Cynthia Brown; Carlos Buford; Jenny Sue Rowe v. David Yost, in his official capacity as Ohio Attorney General

RECOMMENDED FOR PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 24a0258p.06
UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT
CYNTHIA BROWN; CARLOS BUFORD; JENNY SUE
ROWE,
Plaintiffs-Appellants,
v.
DAVID YOST, in his official capacity as Ohio Attorney
General,
Defendant-Appellee.
┐
│
│
│
│
│
│
│
│
┘
No. 24-3354
On Petition for Rehearing En Banc
United States District Court for the Southern District of Ohio at Columbus.
No. 2:24-cv-01401—James L. Graham, District Judge.
Argued En Banc: October 30, 2024
Decided and Filed: November 21, 2024
Before: SUTTON, Chief Judge; MOORE, CLAY, GRIFFIN, KETHLEDGE,
STRANCH, THAPAR, BUSH, LARSEN, NALBANDIAN, READLER,
MURPHY, DAVIS, MATHIS, BLOOMEKATZ, and RITZ, Circuit Judges.
_________________
COUNSEL
ARGUED EN BANC: Kelsi Brown Corkran, INSTITUTE FOR CONSTITUTIONAL
ADVOCACY AND PROTECTION, Washington, D.C., for Appellants. T. Elliot Gaiser, OFFICE
OF THE OHIO ATTORNEY GENERAL, Columbus, Ohio, for Appellee. ON SUPPLMENTAL
BRIEF: Kelsi Brown Corkran, Elizabeth R. Cruikshank, Alexandra Lichtenstein, William
Powell, INSTITUTE FOR CONSTITUTIONAL ADVOCACY AND PROTECTION,
Washington, D.C., Mark R. Brown, CAPITAL UNIVERSITY, Columbus, Ohio, Oliver Hall,
CENTER FOR COMPETITIVE DEMOCRACY, Washington, D.C., for Appellants. T. Elliot
Gaiser, Katie Rose Talley, OFFICE OF THE OHIO ATTORNEY GENERAL, Columbus, Ohio,
for Appellee. ON AMICUS BRIEF: Tobias S. Loss-Eaton, Naomi Igra, Stephen Chang,
Madison Ferraro, SIDLEY AUSTIN LLP, Washington, D.C., for Amici Curiae.
The court delivered a PER CURIAM opinion. THAPAR, J. (pp. 8–23), delivered a
separate concurring opinion. MOORE, J. (pp. 24–36), delivered a separate dissenting opinion, in
which CLAY, STRANCH, and DAVIS, JJ., concurred, and KETHLEDGE and RITZ, JJ.,
>

-- 1 of 38 --

No. 24-3354 Brown et al. v. Yost Page 2
concurred in Part II. KETHLEDGE, J. (pp. 37–38), also delivered a separate dissenting opinion,
in which MOORE, CLAY, STRANCH, DAVIS, and RITZ, JJ., concurred.
_________________
OPINION
_________________
PER CURIAM. At issue in this case is whether a request for a preliminary injunction
focused on the 2024 election—arising from a challenge to Ohio’s process for placing constitutional
initiatives on the ballot—is moot. It is. The targeted election has come and gone, making this
request for preliminary relief moot.
Since 1912, Ohio has allowed its citizens to amend the Ohio Constitution through ballot
initiatives. Ohio law requires proponents of initiatives to satisfy several requirements before their
proposed constitutional amendment appears on the ballot. At the outset, the proponents must
submit their proposed amendment, a summary of it, and 1,000 signatures to the Ohio Attorney
General, who then must decide whether the summary is a “fair and truthful statement of the
proposed law or constitutional amendment.” O.R.C. § 3519.01(A). If the Attorney General rejects
the summary, the proponent may seek review in the Ohio Supreme Court. Id. § 3519.01(C). Once
the proponent obtains approval of the summary, the Ballot Board must examine the initiative to
ensure that it does not violate Ohio’s single-subject requirement. Id. § 3505.062(A). With the
Board’s approval in hand, the proponent of the initiative may start the process of presenting the
proposal and summary to Ohio citizens to obtain the roughly 400,000 signatures needed to put the
amendment on the ballot. Ohio Const. art. II, §§ 1a, 1g; O.R.C. § 3519.05(A). Signatures are due
to the Secretary of State at least 125 days before election day, which was July 3, 2024 for the last
election cycle. Ohio Const. art. II, § 1a; O.R.C. § 3519.16(B). If the Secretary approves the
signatures, the Ballot Board prescribes and certifies to the Secretary the language for the proposed
amendment. Ohio Const. art. II, §§ 1e, 1g; O.R.C. § 3505.062(D). Proponents then prepare
arguments or explanations in favor of their proposal, and the General Assembly or Governor names
individuals to prepare arguments or explanations against. Ohio Const. art. II, § 1g; O.R.C.
§ 3519.03(A). Once these requirements are met, the amendment, as written by the Ballot Board,

-- 2 of 38 --

No. 24-3354 Brown et al. v. Yost Page 3
is placed on the ballot for the next general or regular election. Ohio Const. art. II, §§ 1a, 1g; O.R.C.
§ 3519.16.
This appeal arises from a ballot proposal initiated by Cynthia Brown. In February 2023,
she sought to amend the Ohio Constitution to eliminate governmental immunities or defenses in
certain state-law causes of action. That month, she submitted her proposed amendment and
summary to the Attorney General. He rejected the summary as not “fair and truthful.” In total,
she has tried and failed eight times to obtain his approval. After the Attorney General rejected her
proposed summary in March 2024, Brown sought mandamus relief in the Ohio Supreme Court.
With less than four months left before the July 3rd deadline to collect signatures to place her
amendment on the ballot for the November 2024 election, Brown moved for expedited review.
The state court declined to expedite the case, and Brown voluntarily dismissed the action. See
O.R.C. § 3519.01(A), (C); State ex rel. Brown v. Yost, 234 N.E.3d 472 (Ohio 2024) (unpublished
table decision).
Brown turned to federal court for relief. She sued the Attorney General in district court,
alleging that Ohio’s initiative procedures violate the First Amendment facially and as applied. Her
facial challenge complains that “Ohio’s delegation to Defendant of the authority to reject
summaries of proposed constitutional amendments pursuant to O.R.C. § 3519.01(A) coupled with
its failure to provide for immediate judicial review and resolution in the Ohio Supreme Court under
O.R.C. § 3519.01(C)” violates “the First Amendment” in all of its applications. R.1 at 10–11 ¶ 44.
Her as-applied challenge complains that the Ohio Supreme Court’s “refus[al] to expedite review”
and the Attorney General’s “formal[] object[ion]” to it violate her free-speech rights. R.1 at 11
¶¶ 54–55. She asked the district court to declare § 3519.01(A) and (C) unconstitutional and to
enjoin the Attorney General from reviewing summaries without the option of expedited judicial
review.
On March 27, 2024, with the upcoming November 5th election looming, Brown moved for
temporary relief in the district court. She sought to preliminarily enjoin the Attorney General “to
certify Plaintiffs’ summary to Ohio’s Ballot Board” because “[t]he November 2024 election is
closely approaching and injunctive relief is needed to ensure the initiative has a fair chance of
being included on the ballot.” R.2 at 19–20. The district court denied her motion, and she

-- 3 of 38 --

No. 24-3354 Brown et al. v. Yost Page 4
appealed. A divided panel of this court granted the preliminary injunction, reasoning (among other
things) that § 3519.01(A) and (C) likely impose a severe burden on Brown’s “ability to advocate
for [her] initiative” under the Anderson-Burdick line of cases and the Attorney General is likely to
fail strict-scrutiny review. Brown v. Yost, 103 F.4th 420, 440–43 (6th Cir. 2024). The Attorney
General petitioned for rehearing en banc. The full court granted the petition and vacated the panel
opinion. 104 F.4th 621 (6th Cir. 2024). Neither party moved to expedite the briefing, the oral
argument, or the resolution of the case before the full court.
At this point, the November 5, 2024 election has ended, which prompts this threshold
question: Is Brown’s request for a preliminary injunction moot?
Article III of the United States Constitution extends the “judicial Power” only to “Cases”
and “Controversies.” U.S. Const. art. III, § 2. A claimant must meet this “cradle-to-grave
requirement” from the outset of a case to its conclusion. Fialka-Feldman v. Oakland Univ. Bd. of
Trs., 639 F.3d 711, 713 (6th Cir. 2011). An appeal becomes moot if intervening events occur that
make it “impossible” for the relevant federal court to grant any “effectual relief.” Church of
Scientology of Cal. v. United States, 506 U.S. 9, 12 (1992) (citation omitted).
In gauging the potential mootness of an appeal, we must focus on the request for relief in
front of us. That explains why a preliminary injunction may become moot on appeal even as the
underlying controversy and the request for a permanent injunction remain live. Ohio v. EPA, 969
F.3d 306, 309 (6th Cir. 2020). To that end, we have consistently distinguished “between mootness
as to a preliminary-injunction appeal and mootness as to the case as a whole.” Resurrection Sch.
v. Hertel, 35 F.4th 524, 528 (6th Cir. 2022) (en banc) (quoting Ohio, 969 F.3d at 309). So has the
U.S. Supreme Court: “Because the only issue presently before us—the correctness of the decision
to grant a preliminary injunction—is moot, . . . the case must be remanded to the District Court for
trial on the merits.” Univ. of Tex. v. Camenisch, 451 U.S. 390, 394 (1981).
Brown’s appeal from the denial of a preliminary injunction is moot. Just as a claimant is
the master of her complaint, Brown is the master of her requests for interim relief—whether for a
preliminary injunction or a temporary restraining order. In this instance, Brown unmistakably
trained her request for immediate relief on the upcoming election scheduled for November 5, 2024.

-- 4 of 38 --

No. 24-3354 Brown et al. v. Yost Page 5
She sought a preliminary injunction requiring Yost to certify her March summary because “[t]he
November 2024 election is closely approaching and injunctive relief is needed to ensure the
initiative has a fair chance of being included” on that ballot. R.2 at 19; see also R.2-1 at 1.
Throughout her motion for “preliminary” relief, she pointed to the November 2024 election. She
explained to the district court that the “November 2024 election is closely approaching,” R.2 at
19, that she has “less than four months to collect the 400,000-plus signatures they must submit by
July [3], 2024,” R.2 at 3, that “Ohio’s ballot clock is ticking,” R.2 at 19, and that “[t]ime is of the
essence,” R.2 at 3. Removing any doubt about the relief she sought in the preliminary injunction,
Brown gave the proposed amendment an effective date of “January 1, 2025,” after it “appear[ed]
on the November 2024 ballot.” Brown Supp. Br. 4; see also R.1 (Exhibit 3) at 5. Nowhere in
these requests for immediate relief does she say that she would submit her March summary and
amendment for some future election. Even Brown acknowledges on appeal the limited nature of
the relief she sought: “an order permitting [her] to proceed with petition circulation in time for
[her] proposed amendment . . . to be placed on the ballot for the November 2024 election.” Brown
Supp. Br. 2.
Now that the November 2024 election is over, her request for relief, like many a 2024
election yard sign, has become stale. Ordering Yost to certify that summary now will not give
Brown the relief she seeks because the window to appear on the 2024 ballot has closed. Without
power to rewind the clock, we cannot give Brown the relief she sought in her preliminary
injunction. Given the “nature of the relief sought,” the matter is “moot because the . . . election is
over.” Brockington v. Rhodes, 396 U.S. 41, 43 (1969) (per curiam).
We appreciate that the underlying lawsuit is not moot. Her request for permanent
injunctive relief is not limited to 2024. R.1 at 12. Hence the passing of the November 2024
election does not undercut the live nature of the dispute pending in the district court. But it does
mean that Brown has passed the use-by date of her interim request for relief. See Ohio, 969 F.3d
at 308–10; Resurrection Sch., 35 F.4th at 528.
All of this explains why it is “impossible” for us to grant any “effectual relief” to Brown
at this point, Church of Scientology, 506 U.S. at 12, and why Brown does not have a “concrete
interest, however small,” in obtaining relief on her preliminary injunction, Chafin v. Chafin,

-- 5 of 38 --

No. 24-3354 Brown et al. v. Yost Page 6
568 U.S. 165, 172 (2013) (citation omitted). Sure, a ruling in her favor in this case might advance
the general cause of her underlying non-moot lawsuit by giving her a “head start” on collecting
signatures for November 2025. Yost Supp. Br. 4. But it would not advance the specific cause of
the request for “relief” before us—an order seeking relief with respect to a summary meant by its
terms only for the 2024 election. In an “adversarial system of adjudication, we follow the principle
of party presentation.” United States v. Sineneng-Smith, 590 U.S. 371, 375 (2020). Only in her
request for a permanent injunction does Brown seek relief for elections after 2024. Parties are
limited to the preliminary relief they requested. See Resurrection Sch., 35 F.4th at 530. It thus
makes no difference that an approved ballot summary may be used in future elections. In this
instance, Brown’s proposal was limited to the November 2024 election and came with a now-
inaccurate effective date of “January 1, 2025.”
Any other approach runs the risk of collapsing the distinction between preliminary and
permanent injunctions. It is the rare grant of preliminary relief for a claimant that would not aid
her underlying request for permanent relief. We know of no federal court, and the parties at any
rate have not identified one, that permits the non-moot nature of an underlying request for
permanent relief to save a moot request for preliminary relief. Much as we should squint carefully
at any request for preliminary relief to determine whether it could benefit the claimant, we cannot
close our eyes when the window for relief actually sought by the claimant has ended.
It’s true that a federal court has a “virtually unflagging obligation” to exercise the
jurisdiction given it. Colo. River Water Conservation Dist. v. United States, 424 U.S. 800, 817
(1976). But this principle asks a question; it does not answer it. It’s just as accurate to say that a
federal court has an “unflagging obligation” to determine whether it has Article III jurisdiction
over a dispute—save that this imperative is categorically unflagging, not “virtually” unflagging.
The “narrow” exception for disputes that are “capable of repetition, yet evading review”
does not spare this appeal. Gerstein v. Pugh, 420 U.S. 103, 110 n.11 (1975). Rather than evade
review, Brown’s First Amendment challenges in the underlying lawsuit are on track to receive it
before the next general election: November 2025. On remand, Brown remains free to seek
expedited resolution of the permanent injunction request. Nothing prevents either party from
seeking expedited review of an adverse decision. And nothing prevents either party from seeking

-- 6 of 38 --

No. 24-3354 Brown et al. v. Yost Page 7
expedited en banc review. All of this by the way allows the district court to consider several other
intervening developments: Brown’s subsequent submission of a new summary intended for
Ohio’s November 2025 election, the Attorney General’s rejection of it, a recent decision by the
Ohio Supreme Court that undermines one of the Attorney General’s prior rejections, State ex rel.
Dudley v. Yost, -- N.E.3d -- , 2024 WL 4610503 (Ohio 2024) (per curiam), a recent order by the
Ohio Supreme Court requiring a response by the Attorney General, and the developing nature of
both parties’ theories with respect to the nature of the free-speech rights at issue. Because these
issues all are “primarily if not entirely legal,” the federal courts stand ready to resolve them
quickly. Ohio, 969 F.3d at 309.
All of this may explain the ambivalence of the party with her free-speech rights at stake—
Brown—in responding to our request for supplemental briefing about mootness. While she
maintains that the appeal remains live, she concedes that her requested relief “became outdated”
after the “July deadline for the November 2024 election passed,” and “do[es] not object to
dismissal” as it would allow her to “seek relief tailored to [her] current efforts to place the[]
proposed amendment on the ballot in 2025.” Brown Supp. Br. 1, 5. We, too, agree that there is
ample time to ensure that her First Amendment rights are aired in time for the November 2025
election, and we stand ready to ensure that happens.
Because Brown did not have an opportunity to obtain appellate review of the district court’s
decision denying her request for preliminary relief, we vacate that decision. See United States v.
Munsingwear, Inc., 340 U.S. 36, 39–41 (1950).
For these reasons, we dismiss the appeal as moot.

-- 7 of 38 --

No. 24-3354 Brown et al. v. Yost Page 8
_________________
CONCURRENCE
_________________
THAPAR, Circuit Judge, concurring in the denial of the preliminary injunction. Plaintiffs
sought a preliminary injunction to force the Ohio Attorney General to certify their ballot initiative.
Their motion has two basic problems. First, to the extent they are seeking relief for the November
2024 election, I agree with my colleagues that their motion is moot. Second, even if they are
seeking prospective relief for future elections, their motion fails because they have not shown that
they are likely to succeed on the merits.
I.
Whether plaintiffs’ motion for a preliminary injunction is now moot depends on what relief
they sought in the district court. There, plaintiffs asked for a preliminary injunction directing the
Attorney General to certify the March 2024 summary of their amendment proposal as fair and
truthful. They tied that March certification request to their desire to get on the November 2024
election ballot. Plaintiffs added that time was “of the essence” so that they could get on the ballot
come November 2024. R. 2, Pg. ID 45. In addition, plaintiffs’ March 2024 summary and proposed
amendment text listed the effective date of the amendment as January 1, 2025. That effective date
assumed that the proposed amendment would be on the November 2024 ballot. Viewed in context,
then, it seems that plaintiffs’ request was geared towards circulating their proposal in time for the
November 2024 election. But that election is now a thing of the past. So as the majority opinion
thoughtfully lays out, their motion could be moot.
But as both parties point out, the Attorney General’s summary certifications roll over from
election to election and have no expiration date. See Ohio Const. art. II, § 1a; Ohio Rev. Code
§ 3519.01. And as the Attorney General readily admits, once plaintiffs get the certification, they
can start collecting signatures for 2025 or a future election. So, it seems that our ordering the
Attorney General to certify the March 2024 summary may give plaintiffs some “effectual relief,”
which means their motion wouldn’t be moot. Chafin v. Chafin, 568 U.S. 165, 172 (2013) (citation
omitted).

-- 8 of 38 --

No. 24-3354 Brown et al. v. Yost Page 9
In short, the mootness question is thorny. But even if plaintiffs’ request isn’t moot, they
still wouldn’t be entitled to a preliminary injunction. That’s because they can’t show that they
would be likely to succeed on the merits of their First Amendment claim.
II.
As plaintiffs admitted at argument, their theory of First Amendment injury has shifted
throughout this litigation. When plaintiffs first filed their complaint and motion for preliminary
injunctive relief, they repeatedly stressed the lack of expedited state-court review of the Attorney
General’s summary certification decisions. They cited this delay as a key reason why their First
Amendment rights were being violated. Before the en banc court, plaintiffs did refer to that lack
of timely judicial review as a problem. But that issue took a backseat to a different First
Amendment argument—that the Attorney General is exercising content-based, editorial control
over plaintiffs’ core political speech. Plaintiffs pressed this theory at oral argument. So, when we
compare plaintiffs’ original complaint and motion for preliminary injunctive relief with their
appellate stage briefing and oral argument, their First Amendment arguments have not been
consistent. That’s reason enough to deny them a preliminary injunction.
In addition, there is no longer any exigency. Plaintiffs can go back to the district court and
seek an injunction for the 2025 election. When they do so, they can more precisely articulate the
alleged First Amendment issue. The district court and this court can then rule on that motion with
the benefit of fuller, more consistent briefing.
And even if we overlooked plaintiffs’ shifting arguments and assessed only the strongest
ones, they still are unlikely to succeed.
Start with the basics. The First Amendment protects Americans’ right to speak to one
another about matters of public concern. But the First Amendment doesn’t give anyone a right to
have their speech carry the force of law.
Therefore, when we assess First Amendment challenges to states’ regulations of initiative
processes, we must distinguish between two types of regulations: (1) regulations of citizens’
ability to speak to one another as part of that process, and (2) regulations of citizens’ ability to

-- 9 of 38 --

No. 24-3354 Brown et al. v. Yost Page 10
make law as part of that process. The former likely poses a serious First Amendment problem and
would be subject to strict scrutiny. The latter likely does not even implicate the First Amendment
and would be subject to no First Amendment scrutiny at all.
A.
The First Amendment protects Americans’ right to persuade. If a state opts to create a
ballot initiative process, the First Amendment limits the extent to which the state can regulate
“advocacy itself” that takes place within the initiative process. Initiative & Referendum Inst. v.
Walker, 450 F.3d 1082, 1099 (10th Cir. 2006) (en banc) (McConnell, J.). States don’t have free
rein to dictate who can speak or how they speak when citizens try to persuade one another of an
initiative’s merits. Id. When it comes to “interactive communication concerning political change,”
First Amendment protections are at their “zenith.” Buckley v. Am. Const. L. Found., Inc., 525 U.S.
182, 186–87 (1999) (citation omitted). So, a state’s efforts to limit private expression by those
trying to convince others to vote a certain way likely raises serious First Amendment problems—
whether that expression takes the form of one-on-one conversations with members of the public
or the mass distribution of political handbills. Meyer v. Grant proves this point. There, the
Supreme Court invalidated a Colorado law that banned citizens from paying others to circulate
petitions to amend the state constitution. 486 U.S. 414, 416 (1988). The Court reasoned that the
First Amendment protects citizens’ “right not only to advocate their cause but also to select what
they believe to be the most effective means for so doing.” Id. at 424. Therefore, Colorado’s paid
circulation ban not only implicated the First Amendment; it triggered strict scrutiny because it
directly regulated core political speech. Id. at 420–21.
But the First Amendment doesn’t give Americans a right to legislate. Again, start with the
basics. Governments can restrict legislators’ speech during the lawmaking process and can limit
the laws they pass. For example, to this day, states like Ohio require their legislators to “confine”
their floor speeches to the “question under debate.” See, e.g., Ohio House of Representatives Rule
48(a); Michigan House of Representatives Standing Rule 28; Tennessee House of Representatives
Rule 18. Such content-based restrictions on legislators’ speech within the lawmaking process
aren’t new. Colonial assemblies had rules barring “profanity” and “unseemly, unnecessary, and
tedious debate.” Keith E. Hamm & Peverill Squire, 101 Chambers: Congress, State Legislatures,

-- 10 of 38 --

No. 24-3354 Brown et al. v. Yost Page 11
and the Future of Legislative Studies 18 (2005) (citation omitted). That’s unsurprising: The First
Amendment does not “confer[] a right to use governmental mechanics to convey a message.” Nev.
Comm’n on Ethics v. Carrigan, 564 U.S. 117, 127 (2011). Similarly, governments can restrict
what sorts of laws legislators may enact. Federal and state constitutions say legislators can’t pass
some laws—like giving out noble titles or stripping jury trial rights. See, e.g., U.S. Const. art. I,
§ 9; Ohio Const. art. I, § 5. Likewise, governors can veto laws legislatures passed. Neither
restrains legislators’ speech—just the legal effects of their conduct.
The same logic applies to the petition process. Governments can restrict what citizens can
propose through the official initiative process. When initiative proponents submit their proposals
for certification, they’re like the legislators in the state house who are subject to no-swearing and
“question under debate” rules. So too can governments restrain what sorts of petitions can
ultimately become law. If an initiative garners enough signatures and otherwise complies with
applicable rules, it becomes the law. So, when the government regulates the content of initiative
petitions and their summaries, it isn’t regulating private speech based on content—it’s regulating
what sorts of laws citizens can enact.
Therefore, a state likely doesn’t have to worry about the First Amendment at all—let alone
satisfy strict scrutiny—when it regulates an initiative petition’s content or its summary’s content.
See Walker, 450 F.3d at 1099–1100.
B.
Indeed, when it comes to initiative processes, states have long been free to determine that
a wide array of subjects are off limits. Once a state makes that determination, it can impose various
content-based checks on initiative proposals and their summaries either before or after signature
collection. In short, Ohio isn’t in a league of its own.
Consider a few examples. In Montana, much like in Ohio, the Attorney General reviews
proposed initiatives “for legal sufficiency.” Mont. Code Ann. § 13-27-226(1). Whether a proposal
is legally sufficient hinges in part on its content—including the accuracy of the proposal’s
summary, which must be “true and impartial.” Id. § 13-27-212(1). If the Attorney General deems
the proposal legally insufficient, the proposal can’t reach the ballot. Id. § 13-27-218(8)(a). And

-- 11 of 38 --

No. 24-3354 Brown et al. v. Yost Page 12
just like in Ohio, Montanans can challenge the Attorney General’s legal sufficiency determination
in court. Id. § 13-27-605.
Maine law includes similar provisions. In that state, citizens can propose legislation for
the legislature to enact. Me. Const. art. IV, Pt. 3, § 18. To do so, they first need to file their
proposed law’s text and summary with the Secretary of State. Me. Rev. Stat. Ann. tit. 21-A, § 901.
The Secretary can reject the citizens’ proposal if he concludes that its titles or headnotes don’t
“objectively reflect” the proposal’s “content.” Id. § 901(3-A)(B)(2). Much like in Ohio and
Montana, Maine voters named in the application can challenge the Secretary’s decision in state
court. Id. § 901(7).
Other states restrict citizens’ ability to collect signatures and propose amendments based
on the content of their proposed amendments. In Massachusetts, the people have “reserve[d]” to
themselves the power to initiate amendments to their state constitution. Mass. Const. amend. art.
48, Pt. 1, Def. But various content-based restrictions limit their ability to propose initiatives: They
can’t propose amendments relating to religion, abolishing courts, or infringing on the jury trial
right, just to name a few. Id. Init., Pt. 2, § 2. After initiative proponents get an initial ten signatures,
the Attorney General ensures that the proposed amendment doesn’t touch on one of those off-
limits buckets of content. Id. Pt. 2, § 3. Once the Attorney General has made that determination
and the proponents have filed their proposal with the Secretary of State, they collect more
signatures to get their proposal on the ballot. Id. Pt. 4, § 2; see also Herrmann v. Att’y Gen., 208
N.E.3d 727, 732 (Mass. 2023) (describing this process). After the proponents complete these
steps, their proposal appears on the ballot only if one-quarter of state legislators approve it in
successive legislative sessions. Mass. Const. amend. art. 48, Init., Pt. 4, §§ 4–5. Massachusetts’
many restrictions on the content of initiative petitions—and its empowering of its Attorney General
to enforce those restrictions—have withstood First Amendment challenges. See Wirzburger v.
Galvin, 412 F.3d 271, 275–80 (1st Cir. 2005).
Massachusetts’ content-based restrictions on its citizens’ lawmaking powers aren’t unique.
Missouri’s state constitution prohibits citizens from using the initiative process “for the
appropriation of money other than of new revenues created and provided for thereby.” Mo. Const.

-- 12 of 38 --

No. 24-3354 Brown et al. v. Yost Page 13
art. III, § 51. And when an initiative petition is submitted to the Secretary of State, he must first
ensure that it substantively complies with the Missouri constitution. Mo. Ann. Stat. § 116.120(1).
In Wyoming, citizens can use the initiative process to pass legislation, but they can’t use it
“to dedicate revenues, make or repeal appropriations, create courts, define the jurisdiction of courts
or prescribe their rules, enact local or special legislation, or enact that prohibited by the constitution
for enactment by the legislature.” Wyo. Const. art. III, § 52(g). And only after the Secretary of
State has certified that the petition is “in proper form” can proponents circulate it to their fellow
citizens to obtain the required number of signatures. Id. art. III, § 52(b)–(c).
In other states, initiative proponents must include certain disclaimers and other pieces of
information on the forms that they ask their fellow citizens to sign. For example, in Utah, if an
initiative proposes a tax increase, proponents must identify the current tax rate and specify by how
much they hope to raise it. Utah Code Ann. § 20A-7-203(2)(b).
The fact that some of these state procedures provide for content-based review of initiative
proposals after signature collection, while Ohio’s provides for that review before the vast bulk of
signature collection, is irrelevant from the perspective of the First Amendment. Whatever the
timing of the content-based review, the state is limiting which signed proposals can be given
legally operative effect. Only proposals that clear the state’s gatekeeping hurdle can become law.
To sum up: States enjoy broad leeway to regulate the content of initiative proposals and
their summaries. Why? States can prohibit citizens from passing laws on some topics and can
stop initiatives purporting to pass such laws. See, e.g., Mass. Const. 48, Init., Pt. 2, § 2;
Wirzburger, 412 F.3d at 271, 274; Marijuana Pol’y Project v. United States, 304 F.3d 82 (D.C.
Cir. 2002). States can also provide that amendments with inaccurate summaries can’t become law.
See, e.g., Mont. Code Ann. § 13-27-212(1). Neither procedure restricts citizens’ ability to speak—
only their ability to “wield a legislative power.” Chula Vista Citizens for Jobs & Fair Competition
v. Norris, 782 F.3d 520, 530–31 (9th Cir. 2015) (en banc) (emphasis omitted).

-- 13 of 38 --

No. 24-3354 Brown et al. v. Yost Page 14
C.
Ohio’s law regulates citizens’ power to make law, not their right to speak. The fair-and-
truthful certification provision lays out a threshold requirement that Ohioans must meet before
they advocate their position to would-be signatories however they see fit. After citizens obtain an
initial 1,000 signatures, the summary certification requirement poses a gatekeeping hurdle that
they must clear before they continue to speak as part of the initiative process. The law doesn’t
regulate speech that takes place within the initiative circulation process itself, as the Colorado law
at issue in Meyer did. Recall that in that case, Colorado had prevented citizens from using
particular speakers—paid circulators—to get their pro-reform message out. By contrast, Ohio
doesn’t say who can speak or how they can speak when they engage with their fellow citizens.
See Walker, 450 F.3d at 1099.
For example, Ohio cannot bar someone from engaging in “unfair” or “untruthful”
conversations when they are seeking signatures on the campaign trail. Indeed, proponents can say
whatever they please to get their message out. And even if the Ohio Attorney General declines to
certify plaintiffs’ summary, plaintiffs remain free to persuade their fellow Ohioans of their cause.
They also can circulate a petition to this effect, collect signatures, and call for constitutional
change. But when those same proponents show up and ask Ohio to give their proposal legally
operative effect, Ohio can restrict the language of the law itself—or in this case the summary. By
refusing to certify plaintiffs’ summary, the Attorney General merely prevents that sort of speech
from having legal effect under state law. Such a process has friends in other states, and it doesn’t
have an enemy in the First Amendment.
D.
If one were to compare Ohio’s law with the examples from other states like Montana and
Massachusetts, one might still fault Ohio’s fair-and-truthful certification provision as follows: It’s
one thing for a state to prohibit citizens from making law on certain topics, but it’s quite another
for the state to veto inaccurate summaries of citizen petitions when the initiative’s topic is fair
game under state law. But that’s a distinction without a constitutional difference. They’re both
content-based restrictions on the initiative process. The state may be concerned with different

-- 14 of 38 --

No. 24-3354 Brown et al. v. Yost Page 15
sorts of content of citizen-initiated laws—maybe the proposal deals with a particular policy area,
maybe the summary inaccurately reflects the proposal. It doesn’t matter. The point is that the
state is regulating what sorts of laws its citizens can enact rather than directly regulating their
expression.
Similarly, a First Amendment challenge to Ohio’s law for vesting the Attorney General
with too much discretion misses the mark. At first blush, this sort of argument has force since the
last thing the First Amendment allows for are prior restraints or vague prohibitions. But again,
states can restrict what sorts of laws and amendments their citizens can pass by way of initiatives.
And the First Amendment doesn’t stop a state from delegating discretion to an executive official
like the state Attorney General to limit or prohibit citizen lawmaking. When Attorney General
Yost enforces the fair-and-truthful certification provision, he isn’t directly regulating speech; he’s
restricting what sorts of citizen initiative proposals can become law. Schmitt v. LaRose, 933 F.3d
628, 638 (6th Cir. 2019). At any rate, his exercise of discretion isn’t unreviewable: Plaintiffs can
turn to the state courts for legal relief if they think the Attorney General is acting outside the bounds
of his statutory authority. See Ohio Rev. Code § 3519.01(C). In fact, plaintiffs are doing precisely
that right now. See Doc. 63, Pg. 2.
E.
Because Ohio’s law doesn’t directly regulate political speech made within the initiative
process, it certainly shouldn’t be subject to strict scrutiny. But should it be subject to any level of
First Amendment scrutiny? No.
Different courts have come out differently on the question of what tier of scrutiny—if
any—should apply to content-based regulations of initiative processes.
When confronted with a First Amendment challenge like this one, the First Circuit applied
intermediate scrutiny under United States v. O’Brien, 391 U.S. 367 (1968). Wirzburger, 412 F.3d
at 275. Intermediate scrutiny applies when a law regulates conduct that contains both speech and
nonspeech elements. O’Brien, 391 U.S. at 376–77. In Wirzburger, plaintiffs argued that the
Massachusetts constitutional provisions excluding citizen initiatives on certain topics like religion
and jury trials set forth impermissible content-based restrictions on their core political speech. The

-- 15 of 38 --

No. 24-3354 Brown et al. v. Yost Page 16
First Circuit rejected their argument. It reasoned that the “difficulty” with plaintiffs’ argument
was “that a state initiative procedure, although it may involve speech, is also a procedure for
generating law, and is thus a process that the state has an interest in regulating, apart from any
regulation of the speech involved in the initiative process.” Wirzburger, 412 F.3d at 275.
Massachusetts’ interest in controlling its lawmaking process wasn’t related to the expression that
takes place as part of the initiative process. Id. at 278. Still, the court concluded that the state’s
content-based restraints on its initiative process restricted speech—namely, one-on-one
communications between proponents and would-be signatories. Id. at 279. As a result, the court
subjected Massachusetts’ exclusions of certain topics from the initiative process to O’Brien’s
intermediate scrutiny standard. Id. The First Circuit determined that those exclusions cleared
O’Brien’s bar. Id.
By contrast, the en banc Tenth Circuit concluded that content-based restraints on citizens’
initiative lawmaking powers shouldn’t be subject to any First Amendment scrutiny, including
O’Brien’s standard. See Walker, 450 F.3d at 1085, 1101–04. In Walker, the Tenth Circuit affirmed
the dismissal of plaintiffs’ challenge to Utah’s supermajority voting requirements for initiatives
relating to wildlife management. Id. at 1085. Writing for the majority, then-Judge McConnell
rejected Wirzburger’s reliance on O’Brien. He explained that O’Brien’s intermediate scrutiny
standard doesn’t apply “to structural principles of government making some outcomes difficult or
impossible to achieve.” Id. at 1102. It’s one thing to regulate expressive conduct; it’s another
thing to make that expressive conduct less likely to result in a new law being passed. Id.
The Tenth Circuit has the better of the argument. Ohio’s law directly impacts only the
legal effects of the speech that initiative proponents make as part of the circulation process. As a
panel of this court observed, “Ohio’s ballot-initiative laws . . . do not directly restrict core
expressive conduct; rather, the laws regulate the process by which initiative legislation is put
before the electorate, which has, at most, a second-order effect on protected speech.” Schmitt, 933
F.3d at 638. Again, Ohio isn’t barring initiative proponents from speaking or collecting signatures.
It’s merely imposing a threshold requirement that proponents must satisfy if they want their speech
and signature collection to have legally operative effect. Understood in this light, Ohio’s law isn’t

-- 16 of 38 --

No. 24-3354 Brown et al. v. Yost Page 17
regulating the expressive component of the initiative process at all. So, it wouldn’t make sense to
subject it to First Amendment scrutiny.
And to echo then-Judge McConnell’s logic in Walker: Ohio’s fair-and-truthful
requirement doesn’t burden speech just because it impedes state constitutional reform. 450 F.3d
at 1102. Yes, Ohio’s law may make it less likely that Ohioans will achieve constitutional reform,
but “it does not follow that all such limitations are challengeable under the First Amendment.” Id.
Also, unelected federal judges like us have additional reason to pause before subjecting
states’ regulations of their own initiative procedures to intermediate scrutiny. By applying
O’Brien, we necessarily must pass on whether the state’s interest is “substantial.” That’s
worrisome, because it entails “an assessment of the virtues and vices of the particular initiatives
that are affected by the limitation.” Id. at 1103; cf. Nat’l Republican Senatorial Comm. v. Fed.
Election Comm’n, 117 F.4th 389, 400 (6th Cir. 2024) (Thapar, J., concurring) (critiquing the tiers
of scrutiny for “giving judges free rein to weigh a statute’s means and ends”). Unless the First
Amendment demands otherwise, it’s up to the people and elected representatives of each state, not
us unelected federal judges, to determine how to best structure their legislative processes.
Speaking of our limited role as a federal court, consider what would follow if we were to
subject Ohio’s law or the Attorney General’s enforcement to First Amendment scrutiny and then
find a First Amendment violation.
If we were to conclude that the law itself was unconstitutional, we’d be telling Ohio that
its selected means of regulating what amendments its citizens can enact via initiatives are
impermissible—even though the Constitution doesn’t require Ohio to set up an initiative process
at all. Thompson v. DeWine, 976 F.3d 610, 615 (6th Cir. 2020) (per curiam). We’d also throw the
constitutionality of a wide array of analogous state provisions into doubt. If Ohio’s fair-and-
truthful certification provision doesn’t pass constitutional muster, presumably the content-based
limits on initiatives and summaries imposed by states like Massachusetts, Maine, and Montana
wouldn’t, either.
And what if we held that only the Attorney General’s enforcement of the provision violated
the First Amendment? Could we somehow claim that the Attorney General incorrectly interpreted

-- 17 of 38 --

No. 24-3354 Brown et al. v. Yost Page 18
state law and that error somehow offended the First Amendment? On what basis would we
conclude as much?
The Constitution doesn’t give us judges the power to dictate the ins and outs of Ohio’s
initiative process. States have “important regulatory interests in combating fraud” in their
elections and initiative processes. Little v. Reclaim Idaho, 140 S. Ct. 2616, 2617 (2020) (Roberts,
C.J., concurring in the grant of stay). And they “retain ‘considerable leeway to protect the integrity
and reliability of the initiative process.’” Id. at 2616 (quoting Buckley, 525 U.S. at 191). In
enacting the fair-and-truthful certification provision, Ohio stayed within the bounds of its
discretion. If we concluded otherwise, we’d exceed ours.
F.
Finally, public forum precedents don’t bear on Ohio’s fair-and-truthful certification
provision. Public forums are dedicated to fostering speech and debate. See Marijuana Pol’y
Project, 304 F.3d at 86 (collecting cases). Ballots are geared towards structuring democratic
elections, not fostering debate. Cf. Timmons v. Twin Cities Area New Party, 520 U.S. 351, 363
(1997). The same goes for ballot initiative processes: They may “stimulate[] First Amendment-
protected debate and discussion,” but the “legislative act” of voting in the initiative process isn’t a
public forum. Marijuana Pol’y Project, 304 F.3d at 87.
III.
The principal dissent challenges (1) my reliance on the distinction between regulations of
speech and regulations of citizen lawmaking and (2) my characterization of Ohio’s law as a
regulation of citizen lawmaking. I address each concern in turn.
A.
The principal dissent writes that, in the ballot initiative context, Supreme Court precedent
prevents us from distinguishing regulations of citizens’ ability to make law from regulations of
citizens’ ability to speak. The dissent argues: “The Supreme Court has never cordoned off the
ballot-initiative process from First Amendment scrutiny. To the contrary, the Supreme Court has
indicated that regulations of the ballot-initiative process should be analyzed in the same terms as

-- 18 of 38 --

No. 24-3354 Brown et al. v. Yost Page 19
other election-related regulations.” Moore Dissenting Op. 32. This statement implies that every
regulation of state ballot initiative processes should be subject to First Amendment scrutiny. The
Supreme Court has never endorsed this sweeping proposition.
Consider the case that the dissent cites in support of its argument: Buckley v. American
Constitutional Law Foundation, Inc., 525 U.S. 182 (1999). That case doesn’t support the
proposition that a state’s regulation of its ballot initiative process automatically triggers some form
of First Amendment scrutiny. There, the Court invalidated various Colorado laws regulating
speech made within its initiative process. For example, one of those laws required petition
circulators to be registered voters. The Court reasoned that, much like the paid circulator ban set
aside in Meyer, this provision restricted proponents’ ability to spread their message and decreased
the size of their audience. Id. at 194–95. In other words, the regulation burdened political
expression. Id. at 195. Therefore, the Court subjected the law to First Amendment scrutiny. Id.
at 197. The fact that the state was regulating speech, not that it was regulating in the ballot initiative
context, is what triggered First Amendment scrutiny.
Buckley thus stands for the unremarkable proposition that regulations of political
expression trigger First Amendment scrutiny. Ohio’s law, by contrast, does not regulate political
expression. Plaintiffs here remain free to make the same amount of political expression, using the
same methods, as they would be able to make without the certification provision. Plaintiffs can
circulate their petition and their summary to whomever they please. The only difference is that
Ohio has refused to endow that expression with legislative significance. Nothing in Buckley
suggests that the First Amendment is implicated by a state’s regulation of the processes by which
citizens may engage in lawmaking.
To be sure, the Buckley Court cited other First Amendment cases about election regulations
like Anderson v. Celebrezze, 460 U.S. 780 (1983). But in doing so, Buckley did not demand that
all ballot initiative regulations be subject to First Amendment scrutiny. Rather, Buckley cited those
cases to support the uncontroversial proposition that “no litmus-paper test” could easily resolve
First Amendment challenges in the electoral context. Buckley, 525 U.S. at 192 (citation omitted).
That’s why the Court felt compelled to “detail” why, just “as in Meyer,” the laws at issue

-- 19 of 38 --

No. 24-3354 Brown et al. v. Yost Page 20
“significantly inhibit[ed] communication with voters about proposed political change”—thereby
triggering First Amendment scrutiny. Id.
Moreover, if the principal dissent were correct that “the Supreme Court has indicated that
regulations of the ballot-initiative process should be analyzed in the same terms as other election-
related regulations,” then at least four sitting Supreme Court Justices are curiously unaware of
their own Court’s precedent. In Little v. Reclaim Idaho, the district court had ordered Idaho to
certify a citizen-initiated ballot measure, even though the initiative’s proponents hadn’t obtained
the requisite number of signatures under state law. 140 S. Ct. 2616 (2020). The Supreme Court
stayed the district court’s injunction. Concurring in the grant of the stay, Chief Justice Roberts
(joined by Justices Alito, Gorsuch, and Kavanaugh) expressed doubts that neutral, procedural
ballot-initiative regulations like signature requirements implicated the First Amendment at all. His
opinion observed that states have “considerable leeway to protect the integrity and reliability of
the initiative process.” Id. at 2616 (Roberts, C.J., concurring in the grant of stay) (quoting Buckley,
525 U.S. at 191). He then emphasized that “[n]othing in the Constitution requires Idaho or any
other State to provide for ballot initiatives.” Id. at 2617. Only then did his opinion proceed to
“assum[e]” for the sake of argument that neutral, procedural ballot initiative regulations like
signature requirements might “implicate the First Amendment” at all. Id. If the principal dissent
were correct, then four Justices were wrong to imply that certain ballot initiative regulations are
not subject to any First Amendment scrutiny.
The principal dissent’s analysis also conflicts with the en banc Tenth Circuit’s decision in
Walker. There, the Tenth Circuit cited the relevant Supreme Court precedents like Meyer and
Buckley and drew from them a common-sense rule: “Although the First Amendment protects
political speech incident to an initiative campaign, it does not protect the right to make law, by
initiative or otherwise.” Walker, 450 F.3d at 1099.
In sum, I don’t read the Supreme Court to have “cordoned off the ballot-initiative process
from First Amendment scrutiny.” Moore Dissenting Op. 32. But at the same time, the Supreme
Court also hasn’t mandated that all ballot initiative regulations be subject to First Amendment
scrutiny. Instead, the Court has focused on whether ballot initiative regulations restrict speech.

-- 20 of 38 --

No. 24-3354 Brown et al. v. Yost Page 21
Because the First Amendment concerns itself with government regulation of the freedom of
speech, I propose that we do the same.
B.
Accepting for the sake of argument that regulations of the lawmaking process do not
implicate the First Amendment, the principal dissent nevertheless argues that Ohio’s law regulates
speech. The dissent rejects my contention that Ohio’s law merely regulates citizens’ ability to
make law.
In doing so, the dissent relies on Doe v. Reed, 561 U.S. 186 (2010). In that case, the
Supreme Court subjected to First Amendment scrutiny a state’s requirement that referendum
signatories disclose their names and contact information. Id. at 194–95. The forced disclosure of
a citizen’s signature on an initiative, the court held, implicated speech. Id. But the state observed
that it had attached legislative consequences to that speech, since each signature was a “legally
operative legislative act” that increased the likelihood of the initiative’s reaching the ballot. Id. at
195. And the state argued that, because it had attached legislative consequences to the speech, the
state could insulate its regulation of that speech from First Amendment scrutiny. Id. The Court
rejected that argument. Id. It reasoned that the state’s “adding . . . legal effect to an expressive
activity” doesn’t take the expressive activity outside the protection of the First Amendment. Id.
In other words, a state can’t use a referendum or initiative process as an excuse to regulate speech
free of the First Amendment’s strictures.
This case isn’t Doe—it’s the inverse of Doe. Here, plaintiffs object to the state’s refusal
to give legal effect to their speech. They argue that the state has violated the First Amendment by
refusing to attach legislative consequences to their initiative proposal. But plaintiffs have no First
Amendment right to require the state to endow their speech with legal significance. Plaintiffs
remain free to speak as they please; they can circulate the text of their proposal and even their
preferred summary however they see fit. Yet they don’t have a First Amendment right to require
the state to bestow the force of law upon their arguments and ideas. Doe merely held that just
because a state has attached legal significance to speech, that doesn’t remove the speech from the
First Amendment’s protection. Id. Nowhere does Doe say that states are required to endow

-- 21 of 38 --

No. 24-3354 Brown et al. v. Yost Page 22
citizens’ speech with legislative effect. That makes sense, since the First Amendment protects a
right to speak—not a right to legislate.
* * *
When faced with First Amendment challenges, we should be wary of putting the
government to an “all-or-nothing choice” that risks resulting in “less speech, not more.” Ark.
Educ. Television Comm’n v. Forbes, 523 U.S. 666, 680 (1998). “Nothing in the Constitution
requires [Ohio] or any other State to provide for ballot initiatives.” Little, 140 S. Ct. at 2617
(Roberts, C.J., concurring in the grant of stay). After all, initiative processes are a relatively recent
phenomenon; states first implemented them at the turn of the twentieth century. Ariz. State
Legislature v. Ariz. Indep. Redistricting Comm’n, 576 U.S. 787, 793–94 (2015) (surveying
history); see also Comm. to Recall Dan Holladay v. Wiley, No. 23-35107, 2024 WL 4551464, at
*5–9 (9th Cir. Oct. 23, 2024) (Bumatay, J., dissenting from the denial of rehearing en banc)
(same).1 Far be it from the federal courts to make a tool of direct democracy so potent that states
may be induced to remove it from citizens’ toolbox of self-governance.
Luckily, the First Amendment doesn’t require us to risk catalyzing such perverse
consequences. States face First Amendment constraints when they directly regulate speech within
the initiative process, but they don’t face First Amendment scrutiny when they structure and limit
what laws and amendments citizens can enact by direct democracy. Thus, I would deny plaintiffs’
motion for a preliminary injunction because they haven’t shown that they’re likely to succeed on
the merits.
Ultimately, Ohioans who are frustrated by the Attorney General’s exercise of his
gatekeeping authority under the fair-and-truthful certification provision must turn to state courts
for legal relief. See Ohio Rev. Code § 3519.01(C). And if that doesn’t work, they can focus their
1The one exception was the Georgia constitution of 1777. See Comm. to Recall Dan Holladay, 2024 WL
4551464, at *6 (Bumatay, J., dissenting from the denial of rehearing en banc). Georgia imposed high barriers—in
terms of both the number of signatures required and how widely distributed throughout the state they had to be—that
initiative proponents had to clear. Id. These onerous requirements “coexisted with a somewhat analogous protection
of public discourse elsewhere in the constitution.” Id. So, in Georgia at the founding, “the protection of public
discourse was not understood to demand flexibility in ballot qualification rules.” Id.

-- 22 of 38 --

No. 24-3354 Brown et al. v. Yost Page 23
efforts on unseating him from office by speaking, writing, and persuading their fellow citizens.
The First Amendment will be there to protect their efforts.

-- 23 of 38 --

No. 24-3354 Brown et al. v. Yost Page 24
_________________
DISSENT
_________________
KAREN NELSON MOORE, Circuit Judge, dissenting. Plaintiffs are proponents of
“Protecting Ohioans’ Constitutional Rights,” a proposed state constitutional amendment that
would create a private right of action without recourse to governmental immunity for state officials
when they violate Ohioans’ state constitutional rights. Because Ohio Attorney General David Yost
has repeatedly refused to certify a summary of their initiative for nearly two years, Plaintiffs have
been unable even to begin collecting signatures in support of their amendment. Plaintiffs have
asked us for a preliminary injunction requiring Yost to certify a summary they submitted to Yost
on March 5, 2024. We can and should grant them relief so that they can exercise their First
Amendment rights while this litigation proceeds.
I. BACKGROUND
Ohio law requires citizens to take several steps before they can place a proposed
constitutional amendment on the ballot. As relevant here, petitioners must collect 1,000 supporting
signatures for their proposed amendment and then submit the proposed amendment, along with a
summary, to the Ohio Attorney General for review. Ohio Rev. Code § 3519.01(A). The Ohio
Attorney General reviews that summary within ten days and determines whether “the summary is
a fair and truthful statement of the proposed . . . constitutional amendment.” Id. If the summary
is fair and truthful, the Ohio Attorney General certifies the petition to the Ohio ballot board for
approval. Id. Only after petitioners have gained these approvals can they begin collecting the
approximately 400,000 signatures they need to get on the ballot. In short: unless the Ohio
Attorney General deems the summary “fair and truthful,” petitioners cannot circulate their petition.
Obtaining approval of Plaintiffs’ petition summary has proven to be a Sisyphean task. Over
the past 21 months, Plaintiffs have submitted a proposed summary to Yost seven times; each time,
Yost found a nit or two to pick. See D. 44 (Pls.’ En Banc Suppl. Br. at 3). For example, Yost
rejected one proposed summary in part because it was “misleading to the extent that it falsely
purports to set forth an exhaustive list of potential venues” when the summary addressed only the

-- 24 of 38 --

No. 24-3354 Brown et al. v. Yost Page 25
venue for lawsuits against one public employee but did not address venue for lawsuits against
multiple public employees who do not live or work in the same county. R.1 (Compl., Ex. 2,
November 17, 2023, Letter at 2) (Page ID #26). This was misleading, according to Yost, even
though it was not in conflict with the proposed amendment itself. See id. That was not the only
problem he found. Yost also identified a purportedly misleading conflict in that the proposed
amendment defined the State to mean “the State of Ohio, including, but not limited to, the offices
of all elected state officers and all departments and other instrumentalities of the State of Ohio,”
and the summary provided only that a cause of action would be available against “the State of
Ohio, its officers, departments and instrumentalities,” omitting specific mention of “offices of all
elected state officers.” Id. at 3 (Page ID #27). Yost added that these were “just a few examples of
the summary’s omissions and misstatements” but declined to give any further guidance on what
(if anything) Plaintiffs could do to succeed. Id. at 5 (Page ID #29).
And, indeed, Yost had not yet aired all his concerns, as his response to the March 5, 2024,
petition would reveal. This time, Yost took issue with the way Plaintiffs implemented his prior
suggestions and, for the first time, challenged the amendment’s title. R. 1 (Compl., Ex. 4, March
14, 2024, Letter at 2) (Page ID #39). He claimed that the title, which Plaintiffs included in prior
submissions, was misleading because it “offers a subjective hypothesis (that eliminating such
defenses will ‘protect’ the constitutional rights of citizens) regarding the proposed amendment in
lieu of an objective description of its character and purport (that it creates a cause of action
notwithstanding those defenses).” Id. Basically, Yost rejected the title because he did not agree
that removing qualified immunity would protect citizens’ constitutional rights.
No end in sight, Plaintiffs challenged Yost’s rejection of the March 5, 2024 summary in
the state supreme court. That too proved difficult: after Yost opposed expedited consideration,
the Ohio Supreme Court refused to supply any expedited review, effectively frustrating Plaintiffs’
efforts to obtain any meaningful relief before the November 2024 election. See D. 44 (Pls.’ En
Banc Suppl. Br. at 4). So, Plaintiffs turned to the federal courts, asserting that Yost’s repeated
refusals, coupled with a lack of timely review in state court, violated their First Amendment rights.1
1Meanwhile, Plaintiffs tried submitting yet another proposed amendment and summary—this one with no
title at all—which Yost rejected for that reason. See D. 44 (Pls.’ En Banc Suppl. Br. at 3). Plaintiffs challenged that

-- 25 of 38 --

No. 24-3354 Brown et al. v. Yost Page 26
Plaintiffs sought a preliminary injunction, which the district court denied. On appeal, a panel of
this court reversed and granted preliminary relief enjoining Yost from enforcing § 3519.01 and
ordering him to send Plaintiffs’ proposed amendment and the most recent summary to the ballot
board for the next phase of the process. Brown v. Yost, 103 F.4th 420, 446, vacated, 104 F.4th
621, 622 (6th Cir. 2024) (en banc). Yost protested to the en banc court, which granted rehearing
and vacated the panel decision, vitiating Plaintiffs’ relief. Brown, 104 F.4th at 622. We heard this
case en banc six days before the November 2024 election. Now, the majority claims it is too
late for us to do anything because the requested preliminary injunction is moot. Respectfully,
I disagree.
II. MOOTNESS
Plaintiffs’ motion for a preliminary injunction is not moot. Plaintiffs requested a
preliminary injunction requiring Yost to “immediately certify Plaintiffs’ citizen-initiative entitled
‘Protection of Ohioans’ Constitutional Rights,’ which was submitted to [Yost] on March 5, 2024,
to the Ohio Ballot Board for further proceedings under Ohio law.” R. 2-1 (Proposed Order) (Page
ID #64). Such an injunction is not inextricably tied to the now-past November 2024 election.
Moreover, granting this relief would still be meaningful to Plaintiffs. Plaintiffs’ petition would
move forward in the ballot-initiative process. And, after review from the Ohio Ballot Board,
Plaintiffs could begin collecting signatures to place their initiative on a future ballot. Affording
Plaintiffs’ requested relief would have the “practical effect” of remediating Plaintiffs’ First
Amendment injury while the case proceeds on the merits.
Under Article III, “federal courts are without power to decide questions that cannot affect
the rights of litigants in the case before them.” DeFunis v. Odegaard, 416 U.S. 312, 316 (1974)
(per curiam) (quoting North Carolina v. Rice, 404 U.S. 244, 246 (1971)). Accordingly, when
intervening events cause parties to lose a “legally cognizable interest in the determination whether
[a] preliminary injunction was properly granted [or denied],” the appeal of the preliminary
rejection in the Ohio Supreme Court. Recently, after holding in another case that the Ohio Attorney General lacks
statutory authority to review titles, State ex rel. Dudley v. Yost, 2024-Ohio-5166, 2024 WL 4610503, at *1 (Ohio Oct.
30, 2024), the Ohio Supreme Court granted review in Plaintiffs’ case and remanded the untitled version to Yost to
conduct a fair-and-truthful review. See State ex rel. Brown v. Yost, 2024-Ohio-5388.

-- 26 of 38 --

No. 24-3354 Brown et al. v. Yost Page 27
injunction is moot, and we must vacate and remand for consideration of any remaining issues in
the case. Univ. of Tex. v. Camenisch, 451 U.S. 390, 394 (1981) (quoting Ammond v. McGahn, 532
F.2d 325, 328 (3d Cir. 1976)). In determining whether parties retain a legally cognizable interest
in the determination, we consider whether preliminary relief would have any “practical effect”
while the lawsuit is pending. Resurrection Sch. v. Hertel, 35 F.4th 524, 528 (6th Cir. 2022) (en
banc).
A preliminary injunction can lose its “practical effect” for three reasons. First, “the terms
of the injunction . . . [may] have been fully and irrevocably carried out.” Camenisch, 451 U.S. at
398. In Camenisch, the University of Texas’s appeal of a preliminary injunction became moot
after the plaintiff student graduated, having received the benefit of a sign-language interpreter
ordered by the courts. Id. at 393, 398. Second, the defendant may have voluntarily undertaken
the requested action with no reasonable possibility of retraction during pendency of the litigation.
Ohio v. U.S. Env’t Protec. Agency, 969 F.3d 306, 308–09 (6th Cir. 2020). Thus, in Ohio, the
state’s application for a preliminary injunction became moot after the EPA repealed and replaced
the rule challenged by Ohio, and there was no reasonable possibility that the rule would be
reinstated during the pendency of the litigation. Id. at 309. Third, intervening events may foreclose
the court from granting relief with any practical effect. So, when a plaintiff seeks preliminary
relief inextricably tied to a certain election, passage of the election renders the appeal of their
preliminary injunction moot. Thompson v. DeWine, 7 F.4th 521, 524 (6th Cir. 2021) (per curiam).
Plaintiffs’ request for preliminary relief does not fall within any of these categories.
Plaintiffs’ request for relief has not been mooted by compliance with a prior injunction, nor by
defendants’ voluntary cessation of the challenged conduct. To the contrary, Yost has not approved
Plaintiffs’ March 5, 2024 initiative summary. Indeed, he has rejected a subsequently submitted
summary. Passage of the November 2024 election also has not precluded us from granting relief.
While Plaintiffs’ expressed goal was collecting sufficient signatures to place their initiative on the
November 2024 general-election ballot, the relief they sought was not inextricably tied to that
election. They merely sought an order requiring Yost to “immediately certify Plaintiffs’ citizen-
initiative,” R. 2-1 (Proposed Order) (Page ID #64), which, the parties tell us, would allow Plaintiffs

-- 27 of 38 --

No. 24-3354 Brown et al. v. Yost Page 28
to proceed in any future election. We require no “time machine” to grant Plaintiffs effective relief
for the duration of the litigation. Thompson, 7 F.4th at 524.
The majority’s disagreement rests on how we construe Plaintiffs’ request for relief.
According to the majority, Plaintiffs’ request for relief turns exclusively on placing this initiative
on the ballot for the November 2024 election, which has already passed. See Maj. Op. at 4–6. In
reaching this conclusion, the majority looks past the actual requested relief—an injunction
requiring Yost to “immediately certify Plaintiffs’ citizen-initiative,” R. 2-1 (Proposed Order) (Page
ID #64)—and gives undue weight to other portions of the record.
First, the majority mistakenly focuses on how Plaintiffs justified their request for
preliminary relief, rather than on the requested relief itself. The majority correctly notes that
Plaintiffs’ arguments originally emphasized the need to obtain a preliminary injunction in time to
collect 400,000 signatures before the November 2024 election. See Maj. Op. at 5. But the
impossibility of gaining access to the November 2024 ballot without emergency relief went
primarily to Plaintiffs’ claim of irreparable harm, not the nature of the relief sought. See R. 2
(Mem. of L. in Supp. of Emergency Relief at 19) (Page ID #61).
Second, the majority places too much emphasis on language in the March 5, 2024 draft
initiative and summary language, both of which indicate an effective date of January 1, 2025.
According to the majority, this language reveals that Plaintiffs’ requested relief was targeted
exclusively at the November 2024 election. Maj. Op. at 5. Clearly, Plaintiffs’ initial goal was to
place this initiative on the November 2024 ballot, and emergency relief was required if they would
have any chance of doing so. They missed that opportunity after the en banc court vacated the
panel majority’s order granting a preliminary injunction. Equally clearly, Plaintiffs still endeavor
to obtain certification of their initiative and to begin collecting signatures. See D. 65 (Pls.’ Letter
Br. at 4–5). Under these circumstances it is not “impossible” to grant “any effectual relief
whatever.” Church of Scientology of Cal. v. United States, 506 U.S. 9, 12 (1992) (citation omitted).
Plaintiffs’ appeal of their motion for a preliminary injunction is not moot.

-- 28 of 38 --

No. 24-3354 Brown et al. v. Yost Page 29
III. MERITS
Because Plaintiffs’ motion for a preliminary injunction is not moot, I would reach the
merits. I continue to believe that Plaintiffs are likely to succeed on the merits of their First
Amendment challenge and that preliminary relief is accordingly warranted.
The Supreme Court has made abundantly clear that circulating a petition in support of a
ballot initiative is “core political speech.” Meyer v. Grant, 486 U.S. 414, 422 (1988). That is
because “circulation of an initiative petition of necessity involves both the expression of a desire
for political change and a discussion of the merits of the proposed change.” Id. at 421. While
states have “considerable leeway to protect the integrity and reliability of the initiative process,”
the First Amendment requires us to “guard against undue hindrances to political conversations and
the exchange of ideas.” Buckley v. Am. Const. L. Found., Inc., 525 U.S. 182, 191–92 (1999).
Typically, we analyze First Amendment challenges to the ballot initiative process under
the framework set forth in Anderson v. Celebrezze, 460 U.S. 780 (1983), and Burdick v. Takushi,
504 U.S. 428 (1992) (“Anderson-Burdick”). The Anderson-Burdick framework requires us to
weigh the “character and magnitude of the asserted injury” against the “precise interests put
forward by the State as justifications for the burden imposed by its rule.” Anderson, 460 U.S. at
789. The level of scrutiny we apply is determined by the magnitude of the burden. If the burden
is severe, the regulation will be upheld only if it is “narrowly drawn to advance a state interest of
compelling importance.” Burdick, 504 U.S. at 434 (quoting Norman v. Reed, 502 U.S. 279, 289
(1992)). In addition to the traditional Anderson-Burdick analysis, if an election law impacts—
directly or indirectly—“core political speech,” then that law may be challenged under Grant and
American Constitutional Law Foundation, and their progeny. For such a law, strict scrutiny is
triggered. See Am. Const. Law Found., Inc., 525 U.S. at 192; Grant, 486 U.S. at 425. When a law
crosses clear boundaries between regulating the ballot initiative process and restricting core
political speech, we have applied a hybrid approach, considering the “character and magnitude”
of the state’s law in light of Grant and its progeny. Citizens for Tax Reform v. Deters, 518 F.3d
375, 380–87 (6th Cir. 2008).

-- 29 of 38 --

No. 24-3354 Brown et al. v. Yost Page 30
In the panel majority opinion, the panel evaluated § 3519.01 under this hybrid approach,
recognizing that the law had aspects of regulation of the initiative process, as well as aspects of a
restriction on core political speech. Whether the statute is more akin to a restriction on core
political speech or a regulation of the initiative process, in either case direct application of Grant
or application of the hybrid approach is similar.
The first step under the Anderson-Burdick framework is to determine the character and
magnitude of Plaintiffs’ injury. When determining the character and magnitude of Plaintiffs’
injury, we must consider “the combined effect of the applicable election regulations,” and not
measure the effect of each statute in isolation. Libertarian Party of Ohio v. Blackwell, 462 F.3d
579, 586 (6th Cir. 2006). States have the “power to ban initiatives”; however, if initiatives are
allowed under state law, states do not have “the power to limit discussion of political issues raised
in initiative petitions.” Grant, 486 U.S. at 425. Circulation of an initiative petition is “core
political speech,” and restrictions on initiative proponents’ ability to advocate for a petition to
obtain signatures are “wholly at odds with the guarantees of the First Amendment.” Id. at 421–
22, 428 (quoting Buckley v. Valeo, 424 U.S. 1, 50 (1976)). “The First Amendment protects
[Plaintiffs’] right not only to advocate their cause but also to select what they believe to be the
most effective means for so doing.” Id. at 424.
In Grant, the Court found that the appellees’ “core political speech” was restricted by the
state’s “refusal to permit appellees to pay petition circulators.” 486 U.S. at 422. That prohibition
“restrict[ed] political expression” because the law “limit[ed] the number of voices who [would]
convey appellees’ message and the hours they [could] speak,” and thereby “limit[ed] the size of
the audience they [could] reach.” Id. at 422–23. Likewise, the Court determined that the state law
burdened the appellees’ “core political speech” because the law “ma[de] it less likely that appellees
w[ould] garner the number of signatures necessary to place the matter on the ballot,” such that the
appellees were unable to “make the matter the focus of statewide discussion.” Id. Here, Plaintiffs’
summary of their proposed constitutional amendment “involves both the expression of a desire for
political change and a discussion of the merits of the proposed change” and, as explained below,
Ohio’s restrictions implicate the same concerns at issue in Grant. Id. at 421.

-- 30 of 38 --

No. 24-3354 Brown et al. v. Yost Page 31
The Supreme Court has identified at least two ways that restrictions on “core political
speech” impose a severe burden: (1) by restricting one-on-one communication between petition
circulators and potential signatories, and (2) by making it less likely that initiative proponents will
be able to garner the necessary signatures to be placed on the ballot “thus limiting their ability to
make the matter the focus of statewide discussion.” Grant, 486 U.S. at 422–23. Both burdens are
implicated in this case.
First, the law restricts one-on-one communication between petition circulators and
potential signatories due to the combined effect of the Ohio Attorney General’s content-based
review of the petition summary and the lack of timely review of his decision making. The law
authorizes the Ohio Attorney General to review the substance of Plaintiffs’ summary of their
proposed constitutional amendment. If Yost determines that the summary is misleading, he
declines to certify the summary and Plaintiffs must alter their summary to conform to his wishes
(petty and self-contradictory as they may be). Should Plaintiffs disagree with his proposed
changes, their only recourse is review in the Ohio Supreme Court, which may—as in this case—
come only months later. In essence, Plaintiffs must speak in the terms that Yost wants them to
speak, or they must wait months or years to begin circulating their petition summary, which
“involves both the expression of a desire for political change and a discussion of the merits of the
proposed change.” Grant, 486 U.S. at 421.
Likewise, the second concern from Grant—that Plaintiffs will be less able to garner the
necessary signatures to be placed on the ballot “thus limiting their ability to make the matter the
focus of statewide discussion,” 486 U.S. at 423, is also present here. Without timely review,
§ 3519.01 allows the Attorney General to reject a summary in perpetuity such that Plaintiffs are
never able to begin collecting signatures in support, much less garner the number of signatures
required. Thus, the time that Plaintiffs will have to speak with potential signatories about the
proposed amendment as well as how widely they can convey the message has been meaningfully
diminished, if not altogether eliminated.
Because the law imposes a severe burden on Plaintiffs’ First Amendment activities, the
law can stand only if it survives strict scrutiny, that is, if the government can demonstrate that the
law is “justified by a compelling state interest and was narrowly tailored in pursuit of that interest.”

-- 31 of 38 --

No. 24-3354 Brown et al. v. Yost Page 32
Kennedy v. Bremerton Sch. Dist., 597 U.S. 507, 525 (2022). Yost has asserted an interest in
“voter education, fraud deterrence, and the integrity of the initiative process and election.” D. 19
(Appellee Br. at 35). Those are compelling interests. See Susan B. Anthony List v. Driehaus, 814
F.3d 466, 473 (6th Cir. 2016). But the law—and Yost’s administration of it—are not narrowly
tailored to those interests. At first blush, the law may appear tailored: it limits Yost’s discretion
to review of the fairness and truthfulness of the summary and gives him just ten days to complete
his review before Plaintiffs have a right of action in the Ohio Supreme Court. Yet, the facts of this
case tell a different story. Yost has issued multiple denials of Plaintiffs’ request for certification,
and Plaintiffs have been unable to obtain any meaningful or timely judicial review in the state
court. Accordingly, it is not likely that this is the least-restrictive means to satisfy the state’s
compelling interests. Because Plaintiffs are likely to succeed on their First Amendment claim, I
would grant the motion for preliminary injunction. Bays v. City of Fairborn, 668 F.3d 814, 819
(6th Cir. 2012).
IV. RESPONSE TO JUDGE THAPAR
Judge Thapar disagrees and argues that Plaintiffs are unlikely to succeed on the merits. His
view, as I understand it, rests on the contention that the First Amendment distinguishes between
“(1) regulations of citizens’ ability to speak to one another as part of [the ballot initiative] process;
and (2) regulations of citizens’ ability to make law as part of that process.” Thapar Op. at 9–10.
He argues that the First Amendment applies to the first but not the second. Id. at 10. Accordingly,
he would exempt all ballot-initiative regulations from the First Amendment, unless those
regulations directly targeted the speech or payment of petition circulators. This position is
incorrect as a matter of law, and it does not determine the outcome of this case.
To begin, Judge Thapar is wrong that the First Amendment does not apply to regulations
of the ballot-initiative process. The Supreme Court has never cordoned off the ballot-initiative
process from First Amendment scrutiny. To the contrary, the Supreme Court has indicated that
regulations of the ballot-initiative process should be analyzed in the same terms as other election-
related regulations. In American Constitutional Law Foundation, the Supreme Court explained
that “‘no litmus-paper test’ will separate valid ballot-access provisions from invalid interactive
speech restrictions; we have come upon ‘no substitute for the hard judgments that must be made.’”

-- 32 of 38 --

No. 24-3354 Brown et al. v. Yost Page 33
525 U.S. at 192 (quoting Storer v. Brown, 415 U.S. 724, 730 (1974)); see id. at 187, 192 n.12;
206–09 (Thomas, J., concurring in the judgment); 215–16 (O’Connor, J., concurring in the
judgment in part and dissenting in part). In setting forth that principle, the Supreme Court relied
on its election-law precedents, which apply even to regulations of the mechanics of the election
process. See id. at 192 (citing Storer, 415 U.S. at 730 (party-affiliation requirements); Timmons
v. Twin Cities Area New Party, 520 U.S. 351, 359 (1997) (third-party ballot access); Anderson,
460 U.S. at 789–90 (candidate filing deadline)).
Construing the “litmus test” referred to in American Constitutional Law Foundation as just
a determination whether the law restricts Plaintiffs’ ability to express themselves about a ballot
initiative, see Thapar Op. at 19–20, overlooks the significant expressive aspects of the ballot-
initiative process. The process of proposing a ballot initiative is bound up with the “the expression
of a desire for political change and a discussion of the merits of the proposed change.” Grant, 486
U.S. at 421. As the First Circuit has explained, “[a] state initiative process provides a uniquely
provocative and effective method of spurring public debate on an issue of importance to the
proponents of the proposed initiative.” Wirzburger v. Galvin, 412 F.3d 271, 276 (1st Cir. 2005).
Regulations that inhibit petitioners from qualifying their initiative for the ballot “indirectly impact
core political speech” by preventing those issues from becoming “‘the focus of statewide
discussion,’” thereby “‘reducing the total quantum of speech on a public issue.’” Angle v. Miller,
673 F.3d 1122, 1133 (9th Cir. 2012) (quoting Grant, 486 U.S. at 423) (alteration omitted).
Additionally, many regulations of the ballot-initiative process—like the regulation at issue in this
case—to some degree affect extrinsic First Amendment rights, such as the right to express oneself
about a ballot initiative or to associate with others of similar persuasion. Applying the Anderson-
Burdick test, as we do in this circuit, see, e.g., Schmitt v. LaRose, 933 F.3d 628, 639 (6th Cir.
2019), is well suited to determining the burden those regulations place on rights within and without
the ballot-initiative process and allows us to calibrate our review accordingly. Judge Thapar has
identified no binding case law from the Supreme Court that would counsel otherwise.2
2Chief Justice Roberts’s brief concurrence in the grant of a stay in Little v. Reclaim Idaho, reached only the
questions (1) whether it was “reasonably likely” that the Court would grant certiorari to resolve a split among the
circuits concerning the application of First Amendment scrutiny to “neutral, procedural regulation[s] [that] inhibit[] a
person’s ability to place an initiative on the ballot,” and (2) whether there was a “fair prospect that the Court [would]

-- 33 of 38 --

No. 24-3354 Brown et al. v. Yost Page 34
Even assuming that a bright line should be drawn between regulation of the ballot-initiative
process and regulation of speech about ballot initiatives, the facts of this case indicate that
§ 3519.01 falls on the speech-regulation side of the line. Recall what is at issue here: a summary
of a proposed ballot initiative. The summary is not the text of the initiative, nor is it the language
that will appear on the ballot, see Ohio Rev. Code § 3505.06(E); it is a description of the proposed
amendment that appears only on the petition that voters sign to place the measure on the ballot, id.
§ 3519.05(A). It is an important tool, Plaintiffs tell us, to spread the message and advocate for
their cause. See D. 44 (Pls.’ En Banc Suppl. Br. at 11 (contending that “[u]nlike language that
appears on the ballot or in the text of the proposed legislation, which might properly be considered
government speech, the summary at issue here is used only during circulation of the petition.
Written by Plaintiffs, the summary is their political speech advocating for their proposed change
. . . .”)). Judge Thapar does not address the distinction between the summary and the initiative
itself. Instead, he collapses the summary with the initiative language when he states that “when
the government regulates the content of initiative petitions and their summaries . . . it’s regulating
what sorts of laws citizens can enact.” Thapar Op. at 11; see id. at 15 (contending that when “Yost
enforces the fair-and-truthful certification provision, he isn’t directly regulating speech; he’s
restricting what sorts of citizen initiative proposals can become law”). But the summary is not a
law that citizens enact. It is political expression, over which the Attorney General has attempted
to exercise editorial control.
That the summary might have some legal significance considering its placement on the
petition does not negate its expressive nature. In Doe v. Reed, the Supreme Court addressed
whether forced disclosure of a referendum petition—specifically, the names and contact
information of the signatories—implicated the First Amendment. 561 U.S. 186, 194–95 (2010).
The Secretary of State argued that “signing a petition is a legally operative legislative act and
therefore ‘does not involve any significant expressive element.’” Id. at 195 (citation omitted). The
Court expressly rejected this argument, reasoning that it did “not see how adding such legal effect
to an expressive activity somehow deprives that activity of its expressive component” particularly
set aside” a court order applying such scrutiny. 140 S. Ct. 2616, 2616–17 (2020) (Roberts, C.J., concurring in the
grant of stay).

-- 34 of 38 --

No. 24-3354 Brown et al. v. Yost Page 35
because “the signature expresses a particular viewpoint.” Id. (“Petition signing remains expressive
even when it has legal effect in the electoral process.”). So too here. The summary of the proposed
ballot initiative expresses at least the viewpoint that its proponents think “that the question should
be considered ‘by the whole electorate,’” and “adding such legal effect to an expressive activity”
does not “deprive[] that activity of its expressive component.” Id. (quoting Grant, 486 U.S. at
421).3
Judge Thapar also attempts to wipe away the First Amendment implications by pointing to
statutes from other states that, he suggests, indicate that “Ohio isn’t in a league of its own” by
authorizing content-based review of the petition summaries. Thapar Op. at 11. But the authority
on which he relies provides scant support for his position. For example, the Montana and Maine
statutes he cites authorize officials to review language that will appear on the ballot itself or
become law, not summaries of laws that petitioners propose. See, e.g., Mont. Code Ann. § 13-27-
212, 213, 226(1)–(3) (ballot statements and “yes and no” statements that appear on ballot); Me.
Rev. Stat. Ann. Tit. 21-A, § 901(3-A) (bill titles and headnotes that form part of proposed law).
Contrary to Judge Thapar’s contention, amici inform us that Ohio is unique in the discretion that
it affords the attorney general to “judge the content of proposed amendment summaries, placing
his opinion between citizens seeking constitutional change and the voters they want to speak to.”
D. 49 (Br. of Direct Democracy Scholars et al. Supp. Pls. & Reversal at 10). Judge Thapar’s
citations to out-of-circuit cases upholding restrictions on the subject matter of ballot initiatives are
similarly inapposite. See Thapar Op. at 11–13.
Perhaps to assuage concerns about the First Amendment implications of this case, Judge
Thapar emphasizes that Ohio does not “bar someone from engaging in ‘unfair’ or ‘untruthful’
conversations when they are seeking signatures on the campaign trail.” Thapar Op. at 14. Nor
does Ohio prevent Plaintiffs from circulating some petition or engaging with fellow citizens on the
3Judge Thapar’s argument that this case is “the inverse of Doe” rests on the mistaken contention that
Plaintiffs’ injury is merely the state’s refusal to give legal effect to their speech. See Thapar Op. at 21–22. As I
explain above, Yost’s continued refusals to certify the ballot-initiative summary burdens Plaintiffs’ ability to circulate
their initiative petition and to use their chosen speech to describe the proposed amendment. That Plaintiffs can
additionally circulate a summary of their own writing does not vitiate the burden imposed by Ohio law and Yost’s
conduct here.

-- 35 of 38 --

No. 24-3354 Brown et al. v. Yost Page 36
topic of eliminating state qualified immunity. Id. The trouble with this argument is that the
Supreme Court has already rejected it: “[t]hat [Plaintiffs] remain free to employ other means to
disseminate their ideas does not take their speech through petition circulators outside the bounds
of First Amendment protection.” Grant, 486 U.S. at 424. “The First Amendment protects
[Plaintiffs’] right not only to advocate their cause but also to select what they believe to be the
most effective means for so doing.” Id.
* * *
In closing, I note that the need for a preliminary injunction is especially acute here. Yost
has rejected Plaintiffs’ summaries time and time again, for reasons that appear increasingly
specious. Meanwhile, Plaintiffs have been unable even to begin circulating their petition. Because
Plaintiffs’ motion for a preliminary injunction is not moot and because they are likely to succeed
on the merits of that motion, I would grant relief. I respectfully dissent.

-- 36 of 38 --

No. 24-3354 Brown et al. v. Yost Page 37
_________________
DISSENT
_________________
KETHLEDGE, Circuit Judge, dissenting. I write to offer an additional reason why this
case is not moot. Some eight times now, on grounds increasingly dubious, Ohio’s Attorney
General has refused to certify any iteration of the plaintiffs’ summaries of their proposed
amendment to the Ohio Constitution. Eventually the plaintiffs moved in the district court for a
preliminary injunction requiring that the AG “immediately certify” the summary they submitted
to him “on March 5, 2024.” Mot. at 1. The district court denied that motion. The result has been
that, for about 21 months now, the plaintiffs have not been able to circulate their petitions—which
is itself core political activity protected by the First Amendment. Buckley v. American
Constitutional Law Foundation, 525 U.S. 182, 186–87 (1999).
We could change that status quo tomorrow, by ordering the AG to approve the March 2024
summary. That was the injunction the plaintiffs requested; and that request by its terms was not
limited to purposes of the 2024 election. See Mot. at 1. Moreover, all the signatures the plaintiffs
obtain would undisputedly remain valid for as long as it takes them to obtain the number of
signatures required by Ohio law; and the plaintiffs want to start collecting those signatures now.
It is therefore not “impossible” for us to grant the plaintiffs “effectual relief” in this appeal. Church
of Scientology of California v. United States, 506 U.S. 9, 12 (1992). By any practical measure, the
contrary is true; and the State and plaintiffs alike, for their part, both say this appeal remains live.
Nothing in the March 2024 summary itself changes these conclusions. True, that summary
states—in the last sentence of many—that the proposed amendment “shall take effect on January
1, 2025.” Compl. Ex. 3, at 2. And that effective date is no longer accurate, given that the recent
election has already passed. But that detail—to the extent it has any legal significance in the first
place—does not leave us without means of providing these plaintiffs with effectual relief. When
adjudicating a request for an injunction, a federal court exercises “the power of a court of equity[.]”
United States v. Swift & Co., 286 U.S. 106, 114 (1932). And courts of equity arose in large part

-- 37 of 38 --

No. 24-3354 Brown et al. v. Yost Page 38
to address inequities wrought by ministerial details like this one. See, e.g., Hall v. Meisner, 51
F.4th 185, 191–92 (6th Cir. 2022).
We can address the detail of the effective date here. Federal courts craft preliminary
injunctions rather than merely approve them. And “[c]rafting a preliminary injunction is an
exercise of discretion and judgment, often dependent as much on the equities of a given case as
the substance of the legal issues it presents.” Trump v. Int’l Refugee Assistance Project, 582 U.S.
571, 579 (2017). “In the course of doing so,” moreover, “a court ‘need not grant the total relief
sought by the applicant but may mold its decree to meet the exigencies of the particular case.’” Id.
at 580 (quoting 11A C. Wright, A. Miller & M. Kane, Federal Practice and Procedure § 2947 (3d
ed. 2013)). Here, striking the summary’s last sentence (and, if need be, the same sentence from
the amendment itself)—as part of an injunction requiring the AG to certify the summary—would
be enough to remove the inaccurate effective date. We have authority to make that revision: a
federal court can “modify an injunction in adaptation to changed conditions[.]” Swift & Co., 286
U.S. at 114.
We therefore face no impossibility of granting effectual relief here. To the contrary, we
can both craft and enter an injunction that allows the plaintiffs to begin collecting signatures for
their ballot initiative—which is the relief they have been seeking to this day. I respectfully dissent.

-- 38 of 38 --