Constitutional Problem-Solving

Joseph Fishkin

In an end-of-the-term flurry that was not lacking in forceful dissents, Chief Justice Roberts’ dissent in Arizona State Legislature v. Arizona Independent Redistricting Commission [Legislature v AIRC] certainly held its own. (In case some readers were unsure how to keep score at home, Justice Scalia, writing separately, opined that the Roberts dissent he joined was “devastating.”) One moment of particularly high dudgeon stood out to me: the Chief’s back-of-the-hand dismissal of the argument of the first page or so of the majority opinion. “The majority begins by discussing policy,” he writes, “I begin with the Constitution.” Elsewhere he accuses the majority opinion of consisting in substantial part of “naked appeals to public policy.” What he’s mainly talking about is the start of Justice Ginsburg’s majority opinion. It reads as follows:

This case concerns an endeavor by Arizona voters to address the problem of partisan gerrymandering—the drawing of legislative district lines to subordinate adherents of one political party and entrench a rival party in power. “[P]artisan gerrymanders,” this Court has recognized, “[are incompatible] with democratic principles.” Vieth v. Jubelirer, 541 U. S. 267, 292 (2004) (plurality opinion); id., at 316 (Kennedy, J., concurring in judgment). Even so, the Court in Vieth did not grant relief on the plaintiffs’ partisan gerrymander claim. The plurality held the matter nonjusticiable. Id., at 281. Justice Kennedy found no standard workable in that case, but left open the possibility that a suitable standard might be identified in later litigation. Id., at 317. 

Legislature v. AIRC is not a partisan gerrymandering suit, but a case about the Elections Clause. That clause says: “The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations, except as to the Places of chusing Senators.” The question in Legislature v. AIRC was whether the phrase “by the Legislature thereof” means independent redistricting commissions like Arizona’s are unconstitutional. The majority and dissent agree that this is what the case is about. But they seem to disagree in an interesting way, not fully spelled out by either side, about the relationship between this question and the question in Vieth.

In Chief Justice Roberts’ view, the problem of partisan gerrymandering is a “policy” problem rather than a constitutional problem. In other words: Partisan gerrymandering is something we might all agree is not good. It would be nice to have less of it. But there are lots of things it would be nice to have less of—traffic jams, for instance. Rain on the Fourth of July. We do not alter our readings of the Constitution to achieve such goals. On this view, reducing partisan gerrymandering might be a “noble” aspiration, but whatever its merits, it is a policy goal external to the Constitution, and it would be improper to allow such a policy goal to distort constitutional interpretation. Roberts draws this line sharply, complaining that the majority opinion shows “greater concern about redistricting practices than about the meaning of the Constitution.”

Justice Ginsburg obviously sees it differently. Her choice to open the opinion with the paragraph quoted above says as much. But Ginsburg never exactly spells out her answer to the “policy” charge. To understand how that side of the argument runs, it helps to remember exactly what happened in Vieth.

Vieth left partisan gerrymandering in a kind of doctrinal limbo: it is a constitutional wrong without a clear remedy. Five Justices (the four liberals + Kennedy) believed that there could be some judicial remedy in the future. The other four did not. But, all nine signed opinions that suggested that partisan gerrymandering is a fundamental problem for our constitutional order—a phenomenon “incompatible . . . with democratic principles.” (That particular sentence, which Ginsburg quotes above, is not from any of the liberal dissents or Kennedy—it’s from Justice Scalia. Scalia is in turn paraphrasing Souter; he adds, “We do not disagree.”). In other words, the divide in Vieth was not really about whether partisan gerrymandering was a constitutional problem. It was about whether courts had a judicially manageable solution. At the moment, it seems like they don’t.

This raises an important question. What is a constitutional problem without a judicial solution? We are living in an era in which the Supreme Court is so super-supreme, so utterly dominant in its authoritative command of the field of constitutional interpretation, that this question may sound almost like a philosophical riddle. (If part of the Constitution falls in the forest and no judiciary is there to hear it, does it make a sound…?) It is actually a little counterintuitive today to think of how any part of the Constitution could be alive and yet not directly enforceable by courts. Holdings of nonjusticiability, in this light, can read almost as though they say that the relevant constitutional provision is mere rhetoric, aspirational talk without legal consequence. In Vieth there were arguably only four votes for nonjusticiability, but even there, the court’s inability to articulate a rule for judicial intervention feels almost like a statement that partisan gerrymandering is not really unconstitutional. In popular culture and even in our present legal culture, we tend these days to think of the real Constitution as the judicially enforced Constitution, so if something is really unconstitutional, that would mean some party can go to court and get an injunction against it, right?

This way of thinking is a big mistake. And it’s one with far-reaching and occasionally surprising consequences, not only outside the courts but also (and this is my focus in the rest of this post), inside the courts. Thinking this way not only impedes non-judicial actors from solving constitutional problems. It also impedes courts from solving constitutional problems.


The Constitution is enforced by every actor in the constitutional scheme, not only by courts. This is not a controversial point. Many smart people have ably made the arguments for legislative, administrative, and popular constitutionalism. This blog post is not the place to rehash them.

In their role in the constitutional scheme, part of what courts do is try to solve constitutional problems. One way to try to solve a constitutional problem is by creating a means of direct enforcement. If one-person-one-vote violates equal protection, then courts can give somebody standing to come in and sue over it, and then grant injunctive relief. But there are other ways to solve a constitutional problem. Many of those other ways involve interpreting other, sometimes-seemingly-unrelated constitutional provisions in such a way as to enable rather than block the efforts of governmental actors other than the federal courts to address the constitutional problem. For example, think of the Court’s landmark decisions reading the Commerce Clause broadly to reach the private conduct of Ollie’s Barbecue or the Heart of Atlanta Motel. Part of the reason the Court interpreted the Commerce Clause the way it did was in order to help solve a constitutional problem—widespread racial discrimination, etched deep into the structure of American society by very numerous private and public actors—that the Court knew was going to be difficult to solve.

One key judgment in any case of this kind is the very first step: the judgment that the problem in question is an actual constitutional problem, a problem that strikes at some pillar of our constitutional scheme.  In those commerce/civil rights cases, the idea was that private as well as public conduct had built economic and social structures that were inimical to constitutional equality principles.  In Vieth, the idea was that the practice of partisan gerrymandering threatens the democratic foundations of our constitutional order.

This is why it matters to Chief Justice Roberts to characterize partisan gerrymandering as a “policy” problem. Distorting constitutional interpretation to achieve our desired policy results is not legitimate. It’s not legitimate because it collapses the distinction between constitutional high politics and ordinary low politics. It’s not as bad as interpreting the Constitution in a way intended to help a judge’s favored political party (that would be even worse, which is why Bush v. Gore draws such special scorn from so many). But it’s not good.

On the other hand, interpreting a constitutional provision in light of efforts to solve or mitigate a constitutional problem is part of what courts do in a constitutional system. That is why Justice Ginsburg begins her opinion with Vieth. What she is saying is: this is not just some policy preference. This is constitutional law.

Constitutional problem-solving is an element in decisions about constitutional interpretation on both the left and the right, although the two sides often differ about which problems really are constitutional problems. For instance, think of the Commerce Clause language in NFIB v. Sebelius—arguably all dicta, but important nonetheless. In that case, the five conservative Justices said they would limit the reach of the Commerce Clause in part because of not-directly-Commerce-related constitutional concerns about individual liberty—the word “broccoli” should come to mind—that loomed large in the background of the case, and appeared directly in the opinions. (The Commerce Clause is a popular vehicle for solving various kinds of constitutional problems, it seems.)

I have been thinking about constitutional problem-solving a lot lately because of a book project I am working on with my colleague Willy Forbath (first installment here) where it looms large. We are writing about a set of constitutional arguments that Americans of many different political stripes have made about equal opportunity. In many cases, the problems these constitutional arguments identify simply cannot be solved through judicial intervention alone, because of the type of problems they are—problems of what we call constitutional political economy. 

For instance, take the problem of oligarchy. Although today we rarely understand our constitutional system in these terms, generations of Americans viewed extreme concentrations of economic and political power as not only a policy problem but a constitutional problem.  The idea, somewhat lost today but we think worth recovering, was that the constitutional order rests on and presupposes a political-economic order. If too few control too much of the economic and political power, then we have an oligarchy rather than the Republic our Constitution contemplates.  If oligarchy is a constitutional problem there are various potential responses, but few of them are the province of courts alone. Courts, by themselves, are not going to create antitrust law, an income tax and an estate tax, labor law, or campaign finance reforms. All of these interventions in our political economy can serve as bulwarks against the consolidation of oligarchic power; that is part of what all of them are there to do. All are primarily legislative creations. But in each of these areas, courts play very important roles—at a minimum, upholding the laws, and usually much more: shaping and reshaping them over time, as the constitutional problems they aimed to address also change over time.

Constitutional problem-solving plays a complex role in constitutional interpretation and adjudication. It is not a license to read principles into the Constitution that are not there, or to ignore principles that are there. It is not a license to override clear commands: even if widespread age discrimination is a pressing constitutional problem, this is not a good enough reason to disregard the rule that the President must be 35 years old. Instead, constitutional problem-solving is something courts do when interpretive work is called for, at the same time that they are doing other forms of interpretive work. In Legislature v. AIRC, despite the dissent’s insistence that the term “Legislature” is completely unambiguous, the real question in the case, and it is not a particularly easy question, is whether that term ought to be read institutionally or functionally. If read functionally—the legislature is the power that legislates—then the Arizona commission is fine; if read institutionally, the Elections Clause requires the Arizona State Legislature to have the authority over districting and the AIRC would lose.

It was lost on none of the Justices writing in Legislature v. AIRC that the institutional reading favored by the dissenters would hard-wire into our constitutional order a commitment to allowing partisans (the legislators) to draw Congressional district lines, with only the judiciary even potentially left to stand in the way of egregious gerrymanders (and likely not even that). The only question was whether acknowledging this bolstered the majority’s constitutional interpretation or revealed its illegitimacy.

Courts can try to go about the work of constitutional interpretation clause by clause and word by word, viewing small units like the word “Legislature” like little scientific specimens, as if interpretation were a science. They can consult dictionaries and the rest of the textualist tool chest. But I actually do not think it is possible to avoid thinking about the larger constitutional problems that an interpretation either helps solve or makes worse. If judges are being honest with themselves, they acknowledge that this is part of what they're doing. The real question is which constitutional problems they think are worthy of attention.
Posted 9:03 PM by Joseph Fishkin [link]