Beaver v. Clingman | Cases | Westlaw

*1 This case presents a challenge to important aspects of the process by which Oklahomans “bring forth those persons desired by their fellow citizens to govern.” Bullock v. Carter, 405 U.S. 134, 148—49 (1972).

Plaintiff Andrea L. Beaver is a resident of Tulsa County and a registered Democratic voter in Tulsa County, Oklahoma.

Plaintiff Floyd Turner is a resident of Tulsa County and a registered Democratic voter in Tulsa County, Oklahoma.

Plaintiff Minelle L. Batson is a resident of Oklahoma County and a registered Democratic voter in Oklahoma County, Oklahoma.

Plaintiff Mary Y. Burnett is a resident of Rogers County and a registered Republican voter in Rogers Country, Oklahoma.

Plaintiff Michael L. Seymour is a resident of Tulsa County and a registered Republican voter in Tulsa County, Oklahoma.

Plaintiff Terry L. Beaver is a resident of Tulsa County and a registered Republican voter in Tulsa County, Oklahoma.

Plaintiff Robert T. Murphy is a resident of Cleveland County, a registered Libertarian voter in Cleveland County, Oklahoma, is a candidate for the Libertarian nomination for United States Representative for the 5^th District of Oklahoma, and is the chair of the Oklahoma Libertarian Party.

Plaintiff Sharon Lynn Atherton is a resident of Tulsa County and a registered Libertarian voter in Tulsa County, Oklahoma.

Plaintiff Roger Bloxham is a resident of Tulsa County, a registered Libertarian voter in Tulsa County, Oklahoma, was a candidate for the Libertarian nomination for Corporation Commissioner in Oklahoma's primary election conducted on August 22, 2000, and is the Libertarian nominee for Corporation Commission in the State of Oklahoma for the general election to be conducted on November 7, 2000.

Plaintiff Steve Galpin is a resident of Oklahoma County, a registered Libertarian voter in Oklahoma County, Oklahoma, and is the Libertarian Voter Registration and Ballot Access Coordinator for the State of Oklahoma.

Plaintiff Richard P. Prawdzienski is a resident of Oklahoma County, a registered Libertarian voter in Oklahoma County, Oklahoma, and was a candidate for the Libertarian nomination for Corporation Commissioner in Oklahoma's primary election conducted on August 22, 2000.

Plaintiff Michael A. Clem is a resident of Tulsa County, a registered Libertarian voter in Tulsa County, Oklahoma, and is a candidate for the Libertarian nomination for United States Representative from the 1^st District of Oklahoma.

Plaintiff Whitney L. Boutin, Jr. is a resident of Cleveland County, a registered Libertarian voter in Cleveland County, Oklahoma, was the leading candidate for the Libertarian nomination for Corporation Commissioner in Oklahoma's primary election conducted on August 22, 2000. Plaintiff Boutin alleges that he withdrew from that candidacy in order to avoid a run-off election because all Oklahoma voters would not be permitted to vote in that run-off election and in order to save taxpayers money which would otherwise be expended on a run-off election which would not be conducted in the manner that the Libertarian Party of Oklahoma desired.

*2 Plaintiff Christopher S. Powell is a resident of Oklahoma County, a registered Libertarian voter in Oklahoma County, Oklahoma, and is a candidate for the Libertarian nomination for the State House of Representatives in District 100.

Plaintiff Charles A. Burris is a resident of Tulsa County and a registered Libertarian voter in Tulsa County, Oklahoma.

Each of the individual plaintiffs are citizens of the State of Oklahoma and are registered voters of the State of Oklahoma who wished to vote in the LPO primary election in Oklahoma on August 22, 2000 and who wish to vote in any future LPO primary or run-off elections conducted in Oklahoma.

Plaintiff the Libertarian Party of Oklahoma (styled in the caption as “a/k/a Libertarian Political Organization”) is, for reasons which are discussed below, not currently recognized in the State of Oklahoma as a political party. It is recognized in the statutory “political organization” category. The Libertarian Party of Oklahoma is sometimes referred to in this order as LPO.

Defendant Michael Clingman is the Secretary of the Oklahoma State Election Board. In that capacity, Secretary Clingman is responsible for administering the election and voter registration laws of the State of Oklahoma. Secretary Clingman has supervisory authority over all county election boards, has responsibility to promulgate, repeal or modify such rules or regulations as he deems necessary to achieve and maintain uniformity in the application, operation, and interpretation of state and federal election laws and to maximize correctness, impartiality, and efficiency in the administration of those election laws. Secretary Clingman is the chief state election official responsible for coordination of state responsibilities under the National Voter Registration Act of 1993.

Defendant Glo Henley is the chairman of the defendant Board.

Defendant Kenneth Monroe is the vice-chairman of the defendant Board.

Defendant Thomas E. Prince is a member of the defendant Board.

The defendant Board is a state board organized under the laws of the State of Oklahoma to operate and perform such duties as are prescribed by law for the purpose of overseeing elections conducted by the state.

Plaintiffs have brought this action under 42 U.S.C. § 1983 (and other statutes), alleging that certain Oklahoma election statutes violate plaintiffs' First and Fourteenth Amendment rights to freedom of political association and free speech. Specifically, plaintiffs challenge 27 O.S. § 1–104(A) and § 1–104(B)(4) as statutes which impermissibly prevent a recognized political party in Oklahoma from opening up its primary and run-off elections to all Oklahoma registered voters regardless of those voters' political party affiliation. Plaintiffs seek a judgment declaring these statutory provisions unconstitutional and request injunctive relief prohibiting defendants from enforcing these provisions. 28 U.S.C. § 2201, 2201.

Defendants assert that this action is moot and, on the merits, deny that plaintiffs are entitled to any declaratory or injunctive relief. The court has concluded, for the reasons which are set forth at length in this memorandum opinion, that this action is not moot. The court has also concluded that, although some of the justifications offered by the state for the existing primary election system are unpersuasive, the plaintiffs have not established that they are entitled, as a matter of constitutional right, to imposition of a party-option open primary election system in Oklahoma.

*3 The parties have stipulated that the following facts are true. Accordingly, the court finds the following stipulated facts (¶¶ 1–36), as drafted by the parties, to be established without further proof:

1. All parties are properly before the court.

2. All parties have been correctly designated.

3. Venue is proper in this court.

4. In the year 2000, the filing period for political party nominations in the State of Oklahoma was on July 10, 11, and 12, 2000.

5. In the year 2000, the filing deadline for contesting a political party's candidates in the State of Oklahoma was by July 14, 2000, at 5:00 p.m.

6. In the year 2000, the political party primary election in the State of Oklahoma was held on August 22, 2000. In-person absentee voting began August 17, 2000.

7. In the year 2000, the run-off election in the State of Oklahoma was held on September 19, 2000.

8. In the year 2000, the general election in the State of Oklahoma was held on November 7, 2000.

9. At present, an Oklahoma voter may only register to vote as Republican, Democratic, Reform, Libertarian, or Independent.

10. Only Oklahoma voters registered as Republicans may vote in the Republican Party primary election. This is by designation of the Oklahoma Republican Party.

11. Only Oklahoma voters registered as Democratic may vote in the Democratic Party primary election. This is by designation of the Oklahoma Democratic Party.

12. The Reform Party has allowed Oklahoma voters who are registered as Reform or Independents to vote in the Reform Party primary election. This is by designation of the Oklahoma Reform Party. However, the Reform Party has never had a primary or run-off election contest in the State of Oklahoma.

13. The Libertarian Party, pursuant to Oklahoma statutes, allows Oklahoma voters who are registered as Libertarian or Independents to vote in the Libertarian Party primary election. This is by designation of the Oklahoma Libertarian Party. The Libertarian Party would like to allow all Oklahoma voters, regardless of their political affiliation, to vote in Libertarian Party primary elections.

14. On May 31, 2000, the Voter Registration and Ballot Access Coordinator for the Libertarians in the State of Oklahoma turned in a designation to the Secretary of the Oklahoma State Election Board notifying the Oklahoma State Election Board that the Libertarian Party of Oklahoma consented to voting by all registered Oklahoma voters—whether they were Libertarian, Independent, Republican, Democratic, or Reform—to vote in the Libertarian primary and run-off elections in the State of Oklahoma for the year 2000.

15. On May 31, 2000, Lance Ward, then Secretary of the Oklahoma State Election Board, accepted the Libertarian Party of Oklahoma's request that Independents could vote in the Libertarian Party primary and run-off elections in the year 2000, but did not accept said designation from the Libertarian Party of Oklahoma as to permitting Republican, Democratic, Reform, or any other recognized political party or political association voters voting in the Libertarian Party primary and run-off elections for the State of Oklahoma for the year 2000.

*4 16. On June 21, 2000, the Libertarian Party of Oklahoma was notified that it had achieved political party status in the State of Oklahoma pursuant to OKLA. STAT. tit. 26, § 1–108. Prior to June 21, 2000, the Libertarian Party of Oklahoma was recognized as a political organization pursuant to OKLA. STAT. tit. 26, § 1–109(B).

17. There were 141 registered Libertarian Party voters in the State of Oklahoma as of November 5, 1996.

18. In the year 1996, the Libertarian Party of Oklahoma permitted Independent voters to vote in the Libertarian primary election conducted on August 27, 1996. The total vote cast in the Libertarian primary election for United States Senator on August 27, 1996, was 2,940 votes. Since only Libertarians and Independents could vote in the Libertarian Party's primary, and since there were no more than 141 registered Libertarian Party voters, the vote in the Libertarian Party's primary election on August 27, 1996, for United States Senator, means that at least 95% of the voters casting votes in the aforesaid Libertarian Party primary election were Independent voters.

19. There are currently more than 2,100 voting precincts in the State of Oklahoma. Pursuant to OKLA. STAT. tit. 26, §§ 2–123 and 1–128.1, each precinct must have a minimum of three officials working an election, totaling more and 6,300 precinct workers working each election.

20. Oklahoma statutes require that the Oklahoma Election Board provide training for all members and employees of each county election board every two years. In addition, each precinct official is to receive in-person training every two years. The training conducted by the State Election Board and each respective County Election Board for the 2000 election cycle has already been completed.

21. There were 360 registered Libertarian Party voters in the State of Oklahoma as of June 30, 2000.

22. In the year 2000, there was no political party primary election conducted for the Reform Party in Oklahoma.

23. In the year 2000, the only statewide political party primary election conducted in the State of Oklahoma on August 22, 2000, involved the Libertarian Party primary for Corporation Commissioner.

24. There were three candidates for the Libertarian nomination for Corporation Commissioner in Oklahoma at the primary election to be conducted on August 22, 2000, to-wit: Richard Prawdzinski of Edmond, Oklahoma; Whitney L. Bootin, Jr. of Norman, Oklahoma; and Roger Bloxham of Tulsa, Oklahoma.

25. There were no statewide primary elections for the Republican or Democratic parties of Oklahoma for the primary election held on August 22, 2000.

26. In the year 2000, the Republican Party held primary elections in the State of Oklahoma on August 22, 2000, for United States Representative for Districts 1, 2, 4, and 5, but not in Districts 3 and 6.

27. In the year 2000, the Democratic Party held primary elections in the State of Oklahoma on August 22, 2000, for United States Representative for Districts 1, 2, and 6, but not in Districts 3, 4, and 5.

*5 28. There were several primary and runoff elections conducted by the Republican and Democratic parties in various State Senate and State House of Representative districts in Oklahoma in 2000 as well as local county and municipal races.

29. Preparation and setup for the 2000 primary election ballots began on July 13. Lance Ward, then Secretary of the Oklahoma State Election Board, expected ballot proofing to begin no later than July 18 and ballot printing for the 2000 primary to begin no later than July 20.

30. The Libertarian Party of Oklahoma has passed By-laws and Resolutions which permit all registered Oklahoma voters—whether they be Republican, Democratic, Reform, Independent, or Libertarian—to vote in any primary or run-off election conducted by the Libertarian Party in any election in the year 2000 or in any future primary or run-off election conducted in the State of Oklahoma.

31. If the Libertarian Party of Oklahoma were allowed to open its primary election to all voters at the individual voters choice, each voter would only be able to receive one ballot selection in order to cast his or her vote.

32. The Libertarian Party of Oklahoma has been on the Oklahoma ballot in every presidential election from 1980 through 2000, and has had primary elections in 1980, 1996, and 2000.

33. There were 770 registered Libertarian voters in the State of Oklahoma as of January 15, 2001.

34. The Libertarian Party of Oklahoma is currently recognized as a political organization in the State of Oklahoma pursuant to OKLA. STAT. tit. 26, § 1–109(B).

35. There were 286 registered Libertarian voters in the State of Oklahoma as of November 1, 2002.

36. In the year 2004, the primary election will be held on August 24, 2004, the run-off election will be held on September 21, 2004, and the general election will be held on November 2, 2004.

The parties also stipulated that the following factual matters, though not admitted, would not be contested at the trial:

37. The Libertarian Party of Oklahoma feels that it would be advantageous to it as a political party and important to its members and all voters of Oklahoma if their rights to political association and political speech pursuant to the First and Fourteenth Amendments to the United States Constitution were furthered by opening up the Libertarian Party primary and run-off elections to all Oklahoma voters regardless of political affiliation.

38. The Libertarian Party of Oklahoma and its members feel that opening up of the Libertarian Party of Oklahoma's primary and run-off elections would help them reach out to Libertarian oriented voters of other political affiliations, produce a more viable Libertarian candidate for the general election, create favorable publicity for the Libertarian Party, and encourage further expansion and growth of the Libertarian philosophy. It has also been the experience of most Libertarians that many individuals not formally affiliated with the Libertarian Party share Libertarian views on certain issues so as to support the advancement of common political objectives favored by the Libertarians.

*6 39. There are a number of voters in the State of Oklahoma who are registered in political parties other than the Libertarian Party who wished to vote in the Libertarian Party's primary and run-off elections in the State of Oklahoma in the year 2000 and wish to vote in future primary and run-off elections to be conducted by the Libertarian Party.

40. The Libertarian Party of Oklahoma intends to seek political party recognition in the State of Oklahoma, pursuant to OKLA. STAT. tit. 26, § 1–108, and run candidates for elective office in the election to be conducted in Oklahoma in the year 2004.

41. There are a number of Libertarians who are interested in running in Libertarian Primaries to be conducted in the State of Oklahoma in the year 2004. Among these individuals are Roger Bloxham, Lynn Atherton, and Robert Murphy, all of whom desire to seek the Libertarian nomination for U.S. Senator from Oklahoma in the year 2004.

On the basis of the evidence before it, the court makes the following additional findings of fact:

1. The factual matters set forth in ¶¶ 37–41 in part II, above, to which the parties do not stipulate but which were not contested at trial, are well supported by the evidence and are accordingly found by the court to be true.

2. The Libertarian Party has been an active and energetic, albeit largely unsuccessful participant in Oklahoma politics for more than 20 years. As is set forth elsewhere in the court's findings, at various times, the Libertarian Party has enjoyed official statutory party status. At other times, the status of the Libertarian Party has receded to that of a political organization.

3. In 1980, the Libertarian Party attained full party status (e.g ., statutory political party status) for the first time. Once a party attains recognized political party status, it must satisfy statutory requirements in order to retain that recognition and thus to retain its place on the ballot. Specifically, a recognized political party whose nominee for governor or nominees for electors for president and vice-president (as the case may be) fail to receive at least 10% of the total votes cast for those offices at a general election will loose its status as a recognized political party. The erstwhile political party becomes a statutory political organization. Oklahoma law provides (26 O.S. § 1–109) that status as a political organization will terminate four years after the date that the political party ceases to be recognized or when the political organization regains recognition as a political party (whichever is earlier). The 10% retention requirement is a significant hurdle for minor party ballot access in Oklahoma.

4. The Libertarian Party has lost its recognized status by virtue of its failure to satisfy the retention requirement set forth in Oklahoma law, as applied to the 2000 election cycle.² The Libertarian Party intends to regain full political party status for the 2004 election cycle. Several Libertarian Party candidates plan to run for United States Senate in 2004, making it likely that there will be a primary election with various Libertarian Party candidates competing for the right to stand for election to the United States Senate in the November, 2004 general election.

*7 5. Although the Libertarian Party has not participated in every primary election and has not fielded candidates for all offices in any primary election, the Libertarian Party has been a fairly consistent participant in Oklahoma's primary elections. Robert T. Murphy, who has been active in Libertarian Party politics, has been a Libertarian Party candidate for elective office several times. He ran for United States Senate in 1980 and 1984 as a Libertarian candidate. He ran as a candidate for Congress in 2000. (It was not necessary for Mr. Murphy to run in a primary election in those three elections.) Mr. Murphy intends to run for United States Senate in 2004. Mr. Murphy wishes to have Republicans and Democrats eligible to vote at the primary election stage of his intended 2004 effort to win a United States Senate seat.

6. A Libertarian Party primary election for the United States Senate seat is likely in 2004.

7. In 2000, Richard Prawdzienski ran for Oklahoma Corporation Commissioner as a Libertarian Party candidate. In that election, the Libertarian Party allowed independent candidates to vote in its primary.

8. Floyd Turner, an Oklahoma resident and registered Democrat desired to vote in the Libertarian primary election for Oklahoma Corporation Commissioner in the 2000 election cycle. Of course, the law precluded him from doing so. Mr. Turner does not intend to relinquish his status as a registered Democrat in order to avail himself of the opportunity to vote in the Libertarian primary.

9. The court finds, on the basis of the LPO's consistent record of political activity in Oklahoma and on the basis of the uncontroverted testimony with respect to the plans of the LPO and of the individuals who plan to run as Libertarian candidates, that it is likely that the LPO will once again regain recognized party status and will, accordingly, participate in the 2004 election cycle. The court further finds that it is likely that there will be competition among Libertarian candidates for office which will necessitate a primary election for the Libertarian Party. The court accordingly finds that it is likely that the issues which are presented by the LPO's expressed intent to invite voters affiliated with other parties to participate in its primary are likely to recur in the 2004 election cycle.

10. Compared to other states, it is relatively easy, in Oklahoma, for an independent candidate for public office to gain access to the ballot but relatively difficult for a political party to establish and maintain party status (with resultant ballot access).

11. There are at least four types of primary elections in the United States (perhaps with some variations within these general species). The four general types of political primary may be referred to as the closed primary, the semi-closed primary, the open primary and the blanket primary. In the closed primary election system, there is a separate primary ballot for each party. No crossover voting is allowed. In the semi-closed primary system, there is a separate primary ballot for each party, but independents are permitted to vote in a primary election if the party chooses to permit voting by independents. In the open primary system, all voters may, regardless of registration or party affiliation, choose the primary in which they will vote in that primary election. In the blanket primary system, each voter can choose which primary he or she will vote in for each separate office. Within this nomenclature,³ Oklahoma has a semi-closed primary. The primary election format to which the plaintiffs assert they are constitutionally entitled will be referred to as the party-option open primary.

*8 12. A party-option open primary can help a minor party get established, at least in a broad electoral sense. Nationally, there are 23 states which effectively have open primaries. Utah, Alaska and Rhode Island (Rhode Island by virtue of the decision in Cool Moose Party v. State of Rhode Island, 183 F.3d 80 (1^st Cir.1999)) are the only states which have party-option open primaries.

13. In the 23 open primary states, there were 41 Libertarian primary elections between 1976 and 2002. In 40 of the 41 Libertarian primary elections, the Libertarian candidates garnered 1% or less of the total votes cast in the primary.

14. In 2002, the Utah Democratic⁴ Party opened its contested primary in the First Congressional District to members of all parties. Even though the Democrats have only about 20% of party registration in Utah, 24% of the voters chose to vote in the Democratic primary. Thus, Democrats attracted a fair number of Republicans and Independents. This is an example of the base-broadening effect which the LPO seeks to establish in Oklahoma.

15. It is the political judgment of the Libertarian Party of Oklahoma that the advantages of a party-option open primary outweigh the disadvantages of that format. The Libertarian Party has no plans to change its bylaws in this respect.

16. It is plaintiffs' position that the Libertarian Party is entitled, as a matter of constitutional right, to declare its primary open or closed and to make the same choice with respect to any runoff election which may ensue. In turn, a voter who accepted the invitation to cross over in the primary would be afforded the same choice at the runoff stage, as long as the party in whose runoff he elected to vote had opted to open its primary.

17. In an open primary, the candidates must appeal to the voters of all parties. Plaintiffs' political science expert, Richard Winger, acknowledged that voters in an open primary are, in the aggregate, less likely than they would be in a closed primary to support the ideals of the parties in whose primaries they choose to vote on primary election day.

18. The defendant, Michael Clingman, Secretary of the Oklahoma State Election Board since August, 2001, testified that voter confusion and “raiding” would result from a party-option open primary system. Voter confusion would result from the application of varying rules on primary election day, depending upon the choices made by the parties as to whether to open their primaries. According to Mr. Clingman, the raiding phenomenon, which would occur with a party-option open primary, would put one party in a position to affect the choice of the candidates of another party.⁵ Under Oklahoma election law, a political party cannot determine what individuals will file for office under its banner. Thus, with the phenomenon of raiding, Mr. Clingman suggests that a party-option open primary would give non-members of a minor party the ability to overtake the minor party as a flag of convenience, thus affecting the ideology of the ticket representatives of the party. In this sense, crossover voting would, Mr. Clingman suggests, dilute party identity and impair party identification.

*9 19. Lance Ward, the former Secretary of the Oklahoma State Election Board (having served in that capacity for 13 years) also expressed concern about the voter confusion which he asserts would result from a party-option open primary. Mr. Ward believes that the potential for confusion resulting from differences between parties in any given election would be compounded by differences in treatment of voters from one election to the next.

20. The legislation permitting registered Independents to vote in primary elections was enacted in 1987 (presumably in the wake of the Tashjian decision in 1986), but was first implemented in the 1996 general election. Mr. Ward confirmed that this was a relatively simple change to implement.

21. Mr. Ward suggested that the potential for voter confusion would be maximized if a party opted to open its primary but close its runoff. He raised the possibility that this system could result in the casting of a number of irregular ballots sufficient to make the outcome of a primary election mathematically uncertain. As described by Mr. Ward, this problem results from changing the ballots voters get from election to election, perhaps within a three week period (between a primary and a runoff primary).

22. The court is not persuaded that significant voter confusion would result from a party-option open primary. As indicated by plaintiffs' exhibits 1 and 2, very simple rules for voting eligibility can be posted at polling places when the primary and runoff elections are conducted.

23. In Atherton v. Ward, 22 F.Supp.2d 1265 (W.D.Okla.1998), the Libertarians sued the State Election Board for the purpose of establishing a right to register as Libertarians even though the LPO was not then a recognized political party in Oklahoma. The defendants predicted that if the Libertarians were granted the relief they sought, the result would be administrative problems amounting to a “parade of horrors,” in Judge Alley's words. 22 F.Supp.2d 1268. At the trial of the case at bar, Lance Ward, the lead defendant in Atherton, could offer no evidence that his predictions, in Atherton, of administrative problems had come true. Likewise, Mr. Ward could not cite any instance, from any of the states which have party-option open primaries, in which difficulties were encountered in informing the voters as to which parties' primary elections and which parties' primaries were closed. This court finds in this case, as did Judge Alley in Atherton, that the granting of the relief sought by the plaintiffs would not result in any substantial, let alone insuperable, administrative burden.

24. The extent to which crossover voting (e.g., Republicans and Democrats voting in the Libertarian primary) would occur if Oklahoma had a party-option open primary is highly speculative. For the same reason, the effect of that crossover voting on the “donor” parties is a highly speculative matter.

25. Mr. Clingman asserts, in substance, that the state has an interest in seeing to it that the parties' candidate selections are the result of the choices of persons of like ideology and that, under the Oklahoma election scheme, the safety valve is the ease with which a candidate can run as an independent.

*10 26. Mr. Ward also cited as an evil resulting from a party-option open primary system the possibility that major party candidates might tailor their positions to attract voters who might otherwise abandon their party's primary. Another evil which would result from a party-option open primary system, according to Mr. Ward, is “swamping.” Swamping is the control of a minor party's candidate selection by voters affiliated with other parties. The phenomenon of swamping was vividly described by the Supreme Court in California Democratic Party v. Jones, 530 U.S. 567, 578 (2000), where the Court noted that, in California's blanket primary, the total number of votes cast for party candidates in some races was more than double the total number of registered party members.

27. Because of Oklahoma's liberal filing rules, the state's professed concern for protection of the political identities of the parties is not fully expressed in Oklahoma's election rules. For instance, in the 2002 election in Oklahoma's third district, no Democrat filed against the Republican incumbent. Consequently, at the end of the filing period, a radical left wing or right wing candidate could have filed for Congress as a Democrat (assuming he had registered as a Democrat before the deadline; see 26 O.S.2001 § 5–105), conferring upon himself status as the Democratic Party standard bearer.

28. In terms of the precise issues in this case relating to the effect of implementation of a party-option open primary in Oklahoma, the expert testimony was both highly interesting and highly speculative, although both of the retained experts ventured some opinions that were plausible, at least in the sense that they were not counterintuitive. The experts also testified to some historical facts which were informative for this court and which may likewise be informative for a reviewing court. In addition to the testimony of their retained expert, the defendants presented expert testimony from Lance Ward, a former Secretary of the State Election Board, and from the defendant Michael Clingman, the current Secretary.

29. The opinions of Richard Winger, called by the plaintiffs, are summarized in defendants' exhibit 12. The opinions of Dr. Robert Darcy, called on behalf of the defendants, are summarized in defendants' exhibit 8. Mr. Winger and Dr. Darcy were well-qualified to address the matters which they addressed as expert witnesses.

30. Elsewhere in this Memorandum Opinion, the court discusses the issues and contentions which are essentially irrelevant to the court's disposition of this case and has likewise discussed those matters which are decisive. In light of the court's parsing of the issues, point-by-point resolution of the conflicts in the expert testimony is not necessary.

31. Although the court relies principally upon the affidavit of Richard Winger and on the transcript of his trial testimony as a record of the theories he expounded in this case, some aspects of his testimony deserve comment here. Mr. Winger testified persuasively that, under Oklahoma's ballot access rules, it is relatively difficult for minor parties to get established in Oklahoma politics. He pointed out that no Libertarian or member of any other minor party has been elected to any Oklahoma public office in more than 25 years and that, in more than 25 years, no minor party has had more than two years and five months of continuous legal recognition. This, of course, has consequences with respect to party membership. Party membership is, for instance, much higher in Kansas, where the Libertarian Party has been able to maintain its recognized status continuously.

*11 32. Mr. Winger also testified persuasively that opportunistic candidacies, by candidates who do not share a party's ideology but join the party so that they can get on the ballot, are relatively unlikely in Oklahoma because, in Oklahoma, it is quite easy for independent candidates (for offices other than president) to get on the ballot. They simply need to pay the filing fee. For that reason, it is significantly less likely in Oklahoma that opportunistic candidacies, with resultant electoral raiding, will occur.

33. Mr. Winger also described in some detail the differences between Oklahoma law and California law with respect to independent candidacies and ballot access for minor parties. In California, according to Mr. Winger, it is much easier for minor parties to gain ballot access and much more difficult for independent candidates to run for public office. On this basis, Mr. Winger explained in understandable terms the reasons for which the Libertarians' strategy in Oklahoma, with respect to the party-option open primary, is markedly different from the Libertarians' strategy in California (as demonstrated by the opinion in California Democratic Party v. Jones ).

34. Dr. Robert Darcy, defendants' academic expert, opined that swamping has already occurred in Oklahoma in that independent voters have vastly outnumbered registered Libertarians in the Libertarian primaries. Dr. Darcy is also of the opinion that permitting Republicans and Democrats to vote in the Libertarian primary would diminish if not destroy the Libertarians' control of selection of their own political candidates. He estimated that 2–4% of Republican and Democratic voters would opt to vote in the Libertarian primary. He also opined that opening the Libertarian primary to Republican and Democratic voters would have an effect on the primaries from which the crossover voters were lured.

35. Thus, according to Dr. Darcy, modification of Oklahoma's primary election structure, which is presently designed to eliminate swamping and raiding, would have an effect on the group from which the crossover voters came as well as the group which they joined. The effect on the forsaken party's primary would depend upon the closeness of the primary, the nature of the race and the voting characteristics of the crossover group of voters. These matters are, of course, highly speculative.

36. Dr. Darcy offered some predictions as to how some registered Republicans and Democrats might behave as voters if offered the opportunity to vote in the Libertarian primary while retaining their Republican or Democratic registration. One of the major premises underlying Dr. Darcy's testimony was that the Libertarian primary elections would likely become essentially meaningless by virtue of the numbers of Republicans and Democrats who might cross over to vote in a Libertarian primary if invited to do so. The impact of these prognostications is substantially reduced, however, by the fact that, under Oklahoma election law, all registered Libertarians are purged from the voter rolls as Libertarians when the Libertarian Party, as it typically does, forfeits its recognized party status following the presidential election cycle. Thus, it is highly likely that the ranks of independents and, indeed, of registered Republicans and Democrats, contain numerous voters who sympathize with the LPO but who simply do not wish to go through the motions of re-registering every time they are purged from the rolls.

*12 37. Of course, the focal point of the experts on both sides was the question of the effect, in Oklahoma, of implementation of a party-option open primary in which the LPO would opt to invite voters affiliated with other parties to vote in the Libertarian primary. This is a scenario which has not occurred in Oklahoma. Consequently, the experts' predictions are based entirely upon their assumptions and hypotheses. As has been noted, the experts' opinions as to the effect of implementation of a party-option open primary are speculative. Some well-designed and undoubtedly very expensive exit polling would have reduced the speculativeness of the experts' predictions. However, in light of the court's conclusions as to the dispositive legal issues in this case, there was and is no real need for detailed predictions, with a high degree of certainty, as to the practical effect of implementation of a party-option open primary. As has also been noted, some of the historical facts which have been cited by the experts are of interest and are, of course, more reliable than predictions as to future voter behavior. One notable example is the fact, testified to by Dr. Darcy and shown by his report, that, in the August, 2000 primary election in Oklahoma, nearly 24% of the races were decided by a margin of 5% or less. Dr. Darcy also testified that the 2002 primary election results, in terms of closeness of the margins of victory, were essentially the same. This suggests to the court, and the court finds, that the institution of a party-option open primary format in Oklahoma as sought by the plaintiffs would likely affect the outcome of some primary elections. Viewing the matter from the perspective of the minor parties, the evidence does not persuade the court that the presumed evils of raiding and swamping would be highly likely to occur. Raiding and swamping are phenomena which, in the classic examples, affect the minor party which chooses to open its primary to voters affiliated with other parties. On the other hand, the historical information which is in hand with respect to the number of very close major party primary races in Oklahoma indicates convincingly that, at the margin, implementation of a party-option open primary system would affect Oklahoma primary elections in a way not permitted or intended by the system currently in place.

The LPO is an active organization with political plans for the 2004 election cycle. It is likely that LPO candidates will vie for the Libertarian nomination for United States Senate, and perhaps for other offices, in the 2004 cycle. This is a controversy which is capable of repetition yet evading review. Therefore, this case is not moot. See Democratic Party of the United States v. Wisconsin, 450 U.S. 107, 115, n. 13; Storer v. Brown, 415 U.S. 724, 737, n. 8 (1974); American Party of Texas v. White, 415 U.S. 767, 770, n. 1; all quoting or citing Rosario v. Rockefeller, 410 U.S. 752, 756, n. 5 (1973).

*13 Constitutional challenges to specific provisions of a state's election laws ... cannot be resolved by any litmus-paper test to separate valid from invalid restrictions. Tashjian v. Republican Party of Connecticut, 479 U.S. 208, 213 (1986) (quoting the rule from Anderson v. Celebrezze, 460 U.S. 780, 789 (1983), in turn quoting Storer v. Brown, 415 U.S. 724, 730(1974)).

A 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 justifications for the burden imposed by its rule. In passing judgment, the court must not only determine the legitimacy and strength of each of those interests, it also must consider the extent to which those interests make it necessary to burden the plaintiff's rights. Tashjian, 479 U.S. at 214 (quoting Anderson, 460 U.S. at 789).

Regulations imposing severe burdens on plaintiffs' rights must be narrowly tailored and advance a compelling state interest. Lesser burdens, however, trigger less exacting review, and a state's important regulatory interests will usually be enough to justify reasonable nondiscriminatory restrictions. Timmons v. Twin Cities Area New Party, 520 U.S. 351, 358 (1997) Nor are elaborate, empirical verifications of the weightiness of the state's asserted justifications required. Id. at 364.

The mere fact that a state's system creates barriers tending to limit the field of candidates from which voters might choose does not of itself compel close scrutiny. Burdick v. Takushi, 504 U.S. 428, 433 (1992) (quoting Bullock v. Carter, 405 U.S. 134 (1972). When rights are subjected to severe restrictions, the regulation must be narrowly drawn to advance a state interest of compelling importance. But when a state election law provision imposes only reasonable, nondiscriminatory restrictions upon the First and Fourteenth Amendment rights of voters, the states' important regulatory interests are generally sufficient to justify the restrictions. Burdick v. Takushi, 504 U.S. 428, 434 (1992)(quoting Anderson, 460 U.S. at 788; applying this balancing standard to evaluation of Hawaii's ban on write-in ballots).

Faced with the inherent tension between a political party's right of association and a state's power to regulate elections, the Supreme Court has endorsed a flexible standard of review. Cool Moose Party v. State of Rhode Island, 183 F.3d 80, 82 (1999).

There must be more than a minimal infringement on the rights to vote and of association before strict judicial review is warranted. Nader v. Schaffer, 417 F.Supp. 837, 849 (1976), aff'd 429 U .S. 989 (1976) (upholding, as reasonably related to accomplishment of legitimate state goals, Connecticut statute providing that no person may vote in a party primary unless the voter is on the last-completed enrollment list of that party).

*14 These rules are succinctly summarized in “Elections–Nominations and Primary Elections: The Supreme Court Finds that California's ‘Blanket Primary’ Violates Political Parties' First Amendment Right of Association,” 77 N.D. L.Rev. 827, 830 (2001), as follows:

The following Tenth Circuit authorities are also helpful to the court's understanding of the standard by which it evaluates plaintiffs' challenge to Oklahoma's election statutes. Campbell v. Davidson, 233 F.3d 1229, 1233 (2000)(applying the “flexible standard of Burdick” to find that Colorado's election law requirement that unaffiliated candidates must be registered to vote was not a valid exercise of the state's regulatory power); Campbell v. Buckley, 203 F.3d 738, 742–46 (10^th Cir.2000)(including an informative discussion and determining that the “flexible approach” referred to as “the balancing test,” rather than the strict scrutiny test, applied to a evaluate an unsuccessful challenge to Colorado's ballot initiative law); and Coalition for Free and Open Elections, Prohibition Party v. McElderry, 48 F.3d 493, 496–97 (10^th Cir.1995)(applying the “balancing test” to determine an unsuccessful challenge to the validity of an Oklahoma statute which banned write-in voting in presidential and vice-presidential elections).⁶

As is discussed above, the court must first consider the character and magnitude of the asserted injury to the plaintiffs' First and Fourteenth Amendment associational rights. The court is then required to identify and evaluate the justifications put forward by the state for the burdens complained of. The court will generally discuss the asserted injury and the proffered justifications, following which the court will analyze the matter in more detail.

The plaintiffs in this action include the LPO, as well as individuals registered as Republicans, Democrats or Libertarians. Plaintiffs assert that they “are pro-choice as to the choice to be made by political parties and individual voters on voting in political parties' primary elections....” Plaintiffs' Amended Trial Brief, at 11. Going one step further than Tashjian v. Republican Party of Connecticut, 479 U.S. 208 (1986), and consistently with Cool Moose Party v. State of Rhode Island, 183 F.3d 80 (1^st Cir.1999), plaintiffs assert that Oklahoma's semi-closed primary election system unconstitutionally burdens their freedom to associate with others to advance their political beliefs. Id. at 12. Plaintiffs' claim is elegantly simple and facially appealing. The LPO and the Libertarians assert that they should be able to invite registered Republicans and Democrats into their primary if they choose to do so. The Republican and Democratic plaintiffs correspondingly assert that they should have the option to vote in the Libertarian primary if invited to do so. Accordingly, plaintiffs assert that Oklahoma's semi-closed primary system violates their “constitutional rights to political association and free speech.” Id. at 17.

*15 There is no room for doubt that the “freedom of association protected by the First and Fourteenth Amendments includes partisan political organization.” Tashjian at 214. The First Amendment interests which are at stake here are made all the more important by the fact that the primary election is a crucial stage in biennial cycle of political activity in Oklahoma; hence, the rights of the parties mean little if they are not protected at the primary election stage. Tashjian at 216. The decisions of the Supreme Court “vigorously affirm the special place the First Amendment reserves for, and the special protection it accords, the process by which a political party selects a standard bearer who best represents the party's ideologies and preferences.” Jones, at 575 (internal quotation marks omitted).

This case, of course, involves a dimension which the Supreme Court noted was absent in Tashjian. The competing interests in the case at bar are not just those of the party and the state. We have, instead, the “different combination of considerations” which is raised by the assertion that the court should compel the state to permit the LPO to open its primary to voters affiliated with other parties as well as to independents. Tashjian at 224, n. 13.

The defendants assert that imposition of a party-option open primary format in Oklahoma would result in voter confusion and that the resultant crossover voting would adversely affect the other parties' primaries, with possible effects on the outcome of the other parties' primary races. Amended and Supplemental Trial Brief of Defendants, at 6—7. Elaborating on its arguments, the state asserts that it has an interest in “protecting the integrity of the election process” and that that protection includes “preserving the political parties as viable and identifiable interest groups, insuring that the results of a primary election ... accurately reflect the voting of the party members and prevent[ion][of] confusion or misleading of the general electorate to the extent that the voting public often relies on party labels to make their choice.” Id. at 10. In a display of gallantry not often found in modern day litigation, the state also argues at length that the LPO should lose this lawsuit for its own good. Id. at 11—13. The adverse effects from which the state says the LPO must be protected are swamping and raiding, as discussed above.

(a) Preliminary discussion.

In Tashjian, the Supreme Court held that the party's right to define its associational boundaries were impermissibly infringed by the state's prohibition on participation in the party's primary election by unaffiliated voters. Tashjian at 225. The case at bar presents (as did Cool Moose ) the situation which the Tashjian Court anticipated in its footnote 13:

*16 Tashjian, at 554, n. 13.

The associational rights which were decisive in Tashjian were tested from the opposite direction in Jones. In Jones, the Court held that California's blanket primary in which any primary election voter could, regardless of party affiliation, vote for any candidate for any office—violated the parties' right to define themselves as political associations. Id. at 577.

Tashjian and Jones provide convenient bookends for constitutional analysis, but generalizations based on those cases—regardless of which way the generalizations may cut—are treacherous. This is, in part, a result of the fact that the party option feature of plaintiffs' proposed format for open primaries undermines potent arguments which might otherwise be advanced in support of the challenged legislation. In this respect, Tashjian stands in stark contrast with Nader v. Schaffer, 417 F.Supp. 837 (D.Conn.1976), summarily aff'd, 429 U.S. 989 (1976). In holding Connecticut's closed primary unconstitutional in Tashjian, the Court carefully noted that, in Nader, unlike Tashjian, the Republican Party opposed the plaintiff's effort to participate in the party's primary. Tashjian, at 212. The granting or withholding of a party's consent to participation in its primary by strangers is clearly significant for First Amendment purposes. Hence, many of the considerations which were decisive in Jones, where participation in the primary by strangers was legally compelled, have less force here. On the other hand, the case at bar affects political parties other than the party challenging the legislation in a manner and to an extent that was not present in Tashjian, but perhaps was anticipated in Tashjian's footnote 13. Tashjian and Jones thus provide a useful beginning point, but surely not the ending point, for any analysis of the issues in this case.

Tashjian and Jones teach that state-compelled political association (Jones ) and state-precluded political association (Tashjian ) can impair the associational rights guaranteed by the First and Fourteenth Amendments. These associational rights are clearly implicated in this case, because, regardless of whether it is “politically neutral,” Burdick v. Takushi, 504 U.S. 428, 438 (1992), primary election legislation, by its very nature, regulates political behavior—it necessarily has “the effect of channeling expressive activity at the polls,” id., with consequent inevitable effects on the freedom of the contending political parties to advance their political goals as they see fit.

Resolution of a First Amendment challenge to the political regulation which is in question here also requires, as a preliminary matter, that the court take notice of some aspects of this case which may go far to inform an analysis of the burdens complained of by the plaintiffs and the justifications put forward by the state. As will be seen, there is more to this case than a single-axis analysis of whether we have state preclusion of political association that amounts to an unconstitutional deprivation of associational rights. Moreover, as relevant at this stage of the court's analysis, Tashjian and Jones teach, if they teach nothing else, that cases like the case at bar must be carefully analyzed in light of the totality of the circumstances presented by the state's primary election scheme (a point which is aptly illustrated by the fact that, for reasons which were satisfactorily explained by Richard Winger, an expert witness for the plaintiffs, the Republican and Democratic participation which the LPO so ardently seeks in Oklahoma was anathema to the California Libertarians in Jones ). The court must thus “consider the facts and circumstances behind the law,” Storer v. Brown, 415 U.S. 724, 730 (1974), as was done, for example, in Coalition at 495—96, so that the constitutionality of the challenged provisions may be considered in the appropriate factual context as well as against the “backdrop of Oklahoma's other [primary election] laws.” Id. at 496.

*17 One highly conspicuous facet of this case has already been mentioned: the interests of other political parties are implicated here in a way which was simply not present in Tashjian. In Tashjian, the Connecticut Republican Party successfully asserted a constitutional right to invite unaffiliated voters to participate in its primary election. Imposition of the primary election format which was sought in Tashjian could have had only an indirect effect on the other political parties. None of the other parties were placed in the position of contributing voters to the Libertarian Party while maintaining a home for them on their registration rolls. In contrast, the party-option open primary format sought by the Libertarians in the case at bar would directly affect the other political parties and would make registered party affiliations significantly less meaningful in the Oklahoma primary election system.

Also significant is the fact that Oklahoma voters who are registered with a political party other than the LPO but wish to vote in the Libertarian primary can do so quite easily under the existing legislative scheme. They can change their registration to Libertarian or independent, and are entitled to do so not later than July 1 of any election year. 26 O.S.2001 § 4–119. To be sure, the value of the option to change registration is reduced somewhat by the fact that the deadline for a change in registration falls just before the candidate filing period. 26 O.S.2001 § 5–110. It is also true, however, that the identities of the major party candidates (at least the viable ones) are almost always known well before the filing period begins. Voters who may be sufficiently disenchanted with the major parties to change their registration typically have ample opportunity to develop that disenchantment before the time for change of registration expires. Oklahoma's system thus imposes no substantial restrictions on the voter who may be nomadic in his political preferences. When that fact is borne in mind, it becomes apparent that the sharpest conflict presented by this case is the party vs. party conflict, rather than a state vs. voter conflict.

A final matter which is notable at this point is the role which the state plays under the legislation which is challenged in this case. Many cases in which state election laws are challenged involve access to the electoral competition. See, e.g., Bullock v. Carter, 405 U.S. 134 (1972); Libertarian Party of Oklahoma v. Oklahoma State Election Board, 593 F.Supp. 118 (W.D.Okla.1984). This case, in contrast, involves the rules which govern the competition after all the contending candidates, and their political parties, have entered the field of play. In the former case, the state is a political gatekeeper—a role that assuredly invites skepticism. See, e.g., Bullock and Williams v. Rhodes, 393 U.S. 23 (1968). In this case, the state's role is more akin to that of a referee.⁷ Thus, the challenged legislation is not aimed at minor parties in general, as in ballot access cases, or at the LPO in particular. As has been noted, it is politically neutral, and thus free of any obvious oligopolistic taint. Indeed, the prize sought in Oklahoma by the LPO—participation in the Libertarian primary by Republicans and Democrats—is that which the California Libertarians found repugnant in Jones. Thus we have here a nondiscriminatory restriction which the state's “important regulatory interests are generally sufficient to justify.” Burdick v. Takushi, 504 U.S. 428, 434 (1992) (internal quotation omitted).

*18 (b) The state's arguments based on the possibility of voter confusion and administrative difficulties, and paternalistic concerns, are unpersuasive.

Looking first to the justification which may be most quickly dismissed, the court rejects the state's argument that the semi-closed primary system should be perpetuated because the LPO would be harmed by giving it the option to open its primary. This paternalistic approach has been unequivocally rejected by the Supreme Court. Neither the state nor the court may constitutionally substitute its own judgment for that of the party as to the strategy best calculated to bring about electoral success. Tashjian at 554. Thus, even if the party's chosen strategy consisted of “self-destructive acts, that would not justify a state substituting its judgment for that of the party.” Eu v. San Francisco County Democratic Central Committee, 489 U.S. 214, at 227–28 (1989). The state will fare no better with paternalistic justifications in the case at bar.

The specter of electoral raiding, as proffered by the state, is equally unpersuasive, both factually and legally. Raiding is a practice in which voters in sympathy with one party temporarily align themselves with another party so as to influence or determine the results of the other party's primary. Tashjian at 219; Rosario v. Rockefeller, 410 U.S. 752, 760 (1973). As a factual matter, as has been discussed, the court is in a very poor position to assay the purity of the Libertarian voting electorate because the Libertarian registration rolls are periodically purged, with the result that, to an inherently unmeasurable extent, voters who have Libertarian political inclinations will either come to rest on the roll of unaffiliated voters or will register with one of the major parties for no reason other than to avoid repeatedly being purged to unaffiliated status. No instance has been cited in which, in any jurisdiction, a candidate has effectively hijacked the Libertarian Party for opportunistic reasons. The court is unable to find any significant possibility that opportunistic voters with no real allegiance to Libertarian principles would raid the Libertarian Party. (This is, of course, aside from the fact that the raiding argument is, at bottom, another paternalistic argument of the kind which has found little support from the Supreme Court.)

The state also argues that “swamping” is an evil to be avoided and that swamping would result from imposition of a party-option open primary format. Reduced to its essentials, swamping is the voting phenomenon by which a party's philosophical or political identity (or perhaps both) is lost as a result of primary election participation by non-party members in numbers vastly exceeding the number of voting party members. Referring to crossover voting in California's blanket primary in Jones, the Supreme Court noted that “these substantial numbers of voters who help select the nominees of parties they have chosen not to join often have policy views that diverge from those of the party faithful. Jones, at 578. By definition, swamping affects only minor parties whose registered voter turnout is small in comparison to the available pool of primary election voters. Thus, if swamping is an evil to be avoided, it is an evil which affects only minor parties, in this case the Libertarian Party. Viewing the matter in this light, the court is, for the reasons discussed above with respect to raiding, disinclined to give decisive effect to an essentially paternalistic justification for Oklahoma's present semi-closed primary system.

*19 As has been discussed, the state argues that the proposed party-option open primary format would lead to intolerable voter confusion. The court has resolved the voter confusion argument against the state on the facts (see part III, ¶¶ 21—23, above). The court will note, however, that as a legal matter, voter confusion has provided a slender reed upon which to rest a defense of primary election rules. On one hand, the state has unquestioned authority to control the election process. Tashjian at 217. However, this authority “does not extinguish the state's responsibility to observe the limits established by the First Amendment rights of the state's citizens.” Id. Thus, the 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.” Id. at 221–22 (internal quotation marks omitted). The court has no doubt that the need to avoid voter confusion would become constitutionally relevant if the court were confronted with a proposal which would clearly invite chaos on election day. We have no such situation here. The factual weakness of the state's voter confusion argument fortifies the court's conclusion that that argument is also legally without merit.

Finally, on the subject of administrative difficulties, Messrs. Ward and Clingman both alluded to administrative concerns in their testimony. The concerns voiced by Messrs. Ward and Clingman with respect to administrative difficulties are related to the voter confusion argument also advanced by the state. The essential argument is that if different voting eligibility rules applied to the various parties whose ballots would be available on primary election day, precinct-level snafus could create the situation which is the nightmare of any election official—the casting of questionable ballots in a number sufficient to render the outcome mathematically uncertain. While the court does not take lightly the suggestion that the engrafting of additional complications onto the primary election system might engender difficulties in administering that system throughout Oklahoma's 2251 precincts, the court has simply been provided no basis for a finding that the implementation of a party-option open primary system would create insuperable administrative burdens.

(c) The state has adopted a politically neutral primary format which does not unconstitutionally burden the plaintiffs' First and Fourteenth Amendment Rights.

As is discussed in part IV(A)(2), above, the court has concluded that because the challenged legislation is politically neutral, nondiscriminatory, and “regulatory” at least in the same sense that the write-in ban in Burdick was regulatory, the appropriate test is a balancing test, as was applied in Timmons, Burdick and Tashjian. The court hastens to add, however, that, on the facts of this case, application of a balancing test certainly does not foreshadow the final result. This case aptly illustrates the Supreme Court's observation in Timmons that “[n]o bright line separates permissible election-related regulation from unconstitutional infringements on First Amendment freedoms.” Timmons, at 359.

*20 “It is well settled that partisan political organizations enjoy freedom of association protected by the First and Fourteenth Amendments,” Eu v. San Francisco County Democratic Central Committee, 489 U.S. 214, 224 (1989). At the heart of this right of association is the constitutionally protected “freedom to join together in furtherance of common political beliefs,” Jones, at 574 (internal quotation omitted), which is “protected by the Fourteenth Amendment from infringement by any State.” LaFollette, at 121.

The protected activity which is involved in this case is not mere expression in the abstract sense. It is not mere debate, as important as that can be. Instead, it is the candidate selection phase of the political cycle—“the crucial juncture at which the appeal to common principles may be translated into concerted action, and hence to political power in the community.” Tashjian, at 216. See also, Jones at 581 (candidate selection process is “the basic function of a political party”).

Voters, as well as parties and their candidates, are also affected by Oklahoma's primary election scheme, whether it be plaintiffs' scheme or the state's. This is, to some extent, a result of the fact that “the rights of voters and the rights of candidates do not lend themselves to neat separation.” Bullock v. Carter, 405 U.S. 134, 143 (1972). See also, Cousins v. Wigoda, 419 U .S. 477, 488 (1975); Ripon Society v. National Republican Party, 525 F.2d 567, 585 D.C.Cir.1975); Nader at 845 (“An attempt to interfere with a party's ability to maintain itself is simultaneously an interference with the associational rights of its members.”)

These bedrock principles define the interests which are in play from the plaintiffs' vantage point.

With tenacity which the court cannot but admire, the LPO has struggled for more than twenty years in the courts and elsewhere in the public arena to gain a political foothold in Oklahoma. See, e.g., Artunoff v. Oklahoma State Election Board, 687 F.2d 1375 (10^th Cir.1982) (Libertarian Party challenge to ballot access rules); Libertarian Party of Oklahoma v. Oklahoma State Election Bd., 593 F.Supp. 118 (W.D.Okla.1984) (same); Atherton v. Ward, 22 F.Supp.2d 1265 (W.D.Okla.1998) (Libertarian Party challenge to registration rules). It is now the judgment of the party's leadership (a judgment which no court may second guess, see p. 34, above) that the LAO's political ascent in Oklahoma will be facilitated by inviting members of other parties to give LPO politics a try by voting in the LPO primary without abandoning their registered party affiliation.

The LAO's freedom to implement its chosen strategy has collided with Oklahoma's semi-closed primary format. Subject to constitutional limitations and other exceptions not relevant here, it falls to the states to regulate elections. See, e.g. U.S. Const. Art. 1 § 4, Art. II, § 1; Bush v. Gore, 531 U.S. 98, 104 (2000). Even as to primaries, the role of government is unavoidable. Burdick, at 433 (“Common sense, as well as constitutional law, compels the conclusion that government must play an active role in structuring elections”); The Federalist No. 59 (1788) (“[I]t will therefore not be denied that a discretionary power over elections ought to exist somewhere.”). However, as the LPO would be quick to point out, where statutory law is concerned, the state is the creature of the parties in power, e.g., Williams v. Rhodes, 393 U.S. 23, 31—32 (1968), and is thus a bit suspect as a disinterested rule maker. It is necessary, therefore, to carefully evaluate those justifications for the challenged legislation which the state may plausibly offer (the court having previously discarded the implausible ones).

*21 The right “to vote in any manner” and the right to associate for political purposes through the ballot are not absolute. Burdick at 433. Although the states are not permitted to “completely insulate the two-party system from minor parties' or independent candidates' competition and influence,” Timmons, at 367, they do have a strong interest in the stability of their political systems. Id. at 366. The state's interest in political stability has been described as “not only permissible, but compelling.” Storer at 736.

In Timmons, the Court had before it a challenge to Minnesota's anti-fusion rule, which prohibited a candidate from appearing on the ballot as the candidate of more than one party. Timmons, at 353—54. In a conclusion which resembles one of the LAO's arguments in this case, the Court of Appeals had held that the anti-fusion rule kept “the New Party from developing consensual political alliances and thus broadening the base of public participation in and support for its activities.” Id. at 360 (internal quotation marks omitted). The state argued that it had an interest in avoiding voter confusion and overcrowded ballots, preventing party splintering and disruptions of the two-party system, and being able to clearly identify the election winner. Id . at 364. The Supreme Court held that the party's First and Fourteenth Amendment interests were outweighed by the state's interest in enacting “reasonable election regulations that may, in practice, favor the traditional two-party system.” Id. at 367. In dissent, Justice Stevens argued that the “true basis for the Court's holding [was] the interest in preserving the two-party system.” Id. at 377. The majority saw no reason to quibble with this critique, and in fact embraced the proposition that, for the sake of tempering “the destabilizing effects of party-splintering and excessive factionalism,” the state may nurture “a healthy two-party system” as long as there are no “unreasonably exclusionary restrictions.” Id. at 367. It is thus clear that legislation which favors the two-party system is not, for that reason alone, on thin constitutional ice.

On the subject of the integrity of the electoral process, Jones teaches that state laws which may cause parties to skew their positions from those which they might otherwise take are constitutionally suspect. Jones, at 582. In Jones, the law in question was the blanket primary law which permitted any voter to vote for any candidate in any party's primary. This scheme was unsuccessfully defended by the state on the basis, among others, that the blanket primary forced the parties to moderate their positions, a justification which was emphatically rejected as “a stark repudiation of freedom of political association.” Id. In the case at bar, the primary election format which plaintiffs assert is constitutionally required would have a quality which was identified as constitutionally suspect in Jones: the parties whose voters would be invited by the LAO into its primary would be induced, in an effort to retain voters who might be tempted to taste Libertarian politics without changing their registration, to skew their positions from those which they “would choose if left to their own devices.” Id. Of course, considerations which were influential in Jones must be carefully assessed for applicability to this case, because, in Jones, the open primaries were forced rather than optional. Hence, in Jones, the rules were applied to all parties whether they liked the rules or not, whereas, in the case at bar, the rule contended for by the plaintiffs would be applied to the LAO by consent and to the other parties without their consent. On this point, Tashjian becomes instructive. In Tashjian, the Court was at pains to distinguish between rules which “prevent[ed] the disruption of the political parties from without” and rules which “prevent[ed] the parties from taking internal steps affecting their own process for the selection of candidates.” Tashjian at 224 (emphasis added). The primary election format sought by plaintiffs in the case at bar is a hybrid—it would “disrupt” other parties “from without” but would amount to an “internal step” for the LAO. The Oklahoma legislature has plainly laid down a bright-line rule as to the significance of registered party affiliation. Although this rule is at odds with the plaintiffs' preferred primary election format, the court concludes that Oklahoma's approach to primary elections embodies choices which the Oklahoma legislature was free to make without violating the First and Fourteenth Amendments.

*22 It is clear that the “important regulatory interests,” Burdick at 434, which may outweigh associational rights in the election law context may encompass matters of structure, stability and party integrity, as opposed to justifications, which this court has rejected, based on administrative convenience and avoidance of voter confusion. See, e.g., Timmons at 366—67; Burdick at 433—34 and 438; Storer at 736. In the case at bar, unlike Jones, granting the relief sought by plaintiffs would impose an unwanted rule on parties other than the LAO. The state is entitled, at least in the regulatory sense which has been countenanced by the Supreme Court, to balance the political interests which are in play at the primary election stage of the election cycle by providing for a semi-closed primary. The judgments that are implicit in this format are judgments that the legislature was free to make, and it is clear under the relevant Supreme Court authorities that it is no answer to argue that the semi-closed primary format simply shores up the two-party system and values stability over the right of political association—especially where, as here, a voter who would like to try LAO politics may change his registered party affiliation only a few weeks before the primary election. (Cf. Rosario, where the Supreme Court declined to void the eight-to eleven-month cutoff for changes of registration for voting in New York's closed primary). Oklahoma's rules do not place burdensome requirements on the Republican and Democratic plaintiffs who, as would-be crossover voters, assert a constitutional right to participate in the primary of a party to which they do not belong (and, in the case of plaintiff Floyd Turner, a party which he emphatically refuses to join). Jones “stands, at least in part, for the proposition that the party's right to define its membership becomes meaningless if it cannot limit membership to those individuals who are willing to identify themselves with the party and its message.” Borden, Primary Elections, 38 Hard. J. Legis. 263, 277 (2001).

Timmons, in which the Court upheld Minnesota's anti-fusion rule, is again instructive. Although the two situations are not entirely congruent,⁸ the fact that a state may bar one party from poaching another party's candidates without violating the First and Fourteenth Amendments plausibly suggests that the state may bar a party from poaching another party's primary election voters. The Court, in Timmons, answered the New Party's complaint by noting that the New Party was free to attempt to persuade the candidate to switch parties. Id. at 360. That practical remedy is equally available here, and would be equally efficacious in eliminating any constitutional violation which might otherwise appear. State-enforced party loyalty, or disincentives for disloyalty, may seem suspect, but they have received the sanction of the Supreme Court in cases in addition to Storer. Cf. Nader v. Schaffer, 417 F.Supp. 837, 847 (D.Conn.1976), aff'd 429 U.S. 989 (1976): “[A] state legitimately may condition one's participation in a party's nominating process on some showing of loyalty to that party,” and Lippit v. Cipollone, 404 U.S. 1032 (1972), affirming, without opinion, a decision upholding a statute banning primary election candidacies of those who had voted in another party's primary within the preceding four years.

*23 The challenged provisions amount to “politically neutral regulations that have the effect of channeling expressive activity at the polls.” Burdick at 438. Because the court has concluded that this legislation falls within the limits established by the First and Fourteenth Amendments, the court respectfully disagrees with the conclusion reached in Cool Moose, which was decided by the First Circuit with the benefit of Supreme Court's decision in Tashjian but without the benefit of the Court's decision in Jones. The court in Cool Moose applied a balancing test, an approach with which this court agrees, as is discussed in part IV(A)(2), above. The First Circuit's comments in applying the balancing test may go far to explain the result reached by the court. The court noted arguments which could have been, but were not, made by the state in support of the Rhode Island closed primary. Id. at 86. The court went on to say that the state misconceived the nature of the Cool Moose Party's challenge, and expressly declined to bolster the state's feeble arguments on appeal. Id. at 86—88 (“We will not invoke justifications out of the whole cloth on appeal.”). The court was left with “raiding” and even weaker administrative concerns as the only justifications offered to sustain the legislation, and those justifications were as unpersuasive with the First Circuit as they are with this court. None of the party integrity concerns which were decisive a year later in Jones were discussed.

Accordingly, balancing the interests which are at stake here, and recognizing the state's prerogatives in regulating elections even where that regulation has the effect of reinforcing party identity and shoring up the two-party system, the court concludes that the plaintiffs have not made their case, factually or legally, for imposition of a party-option open primary system in Oklahoma.

The court commends the parties and their counsel for their outstanding presentations, both in writing and at trial. Falling, as it does, in the void between Jones on one hand and Tashjian on the other, this case invites appellate review of the court's resolution of the difficult issues presented. By separate document, judgment will be entered in favor of the defendants.