How Emphatically is it the Province and Duty of the Judiciary to Say What the Law Is?

During the first week of constitutional law as conventionally taught in most American law schools (and as I teach it), students learn that, in the words of Chief Justice John Marshall speaking for the Supreme Court in Marbury v. Madison:

It is emphatically the province and duty of the Judicial Department to say what the law is. Those who apply the rule to particular cases must, of necessity, expound and interpret that rule.

Marshall made that statement in the course of justifying the practice of judicial review of congressional legislation for consistency with the Constitution. Although there is a widespread myth that the Court invented judicial review in Marbury, Marshall did not break new ground. State courts had been practicing judicial review for over a decade, and before that a kind of proto-judicial review existed for decades in the colonies--as work by legal historians such as (my former colleagues) Barbara Black and Philip Hamburger, as well as (my law school classmate) Mary Bilder has shown. Evidence from the Constitutional Convention--especially the decision to reject Madison's proposed Council of Revision as unnecessary--as well as Hamilton's Federalist 78 (which Marshall all but plagiarized in Marbury) also make clear that usurpation is a myth.

The real controversy around judicial review has always concerned scope--which I mean in two senses. As my co-blogger Professor Segall frequently reminds readers, judicial review in the early Republic was defended and (more commonly than now) applied on the assumption that invalidation was reserved for cases of clear violations. That conception of judicial review has been most closely associated with James Bradley Thayer and, in modern times, with notions of judicial restraint.

Here I want to focus on a related aspect of scope, concerning who is bound by judicial rulings. The most articulate versions of judicial review skepticism--voiced by the likes of Thomas Jefferson and Abraham Lincoln (in reference to Dred Scott in his first inaugural)--have accepted that courts get to say what the law is in cases within their jurisdiction but have pushed back on the notion that such judicial pronouncements are binding (except as a matter of prudence) on other governmental actors. In the 1990s and early 2000s, legal scholars (especially Larry Kramer but to some extent also Robert Post, Reva Siegel, and others) sought to revive this notion of "departmentalism"--under the rubric of what came to be called popular constitutionalism. They argued for judicial review without judicial supremacy.

Most versions of popular constitutionalism took aim at constitutional interpretation in particular. Under this view, Marshall was not wrong to see constitutional interpretation as just like statutory interpretation for the work of the courts, but the People have a special relationship with the Constitution as higher law. A popular constitutionalist would thus think that there should not be judicial supremacy with respect to constitutional interpretation but that judicial supremacy with respect to statutory interpretation is unobjectionable. After all, if the People don't like how the Court has construed a statute, their representatives in Congress can always change the statute. This version of popular constitutionalism--which strikes me as the dominant version--thus depends on how difficult our Constitution is to amend relative to the making of ordinary law.

But what if there is another way in which Marshall's claim is misunderstood or overblown? What if we ask whether there really is or ought to be judicial supremacy even in matters of statutory interpretation?

I raise that question a propos of the discussion yesterday of the overruling of Chevron deference in the Loper Bright case at the PLI conference on the most recent Supreme Court Term. As I noted on Monday, I was responsible for the affirmative presentation of the case. Somewhat to my surprise, one of the other panelists said he thought Loper Bright was rightly decided because Chevron was inconsistent with the Marbury "say what the law is" function of the judiciary--even though, as I discussed here, it's not clear that the Chief Justice's lead opinion in Loper Bright rests on the Marbury principle. True, the Chief cites Marbury a number of times, but the holding more clearly rests on the Court's conclusion that Chevron deference is inconsistent with the scope of reviewing authority given to courts under the Administrative Procedure Act. 

But let's play the argument out anyway. Certainly courts don't have the exclusive power to say what the law is, even in statutory cases. Suppose Congress enacts a new statute creating rights and imposing duties without any sort of administrative function. Immediately, lawyers for clients subject to the new law will need to tell those clients what the law is. If a party is eventually involved in litigation as either a plaintiff or a defendant (in a civil or, where applicable, criminal case), the party's lawyer's view of what the law means will have to yield to that of a court, but that doesn't mean that the lawyer didn't have an initial duty (to the client) to say what the law is. The same is true for government lawyers, including those inside federal agencies where a law tasks an agency with enforcing some law.

Of course, Chevron didn't just say that agencies are co-equal with the judiciary in the interpretation of statutes--as departmentalism says with respect to the Constitution. Chevron made agency interpretations of statutes superior to those of courts, at least where the courts judged the relevant statute ambiguous and the agency interpretation reasonable. That's a step beyond the critique of Marbury.

Or is it? In Loper Bright, the majority acknowledges that Congress can expressly delegate discretion to agencies. Even Justice Thomas, who writes separately to state clearly his view that Chevron deference is  unconstitutional, apparently accepts the idea of express delegation. After all, he concurs in full with the majority opinion. But what is an express delegation?

Suppose the following two statutes:

(A) The EPA shall set permitting standards for stationary sources of air pollution.

(B) The EPA shall set permitting standards for stationary sources of air pollution and is hereby granted discretion to determine what fraction, up to the whole, of a geographic area to count as a single stationary source.

Statute (A) is a simplified version of the statute in Chevron itself. If the case were to arise for the first time today, Loper Bright entails that the courts themselves--and not the EPA--would decide what counts as a stationary source. Either the Carter administration approach (each smokestack is a source) or the Reagan approach (the whole plant is a single source) or neither, but not both, could qualify. However, under Statute (B), the EPA could adopt either approach because it's an express delegation.

But now suppose the following variation:

(C) The EPA shall set permitting standards for stationary sources of air pollution and, insofar as the term "stationary source" is ambiguous, the EPA is hereby also granted discretion to define it more precisely by deciding what fraction, up to the whole, of a geographic area to count as a single stationary source.

Is (C) invalid under the version of Loper Bright that depends on Marbury because it grants to a federal agency the power to say what the law is? Justice Thomas says in his concurrence that the problem with Chevron deference is that it violates separation of powers: it gives to agencies a power that belongs with judges. If that really is a problem, then Congress can't cure it by expressly granting agencies a power that belongs to judges.

But how can (C) be unconstitutional when it is functionally and almost literally identical to (B), which every Justice in the Loper Bright majority--including Justice Thomas--is committed to accepting as a valid express delegation? The only difference between (B) and (C) is that in (C) the statute conceptualizes the grant of discretion to the agency as authority to decide the meaning of stationary source. It's the same exact discretion, however.

What I believe these examples show is that the line between the meaning of a statute--which the Marbury principle ostensibly reserves to judges to decide--and the application of a statute to various particular contexts--which every Justice in the Loper Bright majority allows can be assigned to agency discretion--is fuzzy and highly dependent on how Congress words its statutes. Thus, acceptance of the proposition that agencies can legitimately exercise delegated discretion in how to apply statutes entails that they can effectively say what the law is.

To be sure, I can imagine that the Roberts Court might actually say that (B) is permissible but (C) isn't. If it were to do so, however, that would be an acknowledgment that the reservation to the judiciary of the power to say what the law is as a basis for rejecting Chevron deference rests on empty formalism.

I want to add a further and separate reason to reject the view that the say-what-the-law-is function of the judiciary rules out deference to agencies: the legitimacy of Thayerism. Thayer thought that he was describing constitutional practice as deferential--as authorizing judicial invalidation of statutes only when they were clearly unconstitutional. Whether that was an accurate or inaccurate account of the first century of American constitutional law, as Cass Sunstein has explained, Thayer was highly influential for many decades. Even today, some of the most revered Justices (like the second Justice Harlan) are praised precisely for their practice of Thayerian restraint--i.e., for their deference to other bodies. And while Thayerism has fallen out of fashion, that may be because it is difficult for judges to restrain themselves (as Sunstein argues), not because Thayerism is seen as illegitimate.

Yet neither Thayer nor those who followed in his footsteps thought his clear-incompatibility standard for invalidation of legislation violated Marbury. On the contrary, they understood it as what Marbury entailed. Thayerism thus validates the idea that the judicial province and duty to say what the law is can co-exist with a practice of deferring to non-judicial actors.

Indeed, in Marbury itself, Marshall recognized a category of "cases in which the Executive possesses a constitutional or legal discretion" in which "their acts are only politically examinable." That statement is the root of the modern political question doctrine, under which courts grant complete deference to the decision of legal questions--such as whether an impeachment trial satisfied the constitutional requirements--to the determinations of other branches. Whatever else is entailed by the judiciary's province and duty to say what the law is, it does not rule out deference.

QED.