FOR PUBLICATION

   UNITED STATES COURT OF APPEALS
        FOR THE NINTH CIRCUIT

ANDREW TETER; JAMES GRELL,                 No. 20-15948

              Plaintiffs-Appellants,          D.C. No.
 v.                                        1:19-cv-00183-
                                             ACK-WRP
ANNE E. LOPEZ; DARRYL NG,

              Defendants-Appellees.          OPINION

       Appeal from the United States District Court
                for the District of Hawaii
         Alan C. Kay, District Judge, Presiding

           Argued and Submitted June 25, 2024
                  Seattle, Washington

                  Filed January 22, 2025

 Before: Mary H. Murguia, Chief Judge, and Ronald M.
 Gould, Jacqueline H. Nguyen, Ryan D. Nelson, Eric D.
Miller, Bridget S. Bade, Daniel P. Collins, Kenneth K. Lee,
Lawrence VanDyke, Gabriel P. Sanchez and Ana de Alba,
                      Circuit Judges.

                 Opinion by Judge Miller;
            Partial Dissent by Judge VanDyke;
                 Dissent by Judge Collins
2                         TETER V. LOPEZ


                          SUMMARY*


                      Second Amendment

   The en banc court vacated the district court’s summary
judgment for the Hawaii Attorney General and the Hawaii
Sheriff Division Administrator in an action involving a
Second Amendment challenge to Hawaii’s statute
prohibiting butterfly knives, and remanded for further
proceedings.
    While the litigation was pending, the Hawaii Legislature
amended the challenged statute, Hawaii Revised Statutes
§ 134-53(a). Although the new statute continues to impose
some restrictions on butterfly knives, it no longer prohibits
them.
    The en banc court concluded that section 134-53(a) has
been sufficiently altered so as to present a substantially
different controversy from the one the district court
originally decided. The statutory amendment gave plaintiff
everything he hoped to achieve in this litigation. Because no
further relief could be granted, the case was moot, and the
court lacked Article III jurisdiction. The amended statute
does not restrict the acquisition, possession, and use of
butterfly knives, except insofar as a different subsection now
prohibits their possession or use by someone engaged in the
commission of a separate felony or misdemeanor. Nor does
the statute prohibit carrying butterfly knives. Although it
does prohibit carrying concealed butterfly knives, plaintiff
made clear in the district court that he sought the right to

*
 This summary constitutes no part of the opinion of the court. It has
been prepared by court staff for the convenience of the reader.
                         TETER V. LOPEZ                        3


carry a butterfly knife openly. Under the amended statute, he
has that right. On remand, plaintiff may assert whatever
claims remain available to him under the new statute.
    Dissenting in part, Judge VanDyke agreed with Judge
Collins that this case is not moot for the general reasons that
Judge Collins provides in his separate dissent. Judge
VanDyke concurred in the decision to remand this case for
further proceedings but respectfully dissented from that
portion of the majority’s decision that simply vacated the
district court’s judgment without taking any additional
action directed at the court’s prior en banc order
ministerially vacating the panel opinion. To discourage
parties from using mootness strategically, Judge VanDyke
would (1) reinstate the panel opinion by vacating the prior
vacatur order, (2) address those legal questions that are
indisputably relevant to any possible reformulation of
plaintiffs’ claims on remand, and (3) remand the matter to
the district court for plaintiffs to pursue additional challenges
to Hawaii’s reformulated butterfly knife law, should they
choose to do so.
    Dissenting, Judge Collins, joined by Judge Lee, stated
that it is the appellees’ burden to establish that re-enactment
of a similar law cannot reasonably be expected to occur.
They have not carried that formidable burden, given that
there are strong reasons to suspect that the very lawsuit at
hand was the impetus for the legislative amendment.
Because this case has not been shown to be moot, Judge
Collins would proceed to the merits. On the merits, he would
adhere to the views expressed in the panel opinion, including
the holding that bladed weapons facially constitute arms
within the meaning of the Second Amendment.
4                     TETER V. LOPEZ


                       COUNSEL

Alan A. Beck (argued), Law Offices of Alan Beck, San
Diego, California; Stephen D. Stamboulieh (argued),
Stamboulieh Law PLLC, Olive Branch, Mississippi; for
Plaintiffs-Appellants.
Robert T. Nakatsuji (argued), First Deputy Solicitor
General; Ryan M. Akamine and Caron M. Inagaki, Deputy
Attorneys General; Kalikoonalani D. Fernandes, Deputy
Solicitor; Kimberly T. Guidry, Solicitor General; Clare E.
Connors and Holly T. Shikada, Attorneys General of
Hawaii; Office of the Attorney General Hawaii, Honolulu,
Hawaii; Reedy Swanson (argued), Neal K. Katyal, and Dana
A. Raphael, Hogan Lovells US LLP, Washington, D.C.; for
Defendants-Appellees.
Kevin O'Grady, Law Office of Kevin O'Grady LLC,
Honolulu, Hawaii; David T. Hardy, David T. Hardy PC,
Tucson, Arizona; for Amicus Curiae Hawaii Firearms
Coalition.
Cody J. Wisniewski, Mountain States Legal Foundation,
Lakewood, Colorado, for Amicus Curiae Mountain States
Legal Foundation.
John W. Dillon, Dillon Law Group APC, Carlsbad,
California, for Amici Curiae San Diego County Gun Owners
Political Action Committee, Firearms Policy Coalition, and
Knife Rights Foundation Inc..
Michael R. Dreeben and David K. Roberts, O'Melveny &
Myers LLP, Washington, D.C.; Danielle R. Feuer,
O'Melveny & Myers LLP, New York, New York; Wendy F.
Hanakahi and Pamela W. Bunn, Dentons US LLP, Honolulu,
Hawaii; William J. Taylor Jr., Lisa M. Ebersole, Carina B.
                      TETER V. LOPEZ                     5


Gryting, and Janet Carter, Everytown Law, New York, New
York; for Amici Curiae Gun-Violence Prevention Groups
and Policy Researchers.
John D. Echeverria, Deputy Attorney General; Mica L.
Moore and Helen H. Hong, Deputy Solicitors General;
Michael J. Mongan, Solicitor General; Rob Bonta, Attorney
General of California; Los Angeles, California; Kristen K.
Mayes, Attorney General of Arizona, Phoenix Arizona;
Philip J. Weiser, Attorney General of Colorado, Denver,
Colorado; William Tong, Attorney General of Connecticut,
Hartford, Connecticut; Kathleen Jennings, Attorney General
of Delaware, Wilmington, Delaware; Kwame Raoul,
Attorney General of Illinois, Chicago, Illinois; Anthony G.
Brown, Attorney General of Maryland, Baltimore,
Maryland; Andrea J. Campbell, Attorney General of
Massachusetts, Boston, Massachusetts; Dana Nessel,
Attorney General of Michigan, Lansing, Michigan; Keith
Ellison, Attorney General of Minnesota, St. Paul,
Minnesota; Matthew J. Platkin, Attorney General of New
Jersey, Trenton, New Jersey; Letitia James, Attorney
General of New York, New York, New York; Ellen F.
Rosenblum, Attorney General of Oregon, Salem, Oregon;
Michelle Henry, Attorney General of Pennsylvania,
Harrisburg, Pennsylvania; Peter F. Neronha, Attorney
General of Rhode Island, Providence, Rhode Island; Robert
W. Ferguson, Attorney General of Washington, Olympia,
Washington; Brian L. Schwalb, Attorney General of the
District of Columbia, Washington, D.C.; for Amicus Curiae
States of California, Arizona, Colorado, Connecticut,
Delaware, Illinois, Maryland, Massachusetts, Michigan,
Minnesota, New Jersey, New York, Oregon, Pennsylvania,
Rhode Island, Washington, and the District of Columbia.
6                     TETER V. LOPEZ


Daniel M. Gluck, Deputy Corporation Counsel; Dana M. O.
Viola, Corporation Counsel; Honolulu City & County
Department of the Corporation Counsel, Honolulu, Hawaii,
for Amicus Curiae City & County of Honolulu.
Joseph G.S. Greenlee and Erin M. Ernhardt, National Rifle
Association of America, Institute for Legislative Action,
Fairfax, Virginia, for Amicus Curiae National Rifle
Association of America.
Edward A. Paltzik, Serge Krimnus, and Meredith Lloyd,
Bochner PLLC, New York, New York, for Amicus Curiae
Second Amendment Foundation.
Benjamin E. Lowenthal, Law Office of Benjamin E.
Lowenthal, Honolulu, Hawaii, for Amicus Curiae the Office
of the Public Defender.
                        TETER V. LOPEZ                      7

OPINION

MILLER, Circuit Judge:

     This case involves a Second Amendment challenge to
Hawaii’s statute prohibiting butterfly knives. While the
litigation was pending, the Hawaii Legislature amended the
challenged statute. Although the new statute continues to
impose some restrictions on butterfly knives, it no longer
prohibits them. We conclude that the amended statute is not
“substantially similar” to the version originally challenged.
Board of Trs. of Glazing Health & Welfare Tr. v. Chambers,

 (9th Cir. 2019) (en banc). This case is
therefore moot, and we vacate the district court’s judgment.
Because “the plaintiff[s] may have some residual claim
under the new framework,” we remand for further
proceedings. New York State Rifle & Pistol Ass’n v. City of
New York, 

 (2020) (quoting Lewis v.
Continental Bank Corp., 

 (1990)).
                              I
    A butterfly knife, also known as a “balisong,” is a knife
with a folding handle that covers the sharp edge of the blade
when the knife is closed. Unlike an ordinary pocketknife, a
butterfly knife has a handle that is split into two components
that rotate in opposite directions to open. This design allows
a user to expose the blade of the knife by flipping it open
with one hand.
   In 1999, the Hawaii Legislature prohibited the
possession of butterfly knives. See 

, § 1. Specifically, Hawaii Revised Statutes § 134-53(a)
(1999) provided: “Whoever knowingly manufactures, sells,
8                       TETER V. LOPEZ


transfers, possesses, or transports in the State any butterfly
knife . . . shall be guilty of a misdemeanor.”
    Andrew Teter is a Hawaii resident who wishes to own
butterfly knives. Together with James Grell, a similarly
situated Hawaii resident, Teter brought this action in 2018
under 

 against the Hawaii Attorney General
and the Hawaii Sheriff Division Administrator (collectively,
“the Attorney General”), alleging that the ban on butterfly
knives in section 134-53(a) violated the Second
Amendment. In the complaint, Teter asserted “an as-applied
and facial challenge to the applicable Hawaii laws which
prevent [Teter] from owning butterfly knives.” He sought an
injunction against Hawaii’s “policies generally banning the
acquisition, possession, carrying and use of butterfly
knives,” as well as a declaration “that the State of Hawaii’s
ban on the ownership of butterfly knives violates the Second
Amendment.”
    Both parties moved for summary judgment. At a hearing
on those motions, the district court observed that our
decision in Peruta v. County of San Diego foreclosed any
Second Amendment challenge to a prohibition on the
concealed carrying of butterfly knives. 

(9th Cir. 2016) (en banc). Teter agreed that his challenge
focused on “the right to possess a butterfly knife in [the]
home, as well as the right to carry it openly in public.”
    The district court granted summary judgment to the
Attorney General. The court described Teter’s argument as
one that the statute is “unconstitutional only as applied to
law-abiding citizens seeking to possess butterfly knives in
their homes or to openly carry them in public.” After
determining that “the statute does not severely burden the
core Second Amendment right,” the district court applied
                         TETER V. LOPEZ                         9


intermediate scrutiny under then-controlling precedent.
Because the district court identified a reasonable fit between
the statute and the State’s substantial interest in public
safety, it held that the statute survived intermediate scrutiny.
     Teter appealed. While the appeal was pending, the
Supreme Court decided New York State Rifle & Pistol Ass’n,
Inc. v. Bruen, 

 (2022). In that case, the Court
rejected the approach of applying varying levels of scrutiny
to different regulations of arms. 

. It instead held “that
when the Second Amendment’s plain text covers an
individual’s conduct, the Constitution presumptively
protects that conduct,” and a regulation may be justified only
if it “is consistent with this Nation’s historical tradition of
firearm regulation.” 

    A three-judge panel of this court reversed the district
court’s decision. Teter v. Lopez, 

 (9th Cir. 2023).
Applying Bruen, the panel held that “the possession of
butterfly knives is covered by the plain text of the Second
Amendment.” 

. It stated that although the
“‘problem’ of easily concealable, foldable knives being used
in crimes” has existed since the 18th century, the Attorney
General “cite[d] no analogues in which Congress or any state
legislature imposed an outright ban on the possession of
pocketknives to remedy this problem near 1791 or 1868.” 

at 954 (quoting Bruen, 

). The panel therefore
concluded that section 134-53(a) violated the Second
Amendment. 

   The Attorney General petitioned for rehearing en banc.
While that petition was pending, Speaker of the Hawaii
House of Representatives Scott Saiki introduced House Bill
2342, which proposed various changes to Hawaii’s weapons
laws, including section 134-53(a). Where the prior version
10                      TETER V. LOPEZ


of section 134-53(a) provided for the punishment of anyone
who “knowingly manufactures, sells, transfers, possesses, or
transports in the State any butterfly knife,” 

(a) (1999), House Bill 2342 sought to narrow that
prohibition to cover only those who “knowingly carr[y]
concealed on the person in the State any butterfly knife,”
H.B. 2342, 32nd Leg., Reg. Sess. (Haw. 2024) (as
introduced). The bill was referred to the House Committee
on Judiciary and Hawaiian Affairs.
    On February 22, 2024, the Committee voted
unanimously to recommend that House Bill 2342 be passed
with an amendment. As amended, the bill proposed to revise
section 134-53(a) to provide: “Whoever knowingly carries
concealed on the person, or in a bag or other container
carried by the person, any butterfly knife shall be guilty of a
misdemeanor.” H.B. 2342, 32nd Leg., Reg. Sess. (Haw.
2024) (as reported by the H. Comm. on Judiciary &
Hawaiian Affs.).
    The same day, we granted rehearing en banc, thereby
vacating the three-judge panel’s opinion. Teter v. Lopez, 

 (9th Cir. 2024).
    On March 5, the Hawaii House of Representatives
passed House Bill 2342 with the amendment proposed by the
Committee. The Hawaii Senate subsequently passed the bill
as well, and on May 13, the Governor signed it into law.

; see 

 § 6 (amending section 153-
54). The bill took effect immediately upon its enactment. Id.
§ 10.
                              II
    We begin—and end—by considering whether we have
jurisdiction. Article III of the Constitution limits the
                        TETER V. LOPEZ                      11


jurisdiction of federal courts to actual “Cases” and
“Controversies.” U.S. Const. Art. III, § 2. We have no
authority to “decid[e] legal disputes or expound[] on law in
the absence of such a case or controversy.” Already, LLC v.
Nike, Inc., 

 (2013).
     To permit us to exercise jurisdiction, “an actual
controversy must be extant at all stages of review, not merely
at the time the complaint is filed.” Arizonans for Off. Eng. v.
Arizona, 

 (1997) (quoting Preiser v. Newkirk,

 (1975)). “No matter how vehemently the
parties continue to dispute the lawfulness of the conduct that
precipitated the lawsuit, the case is moot”—and we lack
jurisdiction—“if the dispute ‘is no longer embedded in any
actual controversy about the plaintiffs’ particular legal
rights.’” Already, LLC, 

 (quoting Alvarez v.
Smith, 

 (2009)).
    That does not mean “that a defendant may ‘automatically
moot a case’ by the simple expedient of suspending its
challenged conduct after it is sued.” FBI v. Fikre, 

 (2024) (quoting Already, LLC, 

).
When a defendant has voluntarily ceased its challenged
conduct, “[a] controversy may remain to be settled” because
“[t]he defendant is free to return to [her] old ways” and may
do so once the litigation ends. United States v. W. T. Grant
Co., 

 (1953); see also Steel Co. v. Citizens
for a Better Env’t, 

 (1998). And as the
Supreme Court recently made clear, that is true “for
governmental defendants no less than for private ones.”
Fikre, 

.
   But at the same time, we have held that “the repeal,
amendment, or expiration of challenged legislation is
generally enough to render a case moot and appropriate for
12                      TETER V. LOPEZ


dismissal.” Board of Trs. of Glazing Health & Welfare Tr.,

. After all, in a case such as this one, the
defendant is an executive official (here, the Hawaii Attorney
General), and a change in the statute results from the actions
of an independent branch of government (here, the Hawaii
Legislature) rather than from the defendant. See Chemical
Producers & Distribs. Ass’n v. Helliker, 

, 879
(9th Cir. 2006) (“[L]egislation is attributed to the legislature
alone.”), overruled on other grounds by Board of Trs. of
Glazing Health & Welfare Tr., 

. The
Attorney General has ceased to enforce the challenged law
because it no longer exists; whether or not she might wish to
resume enforcement, she cannot do so unless the Hawaii
Legislature first reenacts it.
    The realities of the legislative process—a process
requiring coordinated action by a multi-member body
representing a diverse array of interests—make it unlikely
that a legislature will strategically moot a case only to
“return to [its] old ways” when the litigation is over. W. T.
Grant Co., 

. Specifically, the time and
resource constraints of the legislative process, combined
with the multiplicity of legislative priorities, suggest that
legislatures lack the agility and coordination that would be
required to strategically moot litigation. Accordingly, we
“presume that the repeal, amendment, or expiration of
legislation will render an action challenging the legislation
moot, unless there is a reasonable expectation that the
legislative body will reenact the challenged provision or one
similar to it.” Board of Trs. of Glazing Health & Welfare Tr.,

. The presumption can be overcome by
showing, for example, that the legislative body has
announced its intention to reenact the law at issue. City of
Mesquite v. Aladdin’s Castle, Inc., 

, 289 & n.
                        TETER V. LOPEZ                       13


11 (1982). But when the presumption applies, it satisfies the
defendant’s burden to show that enforcement “cannot
‘reasonably be expected to recur.’” Fikre, 

(quoting Friends of the Earth, Inc. v. Laidlaw Environmental
Services (TOC), Inc., 

 (2000)).
     Because the Hawaii Legislature amended section 134-
53(a) while this litigation was pending, we presume that
Teter’s challenge to that statute is moot, and we ask whether
Teter has rebutted that presumption by establishing that there
is “a reasonable expectation that the legislative body is likely
to enact the same or substantially similar legislation in the
future.” Board of Trs. of Glazing Health & Welfare Tr., 

. Hawaii has already passed an amended law, so
our inquiry is whether the law as amended is “substantially
similar” to the original. 

 We assess substantial similarity
by reference to the claims advanced in the litigation, asking
whether the amended law burdens the plaintiff “in the same
fundamental way” as the original. Northeastern Fla.
Chapter of Associated Gen. Contractors of Am. v. City of
Jacksonville, 

 (1993); see also 

 at 662
n.3.
    Section 134-53(a), as amended, does not burden Teter
“in the same fundamental way” as the prior statute and
therefore is not “substantially similar” in any respect that is
relevant to this litigation. The complaint challenged “the
applicable Hawaii laws which prevent [Teter] from owning
butterfly knives,” and it sought a declaration that the State’s
“ban on the ownership of butterfly knives violates the
Second Amendment.” The amended statute does not
prohibit—or even limit—the ownership of butterfly knives.
The complaint further sought an injunction against the
State’s “policies generally banning the acquisition,
possession, carrying and use of butterfly knives.” Those
14                      TETER V. LOPEZ


policies also no longer exist: The amended statute does not
restrict the acquisition, possession, and use of butterfly
knives, except insofar as a different subsection now prohibits
their possession or use by someone “engaged in the
commission of a separate felony or misdemeanor,” 

(b), a prohibition Teter does not
challenge. Nor does the statute prohibit carrying butterfly
knives. Although it does prohibit carrying concealed
butterfly knives, Teter made clear in the district court that he
sought “the right to carry [a butterfly knife] openly.” Under
the amended statute, he has that right.
    We conclude that section 134-53(a) “has been
‘sufficiently altered so as to present a substantially different
controversy from the one the District Court originally
decided.’” Northeastern Fla. Chapter of Associated Gen.
Contractors of Am., 

 n.3 (quoting 

(O’Connor, J., dissenting)). Or, put another way, the
statutory amendment gave Teter “everything [he] hoped to
achieve” in this litigation. Chemical Producers & Distribs.
Ass’n v. Helliker, 463 F.3d at 876. Because we can grant no
further relief, the case is moot, and we lack Article III
jurisdiction. See Diffenderfer v. Central Baptist Church of
Miami, Fla., Inc., 

, 414–15 (1972) (per curiam)
(describing “a declaratory judgment that the now repealed
[statute] is unconstitutional” as a form of relief that “is, of
course, inappropriate now that the statute has been
repealed”).
   Teter resists that conclusion, arguing that two challenges
to section 134-53(a) survive its amendment. Neither
challenge is properly before us.
   First, Teter argues that the amended law effectively
prohibits carrying butterfly knives in any manner. That is so,
                        TETER V. LOPEZ                      15


he says, because “butterfly knives cannot be carried openly
in any realistic manner,” so a prohibition on concealed
carrying amounts to “a de facto ban on the carry of butterfly
knives.” In support of that assertion, he points to a statement
in his expert’s declaration that butterfly knives “do not have
a clip because it would interfere with manipulating the
handles.” Teter suggests that a clip is necessary to carry a
butterfly knife, but the expert never said that. The expert was
not asked to opine on the distinction between concealed
carrying and open carrying, and nothing in the expert’s
declaration establishes that a person could not openly carry
a butterfly knife.
    Second, Teter argues that even a prohibition on
concealed carrying violates the Second Amendment. As we
have already explained, Teter expressly waived that theory
below. The district court accurately captured Teter’s
position: The prohibition on butterfly knives was
“unconstitutional only as applied to law-abiding citizens
seeking to possess butterfly knives in their homes or to
openly carry them in public.” It was not until oral argument
before the en banc court that Teter argued that a prohibition
on concealed carrying is unconstitutional. We do not
consider Teter’s newly articulated challenge to the
prohibition on carrying concealed butterfly knives.
    In saying that these challenges are not properly before
us, we do not suggest that they are forever foreclosed. Until
the statute was amended, Teter had no reason to develop the
record on the distinction between concealed and open
carrying; with further factual development, perhaps he could
show that openly carrying a butterfly knife is indeed
impractical. Likewise, Teter’s waiver of a challenge to the
ban on concealed carrying might arguably be excused by the
change in law that occurred during the pendency of this
16                      TETER V. LOPEZ


appeal. Cf. Romain v. Shear, 

 (9th Cir.
1986) (explaining that although we “will generally not
consider an issue raised for the first time on appeal,” we may
do so “when a new issue arises while [the] appeal is pending
because of a change in law”). At the time, Teter’s position
was dictated by controlling circuit precedent establishing
that “the Second Amendment does not protect, in any
degree, the right of a member of the general public to carry
a concealed weapon in public.” Peruta, 

. But
that precedent is now subject to reconsideration in light of
Bruen. See Baird v. Bonta, 

 (9th Cir.
2023) (recognizing Bruen’s abrogation of Peruta).
     This is precisely the situation the Supreme Court
confronted in New York State Rifle & Pistol Ass’n v. City of
New York. There, the plaintiffs challenged a New York City
rule restricting the transport of firearms, and, during the
pendency of the litigation, the City amended its rule. 

. The amendment mooted the plaintiffs’ “claim
for declaratory and injunctive relief with respect to the City’s
old rule,” but the plaintiffs argued “that the new rule may
still infringe their rights.” 

. The Court explained
that the “ordinary practice in disposing of a case that has
become moot on appeal is to vacate the judgment with
directions to dismiss.” 

 (quoting Lewis, 

).
But “where the mootness is attributable to a change in the
legal framework governing the case, and where the plaintiff
may have some residual claim under the new framework that
was understandably not asserted previously,” a court should
instead “vacate the judgment and remand for further
proceedings in which the parties may, if necessary, amend
their pleadings or develop the record more fully.” 

(quoting Lewis, 

).
                        TETER V. LOPEZ                      17


    We follow that course here and remand without
instructions to dismiss. On remand, Teter may assert
whatever claims remain available to him under the new
statute. We express no view on the appropriate resolution of
the Second Amendment issues those claims might present.
“[T]he cardinal principle of judicial restraint—if it is not
necessary to decide more, it is necessary not to decide
more—counsels us to go no further.” PDK Labs. Inc. v.
United States Drug Enf’t Admin., 

 (D.C.
Cir. 2004) (Roberts, J., concurring in part and concurring in
the judgment).
    There remains the question whether to vacate the district
court’s judgment. When a case has become moot, our normal
practice is to vacate the judgment. See New York State Rifle
& Pistol Ass’n, 

; United States v.
Munsingwear, Inc., 

 (1950); Chemical
Producers & Distribs. Ass’n, 463 F.3d at 878. Teter asks us
to depart from that practice because, under the “equitable
tradition of vacatur,” it is not always appropriate to vacate
“when mootness results from unilateral action of the party
who prevailed below.” United States Bancorp Mortg. Co. v.
Bonner Mall P’ship, 

 (1994). But the party
who prevailed below is the Attorney General, an executive
officer. As we have already explained, she did not cause this
case to become moot; the Hawaii Legislature did. See Valero
Terrestrial Corp. v. Paige, 

 (4th Cir. 2000)
(“[D]efendant state executive officials are ‘in a position akin
to a party who finds its case mooted by “happenstance,”
rather than events within its control.’”) (quoting National
Black Police Ass’n v. District of Columbia, 

 (D.C. Cir. 1997)); accord Khodara Env’t, Inc. ex rel.
Eagle Env’t L.P. v. Beckman, 

 (3d Cir.
18                      TETER V. LOPEZ


2001). The mootness of the case is therefore not attributable
to any action of the prevailing party.
     Teter suggests that the Hawaii Legislature’s amendment
of section 134-53(a) represents an effort to manipulate our
jurisdiction. We presume “that a legislative body is acting in
good faith in repealing or amending a challenged legislative
provision.” Board of Trs. of Glazing Health & Welfare Tr.,

. Although that presumption is rebuttable,
Teter has identified nothing in the record—or anywhere else,
for that matter—supporting his assertion that the Legislature
amended section 134-53(a) for an improper purpose or that
it is likely to reenact the prohibitions that existed when this
case began. To the extent that the Legislature was motivated
by the panel’s decision to adopt a statute that would be more
likely to be upheld, we see nothing nefarious in its effort to
comply with the Supreme Court’s or this court’s decisions.
See American Libr. Ass’n v. Barr, 

 (D.C.
Cir. 1992) (“[P]assing legislation designed to repair what
may have been a constitutionally defective statute . . . .
represents responsible lawmaking, not manipulation of the
judicial process.”); see also National Black Police Ass’n,

. In any event, Teter’s opposition to vacatur
makes little sense given that Teter lost in the district court,
so any ongoing effect the judgment might have could only
be harmful to him.
     The parties shall bear their own costs on appeal.
     VACATED and REMANDED.
                        TETER V. LOPEZ                     19


VANDYKE, Circuit Judge, dissenting in part:

     When a three-judge panel unanimously and correctly
declared Hawaii’s complete ban on butterfly knives
unconstitutional, Teter v. Lopez, 

, 954–55
(9th Cir. 2023), this circuit responded the way it always does
when a Second Amendment claim is vindicated: we called
the case en banc. And after the en banc vote played out in
all-too-predictable fashion, the Chief Judge issued an
administrative order not only granting en banc review, but
also vacating the panel’s opinion. Teter v. Lopez, 

 (9th Cir. 2024).
     Soon after we took this case en banc, Hawaii narrowed
the scope of its butterfly knife prohibition and then argued
its amendments mooted plaintiffs’ claims. Concluding the
amendments allow the very conduct—and conveniently only
that conduct—in which plaintiffs sought to engage, the en
banc panel now concludes the case is moot and remands for
plaintiffs to pursue any residual challenge they may have to
the amended law. See N.Y. State Rifle & Pistol Ass’n v. City
of New York, 

 (2020).
    As the en banc proceedings in this case demonstrate, this
court’s practice of automatically vacating panel opinions
upon the grant of rehearing en banc creates perverse
incentives for government defendants. By amending their
challenged laws only after this court grants rehearing en
banc, parties like Hawaii can strategically deploy mootness
to lock in the effect of our auto-vacatur without the risk of
losing on the merits before the en banc court. Such mischief
is strikingly like—basically the mirror image of—the kind
our Munsingwear precedents are designed to discourage.
20                      TETER V. LOPEZ


    I agree with Judge Collins that this case is not moot for
the general reasons that he provides in his separate dissent.
I write separately to state that even if the majority was
correct that this case is moot, the remedy that Judge Collins
identifies—reinstating the panel’s judgment and remanding
the case—would still be the proper remedy here. The
majority’s business-as-usual response to mootness here is
woefully inadequate to eliminate the perverse incentives
outlined above.
    Thus, to actually discourage parties from using mootness
strategically, I would (1) reinstate the panel opinion by
vacating our prior vacatur order, (2) address those legal
questions that are indisputably relevant to any possible
reformulation of plaintiffs’ claims on remand, and
(3) remand the matter to the district court for plaintiffs to
pursue additional challenges to Hawaii’s reformulated
butterfly knife law, should they choose to do so. For these
reasons, I concur in the decision to remand this case for
further proceedings, but I respectfully dissent from that
portion of the majority’s decision that simply vacates the
district court’s judgment without taking any additional
action directed at our prior order ministerially vacating the
panel opinion.
                              I.
    Under United States v. Munsingwear, Inc., vacatur “is
commonly utilized … to prevent a judgment, unreviewable
because of mootness, from spawning any legal
consequences.” 

 (1950). “Vacatur is in order
when mootness occurs through … the ‘unilateral action of
the party [that] prevailed in the lower court.’” Arizonans for
Off. Eng. v. Arizona, 

, 71–72 (1997) (quoting
U.S. Bancorp Mortg. Co. v. Bonner Mall P’ship, 513 U.S.
                             TETER V. LOPEZ                              21


18, 23 (1994)). Such vacaturs are meant to prevent
procedural gamesmanship from parties hoping to “secur[e]
‘a favorable judgment, tak[e] voluntary action that moots the
dispute, and then retain[] the benefit of the judgment.’” City
& Cnty. of San Francisco v. U.S. Citizenship & Immigr.
Servs., 

 (9th Cir. 2021) (VanDyke, J.,
dissenting) (quoting Arizonans for Off. Eng., 

).
    On the other hand, “[b]oth the Supreme Court and this
court have recognized exceptions to this practice if the party
seeking appellate relief … is the cause of subsequent
mootness.” NASD Disp. Resol., Inc. v. Jud. Council of State
of Cal., 

, 1069 (9th Cir. 2007) (citation and
internal quotations omitted). The Supreme Court, for
example, will not vacate a lower court judgment “[w]here
mootness results from settlement” because “the losing party
has voluntarily forfeited his legal remedy by the ordinary
processes of appeal or certiorari, thereby surrendering his
claim to the extraordinary equitable remedy of vacatur.”
U.S. Bancorp, 

. In other words, “if a party lost
below, but does something intentional to moot its case while
the appeal is pending, you don’t need to worry about that
losing party deliberately mooting the case” to retain the
benefit of a favorable judgment it never had. City & Cnty.
of San Francisco, 

 (VanDyke, J.,
dissenting).1

1
  Of course, the mere fact that a party lost below is not always a failproof
guarantee that it will have no incentive to strategically moot a case. That
is why I suggested in one of my several dissents in the public charge
litigation that when an administration abandons its defense of a rule for
purposes of political expediency, we should discourage its efforts to
engage in “rulemaking-by-collusive-acquiescence” by using
Munsingwear vacatur. City & Cnty. of San Francisco, 

,
752–55 (VanDyke, J., dissenting).
22                      TETER V. LOPEZ


    Thus, in both the lower court judgments that appellate
courts choose to vacate and those they leave in place, it is
sufficiently clear that our Munsingwear precedents are
predominantly motivated by a desire to discourage parties
from taking strategic action designed to lock in the effect of
a favorable judgment. See U.S. Bancorp, 

(“The principal condition to which we have looked is
whether the party seeking relief from the judgment below
caused the mootness by voluntary action.”).
    The procedural posture of this case, of course, differs in
a few ways from the typical case in which Munsingwear
vacatur would be in order. For one, Munsingwear is usually
meant to resolve the equitable problems caused by
intervening mootness while the judgment of a lower court is
appealed to a higher court. But here, where the mootness
has occurred entirely within the confines of the Ninth Circuit
as the case passed from the panel into the hands of the en
banc court, the panel opinion was not “appealed” to a higher
court.
    And second, like the cases to which the U.S. Bancorp
exception to vacatur usually applies, it was Hawaii, the loser
before the panel and “the party seeking [en banc] relief,”
that was “the cause of subsequent mootness.” NASD Dispute
Resolution, 488 F.3d at 1069. Usually, such a posture would
mean that the equities necessitating Munsingwear vacatur do
not apply.
    There is, however, at least one more unique aspect of this
case that, notwithstanding the differences outlined above,
weighs strongly in favor of vacatur: this court’s curious
practice of automatically vacating a panel opinion upon the
grant of rehearing en banc. Under normal circumstances,
when a case is appealed from district to circuit court or when
                        TETER V. LOPEZ                      23


the Supreme Court grants review of a circuit court decision
through certiorari, the opinion and judgment of the lower
court remain in place on appeal. Even if an appellate court
grants a temporary reprieve in the form of a stay, such orders
do not wholly vacate a lower court judgment. Instead, they
merely suspend enforcement of that judgment for a time,
usually until appellate review is complete. See, e.g., Ohio v.
EPA, 

, 300 (2024) (“The applications for a stay
… are granted. Enforcement … shall be stayed pending the
disposition of the applicants’ petitions for review ….”).
    This court’s en banc procedure, however, departs from
this usual appellate process. Instead of leaving the results of
the three-judge panel in place or even just staying the results
pending en banc review, this court has developed a practice
of automatically vacating the three-judge panel’s opinion in
the order granting en banc review. Usually, this task is
accomplished in just a few short words—“The three-judge
panel opinion is vacated,” and nothing else—without
citation to any authority upon which the vacatur order is
grounded. See, e.g., Teter, 

; see also United
States v. Duarte, 

 (9th Cir. 2024) (same). The
judges who have been drawn for the en banc panel do not
vote on whether to vacate the panel opinion—the court just
issues a ministerial order under the Chief Judge’s name
taking the case en banc and vacating the panel opinion. See,
e.g., Teter, 

.
                              II.
    Before explaining why these circumstances implicate
Munsingwear, it is worth noting that the absence of any
citation supporting our en banc vacaturs is ultimately
unsurprising. That is because our vacatur practice (1) is a
relatively new addition to our en banc procedure that
24                       TETER V. LOPEZ


(2) until very recently was not explicitly grounded in the
plain text of any rule or general order of this circuit and (3) is
also arguably inconsistent with our limited en banc model.
     First, such vacaturs appear to have a limited historical
pedigree. As far as I can tell, we haven’t always
automatically vacated panel opinions upon the grant of
rehearing en banc. In fact, some of our earlier rules carried
a strong presumption against such vacaturs. Before 1987,
for example, our General Orders provided that “[t]he opinion
of the three-judge panel should not be withdrawn absent
exceptional circumstances” (emphasis added). See, e.g.,
Dir., Off. of Workers’ Comp. Programs, U.S. Dep’t of Lab.
v. Cargill, Inc., 

 (9th Cir. 1983) (vacating
and withdrawing only after disapproving of the panel
opinion on its merits in published opinion). That rule was
later modified to provide that “the en banc panel may
withdraw the panel opinion” “[w]hen the court votes to take
a case en banc following the publication of a panel decision,”
(emphasis added), but even then, vacatur remained
discretionary and required a separate decision of the limited
en banc panel. See, e.g., Lear Siegler, Inc., Energy Prods.
Div. v. Lehman, 

 (9th Cir. 1989)
(withdrawing part of a panel opinion after disagreeing with
its rationale).
    Even after the court began to act directly on panel
opinions in the orders granting en banc review, it still
stopped short of explicitly requiring vacatur. See, e.g., In re
Schwartz-Tallard, 

 (9th Cir. 2014) (“The
three-judge panel opinion shall not be cited as precedent by
or to any court of the Ninth Circuit.”). This “shall not be
cited as precedent” language, which has since fallen out of
use, was more consistent with our rules than the “opinion is
vacated” language we currently employ. When the three-
                         TETER V. LOPEZ                        25


judge-panel opinion here was vacated, General Order 5.5d,
for example, read:

        If a majority of the judges eligible to vote on
        the en banc call votes in favor of en banc
        consideration, the Chief Judge shall enter an
        order taking the case en banc pursuant to
        Circuit Rule 35-3. The three-judge panel
        opinion shall not be cited as precedent by or
        to this Court or any district court of the Ninth
        Circuit, except to the extent adopted by the en
        banc court.

(emphasis added). Our advisory committee notes included
substantially similar language. Circuit Advisory Committee
Note 3 to Rule 35-1 to 35-3 (“The three-judge panel opinion
shall not be cited as precedent … except to the extent
adopted by the en banc court.”). And while we have
renumbered and revised the relevant advisory committee
note to replace the “shall not be cited as precedent” language
with a provision explicitly noting that “[t]he three-judge
panel opinion is vacated,” such changes did not go into effect
until December 2024.2
    Finally, as a purely practical matter, our automatic
vacatur practice also puts the cart before the horse in that it
prescribes a certain remedy—vacatur—without any
guarantee that the en banc court will disagree with the three-
judge panel on the merits. While other circuits auto-vacate
panel opinions when granting rehearing en banc, such an
approach makes sense when every active judge both votes

2
  See United States Court of Appeals for the Ninth Circuit, Public
Comment Package March 2024 (Mar. 25, 2024), available at:
https://www.ca9.uscourts.gov/rules/proposed-rules-and-amendments/.
26                      TETER V. LOPEZ


whether to grant en banc rehearing and sits on the en banc
panel. See, e.g., 5th Cir. R. 41.3 (“Unless otherwise
expressly provided, the granting of a rehearing en banc
vacates the panel opinion and judgment of the court and
stays the mandate.”). But unlike the approach of other
circuits, in which the initial en banc vote may presumptively
predict the views of the full en banc court, our circuit’s
unique practice of randomly drawing an eleven-member
panel means the en banc panel’s views may not always
match the full court’s initial vote. Even assuming the full
court mostly votes to rehear cases to correct perceived errors,
our automatic vacatur practice cannot readily account for
these potential mismatches between the full court’s and the
en banc panel’s views of the merits.
                             III.
    But more important than these historical and practical
problems with our automatic vacatur practice are the
perverse incentives it creates for parties like Hawaii. As
noted above, a core function of Munsingwear vacatur is “to
prevent a party from securing ‘a favorable judgment,
tak[ing] voluntary action that moots the dispute, and then
retain[ing] the benefit of the judgment.’” City & Cnty. of
San Francisco, 

 (VanDyke, J., dissenting)
(quoting Arizonans for Off. Eng., 

). All one
must do is replace “judgment” with “vacatur” in the
above-quoted passage, and it perfectly describes the
circumstances presented here.
   For government defendants trying to rid themselves of
adverse panel precedent, playing the panel-erasure game is
easy: all they need to do is wait until after they have secured
                             TETER V. LOPEZ                             27


rehearing en banc,3 thereby ridding themselves of the panel
opinion by automatic vacatur, and then strategically amend
challenged laws to moot the plaintiffs’ claims. Then, such
government defendants can sit back and relax, assured that
their amendments will shield them from the risks associated
with any en banc reconsideration of the merits. And once
the en banc panel compliantly plays along and dismisses the
case, the government is then free to reenact the very law the
panel only recently declared unconstitutional.
    If the government were really thinking hard about it, it
could even wait to learn the composition of the en banc panel
before making a final decision about whether to moot the
case. If the en banc panel draw seems favorable, the
government can take the opportunity to continue litigating
the en banc proceedings to their conclusion and obtain a
favorable decision on the merits. But if the en banc panel
draw appears bad from the government’s perspective, it can
then nullify its downside risk by strategically employing
mootness. Heads the government wins, tails the plaintiffs
lose. While one could hardly blame government attorneys

3
   For cases in the Ninth Circuit where the panel found the Second
Amendment violated, this step is usually accomplished simply by filing
a petition for rehearing en banc. Our court will bend over backward to
handle the rest. Actually, an en banc petition isn’t even necessary, since
our court is more than willing to sua sponte take such a case en banc even
when the government has not requested it. See, e.g., McDougall v. Cnty.
of Ventura, 

 (9th Cir. 2022) (granting en banc rehearing
after a sua sponte en banc call). Indeed, if the government does not want
a case like this to be taken en banc, it usually isn’t enough to simply not
file an en banc petition. The government has to beg our court not to sua
sponte take the case en banc. See, e.g., Order Directing Issuance of the
Mandate, Baird v. Bonta, No. 23-15016, Dkt. 42 (Nov. 15, 2023)
(surviving a sua sponte en banc call only after the government urged the
court not to rehear the case en banc).
28                            TETER V. LOPEZ


for deploying a winning gambit so simple to execute, this
court should not encourage it. See City of New York, 

, 364–70 (Alito, J., dissenting) (lamenting that
“the Court permitt[ed] [its] docket to be manipulated in a
way that should not be countenanced”).
    Citing our prior decision in Board of Trustees of Glazing
Health & Welfare Trust v. Chambers, 

 (9th
Cir. 2019) (en banc), the majority repeatedly asserts that we
must give a government’s voluntary cessation of challenged
conduct “more solicitude” than voluntary cessation by
private parties. First, as Judge Collins argues in his separate
dissent, the Supreme Court’s decision in FBI v. Fikre, 

 (2024), is irreconcilable with our court’s holding in
Glazing Health. And this en banc panel is not bound by
Glazing Health in any event. We easily could have
reconsidered whether states like Hawaii are entitled to
Glazing Health’s presumption of regularity (at least in
Second Amendment cases) given their well-documented
intransigent responses to the Supreme Court’s decision in
New York State Rifle & Pistol Ass’n, Inc. v. Bruen, 

 (2022).4


4
   Cf. State v. Wilson, 

, 453–54, 459 (Haw. 2024)
((1) criticizing Bruen for its “fuzzy ‘history and traditions’ test” that “ad-
libs a ‘history only’ standard,” “dismantles workable methods, and
creates “a dangerous way to look at the federal constitution,”
(2) asserting that “[h]istory by historians quickly debunked Heller’s
history,” and (3) applying the powerful “spirit of Aloha” to conclude the
Hawaii Constitution does not protect “an individual right to keep and
bear arms”); see also Governor Gavin Newsom, Governor Newsom
Responds to Supreme Court Decision on Concealed Carry (June 23,
2022),            https://www.gov.ca.gov/2022/06/23/governor-newsom-
responds-to-supreme-court-decision-on-concealed-carry/
                           TETER V. LOPEZ                            29


     But putting that aside, the issue of whether Hawaii was
entitled to “more solicitude” than any other litigant shouldn’t
make any difference because Hawaii’s motivations
ultimately should matter very little to the question of how
the court should remedy the problems caused by the
intervening mootness.         Whether Hawaii intended to
strategically moot this case for the reasons described above
or not, it has nevertheless taken all the steps necessary to
unfairly benefit from our automatic vacatur practice. Its
actions have wiped the record of any judicial declaration that
its absolute ban on butterfly knives was unconstitutional, and
now that these en banc proceedings have concluded, Hawaii
is free to reenact its prior, unconstitutional ban in toto—all
without so much as a single word of analysis from the en
banc court about the underlying merits.
    This situation cries out for some sort of equitable vacatur
under Munsingwear, but the question is: What to vacate?
Reflexively vacating the district court judgment as the
majority does today does not address the perverse incentives
created by our automatic vacatur practice and ironically only
perpetuates the kind of problem Munsingwear is designed to
address. But we need not artificially limit ourselves to
vacating only the district court judgment “[b]ecause
[vacatur] is rooted in equity,” and thus “the decision whether
to vacate turns on ‘the conditions and circumstances of the
particular case.’” Azar v. Garza, 

 (2018)

(characterizing Bruen as “reckless” and “radical” and asserting that his
“[a]dministration has been working closely with the Attorney General
and the legislature for months” to pass a slate of “16 new gun safety
bills” in response). These statements and many others like them
obviously undercut any judicial fiction that all state governments are
“acting in good faith” in response to the Supreme Court’s Second
Amendment jurisprudence.
30                      TETER V. LOPEZ


(per curiam) (citation omitted); see also Dilley v. Gunn, 

 (9th Cir. 1995) (“U.S. Bancorp makes clear
that the touchstone of vacatur is equity.”).
    To correct the perverse incentives created in
circumstances like these, I would simply vacate the order
that created those incentives in the first place: our prior
vacatur order itself. Doing so would have reinstated the
panel opinion, leaving it suitable to “be cited as precedent by
or to this Court,” General Order 5.5d, and therefore would
have removed any incentive to strategically misuse the
mootness limitations on this court’s Article III jurisdiction
in the manner described here.
    In addition to correcting the equitable incentives
associated with our en banc procedure, reinstating the panel
opinion would have had at least two additional ancillary
benefits. First, it would have facilitated the special
solicitude the majority gives to Hawaii’s voluntary
cessation. If we must assume that Hawaii “is acting in good
faith in repealing or amending” its prior law, Glazing Health,

, then presumably Hawaii would have no
objection to reinstating the panel’s opinion because, as it
argues in its suggestion of mootness, that opinion apparently
does not cast doubt on any part of its amended butterfly knife
ban. Or to put it more succinctly using the words of a certain
former president: “Trust, but verify.”
    And second, reinstating the panel opinion would have
assured that the panel’s lengthy and well-reasoned opinion
contributed to a developing legal issue that has recently
divided courts across the country. Compare Commonwealth
v. Canjura, 

 (Mass. 2024) (applying Bruen
to conclude that Massachusetts’s ban on carrying
switchblade knives violates the Second Amendment), with
                         TETER V. LOPEZ                        31


Knife Rights, Inc. v. Bonta, 

 (S.D. Cal.
Aug. 23, 2024) (reaching the opposite conclusion as to
California’s ban). “Judicial precedents,” after all, “‘are not
merely the property of private litigants,’ but are ‘valuable to
the legal community as a whole.’” Dickens v. Ryan, 

 (9th Cir. 2014) (en banc) (quoting U.S.
Bancorp, 

). We should not “force future
panels to duplicate our efforts by re-deciding issues we have
already resolved within the contours of [A]rticle III.” 

                               IV.
     Finally, one might object that by vacating our prior
vacatur order and reinstating the panel opinion, the en banc
court would be improperly making law in a moot case where
it had no jurisdiction to do so. That objection fails for at least
two reasons.
    To begin, this court has long recognized its authority to
take a moot case en banc solely for the purpose of vacating
a panel decision. See United States v. Payton, 

 (9th Cir. 2010) (recognizing that the “refusal to vacate
the [panel’s] decision after it has become moot … leaves
open the opportunity to seek an en banc rehearing for the
purpose of vacating our decision”). The decision whether to
do so, like any other question of Munsingwear vacatur, “is
within our discretion based on equity.” Dickens, 

 (quoting Payton, 

). One of the key
factors we look to in exercising such discretion is whether
the panel got it right. Parsons v. Ryan, 

, 572
n.1 (9th Cir. 2015) (Ikuta, J., dissenting) (“[W]e can vacate
the decision to avoid having the panel’s serious
misinterpretations of Supreme Court [Second Amendment]
jurisprudence become the law of our circuit.”). As present
circumstances make clear, there is just as much reason to
32                      TETER V. LOPEZ


equitably reinstate a correct panel opinion as there is reason
to equitably vacate an erroneous one.
    Complementing our ability to take a moot case en banc
and vacate it, our court recently reaffirmed that it also has
jurisdiction to issue an entire new opinion even after a case
becomes moot, provided that opinion explains the reasons
for a decision made prior to the event causing mootness.
United States v. Perez-Garcia, 

 (9th Cir. 2024).
Perez-Garcia gave four justifications for issuing its opinion
notwithstanding the intervening mootness, and each is
instructive here.
    First, like in Perez-Garcia, the Teter panel “heard and
conclusively resolved the merits of [the] appeal” when “no
party dispute[d] that [it] had jurisdiction.” 

.
Second, as litigated before the panel, Teter also “properly
present[ed] questions concerning … specific constitutional
rights,” and reinstating the panel order would “not take
further action on the merits of [plaintiffs’] claims. 

(quoting Armster v. U.S. Dist. Ct. for the Cent. Dist. of Cal.,

 (9th Cir. 1986)). Third, for all the
reasons described above, “equity weighs in favor of”
reinstatement, which provides a narrowly tailored solution
to a specific procedural problem posed by our en banc
practice. Id. at 1174. And fourth, the majority’s approach
will “likely force later panels to duplicate [the panel’s]
efforts while confronting the exact same issues,” an
especially salient concern “[i]n light of the extensive and
complicated historical analysis the Second Amendment …
demands.” Id.
   So if this court has authority to (a) take a moot case en
banc and vacate it and (b) issue an opinion in a moot case to
explain the basis for a prior order, then it is difficult to see
                        TETER V. LOPEZ                      33


why we would lack authority to reinstate a previously
vacated panel opinion under similar circumstances. We lack
the will, but not the authority.
    Ultimately, the proper remedy in cases involving this
kind of intervening mootness depends entirely on which way
a court’s default en banc rule points. If, as is currently the
case, administratively vacating the panel opinion upon the
grant of rehearing en banc is the norm, then restoring the
status quo after intervening mootness necessarily requires
equitable reinstatement. But if vacatur were not the norm,
then the court would have to respond to mootness by
considering whether a more traditional Munsingwear
vacatur was in order. The underlying judicial action and the
equitable considerations it is designed to serve are
essentially the same no matter which way the default rule
points. A wooden jurisdictional rule that countenances only
the latter remedy but not the former creates a one-way
mootness ratchet that serves no discernible purpose and
favors only strategic gamesmanship by government
defendants.
                              V.
    For the reasons explained above, this court’s “shoot first,
ask questions later” approach to panel opinion vacatur
creates unique and perverse incentives for government
defendants like Hawaii, who can strategically misuse the
mootness limitations on our Article III jurisdiction to rid
themselves of adverse panel precedent. Whatever Hawaii’s
motivations for what it did here, our court could and should
have mitigated the improper incentives for such mischief—
in this and future cases. All it had to do was adopt a
presumptive policy of vacating our prior vacatur orders,
thereby reinstating the precedential value of the panel’s
34                      TETER V. LOPEZ


opinion, whenever an en banc petitioner takes deliberate
action that moots a case after rehearing en banc is granted. I
respectfully dissent from our failure to do so.


COLLINS, Circuit Judge, with whom LEE, Circuit Judge,
joins, dissenting:

    Judge Lee and I were members of the three-judge panel
that unanimously held that Hawaii Revised Statutes § 134-
53(a)’s complete ban on the possession of butterfly knives
violated the Second Amendment “right of the people to keep
and bear Arms.” See Teter v. Lopez, 

 (9th
Cir. 2023), vacated on grant of rehearing en banc, 

 (9th Cir. 2024). The State of Hawaii has, at least for
now, acquiesced in the bottom-line practical result of that
vacated decision because, even as state officials sought and
obtained rehearing en banc, the Hawaii Legislature amended
the statute to eliminate the “outright ban[]” that was the basis
for our judgment invalidating the statute. Teter, 

. In my view, however, that development does not moot
this case, and I therefore respectfully dissent.
    With respect to Plaintiffs’ specific claim that a total ban
on butterfly knives violates the Second Amendment, I agree
that Hawaii’s amendment of the statute eliminates that
challenged prohibition and thereby grants Plaintiffs,
“outside of litigation,” complete relief as to that specific
issue. FBI v. Fikre, 

 (2024). But because
any such relief to Plaintiffs occurred only through the
voluntary action of the appropriate state actors (namely, the
Hawaii Legislature), it will moot that claim “only if the
defendant can show that the practice cannot reasonably be
expected to recur.” 

 (emphasis added) (simplified).
                       TETER V. LOPEZ                     35


Fikre held that this is a “formidable burden,” 

 (citation
omitted), and that “[i]n all cases”—even cases, such as
Fikre, that involve a governmental defendant—“it is the
defendant’s burden to establish that it cannot reasonably be
expected to resume its challenged conduct” and to make that
showing regardless of “whether the suit happens to be new
or long lingering, and whether the challenged conduct might
recur immediately or later at some more propitious
moment,” 

 (emphasis added) (simplified). Fikre
thus squarely overrules our prior caselaw under which we
“treat the voluntary cessation of challenged conduct by
government officials with more solicitude than similar
action by private parties.” Board of Trs. of Glazing Health
& Welfare Tr. v. Chambers, 

 (9th Cir.
2019) (en banc) (simplified); see also Brach v. Newsom, 

, 12–13 (9th Cir. 2022) (en banc) (same). The
majority recognizes as much, because it acknowledges that,
“as the Supreme Court recently made clear,” the principle
that a defendant’s voluntary cessation of challenged conduct
may not be enough to moot a case “is true ‘for governmental
defendants no less than for private ones.’” See Opin. at 11
(quoting Fikre, 

).
    The majority nonetheless insists that voluntary-cessation
principles do apply differently here, because the relevant
governmental actor is the state legislature and not an
executive official or an administrative agency. In reaching
that conclusion, the majority relies on Chambers, in which
we adopted and applied a more government-friendly version
of the voluntary-cessation doctrine in the context of a
legislative repeal. See Chambers, 

. But in
my view, Chambers does not survive Fikre.
    In Chambers, we began by stating that we afford
“solicitude” to governmental actors when we apply the
36                          TETER V. LOPEZ


voluntary-cessation doctrine, and we concluded that “[f]or
this reason, the repeal, amendment, or expiration of
challenged legislation is generally enough to render a case
moot and appropriate for dismissal.” 

(emphasis added).1 We then acknowledged that, in two
cases, the Supreme Court had nonetheless invoked the
voluntary-cessation doctrine in holding that a repeal of the
challenged law did not moot the dispute in question. 

(citing City of Mesquite v. Aladdin’s Castle, Inc., 

 (1982); Northeastern Fla. Chapter of Associated Gen.
Contractors of Am. v. City of Jacksonville, 

(1993)). We observed that our caselaw construing City of
Mesquite and Northeastern Florida had been inconsistent,
with some decisions applying conventional voluntary-
cessation standards and others applying a strong
presumption that a legislative repeal of the challenged law
moots the dispute. Chambers, 

. With little
analysis, we then resolved that intra-circuit conflict by
adopting what we described as the near-unanimous view of
the other circuits, which required “evidence that [the
government] plans to or already has reenacted the challenged
law or one substantially similar.” 

 (citation
omitted). We therefore held that “legislative actions should
not be treated the same as voluntary cessation of challenged

1
  In support of this syllogism, we cited three cases, but none of them
actually supported it. Two of them held that the amendment or repeal of
the statute in question mooted the dispute, but without considering or
even mentioning the voluntary-cessation doctrine (presumably because
no party invoked it as a ground for avoiding mootness). See Burke v.
Barnes, 

 (1987); Kremens v. Bartley, 

,
127–28 (1977). In the third case, the only mootness exception invoked
was the distinct exception for disputes that are capable of repetition but
evading review; voluntary cessation was again not discussed. See Lewis
v. Continental Bank Corp., 

, 481–82 (1990).
                        TETER V. LOPEZ                     37


acts by a private party.” Id. at 1199. Rather, we said, “we
should assume that a legislative body is acting in good faith
in repealing or amending a challenged legislative provision,
or in allowing it to expire.” Id. We therefore held that, “in
determining whether a case is moot, we should presume that
the repeal, amendment, or expiration of legislation will
render an action challenging the legislation moot, unless
there is a reasonable expectation that the legislative body
will reenact the challenged provision or one similar to it.”
Id. We imposed on the party resisting mootness—typically
a private plaintiff—the burden to affirmatively show that
“there is a reasonable expectation of reenactment.” Id.
    Chambers was doubtful as an original matter, because
both City of Mesquite and Northeastern Florida applied
ordinary voluntary-cessation principles in holding that the
repeals in those cases did not moot the disputes. See
Northeastern Fla., 508 U.S. at 661–63; City of Mesquite, 455
U.S. at 289–90. There is no language in either decision
suggesting that different voluntary-cessation standards apply
to legislative repeals or that the party resisting mootness in
such a case has an affirmative burden to show a reasonable
expectation of re-enactment.         It was the dissent in
Northeastern Florida that argued for a presumption that
legislative bodies act in good faith. Northeastern Fla., 

 (O’Connor, J., dissenting). Notably, in making
that argument, the dissent itself conceded that City of
Mesquite had “required the government to establish that it
cannot be expected to reenact repealed legislation before [the
Court] w[ould] dismiss the case as moot.” 

 (emphasis
added). Moreover, there is no need for any such special rules
governing legislative repeals. If—as Chambers seemed to
think—a legislature’s repeal is unlikely, in most cases, to
have anything to do with attempting to moot a particular
38                      TETER V. LOPEZ


lawsuit, then the government typically should have little
difficulty carrying its burden to show that re-enactment of
similar legislation “cannot reasonably be expected” to occur.
Fikre, 

. But Chambers’ re-allocation of the
burden of proof concerning mootness in legislative-repeal
cases makes a decisive difference in matters—such as this
one—in which there are strong reasons to suspect that the
very lawsuit at hand was in fact the impetus for the
legislative amendment.
    In addition to being questionable as an original matter,
Chambers is further vitiated by Fikre in three respects. First,
Chambers’ presumption that a legislative repeal moots a
challenge to the repealed statute rested dispositively on our
adherence to the general principle that governmental actors
should receive “solicitude” in our application of the
voluntary-cessation doctrine. Chambers, 

(citation omitted). As I have explained, however, Fikre
rejected that fundamental premise, squarely holding that the
governmental defendant in that case had to satisfy the same
“formidable burden” required by settled voluntary-cessation
principles. Fikre, 

 (citation omitted).
Second, in pointing to precedents in which a governmental
defendant had failed to “satisf[y] that formidable standard,”
the Court in Fikre specifically cited City of Mesquite,
thereby further confirming (as I have argued above) that City
of Mesquite applied the same conventional voluntary-
cessation standards that apply in other contexts. 

.
Third, Fikre expressly stated that “[i]n all cases,” the burden
to show whether the challenged conduct can “reasonably be
expected” to recur is on “the defendant[].” 

 Fikre thus
rules out the burden-shifting we adopted in Chambers.
    Although the majority expressly reaffirms Chambers’
rule that a legislative repeal creates a presumption that the
                        TETER V. LOPEZ                       39


case is moot, the majority does not rely on Chambers’
discredited view that, with respect to the voluntary-cessation
doctrine, governmental defendants should generally be
given a “solicitude” that is denied to private defendants.
That broader view, as explained earlier, is flatly contrary to
Fikre, as the majority implicitly recognizes. The majority
instead relies on a substitute rationale for the Chambers
presumption, but it fares no better.
     According to the majority, “[t]he realities of the
legislative process—a process requiring coordinated action
by a multi-member body representing a diverse array of
interests—make it unlikely that a legislature will
strategically moot a case only to return to its old ways when
the litigation is over.” See Opin. at 12 (emphasis added)
(simplified). But, as I noted above in discussing Chambers,
this is simply another way of saying that, in the context of a
legislative repeal, a state defendant should ordinarily have
little difficulty, as a practical matter, carrying the burden of
proof imposed on it by the voluntary-cessation doctrine.
That predictive judgment provides no basis for creating a
special legal rule in which the state defendant may invoke a
presumption as a substitute for carrying its affirmative
burden of proving mootness, thereby shifting the burden of
proof to the plaintiff. Yet that is precisely what the majority
does when it insists that a state defendant can simply rely on
the Chambers presumption to “satisf[y],” see Opin. at 13
(emphasis added), the defendant’s “formidable burden” of
showing that enforcement of the challenged law “cannot
reasonably be expected to recur,” Fikre, 

(simplified). For the reasons I have explained, that special
burden-shifting framework in the legislative context, which
the majority reaffirms, cannot be reconciled with Fikre.
40                         TETER V. LOPEZ


    Accordingly, I conclude that Fikre abrogates Chambers
and that it is the Appellees’ burden here to establish that re-
enactment of a similar law cannot reasonably be expected to
occur. They have not carried that formidable burden in the
“Suggestion of Mootness” they filed in this court, which
instead relies dispositively on Chambers’ shifting of the
burden. Because this case has not been shown to be moot, I
would proceed to the merits.
    On the merits, I adhere to the views expressed in the
panel opinion, including specifically its holdings that
(1) “bladed weapons facially constitute ‘arms’ within the
meaning of the Second Amendment,” Teter, 

;
(2) to the extent that Appellees contend that butterfly knives
may be proscribed because they fall within the “historical
tradition of prohibiting the carrying of dangerous and
unusual weapons,” District of Columbia v. Heller, 

 (2008) (simplified), Appellees “bear[] the burden
of proof” on that issue, Teter, 

; and
(3) Appellees have failed to carry that burden, 

 at 954–55.
I therefore would reinstate the panel’s judgment reversing
the district court’s judgment and remanding the case.2
Because the majority does otherwise, I respectfully dissent.




2
 That would leave open, for remand, any issue of a possible amendment
of the complaint to specifically address whether the amended statute’s
more narrowly drawn provisions also violate the Second Amendment. I
express no view on that point.