legislature “provid[ed] for
tough civil and criminal penalties together with mandatory
treatment,” H.R. Legis. Journal, 187th Gen. Assemb., Reg.
Sess. 1443 (Pa. 2003) (statement of Rep. Turzai), to “mak[e] it
clear that if you are under the influence of alcohol or drugs and
behind the wheel in Pennsylvania, you will be punished,” H.R.
Legis. Journal, 187th Gen. Assemb., Reg. Sess. 1445 (Pa.
2003) (statement of Rep. Harper). Therefore, despite the
misdemeanor label, Pennsylvania’s decision to impose a
mandatory minimum jail term and a maximum penalty of up to




                               22
five years’ imprisonment for a second DUI at the highest BAC
reflects the seriousness of the offense.14

       Holloway received the statutory minimum sentence of
90 days’ imprisonment, 75 Pa. Cons. Stat. Ann. § 3804(c)(2),
and although he was permitted to work, he received a custodial
sentence unlike either of the challengers in 

(“With not a single day of jail time, the punishments
here reflect the sentencing judges’ assessment of how minor
the violations were.”). The legislature’s mandate that repeat
DUI offenders receive at least three months in jail reflects its
judgment that such offenses are serious.


       14
          As one district court analyzing an as-applied
challenge under Binderup aptly observed,

       juxtaposing the Pennsylvania legislature’s use of
       the misdemeanor label with the legislature’s
       simultaneous imposition of a substantial
       imprisonment term creates an inherent
       contradiction: a five-year maximum prison term
       suggests that [the plaintiff’s] predicate offense is
       serious, while the misdemeanor label
       simultaneously undercuts the apparent severity
       by labeling the offense a non-serious.

Laudenslager v. Sessions, 4:17-CV-00330, 

,
at *4 (M.D. Pa. Feb. 13, 2019) (discussing the classification
and maximum sentence for receiving stolen property under
Pennsylvania law). We agree, and for the reasons described
above, conclude that the legislative history elucidates this
contradiction.




                               23
       Pennsylvania is not alone in its decision to severely
punish repeat DUI offenders. 

(“[M]any States . . . have passed laws imposing increased
penalties for recidivists or for drivers with a BAC level that
exceeds a higher threshold.” (citations omitted)). Although
most states do not impose penalties for second DUI offenses
that subject an offender to disarmament under § 922(g)(1),
three states impose penalties that subject misdemeanants who
commit a second DUI at a higher BAC to § 922(g)(1)
disarmament. Moreover, several states grade a second DUI
offense as a felony, thus triggering disarmament. The absence
of a cross-jurisdictional consensus regarding the punishment
for such conduct does not mean the conduct is not serious.
Indeed, states unanimously agree that DUIs are crimes subject
to punishment.

        Holloway suggests that his crimes cannot be so serious
to justify federal disarmament and that to apply § 922(g)(1) to
him would be overinclusive because Pennsylvania law only
disarms DUI offenders at their third offense and permits them
to apply for relief after ten years. This argument ignores the
gradations in Pennsylvania’s DUI laws.                In fact,
Pennsylvania’s prohibition may be broader than § 922(g)(1)
because it applies to all DUIs under 75 Pa. Cons. Stat. Ann.
§ 3802, regardless of punishment. For example, an individual
who commits a third DUI, none at the high or highest BAC,
within a five-year period, is convicted of a second-degree
misdemeanor under 75 Pa. Cons. Stat. Ann. § 3803(a)(2) and
subject to up to two years’ imprisonment under 18 Pa. Cons.
Stat. Ann. § 1104(2). This individual’s third DUI triggers
Pennsylvania’s disarmament statute under 18 Pa. Cons. Stat.
Ann. § 6105(c), but does not trigger § 922(g)(1) because it falls
within § 921(a)(20)(B)’s exception for state misdemeanors




                               24
subject to a term of imprisonment of two years or less.
Holloway’s second DUI, however, subjects him to the federal
provision but not the state provision because his offense was at
the highest BAC, which enhanced the grading of his offense to
a first-degree misdemeanor and exposed him to five years’
imprisonment. Thus, Pennsylvania’s disarmament statute
captures offenders who may not face § 922(g)(1)’s bar and
shows that Pennsylvania meant to disarm a broader swath of
offenders than § 922(g)(1).

       Together, these considerations demonstrate that
Holloway’s DUI conviction constitutes a serious crime,
placing him within the class of “persons historically excluded
from Second Amendment protections.” 

. Because Holloway has not met his burden at the first step
of the analysis to overcome the presumptive application of
§ 922(g)(1),15 § 922(g)(1) is constitutional as applied to him,
and he is not entitled to relief.16

       15
          At the first step of our framework, we do not consider
Holloway’s arguments that he has not committed any offenses
since 2005 or the letters he offered in support of his character
because “[t]here is no historical support for the view that the
passage of time or evidence of rehabilitation can restore
Second Amendment rights that were forfeited.” 

, 354 n.7.
       16
          Because Holloway has not carried his burden at step
one to show he was not convicted of a serious offense, we need
not move on to step two to determine whether the statute as
applied to him survives intermediate scrutiny. We do note,
however, that our precedent is cautious in applying the
intermediate scrutiny test used in First Amendment cases.
Compare N.J. 

n.28 (stating that we do




                              25
                              III

       For the foregoing reasons, we will reverse the order
granting Holloway summary judgment, a declaratory
judgment, and an injunction and remand for the entry of
judgment in favor of the Government.




not incorporate “wholesale” First Amendment jurisprudence
when evaluating Second Amendment challenges), with
Dissenting Op. at 26 (advocating that we import the Supreme
Court’s test for commercial speech cases for Second
Amendment challenges to § 922(g)(1)). In addition, the
dissent’s application of intermediate scrutiny seemingly asks
for a near-perfect fit between the challenged regulation and the
objective, rather than a “reasonable” fit. 

(stating that the “fit between the challenged regulation
and the asserted objective be reasonable, not perfect”).




                              26
FISHER, Circuit Judge, dissenting.
       Driving under the influence of alcohol is undoubtedly a
significant offense deserving of punishment. Yet the principal
question in this case is not whether that offense is “serious” in
the abstract or even as a matter of ordinary understanding.
“Seriousness” here has a discrete legal meaning—that a
conviction of the crime deprives in perpetuity an individual of
an enumerated constitutional right. Under our precedent, these
two categories are distinct, and they must be treated as such.
Just because this question arises under the Second Amendment
does not make our decision any less weighty. If the
circumstances were different, we would assuredly consider
very carefully the legal standard for depriving an individual of
his right to free speech. The majority incorrectly, in my view,
holds that Holloway has not carried his burden at Step One of
the two-step framework established in United States v.
Marzzarella, 

(3d Cir. 2010). Further, because I
conclude that at Step Two, 18 U.S.C. § 922(g)(1) as applied
here does not survive intermediate scrutiny, I must respectfully
dissent.
                                I
        Under the Marzzarella framework, we first determine
“whether the challenged law imposes a burden on conduct
falling within the scope of the Second Amendment’s

. In particular, our precedent requires
the challenger to satisfy the two elements articulated in United
States v. Barton, 

(3d Cir. 2011). He must “identify
the traditional justifications for excluding from Second
Amendment protections the class of which he appears to be a
member,” and then “present facts about himself and his
background that distinguish his circumstances from those of
persons in the historically barred class.” Binderup v. Attorney




                               1
Gen. United States of America, 

, 347 (3d Cir.
2016) (en banc) (plurality opinion) (citing 

); see 

(Hardiman, J., concurring in part and
concurring in the judgments); see also Beers v. Attorney Gen.
United States of America, 

, 157 (3d Cir. 2019)
(adopting this test for an as-applied challenge to 18 U.S.C. §
922(g)(4)).
        In Binderup, ten judges on the fifteen-member en banc
court agreed that, in the context of as-applied challenges to 18
U.S.C. § 922(g)(1), the “historically barred class” is those who
are “unvirtuous” because they have “committed a serious
criminal offense, violent or 

(plurality opinion); see 

(Fuentes, J., concurring in
part, dissenting in part, and dissenting from the judgments); see
also United States v. Donovan, 

, 182 (3d Cir.
2011) (noting that when this Circuit confronts a fractured
decision, we “look[] to the votes of dissenting [judges] if they,
combined with votes from plurality or concurring opinions,
establish a majority view on the relevant issue”).
“Seriousness”—and by extension “unvirtuousness”—
therefore has no independent legal significance. It is a gloss on
the first part of the Barton test—a way of describing the
offenses committed by those historically barred from
possessing a firearm.1

1
  The majority suggests that any discussion of Barton is
misplaced because that decision “has been overruled in nearly
all respects.” Majority Op. at II.B.2 n.11. Yet, even if that is
true, my emphasis throughout this opinion is on a key respect
in which it has not been overruled: that a challenger to the
application of § 922(g)(1) must distinguish his circumstances
from those of the historically barred class. The majority
acknowledges that we must still conduct such an analysis. See




                               2
        The principal question before us today concerns the
application of Barton’s second prong in the § 922(g)(1)
context—that is, how to evaluate whether a challenger’s crime
is sufficiently similar to crimes of the historically barred class
such that he is not entitled to Second Amendment protection.
The Binderup Court divided on this issue, and, for the reasons
detailed below, it remains an open question whether the
multifactor test used in Binderup is binding precedent in our
Circuit—despite the lower courts’ application of it as such. See,
e.g., Williams v. Barr, 

, 370-74 (E.D. Pa.
2019); Holloway v. Sessions, 

, 457-60
(M.D. Pa. 2018). Nevertheless, for reasons I also state, the test
is an appropriate means under our precedent of determining
whether a challenger’s crime is “serious” for purposes of
Marzzarella Step One.
       It is on this latter point—the application of the
multifactor test—that I break with my colleagues in the
majority. They interpret the test’s list of factors to be non-
exhaustive, Majority Op. at II.A, and so they supplement their
analysis of the factors with additional considerations. The
majority appears to concede that at least three of the four
Binderup factors are in Holloway’s favor, but still concludes
that Holloway is not entitled to Second Amendment protection.
Although I agree that we are not bound to consider the four
factors exclusively, I disagree with my colleagues in how they
have applied and supplemented those factors. Simply because
our precedent does not require us to apply the four factors alone
does not mean the determination of “seriousness” is open to

& n.4; see also 

-47
(plurality opinion) (“At step one of the Marzzarella decision
tree, a challenger must prove, per Barton, that a presumptively
lawful regulation burdens his Second Amendment rights.”).




                                3
any legal content. Our precedent does require us to follow the
doctrinal structure established in Barton and adopted in
Binderup. The “seriousness” inquiry is a comparison of the
challenger’s circumstances with those of the historically barred
class. The majority’s analysis, in my view, diverges too far
from this requirement.
                               A
        As it was applied in Binderup, the multifactor test
contains four factors for determining whether an individual’s
crime is sufficiently “serious” to deprive him of his Second
Amendment right. First, the court considers whether the state
classifies the challenger’s disqualifying crime under §
922(g)(1) as a felony or a 

(opinion of Ambro, J.). Second, it determines whether the
challenger’s crime “had the use or attempted use of force as an
element.” 

. Third, also relevant is the sentence the
challenger in fact received. Although the maximum possible
sentence determines whether the crime triggers the § 922(g)(1)
bar, the crime’s “seriousness” for purposes of Second
Amendment analysis turns, in part, on the challenger’s actual
punishment. Finally, the court considers whether there exists a
“cross-jurisdictional consensus regarding the seriousness of
the [challenger’s] crimes.” 

multifactor test
garnered the support of only three judges, it was declared “the
law of our Circuit” under the Supreme Court’s Marks rule. 

      My review of our case law leads me to question this
conclusion. Courts and legal scholars disagree as to the nature




                               4
of the Marks rule and how it is to be applied.2 In particular,
there are multiple possible versions of the rule, and the
Supreme Court’s most recent statement on the matter
acknowledged but declined to resolve this debate. See Hughes
v. United States, 

, 1771-72 (2018). On my
assessment, the multifactor test would be Circuit precedent
under only one of these versions,3 and our Court has not
adopted this interpretation of the Marks rule above the others.4

2
  See Richard M. Re, Beyond the Marks Rule, 132 HARV. L.
REV. 1942, 1947-65 (2019) (providing a helpful survey of the
Marks debate).
3
  This version holds that the concurring opinion representing
the views of the median judge constitutes binding precedent.
See 

(citing MAXWELL L. STEARNS,
CONSTITUTIONAL PROCESS: A SOCIAL CHOICE ANALYSIS OF
SUPREME COURT DECISION MAKING (2000)).
4
  In fact, we have occasionally endorsed a different version of
the rule, which construes it to apply only to those views in an
opinion concurring in the judgment that constitute a logical
subset of broader views expressed in another concurrence in
the judgment. See, e.g., B.H. ex rel. Hawk v. Easton Area Sch.
Dist., 

, 310-13 (3d Cir. 2013) (en banc); Jackson
v. Danberg, 

, 222 (3d Cir. 2010) (“[T]he Marks
framework applies where one opinion is clearly ‘narrower’
than another, that is, where one opinion would always lead to
the same result that a broader opinion would reach.”); Planned
Parenthood of Se. Pa. v. Casey, 

, 693-94 (3d Cir.
1991), aff’d in part and rev’d in part on other grounds, 

(1992); see also 

(explaining
this version of the Marks rule). Under this version of the rule,
the multifactor test would have to constitute a logical subset of
the views expressed in Judge Hardiman’s opinion, which was




                               5
As a result, despite the declaration in Binderup to the contrary,
I do not think Marks requires us to treat the multifactor test as
controlling authority.5
                                B
       Nevertheless, like the District Court, I believe that the
multifactor test should guide the Step One analysis in this case.
On my reading, the four factors reflect an underlying logic that
is consistent with our precedent in Barton and Binderup. Those
cases require us to assess the relation between the challenger’s
“circumstances [and] those of persons historically barred from
Second Amendment protections.” 

; see
also 

-47 (plurality opinion); 

(Hardiman, J., concurring in part and concurring in the
judgments). This comparative exercise demands certain
measures of “seriousness,” and those measures should
naturally be the features—the classification, elements, and
punishments—common to the crimes that traditionally have
qualified the individuals convicted of them for firearm
dispossession. These crimes include felonies, crimes of

the other concurring opinion in Binderup. It is difficult to see
how this is the case.
5
  Nor has any subsequent precedential opinion of this Court
resolved this difficulty by adopting that test. Only three of this
Court’s precedential opinions cite Binderup. None concerns an
as-applied challenge to § 922(g)(1). See Beers, 

;
United States v. Greenspan, 

(3d Cir. 2019); Ass’n
of N.J. Rifle & Pistol Clubs v. Attorney Gen. New Jersey, 

(3d Cir. 2018). However, one recent non-precedential
opinion confronting an as-applied challenge to § 922(g)(1) has
declared the multifactor test controlling authority. See King v.
Attorney Gen. of the United States, No. 18-2571, 

, at *2 & n.2 (3d Cir. July 25, 2019).




                                6
violence, and (as Binderup held)6 some nonviolent
misdemeanors. Further, because neither courts nor scholars
have agreed on the precise contours of this category—and in
particular how “longstanding” a regulation must be for its
violators to be considered part of the historically barred class,
see, e.g., United States v. Skoien, 

, 640-41 (7th Cir.
2010) (en banc); C. Kevin Marshall, Why Can’t Martha
Stewart Have a Gun?, 32 HARV. J.L. & PUB. POL’Y 695
(2009)—the multifactor test has the virtue of permitting a
number of different measures of “seriousness” without making
any one factor dispositive.
       A methodical evaluation of each factor, consistent with
this logic, compels the conclusion reached by the District
Court: that Holloway’s conduct has not removed him from the
scope of Second Amendment protection. In conducting this
analysis, I shall also address the majority’s additional
considerations—the “potential for danger and risk of harm”
posed by the challenger’s crime, Majority Op. at II.B.2, and the
maximum level of punishment Pennsylvania imposes for
Holloway’s second DUI offense, 

noted, I do not
dispute that the majority may supplement the four factors, any
such additions must be—as the four factors are—consistent
with the comparative exercise required by Barton and
Binderup.7

6
  

–49 (plurality opinion); 

–88
(Fuentes, J., concurring in part, dissenting in part, and
dissenting from the judgments)
7
  According to the majority, I argue that “our considerations of
who falls within the historically barred class must be tied to
Barton, and in particular ‘the presence of force or violence in
the challenger’s conduct,’” Majority Op. at II.B.2 n.11. Yet that
is not my argument. At multiple points in this opinion I note




                                7
                                1
       The first factor asks whether the challenger’s crime is a
felony or a misdemeanor. The majority acknowledges that
Pennsylvania classifies Holloway’s second DUI offense as a
misdemeanor, but it points out that the offense “carries . . . a
five-year maximum prison term.” Majority Op. at II.B.2. Yet,
under our precedent, the potential prison term cannot nullify
the relevance of the felony/misdemeanor distinction for
determining whether a crime is “serious” enough to deprive an
individual of his Second Amendment right. A common feature
of the crimes that traditionally have barred an individual from
owning a firearm is that they are classified as felonies.
       For example, in Heller, the Supreme Court warned
specifically that its opinion should not be read to question
“longstanding prohibitions on the possession of firearms by
felons.” District of Columbia v. Heller, 

, 626
(2008) (emphasis added); see also McDonald v. City of
Chicago, 

, 786 (2010) (plurality opinion).

that because of the indefinite nature of the historically barred
class, no one factor can be dispositive. I assert, rather, that the
relevant factors may not be any ones we choose—they must
aid the determination of whether the challenger’s crime is
sufficiently similar to those of the persons historically barred
from firearm possession. This certainly involves historical
analysis (which the majority also engages in), but, as I
mentioned above and restate below, it additionally includes
looking to other measures relevant to making the comparison.
My point, as I go on to detail, is that the majority has given too
much weight to considerations that, however compelling in
other contexts, are irrelevant to the comparative analysis that
the majority itself acknowledges we must conduct. See 

& n.4.




                                8
Congress itself recognized the relevance of the distinction
when it excluded from § 922(g)(1)’s reach misdemeanors
punishable by imprisonment of two years or less. See 18 U.S.C.
§ 921(a)(20)(B). If, as the majority suggests, the maximum
length of the sentence rather than the classification of the crime
is what really matters, then Congress would never have made
an exception for misdemeanors alone. It would either have
amended § 922(g)(1) to cover all crimes punishable by more
than two years’ incarceration or never added § 921(a)(20)(B)
in the first place.8

8
  To the extent one gives it validity, the legislative history
confirms this interpretation. In 1961, Congress amended the
precursor of § 922(g)(1) to prevent the transportation or receipt
of a firearm by all persons convicted of any “crime punishable
by imprisonment for a term exceeding one year”—not just
persons convicted of a “crime of violence,” as had previously
been the case. See Act of Oct. 3, 1961, Pub. L. No. 87-342, 75
Stat. 757, 757 (1961). In the Gun Control Act of 1968,
however, Congress amended §§ 921 and 922 to their present
form. See Pub. L. No. 90-618, 82 Stat. 1213 (1968). The House
bill would have maintained the existing broad language
covering all crimes—both felonies and misdemeanors—
punishable by more than one year of imprisonment. See H.R.
17735, 90th Cong. § 2 (1968). By contrast, the Senate bill
would have made it “unlawful for any person . . . convicted in
any court of a crime punishable as a felony” to transport or
receive any firearm. S. 3633, 90th Cong. § 102 (1968). The
Conference Report noted this discrepancy, declaring the
compromise to be the maintenance of the House language in §
922(g)(1), but adding what became § 921(a)(20)(B). See H.R.
REP. NO. 90-1956, at 28-29 (1968) (Conf. Rep.). Thus, in
creating our current regime, Congress not only wanted to




                                9
       Further, the classification of a crime as a felony has
profound implications for whether a person may possess a
firearm under state law. On my assessment, thirty-two out of
fifty-one jurisdictions (the fifty states and the District of
Columbia) disarm individuals because of a felony conviction.9
That is, they bar for at least some time the possession of a


include misdemeanors as well as felonies in the reach of the
law, but also drew a distinction between the two types of
crimes.
9
  See ALASKA STAT. § 11.61.200(a)(1) (2019); ARIZ. REV.
STAT. ANN. § 13-904(A)(5) (2019); ARK. CODE ANN. § 5-73-
103(a)(1) (2019); CAL. PENAL CODE § 29800(a)(1) (West
2019); COLO. REV. STAT. § 18-12-108(1) (2019); CONN. GEN.
STAT. § 53a-217(a)(1) (2019); DEL. CODE ANN. tit. 11, §
1448(a)(1) (2019); D.C. CODE § 7-2502.03(a)(2) (2019); FLA.
STAT. § 790.23(1) (2019); HAW. REV. STAT. § 134-7(b) (2019);
720 ILL. COMP. STAT. 5/24-1.1(a) (2019); IND. CODE § 35-47-
2-3(h) (2019); IOWA CODE § 724.26(1) (2019); KAN. STAT.
ANN. § 21-6304(a) (2019); KY. REV. STAT. ANN. § 527.040(1)
(West 2019); MD. CODE ANN., CRIM. LAW § 5-622(b) (West
2019); MASS. GEN. LAWS ch. 140, § 131(d)(i)(A) (2019);
MICH. COMP. LAWS § 750.224f(1) (2019); MISS. CODE ANN. §
97-37-5(1) (2019); MO. REV. STAT. § 571.070.1(1) (2019);
NEB. REV. STAT. § 28-1206(1)(a)(i), (2) (2019); NEV. REV.
STAT. § 202.360(1)(b) (2019); N.H. REV. STAT. ANN. § 159:3
(2019); N.M. STAT. ANN. § 30-7-16(A)(1) (2019); N.Y. PENAL
LAW § 400.00(1)(c) (McKinney 2019); N.C. GEN. STAT. § 14-
415.1(a) (2019); OKLA. STAT. tit. 21, § 1283(A) (2019); OR.
REV. STAT. § 166.270 (2019); TEX. PENAL CODE ANN. §
46.04(a) (West 2019); VA. CODE ANN. § 18.2-308.2(A) (2019);
WASH. REV. CODE § 9.41.040(2)(a)(i) (2019); WIS. STAT. §
941.29(1m) (2019).




                             10
firearm precisely because the person was convicted of a crime
labeled a felony. The distinction therefore matters for defining
the historically barred class, regardless of jurisdictional
diversity in the sentence ranges for various crimes.
       As noted, in evaluating the relevance of the
felony/misdemeanor distinction, the majority lends great
weight to the maximum punishment that Pennsylvania imposes
for Holloway’s offense. See Majority Op. at II.B.2. However,
a majority of the en banc Court in Binderup rejected the
significance of that consideration. As Judge Ambro noted
there, prohibitions on the possession of firearms by criminals
are only “presumptively lawful.” 

(opinion of Ambro, J.) (emphasis added) (citing 

& n.26), and in the absence of an explicit
declaration to the contrary, all presumptions are rebuttable. To
hold otherwise would constitute “an end-run around the
Second Amendment,” in effect subjecting such prohibitions to
rational-basis review rather than the heightened scrutiny
demanded when a constitutional right is at stake. 

As a result, the maximum possible sentence for Holloway’s
crime, although a valid consideration, cannot detract from the
relevance of a factor that is consistent with our precedent in
Barton and Binderup.10

10
   I do not, as the majority suggests, read Binderup as
“reject[ing] consideration of a maximum penalty in favor of
the felony/misdemeanor label.” Majority Op. at II.B.2 n.13.
Rather, my point is that the majority cannot invoke the
maximum penalty to discount the relevance of a factor
consistent with the comparative exercise Barton and Binderup
require us to conduct. The dissent in Binderup would have held
the challengers’ crimes “serious” simply because they carry
maximum prison terms exceeding those provided in §§




                              11
        In saying this, I do not question the Pennsylvania
legislature’s judgment that an offense such as Holloway’s
should be punishable by a lengthy prison term. But for the
purposes of answering the question before us today—whether
that offense is “serious” enough to deprive Holloway of his
Second Amendment right—we must look to how his offense
compares with those of the historically barred class. That
involves giving weight to the felony/misdemeanor distinction.
In addition to the sentence it permitted, the Pennsylvania
legislature also chose to punish Holloway’s crime as a
misdemeanor. Indeed, the sentence and the classification are
inseparable—all such misdemeanors in Pennsylvania carry
Holloway’s maximum possible prison term. See 18 PA. CONS.
STAT. § 1104(1) (2019). Even as a simple matter of statutory
interpretation, then, the classification of the crime matters. This
factor therefore weighs in Holloway’s favor.




921(a)(20)(B) and 922(g)(1). 

(Fuentes, J.,
concurring in part, dissenting in part, and dissenting from the
judgments). A majority of the judges rejected such a
categorical approach—and that is a key reason why Binderup
came out as it did. The maximum penalty and the
felony/misdemeanor distinction cannot, therefore, be treated as
mutually exclusive. For this same reason, I agree with the
majority that the maximum punishment is probative of the
offense’s “seriousness.” See Majority Op. at II.B.2. But I think
that fact should be considered under the fourth factor—how
United States jurisdictions generally punish the offense. It is
important, for purposes of the Barton and Binderup
comparison, whether the challenger’s maximum punishment
reflects a jurisdictional consensus or is an outlier.




                                12
                                2
        The second factor asks whether the “[c]hallenger’s
offense had the use or attempted use of force as an element.”

(opinion of Ambro, J.). The majority
concedes that Holloway’s DUI offense does not fulfill this
criterion, see Majority Op. at II.B.2, but it supplements its
analysis by considering the crime’s “potential for danger and
risk of harm to self and others,” 

Marks rule
does not foreclose additions to the multifactor test by a panel
majority, our precedent demands that the “seriousness” inquiry
be a comparative exercise involving the challenger’s offense
and the characteristic features of those crimes that traditionally
have disqualified persons from owning firearms. The virtue of
the second Binderup factor is that it crystallizes in a clear legal
standard the evident historical concern with force and violence.
By contrast, the relevant historical and contemporary
authorities do not support a standard focusing on all conduct
that poses a “potential for danger and risk of harm to self and
others.” 

prominent late eighteenth-century sources
supporting legislative power to bar certain individuals from
owning firearms are the proposals made in the ratifying
conventions of Pennsylvania, New Hampshire, and
Massachusetts. The first of these provides that “no law shall be
passed for disarming the people or any of them, unless for
crimes committed, or real danger of public injury from
individuals.” THE ADDRESS AND REASONS OF DISSENT OF THE
MINORITY OF THE CONVENTION, OF THE STATE OF
PENNSYLVANIA, TO THEIR CONSTITUENTS 1 (Phila., E. Oswald
1787), https://www.loc.gov/item/90898134. It is important to
note that the two categories are interlocking—the provision
captures both convicted criminals and those non-criminals who
pose a “real danger of public injury.” 

of the




                                13
latter phrase in turn suggests that the drafters did not
necessarily have in mind all crimes, but rather those that
manifest a real danger to the public. To this extent, I agree with
the majority’s reading of the text. See Majority Op. at II.B.2
n.11.
        Yet the provision alone does not tell us what “real
danger of public injury” means. Perhaps the best way of
interpreting this historical term is to look to the dispossessory
provisions proposed at the other two conventions. In voting to
ratify the Federal Constitution, New Hampshire’s delegates
also recommended certain amendments to it. Among these was
a provision that “Congress shall never disarm any citizen,
unless such as are or have been in actual rebellion.” 1
JONATHAN ELLIOT, THE DEBATES IN THE SEVERAL STATE
CONVENTIONS ON THE ADOPTION OF THE FEDERAL
CONSTITUTION 326 (2d ed. 1836). Although the Pennsylvania
minority’s “real danger of public injury” was likely meant to
sweep more broadly than New Hampshire’s “in actual
rebellion,” insofar as we are attempting to discover the
limitations the ratifying public would have implicitly placed on
the Second Amendment, the New Hampshire provision
suggests a concern with armed conflict or violence against the
government, rather than with all dangerous acts. In this
context, it is noteworthy that the Pennsylvania minority speaks
of the danger of public, rather than private, injury—a
distinction it explicitly makes elsewhere in the document. See,
e.g., ADDRESS AND REASONS OF DISSENT, at 3 (“The absolute
unqualified command that congress have over the militia may
be made instrumental to the destruction of all liberty, both
public and private . . . .”). From this perspective, it appears the
Pennsylvania antifederalists had in mind something narrower




                                14
than the majority’s standard of “risk of harm to self and
others.”11

11
   The majority does not discuss the New Hampshire proposal.
Nevertheless, it declares this reading of the Pennsylvania
minority’s Address “foreclosed by our precedent” in Beers.
Majority Op. at II.B.2 n.11. It is unclear, though, how Beers’s
interpretation constitutes binding precedent. Beers used the
phrase “real danger of public injury” to hold in part “that the
traditional justification for disarming mentally ill individuals
was that they were considered dangerous to themselves and/or
to the public at 

. By its very terms, this
holding applies to the mentally ill, not to those convicted of
crimes. To the extent Beers found the phrase to apply to all
persons who present a danger to themselves or the public at
large, that finding is dicta. Alternatively, if an interpretation of
“real danger of public injury” can apply precedentially beyond
the context in which it is invoked, then Beers was in fact bound
by Barton’s interpretation, which found the phrase to cover
“those who were likely to commit violent offenses.” 

. It cannot plausibly be argued that Binderup
overruled this aspect of Barton, since the Binderup plurality
opinion emphasized the phrase “crimes committed,” which
precedes “real danger of public injury” in the Address, and
suggested that it was the operative language covering
nonviolent offenses. 

(plurality opinion).
Further, the plurality opinion explicitly stated that it was
overruling Barton “[t]o the extent” that Barton “holds that
people convicted of serious crimes may regain their lost
Second Amendment rights after not posing a threat to society
for a period of time.” 

On any reading, then, the
majority is incorrect to suggest that Beers requires us to
interpret “real danger of public injury” as it does.




                                15
        This understanding is also found in Samuel Adams’s
proposal to the Massachusetts ratifying convention. The
Constitution, he suggested, should never be “construed to
authorize Congress . . . to prevent the people of the United
States who are peaceable citizens from keeping their own
arms.” 3 WILLIAM V. WELLS, THE LIFE AND PUBLIC SERVICES
OF SAMUEL ADAMS 267 (Bos., Little, Brown & Co. 1865).
What Adams meant by “peaceable” can be determined from
the rest of his proposal. He also thought the Constitution should
not be construed “to prevent the people from petitioning, in a
peaceable and orderly manner, the Federal Legislature for a
redress of grievances.” 

to keep arms was linked
to the assembly and petitioning right not only in Adams’s
proposal but also in the Bill of Rights itself. To many late-
eighteenth-century Americans, the arms right in the Second
Amendment helped to ensure that the liberties guaranteed in
the First Amendment would not be eroded by a tyrannical
central government. See AKHIL REED AMAR, THE BILL OF
RIGHTS: CREATION AND RECONSTRUCTION 47-48 (1998).
Thus, in both Adams’s proposal and the Bill of Rights, it is “the
people” who are given the right to petition their government
and to possess arms. That entity, of course, is the one that (as
the Preamble declares) alone has the power to form the
government, and concomitantly to alter or abolish it. In this
context, “peaceable” refers to those individuals who remain a
part of “the people,” and do not independently disturb or take
up arms against its legitimate government. Only “the people”
itself has that ability.
        In sum, the principal historical evidence from the
Founding period suggests that the majority’s “risk of harm”
standard is too broad to serve as a basis for comparison under
our precedent. The correct standard appears to be something
closer to the one used in Binderup, focusing on the presence of




                               16
force or violence in the challenger’s conduct. Notably, in a part
of Barton that remains good law, our Court summarized the
ratifying convention proposals as “confirm[ing] that the
common law right to keep and bear arms did not extend to
those who were likely to commit violent 

(emphasis added); see also Kanter v. Barr, 

,
456 (7th Cir. 2019) (Barrett, J., dissenting) (concluding that
“[t]he concern common to all three” proposals is “threatened
violence and the risk of public injury”).12
       Further, although the majority cites contemporary
authorities to support its standard, these seem to me inapt for
conducting the comparison required by Barton and Binderup.
On my reading, the majority principally relies on an inference
from a colloquial understanding of drunk driving’s

12
   Additional historical evidence from after the Founding
further undercuts the majority’s position. For one, scholars
have found little evidence of categorical bans on firearm
possession in the nineteenth century. The principal means of
gun control in this period appear to have been public-carry
laws. See Saul Cornell, The Right to Keep and Carry Arms in
Anglo-American Law: Preserving Liberty and Keeping the
Peace, 80 LAW & CONTEMP. PROBS. 11, 33-43 (2017);

. In addition, although firearm
dispossession laws became increasingly prevalent in the early
twentieth century, even these foundational statutes cannot
support the majority’s standard. For example, the original
version of § 922(g)(1) made it unlawful for any person
“convicted of a crime of violence” to transport or receive a
firearm. Federal Firearms Act, Pub. L. No. 75-785, § 2, 52 Stat.
1250, 1251 (1938) (codified at 15 U.S.C. § 902 (1940)). On the
background to the Federal Firearms Act’s “crime of violence”
provision, see 

.




                               17
“seriousness” to that offense’s “seriousness” for purposes of
depriving a person of Second Amendment protection. See
Majority Op. at II.B.2. This is a category mistake. If we
conducted a poll of a representative sample of Americans,
asking them whether drunk driving is a serious crime, it is
likely that most would answer affirmatively. Such an appeal to
ordinary meaning has legal purchase in the context of statutory
interpretation because a court there confronts words as adopted
by a procedurally established majority of the people’s elected
representatives. But “serious” for present purposes is not a
statutory, let alone a constitutional, term. It is how a majority
of this Circuit’s judges in Binderup summarized the crimes that
historically have deprived persons convicted of them of the
right to own a firearm. “Serious,” therefore, has a discrete legal
meaning, and the “seriousness” inquiry must be given content
consistent with that meaning. It is a determination of whether
a challenger’s offense is sufficiently similar to those committed
by the historically barred class. Evaluation of the second factor
should be grounded in this legal framework.
        Given the indeterminate nature of the historically barred
class, I do not dispute that current authorities may assist us in
measuring the “seriousness” of a challenger’s offense. But any
such measurement must be consistent with our precedent. To
me, the most relevant contemporary authorities for measuring
“seriousness” are in fact included in the third and fourth
factors: the actors within the criminal-justice system who
confronted the challenger’s offense and imposed a punishment,
and the jurisdictions that penalize the challenger’s conduct as
a crime. As a result, I must conclude that the second factor
weighs in Holloway’s favor.




                               18
                                3
       Although the preceding factors support Holloway, they
are insufficient in themselves to establish whether he is entitled
to Second Amendment protection. Because a majority of the
judges in Binderup held that a nonviolent misdemeanor may
be “serious,” the preceding factors, while probative measures
of “seriousness,” are not dispositive. Yet in the absence of
common features of “serious” nonviolent misdemeanors—and
Binderup did not specify any—we must compare the
punishment for the challenger’s crime with the punishments
for the crimes of the historically barred class. 

(opinion of Ambro, J.). The third and fourth Binderup
factors both accomplish this end.13
       The third factor looks to the sentence the challenger
received. It directs our attention to the unique circumstances of
the challenger’s offense and conviction. Holloway was
arrested in January 2005 after a police officer witnessed him
driving the wrong way down a one-way street. 

. He registered a blood alcohol concentration
(BAC) at the “highest rate” under Pennsylvania law, and
because this was his second DUI offense, he was convicted of

13
  The majority says that “in addition to ascribing high value to
the offense’s felony/misdemeanor label,” I “favor[]” a focus
“on the actual penalty imposed.” Majority Op. at II.B.2 n.12. It
contrasts this view with its own, declaring it “proper to
consider the maximum penalty an offender faces, and not
simply” these other factors. 

have noted, however, I do
not value any one factor above another, and in fact agree with
the majority that the maximum penalty is relevant, though (for
the reasons I state below) I think that such a penalty is most
appropriately, for purposes of the Barton and Binderup
comparison, considered under the fourth factor.




                               19
a first-degree misdemeanor, punishable by up to five years in
prison. 

received the mandatory minimum
sentence, which included three months of confinement on a
work-release program. 

majority finds this factor against Holloway,
emphasizing that, unlike the challengers in Binderup, he
received a punishment that deprived him of his liberty. See
Majority Op. at II.B.2. While this fact is certainly evidence that
Pennsylvania considers Holloway’s offense more significant
than that of Binderup (which was also committed in
Pennsylvania), it does not measure Holloway’s offense against
those of the historically barred class. A factor that considers the
punishment received suggests some deference to the decisions
of those within the criminal-justice system. See 

(opinion of Ambro, J.) (“[P]unishments are selected
by judges who have firsthand knowledge of the facts and
circumstances of the cases and who likely have the benefit of
pre-sentence reports prepared by trained professionals.”).
Here, the actors on the ground did not deem Holloway’s
offense “serious” enough to warrant the maximum penalty that
Pennsylvania law permitted. Rather, the sentencing judge
imposed the lightest punishment that the law allowed—a term
of imprisonment, with work release, considerably shorter than
the qualifying sentences under either § 922(g)(1) or §
921(a)(20)(B). As the District Court noted, Holloway’s
assignment to a work-release program “undergirds the
relatively minor nature of his sentence and suggests that the
sentencing judge did not find Holloway to pose a significant
risk to public safety.” 

.
      For the purposes of the Barton and Binderup
comparison, then, I conclude that those who administered
Pennsylvania’s law did not deem Holloway’s offense “serious”
enough to merit imposition of a sentence on a par with those of




                                20
the historically barred class. The argument that Holloway’s
punishment was greater than anything received by the
Binderup challengers bears more on the final factor than on the
present one. The latter supports Holloway’s claim to Second
Amendment protection.
                                4
        The fourth factor asks whether there is a “cross-
jurisdictional consensus regarding the seriousness of the
[challenger’s] crime[].” 

sentence actually received,
the challenger’s maximum possible punishment similarly
provides a point of comparison with the historically barred
class, but it cannot be assessed by looking to the challenger’s
jurisdiction alone. The fact that the challenger’s crime is
punishable by more than one or two years is the very reason he
is in court; it demonstrates only that one jurisdiction has chosen
to punish his conduct on terms comparable to the punishments
of the historically barred class. More significant is how
jurisdictions generally punish the challenger’s conduct because
such a measure permits a comparison of current appraisal of
the significance of the challenger’s crime with the punishments
imposed on the historical class.
        My review of the DUI laws in all fifty states and the
District of Columbia reveals a notable consensus in how these
jurisdictions punish Holloway’s conduct. Most importantly,
only twelve of these jurisdictions punish such conduct with a
maximum term of imprisonment exceeding one year.14 Of

14
  See CONN. GEN. STAT. §§ 14-227a, 53a-25, 53a-26 (2019);
DEL. CODE ANN. tit. 21, § 4177; tit. 11, § 233 (2019); IND.
CODE §§ 9-30-5-1, 9-30-5-3 (2019); IOWA CODE § 321J.2
(2019); MD. CODE ANN., TRANSP. §§ 21-902, 27-101 to -102
(West 2019); MASS. GEN. LAWS ch. 90, § 24; ch. 274, § 1




                               21
these twelve jurisdictions, seven provide for a maximum
punishment exceeding two years,15 and only four of these
seven classify such a crime as a misdemeanor.16 The other three
jurisdictions, as well as the remaining five that punish the
crime by more than one year of imprisonment, classify it as a
felony. Given these statistics, there is no cross-jurisdictional
consensus that a second DUI offense with a BAC at 0.192% is
“serious” for purposes of Second Amendment analysis. In fact,
the consensus lies in the other direction: a significant majority
of jurisdictions—thirty-nine out of fifty-one—do not consider
Holloway’s second DUI offense to be a crime worthy of
punishment in accord with that of a traditional felony.



(2019); N.Y. VEH. & TRAF. LAW §§ 1192-1193 (McKinney
2019); N.C. GEN. STAT. §§ 20-138.1, 20-138.5, 20-179 (2019);
OKLA. STAT. tit. 47, § 11-902 (2019); 75 PA. CONS. STAT. §§
3803(b)(4), 3804 (2019); 18 PA. CONS. STAT. § 1104 (2019);
S.C. CODE ANN. §§ 56-5-2930, 56-5-2933, 16-1-20, 16-1-100
(2019); VT. STAT. ANN. tit. 23, §§ 1201, 1210; tit. 13, § 1
(2019).
15
   See IND. CODE §§ 9-30-5-1, 9-30-5-3 (2019); MASS. GEN.
LAWS ch. 90, § 24; ch. 274, § 1 (2019); N.Y. VEH. & TRAF.
LAW §§ 1192-1193 (McKinney 2019); N.C. GEN. STAT. §§ 20-
138.1, 20-138.5, 20-179 (2019); OKLA. STAT. tit. 47, § 11-902
(2019); 75 PA. CONS. STAT. §§ 3803(b)(4), 3804 (2019); 18 PA.
CONS. STAT. § 1104 (2019); S.C. CODE ANN. §§ 56-5-2930, 56-
5-2933, 16-1-20, 16-1-100 (2019).
16
   See MASS. GEN. LAWS ch. 90, § 24; ch. 274, § 1 (2019); N.C.
GEN. STAT. §§ 20-138.1, 20-138.5, 20-179 (2019); 75 PA.
CONS. STAT. §§ 3803(b)(4), 3804 (2019); 18 PA. CONS. STAT.
§ 1104 (2019); S.C. CODE ANN. §§ 56-5-2930, 56-5-2933, 16-
1-20, 16-1-100 (2019).




                               22
        The majority finds it sufficient that “states unanimously
agree that DUIs are crimes subject to punishment.” Majority
Op. at II.B.2. Yet as I have emphasized, our precedent dictates
that the relevant measures of “seriousness” are those indicating
how the challenger’s circumstances compare with the
circumstances of the historically barred class. The fact of
punishment alone should not render a crime “serious” enough
to deprive an individual of a constitutional right. In the light of
the evidence presented above, I must conclude that under the
fourth factor, Holloway is not removed from the scope of
Second Amendment protection.
                               ***
        Drunk driving is a dangerous crime. Declaring it not
“serious” for purposes of the Second Amendment in no way
detracts from its “seriousness” in the ordinary understanding
of that word. But that is my point—the two categories are
distinct, and our analysis should reflect that fact. Although
Binderup did not create controlling precedent on the nature of
the “seriousness” inquiry, the legal content of that inquiry must
fulfill the requirements established in Barton and Binderup.
Properly understood and applied, the multifactor test meets
these demands. And in the context of the present case, it leads
me to agree with the District Court that § 922(g)(1) burdens
Holloway’s constitutional right to own a firearm. In this way, I
part with the majority in this case.
                                II
       If a court determines, as I do here, that the challenged
law burdens protected conduct, then Marzzarella’s second step
requires the court to “evaluate the law under some form of
means-ends 

. In Binderup, the same
ten judges who agreed to adopt Marzzarella’s two-step
framework and the “seriousness” standard also accepted the




                                23
application of intermediate scrutiny in as-applied challenges to
§ 922(g)(1). See 

(opinion of Ambro,
J.); 

(Fuentes, J., concurring in part, dissenting in part,
and dissenting from the judgments). Therefore, our precedent
requires the application of intermediate scrutiny in the present
case. See 

.
        Following a long line of Supreme Court case law,
Marzzarella enumerated two elements of intermediate-scrutiny
review. First, the government interest in the enforcement of the
challenged regulation must be “significant, substantial, or

(internal quotation marks omitted).
Second, there must be a “reasonable” fit between the asserted
government interest and the regulation as written or applied.
Id.; see also Ass’n of N.J. Rifle & Pistol Clubs, Inc. v. Attorney
Gen. New Jersey, 

, 119 (3d Cir. 2018) (adopting
this two-part test); Drake v. Filko, 

, 436 (3d Cir.
2013) (same). I will consider each in turn.
                                 A
        The parties do not contest that the government has a
substantial interest in “protecting the public from people who
cannot be trusted to use firearms responsibly.” Appellants’ Br.
at 29. Neither Holloway’s brief nor the District Court’s opinion
even mention this element. Thus, there is no reason to question
whether the government has a substantial interest in enforcing
§ 922(g)(1).
                                 B
       Our primary difficulty lies in determining how to apply
the second element of intermediate-scrutiny review to §
922(g)(1). Binderup established no precedent for how to decide
whether there is a “[reasonable] fit between [§ 922(g)(1)] and
the asserted governmental end.” 

.
Moreover, the standards applied by the judges in that case are




                                 24
not the same as the standard applied by the Court in
Marzzarella. Yet as I detail in Section II.B.1, these standards
are in fact doctrinally consistent with each other. If the
government presents sufficient evidence to support its
enforcement of the regulation at issue, we are then to evaluate
how closely the regulation has been drawn to advance that
interest. This is the standard I apply in Section II.B.2,
concluding that § 922(g)(1) as applied in the present case fails
intermediate scrutiny.
                               1
       There is no binding precedent in our Circuit for the
proper application of intermediate scrutiny to § 922(g)(1). In
Binderup, the opinion announcing the Court’s judgment said
the government “must ‘present some meaningful evidence, not
mere assertions, to justify its predictive . . . judgments’”
regarding the danger presented by the challengers and others
like 

(opinion of Ambro, J.) (quoting
Heller v. District of Columbia, 

, 1259 (D.C. Cir.
2011)). By contrast, in Marzzarella, the Court held that 18
U.S.C. § 922(k) “fits reasonably with [the government’s
asserted] interest in that it reaches only conduct creating a
substantial risk of rendering a firearm 

. Whereas in Binderup, then, the judges were concerned
with the evidence the government put forward, in Marzzarella
the Court focused on the relation between the government’s
asserted interest and the statute’s actual operation.17
      Despite this ostensible difference, these standards are in
fact consistent with each other as a doctrinal matter.

17
   For the same reasons given above with regard to the
multifactor test, I do not think the application of intermediate
scrutiny in Binderup is binding precedent under the Marks rule.




                              25
Marzzarella followed Heller in looking to the Supreme Court’s
First Amendment case law for guidance, calling that doctrine
“the natural choice” for “evaluating Second Amendment

n.4. In particular, for the second
prong of intermediate-scrutiny review—that “the fit between
the challenged regulation and the asserted objective be
reasonable, not perfect”—Marzzarella referred to two of the
Supreme Court’s commercial-speech cases. See 

(citing Lorillard Tobacco Co. v. Reilly, 

, 556
(2001); and Bd. of Trs. of State Univ. of N.Y. v. Fox, 

, 480 (1989)). A brief consideration of commercial-speech
doctrine allows us to see how our Circuit’s Second Amendment
precedent in fact dictates a single standard for subjecting §
922(g)(1) to intermediate scrutiny.18
       The Supreme Court applies a four-step test for
determining whether a regulation of commercial speech
violates the First Amendment. A court must first “determine
whether the expression is protected by the First Amendment,”
and then “ask whether the asserted governmental interest is
substantial.” Cent. Hudson Gas & Elec. Corp. v. Pub. Serv.
Comm’n of N.Y., 

, 566 (1980); see also 

. If the answer to both inquires is affirmative, the
government must then show “that the statute directly advances
a substantial governmental interest and that the measure is
drawn to achieve that interest.” Sorrell v. IMS Health Inc., 564


18
  The majority suggests that I am “advocating that we import”
the commercial-speech standard into the § 922(g)(1) context.
Majority Op. at II.B.2 n.16. To the contrary, I am simply
applying our precedent, mindful that Marzzarella has “guided
how we approach as-applied Second Amendment challenges.”

(plurality opinion).




                              

, 572 (2011) (citing 

; and Cent.

).
        This test bears notable resemblance to our Circuit’s
developing Second Amendment doctrine. For our purposes
here, the third and fourth steps are especially remarkable: they
resemble the standards applied in Binderup and Marzzarella,
respectively. Both are essential means of measuring the fit
between the interest and the regulation. Indeed, the Supreme
Court has said that these steps are not necessarily distinct
inquiries. In as-applied challenges, the question posed at step
three “cannot be answered by limiting the inquiry to whether
the governmental interest is directly advanced as applied to a
single person or entity.” United States v. Edge Broad. Co., 

, 427 (1993). The court must also consider “the
regulation’s general application to others” with the same
relevant characteristics as the challenger. 

result, the
validity of the regulation’s application to the challenger
“properly should be dealt with under the fourth factor of the
Central Hudson test.” 

that, regardless of the
nature of the challenge, the third and fourth steps “basically
involve a consideration of the fit between the legislature’s ends
and the means chosen to accomplish those ends.” 

(internal quotation marks omitted).
       This background clarifies the standard to apply in the
present case. In effect, Binderup concerned the correct
application of the third step—whether the regulation “directly
advances a substantial governmental interest.” 

. The three-judge opinion announcing the judgment
of the Court did not need to advance its inquiry any further,
because it concluded that § 922(g)(1) already failed as applied.
In Marzzarella, however, there was no question whether the
government had presented sufficient evidence to justify its
enforcement action, and so the Court looked to how closely §




                               27
922(k) was drawn to achieve the government’s stated interest,
holding that the statute is not impermissibly overinclusive
because “it reaches only conduct creating a substantial risk of
rendering a firearm 

. As a result, to
my mind Binderup and Marzzarella are doctrinally consistent,
or at least reconcilable, in the light of how the Supreme Court
has elaborated the final two steps of the commercial-speech
test. At Marzzarella Step Two, if we are satisfied with the
evidence supporting the statute’s application, we must then
consider how closely the statute has been drawn to advance the
government’s substantial interest.
                               2
       Applying that standard in the present case, I conclude
that § 922(g)(1) does not survive intermediate scrutiny. I
disagree with the District Court, however, that the government
has failed to produce evidence demonstrating that its
enforcement of the statute directly advances its stated
substantial interest. Rather, the flaw with the government’s
case is that the statute as applied here is “wildly
underinclusive.” Nat’l Inst. Fam. & Life Advocs. v. Becerra,

, 2375 (2018) (quoting Brown v. Entm’t Merchs.
Ass’n, 

, 802 (2011)).
                               a
       In finding that § 922(g)(1)’s enforcement in this case
does not directly advance the government’s substantial
interest, the District Court demanded an excessively
particularized connection between the evidence proffered and
Holloway’s circumstances. Yet, as explained above, we should
not limit our “inquiry to whether the governmental interest is
directly advanced as applied to a single person or entity,” but
also consider “the regulation’s general application to others.”
Edge Broad. 

. The government’s studies in




                               28
Binderup were “obviously 

(opinion of Ambro, J.). They concerned the likelihood of
incarcerated felons to reoffend, though the Binderup
challengers were neither incarcerated nor felons under state
law. And the studies cited recidivism rates not applicable to
individuals in the challengers’ situation. More compelling
studies would have presented evidence relating to individuals
“with the Challengers’ backgrounds.” 

       The government’s expert report in the present case does
exactly that. It offers evidence relating to the features of
Holloway’s biography that are at issue in this case. It refers to
the likelihood of drug and alcohol abuse among repeat DUI
offenders. D. Ct. Docket No. 61-4, at 4. It refers to firearm
purchasers with prior alcohol-related convictions. 

These are the features of Holloway’s biography at issue here.
For the purposes of government policy, barring individuals
with those characteristics from possessing a firearm is
reasonable.
                               b
        As explained above, our inquiry into “reasonable fit”
does not end here. The question is not merely whether it is
reasonable to disarm the challenger because of his conviction,
but whether “the fit between the challenged regulation and the
asserted objective [is] reasonable.” 

(emphasis added). As a result, we must consider, in the context
of this as-applied challenge, how closely § 922(g)(1) has been
drawn to achieve the government’s substantial interest.
       Under this standard, the law appears to be significantly
underinclusive. Holloway’s crimes—a first DUI offense at a
BAC of 0.131%, and a second DUI offense less than three
years later with a BAC of 0.192%—implicate § 922(g)(1) in
only eight of fifty-one jurisdictions (the fifty states and the




                               29
District of Columbia).19 These eight jurisdictions account for
approximately 21% of the United States population.20 On
average, then, only about one in five individuals behaving
exactly as Holloway did would be barred from possessing a
firearm under § 922(g)(1). The statute’s dependence on state
criminal classifications and punishments results in an
underinclusive application that raises constitutional concerns,
regardless of the reasonableness of disarming recidivist DUI
offenders.


19
   See CONN. GEN. STAT. §§ 14-227a, 53a-25, 53a-26 (2019);
IND. CODE §§ 9-30-5-1, 9-30-5-3 (2019); MASS. GEN. LAWS ch.
90, § 24; ch. 274, § 1 (2019); N.Y. VEH. & TRAF. LAW §§ 1192-
1193 (McKinney 2019); N.C. GEN. STAT. §§ 20-138.1, 20-
138.5, 20-179 (2019); OKLA. STAT. tit. 47, § 11-902 (2019); 75
PA. CONS. STAT. § 3804 (2019); S.C. CODE ANN. §§ 56-5-2930,
56-5-2933, 16-1-20, 16-1-100 (2019).
20
   I base this number on the U.S. Census Bureau’s estimated
2019 national and state populations. The estimated population
of the fifty states and the District of Columbia on July 1, 2019
was 328,239,523 persons. See U.S. Census Bureau, National
Population Totals and Components of Change: 2010-2019,
U.S.        DEP’T        COM.         (Dec.     23,       2019),
https://www.census.gov/data/tables/time-
series/demo/popest/2010s-national-total.html. On that same
date, the total estimated population of the eight states where
Holloway’s crimes would implicate § 922(g)(1) was
69,039,328 persons. See U.S. Census Bureau, State Population
Totals and Components of Change: 2010-2019, U.S. DEP’T
COM.                  (Dec.               30,             2019),
https://www.census.gov/data/tables/time-
series/demo/popest/2010s-state-total.html.




                              30
                                c
       The next question is whether this underinclusivity
renders § 922(g)(1) as applied here unconstitutional under
intermediate scrutiny. To my mind, there are two principal
counterarguments to answering this question affirmatively.
Both of them fail.
        First, it might be argued that our precedent remains
unsettled regarding whether underinclusivity is a valid
consideration in the Second Amendment context. Although
Marzzarella allowed that a regulation’s “underinclusiveness
can be evidence that the interest is not significant enough to
justify the 

, the Court was there
referring to underinclusivity in the context of strict, rather than
intermediate, scrutiny. As a result, a future panel majority may
reject a consideration of underinclusivity in intermediate-
scrutiny review. See, e.g., Ass’n of N.J. Rifle & Pistol 

n.28 (“While our Court has consulted First
Amendment jurisprudence concerning the appropriate level of
scrutiny to apply to a gun regulation, we have not wholesale
incorporated it into the Second Amendment.” (citations
omitted)).
       Yet, in constitutional law, underinclusivity follows
necessarily from the evaluation of a fit between means and
ends. And in Marzzarella we explicitly adopted a test that
considers “the fit between the challenged regulation and the
asserted 

; see also 

; Fla. Bar v. Went For It, Inc., 

, 632 (1995);

. The assessment of fit looks to the relation
between the class of persons who come within the scope of the
regulation’s stated objective, and the class of persons actually
affected by the regulation. See, e.g., Joseph Tussman &
Jacobus tenBroek, The Equal Protection of the Laws, 37 CALIF.




                                31
L. REV. 341, 344-53 (1949).21 Under this standard, what
matters is not whether a regulation is specifically
overinclusive, but rather by how much it is either over- or
underinclusive. See, e.g., City of Cincinnati v. Discovery
Network, Inc., 

, 428 (1993) (holding a city
ordinance intended to advance safety and aesthetic interests
unconstitutional because it unjustifiably affected only a small
fraction of operating newsracks, thus constituting an
unreasonable fit between ends and means).
        This generalized inquiry encompasses both
intermediate and strict scrutiny. The difference between those
standards is the degree, rather than the type, of fit—whether
the fit is either “reasonable” or “perfect.” 

; see McCutcheon v. Fed. Election Comm’n, 

, 218 (2014) (plurality opinion) (“Even when the Court is
not applying strict scrutiny, we still require ‘a fit that is not
necessarily perfect, but reasonable; . . . that employs not
necessarily the least restrictive means but . . . a means narrowly
tailored to achieve the desired objective.’” (alterations in
original) (quoting 

)). Intermediate scrutiny


21
   The Court first developed this test in the equal-protection
context, and subsequently imported it into First Amendment
doctrine in the early 1970s. See, e.g., Police Dep’t of Chicago
v. Mosley, 

(1972); Kenneth L. Karst, Equality as a
Central Principle in the First Amendment, 43 U. CHI. L. REV.
20 (1975). It therefore makes sense that when our Court in
Marzzarella began to formulate Second Amendment doctrine,
it called for an evaluation of the challenged law “under some
form of means-end 

, and described that
evaluation as an assessment of “the fit between the challenged
regulation and the asserted objective,” 

.




                               32
simply requires less of a fit between the governmental interest
and the challenged regulation than strict scrutiny does.
       It would be contrary to the logic of this analysis to hold
that under intermediate scrutiny alone a court may not consider
a regulation’s underinclusivity. To be sure, there may be a
compelling reason why the Second Amendment context
precludes such a consideration, but, to my mind, even that
determination must now be left either to this Court sitting en
banc or to the Supreme Court. Because our Court in
Marzzarella adopted a means-ends fit analysis, we have
already decided that underinclusivity is at least a valid
consideration.
         Second, it might be argued that § 922(g)(1) as applied
here falls into one of the contexts in which the Supreme Court
has upheld a regulation despite claims of underinclusivity. In
particular, the Court has acknowledged two principal
defenses—that a distinction drawn by a lawmaking body is in
itself legitimate, and that a legislature is permitted to address a
problem incrementally. See Williams-Yulee v. Fla. Bar, 135 S.
Ct. 1656, 1668-69 (2015) (highlighting these two defenses);
see also McConnell v. Fed. Election Comm’n, 

,
207-08 (2003) (“[R]eform may take one step at a time,
addressing itself to the phase of the problem which seems most
acute to the legislative mind.” (quoting Buckley v. Valeo, 

, 105 (1976))); Burson v. Freeman, 

, 207
(1992) (“States adopt laws to address the problems that
confront them. The First Amendment does not require States to
regulate for problems that do not exist.”).
       These defenses do not support § 922(g)(1) as applied in
the present case. Congress has drawn no distinction between
different types of conduct—the same behavior may activate §
922(g)(1) or not based merely on where that behavior occurred.




                                33
See City of 

(declaring a city
ordinance       unconstitutionally       underinclusive      under
intermediate scrutiny “[b]ecause the distinction [the city] has
drawn has absolutely no bearing on the interests it has
asserted”). For this same reason, it is hard to see how the statute
represents Congress addressing problems as they arise. Section
922(g)(1) sweeps so broadly, covering any person convicted
under state law of a felony or a misdemeanor carrying a
sentence that exceeds two years, that in particular applications
it is underinclusive, curtailing the constitutional rights of some
and not others for the exact same conduct. Far from regulating
for problems that do not exist, Congress is here not even
regulating the vast majority of conduct it apparently deems
problematic.
                               ***
       Ultimately at stake in this case is whether the
government may arbitrarily burden the constitutional right of
some citizens and not others. This equality concern goes to the
heart of constitutional adjudication, regardless of the nature of
the right at issue. As Justice Jackson put it in a different
context:
       The framers of the Constitution knew, and we
       should not forget today, that there is no more
       effective practical guaranty against arbitrary and
       unreasonable government than to require that the
       principles of law which officials would impose
       upon a minority must be imposed
       generally. Conversely, nothing opens the door to
       arbitrary action so effectively as to allow those
       officials to pick and choose only a few to whom
       they will apply legislation and thus to escape the
       political retribution that might be visited upon




                                34
       them if larger numbers were affected. Courts can
       take no better measure to assure that laws will be
       just than to require that laws be equal in
       operation.
Ry. Express Agency v. New York, 

, 112-13 (1949)
(Jackson, J., concurring). When a law, for reasons unrelated to
enforcement discretion, on average punishes the same conduct
only one in five times, such that those chosen individuals are
deprived in perpetuity of a constitutional right, there is not a
reasonable fit between the legislature’s asserted interest and
the challenged regulation.22 If Congress wants to bar all
individuals convicted of a second DUI offense with a BAC
above 0.16% of owning a firearm, then it must do so through
the ordinary channels of democratic lawmaking. At least then
all persons’ constitutional right will be treated equally.
       For these reasons, I respectfully dissent.




22
   Although the majority does not reach Step Two, it observes
that I “seemingly ask[] for a near-perfect fit.” Majority Op. at
II.B.2 n.16. Reasonable minds may differ as to whether
demanding a fit of greater than 21% is to demand near-
perfection.




                               35