Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 1 of 30 1 WO 2 3 4 5 6 IN THE UNITED STATES DISTRICT COURT 7 FOR THE DISTRICT OF ARIZONA 8 9 Arizonans for Fair Elections, et al., Plaintiffs, 10 11 v. 12 Katie Hobbs, et al., 13 No. CV-20-00658-PHX-DWL ORDER Defendants. 14 INTRODUCTION 15 16 In Arizona, the people’s right to enact laws via the initiative process is sacrosanct. 17 This right has been enshrined in Arizona’s constitution since Arizona’s inception, and the 18 debate over whether to adopt it was the “burning” and “most important” question raised 19 during Arizona’s constitutional convention. Whitman v. Moore, 125 P.2d 445, 450 (Ariz. 20 1942). 21 The relevant provisions appear in Article IV of the Arizona constitution. Among 22 other things, Article IV provides that “the people reserve the power to propose laws and 23 amendments to the constitution and to enact or reject such laws and amendments at the 24 polls, independently of the legislature” and that “[u]nder this power ten per centum of the 25 qualified electors shall have the right to propose any measure.” See Ariz. Const., Art. IV, 26 Part 1, § 1(1), (2). Additionally, and most relevant here, Article IV contains detailed 27 requirements concerning the “[f]orm and contents of initiative and referendum petitions” 28 and “verification.” Id. §1(9). Those requirements include the following: Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 2 of 30 1 7 Every initiative or referendum petition . . . shall contain the declaration of each petitioner, for himself, that he is a qualified elector . . . , his post office address, the street and number, if any, of his residence, and the date on which he signed such petition. Each sheet containing petitioners’ signatures shall be attached to a full and correct copy of the title and text of the measure so proposed to be initiated or referred to the people, and every sheet of every such petition containing signatures shall be verified by the affidavit of the person who circulated said sheet or petition, setting forth that each of the names on said sheet was signed in the presence of the affiant and that in the belief of the affiant each signer was a qualified elector . . . . 8 Id. (emphasis added). In other words, the rule in Arizona for over 100 years has been that 9 an initiative proponent must (1) submit a “sheet” containing the signatures of the qualified 10 electors who have agreed to support the initiative, and (2) submit an affidavit from the 11 signature gatherer (also known as the circulator) certifying that he or she was physically 12 present when each qualified elector’s signature was obtained. 2 3 4 5 6 13 Although these requirements arise from the constitution, the Arizona Legislature 14 has enacted statutes that reaffirm and effectuate them. Under A.R.S. § 19-112(A), “[e]very 15 qualified elector signing a petition shall do so in the presence of the person who is 16 circulating the petition and who is to execute the affidavit of verification.” Under A.R.S. 17 § 19-112(C), “[t]he person before whom the signatures, names and addresses were written 18 on the signature sheet”—that is, the circulator—“shall subscribe and swear before a notary 19 public that each of the names on the sheet was signed and the name and address were 20 printed by the elector and the circulator on the date indicated . . . .” Finally, under A.R.S. 21 § 19-121.01, the Arizona Secretary of State must disregard any signature sheets that aren’t 22 accompanied by the required circulator affidavit and disregard any entries on particular 23 sheets in which “the signature of the qualified elector is missing.” 24 121.01(A)(1)(d), (A)(3)(a). Id. § 19- 25 Enter the coronavirus. In this lawsuit, the plaintiffs—a pair of ballot measure 26 committees that hope to place initiatives on the ballot for the November 2020 election, plus 27 an individual Arizona voter who wishes to sign the committees’ initiative petitions 28 (collectively, “Plaintiffs”)—argue that the COVID-19 pandemic has effectively eliminated -2- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 3 of 30 1 their ability to comply with Arizona’s rules requiring in-person signature gathering for 2 initiative petitions. Plaintiffs further note that Arizona has already created a system (known 3 as “E-Qual”) for obtaining electronic signatures from qualified electors and authorizes the 4 use of that system in one specific context—by candidates for statewide and legislative 5 offices who are gathering signatures for candidate nominating petitions. Thus, Plaintiffs 6 seek a declaration that “the provisions governing the signature-gathering requirements for 7 initiative measures under Title 19, Chapter 1 of the Arizona Revised Statutes, violate the 8 First and Fourteenth Amendments to the United States Constitution during the state of 9 emergency caused by the COVID-19 pandemic” as well as an injunction requiring the 10 Secretary of State “to allow the electronic submission of signatures through E-Qual . . . 11 during the state of emergency in Arizona caused by the COVID-19 pandemic” and 12 precluding Arizona’s various county recorders “from striking signatures based solely on 13 their submission in electronic form.” (Doc. 1 at 20-21.) 14 Now pending before the Court is Plaintiffs’ motion for temporary restraining order 15 (“TRO”), which essentially seeks the same injunctive relief sought in the complaint. 16 (Doc. 2.) As explained in more detail below, this motion will be denied and this action 17 will be dismissed due to a lack of jurisdiction. 18 First, Plaintiffs’ complaint and moving papers do not challenge Arizona’s 19 constitutional provisions governing the initiative process. Instead, Plaintiffs only seek to 20 challenge the Arizona statutes requiring in-person signature gathering. This approach 21 raises serious doubts about whether the requested relief would even redress Plaintiffs’ 22 alleged injuries—as noted, the Arizona constitution has always required in-person 23 signature verification. During the TRO hearing, Plaintiffs attempted to belatedly address 24 this issue by arguing that the solicitation of electronic signatures through the E-Qual system 25 could be deemed “substantial compliance” with Article IV of the Arizona constitution. Not 26 only does this argument seem questionable, but Plaintiffs are effectively asking a federal 27 court to make a guess about an unsettled question of state law and then, based on that guess, 28 overturn a century-old state-law election rule. This outcome would be distressing from a -3- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 4 of 30 1 federalism perspective and is precluded by both (1) the rule requiring a party invoking a 2 federal court’s subject matter jurisdiction to establish a likelihood of redressability, not the 3 mere possibility of redressability, and (2) the rule requiring a party seeking a TRO—which 4 is an extraordinary remedy never awarded as a matter of right—to clearly demonstrate that 5 the requested relief is necessary to avoid irreparable injury. 6 Second, Plaintiffs have not, in any event, demonstrated a likelihood of success or 7 even serious questions going to the merits of their First and Fourteenth Amendment-based 8 claims. This is not the first time a litigant has attempted to invoke those provisions to 9 challenge state laws governing the signature gathering process for initiative petitions. 10 Under Ninth Circuit law, such a challenger must show that the law creates a “severe 11 burden” on the ability to successfully place an initiative on the ballot, and burdensomeness 12 is gauged in part by assessing whether a “reasonably diligent” initiative committee could 13 have succeeded despite the law. Here, although it is undeniable that the COVID-19 14 pandemic is currently wreaking havoc on initiative committees’ ability to gather signatures, 15 it is undisputed that some Arizona initiative committees (including one of the committees 16 in this case) had gathered enough signatures to qualify before the pandemic took hold. It 17 is also undisputed that the two committees in this case didn’t start organizing and gathering 18 signatures until the second half of 2019, whereas some of their counterparts began 19 organizing as early as November 2018. Finally, although it is impossible to predict how 20 the pandemic will play out in the coming weeks and months, it is possible that conditions 21 will abate to the point that in-person signature gathering again becomes viable before the 22 July 2020 submission deadline for signatures. 23 demonstrated that Arizona law creates a severe burden that would prevent a reasonably 24 diligent initiative committee from placing its proposed initiative on the ballot. And because 25 Plaintiffs failed to make this showing, the challenged laws are subject to a relaxed form of 26 scrutiny that is easily satisfied by Arizona’s interests in preventing fraud and promoting 27 political speech and civic engagement. 28 On this record, Plaintiffs have not Third, in large part because of the considerations discussed above, the Court does -4- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 5 of 30 1 not believe the issuance of a TRO would properly balance the equities or be in the public 2 interest. Although Plaintiffs are correct that the COVID-19 pandemic constitutes an 3 “extraordinary circumstance[]” that has resulted in “profound” dislocations (Doc. 2 at 10- 4 11), it is also a profound thing for a federal court to rewrite state election laws that have 5 been in place since the 1910s. The difficulty is underscored by the arguments made by 6 some of the defendants in this case who don’t oppose the relief sought by Plaintiffs. Those 7 state and local officials have identified an array of granular policy choices this Court would 8 need to make in order to effectively implement that relief. Such an approach would raise 9 significant separation of powers and federalism concerns and run afoul of the Ninth 10 Circuit’s exhortation that, “[w]hile we are mindful that federal courts have a duty to ensure 11 that national, state and local elections conform to constitutional standards, we undertake 12 that duty with a clear-eyed and pragmatic sense of the special dangers of excessive judicial 13 interference with the electoral process.” Soules v. Kauaians for Nukoli Campaign Comm., 14 849 F.2d 1176, 1182-83 (9th Cir. 1988). 15 16 BACKGROUND A. The Parties 17 The three Plaintiffs in this action are (1) Arizonans for Fair Elections (AZAN) 18 (“AFE”), a non-profit corporation that was formed to promote a ballot initiative known as 19 the Fair Elections Act, (2) Arizonans Fed Up with Failing Healthcare (Healthcare Rising) 20 (“HRAZ”), a non-profit corporation that was formed to promote a ballot initiative known 21 as the Stop Surprise Billing and Protect Patients Act, and (3) Randi Turk, “a qualified 22 elector within the State of Arizona who would like to sign the petitions supported by the 23 Committee Plaintiffs but has not yet done so.” (Doc. 1 ¶¶ 3-5.) 24 The defendants named in the complaint are Katie Hobbs, Arizona’s Secretary of 25 State (“the Secretary), and the county recorders from Arizona’s 15 counties. (Doc. 1 ¶¶ 6- 26 7.) However, after the Secretary made public statements suggesting she would not oppose 27 Plaintiffs’ requests, the state of Arizona (“the State”), represented by the Arizona Attorney 28 General, moved to intervene. (Doc. 46.) That motion was granted over Plaintiffs’ -5- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 6 of 30 1 opposition. (Docs. 59, 61.)1 2 B. Procedural History 3 On April 2, 2020, Plaintiffs filed their complaint and motion for a TRO. (Docs. 1, 4 2.) In support of their motion, Plaintiffs filed declarations from four individuals: (1) 5 Anabel Maldonado, a campaign manager for AFE (Doc. 3); (2) Jessica Grennan, a 6 campaign manager for HRAZ (Doc. 4); (3) Randi Turk, the individual Plaintiff (Doc. 5); 7 and (4) Christopher Gallaway, an employee of a company that “provides campaign-related 8 services to clients seeking to place initiatives on the ballot for voting by the electorate” 9 (Doc. 79). 10 That same day, four different initiative committees filed a corrected petition for 11 special action in the Arizona Supreme Court that raises claims and requests similar to those 12 presented here. See Arizonans for Second Chances Rehabilitation & Safety et al. v. Hobbs, 13 No. CV-20-0098-SA.2 14 On April 10, 2020, the State filed a corrected response to the TRO motion. (Doc. 15 77.) In support of the response, the State provided an affidavit that had been filed by one 16 of the plaintiff-committees in the state-court action. (Doc. 77-1 at 3-8.) 17 Between April 7-13, 2020, some of the officials named as defendants in the 18 complaint also filed responses to Plaintiffs’ motion. In a nutshell, the county recorders 19 from Pinal and Navajo Counties oppose the TRO request (Docs. 65, 72), the county 20 recorder from Pima County “agrees with Plaintiffs that electronic signature gathering for 21 initiative petitions should be temporarily allowed during the pendency of COVID-19 22 restrictions” (Doc. 53), the Secretary likewise “does not oppose the narrow relief sought 23 by Plaintiffs for this election year,” and “[i]ndeed . . . believes that such relief would further 24 25 26 27 28 1 The Speaker of the Arizona House of Representatives and the President of the Arizona Senate also moved to intervene (Doc. 60) but their request was denied (Doc. 75). 2 The State has also intervened in the state-court action, the Supreme Court has issued a schedule that calls for briefing to be completed by April 27, 2020, and the Supreme Court has announced no oral argument will be held. The docket is available at https://apps.supremecourt.az.gov/aacc/appella/ASC/CV/CV200098.pdf. -6- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 7 of 30 1 the public interest by protecting public health while facilitating continuity of democratic 2 processes,” but “requests that the Court place certain limitations on the relief to minimize 3 administrative burden under the current circumstances” (Doc. 78), and the county recorders 4 from Maricopa, Yuma, Mohave, and Santa Cruz Counties “take no position” on the merits 5 of Plaintiffs’ request (Docs. 62, 67, 73, 85). 6 Additionally, between April 10-14, 2020, the Court authorized the filing of amicus 7 briefs by the Arizona Republican Party (Doc. 86), the Arizona Free Enterprise Club (Doc. 8 87), and the Goldwater Institute (Doc. 92). 9 On April 14, 2020, the Court held a telephonic hearing on the motion for a TRO. 10 (Docs. 90, 102.) The bulk of the argument was provided by counsel for Plaintiffs and the 11 State, and counsel for the Secretary and the Maricopa County recorder also provided 12 remarks. (Id.) Media organizations and members of the public were allowed to listen to 13 the hearing telephonically. (Doc. 68.) 14 On April 16, 2020, the State conditionally moved to certify certain questions to the 15 Arizona Supreme Court. (Docs. 99, 100.) Specifically, the State requested that, if the 16 Court determined this case turned on whether signatures gathered through E-Qual would 17 “substantially comply” with the Arizona constitution, the Court certify that issue to the 18 Arizona Supreme Court. (Doc. 100 at 2.) The State further requested that, if the Court 19 found that certain provisions of the Arizona constitution violated the First and Fourteenth 20 Amendments, it certify the question of whether those provisions could be severed from the 21 rest of the constitution. (Id. at 2-3.) Plaintiffs oppose the certification request, arguing that 22 both questions are unnecessary to resolve their motion. (Doc. 101). 23 24 ANALYSIS I. Standing 25 The State asserts that Plaintiffs lack standing to pursue this action. (Doc. 77 at 3- 26 4.) Specifically, the State argues that Plaintiffs have only sought to enjoin the statutory 27 provisions in Title 19 governing signature collection but have not challenged the provisions 28 of Article IV of the Arizona constitution that, by and large, impose the same requirements. -7- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 8 of 30 1 (Id.) In the State’s view, this creates a standing problem—even if Plaintiffs succeed in 2 arguing that Title 19 is unconstitutional, the Arizona constitution would stand and 3 Plaintiffs’ injury would not be redressed. (Id.)3 During oral argument, Plaintiffs responded 4 by acknowledging that they are not challenging Article IV of the Arizona constitution but 5 arguing that the requested relief would still redress their injury because, once Title 19’s 6 requirements are stripped away, the Arizona courts would be free to conclude that 7 gathering electronic signatures via E-Qual during a pandemic qualifies as “substantial 8 compliance” with Article IV’s requirements. 9 Plaintiffs’ arguments are unavailing. “A litigant must demonstrate . . . a substantial 10 likelihood that the judicial relief requested will prevent or redress the claimed injury to 11 satisfy the ‘case or controversy’ requirement.” Duke Power Co. v. Carolina Env’t Study 12 Grp., Inc., 438 U.S. 59, 79 (1978). Thus, “[t]o establish redressability, a plaintiff must 13 show that it is ‘likely, as opposed to merely speculative, that the injury will be redressed 14 by a favorable decision.’” M.S. v. Brown, 902 F.3d 1076, 1083 (9th Cir. 2018) (quoting 15 Lujan v. Defs. Of Wildlife, 504 U.S. 555, 561 (1992)). Here, it is entirely speculative that 16 Arizona courts would conclude that gathering electronic signatures through the E-Qual 17 system constitutes “substantial compliance” with Article IV’s requirements. 18 As an initial matter, it is not entirely clear that Arizona courts would apply a 19 “substantial compliance” standard in this context. It’s true that Arizona courts have 20 frequently stated that “[w]hen considering challenges to the form of initiative petitions, 21 Arizona courts follow a rule of substantial compliance.” Wilhelm v. Brewer, 192 P.3d 404, 22 405 (Ariz. 2008). The origin of this standard dates back to 1914, when the Arizona 23 Supreme Court decided Arizona v. Osborn, 143 P. 117 (Ariz. 1914). The plaintiffs in that 24 case sought to prevent an initiative from being placed on the ballot by invoking a statute 25 26 27 28 Several amici make variants of the same argument. (Doc. 87 at 2 [“The elephant in the room is the Plaintiffs’ failure to discuss the Arizona Constitution, and . . . the E-Qual system . . . cannot be instituted in a manner that is in compliance with the Arizona Constitution’s provisions on the right to initiative itself.”]; Doc. 92 at 5 [“[T]he relief Plaintiffs request would violate the Arizona Constitution, which expressly requires inperson signature gathering for initiatives.”].) 3 -8- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 9 of 30 1 that authorized courts to enjoin the secretary of state from certifying an initiative “[o]n a 2 showing that any petition is not legally sufficient.” Id. at 117. In addressing whether the 3 challenged petition was “legally sufficient,” the court examined the statutory law of states 4 with similar initiative procedures and concluded that the words “legally sufficient” in the 5 statute indicated “the Legislature meant to describe a valid petition, signed by legal voters, 6 and complying substantially, not necessarily technically, with the requirements of the law.” 7 Id. at 118 (quoting Oregon v. Olcott, 125 P. 303, 304 (Ore. 1912)). 8 Osborn has served as the foundation for the Arizona courts’ subsequent application 9 of the “substantial compliance” standard. Kromko v. Superior Court, 811 P.2d 12, 19 10 (Ariz. 1991) (“The term ‘legal sufficiency, as used [in a statute], requires substantial, not 11 necessarily technical, compliance with the requirements of the law.”) (quoting Osborn); 12 Feldmeier v. Watson, 123 P.3d 180, 183 (Ariz. 2005) (citing Kromko); Wilhelm, 192 P.3d 13 at 405 (citing Feldmeier). However, it appears that no case applying that standard 14 expressly rooted it in the Arizona Constitution. Some, like Osborn, instead focused on 15 statutory interpretation. Kromko, 811 P.2d at 19. Thus, whether substantial compliance 16 survives as the applicable standard may be called into question by the Arizona Legislature’s 17 enactment in 2017 of A.R.S. § 19-102.01, which requires strict compliance with statutory 18 and constitutional requirements. It is notable that the Arizona Supreme Court has, thus far, 19 avoided answering that question. Stanwitz v. Reagan, 429 P.3d 1138, 1142 (Ariz. 2018) 20 (“As our decision does not turn on whether the Committee strictly complied with § 19- 21 118(C), we need not determine the constitutionality of the strict compliance requirement 22 of § 19-102.01(A).”). See also Morales v. Archibald, 439 P.3d 1179, 1181 (Ariz. 2019). 23 But even assuming that “substantial compliance” survives as the applicable standard 24 in Arizona, it is entirely speculative that the Arizona courts would deem the gathering of 25 electronic signatures via E-Qual to be substantially compliant with Article IV’s 26 requirements. Again, the text of Article IV requires an initiative proponent to submit an 27 “affidavit of the person who circulated said sheet or petition, setting forth that each of the 28 names on said sheet was signed in the presence of the affiant . . . .” Ariz. Const., Art. IV, -9- Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 10 of 30 1 Part 1, § 1(9). The Arizona courts have repeatedly commented upon the importance and 2 significance of this physical-presence requirement. See, e.g., Stanwitz, 429 P.3d at 1143 3 (“[W]e note that the Arizona Constitution specifically envisions a signature verification 4 requirement . . . and this Court has observed that ‘[t]he circulator is the only person in the 5 process who is required to make a sworn statement and is, therefore, the person under the 6 greatest compulsion to lend credibility to the process.’”) (citation omitted). Thus, even 7 though the “substantial compliance” standard allows Arizona courts to overlook 8 “technical” errors and “errors in petition formalities” if they do not undermine “the 9 purposes of the relevant statutory or constitutional requirements,” Wilhelm, 192 P.3d at 10 405, it is difficult to see how non-compliance with the physical-presence requirement could 11 be disregarded under these standards. Cf. Porter v. McCuen, 839 S.W.2d 521, 523 (Ark. 12 1992) (rejecting request “not to impose too rigid a standard of compliance with the 13 requirement that signatures be obtained ‘in the presence’ of the person circulating the 14 petition” and concluding “that where the signatures are gathered in areas and places while 15 the canvasser is neither physically or proximately present . . . substantial compliance is 16 lacking”). 17 One final point is worth emphasizing. Whether the use of E-Qual could be deemed 18 substantially compliant with Article IV’s requirements is a pure question of state law. It is 19 also a question the Arizona Supreme Court may be asked to decide in the coming weeks in 20 the parallel lawsuit noted above.4 These circumstances amplify the federalism concerns 21 4 22 23 24 25 26 27 28 Although two amici suggest the Court should abstain from hearing this case under the Pullman abstention doctrine (Doc.86 at 7-8; Doc. 92 at 7), Plaintiffs and the State both asserted during oral argument that Pullman abstention is unwarranted. The Court agrees. “Pullman abstention is an extraordinary and narrow exception to the duty of a District Court to adjudicate a controversy that is properly before it,” Porter v. Jones, 319 F.3d 483, 492 (9th Cir. 2003) (citation omitted), and the Ninth Circuit has emphasized that Pullman abstention is particularly inappropriate in cases—like this case—involving First Amendment challenges. Chula Vista Citizens for Jobs & Fair Competition v. Norris, 782 F.3d 520, 528 (9th Cir. 2015) (“It is rarely appropriate for a federal court to abstain under Pullman in a First Amendment case.”) (citation omitted). Additionally, although it might theoretically be possible to certify the “substantial compliance” question to the Arizona Supreme Court, see generally Doyle v. City of Medford, 565 F.3d 536, 543 (9th Cir. 2009) (“[W]e could simply abstain from deciding this case under the Pullman doctrine . . . [but] certification is appropriate in Pullman-type abstention cases . . . .”), there is no need to pursue certification here because the unsettled nature of the question alone is enough to prevent Plaintiffs from meeting their burden of establishing a likelihood of redressability - 10 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 11 of 30 1 that would flow from ruling in Plaintiffs’ favor on a TRO request. Cf. M.S., 902 F.3d at 2 1090 (acknowledging that “[t]he interaction between the federalism limits on a district 3 court’s remedial power . . . and a district court’s power in general to order prospective relief 4 against state executive officials . . . remains an open and contentious area of the law” but 5 concluding that “where, as here, a plaintiff sues state officials seeking intrusive affirmative 6 relief that is incompatible with democratic principles and where there is no basis for the 7 district court to invoke its equitable power, such relief would also violate principles of 8 federalism”). 9 II. Merits 10 Plaintiffs’ failure to establish redressability means this action must be dismissed due 11 to a lack of subject matter jurisdiction—an outcome that, in turn, means Plaintiffs’ TRO 12 request must be denied as moot. See, e.g., Appalachian Voices v. Bodman, 587 F. Supp. 13 2d 79, 83 (D.D.C. 2008) (“[B]ecause the court determines that the plaintiffs lack standing, 14 the court . . . denies as moot the plaintiffs’ motion for a preliminary injunction.”). 15 Nevertheless, to provide a complete record in the event of appellate review, the Court will 16 proceed to analyze the merits of the TRO request. 17 “A preliminary injunction is an extraordinary and drastic remedy, one that should 18 not be granted unless the movant, by a clear showing, carries the burden of persuasion.” 19 Lopez v. Brewer, 680 F.3d 1068, 1072 (9th Cir. 2012) (quotation omitted). See also Winter 20 v. Natural Res. Def. Council, Inc., 555 U.S. 7, 24 (2008) (“A preliminary injunction is an 21 extraordinary remedy never awarded as of right.”). A plaintiff seeking a preliminary 22 injunction must show that (1) he is likely to succeed on the merits, (2) he is likely to suffer 23 irreparable harm without an injunction, (3) the balance of equities tips in his favor, and (4) 24 an injunction is in the public interest. Winter, 555 U.S. at 20. “But if a plaintiff can only 25 26 27 28 and, relatedly, a likelihood that the requested relief is necessary to avoid irreparable harm. Indeed, the parties have not identified any precedent supporting the issuance of a certified question in the middle of a TRO proceeding and such an approach would seem selfdefeating—the whole point of a TRO is that the plaintiff needs immediate relief. That is why the law places such a heavy burden on the party seeking a TRO and authorizes the issuance of a TRO only in cases presenting a clear entitlement to relief. - 11 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 12 of 30 1 show that there are ‘serious questions going to the merits’—a lesser showing than 2 likelihood of success on the merits—then a preliminary injunction may still issue if the 3 balance of hardships tips sharply in the plaintiff’s favor, and the other two Winter factors 4 are satisfied.” Shell Offshore, Inc. v. Greenpeace, Inc., 709 F.3d 1281, 1291 (9th Cir. 2013) 5 (quotation omitted). Under this serious-questions variant of the Winter test, “[t]he elements 6 . . . must be balanced, so that a stronger showing of one element may offset a weaker 7 showing of another.” Lopez, 680 F.3d at 1072. 8 Regardless of which standard applies, the movant “carries the burden of proof on 9 each element of the test.” Envtl. Council of Sacramento v. Slater, 184 F. Supp. 2d 1016, 10 1027 (E.D. Cal. 2000). Further, there is a heightened burden where a plaintiff seeks a 11 mandatory preliminary injunction, which should not be granted “unless the facts and law 12 clearly favor the plaintiff.” Comm. of Cent. Am. Refugees v. INS, 795 F.2d 1434, 1441 (9th 13 Cir. 1986). A court should not issue such an injunction “unless extreme or very serious 14 damage will result.” Marlyn Nutraceuticals v. Mucos Pharma Gmbh & Co., 571 F.3d 873, 15 878-89 (9th Cir. 2009). They “are not issued in doubtful cases.” Id. 16 A. 17 Likelihood of Success 1. Appropriate Standard 18 It is necessary to begin by identifying the correct test governing Plaintiffs’ claims. 19 The State argues, and Plaintiffs seem to agree, that the Anderson/Burdick framework 20 supplies the relevant test.5 That framework is a flexible approach that balances the severity 21 of the restriction against the government’s purported interest. Dudum v. Arntz, 640 F.3d 22 1098, 1105-1106 (9th Cir. 2011). Severe restrictions trigger strict scrutiny, but less-than- 23 severe restrictions only require the government to demonstrate “important regulatory 24 interests.” Id. 25 The State cites Dudum as establishing that “all constitutional challenges to election 26 regulations are governed by” the Anderson/Burdick framework. (Doc. 74 at 8.) Dudum, 27 28 5 This framework draws its name from Anderson v. Celebreeze, 460 U.S. 780 (1983), and Burdick v. Takushi, 504 U.S. 428 (1992). - 12 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 13 of 30 1 however, focused specifically on “[r]estrictions on voting,” 640 F.3d at 1105,6 and the 2 Ninth Circuit has suggested in subsequent decisions that ballot access regulations (and, in 3 particular, regulations governing the signature gathering process for initiative petitions) 4 raise unique issues that aren’t present in pure voting restriction cases. Angle v. Miller, 673 5 F.3d 1122, 1130 (9th Cir. 2012) (“Although . . . a district-by-district system of counting 6 votes in a statewide election would violate equal protection, . . . district-by-district counting 7 of signatures obtained to qualify an initiative for the ballot [does not] present[] the same 8 problem. Votes and petition signatures are similar in some respects, but ballot access 9 requirements and elections serve different purposes.”) (citations omitted). Thus, before 10 turning to the parties’ specific arguments, it is helpful to begin by summarizing the two 11 most analogous Ninth Circuit decisions involving challenges to state laws governing the 12 signature gathering process for initiative petitions. 13 First, Prete v. Bradbury, 438 F.3d 949 (9th Cir. 2006), involved a challenge to an 14 Oregon law that “prohibit[ed] . . . payment to electoral petition signature gatherers on 15 a . . . per signature basis.” Id. at 951. The plaintiffs argued this law violated the First 16 Amendment but the district court rejected their challenge and the Ninth Circuit affirmed. 17 Id. The court began by noting that, although “the circulation of initiative and referendum 18 petitions involves core political speech,” state regulation in this area is “inevitabl[e]” and 19 “States allowing ballot initiatives have considerable leeway to protect the integrity and 20 reliability of the initiative process.” Id. at 961 (citations and internal quotation marks 21 omitted). Thus, the court held that the plaintiffs’ First Amendment challenge was governed 22 by the same general test later articulated in Dudum: severely burdensome regulations must 23 pass strict scrutiny, but less burdensome regulations trigger less exacting review. Id. As 24 for the first part of this test—whether the law created a “‘Severe’ or ‘Lesser’ Burden”— 25 the court concluded it created a lesser burden because, among other things, (1) the 26 6 27 28 Specifically, Dudum involved a challenge to San Francisco’s practice, following the its adoption of the instant runoff voting method, to restrict the number of rankings on each ballot to three. 640 F.3d at 1100-02. The Dudum plaintiffs argued this limitation was unconstitutional because it had the practical effect of disenfranchising certain voters and/or diluting certain votes. Id. at 1107-14. - 13 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 14 of 30 1 declarations proffered by the plaintiffs to illustrate the supposed difficulty of gathering 2 signatures under the law were based on “unsupported speculation” (id. at 964-65), and (2) 3 at least one referendum petition qualified for the ballot after the enactment of the law and 4 it had a relatively low signature error rate (id. at 966-67). This finding, in turn, meant that 5 the challenged law was only subject to “less exacting review,” and the court concluded that 6 Oregon’s “important regulatory interest in preventing fraud and its appearances in its 7 electoral processes” was sufficient to insulate the law from constitutional challenge. Id. at 8 969-71.7 9 Next, Angle involved a challenge to a Nevada law that required the proponents of a 10 ballot initiative to obtain signatures from at least 10% of the registered voters in each of 11 Nevada’s congressional districts. 673 F.3d at 1126-27. The plaintiffs argued this law 12 violated the First Amendment but the district court rejected this claim and the Ninth Circuit 13 affirmed.8 Id. at 1127. The court analyzed the claim under the same test it had applied in 14 Prete—that is, an initial assessment of whether the regulation resulted in a “severe 15 burden[]” on the plaintiffs’ First Amendment rights, then the application of either strict 16 scrutiny or “less exacting review” depending on the outcome of that assessment. Id. at 17 1132. 18 During the first step of that analysis, the court noted that “restrictions on the 19 initiative process” have the potential to create two different types of First Amendment 20 burdens: first, they can “restrict one-on-one communication between petition circulators 21 and voters,” and second, they “can make it less likely that proponents will be able to garner 22 the signatures necessary to place an initiative on the ballot, thus limiting their ability to 23 make the matter the focus of statewide discussion.” Id. (citations and internal quotation 24 7 25 26 27 28 The Prete court also noted that Oregon law (like Arizona law) requires that “[p]etition circulators must certify that the signatures on the petitions were obtained in the presence of the circulator . . . .” Id. at 969 n.26. 8 The plaintiffs in Angle also raised an unsuccessful equal protection challenge to the Nevada law, id. at 1127-32, but it is unnecessary to summarize the Ninth Circuit’s equal protection analysis because this case does not involve an equal protection challenge. The absence of an equal protection claim also distinguishes this case from Idaho Coal. United for Bears v. Cenarrusa, 342 F.3d 1073 (9th Cir. 2003), which invalidated an Idaho law governing the signature requirements for initiative petitions on equal protection grounds. - 14 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 15 of 30 1 marks omitted). The court concluded the first category was inapplicable because the 2 challenged Nevada law “does not restrict one-on-one communication between petition 3 circulators and votes” and indeed “likely increases the total quantum of speech on public 4 issues, by requiring initiative proponents to carry their messages to voters in different parts 5 of the state.” Id. at 1132-33 (citation and internal quotation marks omitted). As for the 6 second category, the court stated that the test was whether a “reasonably diligent” initiative 7 campaign could have secured a place on the ballot despite the challenged regulation. Id. 8 at 1133 (citing Nader v. Brewer, 531 F.3d 1028 (9th Cir. 2008)). Although the plaintiffs 9 had submitted declarations asserting that the regulation would inhibit their ability to place 10 initiatives on the ballot, the court concluded these affidavits were “too vague, conclusory 11 and speculative” and also noted that the plaintiffs had failed to present evidence that “other 12 initiative proponents have been unable to qualify initiatives for the ballot as a result of the 13 [challenged law].” Id. at 1133-34. Given this backdrop, the court applied less exacting 14 scrutiny and concluded the law passed constitutional muster because it was supported by 15 important regulatory interests—Nevada’s interests in ensuring state-wide support for 16 initiatives and avoiding voter confusion. Id. at 1134-36. 17 In sum, under Prete and Angle, Plaintiffs bear the initial burden of showing that the 18 challenged provisions of Title 19 impose a severe burden on their First Amendment rights. 19 Such a burden may take the form of an impediment on their ability to engage in one-on- 20 one communication or a limitation on their ability to actually earn a place on the ballot. If 21 Plaintiffs demonstrate the existence of severe burden, strict scrutiny applies, but if not, a 22 relaxed degree of scrutiny applies. See also Ariz. Green Party v. Reagan, 838 F.3d 983, 23 985 (9th Cir. 2016) (“Ballot access litigation follows a common pattern. The scrutiny 24 courts employ . . . turns on the severity the law imposes on . . . First and Fourteenth 25 Amendment rights. The plaintiff bears the burden of showing the severity of the burden 26 on those constitutional rights; evidence that the burden is severe, de minimis, or something 27 in between, sets the stage for the analysis by determining how compelling the state’s 28 interest must be to justify the law in question.”); id. at 988 (“This is a sliding scale test, - 15 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 16 of 30 1 where the more severe the burden, the more compelling the state’s interest must be, such 2 that a state may justify election regulations imposing a lesser burden by demonstrating that 3 the state has important regulatory interests.”) (quotation omitted).9 4 2. Whether The Burden Is Severe 5 As noted, initiative-related regulations can create two different types of First 6 Amendment harms: (1) they can inhibit one-on-one communication with voters, and/or (2) 7 they can interfere with the proponents’ ability to secure a place on the ballot. Although 8 Plaintiffs focus primarily on the second type of harm in their complaint and moving 9 papers,10 some portions of the complaint can be interpreted as alleging the first type of 10 harm. (See, e.g., Doc. 1 ¶¶ 85-87 [arguing that “[t]he First Amendment is at the core of 11 petition circulation . . . because it involves interactive communication concerning political 12 change,” that such “advancement of beliefs and ideas[] is an inseparable aspect of the 13 liberty assured by the Due Process Clause of the Fourteenth Amendment,” and that “[t]he 14 Secretary’s strict enforcement of A.R.S. § 19-112 . . . unduly burdens the public’s right to 15 engage in political speech during the COVID-19 pandemic”). 16 To the extent Plaintiffs intended to allege the first type of harm, their claim is 17 unavailing. The in-person signature requirements of Title 19 affirmatively promote speech. 18 See, e.g., Angle, 673 F.3d at 1132-33 (stating that Nevada’s signature requirements “likely 19 increase[d] the total quantum of speech on public issues, by requiring initiative proponents 20 to carry their messages to voters in different parts of the state”). To the extent Plaintiffs 21 aren’t currently able to engage in face-to-face interaction with qualified electors, that’s the 22 fault of the COVID-19 pandemic, not the Title 19 requirements. It’s only when a state law 23 bars certain individuals from serving as petition circulators that the first category of First 24 Amendment harm might arise. Compare Meyer v. Grant, 486 U.S. 414, 422-23 (1986) 25 9 26 27 28 In any event, the level of review is something more than rational basis. Pub. Integrity All., Inc. v. City of Tucson, 836 F.3d 1019, 1025 (9th Cir. 2016) (en banc). 10 See, e.g., Doc. 1 ¶ 78 (“Absent injunctive relief, Plaintiff[s] will suffer irreparable injury . . . [because] it is impossible for Plaintiffs [to] obtain signatures during the pandemic to qualify their measure for the ballot.”); Doc. 2 at 14 (“The injury to the Plaintiffs is the Plaintiffs not being able to obtain the required number of signatures by the July 2, 2020 deadline . . . .”). - 16 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 17 of 30 1 (holding that a Colorado law was unconstitutional in part because “it limits the number of 2 voices who will convey appellees’ message and the hours they can speak”). 3 Plaintiffs’ arguments concerning the second category of harm present a closer call. 4 Plaintiffs contend they were forced to stop collecting signatures in mid-March 2020 due to 5 the outbreak of the pandemic. At the time of cessation, AFE had collected 110,033 6 signatures (Doc. 3 ¶ 29) and HRAZ had collected 273,786 signatures (Doc. 4 ¶ 17). During 7 this election cycle, Arizona law requires at least 237,645 valid signatures—meaning that 8 HRAZ has already collected more than the minimum number—but Plaintiffs have 9 submitted a declaration from Christopher Gallaway, their professional organizer, who 10 avers that, “[i]n my experience, ballot initiative committees should obtain one and a half 11 to two times the number of required signatures to account for signature sheets that may be 12 eliminated and signatures that may be stricken for deficiencies. In this case, it means that 13 [Plaintiffs] should obtain between 356,468 to 475,290 signatures.” (Doc. 79 ¶ 27.) Mr. 14 Gallaway also states in his declaration that, but for the COVID-19 outbreak, his firm “could 15 have gathered an average of 160,000 signatures per initiative campaign” during the period 16 between March 11, 2020 and April 30, 2020, which is period of time during which 17 government entities have ordered or recommended social distancing. (Id. ¶ 47.) Finally, 18 Plaintiffs and Mr. Gallaway all state in their respective declarations that it is impossible as 19 a practical matter to gather any new signatures while the various government orders remain 20 in effect. (Doc. 3 ¶¶ 31-33; Doc. 4 ¶¶ 19-25; Doc. 79 ¶¶ 35-46.) 21 In response, the State offers four reasons why the burdens on Plaintiffs arising from 22 Title 19’s signature-gathering requirements shouldn’t be considered severe. (Doc. 77 at 5- 23 9.) Three of those arguments are unpersuasive. First, the State argues that Plaintiffs 24 haven’t really shown that it’s impractical to keep gathering in-person signatures in the 25 present environment because “[n]one of the Plaintiffs[’] declarants appear to consider 26 genuinely measures such as using single-use signature sheets, social distancing, or 27 scheduling petition signing in advance at prepared and sanitary locations. Perhaps those 28 measures would be sufficient. Perhaps not. But Plaintiffs bear the burden of proof, and - 17 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 18 of 30 1 their declarations are simply too thin a reed to satisfy it.” (Id. at 6.) The Court appreciates 2 the State’s emphasis on the burden of proof, but Plaintiffs’ averments on this point are 3 compelling and consistent with common sense. The fact the TRO hearing was held 4 telephonically, pursuant to a recent General Order postponing most court hearings in the 5 District of Arizona and requiring that all others be held “in the safest manner possible” and 6 “scheduled and conducted by video teleconference or telephone to the extent possible” (D. 7 Ariz. G.O. 20-17), underscores this point. 8 Second, the State contends that, because the challenged Title 19 requirements “are 9 viewpoint-neutral and even handed[,] applying to all initiatives regardless of their subject 10 matter or position,” this viewpoint neutrality “militates against finding a severe burden.” 11 (Doc. 77 at 8.) The problem with this argument is that it conflates First Amendment 12 principles that apply in other contexts with the specific inquiry required under Angle— 13 whether the burden created by the challenged regulation should be considered “severe” 14 because it “make[s] it less likely that proponents will be able to garner the signatures 15 necessary to place an initiative on the ballot.” 673 F.3d at 1132.11 Angle itself involved a 16 viewpoint-neutral law, yet the Ninth Circuit didn’t suggest that a different or less stringent 17 test applied due to its neutrality. 18 Third, during oral argument, the State asserted—in response to a hypothetical 19 question about whether a pandemic that persisted across an entire election cycle could be 20 considered a severe burden—that “it’s important to note that there’s no constitutional right 21 to make laws by petition at all. And, in fact, most states do not do so. So if that were 22 actually the case, Arizona would for that election cycle simply become like most other 23 states in not having an opportunity to law make by initiative.” (Doc. 102 at 42-43.) This 24 argument, like the previous one, is foreclosed by Angle, which repeatedly emphasized that, 25 although a state is not required to allow its citizens to enact legislation through the initiative 26 11 27 28 The case cited by the State in support of its viewpoint-neutrality argument, Chamness v. Bowen, 722 F.3d 1110 (9th Cir. 2013), involved a constitutional challenge to a California law that “eliminat[ed] party primaries and general elections with partynominated candidates, and substitute[ed] a nonpartisan primary and a two-candidate runoff.” Id. at 1112. - 18 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 19 of 30 1 process, a state that chooses to make that process available must not restrict it in an 2 unconstitutional manner. See, e.g., 673 F.3d at 1127-28 (although a “state may decline to 3 grant a right to legislate through ballot initiatives . . . when a state chooses to give its 4 citizens the right to enact laws by initiative, it subjects itself to the requirements of the 5 Equal Protection Clause”) (citations and internal quotation marks omitted); id. at 1133 6 (although “[t]here is no First Amendment right to place an initiative on the ballot . . . we 7 assume that ballot access restrictions . . . trigger strict scrutiny[] when they significantly 8 limit the ability of initiative proponents to place initiatives on the ballot”); id. at 1133 n.5 9 (“The state’s power to ban initiatives thus does not include a lesser power to restrict them 10 in ways that unduly hinder political speech”). 11 The State’s final argument—diligence—has more force. The State notes that 12 Plaintiffs could have begun organizing and gathering signatures in November 2018 (as at 13 least one other initiative committee did) yet didn’t file the necessary registration paperwork 14 with the Secretary until August 20, 2019 (HRAZ) and October 30, 2019 (AFE), thereby 15 wasting between 45% and 55% of the 20-month election cycle. (Doc. 77 at 5-7 & nn.3, 6.) 16 In contrast, the State notes that the government-issued social distancing guidelines arising 17 from the COVID-19 pandemic, which came into effect on March 11, 2020, will cover only 18 7.5% to 12.5% of the election cycle, depending on whether they remain in effect through 19 April 30, 2020 or May 31, 2020. (Id.) And the States notes that a different committee, 20 which is serving as one of the plaintiffs in the state-court lawsuit, had already gathered 21 around 300,000 signatures by the time of the pandemic outbreak. (Id. at 7, citing Doc. 77- 22 1 at 5 ¶ 5.) Given all of this, the State concludes: “It was Plaintiffs’ choice—not the 23 State’s—to procrastinate and dither away time that might later become critical. Plaintiffs’ 24 delay absolutely dwarfs the time period that COVID-19 is likely to affect their signature 25 gathering efforts. . . . [I]t appears likely that Plaintiffs could have qualified for the ballot 26 had they exerted reasonable diligence.” (Id. at 6-7.) 27 The Court agrees with the State that, on this record, Plaintiffs have not established 28 that the Title 19 requirements create a “severe burden” on the ability to place an initiative - 19 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 20 of 30 1 on the ballot. As noted, “the burden on plaintiffs’ rights should be measured by whether, 2 in light of the entire statutory scheme regulating ballot access, ‘reasonably diligent’ 3 initiative proponents can gain a place for their proposed initiative on the ballot.” Angle, 4 673 F.3d at 1133 (quotation omitted). The party challenging the regulation bears the 5 burden of establishing severity. 6 supporting evidence,” is insufficient to demonstrate that the statutory scheme results in a 7 severe burden. Angle, 673 F.3d at 1134; Prete, 438 F.3d at 964 (rejecting “unsupported 8 speculation” as insufficient to demonstrate severe burden). Reagan, 838 F.3d at 989. “Speculation, without 9 Here, a “reasonably diligent” committee could have placed its initiative on the 10 November 2020 ballot despite the Title 19 requirements and the COVID-19 outbreak. It 11 is notable that Plaintiffs’ declarations fail to provide any explanation (let alone 12 justification) for why they waited so long to begin organizing and gathering signatures. 13 The State has presented evidence that at least one Arizona initiative committee began that 14 process in November 2018, yet the two committees in this case waited until the second half 15 of 2019, thereby missing out on essentially a year’s worth of time to work toward the 16 237,645 signature cutoff.12 Moreover, notwithstanding that delay, one of the Plaintiffs was 17 able to gather over 270,000 signatures—much more than the amount required under state 18 law, albeit not enough to provide the buffer recommended by Mr. Gallaway—in the few 19 months it was operating, and a different Arizona committee was able to gather around 20 300,000 signatures during the same abbreviated timeframe. All of this strongly suggests 21 that, had Plaintiffs simply started gathering signatures earlier, they could have gathered 22 more than enough to qualify for the ballot before the COVID-19 pandemic started 23 interfering with their efforts. 24 12 25 26 27 28 The Court recognizes there may be sound reasons for an initiative committee to delay the ramp-up process until the latter part of the election cycle and that it may be the norm for committees to engage in such delay. The difficulty here is that Plaintiffs haven’t proffered any evidence of those reasons and norms in their declarations. (Doc. 102 at 33 [Plaintiffs’ acknowledgement that “we haven’t provided evidence that it’s typical”].) It’s hornbook law that the party seeking a TRO bears the burden of establishing a clear entitlement to relief, and Prete and Angle emphasize that, even outside the TRO context, the party raising a constitutional challenge to a state electoral law must present nonspeculative evidence in support of its claim that the law creates a severe burden. - 20 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 21 of 30 1 This analysis, to be clear, should not be interpreted as a criticism of Plaintiffs. They 2 are hardly the only members of our community who failed to anticipate and plan for a once- 3 in-a-century pandemic. But the relief they are seeking in this case is profound—the 4 displacement of a bedrock component of Arizona law. Such laws should not be wantonly 5 overturned, and that is why courts (including the Ninth Circuit) require parties raising 6 constitutional challenges to state ballot access laws to show not only that they have been 7 thwarted by the law, but that a reasonably diligent party would have been thwarted, too. 8 Thus, in Prete, the Ninth Circuit rejected the challenge to Oregon’s law in part because a 9 different referendum campaign was able to “qualif[y] for the . . . Oregon ballot, after the 10 passage of [the challenged law]”—an outcome that “weighs against plaintiffs’ claim.” 438 11 F.3d at 967. Similarly, in Angle, the Ninth Circuit rejected the challenge to Nevada’s law 12 in part because the plaintiffs “have not presented any evidence” that “other initiative 13 proponents have been unable to qualify initiatives for the ballot as a result of the 14 [challenged law].” 673 F.3d at 1134.13 15 Finally, the Court is unpersuaded by Plaintiffs’ citation of three decisions in which 16 other courts issued emergency injunctive relief in an attempt to address unanticipated 17 electoral dislocations. As an initial matter, none of those decisions were issued by courts 18 in the Ninth Circuit and none applied the Ninth Circuit standards addressed above. 19 Additionally, each is distinguishable for other reasons. 20 First, in Florida Democratic Party v. Scott, 215 F. Supp. 3d 1250 (N.D. Fla. 2016), 21 the plaintiffs sought a TRO requiring an emergency extension of Florida’s voter 22 registration deadline because “[j]ust five days before that deadline . . . Hurricane Matthew 23 bore down and unleashed its wrath on the State of Florida. Life-threatening winds and rain 24 forced many Floridians to evacuate or, at a minimum, hunker down in shelters or their 25 homes.” Id. at 1254. The district court concluded that because “Florida’s statutory 26 For these reasons, Plaintiffs’ contention that they “lost the opportunity to collect up to 160,000 signatures between March 11, 2020, and April 30, 2020” due to the combination of Title 19 and the COVID-19 responses (Doc. 79 ¶ 47) misses the mark. Even assuming that 160,000 figure is accurate, a reasonably diligent campaign wouldn’t have needed to put all of its eggs in the March/April basket. 13 27 28 - 21 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 22 of 30 1 framework would categorically deny the right to vote” to those who failed to register before 2 the hurricane, it amounted to a severe burden that warranted strict scrutiny. Id. at 1257. 3 Second, Plaintiffs cite Democratic Nat’l Comm. v. Bostelmann, 2020 WL 1320819 4 (W.D. Wis. 2020). There, the plaintiffs sought a TRO extending the date by which 5 individuals could register to vote electronically, rather than in-person, in light of the 6 COVID-19 pandemic. The court, relying on Florida Democratic Party, found that the 7 existing voter registration deadline presented an “excruciating dilemma that will soon be 8 faced by eligible voters . . . [to] either venture into public spaces, contrary to public 9 directives and health guidelines or stay at home and lose the opportunity to vote.” Id. at The court concluded that the dilemma presented “an undue burden” and that 10 *5. 11 Wisconsin’s proffered reasons for maintaining the deadline were insufficient to justify it. 12 Id. at *5-*6. 13 The key difference between those cases and this one is that Florida Democratic 14 Party and Bostelmann were both voting restriction cases, while this is a ballot access case. 15 As discussed above, the overarching standard is the same, but the key distinction comes in 16 how courts analyze whether a severe burden is present. When assessing whether a state 17 law presents a severe burden on ballot access, courts in the Ninth Circuit look to whether 18 a reasonably diligent initiative proponent would have been able to get the issue on the ballot 19 despite the law. Regulations on the right to vote, in contrast, do not apply the “reasonably 20 diligent” standard. See, e.g., Pub. Integrity All., Inc. v. City of Tucson, 836 F.3d 1019, 21 1023-1027 (9th Cir. 2016); Dudum, 640 F.3d at 1105-07. See also Angle, 673 F.3d at 1130 22 (“Votes and petition signatures are similar in some respects, but ballot access requirements 23 and elections serve different purposes.”). Notably, neither Florida Democratic Party nor 24 Bostelmann examined whether a reasonably diligent voter would have already registered 25 by the time the respective crises struck.14 26 14 27 28 Additionally, Plaintiffs fail to cite recent decisions by other federal courts in Wisconsin denying similar claims for relief. Taylor v. Milwaukee Election Comm’n, 2020 WL 1695454, *9 (E.D. Wisc. 2020) (denying motion for preliminary injunction to postpone the Wisconsin election in light of the COVID-19 pandemic: “[I]t appears that tomorrow morning, those who have not yet voted will face a grim choice: go out to the polling places (the ones that are open) and risk being exposed to the virus or spreading it - 22 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 23 of 30 1 Plaintiffs’ final case is Faulkner v. Va. Dep’t of Elections, CL-20-1456 (Va. Cir. Ct. 2 2020). There, a Virginia trial court enjoined enforcement of a Virginia statute requiring a 3 political candidate to gather a certain number of signatures in order to appear on the 2020 4 primary ballot. Id. at 3. The court found that, in light of the ongoing COVID-19 pandemic, 5 the statute imposed a “significant” burden and was therefore subject to strict scrutiny. Id. 6 at 2-3. Absent from the court’s order, however, is any discussion of whether a reasonably 7 diligent candidate would have acquired enough signatures in spite of the statute. Id. 8 Regardless of the reason for that omission, this Court is bound by Ninth Circuit law, which 9 requires an examination of what a reasonably diligent proponent would have accomplished 10 11 in the same circumstances. 3. Important Regulatory Interest 12 “Because [Plaintiffs] have not shown that [Title 19] imposes severe burdens, the 13 state need show only that the rule furthers an important regulatory interest.” Angle, 1134- 14 35. Additionally, the State need not demonstrate that the rule is narrowly tailored to 15 promote that interest. Dudum, 640 F.3d at 1114 (“[W]e emphasize that the City is not 16 required to show that its system is narrowly tailored . . . . [W]hen a challenged rule imposes 17 only limited burdens on the right to vote, there is no requirement that the rule is the only 18 or the best way to further the proffered interests.”). 19 The Supreme Court has recognized that Arizona (like other states) “indisputably has 20 a compelling interest in preserving the integrity of its election process. Confidence in the 21 integrity of our electoral processes is essential to the function of our participatory 22 democracy.” Purcell v. Gonzalez, 549 U.S. 1, 4 (2006) (citation omitted). Here, the 23 provisions of Title 19 are clearly aimed at maintaining the integrity of the initiative process. 24 The Arizona Legislature has declared that “strict compliance with the constitutional and 25 26 27 28 to their friends and neighbors, or forego one of the most sacred rights of citizenship—the right to have a say in the governance of their communities, their state and their nation. ‘Extraordinary’ is a feeble description of the circumstances that appear to be leading to that choice. But this court must hold . . . that this federal court does not have the authority ‘to act as the state’s chief health official’ by making the decision that needs to be made to put the health and safety of the community first. . . . The court DENIES the plaintiffs’ motion for preliminary injunction.”). - 23 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 24 of 30 1 statutory requirements for the referendum process and in the application and enforcement 2 of those requirements provides the surest method for safeguarding the integrity and 3 accuracy of the referendum process.” A.R.S. § 19-101.01 (emphasis added). It has made 4 the same finding with respect to the initiative process. Id. § 19-102.01(A). And the Arizona 5 Supreme Court has held that Title 19’s regulation of signature gatherers “represents a 6 reasonable means of fostering transparency . . . and mitigating the threat of fraud or other 7 wrongdoing infecting the petition process.” Stanwitz, 429 P.3d at 1144. 8 Plaintiffs do not appear to seriously contest this point. They acknowledge the 9 State’s interest in “ensur[ing] that the hundreds of thousands of required signatures on the 10 ballot come from qualified electors.” (Doc. 2 at 12.) Further, they don’t argue that Title 11 19’s requirements would fail under a standard of review below strict scrutiny. (Id. [“Under 12 ordinary circumstances, requiring [in-person] gathering of signatures would provide 13 somewhat of a burden, but it could arguably be justified by the State’s interest in preventing 14 fraud . . . .”].) Thus, it would seem that all agree that the State has an important regulatory 15 interest that satisfies the second component of the analysis. See also Prete, 438 F.3d at 16 969 (Oregon’s law restricting use of paid signature gatherers was supported by Oregon’s 17 “important regulatory interest in preventing fraud and its appearances in its electoral 18 processes”).15 19 Finally, the State also argues that Title 19’s requirements further the “significant 20 interest in promoting dialogue by requiring proponents of initiatives to individually engage 21 signers and in doing so provide opportunity for meaningful discussion.” (Doc. 77 at 10.) 22 Although neither party has cited any case addressing whether this interest qualifies an 23 important regulatory interest for purposes of the Prete/Angle framework, common sense 24 25 26 27 28 Given this conclusion, the Court need not resolve amicus’s contention that the EQual system would be worse than the Title 19 requirements at preventing fraud (Doc. 86 at 2 [“E-Qual Is Highly Susceptible To Fraud”]) or Plaintiffs’ rejoinder during oral argument that E-Qual may be better than the Title 19 requirements at preventing fraud. As noted, when an electoral rule doesn’t create a severe burden, the state isn’t required to show that it constitutes the best way to promote the regulatory interest at issue. Dudum, 640 F.3d at 1114 (if relaxed scrutiny applies, “there is no requirement that the rule is the only or the best way to further the proffered interests”). 15 - 24 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 25 of 30 1 suggests it should qualify and Plaintiffs’ counsel did not dispute, during oral argument, that 2 it should qualify.16 3 4. Conclusion 4 For the reasons stated above, Plaintiffs have not demonstrated a likelihood of 5 success on the merits or even a substantial question going to the merits. This failure, 6 standing alone, requires the denial of their request for a TRO. 7 B. Irreparable Injury 8 “A plaintiff seeking a preliminary injunction must demonstrate that irreparable 9 injury is likely in the absence of preliminary relief. Mere possibility of harm is not 10 enough.” Enyart v. Nat. Conference of Bar Exam’rs, 630 F.3d 1153, 1165 (9th Cir. 2011) 11 (citation omitted). 12 The Court has already determined that Plaintiffs lack standing because it is 13 speculative, as opposed to substantially likely, that the relief sought in their complaint 14 would actually redress their injuries in light of their failure to challenge Article IV of the 15 Arizona constitution. This means Plaintiffs also cannot demonstrate they will likely suffer 16 irreparable injury in the absence of a TRO. Cf. Hispanic Affairs Project v. Perez, 141 F. 17 Supp. 3d 60, 67 n.6 (D.D.C. 2015) (“Whether the requested injunctive relief can redress 18 plaintiff Llacua’s injuries . . . dovetails with consideration of the showing of irreparable 19 harm . . . .”). 20 Additionally, putting aside the Article IV issue, it is unclear whether each 21 committee’s inability to gather in-person signatures during the pandemic will cause it to 22 fail to secure a place on the November 2020 ballot. As for AFE, it only gathered 110,333 23 signatures before March 11, 2020—which is only 46% of the required minimum figure of 24 237,645 and only 31% of the “buffer” figure of 356,468 provided by Mr. Gallaway—and 25 the Court is dubious of Plaintiffs’ assertion that AFE would have gathered 160,000 26 16 27 28 Instead, Plaintiffs’ counsel disputed the State’s contention that shifting to electronic signature gathering through the E-Qual system would result in less speech, arguing that EQual would result in meaningful exchanges of speech, too. But again, “there is no requirement that the rule is the only or the best way to further the proffered interests.” Dudum, 640 F.3d at 1114. - 25 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 26 of 30 1 additional signatures during the March/April timeframe. The only evidence Plaintiffs have 2 proffered in support of this anticipated increase in AFE’s collection rate is the assertion in 3 Mr. Gallaway’s declaration that “[b]y the week of February 22, 2020, FieldWorks, on 4 behalf of each Ballot Initiative Committee, was in full swing gathering signatures and 5 ahead of schedule to meet the goals for obtaining a sufficient number of signatures.” (Doc. 6 79 ¶ 32.) This assertion is conclusory and lacks foundation. The Court has no basis for 7 understanding what it means for a committee to be “in full swing” or why gathering only 8 31-46% of the required signatures during the first 16 months of a 20-month collection cycle 9 could nevertheless be considered “ahead of schedule.” On these facts, Plaintiffs have not 10 demonstrated that it is likely, as opposed to merely possible, that but-for the COVID-19 11 pandemic, AFE would have gathered at least 237,645 signatures, and perhaps as many as 12 356,468 signatures, by the July 2, 2020 cutoff and therefore qualified for the November 13 2020 ballot. 14 As for HRAZ, the issue is that it had already gathered 273,786 signatures before the 15 COVID-19 restrictions came into force. This is tens of thousands more than the required 16 minimum figure. Although the Court appreciates Mr. Gallaway’s observation that prudent 17 ballot initiative committees “should” obtain additional signatures, above and beyond the 18 required figure, to account for the possibility that some of the signatures will be deemed 19 invalid and stricken, this is not the same thing as cognizable evidence establishing a 20 likelihood that HRAZ’s 273,786 signatures will prove insufficient. 21 inappropriate to issue a TRO based on speculation concerning its necessity. Winter, 555 22 U.S. at 22 (“Issuing a preliminary injunction based only on a possibility of irreparable harm 23 is inconsistent with our characterization of injunctive relief as an extraordinary remedy that 24 may only be awarded upon a clear showing that the plaintiff is entitled to such relief.”).17 It would be 25 17 26 27 28 The Court appreciates the difficulty of demonstrating irreparable harm in this context—a committee with too few signatures cannot establish it, and neither can a relatively better-off committee that is seeking preliminary injunctive relief in an abundance of caution. But this difficulty is a function of the TRO standards, which rightly recognize that because “[a] preliminary injunction is an extraordinary and drastic remedy,” one “should not be granted unless the movant, by a clear showing, carries the burden of persuasion.” Lopez, 680 F.3d at 1072. - 26 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 27 of 30 1 Finally, as for Ms. Turk, although the Court has already stated that it agrees with 2 Plaintiffs’ contention that the COVID-19 pandemic is currently interfering with the 3 committees’ ability to gather in-person signatures en masse, it doesn’t follow that an 4 injunction is required so Ms. Turk can personally sign the committees’ petitions. There is 5 plenty of time between now and July 2, 2020 for Ms. Turk to make arrangements, while 6 adhering to social distancing requirements, to sign each committee’s petition in the 7 presence of a circulator. 8 C. Balance of Equities 9 “[D]istrict courts must give serious consideration to the balance of equities.” Earth 10 Island Inst. v. Carlton, 626 F.3d 462, 475 (9th Cir. 2010) (citation omitted). In doing so, 11 courts must consider “all of the competing interests at stake.” Id. 12 Plaintiffs’ arguments concerning the balance-of-equities factor are all predicated on 13 the assumption that the challenged provisions of Title 19 create a severe burden on 14 Plaintiffs’ First Amendment rights and will preclude Plaintiffs from qualifying for the 15 November 2020 ballot. (Doc. 2 at 13-15.) But as discussed above, that assumption is 16 unfounded. 17 On the other hand, a state “suffers an irreparable injury whenever an enactment of 18 its people or their representatives is enjoined.” Coal. For Econ. Equity v. Wilson, 122 F.3d 19 718, 719 (9th Cir. 1997). Additionally, it is significant that Plaintiffs are seeking to enjoin 20 the State’s election rules midway through the election cycle. The Supreme Court has 21 repeatedly instructed courts to exhibit caution when faced with such requests. See, e.g., 22 Purcell, 549 U.S. at 4 (“Faced with an application to enjoin operation of voter identification 23 procedures just weeks before an election, the Court of Appeals was required to weigh, in 24 addition to the harms attendant upon issuance or nonissuance of an injunction, 25 considerations specific to election cases . . . .”). 26 At bottom, the balance-of-equities factor weighs against issuing a TRO. 27 D. 28 When considering whether to grant preliminary injunctive relief, courts must Public Interest - 27 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 28 of 30 1 consider the requested injunction’s impact on the public interest. Stormans, Inc. v. Selecky, 2 586 F.3d 1109, 1138-39 (9th Cir. 2009). When a proposed injunction is narrow and limited 3 to the parties, the public interest “will be at most a neutral factor in the analysis.” Id. at 4 1139. “If, however, the impact of an injunction reaches beyond the parties, carrying with 5 it a potential for public consequence, the public interest will be relevant to whether the 6 district court grants the preliminary injunction.” Id. 7 Here, Plaintiffs seek an injunction that would apply to all “initiative and petition 8 proponents and supporters.” (Doc. 2 at 2.) Thus, the injunction they seek extends to the 9 public at large. 10 The public interest weighs against issuing such a TRO. Although the public has a 11 strong interest in enacting laws through the initiative process, and although the Court is 12 loathe to take any action (or inaction) that would expose Arizonans to an increased risk of 13 harm during these challenging times, the signature requirements Plaintiffs seek to displace 14 have been a part of Arizona’s constitutional and electoral landscape for over a century. 15 These requirements reflect a considered judgment, which has stood the test of time, about 16 how best to prevent electoral fraud and promote civic engagement. The public has a strong 17 interest in the continued adherence to such requirements, even during challenging times. 18 The State has also explained that, due to unique features of Arizona law, it is 19 extremely difficult to amend a law that was enacted via the initiative process. (Doc. 77 at 20 10.) This underscores the public’s interest in adhering to a sound initiative process, with 21 time-tested procedures to prevent fraud and promote civic engagement, in every election, 22 not just every election except this one. 23 Finally, although Plaintiffs’ complaint and moving papers suggest it would be easy 24 for the Court to simply decree that the use of the E-Qual system is now permissible for 25 initiative-related signature gathering, other parties paint a different and more complicated 26 picture. For example, the Secretary (who supports Plaintiffs’ request) included in her 27 response to the TRO motion a “request[] that the Court place certain limitations on the 28 relief to minimize the administrative burden under the current circumstances,” including - 28 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 29 of 30 1 “specify[ing] whether the 5% sample of signatures required by A.R.S. § 19-121.01(B) to 2 be randomly drawn from all signatures submitted for a petition should include E-Qual 3 signatures or only those signatures submitted on a hard-copy petition sheet.” (Doc. 78 at 4 2, 8.) During oral argument, the Secretary’s counsel explained that further clarification 5 was needed on this issue because there are two ways to draw the random sample but, if E- 6 Qual signatures are considered presumptively valid, “it could make sense from a policy 7 perspective” to require that samples be drawn only from the hard-copy signature sheets. 8 (Doc. 102 at 61.) In response to this argument, counsel for the Maricopa County recorder 9 (who takes no position on Plaintiffs’ request) argued the Court shouldn’t consider E-Qual 10 signatures to be presumptively valid and instead should require the sample to be drawn 11 from both sets of signatures. (Id. at 64-65.) 12 A consistent theme in this order is that Plaintiffs’ request raises significant 13 federalism and separation-of-powers concerns. This exchange underscores those concerns. 14 The people of Arizona, through their elected representatives (or, perhaps, through the 15 initiative process), should be the ones making policy choices about how to draw signature 16 samples and whether to treat signatures generated through the E-Qual system as 17 presumptively valid. The Court does not, in any way, fault the Secretary or the recorders 18 for seeking clarification on those issues—it is appropriate and prudent to seek clarity so 19 elections can run smoothly—but a federal judge should not be making those choices on the 20 fly as part of a TRO proceeding. The public’s interest in having such policy choices made 21 through appropriate channels is an additional factor weighing against relief. 22 Accordingly, IT IS ORDERED that: 23 (1) 24 Plaintiffs’ complaint (Doc. 1) is dismissed without prejudice due to a lack of subject matter jurisdiction. 25 (2) Plaintiffs’ motion for a TRO (Doc. 2) is denied as moot. 26 (3) The motion for reconsideration filed by the House Speaker and Senate 27 28 President (Doc. 82) is denied as moot. (4) The State’s conditional motion to certify questions to the Arizona Supreme - 29 - Case 2:20-cv-00658-DWL Document 103 Filed 04/17/20 Page 30 of 30 1 Court (Docs. 99, 100) is denied as moot. 2 3 (5) The Clerk of Court shall terminate this action and enter judgment accordingly.18 4 Dated this 17th day of April, 2020. 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 18 Although it might be appropriate, under other circumstances, to afford a plaintiff who has failed to establish standing an opportunity to amend the complaint in an attempt to cure the deficiency, Plaintiffs made clear during oral argument—and reaffirmed in a post-argument filing—that their intention is to challenge the Title 19 requirements without challenging Article IV of the Arizona constitution. (Doc. 101 at 1 [“Plaintiffs have not challenged Article IV, Part 1, Section 1(9). . . . The Court does not need to rule on the application of Arizona’s Constitution to E-Qual to decide this case. Nor have Plaintiffs asked it to do so.”]; Doc. 102 at 22 [Plaintiffs’ confirmation during oral argument that the “omission from [Plaintiffs’] moving papers and complaint [of] any challenge to the constitutional provisions within the Arizona Constitution . . . was intentional”].). Thus, granting leave to amend would be futile. Cf. Carrico v. San Francisco, 656 F.3d 1002, 1008 (9th Cir. 2011) (“Appellants’ principal argument is that the allegations of their amended complaint are sufficient to confer standing . . . . [T]hey do not . . . propose any specific allegations that might rectify their [lack of standing] . . . . Accordingly, we deny leave to amend as futile.”). Additionally, because the dismissal of the complaint (like all dismissals based on a lack of jurisdiction) is without prejudice, Plaintiffs could re-file suit if a future decision by the Arizona Supreme Court in the parallel case changes the statelaw landscape concerning redressability. - 30 -