Margarita LÓPEZ TORRES, Steven Banks, C. Alfred Santillo, John J. Macron, Lili Ann Motta, John W. Carroll, Philip C. Segal, Susan Loeb, David J. Lansner, and Common Cause/NY, Plaintiffs,
v.
NEW YORK STATE BOARD OF ELECTIONS; Neil W. Kelleher, Carol Berman, Helena Moses Donohue, and Evelyn J. Aquila, in their official capacities as Commissioners of the New York State Board of Elections, Defendants,
New York County Democratic Committee, New York Republican State Committee, Association of Justices of the Supreme Court of the State of New York, Association of Justices of the Supreme Court of the City of New York, and Justice David Demarest, Individually, and as President of the State Association, Defendant-Intervenors,
Attorney General of the State of New York, Statutory Intervenor.
No. 04 CV 1129(JG).
United States District Court, E.D. New York.
January 27, 2006.*213 Brennan Center for Justice at NYU School of Law, New York, NY, By: Frederick A.O. Schwarz, Jr., Deborah Goldberg, Jeremy Creelan, Adam H. Morse, for Plaintiff.
Arnold & Porter LLP, New York, NY, By: Kent A. Yalowitz, Anand Agneshwar, S. Jeanine Conley, Angela Givens, for Plaintiff.
Akin Gump Strauss, Hauer & Feld LLP, New York, NY, By: Steven M. Pesner, Andrew J. Rossman, James P. Chou, *214 James E. d'Auguste, Vincenzo A. DeLeo, Jamison A. Diehl, for the New York County Democratic Committee.
Arthur W. Greig, New York, NY, for the New York County Democratic Committee.
Robert Alan Muir, Associates LLP, Brooklyn, NY, By: Robert Allen Muir, for New York State Republican Committee.
Special Counsel to State, Board of Elections, Albany, NY, By: Todd D. Valentine, for State Board of Elections.
Stroock & Stroock & Lavan LLP, New York, NY, By: Joseph L. Forstadt, Ernst R. Rosenberger, Kevin J. Curnin, David Sifre, Mary A. Gorman, for Associations of New York State Supreme Court Justices In the City and State of New York, Honorable David Demarest.
MEMORANDUM AND ORDER INCLUDING PRELIMINARY INJUNCTION
GLEESON, District Judge:
The plaintiffs in this case claim that New York State's electoral process for the office of Supreme Court Justice violates the First and Fourteenth Amendments to the United States Constitution. Specifically, they claim that the system both deprives voters of the right to choose their parties' judicial candidates and imposes insurmountable burdens on challenger candidates who seek a major party nomination without the support of local Democratic or Republican Party leaders.
The plaintiffs seek a declaration that New York's judicial convention system is unconstitutional and a preliminary injunction directing the New York State legislature to create a new system. In the meantime, and for as long as the legislature fails to do so, the plaintiffs request that this Court direct that Supreme Court Justices be nominated through direct primary elections.
In late 2004, a hearing was held on the plaintiffs' motion for a preliminary injunction. The hearing spanned 13 days; 24 witnesses testified, and more than 10,000 pages of documentary exhibits were received into evidence. Oral argument was held on November 18, 2004. The parties have submitted 494 pages of proposed factfindings and legal conclusions.
As discussed below, the plaintiffs have demonstrated convincingly that local major party leaders not the voters or the delegates to the judicial nominating conventions control who becomes a Supreme Court Justice and when. The highly unusual processes by which that extremely important office is filled perpetuate that control, and deprive the voters of any meaningful role. The result is an opaque, undemocratic selection procedure that violates the rights of the voters and the rights of candidates who lack the backing of the local party leaders.
Accordingly, the motion for a preliminary injunction is granted. Specifically, the defendants are enjoined from enforcing New York Election Law § 6-106, and from using the existing procedures set forth in New York Election Law § 6-124 for major party nominations for the office of Supreme Court Justice. Until the New York legislature enacts another electoral scheme, such nominations shall be made by primary election. The petitioning requirements that will attend those primary elections shall be set forth in a subsequent order, after the parties have had the opportunity to be heard, afforded by a schedule set forth below.
FACTUAL BACKGROUND
A. The Uniqueness of New York's Electoral Process for the Office of Supreme Court Justice
The Supreme Court of the State of New York is not the state's highest court. *215 Rather, it is the trial court of general jurisdiction. See N.Y. Const. Art. VI, § 7. Justices of the Supreme Court hear civil and criminal cases across the state. They also serve, when appointed to do so by the Governor, in the four Appellate Divisions, which hear appeals as of right from trial court decisions and from administrative agencies. N.Y. Const. Art. VI, § 4. The Constitution of the State of New York provides, in Article VI, § 6(c), that "[t]he justices of the supreme court shall be chosen by the electors of the judicial district in which they are to serve." The same provision fixes the term of office at 14 years.
The United States Constitution does not prescribe any particular method by which the states must choose their judges. The states have established a wide array of methods, including appointment, nonpartisan election, partisan election, convention, retention election and various combinations of those methods. New York is one of 39 states that elect some or all of their judges of trial courts of general jurisdiction. Eighteen of those states select all of their general jurisdiction trial judges through nonpartisan elections that involve an initial election on primary day and then, if necessary, a run-off between the top candidates on the day of the general election. No party labels appear with the candidates' names on the ballots. Nine of the 39 states, including New York, select all of their general jurisdiction trial judges through partisan elections.[1] Six of the 39 states use an initial merit appointment system, followed by a retention election at or near the end of the judge's term. In a retention election, the judge faces no opponent; rather, the voters determine whether she should be retained for an additional term. Finally, of the 39 states that elect their judges, six employ various combinations of the three selection methods already identified.
New York is unique in its use of a convention system to select nominees for election to its trial court of general jurisdiction (i.e., the Supreme Court). All major party nominations for Supreme Court Justice are required by law to be made by judicial district nominating conventions. Specifically, Section 6-106 of the New York Election Law reads as follows: "Party nominations for the office of justice of the supreme court shall be made by the judicial district convention."[2]
Of the 33 states that elect some or all of their general jurisdiction trial court judges in contestable elections (i.e., not including the six states that use retention elections alone), all but one give challenger candidates who lack the backing of party leaders an opportunity to be placed directly on a primary election ballot by filing a notice, gathering a reasonable number of signatures, paying a filing fee, or fulfilling some combination of these requirements. The one exception is New York.
The convention system at the heart of this case not only distinguishes New York from every other state, it also distinguishes *216 the office of Supreme Court Justice in New York from every other elective judicial office in the state. All other elected judges in New York State are nominated in a direct primary election, rather than in a judicial convention. And though judges of the Court of Claims, County Court, Surrogate's Court, Family Court, Civil Court, and Criminal Court may be appointed to serve as "Acting Supreme Court Justices," exercising the full powers, duties and jurisdiction of a Supreme Court Justice, none of these inferior court judges is selected by a process using a convention system for nominations.
Indeed, with respect to all elective offices in the state except Supreme Court Justice, New York allows candidates to petition onto a primary ballot by gathering signatures among the voters.[3] By contrast, New York Election Law § 6-106 prescribes the judicial nominating convention as the sole means of obtaining a nomination for the office of Supreme Court Justice. The absence of any alternate routes onto the ballot as a major party candidate, and the absence of primary elections decided by voters, makes Supreme Court Justice unique among New York State's elective offices.
The electoral process for Supreme Court Justice is unique not merely because it affords but a single path to the nomination, i.e., the judicial district nominating convention, but also because the path is exceedingly difficult to navigate, as described below.
B. The Mechanics of the Process
The electoral process for Supreme Court Justice has three official steps: (1) the election of judicial district convention delegates and alternates; (2) the judicial district convention; and (3) the general election. However, the true nature of the process is far more complex than a list of its formal phases suggests. The facts below describe the Supreme Court Justice selection process as I find it to be in actual operation.
First, the selection of delegates and alternate delegates to the judicial district nominating conventions is dominated by party leadership. The unique, and uniquely burdensome, features of that process virtually guarantee that result. The argument advanced by the New York State Republican Committee that a candidate for the office of Supreme Court Justice who lacks party leader support can clear all the hurdles necessary to elect supportive delegates to the convention is unsupportable.
Second, the alternative argument (advanced by the other defendants) that challenger candidates need not elect their own *217 supportive delegates because they can lobby the delegates elected by their party's machinery is just as unsupportable. The structural and practical impediments are insurmountable.
Third, the nominating conventions themselves are certainly not the places where important decisions get made. Rather, they are brief, rote, formal stamps of approval given to decisions made elsewhere.
Fourth, the general elections, ostensibly the culmination of the democratic process, play almost as minor a role in the selection of Supreme Court Justices as do the conventions. In most places, the nominees of a single party (either Democratic or Republican) win all or virtually all of the time. In others, cross-nominations by those parties deprive the general election of any contest. Contested elections for Justice of the Supreme Court are the exception, not the rule.
If, as the foregoing summary suggests, neither the voters nor the delegates to the judicial nominating conventions are evaluating candidates for Supreme Court Justice and selecting nominees for that office, then who is? The fifth section below discusses how the county leaders,[4] in tandem with district leaders,[5] control the process. In effect, they decide who becomes a Supreme Court Justice in New York courtrooms.
1. The Party Leaders' Domination of the Election of Convention Delegates and Alternates
New York is divided into 12 judicial districts. Each includes numerous Assembly Districts, in whole or in part. Judicial districts generally include more than one county, but at present, no counties are divided between two or more judicial districts. A map depicting the 12 judicial districts, the 150 Assembly Districts and the 62 counties is attached hereto as Appendix A.[6]
New York law does not require that Supreme Court Justices reside, either before or after election, in the judicial district in which they are elected. Also, a justice may be assigned to sit in a district other than the one from which he or she was elected. The only legal requirements for office are that the candidate be a citizen of the United States, reside within New York State, and have been admitted to practice law in the State for at least ten years.
*218 The number of authorized Supreme Court Justice positions in each judicial district is set forth below:
Judicial Authorized Duly Elected District Supreme Court Justices First 38 Second 52 Third 15 Fourth 13 Fifth 17 Sixth 10 Seventh 18 Eighth 26 Ninth 26 Tenth 47 Eleventh 38 Twelfth 24
The number of elections each year in each judicial district varies, depending on the expiration of sitting justices' terms and other factors. During the period from 1990 through 2002, 52 justices were elected (or re-elected) in the First District.
The judicial nominating conventions are governed by § 6-124 of the Election Law, which is set forth in its entirety in the margin.[7] Candidates for the office of Supreme Court Justice are nominated at large by delegates from within the particular judicial district. However, those delegates are not elected at large. Rather, they are elected in separate races in each Assembly District ("AD"), or part thereof, within the judicial district. This odd feature of the judicial nominating convention process contributes significantly to the heavy burden it places on those who seek major party nominations for Supreme Court Justice without the support of the party's district leaders and county leaders. Such challenger candidates face the prospect of as many delegate races as there *219 are ADs in the judicial district at least nine and as many as 24.
The sheer numbers of delegates and alternate delegates[8] add to the burden. Section 6-124 allows the political parties to determine the number of delegates and alternate delegates at their judicial conventions, so long as they are allocated among the ADs substantially in proportion to the votes cast in that AD for the party's candidate (and on the party's ballot line) in the previous gubernatorial election. The number of alternates from any district may not exceed the number of delegates. The Democratic Party currently allots each AD one delegate and one alternate delegate, plus an additional delegate and alternate for each 2,500 votes cast in the prior election. The Republican Party follows the same formula for eight of the twelve judicial districts, but uses 4,000 votes (rather than 2,500) as the divisor in the Third and Fourth Judicial Districts, and 5,000 in the Seventh and Eighth Judicial Districts.
In 2004, these formulas produced the following numbers of delegates and alternates in each judicial district:
Democratic Republican
Number of Assembly
Judicial Total Number Total Number Total Number Total Number Districts within the
District of Delegates of Alternates of Delegates of Alternates Judicial District
3 41 41 50 50 11
4 32 32 54 54 10
5 35 35 81 81 12
Non- 6 24 24 66 66 9
NYC 7 43 43 48 48 11
8 62 62 62 62 13
9 76 76 139 139 17
10 102 102 185 185 21
1 93 93 57 57 12
2 124 124 98 98 24
NYC 11 87 87 77 77 18
12 63 63 32 32 11
Thus, frequently as many as six or seven delegates are elected from each AD to attend the nominating convention for Supreme Court candidates. These high numbers of delegates constitute another distinctive feature of New York's judicial *220 nominating conventions. Only two state committee members are elected from each AD to the statewide conventions in June that designate candidates for Governor, U.S. Senator, and other statewide offices. The presidential primaries elect approximately five or six delegates per congressional district. While that is approximately the same number of delegates as the average AD in the Second Judicial District sends to the district's judicial nominating convention, that does not account for an almost equal number of alternate judicial delegates, and in any event congressional districts are typically more than three times as large as ADs. By delegating to the major parties the right to determine the number of delegates and alternate delegates, the Election Law has enabled those parties to make a challenger candidate's effort to elect a majority of delegates more difficult. In the Second Judicial District in 2004, for example, such a challenge would require running approximately 250 candidates across 24 ADs and two counties. The sheer number of people a challenger must recruit to run for the office of delegate and alternate delegate is a significant burden in itself.
Moreover, unlike presidential delegates, New York's judicial delegates cannot signify on the primary ballot an allegiance to a specific candidate. As a result, even if a Supreme Court challenger candidate were to petition successfully to place supportive delegate candidates on the ballot in each AD within the judicial district, he or she would then be required to inform the voters through campaign literature or advertising of those delegates' allegiance. And because there would not be just one such race, but at least nine and up to 24, the challenger would have to replicate that public education campaign for each distinct slate of delegates in each AD throughout the judicial district.
In this regard, I reject defendants' contention that a challenger would need to run slates only in enough hand-picked ADs to elect a majority of convention delegates. Even if that were true, the challenger's burden would be severe, and likely insurmountable. But it is not true; a challenger candidate would need to field delegates in all or virtually all the ADs to have a realistic chance of prevailing in enough races to obtain a majority of delegate support at the convention. This is true not only because a challenger candidate is not likely to win every delegate race under normal circumstances, but also because the prospect of competing against the party leaders elevates the degree of difficulty. A challenger candidate's strategy would have to include a plan to occupy the organization's base of support. In testimony I credit, Henry Berger stated that such a candidate must field slates even in ADs she is unlikely to win in order to occupy those loyal to the party leaders, and thus keep them from working against her in other ADs, where success is more likely.
In addition to the hurdles established by the electoral scheme and the delegate allocation process, the petitioning rules are onerous. To place an individual (or slate of individuals) on the ballot as a Democratic or Republican candidate for delegate or alternate delegate, 500 valid signatures of party members must be gathered in each AD.[9] As a practical matter, in light of the brief period of time in which petitions may circulate (37 days) and of the rules regarding who may sign them and who may witness the signatures, 1000 to 1500 signatures *221 per AD are necessary to ensure that legal challenges will be fended off. Thus, a challenger candidate for the Supreme Court in Brooklyn or Staten Island would need to gather 24,000 to 36,000 signatures drawn equally from the 24 ADs in the district. And since party members may lawfully sign just one designating petition for judicial delegates, the pool of eligible signers shrinks each time a petition is signed.
These features of New York's electoral system render any effort by a challenger candidate to field slates of supportive delegates and alternates virtually impossible. Indeed, it is considerably easier for an aspiring Supreme Court Justice to petition herself onto the ballot for the office of Mayor of New York City (7,500 signatures 15,000 to 22,500 as a practical matter from anywhere in the city, see N.Y. Elec. Law § 6-136(a)) than it is to petition onto the ballot slates of delegates to a judicial nominating convention. In any event, I reject the Republican State Committee of New York's insistence that it is not unduly burdensome for a challenger candidate for Supreme Court Justice to run her own slates of delegates to the judicial nominating conventions. See, e.g., Tr. 151-55 (cross-examination of Berger); Tr. 465 (cross-examination of Carroll); Tr. 2434 (oral argument). I find it significant that the other defendants apparently reject it as well. Justice Helen Freedman, a witness called by the Associations of Supreme Court Justices of the City and State of New York, testified that it would have been prohibitively expensive to field her own petitions for convention delegates, and there was "just no way" she could have run her own slates against the "county endorsed candidates." Tr. 1475-76; 1782-84. Douglas Kellner, who was called by the New York County Democratic Committee, went further, testifying that any notion that challenger candidates might run their own slates of delegates "twists on its head the system that the Legislature set up." Tr. 1567-68.
By contrast, the petitioning process is rather easy for the major party organizations. Slates of judicial delegates are included on omnibus petitions on which signatures are obtained for other candidates seeking other offices. The county and district leaders can easily mobilize the resources necessary to conduct the petition drives throughout the judicial districts because they are collecting signatures in all of those ADs anyway, for a variety of other party and public offices. Thus, when deciding who to run for the array of offices set for election, the party leaders also decide who to run as judicial delegates and alternates.
As a result, contests for the office of delegate and alternate delegate are rare. The party leaders use the massive apparatus of their respective major party to file the necessary petitions across the judicial district. Their slates of candidates for delegates and alternate delegates are usually selected without opposition. And when only one slate of candidates per AD files petitions, those candidates are "deemed elected" and do not appear on the September primary ballot. As a result, most voters in most years across the state do not even see the names of delegates or alternate delegates on the primary ballot when they vote. In New York City, which has more contested delegate elections than any other part of the state, only 12.7% of the judicial delegate seats filled by Democrats and Republicans were contested at the September elections from 1999 through 2003. For Democrats, the dominant party within New York City, the figure was only 19.3% during that period. In short, the extent of the party leaders' control over who becomes a delegate or alternate delegate varies by judicial district, but the fact that the voters play virtually no role in *222 selecting the majority of delegates and alternates is constant throughout the state.
Some of the defendants make much of the infrequent exceptions to that general rule. Most have occurred in the First District,[10] which in this respect and all others the defendants seek to showcase above the other 11 districts. But these exceptions are not helpful to the defendants for two reasons. First, their infrequency proves the general rule that contested elections for the positions of delegate and alternate delegate at judicial nominating conventions do not occur. A summary of the available evidence on the subject is set forth in the margin.[11] Second, when such contests do occur, they are almost always the result of localized disputes between entrenched factions of the party, not of challenger candidates' efforts to obtain support of party members. More specifically, the overwhelming majority of these disputes involve rival Democratic clubs in Manhattan. These clubs' members, who must pay dues, represent approximately 1.3% of the registered Democrats in Manhattan. That there are occasional fights for control at the upper levels of the political structure of an AD or a county does not mean the delegate selection process is open to a challenger candidate who seeks her party's nomination. To the contrary, it is yet another window into how closed the process is to the ordinary voters. Only those few who take the extra step of joining a *223 club in the borough of Manhattan can actually have a role, albeit a minor one, in the selection of judicial convention delegates.
The defendants place significant weight on the fact that William Allen and Alan Flacks, a district leader and a community activist, respectively, in the First District, ran successfully for delegate in opposition to the delegate slate sponsored by local party leaders or a political club. But the argument conflates the virtually impossible task of running slates of delegates and alternates across a judicial district with running for delegate in a particular AD. Plaintiff Margarita López Torres, for example, did not seek the office of judicial delegate, or to become a gadfly at a convention otherwise populated by delegates placed there by the party organization. She sought the office of Supreme Court Justice. If her ability to seek the Democratic Party nomination for that office is extinguished by a process that violates the First Amendment, it is no answer to say that she could run for the office of judicial delegate.[12]
In sum, the model advocated by the New York State Republican Committee i.e., if a challenger candidate wants her party's nomination for the office of Supreme Court Justice, she need only go out and elect enough delegates who will vote for her at the convention defies reality. It is not feasible.
2. The Structural and Practical Impediments to the Lobbying of Delegates
The other defendants posit a different model. A challenger candidate need not get her own slates of delegates and alternates elected because all she needs to do is lobby the ones the party leaders get elected. According to this model, New York's concededly unique judicial nominating convention system was created to "allow an elected body of informed delegates to consult, deliberate and choose Supreme Court nominees who best reflect the interests and values of the delegates' constituents." Def. Proposed Findings of Fact ¶ 31.
I reject this contention as well. It does not describe how the system in fact works. In reality, with very few exceptions, district leaders and county leaders select the delegates and alternate delegates, who, without consultation or deliberation, rubber stamp the county leaders' choices (or "package" of choices) for Supreme Court Justice. Most delegates have strong ties to the district leaders who select them, and sometimes work for them as well.
The defendants contend that there is no evidence that delegates are explicitly directed how to vote. That is incorrect,[13] but the more important fact is there is no need for an express directive. Most of the time that is because there is no choice the party leaders' candidate or package of candidates is all there is to vote on. On *224 the infrequent occasion that there is a competing nomination, the political dynamic ensures that the district leaders and their delegates from each AD virtually always support the candidates backed by the party leaders. They do not and will not jeopardize their ongoing, multi-faceted relationships with other district leaders and the county leaders over a candidate for Supreme Court Justice.
The power of this political dynamic was demonstrated in the Second District in 2002. For the seventh consecutive year, State Senator Martin Connor was asked to convene the judicial nominating convention. He knew the delegates well, and they respected him and his views. Senator Connor was uncomfortable presiding at the 2002 convention, however, because Clarence Norman (then the Democratic Party's county leader) and the other Democratic Party leaders had decided to make an unqualified person a Supreme Court Justice. In Senator Connor's view, it was a horrible choice; the person was already a bad judge on a lower court, and was obnoxious and rude to lawyers to boot. Senator Connor was outraged, convinced that the nomination was bad for the Supreme Court bench. But he said nothing. He dared not tell the delegates that the county leader's choice would be a disaster. Instead, after he convened the nominating convention, he declined to continue as its chair and ducked out, saying nothing. The candidate he objected to was nominated and became a Supreme Court Justice. Why was he afraid to voice objection? Because he wanted the county leader to support his choice in a State Senate race.
If the minority leader of the State Senate, who testified that he "had five offices and 150 staff members," Tr. 2208, is admittedly unable to "speak truth to power" despite being outraged by the county leader's decision to foist a horrible candidate on a convention of delegates, Tr. 2263, it is hardly likely that mere delegates will refuse to do the county leader's bidding in the ordinary course of events. Explicit voting commands are not needed to keep the delegates in line.[14]
In short, the party leaders select the delegates and, as discussed more fully below, they also select the judicial candidates for the delegates to nominate. But the problem with this second model is not limited to the fact that the delegates and alternate delegates do not actually perform the deliberative, consultative, informed roles the defendants ascribe to them. Even if the delegates were amenable to such tasks and attempted to perform them, the time frame available to a challenger candidate is so unreasonably brief that it would doom them to failure. Moreover, in two key districts, that time frame has been deliberately compressed even further by the county leaders.
New York law requires that judicial nominating conventions "be held not earlier than the Tuesday following the third Monday in September preceding the general election and not later than the fourth Monday in September preceding such election." N.Y. Elec. Law § 6-158(5). Thus, delegates are selected at the parties' primaries in the first two weeks of September, and the judicial convention occurs in the third week of September. That leaves virtually no time for lobbying dozens, if not hundreds, of delegates and alternates.
Superimposed upon the statutory time line in the First and Second Judicial Districts is the further constraint of the judicial screening panels. Candidates in those *225 districts must be "reported out" of the screening panel, i.e., approved as qualified, to be eligible for nomination. Since the county leaders ensure that the screening panels do not report out candidates until after Labor Day, any lobbying of prospective delegates prior to that time could turn out to be a waste of time. And given the extremely large numbers of delegates and alternates in those districts, the time frame established by state law and party rules simply does not allow a meaningful opportunity for candidates for the Supreme Court to argue their case to the delegates.[15]
Defendants make much of the fact that, in the First and Second Districts, candidates for Supreme Court Justice mail their qualifications and other materials to the delegates or even invite them to receptions. But although those contacts are certainly part of the process, it distorts reality to suggest that they reflect efforts to persuade delegates how to exercise their votes at the convention. In fact, the party leaders decide who the delegates will vote for at the conventions. In testimony I credit, Henry Berger stated that the contacts between candidates and delegates are simply part of a ritual that confirms those party leaders' decisions. The contacts also confirm to the delegates that they are part of the process, which in turn makes it easier for the district leaders to keep the delegates in line so the nomination process runs smoothly.
Indeed, I credit Berger's testimony in its entirety. To the extent it was contradicted by the defendants' witnesses, I reject the latter testimony. I specifically reject the description of the process described by Arthur Schiff, who has been a Democratic district leader in Manhattan since 1993 and a delegate in the First District's judicial nominating convention about ten times. Called by the New York County Democratic Committee to refute, among other things, Berger's testimony that the delegates vote in accordance with the party leaders' wishes, Schiff accurately described the narrow time frame imposed by state law and major party rules. There "is not a great deal of lobbying done [by Supreme Court candidates] prior to . . . the screening panel['s]" report, Tr. 1298,; rather, they "do their politicking" among the delegates in a brief period that usually lasts only two weeks and "certainly not more than three weeks." Tr. 1299. However, a lot happens in that narrow window of time, according to Schiff. First, all the candidates are "allowed or persuaded," Tr. 1292, to go out and make their case to the nearly 200 delegates and alternates. After lobbying all those people, candidates then go to the county leader, and try to convince him they have the most support. Where more than one vacancy exists, the county leader "tries to act as a kind of facilitator" in this process, Tr. 1293, ensuring that those candidates who have the most support become nominated, and that fairness is served as well. Also during this compressed time period, and before the convention begins, candidates who fail to garner majority support achieve "an understanding" of that fact and an acceptance of it, which is why they almost invariably withdraw their candidacies before or at the convention. Tr. 1294-95.
To this flurry of alleged activity, defendants' counsel add yet another task after lobbying the delegates and alternates, the candidates "engage in logrolling," presumably *226 using their own minority bloc of supporters to obtain support from other such blocs in multiple-vacancy situations. See Def. Reply Findings at ¶ 19 ("As explained in . . . this part of Mr. Schiff's testimony, the final slate is determined through a dynamic and fluid process whereby judicial candidates actively lobby delegates and engage in logrolling in the two to three-week time period before the convention.") In fact, Schiff's testimony made no mention of logrolling, and there is no persuasive evidence that it occurs among judicial delegates. But the essential point is larger than that: there is no "dynamic and fluid process," and there would be insufficient time for one even if the participants were so inclined.[16]
Concededly, there is not a rich history of challenger candidates even attempting to make their case to the delegates selected by the party leaders. But plaintiff Margarita López Torres has tried it, and her experience in the Second District does not bode well for other such candidates, especially those who deign to run slates of judicial delegates against those fielded by the party leadership. Beginning in March of 2003, López Torres wrote repeatedly to the Kings County Democratic County Committee to learn three basic things: (1) the date, time and place of the convention; (2) the names of the delegates, so she could lobby them; and (3) whether she could address the delegates at the convention. She did not hear from its Executive Director, Jeffrey C. Feldman, until September 4, 2003, after she once again requested the information.
Feldman's response is difficult to reconcile with the defendants' gauzy characterizations of a democratic process open to all party members who seek the office of Supreme Court Justice. Def. Reply Findings ¶ 26. He began by mocking the request for a list of delegates to lobby: "I erroneously believed that a learned jurist, such as yourself, would be well aware that Delegates and Alternate Delegates to the Democratic Judicial Convention stand for independent election in the Primary Election, yet to be held." López Torres Decl., Ex. 6.[17] Thus, no such list existed "anywhere in the world," Feldman helpfully added. Id. *227 As for López Torres's inquiry about addressing the convention, Feldman wrote as follows: "I suffer from the innocent belief that the floor of the Convention is open, only, to elected Delegates and their successors. I am not aware of any Convention in my thirty (30) years of attendance, which permitted a non-accredited member to be accorded the privilege of the floor. . . ." Id. In closing, Feldman "note[d] for the record that" López-Torres's fax machine was "in violation of Federal Communications Commission regulations" and admonished her to bring it into compliance. Id. He then chastised her for mailing him so many (unanswered) letter requests, and spending "copious sums in postage from, presumably, your political committee." Id.
In sum, lobbying judicial convention delegates and alternates is not a realistic route to a nomination even for diligent candidates if they lack the support of party leaders. The defendants insist that they proved otherwise through the testimony of six sitting Supreme Court Justices. Specifically, defendants contend that these witnesses (four from the First District, one each from the Fourth and the Eighth) show that delegates can be successfully lobbied, and nominations obtained, without the support of party leaders. The defendants further contend that unless these sitting justices are branded as perjurers, the plaintiffs must lose. Def. Reply Findings at 2. The latter contention is both incorrect and extremely unfair to the testifying justices; the former unfairly characterizes their testimony.
First, the witnesses from the First District confirmed that, without the support of county leader Farrell, a candidate for Supreme Court Justice cannot succeed. Second, the defendants' characterization of their testimony obfuscates the critical distinction between delegates and district leaders. I find that to the extent it sheds meaningful light on the selection process, the testimony at issue supports the plaintiffs' contention that the district leaders (together with Farrell), not the delegates, select the Supreme Court Justice nominees.
After Justice Phyllis Gangel-Jacob failed to secure party leader support in 1991,[18] her spouse became a district leader. By 1993, Justice Gangel-Jacob became part of Farrell's "package" of judges, so she was nominated and elected. Her husband stepped down as district leader after she took office. In short, Justice Gangel-Jacob did not, as defendants contend, obtain her nomination over the opposition of the party leaders, or of Farrell himself.[19]
Neither did Justice Alice Schlesinger. In 1991 or 1992, she commenced her efforts *228 to become an elected Supreme Court Justice. In 1993, her husband became a district leader. By 1999, she and her husband had achieved Farrell's support, and she became a Supreme Court Justice.
Justice Helen Freedman's testimony, like that of her colleagues on the Supreme Court bench in the First District, reveals the central importance of obtaining the support of district leaders.[20] Her lobbying was not confined to the narrow window each year set aside for delegate lobbying. Rather, it also occurred during periods of the year when the delegates were not even identifiable.[21] She eventually became part of Farrell's package in 1988, a year in which he publicly threatened to retaliate against anyone who sought to interfere with three African-American members of his five-candidate package.
To be sure, Justice Freedman's efforts to obtain the support of party leaders included substantial contact with delegates. In the setting of the First District, where Democratic club activity is intense, efforts to appeal to the real decision-makers the district leaders and the county leader inevitably involve contacts with the delegates the party apparatus selects. And for reasons expressed above, certain ritualistic contacts with delegates as the nomination season is under way is a required part of the process. But Justice Freedman's testimony, which I credit, does not alter my findings that (1) neither the voters nor the delegates play a significant role in the nomination of Supreme Court Justices; and (2) an aspiring candidate cannot obtain a major party nomination without the backing of the party leaders, in the First District or elsewhere in the state.
Justice Joseph Sise's testimony is discussed more fully below. Suffice it to say here that lobbying delegates had almost nothing to do with his nomination. Similarly, Justice Robert Lunn secured his nomination in the Eighth District by appealing to the county leader of the Conservative Party in Monroe County, who, as discussed more fully below, played the decisive role in Republican nominations for Supreme Court Justice. Though Justice Lunn, like Justice Sise, campaigned among the voters themselves, he neither lobbied the delegates to the judicial convention in the manner claimed by the defendants nor needed to in order to become the Republican nominee.[22]
*229 3. The Nominating Conventions
The judicial nominating conventions themselves are perfunctory, superficial events. They do not determine candidates, but rather formally endorse determinations made elsewhere. The extremely high absentee rate reflects, among other things, the fact that delegates to the conventions know their votes do not really matter. Many choose not to participate in what is, ostensibly, the denouement of the judicial nomination process because it is widely known that the purpose of the convention is to rubber stamp the major party leaders' choices for Supreme Court Justice.
The 2001 Democratic Party convention in the Second District illustrates the disparity between what may appear to happen at a convention and what actually happens. The minutes of the convention and the testimony of its chairperson reveal that the following events occurred:
The convention was called to order by State Senator Martin Connor.
The conventioneers recited the Pledge of Allegiance.
Senator Connor read two letters one from the state chair to the convener (typically about eight pages long) and one from the state chair to the delegates.
Senator Connor announced that nominations were in order for the office of temporary chair.
People were recognized to make a nomination and a second for that office.
Further nominations for temporary chair were called for.
Persons were appointed as tellers.
A vote was called on the temporary chair nomination. The 77 delegates were polled one-by-one.[23]
The results of the election for temporary chair were declared. Senator Connor was elected.
The temporary chair's oath of office was administered.
Nominations for the office of temporary secretary were solicited.
A nomination and a second for that office were received.
Further nominations for that office were invited, and such nominations were closed.
Another vote was taken, this time for temporary secretary.
The election results were announced, and the temporary secretary subscribed to an oath of office.
A motion was made, and seconded, that the temporary officers be made permanent officers.
A third vote was held, and that motion carried.
The permanent chair (Senator Connor) and permanent secretary once again subscribed to an oath of office.
*230 Nominating and seconding speeches for the office of Supreme Court Justice were invited. The delegates were advised that time limits of two minutes and one minute, respectively, applied to such speeches.
Senator Connor announced how, pursuant to the New York Election Law, it would be lawful to have a vote viva voce if only one candidate were nominated per vacancy.
A group of six people were nominated to the office of Supreme Court Justice. There were six vacancies to be filled.
Someone else seconded those nominations.
Further nominations were called for.
No further nominations having been made, the chair declared that nominations were closed.
A vote was held, viva voce, on the six nominees.
The chair announced that all six had been elected.
A motion was made, and seconded, that certain people be nominated to a committee to fill vacancies.[24]
A vote was held on the committee to fill vacancies.
The results of that vote were announced.
Senator Connor declared the convention adjourned.
That looks like a lot of work. But the entire convention lasted only 20 minutes. Senator Connor testified, with obvious pride, that he did it even faster one year. He explained the seemingly impossible task as follows: "I read very fast, I did it [New York State] Senate style." Tr. 2230.
Although the 2001 Democratic convention in the Second District may have been among the speediest,[25] it was representative of all such conventions in making clear that the decisions of who becomes a Supreme Court Justice are only ratified at the conventions. They are made elsewhere. Not even the defendants contend otherwise.
4. The General Elections
The last stage in the process of selecting Supreme Court Justices is the general election in November. For the most part, the general election is a formality. In New York City (the First, Second, Eleventh and Twelfth Districts), the Democratic Party nominees are always elected. Outside the city, there are judicial districts (the Fourth and Sixth) where the Republican nominees are always elected and others (the Seventh and Ninth) where they usually are.
Indeed, based on data from 1990-2002, 47% of the races for Supreme Court Justice were not even contested, i.e., there was only one candidate from the Democratic or Republican Parties running. In the First District, 85% of the races were uncontested, as were 91% and 57% of the races in the Sixth and Eleventh Districts, respectfully. Eight of the 12 districts have uncontested races at least half the time.
Cross-endorsements are very common. Ninety-five percent of the winners in the Twelfth District during the 1990-2002 period were endorsed by both the Democratic and Republican Parties, as were 51% of the candidates in the Second District. In districts that are not dominated by a single *231 party, the Democratic Party and the Republican Party essentially divvy up the judgeships through cross-endorsements.[26]
The relatively few contested elections are not competitive. If a competitive election is defined as one in which the second-place vote-getter receives at least 80% of the winner's total, only 2% of the elections in New York City from 1990-2002 were competitive. In all, 76% of the general elections across the state during that period were either uncontested or uncompetitive.
Finally, candidates without either the Democratic or Republican nomination have no chance of being elected anywhere. Though defendants have emphasized that a challenger candidate can petition onto the general election ballot, no such independent candidate has ever succeeded.
5. The Democratic and Republican Party Leaders Select the Nominees
As discussed above, neither of the two models advanced by the defendants is accurate. The first an aspiring Supreme Court Justice can elect supportive delegates to the nominating convention is especially far-fetched. The labyrinthine, burdensome procedures preclude it, and further ensure that the entrenched party leaders can install as delegates persons who will do their bidding. For the latter reason, and for the additional reason that the available time frame is preclusively short, the second model also fails. Faced with an inability to get supportive delegates elected, it is cold comfort for a challenger candidate to be told not to worry, she can try to convince the delegates selected by the party leaders. The current system for selecting judges does not work that way and is not structured to work that way. As discussed further below, Margarita López Torres demonstrated in 2003 that indisputable qualifications for the job and immense popularity among the candidate's fellow party members are neither necessary nor sufficient to get the party's nomination. Something different is required: the imprimatur of the party leadership.
As a result, almost all Supreme Court Justice nominations in New York State are uncontested. There is no evidence of a single successful challenge to candidates backed by the party leaders.[27]
The evidence at the hearing did not focus equally on all 12 judicial districts. The most dominant subject by far was the First District, followed by the Second District. To a much lesser extent, conditions in some of the upstate (using that term as it is understood in Brooklyn) districts were addressed.
a. The First District
The First District's selection process is the best in the state because of its screening panel, which is discussed further below, and because of the intense involvement of Manhattan Democratic clubs in judicial politics. Those features of the *232 nomination process in the First District have resulted in a measure of quality control that is unmatched elsewhere in the state. Nevertheless, even in the First District, New York County's Democratic county leader controls the process. Farrell assembles a package of candidates for presentation to the convention, and the delegates approve it. Candidates who Farrell decides should be nominated get nominated.
Since Farrell became the county leader in 1981, no Democrat has ever been nominated for Supreme Court Justice in the First District over his objection. Farrell himself explained why, not in this case, but in his deposition in France v. Pataki:[28] "I have the votes to be able to kill someone in other words, I can't guarantee I can always make you, but I can surely block you. No one wants to get me angry, so they will not go against me until they have nothing to lose." Pl. Ex. 99(B) at 193 (emphasis added).
Farrell further testified at the same deposition as follows:
I basically think I would keep the nominating conventions, because and I am now speaking from a self-serving point of view, as opposed to whether it is best in terms of electing minorities. I take the position that it works best for me, because it gives me a better chance to control what goes on, and that, I think, is important in terms of doing some of the things we have done.
Pl. Ex. 99(A) at 123:16-25 (emphasis added).
Farrell possesses this degree of control because he directly controls a sizeable bloc of delegates himself and because he controls the district leaders who select (and control) the remaining delegates.[29] Those district leaders almost always follow the wishes of the county leader. Those seeking to influence the outcome of the process deal with the district leaders and the county leader, not the delegates.
Defendants acknowledge that Farrell appears to control the process, and claimed to do so under oath in the France case, but contend that he really does not. The factual support for this claim is thin. First, defendants contend when Farrell has said he controls the process, he really "mean[t] having an appearance of naming the winner." Tr. 1663 (emphasis added). But that's not what Farrell himself has said, repeatedly. He has made it clear that no one can become a Supreme Court Justice in the First District without his approval, and that he will punish those who seek to derail his package of candidates. In short, he has asserted that he exercises actual control, not just the appearance of control.
Second, the defendants have resorted to various metaphors to advance their claim that appearances can be deceiving. In the more than 20 years Farrell seems to have been making Supreme Court Justices, he has really been picking the winners of horse races after the races were over. Tr. 1663. Through the use of "smoke and mirrors," Farrell makes everyone think he actually picks Manhattan's Supreme Court Justices, but that's merely "what he wants *233 the perception to be." Tr. 1711. Though it looks for all the world to see like Farrell and his district leaders control the process with an iron fist, when the curtain is pulled back, Farrell is revealed to be nothing more than the Wizard of Oz. Tr. 1736-1740. The reality, defendants contend, is that Farrell's decisions reflect a grass-roots consensus, expressed by the delegates through the district leaders to him as the county leader, as to which candidates should become Supreme Court Justices.
Based on the record before me, I find that county leaders in general, and Farrell in particular, actually wield enormous and dispositive power in the process by which Justices of the Supreme Court are selected.[30] No one can get elected Supreme Court Justice in the First Department without Farrell's support. And since the heart of this case is the plaintiffs' claim that county leaders, along with district leaders, decide who becomes a Justice of the Supreme Court, and the plaintiffs have produced ample evidence to support that claim, the defendants' failure to call Farrell or any other county leader as a witness is striking. Even some of the witnesses who testified in defense of the process did not know how much control the county leader has over it.[31] If it were really true that the county leaders' apparent stranglehold over the process were just an illusion, one would think they or at least one of them would take the witness stand and say so. If the widely-held belief that the county leaders' control of the system is simply the result of their "smoke and mirrors" act, the defendants should have called as witnesses the architects of what they now claim is a deliberately-created popular misconception. Their only explanation for failing to do so that the county leaders may be called at a trial on the merits if a preliminary injunction is granted, see Tr. 2405 is no explanation at all.
b. The Second District
From the 1960s, when the Kings County Democratic Party was under the leadership of Meade Esposito, through the time of the evidentiary hearing in this case, when Clarence Norman was the county leader, the county leader in Kings County has selected the Supreme Court Justices in the Second District. No one "recommended" by that county's organization has failed to obtain the nomination. Candidates opposed by that organization do not stand a chance.
The record of financial contributions by candidates for Supreme Court Justice to political groups controlled by Norman has fostered not only the (accurate) perception that he, rather than the voters or delegates, controlled the selection of the justices, but the further perception that he used the wrong criteria in making his decisions.[32] For the purposes of this case, only the fact of the county leader's control matters, and it has been established overwhelmingly.
The experience of plaintiff Margarita López Torres places in clear relief the *234 control exercised by Norman and the district leaders who answered to him while he was county leader. It also reveals most of the flaws in the process by which Supreme Court Justices are selected in New York.
López Torres is a 1979 law school graduate who spent her first 13 years in the profession as a legal services attorney and an attorney for the City of New York. In 1992, she became a candidate for a county-wide seat on the Civil Court of the City of New York. She received the support of the Kings County Democratic County Committee, which assisted her onto an uncontested primary ballot, and of course she was elected at the general election in November 1992.
López Torres was off to a great start in her judicial career. The path to the office of Supreme Court Justice, at least in the First and Second Districts, typically runs through the Civil Court. Shortly after she was elected, however, López Torres lost her way. It began when she was told by Norman, the county leader, and Vito Lopez, her district leader, to hire a particular young attorney as her court attorney. The directive came in a November 4, 1992 letter to López Torres from Steven Cohn, a Brooklyn Democratic Party official. The letter stated as follows:
Congratulations on your election. Clarence [Norman] and Vito [Lopez] have asked me to refer this wonderful gentleman to you as your Law Secretary.
Please be so kind as to interview him and obtain the necessary paperwork for his employment.
Thank you.
Pl. Ex. 85. The letter attached a resume. Defendants contend that Cohn was merely suggesting a candidate to López Torres. The text of the letter, particularly the direction to obtain the paperwork necessary for the young man's employment, supports López Torres's view that it was stronger than a suggestion. At any rate, the context in which the letter was written cannot be ignored, and that context compels the conclusion (and I find) that the party leaders, having in their view placed López Torres on the bench, also felt entitled to place employees in her chambers. Their subsequent conduct, discussed below, provides further support for that finding.
The position of court attorney is critical to the Civil Court judge; the lawyer who occupies it provides help in all phases of managing and deciding cases. Upon inquiry into the credentials and references of the young man she was told to hire, and after interviewing him, López Torres concluded, reasonably, that he was unqualified for the position. So she hired a qualified attorney for the position instead.
This was perceived by Norman as an act of defiance. In December of 1992, Norman chastised López Torres for not hiring the young lawyer he had sent to her, telling her that she did not understand how the process worked. He directed her to fire the attorney she had hired. When López Torres refused, Norman told her she would not become a Supreme Court Justice.
At approximately the same time, Lopez, the district leader who had assisted López Torres in becoming a Civil Court judge, angrily confronted López Torres about her refusal to hire the attorney sent to her by "County," i.e., Norman. Her ungratefulness to Norman had made Lopez look bad. To fix that problem, Lopez directed López Torres to redeem herself by firing her court attorney and hiring "County's" choice. She again refused.
In June 1995, Lopez gave López Torres another opportunity for redemption: if she hired Lopez's daughter, a recent law school graduate, as her court attorney, Lopez would get López Torres nominated *235 to fill an upcoming vacancy on the Supreme Court that the party leadership had earmarked for a "Latino." López Torres declined, refusing to fire the qualified attorney she had initially hired to the position.[33] From that point forward, Lopez never supported the judicial aspirations of López Torres, and indeed he worked against her in 2002, when she was reelected to the Civil Court.
In 1997, López Torres first sought the Democratic nomination to the office of Supreme Court Justice. She requested a meeting with Norman, who met her on August 22, 1997, in Junior's Restaurant on Flatbush Avenue in Brooklyn. Norman reminded López Torres that her failures to hire as her court attorney the people sent to her by party leaders had been a serious breach of protocol. López Torres replied that she was at that point between court attorneys, and was willing to consider a qualified applicant referred by Norman. Norman told López Torres that she needed to obtain the support of the "Latino" district leaders.
Three weeks later, however, Norman placed an urgent call to López Torres, demanding that she remove her name from consideration at the upcoming nominating convention. Her failure to do so, Norman declared, would be a direct challenge to him. An open convention, involving the competing nominations of more candidates than vacancies, was "not the way it works," according to Norman. López Torres Decl. ¶ 19. López Torres declined, expressing the naive view that she had a right to seek the nomination at the convention even without Norman's support. At the convention itself shortly thereafter, Norman was proved correct, as not a single delegate proposed López Torres for nomination.
López Torres tried again in 1998, when she applied to, and was interviewed by, the Kings County Judicial Screening Committee. In late September, before the nominating convention, she tried repeatedly to ascertain the results of that process, i.e., whether the screening committee had reported her as qualified. Neither Jerome Karp, the chair of the committee, nor Norman would provide that information, so López Torres ended her bid for the nomination.
In January 2002, López Torres renewed her effort to obtain the Democratic nomination for Supreme Court Justice. She contacted Norman and Karp to tell them so and to commence the screening process. Karp responded that the screening panel would consider only candidates referred by Norman. Norman made it clear that he would not make such a reference. In February and again in May of 2002, he told López Torres that his concerns with her candidacy had nothing to do with her qualifications, but rather arose out of her disloyalty her failure to hire the people sent to her by the party leadership and her refusal to withdraw her candidacy in 1997.
Without the support of Norman, López Torres's efforts to obtain a nomination at the 2002 convention were doomed. Her name was placed in contention by a supportive delegate, but the overwhelming majority of delegates voted instead for the package of candidates supported by Norman.[34]
*236 Norman was not content to simply block López Torres's effort to be elected as a Supreme Court Justice. In that same year, she was up for re-election to the Civil Court, and he chose to have the Democratic Party in Kings County support another nominee, an unusual tactic that evidenced the party leadership's overt hostility toward López Torres. Elections for that judicial office, as discussed earlier, involve a primary, in which a challenger candidate like López Torres can appeal directly to the voters. López Torres did just that, and she not only prevailed over the party leaders' preferred candidate in the primary, but she received more votes in the general election for Civil Court (200,710) than any of the Democratic candidates for Supreme Court Justice received on the same day.
In January of 2003, López Torres again wrote to Norman and Karp, declaring her candidacy for the office of Supreme Court Justice and requesting to be considered by the screening committee. At that time, it remained the policy of the committee to screen candidates only at the request of Norman or his Republican counterpart. Norman, who faced mounting criticism of the process, allowed the screening committee to interview López Torres. However, Norman still refused to support López Torres. In a meeting on June 6, 2003, he reiterated his displeasure at her refusal to withdraw her candidacy in 1997. He also stated that López Torres did not have sufficient support. Since she had been the leading vote-getter among elected judges in Brooklyn just six months earlier, Norman was not referring to support in the electorate. Rather, he was referring to the only support that matters when it comes to Supreme Court Justice candidates in the Second District his own and that of the district leaders. Norman summed up his position by telling López Torres that "County," i.e., the leadership of the Kings County Democratic County Committee, would only support candidates who support "County." López Torres Decl. ¶ 42.
As described above, López Torres's efforts to identify and lobby the delegates to the 2003 convention were frustrated by both the timing of the relevant events and the antipathy of the party leadership. At the convention, two delegates attempted to nominate López Torres. The effort failed, and the convention nominated the package of candidates endorsed by Norman. In a letter to his fellow district leaders, District Leader Ralph Perfetto explained that he voted against the "highly qualified" López Torres because she was an "ingrate;" she had "courted Vito Lopez to support her for Civil Court, but then decided she didn't need him anymore and denied his daughter a job." Pl. Ex. 10. "Too many judges have the same attitude towards District Leaders," Perfetto lamented. Id. He offered to take a polygraph test on the question whether Norman and Lopez had influenced his vote, and denied "any `deals' or conspiracy." Id.
López Torres's seven-year effort[35] to obtain her party's nomination for Supreme Court Justice is the selection process in microcosm. The path to the office of Supreme Court Justice runs through the *237 county leader of the major party that dominates in that part of New York State. Without his or her support, neither superior qualifications nor widespread support among the party's registered voters matters.
c. The Third District
In the Third District, the dominant county leaders, particularly those in Albany and Rensselaer Counties, choose the convention delegates and the nominees for Supreme Court Justice. These leaders also frequently enter into agreements with Republican Party leaders to cross-endorse each others' candidates.
d. The Fourth District
The Associations of New York State Supreme Court Justices in the City and State of New York called Justice Joseph Sise as a witness. Justice Sise's testimony demonstrated that the Republican Party county leaders in the Fourth District not the judicial convention delegates or the voters selected him as the nominee for Supreme Court Justice in 1998.
When Justice Sise learned in early 1998 that the judgeship traditionally set aside by party leaders for a Montgomery County resident would soon be vacated, Justice Sise (who resides there) told his Republican county leader that he wanted the position. On May 27, 1998, the Montgomery County Republican committee endorsed him. By the first week of June, a meeting was arranged for him to be introduced to the ten other county leaders in the Fourth District. Justice Sise already knew the county leader from Schenectady County the second most populous county in the district because he was his brother. By the end of June, there were two candidates (Justice Sise and one other) for two vacancies. When Justice Sise received the nomination at the convention, he expressed his appreciation to the eleven county leaders.
The defendants paint an entirely different picture of Justice Sise's successful effort to become a Supreme Court Justice. He was "the very model of efficiency and industry in seeking support district-wide from the grass roots on up," they say. Def. Reply Findings at ¶ 90. In the summer of 1998, he "traveled extensively to all corners of the sprawling district; mailed literature; used radio advertising; gave out magnets, buttons, balloons and other name recognition materials; used lawn signs . . . attended political and community events; and went to all nine of the 4th Judicial District's county fairs." Id. "[T]he whole purpose of this concerted outreach," defendants conclude, "was to win delegate support." Id. at ¶ 91.
Justice Sise's testimony undermines the defendants' position in two ways. First, the facts do not support the defendants' conclusion. Justice Sise worked very hard in his effort to become a Supreme Court Justice, but the one thing he did not concern himself with was judicial delegates. He courted the county leaders, and he engaged in his grass roots campaigning only because he already had the county leaders' support. His only mention of delegates was his statement that he ran into them "at different events," presumably the county fairs and other community events he attended. Tr. 1503-04. In short, winning delegate support was not Justice Sise's "whole purpose" or even a subsidiary purpose.
More importantly, the defendants' defective factual thesis with regard to the Sise election is flatly contrary to their claim about how New York's judicial conventions are supposed to work. According to the defendants, the basic purpose of the conventions is to avoid precisely the sort of grass roots campaign Justice Sise conducted. "The system isn't designed for individual candidates to be campaigning *238 directly among the voters." Tr. 1567 (Kellner). Indeed, the defendants have claimed that a campaign like Justice Sise's is one of the evils the convention system prevents. See, e.g., Tr. 214-15 (cross of Berger). In its place, they claim, is a "system . . . designed for the voters to select delegates who will perform th[e] function of screening candidates and deciding who the party's candidate should be." Tr. 1567 (Kellner).
In sum, I have no answer to the defendants' rhetorical question of why Justice Sise would campaign like he was running for mayor if he had already been assured by county leaders of a nomination for Supreme Court Justice. And I reject their false dichotomy, i.e., that I cannot find that he was assured of the nomination without branding Justice Sise a liar. See Def. Reply Finding at 36. Once again, this tactic is unfair to Justice Sise, who was an earnest and wholly credible witness, and plainly qualified for the position he holds. His testimony, however, helps the plaintiffs, not the defendants.
e. The Seventh Judicial District
By the early 1990s, the Republican Party's power base in the Rochester area was concentrated in the Monroe County government. The Democratic Party's power base was concentrated in the city government of Rochester. Because the Conservative Party nomination often affected the balance of political power in the county, the Republican Party county leader gave the Conservative Party county leader control over Republican Supreme Court nominations in exchange for Conservative Party support for the Republican leader's chosen candidates for the county legislature. Consequently, Republican candidates for Supreme Court Justice had to obtain the support of the Conservative Party county leader to be considered for the Republican Party's nomination. In effect, the Republican nomination process for Supreme Court effectively ended at the Conservative Party dinner held in May, when people "[found] out who the candidate [was] going to be." Tr. 388.
Former Rochester City Court Judge John Regan's efforts to compete for the Republican nomination illustrate the extent to which the Monroe County Conservative and Republican Party leaders control the nomination process. Regan had angered the Conservative Party leader by informing Democratic voters that the Democratic candidate for City Court, who was supported by the Conservative Party, was not a genuine Democrat. In retaliation, the Conservative Party leader would not support Regan's nomination for Supreme Court in 1994. Despite significant popularity among the voters and the Republican Party county committee members across the district, Regan was told by the Republican county leader that he controlled the judicial convention delegates, and "that's all that matters." Regan Decl. ¶ 12. Regan nevertheless contacted many of the Republican Party county committee members from each of the ADs in the district to seek their support for his candidacy. They indicated to Regan that they had no choice but to back the Republican Party Leader's choice of nominee at the convention, which they did.
f. The Eighth Judicial District
The Eighth District is the only place in the state in the past 25 years where a Supreme Court Justice candidate received a major party nomination over the objection of the dominant county leader. That unique event was the product of a broader struggle for control within the Erie County Democratic Party that two years later resulted in the ouster of the county leader. It was also the only time the votes of the delegates at the judicial nominating conventions in the district have had any significance *239 other than ratifying the county leader's choices.
g. The Ninth Judicial District
The Republican Party county leaders control the selection of judicial delegates and alternates in the Ninth District. The delegate allocation formula accords the most power to the county leader in Westchester County, where 45% of the district's registered Republicans reside. At the time of the hearing, 17 of 23 sitting Supreme Court Justices (not including those designated as Appellate Division Justices) in the district were from Westchester County. An agreement among all the county organizations in the 1980s, however, has prevented any further imbalance in favor of Westchester by allowing each county leader to fill vacancies arising in his or her county.
The nominees for Supreme Court Justice in the Ninth District are selected by the county leader in the county where the justice will sit. The only way to become nominated is to obtain the sponsorship of that county leader; if the vacancy happens to be in one of the four counties other than Westchester, the blessing of the leader in Westchester County is required as well.
The conventions in Westchester are mere formalities, as they are throughout the state. There is no lobbying of delegates beforehand and no deliberation at the conventions themselves. Indeed, there is nothing to deliberate, as the delegates know their role is to ratify the party leaders' choice of candidates.
C. Screening Panels
In the First and Second Judicial Districts, screening panels play a role in the selection of Supreme Court candidates. These panels are not required or addressed by New York law.
The screening panel in the First District has had a salutary effect on judicial selection. Specifically, it has made it much more likely that the candidates selected by the county leaders are highly qualified. The panel is constituted pursuant to a "double-blind" procedure that ensures an important measure of independence. Though it is not free of political influence altogether, the county leadership deserves praise for having created it many years ago and for abiding by its decisions.
Still, the most important thing to be said about screening panels is why they are necessary. Robert Levinsohn, a defense witness who was instrumental in the creation of the First District panel, testified that screening panels are needed to counter the manipulation by county leaders of the process of judicial selection. In the First District, the panel was created in 1976 because of the closed nature of process, in which the judicial delegates simply followed the wishes of the party leaders, with no public input in the selection of the judges. Those characteristics of the process continue to exist today throughout the state.[36]
The screening panel in the Second District, at least until recently, was almost farcical. In 2002, plaintiff Margarita López Torres wrote to the chair of the panel, Jerome Karp, expressing her interest in a nomination for Supreme Court Justice and seeking guidance about the nomination process. Karp promptly wrote back, advising López-Torres that the screening panel considered only people referred to it by the county leader. That sort of screening panel is not likely to reduce a county leader's manipulation of the judicial selection process.
*240 When plaintiff Philip Segal, a former Family Court judge, was interviewed by the Second District screening panel in the early 1990s, he was asked an unusual question: could he identify who his district leaders were? When he said no, the panel members broke out laughing.
Escalating criticism of the judicial selection process in the Second District brought changes to its screening panel in 2003. Entities other than party officials now play a role in selecting the membership of the screening panel. People other than those selected by Norman may now apply. The results are hotly debated, in part because members of the panel have resigned and because Norman successfully pressured the delegates to the 2004 judicial nominating convention to re-nominate a sitting Supreme Court Justice the screening panel found to be unqualified. But the outcome of that debate is not relevant to the outcome of this motion. The issue in this case is whether the voters are accorded their rightful role in the selection of Supreme Court Justices. If they are not, that constitutional defect cannot be remedied by a screening panel, even if it has integrity and plays a meaningful role in the quality of the judges selected.
DISCUSSION
Because they seek to enjoin the judicial nominating convention system prescribed by New York Election Law §§ 6-106, 6-124, and 6-158(5), the plaintiffs request injunctive relief that would affect "government action taken in the public interest pursuant to a statutory or regulatory scheme," Beal v. Stern, 184 F.3d 117, 122 (2d Cir.1999), and would "alter rather than maintain the status quo," Rodriguez v. DeBuono, 175 F.3d 227, 233 (2d Cir. 1999). Accordingly, they must establish (1) that absent the requested injunctive relief they will suffer irreparable harm and (2) a "clear" or "substantial" likelihood they will succeed on the merits. No Spray Coalition, Inc. v. City of New York, 252 F.3d 148, 150 (2d Cir.2001) (per curiam).
A. Irreparable Harm
The defendants argue the plaintiffs cannot show irreparable harm because challenger candidates for the office of Supreme Court Justice "have more than adequate ballot access through several means." Def. Proposed Conclusions of Law 3. In support, the defendants marshal, in miniature, their arguments as to why New York's system of nominating candidates for Supreme Court Justice comports with the Constitution. In other words, the defendants argue that the plaintiffs cannot show irreparable harm because their claims fail on the merits.
Although it is of course true there is no "harm" to the plaintiffs if the challenged nominating scheme is consistent with the Constitution, this manner of argument is hardly illuminating. Arguments as to the validity of the plaintiffs' claims whether they have shown a violation of their constitutional rights are more properly addressed to the plaintiffs' likelihood of success on the merits. The separate, and preliminary, inquiry of whether the plaintiffs have made a showing of "irreparable harm" turns not upon the validity of the plaintiffs' claims but upon the nature of the injury alleged: is it the kind of harm that warrants preliminary injunctive relief? As to that question, the defendants put up no genuine dispute. Assuming the plaintiffs' claims have merit, the harm alleged a violation of the plaintiffs' fundamental right to vote and to compete for elective office is properly remedied pending trial on the merits by a preliminary injunction.[37]See 11A Charles Alan Wright *241 & Arthur R. Miller, Federal Practice and Procedure § 2948.1 (3d ed. 1998) ("When an alleged deprivation of a constitutional right is involved, most courts hold that no further showing of irreparable injury is necessary.").
B. Likelihood of Success of the Merits
1. Threshold Issues
The defendants advance two threshold arguments as to why the plaintiffs are not likely to succeed. I address them in turn below, rejecting each.
a. Compulsory Joinder
The defendants argue the court cannot properly afford the plaintiffs complete relief because they have failed to join an indispensable party, namely, the New York State Democratic Committee. See Fed.R.Civ.P. 19(a). As the defendants point out, "[t]he statutory provisions that Plaintiffs are challenging are contained not only in the statutes that Plaintiffs seek to enjoin, but also in the rules of the Democratic Party of The State of New York, which is not a party to this case." Def. Proposed Conclusions of Law 11. Specifically, as discussed above, the Democratic Party's rules prescribe, within limits set by statute, the number of delegates and alternate delegates each Assembly District sends to the judicial nominating conventions. Such political party rules, the defendants argue, "will still be operative even if the statutory provisions are struck down because, as the Supreme Court has made clear, a state cannot vitiate the rules by which a political party selects its nominee." Id., citing Cal. Democratic Party v. Jones, 530 U.S. 567, 120 S.Ct. 2402, 147 L.Ed.2d 502 (2000).
The defendants provide no pinpoint citation for this proposition, perhaps because Jones does not support that sweeping prohibition. See Jones, 530 U.S. at 572, 120 S.Ct. 2402 ("States have a major role to play in structuring and monitoring the election process, including primaries."). To be sure, the Democratic Party has important First Amendment rights of its own. Nominating candidates for public office is "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 v. Republican Party of Conn., 479 U.S. 208, 216, 107 S.Ct. 544, 93 L.Ed.2d 514 (1986), and all political parties accordingly have a right "to select a standard bearer who best represents the party's ideologies and preferences." Eu v. San Francisco County Democratic Cent. Comm., 489 U.S. 214, 224, 109 S.Ct. 1013, 103 L.Ed.2d 271 (1989) (internal quotation marks omitted). *242 The defendants' argument here, however, amounts to a claim that any remedy imposed in this case, including, for example, requiring a primary, would be an unconstitutional violation of the major parties' right to freedom of association. I disagree. The Supreme Court has repeatedly deemed it "`too plain for argument' . . . that a State may require parties to use the primary format for selecting their nominees, in order to assure that intraparty competition is resolved in a democratic fashion." Jones, 530 U.S. at 572, 120 S.Ct. 2402, quoting Am. Party of Tex. v. White, 415 U.S. 767, 781, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974).
Any relief granted in this case would be binding upon the defendant New York State Board of Elections, which is charged with the "authority and responsibility for the execution and enforcement of all laws relating to the elective franchise." Matter of New York State Bd. of Elections, 49 A.D.2d 806, 373 N.Y.S.2d 420, 423 (4th Dep't 1975); see also Schulz v. Williams, 44 F.3d 48, 61 n. 13 (2d Cir.1994) (noting it is "well-settled that a state official may properly be made a party to a suit seeking to enjoin the enforcement of an allegedly unconstitutional act if that official plays some role in the enforcement of the act," and that the "state Board's members . . . have the requisite `special relation' to the contested provision [of the New York Election Law] to render them proper defendants"). The defendants point to no colorable constitutional ground upon which the State Democratic Party might lawfully disregard the State Board's authority to enforce all forms of relief the plaintiffs seek, including an order requiring the major parties to nominate their candidates for Supreme Court Justice by direct primary election. The defendants' claim that the relief sought by the plaintiffs would be "wholly ineffectual," Def. Proposed Conclusions of Law 11, is therefore without merit.
b. Preclearance Under the Voting Rights Act
The defendants next argue that "statewide changes to New York's election laws governing nomination of Supreme Court justices would . . . require preclearance by the Civil Rights Division of the United States Department of Justice. . . . which has not been sought." Def. Proposed Conclusions of Law 12. As the plaintiffs correctly point out, the requirement of preclearance applies only to potential changes made by the New York State legislature. Pl. Reply Conclusions of Law 7. "A decree of the United States District Court," however, "is not within reach of Section 5 of the Voting Rights Act." Connor v. Johnson, 402 U.S. 690, 691, 91 S.Ct. 1760, 29 L.Ed.2d 268 (1971). The plaintiffs therefore need no preclearance in order to seek relief from this Court. In the event the New York State legislature makes changes as a result of this opinion, preclearance may properly be sought at that time.
2. The Merits
The plaintiffs bring their challenge both as voters and as would-be candidates for the office of Supreme Court Justice. The right to vote "is of the most fundamental significance under our constitutional scheme," Burdick v. Takushi, 504 U.S. 428, 433, 112 S.Ct. 2059, 119 L.Ed.2d 245 (1992) (internal quotation marks omitted); "[o]ther rights, even the most basic, are illusory if the right to vote is undermined." Wesberry v. Sanders, 376 U.S. 1, 17, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964). Notwithstanding the primacy of the citizens' right to vote, the Constitution of the United States does not impose upon the State of New York an obligation to select its judges by popular election. If, however, the State of New York "chooses to tap the energy and legitimizing power of the democratic *243 process" in selecting its Supreme Court Justices, "it must accord the participants in that process . . . the First Amendment rights that attach to their roles." Republican Party of Minn. v. White, 536 U.S. 765, 788, 122 S.Ct. 2528, 153 L.Ed.2d 694 (2002) (internal quotation marks omitted).
The Constitution of the State of New York requires that "[t]he justices of the supreme court shall be chosen by the electors of the judicial district in which they are to serve." N.Y. Const. art. VI, § 6(c). In carrying out that command, the New York legislature has provided the major parties an official role in the election process. Although New York law for a brief time required the major parties to nominate candidates for the Supreme Court by direct primary, see N.Y. Laws of 1914, ch. 5, the prevailing method, since 1921, has been to select Supreme Court nominees through the nominating conventions challenged in this case.
a. The Governing Standard
The plaintiffs argue that New York's electoral scheme for nominating candidates for the office of Supreme Court Justice infringes both the right of rank-and-file members of major parties to participate in the nominating process and the right of potential candidates to compete for the nomination. In evaluating this claim, the first question is whether New York's judicial convention system "should be sustained if it can be shown to have some rational basis, or whether it must withstand a more rigid standard of review." Bullock v. Carter, 405 U.S. 134, 143, 92 S.Ct. 849, 31 L.Ed.2d 92 (1972). As the Supreme Court has often explained, a "flexible" standard guides this inquiry:
A court considering a challenge to a state election law must weigh the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate against the precise interests put forward by the State as justifications for the burden imposed by its rule, taking into consideration the extent to which those interests make it necessary to burden the plaintiff's rights. . . .
[W]hen those 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 State's important regulatory interests are generally sufficient to justify the restrictions.
Burdick, 504 U.S. at 434, 112 S.Ct. 2059 (internal quotation marks and citations omitted).
Many of the plaintiffs' arguments focus upon the rights of candidates rather than upon the rights of the voters themselves. Of course, "laws that affect candidates always have . . . [a] correlative effect on voters." Bullock, 405 U.S. at 143, 92 S.Ct. 849. That is, "[b]y limiting the choices available to voters, the State impairs the voters' ability to express their political preferences." Ill. State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 184, 99 S.Ct. 983, 59 L.Ed.2d 230 (1979). In evaluating such barriers to candidates, therefore, "it is essential to examine in a realistic light the extent and nature of their impact on voters." Bullock, 405 U.S. at 143, 92 S.Ct. 849.
b. The Plaintiffs' Theory
The plaintiffs' theory of the case is as follows: For any Supreme Court candidate who is not supported by party leadership, the burdens of getting her own delegate and alternate delegate candidates on the ballot in enough assembly districts to command *244 a majority at the convention are insurmountable. Moreover, the delegates and alternates selected and installed in office by the party leaders simply do their leaders' bidding. They are not amenable to persuasion by challenger candidates, and there is insufficient time for such persuasion in any event. Accordingly, the plaintiffs argue, challenger candidates have no way of presenting to the voters a set of alternatives to the judicial candidates chosen by the party leadership, and voters in turn have no opportunity to review the party leadership's decisions.
c. Challenger Candidates
The defendants argue that the plaintiffs' theory is unsound as a matter of constitutional law. They begin by attacking the plaintiffs' "challenger candidate paradigm," which they contend is "problematic" both because it is vaguely defined and because it "asks this Court to weigh the Constitutionality of the convention system through a very peculiar set of circumstances." Def. Findings at ¶¶ 5-7. I disagree. As Chief Judge Korman explained in Molinari v. Powers, considering a challenge by Republican presidential candidate John McCain, who lacked the support of party leadership:
The Republican State Committee controls a massive apparatus of over 10,000 Republican elected officials and workers that are selected in each town, county and congressional district throughout New York. Each county chairperson can easily mobilize the resources necessary to conduct a petition drive: recruiting potential delegate candidates, quickly obtaining up-to-date and usable walking lists of registered voters from the county boards of elections, supplying an army of dedicated Party workers who will gather signatures, making signers available at Party functions and providing persons who will review opposing petitions for technical defects. The Republican State Committee delivered this apparatus to its favored candidate, Governor Bush, who then had the luxury of being able to obtain the necessary ballot access signatures relatively painlessly [and] without having to [expend] scarce campaign resources. . . .
82 F.Supp.2d 57, 62 (2000).
The same general phenomenon prevails with the county organizations at issue in this case. Such grass roots organizations are in many ways the lifeblood of our politics, but the First Amendment prohibits them, once established, from engaging in anticompetitive conduct that freezes the political status quo by establishing unduly onerous procedures. Williams v. Rhodes, 393 U.S. 23, 31, 89 S.Ct. 5, 21 L.Ed.2d 24 (1968) (striking down Ohio law giving "the two old, established parties a decided advantage over any new parties struggling for existence" and thereby severely burdening "both the right to vote and the right to associate"); Storer v. Brown, 415 U.S. 724, 729, 94 S.Ct. 1274, 39 L.Ed.2d 714 (1974) (noting the prohibition on "effective political monopoly"); Jenness v. Fortson, 403 U.S. 431, 438, 91 S.Ct. 1970, 29 L.Ed.2d 554 (1971) (same). If the petitioning requirements for getting delegates on the ballot are so burdensome that only an extensive, pre-existing organization can satisfy them, then the voters will be virtually eliminated from the machinery of choice at the nomination stage. The burdens of a statutory scheme restricting ballot access, therefore, must be judged from the viewpoint of a "reasonably diligent . . . candidate" who lacks the support of "a massive apparatus" controlled by party leadership. Molinari, 82 F.Supp.2d at 70-71; cf., Storer, 415 U.S. at 742, 94 S.Ct. 1274 ("Could a reasonably diligent independent candidate be expected to satisfy the signature requirements, or will it be only rarely that the unaffiliated candidate will succeed in getting on the ballot?"). *245 That is precisely the sense in which the plaintiffs use the term "challenger candidate," and the meaning I intend by my use of the phrase in this opinion.
d. Access to the General Election Ballot
The defendants next argue that "there is no constitutional violation where the electoral scheme provides for alternative means of access to the relevant ballot." This much is true. See Lubin v. Panish, 415 U.S. 709, 716, 94 S.Ct. 1315, 39 L.Ed.2d 702 (1974) ("In the absence of reasonable alternative means of ballot access, a State may not, consistent with constitutional standards, require from an indigent candidate filing fees he cannot pay."); LaRouche v. Kezer, 990 F.2d 36, 39 (2d Cir.1993) (analyzing statutory scheme providing two means of ballot access and holding that "if either alternative would be constitutional standing alone, the other must be viewed as broadening the opportunities for ballot access and is a fortiori constitutional"). As the defendants point out, the New York scheme challenged here provides "any would-be candidate the option of . . . petitioning directly onto the general election ballot" as an independent, and the prerequisites for doing so have been "repeatedly sustained." Def. Proposed Conclusions of Law 19, 29; see, e.g., Kuntz v. N.Y. State Senate Bd. of Elections, 113 F.3d 326, 327-29 (2d Cir.1997) (upholding similar signature requirements for independent congressional candidates).
The defendants then argue that the general election ballot must be "the relevant ballot" in this case because there is no primary ballot for Supreme Court candidates. The "primary" is only for delegates, and the petitioning requirements for that office, considered individually, are not particularly burdensome.
The defendants' argument fails because the right the plaintiffs assert is not to run for delegate or to access the general election ballot as an independent, but rather to compete for their major party's nomination for Supreme Court Justice by garnering support among the rank-and-file members. The issue is whether the First Amendment guarantees the plaintiffs that opportunity. I conclude it does.
I recognize that at least in form, petitioning onto the general election ballot as an independent and winning a major party's nomination gain a candidate the same thing: a line on the ballot and an opportunity to appeal to the voters. But in the reality of politics, the two paths are worlds apart. As I have explained, nearly all Supreme Court Justice "races" in New York State are overwhelmingly controlled by one party the Democrats in New York City and the Republicans in most of the rest of the state. The facts in this case, viewed "in a realistic light," Bullock, 405 U.S. at 143, 92 S.Ct. 849, require the conclusion that the partisan nomination process prescribed by New York law functions as an "integral part of the election machinery." United States v. Classic, 313 U.S. 299, 318, 61 S.Ct. 1031, 85 L.Ed. 1368 (1941). As the Supreme Court held in considering the right to vote for members of the House of Representatives, specifically granted in Article I, Section 2:
Where the state law has made the primary an integral part of the procedure of choice, or where in fact the primary effectively controls the choice, the right of the elector to have his ballot counted at the primary, is likewise included in the right [to vote]. . . . Here . . . the right to choose a representative is in fact controlled by the primary because . . . the choice of candidates at the Democratic primary determines the choice of the elected representative. Moreover, we cannot close our eyes to the fact . . . that the practical influence of the choice of candidates at the primary may be so *246 great as to affect profoundly the choice at the general election even though there is no effective legal prohibition upon the rejection at the election of the choice made at the primary and may thus operate to deprive the voter of his constitutional right of choice.
Classic, 313 U.S. at 318, 61 S.Ct. 1031.
The Supreme Court has adhered to this view in the context presented here the right to vote under the First and Fourteenth Amendments and rejected the very line of argument the defendants proffer. Considering a challenge to a candidate filing fee in a primary election for state office when no such fee existed for the general election, the Court explained:
Instead of arguing for the reasonableness of the exclusion of some candidates, appellants rely on the fact that the filing-fee requirement is applicable only to party primaries, and point out that a candidate can gain a place on the ballot in the general election without payment of fees by submitting a proper application accompanied by a voter petition. Apart from the fact that the primary election may be more crucial than the general election in certain parts of Texas, we can hardly accept as reasonable an alternative that requires candidates and voters to abandon their party affiliations in order to avoid the burdens of the filing fees imposed by state law.
Bullock, 405 U.S. at 146-47, 92 S.Ct. 849. I agree with the plaintiffs that the reasoning in Classic and Bullock is not limited to primary elections, but rather applies to party nominations generally, regardless of the form. In short, states need not make party nominations an integral part of the machinery of an election, but once they do, they must accept the consequent burden of ensuring that the nomination process does not "operate to deprive the voter of his constitutional right of choice." Classic, 313 U.S. at 318, 61 S.Ct. 1031. Particularly where, as here, there are established areas of one-party rule, voters and candidates have a right to participate meaningfully in the nomination process, which includes a realistic opportunity to challenge the selections of party leadership.
The defendants contend that this conclusion reduces to a declaration, contrary to Supreme Court precedent and longstanding tradition, that voters have a constitutional right to a primary, or at least to a convention that "resembles a primary . . . as closely as one could imagine," Morse v. Republican Party of Va., 517 U.S. 186, 236, 116 S.Ct. 1186, 134 L.Ed.2d 347 (Breyer, J., concurring in the judgment). Def. Proposed Conclusions of Law 30. They cite American Party of Texas v. White, in which the Supreme Court found it "too plain for argument . . . that the State may . . . insist that intraparty competition be settled before the general election by primary election or by party convention." 415 U.S. 767, 781, 94 S.Ct. 1296, 39 L.Ed.2d 744 (1974) (emphasis supplied). But White suggests only that states may afford voters and candidates their constitutional rights in a convention system of nomination. It hardly follows that New York's unique judicial convention system is necessarily constitutional. The constitutional analysis is not categorical; just as some primary systems fail to pass constitutional muster, see Cal. Democratic Party v. Jones, 530 U.S. 567, 120 S.Ct. 2402, 147 L.Ed.2d 502 (2000), the New York system does not elude scrutiny by virtue of its falling under the broad rubric of "party convention." Rather, the standard for evaluating the plaintiffs' claims is a flexible one, not readily susceptible to such per se rules. Anderson v. Celebrezze, 460 U.S. 780, 789, 103 S.Ct. 1564, 75 L.Ed.2d 547 (1983). The relative interests of the voters, the candidates, the parties, and the state must all be evaluated.
*247 In a loose sense, then, the defendants are correct that "[n]othing in the United States Constitution requires that once a state decides to make an office elective, the electoral system must ratify the direct and unmediated preferences of a plurality of voters." Def. Proposed Conclusions of Law 31-32. Indeed, under certain circumstances, "standing between the individual voter and the eventual nomination of a candidate may be numerous party rules and procedures so that the will of the majority of the electorate expressing a . . . preference[,] and the selection of delegates may be only partially translated into the actual nomination." Bachur v. Democratic Nat'l Party, 836 F.2d 837, 842 (4th Cir. 1987).
There comes a point, however, where the barriers between the voice of the voters and the results of the nominating process become so high as to overshadow any legitimate interest the state or the party may have in distancing the elective office from the electorate. An election must always retain the essential qualities of an election, and in areas of one-party rule, voters' and candidates' rights of access to the nomination stage of the electoral process take on greater importance.
In Ripon Society, Inc. v. National Republican Party, 525 F.2d 567 (D.C.Cir. 1975), the leading case upon which the Fourth Circuit relied in Bachur, the D.C. Circuit considered a challenge to the delegate allocation formula adopted by the National Republican Party for its quadrennial convention. The formula granted extra delegates to states that had voted for the Republican candidate in the last presidential election, and the challengers argued the formula thus denied them equal protection of the laws. The court upheld the formula, but Judge McGowan's explanation for the en banc court was careful to note a crucial caveat:
We are keenly aware that as a practical matter, the ultimate choice of the mass of voters is predetermined when the nominations have been made. If the right to vote is a right to true participation in the elective process, then it is heavily implicated in the nomination process. We do not deny this, but rest our judgment on the view that, as between that right and the right of free political association, the latter is more in need of protection in this case. . . . Of . . . practical importance[, however,] is the fact that there are two major parties, which for the time being at least are in intense competition. Persons not heard in one party may be welcomed in the other, and if there are enough such defections, the offending party may lose the general election, as both parties must be well aware.
. . .
[W]e have said that voting rights are not as heavily implicated in a nomination as in an election. It might be otherwise in a case where there is only one party with a realistic chance to win the election, and where a vote in the nominating process is the only effective vote that can be cast.
Id. at 586, 588. This is just such a case: "a vote in the nomination process is the only effective vote that can be cast." Id. at 586. Regardless of whether the state has chosen to employ a primary or a convention, there is an inverse relationship between the degree to which the nomination, rather than the general election, dictates the eventual occupant of the office and the permissible distance "between the individual voter and the eventual nomination of a candidate" due to "party rules and procedures." Bachur, 836 F.2d at 842. Because nominations, not general elections, are the critical determinant in electing Supreme Court Justices in New *248 York, more open and effective participation by voters must be allowed at the nomination stage, and candidates must be permitted an effective means of appealing to voters when it counts.
e. The Severity of the Burdens
Accordingly, the operative inquiry, to which I now turn, is whether the burdens imposed by the New York system, both on effective participation by voters and on candidates in appealing to them, are "severe" in practical application. In the context of a challenge to ballot access laws, the inquiry into whether the burdens are "severe" is stated more particularly as:
[C]ould a reasonably diligent independent candidate be expected to satisfy the signature requirements, or will it be only rarely that the unaffiliated candidate will succeed in getting on the ballot? Past experience will be a helpful, if not always an unerring, guide: it will be one thing if independent candidates have qualified with some regularity and quite a different matter if they have not.
Storer v. Brown, 415 U.S. 724, 742, 94 S.Ct. 1274, 39 L.Ed.2d 714 (1974). As applied to this case, this inquiry produces the following questions: Could a reasonably diligent challenger candidate for Supreme Court Justice succeed in getting her own delegates and alternates on the ballot in each Assembly District? See Molinari v. Powers, 82 F.Supp.2d 57, 70-71 (2000) (applying the words "independent" and "unaffiliated" in Storer to mean independent of party leadership). If not, could she succeed in lobbying the delegates installed by the party leaders?
As for the first question, to say that challenger candidates for the Supreme Court "only rarely" succeed in getting their own slates of delegates on the ballot across the judicial district "gives euphemism a bad name." Koszola v. FDIC, 393 F.3d 1294, 1302 (D.C.Cir.2005). It never happens. Only the New York State Republican Committee has argued that it is even feasible. It contends that a challenger candidate can run delegate slates in selected ADs to either get a majority of delegates or at least a sufficient bloc to get the candidate into a multiple-candidate "package" by logrolling with other groups of delegates.
I have already rejected these assertions on factual grounds. I further note that the logrolling theory presupposes a severe burden on the rights of both voters and the challenger candidates. It suggests a potential way to work around the onerous petitioning requirements, a strategy by which a savvy candidate might win a nomination despite the insuperable barriers to electing delegates who would nominate her directly. Even if the process were amenable to such a strategy, it could not alter the fact that the burdens on ballot access are severe.
All but one of the defendants concede the system is not designed for challenger candidates to run their own slates of delegates. Rather, they argue, "the `voters of the party' to whom candidates `present their case' at the nominating phase of the judicial selection process are the duly elected judicial delegates," not the rank-and-file members. Def. Proposed Conclusions of Law 38. Kellner's testimony in this regard bears repeating:
The system isn't designed for individual candidates to be campaigning directly among the voters. The system is designed for the voters to select delegates who will perform that function of screening candidates and deciding who the party's candidate should be. So the idea that an individual candidate would go out and recruit delegate candidates and run delegates pledged to that candidate in the primary is not the system and it twists the design of the system on its head.
*249 Tr. 1567. As discussed above, this theory is also grounded on a faulty factual premise. Specifically, New York's judicial convention does not actually work, and was not designed to work, in the manner described by these defendants. The process of placing delegates at the convention is so difficult, except in isolated instances in a single AD, that only the party organization can accomplish it. The party leaders select delegates who do their bidding. In addition, the time frame in which judicial candidates are able to contact delegates is deliberately kept short. Delegates are elected less than three weeks before the convention. Even when they are unopposed, they may only be identified by examining records at the Board of Elections after the second week of July. The timing of the screening panels in two of the most populous districts also works to narrowly constrict the time available to lobby delegates.
I discuss below the merits of the State's asserted interests in deliberately precluding candidates from appealing directly to voters in the nominating process. It suffices to say here that I conclude, based on the findings of fact discussed above, that the burdens on candidates and voters under New York's convention system are severe. Reasonably diligent candidates who lack the support of entrenched party leaders stand virtually no chance of obtaining a major party nomination, no matter how qualified they are and no matter how much support they enjoy among the registered voters of the party.
Chief Judge Korman has twice struck down petitioning requirements analogous to those at issue in this case in favor of the rights of challenger candidates who were not supported by an established party organization in their efforts to get on the ballot. Rockefeller v. Powers, 917 F.Supp. 155 (E.D.N.Y.1996), aff'd, 78 F.3d 44 (2d Cir.1996), reviewed the process for petitioning delegates onto the ballot in each of New York's 31 congressional districts in the Republican primary for President of the United States. In each district, candidates were required "to obtain signatures from 5% or 1250, whichever is less, of the enrolled Republican voters in the district." 917 F.Supp. at 160. After observing that these signature requirements, in the aggregate, were dramatically higher than those in any other state, Chief Judge Korman explained that New York's highly technical system for challenging the validity of signatures greatly increases the actual number of signatures a candidate needs to obtain. Id. at 160-61. He also cited a number of other factors contributing to the difficulty of getting on the ballot:
New York law prohibits a person from signing petitions for more than one candidate and invalidates any signatures given after the first. See N.Y. Elec. Law § 6-134(5). As the petitioning phase proceeds, the number of people still available to sign petitions decreases and the difficulty in finding them increases. The thirty-seven day petition gathering period is, on its face, relatively short and, making matters worse, the petitioning phase now falls during the period from Thanksgiving to just after New Year's Day. As a result, petition gatherers were hampered by holidays and vacations, short days, and winter storms.
Id. at 161. The "most troubling" confirmation that these burdens in fact precluded candidates who lacked the support of the party organization from competing for New York's delegates to the National Convention was "[t]he rarity of a candidate who [was] willing and able to target all or almost all of the districts in New York. . . . History shows that the candidate who is not favored by the Party will never target every district, unless he is Steve Forbes," id. at 163, that is, unless he has a vast personal fortune at his disposal. The Second *250 Circuit agreed that these burdens, taken together, were severe and unjustifiable, and affirmed. Rockefeller v. Powers, 78 F.3d 44 (1996).
I recognize that the Second Circuit has since indicated that its holding in Rockefeller was dependent upon the "special circumstances of that case." Prestia v. O'Connor, 178 F.3d 86, 87 (2d Cir.1999). In Prestia, the court affirmed the denial of a request to enjoin numerically identical signature requirements in a congressional election, which involved only one district rather than all 31 as in Rockefeller.
Unlike in Prestia but as in Rockefeller, candidates for Supreme Court must run delegates in multiple ADs as many as 24 in the Second District. Also as in Rockefeller, the burdens here are comparatively more severe, both with respect to other elective offices (judicial and otherwise) in New York State and with respect to other procedures for electing trial judges across the country. And challenger candidates are never successful in winning the nomination. Indeed, the burdens here are more severe than those in Rockefeller, in part because the challengers here face the tremendous additional burden of educating voters as to which delegates plan to support them at the convention.
During the 2000 presidential election, Chief Judge Korman again entertained a challenge to the petitioning requirements for access to the ballot in the Republican presidential primary, and again concluded the aggregate burdens of the ballot access requirements were unconstitutionally severe. Molinari v. Powers, 82 F.Supp.2d 57 (E.D.N.Y.2000). I need not review Molinari in great detail here because, as the plaintiffs point out, the analysis involved "a straightforward elaboration on Rockefeller." Pl. Reply Conclusions of Law 12.
In sum, precedents from this Circuit and from other cases in this district amply support the conclusion that the burdens on ballot access found in this case are severe.
f. The State's Asserted Interests
In order to survive constitutional scrutiny, the convention scheme must be narrowly drawn to serve a compelling state interest. Lerman v. Bd. of Elections, 232 F.3d 135, 145 (2d Cir.2000). The defendants argue the convention system is uniquely designed simultaneously to (1) protect the major parties' rights to freedom of association; (2) promote geographic diversity; (3) promote racial diversity; and (4) protect incumbent Supreme Court Justices. They further contend the New York judicial convention scheme "is the only system designed to effectuate all of these legitimate and compelling state interests and, therefore, serves a compelling state interest." Def. Proposed Conclusions of Law 48. I consider these interests in turn.
The state's interest in protecting the major parties' associational rights is indeed compelling. But the major parties do not have an absolute right under the First Amendment to select their nominees through a convention process no matter how much that process impinges on the rights of challenger candidates and the party's registered voters. Indeed, in appropriate circumstances, "a State may require parties to use the primary format for selecting their nominees, in order to assure that intraparty competition is settled in a democratic fashion." Cal. Democratic Party v. Jones, 530 U.S. at 572, 120 S.Ct. 2402. Thus, when the interest in voter participation is asserted by the state on the voters' behalf, it outweighs the major parties' generalized interest in autonomy. Surely the result is no different when the interest is asserted by the voters directly.
The only specific interest the defendants advance in this regard is in avoiding party *251 raiding, i.e., the sabotage of the nomination process by members of another party. That interest is surely legitimate, and the convention system does advance it. But it is by no means narrowly tailored to do so. A primary election (or any other nomination mechanism, for that matter) open only to registered members of the party would protect against raiding just as well. The fact that the convention system excludes non-members of the party from voting or running for delegate has nothing to do with whether the State has a compelling interest in permitting the exclusion by a party of its own members.[38] The state's interest in party associational rights cannot justify the severe burdens imposed on voters and candidates by the convention system.
The defendants characterize the state's interests in geographic and racial diversity as legitimate, though they do not go so far as to call them compelling. Their notion of geographic diversity is not precisely defined, but basically it means that there should be a respectable amount of judicial representation from each of certain smaller geographic regions within the large judicial districts. The convention system is hardly designed to promote this interest. Delegates elected from ADs in which they need not even reside select justices who need not even be residents of the districts in which they are later elected. The justices can be (and apparently are) assigned to sit elsewhere in the state. See Tr. 1513 (Justice Sise, though elected in Fourth District, was assigned for the 2000-2001 term to New York City). In any event, as with party raiding, this interest does not justify precluding voters from participating in the process. The more direct and democratic way to serve geographical diversity is to define the geographic areas from which representation is desired, draw lines around them, and have the voters within them select the nominees.
With respect to racial diversity, the defendants style the judicial convention system as a species of affirmative action, arguing that it "ensures that minorities have the opportunity to win the nomination and a seat on the Supreme Court because by its nature, the convention process is a representative system that fosters coalition-building." Def. Proposed Conclusions of Law 47. It is not clear whether the defendants mean to assert that racial or ethnic minorities are likely to be denied equal protection of the laws under a system in which voters directly participate, and that the convention system is a prophylactic means of remedying latent discrimination. At any rate, the plaintiffs assume that this is a legitimate state interest, and so will I. However, the evidence the defendants have marshaled in support of this claim shows little more than that in New York City, where racial minorities exist in sufficient numbers that minority candidates do well in primary elections for Civil Court and other public offices, such candidates have also achieved success in obtaining Supreme Court nominations through the convention system. In other parts of the state, where minorities are present in much fewer numbers, minority representation among Supreme Court Justices is hardly remarkable.
The defendants' expert on this issue, Dr. Michael Hechter, was unpersuasive. He testified that direct participation by the voters, as in a primary system, would fail to produce minority Supreme Court Justices in districts where a majority of the voting population is white. Because Supreme *252 Court elections are low visibility, Dr. Hechter testified, he assumed that each voter will always vote for a candidate sharing the voter's own race or ethnicity. As mentioned above, Dr. Hechter and the defendants cast the convention system as an enlightened and benevolent antidote to that phenomenon. The logrolling it fosters, they contend, permits minority candidates who cannot command the support of a majority of delegates to get nominated.
I reject these arguments for several reasons. First, curbing voter participation is not the only mechanism by which to promote racial diversity on an elected bench. As the plaintiffs propose, allowing cumulative voting[39] in primary elections involving multiple judgeships would permit minority groups of any kind racial, geographic, ideological, or otherwise to concentrate their support behind the candidate most to their liking. Similarly, a districting scheme that allows geographically compact minority populations an opportunity to elect their candidates of choice would achieve the same result. Thus, although it is impossible to predict with certainty whether such measures would result in a greater number of minority justices, the convention system is not narrowly tailored to serve the State's interest in racial diversity.
Second, Dr. Hechter's thesis about the results of direct voter participation is flatly incorrect. It is refuted by the results of direct elections for other judicial offices in the state. Moreover, though Dr. Hechter assumed voters will always vote for judges based on race or ethnicity, he observed in his lengthy report that voters generally do not know the race of the candidates. Confronted with that anomaly, Dr. Hechter's revised "best assumption would be [the voters] flip a coin, essentially make a random decision or not vote at all." Tr. 1193.
Third, a factual assumption critical to Dr. Hechter's thesis about conventions that they are a breeding ground for logrolling among delegates is simply incorrect. As discussed above, I find otherwise.[40]
Finally, the defendants' flawed contentions regarding diversity are of a piece *253 with a theme they sounded throughout these proceedings: that the voters cannot be trusted to make the right decisions. It surfaced in numerous ways. Justice Freedman testified that the current system "insulate[s]" Supreme Court Justices from voter retribution based on unpopular decisions. Tr. 1776. Justice Abdus-Salaam suggested that the system increases the likelihood in post-September 11 New York that qualified persons with Muslim names will ascend to and remain on the Supreme Court bench. Tr. 1889. More direct participation by the voters would mean they would "actually have to listen to political speeches . . . and discern between different candidates competing for the party's nomination and make a decision." Tr. 214. They would have to educate themselves, which is unnecessary under the current system. Tr. 1193.
These concerns, and others expressed by the defendants as well, are justifications for eliminating elections as a method for selecting judicial officers, not for depriving the voters of their rights where elections are required. Does increased voter participation mean candidates may have to "pander[][to] individuals or corporations who will support their candidacy," Tr. 219, and that voters may "go into the voting booth not knowing anything about the candidate," or vote based on an "adverse reaction to [an] unpopular decision?" Tr. 216. I doubt it, as those evils do not appear to have afflicted the primary or general elections for New York's other elective judicial offices. But if they do, the answer is an appointive system for judges. And under New York's Constitution, the system must be elective.
Along those lines, the defendants' last argument in support of the convention system is that it "ensures that qualified incumbent Supreme Court justices remain on the bench without having to (a) expend significant sums of money for a primary campaign, (b) re-engage in politics after being politically inactive for fourteen years, and (c) fear adverse consequences of making politically unpopular but correct decisions." Def. Proposed Conclusions of Law 47. These interests in protecting the independence of sitting judges from the potential conflicts of interest inherent in soliciting political contributions and from the sways of public opinion are undoubtedly compelling. As Justice Stevens recently explained, the creed of the judicial officer has long been, in the words of Sir Matthew Hale:
That popular opinion or court applause or distaste have no influence in anything I do, in point of distribution of justice. . . .
Not to be solicitous what men will say or think, so long as I keep myself exactly to the rule of justice.
Rep. Party of Minnesota v. White, 536 U.S. at 798, 122 S.Ct. 2528 (Stevens, J., dissenting) (quoting 2 J. Campbell, Lives of the Chief Justices of England 208 (1873) (quoting Hale's Rules For His Judicial Guidance, Things Necessary to be Continually Had in Remembrance)).
Moreover, the convention system, shielded from the voters as it is, furthers these interests quite well. The evidence is that incumbents, should they wish to continue serving, are re-nominated as a matter of course at each judicial convention,[41] and all candidates, not just incumbents, are spared the expense of having to campaign in a primary. In addition, the convention system narrows the number of people whose favor a judicial candidate must court from many thousands of voters to no *254 more than 300 or so delegates and alternates, and even conveniently compresses the timeframe in which such contacts may be made.
Whether a state that has established a system for electing judges has the authority systematically to remove the voters from the determinative elements of that process in the name of judicial independence and impartiality presents a question of considerable difficulty, upon which the Supreme Court has recently shed some light. In Republican Party of Minnesota v. White, the Court struck down a Minnesota law prohibiting judicial candidates from "announcing their views on disputed legal and political issues" during the course of campaigning for office. 536 U.S. at 788, 122 S.Ct. 2528. Noting that "opposition [to having elected judges] may be well taken (it certainly had the support of the Founders of the Federal Government)," the Court nonetheless held that "the First Amendment does not permit [the state] to achieve its goal [of judicial independence] by leaving the principle of elections in place while preventing candidates from discussing what the elections are about." Id. at 787-88, 122 S.Ct. 2528. Disagreeing with the Court's "unilocular, `an election is an election"' approach, four justices were persuaded by arguments like those the defendants advance here. The dissenting justices would have permitted states greater flexibility to fashion judicial elections, "differentiat[ing] between elections for political offices, in which the First Amendment holds full sway, from elections designed to select those whose office it is to administer justice without respect to persons." 536 U.S. at 805, 122 S.Ct. 2528 (Ginsburg, J., dissenting). The Court rejected that proposal, concluding that "the greater power to dispense with elections altogether does not include the lesser power to conduct elections under conditions of state-imposed voter ignorance." 536 U.S. at 788, 122 S.Ct. 2528 (internal quotation marks omitted). Justice O'Connor, writing separately, dismissed the State's asserted interest in judicial independence out of hand: "If the State has a problem with judicial impartiality, it is largely one the State brought upon itself by continuing the practice of popularly electing judges." 536 U.S. at 792, 122 S.Ct. 2528 (O'Connor, J., concurring).
In my view, Republican Party of Minnesota stands for a principle of transparency. A State may not choose to have judicial elections and then stifle the electoral process, whether by suppressing the speech of candidates or, as in this case, by creating electoral practices that effectively keep candidates out of contention entirely, in the name of protecting the judicial office from politics. A state may decide whether or not voters will be the best choosers of judges. But it may not say one thing "The justices of the supreme court shall be chosen by the electors," N.Y. Const. art. VI, § 6(c) and do quite another, as they have here by effectively transferring the power to choose to major party leaders. Put simply, as applied to this case, the state may not pass off the will of party leaders as the will of the people.[42] Because that is exactly what the New York judicial convention system does, it violates the First Amendment.
I have assumed in this discussion that the defendants do not advance the justification of incumbent protection for the sake of incumbent protection itself, but rather for the broader goals they say incumbent protection is meant to accomplish, namely, protecting judicial independence and impartiality. Insofar as the defendants consider *255 incumbent protection desirable apart from those values, the interest is of course much less compelling, and, in any event, the convention system is not narrowly tailored to advance it. Allowing incumbents to stand for "unopposed retention elections in which voters are asked whether the judges should be recalled," Republican Party of Minnesota, 536 U.S. at 791, 122 S.Ct. 2528 (O'Connor, J., concurring), would serve that interest just as well. As Justice Freedman observed, retention elections would not be, in substance, a big change from the current system, in which sitting justices typically receive cross-endorsements at re-election time.
I also reject the argument, which the defendants made throughout the course of the proceedings but abandoned in their proposed conclusions of law, that the convention system is justifiable because it is less expensive for judicial candidates than a direct primary system would be. Although it is of course true that persuading voters in a competitive race can be an expensive proposition, again, that expense follows from the choice to have an elected judiciary. A state may not choose to have elections and then design procedures to render them uncompetitive at the critical stage because having candidates communicate with voters is too costly or the fund-raising it requires too corrupting. The United States Supreme Court has thus far allowed significant leeway, at least for the federal government, to regulate campaign fundraising through limiting contributions on the ground that such reforms protect the integrity of the democratic process. See McConnell v. Fed. Election Comm'n, 540 U.S. 93, 124 S.Ct. 619, 157 L.Ed.2d 491 (2003). New York's convention system, however which keeps the voters, not just the donors, at a safe distance from the elections throws the baby out with the bath water.
I note finally that it is of no consequence that the party leaders discussed herein are themselves periodically subject to election by party members. A system in which an elected official has, as one of her many official duties, the job of selecting the members of the judiciary is the very definition of an appointive system.[43] The Constitution of New York, however, could not state more plainly that an elective system is required. N.Y. Const. art. VI, § 6(c). In any event, New York law does not provide county and district leaders any official role in the nomination process (unless they happen to have also been elected as delegates). Therefore, voters cannot be expected to provide any check on party leaders' judicial selections by voting those leaders out of office for exercising poorly a discretion with which they are not vested.
* * * * * *
Based on the substantial body of evidence before me, I conclude that the plaintiffs have made a compelling showing that the New York system is designed to freeze the political status quo, in which party leaders, rather than the voters, select the Justices of the Supreme Court. By preventing competition among candidates and deterring voter participation, the system is successful in fact at achieving that goal.
The choice of a permanent remedy for this constitutional violation does not fall to me, but rather to the legislature of New York State. I have no authority to direct the legislature to take up the matter immediately, as the plaintiffs have requested. In order to afford the plaintiffs relief, therefore, a temporary remedy, lasting only until the legislature enacts a new *256 method of electing Supreme Court Justices, must be imposed. I conclude that the least intrusive course is to require a direct primary election, the nomination mechanism currently in place for numerous elective judicial posts. See also N.Y. Elec. Law § 6-110 ("All other party nominations of candidates for offices to be filled at a general election, except as provided for herein, shall be made at the primary election."). This method preserves for the major parties an official role in the process, and of course permits party leaders and their organizations to work to support the candidates of their choosing. The chief virtue of the primary system, however, is that it "assure[s] that intraparty competition [will be] resolved in a democratic fashion." Cal. Democratic Party v. Jones, 530 U.S. at 572, 120 S.Ct. 2402.
I have considered what number of signatures should be required for designation upon the primary ballot in each judicial district, and I am inclined to follow the structure of Election Law § 6-136(2).[44] Because I have not yet had the benefit of briefing from the parties on the matter, however, I will defer a final decision as to the appropriate petitioning requirements until the parties have had an opportunity to submit letter briefs. They shall do so on or before February 3, 2006. Each side may reply to the other's submission on or before February 8, 2006.
CONCLUSION
For the reasons set forth above, the plaintiffs' application for a preliminary injunction is granted. Because I find the plaintiffs are likely to succeed on their First Amendment claim, which fully supports the relief they request, I need not consider their equal protection claim at this time.
The defendants are hereby enjoined from enforcing New York Election Law § 6-106, and from using the procedures set forth in § 6-124. With those provisions inoperative, the nomination of Supreme Court Justices shall be by primary election until the legislature of the State of New York enacts a new statutory scheme. After reviewing counsels' submissions regarding petitioning requirements, a supplemental order will issue on that subject.
So ordered.
*257
NOTES
[1] The others are Alabama, Illinois, Louisiana, Ohio, Pennsylvania, Tennessee, Texas and West Virginia. In Illinois and Pennsylvania, the judges are initially elected in partisan elections but face only retention elections after their first term.
[2] The term "party" in § 6-106 and the phrase "major party" as used in this decision refer only to the political parties that enjoy "party" status as defined in New York Election Law § 1-104(3), i.e., the political parties whose gubernatorial candidate received at least 50,000 votes in the last election. There are currently five such parties: Republican, Democratic, Independence, Conservative and Working Families. Political parties that have not met that 50,000 vote threshold in the most recent race for governor are termed "independent bodies" under New York law. See New York Election Law § 1-104(12).
[3] New York State's Assembly and Senate candidates, as well as candidates for Civil Court (in New York City), City Courts (in upstate cities such as Rochester and Albany), and County Courts (outside New York City), all may obtain a place on the primary ballot by gathering signatures directly from voters. See N.Y. Elec. Law §§ 6-104, 6-108, 6-110, 6-136, 6-168. Similarly, in statewide elections for U.S. Senator, Governor, Lieutenant Governor, Attorney General, and State Comptroller, candidates can obtain a place on the Democratic or Republican Party primary election ballot by petitioning voters directly, as well as by obtaining votes at a convention. Id.
For statewide offices in New York, the major parties hold conventions in June. Candidates obtaining at least 25% of the convention delegates' votes are automatically placed on the primary ballot. The 25% threshold ensures that even if one candidate has the support of a majority of state committee members at the convention, a challenger candidate can still earn a place on the ballot at the party's primary election. Furthermore, as mentioned above, even if a statewide candidate fails to obtain at least 25% of the convention votes, she can still earn a place on the primary ballot through direct petitioning after the June convention.
[4] I use the phrase "county leader" in this opinion to refer to the person who exercises the executive authority over his or her political party in the relevant county. The position can have different titles.
[5] In New York City, two district leaders per AD are elected to two-year terms by the registered major party members on primary day. See N.Y. Elec. Law § 2-110. Outside the city, district leader positions exist if the rules of the county committee so provide. Id.
[6] The 12 districts (and the counties they include) are as follows: 1st Judicial District (New York); 2nd Judicial District (Kings, Richmond); 3rd Judicial District (Albany, Columbia, Greene, Rennselaer, Schoharie, Sullivan, Ulster); 4th Judicial District (Clinton, Essex, Franklin, Fulton, Hamilton, Montgomery, St. Lawrence, Saratoga, Schenectady, Warren, Washington); 5th Judicial District (Herkimer, Jefferson, Lewis, Oneida, Onondaga (including Syracuse), Oswego); 6th Judicial District (Broome, Chemung, Chenango, Cortland, Delaware, Madison, Otsego, Schulyer, Tioga, Tompkins); 7th Judicial District (Cayuga, Livingston, Monroe (including Rochester), Ontario, Seneca, Steuben, Wayne, Yates); 8th Judicial District (Allegany, Cattaraugus, Chautauqua, Erie (including Buffalo), Genesee, Niagara, Orleans, Wyoming); 9th Judicial District (Dutchess, Orange, Putnam, Rockland, Westchester); 10th Judicial District (Nassau, Suffolk); 11th Judicial District (Queens); and 12th Judicial District (Bronx). N.Y. Jud. Law § 140.
[7] Conventions; judicial
A judicial district convention shall be constituted by the election at the preceding primary of delegates and alternate delegates, if any, from each assembly district or, if an assembly district shall contain all or part of two or more counties and if the rules of the party shall so provide, separately from the part of such assembly district contained within each such county. The number of delegates and alternates, if any, shall be determined by party rules, but the number of delegates shall be substantially in accordance with the ratio, which the number of votes cast for the party candidate for the office of governor, on the line or column of the party at the last preceding election for such office, in any unit of representation, bears to the total vote cast at such line or column in the entire state. The number of alternates from any district shall not exceed the number of delegates therefrom. The delegates certified to have been elected as such, in the manner provided in this chapter, shall be conclusively entitled to their seats, rights and votes as delegates to such convention. When a duly elected delegate does not attend the convention, his place shall be taken by one of the alternates, if any, to be substituted in his place, in the order of the vote received by each such alternate as such vote appears upon the certified list and if an equal number of votes were cast for two or more such alternates; the order in which such alternates shall be substituted shall be determined by lot forthwith upon the convening of the convention. If there shall have been no contested election for alternate, substitution shall be in the order in which the name of such alternate appears upon the certified list, and if no alternates shall have been elected or if no alternates appear at such convention, then the delegates present from the same district shall elect a person to fill the vacancy.
N.Y. Elec. Law § 6-124. At present, the Democratic Party in New York State does not divide Assembly Districts for purposes of electing delegates and alternates, except where the AD crosses judicial district lines. The Republican Party divides ADs along county lines, as well as along judicial district lines, for purposes of electing delegates and alternates.
[8] As noted above, if elected delegates fail to attend the convention, alternates elected from the same AD are called upon to serve as voting delegates. Absentee rates for judicial delegates are high. The rate for 16 conventions outside New York City between 1995 and 2002 was almost one-third. Within New York City, in 1999 and 2000, the two years for which data are available, the absentee rate for Democratic Party judicial conventions was 24.5%. For the three Republican Party conventions between 1999 and 2002 for which data are available, the average absentee rate was 69.1%.
[9] Strictly speaking, the signature requirement is the lesser of 500 or five percent of the voters enrolled in the party in the AD. See New York Election Law § 6-136(3). However, all parties agree that the 500-signature requirement is the only one at issue in the case.
[10] For example, in 1971, an insurgent slate of delegates was elected in the 67th AD. (There has not been a contested election of delegates in that AD since.) The 66th AD, in Greenwich Village, contains three rival Democratic Clubs, which commonly have fielded rival slates of delegates. And except in the occasional years when "for strange reasons peace breaks out in the 68th" AD in East Harlem, rival clubs there run separate slates as well. Tr. 47. Less common are similar skirmishes among rival Democratic clubs in the 64th, 69th and 75th ADs, and long ago in the 74th AD.
[11] First District: From 1999 through 2003, there were 518 delegate positions in the Democratic Party. Although this district produced more contested elections for delegates than any other, only 212 (40.9%) were filled through contested elections. There were no years in which a majority of the delegate positions were filled through contested elections. In the same time period, there were 241 Republican delegate positions. None of the Republican delegates were elected in contested elections. Second District: From 1999 through 2003, the Democratic Party had 670 delegate positions. Only 65 (9.7%) were filled through contested elections. In the same time period, of the 442 Republican delegate positions, only 8 (1.8%) of the Republican delegates were elected in contested primaries. Third District: From 1999 through 2002, no contests for judicial delegates or alternates took place in Albany County, which includes 35.4% of the voting-age population of the Third Judicial District. Fourth District: Justice Joseph Sise testified that he was not aware of any contested primaries for judicial delegates in the Fourth District. Seventh District: Judge John Regan testified that his effort to run delegate candidates in 1994 was "the only contested delegate race within the Seventh District of which I am aware in the many years I have been in Rochester." Regan Decl. ¶ 23; Tr. 396-97. Eighth District: With the exception of 2000 and 2004, there were no contested elections for delegates in this district at least from 1995 through 2004. Ninth District: Benjamin Ostrer testified that there has never been a contested delegate or alternate race in this district. Tr. 1407. Tenth District: There were no contested delegate or alternate delegate elections for either party between 1999 and 2002. Eleventh District: From 1999 through 2003, only 27 (5.4%) the Democratic Party's 498 delegate positions were filled through contested elections. In the same time period, there were 338 Republican delegate positions, four of which were elected in contested primaries. Twelfth District: From 1999 through 2003, there were 340 delegate positions in the Democratic Party. Of those, 87 (25.6%) were filled through contested elections. In the same time period, none of the 137 Republican delegate positions involved a contested primary.
[12] After first describing the nomination process for Supreme Court Justice as a "roadblock" between the uninformed general public and judicial elections, Tr. 2046, Allen testified that (1) the process involves judicial candidates seeking the support of convention delegates; (2) he was not familiar with the fact that the Democratic County Leader, New York State Assemblyman Herman "Denny" Farrell, Jr., presents a "package" of candidates to the convention; and (3) Farrell exercises no leadership with respect to the candidates who become nominated. Tr. 2047-53. I credit the roadblock testimony and specifically reject the other listed assertions. I further note that Allen's statement that Farrell is "just a delegate like everybody else," Tr. 2051, is inconsistent with the testimony of every other witness in the case who had knowledge of First District nominations.
[13] See Tr. 621-22 (López-Torres) (overhearing such a direction); Tr. 233-34 (Berger) (giving one).
[14] I do not mean to suggest that Senator Connor's blunt assessment of the candidate he found offensive in 2002 was accurate. Indeed, there is no evidence at all on that issue. The point is not whether Senator Connor's evaluation was correct, but that he felt constrained to keep it to himself.
[15] By contrast, the party conventions that designate candidates for statewide office are held in June, but the identities of the delegates to those conventions (i.e., the state party committee members) are known for at least eight months prior to the conventions. Candidates for these offices thus have ample opportunity to contact, lobby, and obtain commitments from party delegates if they are not assured of sufficient support by party leaders.
[16] Implicitly recognizing the implausibility of the "subtle process" described by Schiff, defense counsel tried gamely, albeit with limited success, to get him to enlarge the time frame within which the alleged lobbying of delegates occurs. See Tr. 1308-12.
As a practical matter, an aspiring Supreme Court Justice can often determine, with a few hours' effort at the Board of Elections, the identities of the potential delegates (and alternates) after the second week of July, when the nominating petitions must be filed at the Board of Elections. Since most delegate slates are unopposed, the pool of delegates is usually identifiable at that time. This additional time is of no avail to a diligent challenger candidate. Putting aside the other impediments to lobbying described in the text, these extra weeks in the height of the summer do not allow any effective lobbying. Indeed, the Democratic organization in the First District, after experimenting with different screening panel reporting dates in the 1980s, fixed on a post-Labor Day date for the explicit purpose of narrowing the lobbying time frame to a two- or three-week period after the summer vacation period was over. See Tr. 1672-73.
[17] Pursuant to an order dated September 1, 2004, on the consent of all parties, the direct testimony of most witnesses was presented at the hearing by having the witness adopt his or her declaration. See also Tr. 439 (declarations of testifying witnesses are evidence). Exceptions to this procedure occurred for threshold questions, such as the qualifications of an expert (see, e.g., Tr. 18-38 (Berger)), and for any amendments to the declarations.
The treatment of hearsay had a tortured history during the hearing. The defendants made hearsay objections to the declarations of the plaintiffs' fact witnesses. Initially, that produced awkward examinations in which witnesses were probed on cross-examination as to specific statements in their declarations. See, e.g., Tr. 355-60. That, in turn, produced an oral direction to present, in advance of a witness' testimony, a marked copy of his or her declaration identifying the challenged statements. See Tr. 360-61; 405-06.
It soon became clear that the defendants were relying heavily (and properly, in my view) on the same sort of hearsay to which they had been objecting. As a result, I suggested that their time-consuming parsing of the plaintiffs' witness declarations for hearsay might not be useful. See Tr. 474-75; see also Tr. 612. Plaintiffs finally raised the argument that the hearsay rule does not apply during a hearing on a motion for preliminary injunction. See Tr. 595-97. The issue was briefed and I ruled in their favor. Tr. 981.
[18] Justice Gangel-Jacob was not supported initially because she had antagonized Farrell by opposing and defeating the organization's preferred candidate for Civil Court in 1984. The 1984 events are additional evidence that a challenger candidate can receive her party's nomination if given an opportunity to appeal directly to the voters in a primary election.
[19] The defendants accuse the plaintiffs of being "disingenuous and sexist" in ascribing significance to the fact that Justice Gangel-Jacob's husband was a district leader. Def. Reply Findings at 31-32. The accusation is unfair. Where, as here, a party's factual thesis is that the county leader and his or her district leaders not the voters or convention delegates select nominees, it is obviously permissible to suggest that a critical intervening event between a candidate's 1991 failure to get the nomination and her 1993 success was her spouse's election as a district leader. That a particular candidate is a woman and married to a party leader is just as relevant as, for example, a male candidate being the brother of one. Justice Sise fits the latter description, and the defendants do not complain about the plaintiffs' reliance on that relationship.
[20] See Tr. 1754 ("I don't know that I knew so many delegates then. I just talked to people who were there, [who] may have been delegates[,] and I talked to district leaders."); Tr. 1755 (a friend obtained a list of "judicial delegates and a list of all the district leaders"); Tr. 1759 (in response to a question whether she thought delegate support would make the difference, Justice Freedman responded, "Delegate and district leader support, yes. . . ."); Tr. 1761 (At county dinner in early 1988, "I tried to talk to anyone who I thought might be a delegate. I spoke to district leaders.").
[21] See, e.g., Tr. 1751, 1754 (attending fundraisers beginning nine months before the election).
[22] As for Justice Sheila Abdus-Salaam's testimony, there are insufficient facts in the record about her nomination in 1993 to make reliable factfindings in this regard. Farrell called her to say he would not support her, and then he called her not long afterward to say he would. In between, Justice Abdus-Salaam "started making a few calls" to delegates. Tr. 1864. These events instilled in Justice Abdus-Salaam a belief that Farrell was responding to a "grass roots groundswell" of support for her. Tr. 1865. I find that Justice Abdus-Salaam truthfully testified to her belief, but there are no facts before me to support its accuracy. Defense counsel gamely sought to fill that void with a leading question that had no basis in the evidence. Tr. 1897 (Q: "[Farrell changed his mind] because you were able to get support throughout the city, the upper west side, the upper east side, Chelsea, the Village, not in your home assembly district[], is that true?"). The Justice's polite response, see id. ("That is apparently so"), is not a substitute for evidence, and there is no evidence of who she called, where they were from, or why or how quickly Farrell changed his mind.
[23] This was unclear from the testimony. Senator Connor testified both that the 77 delegates were polled individually and that the whole process took no more than two seconds, explaining that "in the New York State Senate you could call 77 senators' name[s] and pass 50 bills in 15 minutes." Tr. 2226.
[24] See New York Election Law 6-126(2) ("The convention may appoint a committee to nominate candidates to fill vacancies in nominations made by the convention and caused by the death, declination or disqualification of a candidate.").
[25] There have been faster ones. In 1996, under the supervision of Gerald P. Garson, Esq., the Democratic convention in the Second District performed the tasks listed above and nominated eight judges, all in 11 minutes.
[26] For example, in the Third District, there were ten elections during the period from 1990-2002. In five of them, including the three most recent ones, the two major parties cross-endorsed the candidates. Similarly, in the Fifth District, the two major parties cross-endorsed candidates for the single open position in 1993 and 1997, and for one of the two open seats in 1994 and 2000. Most (20 out of 34) winning candidates in the Eighth District in 1990-2002 were cross-endorsed. In the Tenth District, Republican nominees prevail more than Democratic ones, but some Democratic candidates have won.
[27] The 2002 nomination of a candidate opposed by the Democratic county leader in Erie County, discussed below, is not an exception. In that race, both candidates were backed by significant factions of a fractured county organization.
[28] France was an unsuccessful challenge to the Supreme Court Justice selection process pursuant to § 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973. See 71 F.Supp.2d 317 (S.D.N.Y.1999).
[29] There are occasional exceptions to this rule. As discussed in subsection B.1, supra, some delegates are actually selected by the votes of dues-paying members of Democratic clubs. But Farrell's control over the district leaders always affords him sufficient control over enough delegates to allow him to control nominations at the conventions.
[30] Kellner testified that county leaders also have input in the Office of Court Administration's selection of Administrative Judges. Tr. 1745.
[31] See e.g., Tr. 1768-69 (Justice Freedman) ("I assume that the county leader was somewhat involved at that point, but I'm not sure what how much he controlled it or didn't control it at that point."); Tr. 1260 (district leader Schiff) ("I'm really not certain exactly what involvement the county leader has in the process.")
[32] See Report to the Chief Judge of the State of New York by the Commission to Promote Public Confidence in Judicial Elections (June 27, 2004), at 14 & App. G-3. The report is in evidence as Pl. Ex. 78.
[33] Lopez's daughter was hired by another Civil Court judge in Brooklyn, who was subsequently nominated and elected to the office of Supreme Court Justice.
[34] While she was petitioning to get herself on the primary ballot for Civil Court in 2002, López Torres tried to include on her petitions slates of delegates to the judicial convention as well. It was a difficult task. Other elected officials who appeared with her on joint petitions and who supported her for Civil Court were reluctant to support her delegate states because Supreme Court Justice positions were supposed to be "decided by the district leader and the County leader." Tr. 695. López Torres amassed many signatures (about 30,000), but they were virtually all from only eight ADs. Only 17 of the 47 delegates on the petitions, 15 of whom (from two ADs) ran without opposition, were elected.
The defendants' claim that López Torres simply didn't try hard enough to obtain the nomination (in 2002 and other years), like their attack on her qualifications, is meritless.
[35] López Torres was again a candidate for Supreme Court Justice in 2004, but she was not nominated at the convention.
[36] As Douglas Kellner, one of the defendants' experts, acknowledged, other districts in New York operate in precisely the "smoke-filled room" manner that existed in Manhattan before reformers instituted the First District's screening panel. Tr. 1630-31.
[37] The defendants' argument that "because there is no injury," the plaintiffs lack standing suffers from the same circularity. See Def. Proposed Conclusions of Law 9-10. I reject it in like fashion. Voters have standing when they "are asserting a plain, direct, and adequate interest in maintaining the effectiveness of their votes." Baker v. Carr, 369 U.S. 186, 208, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962) (internal quotation marks omitted). A former and would-be candidate such as López-Torres has standing to challenge ballot access requirements. See, e.g., Storer v. Brown, 415 U.S. 724, 738 n. 9, 94 S.Ct. 1274, 39 L.Ed.2d 714 (1974) (candidates have "ample standing to challenge the signature requirement").
I note also that, contrary to the defendants' assertion, this case falls within a long-recognized exception to the doctrine that a plaintiff bringing a facial challenge must "establish that no set of circumstances exists under which the [statute] would be valid." United States v. Salerno, 481 U.S. 739, 745, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987); see also Lerman v. Bd. of Elections, 232 F.3d 135, 144 (2d Cir.2000). Because "the plaintiffs challenge a [statutory scheme] regulating the ability to engage in interactive political speech and association activity," the overbreadth doctrine applies, under which "the plaintiffs need only demonstrate a substantial risk that application of the provision will lead" to a deprivation of First Amendment right. Id.
[38] I note that nothing in the plaintiffs' case seeks to prohibit the party leadership from publicly endorsing and working to support its chosen candidates. See Eu v. San Francisco County Democratic Cent. Comm., 489 U.S. 214, 109 S.Ct. 1013, 103 L.Ed.2d 271 (1986).
[39] Professor Richard Pildes has described cumulative voting as follows:
[E]ach voter is given as many votes to cast as there are seats to be filled. Voters are free to distribute their votes among candidates in any way they choose. This approach enables voters to express not just their raw preferences, but the intensity with which those preferences are held. In a five-way race, for example, a voter can cast one vote for each candidate, vote three times for one candidate and twice for a second, or cast all his votes for one candidate. In this way, minority groups with common interests and strong preferences for a particular candidate can ensure her election, even in the face of a hostile majority.
Richard H. Pildes, Gimme Five: Non-gerrymandering Racial Justice, The New Republic (March 1, 1993).
[40] In other respects besides diversity, Dr. Hechter's conclusions support the plaintiffs more than the defendants. For example, he concluded that:
The consequences of the New York judicial convention are well-appreciated. In politically unified districts it is difficult to attain the nomination without the support of the party leadership. This high barrier to entry significantly narrows the pool of candidates. Party leaders have great influence over the nomination process when their party is unified. Indeed, judicial conventions generally take little time. Decisions are often unanimous. Since insurgent candidates are only likely to be nominated when the party is internally divided, rates of incumbency are high.
Pl. Ex. 69 at ¶ 9 (emphasis added). Dr. Hechter also concluded that (i) "district leaders effectively select delegates to the judicial convention," see ¶¶ 60-63, and (ii) "county leaders are considerably more influential than district leaders," although their influence varies across boroughs. Id. at ¶¶ 64-71.
[41] Indeed, the current system may protect these interests too much. As discussed above, the county leader in 2004 in the Second District required the convention to re-nominate a sitting justice that the recently-strengthened screening panel had found unqualified.
[42] Even the dissenters in Republican Party of Minnesota "recogniz[ed] that the influence of political parties is incompatible with the judge's role." 536 U.S. at 804, 122 S.Ct. 2528 (Ginsburg, J., dissenting).
[43] Delegates to a convention are not elected officials in this sense, as they have no other duty than to select judges.
[44] Under that scheme, because some judicial districts are comprised of "more than one . . . county," § 6-136(2)(k), but no counties are divided between two or more judicial districts, candidates would be required to collect a number of signatures equal to "five per centum, as determined by previous enrollment, of the then enrolled voters of the party residing with the [judicial district]," § 6-136(2), but "not to exceed the aggregate of the signatures required" for the counties contained in the judicial district, § 6-136(2)(k). Thus, in the First District, which contains only New York County, the maximum number of required signatures would be four thousand under § 6-136(2)(b), which applies to "any office to be filled by all the voters of any county . . . within the city of New York." The same goes for the Eleventh and Twelfth Districts, which are co-terminous with Queens and Bronx Counties, respectively. In the Second District, which is comprised of Kings and Richmond Counties, eight thousand signatures would be required. In the remaining districts, the maximum number of required signatures would depend upon the population of each county ("according to the last preceding federal enumeration") in the district under 6-136(2)(d), (e), and (f) (i.e., for counties containing more than 250,000 inhabitants, 2000 signatures; for counties containing less than 250,000 but more than 25,000 inhabitants, 1000 signatures; for counties containing less than 25,000 inhabitants, 500 signatures).