with whom Justice Blackmun and Justice Rehnquist join, dissenting.
Under Wisconsin law, the Wisconsin delegations to the Presidential nominating conventions of the two major political parties are required to cast their votes in a way that *127reflects the outcome of the State’s “open” primary election. That election is conducted without advance party registration or any public declaration of party affiliation, thus allowing any registered voter to participate in the process by which the Presidential preferences of the Wisconsin delegation to the Democratic National Convention are determined. The question in this case is whether, in light of the National Party’s rule that only publicly declared Democrats may have a voice in the nomination process, Wisconsin’s open primary law infringes the National Party’s First Amendment rights of association. Because I believe that this law does not impose a substantial burden on the associational freedom of the National Party, and actually promotes the free political activity of the citizens of Wisconsin, I dissent.
I
The Wisconsin open primary law was enacted in 1903. 1903 Wis. Laws, ch. 451. It was amended two years later to apply to Presidential nominations. 1905 Wis. Laws, ch. 369. See 93 Wis. 2d 473, 492, 287 N. W. 2d 519, 527 (1980). As the Wisconsin Supreme Court described in its opinion below:
“The primary was aimed at stimulating popular participation in politics thereby ending boss rule, corruption, and fraudulent practices which were perceived to be part of the party caucus or convention system. Robert M. La Follette, Sr., supported the primary because he believed that citizens should nominate the party candidates; that the citizens, not the party bosses, could control the party by controlling the candidate selection process; and that the candidates and public officials would be more directly responsible to the citizens.” Ibid.
As noted in the opinion of the Court, the open primary law only recently has come into conflict with the rules of the National Democratic Party. The new Rule 2A was enacted *128as part of a reform effort aimed at opening up the party to greater popular participation. This particular rule, however, has the ironic effect of calling into question a state law that was intended itself to open up participation in the nominating process and minimize the influence of “party bosses.”
II
The analysis in this kind of First Amendment case has two stages. If the law can be said to impose a burden on the freedom of association, then the question becomes whether this burden is justified by a compelling state interest. E. g., Bates v. Little Rock, 361 U. S. 516, 524 (1960). The Court in this case concludes that the Wisconsin law burdens associational freedoms. It then appears to acknowledge that the interests asserted by Wisconsin are substantial, ante, at 120-121, but argues that these interests “go to the conduct of the Presidential preference primary — not to the imposition of voting requirements upon those who, in a separate process, are eventually selected as delegates,” ante, at 125. In my view, however, any burden here is not constitutionally significant, and the State has presented at least a formidable argument linking the law to compelling state interests.
A
In analyzing the burden imposed on associational freedoms in this case, the Court treats the Wisconsin law as the equivalent of one regulating delegate selection, and, relying on Cousins v. Wigoda, 419 U. S. 477 (1975), concludes that any interference with the National Party’s accepted delegate-selection procedures impinges on constitutionally protected rights. It is important to recognize, however, that the facts of this case present issues that differ considerably from those we dealt with in Cousins.
In Cousins, we reversed a determination that a state court could interfere with the Democratic Convention’s freedom to *129select one delegation from the State of Illinois over another. At issue in the case was the power of the National Party to reject a delegation chosen in accordance with state law because the State’s delegate-selection procedures violated party rules regarding participation of minorities, women, and young people, as well as other matters. See id., at 479, n. 1. The state court had ordered the Convention to seat the delegation chosen under state law, rather than the delegation preferred by the Convention itself. In contrast with the direct state regulation of the delegate-selection process at issue in Cousins, this case involves a state statutory scheme that regulates delegate selection only indirectly. Under Wisconsin law, the “method of selecting the delegates or alternates [is] determined by the state party organization,” Wis. Stat. § 8.12 (3)(b) (1977). Wisconsin simply mandates that each delegate selected, by whatever procedure, must be pledged to represent a candidate who has won in .the state primary election the right to delegate votes at the Convention.1
In sum, Wisconsin merely requires that the delegates “vote in accordance with the results of the Wisconsin open primary.” Ante, at 126. While this regulation affecting participation in the primary is hardly insignificant, it differs substantially from the direct state interference in delegate selection at issue in Cousins. This difference serves to emphasize the importance of close attention to the way in which a state law is said to impose a burden on a party’s freedom of association. Cf. Marchioro v. Chaney, 442 U. S. 191, 199 (1979). All that Wisconsin has done is to require the major parties to allow voters to affiliate with them — for the limited purpose of participation in a primary — secretly, in the pri*130vacy of the voting booth.2 The Democrats remain free to require public affiliation from anyone wishing any greater degree of participation in party affairs. In Wisconsin, participation in the caucuses where delegates are selected is limited to publicly affiliated Democrats. Brief for Appellee Democratic Party of Wisconsin 19. And, as noted above, the State’s law requires that delegates themselves affirm their membership in the party publicly.
In evaluating the constitutional significance of this relatively minimal state regulation of party membership requirements, I am unwilling — at least in the context of a claim by one of the two major political parties — to conclude that every conflict between state law and party rules concerning participation in the nomination process creates a burden on associational rights. Instead, I would look closely at the nature *131of the intrusion, in light of the nature of the association involved, to see whether we are presented with a real limitation on First Amendment freedoms.
It goes without saying that nomination of a candidate for President is a principal function performed by a national political party, and Wisconsin has, to an extent, regulated the terms on which a citizen may become a “member” of the group of people permitted to influence that decision. If appellant National Party were an organization with a particular ideological orientation or political mission, perhaps this regulation would present a different question.3 In such a case, the state law might well open the organization to participation by persons with incompatible beliefs and interfere with the associational rights of its founders.
The Democratic Party, however, is not organized around the achievement of defined ideological goals. Instead, the major parties in this country “have been characterized by a fluidity and overlap of philosophy and membership.” Rosario v. Rockefeller, 410 U. S. 752, 769 (1973) (Powell, J., dissenting). It can hardly be denied that this Party generally has been composed of various elements reflecting most of the American political spectrum.4 The Party does take positions *132on public issues, but these positions vary from time to time, and there never has been a serious effort to establish for the Party a monolithic ideological identity by excluding all those with differing views. As a result, it is hard to see what the Democratic Party has to fear from an open primary plan. Wisconsin’s law may influence to some extent the outcome of a primary contest by allowing participation by voters who are unwilling to affiliate with the Party publicly. It is unlikely, however, that this influence will produce a delegation with preferences that differ from those represented by a substantial number of delegates from other parts of the country. Moreover, it seems reasonable to conclude that, insofar as the major parties do have ideological identities, an open primary merely allows relatively independent voters to cast their lot with the party that speaks to their present concerns.5 *133By attracting participation by relatively independent-minded voters, the Wisconsin plan arguably may enlarge the support for a party at the general election.
It is significant that the Democratic Party of Wisconsin, which represents those citizens of Wisconsin willing to take part publicly in Party affairs, is here defending the state law. Moreover, the National Party’s apparent concern that the outcome of the Wisconsin Presidential primary will be skewed cannot be taken seriously when one considers the alternative delegate-selection methods that are acceptable to the Party under its rules. Delegates pledged to various candidates may be selected by a caucus procedure involving a small minority of Party members, as long as all participants in the process are publicly affiliated. While such a process would eliminate “crossovers,” it would be at least as likely as an open primary to reflect inaccurately the views of a State’s Democrats.6 In addition, the National Party apparently is quite willing to accept public affiliation immediately before primary voting, which some States permit.7 As Party affiliation becomes this easy for a voter to change in order to participate in a particular primary election, the difference between open and closed primaries loses its practical significance.8
*134In sum, I would hold that the National Party has failed to make a sufficient showing of a burden on its associational rights.9
B
The Court does not dispute that the State serves important interests by its open primary plan. Instead the Court argues that these interests are irrelevant because they do not support a requirement that the outcome of the primary be binding on delegates chosen for the convention. This argument, however, is premised on the unstated assumption that a nonbinding primary would be an adequate mechanism for pursuing the state interests involved. This assumption is unsupportable because the very purpose of a Presidential primary, as enunciated as early as 1903 when Wisconsin passed its first primary law, was to give control over the nomination process to individual voters.10 Wisconsin cannot do this, and still pursue the interests underlying an open primary, without making the open primary binding.11
*135If one turns to the interests asserted, it becomes clear that they are substantial. As explained by the Wisconsin Supreme Court:
“The state’s interest in maintaining a primary and in not restricting voting in the presidential preference primary to those who publicly declare and record their party preference is to preserve the overall integrity of the electoral process by encouraging increased voter participation in the political process and providing secrecy of the ballot, thereby ensuring that the primary itself and the political party’s participation in the primary are conducted in a fair and orderly manner.
“In guaranteeing a private primary ballot, the open primary serves the state interest of encouraging voters to participate in selecting the candidates of their party which, in turn, fosters democratic government. Historically the primary was initiated in Wisconsin in an effort to enlarge citizen participation in the political process and to remove from the political bosses the process of selecting candidates.” 93 Wis. 2d, at 512-513, 287 N. W. 2d, at 536-537 (footnote omitted).
The State’s interest in promoting the freedom of voters to affiliate with parties and participate in party primaries has been recognized in the decisions of this Court. In several cases, we have dealt with challenges to state laws restricting voters who wished to change party affiliation in order to participate in a primary. We have recognized that voters have a right of free association that can be impaired unconstitutionally if such state laws become too burdensome. In Rosario v. Rockefeller, 410 U. S. 752 (1973), the Court upheld a *136registration time limit, but emphasized that the law did not absolutely prevent any voter from participating in a primary and was “tied to a particularized legitimate purpose” of preventing “raiding,” 12 id., at 762. In Kusper v. Pontikes, 414 U. S. 51 (1973), we struck down an Illinois law that prevented voters who had participated in one party’s primary from switching affiliations to vote in another party’s primary during the succeeding 23 months. We concluded that such a law went too far in interfering with the freedom of the individual voter, and could not be justified by the State’s interest in preventing raiding.
Here, Wisconsin has attempted to ensure that the prospect of public party affiliation will not inhibit voters from participating in a Democratic primary. Under the cases just discussed, the National Party’s rule requiring public affiliation for primary voters is not itself an unconstitutional interference with voters’ freedom of association. Nader v. Schaffer, 417 F. Supp. 837 (Conn.) (three-judge court), summarily aff’d, 429 U. S. 989 (1976). But these cases do support the State’s interest in promoting free voter participation by allowing private party affiliation. The State of Wisconsin has determined that some voters are deterred from participation by a public affiliation requirement,13 and the validity of that concern is not something that we should second-guess.14
*137III
The history of state regulation of the major political parties suggests a continuing accommodation of the interests of the parties with those of the States and their citizens. In the process, “the States have evolved comprehensive, and in many respects complex, election codes regulating in most substantial ways, with respect to both federal and state elections, the time, place, and manner of holding primary and general elections, the registration and qualifications of voters, and the selection and qualification of candidates.” Storer v. *138Brown, 415 U. S. 724, 730 (1974).15 Today, the Court departs from this process of accommodation. It does so, it seems to me, by upholding a First Amendment claim by one of the two major parties without any serious inquiry into the extent of the burden on associational freedoms and without due consideration of the countervailing state interests.
The delegates selected must be approved by the candidate they are to represent, Wis. Stat. § 8.12 (3) (b) (1977), and must pledge that they are affiliated with the candidate’s party and will support their candidate until he or she fails to receive at least one-third of the votes authorized to be cast at the Convention, §8.12 (3)(c).
It is not fully accurate to say, as the Court does, that the “election laws of Wisconsin allow non-Democrats — including members of other parties and independents — to vote in the Democratic primary.” Ante, at 110-111. The Wisconsin statute states that “[i]n each year in which electors for president and vice president are to be elected, the voters of this state shall at the spring election be given an opportunity to express their preference for the person to be the presidential candidate of their party.” Wis. Stat. § 8.12 (1) (1977) (emphasis added). Thus, the act of voting in the Democratic primary fairly can be described as an act of affiliation with the Democratic Party. The real issue in this case is whether the Party has the right to decide that only publicly affiliated voters may participate.
The situation might be different in those States with “blanket” primaries — i. e., those where voters are allowed to participate in the primaries of more than one party on a single occasion, selecting the primary they wish to vote in with respect to each individual elective office. E. g., Wash. Rev. Code § 29.18.200 (1976). Cf. 93 Wis. 2d 473, 504, 287 N. W. 2d 519, 532 (1980) (“[T]he legislature has taken steps to encourage voters to participate in the primary of their party and to discourage a voter of one party from being tempted to vote in the primary of another party. Limiting voters to only one party’s ballot discourages voters from voting on a ballot of a party other than their own, because in order to do so they would have to sacrifice their opportunity to participate in their own party’s selection process”).
Compare NAACP v. Alabama ex rel. Patterson, 357 U. S. 449, 462-463 (1958), where the Court was careful to assess the effect of a membership disclosure requirement on associational freedoms in light of the particular nature of the organization involved and the likely responses of those opposed to its aims.
See R. Horn, Groups and the Constitution 103-104 (1956); A. Campbell, P. Converse, W. Miller, & D. Stokes, The American Voter 183-187, 543 (1960); Developments in the Law: Elections, 88 Harv. L. Rev. 1111, 1166 (1975). The Charter of the National Democratic Party states that it is “open to all who desire to support the party and ... be known as Democrats.” Art. Ten, § 1.
This perception need not be taken as a criticism of the American party structure. The major parties have played a key role in forming coalitions and creating consensus on national issues. “Broad-based political parties supply an essential coherence and flexibility to the American political *132scene. They serve as coalitions of different interests that combine- to seek national goals.” Branti v. Finkel, 445 U. S. 507, 532 (1980) (Powell, J,, dissenting). As Professor Ranney has written:
“[E]ach party has sought winning coalitions by attempting, accommodations among competing interests it hopes will appeal to more contributors and voters than will the rival accommodations offered by the opposition party. This strategy, it is conceded, has resulted in vague, ambiguous, and overlapping party programs and in elections that offer the voters choices between personalities and, at most, general programmatic tendencies, certainly not unequivocal choices between sharply different programs. But this ... is not a vice but a virtue, for it has enabled Americans through all but one era of their history to manage their differences with relatively little violence and to preserve the world’s oldest constitutional democratic regime.” A. Ranney, Curing the Mischiefs of Faction 201 (1975).
See Comment, The Constitutionality of Non-Member Voting in Political Party Primary Elections, 14 Willamette L. J. 259, 290 (1978) (“Independents and members of other parties who seek to participate in a party primary will do so precisely because they identify with the community of interest, if indeed one exists. Their very motive for participating in the primary would be to associate with a party presenting ‘candidates and issues more responsive to their immediate concerns’”), quoting Rosario v. Rockefeller, 410 U. S. 752, 769 (1973) (Powell, J., dissenting).
The unrepresentative nature of the delegate selections produced by caucuses is suggested by differences between the results of caucuses and nonbinding primaries held in the same State. See n. 11, infra.
E. g., Tenn. Code Ann. § 2-7-115 (b) (2) (1979). See Developments in the Law, supra n. 4, at 1164.
As one scholar has stated:
“The distinctions between open and closed primaries are easy to exaggerate. Too simple a distinction ignores the range of nuances and varieties within the closed primary states, which after all do account for 82 percent of the states. Take the case of Illinois. Voters do not register as members of a party; at the polling place they simply state their party preference and are given the ballot of that party, no questions asked. Because Illinois voters must disclose a party preference before entering the voting booth, their primary is generally considered 'closed.’ One would be hard put, however, to argue that it is in operation much different from an open *134primary.” F. Sorauf, Party Politics in America 206 (4th ed. 1980) (hereinafter Sorauf).
Of course, the National Party could decide that it no longer wishes to be a relatively nonideological party, but it has not done so. Such a change might call into question the institutionalized status achieved by the two major parties in state and federal law. It cannot be denied that these parties play a central role in the electoral process in this country, to a degree that has led this Court on occasion to impose constitutional limitations on party activities. See Smith v. Allwright, 321 U. S. 649 (1944); Terry v. Adams, 345 U. S. 461 (1953). Arguably, the special status of the major parties is an additional factor favoring state regulation of the electoral process even in the face of a claim by such a party that this regulation has interfered with its First Amendment rights.
See, e. g., Sorauf 204 (“it was an article of faith among [the Progressives] that to cure the ills of democracy one needed only to prescribe larger doses of democracy”).
Any argument that a nonbinding primary would be sufficient to allow individual voters a voice in the nomination process is belied by the fact that such a primary often will be ignored in later, nonprimary delegate-selection processes. In 1980, for example, Vermont’s nonbinding open pri*135mary produced a lopsided victory, 74.3% to 25.7%, for President Carter over Senator Kennedy. 38 Cong. Q. Weekly Rep. 647 (1980). Party caucuses then produced a state delegation to the Democratic Convention that favored Kennedy over Carter by 7 to 5. Id., at 1472.
“Raiding” refers to primary voting by members of another party who are seeking to encourage their opponents to select a less desirable or strong candidate. It does not appear to be a problem in Wisconsin. See 93 Wis. 2d, at 506, 287 N. W. 2d, at 533 ("The petitioner and respondents agree that raiding is not a significant problem and that neither the Wisconsin open primary nor the declaration required by Rule 2A prevents ‘raiding’ ”).
A related concern is the prevention of undue influence by a particular political organization or “machine.” The Progressives who promoted the idea of a primary election perceived a need to combat political professionals who controlled access to governmental power. See A. Lovejoy, *137La Follette and the Establishment of the Direct Primary in Wisconsin 7-8 (1941) (“avowed purpose” was “the elimination of the boss from the American political scene”); id., at 97 (“Because of their faith in the American people, the Progressives sought to cure the ills of democracy with more democracy. . . . For the first time the middleman was eliminated between the people and their representatives”); Sorauf 203-204. The open primary carries this process one step further by eliminating some potential pressures from political organizations on voters to affiliate with a particular party. Although one well may question the wisdom of a state law that undermines the influence of party professionals and may tend to weaken parties themselves, the state interests involved are neither illegitimate nor insubstantial. As noted supra, at 133, the Democratic Party of Wisconsin has filed a brief in support of the validity of the Wisconsin plan.
A more difficult question in this case is whether Wisconsin can satisfy the second component of the “compelling interest test” — whether it can show that it has no “less drastic way of satisfying its legitimate interests.” Kusper v. Pontikes, 414 U. S. 51, 59 (1973). The answer to this question depends in many cases on how the state interest is conceived. Here, a state interest in protecting voters from the possible coercive effects of public party affiliation cannot be satisfied by any law except one that allows private party affiliation. On the other hand, if the state interest is described more generally, in terms of increasing voter freedom or participation, there may well be less “drastic" alternatives available to Wisconsin. Because of my conclusion that there is no significant burden on the associational freedoms of appellant National Party in this case, and because the Court’s analysis does not reach this question, I express no view on whether the State has shown a sufficient interest in this particular method of regulating the electoral process to satisfy a less-drastic-means inquiry.
The Court concedes that the States have a substantial interest in regulating primary elections. Ante, at 124, n. 28, 126. The power of the States in this area derives from the specific constitutional grant of authority to the States to "appoint, in such Manner as the Legislature thereof may direct” Presidential electors, U. S. Const., Art. II, § 1, cl. 2, as well as from the more general regulatory powers of the States. See Cousins v. Wigoda, 419 U. S. 477, 495-496 (1975) (RehNQUist, J., concurring in result).