In a decision this morning, the New York Court of Appeals unanimously ruled against the town of Newburgh, the defendant in an ongoing vote dilution case under the New York Voting Rights Act (NYVRA). Newburgh argued that the NYVRA’s vote dilution provision violates the Equal Protection Clause because, to comply with the law, jurisdictions must consider race. The court declined to address Newburgh’s merits argument, holding instead that, as a municipality of New York, the town lacks the capacity to mount a facial constitutional challenge. Harvard Law School’s Election Law Clinic represents the plaintiffs, and I argued the case before the court.

This case presents the question whether the Town and Town Board of Newburgh— subordinate governmental entities created by, divisible by and even extinguishable by the State Legislature—can maintain this facial constitutional challenge to the vote dilution provision of the New York Voting Rights Act (“NYVRA”) (codified at Election Law § 17-200 et seq.). They cannot. . . .

Plaintiffs allege that (1) voting patterns in Newburgh are racially polarized and (2) the at-large election system effectively disenfranchises Black and Hispanic voters, who cannot elect candidates of their choice or influence the outcome of elections. Plaintiffs seek a declaration that Newburgh’s use of an at-large election system violates Section 17-206 and an injunction ordering Newburgh to implement either a districting plan or an alternative method of election for the 2025 Town Board election.

Newburgh moved for summary judgment on the bases that (1) Section 17-206 is facially unconstitutional because it violates the Equal Protection Clause of both the U.S. and New York Constitutions and (2) its Town Board elections comply with the NYVRA. . . .

The longstanding rule in New York is that political subdivisions—as creatures of the State that “exist[] by virtue of the exercise of the power of the State through its legislative department”—cannot sue to invalidate State legislation (City of New York v State of New York, 86 NY2d 286, 289-290 [1995] . . . .

Just as the legislature has the power to create entities to perform its functions, it has the power to change, and even destroy, those entities. Separation of powers principles accordingly demand that courts do not interfere in legislative disputes raised by legislative subordinates. Those principles are the bedrock of our federal and State Constitutions alike. . . .

Newburgh’s challenge to the NYVRA does not fall within the dilemma exception. Whatever might be said as to a municipality’s ability to bring an as-applied challenge, showing that it will be forced to take a course of action that is unconstitutional, Newburgh is pursuing a facial invalidity claim. . . . For a facial constitutional challenge, principles of “judicial restraint” (World Trade Ctr., 30 NY3d at 385) counsel strongly against permitting subordinate units of state government from using the judiciary to second-guess the wisdom of enacted legislation. A municipality’s authority to raise a challenge to a State law is at its lowest ebb when that challenge is a facial constitutional challenge, seeking to invalidate a statute in all possible applications, not merely because it allegedly placed the particular municipality in an allegedly untenable position. . . .

Newburgh’s arguments about why we should hold that it meets a dilemma exception fail to persuade us. Newburgh has not shown that compliance with the NYVRA would force it into taking an unconstitutional action. The litigation has yet to even proceed to trial, making presently unknown: (1) whether Newburgh would face any liability; and (2) in the event it did, what a court would require it to do. The NYVRA’s vote-dilution provision leaves courts wide latitude in designing remedies, so that to prevail on its facial challenge, Newburgh would have to show that “every conceivable application” of the NYVRA—i.e., every possibly remedy a trial court could order—would force it to take an unconstitutional act . . . .

Newburgh contends that because, in its view, the NYVRA violates the U.S. Constitution, the Supremacy Clause overcomes New York’s bar prohibiting its subordinate local governments from suing it. Newburgh offers no authority for that novel proposition, which would authorize every local governmental entity to sue to challenge as unconstitutional any State legislation arguably affecting that subordinate entity. . . .

Newburgh argues that “any alteration of its race-neutral, at-large election system in order to comply with the NYVRA’s vote-dilution provisions would be unconstitutional.” But that contention, as explained by counsel at oral argument (see oral argument tr at 8-12), rests on the proposition that a mere finding of liability itself would place Newburgh in the position of violating the Constitution or obeying the order of the court—when there is no order of the court compelling it to do anything. And in any event, several of the potential remedies mentioned by the NYVRA to redress a finding of vote dilution—such as longer polling hours or enhanced voter education—cannot reasonably be described as alterations of an at-large election system.

I also have no idea whether there’s any merit in DOJ’s racial gerrymandering challenge to California’s redistricting via Prop. 50. Assessing those claims would require development of a full factual record that does not yet exist.

But I wanted to respond to a legal argument my friend Rick Hasen recently made in Slate about that challenge that I believe is wrong.  Rick asserts that DOJ has a “huge problem:” even if the legislature designed various districts as unconstitutional racial gerrymanders, he argues that doesn’t matter because CA voters ultimately approved the map.  And he argues (and here I agree) voters approved the map for partisan political purposes.  Thus, his argument runs, race cannot have predominated in the design of specific districts and hence DOJ’s racial gerrymandering challenge to various districts must fail.

I do not think voter approval can turn an unconstitutional set of districts into constitutional ones. Voter approval cannot “cure” a map that would otherwise violate the Constitution.  Suppose, for example, the mapmakers had designed districts with a racially discriminatory intent.  But when voters approve the map, the entire campaign is based on partisan appeals.  The voters themselves might not have had a racially discriminatory intent (leave aside the complexity of assigning purposes in a direct democracy process). Nonetheless, I’m confident the map would still be struck down, because a map designed for racially discriminatory reasons remains unconstitutional even if voters approve it.

Similarly, if the mapmakers design a map that violates the “results” test of Section 2 of the VRA, the map still violates Sec. 2 — even if voters must approve the map before it takes effect.

There is no reason racial gerrymander claims are any different. Racial gerrymandering doctrine prevents a State, in the absence of “sufficient justification,” from “separating its citizens into different voting districts on the basis of race.” Bethune-Hill v. Virginia State Bd. of Elections, 580 U. S. ___, ___ (2017) (slip op., at 6). Absent sufficient justification, race cannot be the predominant factor in how districts are designed. If districts are designed as unconstitutional racial gerrymanders, they do not become unracially gerrymandered merely because voters approve the final map. In the Texas litigation, for example, if voters in Texas had approved the new maps, after a campaign proclaiming the redistricting was done for purely partisan purposes, that would surely not have affected the district court’s recent decision that six of those districts were unconstitutional racial gerrymanders.

Moreover, voters do not design districts. They approve a map. But a racial gerrymandering claim does not challenge a “map” as a whole.  Those challenges must be made to specific districts.  Voter approval of a map does not change the way specific districts have been designed. The harm in racial gerrymandering cases, under the doctrine, is that race has been the predominant factor in how a district has been designed.

As I say, I have no idea whether the facts support DOJ’s racial gerrymandering claim.  But if those districts are racial gerrymanders, they don’t magically become unracially gerrymandered merely because voters have approved the final map. 

I have no view at this stage whether the majority is correct that six districts in Texas’s re-redistricting are unconstitutional racial gerrymanders. The dueling opinions are long; the issues are highly fact dependent. But I wanted to note that the decision is based on the anti-racial gerrymandering doctrine that the Court first established in Shaw v. Reno (1993).

At the time Shaw was decided, many voting-rights groups were sharply critical of the decision. But in the years since, Shaw has been used successfully over and over again by voting-rights groups and their allies to invalidate racially gerrymandered maps. In the Texas case, the lead plaintiff was the League of United Latin American Citizens (LULAC), which has been the major group litigating on behalf of Latino voting rights in Texas for decades.

If the majority is indeed right that these districts were racially gerrymandered, it would be far more difficult, and frequently impossible, to invalidate them on some other basis in the absence of Shaw’s constraint on racial gerrymandering. In Texas, the plaintiffs also brought vote dilution and intentional discrimination claims. But racial gerrymandering frequently happens without vote dilution taking place. And voting-rights plaintiffs have long been concerned about the difficulty of proving intentional discrimination.

The court decided the Texas case based on Shaw and the racial gerrymandering claims precisely because those claims are easier to prove. Having invalidated the districts on that basis, the court did not address these other claims.