2018 is promising to be far more consequential than your average midterm election year. A number of landmark Supreme Court and state court decisions could literally transform parts of the country’s political landscape. Beginning with Gill v Whitford, the case heard by the Supreme Court last October alleging that the state of Wisconsin’s legislative districting plan diluted the votes of Wisconsin Democrats, and a similar suit brought by Maryland Republicans, along with districting plans recently thrown out in North Carolina and Pennsylvania, the practice of extreme partisan gerrymandering is now being challenged on multiple fronts.
Unfortunately, while few plans (Pennsylvania is an important exception) will likely be redrawn in time for state primaries, the Supreme Court’s opinions in Gill and Benisek (Maryland), expected in June, will keep the issue salient throughout Congressional campaigns, regardless of the outcome. In an era characterized as “post-truth,” with claims of “fake news” polluting our discourse, there is something redemptive to following this legal and scientific debate. Arguably the two strongest truth-generating discourses of our civilization, science and the law, are intersecting on this question of how to fairly map out areas for electoral conflict. It reminds us of just how precious truth is to this whole civilizing project.
First, the question of which constitutional principles are at stake is being addressed. In previous gerrymandering cases and those involving enforcement of the Voting Rights Act, claims of voting rights violations have relied on the Equal Protection Clause and a demonstration that a districting plan dilutes the value of some votes over others. Whitford followed this path in part, as did the North Carolina case, and the Pennsylvania plaintiffs, in State Supreme Court, successfully argued that their plan violated the equal protection guarantee in the state constitution. However, plaintiffs in that case were also successful in getting the plan thrown out on the grounds that it violated state free speech protections.
This potential of a First Amendment protection against gerrymandering was hinted at by Justice Kennedy in the Supreme Court case that first removed the threat of judicial restraint from state legislatures seeking maximum partisan advantage, Vieth v Jubelier. Taking the hint, plaintiffs in Gill also claimed a free speech violation, as did the Maryland Republicans in Benisek and plaintiffs in the successful North Carolina case. Two courts, one state and one federal, have thus already declared partisan gerrymanders unconstitutional based on violations of equal protection and free speech.
A third approach also received judicial support in the North Carolina case, but was rejected by a federal court in Pennsylvania. Plaintiffs in North Carolina successfully argued that extreme partisan gerrymandering of Congressional districts violates Article I, section 4 of the Constitution, the Elections Clause, which delegates to Congress federal authority over the “Times, Places and Manner of holding elections for Senators and Representatives…” The logic of this claim rests on an interpretation of the Clause as a safeguard that protects the House of Representatives against manipulation from state legislatures. In a related vein, during Gill arguments, Justice Gorsuch pondered whether or not these cases actually fell under the Guarantee Clause, which obliges the federal government to provide a republican (majority rule) form of government in each of the states.
Wherever the Constitutional violation might ultimately be grounded, the actual intersection of law and science comes in the standards used to establish the truth of a constitutional violation having occurred. Indeed, language between the two fields has become a little slippery here, as there are scientific standards that scientists develop, and constitutional standards developed through case law, such that the attempt by scientists to “constitutionalize” their standards or “quantify” the Constitution has become a familiar criticism of defendants. Nevertheless, most scientists are quite clear, as were professors Gaddie and Grofman in their Gill (and other) Amicus Brief, that “social scientists provide metrics, courts specify standards.”
The scientific gold metric that has emerged over the last several decades is partisan symmetry. Given that proportionality (where a party’s seat share is equivalent to its vote share) is currently a higher standard than what the Constitution demands, a symmetry test measures how disproportionality is allocated between parties. If one party is disadvantaged, as in the extreme case where one party wins a majority of the vote but fails to win a majority of seats, that districting plan is asymmetric.
Thanks to advances over recent decades, symmetry can be measured numerous ways, with bias, median-mean gap and efficiency gap being the most widely cited in these cases. These are complimentary measures in so far as they converge on clear cut examples of partisan gerrymanders. Further, advances in spatial and computational science now make it possible to assess how durable a gerrymander is through simulated vote swings and the sampling of alternative plans that meet standard districting criteria.
“Advances in statistical and empirical theory and application” like these, noted Judge Wynn (North Carolina) in one of the more eloquent decisions to date, “have the potential to allow parties, experts, and amici to provide courts with more rigorous and probative evidence,” and he warned against those who would dismiss science as gobbledygook: “The Constitution does not require the federal courts to act like Galileo’s Inquisition and enjoin consideration of new academic theories, and the knowledge gained therefrom, simply because such theories provide a new understanding of how to give effect to our long-established governing principles.”
Establishing the truth
Just how the Supreme Court will ultimately apply the science is not clear, but if they do, it is likely to take some form of three-part standard that uses the test of symmetry as evidence of effect, and a statistical estimate of how durable, or unresponsive, a plan is to vote swings compared to alternative or historical plans, which could also serve to demonstrate intent and/or invalidate alternative explanations for the degree of observed asymmetry. These sorts of applications were developed by the lower courts, in addition to use of non-quantitative evidence of intent, including, in the case of North Carolina, public acknowledgement of the partisan intent of the designers. It won’t always be that easy.
The final truth to be established through this discourse is “how much is too much?” While it is unlikely that a constitutional standard against gerrymandering will tolerate no asymmetry, or “as little as possible,” courts are not likely to overturn just the few extreme cases like Wisconsin or Virginia, where majority rule has repeatedly been violated. The crucial question is whether any voter whose party loses at least one seat through gerrymandering is entitled to relief. Whatever the decision, science and the law will continue this truth-generating application of facts to principles, in a way that hopefully upgrades our capacity to democratically address collective disagreements.
Support from UCS members make work like this possible. Will you join us? Help UCS advance independent science for a healthy environment and a safer world.