Callais: The Good, the Bad, and the Ugly

The Supreme Court’s recent re-interpretation of Section Two of the Voting Rights Act in Louisiana v. Callais has produced a lot of condemnation. As I discussed on the podcast, my view on the legal issues is more mixed, and I thought I would put down a few thoughts since I haven’t seen this written elsewhere.

The Good: Strangely, since we are supposedly living in the age of textualism, I have not seen that much written about whether Callais provides a plausible or even good account of the text of Section Two. In my view, this is actually one of the strongest points for the opinion.

Here is the text of Section Two:

(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 10303(f)(2) of this title, as provided in subsection (b).

(b) A violation of subsection (a) 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) 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.

That is a big wall of text. But some key phrases that emerge are that this is a test based on the “results” of the state voting law (rather than its intent), where "results” are then specified to ask whether the processes leading to the primary or election are “equally open to participation by members of a class of citizens protected by subsection (a).” And equal-openness is in turn specified as asking whether members of that class “have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.”

So this is a comparative results test — there is some baseline opportunity to participate and elect chosen representatives, and the statute requires protected groups to have at least those same opportunities.

Of course, comparative results tests always prompt the question: what is the baseline? It is a famous critique of such tests that there is no a priori, naturally-given baseline. A baseline must be chosen and defended.

One possible baseline would be proportional representation — members of a protected class should presumptively elect candidates in proportion to their population in the state. This may be somewhat intuitive, but it does run into the final proviso: "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.”

Now there are still ways to try to reconcile the provision and the proportionality principle (for instance the proviso speaks about the race of the elected officials rather than the race of the voters), but in any event it is worth asking whether there are other sensible baselines.

Another possible baseline would be a “similarly-situated” analysis. It would ask whether members of the protected class have the same opportunities as other citizens who share similar characteristics except for race. If so, then any negative “results” would not be “on account of” race, but on some other account.

An analysis like this is especially potent because people vote for the losing side in elections all the time — sometimes quite persistently — and sometimes for reasons that do not make their losses unlawful. It might be that the person has an unpopular view like libertarianism; it might be that the view is popular but that other people who share the view are distributed or concentrated in various ways; and so on. The question is how to read the VRA to isolate whether being on the losing side is the result of race.

A version of this baseline was described almost twenty years ago in a Seventh Circuit opinion written by Judge Frank Easterbrook. And it is a version of this baseline that the Court seems to adopt in Callais. And it seems to me to be a natural way to understand the text of Section Two of the VRA.

The best objection to this bottom line, I think, comes not from the text of the VRA but from the four decades of precedents interpreting it. Stare decisis in such areas is always fraught, but throughout those four decades other areas of the Court’s public law jurisprudence have been trending more and more towards a general principle of government colorblindness. Indeed, election law scholars have pointed out for decades that election law was anomalous in being an exception to this general principle. Callais goes some distance toward resolving the anomaly.

The Bad: Despite the defensible textual interpretation at the heart of Callais, the opinion is quite confusingly and strangely drafted in many other respects. To name just a few:

• The opinion implausibly and confusingly claims not to overrule Allen v. Milligan, even though more or less the same basic claim was made in that case, just in a much more confusing fashion. The subsequent GVR in Allen v. Caster exacerbates this problem.

• The opinion can be read to say that the discriminatory results test is really just a way of looking for an ultimate form of discriminatory intent. It can also be read not to say that. In the part of the opinion that responds to this charge from the dissent, the Court is so ambiguous that I suspect the ambiguity is intentional.

• The opinion is confusing in its gloss of the Shaw v. Reno standard. The cases had previously said that race discrimination in redistricting provoked strict scrutiny only if it “predominated.” Callais is ambiguous about whether it retains the predominance test.

• The opinion is also somewhat confusing in its treatment of the question of whether adherence to Section Two of the VRA constitutes a compelling interest that can justify race discrimination. It says several times that in previous cases this interest was assumed and that in this case they will finally have to decide whether it is a compelling interest. But it is hard to find the part of the opinion that actually does decide it, although the opinion says it decides it.

• Finally, this is the least of the issues, but the opinion is also somewhat confusing in its gloss on the strict scrutiny precedents. As of Students for Fair Admissions, the Court said that there were three compelling interests that could justify race discrimination under strict scrutiny — diversity in higher-education, remediating specific instances of illegal discrimination, and avoiding imminent risks to human safety. In Callais the Court reduced this list to the last two (remediation and safety) before adding adherence to the (properly interpreted) VRA. This would make sense if SFFA had held that diversity in higher education is not a compelling interest and that Grutter, Fisher, etc. were thereby overruled. But SFFA did not say that, at least in so many words. It would also make sense if one thinks that SFFA must be read to say that implicitly, and thus the Court is now making explicit what SFFA left implicit. But Callais simply does not address this, making it unclear whether it is implicitly endorsing this implicit reading, or just unaware of the ambiguities in the case law.

None of these points necessarily undermines the core textual analysis in Callais, but on the other hand, none of them is necessary to Callais’s holding either. They seem like unforced errors.

The Ugly: But in my view the most troubling thing about Callais is none of the above. It is the Court’s treatment of the role of partisanship in districting, and of the ability of avowed partisan gerrymandering to defeat any claim of racial gerrymandering. It is one thing to say that one must control for compactness, geographic distribution, etc. in deciding whether there has been a discriminatory effect under the VRA. But to say that one must also control for partisan goals may allow the dismantling of every blue majority-minority district in a red state.

There are many varieties of partisan gerrymandering — some focus on incumbent protection, some protect seats with a high margin of safety, some accept a lower margin of safety in exchange for a chance at more seats. This gives gerrymanderers many degrees of freedom in justifying a map against a VRA challenge. Additionally, many have pointed out that voters choose parties in part for reasons related to race. This makes the opposition of partisan gerrymandering to racial effect potentially artificial.

To be sure, there is some logic to Callais’s treatment of partisan gerrymandering. If districting for partisan purposes is just as permissible and just as legal as districting for any other traditional purpose, then the similarly-situated analysis would seem to extend to partisanship. And Rucho v. Common Cause famously (or infamously) eliminated constitutional claims against partisan gerrymandering.

But Rucho did not quite say that districting for partisan purposes is just as permissible and just as legal as districting for any other traditional purposes. It held that there were no justiciable objections to those partisan purposes — which is not the same thing as saying that they are totally cool and totally lawful. The slide from saying that partisan gerrymandering is non-justiciable to saying that it can defeat other justiciable challenges is seemingly technical, but it supercharges the effect of Callais.

Now that slide may have been an inevitable consequence of Rucho. There is not an easy space in constitutional doctrine for the category: “this purpose in no way invalidates legislative action but cannot be used to justify legislative action.” And the slide did not begin with Callais — Justice Alito’s 2024 opinion for the Court in Alexander v SC NAACP began with a similar claim about the lawfulness of partisan gerrymandering and its ability to rebut claims of racial gerrymandering.

So while there is something very disquieting about extending Rucho in this way, maybe there is no alternative if Rucho is correct. It does highlight, perhaps, the appeal of the old regime managed by Justice Kennedy under Vieth, in which partisan gerrymandering claims would never in practice succeed, but were never ruled out in theory. While intellectually unsatisfying and maddeningly uncertain, that regime had the advantage of keeping partisan gerrymandering in the “not-invalidating-but-not-justifying” box.

One other alternative to the current combination would have been to say that while partisan gerrymandering is not a forbidden criteria, it also does not count as a “traditional” districting criterion under the gerrymandering cases. Note, for example, that the Easterbrook opinion linked above, written pre-Rucho, did not describe partisan gerrymandering as one of the possible districting criteria one should control for. The summary-affirmance in Cox v. Larios could also be read to stand for the not-justifying-even-if-not-forbidden proposition.

It could be that the combination of Callais and supercharged partisan gerrymandering is legally correct. But it is not a pretty sight.