STRAUB, Circuit Judge, with whom Judge SACK joins, concurring in part and concurring in judgment.

We concur in the result reached by the majority and in its reasoning that the evidence of legislative intent weighs decisively against applying the Voting Rights Act to New York Election Law § 5-106. We also, specifically, concur in the majority's limited remand for purposes of determining whether plaintiffs have stated a vote dilution claim on behalf of non-disenfranchised residents, based on New York's system of apportioning electoral districts. We do not join in any holding that a clear statement rule applies here, as we believe such a rule, in addition to being unnecessary to the disposition of this case, would be inappropriate in the voting rights context.

SACK, Circuit Judge, with whom Judge STRAUB joins with respect to Part I, concurring in part and concurring in judgment.

I

I write separately principally to respond to a portion of the dissent of my colleague, Judge Calabresi, in which he concludes that Judge Straub and I, being "motivated in large part by skepticism that Congress could have intended the result that the plaintiffs urge," Dissent of Judge Calabresi, post, at 365-66, have confused the views of the current Congress with that of the Congress that enacted the Voting Rights Act and its 1982 amendments. Judge Calabresi may or may not be right on the merits of this case, but in this respect he is surely wrong. What the current Congress, or a Congress that will next consider voting rights, thinks or will think is Congress's business, not ours. The question here is what the statute means.

It is in response to that question that we join Judge Cabranes's opinion insofar as it concludes that Congress did not intend the Voting Rights Act to encompass the type of felon disenfranchisement law at issue in this case. By amending Section 2 of the Voting Rights Act in 1982, Congress apparently sought to make clear that Section 2 covered both vote denial and vote dilution, and to reinstate the "results test," which Congress found to be the rule developed in the voting rights case law before the Supreme Court's decision in City of Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), particularly White v. Regester, 412 U.S. 755, 766, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973) and Zimmer v. McKeithen, 485 F.2d 1297 (5th Cir. 1973). Whatever one's view of the uses and misuses of legislative history to enlighten, or even to contradict plain meaning, insofar as the 1982 amendment was aimed specifically at overruling Bolden and reinstating previous case law, it is not that difficult to discern what Congress meant to do when it said what it said. See United States v. Rybicki, 354 F.3d 124, 136 (2d Cir. 2003) (en banc), cert. denied, 543 U.S. 809, 125 S.Ct. 32, 160 L.Ed.2d 10 (2004) (reasoning that when a statute is adopted for the purpose of overruling a Supreme Court decision which in turn overruled previous law on the subject, Congress likely intended to reinstate the previous law).

The case law that Congress sought to reinstate, so far as we can tell, involved cases in which "designedly or otherwise, an apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population." Zimmer, 485 F.2d at 1304 (internal quotation marks omitted; alterations incorporated) (quoted in S. Rep. 97-417 (1982), *23, 1982 U.S.C.C.A.N. 177, **200). As such, the White-Zimmer factors, under which a court assesses the result of a voting law under the totality of the circumstances, seem to us to be particularly suited — even if not exclusively so — to claims of apportionment and legislative districting. In amending the Voting Rights Act in 1982, Congress told the courts to resume doing what they had done in the White-Zimmer cases. There is no indication that any court pre- Bolden thought of using, let alone actually used, the White-Zimmer factors to strike down felon disenfranchisement laws. Cf. Rybicki, 354 F.3d at 138 (explaining that the phrase "intangible right of honest services" should be understood as codifying prior case law, and that "Congress was recriminalizing mail- and wire-fraud schemes to deprive others of that 'intangible right of honest services,' which had been protected before McNally [v. United States, 483 U.S. 350, 107 S.Ct. 2875, 97 L.Ed.2d 292 (1987)], not all intangible rights of honest services whatever they might be thought to be" (emphasis in original)).

The plain meaning argument — as articulated by Judge Parker in his powerful opinion — is of course a compelling one. But none of our colleagues yet, so far as we can see, has presented a persuasive reason to think that if the language of section 2 covers felon disenfranchisement laws as applied to persons duly imprisoned, the result was intentional rather than an accident resulting from the breadth and generality of language used to accomplish specific, identifiable goals. At the end of the day, the notion that by using such language Congress "accidentally meant to" address the long-standing, nearly universal felon disenfranchisement laws recognized in the Fourteenth Amendment, just does not make sense to us. "Boggling," perhaps, is in the mind of the beholder. See Dissent of Judge Calabresi, post at 367.

II

Although I don't think it to be dispositive, I am also troubled that the central concern of these plaintiffs, obviously of enormous importance, seems to be tangential at best to the relief they seek under the Voting Rights Act. The plaintiffs assert that, in New York, criminal law-enforcement, prosecution, and punishment are corrupted by a deeply ingrained institutional bias against Blacks and Latinos. They "allege that the stark differences in incarceration and probation rates for Blacks and Latinos in New York, as opposed to Whites, have resulted from discrimination in New York's criminal justice system." Dissent of Judge Parker at 343. Perhaps giving the plaintiffs the right to vote from prison would have some positive impact on that system, but it would be a highly unusual remedy, in light of the ubiquitous state felon disenfranchisement laws, addressing the pleaded wrong very indirectly, indeed. It gives me pause: Did Congress mean, through the use of the general language of the Voting Rights Act, to seek to address the problem of racial discrimination in the states' administration of criminal justice?

III

I note, finally, that there may well be important distinctions between the impact of the felon disenfranchisement statute at issue here and those that provide for lifetime disenfranchisement, whether or not the difference matters for purposes of the Voting Rights Act analysis. They appear to me to be very different: One type (the one before us) appears to be connected to the issue of punishment — the removal of criminals from society for a term, for purposes of punishment, the safety of society, "rehabilitation" and the like. The other seems ultimately untethered to punitive segregation, since the disenfranchisement persists long after termination of the segregation, at which point punitive ends seem no longer to be at issue. Today's decision, of course, explicitly leaves the issue open. See Maj. Op. at 314.

B.D. PARKER, JR., Circuit Judge, dissenting.

The essence of a Section 2 claim under the Voting Rights Act ("VRA"), 42 U.S.C. § 1973, is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality along racial lines in the opportunities of voters to elect their preferred representatives. See Thornburg v. Gingles, 478 U.S. 30, 47, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). The majority, in my view, reaches the wrong result about § 2 for a host of reasons, not the least of which is that it attacks the wrong question. The operative inquiry on this appeal is not whether a historic policy of felon disenfranchisement, read next to odds and ends from legislative histories, indicates Congress's intention to exclude felon disenfranchisement laws from the coverage of the VRA. Rather, this appeal begins and ends with the simple question of whether we should read an unambiguous remedial statute, intended to have, as the Supreme Court has emphasized, the "broadest possible scope," to allow the Hayden plaintiffs' claims to go forward. Allen v. State Bd. of Elections, 393 U.S. 544, 567, 89 S.Ct. 817, 22 L.Ed.2d 1 (1969). I believe we should.

The amended complaint alleges that the stark differences in incarceration and probation rates for Blacks and Latinos in New York, as opposed to Whites, have resulted from discrimination in New York's criminal justice system. Under New York law, incarcerated felons and felons on parole cannot vote. See N.Y. Election Law § 5-106. Section 2(a) of the VRA provides that "[n]o voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State . . . in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color. . . ." 42 U.S.C. § 1973(a) (emphasis added). New York Election Law § 5-106 imposes a voting qualification. It denies those convicted of felonies the opportunity to vote. The Hayden plaintiffs allege that § 5-106 violates § 2(a) because it results in a denial or dilution of the right to vote on account of race. Because I believe these contentions state a claim upon which relief can be granted, I respectfully dissent.

A.

For purposes of this appeal we must treat as unassailable the Hayden plaintiffs' allegations that New York systematically discriminates against its minority citizens through its felony disenfranchisement law. This case was resolved in the district court on a Rule 12(c) motion for judgment on the pleadings. Consequently, we must accept the allegations in the complaint as true and benefit the plaintiffs with all reasonable inferences. Greco v. Trauner, Cohen Thomas, L.L.P., 412 F.3d 360, 363 (2d Cir. 2005). Under that Rule, the Hayden plaintiffs are entitled to prevail unless it appears beyond doubt that they can prove no facts in support of their claims which would entitle them to relief. Id.

Though the Hayden plaintiffs may have a difficult time mustering sufficient proof, they are, at the very least, entitled to that opportunity. The Supreme Court has left us no room whatsoever to question this approach: "Even if serious problems lie ahead in applying the 'totality of circumstances' standard described in [VRA] § 2(b), that task, difficult as it may prove to be, cannot justify a judicially created limitation on the coverage of the broadly worded statute, as enacted and amended by Congress." Chisom v. Roemer, 501 U.S. 380, 403, 111 S.Ct. 2354, 115 L.Ed.2d 348 (1991).

The majority concludes that felon disenfranchisement laws are immune from VRA scrutiny, no matter how discriminatory the effects of those laws might be. A review of the plaintiffs' allegations, which state a paradigmatic claim of discriminatory disenfranchisement, demonstrates the impropriety of treating this case differently than any other VRA case. Even if the following allegations are true, the majority finds, they do not offend the VRA in the slightest.

The First Amended Complaint is replete with allegations of racial disparities resulting from the operation of New York Election Law § 5-106. It alleges that Blacks and Latinos are prosecuted, convicted, and sentenced to incarceration at rates substantially disproportionate to those of Whites. Hayden First Amended Complaint at ¶ 61. It also alleges that while Blacks and Latinos make up 86% of the total current prison population and 82% of the current parolee population, they comprise only 31% of New York State's overall population. Hayden First Amended Complaint at ¶ 64. It further alleges that Blacks and Latinos are sentenced to incarceration at substantially higher rates than Whites, and Whites are sentenced to probation at substantially higher rates than Blacks and Latinos. Hayden First Amended Complaint at ¶ 66. The collective impact of these statistics, according to the First Amended Complaint, is that Blacks and Latinos comprise nearly 87% of those currently denied the right to vote pursuant to New York Election Law § 5-106. Hayden First Amended Complaint at ¶ 68.

There is no substantive record in this case. None of us knows what reality — if any — exists behind these allegations. Significantly, each of the other circuits that has wrestled with this issue has done so on the basis of a fully-developed summary judgment record, not simply on the pleadings. For instance, in support of their vote denial claim, the plaintiffs in Farrakhan v. Washington, 338 F.3d 1009 (9th Cir. 2003), reh'g en banc denied, 359 F.3d 1116 (9th Cir. 2004), presented "statistical evidence of the disparities in arrest, bail and pre-trial release rates, charging decisions, and sentencing outcomes in . . . Washington's criminal justice system," submitted expert declarations, reports, and studies and relied on them to show "the extent to which these disparities could be attributed to racial bias and discrimination," and offered evidence of the "tenuous policy justifications for Washington's felon disenfranchisement law," and of the "discriminatory intent that guided the enactment of these laws." Id. at 1013. The plaintiffs in the Eleventh Circuit presented a similarly robust summary judgment record. See Johnson v. Governor of Fla., 405 F.3d 1214, 1230 n. 31 (11th Cir. 2005) (en banc); id. at 1241-42 (Wilson, J., concurring in part and dissenting in part); see also Johnson v. Governor of Fla., 353 F.3d 1287, 1293, 1294-96, 1301-02, 1304-06 (11th Cir. 2003) (analyzing the considerable record of statistical evidence and expert reports presented by plaintiffs in support of their claims), vacated by 377 F.3d 1163 (11th Cir. 2004).

Here, by contrast, we have no analysis of the disparities in New York's criminal justice system. We have no facts to think about. We have no expert reports or studies on whether these disparities can be attributed to racial bias or to other factors. Consequently, none of us can say with the requisite certainty what actually occurred in New York to create the anomalies catalogued in the First Amended Complaint. The Hayden plaintiffs want the opportunity to eliminate other reasons for these disparities and to prove that their inability to vote is the product of discrimination in the criminal justice system. If they do so, they could well prove a violation of the VRA.

Suppose, for example, they were able to demonstrate that the dramatically different incarceration rates for minorities and Whites in New York were largely driven by drug convictions and reflected the manner in which law enforcement resources were deployed in the "war on drugs." Suppose they showed that law enforcement officials (and task forces) concentrated resources on street-level users/dealers of heroin and crack cocaine in minority neighborhoods (because the problems were worse and arrests were easier in such areas) but, at the same time, devoted comparatively little attention to areas where Whites were abusing those same illegal drugs at the same rates (and powder cocaine at higher rates). Suppose they also showed that Whites received probation three times as frequently as similarly situated Blacks or Latinos for similar crimes. Neither showing is remotely beyond the realm of possibility in New York, and I believe this type of proof could constitute some evidence of a VRA violation. Yet this would be the very type of undertaking that the majority would preclude. Whether or not the Hayden plaintiffs ultimately succeed is, at this point, irrelevant. The issue is whether the VRA affords them the opportunity to make the effort. If the majority is correct, the Congresses that passed the VRA in 1965, and extended it in 1982, did not intend for the statute to reach situations such as these. I do not believe those Congresses intended any such thing; and I reach that conclusion simply by reading the words they used.

No one disputes that states have the right to disenfranchise felons: § 2 of the Fourteenth Amendment makes that clear. See U.S. Const. amend. XIV, § 2 ("when the right to vote at any election for the choice of electors for President and Vice President of the United States, Representatives in Congress, the Executive and Judicial officers of a State, or the members of the Legislature thereof, is denied to any of the male inhabitants of such State . . . or in any way abridged, except for participation in rebellion, or other crime, the basis of representation therein shall be reduced. . . ."); see also Richardson v. Ramirez, 418 U.S. 24, 54, 94 S.Ct. 2655, 41 L.Ed.2d 551 (1974).

But, as will be explored below, the fact that felon disenfranchisement statutes may sometimes be constitutional does not mean they are always constitutional. And certainly, the fact that they sometimes may be constitutional does not mean that they are immune from congressional regulation. In any event, this case is largely about the Fifteenth, not the Fourteenth Amendment. Section 1 of the Fifteenth Amendment makes it clear that states may not disenfranchise on the basis of race. See U.S. Const. amend. XV, § 1 ("The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude."). Section 2 of the VRA, as amended in 1982, also makes it clear that states may not disenfranchise on the basis of race, even unintentionally. See Reno v. Bossier Parish Sch. Bd., 520 U.S. 471, 482, 117 S.Ct. 1491, 137 L.Ed.2d 730 (1997) ("When Congress amended § 2 [of the VRA] in 1982, it clearly expressed its desire that § 2 not have an intent component.") (emphasis omitted); Thornburg, 478 U.S. at 43-44, 106 S.Ct. 2752 (explaining that the 1982 amendments to § 2 of the VRA were a rejection of the position of the plurality in Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), which "required proof that the contested electoral practice or mechanism was adopted or maintained with the intent to discriminate against minority voters").

Thus, as Judge Wilfred Feinberg, writing for half of our evenly divided en banc Court in Baker v. Pataki, 85 F.3d 919 (2d Cir. 1996) (en banc), said: "While a State may choose to disenfranchise some, all or none of its felons based on legitimate concerns, it may not do so based upon distinctions that have the effect, whether intentional or not, of disenfranchising felons because of their race." Id. at 937. If the majority is correct, this proposition is false.

B.

In order to justify its failure to apply the plain language of VRA § 2, the majority must find ambiguity in that provision's pellucid language. Tellingly, the majority never attempts to argue that § 2(a) is ambiguous, instead stating that "[w]e are not convinced that the use of broad language in [§ 2(a)] necessarily means that the statute is unambiguous with regard to its application to felon disenfranchisement laws." Maj. Op. at 315. But this statement is not a finding of ambiguity. The majority acknowledges that "there is no question that the language of § 1973 is extremely broad . . . and could be read to include felon disenfranchisement provisions. . . ." Id. The real question is whether it can be read any other way. Conspicuously absent from the majority opinion is so much as a hint of any intelligible reading under which § 5-106 is not a "voting qualification or prerequisite to voting or standard, practice or procedure." (What else on earth could § 5-106 possibly be?)

The majority cites Robinson v. Shell Oil Co., 519 U.S. 337, 117 S.Ct. 843, 136 L.Ed.2d 808 (1997), which states that "[t]he plainness or ambiguity of statutory language is determined by reference to the language itself, the specific context in which that language is used, and the broader context of the statute as a whole." Id. at 341, 117 S.Ct. 843 (finding that "the term 'employees' . . . is ambiguous as to whether it excludes former employees"). Robinson did not suggest that ambiguity could be determined by reference to legislative history or telling silences. More importantly, the majority disregards the stern warning that accompanied the passage it quoted: "Our inquiry must cease if the statutory language is unambiguous and the statutory scheme is coherent and consistent." Id. at 340, 117 S.Ct. 843 (emphasis added) (internal quotation marks omitted). The majority's inquiry plunges ahead with no finding that the statutory language is ambiguous, in direct disregard of the Supreme Court's instructions.

Finding no support in the text and no ambiguity, the majority turns to various secondary and tertiary sources. In each instance, the source is employed to try to show that Congress did not mean what it said. The most prominent of the extra-textual sources enlisted by the majority are "context," "background assumptions supplied by other statutory and Congressional wording," "history," and "manifestations of intent by Congress at the time of the VRA's enactment and thereafter." Maj. Op. at 315. Rather than follow the majority into this slough, I believe we should heed the Supreme Court's instruction that when "the meaning of [a statute's] text is plain and unambiguous, we need not accept [an] invitation to consider the legislative history." Whitfield v. United States, 543 U.S. 209, 215, 125 S.Ct. 687, 160 L.Ed.2d 611 (2005).

To begin, resort to these types of sources is conspicuously inappropriate because the VRA, while broad, is not ambiguous. Tellingly, the very panel that decided Muntaqim v. Coombe, 366 F.3d 102 (2d Cir. 2004), conceded this point. The original Muntaqim panel said "section 1973, while vague, does not seem ambiguous." Id. at 128 n. 22 (citing 42 U.S.C. § 1973). Given this concession, our responsibility, as we have been reminded time and time again, is to apply the statute as Congress wrote it. Accordingly, "we begin with the understanding that Congress says in a statute what it means and means in a statute what it says." Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6, 120 S.Ct. 1942, 147 L.Ed.2d 1 (2000) (internal quotation marks omitted). "[I]n interpreting a statute a court should always turn first [to this] one, cardinal canon before all others. . . . When the words of a statute are unambiguous, then, this first canon is also the last: judicial inquiry is complete." Conn. Nat'l Bank v. Germain, 503 U.S. 249, 253-54, 112 S.Ct. 1146, 117 L.Ed.2d 391 (1992) (internal quotation marks omitted).

The majority's counter-weight to all of this authority is Justice Holmes' language from the 1920s to the effect that the plain meaning of the text does not necessarily "preclude consideration of persuasive evidence if it exists." Boston Sand Gravel Co. v. United States, 278 U.S. 41, 48, 49 S.Ct. 52, 73 L.Ed. 170 (1928). I certainly dispute that the majority's evidence is anything close to "persuasive," but, regardless, this proposition has not been good law for many years. The modern view is that the "authoritative statement" of Congress's intent is "the statutory text, not the legislative history or any other extrinsic material." Exxon Mobil Corp. v. Allapattah Servs., Inc., ___ U.S. ___, 125 S.Ct. 2611, 2626, 162 L.Ed.2d 502 (2005); see also United States v. Wells, 519 U.S. 482, 509-10, 117 S.Ct. 921, 137 L.Ed.2d 107 (1997) (Stevens, J., dissenting) ("Congress [in the past] looked to the courts to play an important role in the lawmaking process by relying on common-law tradition and common sense to fill gaps in the law — even to imply causes of action and remedies that were not set forth in statutory text. . . . The Court's approach to questions of statutory construction has changed significantly since that time."). See generally William N. Eskridge, Jr., The New Textualism, 37 UCLA L.Rev. 621 (1990). The majority's variation on Justice Holmes' position is its view that it is not bound by the plain language of the VRA because "[t]he plain meaning of legislation should be conclusive, except in the 'rare cases [in which] the literal application of a statute will produce a result demonstrably at odds with the intentions of its drafters.'" United States v. Ron Pair Enters., Inc., 489 U.S. 235, 242, 109 S.Ct. 1026, 103 L.Ed.2d 290 (1989) (quoting Griffin v. Oceanic Contractors, Inc., 458 U.S. 564, 571, 102 S.Ct. 3245, 73 L.Ed.2d 973 (1982)); accord Stoltz v. Brattleboro Hous. Auth. (In re Stoltz), 315 F.3d 80, 89 (2d Cir. 2002).

But this case does not present a "rare" or "exceptional" circumstance where following the rules would lead to a result " so bizarre that Congress could not have intended it." Demarest v. Manspeaker, 498 U.S. 184, 191, 111 S.Ct. 599, 112 L.Ed.2d 608 (1991) (emphasis added) (internal quotation marks omitted). Indeed, just the opposite is true. In Allen v. State Board of Elections, supra, the Supreme Court observed that § 2(a) of the VRA, as originally drafted, included only a prohibition against any voting "qualification or procedure." 393 U.S. at 566, 89 S.Ct. 817. After concerns were raised that this language may not be sufficiently expansive to cover all practices that could deny individuals the right to vote on account of their race, Congress expanded the language in the final version to include any "voting qualifications or prerequisites to voting, or standard, practice, or procedure." Id. at 567, 89 S.Ct. 817 (quoting 42 U.S.C. § 1973(a)). The Supreme Court viewed this change as "[i]ndicative of [Congress's] intention to give the Act the broadest possible scope." Allen, 393 U.S. at 566-67, 89 S.Ct. 817.

In keeping with the general principle that courts should interpret the VRA to provide "the broadest possible scope in combating racial discrimination," Chisom, 501 U.S. at 403, 111 S.Ct. 2354 (internal quotation marks omitted), the Supreme Court has interpreted § 2 to apply to a wide variety of election and voting practices. See, e.g., Growe v. Emison, 507 U.S. 25, 40, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993) (holding that VRA should be interpreted broadly to cover single member district plans); Houston Lawyers' Ass'n v. Att'y Gen. of Tex., 501 U.S. 419, 428, 111 S.Ct. 2376, 115 L.Ed.2d 379 (1991) (same for the method of electing state trial judges); Chisom, 501 U.S. at 404, 111 S.Ct. 2354 (same for the method of electing state appellate court judges); see also Voinovich v. Quilter, 507 U.S. 146, 154, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993) (assuming, without deciding, that VRA covers claims of "influence" dilution). Reading the section to cover felon disenfranchisement laws is fully consistent with this line of cases. Since everyone agrees that Congress amended § 2 in 1982 in an express effort to broaden the protection afforded by the VRA, we are hard pressed to understand the majority's conclusion that, at that same time, Congress, without comment, intended to except an important voting test from that protection. See Chisom, 501 U.S. at 403-04, 111 S.Ct. 2354.

Although the majority would prefer to look beyond the plain language of § 2(a), "judicial review must end at the statute's unambiguous terms." Brodie v. Schmutz ( In re Venture Mortgage Fund, L.P.), 282 F.3d 185, 188 (2d Cir. 2002). Though some may question the wisdom of Congress's decision to enact a statute that permits challenging felon disenfranchisement laws, we are judges, not policy-makers. If different policy-based outcomes are generally desired, Congress may always amend the VRA. But that is its job, not ours.

C.

The majority nonetheless concludes that we are relieved of the obligation to respect the text of VRA § 2, and, in construing it, may look behind its words for essentially three reasons: (1) the explicit approval given felon disenfranchisement laws in the Fourteenth Amendment; (2) various pieces of legislative history indicating the VRA was not intended to reach such laws; and (3) the long history and continuing prevalence of felon disenfranchisement provisions throughout the United States. See Maj. Op. at 315-16. The common denominator of each of these reasons is that it is factitious.

1. Section 2 of the Fourteenth Amendment

The primary point of departure for the majority is the mention of felon disenfranchisement in § 2 of the Fourteenth Amendment. U.S. Const. amend. XIV, § 2. The majority's reliance on this provision is misplaced. The Constitution does not endorse felon disenfranchisement when it declines to prohibit that practice, any more than the Constitution endorses felon enslavement when the Thirteenth Amendment states: "Neither slavery nor involuntary servitude, except as punishment for crime whereof the party shall have been duly convicted, shall exist within the United States. . . ." U.S. Const. amend. XIII, § 1 (emphasis added). Declining to prohibit something is not the same as protecting it.

Nor has the Supreme Court read any special immunity for felon disenfranchisement into the Fourteenth Amendment. In Richardson, the Court considered a claim that § 1 of the Fourteenth Amendment prohibited felon disenfranchisement altogether. 418 U.S. at 33, 94 S.Ct. 2655. The Court rejected this claim on the reasonable basis that the Fourteenth Amendment would not need an exception for felon disenfranchisement in § 2 if it intended to prohibit it altogether in § 1. Id. at 43. Richardson did not grant felon disenfranchisement immunity against any other ground of invalidity; it simply held that denying the vote to felons was not a per se violation of the Fourteenth Amendment.

Unquestionably, Richardson did not authorize the use of felon disenfranchisement for discrimination, as the Supreme Court made clear in Hunter v. Underwood, 471 U.S. 222, 105 S.Ct. 1916, 85 L.Ed.2d 222 (1985). See Baker, 85 F.3d at 937. Hunter involved an attack under 42 U.S.C. § 1981 and § 1983 on a provision in the Alabama Constitution of 1901 that intentionally disenfranchised Blacks. Hunter, 471 U.S. at 223-24, 105 S.Ct. 1916. The Supreme Court unanimously held that the provision was unconstitutional, and in so doing addressed the relationship between § 2 of the Fourteenth Amendment and felon disenfranchisement as follows:

The single remaining question is whether [the Alabama provision] is exempted from the operation of the Equal Protection clause of § 1 of the Fourteenth Amendment by the "other crime" provision of § 2 of that Amendment. . . . [W]e are confident that § 2 was not designed to permit the purposeful racial discrimination attending the enactment and operation of [the Alabama provision] which otherwise violates § 1 of the Fourteenth Amendment.

Hunter, 471 U.S. at 233, 105 S.Ct. 1916. As Judge Feinberg so correctly observed: "[b]efore Hunter, it might have been plausibly argued that § 2 of the Fourteenth Amendment carves all aspects of felon disenfranchisement out of the purview of the Equal Protection Clause, which is in the very same amendment." Baker, 85 F.3d at 936. But because the "other crimes" provision of § 2 of the Fourteenth Amendment did not foreclose the Equal Protection challenge to Alabama's felon disfranchisement law in Hunter, it necessarily follows that Congress also has the power under the Fourteenth Amendment to regulate potentially discriminatory felon disenfranchisement statutes, as it did when it passed the VRA.

More to the point, § 2 of the Fourteenth Amendment — which expressly contemplated and essentially sanctioned racially discriminatory voting qualifications — in no way diminishes Congress's power to enforce the Fifteenth Amendment. The majority fails to appreciate that the operative provisions of the VRA were enacted pursuant to Congress's power under the Fifteenth Amendment. See Chisom, 501 U.S. at 383, 111 S.Ct. 2354 ("The preamble to the Voting Rights Act of 1965 establishes that the central purpose of the Act is to enforce the fifteenth amendment to the Constitution of the United States.") (internal quotation marks omitted); see also South Carolina v. Katzenbach, 383 U.S. 301, 308, 86 S.Ct. 803, 15 L.Ed.2d 769 (1966) ("Congress assumed the power to prescribe these remedies from § 2 of the Fifteenth Amendment, which authorizes the National Legislature to effectuate by 'appropriate' measures the constitutional prohibition against racial discrimination in voting.").

Moreover, when it added the results test of VRA § 2 in 1982, Congress invoked its powers to enforce by "appropriate legislation" both the Fifteenth Amendment's guarantee that no citizen will be denied the right to vote on account of race and the Fourteenth Amendment's guarantee of racial equality. See S.Rep. No. 97-417, at 27, 39 (1982), as reprinted in 1982 U.S.C.C.A.N. 177, 205, 217. Thus, the results test is grounded in both the Fourteenth and the Fifteenth Amendments. In most instances, it is immaterial whether Congress's enforcement powers are grounded in one amendment or the other. This case, however, is different.

Congress's enforcement powers under the Fourteenth and Fifteenth Amendments overlap in certain areas. Nevertheless, the Fifteenth Amendment provides a separate, distinct source of congressional power to regulate racially discriminatory voting restrictions. See, e.g., Rice v. Cayetano, 528 U.S. 495, 522, 120 S.Ct. 1044, 145 L.Ed.2d 1007 (2000) (holding that a voting qualification that satisfies the Fourteenth Amendment's one-person, one-vote rule may nevertheless violate the Fifteenth Amendment's "race neutrality command," because the "Fifteenth Amendment has independent meaning and force"); see also Neal v. Delaware, 103 U.S. 370, 389, 26 L.Ed. 567 (1881) ("Beyond question the adoption of the Fifteenth Amendment had the effect, in law, to remove from the State Constitution, or render inoperative, that provision which restricts the right of suffrage to the white race."); United States v. Reese, 92 U.S. 214, 218, 23 L.Ed. 563 (1876) ("[T]he [ Fifteenth] amendment has invested the citizens of the United States with a new constitutional right which is within the protecting power of Congress. That right is exemption from discrimination in the exercise of the elective franchise on account of race, color, or previous condition of servitude.").

While § 2 of the Fourteenth Amendment permitted states to deny voting rights to their citizens (but threatened to reduce their representation if they did), the Fifteenth Amendment took a more sweeping approach: It expressly prohibited states from abridging the right to vote "on account of race." U.S. Const. amend. XV § 1. Even if the reference to criminal disenfranchisement in § 2 of the Fourteenth Amendment could somehow be construed to limit Congress's power under that Amendment to reach criminal disenfranchisement laws with racially discriminatory results, the Fifteenth Amendment's later, comprehensive ban on race discrimination in voting could not possibly be reasonably read in that way, since it is unqualified.

The text and the legislative history of the Fifteenth Amendment demonstrate that it did not simply mimic § 2 of the Fourteenth Amendment, but, instead, broke new ground by instituting a ban on any disenfranchisement based on race. Recall that § 2 of the Fourteenth Amendment was a structural successor to the Three-Fifths Clause of the original Constitution. See Richardson, 418 U.S. at 43-45, 94 S.Ct. 2655; see also U.S. Const. art. I, § 2, cl. 3 ("Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers, which shall be determined by adding to the whole Number of free Persons . . . three fifths of all other Persons."), amended by U.S. Const. amend. XIV, § 2. The concern of the Republican legislators who drafted § 2 of the Fourteenth Amendment was that, with the slaves emancipated, the Three-Fifths Compromise would be inoperative, freedmen would therefore be counted fully in decennial censuses, and the former slave states would soon be entitled to enough additional representatives to threaten Republican control of the federal government. See Richardson, 418 U.S. at 73-74, 94 S.Ct. 2655 (Marshall, J., dissenting); Foner, supra note 3, at 251-61. Section 2 of the Fourteenth Amendment explicitly contemplated that states would deny voting rights to their citizens, including through race-based disenfranchisement, and therefore adjusted the states' representation in Congress to ensure that Southern White Democrats could not control the national government by disenfranchising Blacks. See Foner, supra note 3, at 252, 255, 259. As the Supreme Court observed, the framers of § 2 of the Fourteenth Amendment "were primarily concerned with the effect of reduced representation upon the States, rather than with the two forms of disenfranchisement which were exempted from that consequence." Richardson, 418 U.S. at 43, 94 S.Ct. 2655.

In practice, § 2 of the Fourteenth Amendment was immediately succeeded by the Military Reconstruction Acts of 1867, which essentially prohibited any state from reentering the Union if it barred Blacks from voting in state constitutional conventions. See, e.g., An Act To Provide for the More Efficient Government of the Rebel States, ch. 153, § 5, 14 Stat. 428, 429 (Mar. 2, 1867) (effectively enfranchising "male citizens of said State, twenty-one years old and upward, of whatever race, color or previous condition, who have been resident in said State for one year previous to the day of such election, except as may be disenfranchised for participation in the rebellion or for felony at common law" to elect delegates for state constitutional conventions). In effect, by means of the Military Reconstruction Acts of 1867, Southern states that denied the franchise to Blacks would be denied any representation in Congress. Then, with Southern Blacks voting in large numbers by the election of 1868, Congress enacted the Fifteenth Amendment to enfranchise Blacks in Northern and border states, entrench the franchise for Black males nationwide, and prevent any future retrenchment. See generally Gillette, supra note 3. The Fifteenth Amendment thus provided an additional tool for rooting out discrimination in voting.

In sum, the Fifteenth Amendment was enacted, among other reasons, precisely because the Fourteenth Amendment — including § 2 — did not prohibit states from disenfranchising Blacks. Section 2 of the Fourteenth Amendment, after all, expressly contemplates that a state might bar Blacks from the polls; it simply exacts a price in reduced representation. The Supreme Court in United States v. Reese, supra, writing only five years after the ratification of the Fifteenth Amendment, put it simply: "Previous to this [ Fifteenth] amendment, there was no constitutional guaranty against this discrimination: now there is." 92 U.S. at 218.

The Fifteenth Amendment was thus not an extension or continuation of § 2 of the Fourteenth Amendment. It took a diametrically different approach and conferred a right not previously secured by the Constitution. While the Supreme Court has not found felon disenfranchisement laws to be per se unconstitutional under the Fourteenth Amendment, it has never suggested that they are insulated from challenges under a statute passed to enforce the Fifteenth Amendment's ban on racial discrimination in voting. The claim at issue in this case is, therefore, firmly imbedded in the Fifteenth Amendment and cannot be made to disappear by citing § 2 of the Fourteenth Amendment.

2. Legislative history of the VRA

"Judicial investigation of legislative history has a tendency to become . . . an exercise in looking over a crowd and picking out your friends." Allapattah Servs., 125 S.Ct. at 2626 (internal quotation marks omitted). The majority is no exception, having drafted a number of "friends" to help it around unambiguous statutory language. See Maj. Op. at 317-23. However, even assuming justification exists for disregarding the language of a statute — a proposition I disfavor — the committee reports, floor debates, and stray comments cited by the majority do not, in my view, trump the text of VRA § 2(a).

The majority first looks to the legislative history of § 4 of the VRA, for assistance in explicating § 2(a) of the Act. See Maj. Op. at 318-19. The probability of useful results from such an exercise is nearly always extremely low because the legislative history of one section of an expansive statute such as the VRA is typically of no value in understanding another, entirely different section. Section 4(a) of the VRA banned the use, in jurisdictions with a demonstrated history or racial discrimination, of "any test or device" to limit the ability to vote in order to "assure that the right of citizens of the United States to vote is not denied or abridged on account of race or color." 42 U.S.C. § 1973b(a)(1). Section 4(c) defined the phrase "test or device" as

any requirement that a person as a prerequisite for voting or registration for voting (1) demonstrate the ability to read, write, understand, or interpret any matter, (2) demonstrate any educational achievement or his knowledge of any particular subject, (3) possess good moral character, or (4) prove his qualifications by the voucher of registered voters or members of any other class.

42 U.S.C. § 1973b(c).

Thus, § 4 and § 2(a) employ starkly different language that dramatically distinguishes their scope. Section 4's use of the narrow terms "any test or device," 42 U.S.C. § 1973b(a)(1), is not comparable to § 2(a)'s use of the broad language "[n]o voting qualification or prerequisite to voting or standard, practice, or procedure." 42 U.S.C. § 1973(a). Second, § 4 and § 2 serve separate functions and operate differently. Section 4 imposes an outright ban on tests or devices, see 42 U.S.C. § 1973b(a)(1), while § 2(a) creates a "results" test, which requires investigating and weighing numerous factors. See 42 U.S.C. § 1973(b). Given this outright ban, one can understand why Congress would want to narrow the category of voting mechanisms falling under § 4(c) relative to § 2(a), where a plaintiff need only demonstrate discriminatory results. Third, the legislative history of one provision enacted in 1965 (§ 4) has nothing to say about Congress's intentions when amending a different provision (§ 2), seventeen years later in 1982. Equivocal fragments from legislative history should not obscure the fact that, from its inception and particularly through its amendment in 1982, Congress intended that § 2, unlike § 4, be given the broadest possible reach, as the text it chose makes clear. See Chisom, 501 U.S. at 403, 111 S.Ct. 2354; Allen, 393 U.S. at 566-67, 89 S.Ct. 817.

Not only is the majority's reliance on the legislative history of § 4 problematic, its reliance on floor statements concerning § 4(c) compounds the problems. See Maj. Op. at 318-19. Floor statements are among the most dangerous and least reliable forms of legislative history. As the Supreme Court has warned, floor statements from a handful of Senators "cannot amend the clear and unambiguous language of a statute," and there is "no reason to give greater weight to the views of two Senators than to the collective votes of both Houses, which are memorialized in the unambiguous statutory text." Barnbart v. Sigmon Coal Co., 534 U.S. 438, 457, 122 S.Ct. 941, 151 L.Ed.2d 908 (2002); see also Crosby v. Nat'l Foreign Trade Council, 530 U.S. 363, 390-91, 120 S.Ct. 2288, 147 L.Ed.2d 352 (2000) (Scalia, J., concurring in the judgment) ("[T]he statements of individual Members of Congress (ordinarily addressed to a virtually empty floor) . . . [are not] a reliable indication of what a majority of both Houses of Congress intended when they voted for the statute before us. The only reliable indication of that intent — the only thing we know for sure can be attributed to all of them — is the words of the bill that they voted to make law.").

The danger lies in allowing "clear statutory language to be materially altered by such colloquies [by individual Congressmen], which often take place before the bill has achieved its final form," because this might "open the door to the inadvertent, or perhaps even planned, undermining of the language actually voted on by Congress and signed into law by the President." Regan v. Wald, 468 U.S. 222, 237, 104 S.Ct. 3026, 82 L.Ed.2d 171 (1984). Finally, when compared to committee reports that arguably "represent the considered and collective understanding of those Congressmen involved in drafting and studying proposed legislation," floor statements "reflect at best the understanding of individual Congressmen." Zuber v. Allen, 396 U.S. 168, 186, 90 S.Ct. 314, 24 L.Ed.2d 345 (1969); see also Fidelity Fin. Servs., Inc. v. Fink, 522 U.S. 211, 220, 118 S.Ct. 651, 139 L.Ed.2d 571 (1998) ("Whatever weight some Members of this Court might accord to floor statements about proposals actually under consideration, remarks that purport to clarify 'related' areas of the law can have little persuasive force, and in this case none at all."); 2A Sutherland Statutory Construction § 48.13 ("Statements by individual members of the legislature about the meaning of provisions in a bill . . . are generally held not to be admissible as aids in construing the statute. . . .").

In addition to looking at the legislative history from 1965 of another portion of the VRA, the majority also relies on Congress's silence on felon disenfranchisement in the legislative history of the 1982 amendments to § 2. See Maj. Op. at 320-21. The majority's use of silence is even worse than its use of floor debates, and the Supreme Court has repeatedly cautioned against drawing inferences from silence. See, e.g., Whitfield, 543 U.S. at 216, 125 S.Ct. 687; Wells, 519 U.S. at 497, 117 S.Ct. 921; Harrison v. PPG Indus. Inc., 446 U.S. 578, 592, 100 S.Ct. 1889, 64 L.Ed.2d 525 (1980). As we opined in Lander v. Hartford Life Annuity Insurance Co., 251 F.3d 101 (2d Cir. 2001), "'[i]t would be a strange canon of statutory construction that would require Congress to state in committee reports or elsewhere in its deliberations that which is obvious on the face of a statute.'" Id. at 110 n. 5 (quoting United States v. Koh, 199 F.3d 632, 636-37 (2d Cir. 1999)); see also Harrison, 446 U.S. at 592, 100 S.Ct. 1889 ("In ascertaining the meaning of a statute, a court cannot, in the manner of Sherlock Holmes, pursue the theory of the dog that did not bark.") (citing Arthur Conan Doyle, The Silver Blaze, in The Complete Sherlock Holmes (1938)). Contrary to the majority's assertion, therefore, there is nothing especially telling about Congress's silence. Moreover, Congress was neither silent nor subtle on the main question that concerns us here: whether it intended to permit discrimination in voting in any circumstance. The purpose of VRA § 2, as amended, is "to prohibit any voting practice, or procedure [that] results in discrimination." S.Rep. No. 97-417, at 2 (1982), 1982 U.S.C.C.A.N. at 179 (emphasis added). Congress could hardly have been clearer that its intention was to pass a broad remedial statute that guaranteed that no discriminatory law or device could escape scrutiny. As the Supreme Court has noted, "Congress enacted the Voting Rights Act of 1965 for the broad remedial purpose of ridding the country of racial discrimination in voting." Chisom, 501 U.S. at 403, 111 S.Ct. 2354 (internal quotation marks omitted). The legislative intent that matters here is Congress's intent for § 2 to apply as broadly as possible.

3. Historical anecdotes

Nor does the fact that felon disfranchisement has been a long-standing and widely-used device (it all started in Athens, we are told) mean that VRA § 2(a) does not apply. See Maj. Op. at 316-17. For starters, this is an odd argument indeed, since the very purpose of § 2(a) was to address long-standing, widely-used devices that impacted minority voting. Felon disenfranchisement laws are not the only voting qualification that existed before the Civil War; so did poll taxes, property requirements, and literacy tests to name just a few. See generally Keyssar, supra note 3, at 53-76. The majority's arguments about felon disenfranchisement apply equally to each of them. Since no one would doubt that VRA § 2(a) encompasses these voting qualifications, I am at a loss to understand why felon disenfranchisement should be treated differently.

The VRA, as remedial legislation, was purposefully broad and flexible to accommodate changing circumstances impacting minority voting. This country's more recent history of exploding prison populations and dramatically rising incarceration rates is such a circumstance. The Bureau of Justice Statistics issues reports on state and federal prison populations. See Bureau of Justice Statistics, Prisoners Under State or Federal Jurisdiction 1977-2004, from National Prisoner Statistics data series (NPS-1) (updated Dec. 6, 2005), at http://www.ojp.usdoj.gov/bjs/data/ corpop02.csv. These reports show dramatic trends in the last thirty years. In New York, for example, the state and federal prison population at the end of 1977 was 19,367 persons. Id. At the end of 1999, only twenty-two years later, New York's state and federal prison population was more than three and one-half times larger, at 72,899 persons. Id. The national trends are even more dramatic. Nationwide, the state and federal prison population has increased from 300,024 persons at the end of 1977, to 1,496,629 persons at the end of 2004 — nearly five times larger in only twenty-seven years. Id. Researchers estimate that the total disenfranchised population has more than doubled from 1976 to 2000. See Christopher Uggen Jeff Manza, Democratic Contraction? Political Consequences of Felon Disenfranchisement in the United States, 67 Am. Soc. Rev. 777, 782 (2002).

When viewed against this historical backdrop, Congress's general silence on felon disenfranchisement with respect to VRA § 2(a) makes perfect sense. The relationship between felon disfranchisement and minority voting is fundamentally different now than forty years ago when the VRA was originally passed, or even almost twenty-five years ago when the 1982 amendments were passed. The stray comments in legislative history on which the majority heavily relies are typically pulled from the 1960s and 1970s. Nearly all predate the 1982 amendments to the VRA, and nearly all were generated well before the explosive growth in this nation's prison population. It is hardly surprising that legislators did not focus on felon disfranchisement and minority voting during the debates surrounding the passage of the VRA, or even during its amendment in 1982, since the problem as it currently manifests itself did not exist.

And again, while Congress may not have given much thought to the specific question of felon disenfranchisement, it certainly made clear its intent to outlaw any practice that resulted in discrimination in voting. To demand, as the majority does, that Congress mention every possible application of a statute in the legislative history is to prohibit Congress from passing broad remedial statutes that protect against methods of discrimination yet to be invented. The Congresses that created the VRA knew that discrimination takes many forms, and that only broad statutes — statutes that do not enumerate, in their text or in their history, every form of discrimination against which they protect — can be flexible enough to address new challenges. See H.R. Rep. No. 89-439, at 10 (1965), as reprinted in 1965 U.S.C.C.A.N. 2437, 2441 ("[E]ven after apparent defeat resisters seek new ways and means of discriminating. Barring one contrivance too often has caused no change in result, only in methods.").

D.

Next, a minority of this Court turns to the clear statement rule. This canon of interpretation requires Congress to make its intention "unmistakably clear" when enacting statutes that would alter the usual constitutional balance between the federal government and the states. The minority's concern is that application of the VRA to New York Election Law § 5-106 would intrude into three important state functions: (1) the regulation of the franchise; (2) the state's authority to craft its criminal law; and (3) the regulation of correctional institutions. See Min. Op. at 326. The minority reasons that, to the extent the VRA would affect this balance if applied to felon disenfranchisement statutes, it must be construed not to encompass such provisions if it is unclear whether Congress intended the VRA to apply to such laws.

For several reasons, the clear statement rule does not apply. First, for it to apply, ambiguity must exist, and § 2(a) is unambiguous. See, e.g., Penn. Dep't of Corr. v. Yeskey, 524 U.S. 206, 212, 118 S.Ct. 1952, 141 L.Ed.2d 215 (1998) ("in the context of an unambiguous statutory text," arguments concerning whether Congress has made its intention clear are "irrelevant"); Salinas v. United States, 522 U.S. 52, 60, 118 S.Ct. 469, 139 L.Ed.2d 352 (1997) (" Gregory itself . . . noted [that] the principle it articulated did not apply when a statute was unambiguous.") (citing Gregory v. Ashcroft, 501 U.S. 452, 467, 111 S.Ct. 2395, 115 L.Ed.2d 410 (1991)); Hilton v. S.C. Pub. Rys. Comm'n, 502 U.S. 197, 206, 112 S.Ct. 560, 116 L.Ed.2d 560 (1991) (citing Supreme Court cases that "describe the plain statement rule as a rule of statutory construction to be applied where statutory intent is ambiguous") (internal quotation marks omitted).

Second, even if VRA § 2(a) were ambiguous, the clear statement rule would still not apply because the provision is broadly worded, and the rule does not apply to broadly worded remedial statutes. As the Supreme Court has made clear, a broad statute satisfies the clear statement rule even though it does not enumerate all of its intended applications: "[T]he fact that a statute can be 'applied in situations not expressly anticipated by Congress does not demonstrate ambiguity. It demonstrates breadth.'" Yeskey, 524 U.S. at 212, 118 S.Ct. 1952 (quoting Sedima, S.P.R.L. v. Imrex Co., 473 U.S. 479, 499, 105 S.Ct. 3275, 87 L.Ed.2d 346 (1985)). Congress used language in § 2 that was deliberately broad and generic. It is well-settled that "[a] statute can be unambiguous without addressing every interpretive theory offered by a party. It need only be 'plain to anyone reading the Act' that the statute encompasses the conduct at issue." Salinas, 522 U.S. at 60, 118 S.Ct. 469 (quoting Gregory, 501 U.S. at 467, 111 S.Ct. 2395). Congress could hardly have been expected to have enumerated every conceivable voting qualification, prerequisite, practice, or procedure to which the statute could apply in the text, or even the legislative history, of § 2(a). To do so would have left the states free to devise new means to discriminate that were not listed. To hold that Congress did not intend the VRA to cover felon disenfranchisement statutes is to hold that Congress actually intended to allow some forms of race-based voter disenfranchisement. Such a result I find improbable — indeed inconceivable.

Third, the clear statement rule cannot be justified by contending that unless it is applied, the VRA would improperly interfere with "sensitive domains" such as the core state function of regulating the franchise. See Min. Op. at 325, 326-27. This contention overlooks the quite obvious fact that the very purpose of the VRA was to impose Congressional regulation on the traditional state function of regulating voting. See Lopez v. Monterey County, 525 U.S. 266, 284-85, 119 S.Ct. 693, 142 L.Ed.2d 728 (1999) ("In short, the Voting Rights Act, by its nature, intrudes on state sovereignty. The Fifteenth Amendment permits this intrusion, however. . . ."). Time and time again, the Supreme Court has held that state voting requirements are comfortably within Congress' reach under the VRA. See id. at 278, 119 S.Ct. 693 (holding that VRA's pre-clearance requirement could be applied to voting changes adopted by a non-covered state if the changes had an effect on a covered county); Young v. Fordice, 520 U.S. 273, 283-84, 117 S.Ct. 1228, 137 L.Ed.2d 448 (1997) (holding that state's new plan for separate state and federal voter registration was a discretionary change and therefore required pre-clearance under the VRA); Bush v. Vera, 517 U.S. 952, 976, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996) (applying VRA to state's drawing of voting districts); Chisom, 501 U.S. at 403-04, 111 S.Ct. 2354 (holding that § 2(a) of the VRA applies to state laws regarding election of state court judges); Katzenbach v. Morgan, 384 U.S. 641, 658, 86 S.Ct. 1717, 16 L.Ed.2d 828 (1966) (holding that state's English literacy voting requirement could not be enforced to the extent it was at odds with the VRA). Felon disenfranchisement is no more a core state function than any of these examples.

Fourth, while it is correct that the states possess the primary authority for defining and enforcing the criminal law, United States v. Lopez, 514 U.S. 549, 561 n. 3, 115 S.Ct. 1624, 131 L.Ed.2d 626 (1995), the short — and conclusive — answer is that New York Election Law § 5-106 is not a criminal law. It is a voting law found in New York's Election code, not among its criminal laws. As Judge Friendly pointed out, "[d]epriving convicted felons of the franchise is not a punishment but rather is a 'nonpenal exercise of the power to regulate the franchise.'" Green v. Bd. of Elections, 380 F.2d 445, 450 (2d Cir. 1967) (quoting Trop v. Dulles, 356 U.S. 86, 97, 78 S.Ct. 590, 2 L.Ed.2d 630 (1958) (plurality opinion)). See also N.Y. Election Law § 1-102 ("This chapter shall govern the conduct of all elections at which voters of the state of New York may cast a ballot for the purpose of electing an individual to any party position or nominating or electing an individual to any federal, state, county, city, town or village office, or deciding any ballot question submitted to all the voters of the state or the voters of any county or city, or deciding any ballot question submitted to the voters of any town or village at the time of a general election."); N.Y. Penal Law § 1.05 ("The general purposes of the provisions of this chapter are: . . . To proscribe conduct which unjustifiably and inexcusably causes or threatens substantial harm to individual or public interests. . . .").

Fifth, the clear statement rule is particularly inappropriate in the context of the VRA, which was enacted and amended pursuant to Congress's powers under both the Fourteenth and Fifteenth Amendments, as explained in Part C.1, supra. Contrary to the suggestion of some members of this Court, see Min. Op. at 326, the seismic shift created by the Fourteenth and Fifteenth Amendments clearly altered the federal-state balance in an attempt to address a truly compelling national interest — namely, reducing racial discrimination perpetuated by the states. Indeed, these Amendments "were specifically designed as an expansion of federal power and an intrusion on state sovereignty." Gregory, 501 U.S. at 468, 111 S.Ct. 2395 (quoting City of Rome v. United States, 446 U.S. 156, 179, 100 S.Ct. 1548, 64 L.Ed.2d 119 (1980)); see also Mitchum v. Foster, 407 U.S. 225, 238 n. 28, 92 S.Ct. 2151, 32 L.Ed.2d 705 (1972) (recognizing the "basic alteration in our federal system wrought in the Reconstruction era through federal legislation and constitutional amendment" such as the Fourteenth and Fifteenth Amendments). In sum, any shift in the federal-state balance of power that would purportedly result from applying VRA § 2 to New York Election Law § 5-106 would not occur as a result of the resolution of this case. That shift occurred more than 130 years ago when the Reconstruction Amendments were passed and ratified.

Finally, were a clear statement required, VRA § 2(a) supplies it: "No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color. . . ." 42 U.S.C. § 1973(a). Since § 2(a) covers all voting qualifications, it indisputably covers felon disenfranchisement laws like New York Election Law § 5-106. If anything is clear from the legislative history of the VRA it is that Congress intended to eliminate all race-based disfranchisement, no matter the means by which it was achieved.

E.

Next, I address Judge Walker's effort to import the Supreme Court's City of Boerne v. Flores, 521 U.S. 507, 117 S.Ct. 2157, 138 L.Ed.2d 624 (1997), line of cases to § 2 of the Fifteenth Amendment, and to issues of race and voting. This effort hinges on an erroneous belief that courts must first define with "precision" the scope of the constitutional right that Congress seeks to enforce. See Concurring Op. of Judge Walker at 330. His concurrence contends that, if taken at its word, VRA § 2(a) would be unconstitutional, and that a plain reading of the statute should be avoided for that reason. He believes, after defining "with precision" the constitutional right at issue, the reviewing court would look to see whether Congress explicitly identified a history and pattern of unconstitutional violations of that particular, precisely defined right, and finally, the court would assess Congress's chosen means of addressing these violations to determine whether its remedy was a congruent and proportional response to those violations. See Concurring Op. of Judge Walker at 330.

With all due respect, this "define with precision" test as applied to issues like voting and race was invented from whole cloth. It finds no analogue in jurisprudence, and no grounding in Boerne and its progeny. This is so for a number of reasons. First, Judge Walker cites Board of Trustees of University of Alabama v. Garrett, 531 U.S. 356, 121 S.Ct. 955, 148 L.Ed.2d 866 (2001), in support of his "define with precision" test. See Concurring Op. of Judge Walker at 330. Garrett actually stated that courts should "identify with some precision the scope of the constitutional right at issue." 531 U.S. at 365, 121 S.Ct. 955 (emphasis added). In Garrett, this process required the Supreme Court to "examine the limitations § 1 of the Fourteenth Amendment places upon States' treatment of the disabled." Id. If we were to actually follow the Supreme Court's guidance in Garrett, we would examine the limitations § 1 of the Fourteenth Amendment and § 1 of the Fifteenth Amendment place on a state's treatment of race and voting. When this is done, it is clear that the scope of Congress's enforcement authority is at its zenith when protecting against discrimination based on suspect classifications (such as race), or when protecting fundamental rights (such as voting). Unlike the rights of the disabled at issue in Garrett, the scope of the constitutional rights at issue here are expansive, and not subject to artificial narrowing.

The "define with precision" test is an invitation for judges to impress their subjective judgments on statutory texts. Accordingly, in this case, Judge Walker has chosen to narrowly define (to define away?) the right at issue as the right of state prisoners and parolees to vote. See Concurring Op. of Judge Walker at 330-31. Contrary to this reading of the Boerne cases, the scope of the right that Congress seeks to enforce should come from the text of the statute, not from the air or from a judge's preferences. Thus, in stark contrast to Judge Walker, Congress itself defined the right as the right to be free from any "voting qualification or prerequisite to voting or standard, practice, or procedure . . . imposed or applied by any State . . . in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color. . . ." 42 U.S.C. § 1973(a) (emphasis added).

If a statute is unambiguous, a judge's role is to read and to apply it, not to rewrite it through an illusory and subjective search for "precision." If judges are free to search broad, remedial statutes for "precision" and enforce them according to what they think they find, then the coverage of such statutes is in the hands of judges, not legislators. If taken seriously, this "define with precision" test would call into question Congress's ability to pass most types of broad remedial legislation. If that test were the standard, Congress, to insulate an enactment from constitutional objection, would have to anticipate and document the need for virtually all applications of a remedial statute. This approach is, of course, completely contrary to what remedial legislation is intended to do: protect rights from encroachment by unforeseen as well as by known means.

Fortunately, the "define with precision" test has no takers. The Supreme Court has affirmed that "Congress may enact so-called prophylactic legislation that proscribes facially constitutional conduct, in order to prevent and deter unconstitutional conduct." Nev. Dep't of Human Res. v. Hibbs, 538 U.S. 721, 727-728, 123 S.Ct. 1972, 155 L.Ed.2d 953 (2003). Moreover, when Congress "seeks to remedy or prevent unconstitutional discrimination," § 5 of the Fourteenth Amendment and § 2 of the Fifteenth Amendment authorize Congress "to enact prophylactic legislation proscribing practices that are discriminatory in effect, if not in intent." Tennessee v. Lane, 541 U.S. 509, 520, 124 S.Ct. 1978, 158 L.Ed.2d 820 (2004). The Supreme Court has repeatedly referred to this enforcement power as a "broad power indeed." Lane, 541 U.S. at 518, 124 S.Ct. 1978 (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 732, 102 S.Ct. 3331, 73 L.Ed.2d 1090 (1982)) (citing Ex parte Virginia, 100 U.S. 339, 346, 25 L.Ed. 676 (1879)).

Obviously, this broad power is not without limits. Since its 1997 decision in Boerne, the Supreme Court has issued several decisions concerning the scope of Congress's authority to enact prophylactic legislation under § 5 of the Fourteenth Amendment. The Supreme Court has held, for example, that Congress lacked the power under § 5 to abrogate state sovereign immunity to suit for discrimination on the basis of disability, age, and religion. See Garrett, 531 U.S. at 374, 121 S.Ct. 955 (disability); Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 91, 120 S.Ct. 631, 145 L.Ed.2d 522 (2000) (age); Boerne, 521 U.S. at 536, 117 S.Ct. 2157 (religion). The Supreme Court's decisions turned on a lack of "congruence and proportionality," id. at 530, 117 S.Ct. 2157; in other words there was insufficient evidence of discrimination to justify federal intrusion into matters traditionally regulated by the states.

By contrast, despite any number of opportunities, the Supreme Court has never questioned the constitutionality of the VRA. Indeed, the Supreme Court has referred to it as the paradigm of appropriate remedial legislation. See Hibbs, 538 U.S. at 738, 123 S.Ct. 1972 (upholding the FMLA and likening it to the VRA, which the Court described as a "valid exercise of Congress' § 5 power"); Garrett, 531 U.S. at 373, 121 S.Ct. 955 ("The ADA's constitutional shortcomings are apparent when the Act is compared to Congress' efforts in the Voting Rights Act of 1965 to respond to a serious pattern of constitutional violations."); Fla. Prepaid Postsecondary Educ. Expense Bd. v. Coll. Sav. Bank, 527 U.S. 627, 640, 119 S.Ct. 2199, 144 L.Ed.2d 575 (1999) (distinguishing the Patent Remedy Act from the VRA on account of the "undisputed record of racial discrimination confronting Congress in the voting rights cases"); Boerne, 521 U.S. at 518, 117 S.Ct. 2157 ("[M]easures protecting voting rights are within Congress' power to enforce the Fourteenth and Fifteenth Amendments, despite the burdens those measures place on the States.").

The distinction between the VRA and other remedial legislation rests on three obvious and interrelated factors: One, the right to vote free of racial discrimination is a "fundamental principle" of the Constitution. Rice, 528 U.S. at 512, 120 S.Ct. 1044; see Morgan, 384 U.S. at 654, 652, 86 S.Ct. 1717 (describing the right to vote as "precious and fundamental" and "the right that is preservative of all rights") (internal citations omitted). Two, the country's long and persistent history of discrimination gives Congress much greater latitude in fashioning appropriate remedies for racial discrimination in voting than for other types of discrimination. See Boerne, 521 U.S. at 526, 117 S.Ct. 2157 (acknowledging "the necessity of using strong remedial and preventive measures to respond to the widespread and persisting deprivation of constitutional rights resulting from this country's history of racial discrimination"); see also Lane, 541 U.S. at 518-19 n. 4, 124 S.Ct. 1978 (recounting that federal legislation meant to remedy racial discrimination in voting has repeatedly withstood federalism-related scrutiny); id. at 561, 124 S.Ct. 1978 (Scalia, J., dissenting) ("Giving § 5 more expansive scope with regard to measures directed against racial discrimination by the States accords to practices that are distinctively violative of the principal purpose of the Fourteenth Amendment a priority of attention that this Court envisioned from the beginning, and that has repeatedly been reflected in our opinions."); Kimel, 528 U.S. at 83, 120 S.Ct. 631 (distinguishing age discrimination from racial discrimination because "[o]lder persons . . . unlike those who suffer discrimination on the basis of race . . . have not been subjected to a history of purposeful unequal treatment") (internal quotation marks omitted).

And three, it is easier for Congress to show a pattern of state constitutional violations when the remedial legislation is targeted at a classification that is subject to heightened judicial scrutiny, such as race. See Hibbs, 538 U.S. at 736, 123 S.Ct. 1972 ("Because the standard for demonstrating the constitutionality of a gender-based classification is more difficult to meet than our rational-basis test . . . it was easier for Congress to show a pattern of state constitutional violations. Congress was similarly successful in South Carolina v. Katzenbach, where we upheld the Voting Rights Act of 1965: Because racial classifications are presumptively invalid, most of the States' acts of race discrimination violated the Fourteenth Amendment.") (citing Katzenbach, 383 U.S. at 308, 86 S.Ct. 803) (internal citation omitted); see also Lane, 541 U.S. at 529, 124 S.Ct. 1978. Thus Congress was comfortably within its expansive enforcement powers when enacting VRA § 2(a).

Nor is the fact that Congress did not make specific findings of racial discrimination through felon disenfranchisement persuasive. Congress specifically found that racial discrimination in voting has proven resilient in the face of narrow prohibitions, in light of the ingenuity of state and local efforts to erect novel and subtle barriers to minority voting. See S.Rep. No. 97-417 at 6, 1982 U.S.C.C.A.N. at 183 ("The ingenuity of such schemes seems endless."); H.R. Rep. No. 89-439, at 10 (1965), as reprinted in 1965 U.S.C.C.A.N. 2437, 2441 ("[E]ven after apparent defeat resisters seek new ways and means of discriminating. Barring one contrivance too often has caused no change in result, only in methods."). Congress determined that it was necessary to enact broad and flexible remedies to eliminate discrimination that persisted for more than 100 years after the ratification of the Fifteenth Amendment. It therefore provided that "[n]o voting qualification . . . shall be imposed" if it "results" in the abridgement of the right to vote on account of race or color. 42 U.S.C. § 1973(a).

To require Congress to catalogue a history of discrimination by specifically enumerating every conceivably discriminatory practice in order to regulate it would run directly counter to the purpose of the VRA, which was to ban discrimination not only by currently used means, but by any others that could be devised by the states. See Allen, 393 U.S. at 566, 89 S.Ct. 817 (noting that Congress expanded the language of the Act to give it "the broadest possible scope" in response to the concern that an earlier draft "was not broad enough to cover various practices that might effectively be employed to deny citizens their right to vote"). Accordingly, Congress has never been required to find that a particular voting practice had resulted in discrimination before it could ban or regulate such a practice, and the VRA has been properly applied to many discriminatory voting devices and practices that were not mentioned in the VRA's legislative history. See, e.g., Harris v. Graddick, 593 F.Supp. 128, 132-33 (M.D.Ala. 1984) (applying VRA § 2 to state's hiring of a disproportionately small number of Black poll workers); Brown v. Dean, 555 F.Supp. 502, 504-06 (D.R.I. 1982) (applying VRA § 2 to location of state's polling places and finding a "constructive disenfranchisement" of minority voters); Brown v. Post, 279 F.Supp. 60, 64 (W.D.La. 1968) (applying VRA § 2 to state's failure to provide voters with absentee ballots).

In sum, Judge Walker's view that Congress lacks the authority to reach felon disenfranchisement statutes that result in the denial of the right to vote on account of race is wrong. To adopt that view is to conclude that there are some forms of race-based voter discrimination that are beyond Congress's reach, a proposition that, as we have seen, is not correct.

* * *

For these reasons, I respectfully dissent.

CALABRESI, Circuit Judge, dissenting.

The majority opinion is learned, thoroughly researched, well-written, and restrained — but it is almost totally irrelevant to the question presented in this case. The majority demonstrates beyond peradventure that Congress did not intend the Voting Rights Act to prohibit felon disenfranchisement categorically, as that statute categorically prohibits, for instance, the use of literacy tests and "good moral character" requirements in certain jurisdictions. See 42 U.S.C. §§ 1973b(a), (c). And if the plaintiffs were, in fact, arguing that the Voting Rights Act erects a per se ban on felon disenfranchisement, I would readily join the majority. But, of course, this is not the plaintiffs' position. Rather, they contend that the Voting Rights Act prohibits felon disenfranchisement laws that result in the denial or dilution of voting rights on the basis of race. Their complaint alleges that New York Election Law § 5-106 has precisely this result, and, as Judge Parker notes in his dissent, we are bound to accept this allegation as the gospel truth for purposes of New York's motion for judgment on the pleadings. Nothing in the majority opinion — nor, for that matter, in the concurrences — gives a single reason to suggest that Congress did not intend the Voting Rights Act to do what its plain language says and bar felon disenfranchisement statutes that result in racial discrimination. I, therefore, see no basis for depriving the plaintiffs of the right to prove their allegations.

I join Judge Parker's well-reasoned dissent. I write separately simply to emphasize how completely, in my view, the majority's analysis misses the actual issue of the case, and to muse as to what could have led my learned colleagues to affirm the district court when, as Judge Parker's dissent demonstrates, all of the canonical indicators of legislative intent — statutory text, context, and legislative history — so clearly point in the opposite direction.

I

The bulk of the majority opinion is devoted to discussing (1) congressional statements that § 4(c) of the Voting Rights Act (which bans some practices outright) did not reach felon disenfranchisement, (2) failed efforts to amend the statute to ban felon disenfranchisement outright, and (3) congressional actions that presuppose the validity of felon disenfranchisement in some circumstances. Whatever interest such subjects may hold in the abstract, they have no more bearing on the issue at hand than do the punitive practices of ancient Greeks or medieval Britons, which the majority opinion also canvasses.

The majority's discussion of § 4(c) illustrates the point. That provision categorically forbids the use of "any test or device," including "good moral character" requirements, in covered jurisdictions. See 42 U.S.C. § 1973b(c). In other words, such tests and devices are forbidden whether or not they can be shown to have discriminatory results. The majority makes much of legislative history showing that Congress did not intend § 4(c) to forbid felon disenfranchisement. True enough. Felon disenfranchisement is not prohibited in the absence of a showing that it brings about discriminatory results. But the statements in legislative history that felon disenfranchisement is not banned by § 4(c) cannot be taken to imply a wholesale carve-out that exempts felon disenfranchisement from Voting Rights Act scrutiny altogether, as the majority asserts. Rather, such legislative statements simply make the uncontroversial point that felon disenfranchisement laws are not "good moral character" requirements within the meaning of § 4(c). That, however, is not the issue before us. The fact that race-neutral felon disenfranchisement is permissible under § 4(c) tells us nothing at all about whether § 2 allows racially discriminatory felon disenfranchisement. And, as Judge Parker shows, the language of § 2(a) makes perfectly plain that such discriminatory disenfranchisement is barred.

The majority also tells us that some members of Congress tried, unsuccessfully, to amend the Voting Rights Act "to prohibit the States from denying the right to vote in Federal elections to former criminal offenders who have not been convicted of any offense related to voting or elections and who are not confined in a correctional institution." Maj. Op. at 319 (quoting H.R. 15049, 92d Cong. (1972)). Once again, this does indeed imply that the Voting Rights Act does not, of itself, prohibit all felon disenfranchisement — a fact that no one disputes. How the majority moves from the fact that Congress declined to proscribe race-neutral felon disenfranchisement to the conclusion that Congress intended to exempt racially discriminatory felon disenfranchisement from the coverage of the Voting Rights Act is beyond me. It is perfectly clear that voting practices and procedures that are not per se impermissible under the Voting Rights Act — at-large voting and multi-member districts, for instance — violate the statute when they produce discriminatory results. See Thornburg v. Gingles, 478 U.S. 30, 48, 80, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986) (recognizing that "[m]ultimember districts and at-large election schemes . . . are not per se violative of minority voters' rights," but affirming the district court's conclusion that the multimember electoral structure at issue did violate § 2 because it resulted in racially discriminatory vote dilution) (internal quotation marks and alteration omitted). And so it is with felon disenfranchisement laws. The majority has shown that some members of Congress thought felon disenfranchisement to be so inimical to voting rights as to try to forbid the practice even in the absence of discriminatory effects. The idea that such congressional efforts somehow imply the existence of a "safe harbor" for felon disenfranchisement statutes that do result in racially discriminatory denials or dilutions of voting rights is, it seems to me, risible.

Nor do subsequent enactments that presuppose the validity of felon disenfranchisement laws ( e.g., a provision of the Help America Vote Act of 2002, 42 U.S.C. § 15483(a)(2)(A)(ii)(I)), or bills that seek to limit felon disenfranchisement ( e.g., the Ex-Offenders Voting Rights Act of 2005, H.R. 663, 109th Cong. (2005)), suggest in the slightest that Congress understands discriminatory felon disenfranchisement to be consistent with the Voting Rights Act.

Still, the majority concludes, largely based on the statutory history recounted above, that "the Voting Rights Act does not apply to felon disenfranchisement provisions." Maj. Op. at 322. What is behind this remarkable decision to buck text, context, and legislative history in order to insulate a particular racially discriminatory practice from an anti-discrimination rule of general applicability?

II

I believe the majority opinion and, perhaps especially, the concurrence of Judges Straub and Sack are motivated in large part by skepticism that Congress could have intended the result that the plaintiffs urge. But it is important here, in talking about congressional intent, to distinguish between the enacting Congress and the current Congress. It is, of course, the legislative intent of the enacting Congress — not the current Congress — that is controlling. See, e.g., European Cmty. v. RJR Nabisco, Inc., 355 F.3d 123, 136 (2d Cir. 2004) ("[E]xpressions of legislative intent made years after the statute's initial enactment are entitled to limited weight under any circumstances. . . ."), rev'd on other grounds, 544 U.S. 1012, 125 S.Ct. 1968, 161 L.Ed.2d 845 (2005); see also United States v. Southwestern Cable Co., 392 U.S. 157, 170, 88 S.Ct. 1994, 20 L.Ed.2d 1001 (1968) ("[T]he views of one Congress as to the construction of a statute adopted many years before by another Congress have very little, if any, significance.") (internal quotation marks omitted). And I see no reason to think that the 97th Congress, which was responsible for the "dramatic substantive transformation" of the Voting Rights Act in 1982, meant the expansive prohibition of discriminatory results it enacted to apply in any other way than precisely as written. Thomas M. Boyd Stephen J. Markman, The 1982 Amendments to the Voting Rights Act: A Legislative History, 40 Wash. Lee L.Rev. 1347, 1347 (1983). The fact that the 109th Congress in the year 2006, if asked, might very well choose not to invalidate felon disenfranchisement laws that produce discriminatory results in no way indicates that the very different Congress of a generation ago made, or would now make, the same choice.

It may be that the current Congress would rather not see state felon disenfranchisement statutes invalidated under the Voting Rights Act even when these resulted in racial discrimination. And some scholars, myself included, have suggested that it might be a good idea if, as a starting point, in certain circumstances, courts were permitted to read the law according to what they perceived to be the will of the current Congress, rather than that of a long-gone-by one. See Guido Calabresi, A Common Law for the Age of Statutes (1982); see also William N. Eskridge, Jr., Dynamic Statutory Interpretation (1994). But whatever the merits of such an arrangement in the abstract, it is simply not a part of our legal system.

Moreover, the only justification for such statutory updating, even in theory, is that the legislature may be stalled by inertia, with the result that an outmoded and antimajoritarian statutory scheme becomes fixed in stone. See Calabresi, A Common Law for the Age of Statutes 120-21. This is hardly the case here, for important portions of the Voting Rights Act will be before Congress next year for reauthorization. See 42 U.S.C. §§ 1973b(a)(8), 1973aa-1a(b)(1). The reauthorization process removes most inertial barriers to lawmaking, so that if a majority in Congress wishes to amend the statute to permit even discriminatory felon disenfranchisement, it will be able to do so easily. Cf. William N. Eskridge, Jr., et al., Cases and Materials on Legislation: Statutes and the Creation of Public Policy 66-67 (2001) (discussing procedural obstacles that ordinarily can prevent the enactment of laws that have the support of a legislative majority).

Significantly, any such amendment will take effect long before Hayden and the other class plaintiffs, whose case is only at the pleading stage, ever reach a ballot. Thus, the fact that Congress will soon take up the Voting Rights Act removes even the most academic of reasons for the Court to take the matter into its own hands and impose a statutory construction deeply at odds with the language of the statute and with the intention of the original enacting Congress "to give the Act the broadest possible scope." Allen v. State Bd. of Elections, 393 U.S. 544, 567, 89 S.Ct. 817, 22 L.Ed.2d 1 (1969). Our unwillingness to let the statute apply as it is written unless and until Congress changes it in 2007 is, to me, mind-boggling. Accordingly, I respectfully dissent.

* * *

SOTOMAYOR, Circuit Judge, dissenting.

I join in Judge Parker's dissent, and write this separate opinion only to emphasize one point. I fear that the many pages of the majority opinion and concurrences — and the many pages of the dissent that are necessary to explain why they are wrong — may give the impression that this case is in some way complex. It is not.

It is plain to anyone reading the Voting Rights Act that it applies to all "voting qualification[s]." And it is equally plain that § 5-106 disqualifies a group of people from voting. These two propositions should constitute the entirety of our analysis. Section 2 of the Act by its unambiguous terms subjects felony disenfranchisement and all other voting qualifications to its coverage.

The duty of a judge is to follow the law, not to question its plain terms. I do not believe that Congress wishes us to disregard the plain language of any statute or to invent exceptions to the statutes it has created. The majority's "wealth of persuasive evidence" that Congress intended felony disenfranchisement laws to be immune from scrutiny under § 2 of the Act, Maj. Op. at 322, includes not a single legislator actually saying so. But even if Congress had doubts about the wisdom of subjecting felony disenfranchisement laws to the results test of § 2, I trust that Congress would prefer to make any needed changes itself, rather than have courts do so for it.

I respectfully dissent.

KATZMANN, Circuit Judge, dissenting.

As a matter of public policy, I do not doubt that it may well be rational for the state to enact laws — applied fairly and equally, regardless of race, creed, color, religion or gender — that deprive felons of the right to vote while serving their sentences. At the same time, I am mindful that the task of a court is not to make policy or to substitute its views for that of political branches, but rather to be faithful to the words of a statute. In the case at hand, Congress stated:

No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color . . . as provided in subsection (b) of this section.

42 U.S.C. § 1973(a). I believe that this language of the Voting Rights Act is facially unambiguous and that New York Election Law § 5-106 imposes a "voting qualification or prerequisite to voting or standard, practice, or procedure." In my view, the plain text of the Voting Rights Act therefore encompasses the claim that appellants have pleaded, and thus allows appellants to try to show a violation. But in construing a broad, unambiguous statute, my inquiry as to its meaning does not necessarily end with the text.

To be sure, in the ordinary course, caselaw instructs that courts should be loathe to go beyond the text of an unambiguous statute. See, e.g., Lee v. Bankers Trust Co., 166 F.3d 540, 544 (2d Cir. 1999) ("It is axiomatic that the plain meaning of a statute controls its interpretation and that judicial review must end at the statute's unambiguous terms.") (internal citations omitted). However, that is not to say that a court must always and slavishly look only to the text of a statute when seeking to understand its meaning, just because the text is seemingly clear in isolation. A statute that is broadly worded, albeit facially unambiguous, can be reasonably construed in a variety of ways consistent with its text, but construing text without context can lead to results that Congress did not intend or that are even absurd. I am of the view that legislative history — in reliable form, such as conference committee reports and the floor statements of managers — can be useful in discerning legislative meaning when the statutory text is susceptible to different interpretations and the plain language alone yields results that are surprising or absurd (of course, where further investigation reveals that Congress did intend such results, courts must be faithful to Congress's meaning). See William N. Eskridge, Jr., et al., Legislation and Statutory Interpretation 300-01 (2000); Robert A. Katzmann, Courts and Congress 62-68 (1997).

Thus, although we agree as to the outcome, I part company with Judge Parker's powerful dissent as to interpretive method. That is, I do not agree that the majority engages in an improper inquiry by examining the legislative history of the Voting Rights Act with respect to felon disenfranchisement policies, notwithstanding that the broadly worded and facially unambiguous text of 42 U.S.C. § 1973(a) literally covers such policies as readily as it does any others that result in the denial of the right to vote on account of race or color. Because § 1973(a) does not speak specifically to the question we address today, if I saw clear evidence in the authoritative legislative history that the Congress that enacted it intended to exclude felon disenfranchisement policies from its reach, I would so construe it. But when we look to the authoritative legislative history of § 1973(a)'s enactment in 1982 — the materials directly relevant to our present inquiry, under conventional methods of statutory analysis — or even to activity in the immediately preceding Congress, we find complete silence as to whether Congress intended to exclude felon disenfranchisement policies from its reach. Surely, the silence of enacting legislators cannot overcome the unambiguous and broadly worded provisions of a statute that was meant to apply to a multitude of state policies not specifically enumerated in its text — notwithstanding the majority's references to congressional activity in legislative sessions far removed from the Congress that enacted § 1973(a) in 1982, as well as to Congress's assumptions in enacting unrelated legislation.

I see no precedent for not following the plain language under these circumstances. What we have here cannot be that rare situation where "the literal application of a statute will produce a result demonstrably at odds with the intention of its drafters," United States v. Ron Pair Enters., Inc., 489 U.S. 235, 242, 109 S.Ct. 1026, 103 L.Ed.2d 290 (1989) (internal quotations omitted), because we have literally no evidence as to the intention of the drafters of the 1982 provision specifically with respect to felon disenfranchisement policies. Moreover, because reasonable people can differ over the wisdom of extending the Act's coverage to felon disenfranchisement policies, such a result cannot be considered "absurd" or "unthinkable." See Green v. Bock Laundry Machine Co., 490 U.S. 504, 527, 109 S.Ct. 1981, 104 L.Ed.2d 557 (Scalia, J., concurring). If Congress has a different view, then it is free to amend § 1973(a) accordingly. But this court has no license on its own to do so.

Cases of statutory interpretation are sometimes decided by judges' differing views as to when it is appropriate to look beyond the plain text of the statute and examine authoritative legislative history, but this is not such a case. Whether one stops with the plain language of an unambiguous provision, as Judge Parker advocates, or looks at the authoritative legislative history in addition, as I do, the result should be the same.