IN THE RECENT decision of the Supreme Court on the constitutionality of the Affordable Care Act, National Federation of Independent Business v. Sebelius, the opinions cited a myriad of precedents, or past decisions of the Supreme Court. Every Justice who wrote claimed that his or her position was consistent with, even dictated by, the precedents. This is an annual sight in Washington toward the end of June when the decisions on politically charged cases are delivered: as surely as the swallows return to Capistrano, so too the Justices turn to precedent to justify their decisions.
But the ritual creates a number of puzzles. What of the claims on both sides that the other’s position was unprecedented? In the health care case, five Justices would have invalidated the Act’s individual mandate as beyond Congress’s power to regulate interstate commerce (although a different majority actually upheld the Act on a different ground, as a constitutionally permissible tax). Those five Justices said that congressional regulation of “inactivity”—a congressional mandate that “inactive” citizens, who rely on emergency care, must purchase health insurance—was unprecedented and therefore suspect. On the other hand, four Justices said that the very distinction between activity and inactivity drawn by the five was itself unprecedented, lacking any basis in the Court’s previous decisions. Could both sides have been right? If so, is it because they were referring to different types of “precedent”?
Michael Gerhardt has produced a learned overview of contemporary debates in legal theory and political science over the role of precedent, or past official decisions, in the legal system. Gerhardt’s book was published before the health care decision came down, but the debates that he canvasses are and always will be with us. Do precedents constrain judges? Even if they constrain lower-court judges in run-of-the-mill cases, do they constrain the Justices, especially in the politically fraught cases decided toward the end of June? What of “precedents” created by non-judicial officials through their own decisions—do the same questions apply, or should we think about them in a different way altogether?
Gerhardt offers sensible mainstream judgments on these issues, though there is something of a Goldilocks problem with the book: Gerhardt wants an account of precedent that is neither too hot nor too cold, so it ends up lukewarm. As against the traditional lawyers, who assert the influence of precedent, Gerhardt emphasizes several points. Precedent has limited constraining force, especially in hard cases at the Supreme Court level; precedent serves multiple purposes—heuristic, justificatory, rhetorical—in addition to constraint; non-judicial precedents are collectively far more important than judicial ones; and the significance of a precedent is determined not by the official who issues it, but by the happenstance of subsequent official decisions, political circumstances, and social movements. Still, as against the many political scientists who are skeptical that precedent has any constraining force at all, Gerhardt argues that judges have self-interested incentives to maintain a norm of precedent through cooperation. The “golden rule,” he suggests, is that “the justices recognize the need to give the same level of respect to the precedents of others as they expect their preferred precedents to deserve.” (Here Gerhardt implicitly appeals to a game-theoretical account of precedent as an emergent norm based on reciprocity, pioneered by law-and-economics scholars such as Erin O’Hara.) He documents that it is very rare for the Court to overrule a precedent, or for an individual Justice to call for overruling one. Indeed, Gerhardt goes on to argue that there are “super-precedents” with more than ordinarily binding force; Marbury v. Madison is an example, the decision in 1803 that is conventionally cited as having established judicial authority to decide whether laws are constitutional.
Gerhardt’s arguments against the traditional lawyers are more persuasive than his arguments against the political scientists. All of human history bears witness that where political actors play tit-for-tat, the golden rule of reciprocity is but one possible equilibrium; the other equilibrium, darker and perhaps more common, is that if I gain the upper hand, I will gleefully do unto you exactly what you would do unto me if the balance of power were reversed. Many of the Justices who voted to restrict Congress’s power to regulate national health care voted a few years back to uphold Congress’s power to regulate medical marijuana, or to detain sex offenders; and so it goes, on and on. The Justices are not particularly faithful to their own precedents, let alone to those written by others. The evidence that a norm of reciprocity in fact operates at the Court is contestable, to say the least.
If the Justices rarely overrule precedents, moreover, it may be because they rarely need to. It is the unusual case that cannot be put aside as irrelevant (“distinguished,” as lawyers say), at least with enough effort and ingenuity. In high-stakes cases, the effort is worth it, and the Justices have a crowd of ingenious clerks and counsel to help them make it all come out right. The health care opinions feature the most useful tool in the legal casuist’s toolbox, which is the introduction of an entirely new conceptual dimension unanticipated by the earlier precedents. “Sure, the cases say that Congress may regulate economic matters,” certain Justices reasoned, “but this is a regulation of economic inactivity rather than economic activity, and we’ve never said that was OK.” Welcome to the new paradigm.
It is this ability to introduce new and unforeseen dimensions that puts today’s precedent at the mercy of the future. Randy Barnett, a libertarian legal scholar and advocate—the genial Victor Frankenstein who stitched together and galvanized the activity/inactivity distinction—has suggested that even though the Affordable Care Act was upheld, the opinions handed his side a long-run victory by announcing a more restrictive set of Commerce clause doctrines. He lost the battle, but he might still win the war. But Barnett is too shrewd not to understand that judicial opinions by themselves cannot produce long-run victories. A future majority could uphold a future statute falling within the semantic scope of the rules that Barnett likes, by introducing some distinction that neither Barnett nor I can now anticipate. “Sure, the cases say that Congress may not regulate inactivity, but this is a regulation of a special type of inactivity, type X, and we’ve never said that wasn’t OK.” Whether the gambit will work will be a function of the then-prevailing circumstances, not of what some Justices said back in 2012. At the Supreme Court level, in hard cases at least, constitutional law is a game without legal rules; the rules are created and enforced by other means.
So what are the rules, and where do they come from? The precedents that mattered in the health care cases were not legal precedents in the narrow sense, the doctrinal rules set out in opinions and dissected at length in constitutional law treatises. What mattered were normatively colored expectations, held by public opinion and (derivatively) by public officials, about how political actors, such as the Justices, do and should behave. These expectations are shaped in part by past episodes—decisive showdowns between branches of government that are refereed by public opinion. The result of such a showdown is a constitutional “convention,” to use the term developed in Commonwealth legal and political theory. Some conventions have fuzzy boundaries and contestable significance, but by and large they are less malleable and more difficult to manipulate than judicially crafted rules, in part because of the high salience of the showdowns that create them and in part because they are typically less elaborate and technical.
It is a convention that Supreme Court justices should cite a great deal of legal precedent, even in great and hard cases. Since the showdown between the New Deal and the Court in the 1930s, however, it is also a convention that the Court should not invalidate major social welfare statutes enacted by the federal government. The only statutes invalidated on Commerce clause grounds since 1937 have involved minor or collateral federal regulation outside the core domain of national social insurance and social welfare, whereas the Affordable Care Act was serious business, squarely in the conventional core of federal competence. (In another part of the decision, the Justices invalidated an expansion of the cooperative state-federal Medicaid program on the ground that the statute “coerced” states into participating, but the decision leaves open the option that the federal government can in effect bribe states to accept the expansion, so the invalidation does not foreclose attainment of the federal aim.) That set of unwritten norms is not found in terms anywhere in the reports of Supreme Court opinions, but it seems likely to have weighed on the Court’s decisive voter, Chief Justice Roberts.
It isn’t helpful to say, as some legal scholars do, that “politics” determines what the Court does. There is a sense in which that is true, but the category of “politics” is too crude. It includes too many things that are not alike, ranging from the grandiloquent constitutional discourse of the 1936 court-packing debate to a Chicago mayoral election. Even if, as I have suggested, Chief Justice Roberts’s decision was not plausibly driven by lawyerly parsing of the language in previous Commerce clause decisions, this does not mean that it was the same sort of decision as Mitt Romney’s decision to oppose the Affordable Care Act. Between “politics” in that sense and legal precedent there is this third category of convention. An official who breaches a convention will be thought to have violated the unwritten rules of the game, not merely to have made just another move within the game.
The deep structural conventions of American constitutional politics are the real super-precedents. Marbury is just a synecdoche for an elaborate set of conventions that structure the metes and bounds of judicial power in the United States, and would continue to exist even if Marbury’s precise holding—which few constitutional lawyers remember—were miraculously wiped off the books. Gerhardt discusses non-judicial precedents, but his examples are mostly small-bore, and one wishes he had brought his formidable knowledge of constitutional history to bear more directly on the mechanisms that underpin the conventions of the American legal system. Legal theorists have recently begun to focus on these, but political scientists have done relatively little. From the social point of view, a great deal of understanding could be obtained by even a modest reallocation of academic attention toward conventions and away from precedent-parsing (by academic lawyers) or data-crunching (by academics who study judicial behavior in politics departments). Gerhardt’s book does some of this crucial work, but there is much more to be done.
Adrian Vermeule is John H. Watson Professor of Law at Harvard Law School. His most recent book is The System of the Constitution.