View PDF Version
Nos. 98-405 and 98-406
In the Supreme Court of the United States
OCTOBER TERM, 1998
JANET RENO, ATTORNEY GENERAL, APPELLANT
BOSSIER PARISH SCHOOL BOARD
GEORGE PRICE, ET AL., APPELLANTS
BOSSIER PARISH SCHOOL BOARD
ON APPEALS FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
REPLY BRIEF FOR THE FEDERAL APPELLANT
SETH P. WAXMAN
Counsel of Record
Department of Justice
Washington, D.C. 20530-0001
In the Supreme Court of the United States
OCTOBER TERM, 1998
JANET RENO, ATTORNEY GENERAL, APPELLANT
BOSSIER PARISH SCHOOL BOARD
GEORGE PRICE, ET AL., APPELLANTS
BOSSIER PARISH SCHOOL BOARD
ON APPEALS FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
REPLY BRIEF FOR THE FEDERAL APPELLANT
Appellee argues (Bd. Br. 8-11)1 that this case is moot because the next regularly scheduled School Board election will not be held until 2002, by which time it should have adopted a new redistricting plan. As we have explained in our brief in opposition to appellee's motion to dismiss or affirm (at 1-3), appellants retain a live interest in the outcome of this litigation. In Section 5 of the Voting Rights Act of 1965, 42 U.S.C. 1973c, Congress granted voters (and the Attorney General) a statutory right against the implementation of voting changes that have not been properly precleared.2 Thus, if this Court reverses the district court's preclearance judgment, voters in Bossier Parish might well be entitled to a special election held under a lawful plan that complies fully with Section 5. That lawful plan might be developed by the Board and precleared by the Attorney General or the district court, or (should the Board fail to hold a special election on its own initiative) elections might be ordered by a federal court under a plan fashioned by that court as a remedy for the Board's violation of Section 5.
Appellee suggests (Bd. Br. 11) that, if the district court's preclearance judgment is reversed and new elections are ordered, minority voters in the Parish would receive no benefit because its previous 1980s plan was little different from the 1992 plan, in terms of its effect on minority voting rights. That argument proceeds from the incorrect assumption that there would be no objection to holding elections under the 1980s plan. In fact, the 1980s plan is severely malapportioned. See J.S. App. 171a-172a. It is appropriate to assume that, if the district court's judgment is reversed and the Board then chooses or is ordered to hold a special election, the Board would not hold the election under an unconstitutional plan; and, if the Board attempted to do so, use of the 1980s plan would likely be promptly challenged in district court. Cf. J.A. 41-42 (prior equal-apportionment challenge to implementation of 1980s plan).
Even if this case were moot, the appropriate action would be for the Court to vacate the lower court's judgment and to remand the case with instructions to dismiss the complaint. See Gov't Br. in Opp. to Mot. to Dism. or Aff. 2-3 n.1. Appellee erroneously relies (Bd. Br. 11-12 n.10) on U.S. Bancorp Mortgage Co. v. Bonner Mall Partnership, 513 U.S. 18, 26 (1994) to argue that the Court should dismiss the appeals rather than vacate and remand. In that case, however, the controversy became moot because the parties had voluntarily settled the case. In such a situation, "the losing party has voluntarily forfeited his legal remedy by the ordinary processes of appeal or certiorari, thereby surrendering his claim to the equitable remedy of vacatur." Id. at 25. The Court reaffirmed, however, that "[a] party who seeks review of the merits of an adverse ruling, but is frustrated by the vagaries of circumstance, ought not in fairness be forced to acquiesce in the judgment." Ibid. That is the case here, where the government and the Price appellants have appealed twice from the district court's decision to preclear the 1992 plan.
2. The Proper Scope of Section 5.
a. Appellee's attempt to argue that the purpose prong of Section 5 is limited to retrogressive intent leads it to the remarkable assertion that there is little if any connection between Section 5 and the Fifteenth Amendment. Bd. Br. 25-28; see Bd. Br. 27 ("there is very little congruence between the Constitution and § 5"); Bd. Br. 28 n.21 ("nothing in the legislative history indicates that § 5 reaches 'racially motivated voting changes' that violate the Constitution"). The Voting Rights Act of 1965 itself, however, stated that its principal animating purpose was "[t]o enforce the fifteenth amendment to the Constitution of the United States." Pub. L. No. 89-110, 79 Stat. 437.3
The weakness of appellee's effort to divorce Section 5 from the Fifteenth Amendment is amply demonstrated by the outlandish consequences of its argument. Under appellee's theory, if in 1965 a town had (by law or practice) effectively barred all blacks from voting in town elections, and had then, after the enactment of the Voting Rights Act, enacted legislation with the purpose of accomplishing the same result--for example, by altering the jurisdiction's boundaries to exclude all blacks from residency (cf. Gomillion v. Lightfoot, 364 U.S. 339 (1960))-the Attorney General or the federal courts would have been required to preclear that voting change, because it would not have had the purpose or effect of making the position of blacks in the town worse. Or, if an all-white town enacted legislation prohibiting blacks from voting, that legislation would also have to be precleared for the same reason, even though it would flatly violate the Fifteenth Amendment (cf. City of Pleasant Grove v. United States, 479 U.S. 462 (1987)).4 The Court should reject a construction of Section 5 that leads to results so demonstrably at odds with Congress's intent.5
We have also pointed out (Gov't Opening Br. 20-23) that appellee's theory is inconsistent with the specific purpose behind the preclearance requirement of Section 5, to eliminate the pattern by which jurisdictions simply replaced one unconstitutional voting practice, struck down by the courts, with another one designed to accomplish the same result, requiring further litigation by the Attorney General and private parties to enjoin the replacement plan. Appellee argues (Bd. Br. 40 n.30) that a district court would not permit a covered jurisdiction to substitute one discriminatory plan for another. A jurisdiction might well try to accomplish just such substitution, however. Under standard rules governing constitutional litigation, if a district court strikes down one voting plan as unconstitutional, it is up to the covered jurisdiction in the first instance to develop a new plan. See Chapman v. Meier, 420 U.S. 1, 27 (1975). The government and voters would then be required to go back to the district court, and would bear the burden to prove that the substitute plan was unconstitutional in order to prevent its implementation-exactly the unsatisfactory situation before Section 5 was enacted. Even if the jurisdiction's effort was eventually stymied by the district court, it would have successfully delayed minority voters' enjoyment of the full exercise of their right to vote. That prospect is impossible to square with Congress's specific objective in Section 5, "to shift the advantage of time and inertia from the perpetrators of the evil to its victims." South Carolina v. Katzenbach, 383 U.S. 301, 328 (1966).
b. The plain language of Section 5 prohibits enforcement of a voting change enacted with the "purpose" of "denying or abridging the right to vote on account of race" (42 U.S.C. 1973c)-language that straightforwardly reaches a voting change enacted with the purpose to discriminate against black voters. Appellee presents several unpersuasive arguments to avoid the thrust of this language. It argues (Bd. Br. 26), for example, that "on account of race," as used in Section 5, cannot refer to unconstitutional, purposeful racial discrimination because Congress used the same language in amended Section 2 of the Act, 42 U.S.C. 1973, which does not contain a purpose requirement. But Section 2 does reach purposeful racial discrimination in official voting practices, see City of Mobile v. Bolden, 446 U.S. 55, 60-61 (1980) (plurality opinion) (emphasizing close connection between Section 2 and Fifteenth Amendment), even though, like Section 5, it reaches more broadly as well. The fact that neither statute reaches only purposeful discrimination hardly suggests that purposeful discrimination is outside either one.6
Appellee also argues (Bd. Br. 19-20) that, because Beer v. United States, 425 U.S. 130, 141 (1976), held that the effect prong of Section 5 reaches only voting changes having a retrogressive effect, the purpose prong must also be limited to changes enacted with the intent to retrogress. We have explained, however, that Beer's construction of the effect prong of Section 5 reflects concerns about that statute's reach beyond the moorings of the Constitution. See Gov't Opening Br. 29-32. Beer does not suggest that the separate purpose prong of Section 5 fails to reach all voting changes enacted with an unconstitutional, racially discriminatory purpose. To the contrary, the Court observed in Beer that a voting change "could be a substantial improvement over its predecessor in terms of lessening racial discrimination, and yet nonetheless continue so to discriminate on the basis of race or color as to be unconstitutional." 425 U.S. at 142 n.14.
In an effort to avoid the force of that language in Beer, appellee suggests (Bd. Br. 35) that Beer merely observed that "a reapportionment which satisfies § 5 may nonetheless violate the Constitution." But the Court in Beer, immediately after the language quoted above, proceeded to observe that the government had "made no claim" that the districts at issue in that case were unconstitutional. See 425 U.S. at 142 n.14. The Court discussed the Constitution precisely because of its relevance to the standards for preclearance under Section 5. The Court also stated in Beer that an ameliorative plan "cannot violate § 5 unless [it] so discriminates on the basis of race or color as to violate the Constitution." Id. at 141 (emphasis added). Thus, the Court's discussion about unconstitutional voting changes related directly to its construction of Section 5. Congress, moreover, codified that precise discussion in Beer when it subsequently reenacted Section 5 in 1982. See Gov't Opening Br. 29.
Furthermore, if appellee's exceedingly narrow construction of Section 5's purpose prong were correct, it is difficult to see why Congress would have adopted it.7 As a practical matter, the purpose prong as so construed would add little if anything to the retrogression analysis required under the effect prong. In almost every case, the Section 5 inquiry would be effectively exhausted by an analysis of the effects of a voting change to determine whether the change was retrogressive. If the change was retrogressive, then preclearance would be denied without any consideration of the change's purpose; but if the change did not have a retrogressive effect, then (in appellee's view) preclearance could be denied only if the covered jurisdiction had enacted the voting change in an unsuccessful effort to achieve retrogression. Nothing in the text, legislative history, or decisions of this Court construing Section 5 suggests that the purpose prong has such a trivial reach, limited to the case of the incompetent retrogressor.8 Cf. United States v. Albertini, 472 U.S. 675, 682-683 (1985) (rejecting construction of statute that would render clause "almost superfluous").
c. Appellee's effort to wave away this Court's precedents fares no better. On this point we refer the Court to our opening brief (at 24-29), but we note that appellee's effort (Bd. Br. 28-29) to recharacterize City of Pleasant Grove, supra, is particularly strained. In that case, the Court denied preclearance to annexations by an all-white town of vacant land and land populated only by whites for the purpose of "provid[ing] for the growth of a monolithic white voting block, thereby effectively diluting the black vote in advance." 479 U.S. at 472. The Court could not have decided the case on the basis, suggested by appellee (Bd. Br. 29), that the annexation of land then populated only by whites could have made "minority voters worse off than they were prior to the annexation," for there were no minority voters in the City of Pleasant Grove to be made "worse off."9
Appellee appears to acknowledge (Bd. Br. 30 & n.23) that Busbee v. Smith, 549 F. Supp. 494 (D.D.C. 1982), aff'd mem., 459 U.S. 1166 (1983), rejected the position it is now advancing, but it suggests that the Court should disregard that decision because the Court mistakenly overlooked the possibility that the voting change considered there caused minor retrogression in one of the two districts at issue. We have explained (Gov't Opening Br. 27-28), however, that the appeal in Busbee was presented to this Court on precisely the opposite assumption, viz., that the plan at issue had no retrogressive purpose or effect.10 Nor did the government argue in that case that the lower court's decision should be affirmed on the alternate basis that the plan in fact had a retrogressive effect. Appellee's reading of Busbee should be rejected because "[q]uestions which merely lurk in the record are not resolved" by summary affirmances, and "no resolution of them may be inferred." Illinois State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173, 183 (1979) (citations and internal quotation marks omitted).
d. Finally, to avoid the conclusion that Section 5 reaches unconstitutional, intentional vote dilution, appellee argues (Bd. Br. 19) that vote dilution is an inherently relative concept, and so suggests that the Department of Justice, in concluding that the 1992 plan was dilutive, must have been comparing that plan to the NAACP plan.11 In determining whether a plan has an unconstitutional, racially discriminatory purpose, however, the Justice Department does not simply compare it to other, possible plans; indeed, in this case, the Department informed the Board that it "is not required by Section 5 to adopt any particular plan." J.S. App. 235a. Rather, the Department undertakes a fact-intensive, case-specific analysis based on Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 265-266 (1977), of the circumstances under which district boundary choices have been made, to determine whether or not those choices reflect an intent to minimize or cancel out minority voting strength within particular communities. That analysis takes into account whether legitimate, nondiscriminatory governmental purposes support the jurisdiction's asserted reasons for selecting those boundary lines. And while Section 5 does place the burden on the covered jurisdiction to show that its plan lacks a discriminatory purpose, Congress plainly did not intend that burden to be impossible for a covered jurisdiction to meet.12 If the covered jurisdiction puts forward evidence showing that its voting change is not retrogressive and raising no concerns under the Arlington Heights framework for analyzing discriminatory purpose, the voting change is likely to be precleared, either by the Attorney General or the court, at least absent other evidence.13
3. The Board's Discriminatory Intent In Adopting The 1992 Plan.
Appellee argues (Bd. Br. 12-13) that the district court did not rule that the purpose prong of Section 5 is limited to retrogressive intent, but also considered whether the Board had a discriminatory (but not retrogressive) intent in adopting the 1992 plan, and found (Bd. Br. 43-50) that no such discriminatory intent was present. We have explained that the district court's opinion, although unclear, is better understood as limiting the scope of its inquiry to retrogressive intent. Gov't Opening Br. 41-42. Even if the district court did consider the question of discriminatory (but not retrogressive) intent, any findings that it may have made on that question cannot be sustained, because they were not made pursuant to the appropriate legal analysis, and are clearly erroneous in any event.
a. As we have explained (Gov't Opening Br. 42-43), the district court failed to apply the analytical framework established in Arlington Heights, 429 U.S. at 265-266, to determine whether the Board acted with a discriminatory (but not retrogressive) purpose. The court's discussion of the evidence under Arlington Heights related only to retrogressive intent, and it made only summary reference to the question of an otherwise discriminatory intent. Moreover, any finding that the Board acted without a discriminatory intent is impossible to square with other findings of the district court, such as its acknowledgment that the Board was motivated by a "tenacious determination to maintain the status quo," that the Board "departed from its normal practices," and that the Board "did not welcome improvement in the position of racial minorities." J.S. App. 7a. At a minimum, therefore, a remand would be required for the district court to evaluate the evidence under the correct legal standard.14
b. In any event, appellee's effort to defend the district court's "finding" falls well wide of the mark. Appellee makes essentially three arguments. First, it contends that the Board was required to adopt the Police Jury plan, and to reject any other plan, because of its supposed obligation to adopt a plan before December 31, 1992, without splitting any precincts. Bd. Br. 44-46. Second, it argues that the Board properly rejected the NAACP plan because that plan would have required the creation of an inordinate number of new precincts in order to develop majority-black districts. Bd. Br. 47-48. Third, it maintains that the 1992 plan did not dilute black voting strength. Bd. Br. 3-4 n.3, 6 n.5, 44. All three arguments fail.
First, the supposed need to develop a plan that would avoid any precinct splits could not have motivated the Board to adopt the Police Jury plan. The Board initially had little interest in adopting the Police Jury plan because that plan failed to respect its traditional priorities in redistricting- incumbency protection and location of schools in districts. See Gov't Opening Br. 36.15 But the Board's cartographer Gary Joiner predicted at trial that, as a practical matter, any plan other than the Police Jury plan that would be, as he put it, "as strong as this one" (meaning the Police Jury plan) would require splitting precincts. See J.A. 271.16 And indeed, when Joiner met with the Board in September 1991, after the Police Jury had adopted its plan, he distributed precinct maps because, he explained, the Board would have to "work with the Police Jury to alter precinct lines." J.S. App. 174a. Nothing in the record suggests that Joiner and the Board believed that they could not ask the Police Jury to alter precincts after December 31, 1992, or that at the time they believed themselves under an obligation to redistrict before that date (since the next Board election was not until 1994). See id. at 172a, 173a.
Second, in criticizing the NAACP plan for requiring the creation of too many new precincts, appellee mistakenly assumes that the relevant question is why it rejected the NAACP plan; but the pertinent question is whether it acted with discriminatory intent when it adopted the Police Jury plan, instead of (for example) exploring some other option that would not have minimized blacks' electoral opportunity. In fact, the Board could have drawn a plan containing two majority-black districts with as few as 46 total precincts- only 3 precincts more than the number in 1990, and 10 precincts fewer than in the Police Jury plan. J.A. 236-237. Furthermore, appellee significantly exaggerates both the number and the cost of additional precincts that would have been required by the NAACP plan. Appellee asserts (Bd. Br. 4) that the NAACP plan would have split existing precincts 65 times, but it is important to understand that this does not mean that 65 new precincts would have been created, for many areas cut out of existing precincts could have been consolidated with each other or with other precincts-an option that Louisiana law permits. J.A. 380 (La. Rev. Stat. § 18:425.1 (West Supp. 1999)).17 Such consolidations could have addressed any significant concern about increased costs. The record gives no indication, moreover, that the Board explored the costs that would be occasioned by such precinct splits, or ways to alleviate them.
Finally, in an effort to wriggle out of its concession and stipulations to the effect that the 1992 plan did dilute blacks' voting strength in the Parish (see Gov't Opening Br. 38-39), appellee argues that the record does not establish either that it was obvious that a reasonably compact majority-black district could be drawn in the Parish, or that the Parish experienced racially polarized voting. Both suggestions are wrong, even aside from the stipulations. Contrary to appellee's assertion (Bd. Br. 3-4 & n.3), Board members were aware that blacks were concentrated in certain areas, and most members also knew where those areas were. J.A. 94-100, 104-105, 109-110, 113-114, 116-124.
In addition, while Dr. Engstrom's report (J.A. 163-174) acknowledged the data limitations for doing ecological regression and extreme-case analyses for most of the elections he analyzed (almost all of the elections involved too few precincts for a reliable ecological regression analysis and no precinct that was homogeneously black), that does not suggest that his report could not validly conclude that racially polarized voting exists in the Parish. A regression analysis of the only interracial parish-wide race for local office in recent years (the 1988 primary election for the 26th Judicial District Court) revealed a high degree of racial polarization: 79.2% of black voters supported the unsuccessful black candidate, while only 28.9% of white voters did so. J.S. App. 202a-203a; J.A. 166-167. Dr. Engstrom explained (J.A. 165-167) that it is appropriate to consider the results of parish-wide elections where, as here, many districts contain too few precincts to obtain reliable estimates using ecological regression analysis of elections held in individual districts, and appellee introduced no expert testimony to the contrary, even on remand. Moreover, Dr. Engstrom was able to conclude, by examining results in homogeneously white precincts, that, in several School Board and other elections, white voters did not support black candidates. J.A. 168-170, 172-174. In fact, "[o]f the 14 elections since 1980 in which black candidates [ran] against white candidates for a single-member district or for mayor, only two candidates * * * won," and those successes were affected by a unique circumstance, the presence of Barksdale Air Force Base. See J.S. App. 206a-207a; Gov't Opening Br. 4-5 n.2.
4. Appellee's Effort To Introduce Extra-Record Information.
Appellee continues to attempt to rely on extra-record information showing that, since the enactment of the 1992 plan, blacks have been elected to the School Board (Bd. Br. 5-6). Appellee was expressly offered the opportunity to reopen the record on remand to introduce evidence about the 1996 elections, but expressly declined to do so. J.S. App. 1a. It should not now be allowed to avoid the consequences of that decision.
As the district court concluded (J.S. App. 1a-2a & n.1), without being subjected to adversary testing and placed in context, those election results have no probative value.18 They have not been subjected to the expert analysis of racial polarization and voter turnout that was conducted regarding previous elections. See id. at 201a-210a. Without such close analysis, it is impossible to draw reliable conclusions about the 1996 and 1998 election results. As this Court has previously cautioned, the fact that some blacks have been elected does not mean that either racially polarized voting or vote dilution has suddenly disappeared. See Thornburg v. Gingles, 478 U.S. 30, 75-76 (1986). Also, "the fact that racially polarized voting is not present in one or a few individual elections does not necessarily negate the conclusion that the district experiences legally significant bloc voting." Id. at 57. Success of a minority candidate may be attributable to "special circumstances, such as the absence of an opponent [or] incumbency," ibid.; it may also be attributable to an effort to influence the outcome of ongoing voting-rights litigation, see id. at 76 n.37.
Indeed, there is reason to believe that a full analysis would lead to the conclusion that such "special circumstances" were present in the 1998 elections of all three black Board Members. Our limited review of the 1998 election results shows that one of the successful candidates, Kenneth Wiggins, was first appointed by the Board in 1997 to fill a vacant seat (which might have been an effort to influence this litigation), and then won election as an incumbent in 1998. Julian Darby and Vassie Richardson ran unopposed as incumbents in 1998. Darby was previously elected in 1996 from a district that, we have explained, has historically been somewhat less influenced by racial polarization because of the presence of Barksdale Air Force Base,19 and his only opponent in the 1996 election was also black, a situation that is of limited utility in analysis of racially polarized voting patterns. In 1996, Richardson won election, in a district with the highest percentage of black voting-age population in the Parish, by only 35 votes, out of 1683 votes cast. Also, in three other Board elections held in 1996 and 1998, black candidates were defeated by white candidates.20 This Court should therefore decline to draw any conclusions about racially polarized voting or vote dilution from the 1996 and 1998 elections.
* * * * *
For the foregoing reasons, and for those set forth in our opening brief, the judgment of the district court should be reversed.
SETH P. WAXMAN
1 As used in this reply brief, "Bd. Br." refers to appellee's brief on the merits in this Court.
2 In a related context, this Court has consistently held that "[t]he . . . injury required by Art. III may exist solely by virtue of 'statutes creating legal rights, the invasion of which creates standing.'" Lujan v. Defenders of Wildlife, 504 U.S. 555, 578 (1992); see Trafficante v. Metropolitan Life Ins. Co., 409 U.S. 205, 208-212 (1972).
3 See also South Carolina v. Katzenbach, 383 U.S. 301, 324-327 (1966) (explaining that Congress enacted Section 5 to enforce the Fifteenth Amendment); id. at 337 (upholding Section 5 as "a valid means for carrying out the commands of the Fifteenth Amendment"); S. Rep. No. 162, 89th Cong., 1st Sess., Pt. 3, at 17 (1965) (joint statement of 12 members of Judiciary Committee) ("The proposed legislation implements the explicit command of the 15th amendment."); H.R. Rep. No. 439, 89th Cong., 1st Sess. 6 (1965) ("The bill, as amended, is designed primarily to enforce the 15th amendment to the Constitution of the United States and is also designed to enforce the 14th amendment.").
4 In suggesting that such results would be unexceptionable, appellee argues (Bd. Br. 20) that Beer v. United States, 425 U.S. 130 (1976), held that "[d]eliberate maintenance of an at-large system for purely discriminatory reasons does not offend § 5." See also Bd. Br. 23 n.17 (noting that Beer held that New Orleans' maintenance of two at-large seats did not implicate Section 5). The discussion in Beer to which appellee refers concerned the distinct principle, unrelated to this case, that voting practices that are not changed are not subject to Section 5 at all. See 425 U.S. at 138-139. That argument is unavailing in this case; the 1992 School Board plan at issue here is, of course, a voting change. For the same reason, appellee errs in arguing (Bd. Br. 38) that the Attorney General and this Court were "forced to preclear the two at-large seats retained in the Beer reapportionment plan." The Attorney General and the Court were not forced to preclear those seats; rather, they were not subject to preclearance at all, because their retention did not constitute a voting change. See 425 U.S. at 139. It is true, of course, that because Section 5 is limited to voting changes, it cannot be used to root out all unconstitutional voting discrimination. Congress's decision to tailor the powerful remedy of preclearance to voting changes, however, reflects its particular purpose in Section 5 to prevent covered jurisdictions from avoiding the dictates of the Constitution by replacing one unconstitutional voting plan with another. See Gov't Opening Br. 20-23.
5 Appellee erroneously contends that Beer and City of Lockhart v. United States, 460 U.S. 125 (1983), concluded that Section 5 "did not in any way prohibit" the perpetuation of existing discrimination (see Bd. Br. 22-23 & n.17). Both cases evaluated the plans at issue only under the effect prong of Section 5, and not the purpose prong, and found no retrogressive effect. While it is true that a voting change may be denied preclearance under the effect prong of Section 5 only if it makes the position of minorities worse than before, no decision of this Court suggests that a voting change should be precleared if it has the purpose of reinforcing existing racial discrimination in official voting practices. See South Carolina, 383 U.S. at 315-316 (noting that purpose of Section 5 was to prevent covered jurisdictions from "perpetuat[ing]" voting discrimination).
6 Moreover, when the 1982 Senate Report to which appellee cites explained that the phrase "on account of race" does not refer to purposeful discrimination, it stressed that Section 5 expressly covers voting changes enacted with the purpose to discriminate on account of race, by its use of the separate term "purpose." The point made by the Senate Report was that enactments having the "effect" (or "result") of denying or abridging the right to vote "on account of race" should be covered in Section 2, and that, in the context of effects as well, the phrase "on account of race" does not refer to purposeful discrimination. See S. Rep. No. 417, 97th Cong., 2d Sess. 27-28 n.109 (1982).
7 In a particularly unpersuasive example designed to show that its construction of Section 5's purpose prong is not more narrow than the Constitution, appellee argues (Bd. Br. 27) that, if a jurisdiction decided to eliminate a majority-black district "for purely race-neutral reasons," such an action would have a retrogressive purpose, but not a discriminatory purpose in violation of the Constitution. It is difficult to see, however, that the jurisdiction's purpose in enacting such a provision, if "race-neutral," would be retrogressive (even though the effect might be). As this Court explained in Personnel Administrator v. Feeney, 442 U.S. 256, 279 & n.24 (1979), under the analytical framework of Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 265-266 (1977), a prohibited "purpose" implies that the decisionmaker took action "at least in part 'because of,' not merely 'in spite of,' its adverse effects upon an identifiable group." If, therefore, appellee's hypothetical jurisdiction indeed acted for "race-neutral reasons," it would not have acted "because of" the plan's retrogressive effect.
8 To the contrary, the design of the statute suggests that the purpose prong was intended to reach unconstitutional, racially discriminatory enactments, and that the effect prong was designed to reach in addition those enactments that, because of their retrogressive effect on minorities' position, would impede their ability to overcome the remaining effects of past discrimination. See City of Rome v. United States, 446 U.S. 156, 175-178 (1980). In addition, some voting changes are more readily susceptible to analysis under the purpose prong than under the effect prong. For example, when a jurisdiction creates a new elected office or position, or chooses an election method for a new governing body, it may be difficult to determine whether the change is retrogressive, and a purpose analysis may be more fruitful to determine whether the change implicates Congress's concerns in Section 5. See 28 C.F.R. 51.54(b)(4).
9 Appellee suggests (Bd. Br. 30) that the lower court echoed City of Pleasant Grove when, in evaluating retrogressive purpose, it considered and rejected the possibility that the Board might have been motivated to break up black "voting blocks before they could be established" (emphasis omitted). This argument, we note, is impossible to square with appellee's other argument-based on the same language in the district court's opinion-that the district court was considering discriminatory, and not just retrogressive, intent when evaluating the evidence (see Bd. Br. 14-15). Moreover, the district court was expressing the view that a purpose to "divide and conquer the black vote" (J.S. App. 6a) might actually be a retrogressive purpose if there were evidence that other aspects of the Board's earlier voting plan permitted black political gains; but of course that was not so, because the previous plan was also dilutive of black votes.
10 See 549 F. Supp. at 516 (district court's finding that "there is no retrogression" and thus, "technically, the voting plan does not have a discriminatory effect, as that term has been construed under the Voting Rights Act") (citing Beer); see also Busbee, J.S. at i (question presented assumes no retrogressive purpose or effect); id. at 7 (arguing that the plan "significantly enhanced black voting strength" in one district while "maintaining an influential level of black voters" in the other); id. at 22 (citing district court finding of no retrogression).
11 In its discussion of this point, appellee introduces the novel concept of "§ 5 dilution," by which it apparently means retrogression in the context of redistricting. See Bd. Br. 19-20.
12 There is no basis for appellee's suggestion (Bd. Br. 39) that, if Section 5 does require the Attorney General and the preclearance court to consider unconstitutional purpose, then the burden should be on the government to show a constitutional violation. The statute places the burden of demonstrating the absence of discriminatory purpose on the jurisdiction. See City of Pleasant Grove, 479 U.S. at 479; City of Rome, 446 U.S. at 183. Appellee's reliance on Miller v. Johnson, 515 U.S. 900, 916-917 (1995), is misplaced, for that case arose in the context of litigation by private parties challenging a districting plan as unconstitutional under the Fourteenth Amendment. In such constitutional litigation-as generally in civil litigation-the burden is on the plaintiff to establish all the elements of the cause of action to prevail. The fact that the Department of Justice--during the time in which it maintained that a "clear violation" of Section 2 also required denial of preclearance under Section 5 (see J.S. App. 32a)-assumed the burden of proving a Section 2 violation (see Bd. Br. 39) does not suggest that the statute places on the Department the burden of proving a constitutional violation, and the Department has not previously taken the position that it has that burden. While the Department previously believed that Beer placed on it the burden of proving a Section 2 violation in the Section 5 preclearance context, this Court's decisions, especially City of Pleasant Grove and City of Rome, make clear that Section 5 places the burden of proving the absence of discriminatory intent on the covered jurisdiction.
13 Appellee also suggests (Bd. Br. 36-37) that Congress could not have authorized the Attorney General or the preclearance court to decide constitutional questions in the Section 5 context and then also authorize a subsequent constitutional challenge to a precleared voting change. The Attorney General might, however, not interpose an objection for a variety of reasons, including the possibility that the covered jurisdiction had not submitted all the relevant evidence to her (as was the case with the Police Jury's submission of its plan, see Gov't Opening Br. 6); the government or private parties might later discover evidence showing that the plan had been enacted with a discriminatory purpose, and pursue Section 2 or constitutional litigation on that basis. Further, although private parties are often allowed to intervene in Section 5 litigation, there is no necessity that they be present, and preclearance cases in the courts are often litigated only against the government. By allowing a subsequent constitutional challenge to be brought, even by private parties, and even after a voting change has been precleared, Section 5 balances the interest of the covered jurisdiction in implementing its voting change promptly with the interest of voters in being free of unconstitutional voting changes.
14 Appellee again suggests (Bd. Br. 14) that, when the district court discussed the 1992 plan's "dilutive impact," it must have been addressing discriminatory intent generally, and not just retrogressive intent, because it understood that this Court had used the term "dilutive impact" to refer to a discriminatory plan, rather than a retrogressive plan. That suggestion is plainly wrong, for the reason explained in our brief in opposition to appellee's motion to dismiss or affirm (at 5 n.3).
15 In an effort to avoid the effect of its stipulations that the Board was traditionally concerned with incumbency protection (J.S. App. 171a, 172a), appellee suggests (Bd. Br. 48) that one member of each pair of incumbents placed in the same district under the Police Jury plan had "already" decided not to run for reelection. The parts of the record on which appellee relies, however, establish only that one member of each pair had decided not to run for reelection by the time discovery was taken in 1994-not when the plan was adopted in 1992. It is hardly surprising that one of each pair thrown together in a new district eventually decided not to challenge the other incumbent; but that only shows that the Police Jury plan in fact disserved incumbency protection, which the record as a whole demonstrates was one of the Board's traditional priorities.
16 Indeed, given that there was great variation among the size of the precincts under the Police Jury plan, and that some of them were quite large (one had 5440 people) while others were quite small (one had 72 people), it would have been very difficult, if not impossible, to draw any plan other than the Police Jury plan that would meet equal- apportionment requirements without breaking at least some of the precincts that formed the building blocks of that plan. See J.A. 497-499.
17 For the figure of 65 precincts, appellee relies on its Exhibit 11 (J.A. 455-496), a table that was not the subject of testimony or other explanation below. On its face, the exhibit does not suggest that 65 new precincts would need to be created under the NAACP plan. The exhibit identifies 65 precinct "cuts," but 13 of those "cuts" contain no population, and many others contain very small population totals. The "cuts" therefore could readily have been remedied by consolidation with other precincts.
18 The fundamental question is what the Board in 1992 expected and desired to be the consequences, for minority voting rights, of its redistricting plan. If, as the record otherwise establishes, the Board adopted that plan with a discriminatory purpose, the fact that its purpose may not have been entirely successful does not entitle it to preclearance of the plan.
19 See Gov't Opening Br. 4-5 n.2. The Board's District 10 has the same lines as the district represented by Julian Darby's brother, Jerome, on the Police Jury. See J.S. App. 196a-198a.
20 In 1996, black candidates were defeated by white candidates in run-off elections for Districts 1 and 7 (the latter has the second-highest percentage of black voting age population in the Parish). In 1998, a black candidate was defeated by a white candidate in District 3. Also, Jerome Blunt, appointed by the Board in 1992 to fill a vacant seat while the Board was considering redistricting plans and sworn in on the day that the Board voted its intent to adopt the Police Jury plan, was shortly thereafter defeated by a white challenger in a special election. J.S. App. 179a; see id. at 133a-134a n.9 (Kessler, J., dissenting) (observing that Board "appointed [Blunt] to fill a seat that they knew he would be unable to hold, hoping to quell the political furor over adoption of the Police Jury plan").