FOR PUBLICATION
UNITED STATES COURT OF APPEALS FOR THE NINTH CIRCUIT
KARI LAKE; MARK FINCHEM, No. 23-16022
Plaintiffs, D.C. No. 2:22-cv- and 00677-JJT
ANDREW D. PARKER; PARKER DANIELS KIBORT, LLC; KURT B. ORDER OLSEN; OLSEN LAW, PC, Counsel for Plaintiffs,
Appellants, v.
BILL GATES, as a member of the Maricopa County Board of Supervisors; CLINT HICKMAN, as a member of the Maricopa County Board of Supervisors; JACK SELLERS, as a member of the Maricopa County Board of Supervisors; THOMAS GALVIN, as a member of the Maricopa County Board of Supervisors; STEVE GALLARDO, as a member of the Maricopa County Board of Supervisors,
Defendants-Appellees, 2 LAKE V. GATES
and
ADRIAN FONTES, Arizona Secretary of State; MARICOPA COUNTY BOARD OF SUPERVISORS; REX SCOTT, as a member of the Pima County Board of Supervisors; MATT HEINZ, as a member of the Pima County Board of Supervisors; SHARON BRONSON, as a member of the Pima County Board of Supervisors; STEVE CHRISTY, as a member of the Pima County Board of Supervisors; ADELITA GRIJALVA, as a member of the Pima County Board of Supervisors; PIMA COUNTY BOARD OF SUPERVISORS,
Defendants.
Filed August 21, 2025
Before: Kim McLane Wardlaw, Ronald M. Gould, and Patrick J. Bumatay, Circuit Judges.
Order; Dissent by Judge VanDyke LAKE V. GATES 3
SUMMARY *
Sanctions
The panel denied a petition for panel rehearing and denied a petition for rehearing en banc in a case concerning Arizona’s voting system in which the panel affirmed the district court’s sanctions order under Rule 11 of the Federal Rules of Civil Procedure and 28 U.S.C. § 1927 against plaintiffs’ lead attorneys. Dissenting from the denial of rehearing en banc, Judge VanDyke, joined by Judges Callahan, R. Nelson, Collins, Lee and Bumatay, wrote that two reasons independently made this case worthy of en banc review. First, the district court and the panel badly misapplied the standards for finding the attorneys’ conduct sanctionable by reading the complaint out of context and in the light least favorable to plaintiffs. Second, this court’s refusal to grant en banc review will be construed as implicitly blessing the district court’s weaponization of sanctions to chill politically disfavored litigation.
ORDER
Judge Wardlaw and Judge Gould voted to deny the petition for panel rehearing and the petition for rehearing en banc. Judge Bumatay voted to grant both the petition for panel rehearing and the petition for rehearing en banc. A
* This summary constitutes no part of the opinion of the court. It has been prepared by court staff for the convenience of the reader. 4 LAKE V. GATES
judge of the court requested a vote on whether to rehear the matter en banc. The matter failed to receive a majority of votes of the active judges in favor of en banc rehearing. Fed. R. App. P. 40. Judge Desai was recused from the vote. The petition for panel rehearing and rehearing en banc, Dkt. 38, is DENIED.
VANDYKE, Circuit Judge, joined by CALLAHAN, R. NELSON, COLLINS, LEE, and BUMATAY, Circuit Judges, dissenting from the denial of rehearing en banc:
The panel decision in this case upheld a sanctions order under Rule 11 of the Federal Rules of Civil Procedure and 28 U.S.C. § 1927 against attorneys Andrew Parker and Kurt Olsen. Parker and Olsen (collectively, “Lead Attorneys”) represented plaintiffs Kari Lake and Mark Finchem in election-related litigation. As the district court candidly acknowledged, the sanctions were intended to “send a message” to similar litigants in election-based lawsuits and to discourage litigation disfavored by the court. Zealous to safeguard the “public trust,” the district court read plaintiffs’ complaint out of context and in the light least favorable to plaintiffs; imposed a heightened requirement that Lead Attorneys conduct “significant” pre-filing inquiries on the basis of their clients and their cause; levied sanctions on the ground that plaintiffs made claims that, as even the district court itself recognized, the complaint never actually stated; and badly misapplied the governing legal standards. Lake v. Hobbs, 643 F. Supp. 3d 989, 998, 1013 (D. Ariz. 2022), aff’d in part, rev’d in part sub nom., Lake v. Gates, 130 F.4th 1054 (9th Cir. 2025), and aff’d sub nom., Lake v. Gates, 130 F.4th 1064 (9th Cir. 2025). LAKE V. GATES 5
This case involved legal claims that might charitably be characterized as aggressive. It was a Hail Mary legal theory, especially as to standing. But we encounter Hail Mary legal theories regularly in our court in a variety of contexts, and while they almost always lose, they don’t get sanctioned just because they are longshots. Cf. Nuclear Regul. Comm’n v. Texas, 145 S. Ct. 1762, 1776 (2025) (characterizing a particular legal claim as “essentially a Hail Mary pass—and in court as in football, the attempt rarely succeeds” (quoting Nyunt v. Chairman, Broad. Bd. of Governors, 589 F.3d 445, 449 (D.C. Cir. 2009))). Many cases are dismissed because the asserted injuries are too speculative to support Article III standing. A great many more are dismissed for failure to state a claim on which relief can be granted. The law has no lack of tools short of sanctions to deal with speculative claims, adventurous legal theories, and imprecisely drafted complaints. Again, our circuit entertains cases with exceedingly improbable claims on a routine basis, which are usually (but not inevitably, which is probably why hope springs eternal) dispatched using any of the panoply of available mechanisms. If the run-of-the-mill Hail-Mary claims we routinely encounter are not sanctionable, neither were the claims in this case. 1
1 Other courts across the country agree that even longshot and improbable claims are not subject to sanctions, including in the election law context. See, e.g., Moss v. Bush, 105 Ohio St. 3d 458, 458–60 (2005) (declining sanctions despite allegations that were deemed “highly improbable and potentially defamatory, inflammatory, and devoid of logic,” including claims of “alleged fraud in the casting and counting of absentee ballots and alleged individual election incidents occurring throughout the state”); Donald J. Trump for President, Inc. v. Boockvar, 502 F. Supp. 3d 899, 906–10 (M.D. Pa. 2020) (declining to impose sanctions for a complaint characterized by the court as “Frankenstein’s Monster” due to it being “haphazardly stitched together from two distinct 6 LAKE V. GATES
Two reasons independently made this case worthy of en banc review. First, the district court and the panel badly misapplied the standards for finding attorney conduct sanctionable. The district court flatly misread the allegations in plaintiffs’ complaint. While the complaint never actually said that Arizona did not use paper ballots—a fact that the district court even acknowledged in its sanctions order—the district court nevertheless found such a claim implied in the complaint (and thus sanctionable). But the context of the complaint confirms what its plain language makes clear: The attorneys never argued that Arizona did not use paper ballots. Although the complaint may not have been drafted with perfect precision, the district court reached the alternative conclusion only by repeatedly going out of its way to construe the complaint in the light least favorable to plaintiffs. Read in context, the complaint cannot be plausibly construed as asserting what it never said. Penumbras, emanations, and acontextual implications should be insufficient to warrant sanctions under Rule 11, and the district court abused its discretion in concluding otherwise. The panel majority ratified those errors, and in doing so reinforced the district court’s departure from the Rule 11 standard and our case law interpreting that Rule. Second, the district court boldly proclaimed that it levied sanctions on Lead Attorneys with the hope that doing so would “send a message” to deter future litigants with similar
theories in an attempt to avoid controlling precedent,” and said to be seeking a “drastic remedy in the contest of an election” based on “strained legal arguments without merit and speculative accusations”); Hunter v. Earthgrains Co. Bakery, 281 F.3d 144, 153 (4th Cir. 2002) (reversing a sanctions order as unwarranted despite the district court’s characterization of the complaint as “utter nonsense” against an attorney with a prior sanctions history). LAKE V. GATES 7
claims—or, put bluntly, to deter a specific type of election litigation. Setting aside the myriad legal problems posed by this action—not the least of which is making a hash of the Rule 11 standard—that just looks bad. And even if the inference is unwarranted, this court’s refusal to grant en banc review will be construed by many as implicitly blessing the district court’s weaponization of sanctions to chill politically disfavored litigation. Who could blame them? Cudgeling attorneys into abandoning unpopular claims and clients is not what sanctions are for. While not authoritative here, see Snead v. Metro. Prop. & Cas. Ins. Co., 237 F.3d 1080, 1091 (9th Cir. 2001), the Arizona Supreme Court astutely observed that “[b]y sanctioning parties and their lawyers for bringing debatable, long-shot complaints, courts risk chilling legal advocacy and citizens raising ‘questions’ under the guise of defending the rule of law,” Ariz. Republican Party v. Richer, 547 P.3d 356, 370 (Ariz. 2024). “Even if done inadvertently and with the best of intentions, such sanctions present a real and present danger to the rule of law.” Id. And that “danger to the rule of law” is all the more present when the judge issuing the sanctions boldly proclaimed that such a chilling effect is an intended feature, not a bug. Id. We should have taken this case en banc to rectify these abuses and make clear that Article III judges are to adjudicate cases without fear or favor, remaining scrupulously neutral toward all litigants—especially in politically charged cases where the public is watching. I respectfully dissent from our failure to do so. I. The underlying dispute in this case concerns the use of electronic voting systems in Arizona elections and the 8 LAKE V. GATES
potential for manipulation of those systems. Lake v. Gates, 130 F.4th at 1067. Plaintiffs Kari Lake and Mark Finchem, candidates in Arizona’s 2022 general election, filed a complaint in federal district court alleging that Arizona’s voting infrastructure insufficiently protected the rights of voters—specifically, that Arizona’s electronic tabulation systems were susceptible to hacking. Id. They also sought a preliminary injunction prohibiting the use of electronic voting systems in Arizona elections. Id. Their claims failed. The district court dismissed the complaint for lack of Article III standing in August 2022, and this court affirmed in a unanimous opinion. Lake v. Fontes, 83 F.4th 1199, 1204 (9th Cir. 2023) (per curiam). But the district court did not stop there. On December 1, 2022, it granted defendants’ motion for sanctions under Federal Rule of Civil Procedure 11(c) and 28 U.S.C. § 1927 against Lead Attorneys, ordered Lead Attorneys to pay defendants’ attorneys’ fees, and ordered the parties to file memoranda regarding the proper amount of attorneys’ fees. Lake, 643 F. Supp. 3d at 1012–13. After oral argument, the district court issued an order holding Lead Attorneys (and their respective law firms) jointly and severally liable for $122,200 in attorneys’ fees. On March 14, 2025, a divided panel of this court affirmed the district court’s sanctions order. Lake, 130 F.4th at 1067. Plaintiffs sought en banc review, which a majority of our court has now declined to grant. In refusing to correct the panel’s opinion, our court has left in place a decision that openly weaponizes sanctions to chill disfavored litigants and litigation. LAKE V. GATES 9
II. An award of Rule 11 sanctions is reviewed for abuse of discretion. Montrose Chem. Corp. of Cal. v. Amer. Motorists Ins. Co., 117 F.3d 1128, 1133 (9th Cir. 1997). A district court “abuse[s] its discretion if it base[s] its ruling on an erroneous view of the law or on a clearly erroneous assessment of the evidence.” Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 405 (1990). To impose sanctions under Rule 11, a district court must determine that a pleading is “both baseless and made without a reasonable and competent inquiry.” In re Keegan Mgmt. Co., Sec. Litig., 78 F.3d 431, 434 (9th Cir. 1996) (quoting Townsend v. Holman Consulting Corp., 929 F.2d 1358, 1362 (9th Cir. 1990)). Sanctions under 28 U.S.C. § 1927 “must be supported by a finding of subjective bad faith.” Blixseth v. Yellowstone Mountain Club, LLC, 796 F.3d 1004, 1007 (9th Cir. 2015) (citation omitted). “[B]ad faith is present when an attorney knowingly or recklessly raises a frivolous argument or argues a meritorious claim for the purpose of harassing an opponent.” Id. (citation omitted). Whether an attorney acted recklessly or in bad faith is a factual finding that is reviewed for clear error. See Pac. Harbor Cap., Inc. v. Carnival Air Lines, Inc., 210 F.3d 1112, 1117 (9th Cir. 2000). III. Plaintiffs’ ultimate inability to establish standing necessarily doomed their election claims from the start. “But asserting an unpersuasive claim is different from asserting a sanctionable one.” Lake, 130 F.4th at 1071 (Bumatay, J., dissenting). It is true the complaint in this case (like perhaps the complaints in many of our cases) may not have been a paragon of clarity and incisive analysis. And it is also true that Lead Attorneys may have “played hardball” with the 10 LAKE V. GATES
state (again, an apt metaphor for many of the advocates who practice before us). Id. Yet “nothing they did was deceptive, intentionally false, or beyond the bounds of zealous advocacy.” Id. Read with a modicum of context and an ounce of charity, the complaint challenges the reliability of Arizona’s use of electronic systems to count ballots. The district court’s sanctions order was premised on its conclusion that the complaint said that Arizona did not use paper ballots. But the complaint never said that. And as Judge Bumatay’s dissent explains, “no party to the litigation was fooled. Arizona’s attorneys fully understood the nature of the claims.” Id. Undaunted, the district court pressed on, ostensibly “concerned that the public might misconstrue [plaintiffs’] claims.” Id. But whether some unnamed member of the public might misunderstand technicalities and legalese in a complaint is obviously not the standard for imposing sanctions under Rule 11. Yet perhaps the most disquieting aspect of this case remains the district court’s express declaration that it was sanctioning Lead Attorneys to “send a message” to other litigants who may raise similar election-law disputes. Lake, 643 F. Supp. 3d at 1013. Sanctions are not a tool for punishing disfavored litigants bringing disfavored claims or their attorneys. To the contrary, this court has already warned of the dangers posed by the abuse of Rule 11 sanctions—including, presciently, warning that such sanctions might be used to “chill vigorous advocacy.” Townsend v. Holman Consulting Corp., 929 F.2d 1358, 1364 (9th Cir. 1990) (en banc). And now this prediction is manifest. The panel opinion’s message is “loud and clear: challenge an election, and judges stand with sanctions at the LAKE V. GATES 11
ready if they disapprove of your claim.” Lake, 130 F.4th at 1071 (Bumatay, J., dissenting). A. Rule 11 sets a “low bar” for attorneys to clear and should only be utilized in unusual situations. See Strom v. United States, 641 F.3d 1051, 1059 (9th Cir. 2011). As we’ve emphasized before, “Rule 11 is an extraordinary remedy, one to be exercised with extreme caution.” Operating Eng’rs Pension Tr. v. A-C Co., 859 F.2d 1336, 1345 (9th Cir. 1988). In this instance the district court based its erroneous imposition of sanctions on four categories of conduct by Lead Attorneys: (1) allegations in the complaint about the use of paper ballots; (2) allegations in the complaint about the testing of Arizona’s election equipment; (3) reliance on speculation and conjecture; and (4) failure to conduct a reasonable pre-filing inquiry. None of this conduct was sanctionable, and the district court abused its discretion in concluding otherwise. 1. The district court’s sanctions order was premised on the idea that the complaint falsely stated that Arizona voters do not cast paper ballots. But the complaint never said that. The thrust of plaintiffs’ challenge was aimed at the alleged infirmities in the security of Arizona’s system for counting votes and the desire “to have their ballots, and all ballots cast together with theirs, counted accurately and transparently.” In the process of making their argument, plaintiffs alleged that electronic vote-counting machines presented greater risks from hacking and, accordingly, that using electronic vote-counting systems made the electoral process 12 LAKE V. GATES
vulnerable to manipulation. Plaintiffs then argued that Arizona’s current electronic vote-counting systems should be replaced by one in which the votes are counted by humans. To be sure, plaintiffs certainly argued that hand ballots must also be part of an accurate and transparent voting system, but their emphasis on vote tabulation (not vote casting) belies any conclusion that their argument implies that Arizona does not use paper ballots. The district court reached the alternative conclusion only by twisting and contorting plaintiffs’ arguments and drawing all inferences against Lead Attorneys. For example, the district court claimed that Lead Attorneys’ request for injunctive relief was “entirely frivolous because [Arizona is] already doing what [they] want the [State] to do.” Lake, 643 F. Supp. 3d at 998. But as just explained, the complaint argued for replacing the electronic-tabulating system with one in which votes are tabulated by humans. Everyone agrees that Arizona has not adopted the system plaintiffs sought and continues to rely on electronic tabulation. So the district court’s characterization of plaintiffs’ request for injunctive relief as requesting things that Arizona is “already doing” is transparently wrong. Given the district court’s imposition of sanctions on Lead Attorneys for supposedly arguing that Arizona did not use paper ballots, it might come as a surprise that the district court could not identify a single instance in the complaint where plaintiffs expressly make this allegation. Lake, 643 F. Supp. 3d at 998 (“Plaintiffs argue that ‘none of these paragraphs say that Arizona does not use paper ballots.’ … That is true only in the most facile sense.”). This alone is sufficient to demonstrate that the district court abused its discretion. Despite explicitly recognizing that plaintiffs were correct to argue that the complaint did not say LAKE V. GATES 13
that “Arizona does not use paper ballots,” the district court eschewed reliance on clear language from the complaint and instead cherry-picked isolated snippets and contorted logic to conclude that “Plaintiffs requested that the Court order Arizona to do something that they contend it is not currently doing: to use paper ballots.” Id. Consider some examples of the district court’s creative reading of the complaint. First, it faulted paragraph 153 of the complaint for stating that “Plaintiffs seek for the Court to Order, an election conducted by paper ballot, as an alternative to the current framework” and construed that statement as affirmatively asserting that Arizona elections do not use paper ballots. Id. That is wrong as a matter of both context and logic. The district court apparently disregarded the header that immediately preceded the statement: “Voting on Paper Ballots and Counting Those Votes by Hand Is the Most Effective and Presently the Only Secure Election Method.” Immediately after that statement, the complaint included nine bullet points detailing how votes cast on paper ballots should be counted by hand. Read in context, the complaint was again advocating both paper ballots and human tabulation. At bottom, plaintiffs wanted the entire election to be conducted by hand-counted paper ballots. It is undisputed that Arizona does not conduct elections entirely with paper ballots, nor does Arizona hand-count all the ballots that are done with paper, so plaintiffs were clearly not requesting procedures that the state was already following. The district court committed a classic part-to-whole fallacy by taking plaintiffs’ (true) belief that Arizona employed a voting system relying on some non-paper ballots to mean that plaintiffs were asserting Arizona employed a system with no paper ballots. Finally, the district court also improperly 14 LAKE V. GATES
assumed that the “alternative” system referred to in paragraph 153 must mean a different system in all respects from the current system. But an “alternative” framework need not be different in every respect. The fact that the proposed “alternative” framework includes votes cast on paper ballots—just like part of the current practice—does not even suggest, let alone assert, that paper ballots are not already used for some voters. Next, the district court took aim at paragraph 7 of the complaint. That paragraph reads, in full:
Through this Action, Plaintiffs seek an Order that Defendants collect and count votes through a constitutionally acceptable process, which relies on tried and true precepts that mandate[] integrity and transparency. This includes votes cast by hand on verifiable paper ballots that maintains voter anonymity; votes counted by human beings, not by machines; and votes counted with transparency, and in a fashion observable to the public.
From this paragraph the district court plucked a single phrase—“[t]his includes votes cast by hand on verifiable paper ballots that maintains voter anonymity”—to suggest that Lead Attorneys claimed Arizona voters do not currently use paper ballots. Once again, that is wrong as a matter of logic. Read in context, the paragraph is advocating two things in conjunction: (1) voting by paper ballot and (2) vote counting by humans. It is “undisputed that Arizona did not hand count votes.” Lake, 130 F.4th at 1073 (Bumatay, J., dissenting). So plaintiffs were proposing a different voting LAKE V. GATES 15
system—one with both paper ballots and human tabulation—which did not then exist. Id. This does not mean paper ballots were not already partially in use. Lastly, the district court found fault with paragraphs 58 through 60 of the complaint. To start, the complaint alleged that “[b]illions of federal dollars were spent to move states, including Arizona, from paper-based voting systems to electronic, computer-based systems.” The district court took issue with these paragraphs because they described Arizona’s shift from an “auditable paper-based system” to a “computer-based system.” Lake, 643 F. Supp. 3d at 998. In the district court’s reading, these allegations were not merely imprecise, they were actually false because they “more than impl[y] a transition away from paper ballots.” Id. at 999. Once again, the district court disregarded the relevant context and went out of its way to construe the allegations in the complaint in the light least favorable to Lead Attorneys. Paragraph 57 of the complaint itself states that “Arizona intends to rely on electronic voting systems to record some votes and to tabulate all votes cast in the State of Arizona in the 2022 Midterm Election.” So once again, the complaint never alleged that no paper ballots will be used in Arizona— it instead alleged that some votes will be recorded electronically, and all votes will be tabulated electronically. All of this was inarguably true. In fact, the complaint had already explained why at least some voters must vote by electronic means: “[v]oters who may have hearing or visual impairments may cast their votes with the aid of electronic ballot marking devices.” Thus, the complaint unequivocally recognized that Arizona uses electronic voting systems to record only “some” votes (i.e., those cast by voters with disabilities). The clear implication then is that the rest of the votes are cast in another way (i.e., on paper ballots). 16 LAKE V. GATES
Regardless, although the complaint plainly contemplates that some ballots are cast on paper, the gravamen of the complaint was not about how votes are cast at all. Rather, the thrust of the complaint concerned allegations regarding how votes are counted. It alleged—correctly and uncontested—that “all” votes in Arizona are tabulated by “electronic voting systems.” Lake, 130 F.4th at 1073–74 (Bumatay, J., dissenting). Indeed, the complaint explicitly stated that “[e]very county in Arizona intends to tabulate votes cast … through optical scanners.” And to state the obvious, “optical scanners” scan paper ballots. The complaint also recognized this fact of how optical scanners work in arguing against their use, directly stating that “[e]very county in Arizona … can simply and securely count votes cast on paper ballots without using centralized machine-counting or computerized optical scanners.” Once again, the complaint’s focus was on opposing the use of optical scanners—not disputing the existence of paper ballots. It is thus irrelevant whether the complaint mentioned “paper ballots.” Sanctions should be reserved for false statements; not for situations where judges simply wish the parties had said more. The upshot is that optical scanners necessarily require the use of paper ballots. The complaint recognized as much. So “any confusion on this point was entirely of the district court’s own making.” Lake, 130 F.4th at 1074 (Bumatay, J., dissenting). And, consequently, the district court abused its discretion in ripping the allegations out of context, reading the arguments in the light least favorable to Lead Attorneys, and contorting the complaint to find that the use of the terms “computerized voting” and “electronic voting systems” must be construed to mean a complete lack of paper ballots. See Lake, 643 F. Supp. 3d at 999, 1001. The district court’s LAKE V. GATES 17
handling of the paper-ballot allegations was grievously wrong, and the panel opinion’s ratification of those errors and deviation from the Rule 11 standard on this basis alone warranted this court’s en banc review. I cannot emphasize this enough: The district court was transparently wrong in its characterization of the Lead Attorneys’ allegations regarding how Arizona used and counted paper ballots. The district court sanctioned Lead Attorneys based on its own blatant misreading of their complaint. This is egregious. After all, if sanctioning Lead Attorneys on the grounds that their complaint alleged something that—by the district court’s own admission— cannot actually be found in the text of the complaint does not qualify as abusing one’s discretion, what does? For better or worse, we live in a time when many citizens believe (rightly or wrongly) that there is rampant election fraud and abuse. We cannot afford the perception that our federal courts are anything but scrupulously impartial in those partisan (and often heated) disputes. Our refusal to correct this error will generate precisely the opposite perception. 2. The district court also found the complaint’s allegations that Arizona’s tabulation machines are not subjected to “objective evaluation” or “neutral, expert analysis” as sanctionable. But in doing so the court plainly misapplied the Rule 11 standard. Id. at 1002 (quoting paragraphs 20 and 57 of the complaint). Under Rule 11, a complaint does not have to be entirely uncontradicted. Rather, it merely requires that the allegations “have” or “will likely have evidentiary support.” Fed. R. Civ. P. 11(b)(3). Eschewing the Rule 11 standard, the district court found the complaint’s allegations that Arizona’s tabulation 18 LAKE V. GATES
machines are not subject to “objective evaluation” or “neutral, expert analysis” were false on the ground that the Arizona Secretary of State—a defendant in the case—had tested the equipment and because a company accredited by a federal election commission had also conducted testing. Lake, 643 F. Supp. 3d at 1002–03. To state the obvious, the fact that a defendant in a case disputes the allegations in a complaint does not demonstrate that those allegations are false. If it did, every complaint would be found riddled with false allegations and practically every plaintiff’s attorney would be subject to sanctions. And the fact that the Secretary of State—again, a defendant in this case—had tested the equipment does not facially disprove the complaint’s allegations, since “the whole point of the complaint was to request ‘objective’ and ‘neutral’ testing— not simply relying on the defendant’s assurances.” Lake, 130 F.4th at 1074 (Bumatay, J., dissenting). The Rule 11 standard does not require one party to take their adversary’s word at face value. Id. The complaint did not claim no testing was done; it challenged whether the testing was sufficiently “objective” and “neutral.” In support, plaintiffs adduced allegations that the company that manufactures the optical scanners has steadfastly “refused to disclose its software and other parts of its electronic voting system in order to subject it to neutral expert evaluation.” Of course, whether testing is appropriately “neutral” and “objective” is necessarily a question of judgment and prudence, “not easily reduced to binary determinations of truth or falsity.” Id. Perhaps unsurprisingly then, even the district court could not bring itself to find these allegations false; rather, it merely weighed the evidence and found the evidence that Arizona had its equipment tested by an accredited laboratory more LAKE V. GATES 19
compelling. But a complaint does not leave an attorney subject to Rule 11 sanctions just because it is contradicted. Rule 11 likewise does not permit a district court to weigh evidence, find one party’s evidence more compelling, and conclude that the less compelling argument is therefore sanctionable. Rule 11 merely requires that the attorneys have conducted “an inquiry reasonable under the circumstances” to permit them to certify that, “to the best of [their] knowledge, information, and belief,” the allegations “have” or “will likely have evidentiary support after a reasonable opportunity for further investigation or discovery.” Fed. R. Civ. P. 11(b). The complaint here satisfied that modest requirement. 3. Next, the district court found Lead Attorneys’ conduct sanctionable because it held that plaintiffs “lacked an adequate factual or legal basis to support the wide-ranging constitutional claims they raised or the extraordinary relief they requested.” Lake, 643 F. Supp. 3d at 1008. This would be a good reason for the district court to find that plaintiffs lack standing to pursue their lawsuit. And indeed, a panel of this court unanimously did so, agreeing that plaintiffs’ alleged injuries were too speculative to meet the strictures of Article III standing. See Lake, 83 F.4th at 1201. But a legal theory that is too speculative to support standing does not connote a legal theory that is sanctionable. It is as unsurprising as it is well-established that “the pleader need not be correct in his view of the law.” Zaldivar v. City of Los Angeles, 780 F.2d 823, 830 (9th Cir. 1986), abrogated on other grounds by Cooter & Gell, 496 U.S. 384. As Judge Bumatay’s dissent from the panel decision persuasively explained, “Rule 11 sanctions don’t apply when the 20 LAKE V. GATES
‘pleader’ has ‘a good faith argument for his or her view of what the law is, or should be.’” Lake, 130 F.4th at 1075 (Bumatay, J., dissenting) (quoting Zaldivar, 780 F.2d at 831). Dismissing a complaint for lack of Article III standing “is not dispositive of the issue of sanctions.” Zaldivar, 780 F.2d at 830. On the contrary, a filing only warrants sanctions when it is “both baseless and made without a reasonable and competent inquiry.” Strom, 641 F.3d at 1059 (cleaned up). “[T]o constitute a frivolous legal position for purposes of Rule 11 sanction, it must be clear under existing precedents that there is no chance of success and no reasonable argument to extend, modify or reverse the law as it stands.” Id. (cleaned up). The upshot is that when “a suit rais[es] a novel issue of law as to which there is no caselaw to the contrary,” Rule 11 sanctions are inappropriate. Id. 4. Finally, the district court found Lead Attorneys’ conduct sanctionable on the grounds that they neglected to conduct a reasonable pre-filing inquiry. But the proper analysis of this final ground follows from the foregoing analysis. Because the complaint did not present clearly false statements or blatantly frivolous arguments, the district court also abused its discretion in making this determination. Another abuse of discretion, however, is not the most worrisome of the district court’s errors on this score. Instead, the more significant of the district court’s errors was its choice to impose a heightened pre-filing inquiry requirement on Lead Attorneys based on the nature of the complaint and the clients that they represented. Indeed, the district court was quite explicit that because Lead Attorneys’ clients were candidates for office and the relief they LAKE V. GATES 21
requested was related to state elections, they were “required” to conduct a “significant pre-filing inquiry.” Lake, 643 F. Supp. 3d at 1009 (emphasis added). The upshot is that the district court deliberately departed from the proper Rule 11 factors to impose additional requirements upon Lead Attorneys unknown to them when they filed their complaint, all in the name of the district court’s alleged “concern” for the “dangers posed by making wide-ranging allegations of vote manipulation in the current volatile political atmosphere.” Id. This kind of selective targeting of certain claims is exactly what our en banc court already warned about in Townsend:
Were vigorous advocacy to be chilled by the excessive use of sanctions, wrongs would go uncompensated. Attorneys, because of fear of sanctions, might turn down cases on behalf of individuals seeking to have the courts recognize new rights. They might also refuse to represent persons whose rights have been violated but whose claims are not likely to produce large damage awards. This is because attorneys would have to figure into their costs of doing business the risk of unjustified awards of sanctions.
929 F.2d at 1363–64. As the foregoing reasons demonstrate, the district court erred repeatedly and egregiously in its interpretation and application of Rule 11, and it abused its discretion by reading the complaint out of context and in the light least favorable to plaintiffs, and by imposing additional pleading 22 LAKE V. GATES
requirements on plaintiffs because of who they are and what they believe. B. The district court’s additional sanctioning of Lead Attorneys under 28 U.S.C. § 1927 for filing a motion for a preliminary injunction was clear error. An attorney may be sanctioned under § 1927 for “multipl[ying] the proceedings in any case unreasonably and vexatiously.” 28 U.S.C. § 1927. Sanctions under § 1927, like sanctions under Rule 11, are “extraordinary” and must be “exercised with extreme caution.” Keegan Mgmt. Co., 78 F.3d at 437 (cleaned up). Section 1927 also requires a finding of “subjective bad faith” such that the attorney must have “knowingly or recklessly raise[d] a frivolous argument.” Id. at 436 (cleaned up). On what grounds did the district court find that Lead Attorneys had acted recklessly in filing their motion for a preliminary injunction? Timing. The district court found that Lead Attorneys acted “recklessly” because they (1) waited seven weeks after filing the complaint to seek a preliminary injunction, and (2) they filed the motion fewer than four months before an election. Those timing factors arguably bear on whether relief was appropriate in the lead-up to an election. See Purcell v. Gonzalez, 549 U.S. 1 (2006). But they certainly do not support a finding of recklessness. Purcell notwithstanding, parties have frequently sought and won relief in election law cases in the months before an election, both before this court and the Supreme Court. See generally, e.g., Republican Nat’l Comm. v. Mi Familia Vota, 145 S. Ct. 108 (2024) (granting in part motion to stay district court’s injunction); see also Mi Familia Vota v. Petersen, 111 F.4th 976 (9th Cir. LAKE V. GATES 23
2024) (vacating motion panel’s stay of district court’s injunction). Finally, it is also important to note that “the district court made no finding that the attorneys here subjectively filed the motion for a preliminary injunction seven weeks after the complaint recklessly or with vexatious intent.” Lake, 130 F.4th at 1076 (Bumatay, J., dissenting). Perhaps intending to patch this hole, the district court offered some impromptu theorizing (in a footnote) that Lead Attorneys’ failure to seek “emergency relief” in this court after the 2022 election “raises questions about the good faith basis for their request for immediate relief.” Lake, 643 F. Supp. 3d at 1011 n.11. But as Judge Bumatay pointed out in dissent, that reasoning completely glosses over the fact that an emergency appeal to our court is reviewed deferentially—not de novo. Lake, 130 F.4th at 1076 (Bumatay, J., dissenting) (citing Int’l Franchise Ass’n, Inc. v. City of Seattle, 803 F.3d 389, 399 (9th Cir. 2015) (“The court does not review the underlying merits of the case, but rather whether the district court relied on an erroneous legal premise or abused its discretion in denying [a party’s] motion for preliminary injunctive relief.”)). Accordingly, there is little that can be gleaned from plaintiffs’ failure to seek an emergency appeal. C. For the foregoing reasons, the district court erred clearly and egregiously by imposing sanctions on Lead Attorneys. But that is not what makes this case truly remarkable. What makes this case most remarkable is that the district court acknowledged that it was imposing sanctions to “send a message” to attorneys who might file a particular type of lawsuit that the court viewed with disfavor. Specifically, the court stated that it wanted to “send a message to those who 24 LAKE V. GATES
might file similarly baseless suits in the future”—suits that, in the court’s view, “further[] false narratives that baselessly undermine public trust at a time of increasing disinformation about, and distrust in, the democratic process.” Lake, 643 F. Supp. 3d at 1013. The district court proclaimed that it would “not condone litigants ignoring the steps that Arizona has already taken toward” the goal of “ensur[ing] that our elections are secure and reliable.” Id. There is thus no meaningful debate that the district court imposed sanctions, based on a clearly misconstrued complaint, after noting its desire to chill litigation that the district court simply disfavored. The district court’s desire to impose sanctions to chill litigation that it disfavors flagrantly violates both the text and purpose of Rule 11. Not one word in the text of Rule 11 empowers judges to “make an example of litigants to reassure the public.” Lake, 130 F.4th at 1076–77 (Bumatay, J., dissenting). Nor, as Judge Bumatay explained in dissent, “does Rule 11 permit monetary sanctions to serve as a message to the public at large.” Id. at 1077. And the Supreme Court has accordingly explained that any sanctions fees awarded must have a “causal link” to a litigant’s misbehavior. Goodyear Tire & Rubber Co. v. Haeger, 581 U.S. 101, 108 (2017) (“[A] court, when using its inherent sanctioning authority (and civil procedures), [needs] to establish a causal link—between the litigant’s misbehavior and legal fees paid by the opposing party.”). Sanctioning attorneys to highlight a district judge’s disagreement with discrete positions on a politically charged issue cannot be squared with the “extreme caution” required under Rule 11. See Operating Eng’rs Pension Tr., 859 F.2d at 1345; Lake, 130 F.4th at 1077 (Bumatay, J., dissenting). Yet that is exactly what the district court did here. LAKE V. GATES 25
IV. Plaintiffs in this case sought to advance aggressive, long- shot legal claims. That is far from unusual. But the district court’s response was far from usual: it imposed sanctions on Lead Attorneys based on an acontextual reading of the complaint; it imposed a heightened requirement that Lead Attorneys conduct “significant” pre-filing inquiries on the basis of their clients and their cause; it concluded that plaintiffs made claims that, as even the district court itself recognized, the complaint never expressly stated; and it significantly departed from the governing legal standards under Rule 11 and 28 U.S.C. § 1927. Worse still, while doing so the district court openly acknowledged its desire that its sanctions order would “send a message” to chill litigants from bringing disfavored political claims. Lake, 643 F. Supp. 3d at 1013. On this score, the Arizona Supreme Court—no stranger to aggressive election litigation—issued a prescient warning:
By sanctioning parties and their lawyers for bringing debatable, long-shot complaints, courts risk chilling legal advocacy and citizens raising “questions” under the guise of defending the rule of law. Even if done inadvertently and with the best of intentions, such sanctions present a real and present danger to the rule of law.
Richer, 547 P.3d at 370. We should have heeded its wisdom. Unfortunately, the panel majority ratified the district court’s many abuses of discretion, and in doing so departed from the Rule 11 standard while implicitly blessing the district court’s weaponization of sanctions against unpopular 26 LAKE V. GATES
claims and disfavored clients. Because I believe that Rule 11 demands more, and the integrity of our judicial system as an impartial arbiter deserves more, I respectfully dissent from the denial of rehearing en banc.