Introduction

“The ungentle law and customs touched upon in this tale are historical . . . .”¹Mark Twain, A Connecticut Yankee in King Arthur’s Court 7 (Henry B. Wonham ed., W. W. Norton & Co. 2018) (1889).

Upon its passage in 1946, the Administrative Procedure Act (“APA”) ushered in a new era of administrative law.²5 U.S.C. §§ 551–559. In the decades since, the APA has come to be regarded as “a sort of superstatute, or subconstitution, in the field of administrative process: a basic framework that was not lightly to be supplanted or embellished.”³Antonin Scalia, Vermont Yankee: The APA, the D.C. Circuit, and the Supreme Court, 1978 Sup. Ct. Rev. 345, 363. Like any other constitution, however, the APA simply does not say that much, leaving a multitude of questions unanswered. And while the APA sets the “mood”⁴See Universal Camera Corp. v. NLRB, 340 U.S. 474, 487 (1951). of administrative law, specific questions in specific cases need specific answers. Accordingly, the APA’s barebones framework has been “embellished” countless times: by Congress,⁵The APA itself affords this option. See 5 U.S.C. § 559 (“Subsequent statute[s] may not be held to supersede or modify [the APA], except to the extent that it does so expressly.”); see also Christopher J. Walker, The Lost World of the Administrative Procedure Act: A Literature Review, 28 Geo. Mason L. Rev. 733, 734 & n.6 (2021) (discussing legislative reforms of the APA itself as well as related congressional action); Aaron L. Nielson, Visualizing Change in Administrative Law, 49 Ga. L. Rev. 757, 784–85 (2015) (similar). by the executive,⁶See Aaron L. Nielson, In Defense of Formal Rulemaking, 75 Ohio St. L.J. 237, 256 (2014) (detailing executive orders which impose procedural restraints above and beyond those required by the APA). by agencies themselves,⁷See Emily S. Bremer, Designing the Decider, 16 Geo. J.L. & Pub. Pol’y 67, 73 (2018) (discussing the broad sweep of agency discretion, and how some agencies choose formal procedures not required by the APA); see generally Emily S. Bremer & Sharon B. Jacobs, Agency Innovation in Vermont Yankee’s White Space, 32 J. Land Use & Envt’l L. 523 (2017). and by the courts.⁸See Walker, supra note 5, at 742–44; see generally Gillian E. Metzger, Embracing Administrative Common Law, 80 Geo. Wash. L. Rev. 1293 (2012).

However, the Supreme Court shut down the judiciary’s ability to embellish or innovate above the APA floor in its 1978 landmark decision Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc.⁹435 U.S. 519 (1978). At least, that is what a surface reading of the Court’s holding might lead you to believe. In slapping down the D.C. Circuit, the de facto headquarters of administrative law,¹⁰See Gillian E. Metzger, The Story of Vermont Yankee: A Cautionary Tale of Judicial Review and Nuclear Waste, in Administrative Law Stories 124, 143 (Peter L. Strauss ed., 2006) (summarizing the D.C. Circuit’s path to becoming the nation’s administrative law court in the 1960s and 1970s). the Court emphasized that “nothing . . . permitted the court to review and overturn the rulemaking proceeding on the basis of the procedural devices employed . . . so long as [the agency] employed at least the statutory minima . . . .”¹¹Vermont Yankee, 435 U.S. at 548 (emphasis added). Such an admonition would seem to foreclose any further judicial tinkering with “hybrid” rulemaking requirements—judge-imposed hurdles for agencies to clear which lie between the APA’s rarely invoked formal rulemaking requirements and the ubiquitous notice-and-comment process of informal rulemaking.¹²See Scalia, supra note 3, at 348, 357–58.

The D.C. Circuit responded to the Court’s all-encompassing ukase by . . . continuing to tinker with hybrid rulemaking requirements. For while the Vermont Yankee holding was (arguably) broad, the D.C. Circuit of the era had perfected the art of narrowing administrative law rulings, thereby safeguarding its own ability to continue judicial innovations in the area.¹³See id. at 359–68. In this case, the path to a narrow reading lay through the long-simmering debate between Judge Harold Leventhal and Chief Judge David Bazelon.¹⁴See Ronald J. Krotoszynski, Jr., “History Belongs to the Winners”: The Bazelon-Leventhal Debate and the Continuing Relevance of the Process/Substance Dichotomy in Judicial Review of Agency Action, 58 Admin. L. Rev. 995, 996 (2006). For years, these rivals argued over the proper forms of hybrid rulemaking and judicial review.¹⁵See id. at 999. Judge Leventhal advocated involved technical review, wherein judges would overturn agency action only for substantive reasons.¹⁶See id. at 1002–04. Chief Judge Bazelon championed a procedural approach, wherein judges would impose participatory processes like cross-examinations and evidentiary hearings.¹⁷See id. at 999.

When the Vermont Yankee hammer fell, it fell on a Chief Judge Bazelon opinion.¹⁸See Nat. Res. Def. Council, Inc. v. NRC, 547 F.2d 633, 657 (D.C. Cir. 1976), rev’d, 435 U.S. 519 (1978). Capitalizing on the opportunity, Judge Leventhal and his followers declared victory, attributing the Court’s vitriol to a specific repudiation of procedural hybrid rulemaking, and reading in an implicit endorsement of the substantive approach.¹⁹See Metzger, supra note 10, at 160. Thus armed, the circuit set about strengthening and expanding its substantive review doctrines over the decades that followed.²⁰See Krotoszynski, supra note 14, at 996. Remarkably, despite these doctrines’ disconnect from the text of the APA, and the practical distortions they cause, the Supreme Court has largely acquiesced in this arrangement.²¹While then-Judge Kavanaugh recognized the tension between established D.C. Circuit precedent and Vermont Yankee during his time on that bench, he has not yet disrupted this uneasy status quo. See Am. Radio Relay League, Inc. v. FCC, 524 F.3d 227, 246 (D.C. Cir. 2008) (Kavanaugh, J., concurring in part, concurring in the judgment in part, and dissenting in part) (explaining that the Portland Cement doctrine “stands on a shaky legal foundation”). Those administrative lawyers who have waited for Vermont Yankee II since shortly after Vermont Yankee I have thus far waited in vain.²²See Paul R. Verkuil, Judicial Review of Informal Rulemaking: Waiting for Vermont Yankee II, 55 Tul. L. Rev. 418, 419 (1981); Jack M. Beermann & Gary Lawson, Reprocessing Vermont Yankee, 75 Geo. Wash. L. Rev. 856, 858–59 (2007). But see Perez v. Mortgage Bankers Ass’n, 575 U.S. 92, 100–01 (2015) (invalidating a D.C. Circuit doctrine which required agencies to proceed by notice-and-comment to reverse longstanding interpretive positions); Cass R. Sunstein & Adrian Vermeule, The New Coke: On the Plural Aims of Administrative Law, 2015 Sup. Ct. Rev. 41, 45 (describing Perez as “Vermont Yankee II”).

While the details of the Leventhal-Bazelon debate have largely faded from memory, the Leventhalite reading of Vermont Yankee lives on as legal canon.²³See Krotoszynski, supra note 14, at 996. The hybrid rulemaking we know has therefore been profoundly shaped, flaws and all, by this four-decade-old act of interpretation. This Comment argues that, viewed in context, Vermont Yankee is properly read as a broader denunciation of hybrid rulemaking, rejecting outright the substance-process dichotomy of the Leventhal-Bazelon debates. This conclusion is reached via a brief trek to a hypothetical world of hybrid rulemaking, wherein the Supreme Court instead takes up, and slaps down, a Judge Leventhal-led opinion in Connecticut Yankee.²⁴There was a real Connecticut Yankee Power Corporation, which built and operated the Connecticut Yankee Nuclear Power Plant along the Connecticut River. The plant operated from 1968 to 1996. Connecticut Yankee Nuclear Power Plant, Wikipedia, https://perma.cc/S6UJ-NKKY (Dec. 20, 2022).

Part I of this Comment recounts the history of the hybrid rulemaking world we know. It begins with a brief APA primer before moving on to the jurisprudential battles leading up to Vermont Yankee. Part II describes the decision itself and surveys hybrid rulemaking’s survival in the decades that followed. Part III imagines a hybrid rulemaking world that might have been. After a brief thought experiment in the realm of Connecticut Yankee, it explores advantages such rulemaking might offer. Part IV ponders how far apart the worlds of Vermont Yankee and Connecticut Yankee really are, and considers how this new perspective could improve rulemaking.

I.      Evolving Administrative Procedure: The World We Knew

“[I]t only required a chapter or so to bring it down to date.”²⁵Twain, supra note 1, at 363.

Before venturing into the unknown, we review the familiar. A rehearsal of the APA’s history and text lays the foundation, from which this Part explores the D.C. Circuit’s hybrid rulemaking jurisprudence through the 1960s and 1970s.

A.      Yet Another APA Primer

“There is a profound monotonousness about its facts that baffles and defeats one’s sincerest efforts to make them sparkle and enthuse.”²⁶Id. at 225.

In many ways, administrative law has long outgrown the relatively limited ground marked out by the text of the APA. Nevertheless, the document remains a vital foundation, supporting and informing the development of administrative law even when doctrine is disconnected from the APA’s text. Accordingly, it pays to review some of the APA’s history as well as the principles it eventually instantiated. With the mood set, this Section then proceeds to the text: the statutory minima of formal and informal rulemaking and the guarantee and process for judicial review.

1.      A Foundational Formula

“So they took it, handling it as cautiously and devoutly as if it had been some holy thing come from some supernatural region . . . .”²⁷Id. at 227.

The APA’s path from passage to semiconstitutional status was relatively short, but the actual battle to craft and pass the bill was arduous. Coming out of the New Deal expansion of the administrative state, conservatives primarily focused on formalizing agency adjudication procedures.²⁸See Paul R. Verkuil, The Administrative Procedure Act at 75: Observations and Reflections, 28 Geo. Mason L. Rev. 533, 536 (2021). Dean Roscoe Pound was emblematic of this minority view, harshly criticizing agencies and “administrative absolutism” and advocating for more judicial control of such procedures.²⁹Id. at 533, 539. The majority view, championed by Professor Walter Gellhorn, saw agencies as generally well-behaved and certainly a far cry from Pound’s “hysterical stirrings.”³⁰Paul R. Verkuil, Walter Gellhorn & Kenneth Culp Davis, Present at the Creation: Regulatory Reform Before 1946, 38 Admin. L. Rev. 511, 514 (1986). Regardless of the quality of agency action, the Gellhorn view held, agency processes were simply too varied to be captured and controlled by a generalized statute such as the APA.³¹See Verkuil, supra note 28, at 534–35, 538.

But if the majority view of the APA “founding fathers” was to have no APA at all, where did it come from? As Professor Paul Verkuil recounts it, the pugnacious Pound was eventually sidelined and replaced with the more accommodationist Carl McFarland.³²Id. at 538. McFarland managed to move the American Bar Association’s administrative law committee away from Pound’s outright rejection of regulatory administration and towards procedural compromise.³³Id. at 535–38. With both Congress and the Roosevelt administration spurred to study the issue, the Gellhorn “majority” position to maintain the status quo was untenable.³⁴See George B. Shepherd, Fierce Compromise: The Administrative Procedure Act Emerges from New Deal Politics, 90 Nw. U. L. Rev. 1557, 1580–88 (1996). McFarland’s committee managed to get a statute drafted and put before Congress, instantly becoming the only real candidate for administrative change.³⁵See Verkuil, supra note 28, at 535–36. After the brief interlude of World War II, both houses unanimously passed the APA, which was signed into law in 1946.³⁶Administrative Procedure Act, Pub. L. No. 79-404, 60 Stat. 237 (1946); see Verkuil, supra note 28, at 534.

The APA’s stature as a constitution for administrative law was enshrined shortly after its passage.³⁷See Wong Yang Sung v. McGrath, 339 U.S. 33, 40 (1950). Just four years later, Justice Robert H. Jackson famously summarized the APA’s settlement: “The [APA] thus represents a long period of study and strife; it settles long-continued and hard-fought contentions, and enacts a formula upon which opposing social and political forces have come to rest. It contains many compromises and generalities and, no doubt, some ambiguities.”³⁸Id. at 40–41. Justice Felix Frankfurter described the APA as “express[ing] a mood” which “must be respected, even though it can only serve as a standard for judgment and not as a body of rigid rules assuring sameness of applications.”³⁹Universal Camera Corp. v. NLRB, 340 U.S. 474, 487 (1951).. These statements are reminiscent of the Gellhorn position, leaving room for agencies and judges to maneuver with limited respect for the text.⁴⁰See Verkuil, supra note 28, at 535. This understanding has informed administrative law ever since, with some scholars estimating that ninety percent of administrative law is judge-made common law.⁴¹See John F. Duffy, Administrative Common Law in Judicial Review, 77 Tex. L. Rev. 113, 115 (1998); see generally Metzger, supra note 8.

But while Gellhorn’s majority position won early victories in the constitutional treatment of the APA, there was no rout, as the existence of the document itself testifies to: the text is there, and it has meaning.⁴²See Evan D. Bernick, Envisioning Administrative Procedure Act Originalism, 70 Admin. L. Rev. 807, 834 (2018) (examining the APA’s history and how much meaning the text can bear); see generally Walker, supra note 5. And while lip service to the APA may be the norm, the Supreme Court occasionally reminds its colleagues that the text truly is foundational, as it did in Vermont Yankee.⁴³See Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 523–27 (1978).

2.      Statutory Minima: Formal and Informal, Rulemaking and Adjudication

“Indeed, there was too lightsome a tone of flippancy all through the paper.”⁴⁴Twain, supra note 1, at 224.

The APA’s text, and the statutory minima to which Justice Rehnquist referred in Vermont Yankee,⁴⁵See Vermont Yankee, 435 U.S. at 523–25. are not especially lengthy.⁴⁶See 5 U.S.C. §§ 551–559. Within its limited space, the Act draws distinctions along two axes: on the types of administrative procedures, rulemaking is distinct from adjudication, and on the rigor of those procedures, formal is distinct from informal (though the Act does not use that terminology).⁴⁷See 5 U.S.C. §§ 553, 554, 556, 557. The APA also has many “escape hatches,” exceptions which allow an agency to avoid these quarters altogether.⁴⁸See 5 U.S.C. § 553(b); Verkuil et al., supra note 30, at 522. While this Comment is concerned primarily with rulemaking, a brief discussion of adjudication and “regulatory dark matter” will prove useful later.⁴⁹See Clyde Wayne Crews Jr., Competitive Enter. Inst., Mapping Washington’s Lawlessness 2016: A Preliminary Inventory of “Regulatory Dark Matter” 1 (2015).

The APA defines “rule” in terms similar to how one might describe legislation: “[T]he whole or a part of an agency statement of general or particular applicability and future effect . . . .”⁵⁰5 U.S.C. § 551(4). Rules are the result of agency “rule making,” defined as any “agency process for formulating, amending, or repealing a rule.”⁵¹Id. § 551(5). Adjudications are defined in contradistinction to rulemakings, consisting in “final disposition[s], whether affirmative, negative, injunctive, or declaratory in form” as expressed in orders.⁵²Id. § 551(6).

Section 553 contains the procedures for “informal” or “notice-and-comment” rulemaking.⁵³Id. § 553. To engage in such rulemaking, an agency must perform three tasks. The process starts with a “[g]eneral notice of proposed rule making” which must include “(1) a statement of the time, place, and nature of public rule making proceedings; (2) reference to the legal authority under which the rule is proposed; and (3) either the terms or substance of the proposed rule or a description of the subjects and issues involved.”⁵⁴Id. § 553(b). Next, the agency must allow “interested persons an opportunity to participate in the rule making” by accepting public comments “with or without opportunity for oral presentation.”⁵⁵5 U.S.C. § 553(c). Finally, when issuing its final rules, the agency must include within them a “concise general statement of their basis and purpose.”⁵⁶Id.

The statutory provisions for informal adjudication are even more minimal, which is to say that they largely do not exist. Section 554 discusses the procedures for “every case of adjudication required by statute to be determined on the record after opportunity for an agency hearing,” which denotes formal proceedings.⁵⁷Id. § 554(a). Those adjudications falling outside of § 554 must minimally meet the procedural protections described in § 555’s “[a]ncillary matters” and § 558’s discussion of licensing, a form of adjudication.⁵⁸See id. §§ 555, 558; see also Walker, supra note 5, at 747–48, 747 n.78. Beyond these guidelines, informal adjudication is largely defined (if at all) by agency-specific statutes.⁵⁹See Walker, supra note 5, at 747–51; Richard J. Pierce, Jr., Agency Adjudication: It Is Time to Hit the Reset Button, 28 Geo. Mason L. Rev. 643, 646–47 (2021).

Formal procedures, be they rulemaking or adjudication, are subject to the more rigorous requirements of §§ 556 and 557. When these sections apply, they impose trial-like requirements such as evidentiary hearings⁶⁰5 U.S.C. § 556(b). and cross-examinations,⁶¹Id. § 556(d). limit ex parte communications with the agency,⁶²Id. §§ 554(d), 557(d)(1). and require more detailed legal findings in the final rule or order.⁶³Id. § 557(c)(3)(A). However, the caveat of “when these sections apply” is larger than it may appear, since they usually do not. The Supreme Court effectively sentenced formal APA proceedings to exile in its 1973 decision United States v. Florida East Coast Railway,⁶⁴410 U.S. 224 (1973); see Nielson, supra note 6, at 239–40. where it held that only statutory language mirroring the APA’s “on the record after opportunity for an agency hearing” verbiage invokes formal procedures.⁶⁵See 5 U.S.C. § 553(c); United States v. Fla. E. Coast Ry. Co., 410 U.S. 224, 237–38 (1973). Almost no statutes, enacted either before Florida East Coast Railway or in the decades since, include the requisite words; as a result, formal rulemaking is largely forgotten.⁶⁶See Nielson, supra note 6, at 239–40 (explaining the death of formal rulemaking and arguing for its continuing relevance); Michael Asimow, Admin. Conf. of U.S., Federal Administrative Adjudication Outside the Administrative Procedure Act 59–87 (2019) (setting forth best practices for informal adjudication).

The APA contains exceptions which allow for neither formal nor informal proceedings, one of which is the nebulous category of “interpretative” rules.⁶⁷Id. § 553(b). This exception has been used, and abused, to form a growing body of “regulatory dark matter.”⁶⁸See Crews, supra note 49, at 3–4. Taking many forms, including “agency and presidential memoranda, guidance documents . . . directives, news releases, letters, and even blog posts,” such regulations manage a lot of governing with minimal process.⁶⁹Id. at 3; see id. at 17–21. Executive efforts to formalize dark matter use have failed, so this shadowy corner of the law is likely here to stay.⁷⁰See Exec. Order No. 13,891, 84 Fed. Reg. 55,235 (Oct. 15, 2019) (providing some procedural controls and executive oversight for agency use of guidance documents); Exec. Order No. 13,992, 86 Fed. Reg. 7,049 (Jan. 25, 2021) (revoking E.O. 13,891).

3.      Judicial Review: Why Hybrid Rulemaking, Anyway?

“Everywhere, these black-robed, soft-sandaled, tallow-visaged specters appeared, flitted about and disappeared, noiseless as the creatures of a troubled dream, and as uncanny.”⁷¹Twain, supra note 1, at 175.

Administrative agencies wield vast powers—at times executive, legislative, or judicial, depending on the circumstance.⁷²See, e.g., Kristin E. Hickman & Richard J. Pierce, Jr., Federal Administrative Law 15–17 (3d ed. 2020). To guard against abuse of this power, the APA grants judicial review as a right to anyone “aggrieved by agency action.”⁷³5 U.S.C. § 702. This right is waived only by statute, expressly or impliedly.⁷⁴Id. § 701(a). While this review is relatively deferential—agency action is set aside only for listed defects such as “abuse of discretion”⁷⁵Id. § 706(2)(A). or actions “in excess of statutory jurisdiction”⁷⁶Id. § 706(2)(C).—the courts have endeavored to make it effective.⁷⁷See, e.g., McCarthy v. Madigan, 503 U.S. 140, 146–49 (1992). Generally, this review is limited to the administrative record,⁷⁸See 5 U.S.C. § 706 (“In making the foregoing determinations, the court shall review the whole record or those parts of it cited by a party, and due account shall be taken of the rule of prejudicial error.”). though some circumstances may permit a court to supplement that record.⁷⁹See Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 415–18 (1971).

Regulatory dark matter largely avoids both agency procedural requirements on the front end and judicial review on the back end.⁸⁰See Crews, supra note 49, at 3. Informal rulemaking, at least the bare statutory minima described above, produces little for a judge to review at all.⁸¹See Metzger, supra note 8, at 1348–52. With formal rulemaking largely extinct, judicial imposition of hybrid requirements to effectuate judicial review becomes, if not inevitable, at least understandable.⁸²See id. at 1320–22.

B.      Hybrid Rulemaking Before Vermont Yankee

“Begin here—I’ve already told you what goes before.”⁸³Twain, supra note 1, at 13.

The settlement of New Deal administrative agitation brought about by the APA was short-lived. Adjudication, almost the exclusive mode of agency action before the expansion of the administrative state and the primary target of conservative ire in the push for the APA,⁸⁴See Verkuil, supra note 28, at 536. began losing ground to agency rulemaking.⁸⁵See Scalia, supra note 3, at 376–77. Even before Florida East Coast Railway undercut formal rulemaking, agencies moved towards the notice-and-comment process to take advantage of the efficiency of these “more expeditious administrative methods.”⁸⁶Permian Basin Area Rate Cases, 390 U.S. 747, 777 (1968); see Scalia, supra note 3, at 376. The Supreme Court paved the way for this transformation in several judicially modest decisions which left the form of agency action almost completely within agency discretion.⁸⁷See, e.g., SEC v. Chenery Corp., 332 U.S. 194, 202 (1947) (“In performing its important functions in these respects, therefore, an administrative agency must be equipped to act either by general rule or by individual order. To insist upon one form of action to the exclusion of the other is to exalt form over necessity.”). Lower court judges began adapting familiar administrative common-law doctrines and adjudicatory procedures for this new rulemaking world.⁸⁸See Scalia, supra note 3, at 348–56. The D.C. Circuit, as it developed into the hub of administrative law in the mid-1960s, led the way in this hybrid development.⁸⁹Id. The battles between two members of that court, Chief Judge David Bazelon and Judge Harold Leventhal, would shape administrative law doctrine for decades to come.⁹⁰See Metzger, supra note 10, at 126.

1.      The Hybrid Process “Shadow Doctrine”

“[A]nd not a solitary word of it all could these catfish make head or tail of, you understand . . . .”⁹¹Twain, supra note 1, at 211.

While a lower court openly defying a dictate of the Supreme Court might seem to be headline grabbing in most contexts, it appears to have been the norm for the D.C. Circuit in the 1960s and 1970s.⁹²See Scalia, supra note 3, at 359–68. As agencies relied more heavily on informal procedures, the judges of the circuit seemed determined to have a role in prescribing additional requirements as they deemed necessary.⁹³See id. Often, these attempts at evolving an administrative common law purported to effectuate the APA’s judicial review provisions,⁹⁴See, e.g., Kennecott Copper Corp. v. EPA, 462 F.2d 846, 850 (D.C. Cir. 1972) (“There are contexts, however, contexts of fact, statutory framework and nature of action, in which the minimum requirements of the Administrative Procedure Act may not be sufficient. In the interest of justice, cf. 28 U.S.C. § 2106, and in aid of the judicial function, centralized in this court, of expeditious disposition of challenges to standards, the record is remanded for the Administrator to supply an implementing statement that will enlighten the court as to the basis on which he reached the 60 standard from the material in the Criteria.”). though appeals to broader principles such as “considerations of fairness”⁹⁵O’Donnell v. Shaffer, 491 F.2d 59, 62 (D.C. Cir. 1974). and “the very concept of fair hearing”⁹⁶Am. Airlines, Inc., v. CAB, 359 F.2d 624, 632 (D.C. Cir. 1966). were also common.

The variety and scope of these extra-APA requirements were impressive. The court was happy to make rulemaking’s comment process look more like adjudication in cases like International Harvester Co. v. Ruckelshaus⁹⁷478 F.2d 615 (D.C. Cir. 1973). and Mobil Oil Corp. v. FPC,⁹⁸483 F.2d 1238 (D.C. Cir. 1973). but hybrid requirements reached the rest of § 553 as well.⁹⁹Both International Harvester and Mobil Oil held (in roundabout ways) that adjudicatory-style hearings and cross-examination were required in certain types of rulemaking proceedings. See 478 F.2d at 629–31; 483 F.2d at 1257–60. The court beefed up notice requirements to force the disclosure of all data available to the agency.¹⁰⁰See, e.g., Portland Cement Ass’n v. Ruckelshaus, 486 F.2d 375, 392–93 (D.C. Cir. 1973). Ex parte communications, which the APA’s text forbids only in formal proceedings,¹⁰¹5 U.S.C. § 557(d). received a general prohibition.¹⁰²See, e.g., HBO, Inc. v. FCC, 567 F.2d 9, 51–59 (D.C. Cir. 1977). Cementing the disconnect from the text of the APA, “concise statement” requirements ballooned to include comprehensive responses to important points raised during the comment process, including all relevant technical details.¹⁰³See, e.g., Auto. Parts & Accessories Ass’n v. Boyd, 407 F.2d 330, 338 (D.C. Cir. 1968) (“[I]t is appropriate for us . . . to caution against an overly literal reading of the statutory terms ‘concise’ and ‘general.’ These adjectives must be accommodated to the realities of judicial scrutiny, which do not contemplate that the court itself will, by a laborious examination of the record, formulate in the first instance the significant issues faced by the agency and articulate the rationale of their resolution.”).

Almost all of these “evolutions” beyond the APA minimums occurred squarely in the shadow of Supreme Court cases that had held the opposite. The D.C. Circuit managed to persist in these innovations by writing opinions in “tantalizingly ambiguous” fashion¹⁰⁴Scalia, supra note 3, at 350.: “[t]he pattern of dicta, alternate holdings, and confused holdings out of which the D.C. Circuit’s principle of APA hybrid rulemaking so clearly and authoritatively emerged had the effect, if not the purpose, of assuring compliance below while avoiding accountability above.”¹⁰⁵Id. at 372. While then-professor Antonin Scalia provided many pages of examples of these forays away from (very recent) Supreme Court precedent,¹⁰⁶See id. at 359–67. perhaps one will suffice here. In June 1972, the Supreme Court found that formal rulemaking was not required for “car service rules,” and “therefore, [the proceeding] was governed by the provisions of 5 U.S.C. § 553.”¹⁰⁷United States v. Allegheny-Ludlum Steel Corp., 406 U.S. 742, 742, 757–58 (1972). Since the agency “fully compl[ied] with [informal rulemaking] requirements . . . nothing more was required by the [APA].”¹⁰⁸Id. at 758. Just over a year later, in July 1973, the D.C. Circuit flatly rejected that informal rulemaking was the only judicially enforceable alternative if formal rulemaking was not implicated:

The Commission’s position assumes that there are only two permissible forms of procedures cognizable under the APA, that the two are mutually exclusive, and that their existence precludes the use of any other procedures that lie between them. This rigid interpretation of what is permitted and required under the APA is inaccurate . . . .¹⁰⁹Mobil Oil Corp. v. FPC, 483 F.2d 1238, 1251 (D.C. Cir. 1973).

The D.C. Circuit during this period proved to be “a remarkably ineffective instrument for implementing the underlying principles of interpretation which the Supreme Court opinions quite clearly expressed.”¹¹⁰Scalia, supra note 3, at 363. Nonetheless, for more than a decade such decisions were standard fare for the circuit, regardless of the judges involved.¹¹¹Id. at 348 n.13. Debates about hybrid rulemaking filled not just opinions, which interacted in complex and confusing ways given the often impenetrable holdings,¹¹²Id. at 373 n.128. but journal articles as well.¹¹³Id. at 365 n.97. The debate, however, was not over the propriety of hybrid requirements, but rather what form such requirements should take.

2.      Substance versus Process: The Leventhal-Bazelon Debates

“Here was trouble again—a conflict of authority.”¹¹⁴Twain, supra note 1, at 202.

The D.C. Circuit was a truly remarkable court throughout the 1960s and 1970s. Not only was it host to a collection of “‘judicial all-stars’ who individually would have dominated other courts,” but that stable of all-star judges remained relatively constant.¹¹⁵Matthew Warren, Note, Active Judging: Judicial Philosophy and the Development of the Hard Look Doctrine in the D.C. Circuit, 90 Geo. L.J. 2599, 2607 (2002). As a result, an immense array of writings, both judicial and extrajudicial, illustrate the clashes of judicial philosophy between the court’s members.¹¹⁶See id. at 2607 & n.58. Two primary theories of hybrid rulemaking and the relationship between judges and agencies emerged from these endless disputes.¹¹⁷See generally Krotoszynski, supra note 14. For an interesting analysis of Judge Skelly Wright’s jurisprudence at the time, positing that his focus on congressional intent and democratic accountability constituted a third approach to judicial review of agency action, see Warren, supra note 115, at 2626–31. At the head of one camp: Judge Harold Leventhal, former administrator extraordinaire.¹¹⁸See Warren, supra note 115, at 2607–09. Leading the other camp, perhaps just a camp of one, given his difficult personality and “poisonous” relationships with his colleagues: Chief Judge David Bazelon, legal realist par excellence.¹¹⁹Id. at 2617–21.

Judge Leventhal came to the bench following an impressive career in government, bouncing between agencies through the New Deal and war years and eventually leading a presidential task force on independent agencies.¹²⁰See id. at 2608. Accordingly, his perspective on the relationship between judge and agency—that the two parties were to cooperate as partners to advance congressional interests—is understandable.¹²¹See id. at 2611–13; see, e.g., Greater Bos. Television Corp. v. FCC, 444 F.2d 841, 851–52 (D.C. Cir. 1970) (“[A]gencies and courts together constitute a ‘partnership’ in furtherance of the public interest, and are ‘collaborative instrumentalities of justice.’”) (quoting United States v. Morgan, 313 U.S. 409, 422 (1941)); see also Harold Leventhal, Environmental Decisionmaking and the Role of the Courts, 122 U. Pa. L. Rev. 509, 554 (1974) (“Another objective [of judicial review] is to combine supervision with restraint, making the courts a genuine kind of partner with the agency in the overall administrative process.”). When it came time to review agency action, Judge Leventhal envisioned a limited scope for judges and expressed his “utmost diffidence” when that review required him to remand a decision to an agency.¹²²Int’l Harvester Co. v. Ruckelshaus, 478 F.2d 615, 641 (D.C. Cir. 1973); see Krotoszynski, supra note 14, at 1003–04.

Paradoxically, however, when it came to how judges should approach this limited review, Judge Leventhal insisted on the daunting task of fully understanding the technical matters at issue.¹²³See Warren, supra note 115, at 2615–16. Even with his extensive agency experience, Judge Leventhal himself openly struggled to meet this standard,¹²⁴See Harold Leventhal, Remarks, 7 Nat. Res. Law. 351, 355–57 (1974) (“If there is going to be a real monitoring of the question of whether there has been a hard look in recent decision-making, you have to get into the technicalities in order to understand the problem. And that is something that is very hard to do.”). leading him to propose specialized judicial assistants.¹²⁵See Reuel E. Schiller, Rulemaking’s Promise: Administrative Law and Legal Culture in the 1960s and 1970s, 53 Admin. L. Rev. 1139, 1163 (2001). While his proposals never came to fruition, Judge Leventhal maintained that generalist judges could rise to the challenge using traditional judicial principles and tools to obtain outside expertise, such as special masters.¹²⁶See id.; see also Env’t Def. Fund, Inc. v. EPA, 465 F.2d 528, 541 (D.C. Cir. 1972) (“[T]here is a will in the courts to study and understand what the agency puts before us.”). Indeed, judges must grapple with the substance of the problems before them because, as Judge Leventhal believed, that is what Congress, via the APA, demanded.¹²⁷See Ethyl Corp. v. EPA, 541 F.2d 1, 68 (1976) (Leventhal, J., concurring) (“[I]t is my view that while giving up is the easier course, it is not legitimately open to us at present. . . . Our present system of review assumes judges will acquire whatever technical knowledge is necessary as background for decision of the legal questions.”); see also Krotoszynski, supra note 14, at 1004.

Chief Judge Bazelon, by contrast, did not care overmuch what statute and precedent demanded.¹²⁸See Warren, supra note 115, at 2617–21. Described by one fellow judge as “the quintessential legal realist,” he focused first and foremost on the practical results of law and doctrine.¹²⁹See id. at 2617–18, 2618 n.163, 2621 n.191. Accordingly, “[h]e was willing, indeed even eager, to question accepted judicial truths and doctrines.”¹³⁰Id. at 2618. In areas beyond administrative law, such as Chief Judge Bazelon’s attempted reforms to mental illness doctrine,¹³¹See id. at 2619–20; see also J. Skelly Wright, A Colleague’s Tribute to Judge David L. Bazelon, on the Twenty-Fifth Anniversary of His Appointment, 123 U. Pa. L. Rev. 250, 252 (1974); see generally Martha Minow, Questioning Our Policies: Judge David L. Bazelon’s Legacy for Mental Health Law, 82 Geo. L.J. 7 (1993). his more active judicial style partook of the legal-realism movement’s “faith in empiricism and in the liberating, progressive potential of a properly harnessed social science.”¹³²See Warren, supra note 115, at 2617 n.150. A generalist judge, armed with a properly questioning attitude and a fully disclosed record, could reach determinations both just and socially beneficial.¹³³See id. at 2621.

As legal realists of the day thrilled to the possibility that properly harnessed social science could turn judges into philosopher-kings, Judge Leventhal’s faith that judges could master the mundane technical details of agency action would seem a natural fit.¹³⁴See id. at 2617–20. At least for Chief Judge Bazelon, nothing could be further from the truth.¹³⁵See id. While he may have been the quintessential legal realist, Chief Judge Bazelon was the ultimate skeptic of generalist judges’ ability to master highly scientific material to effectively review agency action.¹³⁶See Krotoszynski, supra note 15, at 999–1000. From Chief Judge Bazelon’s perspective, judges were “technically illiterate.”¹³⁷Ethyl Corp. v. EPA, 541 F.2d 1, 67 (D.C. Cir. 1976) (Bazelon, C.J., concurring). Any attempt by judges to review agency decisions founded on complex scientific or mathematical data was bound to be “dangerously unreliable.”¹³⁸Id.; see also David L. Bazelon, The Impact of the Courts on Public Administration, 52 Ind. L.J. 101, 107 (1976) (“Significant or not, decisions involving scientific or technical expertise present peculiar challenges for reviewing courts. The problem is not so much that judges will impose their own views on the merits. The question is whether they will even know what is happening.”).

While Judge Leventhal saw such an approach as laziness at best, or an abdication of judicial duty at worst,¹³⁹See supra note 127 and accompanying text. Chief Judge Bazelon viewed it as a sort of comparative advantage.¹⁴⁰See Krotoszynski, supra note 15, at 999–1000. Regulated parties, not generalist judges, were best able to meet and contest an agency’s scientific arguments.¹⁴¹See Int’l Harvester Co. v. Ruckelshaus, 478 F.2d 615, 652 (D.C. Cir. 1973) (Bazelon, C.J., concurring) (“[I]n cases of great technological complexity, the best way for courts to guard against unreasonable or erroneous administrative decisions is not for the judges themselves to scrutinize the technical merits of each decision.”). What judges could do, and should do, was ensure that agencies provided enough of a forum for such a contest.¹⁴²See id.; see also Nat. Res. Def. Council, Inc. v. NRC, 547 F.2d 633, 657 (D.C. Cir. 1976) (Bazelon, C.J., concurring) (“I am convinced that in highly technical areas, where judges are institutionally incompetent to weigh evidence for themselves, a focus on agency procedures will prove less intrusive, and more likely to improve the quality of decisionmaking, than judges ‘steeping’ themselves ‘in technical matters to determine whether the agency has exercised a reasoned discretion.’” (quoting Ethyl Corp., 541 F.2d at 66 & n.5)). Ordering hearings and overseeing cross-examination were well within the judicial wheelhouse, and the thoroughly ventilated issues¹⁴³Despite (or perhaps because of) the difficulty in defining what was meant by “ventilation of the issues,” it was a favorite phrase of the D.C. Circuit during this time. See Scalia, supra note 3, at 355 n.55. that resulted from such procedures would be amenable to effective judicial review.¹⁴⁴See Krotoszynski, supra note 15, at 999–1000.

These, then, were the two primary philosophies of judicial review of agency action during the years the D.C. Circuit began embellishing beyond the APA’s text—Judge Leventhal’s insistence on substance and Chief Judge Bazelon’s focus on process.¹⁴⁵See generally id. But while the sides of this war were relatively clear when battle was joined in the law reviews,¹⁴⁶See, e.g., Leventhal, supra note 121; Bazelon, supra note 138; David L. Bazelon, Coping with Technology Through the Legal Process, 62 Cornell L. Rev. 817 (1977). things were rarely so pellucid in the murk that characterized D.C. Circuit opinions of the era.¹⁴⁷See supra note 112 and accompanying text. Since this Comment is already premised on a switcheroo of sorts, I have refrained from the old trick of misattributing quotes between the two sides of an apparently irreconcilable divide before—hey presto!—revealing that the two positions are actually indistinguishable. But rest assured, Dear Reader, that if I had done so, you would have been fooled. Judge Leventhal was occasionally willing to impose those “procedural requirements deemed inherent in the very concept of fair hearing for certain classes of cases, even though no such requirements had been specified by Congress.”¹⁴⁸Am. Airlines, Inc. v. CAB, 359 F.2d 624, 632 (D.C. Cir. 1966). Chief Judge Bazelon, meanwhile, “conceded that consideration of the adequacy of the procedures that an agency followed related back to the substantive questions before the agency.”¹⁴⁹Krotoszynski, supra note 15, at 1000 n.19. Indeed, Chief Judge Bazelon viewed the substance-process dichotomy as a mostly meaningless distinction in the first place.¹⁵⁰See Nat. Res. Def. Council, Inc. v. NRC, 547 F.2d 633, 657 n.8 (D.C. Cir. 1976) (Bazelon, C.J., concurring) (writing that “[t]he logic of [the] position that the ‘deficiency’ here is not with the procedures used to make a record, just with the ‘record generated,’ totally escapes me”) (citations omitted). Citing to Judge Henry Friendly, Chief Judge Bazelon noted that, practically speaking: “it does not really matter much whether a court says the record is remanded because the procedures used did not develop sufficient evidence, or because the procedures were inadequate. From the standpoint of the administrator, the point is the same: [informal rulemaking procedures] will not automatically produce an adequate record.” Id. at 657 (citing Henry J. Friendly, “Some Kind of Hearing,” 123 U. Pa. L. Rev. 1267, 1313–14 (1975)); cf. Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 350 (1989) (noting in the context of NEPA that “these procedures are almost certain to affect the agency’s substantive decision”). The difficulty of drawing the distinction is evident throughout the opinions of the time, and at least one of Chief Judge Bazelon’s colleagues shared this view: Judge Edward Tamm acknowledged that “arguments about whether our focus here is ‘procedural’ or ‘substantive’ may be more semantic than determinative.”¹⁵¹Nat. Res. Def. Council, 547 F.2d at 660 n.8 (Tamm, J., concurring).

3.      Portland Cement: A Foundation for the Future

“To be vested with enormous authority is a fine thing; but to have the on-looking world consent to it is a finer.”¹⁵²Twain, supra note 1, at 59.

Portland Cement Ass’n v. Ruckelshaus,¹⁵³486 F.2d 375 (D.C. Cir. 1973). a 1973 Judge Leventhal opinion, bears special mention here. While far from the first hybrid requirements case,¹⁵⁴See, e.g., Am. Airlines, Inc. v. CAB, 359 F.2d 624, 632 (D.C. Cir. 1966); Marine Space Enclosures, Inc. v. FMC, 420 F.2d 577 (D.C. Cir. 1969); Walter Holm & Co. v. Hardin, 449 F.2d 1009 (D.C. Cir. 1971). it earned a prominent place in the D.C. Circuit’s administrative law jurisprudence, which it retained even after Vermont Yankee. The basic setting of Portland Cement should ring a few bells to anyone familiar with administrative law: William Ruckelshaus, the Administrator of the Environmental Protection Agency, determined that portland cement plants¹⁵⁵For those curious, “portland cement” is simply a common type of cement, named for its resemblance to stone found on the Isle of Portland in Dorset, England. Portland Cement, Wikipedia, https://perma.cc/72XM-2QAY (Jan. 4, 2023). were a stationary source of air pollution within the purview of the Clean Air Act.¹⁵⁶42 U.S.C. § 1857c-6; Portland Cement, 486 F.2d at 378. Accordingly, he published a proposed regulation setting a “standard of performance,” air pollutant emission requirements, which such plants would have to meet.¹⁵⁷Portland Cement, 486 F.2d at 378. Notice of the regulation went out, accompanied by a separate document with the data justifying the new standard.¹⁵⁸See id. A proper comment process followed, with more than two hundred interested parties participating.¹⁵⁹See id. Finally, the agency promulgated a standard with a statement of its basis and purpose, later supplemented to comply with a preexisting hybrid requirement.¹⁶⁰See id. at 378–79. The prior decision was Kennecott Copper Corp. v. EPA, 462 F.2d 846 (D.C. Cir. 1972), which required the agency to provide the bases for critical decisions along with its proposed regulation.

The agency complied with the statutory minima of § 553, even going above and beyond to try to appease the court, to no avail. While interested parties received, and took advantage of, the opportunity of a public comment process, Judge Leventhal found the agency action insufficient.¹⁶¹Portland Cement, 486 F.2d at 402. The “critical defect” was the agency’s failure to make available “in timely fashion—the test results and procedures used on existing plants which formed a partial basis for [the rule].”¹⁶²Id. at 392. In addition, the agency failed to properly respond to “what seem[ed] to be legitimate problems” raised by public comments.¹⁶³Id.

Couched in substantive terms, Judge Leventhal’s justifications for these findings nevertheless impose rather clear procedural requirements on the part of agencies. Regarding the inadequate disclosure of data, the judge held that “[i]t is not consonant with the purpose of a rule-making proceeding to promulgate rules on the basis of inadequate data, or on data that, [to a] critical degree, is known only to the agency.”¹⁶⁴Id. at 393. As for responding to comments, Judge Leventhal asserted that he was “not establishing any broad principle that EPA must respond to every comment made by manufacturers,” only those which the court deemed “significan[t], or at least potential[ly] significan[t].”¹⁶⁵Id. How an agency is to intuit which comments may be significant going forward is unclear.

Portland Cement also provided Judge Leventhal a chance to reiterate his conception of the roles of judges and agencies.¹⁶⁶See id. at 394. Tucked into the discussion of his remand decision was his understanding that, “[in] matter[s] involving the public interest . . . the court and agency are in a kind of partnership relationship for the purpose of effectuating the legislative mandate.”¹⁶⁷Portland Cement, 486 F.2d at 394. Five years later, the Supreme Court would rudely interrupt this “partnership relationship.”¹⁶⁸Id.; see generally Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519 (1978). At least, it would try.

II.      Vermont Yankee: The World We Know

“These people were no easier to please than other nines.”¹⁶⁹Twain, supra note 1, at 343.

This Part recounts Vermont Yankee’s journey from the court of appeals to Justice Rehnquist’s monumental response. It then traces the administrative law aftershocks that followed.

A.      Vermont Yankee

“I had promised myself an easy and zenith-scouring triumph, and this was the outcome!”¹⁷⁰Id. at 215.

Things were looking good for commercial nuclear power in America in 1966. The SL-1 meltdown was five years in the rearview mirror,¹⁷¹See SL-1, Wikipedia, https://perma.cc/Q9YB-LHDN (Jan. 14, 2023). and Three Mile Island was over a decade away.¹⁷²Indeed, construction would not begin on either Three Mile Island operating plant until 1968. See Three Mile Island Nuclear Generating Station, Wikipedia, https://perma.cc/TQA7-9B7E (Jan. 13, 2023). The Atomic Energy Commission (“AEC” or “Commission”), which had wielded near-plenary regulatory control over nuclear power via the Atomic Energy Act of 1954,¹⁷³Atomic Energy Act of 1954, Pub. L. No. 83-703, § 3d, 68 Stat. 919, 922 (codified at 42 U.S.C. § 2013(d)). had more or less figured out the safety requirements for new power plant projects.¹⁷⁴Of the ninety-two presently active American nuclear plants, about a third were granted their operating licenses in the late 1960s or early 1970s. U.S. Nuclear Plant License Information, Nuclear Energy Inst. (Aug. 2022), https://perma.cc/SD5U-RAZM. The burgeoning commercial nuclear power industry rose to meet those requirements, and together agency and industry ushered in a new age of American energy independence.¹⁷⁵See Metzger, supra note 10, at 127. There was little reason, then, to remark on yet another AEC construction permit, this one for a new reactor on the banks of the Connecticut River to be operated by the Vermont Yankee Nuclear Power Corporation.

When Vermont Yankee applied for the operating license necessary to run the plant three years later, the situation had changed only slightly. Congress passed the National Environmental Policy Act of 1969¹⁷⁶Pub. L. No. 91-190, 83 Stat. 852 (1970) (codified at 42 U.S.C. §§ 4331–4335). (“NEPA”), requiring the government to consider the environmental impacts of major federal action, though it was yet unclear how NEPA would apply to previously approved projects, if at all.¹⁷⁷See Metzger, supra note 10, at 128–29. The growing number of nuclear plants had also raised public concerns with how to handle the resulting radioactive waste products.¹⁷⁸See id. at 128 & n.6. Despite these shifts, Vermont Yankee had little reason to suspect that its application was at any particular risk.

The first sign of trouble was the D.C. Circuit’s opinion in Calvert Cliffs’ Coordinating Committee, Inc. v. AEC.¹⁷⁹449 F.2d 1109 (D.C. Cir. 1971). Calvert Cliffs took the AEC to task for incorporating NEPA into its procedures too slowly.¹⁸⁰Id. at 1117, 1119–20. End result: the court required the Commission to perform environmental impact analyses for plants undergoing licensing proceedings at the time of NEPA’s passage.¹⁸¹Id. at 1128–29. The always-cautious AEC did not press the issue.¹⁸²See Metzger, supra note 10, at 129–30. The impact of Calvert Cliffs was felt when Vermont Yankee’s adjudicatory hearings finally took place in late 1971 and early 1972, as a bevy of public intervenors pressed environmental challenges under NEPA, many concerning radioactive waste storage.¹⁸³See Vt. Yankee Nuclear Power Corp., 4 A.E.C. 776, 777, 783–86 (Mar. 14, 1972).

The AEC’s safety and license board that conducted the hearings, however, concluded that longer-term environmental concerns with waste products were inappropriate for individual licensing proceedings.¹⁸⁴See id. at 785–86. The appeal board that reviewed the decision to grant Vermont Yankee’s operating license agreed.¹⁸⁵See Vt. Yankee Nuclear Power Corp., 4 A.E.C. 930, 933–34 (June 6, 1972). The environmental effects of waste produced by the plant, which could be reprocessed or stored in any number of future facilities, were too far removed from the operation of the plant at issue to warrant consideration.¹⁸⁶Id. The AEC declined to review or overturn this decision, apparently clearing the way for Vermont Yankee to finally get to work.¹⁸⁷See Metzger, supra note 10, at 131. Shortly thereafter, the AEC proposed a new rulemaking addressing the precise issue deemed inapplicable to Vermont Yankee: “whether—and how—environmental effects of the nuclear fuel cycle should be considered in individual reactor proceedings.”¹⁸⁸Id.

With no statutory hearing requirement, the AEC was free to proceed under the informal rulemaking requirements of § 553.¹⁸⁹See id. at 132–33. However, prompted by its careful nature and desire to win public trust in nuclear power, the Commission voluntarily adopted additional procedural strictures.¹⁹⁰Id. at 134. The Commission held a public hearing, solicited both oral and written testimony,¹⁹¹Id. and created a three-member hearing board empowered to question witnesses.¹⁹²See id. An extra thirty-day period for supplemental comments followed the hearings.¹⁹³Metzger, supra note 10, at 134. The resulting record weighed in at over five hundred pages.¹⁹⁴Id. at 136.

In the end, the AEC adopted a final rule establishing predetermined reactor waste harm values that could slot into environmental impact statements for future reactor licensing proceedings.¹⁹⁵See Environmental Effects of the Uranium Fuel Cycle, 39 Fed. Reg. 14,188, 14,189 (Apr. 22, 1974). In so doing, it explicitly rejected the alternative of not considering future waste as a factor for individual plant licensing adjudications (i.e. the approach that prevailed for Vermont Yankee).¹⁹⁶See Metzger, supra note 10, at 141. Nevertheless, the AEC determined that the new rule should not apply retroactively given the “relatively insignificant” impact it would have.¹⁹⁷Environmental Effects of the Uranium Fuel Cycle, supra note 195, at 14,190. Vermont Yankee’s operating license had become permanent in February 1973, and with the fuel cycle rule finished and inapplicable, it seemed that the plant was finally in the clear.¹⁹⁸See Vt. Yankee Nuclear Power Corp., 6 A.E.C. 358, 358 (May 23, 1973). Shortly thereafter, both Vermont Yankee’s operating license and the AEC’s final rule were dragged before the D.C. Circuit.¹⁹⁹See Metzger, supra note 10, at 143.

1.      The D.C. Circuit Versus the NRC

“[T]here was sure to be some skeptic on hand to turn up the gas at the crucial moment and spoil everything.”²⁰⁰Twain, supra note 1, at 177.

In defending its final rule, the Nuclear Regulatory Commission (“NRC”), which replaced the AEC following the Energy Reorganization Act of 1974,²⁰¹Pub. L. No. 93-438, 88 Stat. 1233 (1974) (codified at 42 U.S.C. §§ 5801–5891). found itself before Chief Judge Bazelon, Judge Tamm, and the Sixth Circuit’s Judge George Edwards, sitting by appointment.²⁰²Nat. Res. Def. Council, Inc. v. NRC, 547 F.2d 633, 636 (D.C. Cir. 1976). With Chief Judge Bazelon writing not one but two opinions in the case (the majority plus a separate concurrence), the result is perhaps unsurprising. Through a gleeful mix of alternative justifications and principles—in other words, a true classic of the D.C. Circuit administrative law genre—the court revoked Vermont Yankee’s operating license and remanded the fuel cycle rule to the NRC.²⁰³See id. at 641, 655.

Woven throughout the opinion were two separate “bases of decision”: “(1) the inadequacy of the agency’s procedures; and (2) the inadequacy of the record to support the agency decision.”²⁰⁴Scalia, supra note 3, at 354. Chief Judge Bazelon’s desire for adjudicatory procedures shines through relatively clearly in a few places. He characterized the petitioners’ “primary argument” as “rely[ing] . . . on the line of cases indicating that in particular circumstances procedures in excess of the bare minima prescribed by [§ 553] may be required,” followed by a hit parade of the circuit’s hybrid requirement precedents.²⁰⁵Nat. Res. Def. Council, 547 F.2d at 643 & n.23. Some circumstances, “by their very nature, might require particular procedures, including cross-examination.”²⁰⁶Id. at 644. The circuit’s favorite turn of phrase also makes an appearance, as the agency failed to produce a “thorough ventilation of the issues.”²⁰⁷Id.; see Scalia, supra note 3, at 355, n.55.

Judge Tamm’s concurrence attempted to make the procedural nature of the court’s concern clear, or at least clearer. He noted that “[t]he majority appears to require the Commission to institute further procedures of a more adversarial nature than those customarily required [by § 553].”²⁰⁸Nat. Res. Def. Council, 547 F.2d at 658 (Tamm, J., concurring). Instead, Judge Tamm would have grounded a remand explicitly in Overton Park–type substantive concerns: “the deficiency is not with the type of proceeding below, but with the completeness of the record generated.”²⁰⁹Id. at 659. This prompted Chief Judge Bazelon’s unusual separate concurrence, where he outright “reject[s] the implication that any techniques beyond rudimentary notice and comment are needless ‘over-formalization’ of informal rulemaking.”²¹⁰Id. at 655 (Bazelon, C.J., concurring). Subsequent paragraphs lament the expansion of rulemaking into traditionally adjudicative areas and declare the difference between substance and procedure illusory in most circumstances: “[I]t does not really matter much whether a court says the record is remanded because the procedures used did not develop sufficient evidence, or because the procedures were inadequate.”²¹¹See id. at 657 (citing Henry J. Friendly, “Some Kind of Hearing,” 123 U. Pa. L. Rev. 1267, 1313–14 (1975)).

Far outstripping the procedural § 553 discussion, however, is an avalanche of substantive analysis. Indeed, when read in isolation, the majority opinion is remarkably Leventhal-esque. Chief Judge Bazelon not only engaged with the technical material, chiefly the testimony of Dr. Frank Pittman on behalf of the AEC, but grounded his critiques in the substance of the agency’s reasoning process.²¹²See id. at 645–47 (majority opinion). A reviewing court, looking to the rule’s statement of basis and purpose, must find “reasoned response[s]” to meaningful comments.²¹³Id. at 646. If the agency “has adduced no reasoned answers,” a judge could remand the rule as an abuse of agency discretion.²¹⁴Nat. Res. Def. Council, 547 F.2d at 646. Here, the fuel cycle rule’s “extremely vague assurances” represented just such an “insufficient record.”²¹⁵Id. at 653. Judge-ordered imposition of new procedures is explicitly forsworn, as the court “do[es] not presume to intrude on the agency’s province by dictating . . . devices it must adopt to flesh out the record.”²¹⁶Id. Indeed, the gamut from completely new procedures to the same procedures already used, “administered in a more sensitive, deliberate manner,” could apparently pass muster.²¹⁷Id. at 653–54. Intimations that NEPA, rather than just the APA, prompted these conclusions further muddle the actual foundation of the holding.²¹⁸See id. at 654. Ultimately, however, Chief Judge Bazelon concluded that the Commission’s decision to promulgate the final rule was arbitrary and capricious.²¹⁹See id. at 655. In a surprise to many, the remand wound up not back at the NRC, but in a grant of certiorari at the Supreme Court.²²⁰See Metzger, supra note 10, at 151–53.

2.      The Supreme Court Strikes Back

“Well, when I make up my mind to hit a man, I don’t plan out a love-tap; no, that isn’t my way; as long as I’m going to hit him at all, I’m going to hit him a lifter.”²²¹Twain, supra note 1, at 282.

That the Court “spoke loudly and carried a huge club”²²²Beermann & Lawson, supra note 22, at 858. when it decided, yet again, to weigh in on judicial review of agency action is perhaps unsurprising, given the D.C. Circuit’s years of rather creative compliance with previous rulings.²²³See supra note 110 and accompanying text. That the Court would grant certiorari at all was surprising, given the circumstances.²²⁴See Metzger, supra note 10, at 151–53; see also Scalia, supra note 3, at 356–57. First, the government was ambivalent about pursuing the matter further, as reflected in its “Janus-like” brief.²²⁵Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 540 n.15 (1978). This confusion was shared by the Natural Resources Defense Council, which alternately argued for a substantive or procedural reading of the D.C. Circuit opinion. Id. The Solicitor General read the D.C. Circuit’s opinion as a simple, if erroneous, substantive critique of the agency’s reasoning.²²⁶See id.; Metzger, supra note 10, at 152. The NRC, desperate to resume licensing new plants, pushed for certiorari in the hopes of a definitive answer on the minimum hurdles—be they substantive or procedural—required to garner judicial approval.²²⁷See Metzger, supra note 10, at 152–53. In parallel with litigation, however, the NRC had also begun a supplemental rulemaking to fix the fuel cycle rule in compliance with the D.C. Circuit’s ruling.²²⁸See id. That the rule at issue in the case below was, in all likelihood, about to be rendered obsolete raised clear mootness concerns, yet another reason for the Court to reject certiorari.²²⁹See id.

Despite these concerns, the Court ultimately agreed to hear the case.²³⁰See id. A desire to finally provide oversight of the D.C. Circuit through its maddening fog of dicta and alternate holdings suggests one possible motivation.²³¹See Scalia, supra note 3, at 356–57. Interestingly, the Court granted a similarly dubious certiorari petition also involving an appellate ruling that could hold up nuclear licensing in the same term.²³²See Duke Power Co. v. Carolina Env’t Study Grp., Inc., 438 U.S. 59, 68–82 (1978). This trend suggests another possible motivation: a Court interested in cutting red tape around the nation’s booming nuclear industry.²³³See Metzger, supra note 10, at 153. Indeed, in pushing for certiorari, Justice Rehnquist “stress[ed] the continuing impact of the D.C. Circuit’s decision on the licenses of the two plants involved and reactor licensing generally.”²³⁴Id. at 159.

With the case before it, the Court was immediately confronted by the problem which had vexed the government and the NRDC: what exactly did the D.C. Circuit opinion mean?²³⁵Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 539–41 (1978). Was the court’s problem with the fuel cycle rulemaking substantive, as most of the language in the majority opinion seemed to suggest? Or was it procedural, coming as it did from Chief Judge Bazelon, who defended his usual process-focused approach in his separate concurrence? Writing for a unanimous, Court,²³⁶Justices Powell and Blackmun did not participate in the case. Id. at 558. Justice Rehnquist lamented that solving this puzzle was “no mean feat.”²³⁷Id. at 539.

While initially allowing that the “matter [was] not entirely free from doubt,” Justice Rehnquist ultimately decided that the procedural reading was the better one.²³⁸Id. at 541–42. Chief Judge Bazelon’s framing of the issue for the decision below—that the court was “called upon to decide whether the procedures provided by the agency were sufficient to ventilate the issues”—weighed heavily in Justice Rehnquist’s conclusion.²³⁹Id. (quoting Nat. Res. Def. Council, Inc. v. NRC, 547 F.2d 633, 643 (D.C. Cir. 1976)). The largely substantive nature of the lower decision left few other textual hooks for the procedural reading, which might, Justice Rehnquist warned, “initially lead one to conclude that the court was only examining the sufficiency of the evidence.”²⁴⁰Id. at 542. Looking beyond the substantive language to the practical effect of the ruling, however, revealed that “the ineluctable mandate of the court’s decision is that the procedures afforded during the hearings were inadequate.”²⁴¹Id. Indeed, despite his initial claim that the matter was not free from doubt, Justice Rehnquist concluded his discussion by noting that “the remaining [non-substantive] portions of the opinion,” which the Justice did not see fit to identify in any specificity, “dispel any doubt” that procedural concerns were in play.²⁴²Id.

Having decided on the procedural reading of the D.C. Circuit opinion, Justice Rehnquist was utterly unsparing in tearing it apart.²⁴³See id. at 543–49; see, e.g., Metzger, supra note 10, at 160 (noting the Court’s “unqualified and stern language”); Beermann & Lawson, supra note 22, at 858 (characterizing the Court’s opinion as “stridently chastis[ing]” the D.C. Circuit). That judges should leave procedural choices to agencies was “absolutely clear,” and the Court’s precedents “could hardly be more explicit in this regard.”²⁴⁴Vermont Yankee, 435 U.S. at 543–44. This view was fully supported by the APA’s legislative history, which demonstrated that Congress meant the informal rulemaking requirements to be a minimum that agencies, not courts, had discretion to build upon.²⁴⁵Id. at 545–46. And this approach was practical: if judges could second-guess agency proceedings, “judicial review would be totally unpredictable.”²⁴⁶Id. at 546–47. Faced with such uncertainty, agencies “would undoubtedly adopt full adjudicatory procedures in every instance,” eliminating the efficiency that informal rulemaking promised.²⁴⁷Id.

“[P]erhaps most importantly,” Justice Rehnquist concluded, the D.C. Circuit had “fundamentally misconceive[d] the nature of the standard for judicial review of an agency rule.”²⁴⁸Id. at 547. Adequacy of record and formality of process are not coextensive, and a reviewing court should determine whether the administrative record was adequate solely in relation to the procedures required by the APA or another statute.²⁴⁹Id. Requiring an agency to do more “can do nothing but seriously interfere with that process prescribed by Congress.”²⁵⁰Id. at 548. So long as an agency “employed at least the [§ 553] statutory minima,” a reviewing court had no power to “overturn [a] rulemaking proceeding on the basis of the procedural devices employed.”²⁵¹Id. Despite his fidelity to the text of the APA, Justice Rehnquist did identify two possible exceptions: cases with concentrated impacts, raising due process concerns, and cases involving “a totally unjustified departure from well-settled agency procedures of long standing.”²⁵²Id. at 542. Explaining why these “extremely compelling circumstances” would justify procedures beyond the statutory minima is an exercise Justice Rehnquist left to the reader.²⁵³Id. at 542–43.

Finally, the Court turned to the D.C. Circuit’s substantive critiques of the fuel cycle rule.²⁵⁴Id. at 549. While primarily crediting Judge Tamm’s concurrence, Justice Rehnquist grudgingly recognized “intimations” of a substantive problem with the NRC’s actions in Chief Judge Bazelon’s majority opinion.²⁵⁵Id. Since the sufficiency of the record to support the rule was indeterminate, the Court remanded the case for a proper substantive review.²⁵⁶Id. On remand, the D.C. Circuit once again concluded, in an opinion written by Chief Judge Bazelon, that the fuel cycle rule was invalid because it was inadequately supported by the record. See Nat. Res. Def. Council, Inc. v. NRC, 685 F.2d 459, 481, 485 (D.C. Cir. 1982). The Supreme Court once again disagreed, concluding that judges should be at their most deferential when reviewing an agency’s technical predictions in its area of expertise. See Balt. Gas & Elec. Co. v. Nat. Res. Def. Council, Inc., 462 U.S. 87, 103–04 (1983).

B.      The Aftermath: Hybrid Rulemaking Post-Vermont Yankee

“The victory is perfect—no other will venture against me . . . .”²⁵⁷Twain, supra note 1, at 332.

The Court’s unusually blunt language, and the potentially sweeping effect of its opinion on administrative law, assured that the case would become an instant classic.²⁵⁸See Metzger, supra note 10, at 160. Judge Leventhal saw the opinion as a final settlement, in his favor, of his debate with Chief Judge Bazelon.²⁵⁹See Peter L. Strauss, Changing Times: The APA at Fifty, 63 U. Chi. L. Rev. 1389, 1412 & n.68 (1996). Indeed, this understanding would become the canonical reading of the case.²⁶⁰See, e.g., Beermann & Lawson, supra note 22, at 858; Scott A. Keller, Depoliticizing Judicial Review of Agency Rulemaking, 84 Wash. L. Rev. 419, 442 (2009); Krotoszynski, supra note 14, at 996 (stating that Vermont Yankee “definitively rejected process-based review of agency action in favor of substantive ‘hard look’ review”); Warren, supra note 115, at 2631. Supporters of substantive review did not have long to wait for another victory; the Court officially blessed the hard look doctrine, which requires judges to carefully scrutinize the reasoning behind agency action, just five years later.²⁶¹See Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43–44 (1983).

The death of procedural hybrid rulemaking led courts to clear the field in regards to the most apparently “procedural” piece of informal rulemaking, § 553(c)’s public comment process.²⁶²See Beermann & Lawson, supra note 22, at 858 (“[F]ederal courts today do not feel free to require agencies to use oral hearings and cross-examination in informal rulemakings of adjudications without grounding in positive law.”). Meanwhile, the various substantive hybrid requirements which the D.C. Circuit had engrafted onto the notice process and the statement of basis and purpose survived and thrived, with Portland Cement as a prime example.²⁶³See Am. Radio Relay League, Inc. v. FCC, 524 F.3d 227, 246–47 (D.C. Cir. 2008) (Kavanaugh, J., concurring in part, concurring in the judgment in part, and dissenting in part) (criticizing Portland Cement and related cases as inconsistent with the APA but concurring that they are settled circuit precedent); Beermann & Lawson, supra note 22, at 857–60 (listing some of the substantive doctrines which survived Vermont Yankee). Many administrative-law scholars, wishing to recapture the revolutionary efficiency of the informal-rulemaking process, have pined for a Vermont Yankee II, wherein the Supreme Court would presumably bring down the hammer on these substantive APA embellishments.²⁶⁴See Beermann & Lawson, supra note 22, at 858–59 (discussing articles by Paul Verkuil and Richard Pierce). But other than a rather weak attempt to distinguish procedural and substantive hybrid requirements, the Court has largely let the subject alone.²⁶⁵See Pension Benefit Guar. Corp. v. LTV Corp., 496 U.S. 633, 654–55 (1990).

In the absence of new guidance from the Supreme Court, the substantive doctrines chronicled above²⁶⁶See supra Part I.B.1. have entrenched and strengthened.²⁶⁷See, e.g., Thomas O. McGarity, Some Thoughts on “Deossifying” the Rulemaking Process, 41 Duke L.J. 1385, 1385–86 (1992). Over the decades, the conglomeration of substantive judicial review doctrines has sapped much of the efficiency and flexibility out of informal rulemaking, a process known in the literature as “ossification.”²⁶⁸See id. Signs of this process include interminably long notices of proposed rulemaking and “concise” statements of basis and purpose—OSHA’s initial attempt at a COVID-19 vaccine mandate ran a whopping 154 pages, and that was just an interim final rule²⁶⁹See COVID-19 Vaccination and Testing; Emergency Temporary Standard, 86 Fed. Reg. 61,402 (Nov. 5, 2021) (the Supreme Court later granted a stay of the rule, leading OSHA to withdraw it. See NFIB v. Dep’t of Lab., 142 S. Ct. 661 (2022); COVID-19 Vaccination and Testing; Emergency Temporary Standard, 87 Fed. Reg. 3928 (Jan. 26, 2022).—and endless delays which slow, or sometimes outright kill, rulemaking attempts.²⁷⁰For another OSHA example, see OSHA’s and EPA’s nine-year struggle to promulgate a standard for a single chemical. McGarity, supra note 267, at 1388. In combination with the logical outgrowth doctrine—which requires final rules to be a logical outgrowth of the information contained in the notice of proposed rulemaking—ossified notice requirements have further neutered the public comment process.²⁷¹See Aaron L. Nielson, Optimal Ossification, 86 Geo. Wash. L. Rev. 1209, 1216 (2018); Aaron L. Nielson, Sticky Regulations, 85 U. Chi. L. Rev. 85, 97–98 (2018). Agencies are often unable or unwilling to modify a final rule in response to comments, meaningful or otherwise.²⁷²See McGarity, supra note 267, at 1390–92; see generally Phillip M. Kannan, The Logical Outgrowth Doctrine in Rulemaking, 48 Admin. L. Rev. 213 (1996). The ossification debates rage on,²⁷³See generally Stuart Shapiro, Embracing Ossification: With Donald Trump in the White House, Pro-Regulation Forces are Changing Their View on Regulatory Procedure, Regulation, Winter 2018–2019, at 8. and further examination is (fortunately) beyond the scope of this Comment.

Judges and agencies are not the only actors with a hand in shaping rulemaking in the decades since Vermont Yankee.²⁷⁴See generally Nielson, supra note 5. Congress has tinkered with new broadly applicable statutes such as the Freedom of Information Act,²⁷⁵5 U.S.C. § 552. which have a semi-constitutional character similar to the APA and NEPA.²⁷⁶See, e.g., Strauss, supra note 259, at 1392, 1406–07, 1420, 1422 n.113. On a more granular level, Congress has also imposed hybrid processes in some agency-specific organic statutes.²⁷⁷See, e.g., Nielson, supra note 6, at 245, 256. The executive branch has also gotten into the game by overseeing agency action through situationally mandatory reviews by the Office of Management and Budget and the Office of Information and Regulatory Affairs.²⁷⁸See id. at 256, 269 & n.213. Even agencies themselves have, on occasion, voluntarily adopted additional rulemaking procedures above the § 553 floor,²⁷⁹See generally, Bremer & Jacobs, supra note 7. as indeed the AEC had in Vermont Yankee.²⁸⁰Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 528–29 (1978). Despite these attempted reforms, our present world of informal rulemaking is certainly not ideal.²⁸¹See, e.g., McGarity, supra note 267, at 1385–87. What could be the harm in imagining a different, perhaps better, rulemaking world?

III.      Connecticut Yankee: The World That Might Have Been

“You know about transmigration of souls; do you know about transposition of epochs—and bodies?”²⁸²Twain, supra note 1, at 8.

To test the plausibility of the Leventhalite reading and to judge its effects on the rulemaking process, we journey to the hypothetical world of Connecticut Yankee. For want of Mr. Twain’s wit and imagination, and to retain some semblance of objective legal analysis, our look at that world will be brief, but hopefully fruitful. The basic conceit: Judge Leventhal leads the D.C. Circuit panel reviewing the fuel cycle rule and accompanying operating license determination. Justice Rehnquist pens a fiery response. What result?

A.      Leventhal Leads Off

“THE TALE OF THE LOST LAND”²⁸³Id. at 14.

As the Connecticut Yankee Nuclear Power Corporation entered the regulatory process in 1966, it could not have imagined what was to come. Work began along the banks of the Connecticut River following the AEC’s grant of the construction license in 1967, a relatively painless process. The subsequent request for an operating license would not prove as painless. New environmentalist intervenors, emboldened by the D.C. Circuit’s decision in Calvert Cliffs,²⁸⁴Calvert Cliffs’ Coordinating Comm., Inc. v. AEC, 449 F.2d 1109 (D.C. Cir. 1971). challenged the license on the basis of the unresolved problem of storing and processing spent fuel. The AEC dismissed the concerns as inapplicable to individual licensing applications, then turned around and began a rulemaking to address the problem for all future licensing proceedings. Shortly thereafter, both the adjudication and the rulemaking were hailed into the D.C. Circuit.

Arguments were heard before Judges Leventhal, Tamm, and Skelly Wright. Writing for the majority, Judge Leventhal conducted a thorough evaluation of the substance of the fuel cycle rule. In scrutinizing the positions of one AEC expert, Dr. Frank Pittman, Judge Leventhal concluded that the agency had failed to provide the proper basis for his technical conclusions, which formed a crucial piece of the rule. Citing to the Supreme Court’s 1971 decision in Citizens to Preserve Overton Park v. Volpe,²⁸⁵401 U.S. 402 (1971). Judge Leventhal found it to be his duty “to consider whether ‘the decision was based on a consideration of the relevant factors and whether there has been a clear error of judgment.’”²⁸⁶Portland Cement Ass’n v. Ruckelshaus, 486 F.2d 375, 402 (D.C. Cir. 1973) (quoting Overton Park, 401 U.S. at 416). To avoid unnecessary confusion, quotations from the world of Connecticut Yankee are cited to their source in the real world, rather than a facsimile.

In this case, the agency’s failure to provide all relevant information constituted a “critical defect in the decision-making process in arriving at” the fuel cycle rule.²⁸⁷Id. at 392. The Judge reiterated his longstanding views that “the court and agency are in a kind of partnership relationship for the purpose of effectuating the legislative mandate,” and that judges should “remain diffident in approaching problems of this technical complexity.”²⁸⁸Id. at 393–94, 402. Nevertheless, it was simply “not consonant with the purpose of a rule-making proceeding to promulgate rules on the basis of inadequate data, or on data that, [to a] critical degree, is known only to the agency.”²⁸⁹Id. at 393. Judge Leventhal concluded by announcing a general principle: agency “information should generally be disclosed as to the basis of a proposed rule at the time of issuance.”²⁹⁰Id. at 394. Since the AEC failed to meet this requirement, the court remanded both the fuel cycle rule and Connecticut Yankee’s operating license back to the agency.

B.      Rehnquist Responds

“It was pretty severe, but I was nettled.”²⁹¹Twain, supra note 1, at 20.

Connecticut Yankee appealed, and surprisingly, the Supreme Court granted certiorari. Justice Rehnquist, writing for the majority, took the D.C. Circuit to task. He began by recounting the history of the APA, describing its semiconstitutional status as “enact[ing] a formula upon which opposing social and political forces have come to rest.”²⁹²Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 523 (1978) (quoting Wong Yang Sung v. McGrath, 339 U.S. 33, 40 (1950)). Next, he delved into the APA’s text, noting that the Court had generally held that § 553 “established the maximum procedural requirements which Congress was willing to have the courts impose upon agencies in conducting rulemaking procedures.”²⁹³Id. at 524. While the Justice conceded that there may be times when additional impositions by courts were appropriate, such circumstances were necessarily rare.²⁹⁴See id.

Turning to the particulars of the lower court decision, Justice Rehnquist began by rejecting Judge Leventhal’s alleged reliance on Overton Park. That case contemplated a fact-specific inquiry into whether the reasoned decisions of the agency were to be found in the record.²⁹⁵Citizens to Pres. Overton Park v. Volpe, 401 U.S. 402, 419–20 (1971). Ordering additional procedures to fill out the record, beyond those required by § 553, was a step “usually to be avoided;”²⁹⁶Id. at 420. a stricture which could not support Judge Leventhal’s generalized disclosure requirement. Justice Rehnquist preempted the objection that Judge Leventhal had merely engaged in his trademark substantive review, noting that “the ineluctable mandate of the court’s decision is that the procedures” which created the notice of proposed rulemaking “were inadequate.”²⁹⁷Vermont Yankee, 435 U.S. at 542.

The Court’s precedents supporting this conclusion “could hardly be more explicit.”²⁹⁸Id. at 544. United States v. Allegheny-Ludlum Steel Corp.,²⁹⁹406 U.S. 742 (1972). a 1972 Justice Rehnquist opinion, upheld an agency’s findings in the statement of basis and purpose against a substantive attack.³⁰⁰See id. at 758. It did so by again returning to the text of the informal rulemaking provisions, the statutory minima.³⁰¹See id. Florida East Coast Railway, another Justice Rehnquist opinion, reaffirmed a strict reading of the APA’s text by holding that formal rulemaking was triggered only by the magic phrase “on the record after opportunity for an agency hearing.”³⁰²United States v. Fla. E. Coast Ry., 410 U.S. 224, 237 (1973) (referencing APA § 553(c)). Less famously, the case resisted a substantive challenge to a notice of proposed rulemaking, even though “the initial notice of the proceeding by no means set out in detail what the [agency] proposed to do.”³⁰³Id. at 243. Twisting the knife, Justice Rehnquist quoted Judge Leventhal’s friend and colleague, Judge Skelly Wright, for the proposition that the Supreme Court’s “ringing message” of APA textualism had been heard by the D.C. Circuit and then ignored.³⁰⁴J. Skelly Wright, Court of Appeals Review of Federal Regulatory Agency Rulemaking, 26 Admin. L. Rev. 199, 206–07 (1974). As Judge Leventhal’s disclosure requirement found no support in either § 553 or Supreme Court precedent, the judgment was reversed. The NRC walked away with its fuel cycle rule intact, and the Connecticut Yankee plant could finally operate without a legal storm cloud overhead.

C.      Hybrid Rulemaking After Connecticut Yankee

“When the spirit of prophecy comes upon you, you merely cake your intellect and lay it off in a cool place for a rest, and unship your jaw and leave it alone; it will work itself: the result is prophecy.”³⁰⁵Twain, supra note 1, at 234.

The harsh language and potentially sweeping impact of the Connecticut Yankee opinion guaranteed it a place in administrative law textbooks for decades to come, though the actual impact of the decision appeared slight. Chief Judge Bazelon took it as a vindication of his doubts about the substance-process dichotomy, but his D.C. Circuit colleagues did not admit defeat. Instead, they read the case mostly as a reminder, a strong reminder to be sure, that substantive review of agency actions should be a narrow inquiry.³⁰⁶See Citizens to Pres. Overton Park v. Volpe, 401 U.S. 402, 416 (1971). The D.C. Circuit continued its APA embellishments, though more circumspectly, and none were bold enough to try to resurrect Judge Leventhal’s disclosure requirement. Chief Judge Bazelon went on ordering hearings and cross-examinations, though only in those rare circumstances where it was appropriate.

The crystal ball gets hazy here. Without the Portland Cement-style disclosure requirement beefing up judicial review of the notice process, the march of ossification was mitigated somewhat. With agencies permitted, more or less, to get by with just the § 553 required “description of the subjects and issues involved” in a notice of proposed rulemaking,³⁰⁷5 U.S.C. § 553(b)(3). the process retained some of its vaunted flexibility. The comment process was occasionally interrupted by a judicially imposed hearing or cross-examination, but, like all common-law innovations, this eventually became routine and (relatively) predictable. The relaxation of notice requirements brought a concomitant loosening up of the statement of basis and purpose, as a larger universe of possible final rules logically outgrow from more general beginnings. And they all lived happily ever after.

IV.      I-91, or: The Road From Vermont to Connecticut

“Wit ye not the law?”³⁰⁸Twain, supra note 1, at 314.

As we return to the real world, what can we take away from our sojourn? Is it possible to reconcile Connecticut Yankee with Vermont Yankee, the law we actually have? How so, and how much? What effects would embracing such a reading have, and how would those changes interact with other rulemaking reforms?

A.      A Hybrid Reading of Vermont Yankee

“Intellectual ‘work’ is misnamed; it is a pleasure, a dissipation, and is its own highest reward.”³⁰⁹Id. at 242.

A threshold question before we even consider a new reading of Vermont Yankee is: why bother? Why not simply join the chorus of scholars crying out for a Vermont Yankee II? After all, we are all good textualists now.³¹⁰See Harvard Law School, The Antonin Scalia Lecture Series: A Dialogue with Justice Elena Kagan on the Reading of Statutes, YouTube (Nov. 25, 2015), https://perma.cc/NHH9-4R3A. Surely the bare text of the APA should govern, and the various judicial embellishments should be swept aside. Pursuing a new reading of a decades-old precedent to allow more intermeddling with agencies seems counterproductive when judges should get out of the game entirely, no?

While the prospect of a solidly textualist Vermont Yankee II is certainly attractive, it is also a pipe dream. As Professor Gillian Metzger has demonstrated, administrative common law is inevitable.³¹¹See Metzger, supra note 8, at 1320–42, 1325 n.162. Professor Metzger continues on to show that administrative common law is also legitimate, which is a plus, though not particularly relevant here. Administrative common law “plays too important a role in enabling the courts to navigate the challenges of modern administrative government” to ever be completely discarded.³¹²Id. at 1320. The APA requires judicial review in § 702, and judges have a duty to make that review effective.³¹³5 U.S.C. § 702. If the APA and applicable organic statutes do not provide the tools necessary to do so, judges will, and should, create them. As a prime example, recall the D.C. Circuit’s practice through the 1960s and 1970s of essentially ignoring Supreme Court rulings.³¹⁴See Scalia, supra note 3, at 359–68. There, despite “occasional stern rhetoric condemning administrative common law and no express judicial defense,” the court continued to innovate ways to effectuate judicial review largely uninterrupted.³¹⁵Metzger, supra note 8, at 1320. Recall also the survival of Portland Cement-related doctrines beyond Vermont Yankee.³¹⁶See Am. Radio Relay League, Inc. v. FCC, 524 F.3d 227, 245–47 (D.C. Cir. 2008) (Kavanaugh, J., concurring in part, concurring in the judgment in part, and dissenting in part).

If we accept some amount of judicial input into the rulemaking process, we must then determine the form and scope of that input. The potential bounds of that determination, in turn, will be circumscribed by our understanding of Vermont Yankee, the undisputed king of “stern rhetoric condemning administrative common law.”³¹⁷Metzger, supra note 8, at 1320. As a jumping off point, Professors Beermann and Lawson have identified three plausible readings of the opinion.³¹⁸See Beermann & Lawson, supra note 22, at 868–73. The broad reading resembles the wished-for Vermont Yankee II, essentially calling for an originalist understanding of the APA and administrative law circa 1946.³¹⁹Id. The “natural” reading would call for an originalist understanding of the APA “specifically with respect to agency procedures,” effectively proscribing judicially imposed procedures throughout the APA, including in both rulemaking and adjudication.³²⁰Id. at 871. The narrow reading would hew to the facts of the case and proscribe only judicially imposed procedures in the course of a rulemaking after the notice phase.³²¹Id. at 868–73. This narrow reading is closest to the canonical understanding.³²²Id.

What are the features of a Connecticut Yankee–esque reading of Vermont Yankee? For starters, it is broader than the “natural” or narrow reading, since it rejects a strict substance-process dichotomy. Substantive judicial requirements can impose procedures on agencies just as readily as explicit procedural impositions, so a reading confined to proscribing specific procedures cannot be right. At the same time, the Connecticut Yankee reading is not coextensive with Beermann and Lawson’s broad reading, since it envisions a continued, albeit limited, role for administrative common law innovations throughout the informal rulemaking process and the APA generally. The Connecticut Yankee reading, then, can perhaps best be described as the broad reading as it would have been interpreted and implemented by the D.C. Circuit of the 1960s and 1970s. Overt announcements of general, judicially imposed requirements beyond the text of § 553 are proscribed, but case-by-case evolutions of similar requirements over time continue: a hybrid reading for hybrid rulemaking.

While Beermann and Lawson list the broad reading of Vermont Yankee as plausible, they certainly do not consider it the best reading, noting inconsistencies with the Court’s general trend towards substantive review.³²³Id. at 870. Nevertheless, the broad reading is—to coin a phrase—a logical outgrowth of the opinion. Getting there begins with rejecting the substance-process dichotomy, which Justice Rehnquist does. Recall that Chief Judge Bazelon’s lower court opinion explicitly imposed no specific procedures on the agency.³²⁴Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 U.S. 519, 541 (1978). Accordingly, Justice Rehnquist had to look behind the curtain of the substantive requirements to reveal the procedural impositions within: “the ineluctable mandate of the court’s decision is that the procedures afforded . . . were inadequate.”³²⁵Id. at 542. As then-professor Scalia stated in his review of the case, “it is not possible to maintain a complete dichotomy between the procedures used and the adequacy of evidentiary support.”³²⁶Scalia, supra note 3, at 391. One could imagine nearly pure procedural requirements—”the agency head shall sign his name at the bottom of each notice of proposed rulemaking”—just as one could imagine nearly pure substantive requirements—”the agency shall make good rules.” Just about everything else, including the provisions of the APA, fall somewhere in the middle. See also Bremer & Jacobs, supra note 7, at 526–27. Indeed, the latter half of the opinion, concerning a companion case³²⁷Vermont Yankee, 435 U.S. at 527 (citing Aeschliman v. NRC, 547 F.2d 622 (D.C. Cir. 1976)). and employing even more fiery language,³²⁸See Scalia, supra note 3, at 370. effectively drew no distinction, emphasizing that “administrative decisions should be set aside . . . only for substantial procedural or substantive reasons.”³²⁹Vermont Yankee, 435 U.S. at 558.

The canonical reading holds that the Court not only appreciated the substance-process dichotomy, but that its harsh language represented a definitive end of the Leventhal-Bazelon debate on the subject. Some additional context may point away from this background assumption, making the broader reading more likely. First, the clearest source of the Court’s harsh language was the D.C. Circuit’s years of dodging administrative law rulings. It is “the exasperated tone of one not explaining a new point of law but unnecessarily reiterating an old one,” rather than a particular ire towards Chief Judge Bazelon’s procedural philosophy.³³⁰Scalia, supra note 3, at 369. These common-law-style innovations had appeared in the shadows of Supreme Court rulings both substantive and procedural.³³¹See id. at 359–66. Second, the Court’s unusual interest in the case, and therefore some of the strength of the holding, may have stemmed from a particular interest in nuclear power, as evidenced by other cases around the same time and Justice Rehnquist’s own memos recounting his push for certiorari.³³²See Metzger, supra note 10, at 159. Finally, Justice Rehnquist’s jurisprudence reveals a general diffidence towards powerful substantive review doctrines like those advanced by Judge Leventhal. While he did join the State Farm majority, Justice Rehnquist penned the partial dissent which embraced a very narrow view of what an agency needs to do to satisfy the “hard look” doctrine.³³³See Motor Vehicle Mfg. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 57–59 (1983) (Rehnquist, J., concurring in part).

With the benefit of hindsight, many of the Justice’s dire predictions of what would follow from muscular procedural requirements have proven applicable to the various “substantive” doctrines that survived Vermont Yankee. Unpredictability is, at least in some degree, a feature of any common-law system.³³⁴See Vermont Yankee, 435 U.S. at 546 (stating that judicial review would be “totally unpredictable” if judges were permitted to second-guess agency procedures). Were judicial expectations of what agency actions satisfied “hard look” review immediately predictable following State Farm? The answer must be no.³³⁵See, e.g., Jerry L. Mashaw, The Story of Motor Vehicle Manufacturers Association of the U.S. v. State Farm Mutual Automobile Insurance Co.: Law, Science and Politics in the Administrative State, in Administrative Law Stories 334, 374–81, 386–89 (Peter L. Strauss ed., 2006) (discussing philosophical difficulties with the doctrine and recounting inconsistent applications to NHTSA rules in the years following State Farm). It is hard to see why a similar evolution into regularity would be impossible for “procedural” impositions like hearings and cross-examinations.³³⁶Then-professor Scalia suggested that judges would bring institutional competencies to bear on such a project. Scalia, supra note 3, at 386–87 (noting that trying to force every administrative action into the APA framework “eliminates the justification (not to mention the purpose) . . . for restraining the courts from returning to their pre-APA ways of developing an administrative common law—a task they are at least as well equipped, and probably better motivated, to perform than are the substantive committees of Congress which consider procedural issues en passant”). How about the possibility that agencies, in the face of such unpredictability, would be forced to expend maximal resources to insulate every agency action from a remand?³³⁷See Vermont Yankee, 435 U.S. at 546–47. One can sift through the reams of ossification articles to see how that turned out under the various “substantive” doctrines.³³⁸If you insist, see generally McGarity, supra note 267; Jason Webb Yackee & Susan Webb Yackee, Testing the Ossification Thesis: An Empirical Examination of Federal Regulatory Volume and Speed, 1950-1990, 80 Geo. Wash. L. Rev. 1414 (2012); Richard J. Pierce, Jr., Rulemaking Ossification is Real: A Response to Testing the Ossification Thesis, 80 Geo. Wash. L. Rev. 1493 (2012); William S. Jordan, III, Ossification Revisited: Does Arbitrary and Capricious Review Significantly Interfere with Agency Ability to Achieve Regulatory Goals Through Informal Rulemaking?, 94 Nw. U. L. Rev. 393 (2000); Nielson, Optimal Ossification, supra note 271.

Another piece of Judge Leventhal’s philosophy, the supposed agency-court partnership, has fared little better. At the time, Judge Henry Friendly could joke that courts had no problem declaring such a partnership because they, not the agencies, would be the superior partners.³³⁹See Henry J. Friendly, “Some Kind of Hearing,” 123 U. Pa. L. Rev. 1267, 1311 n.221 (1975). Today, with the development of various deference doctrines, that dominance is far from clear.³⁴⁰See generally Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke L.J. 511; Richard J. Pierce, Jr., The Future of Deference, 84 Geo Wash. L. Rev. 1293 (2016); Nicholas R. Bednar & Barbara Marchevsky, Deferring to the Rule of Law: A Comparative Look at United States Deference Doctrines, 47 U. Mem. L. Rev. 1047 (2017). Even the detailed and rigorous technical “steeping” that Judge Leventhal prescribed has atrophied, if it ever really existed at all.³⁴¹See Emily Hammond Meazell, Super Deference, the Science Obsession, and Judicial Review as Translation of Agency Science, 109 Mich. L. Rev. 733, 756–64 (2011) (recounting the development of judicial deference on technical matters).

A continually evolving administrative common law does not necessarily mean a continually increasing share of judicial power.³⁴²See Metzger, supra note 8, at 1346–47. The expansion of hearing and cross-examination requirements, even when judicially imposed, would not render agencies powerless to shape the process.³⁴³See Bremer, supra note 7, at 76–80 (describing agency tools for shaping hearing processes). Chief Judge Bazelon predicted that a judicial focus on agency process might prove less intrusive than rigorous substantive review in the long run.³⁴⁴See Krotoszynski, supra note 14, at 999–1000; supra note 136 and accompanying text. Would any agency today accept the possibility of holding a hearing in exchange for jettisoning the Portland Cement disclosure requirement? Could an administrative common law able to tinker with both agency “substance” and “procedure” effectuate the same level of judicial review in a more efficient way than the processes we live with now? Would agencies be less inclined to escape into regulatory dark matter if informal rulemaking were more efficient? At the very least, these are questions worth asking.³⁴⁵See William J. Brennan, Jr., Introduction, 63 Geo. L.J. 2, 4 (1974) (admiring “Judge Bazelon’s firm conviction that asking the right questions is often a great deal more important than getting the right answers”).

B.      Waiting for Connecticut Yankee

“Ah, please you sir, it hath no direction from here; by reason that the road lieth not straight, but turneth evermore . . . .”³⁴⁶Twain, supra note 1, at 82.

Even if we were inclined to embrace the Connecticut Yankee reading, what good would it do to do so today? We cannot wish into existence an alternative forty years of administrative common law development. The development that occurred, under the narrow Leventhalite reading, appears as unshakeable now as if it were set in portland cement. The Supreme Court seems loathe to revisit the issue, and a congressional overhaul of the APA, such as then-professor Scalia’s fanciful proposal of fifteen-plus different statutory rulemaking procedures,³⁴⁷See Scalia, supra note 3, at 408. is not visible on the horizon.

Change is difficult in the world of administrative law, as it should be.³⁴⁸See Nielson, supra note 5, at 758–61, 818. Sometimes, the best we can do is create the proper atmosphere for change. A questioning attitude and a push for transparency are two tools for creating such an atmosphere. Just as judges should be more transparent about their use of administrative common law,³⁴⁹See Metzger, supra note 8, at 1356–58. it may be time to more openly question the canonical reading of the case that has largely circumscribed that common law development for four decades. Ongoing reform efforts, whether they be in the realm of adjudication³⁵⁰See, e.g., Michael S. Greve, Why We Need Federal Administrative Courts, 28 Geo. Mason L. Rev. 765, 808–13 (2021). or rulemaking,³⁵¹See, e.g., Nielson, supra note 6, at 242. can only be strengthened by such inquiries.

For the crowd waiting for Vermont Yankee II, Connecticut Yankee provides another arrow in the quiver. Delving beyond ongoing issues of law and policy, an appreciation of the Bazelon-Leventhal debates highlights the historical weakness of the canonical Vermont Yankee world we know—a world effectively created by coinflip when Justice Rehnquist attempted to lower the boom on hybrid rulemaking as a whole. And should Vermont Yankee II ever come to pass, fans of Connecticut Yankee and Chief Judge Bazelon’s procedural hybrid rulemaking can rest safe in the knowledge that a fresh growth of administrative common law would be close behind.

Conclusion

“Let the record end here.”³⁵²Twain, supra note 1, at 375.

For four decades the tale of Vermont Yankee has recounted Judge Leventhal’s glorious victory over Chief Judge Bazelon, and the consignment of procedural hybrid rulemaking to the dustbin of history. Looking back now, it is unclear that this reading of the case was right, and even less clear that it was desirable. The alternate history of Connecticut Yankee suggests that a hybrid rulemaking system, carefully circumscribed but with more tools for judicial tinkering, could more efficiently and effectually empower judicial review of agency action. Whether such a system is an option now available to us is an open question. At the very least, however, the thought experiment forces us to consider that Chief Judge Bazelon’s process-based approach may be more than an interesting, perhaps even illuminating, historical artifact. He may have been right.