As most will recall, the Supreme Court has spent the last several years reinventing statutory interpretation in agency cases. That project has now been pithily summarized in the recent opinion in Seven County Infrastructure Coalition v. Eagle County: “As a general matter, when an agency interprets a statute, judicial review of the agency’s interpretation is de novo.” The Court cites to its own 2024 decision in Loper Bright Enterprises v. Raimondo, but even that opinion did not so clearly specify the standard of review as the Seven County Infrastructure Coalition opinion does.

The Court’s statement in Seven County Infrastructure Coalition therefore seems like a remarkably clear distillation of the prevailing anti-deference mood—except that the rest of the opinion goes on to explain why deference is the watchword when it comes to cases brought under the National Environmental Policy Act (NEPA). Like so many of the Court’s recent decisions, Seven County Infrastructure Coalition reveals the Court to be Janus-faced when it comes to operative (as opposed to rhetorical) anti-deference. And we may learn more about the interpretive regime by examining how the Court attempts to square its deferential posture towards NEPA reviews with Loper Bright.

A brief bit of background about NEPA is necessary to set the stage. Under the 1970 statute, agencies are often required to study the environmental impacts of their decisions (in many cases, the decision to approve some infrastructure project, like the rail line in Seven County Infrastructure Coalition). Under longstanding interpretation of the statute, agencies’ environmental reviews (some of which entail relatively more involved and formally designated “environmental impact statements,” or EIS’s) are reviewable by courts. Such review requires courts to assess whether the agency took a “hard look” at the environmental impacts that are reasonably foreseeable as a result of the decision being made. These reviews, and litigation over their sufficiency, have sometimes become fairly complex and lengthy affairs, which has led to an entire movement to reform NEPA in order to streamline infrastructure development. (As is abundantly clear from the numerous unsubstantiated statements about the effects of NEPA, the justices were no doubt aware of, if not even immersed in, this debate.) Whatever one thinks about that debate and the policy implications of the Court’s decision, the Court is of course not supposed to be a policymaker—the questions presented by the case are ordinary ones about the scope of agency discretion in implementing federal statutes.

The core of the Court’s dispute with the D.C. Circuit, which had faulted the Surface Transportation Board for failing to fully consider upstream and downstream environmental effects of making it easier to get oil to market, is the conclusion that NEPA review should be deferential because NEPA itself is “purely procedural” and not meant to be a “substantive roadblock.” From the Court’s point of view, “[s]ome courts have strayed” and have begun to engage in “overly intrusive (and unpredictable) review in NEPA cases.” At the end of the day, the reviewing court’s duty in a NEPA case is far more limited: “Because an EIS is only one input into an agency’s decision and does not itself require any particular substantive outcome, the adequacy of an EIS is relevant only to the question of whether an agency’s final decision . . . was reasonably explained.” That, the Court notes, is governed by the (typically) deferential arbitrary and capricious standard, although the Court has certainly wobbled in its characterization of arbitrariness review in recent cases like Prometheus Radio Project and Ohio v. EPA.

Of course, there is a statute being interpreted here, and under Loper Bright the courts are supposed to determine the meaning of laws using their independent judgment without any deference. Arguably, that is precisely what the lower court did in this case by requiring an agency to analyze indirect effects of project approval that it interpreted as within the scope of NEPA. The successful defense of the Court’s deferential approach in NEPA review therefore depends on the Court’s articulation of a theoretical distinction between the types of questions at issue in NEPA cases and those arising in ordinary interpretive disputes involving agencies.

The Court’s approach is to argue that NEPA’s key requirements—that the EIS be “detailed,” that it identify “significant” environmental impacts and “feasible alternatives,” and that it analyze impacts that fall within the “scope” of NEPA—are delegations of discretion to agencies. Therefore, in fleshing out what “details need to be included in any given EIS,” for instance, the question “does not turn on the meaning of ‘detailed.’” Instead, it is better conceptualized as involving “primarily issues of fact”—the kinds of questions on which the “agency is better equipped” to provide an answer than the court. The Court offers similar rationales for each of the NEPA requirements. As the Court sums up, these kinds of questions require an agency to “make a series of fact-dependent, context-specific, and policy-laden choices about the depth and breadth of its inquiry—and also about the length, content, and level of detail of the resulting EIS.” So long as those choices are within the “zone of reasonableness,” the reviewing court must defer.

It is not exactly impossible to square this analysis with Loper Bright, but doing so probably greatly contracts the domain of agency action that could be plausibly argued to be subject to Loper Bright’s seemingly stringent endorsement of de novo review. Under Loper Bright, sometimes the best reading of a statute is that it “delegates discretionary authority to an agency.” And when that is the case, the reviewing court fulfills its duty to interpret law by determining the scope of that delegation and ensuring that the agency has “engaged in ‘reasoned decisionmaking’ within those boundaries.” Notably, the discretionary task assigned to agencies under NEPA does not seem to fit neatly within any of the three types of delegations specifically flagged by Loper Bright, but it is of a piece with them insofar as it fixates on the broad meaning encompassed by vague, flexible terms like “detailed” and “significant.” The problem—which was already a problem in those important two paragraphs in Loper Bright—is that an extraordinary number of statutory terms might be read to similarly require fact-bound implementation. If that characteristic is enough to create a delegation of discretion reviewable only for reasonableness, it would seem that Loper Bright’s domain may be very small in practice, perhaps encompassing only true syntactic ambiguities (as distinguished from workaday conceptual vagueness).

At the same time, though, the Court leaves open the possibility that certain of these questions might be pulled out of the realm of fact and into the realm of law. It cryptically notes that the “meaning of ‘detailed’ is a question of law to be decided by a court.” It’s just that what goes in an EIS on a retail basis does not typically implicate these legal questions. And, on scope, the Court nods to the fact that the statute does seemingly make some interpretable decisions—namely, that the EIS must assess effects linked in some way to the “proposed agency action.” It’s just that determinations about “where to draw the line” in terms of proximate linkage do not implicate any legal question.  The Court seems to implicitly invoke a distinction between pure questions of law and mixed questions of law and fact, but without doing much in the way of explaining this distinction or how it is supposed to work across the full sweep of the statutory corpus. In fact, on the scope question, the Court does characterize one question as subject to a bright-line statutory rule: NEPA does not require analysis of separate projects engendered by the project under consideration, because that separateness does “break[] the chain of proximate causation between the project at hand and the environmental effects of the separate project.” Why the Court can answer this legal question but the lower court’s indirect effects analysis stepped on the agency’s delegated discretion is a question with no answer.

It is difficult to read Seven County Infrastructure Coalition alongside Loper Bright and not think of Judge Friendly’s characterization of the pre-Chevron world in Pittston Stevedoring Corp. v. Dellaventura. There, he noted that it was “time to recognize, in line with Professor Kenneth Culp Davis’ brilliant discussion, 4 Administrative Law Treatise §§ 30.01-.09 and the corresponding sections in the 1970 Supplement, that there are two lines of Supreme Court decisions on this subject which are analytically in conflict, with the result that a court of appeals must choose the one it deems more appropriate for the case at hand.”

Increasingly, it seems that we have returned to that point in the law’s development. The Court is troubled by stringent judicial review of NEPA analyses in a way that it is simply untroubled by stringent judicial review of many similar agency decisions. The framework built by Loper Bright leaves the Court with this kind of discretion to toggle back and forth between independent judgment and deference as it sees fit. This, more than any coherent theory of interpretation, is our law.

Daniel E. Walters is an Associate Professor at Texas A&M University School of Law.