Narrow decisions, broad goals

Why the South Carolina redistricting case was decided 6 to 3.

In Alexander v. South Carolina State Conference of the NAACP, the Republicans beat the Democrats.

I generally find it crude to characterize hotly contested Supreme Court cases as pitting Republicans against Democrats. But in this case, the characterization is apt.

In 2018, Democrat Joe Cunningham was elected to represent South Carolina’s First Congressional District in an upset win over Republican Katie Harrington, taking 50.6% of the vote. In 2020, Republican Nancy Mace barely defeated Cunningham with 50.6% of the vote—too close for comfort.

After the 2020 census, South Carolina engaged in redistricting. The South Carolina legislature, controlled by Republicans, moved several heavily Black neighborhoods from District 1 to District 6, which was already a predominantly Black district that was guaranteed to vote for a Democrat. This transformed District 1 from a swing district into a easy Republican hold.

Black voters sued, alleging that the district lines had been drawn on the basis of race. The state legislature’s defense was that the lines were drawn based on partisanship, not race—it claimed that the Black voters were moved because those voters were Democrats, and it wanted to gerrymander District 1 in favor of the Republican candidate.

This defense prevailed. The Supreme Court was persuaded that the legislature was trying to ensure that Republicans would win District 1. As a result, the map will stay in effect, so Republicans will, in fact, win District 1. So, yes, the Republicans won this case.

The vote in Alexander was 6 to 3. The majority opinion was written by Justice Alito and joined by the Chief Justice and Justices Thomas (in part), Gorsuch, Kavanaugh, and Barrett. The dissent was written by Justice Kagan and joined by Justices Sotomayor and Jackson.

As I began studying this decision, I decided to look up the Justices’ biographies on Wikipedia, and I noticed a weird coincidence. The six Justices in the majority were all appointed by Presidents from the Republican Party. The three Justices in the dissent were all appointed by Presidents from the Democratic Party. Seriously; you can check the Wikipedia articles yourself if you don’t believe me.

Why did this happen?

Well, two hypotheses immediately come to mind.

1. Idealistic hypothesis: The dueling opinions reflect good-faith philosophical disagreements. The majority opinion applied conservative jurisprudential principles, while the dissent applied progressive jurisprudential principles. Republican appointees tend to be judicial conservatives while Democratic appointees tend to be judicial progressives, so it’s no surprise that the opinion was 6 to 3.

2. Cynical hypothesis: The Justices are partisan. Republican appointees cast votes to help Republicans, while Democratic appointees cast votes to help Democrats.

Both hypotheses are wrong.

Hypothesis #1 is wrong because there’s nothing particularly conservative about the reasoning in the majority opinion, nor anything particularly progressive about the reasoning in the dissent.

Hypothesis #2 is wrong because it’s just wrong. If you think that all nine Justices are political operatives, you do not understand the Supreme Court.

So what happened in Alexander? Here’s what I think.

• Alexander was a close case on the merits: the plaintiffs brought a marginal racial-gerrymandering case, but they won anyway in the district court, and the Supreme Court had to decide whether the deferential “clear error” standard of review canceled out the weaknesses in the plaintiffs’ case. It could have gone either way.

• There were no high-minded philosophical debates. Instead, the case boiled down to a series of sharp, but non-philosophical, disputes about the interpretation of the evidentiary record and of ambiguous statements in prior cases.

• In resolving those disputes in the State’s favor, the majority was influenced by its broader project of preventing litigants from using racial-gerrymandering claims as an end run around its recent decision barring partisan-gerrymandering claims.

• This style of judicial reasoning should be accepted and expected, but not celebrated.

It might be wrong, but is it clearly wrong?

The Alexander plaintiffs alleged that South Carolina’s congressional map was an illegal racial gerrymander. To prevail on that claim, the plaintiffs bore the burden of showing that race was the “predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.”

There was no dispute that large numbers of Black voters were moved from District 1 to District 6. The case boiled down to whether those voters were moved because they were Black, or because they were Democrats.

The legislature argued that it was trying to ensure that District 1 would be reliably Republican during the 2020s. The plaintiffs didn’t dispute that this was the legislature’s goal. Rather, their theory was that the legislature relied on race in order to effectuate that goal. The map-makers had access to voting data from the 2020 election, but the 2020 election was weird, with Trump, COVID, and everything else. According to the plaintiffs, the map-makers drew maps based on their predictions of voters’ behavior during congressional elections in the 2020s, but concluded that “is the voter Black?” was a better predictor than “who did the voter vote for in the 2020 presidential election?”

The plaintiffs’ case was an uphill battle for three reasons.

1. There’s case law stating that legislatures are presumed to act in good faith. If you’re going to argue that state legislatures are both racists AND liars, you better have good evidence.

2. The guy who actually drew the maps fervently denied that he drew the district boundaries based on race.

3. The simplest way to show that the legislature used race rather than partisanship would have been to offer an alternative map. Then, the plaintiffs could have argued: “if the legislature was really using partisan data rather than racial data, as they’re currently claiming, it would have looked like this.” However, the plaintiffs did not provide such a map.

Still, the plaintiffs won in the district court, essentially for the following reasons:

1. The map-maker guy had racial data on his computer, and the district court didn’t believe him when he said he didn’t use it.

2. There was evidence that the percentage of Black voters in District 1 held steady throughout the map-making process, which the district court construed as evidence that the map-maker was targeting a particular racial composition.

3. The plaintiffs put forth expert witnesses who conducted studies purporting to show that the legislature relied on race rather than partisanship. These studies weren’t perfect, but the defense’s expert didn’t do a good job in pointing out those imperfections.

Not the greatest case, but it was enough to persuade the district court. Crucially, the district court made factual findings that favored the plaintiffs—it found, based on the trial record, that the district lines were drawn for predominantly racial reasons. As a result, the Supreme Court applied the “clear error” standard of review to those findings—a highly deferential standard. This made the case difficult: the plaintiffs’ case had flaws, but were those flaws of sufficient magnitude to overcome the deferential standard of review?

The majority held that the plaintiffs’ case was so weak that the district court’s factual findings were clearly erroneous. The majority emphasized that there’s a presumption of good faith afforded to state legislatures. It found that the district court should have drawn an adverse inference from the plaintiffs’ failure to offer an alternative map. It also conducted a microscopic analysis of the plaintiffs’ expert testimony and concluded that it was unreliable. The dissent would have deferred to the district court’s factual findings.

Let me pause and offer an observation. Although the majority and the dissent disagreed on pretty much everything, there was no philosophical disagreement. They both applied the same standard—race can’t be the “predominant factor”—and except for Justice Thomas’s solo concurrence, no one questioned that standard.

The disagreements were lawyerly in nature. For example:

• The majority thought that an adverse inference could be drawn from the plaintiffs’ failure to provide an alternative map, while the dissent didn’t think this.

• The plaintiff’s expert conducted a statistical analysis and concluded that the map-drawer moved Black Democratic voters rather than white Democratic voters into District 6. The majority thought that the plaintiffs’ expert didn’t adequately consider whether moving those Black voters (as opposed to white Democratic voters) was necessary to ensure that District 6 would remain contiguous and compact. The dissent thought that controlling for contiguity and compactness wasn’t necessary because “District 1 was a narrow strip along the Atlantic coast,” so “[n]early everyone within District 1 lived close to the border line,” meaning that “nearly everyone could have been sent to District 6, consistent with contiguity and compactness.”

• The majority thought the plaintiffs should have known throughout the case that the defendants’ defense would be that they were pursuing a partisan gerrymander. The dissent thought that the plaintiffs didn’t have adequate notice of this point until the trial.

None of these disagreements reflects differences in judicial philosophy. It’s not like one can say “because I am an originalist, I conclude that the plaintiffs’ expert didn’t adequately control for contiguity” or “the living Constitution requires to us to recognize that, as a result of District 1’s geography, it’s not necessary to control for contiguity.” Something else is going on.

(Dall-E sometimes makes spelling errors. Close enough.)

Party like it’s 1997

As lawyers love to say in briefs, any doubts from the text are resolved by history.

In 1997, North Carolina drew a new map for its 12th Congressional District. (There were actually four (!!) Supreme Court cases about this one district during the 1990s / early 2000s, but I’ll skip to the punch line.)

Litigation erupted that, other than the difference in Carolina, was remarkably similar to the litigation in Alexander. Lots of Black voters were moved to District 12, and a lawsuit erupted, alleging that the district lines were an illegal racial gerrymander. The legislature defended the map on the ground that it was drawn for partisan, rather than racial reasons. A federal district court held that the map was an unconstitutional racial gerrymander.

In Easley v. Cromartie, 532 U.S. 234 (2001) (this case is conventionally referred to as “Cromartie II” so I’ll call it that) the Supreme Court reversed in a bitterly divided 5-4 decision. The Court held that the district court’s factual findings were clearly erroneous. It found there was inadequate support for the plaintiffs’ theory that the legislature had racial rather than partisan motives.

The Court emphasized that “courts must ‘exercise extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race’”—emphasis in original—especially where “the State has articulated a legitimate political explanation for its districting decision, and the voting population is one in which race and political affiliation are highly correlated.” The Court held that in such cases, “the party attacking the legislatively drawn boundaries must show at the least that the legislature could have achieved its legitimate political objectives in alternative ways that are comparably consistent with traditional districting principles. That party must also show that those districting alternatives would have brought about significantly greater racial balance.” The Court conducted a microscopic analysis of the plaintiffs’ expert testimony and concluded that it was unreliable. The dissent would have deferred to the district court’s factual findings.

Cromartie II is a virtual carbon copy of Alexander. So I guess nothing has changed. Except … Cromartie II was written by Justice Breyer, joined by Justices Stevens, O’Connor, Souter, and Ginsburg. The dissent was written by Justice Thomas, joined by Chief Justice Rehnquist and Justices Scalia and Kennedy. Wait, what?

As it turns out, moving blocs of Black voters can be helpful to either party, depending on who is drawing the maps. When Black voters are moved from a swing district to a landslide district—the situation in South Carolina in 2021—this is helpful to Republicans. When Black voters are moved from a landslide district to a swing district—the situation in North Carolina in 1997—this is helpful to Democrats. In Cromartie II , the Supreme Court was persuaded that the legislature’s goal was to gerrymander the map in favor of the Democrats, and so the Democrats’ gerrymandered map was permitted to go into effect, over the dissent of the conservative wing of the Court.

Fast-forward to Cooper v. Harris, 581 U. S. 285 (2017), which again involved North Carolina’s District 12. The Republican legislature decided to move Black voters into District 12, which was already solidly Democratic, leading to “packing” accusations. So this time, it was Democratic voters alleging a racial gerrymander.

Like in Cromartie II, the plaintiffs didn’t proffer an alternative map, but the district court nonetheless found a racial gerrymander. This time, the Supreme Court affirmed the district court’s decision, holding that the finding of a racial gerrymander wasn’t clearly erroneous. It held that an alternative map wasn’t needed because there was sufficient other evidence of racial motivation.

Of the eight Justices who heard the case (the case was argued prior to Justice Gorsuch’s appointment), seven voted as you’d expect, with the sole exception being Justice Thomas. So, the majority opinion was written by Justice Kagan and joined by Justices Thomas, Ginsburg, Breyer, and Sotomayor, while the dissent was written by Justice Alito and joined by the Chief Justice and Justice Kennedy. The dissent started this way: “A precedent of this Court should not be treated like a disposable household item—say, a paper plate or napkin— to be used once and then tossed in the trash.” This was a reference to Cromartie II’s statement that an alternative map is generally needed to establish a racial gerrymandering claim.

In Alexander, Justice Alito wrote the majority opinion and Justice Kagan wrote the dissent. The majority held that an alternative map generally is needed after all. The dissent accused the majority, in essence, of transforming Cooper into a paper plate or napkin. Justice Thomas—who voted to uphold the finding of a racial gerrymander in both Cromartie II and Cooper—clearly has had enough of these exercises in legislative mind-reading. In a characteristically thoughtful concurrence, he argued that the Court should get out of this business altogether.

What can we take from all this?

Well, as a first cut, I think this history confirms that the specific disagreements between the majority and the dissent in Alexander are not rooted in judicial philosophy. There is nothing inherently progressive or conservative about deference, or lack thereof, to factual findings in racial gerrymandering cases.

Context Matters

OK, so what are these disagreements rooted in? Why were the conservative appointees so enthusiastic about rigorous appellate review in Alexander, when it was the progressive appointees that were so enthusiastic about rigorous appellate review in Cromartie II?

I don’t buy the explanation of “conservative judges help Republicans, progressive judges help Democrats.” There’s an explanation that’s both less cynical and more logical.

Alexander was decided in the wake of Rucho v. Common Cause, 588 U. S. 684, 721 (2019), which held that partisan gerrymandering claims are non-justiciable. When it came out, Rucho was perhaps the single most disliked ruling ever among progressive Court-watchers, although Dobbs has lapped the field. If there is ever again a majority of Democratic appointees on the Court, I suspect that very few opinions from the current Court will be overruled (as opposed to read narrowly), but Rucho will be overruled at the first opportunity.

But Rucho is on the books and the current Court is committed to it. In Alexander, the Court explained: “What the court did— inferring bad faith based on the racial effects of a political gerrymander in a jurisdiction in which race and partisan preference are very closely correlated—would, if accepted, provide a convenient way for future litigants and lower courts to sidestep our holding in Rucho that partisan gerrymandering claims are not justiciable in federal court.” Thus, the majority’s granular review of the factual record was in service of its goal of shoring up Rucho. The dissent, needless to say, did not share that goal.

Cromartie II arose in a very different legal context. Unlike in Alexander, the Court’s conservatives were interested in shoring up a prior decision that struck down a map. To make a long story short, in a previous iteration of North Carolina District 12, the boundaries were intentionally drawn in order to ensure a majority-Black district. In Shaw v. Hunt, 517 U.S. 899 (1996), the Supreme Court held, by a 5-4 vote with the conservatives in the majority, that the map violated the Equal Protection Clause. The district was then re-drawn to look similar—but not identical—to the unconstitutional version of District 12. Cromartie II was a challenge to the new map.

The Court’s conservatives undoubtedly perceived the new map as an end-run around its decision striking down the old map. So of course they voted to uphold a district court decision finding the new map to be a racial gerrymander. Meanwhile, the Court’s progressives were perfectly happy to limit the effect of Shaw.

The iron law of law

This explanation isn’t fully satisfying.

It seems that in both Cromartie II and Alexander, the Justices’ application of the clear-error standard was influenced by the legal context in which the case arose. But should that have happened? Shouldn’t the “clear error” standard be applied the same way across all cases, or at least across all racial-gerrymandering cases, regardless of the broader legal context?

Or, put another way … in both Cromartie II and Alexander, the Justices considered several extremely in-the-weeds disputes about the trial record. To repeat an example from above, in Alexander, the Supreme Court addressed whether an expert’s failure to consider contiguity and compactness in his statistical analysis rendered his testimony unreliable. The broader legal context of Rucho seems totally irrelevant to that in-the-weeds assessment. Should the Justices’ broader ideological project of shoring up (or shoring down) Rucho influence their disposition of the fundamentally non-ideological question of whether this particular expert used a reliable methodology?

More generally, how should we feel about judges deciding non-ideological issues based in part on broader ideological considerations?

Well, we have a few options.

1. We can celebrate this practice. It’s great for judges to consider the broader consequences of their actions!

2. We can bemoan this practice. Judges violate their oaths when they take account of extraneous considerations in deciding narrow legal disputes!

I’m in between. I don’t love this practice, but accept it as the way of the world.

After all, it’s not that different from what I do for a living. When lawyers defend their clients in court, they are ethically obligated to zealously advocate for their clients on all issues in order to serve their clients’ broader goals.

Public-interest lawyers do the same thing. Disputes regularly arise in public-interest litigation that are completely non-ideological—say, whether the opposing party missed a litigation deadline. Still, public-interest lawyers zealously advocate for their clients’ interests on those issues in service of their broader ideological missions.

Lawyers think this way even when they’re not subject to an ethical obligation. Lawyers who support a particular ideology will tend to believe that certain legal positions are correct, even when the correctness of the legal position doesn’t turn on the same considerations as the correctness of the ideology. If (for example) a lawyer supports immigrants’ rights, and a dispute arises over the interpretation of some technical provision of the Immigration and Nationality Act, that lawyer will tend to believe that the pro-immigrant position is legally correct, even if it turns on a statutory-interpretation issue having nothing to do with the lawyer’s philosophical views regarding immigration. I succumb to this on issues I care about. So do you. Everyone does.

Judges are not lawyers. But they are ex-lawyers, and this type of judicial decisionmaking is ubiquitous. Last year, I wrote a post called “Does Standing Follow the Merits,” where I commented on the then-pending student loan cases. Here’s what I said:

In the Supreme Court, it is common for controversial cases to feature hot-button merits issues combined with unrelated, mundane standing disputes. The student-loan cases currently pending before the Court are a good example. Several plaintiffs, including the state of Missouri, filed lawsuits contending that President Biden’s partial forgiveness of student loan debt violated federal law. This is a juicy, consequential issue which will likely divide the Court along typical ideological lines. But before the Court can decide this question, it must decide whether the plaintiffs have standing. The plaintiffs have various theories of standing, but the most promising seems to be that the student-loan plan will harm the Missouri Higher Education Loan Authority (MOHELA) and that an injury to MOHELA counts as an injury to Missouri. The Justice Department contends that MOHELA’s injury doesn’t qualify as an injury to Missouri because MOHELA is a legally separate entity that declined to sue. Missouri retorts that MOHELA really isn’t that separate from Missouri. This is an extremely obscure and random issue that turns on esoteric intricacies of Missouri law.

In principle, the Justices’ view of the merits of the student-loan dispute shouldn’t affect their view of the antecedent question of standing. The Court isn’t even allowed to get to the merits question until it determines that the plaintiffs have standing. Also, there’s no particular reason that the Justices’ views of the merits would track their views of MOHELA’s standing. I have no idea whether it is a judicially conservative or progressive position to say that MOHELA does or does not have standing. So, in a perfect world, we’ll see no correlation between the Justices’ view of standing and their view of the merits. (We’ll find out soon.)

Well, we all know what happened in the case. The Supreme Court decided the merits issue by a 6-3 vote, and it also decided the standing issue by the identical 6-3 vote.

I find it difficult to criticize either the majority’s reasoning or the dissent’s reasoning with much fervor because it’s a close issue. As I see it, the 6-3 split on standing reflects the inevitable tendency of judges’ views on ideological issues to spill over to non-ideological issues, at least when those issues are close. And that’s what happened in Alexander, too.

Back to Alexander

Alexander was a close case composed of multiple close sub-issues.

At the highest level, the case fell somewhere in between Cromartie II and Cooper. It’s easy to write an opinion applying Cromartie II and distinguishing Cooper; it’s easy to write an opinion the other way around.

Alexander’s sub-issues were also close calls. To take one of many examples, the majority drew an adverse inference against the plaintiffs because they didn’t propose an alternative map. The plaintiffs said they didn’t know they had to do that because the legislature didn’t disclose before the trial that it would pursue a partisan-gerrymandering defense. The majority said that the plaintiffs should have predicted this, given that Democrats were complaining the whole time about a partisan gerrymander. The dissent said that the plaintiffs shouldn’t be forced to rely on Democrats’ complaints to predict the defense’s trial strategy. You can see both sides of this.

The majority resolved those close calls in favor of the State, while the dissent would have resolved those close calls in favor of the plaintiffs. The Justices’ broader ideological views about partisan gerrymandering shaded their evaluation of those close calls.

Should that have happened? Well, a judge would have to be some combination of a robot and a saint for that not to happen. Perhaps in a perfect world this type of judicial decision-making would not occur, but inveighing against it is truly tilting at windmills.