Much was at stake during Tuesday’s oral arguments at the Supreme Court: the credibility of democracy, the role of the courts, the limits of gerrymandering. But surprisingly, what was most on trial was social science.
The case was Gill v. Whitford, and at issue was one of the most egregiously gerrymandered electoral maps in history, drawn up in secret by Wisconsin Republicans. The closed-door redistricting committee rejected eight maps as being insufficiently tilted in Republicans’ favor, then settled on one that was so biased that Republicans subsequently won 60% of the seats in the legislature with only 47% of the vote.
The Court has long said that, however distasteful, partisan gerrymandering is usually constitutional—unlike, say, drawing districts based on race. The Court has also said that, theoretically, such finagling may be so egregious that it violates constitutional rights to equal protection, and perhaps to freedom of association and speech. But it has never been able to agree on how much is too much—and today’s arguments illustrated why.
On the face of it, Wisconsin’s map should definitely count. In addition to the secret, biased process and highly slanted outcome, the map was shown, on a series of statistical analyses, to be an extreme outlier. All maps are doctored – but this one was crazy.
And yet, how crazy is too crazy? That’s the problem the Court wrestled with today. The plaintiffs offered a three-part test: single-party rule, single-party slanted districts, and statistically lopsided results. But that third part is particularly nettlesome. Much of the Court’s time was spent dickering over terms like “efficiency gap” and “symmetry” and where the line should be drawn.
“This is a bunch of sociological gobbledygook,” complained Chief Justice Roberts. Wisconsin’s map has an “efficiency gap”—which, basically, measures the difference between its districts and fair ones—of 7%. That’s bad. But what about 5%? Where is the magic number?
Or, in Justice Gorsuch’s damning words, “What are the numbers that we are supposed to read into the constitution?”
Worse, the lower court opinion actually used three different statistical analyses. “This is like my steak rub,” the famously folksy Gorsuch offered. “There’s a little bit of this, and a little bit of that, but I’m not going to tell you how much of each.”
If anyone was worried that Gorsuch’s aw-shucks manner was just a show for his senate confirmation, rest assured—this is who he is.
Well, that’s Gorsuch and Roberts. (True to form, Justice Gorsuch hinted he may outflank the entire court on the Right, and say it shouldn’t even be hearing this kind of constitutional challenge.) And Alito, who seemed even more skeptical. And Justice Thomas, who, silent as always, will surely follow Justice Alito just like he followed the late Justice Scalia. (It appeared that Justice Thomas fell asleep at one point during the argument.)
Meanwhile, Justices Kagan, Sotomayor, Breyer, and Ginsberg all said that if nothing is done, trust in democracy will erode. If you know your district is a sure red or blue one, what’s the point of even voting? And if a minority party is guaranteed a legislative majority for years to come, how is that democratic?
In other words, Gill will come down—like so many other cases—to Justice Kennedy, who seemed to well encapsulate the conundrum facing the Court. On the one hand, Wisconsin’s map is clearly outrageous. On the other hand, how to rule against it without choosing some arbitrary standard, and provoking a flood of litigation to argue about it?
The thing is, while the justices can argue about “sociological gobbledygook” —Justice Breyer called it “computer stuff”—politicos on both sides of the aisle are using that gobbledygook to engineer more and more slanted electoral maps. After all, it isn’t gobbledygook—it’s data. Big Data, to be more precise. And it works.
And that leads to an uneven playing field. Partisan fixers can use big science, but courts can’t. The extremists in Wisconsin and elsewhere use the latest data modeling to lock up districts, but if courts use those same models to evaluate them, now they’re peddling gobbledygook. The conservative reluctance to get into the nerdy, data-filled weeds basically takes the courts out of the game, leaving whoever is in power to do whatever they want.
So, Justice Kennedy, what will it be?
Having followed Kennedy for years, and having scrutinized every expression on his stony poker face during today’s argument, I can say that it would be foolish to predict.
On the one hand, Justice Kennedy is not afraid of muddy standards and case-by-case reasoning. In numerous contexts—abortion and affirmative action, for example—he has written opinions that decline to espouse a clear, bright-line rule but basically say “I know it when I see it.” If he applies that method of reasoning (critics call it a lack of reasoning) to this case, he should throw out Wisconsin’s map.
On the other hand, this is messy even for Justice Kennedy. As Justice Alito pointed out, the “efficiency gap” metric is only three years old, and it leads to numerous “false positives,” finding electoral maps to be biased even when they were actually created by bipartisan commissions. Social science may not be gobbledygook, but it’s not as precise as, say, geology. Does it really justify a court throwing out the product of elected legislators?
I think so—for three reasons.
First, it’s what the “other side” is using. Maybe big data is reliable and maybe it isn’t. But to apply the “efficiency gap” and similar tests is to use the same tools as were used to create the map in the first place. It’s not voodoo – it’s using the same methodologies as what’s being reviewed. And Justice Kennedy is generally not afraid to get into the weeds.
Second, it’s not the only evidence. Justice Kennedy could well out-Kennedy the plaintiffs in Gill by looking not just at the data but also all the evidence of locked committee rooms, the discarded maps, and so on. Maybe the sociological data is just one part of the picture. Maybe it’s the picture in its totality that is clearly damning.
That result would indeed lead to more challenges, but it’s the kind of reasoning Kennedy has used before.
Which leads to the third point: what’s the alternative? Maybe this standard is messy, but the alternative is no standard at all. And if even a map as egregious as Wisconsin’s is immune from review, we ain’t seen nothing yet. Big data is getting more and more sophisticated. We’re going to see more and more anti-democratic gerrymandering. And if the Court says “anything goes,” we’re going to see the further erosion of our democracy.
On the courthouse steps, there were chants of “this is what democracy looks like!” and, more cleverly, “we’re tired of your tampering / stop your gerrymandering.” Former governor Arnold Schwarzenegger urged the Court to “Say Hasta La Vista to gerrymandering.” It was the usual pointless rally.
And yet, maybe they were onto something. Yes, Gill is in part about micro-details and the value of social science. But if the Court gives a blank check to partisan data crunchers, it will sit alongside cases like Citizens United and Bush v. Gore as milestones along the path of democracy’s demise.
As messy as this case is, I don’t think Justice Kennedy will allow that to happen, and we should all pray that he doesn’t.