Sam GONZALEZ, Maria Crosby, and Mariana Correa, Plaintiffs-Appellants, v. CITY OF AURORA, ILLINOIS, Defendant-Appellee.
In the 2000 Census, 32.6% of the population in the City of Aurora, Illinois, identified itself as Hispanic, but of the City's residents who are citizens and old enough to vote only 16.3% are Hispanic. Aurora has 10 single-seat wards, only one of which reliably elects Latino candidates to the City Council. Another ward, although about 66% Latino, has twice elected a black alderman since the redistricting that followed the 2000 Census. When the record was compiled, 2 of the 12 aldermen (there are 2 at-large seats in addition to the 10 wards) were Hispanic. One was elected and the second appointed. Plaintiffs contend in this suit under § 2 of the Voting Rights Act, 42 U.S.C. § 1973, that these numbers are insufficient. They want an injunction compelling the City to redraw the ward boundaries so that Aurora's Latino population is concentrated in three wards, each of which then would be likely to elect a Latino candidate (would be, as plaintiffs say, “Latino effective”).
Plaintiffs start with the proposition that it takes 70% or more Latino population to ensure the election of a Latino candidate. Whatever rule of thumb courts may have used in the 1960s and 1970s for black voters does not apply to Latinos, plaintiffs contend, because Latinos are younger and less likely to be citizens than are blacks and other minorities. Cf. Barnett v. Chicago, 141 F.3d 699 (7th Cir.1998) (discussing the rules of thumb used in voting cases and their doubtful transferability from one minority group to another). Although the City used the rule of thumb that 65% population is enough to make a district “effective” for a minority group, plaintiffs are sure that this won't work. This table shows why 65% may not be enough:
These figures, all from 2000, are the right ones to use. Plaintiffs' estimates about population in 2005 don't matter, because apportionment is based on Census returns. The district court concluded that a ward with 65% or more Latino residents should be deemed sufficient no matter who it elects. If Latinos vote for candidates of other ethnic backgrounds, this means that Aurora is not afflicted by racial bloc voting, rather than that the map deprives Latinos' votes of full effect. The judge added that, with 16% of the eligible population, Latinos would receive 2 seats in a 12-seat legislature under proportional representation. As 2 of the existing 12 members were Latino, the district judge saw no problem under § 2 and granted summary judgment for the City. 2006 WL 681048, 2006 U.S. Dist. Lexis 10677 (N.D.Ill. Mar. 13, 2006), reconsideration denied, 2006 WL 3227893, 2006 U.S. Dist. Lexis 81451 (N.D.Ill. Nov. 3, 2006).
The most striking thing about plaintiffs' brief on appeal is that it neither quotes from nor analyzes the text of § 2. Instead it leaps straight to the “Gingles factors” (from Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986)) and language in a Senate committee report. The statute is not self-defining, so it is understandable that lawyers would turn to secondary sources such as judicial decisions and legislative history. But neither is it irrelevant. It is worth quoting. Section 2(a) says that governments cannot adopt standards, practices, or procedures that “result[ ] in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color”. This sounds like a rule that race and color cannot be used to prevent anyone from voting, or to disregard a vote once cast. Section 2(b), 42 U.S.C. § 1973(b), then adds this famously elliptical language:
A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.
What does it mean to “have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice”? Gingles held that gerrymandering district borders can have this effect even though everyone is entitled to vote, and all votes are counted equally. The Court set out circumstances (the “Gingles factors”) under which clever map-drawing could have this effect and then turned to the Senate committee report for factors to consider if the conditions are met. The district judge found, and we shall assume, that these conditions are satisfied in Aurora: Latinos are sufficiently concentrated geographically that they can form a majority in some districts (whether potential majority status is essential is the question posed in Bartlett v. Strickland, cert. granted, --- U.S. ----, 128 S.Ct. 1648, 170 L.Ed.2d 352 (2008) (to be argued Oct. 14, 2008), but not one we need consider); Latinos are politically cohesive; and, without a large bloc of voters, Latino candidates rarely prevail. This just sets the stage.
Plaintiffs leap from satisfaction of the Gingles factors to the proposition that the City must do what is possible to maximize Latino voters' ability to elect Latino candidates (euphemistically “candidates of their choice”). But neither § 2 nor Gingles nor any later decision of the Supreme Court speaks of maximizing the influence of any racial or ethnic group. (Nor does § 5 of the Act, see Reno v. Bossier Parish School Board, 528 U.S. 320, 120 S.Ct. 866, 145 L.Ed.2d 845 (2000).) Section 2 requires an electoral process “equally open” to all, not a process that favors one group over another. One cannot maximize Latino influence without minimizing some other group's influence. A map drawn to advantage Latino candidates at the expense of black (or white ethnic) candidates violates § 2 as surely as a map drawn to maximize the influence of those groups at the expense of Latinos.
The Supreme Court emphasized in League of United Latin American Citizens v. Perry, 548 U.S. 399, 126 S.Ct. 2594, 165 L.Ed.2d 609 (2006) (LULAC), its most recent § 2 redistricting case, that the Voting Rights Act protects the rights of individual voters, not the rights of groups. See Shaw v. Hunt, 517 U.S. 899, 917, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996). That's why LULAC rejected Texas's argument that it could dilute the votes of Latinos in one part of the state as long as it gerrymandered a Latino-majority district somewhere else. There is a serious problem with any proposal to employ black or Asian or white citizens of some other ethnic background as “fill” in districts carefully drawn to ensure three 70%-Latino wards-wards in which the remaining 30% are (by design) never going to be able to elect a candidate of their choice. How could one explain to this 30% that the political process was “equally open” to them, as § 2 commands? A problem under § 2 arises whenever any person is moved from one district to another to minimize the value of his vote and give an advantage to someone else.
Section 2's requirement of an “equally open” process usually is described as including a prohibition of vote dilution by redistricting. See Abrams v. Johnson, 521 U.S. 74, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997); Bush v. Vera, 517 U.S. 952, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996); and Holder v. Hall, 512 U.S. 874, 114 S.Ct. 2581, 129 L.Ed.2d 687 (1994), in addition to Gingles and LULAC. Plaintiffs want the court to reduce the influence of others to produce an advantage for Latino voters. That may be necessary as a remedy for some earlier vote-dilution exercise, see Shaw v. Hunt, 517 U.S. 899, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996), but the first question we need to ask is whether Latino votes have been diluted by Aurora's map. Diluted relative to what benchmark? Not the maximum influence Latinos could have, surely; as we've explained, no group is entitled to that (and all groups cannot enjoy maximum influence simultaneously). Nor is proportional representation the benchmark. Gingles holds that this is not the statute's objective-that it is not necessary and, LULAC adds, is not sufficient either, if a minority group in one part of a jurisdiction has been thrown to the wolves.
So what benchmarks are possible? One would be the outcome of a race-neutral process in which all districts are compact. Cases in which the Supreme Court has found a problem under § 2 all involve transparent gerrymandering that boosts one group's chances at the expense of another's. See, e.g., Shaw v. Reno, 509 U.S. 630, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993) (an inkblot of a district stretching almost all the way across North Carolina); Miller v. Johnson, 515 U.S. 900, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995) (extensive racial gerrymander violates equal protection clause and cannot be rescued by a desire to satisfy § 2 or § 5); LULAC, 548 U.S. at 441, 126 S.Ct. at 2623 (“entirely new district that combined two groups of Latinos, hundreds of miles apart, that represent different communities of interest”). Nothing remotely similar happened in Aurora. A glance at the 2002 ward map reveals that the districts are compact and regular, with the few rough edges needed to ensure that they have equal population.
Still, although many opinions, of which LULAC is the most recent, emphasize compact districts as the benchmark for a map that does not dilute any group's influence, it is possible to locate even compact districts for political advantage.
Given the very large number of ways that reasonably compact districts of equal population can be drawn, how can a court tell whether a jurisdiction has chosen a particular arrangement in order to advantage one ethnic group over another, diluting the influence of the disfavored group? When the Voting Rights Act was enacted, the answer would have depended on the intent of those who drew the map. See City of Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980). The 1982 amendments replaced intent with effect as the rule of decision, see Gingles, but did not supply a means to test whether a given map was ordinary or abnormal. Today, however, computers can use census data to generate many variations on compact districts with equal population. One could do this exercise a hundred or a thousand times, each time placing the center of the first (or “seed”) district in a different location. That would generate a hundred or a thousand different maps, and the software could easily check these to determine the ethnic makeup of the districts.
Suppose that after 1,000 different maps of Aurora's wards have been generated, 10% have two or three “safe” districts for Latinos and the other 90% look something like the actual map drawn in 2002: one safe district and two “influence districts” where no candidate is likely to win without substantial Latino support. Then we could confidently conclude that Aurora's map did not dilute the effectiveness of the Latino vote. But suppose, instead, that Latinos are sufficiently concentrated that the random, race-blind exercise we have proposed yields three “Latino effective” districts at least 50% of the time. Then a court might sensibly conclude that Aurora had diluted the Latino vote by undermining the normal effects of the choices that Aurora's citizens had made about where to live. Redistricting software can not answer all hard questions, but it provides a means to implement a pure effects test without demanding proportional representation.
Plaintiffs did not conduct such an exercise, however (or, if they did, they didn't put the results in the record). What we can see from the record suggests that Latinos are not concentrated enough to support three “Latino effective” districts without serious gerrymandering. Ward 2 has 2,453 voting-age citizens who identified themselves as Hispanic in the 2000 Census. Wards 7 and 3 have fewer. Wards 1, 4, and 6 all have more than 1,000 citizen, voting-age Latinos. Ward 5 has another 668. In other words, the Latino population is not concentrated in a way that neutrally drawn compact districts would produce three “Latino effective” wards. That may be why plaintiffs have staked their all on a proposal that Latinos are entitled at least to proportional representation via two Latino-effective districts no matter what the consequences of race-blind districting would be. The Voting Rights Act does not require either outcome.
Because plaintiffs lack any evidence of dilution, there is no point in traipsing through the multiple factors mentioned in the 1982 committee reports. Although plaintiffs briefly mention the City's two at-large districts-at-large districts are a traditional means of reducing the influence of minority groups-they did not make much of them in the district court or here, devoting less than two pages of their brief to the subject. The at-large districts predate the 2002 reapportionment and, for all we know, long predate the presence of a substantial Latino population in Aurora. After the 2000 Census, the City increased the number of single-member districts from 8 to 10, reducing the effect of the 2 at-large districts. Any contention that the at-large districts violate § 2 of the Voting Rights Act has been forfeited.
And plaintiffs have no other arguments. They ignore the fact that several wards-at least Wards 3 and 7, and likely wards 1, 4, and 6 as well-contain enough Latino citizens to produce substantial influence. (Many cases, of which LULAC is again the most recent example, hold that “influence districts” count in any assessment of vote dilution. See also, e.g., Johnson v. De Grandy, 512 U.S. 997, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994).) Plaintiffs tried for the big prize (three safe districts) but did not build the sort of factual record that creates a genuine issue for trial, even under the balancing approach of Gingles. They thought, wrongly, that all they had to do to prevail is to show that Ward 7 is not “Latino effective.” That's not enough to condemn the current map. The district court's judgment therefore is
EASTERBROOK, Chief Judge.