IS SECTION 5 OF THE VOTING RIGHTS ACT A VICTIM OF ITS OWN SUCCESS? Samuel Issacharoff* With the 2007 renewal date for section 5 of the Voting Rights Act now approaching, the question must be addressed whether the legal and practical preconditions for this extraordinary statute still exist. This Essay suggests that there were four preconditions necessary for the striking successes that section 5 had in transforming politics in its covered jurisdictions: the urgency of swift intervention to counteract the complete exclusion of black citizens from political life in the South; the ease of the administrative remedy; the absence of political competition in the one-party covered jurisdictions; and the lack of any incentive toward partisan manipulation of the preclearance powers exercised by the Department of Justice. Each of these factors has been changed by the creation of a robust political environment in the jurisdictions covered by section 5, particularly by the establishment of an important core of influential black elected officials. This leads to the question whether the success of section 5 has compromised its mission, as reflected in the major decisions under section 5 following the post-2000 reapportionment. The Essay concludes by questioning whether section 5 has served its purpose and may now be impeding the type of political developments that would have been a distant aspiration when the Voting Rights Act was first passed. INTRODUCTION The approaching renewal date for section 5 of the Voting Rights Act (VRA)1 raises the question whether the preconditions for the successes of this extraordinary statute continue to exist. The Voting Rights Act, as the Supreme Court noted both in its early assessments of the Act's constitutionality and in distinguishing it from the Religious Freedom Restoration Act,2 was premised on a deep sense of national urgency over the exclusion of black Americans from meaningful political participation in significant parts of the country--those areas that were then classified as covered jurisdictions, a classification newly created by section 5. Section 5 placed political life in those jurisdictions under a form of administrative receivership and treated political activity within those areas as subject to a rebuttable presumption that the continued exclusion of blacks from meaningful political opportunity was the dominant feature of all political decisionmaking in those jurisdictions. As the 2007 date for renewal emerges on the political landscape, it is worth reviewing the exceptional role of section 5 of the Voting Rights Act. One of the central features of the 1965 Voting Rights Act was the * Harold R. Medina Professor of Procedural Jurisprudence, Columbia Law School. 1. 42 U.S.C. ? 1973c (2000). The Voting Rights Act is codified in whole in sections 1973 to 1973aa-6. 2. Pub. L. No. 103-141, 107 Stat. 1488 (1993) (codified at 42 U.S.C. ?? 2000bb to bb4). 1710 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1711 suspension of all literacy tests and other devices that had served to disenfranchise Southern blacks since the end of Reconstruction. Beginning in the period of "Redemption" at the end of the 19th century, literacy tests, poll taxes, restricted access to registration, and other such mechanisms had been the hallmarks of the disenfranchisement of the post-Civil War South's black population.3 Such devices were suspended under section 4(a) for a period of five years in any "covered" jurisdiction. Section 4(b) provided the coverage formula to include any jurisdiction that had used a literacy test (or later used English-only election materials in jurisdictions with heavily non-English speaking populations) and where voter registration in presidential elections was below fifty percent of the voting-age population.4 Any jurisdiction falling under the Act's coverage formula was then forbidden by section 5 of the Act to alter any election practice absent either "preclearance" by the Department of Justice or a declaratory judgment action in the District Court for the District of Columbia. The effect of preclearance was to provide a one-way ratchet for minority political gains. Particularly after the Court's 1969 decision in Allen v. State Board of Elections,5 section 5 provided oversight not only on the processes of registering and casting a ballot, but on issues such as annexations and the use of at-large election districts. In Allen, for example, the question presented was whether section 5 would be limited to the ability to cast a vote, or would reach the effectiveness of the vote. In extending the reach of section 5 to include the electoral prospects of minority-preferred candidates, the Court gave invaluable protection to fledgling minority political successes in the early stages of the civil rights era. As expressed by Pamela Karlan, "section 5 contains a natural benchmark that preserves the political gains minority voters have achieved through political or legal action."6 That benchmark is "the status quo that is proposed to be changed."7 The benchmark was preserved by freezing in place local political arrangements and imposing exacting administrative review upon covered jurisdictions. In practice, section 5 coverage denied local jurisdictions a customary range of political decisions--including districting, terms of office, and electoral systems--that were ordinarily subject to what Justice Souter would term the pulling and hauling of everyday politics.8 3. For discussions of the history of the use of such devices to frustrate black voting, see South Carolina v. Katzenbach, 383 U.S. 301, 308-13 (1966); see also Samuel Issacharoff, Pamela S. Karlan & Richard H. Pildes, The Law of Democracy: Legal Structure of the Political Process 546-671 (rev. 2d ed. 2002). 4. 42 U.S.C. ? 1973b(b). Nine states and parts of seven others are currently covered jurisdictions, mostly but not exclusively in the deep South. 28 C.F.R. pt. 51 appx. (2003). 5. 393 U.S. 544 (1969) (extending preclearance beyond ability to vote to encompass forms of governance as well). 6. Pamela S. Karlan, Georgia v. Ashcroft and the Retrogression of Retrogression, 3 Election L.J. 21, 21 (2004). 7. Reno v. Bossier Parish Sch. Bd. (Bossier Parrish II), 528 U.S. 320, 334 (2000). 8. Johnson v. De Grandy, 512 U.S. 997, 1020 (1994). 1712 COLUMBIA LAW REVIEW [Vol. 104:1710 In light of the tremendous political gains for minorities covered by section 5, particularly Southern blacks, it may be perverse to even question the need to extend section 5 after its current sunset in 2007. But, as courts reviewing section 5 have noted since its original implementation, the conditions for the success (and constitutionality) of section 5 involve an intensely practical assessment of the justifications for displacing the normal functions of politics. This Essay addresses the question whether the evolution of politics since the last extension of section 5 in 1982 has altered the conditions for its continued utility as a first-order mechanism to oversee minority participation in the political process. This Essay suggests that four preconditions allowed this extraordinary intervention to work effectively and to retain the clear sense of purpose that permitted it to overcome the normal presumptions of state autonomy and respect for federalism. Two of these preconditions are fairly well understood in analyses of section 5. The first is the urgency and extent of the harm to which Congress addressed itself in placing much of the country under a regime of what amounts to administrative receivership. As expressed in South Carolina v. Katzenbach, "[t]he constitutional propriety of the Voting Rights Act of 1965 must be judged with reference to the historical experience which it reflects."9 That historical experience demonstrated to the Court's satisfaction that the underlying assault on black voting rights constituted a clear violation of the Fourteenth and Fifteenth Amendments.10 Even where the evidentiary basis for the claim of widespread discrimination was more tenuous, as with the nationwide extension of section 4 in 1970, the Court nonetheless returned to its empirical assessment that "Congress was supported by substantial evidence in concluding that a nationwide ban on literacy tests was appropriate to enforce the Civil War amendments."11 The second condition is also widely appreciated and turns on the nature of the extraordinary preclearance mechanism. Section 5 of the Voting Rights Act and its implementing regulations require the grant or denial of preclearance to occur within sixty days of submission--leaving aside the capacity to obtain extensions of that time. The Act's original triggering mechanism related back to section 4 and the initial suspension of literacy tests. As the Act was applied to increasingly complex governance questions, including redistricting, ease of administration and the scope of potential administrative review came to dominate the Court's concern in section 5 cases. These cases can be divided among three major themes. First, in cases such as Morris v. Gressette, the Court limited the preclearance process to its administrative core and held that as a matter of administrative law the decision to preclear was not subject to challenge or judicial review.12 Second, in Beer v. United States and its progeny, the 9. 383 U.S. 301, 308 (1966). 10. Id. at 333-34. 11. Oregon v. Mitchell, 400 U.S. 112, 133 (1970). 12. 432 U.S. 491, 507 (1977). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1713 Court limited the scope of the preclearance inquiry to the issue of whether or not there was retrogression as a result of the proposed alteration in state voting practices.13 Third, in cases such as Presley v. Etowah County Commission, the Court limited the sweep of section 5 to narrowly cover voting practices and to remove questions about the efficacy of governance from the ambit of preclearance.14 Together these three considerations narrowed section 5 to questions that could be addressed through relatively mechanical assessments of voting practices. For example, the Beer retrogression test made it easy to decide that a districting arrangement in a town with a twenty-percent black population that yielded one forty-five-percent black district (out of five) should not be precleared if the prior arrangement had afforded one district that was sixty-five-percent black. The limitation on presumptive black electoral opportunity required little more than the sort of sixth-grade arithmetic that has its own allure in the voting rights jurisprudence.15 The last two considerations turn out to be relatively unexplored in the literature. Both of these depended on the one-party feature of political life in the covered jurisdictions. The third precondition is that blacks in the historic Jim Crow South had no avenue of political redress. In the one-party South, the VRA served as a blunt tool directed to the failure of political competition. Not only were blacks largely disenfranchised, but even if able to vote, they could never aspire to be influential swing voters in a one-party political environment. So long as the Democratic Party remained unchallenged, and so long as that Party was organized around the principles of resistance to any encroachments on Jim Crow, the political system would remain immune to the pressing claims of black citizens. The effect of the emergence of black voters was to force political parties to attempt to expand their political bases by appealing to the newly enfranchised black electorate. Finally, there is the fourth precondition which, to the best of my knowledge, has gone unnoticed in judicial and scholarly accounts of the VRA. As long as the South (to use a convenient shorthand for the bulk of the covered jurisdictions) remained entirely Democratic, the tremendous powers given to the federal government to intercede in local political affairs in the covered jurisdictions could not be used for partisan gain. Section 5 served to ratchet up black political power at the expense of the encrusted white establishment of the Democratic Party. Until the reemergence of the Republican Party in the South, intervention from Washington could alter the racial dimensions of political power, but not the partisan divide. In paradoxical fashion, the more Southern politics continued to be organized around the retrograde isolation and suppres13. 425 U.S. 130, 141 (1976); see, e.g., City of Lockhart v. United States, 460 U.S. 125, 134 (1983). 14. 502 U.S. 491, 504 (1992). 15. See Lucas v. 44th Gen. Assembly, 377 U.S. 713, 750 (1964) (Stewart, J., dissenting). 1714 COLUMBIA LAW REVIEW [Vol. 104:1710 sion of black political interests, the more enlightened and noble would be the intervention from Washington. The question for today is whether the successes of the VRA have compromised its mission. No longer are blacks political outsiders in the covered jurisdictions. To take but a single example, the major Supreme Court cases of the past decade addressing minority representation, beginning with Shaw v. Reno,16 have all arisen from claims of Southern politics being too solicitous of minority political claims. As the responsiveness of Southern legislatures to claims for minority representation reveal (the Supreme Court's intervention in the Shaw cases notwithstanding and the Court's nuanced treatment of Georgia v. Ashcroft 17 confirming), the Southern political process is highly attuned to black political claims. The presumption that Washington represents the only forum for safeguarding black political advancement is tenuous at best. Moreover, the development of real partisan competition in the South, and the clear identity of black interests with the electoral prospects of the Democratic Party, lend a partisan tinge to the intervention from Washington. It is not a surprise that the two section 5 inspired cases in the Supreme Court this past Term, arising in Mississippi18 and Georgia,19 were heavily laden with charges of partisan misuse of the preclearance provisions of the VRA. Those charges were already widely heard in the 1990s round of redistricting. In the post-2000 round, not only are those charges being raised again, but it is increasingly difficult to discern the positive role that the extraordinary powers of section 5 now play. I. RULES, STANDARDS, AND ADMINISTRATIVE PRECLEARANCE Beginning with the Court's first assessment of section 5 in South Carolina v. Katzenbach,20 one clear source of constitutional concern was the intrusiveness of the VRA on local implementation of election laws which had long been seen as a matter of state prerogative. In the intervening decades, the Court has returned time and again to questions of federalism and has demanded far greater justification for federal incursions on matters traditionally reserved for the states. In light of the major federalism decisions of late, City of Boerne v. Flores 21 and Nevada Department of Human Resources v. Hibbs,22 it is certainly possible to revisit the Court's acceptance of the constitutionality of section 5 dating back to the seminal cases of the 1960s and 1970s. Most notably, in City of Boerne the Court 16. 509 U.S. 630 (1993); see also Easley v. Cromartie, 532 U.S. 234, 243-44 (2001); Shaw v. Hunt, 517 U.S. 899, 902-03 (1996); Bush v. Vera, 517 U.S. 952, 957 (1996); Miller v. Johnson, 515 U.S. 900, 907-09 (1995). 17. 539 U.S. 461 (2003). 18. See Branch v. Smith, 538 U.S. 254, 254 (2003). 19. See Ashcroft, 539 U.S. at 461. 20. 383 U.S. 301 (1966). 21. 521 U.S. 507 (1997). 22. 538 U.S. 721 (2003). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1715 struck down the Religious Freedom Restoration Act on the grounds that the legislative record was insufficient to establish the need for federal intervention. As set out in City of Boerne, federalism constraints doomed the lack of "congruence and proportionality" of the Act's sweeping remedial scheme in light of the quality of evidence before the Congress.23 Although in reaching its decision the Court repeatedly distinguished the Religious Freedom Restoration Act from the VRA on the basis of the greater factual justification for the latter, the passage of time might bode poorly for a clear legislative finding of the continued need for section 5. Subsequently, in Hibbs, the Court appeared to back off the sterner implications of City of Boerne, indicating that it might very well carve out a protected area for discrimination concerns along the classic frontiers of suspect classes.24 The fact that Hibbs addressed the application of the Family and Medical Leave Act, a form of protection against sex rather than race discrimination, would appear to give Congress even greater latitude to craft remedial legislation in areas of traditional equal protection strict scrutiny. This issue is both significant and addressed in other work exploring the status of section 5 as the issue of its re-extension in 2007 approaches.25 Beyond the role of the congruence between specific factfinding and the scope of section 5, however, its constitutionality may also be thought to rest on the parsimony of the administrative model it employs. It is difficult to read cases such as Beer and Presley without noting the Court's penchant for defining the scope of permissible section 5 review to that which would be "straightforward," to adopt the terminology of Beer.26 Although these cases arise as matters of statutory interpretation, ample authority supports the proposition that the Court's statutory construction should favor rather than condemn the constitutionality of a statute. Likewise, there is authority for constraining the scope of more open-ended standards--as opposed to less discretionary rules--when placed in the hands of nonjudicial actors.27 Indeed, this issue was presented indirectly in cases such as Miller v. Johnson where the Court addressed the "limited 23. City of Boerne, 521 U.S. at 520. 24. 538 U.S. at 736. 25. See, e.g., Michael J. Pitts, Section 5 of the Voting Rights Act: A Once and Future Remedy, 81 Denv. U. L. Rev. 225 (2003); Richard Hasen, Congressional Power to Renew the Preclearance Provisions of the Voting Rights Act after Tennessee v. Lane, 66 Ohio St. L.J. (forthcoming Feb. 2005) (on file with the Columbia Law Review). 26. Beer v. United States, 425 U.S. 130, 141 (1976); see also Presley v. Etowah County Comm'n, 502 U.S. 491 (1992). 27. See generally Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 591 (1992); Kathleen M. Sullivan, The Supreme Court, 1991 Term-- Foreword: The Justices of Rules and Standards, 106 Harv. L. Rev. 22, 61-69 (1992). As expressed in our casebook on this point, "Can any procedural rule, such as preclearance, function mechanically if it sometimes requires a substantive determination, such as the existence of a 'potential for discrimination,' to decide whether the procedural rule should apply?" Issacharoff et al., supra note 3, at 590. 1716 COLUMBIA LAW REVIEW [Vol. 104:1710 substantive goal" of section 5.28 In deciding that compliance with section 5 could not be a sufficiently compelling state interest to survive equal protection strict scrutiny, the Court refused to allow the administrative processes of section 5 preclearance to reach the multifactored assessments of constitutionality. Although not addressed in constitutional terms, the issue of administrability surfaced directly in Georgia v. Ashcroft. It is difficult to imagine a setting as far from the initial concerns of section 5 as that of the Georgia politics giving rise to the Ashcroft litigation. The issue was whether preclearance should be granted to Georgia's post-2000 redistricting of its legislative districts. Rather than create as many black-dominated election districts as possible and use section 5 to protect against any diminution of concentrated black political power, the new version of the Georgia Democratic Party, with full participation of the now significant black political establishment, sought to capitalize on its control of redistricting to expand the political impact of the black vote. The plan as designed by Democrats in the state legislature kept true to the dual goals of maintaining at least as many majority-minority districts while also attempting to increase Democratic strength in the Senate. Part of the Democrats' strategy was not only to keep the same number of majority-minority districts, but to leverage black voting strength by diminishing the concentration of black voters in minority-dominated districts. Spreading black voters across a broader array of districts would in turn increase the number of so-called "influence" districts, where black voters would be able to exert a significant--if not decisive--force in the election process. As the majority leader testified, "in the past, you know, what we would end up doing was packing. You put all blacks in one district and all whites in one district, so what you end up with is [a] black Democratic district and [a] white Republican district."29 The key to the Senate plan was, in the Court's words, to "unpack"30 the black-concentrated districts, distribute the black vote around more Democratic districts, and actually increase the number of districts that were majority-black in voting-age population.31 But the black voters had to come from somewhere and, in order to maximize perceived Democratic prospects, "the new plan reduced by five the number of districts with a black voting age population in excess of 60%."32 There was no mistaking that this was not a plan imposed on black elected officials, nor was there any mistaking the evident partisan objectives of the plan: The Senate adopted its new districting plan on August 10, 2001, by a vote of 29 to 26. Ten of the eleven black Senators voted for the plan. The Georgia House of Representatives passed the Sen28. 29. 30. 31. 32. 515 U.S. 900, 926 (1995). Georgia v. Ashcroft, 539 U.S. 461, 470 (2003). Id. Id. Id. 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1717 ate plan by a vote of 101 to 71. Thirty-three of the thirty-four black Representatives voted for the plan. No Republican in either the House or the Senate voted for the plan, making the votes of the black legislators necessary for passage. The Governor signed the Senate plan into law on August 24, 2001, and Georgia subsequently sought to obtain preclearance.33 At the same time that the new Senate plan sought to leverage black political strength, it also diminished the locked-in protections of the overwhelmingly black-concentrated districts. The question, therefore, was whether this dispersal of concentrated black voting power was retrogressive under the well-established Beer standard. This was a watershed for section 5 jurisprudence. So long as the question was whether there was an arithmetic diminution in the number of clearly black-controlled districts or in the level of black concentration in those districts, the answer seemed foreordained. At the same time, it would be an irony of historic proportions if the VRA were to emerge as a brake on black political aspirations in the heart of the Deep South. It is no small tribute to the successes of the VRA that it would someday have to confront the independent political ambitions of empowered black voters. The striking feature about Ashcroft was the willingness of the entire Court to abandon the formal Beer standard for retrogression in favor of a more nuanced assessment of the on-the-ground political realities of a jurisdiction. Under Beer, preclearance was limited to a mandate "that the minority's opportunity to elect representatives of its choice not be diminished, directly or indirectly, by the State's actions."34 Even a plausible claim of discriminatory effect was not deemed to be an appropriate basis for the interposition of a Department of Justice objection under section 5, so long as the rather mechanical standard of nonretrogression was satisfied.35 This clear mandate of using section 5 to freeze in place the status quo ante unraveled in Ashcroft. Justice O'Connor introduced a new test for section 5 that decisively abandoned the prior reliance on a straight-line mathematical comparison of minority voting strength before and after. Instead, Justice O'Connor would have the preclearance decision turn on the totality of the political factors that might affect minority political--as opposed to narrowly electoral--prospects. Thus, even diminished numbers of minority elected officials could survive preclearance since "[m]aintaining or increasing legislative positions of power for minority voters' representatives of choice . . . can show the lack of retrogressive effect under ? 5."36 The clear implication is that the preclearance process should be faithful to the nuanced 33. Id. at 471 (citations omitted). 34. Bush v. Vera, 517 U.S. 952, 983 (1996) (emphasis omitted). 35. See City of Lockhart v. United States, 460 U.S. 125, 135 (1983) (permitting a change that simply perpetuated the existing, arguably discriminatory, situation because "[a]lthough there may have been no improvement in their voting strength, there has been no retrogression either"). 36. Ashcroft, 539 U.S. at 484. 1718 COLUMBIA LAW REVIEW [Vol. 104:1710 preferences of minority voters who might sensibly favor having fewer, but more powerful, legislative representatives to successfully electing a greater number of less powerful candidates who would face legislative marginalization. The empirical foundation for this inquiry might include facts as disparate as the seriousness and financing of minority preferred candidates approaching the election and the relative power that they might hold in legislative caucuses after the election. The plurality devoted careful attention to the complexities of Georgia politics, a point to which I will return in the next section, and then stated that, by contrast to the mechanical retrogression test from Beer, "the ability of a minority group to elect a candidate of choice remains an integral feature in any ? 5 analysis."37 No reading of the plurality opinion in Ashcroft can fail to recognize that the Court substituted a highly nuanced totality-of-the-circumstances approach for the relatively rigid Beer retrogression test. Significantly, Justice O'Connor claimed authority for this rendering of section 5 not from any of the preclearance case law, but from her concurrence in the defining case under section 2 of the Voting Rights Act, Thornburg v. Gingles.38 The difference between the two sections of the VRA is significant. As set out above, section 5 is an administrative mechanism designed to lock in place the status quo ante, unless the legislature demonstrates that any proposed changes do not have a retrogressive effect. By contrast, section 2, as amended in 1982, provides for a plenary adjudication of whether a particular electoral practice, such as the use of multimember election districts, is actually discriminatory and should be struck down.39 Rather than assume the benchmark of the prior electoral practice, as under section 5, the test for unlawful minority exclusion under section 2 asks whether, under the totality of the circumstances, a challenged election practice results in minorities having a less than equal opportunity to participate in the political process. Justice O'Connor's view of section 2, as set forth in her dissent in Gingles, resisted the primary focus on polarized voting practices in a particular jurisdiction and instead argued for a much broader inquiry into the entirety of the political factors that promote or resist minority electoral ambitions. Ashcroft provided the means for resurfacing that approach, although now in the administrative posture of section 5 rather than direct litigation under section 2. 37. Id. 38. 478 U.S. 30, 87-89 (1986) (O'Connor, J., concurring). 39. The Senate Report for the 1982 amendments to section 2 of the Voting Rights Act set out "[t]ypical factors" that served to establish a violation of the Act. Section 2 proscribed the use of electoral structures that deny minority voters an equal opportunity to participate in the political process and to elect candidates of their choice. See S. Rep. No. 97-417, at 28-29 (1982). As developed in Justice Brennan's lead plurality opinion in Gingles, the core of the inquiry turned on the presence of polarized voting to identify which electoral structures operated to frustrate minority electoral opportunity. See generally Samuel Issacharoff, Polarized Voting and the Political Process: The Transformation of Voting Rights Jurisprudence, 90 Mich. L. Rev. 1833 (1992). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1719 For the dissent in Ashcroft, the abdication of Beer was grudging and went only so far as to repudiate the per se implications of Beer for any reduction in the number of majority-minority districts in a covered jurisdiction.40 But the introduction of broad-range review prompted what is, in my view, the strongest aspect of Justice Souter's dissent. For Justice Souter, the lack of administrability of the Court's refashioned section 5 is a major unwinding of the statutory scheme. In the first instance, Justice Souter claims--no doubt correctly--that it is unlikely that Congress ever envisioned Justice O'Connor's test for retrogression. While true, this is hardly an argument-stopping claim. Voting rights law is liberally strewn with Court innovations that are either subsequently tacitly adopted by Congress--the Beer standard itself--or, as with the test for vote dilution in Gingles and Johnson v. De Grandy,41 crafted wholly by the Court to bring order and lend coherence to a messy legislative compromise.42 More significant yet may be the parade of horrors associated with the fact that historically section 5 has drawn its force from the ease of administration of a relatively mechanical test for diminution of black electoral concentration. So long as the prototypical harm was defined as keeping black votes concentrated below the level necessary to elect a minority candidate to office, and so long as this harm was defined independently of the extent of polarized voting or turnout or other complicated factors, administrative review by the Department of Justice could be accomplished within the regulatory timeframe and allow elections to proceed if precleared. The breadth of the departure from prior case law under section 5 is evident even from reading the first line of the dissent, which accepted that the Court must ease the mechanical application of Beer.43 But for the critical recasting of section 5, one need look no further than the thirteen citations, by my count, to Justice O'Connor's under-explored concurrence in Gingles. Justice O'Connor's opinion speaks repeatedly of the failure of the district court to capture the complex reality of how politics in Georgia had evolved. The challenge, in effect, was whether section 5 of the Voting Rights Act would serve as a statutory bar to the potential emergence of a "beloved community" of racial harmony, as evocatively expressed by Representative John Lewis in the proceedings below.44 Overlooked, however, was whether anything like the totality-of-thecircumstances test was feasible in the statutory sixty days for review. The dissent asks caustically about consideration being given to the desire for 40. Ashcroft, 539 U.S. at 494-95 (Souter, J., dissenting). 41. 512 U.S. 997 (1994). 42. Gingles, 478 U.S. at 50-51; see also Johnson, 512 U.S. at 1013-14. 43. Ashcroft, 539 U.S. at 492 (Souter, J., dissenting) ("I agree with the Court that reducing the number of majority-minority districts within a State would not necessarily amount to retrogression barring preclearance under ? 5 of the Voting Rights Act of 1965."). 44. Id. at 490 (opinion of the Court). 1720 COLUMBIA LAW REVIEW [Vol. 104:1710 legislative clout or even the charisma of particular candidates in particular districts. But the key challenge is whether such multifactored inquiries defy "reviewable administration."45 Justice O'Connor does not satisfactorily come to terms with this challenge. The difficulty may be that the majority holding depends on Justice Kennedy's indispensable fifth vote. Justice Kennedy uses the occasion to tentatively test the continued constitutional viability of the section 5 scheme. For him, the potential constitutional infirmity lies in the tension between the commanding race consciousness of section 5 and the cautionary notes of Shaw v. Reno and Miller v. Johnson. Thus, in order to rebut the challenge posed by constitutional limitations on race consciousness, Justice O'Connor would have to have confronted the compelling component of equal protection strict scrutiny. To the extent that the constitutionality of the VRA turns on the widespread exclusion and isolation of blacks in the political process, recognition of multiracial politics is potentially destabilizing to the core premises of the statutory scheme. At the same time, however, any attempt to dampen the demanding racial concerns of section 5 may very well reopen the iconic account of the constitutionality of the VRA in City of Boerne. It is difficult to argue for the congruence of a statutory response to the extreme isolation of blacks in the political process at the constitutional plane, while at the same time arguing for the importance of coalitional politics as the basis for statutory implementation. II. GEORGIA BY WAY OF NEW JERSEY The focus on the administrability of section 5 helps illustrate the underlying difficulty in fashioning a statutory design that can account for the nuances of a competitive political process. In moving section 5 from a bright-line rule under Beer to an assessment of the competing political considerations in securing effective black representation, the Court introduced for the first time to section 5 the fine grained calculus of trade-offs of political influence versus descriptive representation. This is likely one reason that Judge Emmet Sullivan, writing for the three-judge district court in Ashcroft, held so firmly to a limited inquiry into whether there was still polarized voting in Georgia and whether the plan met the Beer standard for diminution in black voting strength in the black-majority districts.46 For the majority below and the dissent in the Court, the conclusion was that the Georgia plan should fail because of its reduction of black voting concentrations--regardless of the practical impact that decision may have had on either interracial coalitional politics or on the prospect for black effective political power at the state legislative level. This is perhaps most cogently captured in the rather stark assessment of the goal of 45. Id. at 496 (Souter, J., dissenting). 46. Georgia v. Ashcroft, 195 F. Supp. 2d 25, 73-76 (D.D.C. 2002). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1721 section 5 advanced by the concurring opinion of Judge Harry Edwards in the three-judge court: "Our dissenting colleague argues that ? 5 is satisfied whenever a covered jurisdiction adopts a plan that preserves an 'equal or fair opportunity' for minorities to elect candidates of their choice. This is not an accurate statement of the law."47 Judge Edwards well captures the central tension in the current application of section 5. So long as the preclearance requirement is limited to mechanical inquiries, and so long as the ability to provide an "equal or fair opportunity" for minority electoral success is not a factor in the preclearance equation, the Beer standard could work well as a matter of administrative law. As soon as effectiveness of representation and totality of circumstances are added to the mix, however, the simple regulatory regime begins to break down. The clearest example of this comes by way of comparing the record presented in the Georgia litigation to that of New Jersey in Page v. Bartels.48 Page v. Bartels involved a challenge brought by seventy Republican members of the New Jersey Legislature and several black and Latino voters from two affected counties against a redistricting plan adopted by the New Jersey Apportionment Commission in April 2001. Also named as defendants in Bartels were the five Democratic members of the Apportionment Commission, Professor Larry Bartels (the "eleventh member" of the Commission), the Secretary of State for New Jersey, and the Attorney General of New Jersey. The claim in Bartels was that the proposed redistricting of New Jersey would impermissibly dilute minority voting strength in violation of section 2 of the Voting Rights Act.49 In other words, Page v. Bartels bore the unmistakable recent imprimatur of Republicans challenging a Democratic-inspired redistricting plan in the name of its claimed impact on minority voting prospects. Like the appellees in Ashcroft, the Bartels plaintiffs argued that this infringement was retrogressive and dilutive of minority voting strength, destroying the opportunity for minorities to successfully elect candidates of choice to office. Bartels, however was litigated under section 2 of the Voting Rights Act and thus required an assessment of the totality of factors affecting minority voting prospects in New Jersey, not simply a determination of whether the plan was retrogressive in diminishing minority voting concentrations. This is effectively the standard of review that Justice O'Connor would import into section 5 preclearance analysis in her Ashcroft opinion. The district court, operating under section 2, found that the plan "as structured will encourage franchise participation of New Jersey voters, including African-American and Hispanic voters, and will do so without diluting or impairing minority participation."50 Rather than applying a 47. 48. 49. 50. Id. at 97 (Edwards, J., concurring). 144 F. Supp. 2d 346 (D.N.J. 2001). 42 U.S.C. ? 1973 (2000). Bartels, 144 F. Supp. 2d at 369. 1722 [Vol. 104:1710 COLUMBIA LAW REVIEW mechanical assessment of the percentage of minority voters in districts before and after redistricting, the New Jersey court assessed the electoral prospects of minority-preferred candidates under an intensely local examination of political conditions. The contrast between Judge Edwards's rendition of the objectives of section 5 of the Voting Rights Act and the Bartels court's assessment of VRA standards is directly tied to the operative differences between section 2 and section 5. Whereas section 2 claims are held to a threshold of proof of actual diminution of an effective opportunity to elect, section 5 locks in--at least until Ashcroft--a simple quantitative definition of minority concentrations in specified districts. While the legal regimes under which Ashcroft and Bartels operated may have been different, the factual issues of the cases were similar. Thus, a comparison of the factual records highlights the troubling effects of section 5 rigidity in a more fluid and more competitive political environment. The Bartels plan was certified by the Apportionment Commission by a vote of six to one--the single vote against the plan was cast by the lone Republican member of the Commission present at the meeting.51 Despite maintaining a majority in both houses of the New Jersey legislature, the Republican members were unable to compel the Commission to adopt their preferred redistricting plan over the Bartels plan, which was endorsed by an interracial coalition of Democrats. The challenged aspects of the Bartels plan involved the reconfigured boundaries of four districts in Northern New Jersey: Districts 27, 28, 29, and 34. Under the previous (1991) districting system, the percentages of white (WVAP), black (BVAP), Hispanic (HVAP), and total minority voting-age populations (MVAP) were as shown in Table 1. TABLE 1--NEW JERSEY DISTRICTS, 199152 New Jersey District WVAP BVAP HVAP MVAP 27 28 29 34 31.4% 20.4% 20.9% 76.8% 52.8% 57.4% 48.2% 3.9% 9.4% 16.8% 26.2% 11.3% 68.6% 79.6% 79.1% 23.2% The most salient feature of the Bartels plan was the proposed decrease in the black voting-age population in the three districts in which they were in a majority, such that they would find themselves in the minority in all four districts. The reconfigured boundaries shifted the voting-age populations as shown in Table 2. 51. Id. at 349 n.2. 52. Id. at 353. 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1723 TABLE 2--NEW JERSEY DISTRICTS, BARTELS PLAN53 New Jersey District 27 28 29 34 WVAP 58.0% 30.3% 22.5% 48.2% BVAP 27.5% 48.3% 39.2% 35.3% HVAP 6.6% 14.0% 33.2% 9.8% MVAP 42.0% 69.7% 77.5% 51.8% The Bartels plaintiffs claimed that the decrease in the black votingage population from fifty-three percent to twenty-eight percent in District 27, from fifty-seven percent to forty-eight percent in District 28, and from forty-eight percent to thirty-nine percent in District 29 "reduce[s] or eliminate[s] the opportunity of African Americans . . . to elect legislators of their own race."54 Further, plaintiffs asserted, the proposed increase from four percent to thirty-five percent in District 34 "does not afford sufficient opportunities for minorities to elect their preferred candidates."55 In short, the plaintiffs argued that the new districts were racially polarized, resulting in a "'chilling effect' on racial minorities' participation in the political process when they make up less than 50% of the voters in a given district."56 These arguments were supported by testimony suggesting that the history of racial-bloc voting in these districts made it unlikely that African Americans would win elections.57 In addition, plaintiffs offered testimony suggesting the Bartels plan sacrificed minority voters' rights in favor of gaining more Democratic votes across all districts.58 These claims were buttressed by ample voting rights case law suggesting that districts with minority concentrations between forty percent and fifty-five percent were almost the per se embodiment of minority vote dilution. In response, the defendants argued that the Bartels plan not only did not dismantle the three black majority districts, but that it would in fact provide "even more avenues for the election of minority representatives [to] open[ ] up" by unpacking black voters to other districts.59 Minority officeholders testified that the operative levels of racial-bloc voting had not prevented them from obtaining elected office, even when the racial group to which they belonged did not constitute a majority in their district. To the contrary, crossover voting and coalition-building among the 53. Id. 54. Id. 55. Id. 56. Id. at 358. 57. Id. at 359 (describing testimony of James Loewen, expert witness for plaintiffs). 58. The court noted the testimony of Walter Fields, who consulted with the Republicans during the apportionment process, in which he stated that the Bartels plan "shuffled African-American voters around to create more Democratic districts but less African-American districts." Id. at 354. 59. Id. 1724 COLUMBIA LAW REVIEW [Vol. 104:1710 black, Hispanic, and white voters enabled minority candidates to prevail over white candidates; for this reason minority voting strength had to be considered in its entirety, as opposed to on a district-by-district or race-byrace basis. As the defendants noted, eight of the fifteen black representatives then in the New Jersey legislature were elected from districts with a black voting-age population of less than thirty percent. Additional statistical analyses of general elections by the defendants confirmed that the vote, statewide, was not racially polarized. The court was ultimately convinced by testimony supporting the Bartels plan that expressed confidence it "would increase minority election opportunities over the next ten years."60 The factual scenario at issue in Georgia v. Ashcroft was not unlike that of Bartels. The Georgia Senate's redistricting plan in Ashcroft likewise "unpacked" several districts where minority voters constituted the majority of eligible voters, reducing their strength in all but one district to just over fifty percent. At the same time, however, the proposed redistricting plan increased the strength of minority voters in other "influence" districts, where they would have a reasonable chance of electing candidates of their choice. Nonetheless, the District Court for the District of Columbia did not share the New Jersey court's confidence in the potential for increased minority election opportunities under such a plan and determined that Georgia's plan violated section 5 of the Voting Rights Act. Georgia, the court declared, did not meet its burden of proof that the redistricting plan "does not have the purpose and will not have the effect of denying or abridging the right to vote on account of race or color."61 At bottom, the question in both the Georgia and New Jersey cases was whether the protection of black voting interests was best left to the intervention of legal remedies or to the prospect of political trading and hauling, primarily through coalition politics within the Democratic Party. The New Jersey court refused to read section 2 of the Voting Rights Act as rendering inviolable the concentration of black voters into districts with guaranteed black-descriptive representation, achieved at the expense of the emergence of politically viable cross-racial coalitions. A quick look at the numbers in Georgia reveals the similarity of the question posed there under section 5 to that presented under section 2 in New Jersey. Table 3 lists both the percentage of the black voting-age population (BVAP) in the relevant State Senate districts under the existing districting plan, which served as the benchmark against which the proposed redistricting plan was developed, and the percentages in the new challenged districts under the proposed plan. 60. Id. at 366. 61. Georgia v. Ashcroft, 195 F. Supp. 2d 25, 31 (D.D.C. 2002). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1725 TABLE 3--GEORGIA SENATE DISTRICTS62 Georgia Senate District BVAP Under Benchmark Plan BVAP Under Proposed Plan 2 12 15 22 26 60.58% 55.43% 62.05% 63.51% 62.45% 50.31% 50.66% 50.87% 51.51% 50.80% In comparing the redistricting plans challenged in Ashcroft and Bartels, one sees the similarity in the proposed percentage decreases in minority population from the original districts to the reconfigured districts in both plans. Both significantly reduce black concentrations in previously safe majority-minority districts. Both create districts that under preexisting voting rights law would be almost per se exemplars of dilutionary arrangements. Notably, however, the Bartels plan maintains the greatest deviation in the percentage of black voting population from an existing district to a proposed district: New District 27 under the Bartels plan experiences a 25.3% net decrease of black voters from the former 1991 districting plan. Conversely, under the redistricting plan in Ashcroft, the Senate District that anticipates the greatest decrease in black voting-age population is District 22, with a twelve-percentage-point net reduction from its predecessor district. The fact that the redistricting plan in Ashcroft did not propose decreases at the same rate as those upheld in Bartels only reinforces the impact of the more mechanical section 5 restrictions compared to section 2. Moreover, the trial presentations in Ashcroft virtually mirror those in Bartels. Expert witnesses for the United States in Ashcroft testified to the existence of racially polarized voting in the challenged Senate districts, making it "close to impossible for an African American candidate to be elected" in at least one of the proposed districts.63 The United States also offered expert testimony based on the now standard regression analyses used in voting rights cases to further describe racially polarized voting patterns and the minimal amount of "white crossover."64 Despite Georgia's criticism of the regression analysis because it only considered the level of white crossover voting, and the testimony of Georgia's expert witness, based on alternative statistical analysis, to support the contention that the proposed redistricting plan did not have a retrogressive effect,65 the court nonetheless concluded that it did. The court stated: 62. Id. at 56. 63. Id. at 61. 64. Id. at 70. 65. Id. at 66 (describing expert's conclusion that the "point of equal opportunity is 44.3% BVAP, which means that 'there's a 50-50 chance of electing a candidate of choice' in a district with an open seat and with 44.3% BVAP"). 1726 COLUMBIA LAW REVIEW [Vol. 104:1710 In sum, then, just as cross-racial coalition-building . . . can allow smaller numbers to extend great influence, so too can its antitheses, racial-bloc voting, allow a drop in numbers to become a retrenchment of power. Thus, the more we find evidence of racial polarization in the disputed Senate districts, the more we are persuaded that Georgia has failed to meet its burden of proving that the reductions in African American populations in those districts and in other majority-minority districts has not lessened the ability of its African American voters to effectively exercise their collective right to vote.66 Just as in Bartels, supporters of the Georgia redistricting plan marshaled lay testimony from legislators to demonstrate black support for the redrawn districts and for the contention that the plan did not have a retrogressive effect. Incumbent African American senators confirmed that they "believed the plan gave them and others a 'fair' or 'reasonable' chance of victory in the redrawn districts."67 But unlike the court in Bartels, the Ashcroft court rejected such arguments, declaring that the "retrogression analysis does not ask if a proposed district is 'fair' or whether a possibility exists that a minority preferred candidate would be elected."68 The different outcomes in Georgia (in the district court) and New Jersey ultimately turned on three factors. First, under section 5, Georgia bore the burden of proof and had to overcome the presumption that its redistricting objectives were antithetical to black interests. Second, the courts differed on the conception of polarized voting. For the Ashcroft district court, polarized voting emerges as a yes-no lever. Its presence under section 5 buttressed the presumption that any diminution in minority concentrations would be retrogressive and a violation of the Act. In Bartels, by contrast, the court appears more faithful to the Gingles language concerning the existence of legally significant polarized voting. This term appeared for the first time in Justice Brennan's plurality opinion in Gingles and has not yielded much case law elucidation.69 Under this approach, the question is not whether black and white voters show distinct preferences at some level of statistical significance, a question that is likely to be answered in the affirmative in just about any jurisdiction in the United States, but whether such differences in voting patterns compromise black ability to stand a reasonable chance in the political process overall.70 For the Bartels court, this approach meant providing 66. Id. at 78. 67. Id. at 89. 68. Id. at 91. 69. Thornburg v. Gingles, 478 U.S. 30, 55-58 (1985). 70. In approving the redistricting commission's plan, the Bartels court cited the lack of evidence "that the white majority in the challenged districts votes sufficiently as a bloc to enable it to defeat the minority's preferred candidates." Page v. Bartels, 144 F. Supp. 2d 346, 362 (D.N.J. 2001) (emphasis added). For elaboration of this issue see generally Richard H. Pildes, Is Voting-Rights Law Now at War With Itself? Social Science and Voting Rights in the 2000's, 80 N.C. L. Rev. 1517 (2002), which provides an assessment of the role of coalitional politics in compromising some of the more mechanical tests adopted in 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1727 minority voters in the redrawn districts with "a reasonable opportunity to elect candidates of their choice."71 For the Ashcroft court, by contrast, any type of equal chance or political fairness approach was inconsistent with the stringent demands of section 5. The court would deny preclearance even if the effect of such denial was a predictable overall diminution of black political prospects, and even if the plan in question was not imposed on black voters but emerged as the product of intense negotiations in which black representatives were active and critical participants. Third, the courts differed in their assessments of the objectives of the VRA sections that they confronted. The Ashcroft court, consistent with the overwhelming bulk of section 5 case law, viewed the Act as a bulwark against change. The presumption had to be that of black exclusion and the test therefore was whether black political enclaves had been sufficiently protected. By contrast, the Bartels court embraced a totality-of-thecircumstances approach to assess whether the plan short-circuited black political aspirations or whether it fairly offered a meaningful prospect for influence in an increasingly integrated political environment.72 As a result of these three factors, despite the factual similarities between the two cases, the outcomes at the district court level were diametrically opposed.73 III. POLITICS WITHOUT SECTION 5 Leaving aside for the moment the difference between the legal burdens imposed by sections 2 and 5 of the Voting Rights Act, it is impossible to examine the facts presented in New Jersey and Georgia and not be struck by the similarities. I believe that it is difficult to endorse the outcome in Bartels as a matter of protecting black political opportunity, yet side with the district court majority in Ashcroft in condemning the political deals realized in Georgia. Further, I want to suggest that the Supreme Court, by bringing the outcome of the section 5 analysis into conformity voting rights law. Ironically, both the majority and dissenting opinions in the Supreme Court in Ashcroft invoke this article in support of their positions. See Georgia v. Ashcroft, 539 U.S. 461, 483 (2003); see also id. at 493 (Souter, J., dissenting). 71. Bartels, 144 F. Supp. 2d at 362. 72. For example, the court stated that evidence showing that "African-Americans and Hispanics often vote in support of each other's candidates" in the challenged districts was significant in assessing the political opportunities of both groups. Id. 73. It should be noted that the issue of partisan gerrymandering also lay in the background of the district court's decision in Ashcroft. On subsequent remand from the Supreme Court, the district court again invalidated Georgia's proposed redistricting plan--this time for violating the one person, one vote principal of Reynolds v. Sims, 377 U.S. 533, 568 (1964)--and appointed a special master to draw a new, nonpartisan district map. In finding that the Democratic plan violated the Reynolds test, the district court made clear its concern that the plan had been drafted to manipulate electoral outcomes. See generally Rhonda Cook, Court Redraws Capitol Careers, Atlanta J.-Const., Mar. 16, 2004, at A1; Rhonda Cook, Legislature 2004: Lawmakers, Court's Experts Work on Remap, Atlanta J.-Const., Mar. 3, 2004, at B8. 1728 COLUMBIA LAW REVIEW [Vol. 104:1710 with the section 2 result in Bartels, may have seriously compromised any possible administrative application of the preclearance requirement. The next step, therefore, is to pursue the implications of these two observations. What does it mean that the Act, as previously applied, appears to hamper the very type of coalitional politics that traditional defenders of minority voting rights so adamantly fought to protect in New Jersey? Further, what is the effect of compelling the bureaucratic and intrusive section 5 preclearance mechanism to take account of increasingly fine-grained distinctions in the political realities of covered jurisdictions? In sum, have the altered political environment and the compromised administrability of section 5 done more than call into question its constitutional moorings? Have they also forecast its disutility as a matter of policy? These questions point us decidedly away from constitutional law and direct us instead to fundamental questions about the relationship between law and politics. Section 5 of the Voting Rights Act is a strong variant of what is often termed the Carolene Products 74 view of antidiscrimination law. Although section 5 is obviously a product of congressional rather than court generation, it shares with Carolene Products equal protection law a presumption that a discrete and insular minority group cannot turn to the normal operation of politics to protect its interests. As Justice Souter would have it, the assumption that the normal gives and takes, the obligation and opportunity to "pull, haul, and trade to find common political ground,"75 should not apply to a particular group itself requires some level of exacting scrutiny. I will conclude with three points. First, there appears to be little reason to believe that the law should continue to treat blacks in Georgia distinctly from blacks in New Jersey. The effect of the Supreme Court's decision in Ashcroft is to try to equalize the treatment within the confines of section 5. I suspect that administrative implementation of the Ashcroft standards will not be possible76 and that this administrative impossibility will put renewed pressure on the policy justifications (and the constitutional rationale) for section 5. But Ashcroft does more than just make implementation difficult. The decision creates a dilemma for advocates of the continuation of section 5. To the extent that the decision is seen as an improper construction of section 5, then the exact basis for the continued administrative subordination of politics in Georgia has to be made clear. Put bluntly, why should black voters of Georgia not be permitted the same degree of political opportunity to form coalitions as black voters of New Jersey? Conversely, to the extent that the decision is 74. United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938). 75. Johnson v. De Grandy, 512 U.S. 997, 1020 (1994). 76. For a more optimistic account of the ability of the Department of Justice to integrate Ashcroft into an enforceable administrative framework, see Meghann E. Donahue, Note, "The Reports of My Death Are Greatly Exaggerated": Administering Section 5 of the Voting Rights Act After Georgia v. Ashcroft, 104 Colum. L. Rev. 1651 (2004). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1729 accepted as properly advancing a correct interpretation of section 5, what justifies the extraordinary administrative mechanism that operates to reproduce, within a compressed and rigid time frame, the protections and scope of section 2 in Georgia but not New Jersey? Second, the New Jersey litigation points to a deeper concern about what it means to have properly functioning politics. One way of understanding the tension between the district court decisions in Bartels and Ashcroft is to focus on the nature of the assurance owed to black voters by the respective provisions of the VRA. The New Jersey court accepted the claim that blacks were owed only a fair chance to form coalitions and have a reasonable opportunity to prevail electorally. By contrast, the Georgia court asked for guarantees of minority electoral success, which in turn translated into the maintenance of packed minority districts.77 The effect of Ashcroft is to put pressure on a traditional weakness in the structure of section 5. As originally conceived in terms of a suspension of literacy tests followed by administrative obstacles to the reestablishment of such tests, section 5 was fairly confined to questions of eligibility to cast a vote. Preclearance acted only to ensure individual abilities to vote in the face of the overwhelming obstacles inherited from Jim Crow. Beginning with Allen v. Board of Elections,78 however, preclearance expanded to the effectiveness of the franchise. This was arguably of greater importance in advancing black electoral opportunity than even the dismantling of literacy tests. But Allen introduced an administrative presumption that the electoral mechanisms routinely used in other parts of the country were inherently suspect if adopted by Southern jurisdictions faced with newly enfranchised blacks. I do not intend to echo the arguments of Abigail Thernstrom against this important extension in Allen.79 My claim is only that once section 5 was used to reach electoral mechanisms based upon the likely distributional effect on black electoral success, there was always a risk that a constricted view of permissible political outcomes would infect the preclearance process. I am not the first to express concern that a narrow focus on securing the electability of minority candidates could compromise the range of political accords available to minority voters and thereby, under conditions of mature political engagement, actually thwart minority political gains. This is a claim that others, such as Carol Swain, have raised in the past.80 Much in this debate turns on a difficult empirical assessment of when minorities become full players in the political process. But it is possible to express the prob77. For the most current account of the impact of packed minority districts on increasing Republican gains in the South and diminishing the legislative influence of minority legislators, see David Lublin, The Republican South 101-12 (2004). 78. 393 U.S. 544 (1969). 79. See generally Abigail Thernstrom, More Notes from a Political Thicket, 44 Emory L.J. 911 (1995); Abigail Thernstrom, Shaw v. Reno: Notes from a Political Thicket, 1994 Pub. Int. L. Rev. 35. 80. See Carol M. Swain, Not "Wrongful" by Any Means: The Court's Decisions in the Redistricting Cases, 34 Hous. L. Rev. 315, 318 (1997). 1730 COLUMBIA LAW REVIEW [Vol. 104:1710 lem in terms of the tension between two of the operative notions in the Carolene Products footnote. On the one hand, the Court's famous formulation invokes the principle of special solicitude (granted judicial and not congressional, but I wish to leave that to the side) arising out of discrete and insular status. But, at the same time, this is coupled with the idea that the special solicitude is not simply the product of discreteness, but of the inability to seek redress through the normal operations of the political system. One way of reading Ashcroft is to see the Court suggesting that black voters in Georgia have moved from a world of discrete status meriting protections external to the political system to a situation more closely approximating the normal give and take of politics. Further, what is perhaps most remarkable about Ashcroft is the Court's willingness to consider that black electoral prospects in Georgia could not be divorced from the partisan battles for legislative hegemony. Only a few years earlier, in Easley v. Cromartie, the Court held to the fleeting hope that it could ground constitutional doctrine on hermetic seals between considerations of race and partisanship in North Carolina.81 In Easley, the last of the Shaw line of cases, the Court found that racial considerations did not predominate in drawing a minority-concentrated district on the grounds that partisan gain had also been at play. Ashcroft implicitly rejects this approach in favor of assessing racial impact as inextricably tied to overall political opportunity, a consideration in which partisan representation is an inescapable component. In so doing, Ashcroft looks to the competitive fires of realigned Southern politics to find powerful allies for black interests outside external legal intervention. In contemplating that partisan competition may erode the traditional forms of section 5 review, the Court invites further inquiry into whether the administrative model of preclearance captures the protections necessary in jurisdictions with active partisan competition.82 Or, at the very least, Ashcroft calls into question the continued utility of section 5 for those types of front-burner political decisions (as with redistricting) where political vigilance is likely to be intense. Third, so long as the South was solidly Democratic, there was little partisan gain available from controlling the levers of preclearance. The original scope of the VRA was directed primarily to voter eligibility. While in the aggregate bringing black voters into the political process was likely to alter electoral outcomes, my sense is that relatively few preclearance decisions were likely to be directly outcome determinative. This is a debatable point that may turn on how outcome determinativeness is defined. I suppose that cases involving municipal annexations, or a change in voting from districted to at-large elections, could be categorized as outcome determinative in some indirect sense. But I want to 81. 532 U.S. 234, 257-58 (2001). 82. For further elaboration on this point, see Samuel Issacharoff & Richard H. Pildes, Politics as Markets: Partisan Lockups of the Democratic Process, 50 Stan. L. Rev. 643, 700-07 (1998). 2004] IS SECTION 5 A VICTIM OF ITS OWN SUCCESS? 1731 suggest that the use of preclearance for districting configurations following the decennial census dramatically increases the ability to use preclearance to affect the projected outcomes in terms of partisan representation. I would further suggest, without empirical support as such, that whatever the partisan gain to be had from preclearing or failing to preclear the relocation of voting booths or the change in hours of voting, it is unlikely to be sufficiently obvious or significant as to prompt the intervention of political actors in the normal administrative processes of preclearance. This again raises the question whether the continued application of section 5 might be divided into the administrative side of voting access, where partisan vigilance is likely to be more lax, and matters of extreme partisan concern, such as redistricting, where partisan vigilance is not only likely to be high, but likely to tempt the purported guardians on high in Washington. CONCLUSION This Essay starts with an examination of the changed political environment that has eroded the preconditions for the success of section 5 of the Voting Rights Act. The emerging conclusion is that section 5 has served its purposes and may now be impeding the type of political developments that could have been only a distant aspiration when the VRA was passed in 1965. This Essay further suggests that the combination of an administratively complex standard emerging from Ashcroft together with the strengthened world of partisan competition has called into question the continued utility of administrative preclearance--or at least of preclearance in the world of first-order political decisions such as redistricting. What remains unanswered is what would happen in the absence of section 5. Would blacks still have the same bargaining power or might partisan protection erode in the absence of the special attention that preclearance continues to offer? Professor Karlan, in supporting the continued importance of section 5, argues that the political gains registered by black voters in the South over the past forty years "have all occurred in the shadow of section 5, which has given minority voters and their representatives an invaluable bargaining chip."83 If the burden for change is certainty of outcome, then the status quo always prevails. My suspicion is that the combination of section 2 of the Voting Rights Act, the protections of the Fourteenth Amendment, and the fact of being in the process and at the table would afford much protection. Whether this combination is enough, absent section 5, is certainly debatable. What seems less unclear, however, is the mischief that section 5 can play in stalling coalition politics and inviting politically inspired interventions from outside the covered jurisdictions. 83. Karlan, supra note 6, at 36.