Alabama Legislative Black Caucus v. Alabama, 575 U.S. ___ (2015)

NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press.

SUPREME COURT OF THE UNITED STATES

_________________

Nos. 13–895 and 13–1138

_________________

ALABAMA LEGISLATIVE BLACK CAUCUS, et al., APPELLANTS

13–895 v.

ALABAMA et al.

ALABAMA DEMOCRATIC CONFERENCE, et al., APPELLANTS

13–1138 v.

ALABAMA et al.

on appeals from the united states district court for the middle district of alabama

[March 25, 2015]

Justice Breyer delivered the opinion of the Court.

The Alabama Legislative Black Caucus and the Alabama Democratic Conference appeal a three-judge Federal District Court decision rejecting their challenges to the lawfulness of Alabama’s 2012 redistricting of its State House of Representatives and State Senate. The appeals focus upon the appellants’ claims that new district boundaries create “racial gerrymanders” in violation of the Fourteenth Amendment’s Equal Protection Clause. See, e.g., Shaw v. Hunt,517 U. S. 899–908 (1996) (Shaw II) ( Fourteenth Amendment forbids use of race as “ ‘predominant’ ” district boundary-drawing “ ‘factor’ ” unless boundaries are “narrowly tailored” to achieve a “ ‘compelling state interest’ ” (citations omitted)). We find that the District Court applied incorrect legal standards in evaluating the claims. We consequently vacate its decision and remand the cases for further proceedings.

I

The Alabama Constitution requires the legislature to reapportion its State House and Senate electoral districts following each decennial census. Ala. Const., Art. IX, §§199–200. In 2012 Alabama redrew the boundaries of the State’s 105 House districts and 35 Senate districts. 2012 Ala. Acts no. 602 (House plan); id., at no. 603 (Senate plan) (Acts). In doing so, Alabama sought to achieve numerous traditional districting objectives, such as compactness, not splitting counties or precincts, minimizing change, and protecting incumbents. But it placed yet greater importance on achieving two other goals. See Alabama Legislature Reapportionment Committee Guidelines in No. 12–cv–691, Doc. 30–4, pp. 3–5 (Committee Guidelines).

First, it sought to minimize the extent to which a district might deviate from the theoretical ideal of precisely equal population. In particular, it set as a goal creating a set of districts in which no district would deviate from the theoretical, precisely equal ideal by more than 1%—i.e., a more rigorous deviation standard than our precedents have found necessary under the Constitution. See Brown v. Thomson,462 U. S. 835,842 (1983) (5% deviation from ideal generally permissible). No one here doubts the desirability of a State’s efforts generally to come close to a one-person, one-vote ideal.

Second, it sought to ensure compliance with federal law, and, in particular, the Voting Rights Act of 1965.79Stat.439, as amended,52 U. S. C. §10301 et seq. At the time of the redistricting Alabama was a covered jurisdiction under that Act. Accordingly §5 of the Act required Alabama to demonstrate that an electoral change, such as redistricting, would not bring about retrogression in respect to racial minorities’ “ability . . . to elect their preferred candidates of choice.”52 U. S. C. §10304(b). Specifically, Alabama believed that, to avoid retrogression under §5, it was required to maintain roughly the same black population percentage in existing majority-minority districts. See Appendix B, infra.

Compliance with these two goals posed particular difficulties with respect to many of the State’s 35 majority-minority districts (8 in the Senate, 27 in the House). That is because many of these districts were (compared with the average district) underpopulated. In order for Senate District 26, for example, to meet the State’s no-more-than-1% population-deviation objective, the State would have to add about 16,000 individuals to the district. And, prior to redistricting, 72.75% of District 26’s population was black. Accordingly, Alabama’s plan added 15,785 new individ-uals, and only 36 of those newly added individuals were white.

This suit, as it appears before us, focuses in large part upon Alabama’s efforts to achieve these two goals. The Caucus and the Conference basically claim that the State, in adding so many new minority voters to majority-minority districts (and to others), went too far. They allege the State created a constitutionally forbidden “racial gerrymander”—a gerrymander that (e.g., when the State adds more minority voters than needed for a minor-ity group to elect a candidate of its choice) might, among other things, harm the very minority voters that Acts such as the Voting Rights Act sought to help.

After a bench trial, the Federal District Court held in favor of the State, i.e., against the Caucus and the Conference, with respect to their racial gerrymandering claims as well as with respect to several other legal claims that the Caucus and the Conference had made. With respect to racial gerrymandering, the District Court recognized that electoral districting violates the Equal Protection Clause when (1) race is the “dominant and controlling” or “predominant” consideration in deciding “to place a significant number of voters within or without a particular district,” Miller v. Johnson,515 U. S. 900,913,916 (1995), and (2) the use of race is not “narrowly tailored to serve a compelling state interest,” Shaw II, 517 U. S., at 902; see also Shaw v. Reno,509 U. S. 630,649 (1993) (Shaw I ) (Constitution forbids “separat[ion of ] voters into different districts on the basis of race” when the separation “lacks sufficient justification”); Bush v. Vera,517 U. S. 952–959, 976 (1996) (principal opinion of O’Connor, J.) (same). But, after trial the District Court held (2 to 1) that the Caucus and the Conference had failed to prove their racial gerrymandering claims. The Caucus along with the Conference (and several other plaintiffs) appealed. We noted probable jurisdiction with respect to the racial gerrymandering claims. 572 U. S. ___ (2014).

We shall focus upon four critical District Court determinations underlying its ultimate “no violation” conclusion. They concern:

The Geographical Nature of the Racial Gerrymandering Claims. The District Court characterized the appellants’ claims as falling into two categories. In the District Court’s view, both appellants had argued “that the Acts as a whole constitute racial gerrymanders,” 989 F. Supp. 2d 1227, 1287 (MD Ala. 2013) (emphasis added), and one of the appellants (the Conference) had also argued that the State had racially gerrymandered four specific electoral districts, Senate Districts 7, 11, 22, and 26, id., at 1288.

Standing. The District Court held that the Caucus had standing to argue its racial gerrymandering claim with respect to the State “as a whole.” But the Conference lacked standing to make any of its racial gerrymandering claims—the claim requiring consideration of the State “as a whole,” and the claims requiring consideration of four individual Senate districts. Id., at 1292.

Racial Predominance. The District Court held that, in any event, the appellants’ claims must fail because race “was not the predominant motivating factor” either (a) “for the Acts as a whole” or (b) with respect to “Senate Districts 7, 11, 22, or 26.” Id., at 1293.

Narrow Tailoring/Compelling State Interest. The District Court also held that, even were it wrong about standing and predominance, the appellants’ racial gerrymandering claims must fail. That is because any predominant use of race in the drawing of electoral boundaries was “narrowly tailored” to serve a “compelling state interest,” id., at 1306–1307, namely the interest in avoiding retrogression with respect to racial minorities’ “ability to elect their preferred candidates of choice.” §10304(b).

In our view, each of these determinations reflects an error about relevant law. And each error likely affected the District Court’s conclusions—to the point where we must vacate the lower court’s judgment and remand the cases to allow appellants to reargue their racial gerrymandering claims. In light of our opinion, all parties remain free to introduce such further evidence as the District Court shall reasonably find appropriate.

II

We begin by considering the geographical nature of the racial gerrymandering claims. The District Court repeatedly referred to the racial gerrymandering claims as claims that race improperly motivated the drawing of boundary lines of the State considered as a whole. See, e.g., 989 F. Supp. 2d, at 1293 (“Race was not the predominant motivating factor for the Acts as a whole”); id., at 1287 (construing plaintiffs’ challenge as arguing that the “Acts as a whole constitute racial gerrymanders”); id., at 1292 (describing the plaintiffs’ challenge as a “claim of racial gerrymandering to the Acts as a whole”); cf. supra, at 4–5 (noting four exceptions).

A racial gerrymandering claim, however, applies to the boundaries of individual districts. It applies district-by-district. It does not apply to a State considered as an undifferentiated “whole.” We have consistently described a claim of racial gerrymandering as a claim that race was improperly used in the drawing of the boundaries of one or more specific electoral districts. See, e.g., Shaw I, 509 U. S., at 649 (violation consists of “separat[ing] voters into different districts on the basis of race” (emphasis added)); Vera, 517 U. S., at 965 (principal opinion) (“[Courts] must scrutinize each challenged district . . .” (emphasis added)). We have described the plaintiff’s evidentiary burden similarly. See Miller, supra, at 916 (plaintiff must show that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district” (emphasis added)).

Our district-specific language makes sense in light of the nature of the harms that underlie a racial gerrymandering claim. Those harms are personal. They include being “personally . . . subjected to [a] racial classification,” Vera, supra, at 957 (principal opinion), as well as being represented by a legislator who believes his “primary obligation is to represent only the members” of a particular racial group, Shaw I, supra, at 648. They directly threaten a voter who lives in the district attacked. But they do not so keenly threaten a voter who lives elsewhere in the State. Indeed, the latter voter normally lacks standing to pursue a racial gerrymandering claim. United States v. Hays,515 U. S. 737–745 (1995).

Voters, of course, can present statewide evidence in order to prove racial gerrymandering in a particular district. See Miller, supra, at 916. And voters might make the claim that every individual district in a State suffers from racial gerrymandering. But this latter claim is not the claim that the District Court, when using the phrase “as a whole,” considered here. Rather, the concept as used here suggests the existence of a legal unicorn, an animal that exists only in the legal imagination.

This is not a technical, linguistic point. Nor does it criticize what might seem, in effect, a slip of the pen. Rather, here the District Court’s terminology mattered. That is because the District Court found that racial criteria had not predominated in the drawing of some Alabama districts. And it found that fact (the fact that race did not predominate in the drawing of some, or many districts) sufficient to defeat what it saw as the basic claim before it, namely a claim of racial gerrymandering with respect to the State as an undifferentiated whole. See, e.g., 989 F. Supp. 2d, at 1294 (rejecting plaintiffs’ challenge because “[the legislature] followed no bright-line rule” with respect to every majority-minority district); id., at 1298–1299, 1301 (citing examples of majority-minority districts in which black population percentages were reduced and examples of majority-white districts in which precincts were split).

A showing that race-based criteria did not significantly affect the drawing of some Alabama districts, however, would have done little to defeat a claim that race-based criteria predominantly affected the drawing of other Alabama districts, such as Alabama’s majority-minority districts primarily at issue here. See id., at 1329 (Thompson, J., dissenting) (“[T]he drafters[’] fail[ure] to achieve their sought-after percentage in one district does not detract one iota from the fact that they did achieve it in another”). Thus, the District Court’s undifferentiated statewide analysis is insufficient. And we must remand for consideration of racial gerrymandering with respect to the individual districts subject to the appellants’ racial gerrymandering challenges.

The State and principal dissent argue that (but for four specifically mentioned districts) there were in effect no such districts. The Caucus and the Conference, the State and principal dissent say, did not seek a district-by-district analysis. And, the State and principal dissent conclude that the Caucus and the Conference have consequently waived the right to any further consideration. Brief for Appellees 14, 31; post, at 5–12 (opinion ofScalia, J.).

We do not agree. We concede that the District Court’s opinion suggests that it was the Caucus and the Conference that led the Court to consider racial gerrymandering of the State “as a whole.” 989 F. Supp. 2d, at 1287. At least the District Court interpreted their filings to allege only that kind of claim. Ibid. But our review of the record indicates that the plaintiffs did not claim only that the legislature had racially gerrymandered the State “as” an undifferentiated “whole.” Rather, their evidence and their arguments embody the claim that individual majority-minority districts were racially gerrymandered. And those are the districts that we believe the District Court must reconsider.

There are 35 majority-minority districts, 27 in the House and 8 in the Senate. The District Court’s opinion itself refers to evidence that the legislature’s redistricting committee, in order to satisfy what it believed the Voting Rights Act required, deliberately chose additional black voters to move into underpopulated majority-minority districts, i.e., a specific set of individual districts. See, e.g., 989 F. Supp. 2d, at 1274 (referring to Senator Dial’s testimony that the Committee “could have used,” but did not use, “white population within Jefferson County to repopulate the majority-black districts” because “doing so would have resulted in the retrogression of the majority-black districts and potentially created a problem for [Justice Department] preclearance”); id., at 1276 (stating that Representative Jim McClendon, also committee cochair, “testified consistently with Senator Dial”); id., at 1277 (noting that the committee’s expert, Randolph Hinaman, testified that “he needed to add population” to majority-black districts “without significantly lowering the percentage of the population in each district that was majority-black”).

The Caucus and the Conference presented much evidence at trial to show that that the legislature had deliberately moved black voters into these majority-minority districts—again, a specific set of districts—in order to prevent the percentage of minority voters in each district from declining. See, e.g., Committee Guidelines 3–5; 1 Tr. 28–29, 36–37, 55, 63, 67–68, 77, 81, 96, 115, 124, 136, 138 (testimony of Senator Dial); Deposition of Gerald Dial in No. 12–cv–691 (May 21, 2013), Doc. 123–5, pp. 17, 39–41, 62, 100 (Dial Deposition); 3 Tr. 222 (testimony of Representative McClendon); id., at 118–119, 145–146, 164, 182–183, 186–187 (testimony of Hinaman); Deposition of Randolph Hinaman in No. 12–cv–691 (June 25, 2013), Doc. 134–4, pp. 23–24, 101 (Hinaman Deposition).

In their post-trial Proposed Findings of Fact and Conclusions of Law, the plaintiffs stated that the evidence showed a racial gerrymander with respect to the majority of the majority-minority districts; they referred to the specific splitting of precinct and county lines in the drawing of many majority-minority districts; and they pointed to much district-specific evidence. E.g., Alabama Legislative Black Caucus Plaintiffs’ Notice of Filing Proposed Findings of Fact and Conclusions of Law in No. 12–cv–691, Doc. 194, pp. 9–10, 13–14, 30–35, 40 (Caucus Post-Trial Brief); Newton Plaintiffs’ Notice of Filing Proposed Findings of Fact and Conclusions of Law in No. 12–cv–691, Doc. 195, pp. 33–35, 56–61, 64–67, 69–74, 82–85, 108, 121–122 (Conference Post-Trial Brief); see also Appendix A, infra (organizing these citations by district).

We recognize that the plaintiffs relied heavily upon statewide evidence to prove that race predominated in the drawing of individual district lines. See generally Caucus Post-Trial Brief 1, 3–7, 48–50; Conference Post-Trial Brief 2, 44–45, 105–106. And they also sought to prove that the use of race to draw the boundaries of the majority-minority districts affected the boundaries of other districts as well. See, e.g., 1 Tr. 36–37, 48, 55, 70–71, 93, 111, 124 (testimony of Dial); 3 Tr. 142, 162 (testimony of Hinaman); see generally Caucus Post-Trial Brief 8–16. Such evidence is perfectly relevant. We have said that the plaintiff’s burden in a racial gerrymandering case is “to show, either through circumstantial evidence of a district’s shape and demographics or more direct evidence going to legislative purpose, that race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller, 515 U. S., at 916. Cf. Easley v. Cromartie,532 U. S. 234,258 (2001) (explaining the plaintiff’s burden in cases, unlike these, in which the State argues that politics, not race, was its predominant motive). That Alabama expressly adopted and applied a policy of prioritizing mechanical racial targets above all other districting criteria (save one-person, one-vote) provides evidence that race motivated the drawing of particular lines in multiple districts in the State. And neither the use of statewide evidence nor the effort to show widespread effect can transform a racial gerrymandering claim about a set of individual districts into a separate, general claim that the legislature racially gerrymandered the State “as” an undifferentiated “whole.”

We, like the principal dissent, recognize that the plaintiffs could have presented their district-specific claims more clearly, post, at 6–8, 10–12 (opinion of Scalia, J.), but the dissent properly concedes that its objection would weaken had the Conference “developed such a claim in the course of discovery and trial.” Post, at 6. And that is just what happened.

In the past few pages and in Appendix A, we set forth the many record references that establish this fact. The Caucus helps to explain the complaint omissions when it tells us that the plaintiffs unearthed the factual basis for their racial gerrymandering claims when they deposed the committee’s redistricting expert. See Brief for Appellants in No. 13–895, pp. 12–13. The State neither disputes this procedural history nor objects that plaintiffs’ pleadings failed to conform with the proof. Indeed, throughout, the plaintiffs litigated these claims not as if they were wholly separate entities but as if they were a team. See, e.g., Caucus Post-Trial Brief 1 (“[We] support the additional claims made by the [Conference] plaintiffs”); but cf. post, at 3–12 (Scalia, J., dissenting) (treating separately Conference claims from Caucus claims). Thus we, like the dissenting judge below (who also lived with these cases through trial), conclude that the record as a whole shows that the plaintiffs brought, and their argument rested significantly upon, district-specific claims. See 989 F. Supp. 2d, at 1313 (Thompson, J., dissenting) (construing plaintiffs as also challenging “each majority-Black House and Senate District”).

The principal dissent adds that the Conference waived its district-specific claims on appeal. Cf. post, at 8. But that is not so. When asked specifically about its position at oral argument, the Conference stated that it was relying on statewide evidence to prove its district-specific challenges. Tr. of Oral Arg. 15–16. Its counsel said that “the exact same policy was applied in every black-majority district,” id., at 15, and “[b]y statewide, we simply mean a common policy applied to every district in the State,” id., at 16. We accept the Conference’s clarification, which is consistent with how it presented these claims below.

We consequently conclude that the District Court’s analysis of racial gerrymandering of the State “as a whole” was legally erroneous. We find that the appellants did not waive their right to consideration of their claims as applied to particular districts. Accordingly, we remand the cases. See Pullman-Standard v. Swint,456 U. S. 273,291 (1982) (remand is required when the District Court “failed to make a finding because of an erroneous view of the law”); Rapanos v. United States,547 U. S. 715,757 (2006) (same).

III

We next consider the District Court’s holding with respect to standing. The District Court, sua sponte, held that the Conference lacked standing—either to bring racial gerrymandering claims with respect to the four individual districts that the court specifically considered (i.e., Senate Districts 7, 11, 22, and 26) or to bring a racial gerrymandering claim with respect to the “State as a whole.” 989 F. Supp. 2d, at 1292.

The District Court recognized that ordinarily

“[a]n association has standing to bring suit on behalf of its members when its members would have standing to sue in their own right, the interests at stake are germane to the organization’s purpose, and neither the claim asserted nor the relief requested requires individuals members’ participation in the lawsuit.” Id., at 1291 (quoting Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc.,528 U. S. 167,181 (2000); emphasis added).

It also recognized that a “member” of an association “would have standing to sue” in his or her “own right” when that member “resides in the district that he alleges was the product of a racial gerrymander.” 989 F. Supp. 2d, at 1291 (citing Hays, 515 U. S., at 744–745). But, the District Court nonetheless denied standing because it believed that the “record” did “not clearly identify the districts in which the individual members of the [Conference] reside,” and the Conference had “not proved that it has members who have standing to pursue any district-specific claims of racial gerrymandering.” 989 F. Supp. 2d, at 1292.

The District Court conceded that Dr. Joe Reed, a representative of the Conference, testified that the Conference “has members in almost every county in Alabama.” Ibid. But, the District Court went on to say that “the counties in Alabama are split into many districts.” Ibid. And the “Conference offered no testimony or evidence that it has members in all of the districts in Alabama or in any of the [four] specific districts that it challenged.” Ibid.

The record, however, lacks adequate support for the District Court’s conclusion. Dr. Reed’s testimony supports, and nothing in that record undermines, the Conference’s own statement, in its post-trial brief, that it is a “statewide political caucus founded in 1960.” Conference Post-Trial Brief 3. It has the “purpose” of “endors[ing] candidates for political office who will be responsible to the needs of the blacks and other minorities and poor people.” Id., at 3–4. These two statements (the second of which the principal dissent ignores), taken together with Dr. Reed’s testimony, support an inference that the organization has members in all of the State’s majority-minority districts, other things being equal, which is sufficient to meet the Conference’s burden of establishing standing. That is to say, it seems highly likely that a “statewide” organization with members in “almost every county,” the purpose of which is to help “blacks and other minorities and poor people,” will have members in each majority-minority district. But cf. post, at 3–5 (Scalia, J., dissenting).

At the very least, the common sense inference is strong enough to lead the Conference reasonably to believe that, in the absence of a state challenge or a court request for more detailed information, it need not provide additional information such as a specific membership list. We have found nothing in the record, nor has the State referred us to anything in the record, that suggests the contrary. Cf. App. 204–205, 208 (State arguing lack of standing, not because of inadequate member residency but because an association “lives” nowhere and that the Conference should join individual members). The most the State argued was that “[n]one of the individual [p]laintiffs [who brought the case with the Conference] claims to live in” Senate District 11, id., at 205 (emphasis added), but the Conference would likely not have understood that argument as a request that it provide a membership list. In fact, the Conference might have understood the argument as an indication that the State did not contest its membership in every district.

To be sure, the District Court had an independent obligation to confirm its jurisdiction, even in the absence of a state challenge. See post, at 4–5 (Scalia, J., dissenting). But, in these circumstances, elementary principles of procedural fairness required that the District Court, rather than acting sua sponte, give the Conference an oppor-tunity to provide evidence of member residence. Cf. Warth v. Seldin,422 U. S. 490–502 (1975) (explaining that a court may “allow or [r]equire” a plaintiff to supplement the record to show standing and that “[i]f, after this opportu-nity, the plaintiff’s standing does not adequately appear from all materials of record, the complaint must be dismissed” (emphasis added)). Moreover, we have no reason to believe that the Conference would have been unable to provide a list of members, at least with respect to the majority-minority districts, had it been asked. It has filed just such a list in this Court. See Affidavit of Joe L. Reed Pursuant to this Court’s Rule 32.3 (Lodging of Conference affidavit listing members residing in each majority-minority district in the State); see also Parents Involved in Community Schools v. Seattle School Dist. No. 1,551 U. S. 701,718 (2007) (accepting a lodged affidavit in similar circumstances). Thus, the District Court on remand should reconsider the Conference’s standing by permitting the Conference to file its list of members and permitting the State to respond, as appropriate.

IV

The District Court held in the alternative that the claims of racial gerrymandering must fail because “[r]ace was not the predominant motivating factor” in the creation of any of the challenged districts. 989 F. Supp. 2d, at 1293. In our view, however, the District Court did not properly calculate “predominance.” In particular, it judged race to lack “predominance” in part because it placed in the balance, among other nonracial factors, legislative efforts to create districts of approximately equal population. See, e.g., id., at 1305 (the “need to bring the neighboring districts into compliance with the requirement of one person, one vote served as the primary motivating factor for the changes to [Senate] District 22” (emphasis added)); id., at 1297 (the “constitutional requirement of one person, one vote trumped every other districting principle”); id., at 1296 (the “record establishes that the drafters of the new districts, above all, had to correct [for] severe malapportionment . . .”); id., at 1306 (the “inclusion of additional precincts [in Senate District 26] is a reasonable response to the underpopulation of the District”).

In our view, however, an equal population goal is not one factor among others to be weighed against the use of race to determine whether race “predominates.” Rather, it is part of the redistricting background, taken as a given, when determining whether race, or other factors, predominate in a legislator’s determination as to how equal population objectives will be met.

To understand this conclusion, recall what “predominance” is about: A plaintiff pursuing a racial gerrymandering claim must show that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller, 515 U. S., at 916. To do so, the “plaintiff must prove that the legislature subordinated traditional race-neutral districting principles . . . to racial considerations.” Ibid. (emphasis added).

Now consider the nature of those offsetting “traditional race-neutral districting principles.” We have listed several, including “compactness, contiguity, respect for political subdivisions or communities defined by actual shared interests,” ibid., incumbency protection, and political affiliation, Vera, 517 U. S., at 964, 968 (principal opinion).

But we have not listed equal population objectives. And there is a reason for that omission. The reason that equal population objectives do not appear on this list of “traditional” criteria is that equal population objectives play a different role in a State’s redistricting process. That role is not a minor one. Indeed, in light of the Constitution’s demands, that role may often prove “predominant” in the ordinary sense of that word. But, as the United States points out, “predominance” in the context of a racial gerrymandering claim is special. It is not about whether a legislature believes that the need for equal population takes ultimate priority. Rather, it is, as we said, whether the legislature “placed” race “above traditional districting considerations in determining which persons were placed in appropriately apportioned districts.” Brief for United States as Amicus Curiae 19 (some emphasis added). In other words, if the legislature must place 1,000 or so additional voters in a particular district in order to achieve an equal population goal, the “predominance” question concerns which voters the legislature decides to choose, and specifically whether the legislature predominately uses race as opposed to other, “traditional” factors when doing so.

Consequently, we agree with the United States that the requirement that districts have approximately equal populations is a background rule against which redistricting takes place. Id., at 12. It is not a factor to be treated like other nonracial factors when a court determines whether race predominated over other, “traditional” factors in the drawing of district boundaries.

Had the District Court not taken a contrary view of the law, its “predominance” conclusions, including those concerning the four districts that the Conference specifically challenged, might well have been different. For example, once the legislature’s “equal population” objectives are put to the side—i.e., seen as a background principle—then there is strong, perhaps overwhelming, evidence that race did predominate as a factor when the legislature drew the boundaries of Senate District 26, the one district that the parties have discussed here in depth.

The legislators in charge of creating the redistricting plan believed, and told their technical adviser, that a primary redistricting goal was to maintain existing racial percentages in each majority-minority district, insofar as feasible. See supra, at 9–10 (compiling extensive record testimony in support of this point). There is considerable evidence that this goal had a direct and significant impact on the drawing of at least some of District 26’s boundaries. See 3 Tr. 175–180 (testimony of Hinaman); Appendix C, infra (change of district’s shape from rectangular to irregular). Of the 15,785 individuals that the new redistricting laws added to the population of District 26, just 36 were white—a remarkable feat given the local demographics. See, e.g., 2 Tr. 127–128 (testimony of Senator Quinton Ross); 3 Tr. 179 (testimony of Hinaman). Transgressing their own redistricting guidelines, Committee Guidelines 3–4, the drafters split seven precincts between the majority-black District 26 and the majority-white District 25,with the population in those precincts clearly divided on racial lines. See Exh. V in Support of Newton Plaintiffs’ Opposition to Summary Judgment in No. 12–cv–691, Doc. 140–1, pp. 91–95. And the District Court conceded that race “was a factor in the drawing of District 26,” and that the legislature “preserved” “the percentage of the population that was black.” 989 F. Supp. 2d, at 1306.

We recognize that the District Court also found, with respect to District 26, that “preservi[ng] the core of the existing [d]istrict,” following “county lines,” and following “highway lines” played an important boundary-drawing role. Ibid. But the first of these (core preservation) is not directly relevant to the origin of the new district inhabitants; the second (county lines) seems of marginal importance since virtually all Senate District 26 boundaries departed from county lines; and the third (highways) was not mentioned in the legislative redistricting guidelines. Cf. Committee Guidelines 3–5.

All this is to say that, with respect to District 26 and likely others as well, had the District Court treated equal population goals as background factors, it might have concluded that race was the predominant boundary-drawing consideration. Thus, on remand, the District Court should reconsider its “no predominance” conclusions with respect to Senate District 26 and others to which our analysis is applicable.

Finally, we note that our discussion in this section is limited to correcting the District Court’s misapplication of the “predominance” test for strict scrutiny discussed in Miller, 515 U. S., at 916. It does not express a view on the question of whether the intentional use of race in redistricting, even in the absence of proof that traditional districting principles were subordinated to race, triggers strict scrutiny. See Vera, 517 U. S., at 996 (Kennedy, J., concurring).

V

The District Court, in a yet further alternative holding, found that “[e]ven if the [State] subordinated traditional districting principles to racial considerations,” the racial gerrymandering claims failed because, in any event, “the Districts would satisfy strict scrutiny.” 989 F. Supp. 2d, at 1306. In the District Court’s view, the “Acts are narrowly tailored to comply with Section 5” of the Voting Rights Act. Id., at 1311. That provision “required the Legislature to maintain, where feasible, the existing number of majority-black districts and not substantially reduce the relative percentages of black voters in those districts.” Ibid. (emphasis added). And, insofar as the State’s redistricting embodied racial considerations, it did so in order to meet this §5 requirement.

In our view, however, this alternative holding rests upon a misperception of the law. Section 5, which covered particular States and certain other jurisdictions, does not require a covered jurisdiction to maintain a particular numerical minority percentage. It requires the jurisdiction to maintain a minority’s ability to elect a preferred candidate of choice. That is precisely what the language of the statute says. It prohibits a covered jurisdiction from adopting any change that “has the purpose of or will have the effect of diminishing the ability of [the minority group] to elect their preferred candidates of choice.”52 U. S. C. §10304(b); see also §10304(d) (the “purpose of subsection (b) . . . is to protect the ability of such citizens to elect their preferred candidates of choice”).

That is also just what Department of Justice Guidelines say. The Guidelines state specifically that the Department’s preclearance determinations are not based

“on any predetermined or fixed demographic percentages. . . . Rather, in the Department’s view, this determination requires a functional analysis of the electoral behavior within the particular jurisdiction or election district. . . . [C]ensus data alone may not provide sufficient indicia of electoral behavior to make the requisite determination.” Guidance Concerning Redistricting Under Section 5 of the Voting Rights Act, 76 Fed. Reg. 7471 (2011).

Consistent with this view, the United States tells us that “Section 5” does not “requir[e] the State to maintain the same percentage of black voters in each of the majority-black districts as had existed in the prior districting plans.” Brief for United States as Amicus Curiae 22. Rather, it “prohibits only those diminutions of a minority group’s proportionate strength that strip the group within a district of its existing ability to elect its candidates of choice.” Id., at 22–23. We agree. Section 5 does not require maintaining the same population percentages in majority-minority districts as in the prior plan. Rather, §5 is satisfied if minority voters retain the ability to elect their preferred candidates.

The history of §5 further supports this view. In adopting the statutory language to which we referred above, Congress rejected this Court’s decision in Georgia v. Ashcroft,539 U. S. 461,480 (2003) (holding that it is not nec-essarily retrogressive for a State to replace safe majority-minority districts with crossover or influence districts), and it adopted the views of the dissent. H. R. Rep. No. 109–478, pp. 68–69, and n. 183 (2006). While the thrust of Justice Souter’s dissent was that, in a §5 retrogression case, courts should ask whether a new voting provision would likely deprive minority voters of their ability to elect a candidate of their choice—language that Congress adopted in revising §5—his dissent also made clear that courts should not mechanically rely upon numerical percentages but should take account of all significant circumstances. Georgia v. Ashcroft, supra, at 493, 498, 505, 509. And while the revised language of §5 may raise some interpretive questions—e.g., its application to coalition, crossover, and influence districts—it is clear that Congress did not mandate that a 1% reduction in a 70% black population district would be necessarily retrogressive. See Persily, The Promises and Pitfalls of the New Voting Rights Act, 117 Yale L. J. 174, 218 (2007). Indeed, Alabama’s mechanical interpretation of §5 can raise serious constitutional concerns. See Miller, supra, at 926.

The record makes clear that both the District Court and the legislature relied heavily upon a mechanically numerical view as to what counts as forbidden retrogression. See Appendix B, infra. And the difference between that view and the more purpose-oriented view reflected in the statute’s language can matter. Imagine a majority-minority district with a 70% black population. Assume also that voting in that district, like that in the State itself, is racially polarized. And assume that the district has long elected to office black voters’ preferred candidate. Other things being equal, it would seem highly unlikely that a redistricting plan that, while increasing the numerical size of the district, reduced the percentage of the black population from, say, 70% to 65% would have a significant impact on the black voters’ ability to elect their preferred candidate. And, for that reason, it would be difficult to explain just why a plan that uses racial criteria predominately to maintain the black population at 70% is “narrowly tailored” to achieve a “compelling state interest,” namely the interest in preventing §5 retrogression. The cir-cumstances of this hypothetical example, we add, areclose to those characterizing Senate District 26, as set forth in the District Court’s opinion and throughout the record. See, e.g., 1 Tr. 131–132 (testimony of Dial); 3 Tr. 180 (testimony of Hinaman).

In saying this, we do not insist that a legislature guess precisely what percentage reduction a court or the Justice Department might eventually find to be retrogressive. The law cannot insist that a state legislature, when redistricting, determine precisely what percent minority population §5 demands. The standards of §5 are complex; they often require evaluation of controverted claims about voting behavior; the evidence may be unclear; and, with respect to any particular district, judges may disagree about the proper outcome. The law cannot lay a trap for an unwary legislature, condemning its redistricting plan as either (1) unconstitutional racial gerrymandering should the legislature place a few too many minority voters in a district or (2) retrogressive under §5 should the legislature place a few too few. See Vera, 517 U. S., at 977 (principal opinion). Thus, we agree with the United States that a court’s analysis of the narrow tailoring requirement insists only that the legislature have a “strong basis in evidence” in support of the (race-based) choice that it has made. Brief for United States as Amicus Curiae 29 (citing Ricci v. DeStefano,557 U. S. 557,585 (2009)). This standard, as the United States points out, “does not demand that a State’s actions actually be necessary to achieve a compelling state interest in order to be constitutionally valid.” Brief for United States as Amicus Curiae 29. And legislators “may have a strong basis in evidence to use racial classifications in order to comply with a statute when they have good reasons to believe such use is required, even if a court does not find that the actions were necessary for statutory compliance.” Ibid. (emphasis added).

Here the District Court enunciated a narrow tailoring standard close to the one we have just mentioned. It said that a plan is “narrowly tailored . . . when the race-based action taken was reasonably necessary” to achieve a compelling interest. 989 F. Supp. 2d, at 1307 (emphasis added). And it held that preventing retrogression is a compel-ling interest. Id., at 1306–1307. While we do not here decide whether, given Shelby County v. Holder, 570 U. S. ___ (2013), continued compliance with §5 remains a compelling interest, we conclude that the District Court and the legislature asked the wrong question with respect to narrow tailoring. They asked: “How can we maintain present minority percentages in majority-minority districts?” But given §5’s language, its purpose, the Justice Department Guidelines, and the relevant precedent, they should have asked: “To what extent must we preserve existing minority percentages in order to maintain the minority’s present ability to elect the candidate of its choice?” Asking the wrong question may well have led to the wrong answer. Hence, we cannot accept the District Court’s “compelling interest/narrow tailoring” conclusion.

*  *  *

For these reasons, the judgment of the District Court is vacated. We note that appellants have also raised additional questions in their jurisdictional statements, relating to their one-person, one-vote claims (Caucus) and vote dilution claims (Conference), which were also rejected by the District Court. We do not pass upon these claims. The District Court remains free to reconsider the claims should it find reconsideration appropriate. And the parties are free to raise them, including as modified by the District Court, on any further appeal.

The cases are remanded for further proceedings consistent with this opinion.

It is so ordered.

Appendixes

A

* Senate District 26 excluded from this list

B

State’s Use of Incorrect Retrogression Standard

The following citations reflect instances in either the District Court opinion or in the record showing that the State believed that §5 forbids, not just substantial reductions, but any reduction in the percentage of black inhabitants of a majority-minority district.

C