LULAC v. Perry held that Texas violated Section 2 of the Voting Rights Act when it displaced nearly 100,000 Latino residents from a congressional district in Laredo to protect the Republican incumbent they refused to support. At the same time, the Justices let stand the dismantling of a so-called "coalition" district in Fort Worth where African-American voters comprising a minority of the district's population allegedly enjoyed effective control in deciding the district's representative.
Only Justice Kennedy supported the outcome in both Laredo and Fort Worth. His opinion marks the first time that he, or indeed a majority of the Justices, has identified a Section 2 violation in a merits case since Congress amended the statute in 1982. In so doing, Justice Kennedy posits a distinct conception of the role of race in the districting process, one that prohibits districts drawn to construct racially-defined political communities as well as districts that impair the political power of existing communities. This approach neither removes race from the decision-making calculus nor tempers its prominence, but instead channels its use to designated ends. The approach differs from that taken by the Chief Justice, who lamented this "sordid business, this divvying us by race," but would have done nothing to restrain the institutions most responsible for doing so.
Justice Kennedy's opinion also demonstrates both how partisan gerrymandering shapes claims under the Voting Rights Act, and how the Act may provide a meaningful curb on such gerrymandering. LULAC specifically restrains the use of incumbency protection as a districting principle, at least insofar as its application dilutes the voting strength of a racial minority. More generally, LULAC's identification of racial vote dilution in Laredo but not in Fort Worth suggests a nascent conception of political harm to voters regardless of race when a political system is rigged to block competition. In other words, LULAC suggests that Justice Kennedy may find within the Voting Rights Act itself the standard he has been seeking for managing claims of partisan gerrymandering.
LULAC seeks to do away with two types of racially-informed districts. The first, long targeted by Justice Kennedy for elimination in cases like Miller v. Johnson and Rice v. Cayetano, is a district drawn to create a racially-defined political community that would not otherwise exist. That is what Justice Kennedy saw Texas doing in District 25 where it linked Latino voters on the Mexican border with those living hundreds of miles away in Austin.
Texas argued that the Voting Rights Act required the creation of a district like District 25 to compensate for the displacement of Latino voters in Laredo. What Justice Kennedy saw, however, was a misguided attempt at social engineering that sought to unite "far flung segments of a racial group with disparate interests." Echoing language he has used before, Justice Kennedy lamented the State's failure to respect and to "account for the differences between people of the same race." District 25 was simply too vast and the Latino voters united there too disparate from one another to let the district stand, at least insofar as it was meant to offset what was done in Laredo.
The Chief Justice rightly pointed out the novelty of Justice Kennedy's specific finding that District 25 was insufficiently compact for purposes of Section 2, but the disagreement between the justices transcended this knotty doctrinal point. Joined by Justice Alito, Chief Justice Roberts complained that the Court should not be deciding "which mixes of minority voters should count," particularly since the district court had already deemed the mix in District 25 appropriate. While the Chief Justice then referenced this "sordid business, this divvying us up by race," he found nothing legally wrong with the manner in which Texas divvied up a whole lot of folks by race in District 25 and elsewhere. Indeed, he joined the remarkable portion of Justice Scalia's opinion that deemed District 25 an intentional racial gerrymander that was nevertheless justified, under strict scrutiny, by what Justice Scalia labeled the State's compelling interest in complying with Section 5 of the Voting Rights Act.
Justice Kennedy found it unnecessary to reach this issue, but his discussion of District 25 suggests that he disagreed both that the Voting Rights Act mandated this district and that Congress could constitutionality so require. For Justice Kennedy, the Court had no choice but to require the "rejiggering" of district lines to cure the race essentialism he found manifest in the boundaries of District 25.
Justice Kennedy thought Texas erred, however, not only by aggregating Latino voters in District 25 in the manner that it did; he also found the State's effort to divide Latino voters in Laredo problematic. The reason, he explained, was that Texas fractured Laredo's Latino population to impair the political power of an existing, cohesive, "politically active" and racially-defined community. Doing this, Justice Kennedy concluded, diluted Latino voting strength and, given the inadequacy of District 25, violated Section 2 of the Voting Rights Act.
Underlying this holding is the premise that jurisdictions must respect at least some existing racially-defined communities when they draw district lines. While Justice Kennedy's discussion of District 25 explicitly states that we must "account for the differences between people of the same race," his invalidation of the Laredo district necessarily implies that we respect their similarities as well. Race need not necessarily be a problem to overcome but can instead function as a trait that unites people in positive ways and gives rise to communities that, at least in certain circumstances, warrant our protection.
That, of course, may simply be another way of saying that racial vote dilution is a cognizable harm-hardly a revolutionary proposition. LULAC nevertheless makes clear that Justice Kennedy subscribes to it. So apparently do the Court's two newest justices, both of whom declined to join Justices Scalia and Thomas, who deny that racial vote dilution is ever something courts can effectively remedy. Indeed, the Chief Justice's opinion suggests he would have ordered the reconfiguration of the Laredo district had Texas failed to create what he saw as an adequate substitute in District 25.
To be sure, the Court's invalidation of the Laredo district arguably rests on a narrow reading of the Voting Rights Act. The dispute between Justice Kennedy and the Chief Justice on the role of compactness in District 25 may produce an additional hurdle for Section 2 plaintiffs, although its magnitude is not clear. Of more consequence, LULAC's focus on discriminatory intent suggests the Court's conception of racial vote dilution may be quite restricted. While Section 2 prohibits electoral practices that impair minority political participation regardless of the underlying motivation, Justice Kennedy's opinion specifically noted that the Laredo district "bears the mark of intentional discrimination." If this characterization was critical to the Court's holding, LULAC may in fact do no more than affirm the incontrovertible proposition that the State may not intentionally impair minority political participation.
That LULAC does more, however, is suggested by Justice Kennedy's detailed description of the nature and quality of political participation by Laredo's Latino community and the effect the new boundary had on their ability to participate in the political process. Justice Kennedy depicted a vibrant community, whose members overcame a long history of discrimination and were registering and voting in ever-increasing numbers. He repeatedly described the community as "politically active" and "cohesive," one that was "poised to elect [its] candidate of choice," and one whose mobilization was "made fruitless" by a district line drawn through the middle of Laredo.
Justice Kennedy, moreover, resisted the tendency of many lower court judges to apply narrowly the factors that comprise the Section 2 inquiry. Justice Kennedy did not, for example, discount the importance of past racial discrimination in Texas even though many of the State's Latino residents lived outside the U.S. when much of the discriminatory conduct occurred. Nor did he view increasing rates of Latino voter registration and turnout as a reason to dismiss past discrimination as irrelevant. Justice Kennedy did not discount evidence that voting is racially polarized in southwestern Texas even though partisanship contributed to Anglo support for (and Latino opposition to) Laredo Representative Henry Bonilla. Nor did Justice Kennedy reflexively uphold incumbency protection as a legitimate justification for a challenged district, instead deeming it a tenuous policy choice because its pursuit diluted minority voting strength. Finally, Justice Kennedy labeled Representative Bonilla unresponsive to Latino interests based on the simple fact that Latino voters did not support him, a novel and expansive measure of unresponsiveness that dispenses with the examination of substantive policy and constituent access that has typically prompted lower courts to characterize representatives as responsive.
This represents a vigorous application of Section 2 in Laredo, one that recognizes the right of Latino residents there to mobilize and participate not only as individual voters but as a racially-defined, politically active and cohesive community.
Justice Kennedy seemed more grudging when he assessed a similar, albeit distinct Section 2 claim brought by African-Americans in Fort Worth. His controlling opinion left open the question whether Section 2 protects minority interests in coalition districts, where minority voters comprise less than half the electorate but nevertheless control electoral outcomes. Instead, Justice Kennedy resolved the Section 2 claim by denying that the Fort Worth district dismantled in 2003 had been a true coalition district.
African-American voters in Fort Worth had constituted a majority of the primary electorate in what had been a safely Democratic district. They consistently voted for Martin Frost, an "Anglo Democrat" who repeatedly ran unopposed in the primary and represented the district in Congress for nearly thirty years. Justice Kennedy concluded, however, that these facts did not necessarily make Frost the African-American candidate of choice or establish that black voters controlled his election. After all, the opinion posits, had "an African-American candidate of choice" ever challenged Frost in the primary, white and Latino voters might have participated in greater numbers. The assumption is that Frost would have prevailed in this circumstance, the absence of black support notwithstanding, and consequently, that "Anglo Democrats" necessarily controlled the district.
Improbable perhaps, but this hypothetical scenario nevertheless captures the core reason why we cannot be sure Frost was genuinely minority preferred. The reason is not that he is white, but rather that he was perennially unopposed. The Fort Worth district was not only safely Democratic, it was safely Martin Frost's district, with the absence of competition seeping into the primary itself. And, of course, the district was drawn precisely for that purpose. As Justice Kennedy properly reminds us, Frost himself was the "architect" behind the 1991 Democratic gerrymander to which the Republicans 2003 foray into re-redistricting was largely a response.
In this sense, Martin Frost was the Democratic Henry Bonilla, and the Fort Worth district was the product of the same form of incumbency protection that Justice Kennedy saw as problematic in Laredo. As a matter of formal doctrine, of course, this parallel has no bearing on the question of whether either district gave rise to racial vote dilution. And yet, the role of partisan gerrymandering and specifically incumbency protection in the creation of each district was crucial to Justice Kennedy's analysis of dilution in each location.
Incumbency protection, as pursued in Laredo, destroyed what made the Latino community there worth protecting-hence, Justice Kennedy's repeated references to Laredo's energetic, cohesive, "politically active" Latino community. In Fort Worth, by contrast, incumbency protection prevented the district from becoming a forum in which such an engaged community might emerge. Indeed, propelling Justice Kennedy's skepticism that African-American voters truly preferred Frost is the conviction that the Fort Worth district, created by Martin Frost for Martin Frost, was not an environment in which a meaningful preference for a political candidate could be expressed and, consequently, where a vibrant political community could arise and flourish.
Justice Kennedy deemed the effort to insulate Bonilla flawed because it was meant "to benefit the officeholder, not the voters," and he likely thought Fort Worth was no different. This distinction between benefits to voters and those to officeholders is underdeveloped, but the suggestion seems to be that incumbency protection should promote responsiveness rather than stifle competition. That's an intriguing idea, although an approach more easy to administer and more robust in effect would be to return to the rule set forth in Karcher v. Daggett-the often-cited source of the mantra that incumbency protection is a legitimate districting goal. Karcher itself condoned not the protection of incumbents writ large, but instead the far more restricted desire to draw district lines so as to avoid contests between incumbents.
As formulated in LULAC, Justice Kennedy's distinction between good and bad forms of incumbency protection precludes some manifestations of the practice, at least insofar as they cause racial vote dilution. But the opinion also hints at the idea that these bad manifestations might be unlawful even absent such race-specific harm. Indeed, Justice Kennedy's analysis of racial vote dilution in Laredo and Fort Worth begins to define what might well be the representational harm that partisan gerrymandering produces more generally. When Texas stifled the eager and energized Latino electorate in Laredo, it engaged in a classic form of racial vote dilution. But it also stifled competition, doing harm not just to a racially-informed political community on the cusp of victory, but to all voters in the district for whom vibrant political participation was no longer an option.