• Attorneys



• Cases and Documents



• Alan Morrison Supreme Court Assistance Project



• Consumer Justice Project



• Public Interest FOIA Clinic



• Freedom of Information Clearinghouse



• Internet Free Speech



• Fellowship and Employment Opportunities



• FAQs



• Litigation Group
  E-Update

The decision under review, which is unreported, was rendered by Judge Michael J. Davis of the United States District Court for the District of Minnesota. The complaint challenged the constitutionality of several state statutes, thus giving the district court jurisdiction under 28 U.S.C. Secs. 1331 and 1343(a)(3), (4). This Court has jurisdiction under 28 U.S.C. Sec. 1291, as a final judgment was entered on 16 September 1994 and a notice of appeal filed on 26 September 1994.

The most apposite cases are:

Tashjian v. Republican Party of Connecticut, 479 U.S. 208 (1986);

Eu v. San Francisco County Democratic Central Committee, 489 U.S. 214 (1989);

Anderson v. Celebrezze, 460 U.S. 780 (1983); and

Norman v. Reed, 502 U.S. 279, 112 S. Ct. 698 (1992).

The relevant constitutional provisions are the First and Fourteenth Amendments to the United States Constitution. The relevant statutes are those provisions of Minnesota's election law which appellant is challenging as unconstitutional, namely Minn. Stat. Secs. 204B.06, and 204B.04, subds. 1 and 2, the texts of which appear in the next section of this brief.

Appellant Twin Cities Area New Party (the "New Party" or the "Party"), is a fledgling party with a progressive political agenda. In this suit the New Party is seeking the restoration of an associational right historically available throughout the United States, namely, the right to join another party in nominating the same candidate for office. This electoral strategy is known as "plural nomination," "multi-party nomination," "cross-nomination," or electoral "fusion" (the term that we will use here).

This case was filed in the United States District Court for the District of Minnesota after local election officials refused to accept the New Party's nominating petition for State Representative Andy Dawkins as the party's candidate for state representative from his incumbent district in St. Paul. Their reason was that Mr. Dawkins, who was willing to accept the New Party nomination, was also a candidate for reelection as the candidate of the Democratic-Farmer-Labor ("DFL") party. The New Party was thus barred from nominating him under Minn. Stat. Sec. 204B.06, which states: "An affidavit of candidacy shall state . . . that the candidate . . . [h]as no other affidavit on file as a candidate for any office at the same primary or next ensuing general election." In addition, Minn. Stat. Sec. 204B.04, subd. 1, states that no candidate "certified by a canvassing board as the nominee of any major political party shall be named on any ballot as the candidate of any other political party at the next ensuing general election"; subdivision 2 states that no person who "seeks nomination for any partisan office at a primary shall be nominated for the same office by nominating petition."

The New Party challenged these anti-fusion states as violating its associational rights under the First and Fourteenth Amendments. Named as defendants in their official capacities were Lou McKenna and Joan Anderson Growe. Mr. McKenna is Director of the Ramsey County Department of Property Records and Revenue, in which role he is responsible for administering the Minnesota election law with respect to races for offices lying within that county. Minn. Stat. Secs. 204B.09, subd. 1, and 200.02, subd. 16. Ms. Growe is the Secretary of State (J.A. 5). This brief will refer to these defendants collectively as the "State."

The New Party moved for a preliminary injunction, seeking to compel the State to place Mr. Dawkins on the November 1994 ballot as the New Party's nominee (J.A. 11-12). Responding to the parties' request for a ruling before ballots had to be printed, and treating the papers as cross motions for summary judgment (J.A. 40), the district court granted summary judgment to the State on 14 September 1994 (J.A. 38-39). The court's reasons appear in an amended order and memorandum, which was filed two days later (J.A. 40-50). (See Footnote 1)

B. Statement of facts.

The relevant facts here are not in dispute and fall into two categories: facts concerning the historical incidence and effects of fusion and fusion bans, and the facts giving rise to this litigation. In part 1, we outline the historical record regarding fusion and state bans on fusion. Much of this discussion is exemplified by an uncontroverted affidavit from Walter Dean Burnham, a political science professor from the University of Texas (J.A. 26-37). Part 2 then recites the facts of this case.

1. Relevant historical facts.

While the practice of joint party nomination of candidates may appear novel for present-day Minnesotans, it was familiar to their forebears and remains familiar to residents of other states today. Until the very end of the 19th century, fusion was legal throughout the United States. It was utilized particularly heavily by parties in the Midwest and West in the latter 19th century, figuring in every significant election in that period. See Argersinger, "A Place on the Ballot": Fusion Politics and Antifusion Laws, 85 American Historical Review 287, 288 (1980). Although fusion was banned in Minnesota in 1901, id. at 297-302, it remains a right in New York, Connecticut and other states, with varying degrees of incidence in its exercise.

In New York, the practice has often been decisive in determining electoral outcomes. For example, in the presidential elections of 1940, 1944 and 1960, neither Franklin Delano Roosevelt in the first two races nor John F. Kennedy in the latter race received enough votes on the Democratic line to carry New York state, but they prevailed when their votes on the Liberal line were added to those on the Democratic line (J.A. 30 Par. 11). In the 1980 presidential race, Jimmy Carter received more votes as a Democrat than Ronald Reagan did as a Republican, but Reagan's additional votes on the Conservative Party line allowed him to carry New York. 14 R. Scammon, ed., America Votes 278, 290 (1982). That same year, in a race for the U.S. Senate, Elizabeth Holtzman received more votes on the Democratic line than Alfonse D'Amato received on the Republican line, but D'Amato won the election anyway with the additional votes he gleaned as the nominee of both the Conservative and the Right to Life parties. Id. at 279, 290.

Minor parties have historically relied heavily on fusion to advance the interests of their members. Summarizing the 19th century experience, America's leading electoral historian, Peter Argersinger, notes that:

***[Fusion] helped maintain a significant third party tradition by guaranteeing that dissenters' votes could be more than symbol

ic protest, that their leaders could gain office, and that their demands might be heard. Most of the election victories normally attributed to the Grangers, Independents, or Greenbackers in the 1870s and 1880s were a result of fusion between those third party groups and Democrats.***

Argersinger, supra, at 288. Observers of contemporary electoral practice in states which still permit fusion underscore the importance of the right to explaining the viability of minor parties there. (See Footnote 2)

Professor Burnham's declaration in this case discusses how laws barring fusion disproportionately burden minor parties (J.A. 32, Par. 13), and his view is echoed in all the available historical and political science scholarship on the subject.(See Footnote 3) Moreover, the legislative record in states banning fusion indicates that curbing minor party activity was the central aim of the major party legislators responsible for fusion bans. (See Footnote 4) This aim was substantially accomplished with the proliferation of state bans on the practice. Operating in an environment generally devoid of the fusion option, 20th century minor parties were driven to the periphery of mainstream politics. They became much more candidate-centered, less frequent, and less enduring than their 19th century counterparts. (See Footnote 5)

There is no real dispute about this record or the reasons why it is so. As Professor Burnham notes, the availability of a fusion option allows minor parties to advance the interests of their members within the context of the "first past the post" or "winner take all" election rules of American politics (J.A. 27-28, Pars. 4-5). In a winner-take-all system, political influence depends on the ability to win, or contribute meaningfully to winning, a majority or plurality of votes. By definition, a "minor" party is typically unable to command the majority or plurality of votes needed to win electoral office on its own. If deprived of the ability to join electoral coalitions with other parties, a minor party will suffer from a "wasted vote syndrome"; even voters who support the party's ideology and program will generally decline to show that support at the polls because they are reluctant to "waste" votes on candidates perceived as having no serious chance of winning (J.A. 28, Par. 6). A wedge is driven between the political effectiveness of minor parties, as expressed by support at the polls, and the real level of support their platforms and ideologies enjoy in the population. (See Footnote 6)

Where permitted, both historically and today, fusion is widely used by minor parties because it enables them to avoid this wasted vote syndrome. Fusion permits minor parties to enter into alliances with major parties and participate in potentially winning electoral coalitions around select candidates -- with votes for the candidate on either ballot line aggregated in determining support vis-à-vis rivals. Supporters of minor parties are thus free to express that support, by voting on the line of the minor party, without removing themselves or their vote from the arena of meaningful electoral competition (J.A. 32-33, Par. 13).

If avoiding the wasted-vote syndrome permits minor parties to gain voting support from those who indeed share their views, the most vivid example of how such support permits minor parties meaningful influence is when a candidate is elected to office as the nominee of two parties, one major and one minor, and the margin of victory is smaller than the number of votes he or she received on the minor party's line (J.A. 30-31, Par. 11; J.A. 18, Par. 18). See examples at pp. 4-5, supra. The resulting tally, in which a candidate's vote totals on both party lines are added together, makes it clear that the minor party's support was crucial. This sends a message to the electorate, the candidate, and the major party alike about the need to respond to the minor party's agenda and concerns. Even if the minor party votes are not decisive in this way, however, fusion permits a more accurate voter signal to candidates, parties, and other voters about the real distribution of political sentiment in the electorate.

It is just this sort of message that voters are forbidden to communicate under the statutes at issue here. In effect, a fusion ban consigns minor parties to seeking "protest" votes for candidates whom the voters perceive as having little (if any) chance of winning, and it prevents minor parties from pursuing their agenda and the interests of their members through a strategy of cross-nominating candidates whom voters perceive as more "electable" and who, if elected, may work to advance the goals of that minor party.

In sum, fusion is not an obscure bit of electoral exotica, but a crucial balancing mechanism in our winner-take-all electoral system. Historically, the opportunity to pursue a fusion strategy has been the distinctively American answer to the question of how, within such a system, to assure electoral minorities a fair "availability of political opportunity." Lubin v. Panish, 415 U.S. 709, 716 (1974); Argersinger, supra, at 288-289. Reciprocally, as Professor Burnham notes (J.A. 32-33, Par. 13), the removal of that option by legislatures dominated by the major parties had the (intended) effect of denying minor parties effective and ongoing political expression in all but the most extraordinary situations.(See Footnote 7)

2. Relevant facts of this litigation.

The Twin Cities Area New Party was chartered in the spring of 1993 as a chapter of the national New Party, a progressive party with chapters in over a dozen states (J.A. 13-14). Although it is a fledgling organization, the Party's broad aims are identical to those of established parties: to promote candidates its members judge best represent their views, to use the electoral process to advance its program, and to widen its base support in the general electorate (J.A. 17). To advance these aims New Party members resolved on a mixed electoral strategy which would combine the nomination of candidates solely as candidates of the New Party with (in appropriate cases) the nomination of willing candidates of other parties (J.A. 16).

At a membership meeting held in April 1994, the New Party duly voted to nominate Andy Dawkins, the incumbent DFL State Representative, as the New Party's candidate for House District 65A (located in St. Paul) in the November 1994 general election. Mr. Dawkins stated that he was willing to accept the New Party nomination as well as the DFL nomination, for which he was running unopposed in the DFL primary (J.A. 16, 20).

As a minor political party, the New Party does not nominate candidates in a primary. Minn. Stat. Secs. 204B.03, 204B.07. Instead, its candidates appear on the general election ballot only if the Party files a nominating petition containing the requisite number of signatures for each candidate it has nominated (Minn. Stat. Sec. 204B.08, subd. 3). In July 1994, the New Party gathered the required number of signatures. On 18 July, the Party sought to file its nominating petition with the Ramsey County Department of Property Records and Revenue, which administers Minnesota's election law with respect to races in that county. The petition was rejected (J.A. 14-15). Joan M. Pelzer, Supervisor of Elections/Voter Registration, explained in a letter that the nominating petition was being rejected because Mr. Dawkins had previously filed an Affidavit of Candidacy as the DFL candidate, and under Minn. Stat. Sec. 204B.06, subd. 1, candidates must declare that they have "no other affidavit on file as a candidate for any office at the same primary or next ensuing general election" (J.A. 15, 19).

The New Party filed this suit on 10 August 1994, challenging the validity of this restriction, along with the companion restrictions in Minn. Stat. 204B.04, subds. 1 and 2, which provide that (1) candidates nominated by a major party shall not appear on the ballot as another party's candidate shall appear on the ballot as the candidate of more than one party, and (2) candidates for a major party's primary may not compete in the party's primary and also seek a place on the ballot by nominating petition. Because the deadline for printing ballots was approaching, the New Party sought a preliminary injunction to require the State to list Andy Dawkins as the New Party line as well as the DFL candidate in the November 1994 general election. The State opposed this motion, and, pursuant to Rule 65(a)(2), Fed. R. Civ. P., the parties agreed to treat their moving papers as cross-motions for summary judgment (J.A. 40).

C. The district court's decision.

After hearing argument from counsel, the district court entered an order granting summary judgment to the State on 14 September 1994 (J.A. 38-39). An amended order and memorandum, as well as judgment thereon, were entered two days later (J.A. 40-50).

While recognizing as "beyond question" that the "rights of the citizenry to voluntarily associate themselves for partisan political purposes is among the core values protected by the First and Fourteenth Amendments" (J.A. 43), the district court rejected the Party's claim that Minnesota's fusion ban violated its rights under the First and Fourteenth Amendments.

In reaching this conclusion, the district court purported to apply the framework for considering constitutional challenges to state election laws established by Anderson v. Celebrezze, 460 U.S. 780 (1983). Under the Anderson framework, a reviewing court

***. . . must first consider the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate. It then must identify and evaluate the precise interests put forward by the State as justification for the burden imposed by the rule. In passing judgment, the [c]ourt must not only determine the legitimacy and strengths of those interests, it must also consider the extent to which those interests make it necessary to burden the plaintiff's rights. Only after weighing all these factors is the reviewing court in a position to decide whether the challenged provision is unconstitutional.***

Id. at 789, cited at J.A. 43-44.

With regard to the rights at stake and their burdening under Minnesota law, the district court first recognized that:

***There can be little doubt that at the core of the associational rights protected by the First and Fourteenth Amendments is the right of a political party to select "a standard bearer who best represents the party's ideologies and preferences." . . . . Moreover, "[t]o the degree that a state would thwart these First and Fourteenth interests by limiting the access of new parties to the ballot, we have called for the demonstration of a corresponding interest sufficiently weighty to justify the limitation, and we have accordingly required any severe restriction to be narrowly drawn to advance a state interest of compelling importance."***

J.A. 43-44 (citations omitted).

The court also rejected the State's principal argument, namely, that the ban could be justified under Storer v. Brown, 415 U.S. 724 (1974), which upheld a California law barring independent candidates from seeking office unless they had been disaffiliated from a political party for at least a year. The district court explained that the disaffiliation law at issue in Storer sought to advance the state's interest in preventing "sore losers" from carrying intraparty disputes beyond the primary to the general election. Here, by contrast, Mr. Dawkins had won the DFL party's primary, so he was "not a sore loser but, rather, a willing participant" (J.A. 45).(See Footnote 8)

Nevertheless, the district court found that the particular infringement on Party autonomy in nomination was not sufficient "to trigger a heightened scrutiny of the challenged statutes" (J.A. 47). The court noted that "[t]here is no allegation that the New Party was denied access to the ballot," or that it would need to "engage in rigorous and superhuman planning in order to access to ballot" (J.A. 46). Rather, the New Party "was denied a place on the ballot" "solely because it chose as its candidate one chosen by another party" as well (Id.). From these facts ("[o]n that basis"), the court concluded that the fusion ban is a "politically neutral regulation" (J.A. 46-47) which places only "a minimal burden" on the New Party, which is free "to support whoever they want," except for "nominating candidates with other party affiliations" (J.A. 49). In doing so, the court drew explicit analogy to the regulation at question in Burdick v. Takushi, 112 S.Ct. 2059 (1992) (discussed infra), which upheld a state ban on the right of an individual voter to cast a write-in vote in an uncontested election. Although the district court declared that the Minnesota fusion ban was "neutral," the court did not explicitly consider the New Party's claim that the consequences of the ban's application had a disproportionate impact on minor parties.

Having found the New Party's rights only minimally burdened, the district court went on to conclude that two state interests were sufficiently "compelling" to justify the Minnesota fusion ban (J.A. 47). The first was a concern about voter confusion if Andy Dawkins appeared on the ballot twice, once as the New Party candidate and once as the DFL candidate. The court explained that Minnesota "has a compelling interest in setting out a ballot in which all of the candidates are treated equally and which is a plain statement of the available choices" (J.A. 48).

A second, "equally compelling" interest was held to be "knowing how the winner will be determined." The district court noted that the New Party's fusion strategy contemplated aggregating Mr. Dawkins' vote totals into a single sum, regardless of the party line on which they were cast. The "flaw" the court perceived with this result is that Mr. Dawkins "could lose the election under either party designation, but he could still emerge victorious after the totals were aggregated" (J.A. 48). The court considered the prevention of such a result a "compelling reason" for enacting a fusion ban, so as to assure that the winner of an election receives the voters of a majority (or at least a plurality) of the votes (J.A. 48).

The district court added that it would have reached the same result even under a stricter level of scrutiny, which the court thought was reserved for cases where statutes "have the effect of excluding a Party or a candidate from the ballot" (J.A. 49), and the opinion concluded with the observation that the case presented "matters of policy" which were best left to "the state Legislature [as] the appropriate forum" for their resolution (J.A. 49-50). This appeal followed (J.A. 51).

Relevant Supreme Court decisions establish two key principles which guide the analysis here. First, political parties enjoy a considerable measure of autonomy under the First and Fourteenth Amendments, and states are thus limited in their ability to regulate a party's core rights, which include nominating candidates, deciding how the party should be operated and mapping general electoral strategy. Second, the cases establish that if a state law disproportionately burdens a minor party, the law may be unconstitutional even if it is neutral on its face and seems to apply equally to all parties.

The decision below is fundamentally inconsistent with these principles and should therefore be reversed. The district court correctly identified the balancing test to be used in this case, which requires reviewing courts to identify the nature of the associational right being restricted and to weigh the effect of that limitation against the state's interest in regulating the challenged practice. Severe restrictions on a party's ability to function demands heightened scrutiny and proof of a compelling state interest, while less severe restrictions may be upheld under a more deferential standard.

The district court erred, however, in its analysis. The fusion ban restricts the New Party's ability to select its standard bearer, and the Supreme Court has repeatedly held that one of the core rights enjoyed by political parties is the right to select the candidate who will best advance the party's interests in the electoral arena. Differently put, any attempt by a state to dictate to a party who it can or cannot nominate is inherently suspect, and a state must come forward with proof of a compelling countervailing interest which is narrowly tailored to serve those interests.

In viewing the restriction here as only minimal, the district court failed to appreciate the importance of party autonomy as a constitutionally protected right under the governing case law. The district court also erred when it viewed the burden imposed by the challenged statutes as slight, the theory being that the New Party was still free to nominate virtually anyone it wanted. Even if Minnesota's anti-fusion laws may appear to be neutral and non-discriminatory on their face, their impact falls disproportionately on third parties, which is why close scrutiny is required.

A fusion ban may have no impact whatever on the nominating or electoral strategies of major parties, but it can have a decidedly negative impact on minor parties, which have trouble winning elections on their own because of the concern by voters that they not "waste" their votes on candidates with no realistic chance of winning. Because of this factor, a third party which wants to play an active role in electing candidates to office must make common cause with other parties. Differently put, a state law which tells minor parties they may not pursue a fusion strategy effectively consigns those parties and their supporters to the political margin. Any statute which produces such a result imposes a "severe" infringement on a minor party's core rights, and heightened scrutiny of the reasons for the limitation is required.

While the district court undervalued the party right being restricted, it overvalued the justifications put forth by the State for this ban. The first justification was that voters would be confused if Mr. Dawkins' name appeared on the ballot twice, once as the DFL candidate and once as the New Party candidate. This argument has been routinely rejected in a number of Supreme Court cases, and the State offered no proof that confusion exists in states where fusion is a routine practice, nor did it prove that fusion was a problem during the period that it was legal in Minnesota and many other states. If anything, the presence of a candidate on two party lines allows voters to send parties and candidates a message that is not possible under the current system, namely, that they support a particular candidate and want to see him or her elected, but they do not otherwise support the major party which nominated that candidate.

The State also thought fusion could be properly banned as a means of assuring that voters would know who the winner was and that the person elected to office achieves a majority or at least a plurality of the votes cast. This argument is internally inconsistent and is does not present a compelling reason for the current ban. The New Party is claiming a right to cross-nominate Mr. Dawkins and to have his votes on the DFL and New Party lines aggregated. There is thus no question how the winner will be selected in that scenario: Whoever gets the most votes wins, regardless of whether the votes for that candidate are cast on one or more lines.

Nor did the State have a compelling interest in blocking a scenario under which, say, Mr. Dawkins' Republican opponent receives 41 percent of the vote on the Republican line, Mr. Dawkins receives 40 percent of the vote on the DFL line and another 19 percent of the vote on the New Party line, and Mr. Dawkins is declared the winner. The district court did not suggest that Mr. Dawkins' opponent should be declared the winner in that scenario, so the legality of fusion as an electoral strategy does not thwart the will of the majority (or plurality). What the court seemed to be saying that it is somehow "wrong" for a candidate to obtain an electoral majority on two lines and that the State is justified in forcing the New Party's supporters into a Hobson's choice: either vote for Andy Dawkins as the DFL candidate, or else vote for a separate New Party candidate who may be considerably weaker and less able to advance the New Party's aims. In effect, this argument posits that third parties should stay on the periphery and serve only as a vehicle for protest votes, and that type of policy preference is flatly inconsistent with the First and Fourteenth Amendments.

MINNESOTA'S FUSION BAN VIOLATES THE NEW PARTY'S RIGHTS AS A PARTY UNDER THE FIRST AND FOURTEENTH AMENDMENTS.

A. Standard of review.

The question of whether the challenged statutes are constitutional is an issue of law to be decided under a de novo standard of review.

B. Governing constitutional standards.

For several decades now, the Supreme Court "has recognized the constitutional right of citizens to create and develop new political parties." Norman v. Reed, supra, 112 S. Ct. at 705. Such activity "advances the constitutional interest of likeminded voters to gather in pursuit of common political ends, thus enlarging the opportunities of all voters to express their own political preferences." Id.

The right of citizens to develop political parties derives from the freedom of association granted by the First Amendment, as well as the protection of "fundamental rights" secured by the Equal Protection Clause of the Fourteenth Amendment. Anderson v. Celebrezze, supra, 460 U.S. at 787 n.7. Although these two constitutional provisions are implicated when a state seeks to regulate a party's electoral activities, a reviewing court is to examine the constitutionality of such restrictions under a single unified standard. Id.; see also Norman v. Reed, supra, 112 S. Ct. at 705 n.8.

That standard, as enunciated in Anderson v. Celebrezze and refined in later cases, is essentially a balancing test. A reviewing court is required first to identify the "character and magnitude of the asserted injury" to the party's rights, and then to examine the "precise interests put forward by the State as justifications for the burden imposed by its rule," taking into account "the extent to which those interests make it necessary to burden the plaintiff's rights." 460 U.S. at 789. The "rigorousness" of the inquiry "depends upon the extent to which a challenged regulation burdens First and Fourteenth Amendment rights." Burdick v. Takushi, 112 S. Ct. 2059, 2063 (1992). When those rights "are subjected to 'severe restrictions, the regulation must be 'narrowly drawn to advance a state interest of compelling importance.'" Id., quoting Norman v. Reed, supra, 112 S. Ct. at 705. If a state law imposes only "'reasonable, nondiscriminatory restrictions,' . . . 'the State's important regulatory interests are generally sufficient to justify' the restrictions." Id. at 2064-65, quoting Anderson v. Celebrezze, supra, 460 U.S. at 788.

In part C, we outline the nature of the rights which parties enjoy under the applicable Supreme Court precedents. In part D, we explain why the district court erred in finding that the burdens imposed on the New Party were not substantial enough to require proof of a "compelling" countervailing interest. In part E, we then explain why the district court erred in accepting the justifications proffered by the State.

C. Political parties enjoy significant First and Fourteenth Amendment rights.

The decision below is inconsistent with a long line of Supreme Court decisions which hold that the First and Fourteenth Amendments grant political parties a great deal of autonomy when it comes to performing key functions to advance the interests of party members. Freedom to select the candidate of a party's choice is one of those core rights protected by the Constitution, along with the freedom to make key organizational decisions affecting a party's policy, internal governance and electoral strategy.

The case law makes this point clear in a variety of contexts. Thus, in Cousins v. Wigoda, 419 U.S. 477 (1975), and Democratic Party of the United States v. Wisconsin ex rel. La Follette, 450 U.S. 107 (1981), the Court held that state law could not override rules of the national Democratic Party for selecting that party's delegates to a presidential convention. In Tashjian v. Republican Party of Connecticut, 479 U.S. 208 (1986), the Court held that a state Republican Party had a First Amendment right to adopt a rule allowing non-members to vote in its primary, striking down a closed primary law interfering with that right. In Eu v. San Francisco Democratic Party, 489 U.S. 214 (1989), it invalidated a state law prohibiting political parties from endorsing candidates before a primary. Finally, in Norman v. Reed, supra, the Court enjoined enforcement of an Illinois law which prohibited a minor party from using the name of another, consenting party. In these cases and others, the Supreme Court has repeatedly underscored the specific importance of respecting a party's autonomy when it comes to nominating the party's candidates, and an exacting level of scrutiny is applied to any state election law purporting to limit that right. It is this uninterrupted record that the district court departs from below.

Moreover, even in dealing with the kind of sensible administrative regulations that any electoral scheme must have, the Supreme Court has held that states may not adopt measures that disproportionately burden minor parties. Two cases illustrate this element of the Court's jurisprudence in this area. In Williams v. Rhodes, 393 U.S. 23 (1968), the Court struck down an Ohio ballot access law which discriminated on its face against minor parties. The state defended its law as promoting the two-party system "in order to encourage compromise and stability," to which the Court responded: "[T]he Ohio system does not merely favor a 'two-party system'; it favors the Republicans and the Democrats -- and in effect tends to give them a complete monopoly." Id. at 31-32.

Similarly, in Anderson v. Celebrezze, supra, the Court struck down a facially neutral early filing deadline for Presidential candidates, finding that it disproportionately burdened candidates such as John Anderson, who did not decide to make an independent bid for office until after the deadline. The Court held that a "burden that falls unequally on new or small political parties or on independent candidates impinges, by its very nature, on associational rights protected by the First Amendment." 460 U.S. at 793-94 (emphasis added).

In Anderson, and significantly for this case, Ohio defended in part by noting that five other minor parties had been able to meet the early filing deadline. But the Court found the point unpersuasive, noting that the other five parties (including the Communist Party, Socialist Workers Party and Workers World Party) represented "ideologically committed minor parties" with an electoral strategy focused more on making an ideological statement than playing an active role in deciding which candidate would be elected. 460 U.S. at 791 n.12. While the filing deadline did not affect these sorts of parties, it did disproportionately burden minor parties of the kind led by Anderson, which were not content to operate at the political margin -- and which sought, as the New Party does here, to achieve real political effectiveness. Id.

On its face, Minnesota's fusion ban clearly burdens the New Party's autonomy in nominating the candidate of its choice. Thus, it starkly contravenes the constitutional principles enunciated in Cousins, La Follette, and especially Tashjian, Eu, and Norman. Facially neutral, it is also precisely the sort of restriction that imposes a "disproportionate impact" on minor parties and that the Court for that reason condemned in Williams and especially Anderson. There is no real question in this case that removing the fusion option from electoral politics has a disproportionate impact on minor parties such as the New Party, and the reason derives from the winner-take-all electoral system. (See Footnote 9) To compete effectively in such a system, any party must show the ability to win. Absent coalitions with others, this showing is not generally available to minor parties because (by definition) minor parties command something far short of a plurality of the electorate as supporters.

D. Those party rights are heavily burdened here.

Despite this long record of Supreme Court clarity on the subject, the district court deemed the rights at stake in this case to be insignificantly burdened. The court concluded that the party's right to nominate the candidate of its choice was not significantly affected by a fusion ban, on the theory that the New Party could still nominate anyone it wanted, so long as that person had not sought or received another party's nomination (J.A. 46). Thus, the argument continued, even if the New Party could not nominate Andy Dawkins, there was still a large field of potential candidates, so the fusion ban should be viewed as having a minimal effect and operating as a neutral regulation which does not receive any level of heightened scrutiny. This approach was in error for at least three reasons.

First, the district court got off on the wrong foot by analyzing this issue in light of Burdick v. Takushi, 112 S. Ct. 2059 (1992), which was cited for the proposition that the fusion ban is simply "a reasonable politically neutral regulation which 'serves merely to channel expressive activity at the polls'" (J.A. 47). The problem with this analysis is that Burdick involved the rights of voters, not the rights of parties, and the two are analytically quite distinct.

The district court's comparison of this case to Burdick is inappropriate. The plaintiff in Burdick challenged Hawaii's ban on write-in votes in uncontested races, asserting his "right" as a voter included the option of casting a write-in "protest" vote for Donald Duck. Id. at 2065. The Court found this anti-write-in statute a "very limited" burden on First Amendment freedoms, requiring little stringency in review. The "right" in question was trivial, and it was asserted by an individual voter, not a political party. Id.

Here, much more is at stake. What the New Party is seeking -- and what the fusion ban is burdening -- is not the right of individuals to make an expressive protest. At stake here is a party's ability to pursue a pragmatic electoral strategy designed to elect candidates. The fact that this right is sought by a political party, not just an individual voter, requires an entirely different calculus by the court. The Supreme Court made this point in Tashjian v. Republican Party of Connecticut, which struck down a "closed primary" law that prevented the state Republican party from opening its primary to unaffiliated voters. The statute in Tashjian was functionally the same as the closed primary law which the Court upheld in Rosario v. Rockefeller, 410 U.S. 752 (1973). Indeed, ten years before Tashjian, the Court summarily affirmed a decision upholding the validity of the very same Connecticut statute in Nader v. Schaffer, 417 F.Supp. 837 (D. Conn.), summarily aff'd, 429 U.S. 989 (1976).

Tashjian explained the apparent inconsistently by noting that in Rosario and Nader, the closed primary rule had been challenged by voters who wanted to participate in the primaries of parties over the parties' opposition; by contrast, the law challenged in Tashjian restricted the freedom of a party that wanted to open its primary to non-members. 479 U.S. at 215 n.6. In all three cases, the rights of the parties prevailed and outweighed the rights asserted by individual voters. In Rosario and Nader, the party's right to exclude non-members outweighed the right of the non-members to vote, just as in Tashjian, the right of the party to permit non-members to vote in its primaries prevailed and the state law restriction was held unconstitutional.

Second, the district court considered only the quantitative impact of the fusion ban on the New Party's ability to nominate candidates without considering its qualitative impact on the Party. Differently put, the court asked merely "how many" potential candidates does the fusion ban eliminate, when the Supreme Court's cases indicate that the proper question is whether, to what degree and with what consequence, has party autonomy and self-governance been infringed. Given the core character of this right, and the absolute centrality of nomination rights in its exercise, the Supreme Court has repeatedly emphasized how troubling it finds any restriction on the nomination rights of parties.

In Tashjian, for example, the entire Court, majority and dissenters alike, took it as axiomatic that a political party has the strongest First Amendment interest in nominating candidates of its choice. The majority of the Court explicitly stated that it would be unconstitutional for a state law to limit a party's nominees only to party members, and it drew an analogy to precisely such a restriction to explain its holding that Connecticut could not bar a party from opening its primary to non-members. The Court stated:

***Were the state to . . . provide that only Party members might be selected as the Party's chosen nominees for public office, such a prohibition or potential association with nonmembers would clearly infringe upon the rights of the Party's members under the First Amendment to organize with like-minded citizens in support of common political goals***

479 U.S. at 215-6. And in their dissent from the decision in Tashjian, Justices Scalia and two other justices acknowledged that the freedom of association "encompasses a political party's decisions about the identity of, and the process for electing, its leaders"; that "the ability of [party] members to select their own candidate . . . unquestionably implicates an associational freedom"; and that state laws "restricting the ability of the Party's members to select whatever candidate they desire" would be unconstitutional. Id. at 235-6 (emphasis added).

The Court reaffirmed the virtual inviolability of a political party's nomination powers in Eu, holding that:

***Freedom of association means not only that an individual voter has the right to associate with the political party of her choice, but also that a political party has a right to identify the people who constitute the association, and to select a "standard bearer who best represents the party's ideologies and preferences." . . . Depriving a political party of the power to endorse "suffocates that right . . . at the crucial juncture at which the appeal to common principles may be translated into concerted action, and hence to power in the community.***

Id. at 224 (citations omitted). The Court quoted from the dissent in Tashjian, holding further (at 229-30) that:

***Freedom of association . . . encompasses a political party's decisions about the identity of, and the process for electing, its leaders . . .. The ability of the members of a [political party] to select their own candidate . . . unquestionably implicates an associational freedom.***

Below the Supreme Court, there is one case supporting the district court's conclusion that a fusion ban does not implicate a "severe" restriction on the associational rights of minor parties. In Swamp v. Kennedy, 950 F.2d 383 (7th Cir. 1991), cert. denied, 112 S. Ct. 2992 (1992), the panel majority reached this conclusion in upholding Wisconsin's fusion ban, but we submit that Swamp was wrongly decided and should not be followed here. There are two reasons why we say this.

First, a powerful answer to the majority's analysis in Swamp -- and to the district court's decision here -- was provided in the opinion of Judges Ripple, Posner and Easterbrook, who dissented from the Seventh Circuit's refusal to rehear Swamp en banc. 950 F.2d at 388-89. These three judges wrote that the panel had "deviated in important respects" from the method and analysis used by the Supreme Court in Anderson, Eu and Norman and that those precedents required "a more indepth examination of the associational rights at issue" than the panel had afforded them. With respect to a minor party's liberty interest, which anti-fusion laws impinge, Judges Ripple, Posner, and Easterbrook noted:

***The Supreme Court has recognized that the right of a party to nominate a candidate of its choice is a vital aspect of the party's role in our political structure. The ability to choose the same person as another party is an important aspect of that right. It allows a party to form significant political alliance. When a minor party nominates a candidate also nominated by a major party . . . it may -- and often does -- offer the voters a very real and important choice and send an important message to the candidate. If a person standing as the candidate of a major party prevails only because of the votes cast for him or her as the candidate of a minor party, an important message has been sent by the voters to both the candidate and to the major party.... In short, permitting people to vote for a candidate on one party line rather than another increases the opportunity of both voter and party to be heard and for workable political alliances to be formed. ***

950 F.2d at 388-89.

A second reason why Swamp fails to persuade is that the panel rendered its decision before the Supreme Court's decision in Norman v. Reed and declined to reconsider that ruling in light of Norman. The problem with this omission is that Norman v. Reed provides perhaps the best refutation to the district court's view that no significant infringement is involved if a minor party can nominate anyone it wants except candidates nominated by another party.

In Norman, the Harold Washington Party had achieved a measure of electoral success in the city of Chicago, and adherents of that party's agenda sought to expand the party's base by establishing a Harold Washington Party to field candidates for offices in the suburban parts of Cook County, Illinois, where Chicago was located. The organizers of this suburban party ran headlong into an Illinois statute which barred them from using the name "Harold Washington Party" because it was already in use by an existing party. 112 S. Ct. at 704-06.

The limitation in Norman is analytically indistinguishable from the restriction here. Illinois law allowed the suburban party's organizers to call their party anything they wanted, so long as it was not the name of another party. Thus, the restriction on the suburban party's autonomy was numerically no more restrictive than the one upheld by the district court here. Nonetheless, the Supreme Court treated the Illinois law as a basic assault on that party's freedom, and the restriction was analyzed under the strict scrutiny standard.

Norman provides a convenient jumping-off point for considering the second, related, set of rights enjoyed by minor parties of implication here: the right to operate in a political system free of restrictions disproportionately burdensome to them as minor parties. In Norman, the Court noted that the only way a minor party could avoid the "draconian" effect of this "same name" statute would be to run state-wide, an electoral strategy which "would obviously foreclose the development of any political party lacking the resources to run a state-wide campaign." Id. at 706. The district court's opinion makes no mention of this point, though the parallels are clear. A statute such as the party name restriction in Norman poses no obstacle to the two major parties, which are certainly capable of running candidate state-wide. The statute places a disproportionate burden on smaller, fledgling parties which cannot afford to field a state-wide campaign, the Court viewed the restriction as significant and required Illinois to advance a compelling interest for the limitation.

Even more directly than in Norman, a seemingly neutral or innocuous restriction on party activities has a huge impact on the freedom of minor parties to operate in a political system which dies not discriminate against them or unduly burden in their capacity as minor parties. There is no question that fusion bans damages minor parties in ways that are trivial or even beneficial to the two major parties, and the disproportionality of the burden is evident and huge.

This fact, not addressed at all in the district court's opinion, was not seriously contested in this case -- for the good reason that all available historical evidence supports the point. With fusion, minor parties flourish even in normal periods; without it, they do not. A world war or economic depression may occasion their fleeting emergence, see note 7, supra; at other times, tiny groups of the "ideologically committed" (as Anderson put it) may embrace minor parties as vehicles of expressive protest; but overall the historical record shows that minor parties have little staying power as effective competitors of major parties.

Under Anderson, as buttressed by Tashjian, Eu, and Norman, this is simply not a legislative choice which a reviewing court can countenance, at least without a searching inquiry into the state's reasons for it. The Constitution does not mandate the encouragement of minor parties, but neither does allow major parties to wield their de facto monopoly on legislative power to construct artificial barriers to minor party growth and expression, and the restriction at issue here is a classic example of such a barrier. (See Footnote 10)

E. The purported state interests do not justify this ban.

While the district court ignored Anderson's substantive holding regarding disproportionate burdens, it purported to use Anderson's analytic framework for assessing the constitutionality of state election law burdening core freedoms. As subsequently clarified by Norman v. Reed, supra, 112 S. Ct. at 705, and Burdick v. Takushi, supra, 112 S. Ct. at 2063, this framework does not require that any restriction on associative freedoms in electoral arenas be subject to strict scrutiny, though it does require an inquiry as demanding as the restrictions are severe.

When, as here, there is a "severe" restriction on a core party function or activity within the meaning of Anderson and its progeny, the State is obliged to come forward with "compelling" restrictions for the limitation and to show that they are "narrowly tailored" to achieve that end. In the previous discussion, we have demonstrated why the district court erred in not treating the fusion ban as imposing a "severe" restriction on the New Party's autonomy. In this section, we demonstrate why the district court erred in crediting two of the justifications advanced by the State for this ban. We submit that these rationales cannot withstand the heightened level of scrutiny which is required in this case (nor, for that matter, a more deferential level of scrutiny, should the Court disagree with our analysis up to this point).(See Footnote 11)

The two interests which the district court found sufficient to uphold Minnesota's fusion ban are (1) avoiding voter confusion (2) assuring that "the winner is the winner" --- that the winning candidate has a majority or at least plurality of relevant votes. We answer each point in turn.

Voter confusion: The district court's principal concern, that voters might be confused if Andy Dawkins' name appeared twice on the ballot -- once as the DFL candidate and once as the New Party candidate -- is wholly insubstantial.

First, the Supreme Court has repeatedly shown reluctance to find unsubstantiated threats of voter "confusion" as adequate defenses to state electoral laws otherwise burdening constitutional freedoms. The argument that fusion would confuse voters echoes the argument made by the state in Tashjian in defense of a state legislative bar to parties opening their primaries to independent voters:

The legislature could properly find that it would be difficult for the general public to understand what a candidate stood for who was nominated in part by an unknown amorphous body outside the party, while nevertheless using the party name.

479 U.S. at 220. The Supreme Court roundly rejected this line of argument as patronizing, however, holding that:

***To the extent that party labels provide a shorthand designation of the views of party candidates on matters of public concern, the identification of candidates with particular parties plays a role in the process by which voters inform themselves for the exercise of the franchise. Appellant's argument depends upon the belief that voters can be "misled" by party labels. But "[o]ur cases reflect a greater faith in the ability of individual voters to inform themselves about campaign issues." ***

Id. (citations omitted, emphasis added). Similarly, in Anderson, the Court held that:

***A State's claim that it is enhancing the ability of its citizenry to make wise decisions by restricting the flow of information to them must be viewed with some skepticism. As we observed in another First Amendment context, it is true "that the best means to that end is to open the channels of communication rather than to close them."***

460 U.S. at 798, quoting Virginia Pharmacy Board v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 770 (2976). Here, as in Tashjian, Anderson, and Norman, "[t]he State's legitimate interests in preventing voter confusion and providing for educated and responsible voter decisions in no respect 'make it necessary to burden the [Party's] rights.'" Tashjian, 479 U.S. at 222 (citations omitted). (See Footnote 12)

Second, while fusion may be unfamiliar in contemporary Minnesota politics, its widespread practice in the 19th century and its operation today in states where it is currently permitted provides absolutely no evidence of confusion. Professor Burnham's declaration reviews fusion's history and states unequivocally that "there is no evidence in the literature on 'fusion' politics that multiple party nominations have caused confusion among voters" (J.A. 33, Par. 13). (See Footnote 13) A contrary concern would have substance only if it is shown that the Minnesota electorate is less discerning than its forebears or the residents of other states -- a showing that certainly has not been made here.

Third, any state interest in avoiding voter confusion about the real meaning of electoral choices and again, the provision of such choices is a core and legitimate party function in a functioning -- democracy -- is not even rationally related to a ban on fusion. What a fusion ban forbids is a more precise signalling to voters by parties on the ideological pedigree of candidates and the different bases of their support. This is not merely more information, but precisely the kind of information needed for informed political judgments. Fusion also permits more precise signalling from voters to elected officials. As Judges Ripple, Posner, and Easterbrook noted in Swamp:

***If a person standing as the candidate of a major party prevails only because of the votes cast for him or her as the candidate of a minor party, an important message has been sent by the voters to both the candidate and the major party . . .. Such information is of immense value to the electorate, and it would indeed be salutary for the candidate to know which platform the majority of the voters favor.***

950 F.2d at 389.

Far from curbing voter confusion, then, an anti-fusion law frustrates the sources of its relief.

Assuring that "the winner is the winner": The district court also erred in concluding that the State's interests "in knowing how the winner will be determined" and assuring that elected officials are the choice of a majority or plurality of voters (J.A. 49) are sufficient to justify a ban on fusion. These are indisputably legitimate concerns for the State to have, but they are in no way advanced by a ban on fusion.

Under a fusion system, there is no uncertainty about how the winner will be determined. The fusion candidate's total is the sum of his or her votes on each of the ballots lines on which he or she is nominated. If one candidate's totals exceed the total votes cast for the other candidates, the fusion candidate wins. If not, the fusion candidate loses.

The district court nonetheless reasoned that a decision allowing fusion would interfere with the state's interest in "ensuring that elected officials are the choice of a majority or plurality of voters." Id. In explaining this assertion, the court observed that Mr. Dawkins would have "the opportunity not only to be chosen twice but then to have his totals aggregated so that he could lose the election under either party designation, but he could still emerge victorious after the vote totals were aggregated" (J.A. 49). This observation does not support the court's premise.

To put the court's point in concrete terms, the concern is that if fusion were allowed, Andy Dawkins might receive 40 percent of the vote on the DFL line and his Republican opponent might receive 41 percent of the vote; nonetheless, Mr. Dawkins would be declared the winner if he also received 19 percent of the vote on the New Party line.

It is difficult to see the "flaw" in this system, to use the district court's word (J.A. 49). If general elections are intended to determine which candidate has the majority or plurality of votes, there is nothing wrong with declaring Andy Dawkins the winner of an election in which he garnered 59 percent of the vote, simply because the electorate chose to express its support on two party lines, rather than one. As the Burnham affidavit shows (J.A. 29-31, Pars. 9-11), there is ample precedent for aggregating the votes that are cast for the same candidate on different party lines and declaring as the person who receives the most votes the winner. Despite this wealth of evidence answering the district court's concern, these points were never addressed in the decision under review. (See Footnote 14)

The district court's conclusion thus seems to rest on nothing more than a view that there is something inherently "wrong" with declaring a candidate such as Andy Dawkins the winner in a situation where his vote totals on the DFL line do not exceed the total vote cast for his Republican opponent. At bottom, this really amounts to a view that there is something "wrong" with candidates being elected on two party lines, rather than one. This concern is surely not substantial and merely reflects a view that third parties should be small and ineffectual rather than seek to influence the outcome of elections by cross-nominating specific candidates who advance the third party's agenda. Indeed, if the district court's argument were taken to its logical conclusion, then the candidate receiving 41 percent of the vote in the hypothetical should be declared the winner over a cross-nominated candidate who received 59 percent of the vote, a result which can hardly be defended on the ground that "the winner should be the winner."

Bullock v. Carter, 405 U.S. 134, 145 (1972), on which the district court relied, does not support the State's position. That case involved a candidate's challenge to a state law requiring a post-primary runoff where no candidate obtained a majority of the vote. The Court upheld the State's power to require a run-off to guard against a candidate's emerging from, for example, an 11-candidate primary with a ten percent plurality and still appear a place on the ballot as the party's nominee. (See Footnote 15) This concern is simply not presented by the fusion hypothetical posed by the district court under which Mr. Dawkins would have obtained an actual majority.

Indeed, in this respect, by permitting coalitions between parties around a single candidate, fusion makes majorities more likely, not less, again underscoring the lack even of a rational relation between the challenged statute and the interest offered in its defense.

In the final analysis, the district court's reasoning amounts to little more than a conclusion that things would be somehow different if fusion were allowed: that candidates might appear on the ballot twice, instead of once, and that the votes cast for a given candidate would have to be aggregated if his or her name appeared on more than one line. The fact that things would be different if the New Party wins this suit is hardly a "compelling" interest in preserving the status quo, particularly when one considers the vital role which fusion plays in protecting the associational rights of political minorities and also the lessons of experience, which show no problems with fusion in those states where it is presently allowed. (See Footnote 16))

2 D. Mazmanian, Third Parties in Presidential Elections (1974), notes (at 134-35):

The essential attribute of New York's modified two-party (fusion) system is the options it provides to

more resources

Because Public Citizen does not accept funds from corporations, professional associations or government agencies, we can remain independent and follow the truth wherever it may lead. But that means we depend on the generosity of concerned citizens like you for the resources to fight on behalf of the public interest. If you would like to help us in our fight, click here.