The late former Speaker of the United States House of Representatives, Thomas P. "Tip" O'Neill, was fond of saying that "all politics is local." TIP O'NEILL, ALL POLITICS IS LOCAL (1994). Through its requirement that witnesses to ballot access "designating petitions" be "resident[s] of the political subdivision in which the office or position is to be voted for," N.Y. ELEC. L. § 6-132(2) (McKinney 1998), the State of New York has attempted to elevate this political adage into an affirmative command. The Constitution, however, protects certain forms of political activity from government interference even if they transcend local political boundaries. While the state does have legitimate interests that justify regulating the electoral process-especially when it seeks to protect the integrity of that process-it cannot mandate that all political activity be "local" without demonstrating the required fit between its chosen means of regulation and those legitimate ends.

We conclude that the section 6-132(2) witness residence requirement severely burdens interactive political speech and association rights protected by the First Amendment (as incorporated by the Fourteenth Amendment) without advancing any legitimate or important state interest. Accordingly, we hold this particular statutory requirement unconstitutional on its face.

John Sollazo, a registered member of the Independence Party of New York, sought to compete in the primary election held September 14, 1999, in order to gain the nomination of the Independence Party for the New York City Council seat representing the 50th Council District in Staten Island. In order for a City Council candidate's name to appear on the primary election ballot, New York law requires that candidate to file a "designating petition" containing valid signatures from at least five percent of the registered party members within the district from which the candidate seeks election.1 See N.Y. ELEC. L. §§ 6-118, 6-136(2) (McKinney 1998). In Sollazo's case, therefore, it was necessary for him to obtain 38 valid petition signatures, representing five percent of the 760 registered Independence Party members in the 50th District. Plaintiff Anita Lerman, a resident of the 49th Council District, served as a witness to the signatures on Sollazo's petition, witnessing a total of 58 signatures including those of Plaintiffs Angelo D'Angelo and Pio D'Agostino,2 both of whom are residents of the 50th District. Lerman, however, lives outside the 50th District. Accordingly, defendant Board of Elections in the City of New York ("NYC Board") ruled, on August 3, 1999, that these 58 signatures were invalid, since New York law requires that designating petition signatures be witnessed by either a notary public, a commissioner of deeds, or an individual who not only is "duly qualified" to vote in New York and an "enrolled voter of the same political party" as the petition signers, but also is a "resident of the political subdivision in which the office or position is to be voted for." N.Y. ELEC. L. § 6-132(2) (3) (McKinney 1998). Lerman is neither a notary public nor a commissioner of deeds.

The plaintiffs filed a complaint pro se against the NYC Board, the New York State Board of Elections ("NYS Board"), and Governor George Pataki in August 1999, seeking declaratory and injunctive relief on the grounds that the witness residence requirement in section 6-132(2) violates the First and Fourteenth Amendments on its face by permitting only district residents to be eligible to witness signatures on the candidate's designating petition. The NYC Board moved for dismissal pursuant to Fed. R. Civ. P. 12(b)(6), and after oral argument on August 31, 1999, the District Court (Frederic Block, Judge) delivered an opinion from the bench dismissing Plaintiffs' complaint in its entirety. Judgment was entered on September 3, 1999. Plaintiffs moved this Court for an expedited appeal on the grounds that the primary election was to be held on September 14, 1999; that motion was denied. The primary election took place as scheduled without Sollazo's name on the ballot.

We review the District Court's dismissal of the plaintiffs' complaint de novo. See Posr v. Court Officer Shield No. 207, 180 F.3d 409 (2d Cir. 1999). Since most pro se plaintiffs lack familiarity with the formalities of pleading requirements, we must construe pro se complaints liberally, applying a more flexible standard to evaluate their sufficiency than we would when reviewing a complaint submitted by counsel.3 See, e.g., Hughes v. Rowe, 449 U.S. 5, 9 10 (1980) (per curiam); Haines v. Kerner, 404 U.S. 519, 520 21 (1972) (per curiam). In order to justify the dismissal of the plaintiffs' pro se complaint, it must be "'beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief.'" Haines, 404 U.S. at 521 (quoting Conley v. Gibson, 355 U.S. 41, 45-46 (1957)).

In another, widely noted New York ballot access case that was litigated in the Eastern District of New York after briefing in this appeal had been completed, but before oral argument, the NYS and NYC Boards entered into a stipulation providing that several provisions of the New York Election Law "impose[d] an undue burden on access to the ballot in connection with the 2000 New York State Republican Presidential Primary." Molinari v. Powers, No. 99 Civ. 8447, Stipulation at 2 (E.D.N.Y. Feb. 4, 2000). The parties in that case also agreed to waive their rights to appeal from any order entered by the district court based on that stipulation, see id., and upon acceptance by the district court, the stipulation served as the predicate for that court's order of preliminary injunctive relief. See Molinari v. Powers, 82 F. Supp. 2d 57, 68 69 (E.D.N.Y. 2000) (Korman, C.J.). The district court in Molinari accepted the parties' stipulation "only after independently concluding that the scheme, both in its totality and by virtue of two of its individual but related elements," including the section 6-132(2) witness residence requirement, "places an undue burden on the right to vote under the First Amendment." Id. at 69.

After counsel appeared for the pro seplaintiffs and oral argument was held in this appeal on March 24, 2000, we ordered the parties to submit supplemental briefs addressing all of the issues presented on appeal, including the preclusive effect, if any, of the stipulation and order entered in Molinari. See Lerman v. Board of Elections, No. 99 9015, Order at 1 2 (2d Cir. Mar. 30, 2000). In their supplemental brief, the plaintiffs argue that the defendants in this case4 are indeed precluded from defending the constitutionality of the section 6-132(2) witness residence requirement by virtue of the Molinari stipulation and order. The plaintiffs suggest that

the defendants' behavior looks disquietingly like an attempt to have the best of both worlds-to abandon their defense of § 6 137 in a politically sensitive, vigorously contested case that was closely followed in the press, but avoid the usual precedential effects of doing so by pursuing their defense of § 6-132 in what was a pro se, low profile, appeal.5Supp. Brief of Plaintiffs-Appellants at 11. To allow the defendants to relitigate the constitutionality of the section 6-132(2) witness residence requirement, the plaintiffs maintain, would "give[] the lawsuit 'the aura of the gaming table.'" Id. (quoting Blonder-Tongue Labs., Inc. v. University of Illinois Found., 402 U.S. 313, 338 (1971)). The defendants respond that Molinari has no preclusive effect in this case, since Molinaridid not decide the precise issue that we confront in this case and the conclusion made by Chief Judge Korman in Molinari concerning the constitutionality of the section 6-132(2) witness residence requirement went beyond what was necessary to grant the Molinari plaintiffs the relief they requested.

We are not untroubled by the possibility that the defendants have attempted to evade their stipulation in Molinari that the section 6-132(2) witness residence requirement imposes an "undue burden" on political speech and their waiver of the right to argue otherwise on appeal in that case. However, the preclusion issue is close-and since the merits of this case are not, we dispose of this case on the merits without reaching any conclusion as to the preclusive effect of Molinari.

The NYC Board argues that the plaintiffs' claims are moot, since the September 1999 primary election is over, having taken place without John Solazzo's name on the ballot. However, this contention is mistaken since the plaintiffs' claims fall within the exception to the mootness doctrine for issues "capable of repetition, yet evading review." Meyer v. Grant, 486 U.S. 414, 417 18 n.2 (1988) (challenge to ballot access provisions for ballot propositions is controversy capable of repetition, yet evading review); see Members for a Better Union v. Bevona, 152 F.3d 58, 61 62 (2d Cir. 1998). Both of the two preconditions for invoking this doctrine have been met-the challenged action was too short to be fully litigated prior to its expiration, and there is a reasonable expectation that the same complaining parties would be subject to that same action in the future. See Grant, 486 U.S. at 417 18 n.2; Krislov v. Rednour, 226 F.3d 851, 858 (7th Cir. 2000). Since the issues presented in this case "will persist in future elections, and within a time frame too short to allow resolution through litigation," Fulani v. League of Women Voters Educ. Fund, 882 F.2d 621, 628 (2d Cir. 1989), the NYC Board's mootness argument necessarily fails.

The NYC Board also contends that as a resident of the 49th Council District, Lerman lacks standing under Article III to bring this action because she is "unaffected by the outcome" of the election in the 50th Council District and therefore suffers no injury from the absence of Solazzo from the primary ballot in that district. The District Court expressly declined to address the question, but indicated offhand that it, too, did not think Lerman had standing under Article III.6 The District Court also concluded that plaintiffs lacked prudential standing to challenge the section 6-132(2) witness residence requirement on its face, rather than only as applied to them. Both of these contentions are mistaken.

First, it is clear that Lerman has standing under Article III to assert her own claims under the First Amendment.7 In order to have standing under Article III, Lerman must demonstrate that (1) she has suffered an "injury in fact" that is "concrete and particularized" as well as "actual or imminent," rather than "conjectural or hypothetical"; (2) the injury is "fairly traceable" to the challenged conduct; and (3) it is likely, rather than "merely speculative," that the injury will be redressed by a favorable decision. See Lujan v. Defenders of Wildlife, 504 U.S. 555, 560 61 (1992) (internal quotation marks omitted). The determination of whether Article III standing exists also must comport with the "manner and degree of evidence required at the successive stages of the litigation." Id. at 561. Since this case remains at the pleading stage, all facts averred by the plaintiffs must be taken as true for purposes of the standing inquiry-as they must be for any other issue presented. "[G]eneral factual allegations of injury resulting from [the challenged statute and conduct] may suffice [to support the existence of standing], for on a motion to dismiss we presume that general allegations embrace those specific facts that are necessary to support the claim."8 Id. (internal quotation marks and citation omitted).

Lerman appears rather easily to have met the three requirements set forth by Defenders of Wildlife. Having associated with Solazzo in order to promote his political candidacy and help him gain access to the primary election ballot, Lerman asserts injury in having been deprived of the opportunity to gather signatures in behalf of his candidacy. Cf. Coalition for Sensible & Humane Solutions v. Wamser, 771 F.2d 395, 399 (8th Cir. 1985) (refusal by election officials to appoint individual members of association as voter registrars caused injury-in-fact sufficient to confer standing, by preventing those individuals from registering new voters). Moreover, the NYC Board has acted directly to strike those designating petitions witnessed by Lerman, and in the context of an action "challenging the legality of government action," we must draw some significance from the fact that Lerman is a direct "object of the action . . . at issue." Defenders of Wildlife, 504 U.S. at 561. Given the nature of the defendants' challenged conduct, there should be "little question that the [defendants'] action . . . has caused [her] injury, and that a judgment preventing . . . the action will redress it." Id. at 561 62 (distinguishing such challenges from challenges in which the "asserted injury arises from the government's allegedly unlawful regulation . . . of someone else," a situation in which "much more is needed" to aver standing).

The NYC Board rests its challenge to Lerman's standing primarily on the argument that because she does not live in the 50th District, she is "unaffected by the outcome of an election for a council seat" within that district and therefore suffered no injury from Sollazzo's exclusion from the primary election ballot. However, the NYC Board's suggestion that any cognizable "injury-in-fact" stops at the electoral district's edge misunderstands the nature of the plaintiffs' challenge to section 6-132(2). The cases cited by the NYC Board in support of its argument make that misunderstanding apparent.9 For example, the NYC Board analogizes this case to United States v. Hays, 515 U.S. 737 (1995), which held that individuals living outside a "racially gerrymandered" electoral district lack standing to challenge its apportionment as violating the Equal Protection Clause of the Fourteenth Amendment. In the context of so-called "racial gerrymandering," however, the injury-in-fact is caused by the "special representational harms" that result from being identified and placed in a particular electoral district based on one's race-the underlying source of the injury is the racial classification, not the apportionment as such. Id. at 745; see also id. at 739 (individuals have standing to challenge "racial gerrymandering" if they are district residents or if they "demonstrate[] that they, personally, have been subjected to a racial classification"). Here, Lerman does not assert a representational injury of the sort considered in Hays. Rather, it is section 6-132(2)'s geographic restriction itself that is alleged to cause the injury-in-fact to Lerman's rights to engage in interactive political speech and expressive political association across electoral district boundaries. Indeed, from Lerman's perspective, it is the very fact that these speech and associational interests have nothing to do with district boundaries that renders the witness residence requirement problematic. The injury-in-fact she alleges concerns the very process of engaging in political activity in support of Sollazo's candidacy, and that injury is sufficient to confer standing under Article III.

Second, in addition to their standing to challenge the section 6-132(2) witness residence requirement as applied to their own speech and associational activity, the plaintiffs also have third party (or jus tertii) standing to challenge the witness residence requirement on its face. The question of standing encompasses both constitutional and prudential considerations. See Secretary of State of Md. v. Joseph H. Munson Co., 467 U.S. 947, 955 (1984) ("In addition to the limitations on standing imposed by Art. III's case-or-controversy requirement, there are prudential considerations that limit the challenges courts are willing to hear."); Valley Forge Christian College v. Americans United for Separation of Church and State, 454 U.S. 464, 471 (1982) ("The term 'standing' subsumes a blend of constitutional requirements and prudential considerations . . . ."); Singleton v. Wulff, 428 U.S. 106, 112 (1976) (constitutional requirements under Article III and prudential requirement that plaintiffs be the "proper proponents of the particular legal rights on which they base their suits" are distinct aspects of standing inquiry). The issue of whether a facial challenge may be entertained is one such prudential consideration, "a species of third party (jus tertii) standing" by which "a party seeks to vindicate not only his own rights, but those of others who may also be adversely impacted by the statute in question." City of Chicago v. Morales, 527 U.S. 41, 55 56 n.22 (1999) (plurality opinion); see also Joseph H. Munson Co., 467 U.S. at 956 58 (permitting facial challenge under First Amendment overbreadth doctrine as exception to prudential limitations against third party standing); Broaderick v. Oklahoma, 413 U.S. 601, 611 613 (1973) (same).

The District Court concluded that the plaintiffs lacked standing to raise a facial challenge to the section 6-132(2) witness residence requirement, asserting that in order for that challenge to go forward, the plaintiffs needed to "establish that no set of circumstances exists under which the act would be valid." Lerman v. Board of Elections, No. CV-99-4901, transcript at 7 (E.D.N.Y. Aug. 31, 1999) (citing City of New York v. United States, 179 F.3d 29, 33 (2d Cir. 1999), cert. denied, 120 S. Ct. 932 (2000)). The standard relied upon by the District Court derives from the Supreme Court's statement in United States v. Salerno, 481 U.S. 739, 745 (1987). Salerno, however, does not apply to this case, in which the plaintiffs assert the violation of rights protected by the First Amendment.10 Rather, since the plaintiffs challenge a statute regulating the ability to engage in interactive political speech and associational activity, their standing to challenge the statute on its face is governed by the overbreadth doctrine. See Joseph H. Munson Co., 467 U.S. at 956 57. Under the overbreadth doctrine, the plaintiffs need only "demonstrate a substantial risk that application of the provision will lead to the suppression of speech." National Endowment for the Arts v. Finley, 524 U.S. 569, 580 (1998); see also Broadrick, 413 U.S. at 612 615 (holding that statutes inhibiting the exercise of First Amendment rights may be invalidated on their face if the impermissible applications of the law are substantial when "judged in relation to the statute's plainly legitimate sweep"). Cf. Salerno, 481 U.S. at 745 (noting applicability of overbreadth doctrine to First Amendment claims). When a litigant challenges a statute on its face as overly broad, the prudential limitations against third party standing are relaxed, and the litigant may assert the rights of individuals whose interests might be affected by the statute but who are not before the court.11 See Joseph H. Munson Co., 467 U.S. at 956 57; Broaderick, 413 U.S. at 612.

The plaintiffs clearly have standing under both Article III and the overbreadth doctrine to challenge the section 6-132(2) witness residence requirement, both as applied to them and on its face. We therefore proceed to address the merits of the plaintiffs' claims.

IV. Constitutionality of the Section 6-132(2) Witness Residence Requirement

The district court in Molinari concluded that the section 6-132(2) witness residence requirement is unconstitutional on its face, violating the First Amendment (as incorporated by the Fourteenth Amendment) by infringing upon the plaintiffs' political speech and associational rights and their right to equal protection. See Molinari v. Powers, 82 F. Supp. 2d 57, 73 77 (E.D.N.Y. 2000). We agree substantially with Chief Judge Korman's decision in Molinari, and hold that the witness residence requirement in section 6-132(2) is unconstitutional on its face. See id.; accord Buckley v. American Constitutional Law Found., 119 S. Ct. 636, 661 & n.1 (1999) (Rehnquist, C.J., dissenting) (stating that the Court's decision "invalidates a number of state laws" and that section 6-132, among other state laws, is now in "serious constitutional jeopardy"); Rockefeller v. Powers, 917 F. Supp. 155, 161 (E.D.N.Y. 1996) (Korman, C.J.) (witness residence requirement in section 6-132(2) "performs [no] essential function"), aff'd, 78 F.3d 44, 46 (2d Cir. 1996).

A. The Burden on First Amendment Rights and Degree of Scrutiny to Be Applied

In determining the level of scrutiny to be applied to the section 6-132(2) witness residence requirement, we first must assess the extent to which that requirement burdens First and Fourteenth Amendment rights. See Buckley v. American Constitutional Law Found., Inc., 119 S. Ct. 636, 642 (1999); id. at 649 (Thomas, J., concurring); Burdick v. Takushi, 504 U.S. 428, 434 (1992). When state election laws subject speech, association, or the right to vote to "'severe' restrictions, the regulation must be 'narrowly drawn to advance a state interest of compelling importance.'" Burdick, 504 U.S. at 434 (citation omitted); see also Prestia v. O'Connor, 178 F.3d 86, 88 (2d Cir. 1999), cert. denied, 120 S. Ct. 539 (1999); Schulz v. Williams, 44 F.3d 48, 56 (2d Cir. 1994). By contrast, when a state election law "imposes only reasonable, nondiscriminatory restrictions" upon First and Fourteenth Amendment rights, then "the State's important regulatory interests are generally sufficient to justify the restrictions." Burdick, 504 U.S. at 434 (internal quotation marks omitted); see also Prestia, 178 F.3d at 88. The burden imposed by the challenged regulation is not evaluated in isolation, but within the context of the state's overall scheme of election regulations. See Prestia, 178 F.3d at 88.

Ordinarily, policing this distinction between legitimate ballot access regulationsand improper restrictions on interactive political speech does not lend itself to a bright line or "litmus-paper test," Anderson v. Celebrezze, 460 U.S. 780, 789 (1983); see also American Constitutional Law Found., 119 S. Ct. at 642, but instead requires a particularized assessment of the nature of the restriction and the degree to which it burdens those who challenge it. See Burdick, 504 U.S. at 434; Schulz, 44 F.3d at 56. However, in those cases in which the regulation clearly and directly restricts "core political speech," as opposed to the "'mechanics of the electoral process,'" it may make "little difference whether we determine burden first," since "restrictions on core political speech so plainly impose a 'severe burden'" that application of strict scrutiny clearly will be necessary. American Constitutional Law Found., 119 S. Ct. at 650 (Thomas, J., concurring) (internal citations omitted); see id. at 642 n.12 (opinion of the court); McIntyre v. Ohio Elections Commission, 514 U.S. 334, 345 (1995) (declining to apply severe/lesser burdens balancing test to direct regulation of "pure speech").

The petition circulation activity at issue in this case, while part of the ballot access process, clearly constituted core political speech subject to exacting scrutiny. See American Constitutional Law Found., 119 S. Ct. at 651 (Thomas, J., concurring) (applying strict scrutiny to voter registration requirement for initiative petition circulators); Meyer v. Grant, 486 U.S. 414, 421 22 (1988) ("[T]he circulation of a petition involves the type of interactive communication concerning political change that is appropriately described as 'core political speech.'"); Krislov v. Rednour, 226 F.3d 851, 866 (7th Cir. 2000) (stating that "circulating nominating petitions [for political candidates] necessarily entails political speech"). As the Supreme Court held inGrant, petition circulation "of necessity involves both the expression of a desire for political change and a discussion of the merits of the proposed change." 486 U.S. at 421.

It should be clear, however, that even if one characterizes this restriction on petition circulation as being more directly concerned with the "mechanics of the electoral process" than with speech, requiring us to evaluate the severity of the burdens on political speech and association posed by that regulation before determining the level of scrutiny to be applied, the witness residence requirement severely burdens political speech by "drastically reduc[ing] the number of persons . . . available to circulate petitions." American Constitutional Law Found., 119 S. Ct. at 643; see id. at 651 (Thomas, J., concurring) (restriction on petition circulation severely burdens speech if it "reduces the voices available to convey political messages"); Grant, 486 U.S. at 422 23 (prohibition on paid petition circulators restricts political speech by "limit[ing] the number of voices who will convey appellees' message and . . . mak[ing] it less likely that appellees will garner the number of signatures necessary" to gain access to ballot, thereby limiting their ability to influence public political discussion); Krislov, 226 F.3d at 860 (statute limiting petition circulators to residents of the political subdivision for which the office is to be voted substantially burdens political speech by "preclud[ing] the candidate from utilizing a large class of potential solicitors to convey his message"); Molinari, 82 F. Supp. 2d at 76 (witness residence requirement in section 6-132(2) substantially burdens political speech because it reduces the number of individuals available to circulate designating petitions and, therefore, "reduces the chances that those supporting the candidates will gather signatures sufficient to qualify for the ballot"). Moreover, petition circulation bears an intimate relationship to the right to political or expressive association. The right to political association also "is at the core of the First Amendment, and even practices that only potentially threaten political association are highly suspect." Krislov, 226 F.3d at 860 (quoting McCloud v. Testa, 97 F.3d 1536, 1552 (6th Cir. 1996)) (internal quotation marks omitted). Obviously, the section 6-132(2) witness residence requirement does not ban association between candidates and non-residents of the electoral district altogether-but the statute need not go that far in order to substantially burden the right to political association. "[B]y preventing the candidates from using signatures gathered by [non-resident] circulators . . . , the law inhibits the expressive utility of associating with these individuals because these potential circulators cannot invite voters to sign the candidates' petitions in an effort to gain ballot access." Krislov, 226 F.3d at 861.

Sollazo's particular case highlights the burdens on political speech and association imposed by the section 6-132(2) witness residence requirement. First, there should be little doubt that the witness residence requirement dramatically reduced the number of potential petition circulators available to advance Sollazo's political message. As counsel for Lerman noted at oral argument, there are approximately 170,000 registered members of the Independence Party statewide, but only 760 in the 50th District. The witness residence requirement therefore renders almost 99.5 percent of all Independence Party members ineligible to circulate petitions within the 50th District. The fact that many of these statewide party members might not have had any interest in circulating petitions on behalf of Sollazo has no bearing upon the burden that the residence requirement imposes upon interactive political speech. Even if, as a practical matter, the requirement only prevented Sollazo from using a handful of those statewide party members, "for some minor candidates"-especially for a minor party candidate, such as Sollazo, running in a political subdivision as small as the 50th District-"parting with one or two avid circulators could significantly impact their campaigns," Krislov, 226 F.3d at 862.

Second, while Solazzo formally needed only 38 signatures to qualify for the ballot, as a practical matter a candidate seeking election "needs a surplus of signatures, because they will likely be challenged on any number of grounds, resulting in some, perhaps many, invalidations."12 Krislov, 226 F.3d at 859 60; Molinari, 82 F. Supp. 2d at 75 (noting that the "facially low percentage of signatures" required in each district does not mitigate the witness residence requirement's burden on speech since the "burdens New York places on the petitioning process make it far more difficult to obtain the requisite signatures"). For minor political candidates such as Sollazo, that burden can be particularly severe. While candidates who are well-financed or favored by their party's leadership might have ready access to the resources necessary to ensure that they qualify to appear on the ballot, many candidates-especially those challenging their party's leadership-do not. Cf. NEW YORK STATE COMM'N ON GOV'T INTEGRITY, ACCESS TO THE BALLOT IN PRIMARY ELECTIONS: THE NEED FOR FUNDAMENTAL REFORM 10 11 (1988) (noting high costs of New York ballot access process and accompanying litigation to candidates, especially those who are not supported by party organizations). Furthermore, since voters may not sign designating petitions for "a greater number of candidates for public office or party position than the number of persons to be elected," N.Y. ELEC. L. § 6-134(3) (McKinney 1998), the task for challengers or other minor candidates becomes that much more formidable. Because of this rule, candidates seeking ballot access necessarily face a "shrinking pool" of potential signatories as the petitioning period progresses, whereby those voters who already have signed the petition of another candidate are no longer available to sign another petition.13 Under these circumstances, identifying and obtaining even a small number of signatures can be like trying to find the proverbial needle in a haystack-especially for a candidate, like Sollazo, who seeks the nomination of a minor party whose pool of potential petition signatories may already be quite small from the outset of the petitioning period. See Molinari, 82 F. Supp. 2d at 67 68 (discussing the shrinking pool problem). Thus, even an election law that requires a candidate to obtain "only a relatively small number of signatures" still can severely burden the right to engage in interactive political speech under the First Amendment "if it also preclude[s] the candidate from utilizing a large class of potential solicitors to convey his message, or if it substantially restrict[s] the candidate's ability to choose the means of conveying his message." Krislov, 226 F.3d at 860.

The District Court distinguished American Constitutional Law Foundation by positing a distinction between initiative petition circulators, whom it considered to be engaged in interactive political speech, and candidate petition circulators such as Lerman, whom it considered to be engaged in a "classic ballot access type" of activity. Lerman v. Board of Elections, No. CV-99-4901, transcript at 10 11 (E.D.N.Y. Aug. 31, 1999). It ultimately relied upon this distinction to conclude that the section 6-132(2) witness residence requirement poses only a minimal burden on interactive political speech and, therefore, less scrutiny was required when evaluating that requirement than in American Constitutional Law Foundation. This distinction, however, is rather strained. There is no basis to conclude that petition circulation on behalf of a candidate involves any less interactive political speech than petition circulation on behalf of a proposed ballot initiative-the nature of the regulated activity is identical in each instance. The District Court's distinction depends almost entirely upon its characterization of candidate petition circulation as being "classic," but the fact that candidate elections have a longer history than ballot initiatives or referenda upon petition does not actually establish anything with respect to the burden placed on "interactive political speech." To be sure, the distinction between candidate and initiative campaigns can be relevant in certain circumstances. For example, the government's interest in preventing corruption or the appearance of corruption is sufficiently important to justify limits on contributions to candidate campaigns, but is not sufficiently important when it comes to regulating contributions to initiative campaigns, since there are no candidates to corrupt in those campaigns. Compare Buckley v. Valeo, 424 U.S. 1, 26 27 (1976) (government interest in preventing corruption or the appearance of corruption is "constitutionally sufficient justification" for limits on contributions to candidate campaigns), with First Nat'l Bank of Boston v. Bellotti, 435 U.S. 765, 790 (1978) ("The risk of corruption perceived in cases involving candidate elections . . . simply is not present in a popular vote on a public issue." (citations omitted)). By contrast, the strength of the state's interest in preserving the integrity of the ballot access process itself is no greater when petition circulators advocate in behalf of particular candidates than it is when they advocate in behalf of ballot initiatives. The distinction between candidate and initiative elections, therefore, is not salient in this context.

Indeed, if the distinction between candidate and initiative ballot access petition bears any relevance at all, it may well suggest the opposite conclusion from that reached by the District Court-namely, that the witness residence requirement places an even greater burden on candidate petition circulators than on initiative petition circulators. See Krislov, 226 F.3d at 861 62. Unlike initiative campaigns, which typically concern a single issue, campaigns for political office typically require candidates to advance their positions on a variety of issues. As a result, individuals who circulate petitions on behalf of candidates for office typically "must speak to a broader range of political topics" than those who circulate petitions in support of ballot initiatives, and by limiting the number of individuals available to circulate petitions on behalf of candidates, the section 6-132 witness residence requirement "has the potential to squelch a greater quantity and broader range of political speech than laws which only restrict initiative proponents." Id. Since we decline to rely upon this distinction, we disagree with the District Court's conclusion that the witness residence requirement presents only a minimal burden on political speech and association.

B. Application of Strict Scrutiny to the Section 6-132(2) Witness Residence Requirement

Since the section 6-132(2) witness residence requirement imposes a severe burden on political speech and association, the requirement must therefore be narrowly tailored to advance a compelling state interest in order to pass constitutional muster. See California Democratic Party v. Jones, 120 S. Ct. 2402, 2412 (2000); Buckley v. American Constitutional Law Found., 119 S. Ct. 636, 642 n.12 (1999); Eu v. San Francisco County Democratic Central Comm., 489 U.S. 214, 225 (1989). As in Molinari, the defendants attempt to justify the section 6-132(2) witness residence requirement based on three asserted interests: (1) ensuring integrity and preventing fraud in the electoral process; (2) ensuring that candidates demonstrate "a sufficient modicum of support" before their name is included on the ballot; and (3) ensuring that non-residents may not "impose[] the cost of a primary on the district." Molinari v. Powers, 82 F. Supp. 2d 57, 73 74 (E.D.N.Y. 2000) (internal quotation marks omitted). Again, we agree with Chief Judge Korman's decision in Molinari and conclude that the witness residence requirement does not bear even a rational relationship to any of these three justifications, let alone the narrowly tailored relationship that strict scrutiny demands.

As a general matter, the first interest asserted by the defendants-ensuring integrity and preventing fraud in the electoral process-is unquestionably compelling. See Krislov v. Rednour, 226 F.3d 851, 859 (7th Cir. 2000) ("Because elections must be regulated to remain free from fraud and coercion, some latitude is given to regulations designed to serve these purposes."); Schulz v. Williams, 44 F.3d 48, 57 (2d Cir. 1994) (noting that states are "entitled to take steps to ensure that elections are 'fair and honest'"). And were the defendants able to establish that the use of non-resident petition circulators did, in fact, pose a demonstrable threat to the integrity of the signature collection process, we would be obliged to give greater weight to their argument. However, the fact that the defendants' asserted interests are "important in the abstract" does not necessarily mean that its chosen means of regulation "will in fact advance those interests." Turner Broadcasting Sys., Inc. v. F.C.C., 512 U.S. 622, 664 (1994). And in general, the potential dangers to the integrity of the electoral process are more remote during the signature collection process than at the actual time of balloting on election day. See American Constitutional Law Found., 119 S. Ct. at 648; Grant, 486 U.S. at 427; (WIN) Washington Initiatives Now v. Rippie, 213 F.3d 1132, 1139 (9th Cir. 2000); Krislov, 226 F.3d at 865. Accordingly, we cannot uphold a statutory provision that substantially burdens political speech and association at the petition stage of the electoral process without insisting that the defendants "do more than simply 'posit the existence of the disease sought to be cured.'" Turner Broadcasting Sys., 512 U.S. at 664; see Krislov, 226 F.3d at 865.

The defendants' have failed to suggest any meaningful relationship between the section 6-132(2) witness residence requirement and their interest in protecting the integrity of the signature collection process. With respect to that interest, the "usual justification" for a residence requirement is to ensure that the petition witness be answerable to a subpoena. Molinari, 82 F. Supp. 2d at 73; see also American Constitutional Law Found., 119 S. Ct. at 644. The witness residence requirement in section 6-132(2), however, does nothing to advance that end. Since the local Boards of Elections in New York have statewide subpoena power, see N.Y. ELEC. L. § 3-218(1) (McKinney 1998), the state's purpose is already served by the less burdensome requirements in section 6-132(2) that a petition witness (1) live anywhere within the State of New York, and (2) provide their residence address in an affidavit filed together with the petitions.14 See N.Y. ELEC. L. § 6-132(2) (McKinney 1998); Molinari, 82 F. Supp. 2d at 73; see also American Constitutional Law Found., 119 S. Ct. at 644 (state residence requirement "more precisely achieve[s] the State's subpoena service objective" than voter registration requirement (internal quotation marks omitted)); Krislov, 226 F.3d at 865 (state interest in maintaining the integrity of the electoral process "can be achieved just as easily (and probably more effectively) through" the more narrowly tailored process for challenging invalid signatures).

There also is little reason to believe the defendants' assertion that district residents are more likely to have "some familiarity with persons who sign petitions" or be more qualified to ensure the integrity of the petition circulation process. In an electoral district consisting of many thousands of voters, the likelihood of district residents having any greater "personal familiarity" than non-residents is rather doubtful. Indeed, New York law no longer even requires, as it once did, that petition witnesses attest to personal knowledge of the identity of designating petition signatories. See Molinari, 82 F. Supp. 2d at 73 74 (discussing requirement under predecessor to section 6-132(2) that subscribing witness "know[] each of the voters whose names are subscribed" to the candidate's petition (internal quotation marks and citation omitted)). Further casting doubt on the defendants' asserted justification for the witness residence requirement is the fact that verification of petition signatures has become considerably easier than it once was given the use of signature imaging technology. Under this system, the signature of every registered voter in each of New York's sixty-two counties may be verified rather easily against a digitized version of that voter's signature on file with the NYS Board and the local boards of elections in each county and the city of New York. See NEW YORK STATE BD. OF ELECTIONS, ANNUAL REPORT 14 (1999) (discussing availability of signature imaging technology in every county as of 1999, and its use in helping local boards of elections "to more easily validate a voter's eligibility in each individual election"); see also Schulz, 44 F.3d at 59 (suggesting that existence of "statewide computer system . . . capable of checking the registration status" of all New York voters might, once implemented, "change[] the calculus" by which ballot access restrictions are to be evaluated); Molinari, 82 F. Supp. 2d at 72 (noting that computerization of New York voter registration lists makes it considerably easier to verify identity of petition witnesses and signatories and, therefore, undermines justification for requirement that a petition witness "separately list the town or city in which he or she resides in addition to his or her residence address."). By the same token, if this rationale is understood to mean that district residents are more likely to be aware of the physical boundaries of that district, so as to avoid soliciting invalid signatures from individuals who are not district residents, then the state's interest reduces to mere paternalism, based on the notion that "the State needs to protect candidates because they aren't savvy enough to find solicitors who can read a street map, and thereby refrain from soliciting signatures outside the relevant district." Krislov, 226 F.3d at 864.

Noting the exception to the residence requirement that permits non-resident notaries public or commissioners of deeds to circulate and witness petitions, see N.Y. ELEC. L. § 6-132(3) (McKinney 1998), the defendants urge us, in effect, to characterize petition circulators as quasi-public officials. Such a characterization, however, does not accurately reflect the role actually played by petition circulators in the ballot access process. As partisans of the candidates for whom they are circulating petitions, petition circulators-regardless of whether they are district residents-almost certainly are motivated entirely by an interest in seeing their candidate gain access to the ballot, rather than any interest in ensuring integrity in the signature collection process. See Grant, 486 U.S. at 426 (rejecting argument that paid petition circulator "is any more likely to accept false signatures than a volunteer who is motivated entirely by an interest in having the proposition placed on the ballot"); Molinari, 82 F. Supp. 2d at 73 ("The reality is that all witnesses are partisan petition circulators whose goal is to convince voters to sign the petitions."). Moreover, even if we were to accept the defendants' understanding of petition circulators as quasi-public officials, rather than simply as advocates for a particular candidate, that characterization might just as easily lead us to conclude that district residents are even more likely to be motivated or well-positioned to engage in chicanery than non-residents, since district residents arguably have a stronger self-interest in which candidate ultimately gets elected and are more likely to be enmeshed in the intra-district political circles in which fraud might flourish.

The defendants' arguments based on the other two interests are even weaker, and both fail for similar reasons. Requiring a candidate to have a "modicum of support" within their district before their name appears on the ballot is well-established as a legitimate and important state interest, which helps to "avoid[] confusion, deception, and even frustration of the democratic process." Jenness v. Fortson, 403 U.S. 431, 442 (1971); see also Prestia v. O'Connor, 178 F.3d 86, 88 (2d Cir. 1999), cert. denied, 120 S. Ct. 539 (1999); Schulz v. Williams, 44 F.3d 48, 57 58 (2d Cir. 1994); LaRouche v. Kezer, 990 F.2d 36, 39 (2d Cir. 1993). However, that interest already is advanced by the requirement that candidates obtain a minimum number of signatures by district residents. See N.Y. ELEC. L. § 6-136 (McKinney 1998) (specifying number of signatures required from within a given political subdivision in order to gain access to ballot for various public offices). In light of that requirement, the state's interest in requiring a "modicum of support" from within the district bears no relationship whatsoever to who actually witnesses or circulates the petition. See Grant, 486 U.S. at 425 26; Krislov, 226 F.3d at 865; Molinari, 82 F. Supp. 2d at 74; Rockefeller v. Powers, 917 F. Supp. 155, 164 (E.D.N.Y.), aff'd, 78 F.3d 44, 46 (2d Cir. 1996).

The defendants' final asserted interest-preventing non-resident witnesses from "imposing" the costs of a primary upon a district's residents fails for similar reasons. As with their assertion of the state's interest in ensuring a "modicum of support" from within the district, the defendants' argument with respect to the "imposition of primary election costs" appears to misunderstand the nature of the ballot access process-for candidates only are placed on the ballot if they obtain the requisite number of signatures by registered party members from within the electoral district, no matter where the witnesses live.15 See Molinari, 82 F. Supp. 2d at 74. On the other hand, to the extent that the defendants mean to argue that the witness residence requirement helps to prevent non-residents from influencing politics within the district, that interest does not appear to be legitimate at all. A desire to fence out non-residents' political speech-and to prevent both residents and non-residents from associating for political purposes across district boundaries-simply cannot be reconciled with the First Amendment's purpose of ensuring "the widest possible dissemination of information from diverse and antagonistic sources." Krislov, 226 F.3d at 866 (quoting Buckley v. Valeo, 424 U.S. 1, 49 (1976)); see also Warren v. Fairfax County, 196 F.3d 186, 190 (4th Cir. 1999) (en banc) (invalidating law prohibiting non-residents from using public forum); VanNatta v. Keisling, 151 F.3d 1215, 1218 (9th Cir. 1998) (invalidating law prohibiting candidates from accepting campaign contributions from outside their district), cert. denied, 525 U.S. 1104 (1999). In any event, the strength of this asserted interest in preventing outsiders from influencing politics within the district is undermined by the exception to the residence requirement that permits notaries public and commissioners of deeds-even if they live outside the relevant political subdivision-to circulate and witness candidates' ballot access petitions. See N.Y. ELEC. L. § 6-132(3) (McKinney 1998) (exception for notaries public and commissioners of deeds); Molinari, 82 F. Supp. 2d at 74.

Finally, the argument that the witness residence requirement does not affect Lerman's "absolute right to enter the 50th district . . . [to] express, as vigorously as she desired, her views and opinions about the issues facing the voters in that district and the candidates for the City Council" is irrelevant. "We have consistently refused to overlook an unconstitutional restriction upon some First Amendment activity simply because it leaves other First Amendment activity unimpaired." California Democratic Party v. Jones, 120 S. Ct. 2402, 2412 (2000); see also Grant, 486 U.S. at 424 ("That appellees remain free to employ other means to disseminate their ideas does not take their speech through petition circulators outside the bounds of First Amendment protection."). Cf. American Constitutional Law Found., 119 S. Ct. at 644 ("The ease with which qualified [individuals] may register to vote . . . does not lift the burden on speech" placed by voter registration requirement for petition circulators). The fact that the witness residence requirement "leaves open 'more burdensome' avenues of communication," therefore, "does not relieve its burden on First Amendment expression." Grant, 486 U.S. at 424.

Since the section 6-132(2) witness residence requirement does not bear any relationship to the asserted state interests, we conclude that the statute has no "plainly legitimate sweep" at all, Broaderick v. Oklahoma, 413 U.S. 601, 612 (1973), and is therefore invalid on its face under the overbreadth doctrine. For the same reason, the plaintiffs would prevail even if we applied the Salernostandard, since they sufficiently have demonstrated that "no set of circumstances exists" in which the witness residence requirement constitutionally could be applied.

We hold that the requirement under Section 6-132(2) of the New York Election Law that witnesses to ballot access designating petitions be residents of the political subdivision in which the office is to be voted violates the First Amendment on its face. We therefore REVERSE the District Court's judgment and REMAND with instructions to enter judgment consistent with this opinion.