Corbin K. Barthold is Internet Policy Counsel at TechFreedom and previously served as Senior Litigation Counsel at Washington Legal Foundation.

Prodded by three witches and his wife, a man murders a king and takes his crown. Things head south, and the man is overthrown in his turn. Good plots can be simple. Good themes often aren’t. We know what happens in this play, but what is it about? Fate? (The weird sisters’ prophecies are fulfilled, after all.) Moral order and the importance of station? (Indeed, the divine right of kings?) What are we to make of MacBeth himself? Does his destruction warn us off ambition? Or is he simply too weak-willed for the task at hand? (“Now does he feel his title / Hang loose about him, like a giant’s robe / upon a dwarfish thief.”)

A statute empowers an agency to set “standards of performance” for certain power-plant emissions. Those standards, the statute says, must embody the “best” feasible “system for emission reduction.” For many years, the agency assumes that the statute lets it set standards that spur power plants to reduce emissions. But, prodded by activists and the president, the agency later tries to set “standards of performance” that require coal-fired power plants not to perform better, but to operate less.

We know what happens in West Virginia v. EPA, a 6-3 decision issued by the Supreme Court last week. The Court rules that Section 111(d) of the Clean Air Act does not permit the Environmental Protection Agency to turn power plants off. But what is the decision about? There are three opinions, written by Chief Justice Roberts, for the conservative majority; Justice Gorsuch, joined by Justice Alito, in concurrence; and Justice Kagan, joined by the other two liberals, in dissent. The justices touch on themes of the first importance, including the imprecision (and, thus, inherent contestability) of language, the inscrutability of the popular will, the complexity of modern society, and the limits of expertise. They debate the role of the judiciary, and of the administrative state, in a country whose legislature has collapsed into dysfunction. They dispute the nature of representative democracy. The vagaries of human governance. The shifting sand of politics and power.

Like a great play, West Virginia v. EPA deserves to be read and re-read. The decision features characters—the justices—who speak with eloquence and conviction. Their dialogue is about much more than the events at hand. Some of the questions they grapple with are, ultimately, indeterminate. They are eternal.

The night has been unruly; some say the earth was feverish and did shake. It will take time to unpack the meaning of this remarkable ruling. Let us start, though, by spotting a few of the threads.

Congress—the Justices’ Imaginary Friend

Much of West Virginia v. EPA revolves around stories the justices tell themselves about Congress. If the justices were characters in a novel, Congress would be the family patriarch who dies before Chapter I. His wife insists that he was a loving and generous husband. He was a cold and malicious father, retorts his eldest son. Different experiences, faulty memories, motivated reasoning, and the man’s very real contradictions drive the plot forward.

 What does Congress—a corporate body of 535 people—“intend” when it passes a law? What’s its “vision” for how that law will operate in fifty years? According to Justice Kagan, Congress wrote Section 111(d) “to give EPA plenty of leeway.” In her view, “Congress understood—it had to—that the ‘best system’” for emission reduction “would change over time.” She insists that “Congress wanted and instructed EPA to keep up.” But even if the legislators who passed Section 111(d) understood that the best ways to make power plants perform better would evolve, who says they equated perform better and operate less? And who says they would have wanted an “operate less” approach deployed, via Section 111(d), to restructure the energy industry? As the majority observes, one senator later called Section 111(d) an “obscure, never-used section of the law.”

The Clean Power Plan—the cap-and-trade program containing the Obama-era EPA’s “operate less” command—went after coal-fired power plants as a means of reducing emissions of carbon dioxide. The concern, of course, is that carbon dioxide contributes to the global, and atmospheric, problem of climate change. Signed into law by President Nixon in 1970, Section 111(d) aims to combat “air pollution.” The issue of the day was literal dirty air. Would Congress, or Nixon, have considered carbon dioxide an “air pollutant”? Would they have accepted that Section 111(d) enables regulation, not just of the air (local), but of the climate (global)? There was a time when the EPA thought not. As then-Solicitor General Paul Clement explained in 2006, George W. Bush’s EPA “determined that ‘[greenhouse gases], as such, are not air pollutants under the CAA’s regulatory provisions, including . . . [section] 111,’” and that “‘the term ‘air pollution’ as used in the regulatory provisions cannot be interpreted to encompass global climate change.” Kagan’s claim that Congress “understood” Section 111(d) would be used to combat climate change, or that Congress “wanted” or “instructed” the EPA to adopt something like the Clean Power Plan, is a story. It is a contestable (though convenient) account of what the broadminded legislators living in Kagan’s head “intended.”

The majority, for its part, tells its own tale. “We presume,” declares Chief Justice Roberts, quoting a dissent by then-Judge Kavanaugh, “that ‘Congress intends to make major policy decisions itself, not leave those decisions to agencies.’” At least the majority here merely “presumes” what Congress “intends”—an implicit concession that the justices are weaving a fiction. In any event, notice how the majority’s and the dissent’s stories conflict. Whereas Kagan’s legislators give agencies “leeway,” Roberts’s legislators do not. They take charge. They make tough decisions. In Roberts’s mind, it is “likely” that Congress reserves choices about “basic and consequential tradeoffs . . . for itself.” In Kagan’s mind, “members of Congress often don’t know enough—and know they don’t know enough—to regulate sensibly on an issue,” and so they delegate to others.

These divergent pictures of Congress shape the justices’ conflicting claims about textualism. Roberts’s careful, competent legislators spell out what they mean. When they don’t, therefore, the Court may assume, as part of construing the text, that Congress doesn’t make big pronouncements in imprecise ways. Kagan wants none of that. Her busy (but beneficent) legislators have goals—“general objectives,” in her words—but only hazy notions of how to achieve them. If they get within the ballpark (“best system”), the Court’s role is to assist them (“cap-and-trade? sure, that fits”). The majority’s demands for precision are, in this telling, not textualism. On the contrary, Kagan calls them “get-out-of-text-free cards.” They’re obstructionism. “The current Court is textualist only when being so suits it,” she complains. The Court’s real goal, in her eyes, is to “prevent agencies from doing important work, even though that is what Congress directed.”

Major Questions About Major Questions

West Virginia v. EPA is a landmark decision on the major questions doctrine. “In certain extraordinary cases,” the majority says, a court should be “reluctant to read into ambiguous statutory text the delegation claimed to be lurking there.” In such cases, “more than a merely plausible textual basis for the agency action is necessary.” Instead, an agency “must point to clear congressional authorization for the power it claims.”

Justice Kagan says that the majority “announces the arrival of the ‘major questions doctrine.’” True enough. The name, at least, is of recent vintage. But as Chief Justice Roberts explains, it is a name that captures the Court’s approach to a series of cases in which “agencies assert[ed] highly consequential power beyond what Congress could reasonably be understood to have granted.” The Court has tended to “greet” such assertions “with skepticism.” In his concurrence, Justice Gorsuch finds that major questions and the administrative state were born twins. The earliest case he can find, from 1897, involved the Interstate Commerce Commission—by some accounts the first administrative agency. (The ICC claimed to possess rate-setting authority, and lost. “That Congress has transferred such a power to any administrative body,” the Court wrote, “is not to be presumed or implied from any doubtful or uncertain language.”) Major-questions cases have become more frequent in recent years, but so have agency assertions of far-reaching power.

What does West Virginia tell us about major questions? For one thing, the principle is separate from Chevron deference. Previous decisions had left this point unclear. Take King v. Burwell (2015). “Under [the Chevron] framework,” Burwell said, “we ask whether the statute is ambiguous”—that’s Chevron step one—“and, if so, whether the agency’s interpretation is reasonable”—Chevron step two. Like West Virginia, Burwell went on to discuss “extraordinary cases.” Unlike West Virginia, however, Burwell found merely that in such cases, the Court should hesitate before proceeding to Chevron step two. Not affording an agency’s interpretation a presumption of reasonableness is one thing. Demanding that the agency “point to clear congressional authorization” is quite another. But Roberts and the West Virginia majority never once mention Chevron. So the former approach is out, and the latter approach is in.

We have also learned that major questions stands on both “separation of powers principles” and “a practical understanding of legislative intent.” This is a blurring of lines that were once a little more distinct. Dissenting in Gundy v. United States (2019), Gorsuch argued that the Court, by letting Congress delegate legislative power to agencies, has distorted the separation of powers. That “nondelegation principle” differs, Justice Kavanaugh noted the following term, from major questions, which, although “closely related” to nondelegation, is itself—Kavanaugh asserted—a “statutory interpretation doctrine.” Kavanaugh declared that Gorsuch’s broad nondelegation rule would leave no room for major questions—the idea being that, if Gorsuch had his druthers, a delegation of a major question would by definition be an unconstitutional delegation of legislative power.

Here is how Kavanaugh put it: “[The Gundy dissent] of Justice Gorsuch would not allow . . . congressional delegations to agencies of authority to decide major policy questions—even if Congress expressly and specifically delegates that authority.” But West Virginia establishes that things aren’t so straightforward. Consider this critical sentence at the end of Roberts’s opinion: “A decision of [great] magnitude and consequence rests with Congress itself, or an agency acting pursuant to a clear delegation from that representative body” (emphasis mine). Gorsuch signed on to this sentence, a line that contradicts Kavanaugh’s description of the Gundy dissent. Not only does Gorsuch endorse major questions in West Virginia; he devotes much of his concurrence to bolstering the doctrine’s constitutional pedigree.

This leads to the question of where major questions ends and nondelegation begins. Imagine that Section 111(d) granted the EPA the power to impose the “best feasible system for combating climate change.” That would surely count as “clear congressional authorization” for the Clean Power Plan. Yet one pictures Gorsuch objecting that the phrase contains within it questions so major that Congress may not delegate them to others. Where does he draw the lines between an issue that may go to an agency through “normal” delegation, one that may go to an agency only through “clear delegation,” and one that may not go to an agency at all? In Gundy Gorsuch wrote that “the exact line between policy and details” has “invited reasonable debate.” In West Virginia he reiterates that “what qualifies as an important subject and what constitutes a detail may be debated.” He has yet to flesh out the distinction. Perhaps he won’t. He might never be satisfied that a congressional statement clearly passes a major question to an agency. (“Alright, this new statute says, ‘The EPA shall impose a cap-and-trade program to combat climate change.’ But what does ‘cap-and-trade’ really mean?”)

On that note, no one in West Virginia tells us much about the contours of major questions. It falls to Gorsuch, in concurrence, to produce a list—one that “may not be exclusive”—of elements that trigger the doctrine. Questions of “great political significance” or “earnest and profound debate.” Measures that Congress has “considered and rejected” in other bills. Plans of great economic impact. Intrusions on state authority. But how important, how contested, how poorly received, how costly, how intrusive? We are not told. Similarly, the majority instructs that “clear authorization” is not to be found in “the vague language of an ancillary provision.” And it’s a bad sign if the agency has changed its reading of a law, or “discovered” new authority in an old statute. And the power the agency asserts must fit that agency’s statutorily assigned role. But explaining how these factors would interact in a clear-authorization test is a task left for another day.

The majority has done well to clarify “that the approach under the major questions doctrine is distinct,” and to outline how that doctrine operates. But there is a lot we still don’t know. The signs that a major question is present Kagan derides as “some panoply of factors.” And “clear authorization” is, as best she can tell, to be found “someplace over and above the normal statutory basis we require.”

War in the Kludgeocracy

About a decade ago, the political scientist Steven Teles started using the word kludgeocracy to describe American government. “The term,” Teles explains, “comes out of the world of computer programming, where a kludge is an inelegant patch put in place to be backward compatible with the rest of the system.” Solving problems with duct tape and spare parts meets the needs of the moment. But “when you add up enough kludges,” Teles warns, “you get a very complicated program, one that is hard to understand and subject to crashes.”

Our system of government declined into kludgeocracy for a number of reasons. The key problem, though, is that the legislature won’t legislate. Our political parties are highly polarized. Fewer moderates in Congress means less cross-party cooperation, which means less negotiation and compromise—less lawmaking. We now have what Jonah Goldberg calls a parliament of pundits. Politicians don’t hash out the fine points of legislation in committee; they vilify the opposition on podcasts, social media, and cable news.

In our constitutional order, this qualifies as a major malfunction. The framers of that order set up a pipeline: Congress makes the rules → the president applies the rules → the courts interpret the rules when disputes arise. Remove the first piece, and none of the rest coheres. So when new societal problems arise, but Congress won’t rise to meet them, what are the other two branches to do? Resort to kludges, that’s what.

The facts behind West Virginia v. EPA are an operatic display of kludgeocracy in action. “I’ve got a pen, and I’ve got a phone,” President Obama said. He vowed to work around Congress. Climate change was one of the areas where he made this intention explicit. “If Congress won’t act soon to protect future generations,” he proclaimed during the 2013 State of the Union Address, “I will. I will direct my cabinet to come up with executive actions we can take . . . to speed the transition to more sustainable sources of energy.” By that point, it was already obvious that Congress would not pass a cap-and-trade bill. Accordingly, a few months after the speech, Obama directed the EPA to issue what would become the Clean Power Plan. An arcane provision of an old law would be used by unelected administrators to address a new (or newly recognized) problem.

Not surprisingly, lawsuits ensued. Dozens of parties, and more than half the states, went straight to court. The D.C. Circuit declined to block the Clean Power Plan. The Supreme Court overruled. Then President Trump took office, and his administration replaced the Clean Power Plan with something called the Affordable Clean Energy Rule. Many parties and states challenged that measure. Other states and entities intervened to defend it. The D.C. Circuit vacated the Trump administration’s rule and undid the repeal of the Clean Power Plan. Several parties petitioned the Supreme Court for review. In the meantime, President Biden took office. His administration wanted, not to reimpose the Clean Power Plan, but to issue an updated (and likely more stringent) program. It sought to dismiss the proceedings, but the Court rejected that request and issued West Virginia. This is what kludgeocracy looks like.

Underneath the justices’ opinions is an unacknowledged argument over the kludgeocracy’s effectiveness and validity. “If the current rate of [carbon] emissions continues,” Justice Kagan informs us at the outset of her dissent, “children born this year could live to see parts of the Eastern seaboard swallowed by the ocean.” A crisis is afoot. In response, the EPA has cast the dusty and somewhat cryptic words of Section 111(d) in a new light. So be it. The officials “found in agencies” have “greater expertise and experience” than our legislators, Kagan assures us, and they are here to help. Those who should be wielding power are failing to do so? No matter. “A rational Congress delegates,” Kagan intones. “It enables an agency to adapt old regulatory approaches to new times.” She is sure of this. So sure that she repeatedly questions the motives of the six justices who disagree with her.

The majority and the concurrence persist in defending the original order’s legitimacy. “Agencies have only those powers given to them by Congress,” Roberts reminds us. Crisis or no—for that matter, functional Congress or no—agencies do not rule over us. We live, Gorsuch cries, in “a republic—a thing of the people.” Like it or not, that’s the system. “Admittedly,” he concedes, “lawmaking under our Constitution can be difficult.” That’s not an excuse for technocracy.

It’s true that the framers made lawmaking hard by design. But would they approve of the way Congress operates today? Roberts, Gorsuch, and Kagan seem unwilling to admit that there’s a problem. Roberts swears that major questions arises only in “extraordinary cases,” yet he protests, almost in the same breath, that agency overreach is a “recurring” issue that arises in “all corners of the administrative state.” A modern-day Canute, Gorsuch plans to stem the rising tides by reciting passages of Hamilton and Madison at them. As for Kagan, it would be amusing to trundle her to a time machine, launch her back to 1970, and watch her lecture Congress about what it “intends” to do to tackle global climate change.

“Permitting Congress to divest its legislative power to the Executive Branch,” Gorsuch states, “would dash” the “whole [constitutional] scheme.” The “would” in that sentence is a rhetorical affectation. In West Virginia, the justices fight an undeclared civil war over how power is to be shared, as between the judiciary and the administrative state, within our heavily damaged (if not quite “dashed”) framework. Gorsuch is right about this much: Until Congress gets back to work, “stability w[ill] be lost, with vast numbers of laws changing with every new presidential administration,” and “powerful special interests . . . w[ill] flourish.”

Expertise Versus Politics

Before West Virginia v. EPA was handed down, I wrote in this space that the Court should seek, not to cripple government experts, but merely to keep them focused on applying their special knowledge. “We want experts to stick where possible to addressing difficult technical matters,” I said, “rather than subjects where, as Aristotle put it, ‘certainty is impossible and opinions are divided.’” As they seek to cabin the power of the administrative state, “the justices’ goal . . . should be not to ‘burn it all down,’ but rather to limit administrators’ ability to make nakedly political decisions under the guise of expertise.” By this standard, West Virginia is a great decision.

Justice Kagan agrees that an agency should not stray “out of its lane, to an area where it ha[s] neither expertise nor experience.” But she puts too much stock in what expertise can accomplish. At the end of her dissent, she praises workplace safety standards, automobile airbag requirements, and food plant inspections. But these are fit subjects for expert-guided government regulation. They are, generally speaking, not topics of “great political significance” or “earnest and profound debate.” They don’t resemble the case before her.

The majority has a better grasp of this distinction. “On EPA’s view of Section 111(d),” Roberts observes, Congress “tasked it, and it alone, with balancing the many vital considerations of national policy implicated in deciding how Americans will get their energy.” Yet the “balancing” of those “considerations” is not, at root, a matter of expertise. It is a matter of politics. No intellectual procedure can tell us how to weigh affordable energy against the health of the climate. To figure out what to do, we must have a debate over values. We must use evidence and logic, but also rhetoric and persuasion and moral appeals, to argue about how to proceed in a situation where certainty is impossible and opinions are divided.

When the Frankish general Arbogastes usurped power in the West, the Emperor Theodosius (r. 379-95) had to decide whether to go to war to reunify the Empire. So he sent a eunuch to Lycopolis, a remote town in Egypt, with instructions to find John, a hermit renowned for his clairvoyance, and seek his counsel. As Gibbon recounts in the Decline and Fall:

In the neighborhood of that city, and on the summit of a lofty mountain, the holy John had constructed with his own hands an humble cell, in which he had dwelt above fifty years, without opening his door, without seeing the face of a woman, and without tasting any food that had been prepared by fire or any human art. Five days of the week he spent in prayer and meditation, but on Saturdays and Sundays he regularly opened a small window, and gave audience to the crowd of suppliants who successively flowed from every part of the Christian world. The eunuch of Theodosius approached the window with respectful steps, proposed his questions concerning the event of the civil war, and soon returned with a favourable oracle, which animated the courage of the emperor by the assurance of a bloody but infallible victory.

Now, you might object that this is an insane way to address matters of war and peace. Then again, there are only so many options for dealing with imponderables. Democratic hurly-burly beats hermits atop hills—but not by much. Theodosius—who fought and won his war—had no way to discover the “correct” answer to pressing political questions. And neither do we.