United States Court of Appeals For the Eighth Circuit ___________________________
No. 24-2810 ___________________________
Get Loud Arkansas; Vote.org; Nikki Pastor; Trinity Blake Loper,
lllllllllllllllllllllPlaintiffs - Appellees,
v.
Cole Jester, in his official capacity as Secretary of State; Sharon Brooks, in her official capacity as Commissioner of the Arkansas State Board of Election Commissioners; Jamie Clemmer, in official capacity as Commissioner of the Arkansas State Board of Election Commissioners; Bilenda Harris-Ritter, in her official capacity as Commissioner of the Arkansas State Board of Election Commissioners; William Luther, in his official capacity as Commissioner of the Arkansas State Board of Election Commissioners; James H. Smith, III, in his official capacity as Commissioner of the Arkansas State Board of Election Commissioners; Jonathan Williams, in his official capacity as Commissioner of the Arkansas State Board of Election Commissioners,
lllllllllllllllllllllDefendants - Appellants,
Betsy Harrell, in her official capacity as Benton County Clerk; Becky Lewallen, in her official capacity as Washington County Clerk; Terri Hollingsworth, in her official capacity as Pulaski County Clerk,
lllllllllllllllllllllDefendants.
------------------------------
Center for Election Confidence, Inc.; Honest Elections Project; Restoring Integrity and Trust in Elections, Inc.,
lllllllllllllllllllllAmici on Behalf of Appellant(s), United States,
lllllllllllllllllllllAmicus on Behalf of Appellee(s). ____________
Appeal from United States District Court for the Western District of Arkansas - Fayetteville ____________
Submitted: September 18, 2025 Filed: March 31, 2026 ____________
Before COLLOTON, Chief Judge, ERICKSON and STRAS, Circuit Judges. ____________
COLLOTON, Chief Judge.
Get Loud Arkansas and others sued to enjoin a rule promulgated by the Arkansas State Board of Election Commissioners. Get Loud asserts that the rule, which provides that voter registration applications must include a handwritten or “wet” signature, violates a provision of the Civil Rights Act of 1964, 52 U.S.C. § 10101(a)(2)(B). This provision states that no person acting under color of law shall deny the right of any individual to vote in any election because of an error or omission on a voter registration application that is not material in determining whether the individual is qualified under state law to vote in the election. The district court1 determined that the Arkansas rule likely violated federal law, and entered a preliminary injunction after examining the other relevant factors. The Board appeals, and we affirm.
1 The Honorable Timothy L. Brooks, now Chief Judge, United States District Court for the Western District of Arkansas.
-2- I.
To vote in Arkansas, one must be at least 18 years old, a citizen of the United States, a resident of Arkansas, neither adjudicated mentally incompetent nor convicted of a felony, and registered to vote. Ark. Const. art. 3, § 1(a)(1)-(4); id. amend. 51 § 11(a)(4), (5). Amendment 51 to the state constitution addresses voter registration. To register, an applicant must complete a registration form with a “signature or mark” affirming under penalty of perjury that the applicant satisfies these requirements. Id. amend. 51 § 6(a)(3)(F). A person may submit a voting application himself, submit the application through a third-party registration organization, or register at a designated state registration agency. Id. amend. 51 §§ 5(a); 6(a)(2)(G).
In early 2023, Get Loud Arkansas developed an online voter registration tool. The tool enabled an applicant to complete his voter registration application on a mobile device or computer, and to sign the application digitally using a stylus, finger, or computer mouse. The applicant could then authorize Get Loud to print and submit the form on behalf of the applicant.
Before launching the online registration tool, Get Loud’s executive director contacted the Arkansas Secretary of State to confirm that the tool complied with Arkansas law. Get Loud asked the Secretary whether a wet signature is required on a voter registration application. In an e-mail to Get Loud’s deputy director, an assistant director of elections in the Secretary’s office wrote that the office’s attorneys concluded that Arkansas law does not require a wet signature.
Get Loud introduced the online registration tool in January 2024. After launching the tool, Get Loud observed a significant increase in completed registrations at in-person registration events. Because Get Loud no longer needed to provide voters with paper applications at events, the organization’s voter registration
-3- efforts were more successful and cost-effective. The tool also enabled Get Loud to register voters in all seventy-five counties in Arkansas. On February 26, an article in the Arkansas Times reported that Get Loud had registered 358 voters using the tool and that 78 percent of the voters registered were under twenty years old.
On February 28, two days after the article reported the success of Get Loud’s online tool, Arkansas Secretary of State Thurston instructed county clerks to stop accepting voter registration applications executed by electronic signature. On March 12, Secretary Thurston requested a formal opinion from Arkansas Attorney General Griffin on the legality of digital signatures. Attorney General Griffin issued a formal opinion that “an electronic signature or mark is generally valid under Arkansas law.” General Griffin wrote that “given the historical acceptance of signatures produced through a variety of means, the widespread acceptance of electronic signatures, and the fact that Amendment 51 does not contain any restriction on how a ‘signature or mark’ may be made, I believe that an electronic signature satisfies Amendment 51’s ‘signature or mark’ requirement.”
After receiving the attorney general’s opinion, the Board adopted an emergency rule that narrowed the definition of “signature or mark” to include only wet signatures. The new rule excludes any signature or mark that “utilizes a computer to generate or recreate the applicant’s signature or mark.” The emergency rule required a registrant to make “a handwritten wet signature or handwritten wet mark . . . on a Registration Application Form with a pen or other writing device that is physically moved across the form and that forms the applicant’s signature or mark on the paper form.” The emergency rule took effect on May 4 and expired on September 1. After opening the rule to public comment, the Board replaced the emergency rule with an identical permanent rule that was approved by the Arkansas Legislative Council and effective on September 2.
-4- The rule had an immediate effect on Get Loud’s operations. To ensure that its voter registration activities complied with the rule, Get Loud spent additional time, labor, and money to redesign its digital registration tool, and hired additional staff to register voters with paper applications. Get Loud’s deputy executive director averred that the pace at which the organization was able to register new voters “declined precipitously” after the rule went into effect. The rule effectively forbade Get Loud to use its most successful voter registration tool during the ten weeks leading up to the 2024 federal elections.
On June 5, 2024, one month after the rule went into effect, Get Loud and others sued and alleged that the rule violated § 101 of the Civil Rights Act of 1964. This provision, commonly known as the “Materiality Provision” states that:
No person acting under color of law shall . . . deny the right of any individual to vote in any election because of an error or omission on any record or paper relating to any application, registration, or other act requisite to voting, if such error or omission is not material in determining whether such individual is qualified under State law to vote in such election.
52 U.S.C. § 10101(a)(2)(B).
The district court granted Get Loud’s motion for a preliminary injunction. The court enjoined the election commissioners and county clerks from enforcing the rule and from rejecting voter registration applications on the ground that they were completed with a digital or electronic signature.
The Board appeals on several grounds. We consider the question of Article III standing de novo. In reviewing the issuance of a preliminary injunction, we consider the threat of irreparable harm to the movant, the likelihood that the movant will succeed on the merits, the balance between the harm to the movant and the injury that
-5- an injunction would inflict on other parties, and the public interest. Dataphase Sys., Inc. v. C L Sys., Inc., 640 F.2d 109, 113 (8th Cir. 1981) (en banc). The ultimate decision to grant an injunction is reviewed for abuse of discretion, with factual findings examined for clear error and legal conclusions considered de novo. Osthus v. Whitesell Corp., 639 F.3d 841, 844 (8th Cir. 2011).
II.
A.
To establish Article III standing, a plaintiff must show an injury in fact that is fairly traceable to the defendant’s challenged action and is likely redressable by a favorable judicial decision. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560-61 (1992). An organizational plaintiff satisfies the injury-in-fact requirement if a defendant’s actions directly affected and interfered with the organization’s core business activities. FDA v. Alliance for Hippocratic Medicine, 602 U.S. 367, 395 (2024); see Havens Realty Corp. v. Coleman, 455 U.S. 363, 379 (1982).
Get Loud has established that it suffered an injury in fact. The challenged rule directly interfered with Get Loud’s voter registration activities by prohibiting the use of its online registration tool and forcing Get Loud to collect hand-written signatures to register voters. To comply with the rule, Get Loud spent additional time, labor, and resources to redesign its online tool and hired additional staff. Get Loud has established that these actions were necessary to comply with the rule and that the organization was substantially less effective in registering eligible voters after the rule was implemented. The rule impaired Get Loud’s ability to provide voter registration services, and the organization satisfies the injury-in-fact requirement.
The Board argues that Get Loud cannot manufacture an injury merely by diverting resources in response to a defendant’s action. It is true that an organization
-6- cannot “spend its way into standing,” Alliance, 602 U.S. at 394, but that is not what happened here. The rule effectively forbade Get Loud to use its online registration tool. When regulations “require or forbid some action by the plaintiff,” standing is “usually easy to establish.” Alliance, 602 U.S. at 382. Because Get Loud was forbidden to use its online registration tool, which was an important element of its core business activities, the organization has suffered an injury in fact.
Get Loud has also established causation and redressability. Get Loud’s injury was caused by the rule’s prohibition on digital signatures, and the injury can be remedied by an injunction against enforcement of the rule. Thus, Get Loud satisfies the requirements for Article III standing. Because the organizational plaintiff has standing to seek all of the relief at issue, it is unnecessary to address the standing of other plaintiffs.
B.
The district court determined that the Materiality Provision creates a private right of action enforceable through 42 U.S.C. § 1983. See Health & Hosp. Corp. of Marion Cnty. v. Talevski, 599 U.S. 166, 183-91 (2023). At oral argument, the Board waived any challenge to that ruling:
Frankly, on the private right of action, I recognize there is some pretty compelling authority recognizing the private right of action under the Materiality Provision, including in the Callanen case, which I would respectfully submit the Fifth Circuit’s decision in that case should guide this Court’s analysis.
In Vote.org v. Callanen, 89 F.4th 459 (5th Cir. 2023), the Fifth Circuit agreed with decisions of the Third and Eleventh Circuits that the Materiality Provision creates an individual right that is enforceable under § 1983. Id. at 476-78; see Migliori v. Cohen, 36 F.4th 153, 159 (3d Cir.), judgment vacated as moot sub nom. Ritter v.
-7- Migliori, 143 S. Ct. 297 (2022) (mem.); Schwier v. Cox, 340 F.3d 1284, 1297 (11th Cir. 2003). When asked to elaborate on the Board’s position, counsel confirmed that he was waiving any challenge to the district court’s decision on this issue: “In keeping with the Callanen opinion, which recognized the private right of action, I would be willing to concede that issue here today.” We therefore accept as unchallenged in this case the district court’s conclusion that the Materiality Provision creates an individual right that is enforceable under § 1983.
C.
The Materiality Provision prohibits denial of an individual’s right to vote based on “an error or omission” in a registration document if such error or omission is “not material in determining whether such individual is qualified under State law to vote.” 52 U.S.C. § 10101(a)(2)(B). An error is “material” when it is “important” or has “influence or effect” over a decision maker. Material, Black’s Law Dictionary 1128 (4th ed. 1951). The legal meaning of materiality typically sets a “demanding” standard; a matter is “material” only if a reasonable person “would attach importance to [it] in determining his choice of action.” Universal Health Servs., Inc. v. United States, 579 U.S. 176, 193-94 (2016) (quoting Restatement (Second) of Torts § 538, at 80). Under the Materiality Provision, a state actor cannot deny the right to vote based on an error or omission in a registration form unless the error is “material in determining” whether an applicant is qualified to vote. 52 U.S.C. § 10101(a)(2)(B).
When a person applies to register to vote in Arkansas, the registrant must attest under penalty of perjury that she meets every qualification to vote. Ark. Const. amend. 51 § 6(a)(2)(B). The rule provides that the registrant’s attestation must be made with a handwritten signature or mark on an application.
There is no evidence, however, that Arkansas election officials use the type of signature on a registration form in determining whether an applicant is qualified to
-8- vote. The district court found that “the record evidence shows that the ‘wetness’ of a signature does not affect county officials’ determinations of qualifications at all.” Ample evidence supports the finding. Before the Board adopted the new rule, Arkansas election officials received applications with both wet and digital signatures, and officials did not consider “the type of instrument used in signing or marking a voter registration application as a factor in determining whether the applicant is or is not qualified to vote in Arkansas.” County clerks are instructed to accept voter registration applications with “any type of signature or mark,” even if the signature is illegible. County clerks are not trained to analyze or compare signatures. Training sessions “made clear to officials reviewing voter registrations for signature matches that they are not signature analysts, and they are taught to err on the side of the voter.” As the district court explained, the Board did not “present argument or evidence as to how a wet signature—as compared to a digital signature—aids in determining whether a person” is qualified to vote under Arkansas law.
The Board argues that the wet signature requirement is material in determining a voter’s qualifications because the signature confirms an applicant’s existence and identity. This argument is both forfeited and without merit. The evidence shows that signatures are not used by Arkansas election officials to confirm an applicant’s identity or existence at the registration stage. The Board has not shown that a handwritten signature or “X” confirms an applicant’s identity or existence. There is no guarantee that a person who submits a paper registration form marked with an “X” will do so on behalf of a person who exists or that the identity of the person signing or marking the form will match the name on the application. Even if this argument had been raised in the district court, the record does not support a finding that the type of signature is material in determining a voter’s existence and identity.
The Board also maintains that compliance with the signature rule is material because it is useful in preventing voter registration fraud. The Board, citing decisions concerning the constitutional right to vote, suggests that the court should give
-9- considerable deference to a state’s chosen election procedures and balance the state’s asserted interest with the burden imposed on voters. See Crawford v. Marion Cnty Election Bd., 553 U.S. 181, 197 (2008); Purcell v. Gonzalez, 549 U.S. 1, 4 (2006). The Board’s reliance on constitutional right-to-vote cases is misplaced, however, because the Materiality Provision calls for a different analytical framework. The statute’s text contains no interest-burden balance. To pass muster, a rule must actually be “material in determining” whether applicants are qualified to vote. 52 U.S.C. § 10101(a)(2)(B). In this respect, we find ourselves in agreement with the dissenting opinion in Callanen, 89 F.4th at 491-92 (Higginson, J., dissenting), and part company with the Callanen majority’s reliance in this context on cases interpreting § 2 of the Voting Rights Act or the constitutional right to vote.
The Board cites no evidence to support its claim that a requirement of wet signatures on registration applications is material to preventing voter fraud in Arkansas. The record shows that county clerks do not analyze original wet signatures from registration forms in an effort to detect fraud. Even if county clerks did attempt to analyze signatures, the law allows a voter to register without signature by making a simple mark such as an “X” on the form. State law and practice thus do not employ a wet signature requirement as a means to detect fraud, and the absence of a handwritten signature is not material on this record in determining whether a person is qualified to vote.
The Board belatedly suggests in its reply brief that a wet signature carries a “solemn weight” that is material in preventing voter fraud. There is no evidence, however, that a wet signature is more “solemn” than a digital signature, and no showing that the asserted solemnity of a wet signature deters voter fraud. When a person signs a registration form, he attests to the accuracy of the application under penalty of perjury. The sworn statement on Get Loud’s digital registration tool is identical to the sworn statement on the Secretary’s paper form, so the solemn
-10- consequences of submitting a fraudulent registration form are evident to an applicant affixing either type of mark or signature.
In sum, Get Loud presented substantial evidence that the wet signature requirement is immaterial in determining whether an individual is qualified under state law to vote. The Board did not counter with evidence to undermine Get Loud’s showing. The district court did not abuse its discretion or otherwise err in concluding that Get Loud demonstrated a likelihood of success on the merits of its claim that the rule requiring wet signatures violates the Materiality Provision of the Civil Rights Act.
We also agree with the district court that Get Loud showed a likelihood of irreparable harm. An organization suffers irreparable harm when a defendant’s conduct results in lost opportunities to conduct election-related activities such as voter registration. League of Women Voters of U.S. v. Newby, 838 F.3d 1, 8-9 (D.C. Cir. 2016). Get Loud lost opportunities to register voters before the 2024 elections because it was prohibited from using its online tool and was forced to reallocate resources to comply with the rule.
The remaining injunction factors—the balance of equities and the public interest—merge when the government opposes a preliminary injunction. Nken v. Holder, 556 U.S. 418, 435 (2009). There is no dispute that successful registration of qualified voters is in the public interest, and the rule interfered with otherwise lawful voter registration activities. The new rule prevented Get Loud from using its online voter registration tool and significantly impaired its registration efforts. The State, for example, rejected voter registration applications from two individual plaintiffs, because they applied with a digital signature. The State’s arguments on equities and public interest are unpersuasive where the State previously accepted digital signatures without evidence of fraud or other deleterious consequences.
-11- For these reasons, the district court did not abuse its discretion in enjoining state officials from enforcing the rule requiring a wet signature on a voter registration application. The order of the district court is affirmed. The motion to strike the appellants’ post-argument letter is denied.
STRAS, Circuit Judge, dissenting.
Just because there are other ways to fight fraud does not mean that a method Arkansas chose, the wet-signature rule, is “[im]material.” 52 U.S.C. § 10101(a)(2)(B). Handwritten signatures help ensure that those registering to vote are who they claim to be and understand what they are doing. At the very least, the plaintiffs have not made a “clear showing” to the contrary, Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 22 (2008), so I would let Arkansas continue to require them on voter-registration forms.
I.
Under federal law, state voting rules must be “material.” 52 U.S.C. § 10101(a)(2)(B) (prohibiting the denial of the right to vote due to “error[s] or omission[s]” that are “not material in determining whether such individual is qualified . . . to vote”). The Arkansas State Board of Election Commissioners has conceded that the plaintiffs here can enforce this provision,2 so we must decide
2 It took some procedural gymnastics to get to this point. Nothing today forecloses an argument in a future case that no private cause of action exists to enforce the materiality provision. See Turtle Mountain Band of Chippewa Indians v. Howe, 137 F.4th 710, 713, 719 (8th Cir. 2025) (concluding that § 2 of the Voting Rights Act lacks an individually enforceable right under 42 U.S.C. § 1983 based on its “dual focus on the individuals protected and the entities regulated”); Ark. State Conf. NAACP v. Ark. Bd. of Apportionment, 86 F.4th 1204, 1208–11 (8th Cir. 2023) (holding that § 2 of the Voting Rights Act lacks a private right of action based on its “remedial framework,” which leaves enforcement to the Attorney General); see also Ark. United v. Thurston, 146 F.4th 673, 677–79 (8th Cir. 2025) (reaching the same conclusion about § 208 of the Voting Rights Act). -12- whether a rule requiring voters to put pen to paper and sign their name on a registration form is a “material” voting requirement. Id. The short answer is yes: wet-signature rules have been around in various forms for centuries, and to this day remain a powerful way of preventing fraud.
Before signatures, people used seals, which involved making an impression on a piece of paper, usually with wax or a similar substance. See 4 James Kent, Commentaries on American Law *451; Frederick E. Crane, The Magic of the Private Seal, 15 Colum. L. Rev. 24, 24 (1915). Many documents included them, from government decrees and legal records to commercial agreements and private letters. See 1 William Blackstone, Commentaries *260, *351 (discussing use of the “great seal” on various orders of the king); G.V.B. Mining Co. v. First Nat’l Bank of Hailey, 95 F. 23, 32–33 (9th Cir. 1899) (noting the requirement that “important contracts” executed by a corporation include a “corporate seal”); The Complete Letter-Writer; Or, Polite English Secretary 15–16 (Edinburgh, W. Darling 1778) (discussing the use of seals in private correspondence). When used on legal documents, seals were “a mark of authenticity” that provided “evidence of truth” and an “attestation” of their contents. 2 William Blackstone, Commentaries *304–05. A related function was impressing upon the parties the importance of the commitment they were making. See Bates v. Bos. & N.Y.C.R. Co., 92 Mass. (1 Allen) 251, 254 (1865) (explaining that one “purpose” of seals was to “certainly and effectually fix[]” the parties’ “attention” on the act). They were burdensome, but they made “fraud[] less likely.” Warren v. Lynch, 5 Johns. 239, 246 (N.Y. Sup. Ct. 1810) (Kent, C.J.).
Over time, signatures replaced seals. The transition began with the Statute of Frauds, which required certain types of contracts to be in writing and signed by hand. See An Act for prevention of Frauds and Perjuries, 1677, 29 Car. 2, c. 3 (Eng.) (listing the “prevention of . . . fraudulent Practices” as the reason for it); see also Pratalongo v. Larco, 47 Cal. 378, 383 (1874) (“The English statute of frauds
-13- require[d] certain contracts to be in writing, and to be signed by the party to be charged thereby.”). Although the mechanism was different, the reasons were familiar: “giving authenticity to the whole instrument,” underscoring the significance of the act to the signers, and preventing fraud. Barry v. Coombe, 26 U.S. (1 Pet.) 640, 651 (1828) (explaining the signature requirement under Maryland’s statute of frauds); see Waller v. Waller, 42 Va. (1 Gratt.) 454, 480 (1845) (opinion of Allen, J.) (noting that the purpose of a signature on a will is to “give authenticity to the document”); Warner v. Tex. & P. Ry. Co., 164 U.S. 418, 423 (1896) (discussing the widespread adoption of the Statute of Frauds); In re Fransen’s Will, 2 Pa. 202, 206 (1856) (explaining that the signature was one of the “solemnities of execution”). In the American colonies, legislatures began allowing a “scroll”—a hand-drawn “mark”—instead of a seal on certain documents. Game Place, L.L.C. v. Fredericksburg 35, LLC, 813 S.E.2d 312, 317 & n.16 (Va. 2018) (discussing the historical use of scrolls); see Black’s Law Dictionary 1620 (12th ed. 2024) (defining a “scroll” as “a character affixed . . . in place of a seal,” also known as a “scrawl”). It could be anything from a “flourish, tracing, mark, or design,” Pardee v. Johnston, 74 S.E. 721, 722 (W. Va. 1912) (citation omitted), to the letters “L.S.,” short for “locus sigilli,” which in Latin meant “place of the seal,” Black’s Law Dictionary, supra, at 1125 (defining “locus sigilli”); see Hastings v. Vaughn, 5 Cal. 315, 318 (1855) (describing the initials “L.S.” as “a good seal”).
In the eighteenth and nineteenth centuries, some legal documents, like land deeds, needed both a written signature and a seal. See 2 Blackstone, Commentaries *305–06 (noting that courts sometimes required “signing” and “sealing” for a conveyance); 4 Kent, Commentaries *452–53 (same). It may sound old-fashioned today, but it reflected a reluctance to move away from the “distinction and solemnity of seals.” Warren, 5 Johns. at 247; see Bank of U.S. v. Dandridge, 25 U.S. (12 Wheat.) 64, 96 (1827) (“An instrument was much more certainly authenticated by the seal, than by the name of the maker.”); 4 Kent, Commentaries *453 (arguing that allowing a “flourish with the pen” or a “scroll” to suffice was “destroying the character of seals”). It also reflected the reality that forgery of seals was more
-14- difficult “than a stroke of the pen.”3 Oliver Wendell Holmes, The Common Law 272 (1923). Even into the early twentieth century, seals still had a limited role to play in authenticating documents. See Briggs v. Partridge, 64 N.Y. 357, 361–62 (1876) (recognizing the longstanding rule that a contract bearing the seal of an agent did not bind the undisclosed principal, even though an unsealed contract did); Crowley v. Lewis, 146 N.E. 374, 374 (N.Y. 1925) (continuing to distinguish between sealed and unsealed documents under Briggs despite acknowledging that “the importance of the seal in this state [was] much diminished”).
But by then, handwritten-signature requirements had asserted their dominance, largely out of convenience. See Commonwealth ex rel. Arrott Steam-Power Mills Co. v. Arnold, 29 A. 270, 270 (Pa. 1894) (“With the changes of custom and the diffusion of education the written signature has, in practice, taken the place of the seal as the important element of the certification.”); Strong v. Strong, 140 P.2d 386, 388 (Cal. 1943) (noting how “seals were replaced by signatures”). Even states that kept the old sealing requirement would allow parties to satisfy it by “a mere scroll or flourish of the pen.” Holmes, supra, at 273; see Crane, supra, at 24–25, 32–36 (discussing how scrolls or the letters “L.S.” became substitutes for them). Wax seals, and the burdens associated with them, became relics of the past.
We are now right in the middle of a similar evolution from handwritten signatures to digital ones. Rather than printing a document and signing it with a pen,
3 With seals, fraud took a different form. It was fairly simple to lift a seal from one document and place it on another. See Edward Coke, 2 The Third Part of the Institutes of the Laws of England 15 (London, M. Flesher 1644) (discussing how it was considered a “great misprision” to “take wax lawfully imprinted with the great seal, and fix it to a writing purporting a grant from the king”); 4 William Blackstone, Commentaries *83 (discussing the “knavish artifice” of lifting “wax bearing the impression of the great seal off from one patent, and fix[ing] it to another”). Requiring both a signature and a seal was a way to ensure that the person executing the document was who they purported to be. See 2 Blackstone, Commentaries *305; Wood v. Owings, 5 U.S. (1 Cranch) 239, 247 (1803) (explaining that “at common law,” a deed was both “signed” and “sealed”). -15- a person can type their name, draw a symbol with a stylus, or upload a facsimile of their signature into an online form. See Black’s Law Dictionary, supra, at 1666 (defining an “electronic signature” as “[a]n electronic symbol, sound, or process” on a document, including a “typed name” or “a digital image of a handwritten signature”). In many situations, digital signatures get equal weight. See 15 U.S.C. § 7001(a)(1) (declaring that, for “any transaction in or affecting interstate or foreign commerce[,] a signature . . . relating to such transaction may not be denied legal effect, validity, or enforceability solely because it is in electronic form”).
Many, however, does not mean all. To this day, important documents, including some once needing a seal, call for a handwritten signature. See, e.g., id. § 7003(a)–(b) (excepting “wills, codicils, or testamentary trusts”; papers related to “adoption, divorce, or other matters of family law”; and documents governed by all but a few sections of the Uniform Commercial Code from a provision requiring equal recognition of digital signatures); 8 C.F.R. § 103.2(a)(2) (requiring a “handwritten” signature for certain immigration forms unless the instructions allow e-filing). Just in Arkansas, they include wills, promissory notes, bank loans, and certain court filings. See Ark. Code Ann. §§ 25-32-103(a)–(b), -107 (allowing digital signatures for all documents “relating to a transaction” except “wills, codicils, or testamentary trusts” and promissory notes, bank loans, or other transactions governed by the Uniform Commercial Code); Ark. Sup. Ct. Admin. Ord. No. 21, § 8(b) (calling for an “original signature[]” for court documents made “under penalty of perjury or requiring [the] signature of a notary public”).
The reasons have not changed. For those documents, a wet signature is the something extra that authenticates them, “carries solemn weight,” and prevents fraud. Vote.org v. Callanen, 89 F.4th 459, 489 (5th Cir. 2023) (citation omitted); see Weems v. Smith, 237 S.W.2d 880, 883 (Ark. 1951) (“The purpose of . . . requiring wills to be signed at the end is to prevent fraud.”); James v. Mounts, 644 S.W.3d 425, 432 (Ark. Ct. App. 2022) (“The purpose of a signature is to authenticate the writing to which it is affixed . . . .”), vacated on other grounds, 660 S.W.3d 801 (Ark. 2023). When
-16- and whether to require one has always been a judgment call for policymakers, who must balance security with convenience, the considerations that dominated the centuries-long evolution from seals to scrolls to handwritten signatures. See 4 Kent, Commentaries *452–53 (leaving it up to “municipal regulation” whether to require a signature, a seal, or both). The balance they have struck, at least for now, is that certain documents are just too important for digital signatures. See Callanen, 89 F.4th at 489; Aerotek, Inc. v. Boyd, 624 S.W.3d 199, 204–05 (Tex. 2021) (explaining that the authenticity of a “handwritten, wet-ink signature” can be proven by a “witness familiar with the signatory’s handwriting, or an expert who has compared the signature against a genuine specimen,” but that “these authentication methods may not be applicable to a purely electronic signature”).
To protect the “integrity of the voting process,” Common Cause Ind. v. Lawson, 937 F.3d 944, 962 (7th Cir. 2019), Arkansas believes voter-registration forms deserve the same treatment, see Ark. Const. amend. 51, § 6(a)(3)(F) (requiring a “signature or mark” certifying “under penalty of perjury that the applicant meets each requirement for voter registration”); 108.00.24 Ark. Code R. § 6 (defining a “signature or mark” as “a handwritten wet signature or handwritten wet mark”). As the Board’s director explained, physically signing a registration form is a secure way of “affirm[ing] that [someone is] qualified to be a voter.” It provides a means of “identification” and combats “fraud” more effectively than digital signatures, which have only existed for about three and a half decades. See Jeremiah S. Buckley et al., The Law of Electronic Signatures and Records § 2:3 (May 2024 update) (noting that less than half of the states had electronic-signature laws prior to the drafting of the Uniform Electronic Transactions Act in the mid-1990s); cf. Miller v. Thurston, 967 F.3d 727, 741 (8th Cir. 2020) (stating that requiring “in-person signatures” for initiative petitions “prevent[s] fraud”); Kendall v. Balcerzak, 650 F.3d 515, 526 (4th Cir. 2011) (explaining that requiring a signature for comparison in voting helps “detect[] fraudulent or otherwise improper signatures”). If Arkansas is right, the materiality of the requirement speaks for itself. See 52 U.S.C. § 10101(a)(2)(B).
-17- B.
As the court notes, the word “material” can have different meanings. Sometimes, the definition is “relevant” or “pertinent.” Webster’s Third New International Dictionary 1392 (1961). For example, when evaluating the materiality of a false statement, the bar is low: it must have “a natural tendency to influence, or [be] capable of influencing, the decision of the decisionmaking body to which it was addressed.” Neder v. United States, 527 U.S. 1, 16 (1999) (emphasis added) (citation omitted); see Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986) (defining an issue as “material” for purposes of summary judgment if it “might affect the outcome of the suit” (emphasis added)). Other times, it implies something “of real importance or great consequence[,] substantial[,] . . . essential.” Webster’s Third, supra, at 1392. An example is a prosecutor’s obligation to turn over “material” evidence, which depends in part upon whether “a reasonable probability [exists] that, had [it] been disclosed to the defense, the result of the proceeding would have been different.” United States v. Bagley, 473 U.S. 667, 682 (1985) (plurality opinion) (emphasis added); see Universal Health Servs., Inc. v. United States, 579 U.S. 176, 193–94 (2016).
When it comes to the materiality of “error[s] or omission[s]” in voting, the debate centers on those same two possibilities. On one end is “minimal relevance,” Fla. State Conf. of NAACP v. Browning, 522 F.3d 1153, 1174 (11th Cir. 2008): an “error or omission” need only be relevant to “ascertain[ing] whether [someone] is qualified to vote,” Pa. State Conf. of NAACP Branches v. Sec’y Commonwealth of Pa., 97 F.4th 120, 134 (3d Cir. 2024). On the other is something “closer to outcome- determinative,” which would cast aside all but the important, or even essential, requirements of voting. Browning, 522 F.3d at 1174 (describing it as a “higher burden”).
There is no definition of the word “material” in the Civil Rights Act, but the “statute as a whole” provides a few clues about its meaning. Robinson v. Shell Oil,
-18- Co., 519 U.S. 337, 341 (1997). The first is the opening line, which makes clear that its focus is on protecting the right of “qualified” citizens “to vote at all . . . elections[] without distinction of race, color, or previous condition of servitude.” 52 U.S.C. § 10101(a)(1); cf. Bittner v. United States, 598 U.S. 85, 98 n.6 (2023) (“A preamble, purpose clause, or recital is a permissible indicator of meaning.” (quoting Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 217 (2012))). Then come three specific limitations on state election procedures, including the one covering “error[s] or omission[s] on any record or paper” necessary for voting. 52 U.S.C. § 10101(a)(2)(B). The other two prohibit discrimination in “apply[ing] any standard, practice, or procedure” for elections and the use of “literacy test[s],” both of which were ways of disenfranchising black voters in the Jim Crow South. Id. § 10101(a)(2)(A), (C); see Shelby County v. Holder, 570 U.S. 529, 536 (2013) (discussing the history of literacy tests); Perkins v. City of West Helena, 675 F.2d 201, 210–11 (8th Cir. 1982) (describing other historical standards, practices, and procedures that led to disenfranchisement, like “poll taxes” and a “whites[-]only” primary).
What becomes clear, when reading the surrounding provisions, is that “State law” has a role to play in figuring out who is “qualified . . . to vote,” 52 U.S.C. § 10101(a)(2)(B), but it cannot be a pretext for racial discrimination. One frequently discussed example is making a voter “list the exact number of months and days in his age.” Browning, 522 F.3d at 1173 (citation omitted); see Schwier v. Cox, 340 F.3d 1284, 1294 (11th Cir. 2003) (discussing the same example). A rule like that one is “not material” because it is unrelated to “determining whether [an] individual is qualified . . . to vote” in an election. 52 U.S.C. § 10101(a)(2)(B). A voter may have to prove their age, but whether they can mathematically calculate and explain it in various ways is not a qualification of voting. Consistent with the text, materiality implies something closer to relevance: whether the rule is related to a voter’s qualifications or just a discriminatory requirement masquerading as a legitimate one. See Browning, 522 F.3d at 1174; Pa. State Conf. of NAACP Branches, 97 F.4th at 137 (noting that the materiality provision “target[s]” the discriminatory rejection of
-19- registration forms based on “irrelevant mistakes”); Callanen, 89 F.4th at 482 (explaining that the materiality provision “was written . . . to capture well-disguised discrimination”).
If the rule were otherwise, state legislatures would lose their ability to enact “electoral procedures” of all kinds. Carson v. Simon, 978 F.3d 1051, 1062 (8th Cir. 2020) (per curiam); see U.S. Const. art. I, § 4, cl. 1 (entrusting the “Legislature” of each “State” to “prescribe[]” the “Times, Places and Manner of holding Elections”). Indeed, only the best ones—those that are important or essential—would survive, approaching how the narrow-tailoring requirement of strict scrutiny works. See Reed v. Town of Gilbert, 576 U.S. 155, 171 (2015) (requiring “the Government to prove that the restriction furthers a compelling interest and is narrowly tailored to achieve that interest” (citation omitted)); see also Callanen, 89 F.4th at 478 (rejecting “essential” as “a reasonable meaning” of the materiality provision). Relevance, on the other hand, tolerates imperfection, provided that the rule has some bearing on “determining whether [an] individual is qualified . . . to vote.” 52 U.S.C. § 10101(a)(2)(B).
This one does. No one disputes that a signature or mark of some kind is “material.” Id. Indeed, Arkansas’s Constitution makes one a condition of registration. See Ark. Const. amend. 51, § 6(a)(3)(F) (using it to ensure “that the applicant meets each requirement for voter registration”); id. § 3 (prohibiting unregistered voting). Requiring a signature of any kind reminds “individuals who are trying to register” that they must tell the truth and “actually [be] who they say they are.” Callanen, 89 F.4th at 487; see Alderson v. Steinberg, 137 S.W.2d 925, 927 (Ark. 1940) (“By signing a document which may be placed of record, the person so acting says to the world that he or she . . . did, in fact, execute the instrument.”). Handwritten ones just do it better.
The most obvious advantage is the “solemn weight” of putting an actual pen to paper, which forces potential voters to think twice about lying in a way that
-20- drawing on a screen does not. Callanen, 89 F.4th at 489 (citation omitted) (explaining how physically signing a printed document “has some prospect of getting the attention of many applicants and dissuading false statements that an electronic signature . . . does not”); see Eakin v. Adams Cnty. Bd. of Elections, 149 F.4th 291, 315 (3d Cir. 2025) (“Affixing one’s [handwritten] signature onto a legal document does indeed constitute a solemn act.”). They also create a literal paper trail leading back to the signer. See Aerotek, Inc., 624 S.W.3d at 204–05. Digital signatures, by contrast, cannot be compared or verified in the same way, which makes them harder to authenticate. See id. at 205 (noting the potential problem); see also Cloud Corp. v. Hasbro, Inc., 314 F.3d 289, 296 (7th Cir. 2002) (remarking that “a handwritten signature . . . is better evidence of identity than a typed one” (emphasis removed)).
These fraud-fighting features should be no surprise, given that wet signatures have centuries-old roots in the Statute of Frauds. See 29 Car. 2, c. 3; Thompson v. DeWine, 976 F.3d 610, 614, 618 (6th Cir. 2020) (holding that a wet-signature requirement for ballot initiatives furthered a “legitimate” and “compelling” interest in “preventing fraud by ensuring the authenticity of signatures”). Yet after this case, Arkansas must abandon this tried-and-true method in favor of digital ones.
It is not hard to see how doing so will make fraud easier, faster, and harder to detect. All it would take is finding another document with a valid signature, scanning or taking a picture of it, and pasting it in. Cf. 4 Blackstone, Commentaries *83 (describing a similar practice with seals in which a person lifted the “wax bearing the impression of the great seal off from one patent, and fixed it to another”); supra note 2 (explaining how signature requirements helped prevent this form of fraud). Digital signatures could even turn up online, ready for pasting directly into the form. For potential fraudsters interested in swaying elections, the advantages over copying signatures by hand are obvious. Cf. United States v. Taylor, 159 F.4th 1136, 1140 (8th Cir. 2025) (emphasizing that “[v]oter fraud is no myth” and describing a scheme involving dozens of fraudulent forms, including many with fake signatures); Voting for Am., Inc. v. Steen, 732 F.3d 382, 394 (5th Cir. 2013) (discussing an incident in
-21- which a “Project Vote employee was convicted . . . for submitting more than 400 fake voter[-]registration applications” (first alteration in original) (citation omitted)); see also Jay M. Zitter, Validity of Statute Restricting Voter Registration Solicitations by Third Parties or Organizations, 55 A.L.R.6th 599 § 4 (2010) (discussing a case that recognized how third-party voter registration has “at times become a vehicle for voter fraud and voter[-]registration fraud”).
Even if it is not foolproof, a wet signature provides an added layer of security. This advantage may explain why at least four other states insist on one for voter- registration forms, see Fla. Stat. § 97.053(5)(a)(8) (calling for a handwritten signature unless registering through the Department of Highway Safety and Motor Vehicles); Ga. Code § 21-2-381(a)(1)(C)(i) (requiring one for an “appl[ication] for an absentee ballot by mail”); Okla. Stat. tit. 26, § 4-112(B); Tex. Elec. Code § 13.143(d-2), and three others for petitions or referenda, see Me. Const. art. IV, pt. 3, § 20 (requiring all “written petition[s]” to include the “original signatures of the petitioners”); Ohio Rev. Code § 3501.011(A) (stating that “whenever a person is required to sign or affix a signature” to “any . . . kind of petition,” the signature must be a “written, cursive- style legal mark”); Or. Rev. Stat. § 250.105(1)(e) (“A filed initiative or referendum petition must contain only original signatures.”). Pundits and policymakers can debate whether it is the best approach, but it is remarkable that the court says a practice developed over centuries for just this purpose is immaterial to “determining whether” the person turning in a registration form is really who they say they are and “qualified . . . to vote.” 52 U.S.C. § 10101(a)(2)(B).
The court never meaningfully engages with these justifications. Like the district court, it appears to rely on a lack of “evidence” supporting the Board’s “claim[s]” about its wet-signature rule. Ante, at 10. Yet at the preliminary-injunction stage, the burden “rests entirely with the movant,” Goff v. Harper, 60 F.3d 518, 520 (8th Cir. 1995) (emphasis added), to show that the presence of a wet signature is “not
-22- material,” 52 U.S.C. § 10101(a)(2)(B) (emphasis added). It is not the Board’s job to prove it is. See Mazurek v. Armstrong, 520 U.S. 968, 972 (1997) (explaining that a court may not grant a preliminary injunction “unless the movant, by a clear showing, carries the burden of persuasion” (citation omitted)).
Aside from the court’s burden-shifting error, the evidence it wants is itself immaterial. It claims that a wet-signature requirement must not matter because in “practice” no county clerk uses it to “detect fraud.” Ante, at 10. The problem, of course, is that materiality is about whether something could influence a decision, not whether it does. See Neder, 527 U.S. at 16 (defining “material” as “capable of influencing” a person (emphasis added) (citation omitted)); Kungys v. United States, 485 U.S. 759, 771 (1988) (focusing on “whether the [information is] predictably capable of affecting . . . the official decision” (emphasis added)); see also Basic Inc. v. Levinson, 485 U.S. 224, 234–35 (1988) (explaining that in the securities-fraud context, the focus of the “materiality requirement” is on what a “reasonable investor” would think (emphasis added)).
In the modern era, IRS officials and bank tellers do not pull out a magnifying glass to examine the signatures on every form or check they receive. See, e.g., Cooper v. Union Bank, 507 P.2d 609, 620 (Cal. 1973) (explaining that banks “may handle several thousand instruments bearing third[-]party indorsements in a single day” and it is “commercially unreasonable” to expect them to “undertake foolproof efforts” to verify them). Nor, it is safe to assume, did anyone scrutinize the wax seals on every document. Rather, they offer a way to evaluate claims of fraud when there are other reasons to suspect it. See Aerotek, Inc., 624 S.W.3d at 205 (discussing how to verify a handwritten signature); cf. People v. Marion, 29 Mich. 31, 38 (1874) (describing how the comparison of seals on two documents showed that one was a forgery); Collins v. Carnegie (1834) 110 Eng. Rep. 1373, 1374, 1376 (KB) (accepting a comparison as proof of the authenticity of a seal on a document because “[i]f not, it is difficult to say what proof would be sufficient”). Not to mention the deterrence value they provide, which discourages fraud in the first place. See Callanen, 89 F.4th
-23- at 489 (pointing out that wet signatures “ha[ve] some prospect of getting the attention of many applicants and dissuading false statements”); see also Stephen Mason, The Signature in Law: From the Thirteenth Century to the Facsimile 9 (2022) (explaining that a signature “warn[s] the person affixing [it] that they should take care before committing themselves to the contents of the document”).
The fact that Arkansas allows the use of any mark, even an “X,” does not make the wet-signature rule any less material. Recall that one step in the evolution from seals to signatures was treating certain written marks, called scrolls, as substitutes. See Game Place, L.L.C., 813 S.E.2d at 317. The trend started in Arkansas by the early-to-mid 1800s, see Floyd v. Ricks, 14 Ark. 286, 295 (1853) (explaining that a document with a “scroll” was “adjudged to be sealed”), and culminated in the passage of a provision in the Arkansas Constitution that placed “sealed and unsealed instruments” on equal footing, Ark. Const. of 1874 Schedule § 1 (“Until otherwise provided by law no distinction shall exist between sealed and unsealed instruments, concerning contracts between individuals, executed since the adoption of the Constitution of 1868.”). It should be no surprise that voter-registration requirements follow in the same historical tradition. See Ark. Const. amend. 51, § 6(a)(3)(F) (requiring a “signature or mark”); see also Henry v. Union Sawmill Co., 287 S.W. 203, 204 (Ark. 1926) (accepting a “mark” as “just as effective as if signed by . . . [a] written signature[]”).
Besides, if the court is concerned that accepting a mark sets the bar too low, perhaps any problem lies there, not with requiring the use of a pen to make it. See Watson v. Billings, 38 Ark. 278, 283 (1881) (describing a statute that allowed a mark in place of a signature, but only if the signer was illiterate and someone witnessed it); Ex parte Miller, 3 S.W. 883, 883 (Ark. 1887) (same). If a mark, wet or otherwise, does not provide a “guarantee” that the person who made it is telling the truth, then
-24- by the court’s logic the whole requirement is immaterial, despite its long historical pedigree.4 Ante, at 9. The argument simply proves too much.
It also asks too much. Part of the reason why a wet-signature rule is relevant is that it eliminates one form of fraud: digitally lifted signatures. See Thompson, 976 F.3d at 618 (noting its ability to “prevent[] fraud by ensuring the authenticity of signatures”); see also Kendall, 650 F.3d at 526 (explaining that signature requirements “help[] to make sure that false signatures are not put on the [document] and that unregistered or ineligible voters do not sign it”). The answer to other forms of fraud is to enact measures to deal with them, not try to defeat them all at once. See John Doe No. 1 v. Reed, 561 U.S. 186, 197 (2010) (explaining that states have “considerable leeway to protect the integrity and reliability” of “election processes,” especially when it comes to “efforts to root out fraud” (citation omitted)).
Materiality, after all, is not a high bar. Clearing it requires neither perfection nor a one-size-fits-all solution. See Browning, 522 F.3d at 1175 (“[Section 10101(a)(2)(B)] does not establish a least-restrictive-alternative test for voter[-]registration applications . . . .”); cf. Crawford v. Marion Cnty. Election Bd., 553 U.S. 181, 196 (2008) (“While the most effective method of preventing election fraud may well be debatable, the propriety of doing so is perfectly clear.”). An obviously material requirement like Arkansas’s wet-signature rule does not become immaterial simply because other alternatives exist. See Brnovich v. Democratic Nat’l Comm., 594 U.S. 647, 686 (2021) (recognizing that a state “may take action to prevent election fraud without waiting for it to occur and be detected within its own borders”). As centuries of practice show, fighting fraud is more elusive than it appears. See Burson v. Freeman, 504 U.S. 191, 208 (1992) (noting that “election fraud” is “difficult to detect”).
4 The reason for allowing marks was that not everyone could write their name. See Ex parte Miller, 3 S.W. at 883. It is ironic that the court would twist an accommodation that lets more voters register into a reason to treat the requirement as “[im]material.” 52 U.S.C. § 10101(a)(2)(B). -25- III.
In my view, there has been no showing, let alone a “clear” one, Winter, 555 U.S. at 22, that a wet signature is immaterial, particularly when it comes to “determining whether [an] individual is qualified” to vote, 52 U.S.C. § 10101(a)(2)(B). The materiality provision protects voters from discrimination by guarding against irrelevant barriers to voting. It is not an open invitation for courts to supervise elections. For these reasons, I dissent. ______________________________
-26-