concurring in part and dissenting in part.
I concur with the majority’s holding that the defendants are entitled to summary judgment with respect to the plaintiffs’ Equal Protection claim. Whatever discriminatory motives may have prompted Florida to enact the 1868 criminal disenfranchisement provision, the plaintiffs presented no evidence that intentional discrimination motivated the 1968 Constitutional Revision Committee. As a matter of law, the state met its burden by re-enacting the felon disenfranchisement provision without an impermissible motive, as suggested by Hunter v. Underwood, 471 U.S. 222, 228, 105 S.Ct. 1916, 1920, 85 L.Ed.2d 222 (1985). For the reasons stated by the majority, I concur in affirming the district court’s resolution of this claim.
I write separately, however, to dissent from the majority’s conclusion that racially discriminatory felon disenfranchisement provisions are not cognizable under § 2 of the Voting Rights Act (“VRA”). The majority overstates the case for constitutional avoidance. Because it is possible to harmonize the text of the VRA with the Constitution, we should not stray from the plain text of the statute. See City of Rome v. United States, 446 U.S. 156, 172, 100 S.Ct. 1548, 1559, 64 L.Ed.2d 119 (1980).
*1240I. SCOPE of the Voting Rights Act
Section 2 of the VRA prohibits “voting qualification[s] ... imposed or applied by any State” that “results in a denial or abridgement” of the right to vote “on account of race or color.” 42 U.S.C. § 1973(a). As a comprehensive and expansive remedy for racially discriminatory denials of the right to vote, § 2 reaches a wide variety of electoral practices and schemes.1 The standard for evaluating a § 2 claim is “based on the totality of the circumstances.” 42 U.S.C. § 1973(b). As a purely textual matter, a voting qualification based on felony status that interacts with social and historical conditions to produce a racially discriminatory effect, such as race bias in the criminal justice system, falls within the scope of the VRA. See Thornburg v. Gingles, 478 U.S. 30, 45-47, 106 S.Ct. 2752, 2763-64, 92 L.Ed.2d 25 (1986) (describing interactive standard that accounts for “past and present reality”).
The majority fears that interpreting the VRA in this manner “raises serious constitutional problems.” Consequently, the majority construes the statute to avoid this “conflict,” reading the VRA to preclude challenges to criminal disenfranchisement provisions.
To reach its result, the majority places great stock in § 2 of the Fourteenth Amendment, which reduces a state’s representation in Congress when the state denies its male citizens the right to vote unless it abridges the right “for participation in rebellion, or other crime.” U.S. Const, amend. XIV, § 2. The majority characterizes this clause as sui generis— deeply rooted in our nation’s history and fixed by the text of the Constitution. However, that Section does not constitute an affirmative grant of state power to disenfranchise criminals. Rather, as a Reconstruction Amendment, this Section was intended to punish states that were slow to grant the franchise by reducing their representation in Congress. In holding that states have Unfettered discretion to disenfranchise criminals, the majority relies upon a clause that is an exception to this punishment.
Unlike the majority, I do not see a need to construe the statute in this manner. The “avoidance” canon of construction applies if there is ambiguous statutory language. See Southlake Prop. Assoc., Ltd. v. City of Morrow, 112 F.3d 1114, 1119 (11th Cir.1997). Where, as here, there is no ambiguity, the “avoidance” doctrine should not be employed as a pretext for rewriting clear statutory language. Harris v. Garner, 216 F.3d 970, 984-85 (11th Cir.2000) (citation omitted).
Furthermore, I do not think that § 2 of the Fourteenth Amendment amounts to a right to disenfranchise citizens at will, *1241heedless of the consequences. It is a right only by implication, and therefore does not conflict with Congress’s power to limit criminal disenfranchisement. The Fourteenth Amendment does not define the outer limits of a state’s “right” to disenfranchise criminals, but it is certain that a state’s right to disenfranchise is not absolute. States may not intentionally disenfranchise felons on account of race. Hunter, 471 U.S. at 233, 105 S.Ct. at 1922. Likewise, I believe that a state may not disenfranchise criminals in a manner resulting in a racially discriminatory denial of the right to vote. This is because the VRA reaches conduct for which it may not always be possible to prove purposeful discrimination. The VRA recognizes that discriminatory effects are probative of race bias in electoral schemes and practices.
In sum, § 2 of the Fourteenth Amendment does not conflict with Congress’s attempts to prohibit criminal disenfranchisement that is not racially neutral. That clause does not limit Congress’s power to prohibit a voting qualification that results in a denial of equal access to the electoral process on the basis of race or color. See 42 U.S.C. § 1973. I see no conflict between the Constitution and the VRA in this regard, and therefore I see no reason to interpolate an exemption that does not exist.
I do not quarrel with the state’s discretion to disenfranchise felons as a matter of policy. Rather, I take issue with the majority’s characterization of that discretion. Far from “denying] Florida the discretion to disenfranchise felons,” as the majority fears, states are free “to disenfranchise convicted felons in a racially neutral manner — that is, in a manner that is neither racially motivated nor produces racially disproportionate effects.” Johnson v. Bush, 353 F.3d 1287, 1306 n. 27 (11th Cir.2003). My view is that the Voting Rights Act prohibits criminal disenfranchisement provisions that accomplish denial of the right to vote “on account of race.” 42 U.S.C. § 1973(a).
Importantly, I am not convinced that the plaintiffs have proven their case. The plaintiffs’ statistical evidence raises an inference that the disparate impact of felon disenfranchisement results from the interaction of that scheme with race bias in the criminal justice system and the lingering effects of racial exclusion.2 In a trial on *1242the merits, the defendants would be given every opportunity to present rebuttal evidence to question the methodology of this analysis or to present their own analysis explaining the disparity. I dissent because I believe that the district court’s resolution of the merits was premature and that the plaintiffs were entitled to present their case at trial.
II. Scope of CongRess’s Powee
The majority also suggests that, were a § 2 VRA claim challenging criminal disenfranchisement provisions cognizable, Congress might have exceeded its enforcement powers of the Fourteenth and Fifteenth Amendments. I respectfully disagree with this conclusion.
Congress’s enforcement authority is at its most expansive when protecting against discrimination based on suspect classifications or when protecting fundamental rights. Thus, to carry out the basic objectives of the Fourteenth and Fifteenth Amendments, Congress may enact “prophylactic legislation proscribing practices that are discriminatory in effect, if not in intent.” Tennessee v. Lane, 541 U.S. 509, 124 S.Ct. 1978, 1986, 158 L.Ed.2d 820 (2004) (upholding Title II of the Americans with Disabilities Act as a valid exercise of Congress’s Fourteenth Amendment enforcement power); see also City of Rome, 446 U.S. 156, 100 S.Ct. 1548, 64 L.Ed.2d 119 (upholding VRA’s § 5 preclearance requirement for covered jurisdictions seeking electoral changes as a valid exercise of Congress’s Fifteenth Amendment enforcement power).
Despite the strength of Congress’s remedial enforcement power, it is not without limits. Congress’s remedy must respond to states’ actual violations of a protected right. See Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 89, 120 S.Ct. 631, 649, 145 L.Ed.2d 522 (2000); Fla. Prepaid Postsecondary Educ. Expense Bd. v. Coll. Sav. Bank, 527 U.S. 627, 640, 119 S.Ct. 2199, 2207, 144 L.Ed.2d 575 (1999). Therefore, Congress must identify the “history and pattern of unconstitutional ... discrimination” that it seeks to address, creating a legislative record to support its exercise of enforcement power. Bd. of Trustees of the Univ. of Ala. v. Garrett, 531 U.S. 356, 368, 121 S.Ct. 955, 148 L.Ed.2d 866 (2001). Furthermore, valid § 5 legislation must exhibit “congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.” City of Boerne v. Flores, 521 U.S. 507, 520, 117 S.Ct. 2157, 2164, 138 L.Ed.2d 624 (1997).3
Congress enacted the VRA pursuant to the enforcement clauses of the Fourteenth and Fifteenth Amendments in response to rampant violations of the right to vote. United States v. Bd. of Comm’rs, 435 U.S. 110, 126-27, 98 S.Ct. 965, 976-77, 55 L.Ed.2d 148 (1978). The scope of the VRA and those amendments are not coterminous: As a remedial statute, the VRA reaches beyond what is prohibited by the Fourteenth and Fifteenth Amendments. See, e.g., United States v. Marengo County Comm’n, 731 F.2d 1546 (11th Cir.1984). The VRA not only safeguards a fundamen*1243tal right, voting, but also protects against discrimination based on race, a suspect class. See Lane, 124 S.Ct. at 1992 (noting that because fundamental rights, such as access to the courts, receive heightened scrutiny, review of the legislative record for a pattern of constitutional violations is less searching); Nev. Dep’t of Human Res. v. Hibbs, 538 U.S. 721, 735-36, 123 S.Ct. 1972, 1981-82, 155 L.Ed.2d 953 (2003) (explaining that for suspect classifications, such as sex, that are subject to heightened scrutiny, it is “easier for Congress to show a pattern of state constitutional violations”). The Act is properly considered a remedial statute, passed with the broadest exercise of power allowed under the Fourteenth and Fifteenth Amendments.
The VRA is entitled to a broad reading because Congress has chronicled extensive state violations of the right to vote. In 1965, when first enacting the VRA, Congress documented violations of the Fifteenth Amendment, including “grandfather clauses” that permitted previously registered voters (all white) to register without taking a literacy test, laws restricting the participation in political primaries to whites only, procedural hurdles, racial gerrymandering, improper challenges, and the discriminatory use of tests. See H.R.Rep. No. 89-439, at 8 (1965) (citing Supreme Court decisions). Congress was particularly concerned about states’ use of tests that discriminated against racial minorities, including literacy tests, constitutional interpretation tests, and tests concerning the obligations of citizenship. H.R.Rep. No. 89-439, at 12.
All of these devices worked in concert to depress the registration and turnout rate among voting-age African Americans. For example, prior to the VRA only 6.7% of the African-American voting age population in Mississippi was registered to vote. In Alabama, the registration rate of African Americans lagged behind that of whites by 49.9%. See S.Rep. No. 94-295, at 13 (1975), reprinted in 1975 U.S.C.C.A.N. 774, 779. Perhaps more problematic was the revelation that innovation in discrimination marked the landscape of voting rights. See S.Rep. No. 89-162, at 5 (1965); S.Rep. No. 89-439, at 10 (“[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.”) (citing United States v. Mississippi, 380 U.S. 128, 85 S.Ct. 808, 13 L.Ed.2d 717 (1965), and United States v. Penton, 212 F.Supp. 193 (M.D.Ala.1962)). Congress found specifically that it was impossible to predict the variety of means that would be used to infringe on the right to vote.
In response, Congress passed the VRA, which operates on two levels. First, Congress recognized that some areas of the country had a particularly bad history of voting discrimination. Section 5 of the VRA thus designated these regions as “covered jurisdictions,” requiring them to clear any changes in voting or election laws with either the Attorney General or a federal court in the District of Columbia before putting them into effect. 42 U.S.C. § 1973c. The approval of the Attorney General or the court was conditioned on a showing that the changes would not discriminate in purpose or in effect. Additionally, because Congress understood that violations of voting rights were not confined to covered jurisdictions, it included § 2, a nationwide remedy less intrusive on states’ functions. See S.Rep. No. 97-417, at 41-42 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 219-20.
It is of no moment that the VRA’s legislative record does not contain specific examples of discrimination based on felon status. Boeme and its progeny require that the legislative record show a pattern of state constitutional violations, not that the right at issue be abridged in a particu*1244lar way. See Lane, 124 S.Ct. at 1988-90; Garrett, 531 U.S. at 368-69, 121 S.Ct. at 964-65; Kimel, 528 U.S. at 89, 120 S.Ct. at 648-49; Boerne, 521 U.S. at 530, 117 S.Ct. at 2169. If this were the standard, states would always have one free bite at the apple, which is not what Congress intended when it passed the VRA to deal with voting discrimination “comprehensively and finally.” S.Rep. No. 97-417, at 5, reprinted in 1982 U.S.C.C.A.N. at 182.
Congress has found that racial discrimination in voting has a long history in our country. The remedy that Congress chose to respond to the pattern of state discrimination is to prohibit voting discrimination in whatever form it takes. See Gingles, 478 U.S. at 45 n. 10, 106 S.Ct. at 2764 n. 10 (“Section 2 prohibits all forms of voting discrimination .... ”). This remedy is congruent and proportional to the goal of enforcing the fundamental right to vote. When voting interacts with proof of racial bias in criminal justice, Congress is well within its power to prohibit the resultant discrimination.
. See, e.g., Morse v. Republican Party of Va., 517 U.S. 186, 116 S.Ct. 1186, 134 L.Ed.2d 347 (1996) (exclusion of protected groups from a nominating convention); White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973) (multi-member districts); Arakaki v. Hawaii, 314 F.3d 1091 (9th Cir.2002) (restriction on race of candidate); McMillan v. Escambia County, 748 F.2d 1037 (11th Cir.1984) (majority vote requirement in at-large election primary); United States v. Marengo County Comm’n, 731 F.2d 1546, 1574 (11th Cir.1984) (at-large elections); DeGrandy v. Wetherell, 794 F.Supp. 1076 (N.D.Fla.1992) (single-member districting plans); Dillard v. Town of North Johns, 717 F.Supp. 1471 (M.D.Ala.1989) (selective withholding of candidacy requirement information and forms); Harris v. Siegelman, 695 F.Supp. 517 (M.D.Ala.1988) (refusal to appoint minority registration and election officials); Irby v. Fitz-Hugh, 693 F.Supp. 424 (E.D.Va.1988) (shift from elective to appointive system); Brown v. Dean, 555 F.Supp. 502 (D.C.R.I.1982) (location of polling place); Arroyo v. Tucker, 372 F.Supp. 764 (E.D.Pa.1974) (English-only election system, where residents born in Puerto Rico do not speak, read, write, or comprehend English).
. Judge Tjoflat argues in his concurrence that a plaintiff’s showing of disproportionate effect does not suffice to make out a prima facie case of a § 2 VRA claim. It is far from clear that the language of § 2 VRA claim reinstates the law of the pr e-Bolden cases. Congress could have specified in § 2 that some of the White factors were strongly probative of intent, and that proof of "something more than effect” was the sine qua non of a § 2 claim, regardless of whether the proof was direct or indirect. See White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973). Given Congress’s express language, however, it is plausible that the amended § 2 was meant to ensure racial minorities equal access to the vote, regardless of intent. See, e.g., City of Mobile v. Bolden, 446 U.S. 55, 103-41, 100 S.Ct. 1490, 1519-1539, 64 L.Ed.2d 47 (1980) (Marshall, J., dissenting).
Indeed, it is far from clear that the state of pre-Bolden law did not recognize vote dilution claims based solely on effect. See Burns v. Richardson, 384 U.S. 73, 88, 86 S.Ct. 1286, 1294, 16 L.Ed.2d 376 (1966), (stating that the standard for vote dilution claim is "invidious effect”); Fortson v. Dorsey, 379 U.S. 433, 439, 85 S.Ct. 498, 501, 13 L.Ed.2d 401 (1965) ("It might well be that, designedly or otherwise, a multi-member constituency 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.” (emphasis added)).
In the event that Judge Tjoflat's interpretation is the correct one (that proof of “something more” is required), I do not think that interpretation leads inexorably to the conclusion that a plaintiff's showing of disproportionate effect does not satisfy the summary judgment burden. On the contrary, one of the permissible inferences that can, and must, be drawn from the plaintiffs’ showing here is *1242that the disproportionate effect was caused by race. Thus, I believe that the plaintiffs have carried their summary judgment burden by producing evidence of a disproportionate effect.
. It bears noting that the Supreme Court has yet to determine whether the "congruence and proportionality” test applies to the Fifteenth Amendment. I assume here that it would. Because the VRA is a congruent and proportional remedy, and therefore well within Congress's enforcement power under the Fourteenth Amendment, an inquiry into the Fifteenth Amendment's scope is not required.