(dissenting).
I must respectfully dissent from the opinion of my brothers. Like them, I recognize that a weighted voting system appeared in many representative union constitutions prior to the enactment of the Labor-Management Reporting and Disclosure Act of 1959,1 and that the *31statutory provision in question “was not brought about by any abuses found to have flourished under the aegis of that system.” 2 My disagreement is only with the conclusion that the words “by majority vote of the delegates voting” were intended and must be construed to invalidate this common and concededly fair method of delegate voting at union conventions, insofar as it is used to approve dues increases.3
I think it is fair to say that the majority decision rests principally on the so-called “plain meaning” rule, i. e., that the statutory phrase is capable of but one interpretation. My brethren recognize, as noted above, that elimination of the weighted • voting system was not consciously put forward as a goal of the Act, whose main thrust was to correct the flagrant abuses of power by some union officials which had been uncovered by extensive Congressional investigations. Nor does any reason readily suggest itself for barring the weighted ballot system in connection with dues increases voted at union conventions but in no other respect. It is clear that, with regard to election of officers and, indeed, any other action taken at these conventions, the system is perfectly acceptable. Cf. § 401 (f) of the Act, 29 U.S.C. § 481(f). I can find no support for the majority’s suggestion that dues increases are a particularly “fertile field” for “skulduggery,” to which an especially strict standard should be applied.
We are left then, with the “plain meaning” rule as the only real basis for the majority decision. Conceding that the crucial phrase, taken in isolation, most naturally bears the majority’s interpretation of it, I do not believe that any other reading is absolutely barred. The term used is “majority vote of the delegates voting,” not “vote of a majority of the delegates voting.” It bears the interpretation, perhaps not without some strain, that delegates may have differing voting strength, depending on the size of the locals which they represent. Where there are compelling reasons to give such interpretation, and none to give the more apparent one, it is open to us to do so. The plain meaning rule, after all, “is rather an axiom of experience than a rule of law, and does not preclude consideration of persuasive evidence if it exists.” Boston Sand & Gravel Co. v. United States, 278 U.S. 41, 48, 49 S.Ct. 52, 54, 73 L.Ed. 170 (1928) (Holmes, J.). Again, it has been said that “words are inexact tools at best, and for that reason there is wisely no rule of law forbidding resort to explanatory legislative history no matter how clear the words may appear on 'superficial examination.’ ” Harrison v. Northern Trust Co., 317 U.S. 476, 479, 63 S.Ct. 361, 363, 87 L.Ed. 407 (1943). Of course, the “rule” has an important place as a canon of construction and I do not wish to contest the general proposition that the language actually chosen is the most reliable guide to the legislature’s intention. But it is a myopic view, and one oblivious of precedent, that shuts us off from consideration of the purposes intended to be served by legislation in a case where the “plain meaning” rule leads to incongruous and even unjust results. See, e. g., United States v. American Trucking Assn., 310 U.S. 534, 60 S.Ct. 1059, 84 L.Ed. 1345 (1940); United States v. Dickerson, 310 U.S. 554, 60 S.Ct. 1034, 84 L.Ed. 1356 (1940); United States v. N. E. Rosenblum Truck Lines, 315 U.S. 50, 62 S.Ct. 445, 86 L.Ed. *32671 (1942); N. L. R. B. v. Lion Oil Co., 352 U.S. 282, 77 S.Ct. 330, 1 L.Ed.2d 331 (1957) ; J. C. Penney Co. v. Commissioner, 312 F.2d 65 (2 Cir.1962).
Of the many provisions of the LMRDA, section 101(a) (3) was one of the least controversial. It first appeared in an amendment offered by Senator McClellan on the Senate floor to add a “Bill of Rights” to the labor reform bill which had been reported by the Senate Committee on Labor and Public Welfare (S. 1555, 86th Cong., 1st Sess.). See 105 Daily Cong. Record 5810 (April 22, 1959), reprinted in 2 N. L. R. B. Legislative History of the Labor-Management Reporting Disclosure Act of 1959, 1102 (1959).4 Senator McClellan's amendment provided that in the case of a national labor organization the rates of dues should not be changed, and no new dues, initiation fees or assessments imposed except by a majority vote of the delegates present at a duly constituted ■convention. Senator McClellan explaining this part of his amendment, said only that
In other words, the dues or the initiation fees cannot be increased, as they have been in certain instances. For example, if certain jobs come along, the initiation fees are hiked up. Some of the new members get the new jobs, and the old members who have been members for a long time, get no assignment to work. That has happened in my State, and it is happening in other States, too. It is not right; it is wrong. 2 Leg. Hist. 1104.
'There was no further floor discussion of this part of the amendment prior to the vote on it. On April 24, 1959, Senator Kuchel offered an amended “Bill of Rights” title which reworded most of the provisions to eliminate ambiguities and •overly broad language in some of them. 'The paragraph dealing with dues increases was amended to provide three methods of increasing dues: (1) majority vote of the members at a meeting or (2) a referendum through the mails or (3) in the case of a national or international organization, “by majority vote” at a regular or duly constituted special convention. Senator Kuchel offered no explanation of the changes in this provision, indicating that it was taken “almost verbatim from that part of the McClellan amendment which was applicable * * *” 2 Leg. Hist. 1232. Again, there was no further floor debate until the revised “Bill of Rights” was passed.
The bill then moved to the House. As is well known, three versions of the Labor Reform Bill, H. R. 8342 (the Elliott Bill, reported by the Committee on Education and Labor), H. R. 8400 (the Landrum-Griffin Bill, which was ultimately adopted on the floor) and H. R. 8490 (the Shelley Bill, generally favored by organized labor) were the most important ones considered. The relevant language of all three bills was identical, and was the language finally enacted into law. The principal changes from the Senate version were that only dues increases, rather than all changes in dues, were required to be submitted to membership action, that a referendum vote was specifically authorized for both local and national labor organizations, and that executive boards of national unions were authorized to make interim increases until the next regular convention of the organization. These changes were apparently made in response to criticisms of the Senate Bill voiced by representatives of organized labor at hearings before the House Committee on Education and Labor.
The purpose of the section was expressed in various ways, but emphasis was always placed on ensuring members the “right to participate in deciding upon the rates of dues” (H.Rep. No. 741, 86th Cong., 1st Sess. 7); U. S. Code Congressional and Administrative News 1959, p. 2429; or “to control membership and other fees charged by the union” (id. at 78); or to “protect against exorbitant *33rates or arbitrary changes in dues and fees.” 2 Leg.Hist. 1520 (remarks of Rep. Griffin).
It seems to me clear that Congress intended merely to assure that dues increases could not be imposed by irresponsible or arbitrary union leaders not answerable in any way to the membership at large. It did not set out to dictate to unions a particular method for voting by delegates at national conventions, and, especially, did not have any desire to foreclose national unions from giving more weight to their larger locals on this question. Under the majority’s interpretation of this provision, the only way a national union could do this, and still give even small locals at least one delegate, would often be to increase the number of delegates and the expense of holding conventions, out of all manageable proportions. In the instant case the size of Federation locals varied from 20 (Local 629, Waupaca, Wise.) to 28,000 (Local 802, New York City), so that a fully proportional system of representation would require some 14,000 delegates in an organization with 280,000 members.
In summary, then, I am most reluctant to employ the “plain meaning” rule here as a means of preventing the use of a perfectly normal and accepted means of voting at union conventions, especially since there is no evidence that Congress felt this method was subject to any abuse whatsoever. It is noteworthy that none of the Congressmen and Senators who discussed the bill ever adverted to this problem, or indicated in any way that a change in the practice was desirable. The same is true of the many commentators who have exhaustively analyzed the provisions of the Act, from every possible standpoint. The term “majority vote of the delegates voting” is susceptible to a construction which would allow delegates’ votes to be weighted in accordance with the strength of the different locals they represent. We ought to accept that construction, and reverse the summary judgment herein.
. 73 Stat. 522, 29 U.S.C. § 401 et seq.
. See Leiserson, American Trade Union Democracy 122 (1959). “Though a representative body, the typical convention is identified with the union itself as if it were a general meeting of all the members. On roll call votes, the full membership of each local union is voted by its representatives in some conventions. In most, the same plebiscite effect is achieved by the delegates from each local casting a fixed number of votes proportionate to its membership.”
. Throughout this opinion, the word “weighted” is restricted to the method whereby delegates vote proportionate to the membership of the locals they represent. Nothing herein is to be construed as a blanket approval of all methods of weighted voting.
. Hereinafter cited as Leg.Hist.