“The Game” (or How I Learned to Stop Worrying and Love the Major Questions Doctrine) – Gary Lawson
“The Game” (or How I Learned to Stop Worrying and Love the Major Questions Doctrine)
Gary Lawson*
I have been dubious about the so-called major questions doctrine[1] ever since it emerged as a theorized concept in scholarly discourse in the early 2000s. To be clear: I am not claiming that the practice represented by cases thought to exemplify the major questions doctrine is that novel. It is not.[2] But its theorization as a distinct doctrine dates back only about two decades.[3]
Once the major questions idea took shape, I started experiencing the stages of grief. First came denial: maintaining that there was no such doctrine and that the cases supposedly reflecting it were simply instances of ordinary statutory interpretation. “[A]ll of the relevant cases,” I shouted, “can better be explained on more mundane grounds without positing a free-floating but unstated ‘major issues’ inquiry.”[4] For example, if one looked hard enough at the Food, Drug, and Cosmetic Act as it stood in 2000, one perhaps could glean an implicit tobacco exception to the seemingly categorical definitions of “drug” and “device” that, at first glance, appeared to sweep in tobacco and tobacco products[5]–and perhaps could even glean that exception strongly enough to establish a “clear” meaning of the statute.[6] But one would get there by careful reading of the statute and other relevant enactments, not by asking whether tobacco regulation was important. Of course tobacco regulation was important. Lots of things in the regulatory world are important. Sometimes statutes resolve important things, sometimes statutes resolve trivial things, and sometimes statutes resolve next to nothing. One has to read the statutes to figure out what is going on in any specific instance. All of the so-called “major questions” cases, I staunchly maintained for years, could be explained in that straightforward fashion.
In 2015, denial began looking a bit desperate,[7] and by 2022 it was downright irrational.[8] Yes, Virginia, there is a major questions doctrine. So at that point I moved on to anger.
The major questions doctrine makes no sense as a tool for ascertaining the meaning of law. The notion that Congress universally wants to reserve for itself the resolution of really important matters has never seemed plausible to me as a matter of either anecdotal or theoretical political science. I suppose that makes me an advocate, at least to some extent, of what Joe Postell’s contribution to this symposium calls the abdication hypothesis.[9] A casual glance through the United States Code makes clear, on the face of the statutes, that Congress frequently has no desire to go anywhere near major questions and is more than happy to punt their resolution to executive and judicial actors. Certainly, as Professor Postell points out, sometimes that is not true and Congress truly wants to take on tough and important tasks.[10] But the major questions doctrine overgeneralizes those instances, which can be identified through ordinary statutory interpretation. There is no reason to think that those times when Congress wants to take rather than avoid responsibility can be picked out by generalizations about the importance of the underlying issues. Nor, as Mike Rappaport ably shows in his contribution to this symposium,[11] does it make sense to use the major questions doctrine as a substantive canon of constitutional avoidance. Putting aside the rather large concerns that surround substantive–i.e., meaning-shaping rather than meaning-ascertaining—canons in general[12]: While it may seem tempting to try to protect the Constitution and its underlying principle against legislative subdelegation[13] from a Congress that “gives not a fig for the Constitution,”[14] the way to protect the Constitution is to read and apply it, not to mangle statutes to avoid reading and applying it. There can be occasions when two wrongs make a right, but most of the time two wrongs make two wrongs. Misreading a statute in order to generate a misreading via underenforcement[15] of the Constitution is not something that facially commends itself.
In this article, I am going to skip the stages of bargaining and depression and move straight to acceptance–or at least something vaguely resembling acceptance. I still think that all of my criticisms of the major questions doctrine are sound. But I now suspect that they might be misdirected. While the major questions doctrine is a poor way to ascertain the meaning of the law, that is somewhat beside the point, because the major questions doctrine is not a doctrine designed to ascertain the meaning of the law. It is a doctrine designed to decide cases, and it might be a perfectly fine way to decide cases even if it is not a fine way to ascertain legal meaning. Those are very different things.[16]
The key to understanding this difference, and thus to understanding the real function of the major questions doctrine, is to pose a basic question (which, to give away the ending, is soon going to lead to an even more basic question): What are courts supposed to do when they decide cases?
Federal courts, which are the entities creating and applying the major questions doctrine, exercise “[t]he judicial Power of the United States.”[17] Just as with the “legislative Powers herein granted”[18] vested in Congress by Article I and the “executive Power”[19] vested in the President by Article II, the Constitution vests the “judicial Power” in federal courts without telling us what that vested power is. Given the relative novelty of the “judicial Power” as a distinct governmental power in 1788,[20] that is a noteworthy omission. Nor is it easy to find definitions of judicial power in the founding era. Perhaps the clearest definition was James Wilson’s brief but elegant formulation, identifying the judicial power as courts “applying, according to the principles of right and justice, the constitution and laws to facts and transactions in cases, in which the manner or principles of this application are disputed by the parties interested in them.”[21] That seems like a fair summary of the underlying eighteenth-century assumptions about judicial power.[22] Thus, an obvious (at least to an originalist) answer to the question “What should courts do?” is to say that they should ascertain the relevant law and facts and use those ascertained results to reach a decision.
But while that answer may be obvious, it is not necessarily right, because the question to which it is responding is not actually the real question one needs to ask. Wilson aptly described a way, and perhaps even the normal way, for courts to proceed. Sometimes that is the right way for courts to proceed. But it not the only way for courts to proceed, and it is not always the best way. Wilson’s comment elides something that has to be answered before one even gets to the question how courts should proceed: Why do courts exist at all? What is their essential function? To know how courts should act, one needs to know why courts should act. As with architecture and interpretation, form follows function.[23]
If one both starts and ends with Wilson’s account, one might say that courts exist in order to ascertain the relevant law and facts. But the notion that courts exist primarily to ascertain the meaning of law is a fallacy that comes from Yale–if putting “fallacy” and “Yale” into the same sentence is not redundant. It is the position articulated most clearly by Owen Fiss, who insisted that “the function of the judge . . . is not to resolve disputes, but to give the proper meaning to our public values,”[24] and that the judge’s job “is not to maximize the ends of private parties, nor simply to secure the peace, but to explicate and give force to the values embodied in authoritative texts such as the Constitution and statutes: to interpret those values and to bring reality into accord with them.”[25] If the be all and end all of the judicial process is truly to declare the meaning of the law – either its actual meaning or its Yale-ized Fissian meaning — then of course judges should try to declare the law correctly, and that is a task for which the major questions doctrine is decidedly ill suited.
But that is a Yale point of view. A better account of the fundamental role of courts comes from Harvard—hence the reference to “The Game,” which in this instance Harvard clearly wins—in the person of Lon Fuller. While it is easy to overstate the extent to which Fuller discounted a law-declaring function for courts,[26] he has become associated with the idea that adjudication is “a means of settling disputes or controversies.”[27] Courts on this model exist to resolve disputes, so that parties do not find their own, possibly more violent, means of dispute resolution.[28]
Fuller’s actual position on the role of courts was closer to Fiss’s than posterity has sometimes assumed,[29] but perception is often more important than reality. Just as it is meaningful to speak of “Lockean” ideas that have little to do with the actual writings or thought of John Locke or “Marxist” ideas that Karl Marx would not recognize,[30] it is meaningful to speak of a dispute-resolution model of courts as “Fullerian” even if Lon Fuller would not have endorsed it. In any event, once one understands that courts exist in order to resolve disputes, the role of law ascertainment and declaration in that process becomes complicated. Sometimes that law-declaration function will be crucial. Sometimes it will not.
One possible way to resolve a legal dispute is to ascertain the meaning of relevant legal rules and apply them. But it is not the only way, and it is not always the best way. On many occasions, the best way to resolve a dispute is to avoid ascertaining the meaning of the relevant law.
The most obvious example of law-meaning-avoidance is the classical rule of lenity–and here I mean the classical version, not the watered-down non-version that modern courts purport to apply.[31] The classical version says that no one should be criminally convicted unless the law clearly and unambiguously establishes that the defendant’s conduct violates the law.[32] While that classical rule of lenity can be used as a way of ascertaining legal meaning, by giving a criminal statute the defendant’s preferred meaning unless the government can convince the court otherwise to a high degree of confidence,[33] it is possible to understand the rule as way to decide a case without offering a specific interpretation of the underlying statute.
Think about the proof of facts in a criminal case. Does the fact-finder (whether jury or judge) need to find certain facts in favor of the defendant in order to reach a verdict of “not guilty”? Of course not. All the fact-finder needs to conclude is that the prosecution has not sufficiently proven its version of the facts. The fact-finder does not have to reach any conclusion about what actually happened in the world; it merely has to conclude that it is not sufficiently persuaded that the prosecution’s account is accurate.
One could say the same thing about legal meaning. A court applying the classical rule of lenity can decide a case without saying anything definitive about what the law means. The court only needs to say that the prosecution has failed to establish with sufficient clarity that the law means what the prosecution says it means. That does not establish the defendant’s preferred view as correct. It does not establish any particular interpretation of the statute as correct. It simply reflects the low degree of confidence that the court has in an interpretation that would justify conviction.[34] In essence, the classical version of the rule of lenity treats all criminal cases as “major questions” that require a clear statement from the legislature before statutes are given legal effect in particular cases.
The major questions doctrine that has emerged since the October 2021 Supreme Court term can be thought of in the same way. It is not – or at least does not have to be — a tool for ascertaining legal meaning. It is‑‑or at least can be‑‑a burden of proof rule that makes the government show its authority to act in a certain way, coupled with a standard of proof rule that makes the government show more than merely a plausible argument for its position. Perhaps the government must even show by more than 51-49 that its reading of the supposedly empowering statute is the best reading. The major questions doctrine, conceived in this fashion, is a civil rule of lenity on behalf of subjects of legislative power.
When the major questions doctrine is viewed in this light, the interpretation-based criticisms of the doctrine go away–not because they are wrong but because they are not relevant. The object of the major questions doctrine thus conceived is not to get the correct meaning of the law. The object is to get the correct decision of the specific dispute before the court, which might or might not require or involve getting the correct meaning of the law. That changes the game dramatically. The criteria for choosing decision-making tools are not necessarily the criteria for choosing meaning-ascertaining tools. There might be some overlap, to be sure, but they are not the same things.
If this is even close to the mark, attention then must shift to whether the major questions doctrine makes sense as a decision-making tool. It might or might not, depending on a lot of things.
First, it will only make sense if it is in fact treated as a decision-making tool. That is, the major questions doctrine on this understanding cannot be employed to “fix” the meaning of a statute. Quite to the contrary, the doctrine makes sense only if it avoids definitively addressing the meaning of a statute. A well formulated major questions decision amounts to something like: “OK, agency, maybe the statute means what you say it means. But you haven’t proved to us with sufficient clarity that it means what you say it means. And absent that demonstration, we as a court cannot give legal effect to your action. Come back next time with better proof and maybe you will win, but your proof this time doesn’t measure up.”
Second, the doctrine will only make sense if it makes sense. Is a rule of lenity for civil cases that present major questions a doctrinally and/or normatively sound basis for decision?
The normative question, being a normative question, is outside the scope of legal analysis, and I accordingly have nothing intellectually interesting to say about it at a deep level. If I thought I had intellectually interesting things to say about normative questions, I probably would not be publishing them in short articles in law reviews.[35] Superficially, I simply note that it implements what Randy Barnett has called a “Presumption of Liberty.”[36] As a libertarian in my non-scholarly guise, that strikes me as a fine presumption. But people who like to order other people around will disagree, and, as noted above, that is a dispute about moral and political theory that is unlikely to be resolved in a few pages in a law review article.
Doctrinally, the question is whether the United States Constitution in fact embodies something like Barnett’s presumption of liberty, in which case the major questions doctrine, as I have described it, is constitutionally grounded. For reasons that I have tried to defend at some length elsewhere,[37] I think there is a good case to be made that the Constitution embodies something vaguely resembling that presumption with respect to the federal government, though not necessarily with respect to state governments.[38] Since the major questions doctrine applies only to federal action, that limited presumption is enough to carry the day.
If all (or perhaps even a significant portion) of the foregoing is true, the biggest problem with the major questions doctrine is that is limited to major questions. If there truly is a presumption against the validity of federal governmental action, then the broad-based rule of lenity applies to all exercises of federal power, whether civil or criminal and whether major or minor. What is needed, on those assumptions, is not a major questions doctrine but a questions doctrine.
Perhaps, however, one can view the major questions doctrine as, like the old joke about 10,000 lawyers at the bottom of the Atlantic, a good start. If one thinks that the Constitution calls in the long run for a “questions doctrine,” is it sensible to start with major questions?
Sure. Why not? If one is building Rome, anticipating that it will take more than a day, it makes sense to start with things that are important, like water, farms, roads, and basic infrastructure, leaving hair salons and pubs for later. Similarly, if one is really trying to get the Constitution right, and if one believes that the Constitution embodies a presumption against the exercise of federal power, there are reasons to start with the big stuff, leaving the smaller stuff for later. To be sure, that is a pragmatic rather than strictly legal judgment, and I am not saying definitively that it is the right approach. That would be the kind of normative judgment that I have spent four decades studiously avoiding in scholarly work. But it is at least a plausible position to take. It means that the major questions doctrine is not something to dismiss out of hand.
A trickier question, pertinent to this panel, is whether environmental and civil rights statutes are especially ripe for a Lawsonian major questions doctrine. The possible case for “yes” turns on two considerations, neither of which I can develop here but which might be worth developing elsewhere. One is that those are fields where majorness is easy to find. Agency action in those areas often–not always, but often–has sweeping nationwide effects. If the FTC tries to regulate breakfast cereal advertising, that is not trivial and can affect a lot of people, but it is not akin to the EPA trying to remake the power grid. And if the federal government tries to make schools let teenage boys into the girls’ shower rooms, that is a societal transformation a tad larger even than making fishermen pay for federal monitoring of their boats. None of this is a knock-down argument in favor of categorically treating all environmental or civil rights actions as major questions, but it does suggest that those are likely to be fruitful areas for finding them.
Will the major questions doctrine coming out of the Court bear any resemblance to the rule of lenity approach that I have suggested here? The answer may well be no, in which case I may return to anger, or at least revisit the stages of bargaining and depression. But there is at least a way to think about the major questions doctrine that makes it a potentially valuable tool for decision-making.
* William Fairfield Warren Distinguished Professor, Boston University School of Law. I am profoundly grateful to the Pacific Legal Foundation for hosting a symposium on “Doctrinal Crossroads: Major Questions, Nondelegation & Chevron Deference” at Harvard Law School on January 25, 2024, from which this article comes. I am especially grateful to my fellow panelists Hon. Paul Matey, Alison Somin, and Luke Wake and to the engaging audience at the symposium for thoughtful commentary. Many of the conference participants, most notably Lou Capozzi and Mike Rappaport, have helped shape my thinking on the major questions doctrine over the past few years. Needless to say, I am solely responsible for the carnage that follows.
[1] West Virginia v. EPA, 597 U.S. 697, 723 (2022).[2] Louis J. Capozzi III, The Past and Future of the Major Questions Doctrine, 84 Ohio St. L.J. 191, 197 (2023).
[3] That theorization emerged more or less simultaneously from Jacob Gersen and Cass Sunstein in 2006, see Jacob E. Gersen, Overlapping and Underlapping Jurisdiction in Administrative Law, 2006 Sup. Ct. Rev. 201, 226; Cass R. Sunstein, Chevron Step Zero, 92 Va. L. Rev. 187, 194, 236 (2006); Cass R. Sunstein, Beyond Marbury: The Executive’s Power to Say What the Law Is, 115 Yale L.J. 2580, 2605–06 (2006), though it really grabbed my attention a few years later via Abigail R. Moncrieff, Reincarnating the “Major Questions” Exception to Chevron Deference as a Doctrine of Noninterference (or Why Massachusetts v. EPA Got It Wrong), 60 Admin. L. Rev. 593 (2008).
[4] Gary Lawson, Federal Administrative Law 673 (7th ed. 2016).
[5] 21 U.S.C. § 321(g)(1)(C) (1994) (defining a drug as “articles (other than food) intended to affect the structure or any function of the body”); id. at § 321(h) (defining a device as “an instrument, apparatus, implement, machine, contrivance . . . or other similar or related article, including any component, part, or accessory, which is . . . intended to affect the structure or any function of the body”). Sounds a lot like tobacco and tobacco delivery systems, right? But for how ordinary statutory interpretation, without any substantive canons, might nonetheless indicate that the statute did not cover tobacco, see Gary Lawson, Evidence of the Law: Proving Legal Claims 88-95 (2017). Today the issue is moot; the FDA now has express but limited authority over tobacco. See 21 U.S.C. 387a (2018).
[6] See FDA v. Brown & Williamson Corp., 529 U.S. 120, 126, 133 (2000).
[7] See King v. Burwell, 576 U.S. 473, 485-86 (2015) (talking about “‘extraordinary cases’” (quoting FDA v. Brown & Williamson Corp., 529 U.S. at 123) and adding that “[t]his is one of those cases”); see also Lawson, supra note 4, at 673–74
[8] See Alabama Ass’n of Realtors v. HHS, 141 S.Ct. 2485, 2489 (2021) (per curiam); Nat’l Federation of Independent Bus. v. Dep’t of Labor, OSHA, 595 U.S. 109, 117(2022); West Virginia v. EPA, 597 U.S. 697, 722–24(2022).
[9] See Joseph Postell, Does the Major Questions Doctrine Get Congress Right? 15 HARV. J.L. & PUB. POL’Y: PER CURIAM 1 (2024).
[10] Id.
[11] See Michael B. Rappaport, Replacing the Major Questions Doctrine with Originalist Statutory Interpretation, 17 HARV. J.L. & PUB. POL’Y: PER CURIAM 1 (2024).
[12] See, e.g., Lisa Heinzerling, The Power Canons, 58 Wm. & Mary L. Rev. 1933, 1942 (2017).
[13] This rather casually assumes that there is indeed a constitutional principle regarding legislative subdelegation, which is a hotly contested claim. Many important figures doubt whether any such principle ever existed. See, e.g., West Virginia v. EPA, 597 U.S. 697, 779–81 (Kagan, J., dissenting); Julian Davis Mortenson & Nicholas Bagley, Delegation at the Founding, 121 Colum. L. Rev. 277, 280 (2021); Eric A. Posner & Adrian Vermeule, Interring the Nondelegation Doctrine, 69 U. Chi. L. Rev. 1721, 1722 (2002). I cannot engage that question in my space here. For my most recent thoughts, which are somewhat different from my earlier thoughts, see generally Gary Lawson, A Private-Law Framework for Subdelegation, in The Administrative State before the Supreme Court: Perspectives on the Nondelegation Doctrine 123 (Peter J. Wallison & John Yoo eds., 2022)
[14] Gary Lawson, The Ghosts of Chevron Present and Future, 103 B.U. L. Rev. 1647, 1703 (2023).
[15]As Mike Rappaport points out, see Rappaport, supra note 11, a major questions doctrine might also overenforce the Constitution, depending on the precise content of the Constitution’s subdelegation principle and the precise formulation and application of the major questions doctrine.
[16] See Gary Lawson, Time, Institutions, and Adjudication, 95 B.U. L. Rev. 1793, 1798 (2015).
[17] U.S. Const. art. III, § 1.
[18] Id. art. I, § 1.
[19] Id. art. II, § 1, cl. 1.
[20] By the time of the framing, judicial power as a distinct governmental power was only a few decades old. For most of British legal history before that time, judicial power was an aspect of executive power. Steven Gow Calabresi & Gary Lawson, The U.S. Constitution: Creation, Reconstruction, the Progressives, and the Modern Era 145–46 (2020).
[21] James Wilson, Of Government, in 1 The Works of James Wilson 343, 363 (James DeWitt Andrews ed., 1896).
[22] For lengthy elaboration of this conception of judicial power, see generally Philip Hamburger, Law and Judicial Duty (2008).
[23] See Michael W. McConnell, On Reading the Constitution, 73 Cornell L. Rev. 359, 359 (1988).
[24] Owen M. Fiss, Foreword: The Forms of Justice, 93 Harv. L. Rev. 1, 30 (1979).
[25] Owen M. Fiss, Comment: Against Settlement, 93 Yale L.J. 1073, 1085 (1984).
[26] See Robert G. Bone, Lon Fuller’s Theory of Adjudication and the False Dichotomy between Dispute Resolution and Public Law Models of Litigation, 75 B.U. L. Rev. 1273, 1275 (1995).
[27] Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev. 353, 357 (1978).
[28] See id. at 372.
[29] See Fuller, supra note 27, at 368 (“The proper province of adjudication is to make an authoritative determination of questions raised by claims of right and accusations of guilt.”). A pure dispute-resolution model would say that the proper province of courts is to make an authoritative determination of claims of right or accusations of guilt, not an authoritative resolution of the questions raised by such claims or accusations.
[30] See Gary Lawson, The Ethics of Insider Trading, 11 Harv. J.L. & Pub. Pol’y 727, 763 n.149 (1988).
[31] See Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 298–99 (2012).
[32] See, e.g., U.S. v. Wiltberger, 18 U.S. (5 Wheat.) 76, 95, 104–05 (1820).
[33] See Jeesoo Nam, Lenity and the Meaning of Statutes, 96 So. Cal. L. Rev. 397, 408 (2022) (describing the rule of lenity as a semantic canon).
[34] For more on this account of decision-making, see Lawson, supra note 4, at 115–17.
[35] See Gary Lawson, Originalism Without Obligation, 93 B.U. L. Rev. 1309, 1309–10 (2013).
[36] Randy E. Barnett, Restoring the Lost Constitution: The Presumption of Liberty 5 (2004).
[37] See Gary Lawson, Legal Indeterminacy: Its Cause and Cure, 19 Harv. J. L. & Pub. Pol’y 411, 425–27 (1996).
[38] More precisely, the Constitution assigns the burden of proof to the federal government to show that it has power to act. No such constitutional burden of proof applies to state governmental action. (Although any state’s constitution could impose a similar rule if it so chose.) This does not lead precisely to Barnett’s presumption of liberty, because once the federal actor meets its burden of proof, the burden to show a countervailing “thou shalt not” shifts to the affected person. See id. at 426–28.