dissenting:
With all due respect, I cannot agree with the outcome reached by the majority opinion. I must confess that my approach to the problem is influenced by my conviction that application of § 1973aa-1a(e) of the Voting Rights Act to initiative or recall petitions is inherently perverse; it takes too little account of the incentives that operate on persons circulating and signing, or not signing, such petitions.
I certainly agree with the majority’s proposition that “the Voting Rights Act is to be broadly construed so as to achieve the Act’s objectives.” Supra, p. 918. One major objective is to ensure that citizens of language minorities not be “excluded from participation in the electoral process.” 42 U.S.C. § 1973aa-1a(a). Those who circulate recall petitions, however, have no in-*925eentive to exclude anyone from signing their petitions. There is no way, and no need, to vote “no” on a recall petition itself; those eligible voters who do not sign, for any reason, are effectively counted as “no” votes on the question of whether to have an election.1 The purpose, therefore, of those who circulate recall petitions is to obtain as many signatures as possible in order to precipitate an election that otherwise would not occur. To the extent that they fail to provide translations of their petitions, they take the risk of failure of their enterprise.
It might be argued, however, that minority language voters ought to be able to have the opportunity to sign a petition in their language in order to help precipitate a recall election. It is difficult to see how such an argument can lead to an enforceable right, however. Certainly the circula-tors have no obligation to present a petition to any particular voter. Again, the incentive operating on the circulator is to reach as many potential voters as possible but if, for any number of reasons, the circulator does not reach an eligible voter and provide an opportunity for that voter to sign the petition, it is hard to see how there has been a violation of voting rights remediable by the Voting Rights Act and the courts. It is equally hard to see how the failure to reach potential signers in their own language gives rise to an enforceable right that can stop an otherwise successful recall effort in its tracks.
The plaintiffs in this case present an unusual variant of an eligible voters’ argument. They allege that they signed the petition because of a misrepresentation that it was a request for information rather than a recall petition. This situation is sufficiently extraordinary that it ought not to outweigh the practicalities that will govern most solicitation of signatures for a recall election. There are ample remedies short of enjoining an election to remedy the plaintiffs’ alleged injury. One remedy, employed by one of the plaintiffs here, was to rescind her signature. Another, of course, is to vote “no” in the recall election, where the ballots are required to be printed in both English and Spanish.
The downside of application of § 1973aa-l(a) to initiative and recall petitions is the chilling effect on recalls and initiatives. As the defendants point out, if the Voting Rights Act were to be applied to recall petitions for an office of Orange County, California, petitions would have to be presented in English, Spanish, Vietnamese, Korean and Chinese. It is not at all clear who is to bear the expense of such translation and printing; presumably it would be those who seek the recall. Even aside from the expense, the sheer burden-someness of the effort is likely to chill petition campaigns and make their success extremely unlikely. If, for example, a substantial minority community has enough members to precipitate a recall election by themselves, it is questionable that they should be burdened with the unnecessary duty to solicit numbers of other minority language speakers in their own languages. Those not solicited will have their chance to participate in the recall election. I fear that the majority’s ruling here, rather than opening the electoral process in accord with the intent of the Voting Rights Act, will have a tendency to close it. And the lawsuits it will engender will not be brought by those seeking access to sign a recall petition; they will be brought by plaintiffs like those before us who seek to prevent an election when sufficient signatures have been gathered to precipitate one.
*926Of course, all of these considerations would not carry weight if it were clear that Congress intended § 1973aa-1a(c) to apply to initiative and recall petitions. That intent is not clear, however; indeed, the words of the statute and the decisions of two of our sister circuits point firmly in the opposite direction. As the majority opinion recognizes, § 1973aa-1a(c) imposes its requirement only on a State or political subdivision subject to the applicable provisions of the Voting Rights Act when that State or subdivision “provides ... any ... materials of information relating to the electoral process.... ” (Emphasis added). It is only those materials provided by the State or subdivision that must be translated into the language of the applicable minority groups. Id.
It strains the meaning of these statutory terms to hold, as the majority does, that the State or subdivision “provided” the recall materials merely because they approved them as to form. The petitions themselves originated with, and were supplied by, the non-governmental defendants who caused the petitions to be circulated. Thus the Tenth Circuit held in Montero v. Meyer; 861 F.2d 603, 609-10 (10th Cir. 1988), that initiative petitions were not subject to the requirements of § 1973aa-1(c) because they were not provided by the State. The Eleventh Circuit came to the same conclusion in Delgado v. Smith, 861 F.2d 1489, 1496 (11th Cir.1988). I would follow the lead of these two circuits and hold that § 1973aa-1(c) does not apply to the circulation of recall petitions in the present case.
The majority opinion chooses instead to follow the language of our circuit in Zaldivar v. City of Los Angeles, 780 F.2d 823, 833-34 (9th Cir.1986), opining that a statutorily-required notice of intent to precipitate a recall election was subject to § 1973aa-1a. But the issue in Zaldivar was whether plaintiffs who brought such a claim and their attorneys were subject to sanctions under Fed.R.Civ.P. 11 for filing a frivolous lawsuit. Indeed, the lead sentence of the paragraph of Zaldivar on which the majority relies stated: “Giving section 1973aa-1a the ‘broadest possible scope,’ ... we have no difficulty in concluding that a competent attorney, after reasonable inquiry, could argue in good faith that a notice of intention to recall an office holder provides information relating to the electoral process.” Id. at 833 (citation omitted). Zaldivar’s language, therefore, is subject to interpretation as a statement of the attorney’s good faith argument. Moreover, the discussion was confined to the notice of intent, not the recall petitions themselves, and dealt only with the requirement of § 1973aa-1a that the materials in issue “relat[ed] to the electoral process.” Our case, in contrast, concerns the petitions themselves and the requirement that they be “provided” by the State or its subdivision. See id. Zaldivar is therefore distinguishable.
In any event, the language of Zaldivar relied on by the majority clearly was dictum. The majority points out that views of the scope of dictum vary, and that our court’s solemn pronouncements ought not easily to be disregarded as dictum. With that principle I heartily agree. In the case of Zaldivar, however, there is really no room for dispute. The district court in Zaldivar had ruled in a summary judgment that § 1973aa-la did not apply to the recall process, and it added Rule 11 sanctions against the plaintiffs and their attorneys. By the time the appeal was decided by our court, the election had occurred and it appears to have been undisputed that the case on the merits was moot. If there had been any doubt that our Zaldivar opinion dealt with the merits of the complaint, it was utterly dispelled by the closing words of our discussion. We said:
*927The district court was not persuaded by plaintiffs’ arguments. Under the proper legal standard, we do not review the court’s decision on the Voting Rights issue for legal error. We hold only that plaintiffs’ argument is not frivolous under the first prong of Rule 11.
Zaldivar; 780 F.2d at 834.
I have no difficulty accepting the holding of Zaldivar; I would not consider the plaintiffs in this case or their attorneys to be subject to Rule 11 sanctions for having brought a frivolous claim. I am convinced, however, that we should join the Tenth and Eleventh Circuits in holding that § 1973aa-la does not apply to recall (or in their cases initiative) petitions, for all of the reasons I have set forth. I therefore respectfully dissent from the majority’s opinion.
. The number of signatures needed to precipitate a recall election is calculated as a percentage of the total number of registered voters. Cal. Elec.Code § 11221.