Justice, Democracy, and Law is a recurring series by Edward B. Foley that focuses on election law and the relationship of law and democracy.
Last week, the Supreme Court announced its decision in Louisiana v. Callais, striking down the state’s congressional map as an unconstitutional racial gerrymander, and – in doing so – obliterated the commitment of the Voting Rights Act to racial equality in elections. Although I had several immediate reactions to the decision, I wanted to allow for some time to reflect upon and absorb the court’s ruling. But my bottom line has changed very little from my first impression of Callais: It is a singularly horrendous decision, which misunderstands what the Constitution requires in this context and directly contradicts the text and purpose of the VRA.
***
Justice Elena Kagan’s dissent in Callais describes the court’s decision as the third of a trilogy that eviscerates the VRA, the first two cases being 2013’s Shelby County v. Holder and 2021’s Brnovich v. DNC. But lumping Callais with these other two cases risks concealing how truly egregious Callais is compared to the other two.
Unlike many other election law scholars, I did not find either Shelby County or Brnovich problematic. Shelby County, you will recall, is the 2013 case where the court invalidated the formula that Congress used to determine which states and localities were subject to the so-called “preclearance” provision in Section 5 of the VRA – meaning that these covered jurisdictions were required to submit to the federal Department of Justice or a federal court any proposed changes to their election laws and procedures. The coverage formula, as the court ruled, was indeed woefully out-of-date (having been passed in 1965 and not revised since 1975), with Congress failing to update it at all when reenacting the preclearance provision in 2006. What is more, the court had warned Congress in a 2009 case, NAMUDNO v. Holder, that the court likely would find the coverage formula unconstitutional unless Congress revised it. But Congress refused to heed the court’s warning.
Moreover, the reasoning of the Shelby County majority opinion was defensible. Preclearance is an especially intrusive burden on state sovereignty: ordinarily sovereign states are not required to get preapproval for the laws that they enact. If an enacted law is invalid, its enforcement can be enjoined subsequently; that’s the way the system generally works. Making the states plead in advance “Mother, may I …” to the federal government is out-of-sync with the principles of federalism upon which the United States is founded.
Even more significantly, this arduous preclearance obligation was selective, applying to only some states and not others – that selectivity, after all, was the whole point of the coverage formula. The court was willing to tolerate that selectivity, but only if it was justified for truly problematic state actors. In this respect, the court relied on a kind of equal-protection principle applicable to the states themselves – the states in the union deserve to be treated equally by Congress unless there is a good reason for treating them differently. Many have criticized the court for articulating this equal-treatment-of-states principle, but it is well-grounded in the text and structure of the Constitution, which requires all new states to enter the union on equal terms with existing states. Likewise, the unamendable requirement that each state have the same number of senators (unless a state consents otherwise) signals a core constitutional principle, whether one likes it or not, that states are equal in their sovereignty.
Thus, it was appropriate for the court – especially after its warning – to hold in Shelby County that Congress could not keep imposing preclearance on some states and not others without reconsidering its basis for determining which states were, and were not, subjected to this imposition. After all, by 2013, many uncovered states – like my own state of Ohio – were imposing just as onerous voting restrictions, such as new voter ID requirements, as the states of the former Confederacy.
Brnovich was an even easier decision to defend. That case involved a challenge under Section 2 of the VRA to two Arizona voting rules: first, the obligation of voters living in some Arizona counties who choose to vote in person (rather than by mail) to cast their ballots in their neighborhood precinct rather than some other polling location; and second, the prohibition on a voter’s completed absentee ballot to be mailed or otherwise delivered by anyone other than the voter, a family member, or certain other very narrow categories of individuals. The claim in the case was that these voting rules had the effect of discriminating on the basis of race against minority voters and thus violated the “results” provision of VRA’s Section 2 (which we shall shortly return to), which prohibits any “voting qualification” or “standard, practice, or procedure” that “results in a denial or abridgement of the right of any citizen of the United States to vote on account of race.”
This claim was extremely weak, and the plaintiffs and their attorneys were urged by voting rights groups not to bring such a weak claim to the Supreme Court. The evidence showed that at most only a tiny fraction of all voters, regardless of race, were adversely affected by either rule. To be sure, the miniscule percentages of minority voters harmed by the precinct rule were slightly higher than the comparable percentage of white voters harmed, but this disparity could not be enough to establish that the rules operated as a denial of equal “opportunity” regardless of race “to participate in the political process,” as required by the relevant language of the statute.
And the evidence of any racial disparity caused by the restriction on transmitting absentee ballots was literally nonexistent. As the majority in Brnovich noted, “The plaintiffs were unable to provide statistical evidence showing that [this restriction] had a disparate impact on minority voters.” They argued instead that minority voters used “third-party ballot collection” more often than non-minority voters did. But differential usage rates on a method of delivering ballots, as the court explained, were not enough to conclude that the rule “results in less opportunity to participate in the political process.”
More generally, the court’s opinion in Brnovich was a reasonable response to a statute that was exceptionally unclear in relevant respects. If Section 2 of the VRA had been construed to prohibit any voting rule or procedure that resulted in any differential rates of electoral participation by minority and minority voters, virtually all voting rules and procedures would be vulnerable to invalidation, and that could not have been what Congress intended. For example, suppose it could be shown that keeping the polls open one hour longer on election day would increase the rate of minority voter turnout more than white voter turnout. Does that mean that closing the polls one hour earlier – perhaps reverting to an earlier schedule that had been expanded subsequently – violates Section 2 of the VRA? Not necessarily.
As Alito sensibly wrote for the court in Brnovich, “the degree to which a voting rule departs from what was standard practice when § 2 was amended in 1982 is a relevant consideration.” Because all voting rules are burdensome to at least some extent, Alito explained, “burdens associated with the rules in widespread use when § 2 was adopted are therefore useful in gauging whether the burdens imposed by a challenged rule are sufficient to prevent voting from being equally ‘open’ or furnishing an equal ‘opportunity’ to vote in the sense meant by § 2.” Based on the considerations described above, Alito reasonably found “the modest evidence of racially disparate burdens caused by [Arizona law], in light of the State’s justifications, [compels] the conclusion that the law does not violate §2 of the VRA.”
***
Alito’s opinion for the court in Callais is an altogether different matter. Unlike Shelby County or Brnovich, Callais is an abomination.
Callais purports to interpret VRA’s Section 2, but it destroys the central meaning of the section, converting it into the exact opposite of what Congress meant for it to do. The one thing that is unambiguous about Section 2 is that the 1982 amendment to the section’s text creates a “results” test for determining whether there is liability under the section, replacing the “intent” test that the Supreme Court had previously adopted for Section 2 claims. As the text states, no “standard, practice, or procedure shall be imposed … which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race.” Yet Callais defiantly converts Section 2 back to an intent inquiry rather than a results analysis.
In various passages in the Callais majority opinion, Alito denies substituting an intent test for the statute’s explicit “results” standard, saying that evidence of intent is only to be considered as relevant to the results determination. Specifically, he says that the court’s “interpretation does not demand a finding of intentional discrimination,” while acknowledging that “it imposes liability only when the circumstances give rise to a strong inference that intentional discrimination occurred.” But the bulk of Alito’s opinion belies that the consideration of intent is at all limited in the way he suggests. Instead, it is manifest that intent has become the touchstone of the entire Section 2 inquiry. When discussing the specific facts of the case, Alito faults the Section 2 claimants for “fail[ing] to show an objective likelihood of intentional discrimination based on the totality of circumstances.” Moreover, when setting forth the threshold perquisites that any Section 2 plaintiff must establish before having any chance of prevailing, Alito categorically states that plaintiffs must “demonstrate that the State’s chosen map was driven by racial considerations rather than permissible aims.” It’s undeniable that “driven by racial considerations” is an intent, not results, requirement.
What is worse, Alito sloppily conflates proof of a statutory violation under Section 2 with proof of a constitutional violation under the equal protection clause of the 14th Amendment. Since the 1982 amendment to Section 2, the court has considered claims brought under that section according to an analysis first set forth in a 1986 case called Thornburg v. Gingles. That analysis, which determines whether a legislative map “results” in minority vote dilution, is very different from a separate analysis of intentional racial gerrymandering. Yet, when discussing the Gingles analysis in Callais, Alito asserts that in attempting to show minority vote dilution with “illustrative maps” containing “their desired number of majority-minority districts,” “plaintiffs cannot use race as a districting criterion” because, in Alito’s words (with his own emphasis) “that illustrative map sheds no light on whether the State acted unconstitutionally by not adopting such a map.”
But the issue under Section 2 is not “whether the State acted unconstitutionally” – it’s whether the state violated the statute by failing to adopt a map with an additional district that enables minority voters, in the statute’s words, “to elect representatives of their choice.” This passage in the Callais majority opinion indicates that Alito, and thus the majority of the court, completely lost sight of the actual provision that the opinion purports to interpret.
Even if Callais could be considered as refining a “results” test for Section 2, it’s a test that is entirely inconsistent with the reason why this section was amended in 1982. The amendment was based on the recognition that a legislative districting map may be motivated by wholly valid considerations and still cause minority vote dilution that needs to be undone in order for there to be fair representation regardless of race. For example, districts may be drawn to keep cities and counties intact, or to make the shape of districts compact, or to protect incumbent members of Congress that advantage their constituents with their seniority. But the consequence of these valid considerations may nonetheless be that minority voters have “less opportunity than other members of the electorate to participate in the political process and elect representatives of their choice,” in which case Section 2 requires altering the map to remedy this “violation” of its prohibition against the “denial or abridgement of the right of any citizen of the United State to vote on account of race.” The court’s opinion in Callais, however, obliterates this obligation, allowing states to avoid any Section 2 liability as long as they base their districting on permissible nonracial factors.
Indeed, the court in Callais is explicit on this point, touting that its “interpretation” of Section 2 “does not intrude on States’ prerogative to draw districts based on nonracial factors.” And the court makes a mockery of the VRA by repeatedly proclaiming, based on the 2019 case of Rucho v. Common Cause, that the goal of partisan vote dilution – endeavoring to give members of one political party less opportunity to elect candidates of their choice than members of another party – is a perfectly permissible nonracial factor. For instance, the court asserts that “in considering the constitutionality of a districting scheme, courts must treat partisan advantage like any other race-neutral aim: a constitutionally permissible criterion that States may rely on as desired.”
And then comes the kicker: Only if a Section 2 plaintiff provides “an alternative map that achieves all the State’s objectives—including partisan advantage and any of the State’s other political goals—at least as well as the State’s map,” might the plaintiff be able to prevail. But as a practical matter that will be an impossible burden for a plaintiff to sustain – and one that altogether undermines the purpose of the 1982 amendment – because, following the court’s decision in Rucho, a state can simply justify its minority vote dilution as partisan vote dilution. (Put concretely: because Black voters overwhelmingly vote for Democrats, states can camouflage the disempowerment of Black voters as a desire to disempower Democrats.) In this way, Callais renders Section 2 a nullity.
***
Finally, the majority opinion in Callais makes little sense as a matter of constitutional law. Alito is candid that his purported “interpretation” of Section 2 is based on a desire to avoid rendering the provision unconstitutional. This issue arises from the source of Congress’ authority to enact the VRA’s prohibition against minority vote dilution in the first place. Alito treats this as a 15th Amendment issue, viewing that provision of the Constitution as the only one on which Congress may rely.
The canon of constitutional avoidance – that is, to avoid striking a statute as unconstitutional when another plausible reading can uphold it – is a longstanding and salutary principle of statutory interpretation. But the canon has its limits. Indeed, it was Alito himself who said for the court (in Jenning v. Rodriquez) that “spotting a constitutional issue does not give a court the authority to rewrite a statute as it pleases.” Yet that is exactly what he does for the court in Callais.
According to the 15th Amendment, the ability to “vote shall not be denied or abridged by the United States or by any State on account of race.” As Alito correctly observes, the court has long construed this constitutional language to prohibit only intentional discrimination and not voting rules and procedures that have a racially discriminatory effect. But the court also has long held that Congress, pursuant to its explicitly granted “power to enforce” the amendment “by appropriate legislation,” can enact statutes that go beyond just the elimination of intentional discrimination and instead outlaw results-based discrimination as well. Alito contradicts this longstanding precedent when he asserts that “a law that seeks to enforce the Fifteenth Amendment by prohibiting mere disparate impact would fail to enforce a right that the Amendment secures” and thus be beyond the scope of Congress’ power to enact “appropriate legislation” pursuant to the amendment.
The precedent that clearly established the point that Alito flagrantly contradicts is a unanimous 1976 case, Fitzpatrick v. Bitzer, involving Title VII of the Civil Rights Act of 1964, which prohibits both intentional discrimination and disparate impacts in employment. There, the court had no difficulty concluding that Congress is empowered to impose both types of Title VII liability (intentional and results based) as a means of enforcing the 14th Amendment’s antidiscrimination provision. It would be truly bizarre if Congress had less expansive enforcement power under the 15th Amendment than the 14th, since racial discrimination with respect to voting is an especially pernicious form of discrimination – which is why it is singled out for eradication in the 15th amendment. But Alito’s pronouncement that Congress is powerless to prohibit “mere disparate impact” either implies this inferiority of congressional authority under the 15th amendment relative to the 14th – or else, more ominously, means that the 50-year-old precedent of Fitzpatrick v. Bitzer is no longer good law.
Either way, Alito’s pronouncement is incredibly unsettling – and unwarranted. The court in Callais did not need to consider the question of congressional power to enforce the 15th Amendment. That is because the power of Congress to enact Section 2 of the VRA for the purposes of Callais could have been sustained, not under the 15th Amendment, but under Article I, Section 4 of the Constitution. Congress has full power under Article I, Section 4 to enact laws governing the “time, place, and manner” of congressional elections. Thus, Congress can enact a prohibition against minority vote dilution for congressional districts under a disparate impact theory, without any consideration of discriminatory intent, and not rely on the 15th Amendment at all.
To be sure, for Congress to prohibit minority vote dilution in the districting of state legislatures or city councils requires reliance on the 15th Amendment (since Article I, Section 4 applies only to congressional elections), but Callais did not involve districts for state legislatures or city councils. Because the Supreme Court is supposed to sustain the constitutionality of a congressional statute if Congress has any source of authority to enact it, the court in Callais should have viewed the VRA’s Section 2 as an exercise of power under Article I, Section 4, and set aside any issues involving congressional power under the 15th Amendment for a future case not involving congressional districts.
If the court in Callais had gone that route, it would then have confronted a different constitutional question. As Alito’s opinion emphasizes from start to finish, looming over Callais was the court’s belief in “the general rule that the Constitution almost never permits the Federal Government or a State to discriminate on the basis of race,” which according to the court encompasses “allowing race to play any part in government decisionmaking.” Thus, if the effect of minority vote dilution caused a map to violate Section 2 of the VRA, the court feared, as happened in Louisiana, that the state would be required to intentionally create a new district based on race and that doing so would “force[]” the state “to engage in the very race-based discrimination that the Constitution forbids.” Treating Section 2 of the VRA as legislation empowered by Article I, Section 4, in other words, would have required the court to consider another application of the canon of constitutional avoidance: namely, whether Section 2 could and should be construed to avoid running afoul of the Constitution’s paramount prohibition, according to the court, against any government decisions based on race.
But if the court had entertained this separate constitutional avoidance question, it could have done much less damage to Section 2. The constitutional prohibition on race-based decisions by government actors implicates the remedies that may be imposed for a finding of minority vote dilution. It does not implicate the substantive liability standard, which is to forbid maps that “result” in minority vote dilution. The grave constitutional concern, as Alito sees it, is having to draw a map based on race in order to undo the minority vote dilution. But that grave constitutional concern is about the remedy, not the finding of minority vote dilution itself.
Moreover, it is possible to remedy minority vote dilution without ordering government officials to engage in new race-based districting. (One way is to have the voters themselves choose which congressional districts they wish to join, a system that I have called “self-districting.”) Thus, as an exercise of constitutional avoidance, the court could have disallowed race-based districting as a remedy for a Section 2 violation, but still left untouched the substantive standard of liability for minority vote dilution under it.
But the court did not do that. Instead, by invoking the 15th Amendment as its basis for constitutional avoidance, the court gutted Section 2. This evisceration of Section 2’s fundamental purpose in protecting minority voting rights was totally gratuitous – and thus a travesty of the judicial imperative to effectuate the will of Congress embodied in the text of the statute as much as possible.
For all these reasons, Callais – in sharp contrast to Shelby County and Brnovich – is an indefensible act of judicial overreach. That the court unnecessarily destroyed one of the most important congressional achievements in American history is all the more unconscionable. The only consolation is that there are still race-neutral ways to eradicate congressional districting that unintentionally results in minority vote dilution. But those ways will now have to come in the form of state laws or a new congressional enactment explicitly pursuant to Article I, Section 4, as unlikely as those measures are to occur in the near term.