Structural Indeterminacy and the Separation of Powers

Despite ongoing disagreement about how the Constitution allocates powers among the different branches, the two dominant schools of thought in American separation-of-powers debates—formalism and functionalism—agree on three premises: Certain powers inhere in certain government branches, some powers are vested exclusively in one or another branch, and the judiciary is the final arbiter of separation-of-powers disputes. Disagreement is largely about how powers should be parsed and which should be shared. Yet over the long lifespan of our constitutional tradition, momentous doctrinal upheavals are relatively commonplace. This Article describes four tectonic shifts in separation-of-powers doctrine: Founding-era debates about how to define and blend powers, nineteenth-century debates about the constitutionality of the nascent civil service, Lochner-era debates about legislatures’ authority to define and regulate public utilities, and mid-nineteenth-century debates about the sources of international law. The first, which we call the Inherency Theory, assumes that certain powers and functions are vested by force of the Constitution, are core to a single branch, and are discernible by the judiciary. This is a taxonomical theory of how the Constitution allocates powers, and it animates nearly all of today’s separation-of-powers debates. The second, an Antidomination Theory, denies that the words executive, legislative, and judicial imply any new or distinct powers and instead creates formal separation between the three branches based on the procedures federal actors deploy to enact, enforce, and interpret policy. The third, a rights-based Public Utility Theory, distinguishes between a public sphere that is subject to congressional, presidential, and administrative control, and a private sphere that is not. Recently, this public-private distinction has been marshalled to define the judicial power. Historically, however, it was used to deduce a whole panoply of structural limits, including the constitutionality of agency adjudication and deference. And the fourth, a General Law approach, discerns the limits of government power by reference to the eclectic authority of the common law and right reason. Recovering these theories reveals a rich set of tools for resolving interdepartmental disputes, highlights that current receptions of past settlements are nearly unintelligible without understanding the theoretical context in which they emerged, and suggests that, while different theories have risen and fallen, no one theory of separation of powers has been liquidated in our constitutional tradition.

Table of Contents Show

Introduction

The two dominant schools of thought about the separation of powers—formalism and functionalism[1]—are both committed to a tripartite system under which the judicial branch enjoys a privileged position in the development of constitutional rules. To that end, both think that the Constitution distributes powers among three coequal branches and the states,[2] that some powers are enumerated and others implied, that the separation of powers operates as a restraint on government overreach, and that the judiciary should arbitrate separation-of-powers disputes. Disagreement is primarily about specifics: How many powers does the Constitution imply? What are they? Does the Constitution blend or mix powers among the branches? And what historical moments are relevant in discerning the limits of congressional and presidential power?

Historically, however, the United States has entertained, and sometimes embraced, separation-of-powers theories that reject many of the premises about which modern formalists and functionalists agree. Consider, for example, nineteenth-century debates about the President’s authority to remove civil service officials. Snippets of this long constitutional contest have become canonical moments in the debate about the President’s authority to remove administrative officials. But the removal issue has not always been framed as being about the President’s relationship to Congress. During the nineteenth century, arguments about the President’s authority to fire members of the civil service were often based on policymakers’ views about political science,[3] by appeals to a higher order such as divine providence or right reason, or as primarily implicating the government’s relationship to individual rights. In other words, constitutional theorists advanced unfamiliar structural arguments to try—unsuccessfully—to resolve one of our most familiar structural disputes.

We describe four alternative approaches to resolving interdepartmental disputes.[4] First, some members of the founding generation embraced an Antidomination theory about legislative and executive powers.[5] This theory denied that the terms “executive” or “legislative” implied justiciable limits on government authority, at least beyond certain specifically enumerated powers. It instead stipulated processes by which the legislative and executive branches could enact legally binding policies: Legislation must be enacted through bicameralism and presentment, and the President must act via express statutory or constitutional authorization. The constitutionally relevant fact was that each branch respected the Constitution’s procedural constraints. To be sure, other structural theories travelled alongside this view, but for some early interpreters of the Constitution, the drafters’ political theory focused on the document’s procedural checks as the “practical” elaboration of the core idea, not on metaphysical appeals to the concept of “executive,” “legislative,” or “judicial.”[6]

Second, a rights-based “Public Utility” theory reflected a libertarian commitment to protecting a sphere of private conduct. Courts used the distinction between public rights and public enterprises, on the one hand, and private rights and private enterprises, on the other, to ground the constitutionality of agency deference, adjudication, and rulemaking. Many of today’s separation-of-powers concerns were therefore subsumed under the question of the publicity of the actor subject to regulatory oversight. This reasoning is typically associated with the Lochner era and, to our knowledge, has not been understood as a theory of interbranch relations.[7] Yet, in providing the decisive test to determine when government could control private conduct, the Public Utility idea also came to settle interbranch disputes about legislatures’, agencies’, and courts’ roles in regulating economic activity.

Third, under a Political theory approach, superordinate moral values transcended any individual sovereign order and could be codified as constitutional principles by enlightened judges. This approach has come to be known as the General Law and is, somewhat ironically, now associated with modern positive law theorists.[8] In reality, this mode of constitutional interpretation blended reason, custom, and religion.[9] Judges reasoned their way to the correct scientific constitutional theory based on a body of general and universally applicable rules that often appealed to anachronistic understandings of divine providence.[10]

Fourth, a Social Science theory discerned constitutional limits through technocratic and economic reasoning. In this view, the meaning of the Constitution changed based on evolving practical considerations. For example, within a four-decade period in the nineteenth century, concern about the effectiveness of the civil service was used to justify both presidential removal of civil servants and Congress’s role in regulating removal. Initially, corruption in the civil service justified heightened removal powers and a rotation-in-office system.[11] Then, a decade later, concern about the government’s inability to retain expert administrators caused the tide to shift toward more robust removal protections.[12] These were thought not to reflect incompatible theories of the executive power, but rather an understanding that constitutional principles should evolve with changing economic and political circumstances.[13] Cases from this fourth category have been marshaled to support different theories of departmental limits, many of which have continued relevance in modern formalist and functionalist debates. However, this category was based on a wholly different constitutional cast of mind: one in which structural limitations followed from a political-scientific inquiry that could be answered through both philosophical reasoning and political empirics.

Our taxonomy of separation-of-powers theories is not exhaustive.[14] Nor do we mean to suggest that any particular theory rests on more stable constitutional footing than the others. Just the opposite. Different theories have operated concurrently.[15] For example, both the view that the President enjoys a constitutional power to remove administrative officials at will and the converse idea that legislators can provide tenure protections to members of the civil service have been based on textual and historical arguments, relied on formal distinctions between the different branches, invoked the law of nations as a guide to general public law, and appealed to prudential arguments about how to promote the public welfare.[16]

Although these alternative theories have faded from our legal imagination, they continue to exert a significant, albeit indirect, influence on U.S. constitutional law.[17] Consider a few examples:

1.   Last term, in Loper Bright Enterprises v. Raimondo,[18] the Supreme Court cited the public utility case Saint Joseph Stock Yards v. United States to argue that judicial deference to agency legal determinations violates the separation of powers.[19] The Court used Saint Joseph to suggest that agencies have discretion to make findings of fact but not of law. Yet the Saint Joseph Court was not concerned that a nonlegislative actor exercised legislative power or that a nonjudicial actor exercised judicial power, at least not in the modern sense. It was instead considering when deference to administrative agencies interfered with individual property protections.[20] Though courts are now relying on century-old deference cases to reinvigorate a formalist separation of powers, these cases were originally about administrative discretion to regulate public utilities.[21]

2.   Recently the Supreme Court has invoked a public-private distinction to narrow administrative authority to adjudicate claims.[22] Relying on mid-nineteenth-century cases, the Court has held that non-Article III tribunals should have only limited authority to decide property and contract disputes among private parties.[23] This precisely flips the logic courts used in the early twentieth century when, under something called the “filed rate doctrine,” courts held that agencies, not courts, must adjudicate contract, tort, and antitrust disputes, at least so long as the claim touched on a rate set by a public utility regulator.[24]

3.   Modern scholars of presidential authority often point to a “Decision of 1789” as evidence that the executive power conveys a presidential right to fire inferior administrative officials for any reason.[25] The Decision of 1789 stands for the idea that this structural question was settled by the first Congress. Yet in the early 1800s, in response to concerns about administrative corruption, consecutive Senate reports urged that the Decision of 1789 be “reconsidered by Congress whenever experience should demonstrate the necessity of interposing to arrest the abuse of the power by a less scrupulous and more ambitious Executive.”[26]

These examples draw attention to jurisprudential puzzles that underlie modern separation-of-powers debates. It is not clear, for example, how to decide which period is relevant in settling debates about the constitutionality of adjudication or deference, or why the structural reasoning that underlies the Decision of 1789 has more contemporary purchase than the prudential reasoning that led many of the same constitutional actors to question that earlier settlement.

There exists no conflict-of-law principle for deciding among competing separation-of-powers theories, especially when courts draw from conflicting theoretical frameworks to resolve modern separation-of-powers controversies. Different theories have risen and fallen, but no one theory has been fixed or liquidated through custom or historical practice.[27] Indeed, the same styles of legal reasoning—arguments based on text and structure and history—have supported different constitutional theories. The General Law theory, for example, expressly subordinates questions about ambiguous constitutional terms in order to consult customary practices that reflect exogenous patterns of our political, electoral, and economic lives. The persistence of pluralism in separation-of-powers debates does not suggest that constitutional theory is a confused muddle that cannot be resolved, but rather that the Constitution’s indeterminacy leaves apparent separation-of-powers settlements open to reinterpretation and reinvention.[28]

This Article proceeds in three parts. Part I shows that formalists and functionalists largely agree about the purpose of the separation of powers and the judiciary’s role in settling interdepartmental disputes. Both formalism and functionalism are theories of inherent powers. The Constitution vests certain core powers in a single branch. Legal reasoning is deductive, and the judiciary plays a central role in expounding constitutional meaning. Part II provides an overview of four alternative separation-of-powers theories. Part III analyzes modern constitutional controversies that continue to draw from and co-opt doctrines from older episodes. Our goal, however, is not purely deconstructive. Instead, understanding how and why these alternative theories moved from plausible to implausible over time begins to fill out a more authentic view of the development of our constitutional tradition.

I. Modern Separation-of-Powers Debates

Today most scholars identify with one of two separation-of-powers camps. One, known as formalism, insists that clear, judicially managed distinctions separate the different powers. The other, known as functionalism, accepts that interbranch relations can overlap so long as each branch retains a “core” set of (typically undefined) powers.[29] Yet the two agree on a great deal.[30]

A.      Formalism and Functionalism

The Constitution, we are told, divides power among three federal branches to preserve individual liberty and facilitate good government. That axiom may be true in an abstract sense, but it tells us little about how to resolve actual interbranch disputes that arise when the Constitution’s text does not produce obvious answers. To do this, one needs a theory that yields concrete answers when the branches come into conflict. The most popular modern candidates are formalism and functionalism.

Formalist reasoning often proceeds as though each branch possesses a discrete and stable suite of powers that can be derived from superordinate rules.[31] The act of defining, systematizing, and taxonomizing these powers is the primary means of resolving interbranch conflicts.[32] Separation-of-powers issues arise when an act is either outside a branch’s power or deprives another constitutional actor of its inherent authority. Formalist constitutional interpretation can thus be understood as a two-step process: first, define a branch’s powers; second, determine whether an act either exceeds that branch’s power or impermissibly exercises a power that the Constitution assigns to another branch.

Consider, for example, the nondelegation doctrine, which posits that Congress cannot delegate broad rulemaking authority to administrative agencies.[33] The Constitution assigns legislative power to Congress.[34] According to proponents of the nondelegation doctrine, the legislative power includes the power to enact prospective rules of general applicability.[35] By definition, agency rulemaking consists of rules of general applicability. It therefore violates a formalist separation of powers, since it involves the exercise of legislative power by a non-legislative actor.[36] The same mode of reasoning applies to the President’s authority to remove administrative officials. Article II vests executive powers in the President.[37] According to formalists, that grant conveys exclusive authority to supervise executive departments and thus carries with it the power to appoint and remove personnel.[38] While the appointment power is formally shared with Congress, the Constitution’s silence about removal suggests that the residual power must be the President’s alone. Accordingly, when Congress imposes restrictions on the President’s authority to remove administrative officials, it interferes with the President’s power to execute the law.[39]

Functionalists, by contrast, reject the premise that constitutional grants of power typically exclude other branches from exercising that power.[40] For them, the fact that one branch can wield a given power does not foreclose another branch from doing so.[41] Instead, functionalists tolerate some level of mixing beyond those expressly named in the Constitution. Acts are constitutionally suspect if they interfere with another branch’s “core” powers or if the exercise of a given power disrupts the overall balance of powers.[42] Agency rulemaking is therefore acceptable so long as Congress has authorized the agency to enact rules;[43] removal restrictions are constitutional so long as they do not “impermissibly interfere” with a “purely executive” function;[44] and agency adjudication can be permissible even when the agency exercises “core” judicial functions. The agency just cannot go too far.[45]

The two camps also disagree about how to resolve interdepartmental conflicts. Because formalists interpret the Constitution to demarcate clear lines separating the three federal branches, for them, the allocation of a power to a particular branch implies that the other branches cannot exercise that power.[46] Functionalists, by contrast, are more comfortable with a federal government in which multiple branches exercise the same functions simultaneously. Thus, formalists are skeptical of broad statutory delegations to agencies because the Constitution vests the legislative power to Congress, and only Congress. Functionalists respond that this is an acceptable way to structure the government to control modern social and economic enterprise. Similarly, while formalists argue that the President possesses wide discretion to fire administrative officials as part of their executive power, functionalists think that removal protections are a sensible way to check presidential overreach. And while formalists insist that cases and controversies, especially those involving private rights, must be adjudicated in Article III tribunals, functionalists accept administrative adjudication as an efficient way to resolve highly technical proceedings that often require a unique expertise that only agencies can cultivate over time.

B.      Convergence in Formalism and Functionalism

Despite areas of disagreement, formalists and functionalists share important premises. Both think that the Constitution allocates power among three coequal branches, that some powers are enumerated and others implied, and that the judiciary is (usually) the final arbiter of separation-of-powers disputes.[47] Disagreement is largely about how to parse those powers.

Perhaps the most obvious point of agreement is about the role of the judiciary in mediating separation-of-powers disputes. While academics and some Supreme Court opinions have occasionally flirted with a departmentalist theory that limits judicial review,[48] functionalists accept that the judiciary should demarcate the lines separating each branch. Both approaches thus accept a version of judicial supremacy: It is a judicial prerogative to police executive and legislative overreach. Judges, not members of Congress and not the President, are the primary federal officials charged with expounding constitutional meaning, delineating the boundaries that separate the different branches, and stipulating which powers are shared and which are exclusive.

Similarly, both formalists and functionalists understand the separation of powers to rely primarily on a tripartite federal system in which power is allocated to a legislative, an executive, and a judicial branch, and in which there is some sort of balance between the branches. Agencies may exercise quasi-legislative, quasi-executive, and quasi-judicial powers,[49] but the “quasi” label still consigns to judicial interpreters the role of navigating the separation-of-powers cladogram. Quasi-powers are understood through a process of judicial classification, as acts that are associated with one of the three branches. Agency rules resemble federal legislation and are therefore defined by relation to their legislative cousin. Adjudication resembles judicial review and is therefore a quasi-judicial power. And enforcement resembles prosecution, a prerogative typically housed in the executive branch.

Formalists and functionalists also agree on the civic goals that the separation of powers promotes, though they place different weight on different ends. A common justification for the distribution of powers among the different branches is that it protects individual liberty from the threat of government overreach, an anti-tyranny ideal. On this view, the Madisonian axiom that “ambition must be made to counteract ambition” is intended to limit government: Each branch’s pursuit of power will induce the other branches to restrain it.[50] At other times, the separation of powers has been justified for promoting reasoned deliberation and requiring compromise between competing geographic and political interests.

One way to understand the convergence between formal and functional modes of constitutional reasoning is to notice their shared predilection for definite legal answers.[51] Neither formal nor functional approaches to the separation of powers are “nominalist” or “legal realist.” Rather, each agrees that its chosen approach to constitutional questions corresponds to real features of our public law, even if the two approaches do not always agree on what the relevant ordering principles ought to be. Modern separation-of-powers theorists assume, for example, that there is a thing called Executive Power or, more broadly, a set of constitutional functions that can be discerned and reliably adjudicated. Because these legal features—e.g., the Executive Power or the function of bicameralism and presentment—can be pinned down, each theory predicts fixed answers to future separation-of-powers disputes.

Convergence among separation-of-powers theories presumably reflects complex historical puzzles combining politics, doctrine, and ideologies of elite legal personnel. Whatever their genealogy, however, most separation-of-powers theories are, in the end, formal: They take themselves to “provid[e] practically reasonable normative guidance to [the law’s] addressees in authoritative resolution of conflict and coordination issues that face a political community, thereby enabling the community to realize its aspirations to legality.”[52]

As the next two parts explain, separation-of-powers theories that have arisen in our constitutional history are neither formal nor functional, at least not in the modern sense. Rather than providing ex ante normative guidance that would enable a court to resolve the power-coordination issues that arise in government using familiar legal argot, some of these modes of constitutional argument directly embody a political theory, a science of politics. Rather than enumerating a set of legal rules, of whatever breadth, to guide the resolution of new power-coordination disputes, these theories instead equate the relevant separation-of-powers rules with the most important ideological features of our political community.[53]

Crucially, these theories are less concerned with fixity in legal answers than the principal formal and functional theories of separated powers. They are less interested in “legality,” insofar as that label identifies a kind of legal reasoning that proceeds deductively from known doctrinal premises to predictable and stable legal conclusions over time. And they are less interested in “interpretation,” insofar as that label refers to “revealing an intelligible order in the law, so far as such an order exists.”[54] These theories treat the separation of powers as an appeal to the latest and best modes of political science, not as a legal concept whose content is controlled by the text of the document.[55]

II. Separation-of-Powers Pluralism

Historically, the United States has embraced more varied theories of interdepartmental checks. These include an Antidomination Theory in which the judiciary plays only a limited role in expounding the Constitution’s structural meaning; a rights-based theory in which structural questions are answered by asking if a government intervention deprives a person or business of vested property rights; a general-law theory in which constitutional actors reason deductively about the ideal form of government; and various political science theories in which structural questions are treated as a mode of practical, moral, or common law reasoning about optimal government structure. These theories look quite different from the tripartite framework that animates modern debates, but they all were thought to provide a conceptual framework for defining each branch’s powers and deducing its boundaries.

A.      The Antidomination Theory

Of the four theories described in this Part, the Antidomination Theory bears the most superficial resemblance to modern formalism. Like modern formalism, it accepts a tripartite division of powers, understands certain powers to be exclusive, and recognizes that the separation of powers is designed to preserve liberty by checking government overreach.

The Antidomination Theory treats procedural checks as the principal means of preventing interbranch domination. The theory therefore pursues nondomination of the individual by empowering each branch to resist being brought under the others’ control. The distinguishing feature of the Antidomination Theory is that interdepartmental separation is not based on the powers a branch possesses or the substantive acts in which it engages,[56] but rather on the formal procedures that allow each branch to resist the others.

The theory of Antidomination emerged during the founding era as a defense of the new Constitution. In an intellectual history that Bernard Manin, Nicole Mellow, and Jeffrey Tulis have recently provided,[57] dogged opponents of the proposed Federal Constitution, sometimes called the “Anti-Federalists,” thought the document’s separation of powers was theoretically bankrupt. Armed with enlightenment political theory,[58] the Anti-Federalists argued that the Constitution would not succeed in checking tyranny because it did not separate legislative, executive, and judicial powers. Examples of this mixing were the presidential veto, the congressional war power, and the Senate’s role in appointments. All these allocative decisions, according to the Anti-Federalists, vested powers in the wrong branch. Indeed, the checks on government overreach were anathema to the view that the three great heads of power must be kept entirely separate.

The sharing of powers implied by the proposed Constitution’s system of checks represented “a political error of the greatest magnitude,” because the proper “DIVISION OF POWER” required that “the different jurisdictions are inviolably kept distinct and separate.”[59] The Constitution should accept the “natural” difference between the different branches: No branch could have the “least share of each others jurisdiction.”[60] Unless the error was remedied and the distributions of power made more precise, the Anti-Federalists predicted tyranny. “The Anti-Federalists unremittingly advocated precision and certainty in constitutional matters. They complained repeatedly that the [C]onstitution was ‘incomprehensible’ and ‘indefinite,’ ‘vague and inexplicit.’”[61] The separation of powers, on this view, must observe formal niceties. For government to be accountable, it must also be “simple.”[62]

The Anti-Federalists’ protests drew out an animated and prolific defense from the drafters. Crucially, the Federalists’ defense of the Constitution’s separation of powers did not embrace a vesting thesis. Federalists did not quibble with the Anti-Federalists over questions of degree, and they did not argue that, aside from the express checks, the rest of the three powers are otherwise kept within their proper sphere. Instead, they developed a new theory of interdepartmental checks.

Defenders of the Constitution argued that the Anti-Federalists’ taxonomical exercise—that is, requiring every government power to be paired with the proper label so that no power would be mixed—was a fruitless intellectual endeavor, at least in hard cases. Where the boundaries between sovereign powers were at issue, the taxonomy of powers was “less susceptible of precise limits.”[63] Rather than “simplicity” or accountability, the Federalists thought the Constitution should embrace a different mode of separation in pursuit of a different end. They acknowledged that the Constitution’s separation of powers was a “most difficult task,”[64] not because the Constitution should hew to ideal definitions of the three separate and independent branches, but rather because the relevant task was “to provide some practical security for each [branch], against the invasion of the others.”[65] Checks; not balances.

As we have noted elsewhere, some prominent jurists, like Joseph Story, understood the Federalist Papers to endorse the checks-based reading of the Constitution’s separation of powers.[66] The great maxim of separated powers, he wrote, was

not meant to affirm, that [the powers] must be wholly and entirely separate and distinct, and have no common link of connexion or dependence . . . . The true meaning is, that the whole power of one of these departments should not be exercised by the same hands, which possess the whole power of either of the other departments.[67]

Story’s explanation of the Federalists’ theory of separated powers underscores the practical novelty of their theory.[68] Theirs was not a defense of precise separation.

The Constitution does, of course, contain grants of familiar sovereign powers. For example, Congress possesses the power to regulate interstate commerce and impeach federal officials,[69] while the President holds power as commander in chief to command military operations approved by Congress.[70] Except for these express allocative grants of power, the branches are empty vessels. The watchwords of Antidomination are not “vested powers,” but rather a set of worries about self-judging and the excessive subjection of one branch to the authority of the others. Crucially, a branch that remains able to check the other in a dispute over a common governmental issue is not dominated, even if it chooses not to use its check. But except for specific interdepartmental checks, such as impeachment and appointments, the Federalists did not argue that the Constitution offers neatly parsed distinctions between the branches’ powers. Instead, they saw the principal government evil as excessive consolidation—or domination—of one branch by the others. In Story’s words, the separation of powers is violated if one branch exercises the “whole” power of another.[71] They turned to checking functions in response to the concern that definitions and taxonomy would not provide practical restraints on government overreach.

The Antidomination Theory bears a superficial resemblance to modern formalism. It insists on strict divisions between the different checking powers. It does not recognize blended powers in the interstices between branches: There are no quasi-legislative or quasi-judicial functions to worry about; there are only procedural checks. Indeed, Antidomination accepts that there are only three branches of government, that formal distinctions separate the three branches, and that it is unconstitutional to mix or blend different powers. For example, Congress possesses impeachment and appropriations powers; the President possesses the commander-in-chief power; and the judiciary possesses the power to decide cases or controversies arising under the Constitution.[72] It would violate the separation of powers for impeachment to occur outside the process defined in Article I, for the judiciary to appropriate funds for a new courthouse, for the Senate to instruct the Navy to launch missiles at a foreign adversary, or for the President to issue a judgment in a pending judicial proceeding. These arrogations of power are straightforwardly foreclosed by the Constitution since they would deprive a branch of its capacity to check the others.

But the resemblance to modern formalism is largely superficial. Under the Antidomination approach, the terms “legislative” and “executive” do not possess substantive content beyond (1) the specific grants of power enumerated in the Constitution and (2) the procedures the Constitution provides for enacting policy. No constitutional difficulties arise when a branch exercises its enumerated powers—say, the power to legislate—to enable the other branches to act, at least so long as it respects relevant procedural checks.

Thus, the reason Antidomination can accommodate legislative grants of rulemaking authority is that most such delegations preserve the relevant distinctions between each branch’s checking powers. On this account, “delegation” means simply that Congress directed an agency to give effect to the statute by exercising discretion. So long as Congress (1) follows the procedures enumerated in the Constitution, and (2) respects specific allocative grants in the Constitution’s text, no other formal constraint is breached. Conversely, agency rulemaking is not a “quasi-legislative” act. When agencies promulgate rules, they execute a statutory grant of power; in such cases, Congress decided to require the agent to exercise discretion in executing the law, rather than command a particular mode of enforcement. By definition, then, agencies are always engaged in law-execution, not law-making.

Removal restrictions, too, are legislative acts that underwent bicameralism and presentment and which the President must respect in performing the law-execution function. Of course, a statute that, for example, prevented a President from appointing their own cabinet by forbidding removal of the incumbents[73] may well prevent the President from “tak[ing] Care that the Laws be faithfully executed.”[74] But the express restraints imposed by the Take Care Clause provide more room for legislative restrictions on presidential removal than the positions defended by modern formalists.

In both removal and delegation, Congress exercised its lawmaking function through bicameralism and presentment. If the President vetoed the relevant statute—empowering the President, in the case of delegation, or encumbering the President, in the case of removal—the crucial separation-of-powers question is whether Congress overcame the President’s veto by the required supermajority. Whether an unspoken power of the President or Congress was alienated or arrogated, however, are nonsensical on this approach: Compliance with the constitutionally provided check exhausts the relevant separation-of-powers constraint.

There are, of course, powers necessary for governance that are not enumerated in Article I and Article II. To determine whether such latent powers are properly allocated to one or another branch, one follows the enumerated procedure to assess whether a particular government act is consistent with the Constitution’s system of checks. To put it concretely, a principle of interdepartmental Antidomination, elaborated through various enumerated constitutional checks, constitutes the basic model of separated powers. Constitutional checks serve important political theory ends, such as the goals of classical republican freedom and the avoidance of tyranny. On this account, the Constitution’s decision to blend powers is a virtue. It is how the Constitution ensures that each branch is able to check the others.

The Antidomination approach is thus compatible with outcomes that today are associated with functionalism. But that is not because it accommodates blending or mixing of powers. It is instead because adherents of the Antidomination approach deny that any impermissible blending occurs when a government action follows constitutionally prescribed checks.

While the Antidomination account resists the idea that the branches’ content can be definitively revealed, there is nevertheless a place for an “essential” or “immutable” separation of powers. But these essential powers are enumerated in the Constitution’s text. Unless these checks are breached, there is little for courts to police. Almost everything else is left to statute. The constraints imposed by this checks-centric theory are therefore modest.[75] As Nikolas Bowie and Daphna Renan put this last point, “legislation constitutes the separation of powers; it offers a durable, though not immutable, means of state-building.”[76]

B.      The Public Utility Theory: The Mixing of Rights and Structure

Throughout U.S. constitutional history, strong property-rights protections, often associated with Lochner,[77] have limited the reach of state and federal regulations to protect private property rights. The principal constitutional disputes that arose during this period are typically thought to implicate constitutional rights, not constitutional structure. Perhaps for that reason, they have not made their way into modern separation-of-powers debates. Nonetheless, when the United States was developing a powerful federal bureaucracy in the late nineteenth and early twentieth centuries, issues that today implicate separation-of-powers concerns, such as the constitutionality of agency rulemaking and agency adjudication, were answered by asking whether an entity or activity was “affected with the public interest,” and thus amenable to government control, or whether the regulation interfered with private rights, in which case it was largely immune from government interference.[78]

Modern rights analyses usually start by considering whether a particular government intervention interferes with an individual’s constitutional rights—assuming the question whether the act was ultra vires under the Constitution’s structural provisions. The constitutionally relevant question is the right of the individual, not whether the government policy has been enacted or adopted by the proper actor.[79] However, in the early twentieth century—as state and federal governments began to regulate public utilities, such as railroads, electric suppliers, and telecommunications companies—questions of rights and questions of structure traveled in tandem. Treating freedom of contract and substantive due process as doctrines about rights, rather than about structure, ascribes the modern rights-structure binary on a doctrine that treated rights and structure as nearly synonymous. Substantive due process and freedom of contract delimited the powers of two relevant government actors: those who enacted policy (legislatures, the executive, and agencies) and those who protected against government overreach (judges).

Throughout the Lochner Era, an exception to the Court’s solicitude for freedom of contract existed when courts accepted that certain firms or industries were “affected with a public interest” under the nineteenth-century case Munn v. Illinois.[80] Courts found this affected-with-the-public-interest standard satisfied in three situations. The first was industries that had been regulated at common law. These included inns, ferries, wharfs, and other common carrier businesses.[81] What distinguished this first category from the others is that individuals had a justiciable claim against businesses that refused to serve them even when there was no legislative duty to serve.

The second category encompassed industries that received a special privilege from state or federal regulators, such as an exclusive franchise or the right to exercise eminent domain.[82] In such circumstances, courts permitted government bodies to attach conditions to the privilege.[83] The Supreme Court analogized such regulations to contracts between the state and the firm.[84] Firms received something of value (a special privilege) and provided something in return (submission to regulatory oversight).[85] Each party therefore provided something akin to consideration.

A third category of regulable industries was based not on the common law or a contractarian idea of consent, but rather on courts’ view that such firms became “affected with a public interest” by virtue of some legislative decree.[86] There was considerable debate about the limits of this third category. Pro-regulatory justices and progressive reformers treated it as capacious. For them, it was a legislative prerogative to define which firms were so affected. Anti-regulatory justices, by contrast, often treated the first two categories as a closed set and rejected the third category altogether.[87] Both sets of advocates attributed their views to Munn.[88] At times, courts sought a middle ground between these two positions and reviewed legislative determinations to make sure a firm’s size, monopoly or common carrier status, or some other pragmatic judgment justified government control.[89]

For both pro- and anti-regulatory justices, however, the justiciable question was whether the legislature possessed constitutional authority to regulate private industry notwithstanding corporate property interests. Once courts answered that question in the affirmative—once Munn applied—there was little judicial review of the regulatory choices that legislatures made to control that industry, or of legislatures’ creation of administrative bodies to regulate such industries on their behalf.[90] The early twentieth century saw state and federal legislatures turn to this public utility model to regulate a variety of industries, including buses, cotton gins, milk, eggs, theater tickets, and lemons.[91] These regulatory programs did not invoke the common law or preexisting franchises. They instead rested on an interpretation that Munn swept far more commercial enterprises into the category of public utilities than the fixed set of common law categories would suggest.

1.       Agency Adjudication

In this period, the affected-with-the-public-interest standard supported an approach to agency adjudication that bears only a passing resemblance to the modern standard for cabining agency review of individual disputes. Today’s Supreme Court has provided conflicting messages about whether, and under what circumstances, it is constitutional for agencies to adjudicate disputes between individuals.[92] Regardless of what standard determines the constitutionality of agency adjudication, the justices treat adjudication as constitutionally suspect, and they do so because of separation-of-powers concerns.[93] The question, for them, is whether, and under what circumstances, the Constitution allows Congress to grant a non-Article III tribunal authority to decide property and contract disputes among private parties.[94]

By contrast, in the early twentieth century, the constitutionality of agency adjudication was based on the publicness of the firm in question instead of the publicness of the right that was the subject of controversy. When a regulated firm was affected with the public interest, not only did courts allow agencies to adjudicate private disputes; but under something called the “filed rate doctrine,” they also required that the agency weigh in before courts could review the agency’s decision. When parties tried to bring claims in state or federal court, they were routinely barred from doing so. Courts held that agencies, not courts, must adjudicate contract, tort, and antitrust disputes involving private parties when the claim implicated the regulatory apparatus overseen by a particular state or federal agency.

To understand this doctrine, it is important to understand the complex and slightly anachronistic regulatory approach that characterized the public-utility era.[95] As discussed, in the early twentieth century, courts accommodated legislative interventions for the subset of firms and industries that were affected with the public interest. Many industries that were designated as “affected with the public interest” under Munn were common carriers. That is, either because of common law or congressional or state legislation, they were required to serve all potential customers at regulated rates. The Interstate Commerce Act was the first major federal legislation establishing federal common carrier regulation.[96] It also established the first modern federal administrative agency, the Interstate Commerce Commission (ICC), which regulated the terms and conditions upon which railroads served their customers.[97] Other early federal agencies were the Federal Power Commission, which regulated gas and electric companies,[98] and the Federal Communications Commission, which regulated telecommunications companies.[99] Early approaches to agency adjudication, deference, and rulemaking came primarily from cases about how courts should review decisions made by these agencies.

Courts deployed the filed rate doctrine to determine when, and under what circumstances, adjudication must first occur in an agency. Once a rate had been “filed”—that is, it had been reviewed and approved by an agency authorized by statute to engage in ratemaking—ratepayers were prohibited from collaterally attacking the rate in a judicial proceeding.[100] The doctrine continues to apply today in some regulated industries, such as energy and insurance, where the public-utility regulatory paradigm was never entirely displaced.[101]

When courts began expounding the filed rate doctrine in the early twentieth century, they were not only authorizing agencies to decide claims that sounded in tort and contract, but to do so in cases that involved the largest businesses at the time. Utilities and railroads composed nearly half (fifty-two and forty-two, respectively) of the two hundred firms that Berle and Means famously identified as the largest public firms of the era.[102]

Much of the logic underlying the filed rate doctrine operated at the subconstitutional level. Courts offered prudential and statutory reasons for holding that only agencies could resolve disputes that affected a rate on file with a federal regulator.[103] Because common carriers had to provide nondiscriminatory service to their customers, courts worried that separate judicial enforcement of tort and contract claims could disrupt the entire rate scheme by forcing common carriers to give preferential treatment to some counterparties. For example, in an early filed rate case barring railroad customers from bringing Sherman Act claims in court, the Supreme Court pointed out that, “[i]f a shipper could recover under section 7 of the Anti-Trust Act for damages resulting from the exaction of a rate higher than that which would otherwise have prevailed, the amount recovered might, like a rebate, operate to give him a preference over his trade competitors.”[104] The Court was not moved by customers’ arguments that courts were stripping away the rights Congress had provided in the Sherman Act since parties could challenge price-fixing conspiracies among railroads before the ICC: “If the conspiracy here complained of had resulted in rates which the Commission found to be illegal because unreasonably high or discriminatory, the full amount of the damages sustained, whatever their nature, would have been recoverable in such proceedings.”[105]

Courts were not simply concerned that agencies were more competent at requiring nondiscriminatory service. Courts also understood agency adjudication to be a legislative requirement imposed by the common carrier laws. In one of the earliest filed rate cases, the Supreme Court explained that there is “an indissoluble unity between the provision for the establishment and maintenance of rates until corrected in accordance with the statute and the prohibitions against preferences and discrimination.”[106] The Court regarded agency adjudication as a statutory command: “[U]nless the requirement of a uniform standard of rates be complied with[,] it would result that violations of the statute as to preferences and discrimination would inevitably follow.”[107] The Court explained that agency adjudication was necessary to maintain nondiscriminatory rates:

[I]f it be that the standard of rates fixed in the mode provided by the statute could be treated on the complaint of a shipper by a court and jury as unreasonable, without reference to prior action by the Commission . . . it would come to pass that a shipper might obtain relief upon the basis that the established rate was unreasonable, in the opinion of a court and jury, and thus such shipper would receive a preference or discrimination not enjoyed by those against whom the schedule of rates was continued to be enforced.[108]

The concern was that judicial enforcement of contract or tort claims, as well as ad hoc judicial review of the reasonableness of utility rates, would lead to discriminatory service since parties that succeeded on their tort or contract claims would receive more preferable rates than those that did not.

The Supreme Court barred judicial review of contract claims outside the agency process, holding that “a right” to judicial review is “wholly inconsistent” with the ICC’s administrative power and statutory duty to ensure rate equality and uniformity.[109] The Court further emphasized that “the existence of such a power in the courts, independent of prior action by the Commission, would lead to favoritism, to the enforcement of one rate in one jurisdiction and a different one in another,” and “would destroy the prohibitions against preferences and discrimination, and afford, moreover, a ready means by which, through collusive proceedings, the wrongs which the statute was intended to remedy could be successfully inflicted.”[110]

In other words, the filed rate doctrine was ordinary statutory interpretation but with modern separation-of-powers stakes. In cases where “the Commission had not considered . . . a disputed question” about the reasonableness of utility rates, courts must “remand the case to the Commission to enable it to perform that duty.”[111]

During this period, courts were unmoved by the argument that public-utility agencies were exercising ostensibly legislative, judicial, or executive powers. Consider one filed rate case, in which the Fifth Circuit denied a natural gas customer’s challenge to gas rates. In Mississippi Power & Light v. Memphis Natural Gas, the Fifth Circuit invoked structural arguments to support agency adjudication, not pare it back.[112] The Fifth Circuit described “[r]ate-making” as “a legislative function that the courts will not interfere with, at least until the Commission has exercised the function.”[113] The problem with premature judicial review, the court explained, was that it would “transfer the legislative function of rate-making from the Commission to the courts.”[114]

These decisions do not map on to contemporary understandings of the legislative and judicial powers. The Fifth Circuit described the agency’s ratemaking authority as a legislative power, and it felt that the agency was statutorily required to adjudicate suits between a gas company and its customers to give effect to that legislative power. The structural concern was that the court, not the agency, might intrude upon the legislature. Whether an agency was best placed to undertake rate regulation was a legislative judgment, not a judicial one. That was the case even where that regulation ostensibly infringed on regulated entities’ contract and tort rights.

The characterization of ratemaking and rate regulation as legislative functions appears to have been common in suits seeking judicial adjudication of claims,[115] and it was used to justify—not curtail—administrative review of contract and tort suits. Judges had to abstain from intervening in these disputes in order to respect legislatures’ power to delegate their power to agencies.

Courts did, of course, weigh in on the constitutionality of agency adjudication, but they did so as part of the public-private distinction. The constitutionally relevant fact was not whether a statute respected each branch’s inherent powers, but whether the statute observed the distinction between public and private firms. Thus, agencies could not adjudicate an issue if the firms at issue were not properly “affected with the public interest.” Nor could they adjudicate in a manner that deprived investors in regulated firms of their property rights without due process of law.[116] In other words, the constraints agencies faced had to do with (1) whether they had authority to adjudicate in the first place, and (2) whether any particular adjudicative decision made it impossible for a utility to recover its costs and thus excessively interfered with corporate property rights.[117] Once an entity or issue was affected with the public interest, courts respected legislative decisions about how to structure the adjudicatory process.

2.       Agency Deference

At the end of the nineteenth and beginning of the twentieth centuries, legal disputes about the constitutionality of judicial deference to administrative decisions were also fought over the contours of the public-private distinction. Deference, like agency adjudication, is at the center of modern debates about the Constitution’s structure. This period is notable not only because of the structural theory that emerged to justify deferential review of utility commission findings, but also because the deference framework that emerged in this period remains operative today.[118]

Today, agency deference is thought to raise separation-of-powers concerns, and the Court’s principal doctrinal settlements requiring courts to defer to some agency interpretations of law are being revisited.[119] The Supreme Court in Loper Bright recently overturned Chevron deference, which had instructed courts to defer to agency interpretations of vague or ambiguous statutes so long as the interpretation is reasonable.[120] As one of the amicus briefs in Loper Bright argues, “Chevron’s rule of deference violated basic constitutional principles surrounding the separation of powers and diminished, rather than improved, political accountability regarding statutory meaning.”[121] The Court agreed, stating that “Chevron has . . . become an impediment, rather than an aid, to accomplishing the basic judicial task of ‘say[ing] what the law is.’”[122]

During the heyday of public utility regulation, just as today, judicial deference to administrative agencies was constitutionally suspect, but not because courts policed structural constraints on each branch’s inherent powers. Instead, courts exercised a form of utility-specific judicial review to ensure that regulation did not interfere with private property rights. As one academic put it in 1920, the issue was not whether the regulation invaded the law-making power, but rather what standard of review was needed to protect private property interests.[123]

To be sure, in the decades following Munn, as courts tried to establish a standard for reviewing the decisions made by railroad and energy regulators, they initially entertained challenges grounded both in structural principles based on a theory of interbranch relations, and also on a standard based on the distinction between public and private industries.[124] Gerald Henderson summarized these structural arguments, perhaps mockingly, in an influential 1920 article on railroad regulation:

There is much discussion whether rate fixing is in its nature a “judicial” or a “legislative” function. This was a period in the history of American jurisprudence when discussions of this sort were popular. The separation of powers into executive, legislative, and judicial was looked upon as more than a mere differentiation of functions based upon practical considerations; it was thought to be the manifestation of an inherent truth. The pseudo-philosophy of the period regarded certain governmental acts as in their nature judicial, and hence never to be exercised, under the constitution, by either the legislative or the executive branch. The opponents of legislative rate regulation tried to bring rate fixing into this category.[125]

Despite early “pseudo-philosoph[ic]” attempts at taxonomizing departmental powers, it quickly became settled that that the rate-setting power was “a proper legislative function,”[126] and that “the success of [utilities’] attack on the unreviewability of rate regulations was due most fundamentally to . . . the constitutional guaranty of private property.”[127] Courts rejected a separation of powers based on interbranch relations and instead turned to the public-private distinction to deduce the limits of the legislative and judicial powers.[128]

The early deference cases established a searching standard of judicial review, but the parameters of the legislative power were discerned by reference to public, as opposed to private, property rights. The outer bound of public industry regulation was not the exercise of legislative power by non-legislative entities, but instead confiscatory takings of property. By the 1880s, the Court began reviewing rate decisions, explaining that “[t]his power to regulate is not a power to destroy, and limitation is not the equivalent of confiscation.”[129] The limit of the legislative power was reached when a “law amounts to a taking of private property for public use without just compensation.”[130]

In that same case, Stone v. Farmers Loan & Trust Co., the Supreme Court considered multiple constitutional challenges to agency ratemaking that bear superficial resemblance to modern nondelegation and deference challenges. The underlying question was whether a statute granting an agency ratemaking authority “confers both legislative and judicial powers on the commission, and is thus repugnant to the [C]onstitution.”[131] The Court dismissed this argument, which resembles modern nondelegation and Article III challenges, in a few brief sentences, explaining that “the only limit on the power of the commissioners is the constitutional authority of the state over the subject.”[132]

The Court then turned back to the substantive due process arguments to determine how much deference it should accord the agency and discern the limits of the legislature’s power to set rates.[133] Citing Munn, a majority of the Court in Stone upheld the state commissioners’ rate regulation, while Justices Field and Harlan dissented on the ground that the state had impaired the railroads’ private rights of contract.[134]

The case turned on the publicness of the firm, not the pedigree of the power. Justice Field’s dissent underscored that the state’s argument did not rely on state constitutional law, but rather had been based on the Court’s line of decisions, like Munn, that allowed legislatures to regulate public utilities’ rates.[135] Since he disagreed with the majority’s result, he warned parties “interested in railway property” to not seek remedy through courts, but “by efforts to secure wise and intelligent action from the legislature.”[136]

After Stone v. Farmers & Loan Trust Company, the Supreme Court began to seek a justiciable standard for reviewing agency rate orders. First, in the 1890 case Chicago, Milwaukee & Saint Paul Railway v. Minnesota, the Supreme Court held that utilities were constitutionally entitled to judicial review of the “reasonableness” of rates.[137] By 1894, the Court described deference to agency decisions as a threat to property interests, explaining that “it is . . . a part of judicial duty, to restrain anything which, in the form of a regulation of rates, operates to deny to the owners of property invested in the business of transportation that equal protection which is the constitutional right of all owners of other property.”[138] Personnel changes at the Court meant that Justice Field, previously in dissent, began to command a majority. Former members of the Court’s majorities thought the Court had abandoned the principle that “the regulation and settlement of the fares of railroads and other public accommodations is a legislative prerogative, and not a judicial one.”[139]

While conceding legislatures’ ability to regulate, the Court soon began seeking an actuarial method that would allow courts to determine utility rates.[140] By the late 1890s, in Smyth v. Ames, the Court established a “fair value” test, which held that “the basis of all calculations as to the reasonableness of rates . . . must be the fair value of the property being used [by a railroad] for the convenience of the public.”[141] The Court treated ratemaking as a precise science. Agencies had to consider “the original cost of construction, the amount expended in permanent improvements, the amount and market value of its bonds and stock, the present as compared with the original cost of construction, the probable earning capacity of the property under particular rates prescribed by statute, and the sum required to meet operating expenses.”[142] Cases such as Smyth v. Ames and Chicago, Milwaukee & Saint Paul Railway held that investors still had a private property interest in their utility investment.[143]

In other words, by the turn of the century, the Supreme Court permitted regulation but instead scrutinized the specific actuarial methods agencies used to determine utility rates.[144] The Court affirmed the constitutional basis of this accounting principle numerous times.[145] Thus, while the affected-with-the-public-interest designation authorized public supervision, it did not eliminate investors’ property rights. Courts eventually identified a non-deferential standard of review to vindicate these property rights in an Article III forum.[146]

In the first two decades of the twentieth century, however, the Supreme Court slowly developed a looser understanding of corporate property rights, which allowed it to become more comfortable deferring to agency decisions. For example, the 1906 Hepburn Act amended the Interstate Commerce Act to clarify that ICC orders were self-executing[147]: They automatically went into effect thirty days after becoming final unless they were “suspended or set aside by a Court of competent jurisdiction.”[148] The Hepburn Act did not require a particular mode of judicial review, but, as Thomas Merrill has argued, courts took the Hepburn Act as a reason to transition toward an appellate model of review.[149] Many foundational principles of administrative law, such as the doctrine that courts will conduct de novo review of statutory and constitutional questions but defer to factual findings as long as the agency did not act “arbitrarily,” “contrary to law,” unsupported by substantial evidence, or “unreasonabl[y],” came out of judicial review of ICC decisions.[150]

As with adjudication, courts deferred to agencies for both pragmatic and statutory reasons. For example, courts often pointed out that the administrative “inquiry is essentially one of fact and of discretion in technical matters,” “uniformity can be secured only if its determination is left to the Commission,” and Commission decisions are “reached ordinarily upon voluminous and conflicting evidence, for the adequate appreciation of which acquaintance with many intricate facts of transportation is indispensable, and such acquaintance is commonly to be found only in a body of experts.”[151] Regulated firms did not regain their private right to bargain for whatever rate the market would bear, but they did find a judicial forum in which to test the “reasonableness” of the commissioners’ rates.

As rate regulation became more complex and administrative records more voluminous, courts accepted that agencies were better suited to weigh in on technical matters and to consider the enormous amount of evidence submitted in rate decisions. Surprisingly, however, courts did not impose a more searching standard of review for questions of law. In fact, by the 1920s, when upholding an ICC order, the Court acknowledged that “what construction shall be given to a railroad tariff presents ordinarily a question of law which does not differ in character from those presented when the construction of any other document is in dispute.”[152] This mode of prudential reasoning justified judicial deference to agency interpretations of law:

But words are used sometimes in a peculiar meaning. Then extrinsic evidence may be necessary to determine the meaning of words appearing in the document. This is true where technical words or phrases not commonly understood are employed, or extrinsic evidence may be necessary to establish a usage of trade or locality which attaches provisions not expressed in the language of the instrument. Where such a situation arises, and the peculiar meaning of words, or the existence of a usage, is proved by evidence, the function of construction is necessarily preceded by the determination of the matter of fact. Where the controversy over the writing arises in a case which is being tried before a jury, the decision of the question of fact is left to the jury, with instructions from the court as to how the document shall be construed, if the jury finds that the alleged peculiar meaning or usage is established. But where the document to be construed is a tariff of an interstate carrier, and before it can be construed it is necessary to determine upon evidence the peculiar meaning of words or the existence of incidents alleged to be attached by usage to the transaction, the preliminary determination must be made by the Commission; and not until this determination has been made, can a court take jurisdiction of the controversy.[153]

Courts deferred to agencies about some questions of law because they recognized that agencies possessed specific expertise about railroad and energy rates, and because many questions of rate construction required uniformity across a regulated industry.[154]

Thus, most of what would today be fought about using a separation-of-powers vocabulary was resolved by straightforward statutory construction. As late as 1936, in Saint Joseph Stock Yards, which Justice Gorsuch cited in Loper Bright, the Court would write:

When the Legislature itself acts within the broad field of legislative discretion, its determinations are conclusive. When the Legislature appoints an agent to act within that sphere of legislative authority, it may endow the agent with power to make findings of fact which are conclusive, provided the requirements of due process which are specially applicable to such an agency are met, as in according a fair hearing and acting upon evidence and not arbitrarily.[155]

This did not mean judicial review was at an end. On the question of the right to be free from a regulatory taking, courts owed no deference whatsoever to agencies’ (or legislatures’) factual or legal judgments about confiscation.[156] What today is a dispute over the allocation of power was then resolved as a matter of rights.

C.      Separation of Powers as General Law

Today, functionalists and formalists often assume that they are expounding a Constitution whose legal force derives from the people’s sovereign authority, and whose interpretive principles are both parochial to the American state and can prospectively apply to novel constitutional questions. In earlier periods of our constitutional history, however, constitutional interpretation included a form of moral reasoning that transcended the text of the Constitution, our national borders, and our modern ideas about positive law. On this view, the Constitution, like other bodies of law, incorporated an “unwritten law that was taken to be common throughout the nation rather than produced by any particular state.”[157]

Instead of the pluralism we have described among distinctive approaches to interpreting the Constitution’s grants of power, the general law approach looks to older consensus and customary understanding to try to discern answers to modern structural issues. Scholars, particularly those with originalist intuitions, increasingly refer to this intellectual tradition as the “general law.”[158] Because it finds a customary consensus among a plurality of legal discourses, this theory could perhaps also be labelled the “common law,” the “natural” law, and, in our view, the “law of nations.”[159]

A core feature of this general law approach is that multiple sovereign and juridical entities engage in interpretive practices that give content to constitutional rules and mores. This style of public law reasoning was a deductive enterprise that sought to identify the legal outcomes that follow from the exercise of “right reason.”[160] The idea was that certain principles of public law are legally operable even though they “are not under the control of any single jurisdiction”[161]—they “instead reflect principles or practices common to many different jurisdictions.”[162] As one influential historian of this kind of argument explained, this theory could “support the civil law like a military reserve, so that when the latter has failed on some occurrence which altogether demands a decision in a human court of law, there is recourse to the laws of nature and to the analogy resulting from their comparison.”[163] Indeed, this body of law can “take[] on the force of . . . law” when accurately interpreted by courts.[164]

According to James Whitman, this type of legal reasoning was an amalgam of “inconsistent jurisprudential systems, irreconcilable assumptions, and mismatched intellectual inheritances.”[165] After all, how could one speak of a law that is both derived from “fixed principles of reason” and surveyed from a body of practice?[166]

Whitman has speculated about the motivations of thinkers in this tradition that led them to consult, somewhat paradoxically, both “right reason” and “custom” when writing about the common law. They were

proposing to write about a non-existent subject matter, “customary common law.” It was natural to resort to that “formidable non-entity, the Law of Nature,” by drawing upon the amorphous body of “natural law” thought to date back to the Stoics (and beyond). Second, these jurists had . . . a legitimation problem. It was, to the pre-eighteenth century mind, unlawful to make rules without consulting customary witnesses; and lawyers doing unlawful things need grand sources of authority. What these lawyers chose was the vague and magnificent authority of “reason.”[167]

Thus, the peculiar affection of early constitutional lawyers for arguments that married “right reason” with “practice” was both a contradiction in terms and a pervasive habit of mind.

The “mingling” of right reason and custom pervaded both continental and common law sources at the founding, and, as proponents of the general law contend, American lawyers who parsed the Constitution’s references to great powers continued to rely on this mode of reasoning into the nineteenth century.[168]

Some who think about public law in this way have come to tout its anachronism as a virtue: This is a way of looking at the law that is neither positivist, because it does not ground itself in modern ideas about the constitutive or legislative authority behind the law, nor especially formalist, because it does not presume that general forms named in the Constitution will entail that the government be structured in a specific way. Instead, one can only prove the content of the general law by looking to social practices across time.

Modern originalist scholars have rediscovered this “general law” and are increasingly committed to the idea that judges should resurrect it today.[169] Some urge courts to restore this older intellectual enterprise, and invite us to understand that legal questions “falling outside the laws of any one state [ought] to be caught at the bottom by general law.”[170] They argue that by analyzing the practices of other states and sovereigns[171]—or, more coarsely, by paging through old treatises about the law of nations—one can inductively discern a general law for the Constitution. This general law, the argument goes, will provide the correct answers to enduring constitutional puzzles, and it does so because relevant constitutional actors can identify shared practices and customs that bear on constitutional meaning.[172] For this reason, constitutional silences are not gaps to be filled by the explication of political theory or other interpretive methods, but are instead places where the Constitution “secures,”[173] by leaving in place, the public law “principles”[174] that came before it.

The recent “reviv[al]” of general-law scholarship asks whether “general law” should be rehabilitated for modern constitutional rights and federal courts doctrines.[175] For example, Professors William Baude and Stephen E. Sachs recently interpreted the Eleventh Amendment as leaving in place the general law that preceded it, including structural limitations on jurisdiction “properly derived from the common law and the law of nations.”[176] Professor Sachs has called for the overturning of Erie altogether as part of a project that would revise the law of personal jurisdiction and its related “battles between ‘sovereignty’ and ‘liberty.’”[177]

Another place that the “general law” might furnish separation-of-powers principles is mine-run federal courts questions about the limits of Article III. Before Erie, critics of the general law portrayed the general law as “a transcendental body of law outside of any particular state but obligatory within it unless and until changed by statute,” which erroneously invited “[c]ourts of the United States [to] us[e] their independent judgment as to what it was.”[178] These critiques captured genuine concerns about the pervasive, pre-Erie idea that there was an “august corpus” of principles that courts could access in the general law, notwithstanding its possible consequences for more democratic, legislature-centric, theories of legal change.[179] And, indeed, although references to a general common law pervade early constitutional judgments, so too does the allegation that the invocation of general law licenses an arbitrary eclecticism.[180]

The idea that a general-common-law theory can disclose important “principles” that allow courts to distinguish between “public rights” and “private rights” has experienced a revival both at the Supreme Court and in general-law scholarship. The public-private rights distinction, which largely turns on the idea that the general law will disclose the relevant boundaries, has been offered to delineate the limits of standing doctrine, and to suggest limits to agency adjudication.[181]

One way of understanding the general law is as an argumentative style that was used when arguing about constitutional puzzles. Consider, for example, general-law arguments that appeared in debates around the removal of executive branch officials, also discussed in the next section. Congress became alarmed about increasing abuse of the removal power more than three decades after the Decision of 1789, in connection with President Jackson’s removal practice.[182] In one of many flashpoints of this period, Jackson invoked the removal power to dismiss the Secretary of the Treasury in his quest to destroy the national bank. The Senate refused to print the President’s defense of his actions after it censured him, but it did memorialize Jackson’s general-law argument on the floor: A Senate report recounted Jackson’s argument that “all executive power is vested in the President,” citing Vattel, Grotius, Puffendorf, and Blackstone as support.[183] Senator Daniel Webster, one of Jackson’s critics, did not argue that the treatise writers were irrelevant but instead that they furnished no consensus. The “Executive power,” he wrote, was “not a thing so well known and so accurately defined as that the written Constitution of a limited Government can be supposed to have conferred it in the lump.”[184] In a series of rhetorical questions, he asked the general law proponents of the removal power:

What model or example had the framers of the Constitution in their minds when they spoke of “executive power?” Did they mean executive power as known in England, or as known in France, or as known in Russia? Did they take it as defined by Montesquieu, by Burlamaqui, or by De Lolme?[185]

Indeed, even if the framers thought the “elementary writers” would define unenumerated parts of the executive power, “[a]ll these [writers] differ from one another as to the extent of the executive power of Government.” [186] Webster argued that the text and structure of the document proved that the framers did “not intend . . . a sweeping gift of prerogative.”[187] Another critic, Senator Poindexter, argued that Jackson should have deferred to the views of the framers rather than “resort[ing] to the elementary writers.”[188] This general law approach blended—seemingly without a coherent conceptual or jurisprudential theory—different and potentially incompatible styles of constitutional argument: founding era views, political theory, natural right, and pragmatic ideas about good government.

Other applications of the general law involved international law and foreign affairs.[189] One early example concerned whether the federal treaty-making power could preclude states from confiscating foreign debts.[190] The question is a structural one, since a lawfully executed treaty can supersede contrary state law, and possibly federal statutory law. Thus, the constitutional issue was whether a peace treaty could mandate that no debts would ever “be sequestered, or confiscated . . . on account of national differences and discontents,” notwithstanding contrary state law.[191] Hamilton defended such treaty provisions by relying “in part” upon the “theoretical writers”[192] who “proved, as the dictate of sound reason, that private debts and private property in public funds, are not justly liable to confiscation or sequestration.”[193] Hamilton prevailed: These treaties entered into force, where they became an important feature of the United States’ effort to gain recognition in the international system.[194] And, eventually, the Supreme Court affirmed that the treaty-making power could intrude on state confiscations of land, observing that “the common law in these particulars seems to coincide with the Jus Gentium” on the question whether a property transfer in time of war was enforceable in time of peace.[195]

We choose to focus on foreign affairs and Erie for modern reasons: Much current writing about the general law draws from these two areas. Recent debates about whether federal law should incorporate “customary international law,” for example, have been framed as Erie questions.[196]

The two contexts share a deeper connection, however. Erie’s sweeping rejection of the general-law “fallacy” was part of a broader jurisprudential movement that reshaped the mode of making international and constitutional law alike. By the early twentieth century, many international lawyers came to critique the “arbitrary eclecticism”[197] that characterized international law in the late nineteenth century. International lawyers came to espouse “positivism” regarding how international law is made, and by the early twentieth century, it was fashionable to trace the authoritativeness of international law to the voluntary consent of states.[198] In this moment of jurisprudential taste-changing, international lawyers were critiqued for “appear[ing] to pick and choose from natural and positive law exactly as they think fit.”[199] What results, the midcentury critique continued, was a practice of international lawyering consisting of “[m]utual quotation clubs and . . . ‘ipse-dixitism.’”[200]

Notably, midcentury critics of the general law did not abandon the idea of a customary law altogether. Some instead embraced an inductive approach to customary international law: Custom is derived from systematically observing state practice and other states’ acquiescence to that practice—and only those things.[201] They rejected natural law and other deductive forms of reasoning as producing the “uneasy feeling that . . . [one’s] august teacher is producing a beautiful spiral in the air, coming from nowhere and disappearing in the clouds.”[202] These critics felt that the deductive approach suffered from arbitrary eclecticism and argued that international law should become entirely “inductive,” by which they meant that international law should be determined entirely by observing the voluntary behavior and agreements of states. These accounts of international law emphasize the probative nature of states’ own statements in determining the content of the law, and they search for evidence that states believe that their actions are obliged by law before pronouncing that a given rule has become customary international law.[203]

Setting to one side important disputes over which theory of law is most jurisprudentially defensible, the general-law approach is, at bottom, an anti-fixity theory: Many debates over the place of customary international law and general law frequently turn on claims about their practical virtues. Does the custom-oriented way of looking at the law increase the responsiveness of law to political change, or reliably produce entrenchment? Does it prevent judges from intruding into the political process? Does it better attend to basic rights and structural commitments of the relevant community? Is its focus on discerning practice and belief administrable by judges? In our view, none of these questions has an obvious answer. Neither theory has produced a durable consensus about its normative virtues, and neither is likely to produce a determinate and workable account of the separation of powers.

Indeed, the two strands of law—general public law and customary international law—are discovering similar conceptual impasses. For example, both face the challenge of theorizing normative change over time and the objection of arbitrary eclecticism. A source of law that is based on custom can appear to support “traditional”[204] and atavistic legal outcomes, but it can also be harnessed to quickly change “the law” when legislative options are gridlocked.[205] In both cases, the argument goes, the judge can “look[] over the crowd and pick[] [their] friends.”[206] Both customary-international and general-law theories face the problem of identifying the relevant domain of actors whose practice matters and ensuring all relevant practice is considered and correctly interpreted as a matter of history. Additionally, both face the problem of non liquet—how to identify, and what to do, when evidence of practice runs out;[207] and both face the objection that since customary norms do not harden into law until a sufficient group believes they have become law, legal claims about custom can have a circular quality about them.[208]

More fundamentally, both the customary-international and general-constitutional theories of public law face a basic problem in defending their place in the hierarchy of sources of legal meaning. General public law and the law of nations each could reliably claim that there exist “fundamental” principles,[209] but many of those principles are, to borrow the words of recent general-law theorists, “abstract concepts that lack[] legal specificity.”[210] General and customary law may be “established,” to be sure, but often lack workable “boundaries.”[211] As a result, both bodies of law came to be viewed as intolerably “eclectic” when it came to specifics—that is, when it came to litigation.[212] And because both the general law and customary international law rest on judges’ descriptions of practice, judges are afforded remarkably wide discretion: Both radical change and radical retrenchment could each be justified by presenting evidence of social practices and asserting that those practices conform to the law.[213]

In noting these eclecticism and workability concerns, we note an important anachronism of our own: Much of the functional appraisal of the general law’s virtues would not have made much sense to the historical cast of mind to which the general law can be traced. Nineteenth-century lawyers understood and practiced constitutional interpretation in ways that are often impossible to translate to present constitutional disputes. We disinter this older theory, but only to rebury it.

D.     The Social Sciences Theory: Administration as Political Science

Perhaps more than any other constitutional doctrine, the appointment and removal of executive bureaucrats has persistently evaded constitutional settlement.[214] Recently, the Supreme Court has worked to assign an inherent removal power to the President—at least as to “principal officers,” though perhaps further down the bureaucratic chain as well—pursuant to both a formalist account of “the Executive Power” and a stylized history of the legislation of the first Congress.[215] Interestingly, the removal question remains contested not only as a matter of doctrine, but also as a matter of structural theory.

The large scholarly literature on removal typically focuses on a few historical episodes—the Decision of 1789,[216] Myers,[217] Humphrey’s Executor,[218] and Morrison v. Olson.[219] Most of this work focuses on the Decision of 1789—specifically, whether the first Congress decided that the President has an exclusive removal power; if such a decision should be probative of the Constitution’s meaning; and whether the arguments the first senators made are relevant to today’s separation-of-powers puzzle.

Our aim here is not to reprise the removal debate or challenge a particular interpretation of any particular removal episode, but instead to tease out unusual arguments that were levied to support conflicting constitutional theories about removal. What emerges from this history is not only the absence of a formal doctrinal settlement, but, more importantly, consistent appeals to all the modes of constitutional argument we described above: Appeals to antidomination, political theory, and modern formalism were all marshaled to argue that the President does (or does not) possess an inherent removal power. Yet perhaps the most common nineteenth-century argument about the President’s removal power was an appeal to political science: What was the best rule to establish good governance in light of the period’s idiosyncratic political dynamics?

We should acknowledge at the outset that many of these removal episodes are drawn from Congress, not courts. But it appears to us that congressional partisans were not merely engaged in policy debates; they were arguing about constitutional meaning. First, congressional debates have long been used as evidence of the President’s removal power, both in the Decision of 1789 and during Reconstruction.[220] Second, nineteenth-century senators explicitly contextualized, responded to, and sought to authoritatively interpret the Decision of 1789. They questioned whether an earlier Congress could settle or liquidate such an important constitutional issue, and they insisted that they had a role in continuing to expound constitutional meaning, especially once practical experience highlighted mistakes in the founding generation’s political theory. Third, these debates were framed in legalistic terms. Senators self-consciously thought of themselves as developing and interpreting the Constitution.[221]

The primary takeaway from the removal episodes is that, regardless of whether the first Congress thought it had conclusively resolved the removal issue, several congresses in the early- to mid-nineteenth century clearly thought the question remained open.[222] That consensus permitted distinctive separation-of-powers vocabularies to develop, including some that turned on a set of consciously contingent claims about the proper relationship between Congress, the executive, and the electorate. Put differently, these were arguments about social science, political theory, and rights.[223]

Over the course of the nineteenth century, Congress often sought directly to regulate the removal (or “dismission,”[224] or “displacement”[225]) of executive officials to tackle the executive patronage system. “Patronage” was political shorthand for presidents’ tendency to distribute appointments to friends and party members. Much as today, the Decision of 1789 loomed over Congress’s efforts to limit the patronage,[226] but it was not viewed as an obvious precedent that needed to be dislodged. Much as in the last twenty years of legal scholarship, Congress undertook many elaborate accountings of the votes, and changes of votes, about the Decision of 1789, which illustrate the basic indeterminacy of that early congressional debate.

In these debates, senators argued that the founders did not anticipate the party system or its effects on our politics. According to one 1844 Senate committee report, the question of who holds the removal power had lain dormant between 1789 and 1826. “Little inclination existed” to take up the question, since removal “did not press upon the public mind with the force derived from practical abuse or alarming innovation.”[227]

When Congress did act on removal in the early nineteenth century, it did not feel bound by the text and instead felt that political exigencies provided constitutional justification for experimentation. As another Senate report argued, “the Government itself was an experiment, the success of which depended not alone on an adherence to established constitutional forms, but in a great degree also on the purity and fidelity of the agents to whom the administration of its affairs was intrusted.”[228] For our purposes, the specifics of this experiment are less important than the mode of constitutional reasoning it implies. The proper structure of the federal government was understood to be a matter of study that would be worked out over time.

Indeed, much of the early support for the removal power was pragmatic. The first Congress was not conferring a removal power on the presidency, it was conferring the removal power on a particularly worthy President, George Washington: “Much of the support which the power of removal claimed for the President receives, . . . ‘arises from conferring the power on a man whom all the world admires, and who all know will never abuse it.’”[229]

But evolving political dynamics destabilized whatever earlier settlements may have existed during the Washington Administration. By the 1820s, an increase in the frequency of removal caused the voting public to take a “violent exception” to the emerging patronage system.[230] The concern was that competent officers were being removed to make room for favored appointees of each new presidential administration. Legislative proposals during this period failed, but the opposition they faced turned, once again, on pragmatic arguments. Senators felt that proposed limits on removal were bad policy, not that they violated the Constitution.

Notably, some early nineteenth-century senators who challenged the Decision of 1789 appear to have recognized that, textually and structurally, the Constitution conveyed a removal power to the President. Yet they assumed that they could revise the Constitution and challenge the Constitution’s own political theory simply because the founding generation failed to foresee the rise of the party system. One Senate committee noted: “If a community could be imagined in which the laws should execute themselves . . . [p]arties would be unknown, and the movements of the political machine would but little more disturb the passions of men than they are disturbed by the operations of the great laws of the material world. But this is not the case.”[231] In the “theatre of real life,” laws are “executed by civil and military officers, by armies and navies, by courts of justice, by the collection and disbursement of revenue, with all its train of salaries, jobs, and contracts; and in this aspect of the reality we behold the working of PATRONAGE.”[232] The development of a patronage system made the appointment and removal powers too connected to politics, which made it appropriate for legislative action.

Most alarmingly for these Senators, the growth of the executive bureaucracy began to warp elections: “The power of patronage, unless checked by the vigorous interposition of Congress, must go on increasing, until Federal influence, in many parts of this Confederation, will predominate in elections . . . .”[233] Indeed, an unregulated removal power would draw the presidency closer to the monarchy. Because the President “makes and unmakes” the administrative system, “[h]e chooses from the circle of his friends and supporters, and may dismiss them, and . . . will dismiss them, as often as they disappoint his expectations.”[234] In other words, to them the removal power was problematic because it induced people to vote with their parties, and public servants would come to serve the President rather than the public.

The principal legislative proposal to regulate the patronage in 1826 failed, and the issue was not taken up again for nearly a decade. Future Senate committees, however, took the 1826 bills to “indicate[] very clearly that . . . the power of removal had derived no conclusive sanction, either from the lapse of time or from the apparent acquiescence implied by the silence of the Legislative department.”[235] Since the decision was made by Congress, Congress could “disclaim” it.[236]

In 1835, Congress once more considered restrictions on patronage, citing political realities as a justification for revisiting the issue.[237] One bill would have required the President to state the “fact of the removal . . . to the Senate at the same time that the nomination is made, with a statement of the reasons for which such officer may have been removed.”[238] The bill was a “deliberate expression of opinion, that the Senate ought to be associated with the President as well in acts of removal from, as in appointments to, office. It was an act of reclamation of those co-ordinate powers, which, in 1789, that body had consented, by the casting vote of the Vice President, to yield to the President alone.”[239] Once again, Congress assumed that practical experience was reason enough to revisit prior constitutional settlements: If “the exercise of the dismissing power by the President, which was then pronounced to be a dangerous violation of the Constitution, been restored to Congress? All these pledges have been forgotten. Not one has been fulfilled.”[240]

Congressional actors expressed concern that, during the intervening decade, concerns about bureaucratic corruption increased. Senator Daniel Webster argued that “the power of bestowing office and of taking it away again at pleasure, and from the manner in which that power seems now to be systematically exercised, is productive of serious evils.”[241] After each election, “[a] competition ensues, not of patriotic labors . . . but of complaisance, of indiscriminate support of executive measures, of pliant subserviency, and gross adulation.”[242] Elections had become contests of patronage, not ideas.

These concerns about patronage caused policymakers to reconsider prior settlements about removal. Webster asked if frequent removals “cause[d] a union of purpose . . . among all who hold office, quite unknown in the earlier periods of the Government?”[243] The regulation of appointment and removal was necessary because “men in office have begun to think themselves mere agents and servants of the appointing power.”[244]

For Webster, changing political realities alone justified a congressional reevaluation of the removal power. Webster was concerned that the ideal of the able public servant was vanishing. For him, earlier removal settlements were based on an assumption that officials would have “public virtue”: “Whenever personal, individual, or selfish motives influence the conduct of individuals on public questions, they affect the safety of the whole system.”[245] Again, the constitutional question was tied to the importance of public-oriented administrators to stable government.

Webster thus argued that the classical republican system on which the Constitution was based was turned on its head by frequent use of the removal power. The “plain” theory of our executive bureaucracy, Webster argued, is “that Government is an agency, created for the good of the people, and that every person in office is the agent and servant of the people.”[246] Regulating the President’s removal power “will promote true and genuine republicanism, by causing the opinion of the people . . . to be formed and expressed without fear or favor, and with a more entire regard to their true and real merits or demerits.”[247] In the end, however, Webster’s effort to regulate patronage by limiting the removal power also failed.

In 1844, the Senate tried its hand at regulating removal once more. The Committee on Retrenchment argued that, in the half century since the founding, the government had undergone “a practical trial.”[248] Notably, the Committee felt that the President’s removal power had, in Andrew Jackson’s hands, come in tension with Congress’s power of the purse: “[I]f the President has the power of removing all officers who might be virtuous enough to oppose his base measures, what would become of the liberties of the people? Your TREASURY would fall into his hands; for nobody in that department would dare to oppose him.”[249] The first thirty years of removal were unremarkable: “Removals from office occurred during every administration; but, from Washington to the younger Adams, inclusive, they were made, with few exceptions, for cause, not from caprice.”[250] Yet during Jackson’s presidency, the removal practice changed dramatically. Six times more officers were removed during Jackson’s first term than in the prior four decades.[251] The increased use of a removal power had “convert[ed] the legitimate servants of the people into bondmen of the President—a practice which . . . recognises the public offices as the rewards of partisan zeal and devotion.”[252]

The solution was for Congress to limit the President’s discretion to remove members of the civil service. The Committee argued that the Senate’s connection to popular elections put it in a position to resist politically motivated removals: “Suppose [Congress] to be of opinion that an honest and capable subordinate, of independent political opinions, is a fitter instrument of the public service than a time-serving sycophant . . . —may they not prescribe such terms to the tenure of his office as will assure to the nation the benefit of his services?”[253] An appointment and removal system in which the President can “fetter” his appointees with “political orthodoxy or of personal fealty to himself” was “destructive of the independence, reproachful of the patriotism, humiliating to the pride, degrading to the character of an American citizen.”[254] Since Jackson’s inauguration, “the practice so greatly extended [to] removing from office persons well qualified . . . in order to fill their places with those who are recommended on the ground that they belong to the party in power.”[255]

It is only recently, the Committee continued, “that removals from office have been introduced as a system, and for the first time an opportunity has been afforded of testing the tendency of the practice . . . effecting . . . the character of our political system.”[256] Holding a government office had lately been transformed from a “public trust” to the “spoils of victory.”[257] The result was an “inevitable tendency . . . to convert the entire body of those in office into corrupt and supple instruments of power.”[258] When the executive’s influence over the bureaucracy is “so strong as to be capable of sustaining itself by its influence alone, unconnected with any system of measures or policy, it is the certain indication of the near approach of irresponsible and despotic power.”[259] Once again, the strongest defense of the Senate’s role in the removal power followed not from the text of the Constitution or prior constitutional settlements, but rather from the lived political experience of thirty years of quiet followed by thirty years of partisan removal.

Yet the 1844 effort to regulate patronage also failed, and it took another twenty years for reform to take hold. The first time the emergent political science account actually succeeded in Congress arose in the context of foreign affairs.[260] In 1853, the United States negotiated a treaty with France that established a system of diplomatic “pupils” in both countries and provided that these pupils would have the same international status as other consular agents.[261] In the view of the President who signed the treaty, creating the pupilage would improve the consular service “by the appointment, without regard to partisan considerations, of young men of education and character, holding their places by a tenure independent of administrative changes, and with a degree of permanence . . . [to] allow the government to enjoy the advantages of their special training.”[262]

Congress refused to fund these diplomatic pupils for more than a decade. One opponent contended that “consuls were diplomats, and . . . the best diplomacy was the diplomacy of the backwoods . . . not that which was taught in the diplomatic schools of Europe.”[263] Another thought the system would “merely enable the President to cause his friends’ sons to be educated abroad.”[264] And another opposed the pupilage because it would increase patronage.[265]

The President and the State Department renewed their requests for the creation of consular pupils with increasing urgency, noting Civil War staffing shortages and the increasing need to use non-citizens to perform core diplomatic functions abroad. One commentator marveled at the grip of the spoils system in the defeat of the pupilage funding statute: The pupilage, he wrote, would have recreated but a small fraction of the permanent bureaucracy that characterized the first thirty years of our constitutional government. “Fifty years ago[,] the whole administrative service was permanent; yet in 1856 . . . a man . . . could say that the permanence of twenty-five consular clerks was contrary to the spirit of the government . . . .”[266]

In 1863, when yet another president asked for Congress to authorize the pupil system while pointing to the demands of diplomacy during war, Congress finally acquiesced. The State Department wrote to Congress to explain its need for staff “who shall be of some permanent character,” not merely “political personages.”[267] Supporters of the pupil system thought the examination and tenure protections would allow the State Department to recruit “competent” officials “without being politicians and partisans.”[268] Their basic competence would recommend them for promotion, so that “any Administration that may come into power” would recognize their utility.[269] Supporters pointed out that the American consular service had become much worse than those of other countries “for the reason that we make these continual changes, and that we do not educate our men to the business for which we send them abroad.”[270]

The Secretary of State noted that the practice of removing consular officials at the start of each administration created a system that was prone to abuse. He quoted a consular official at one of the busiest foreign outposts suggesting the risk of fraud: “With clerks of my own selection, I would engage to commit defalcations to the extent of at least one half of the receipts of the office . . . . No man ought to be exposed to so great a temptation as this.”[271] Another Senator explained that the education of the clerks and their tenure protection meant that “frauds cannot be committed without their knowledge, and . . . if properly selected, they would form a check” on abuse by consulates abroad.[272]

Apart from the financial benefits, other supporters of tenure-protected consular posts saw in the pupilage system an antidote to the corrupting effects of partisan appointments. Many of the pupils would be drawn “from our higher institutions of learning” and would thus form “a body of public men fitted, eminently fitted, . . . to sustain the national reputation.”[273] By contrast, consuls did not act in the national interest: “It is their misfortune that they have been thrown from their proper sphere by the blind policy of parties.”[274] One consul wrote to the Senate that most consular officials were appointed without regard for their “fitness” and arrived “know[ing] nothing of their commercial ordinances and policy, or of the treaties of his own or their Government” and “cannot communicate with their authorities.”[275]

To Senators who supported the pupilage, removal protections were crucial to renovating a declining foreign affairs bureaucracy. “No sooner does a consul become familiar with the duties of his office . . . than, by a change of Administration, he is liable to removal and a new man sent out.”[276] The Senator praised tenure protection, in particular, and the Senate amended the statute to provide for-cause removal protection.[277] A series of late amendments provided for-cause removal protection.[278] The Act passed with no material debate over its removal provisions.[279]

By 1864, a small experiment in tenure-protected diplomatic officers had begun. The logic of tenure protection for the diplomatic service was, to contemporary readers, applicable to the entire executive bureaucracy. An editorial in The New York Times, for example, argued that the need for competent officials trained by actual practice was “true . . . of every other department of the National Administration.”[280] The editorial flatly declared that “[n]o change in the subordinate offices is called for by any republican principle”[281] of government. “All subordinate appointments in all departments ought to be permanent.”[282]

Nineteenth-century congressional debates about the removal power bear only a passing resemblance to the formalist and functionalist convictions that underlie today’s debates. Over hundreds of pages, Senate committees canvassed well-trodden arguments in today’s removal debates about whether the Decision of 1789 was decisive[283] or whether “when the Constitution was silent, it became a subject of legislative discretion.”[284] Indeed, by the mid-nineteenth century, no fewer than three Senate reports faulted the first Congress for failing to anticipate the rise of the party system and implored the current Congress to heed the lessons of a more modern political science. The Decision of 1789 must now be “reconsidered by Congress whenever experience should demonstrate the necessity of interposing to arrest the abuse of the power by a less scrupulous and more ambitious Executive.”[285]

Elite legal commentary followed a similar arc. Consider the evolving views of two Adamses, one given in 1835 and the other in 1869. In reply to the first ill-fated effort to regulate the patronage, Charles Francis Adams, the son of John Quincy and grandson of John Adams, wrote a rebuttal on behalf of the “old” Whigs to remind them of the constitutional principles they had forgotten. Adams argued that those who held fast to the principles of the founding understood the proper form of executive power: “[T]he boundaries of executive power are plain; that its extent is well known and can be accurately defined; and . . . no State can be well governed where it is not well known.”[286] A generation later, Charles’s son, Henry Brooks Adams, took the opposite view. Henry Adams thought it apparent that, by his father’s time, the rise of parties destabilized those prior settlements: “While the first five Presidents had in fact formed a continuous government, . . . the last five have been the representatives of so many violent revolutions.”[287] Henry Adams, like his contemporaries, thought the realities of party government required a new system to regulate the executive bureaucracy. He recommended both the adoption of a civil service examination system that would “return to the early practice of the government” of a “permanent” executive bureaucracy and a sparing merit-based use of the removal power.

The removal history we have canvassed came well after the Decision of 1789 and just before the infamous impeachment of President Johnson for violating Congress’s restrictions on removing his principal officers. However, during the forty years from 1826 to 1864, Congress attempted and failed many times to limit the President’s removal power. Each attempt left behind a rich store of constitutional deliberations about whether the Constitution formally lodges the removal power in the President or permits it to be shared along with the rest of the appointment power.

In all these deliberations, Congress assumed that it could restrict the President’s ability to remove administrative officials. The Decision of 1789 had failed to anticipate the rise of parties. Party government had changed the office of the President from the “temporary head of a permanent executive system” to the purveyor of rewards for services rendered to one’s party. Because the civil service could not withstand the turnover caused by the effects of party appointment, and because elections had become warped contests for private gain, these reformers thought the Constitution’s political science permitted Congress to regulate removal. Neither form nor function moved them. Whatever structural arguments had been convincing in 1789 were undermined by practical experience, and they were therefore free to revisit what may well have been an earlier constitutional consensus.

*          *          *

Another way of understanding separation-of-powers pluralism, and especially the considerable overlap between political science and functionalist approaches to constitutional meaning-making, is to recognize that different pragmatic theories can be placed in the functionalist (or at least functionalist-adjacent) camp. One could define functionalism, as we do in Part I.B, as a theory that accepts the existence of three coequal branches and in which the judiciary is the final arbiter of constitutional meaning but which, unlike modern formalism, accommodates some amount of blending between the different powers. Alternatively, one can think of functionalism as a form of inductive reasoning in which constitutional interpretation is elaborated over time, often through pragmatic appeals to political science or political theory, and in which certain constitutional actors enjoy a privileged place in elaborating constitutional meaning.

To that end, the functionalist umbrella could encompass different separation-of-powers approaches that appeal to common sense or political science. For example, the “historical gloss” approach associated with the Court’s decision in Youngstown Sheet and Tube Co. v. Sawyer and Professors Curt Bradley and Trevor Morrison is plausibly a version of functionalism.[288] Under this approach, the federal branches tease out or develop constitutional principles through historical practice and interdepartmental acquiescence. Depending on one’s perspective, historical gloss is either a radical departure from familiar functionalist categories, since it embraces the text’s indeterminacy and understands nonjudicial actors to play a central role in expounding constitutional meaning; or it is a close cousin of functionalism, since it accepts a tripartite framework in which government actors give meaning to underdetermined constitutional issues through historical practice.

David Strauss’s theory of common law constitutionalism can also be thought of as either a variant of functionalism or a rejection of the tripartite framework that underlies modern separation-of-powers discourse. Common law constitutionalism, like historical gloss, recognizes that high-stakes constitutional questions are not, and should not be, settled through the text. According to Strauss, constitutional meaning should not reflect “an implicit adherence to the postulate that law must ultimately be connected to some authoritative source: either the Framers, or ‘we the people’ of some crucial era.”[289] Instead, Strauss identifies and defends a “common law tradition” that “rejects the notion that law must be derived from some authoritative source and finds it instead in understandings that evolve over time.”[290] Strauss’s position is arguably consistent with functionalism, since it accepts that judges elaborate principles for understanding interdepartmental relations inductively over time. But it also preserves space for radical constitutional reinvention insofar as neither text nor history fixes constitutional meaning.

Perhaps the most unsparing rejection of the formalist view is associated with Aziz Rana, who has argued that the idolization of constitutional text is a modern invention and urged constitutional interpreters to resist the mythification of the American founding document.[291] Rana shows that many crucial groups did not think that the written Constitution should enjoy a privileged position in managing our political life during moments of genuine rupture and contestation,[292] and he draws attention to the role that U.S. international hegemony and American “peoplehood” has played in the fetishization of the written document.[293] As a normative matter, Rana accepts that constitutional meaning should be contested and contends that excessive myth-making about the Constitution can forestall necessary reform.[294] But as a descriptive matter about separated powers, his work appears Straussian with a healthy dose of progressive political economy. The Constitution’s mythic place and fixed meanings are historically contingent—developed by American legal elites to predictably protect their own contingent interests after moments of serious political rupture.[295]

At some level, these are semantic questions. Bradley, Morrison, Strauss, and Rana all leave room for unorthodox structural approaches that are subject to revision and reinvention, and they are all part of a pragmatic tradition that allows indeterminate textual provisions to be modified by political exigencies. Going back further still, that tradition has been associated with the legal process school and the legal realists.[296] Depending on semantic sleights of hand, these can be understood as consistent with functionalism or as altogether different modes of constitutional meaning-making insofar as they question whether a tripartite framework is a historical imperative.

III. Deconstructing Modern Separation-of-Powers Settlements

The previous Part emphasized that once-prominent theories about how to allocate powers among the different branches look altogether different from the tripartite structural arguments that underlie ongoing debates about deference, agency adjudication, agency rulemaking, and removal protections. While these arguments have faded from constitutional discourse, they tend to reappear in unexpected places, as modern interpreters of the Constitution pick and choose from the past to shore up modern separation-of-powers arguments.

Modern structural disputes engage in two types of picking and choosing. First, courts simply ignore episodes that should be relevant to contemporary structural debates, selecting certain periods as evidence of a constitutional settlement while ignoring others. It is unclear, for example, why the Decision of 1789 should suggest a more enduring settlement on the removal question than the civil service debates that came twenty years later. Nor is it clear why an 1850s case distinguishing between public and private rights should have more constitutional salience than twentieth-century cases holding that agency adjudication was statutorily required and that structural challenges to agency adjudication were an assault on the legislative power.[297] Why is one precedent, made without aid of several decades of experiments with public utility regulation, entitled to more weight than the product of four decades of constitutional jurisprudence?

Second, courts continue to apply precedents that were grounded in long-discarded separation-of-powers theories, and they do so despite having rejected the logic and institutional equilibria that motivated those decisions. In the early twentieth century, the theories that cabined agency deference and agency adjudication were based on a public-private distinction that applied across numerous other spheres of constitutional debate. Tailoring as to the constitutional right in one sphere was balanced by insisting on a presumed legislative power in another. What legal reason justifies applying just one set of these precedents in modern contexts, and doing so after their underlying theoretical justifications and level of institutional equipoise (i.e., between courts and Congress) have been displaced or disturbed?

The implication is that no theory of separation of powers—certainly not formalism or functionalism—can lay claim to a monopoly on the American legal imagination. Our evidence that the theory of separation of powers will forever resist liquidation is inductive: At least with respect to some of the principal constitutional contests of the present, no separation-of-powers theory has avoided the fate of cycling in and out of favor over the long term. The most that can be said about the more enduring separation-of-powers settlements is that they reflect a form of common law reasoning: Interpreters of the Constitution have revised and reinvented this body of law through judicial lawmaking rather than through discovering the right answers to the Constitution’s more puzzling moments of silence. To be sure, courts will fill lacuna in the Constitution to make sure every necessary power is placed somewhere in the government, but it is doubtful, at least on the available evidence, that courts will produce a durable theoretical consensus about the correct separation-of-powers theory.[298]

A.      Anti-Liquidation

Scholars place both formalism and functionalism at the end of history. Formalism is cast as reformation and restoration, and functionalism as a command of progress. But the reality is that the United States has embraced—sometimes across different historical moments and sometimes concurrently—various intellectual frameworks for understanding the limits of each branch’s powers. If neither formalism nor functionalism explains our constitutional history, it is worth asking how it is even possible to speak of constitutional settlements about the separation of powers.

Aziz Huq and Jon Michaels have argued that courts cycle between rules and standards about the separation of powers,[299] and that courts must attend to normative pluralism in structural constitutional law. As we have shown, accounting for the “neglected role” of normative pluralism must include a genuine engagement with more heterodox separation-of-powers theories that have risen and fallen across our constitutional history.[300] In cycling between separation-of-powers theories, or perhaps cycling toward as-yet-undiscovered theories, we have so far produced no evidence of progress or resolution. At least with respect to the issues we have described, our constitutional crises have not been preludes to a new and enduring consensus, as one might expect in typical scientific revolutions. Put another way, scholars’ efforts to engage in a rational reconstruction of this past to “reconstruct [precedents] in a way that makes them comprehensible because they are now shown as parts of a well ordered though complex whole”[301] appear inconsistent with the available evidence.

It is a mistake, on our account, to pick and choose precedents to settle live structural questions without some confidence that there is an intellectual or theoretical throughline between the modern issue and the earlier precedent. This is important not simply for theoretical coherence in the law, but, more practically, because many separation-of-powers settlements emerged from equilibria that are no longer maintained. Superficially, none of the theories described in Part II has much purchase when it comes to resolving contemporary constitutional debates, and almost no purchase if one’s ambition is to preserve clear lines of accountability across time.[302] Consider, for example, twentieth-century cases that understood rate-making as a legislative power that may be broadly delegated to commissions, and one in which commissions, not courts, must adjudicate the lawfulness of the administratively established rate. These cases delineated the limits of legislative and judicial power, and they were all paired with a constitutional theory designed to ensure that regulated firms were clothed with the public interest, and that these public firms’ residual private property rights were not destroyed.

Modern doctrine picks parts of that settlement (those that favor a legislative rate-making power) while ignoring others (those that subject legislatures’ declarations of publicness to judicial review). For example, public utility cases described in Part II.B were crucial precedents in the New Deal cases that accepted the constitutionality of legislative and administrative regulation of private ordering,[303] but they have not been cited in recent Supreme Court decisions that seek to define the legislative or judicial powers.[304] Conversely, when the Supreme Court overturned Chevron in Loper Bright, it relied on deference cases that were about private rights—not the Article III power to declare what the law is.[305] To rely on the holdings of these cases but ignore the underlying logic is not an act of faithful interpretation. It is reinvention.

The enduring disagreement about the President’s removal power provides another example. In the past few years, a large amount of scholarship has interrogated and deconstructed historical arguments that have claimed that the Decision of 1789 conclusively determined that Article II implies a plenary removal power.[306] Many recent critiques of the unitary executive position are, in our view, convincing, in part because it is more difficult, as an evidentiary matter, to show that a particular constitutional debate was settled and liquidated over time than it is to deconstruct or unsettle that position. One implication of Part II.D’s survey of removal episodes is that it is possible to find evidence of essentially any position on the removal question if one looks hard enough. Constitutional actors have offered arguments supporting and opposing a removal power, and they have done so for nearly every conceivable reason. Sometimes these arguments resembled modern formalist and functionalist intuitions about the implied meaning of the executive power.[307] At other times, they followed from arguments about how to structure good government.[308]

It is therefore difficult to claim that there is evidence of theoretical stare decisis, at least as to separation of powers, in the unstable history of this branch of public law.[309] At most, this history shows that United States’ constitutional actors have embraced a variety of theories about the President’s removal power. This doctrinal point is obvious simply from reading the canonical removal cases, but the lack of separation-of-powers settlements is even more apparent when one examines the modes of constitutional argument that were marshalled during these debates. Beyond the doctrinal flip-flopping about whether Article II’s grant of executive power conveys a removal power, the removal debates also embraced heterodox structural theories. During nineteenth-century civil service arguments, politicians and scholars revisited prior settlements, including the Decision of 1789, because they believed new information suggested a need for a different constitutional rule. Assuming that the Decision of 1789 reflected a constitutional settlement in favor of a strong removal power, shouldn’t it be relevant that, just twenty years later, many of those same politicians reconsidered their prior position out of concern that the President would abuse the removal power to strip the civil service of much-needed expertise? Similarly, shouldn’t it matter that partisans on both sides of the removal question, including during the Decision of 1789 and Myers, argued not from text or structure, but from principles about how to structure an effective federal government? Any claim that the Constitution requires a single rule about removal invariably privileges certain of these episodes and elides others.

Agency adjudication is another example of how constitutional debates arbitrarily periodize structural episodes. A large amount of scholarship has focused on the relevance of the public-private distinction in ongoing debates about the constitutionality of agency adjudication.[310] The debate today centers on whether agencies have constitutional authority to adjudicate private rights disputes, and whether Murray’s Lessee, an 1856 Supreme Court case that is often said to have created the public rights doctrine, holds that agencies have constitutional authority to adjudicate public rights but not private rights.[311] Once academics and courts accept that Murray’s Lessee bears on the constitutionality of agency adjudication, disagreement becomes about defining a coherent category of public rights. For some, public rights involve claims against the government. For others, they refer to “civil actions by the government to enforce obligations created by statute.”[312]

Despite its obvious relevance to that debate, recent academic and judicial commentary largely ignores the public rights cases from the public utility period. In our view, this period is relevant because it marked the growth of a federal administrative state and because the public-private distinction demarcated structural limits on legislative, executive, and judicial powers. When it came to public utilities, the constitutionality of agency adjudication was resolved by distinguishing between public and private firms. Adjudication was permitted when a regulated entity was clothed with the public interest. Of course, the clause “clothed with the public interest” does not perfectly mimic the phrase “public right.”[313] But that only highlights the extent to which modern doctrine is based on discretionary decisions about which periods are relevant: The question—whether a question of right should be adjudicated in court or before an administrative tribunal—was identical.[314]

Importantly, the publicness of the entity did not always have to do with the nature of the right or the existence of a privilege. Sometimes, it was about the business’s market power or importance to the local economy.[315] Thus, a public-private distinction was used to resolve constitutional challenges to agency adjudications in circumstances that look quite different from the public-private distinction that was ostensibly embraced in Murray’s Lessee, and it did so throughout the thirty years when the United States was assembling a modern bureaucracy.

Moreover, as discussed in Part II, the public-private distinction in this period was used to resolve all sorts of constitutional challenges to agency interventions.[316] In addition to agency adjudication, the constitutionality of agency deference and agency rulemaking was also resolved by reference to this distinction.[317] To say that a firm was affected with the public interest not only limited its justiciable Article III causes of action, but also sanctioned broad rulemaking delegations and highly deferential approaches to judicial review, and did so because of a peculiar jurisprudence about property rights. This public-private distinction thus resolved interbranch disputes, but not through a modern taxonomy of Article I or Article II powers. Rather, both Article I and Article II powers were unavailable when it came to private property but authorized an entire menu of administrative interventions once a business was affected with the public interest.

It is difficult to understand why courts should revise a nineteenth-century public-private distinction for adjudication but not for rulemaking or agency deference. Or, cabining the debate to adjudication, it is difficult to understand why a case from 1858 has more doctrinal relevance than the approach courts used when agencies first began hearing disputes between private parties.

But it is difficult to develop a theory about how to apply precedents to modern structural debates when those precedents are based on theories that have since been discarded. For example, the constitutionality of deference emerged as a way of mediating firms’ residual private property rights.[318] Today however, the public-private distinction has little to say about the constitutionality of agency deference. Instead, the relevant constitutional question is whether excessive deference prevents courts or legislatures from exercising their constitutionally assigned powers. To answer these structural questions, courts draw on twentieth-century deference cases that were decided on the basis of a public-private distinction. Thus, while the jurisprudence of this era—the advent of hard-look review, the supposed omnipresence of the nondelegation doctrine before the New Deal—is routinely offered today as evidence of a formalist or functionalist past,[319] the usefulness of those precedents is based on holdings that rested on constitutional theories that were explicitly and emphatically discarded in the 1930s.[320]

We do not know how to choose among these different theories—certainly not by appeals to text, structure, or history. Nor do we know why one theory would be relevant to modern structural debates while others remain forgotten. It should be evident, though, that the courts’ treatment of history is selective, and that they have yet to articulate a theory of precedent that explains why some historical episodes are more important than others, or why it is justified to continue to apply precedents whose underlying logic was abandoned long ago.

B.      Common Law Separation of Powers

The separation-of-powers theories surveyed in Part II can be, and often are, reduced to a couple of words. Antidomination is about “anti-tyranny.” The rights-based regime that arose to regulate utilities turned on the word “publicness” or “bigness.” Social science aims to promote “better politics.” Other values commonly attributed to the separation of powers could be added to this list: liberty, fairness, accountability, inclusion, stability, and efficiency. Given the fate of different separation-of-powers theories, perhaps these values do not extrapolate well into serviceable rules for power allocation. Or perhaps the history in Part II is an ordinary exercise in legal realism: Changing politics yield changing constitutional theories.

Still, it is worth noting that the value- and theory-pluralism written into the constitutional history of separated powers has left us with a hash of doctrines that express little consensus for future constitutional interpreters. Yet these theories yield entirely different methods for answering hard constitutional questions. The Antidomination approach, for example, has little use for laborious effort to parse each power vested by the Constitution. The relevant question is whether the federal actor has respected the Constitution’s procedural checks. The public utility approach discerns structural limits by asking whether a policy or business is affected with the public interest. When this public interest standard is met, all federal actors are authorized to enact policies that would otherwise interfere with private property rights. The social science approach subordinates questions of structure to the empirics of political science.

Despite the different normative justifications for different separation-of-powers theories, we lack any conflict-of-law principle for deciding which value across history should be preeminent or authoritative. Until we arrive at such a principle, separation of powers will remain fragmented and unpredictable. Of course, when we pick and choose among structural theories, we are often searching for evidence of liquidation. Yet over the long course of constitutional history, the reality is that when we choose one constitutional precedent, we create a mishmash of different theories, and in doing so, we discard other plausible options. When we do settle on a precedent, the act of analogizing it to the present often ignores the theoretical underpinnings of the doctrines we so routinely draw on in contemporary constitutional debates.

Separation-of-powers decisions often appear to be based on theoretical arguments about the Constitution’s text, structure, and history. But when judges collate cases to support a case’s outcome, they draw on a mélange of precedents whose underlying theories have fallen out of use, and they ignore other structural precedents that bear on the identical constitutional controversy. Consider, for example, the modern Supreme Court’s use of public utility precedents.

That is not to say that the opinions are wrong. In fact, the selective use of separation-of-powers precedents may be inevitable in a system in which constitutional theories evolve and judges continue to draw from precedent. Even so, it is jarring to see judges apply deference cases in which the constitutional question was whether deference prevented judges from protecting investors’ property interests to a case where the deference challenge is based on the structural concern that agency deference gives Article III power to non-Article III tribunals.

The United States’ separation of powers may be best understood as a common law exercise—one with significant variance in its stability over time.[321] David Strauss has described constitutional interpretation as “a species of custom” that “comes . . . from the . . . law’s evolutionary origins and its general acceptability to successive generations.”[322] We may not share Strauss’s normative commitment to this mode of constitutional interpretation, but, as a descriptive matter, he accurately describes our separation of powers.

Conclusion

The concept of constitutional liquidation is usually traced to James Madison, who recognized that the Constitution contained vague and ambiguous provisions but thought those ambiguities would be resolved—or liquidated—through historical practice.[323] As every constitutional law syllabus makes clear, however, United States constitutional law has undergone momentous doctrinal upheavals, and hard constitutional questions continue to be debated and revisited. If our constitutional history proves anything, then, it is that there is little reason to think our current generation marks the end of constitutional history. Judges and academics will continue to change their minds about jurisprudential issues and structural meaning.

As we have shown, however, structural unsettledness goes deeper than doctrine. It is not simply that judges have changed their minds about what the law says. They have also embraced different structural and jurisprudential theories about how to interpret the Constitution. As we have shown, different normative justifications for separation-of-powers theories have not yielded a conflict-of-law principle for deciding which value across history should be preeminent or authoritative. In one sense, this is a deconstructive point. It is impossible to identify a historically correct doctrinal answer to contested structural questions—at least not without appealing to normative principles that allow one to pick among competing theories. In another sense, however, our heterodox structural traditions can reveal new modes of constitutional history. The common thread in the history of separation of powers is that judges, policymakers, and academics will revisit and reinvent constitutional doctrine and constitutional theory to fit the times. We are not tied to the New Deal formalist-functionalist consensus.

Copyright © 2025 Joshua C. Macey* and Brian M. Richardson**

                *     Associate Professor, Yale Law School.

                **        Associate Professor, Cornell Law School. Thanks for helpful comments to Anya Bernstein, Chris Havasy, Michael Dorf, Ketan Ramakrishnan, Shalev Roisman, Jed Shugerman, and Mila Sohoni. For terrific editorial assistance, we are grateful to Eleanor Bock, Max Friend, Hailey Stewart, and Taylor Thomas.

          [1].     Formalists argue for “an ‘exclusive functions’ interpretation of the relationship between functions and branches,” according to which “each of the three branches has exclusive authority to perform its assigned function,” and “each function is uniquely assigned to one branch.” Thomas W. Merrill, The Constitutional Principle of Separation of Powers, 1991 Sup. Ct. Rev. 225, 231–33 (1991). Functionalists, by contrast, “reject the exclusive functions idea, and believe that many governmental activities can be categorized as falling within more than one function.” Id. at 232.

          [2].     See id. at 231–33 (explaining that functionalists accept “the existence of a nucleus of activities that uniquely belongs to each of the three branches”).

          [3].     See infra Part II.B.

          [4].     Our list is not comprehensive, but it is sufficient to illustrate the existence of theories of separation that are quite unlike the modern fixation on formalism and functionalism.

          [5].     In other work, we adopted the shorthand “anti-domination” to distinguish this theory from its rivals and to note its connection to nondomination accounts of individual freedom. See Joshua C. Macey & Brian M. Richardson, Checks, Not Balances, 101 Tex. L. Rev. 89, 96 (2022). In that account, we drew on recent work by political scientists to understand the Anti-Federalists’ separation-of-powers critique of the Constitution. See also Bernard Manin, The Principles of Representative Government 72 (2002); Jeffrey K. Tulis & Nicole Mellow, Legacies of Losing in American Politics 4 (2018).

          [6].     See Manin, supra note 5, at 72; Tulis & Mellow, supra note 5, at 4–6.

          [7].     Howard Gillman, The Constitution Besieged: The Rise and Demise of Lochner [sic] Era Police Powers Jurisprudence 10, 61 (1993); Stephen A. Siegel, Understanding the Lochner Era: Lessons from the Controversy over Railroad and Utility Rate Regulation, 70 Va. L. Rev. 187, 188–92 (1984).

          [8].     See infra Part II.C.

          [9].     See James Q. Whitman, Why Did the Revolutionary Lawyers Confuse Custom and Reason?, 58 U. Chi. L. Rev. 1321, 1326–27 (1991).

        [10].     See id.

        [11].     See infra Part II.D.

        [12].     See infra Part II.D.

        [13].     See infra Part II.D.

        [14].     See infra Part II.D.

        [15].     For a related argument about indeterminacy at the founding, see William B. Gwyn, The Indeterminacy of the Separation of Powers in the Age of the Framers, 30 Wm. & Mary L. Rev. 263, 263–64 (1989) (“From the time when the idea of separation of powers first became popular in North America, much disagreement prevailed as to what institutional arrangements would satisfy the doctrine.”).

        [16].     See infra Part II.D.

        [17].     See infra Part II.D.

        [18].     603 U.S. 369 (2024).

        [19].     Id. at 387 (citing St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 51 (1936)).

        [20].     See 298 U.S. at 51 (“[T]he Constitution fixes limits to the rate-making power by prohibiting the deprivation of property without due process of law or the taking of private property for public use without just compensation. When the Legislature acts directly, its action is subject to judicial scrutiny and determination in order to prevent the transgression of these limits of power. The Legislature cannot preclude that scrutiny or determination by any declaration or legislative finding.”).

        [21].     See infra Part II.B.

        [22].     See SEC v. Jarkesy, 603 U.S. 109, 127–28 (2024); Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 853 (1986); Stern v. Marshall, 564 U.S. 462, 477 (2011); N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 70 (1982).

        [23].     See sources cited supra note 22.

        [24].     See infra Part II.B.

        [25].     See infra Part II.D.

        [26].     See Cong. Globe, 38th Cong., 1st Sess. 1121 (1864).

        [27].     For an analysis of widely reviled cases, see Jamal Greene, The Anticanon, 125 Harv. L. Rev. 379, 386 (2011) (describing a “set of legal materials so wrongly decided that their errors . . . we would not willingly let die”); cf. William Baude, Constitutional Liquidation, 71 Stan. L. Rev. 1, 9 (2019) (arguing that the Constitution’s “indeterminacies could and would be settled by subsequent practice”).

        [28].     For an argument celebrating the Constitution’s commitment to an evolving separation of powers, see Jon D. Michaels, An Enduring, Evolving Separation of Powers, 115 Colum. L. Rev. 515, 517 (2015); Jon D. Michaels, Constitutional Coup 9 (2017) (arguing that developments in twentieth-century administrative law promote separation-of-powers values); Gillian E. Metzger, Foreword: 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1, 7 (2017).

        [29].     See, e.g., Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573, 579 (1984) (“The continued achievement of the intended balance and interaction among the three named actors at the top of government, with each continuing to have effective responsibility for its unique core function, depends on the existence of relationships between each of these actors and each agency within which that function can find voice.”).

        [30].     See Bijal Shah, A Critical Analysis of Separation-of-Powers Functionalism, 85 Ohio St. L.J. 1007, 1010 (2024) (critiquing both formalism and functionalism for the lack of a “moral philosophy” that can be “coherently applied”); Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U. Pa. L. Rev. 1513, 1518 (1991).

        [31].     Peter L. Strauss, Formal and Functional Approaches to Separation of Powers Questions—A Foolish Inconsistency?, 72 Cornell L. Rev. 488, 489 (1987); Strauss, supra note 29, at 609–16 (providing a classic functionalism account of agency administration and noting a return to “formal” reasoning in removal cases that emphasizes “a radical separation of powers within government, with a concomitant need to place agencies in one or another branch, maximally free from intrusion by the others”).

        [32].     See, e.g., Aditya Bamzai & Saikrishna Bangalore Prakash, The Executive Power of Removal, 136 Harv. L. Rev. 1756, 1762 (2023) (“Congress lacks generic power.”); Gary Lawson, Delegation and Original Meaning, 88 Va. L. Rev. 327, 334 (2002) (“[A] statute that leaves to executive (or judicial) discretion matters that are of basic importance to the statutory scheme is not a ‘proper’ executory statute.”).

        [33].     See Lawson, supra note 32, at 355–72.

        [34].     U.S. Const. art. I, § 1.

        [35].     See Gundy v. United States, 588 U.S. 128, 153 (2019) (Gorsuch, J., dissenting) (“When it came to the legislative power, the framers understood it to mean the power to adopt generally applicable rules of conduct governing future actions by private persons—the power to ‘prescrib[e] the rules by which the duties and rights of every citizen are to be regulated . . . .’” (first alteration in original) (quoting The Federalist No. 78, at 465 (Alexander Hamilton) (Clinton Rossiter ed., 1961))); cf. Kristin E. Hickman, Nondelegation as Constitutional Symbolism, 89 Geo. Wash. L. Rev. 1079, 1085 (2021) (arguing that alternatives proposed to replace the nondelegation doctrine “are quite limited in their scope”).

        [36].     See Gundy, 588 U.S. at 149–79.

        [37].     U.S. Const. art. II, § 1.

        [38].     See Seila Law LLC v. Consumer Fin. Prot. Bureau, 591 U.S. 197, 224 (2020) (quoting Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 496–97 (2010) (“The President ‘cannot delegate . . . the active obligation to supervise . . . ’ because Article II ‘makes a single President responsible for the actions of the Executive Branch.’”).

        [39].     See Bamzai & Prakash, supra note 32, at 1763 (“[T]he Article II vesting of ‘executive power’ is a grant of substantive authority and not merely a reference to the powers subsequently mentioned. The Vesting Clause’s grant of power has several components, one of which is the power to remove executive officers.” (footnote omitted)); see also Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559, 565 (2007) (“Article III . . . strongly implies that neither Congress nor entities within the executive branch can exercise ‘[t]he judicial Power of the United States.’” (second alteration in original)).

        [40].     Merrill, supra note 1, at 231–32 (“All functionalists reject the exclusive functions idea . . . .”).

        [41].     See, e.g., Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 853–55 (1986) (identifying a “core” judicial power but stating that “this conclusion does not end our inquiry”); Morrison v. Olson, 487 U.S. 654, 660, 688–89 (1988) (identifying power at issue as “purely executive” but holding that restrictions on President’s ability to exercise that power constitutional because they do not “impermissibly interfere with the President’s authority under Article II”).

        [42].     Strauss, supra note 29, at 579 (“The continued achievement of the intended balance and interaction among the three named actors at the top of government, with each continuing to have effective responsibility for its unique core function, depends on the existence of relationships between each of these actors and each agency within which that function can find voice.”). For alternative perspectives, see Shalev Gad Roisman, Balancing Interests in the Separation of Powers, 91 U. Chi. L. Rev. 1331, 1338 (2024); Ilan Wurman, Nonexclusive Functions and Separation of Powers Law, 107 Minn. L. Rev. 735, 742–43 (2022) (“[E]xclusive functions must be exercised exclusively by the correlative branch of government, according to the constitutional strictures appertaining to that branch. Nonexclusive functions, however, are nonexclusive not only in that they partake in multiple qualities, but also in that such functions can therefore be exercised by different branches exercising different vested powers.”); John F. Manning, Separation of Powers as Ordinary Interpretation, 124 Harv. L. Rev. 1939, 1948–49 (2011) (proposing a “clause-centered approach” that would displace any “freestanding separation of powers doctrine”).

        [43].     See Cass R. Sunstein, The American Nondelegation Doctrine, 86 Geo. Wash. L. Rev. 1181, 1182 (2018) (describing the “American nondelegation doctrine” under which “[e]xecutive agencies cannot make certain kinds of decisions unless Congress has explicitly authorized them to do so”).

        [44].     Morrison, 487 U.S. at 660, 688–89.

        [45].     Schor, 478 U.S. at 853–55.

        [46].     The formalist perspective often leads to doctrinal puzzles as rights come into conflict. See Shalev Gad Roisman, The Limits of Formalism in the Separation of Powers, 16 J. Legal Analysis 178, 184–85 (2024); Adam B. Cox & Emma Kaufman, The Adjudicative State, 132 Yale L.J. 1769, 1772 (2023).

        [47].     Cf. Z. Payvand Ahdout, Separation-of-Powers Avoidance, 132 Yale L.J. 2360, 2365 (2023).

        [48].     See, e.g., Curtis A. Bradley & Eric A. Posner, The Real Political Question Doctrine, 75 Stan. L. Rev. 1031, 1036–47 (2023) (describing judicial and academic commentary on the political questions doctrine); Robert C. Post, The Supreme Court, 2002 Term—Foreword: Fashioning the Legal Con Law, 117 Harv. L. Rev. 4, 4 (2003); Robert C. Post & Reva B. Siegel, Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act, 112 Yale L.J. 1943, 1967–71 (2003) (“Constitutional interpretation always proceeds within specific institutional contexts that inform both the substance of constitutional rights and the procedural framework within which they are enforced.”); Robert C. Post & Reva B. Siegel, Protecting the Constitution from the People: Juricentric Restrictions on Section Five Power, 78 Ind. L.J. 1, 17–30 (2003) (exploring non-juricentric understandings of the branches’ approaches to constitutional interpretation).

        [49].     Humphrey’s Ex’r v. United States, 295 U.S. 602, 624 (1935) (describing the Federal Trade Commission’s “duties” as “neither political nor executive, but predominantly quasi judicial and quasi legislative”).

        [50].     See The Federalist No. 51, at 322 (James Madison) (Clinton Rossiter ed., 1961) (“Ambition must be made to counteract ambition.”). For a discussion and critique of this theory, see Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2312, 2325–29 (2006). For alternative defenses of different allocative possibilities, see Christopher S. Havasy, Radical Administrative Law, 77 Vand. L. Rev. 647 (2024) (describing “radical” nineteenth-century theories about how to democratize the administrative state); Daniel E. Walters, The Administrative Agon: A Democratic Theory for a Conflictual Regulatory State, 132 Yale L.J. 1, 14 (2022) (arguing for “agonistic democratic theory” that “emphasizes the inevitability of conflict and builds democratic legitimacy around it”).

        [51].     For an example of a formalist approach that wields specific legal answers, see generally Bamzai & Prakash, supra note 32. For an example of a functionalist approach, see Strauss, supra note 29, at 578 (“[F]or any consideration of the structure given law-administration below the very apex of the governmental structure, the rigid separation-of-powers compartmentalization of governmental functions should be abandoned in favor of analysis in terms of separation of functions and checks and balances.”).

        [52].     See Paul B. Miller, The New Formalism in Private Law, 66 Am. J. Juris. 175, 178 (2021).

        [53].     In foregrounding a science of politics as a theory of separated powers, the episodes we describe here are reminiscent of Levinson and Pildes’s call to integrate a modern political science of “parties” into our structural constitutional law. See Levinson & Pildes, supra note 50, at 2325–29 (noting that, in contrast to constitutional scholars, political scientists “have long appreciated political parties’ leading role in enforcing the separation of powers”). Notably, in these episodes, constitutional actors were aware they were defending their separation-of-powers theories not with constitutional formalisms, but rather with claims about necessary updates to, for example, the Senate’s consensus view of political science.

        [54].     We borrow this definition from private law. See Miller, supra note 52, at 179 & n.13 (quoting Stephen A. Smith, Contract Theory 5 (2004)). Constitutional legal theory, of course, is beset by far more dispute about the meaning of legal interpretation.

        [55].     One could describe functionalism as a theory based in political science. See Josh Chafetz, Congress’s Constitution 14 (2017). If that is the case, it is hard to understand functionalist appeals to “core” powers.

        [56].     We draw on political theoretical accounts of the anti-domination principle, but claim no monopoly on that term. Nikolas Bowie and Daphna Renan recently elaborated an account of nondomination in connection with the separation of powers, and postulate its connection to a reactionary, anti-reconstruction turn in twentieth-century Supreme Court jurisprudence. Nikolas Bowie & Daphna Renan, The Separation-of-Powers Counterrevolution, 131 Yale L.J. 2020, 2024 (2022). Anti-domination bears many other meanings, including in the context of economic governance. See, e.g., K. Sabeel Rahman, Democracy Against Domination, 16–29, 116–38 (2016). It also retains a distinctive meaning from ours in neo-republican histories of political thought. See, e.g., Philip Pettit, Republicanism: A Theory of Freedom and Government passim (1999) (elaborating the classic recent account of individual freedom as nondomination); Philip Pettit, On the People’s Terms: A Republican Theory and Model of Democracy passim (2012) (same); Quentin Skinner, A Third Concept of Liberty, 117 Proc. Brit. Acad. 237, 249 (2002) (defining freedom as the right “not to be subject to the power of anyone else”).

        [57].     Manin, supra note 5, at 72; Tulis & Mellow, supra note 5, at 55.

        [58].     See Macey & Richardson, supra note 5, at 113–21.

        [59].     William Penn, Essays (1788), reprinted in 3 The Complete Anti-Federalist 172–73 (Herbert J. Storing ed., 1981).

        [60].     Id.

        [61].     Bernard Manin, Checks, Balances and Boundaries: The Separation of Powers in the Constitutional Debate of 1787, in The Invention of the Modern Republic 32 (Biancamaria Fontana ed., 1994).

        [62].     Id.; see also Macey & Richardson, supra note 5, 119–29 (connecting the Anti-Federalist goal of “simple” government to political accountability).

        [63].     The Federalist No. 48, at 310 (James Madison) (Clinton Rossiter ed., 1961).

        [64].     Id. at 308.

        [65].     Id.

        [66].     Macey & Richardson, supra note 5, at 128–29.

        [67].     2 Joseph Story, Commentaries on the Constitution of the United States 8 (Boston, Hilliard, Gray & Co. 1833).

        [68].     See Macey & Richardson, supra note 5, at 94 & n.14 (describing intellectual historians’ work explaining that the Federalists crafted a novel separation-of-powers theory in part to repudiate the strict formal separation advocated by their political adversaries, the so-called “Anti-Federalists”).

        [69].     U.S. Const. art. I, § 8, cl. 3; id. art. II, § 4.

        [70].     Id. art. II, § 2, cl. 1.

        [71].     Story, supra note 67, at 197.

        [72].     Of course, recognizing the existence of substantive grants does not provide much guidance about how to resolve issues of interbranch conflicts. For a thoughtful take on this issue, see generally Roisman, supra note 42.

        [73].     See, e.g., Tenure in Office Act, ch. 154, 14 Stat. 430 (1867) (“[E]very person holding any civil office to which he has been appointed by and with the advice and consent of the Senate . . . shall be entitled to hold such office until a successor shall have been in like manner appointed and duly qualified . . . .”). The Act exempted the “Secretaries of State, of the Treasury, of War, of the Navy, and of the Interior, the Postmaster-General, and the Attorney General,” who hold their offices “during the term of the [appointing] President . . . and for one month thereafter . . . .” Id. The amended Tenure in Office Act, passed two years later, deleted the removal restrictions on the Cabinet, see Act to Amend Tenure in Office Act, ch. 10, 16 Stat. 6 (1869), though the more extreme tenure restrictions would become central to the Taft Court’s telling of this history in Myers. See Myers v. United States, 272 U.S. 52, 164–75 (1926).

        [74].     U.S. Const. art. II, § 3.

        [75].     Cf. Bowie & Renan, supra note 56, at 2029 (“There is no essential or immutable separation of powers.”). We are more faint-hearted. We view some of the Constitution’s enumerated procedural checks—which sometimes, in our history, have implicated rights concerns—as meaningful constraints, even on a nondomination view.

        [76].     Id.

        [77].     In Lochner and Lochner Era cases, the Supreme Court described itself as protecting individual liberty, saying, for example, that “[t]here is no reasonable ground for interfering with the liberty of person or the right of free contract, by determining the hours of labor, in the occupation of a baker.” Lochner v. New York, 198 U.S. 45, 57 (1905).

        [78].     See Joshua C. Macey & Brian Richardson, The Public Law of Public Utilities, 42 Yale J. Regul. 179, 191–95 (2025).

        [79].     There is, of course, considerable overlap. Consider the constitutionality of agency adjudication, which discerns the limits of nonjudicial adjudication by looking at the rights of the individual.

        [80].     94 U.S. 113, 126 (1876).

        [81].     Charles. Wolff Packing Co. v. Ct. of Indus. Rels., 262 U.S. 522, 535 (1923).

        [82].     These regulators included state legislatures, public utility commissions, and early federal agencies such as the Interstate Commerce Commission and Federal Power Commission.

        [83].     On the franchise model of utility regulation and its essential connection to contract and consent, see generally Werner Troekson, Regime Change and Corruption: A History of Public Utility Regulation, in Corruption and Reform: Lessons from America’s Economic History 259 (Edward L. Glaeser & Claudia Goldin eds., 2006). Courts enforced conditions imposed by such franchises. See, e.g., Thomas v. R.R. Co., 101 U.S. 71, 83–84 (1879) (“[W]here a corporation, like a railroad company, has granted to it by charter a franchise intended in large measure to be exercised for the public good, the due performance of those functions being the consideration of the public grant, any contract which disables the corporation from performing those functions which . . . is a violation of the contract with the State. . . . ‘The corporation cannot absolve itself from the performance of its obligations without the consent of the legislature.’” (citation omitted)); New Orleans Gas-light Co. v. La. Light & Heat Producing & Mfg. Co., 115 U.S. 650, 664–65 (1885) (same); Peoria & Rock Island Ry. Co. v. Coal Valley Mining Co., 68 Ill. 489, 493–94 (1873) (“[I]t was necessary to enlist private enterprise and capital; and, to call it forth, it became necessary to confer rights, privileges and immunities, which were secured to those who might carry out the enterprise and operate the roads. . . . These being the inducements which led to the formation of these bodies, their charters granted privileges and imposed duties on them. . . . Whilst railroads must be protected in all of their rights with the same exactness that individuals are, they must at the same time be held to a rigid performance of their duties to the public.”); Shepard v. Milwaukee Gas Light Co., 6 Wis. 539, 546–50 (1858) (same).

        [84].     Charles K. Burdick, The Origin of the Peculiar Duties of Public Service Companies, 11 Colum. L. Rev. 514, 521 (1911).

        [85].     For a discussion of the importance of the property-privilege distinction in utility cases, see Siegel, supra note 7, at 202 (“Conservatives believed that economic theory pointed to the property-privilege distinction as marking the limit of government’s power over rates.”).

        [86].     See Munn v. Illinois, 94 U.S. 113, 126 (1876).

        [87].     See Macey & Richardson, supra note 78, 199–213.

        [88].     See Munn, 94 U.S. at 131–32 (“They stand . . . in the very ‘gateway of commerce,’ and take toll from all who pass. Their business most certainly ‘tends to a common charge, and is become a thing of public interest and use.’ Every bushel of grain for its passage ‘pays a toll, which is a common charge,’ and, therefore, according to Lord Hale, every such warehouseman ‘ought to be under public regulation, viz., that he . . . take but reasonable toll.’ Certainly, if any business can be clothed ‘with a public interest, and cease to be juris privati only,’ this has been. It may not be made so by the operation of the Constitution of Illinois or this statute, but it is by the facts.”).

        [89].     As we explain below, see infra Part II, one finds opposed views of the breadth of Munn in the jurisprudence of the era. Warring interpretations of Munn became shibboleths of pro- and anti-regulatory jurisprudence in the decades after it was decided. To be sure, however, the Munn Court purported to find a common law basis for its exceptional treatment of industries affected with the public interest. The majority opinion pointed to a section of Lord Hale’s De Portibus Maris, which contained examples of industries “affected with a public interest” that were regulable at common law. See 94 U.S. at 126.

        [90].     Cf. Smyth v. Ames, 169 U.S. 466, 471 (1898) (permitting rate review).

        [91].     See, e.g., Atchison, T. & S. F. Ry. Co. v. R.R. Comm’n of Cal., 283 U.S. 380, 394–97 (1931) (considering railroad regulations); Act of June 30, 1913, § 55, 1913 Ill. Laws 459, 488 (regulating public utilities); Public Service Company Law, No. 854, art. III, §§ 2, 3, 1913 Pa. Laws 1374 (defining and regulating public service companies); Act of Mar. 22, 1915, ch. 176, 1915 Okla. Sess. Laws 354 (regulating cotton gins); Tyson & Brother v. Banton, 273 U.S. 418, 420–21 (1927) (considering theater ticket regulations).

        [92].     Cf. Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 851–52 (1986) (holding that the CFTC could adjudicate a counterclaim based in state law); Stern v. Marshall, 564 U.S. 462, 503 (2011) (finding that a bankruptcy court lacked authority to enter final judgment on a state law counterclaim); N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 87 (1982) (plurality opinion) (holding that the Bankruptcy Reform Act violated Article III by granting bankruptcy judges jurisdiction over all civil proceedings related to Title 11).

        [93].     See, e.g., Northern Pipeline, 458 U.S. at 63–70 (noting Framers’ expectation “that Congress would be free to commit [the adjudication of public rights] completely to nonjudicial executive determination”); see generally Gordon G. Young, Public Rights and the Federal Judicial Power: From Murray’s Lessee Through Crowell to Schor, 35 Buff. L. Rev. 765, 788–89 (1986) (discussing the public-private distinction in the context of adjudication).

        [94].     See William Baude, Adjudication Outside Article III, 133 Harv. L. Rev. 1511, 1513–14 (2020) (“[Article III] refers to the substance of judicial power (which is the power to bind parties and to authorize the deprivation of private rights) . . . .”).

        [95].     See Macey & Richardson, supra note 78, at 191–96.

        [96].     49 U.S.C. § 1101(a)–(b); see Thomas W. Merrill, The Interstate Commerce Act, Administered Contracts, and the Illusion of Comprehensive Regulation, 95 Marq. L. Rev. 1141, 1142–45 (2012).

        [97].     See Merrill, supra note 96, at 1142–45.

        [98].     16 U.S.C. § 792.

        [99].     47 U.S.C. § 154.

      [100].     See Am. Tel. & Tel. Co. v. Cent. Off. Tel., Inc., 524 U.S. 214, 221–24 (1998).

      [101].     See Joshua C. Macey, Zombie Energy Laws, 73 Vand. L. Rev. 1077, 1105–20 (2020) (arguing that the filed rate doctrine has outlived its useful purpose); Jim Rossi, Lowering the Filed Tariff Shield: Judicial Enforcement for a Deregulatory Era, 56 Vand. L. Rev. 1591, 1599 (2003) (“In original design, the [filed rate] doctrine was intended to serve as a sword to protect consumers from monopolistic price discrimination.”).

      [102].     Thomas K. McCraw, In Retrospect: Berle and Means, 18 Revs. Am. Hist. 578, 583 (1990).

      [103].     See, e.g., Tex. & Pac. Ry. Co. v. Mugg & Dryden, 202 U.S. 242, 242 (1906) (denying a coal company’s right to pay contractually agreed upon rate to ship coal and instead requiring the company to pay the filed rate).

      [104].     Keogh v. Chi. & N.W. Ry. Co., 260 U.S. 156, 163 (1922).

      [105].     Id. at 162.

      [106].     Tex. & Pac. Ry. Co. v. Abilene Cotton Oil Co., 204 U.S. 426, 440 (1907).

      [107].     Id. at 439–40.

      [108].     Id. at 440–41.

      [109].     Id.

      [110].     Id.

      [111].     Id. at 444 (“[A] conclusion wholly incompatible with the conception that courts, in independent proceedings, were empowered by the act to regulate commerce, equally with the Commission.”); see also Miss. Power & Light Co. v. Memphis Nat. Gas Co., 162 F.2d 388, 389 (5th Cir. 1947).

      [112].     Miss. Power & Light, 162 F.2d at 390.

      [113].     Id.

      [114].     Id.

      [115].     See, e.g., St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 50 (1936) (“The fixing of rates is a legislative act.”); Colo. Interstate Gas Co. v. Fed. Power Comm’n, 324 U.S. 581, 589 (1945) (“Rate-making is essentially a legislative function.”); Terminal R.R. Ass’n of St. Louis v. United States, 266 U.S. 17, 30 (1924) (“The making of rates is a legislative and not a judicial function.”); Sw. Bell Tel. Co. v. City of San Antonio, 2 F. Supp. 611, 615 (W.D. Tex.), aff’d, 4 F. Supp. 570 (W.D. Tex. 1933), set aside, 75 F.2d 880 (5th Cir. 1935) (“[T]he function of rate-making is purely legislative in its character, to whom the power of fixing rates in detail has been delegated.”); L.A. Gas & Elec. Corp. v. R.R. Comm’n of Cal., 289 U.S. 287, 304–05 (1933) (“The legislative discretion implied in the rate-making power necessarily extends to the entire legislative process, embracing the method used in reaching the legislative determination as well as that determination itself. We are not concerned with either, so long as constitutional limitations are not transgressed.”).

      [116].     See, e.g., St. Joseph Stock Yards Co., 298 U.S. at 51 (“But the Constitution fixes limits to the rate-making power by prohibiting the deprivation of property without due process of law or the taking of private property for public use without just compensation.”).

      [117].     The analogy in contemporary constitutional litigation is regulatory takings. The question presented was often whether a rate was confiscatory under the Takings Clause.

      [118].     For discussions of earlier moments in which judges deferred to agency findings, see Henry P. Monaghan, Marbury and the Administrative State, 83 Colum. L. Rev. 1, 17 (1983) (discussing deference during the Marshall Court); Peter L. Strauss, “Deference” Is Too Confusing—Let’s Call Them “Chevron Space” and “Skidmore Weight,” 112 Colum. L. Rev. 1143, 1156 (2012); Peter L. Strauss, In Search of Skidmore, 83 Fordham L. Rev. 789, 789 (2014) (“[S]ince 1827, the U.S. Supreme Court has repeatedly observed that when a court is interpreting a statute that falls within the authority of an administrative agency, the court in reaching its own judgment about the statute’s meaning should give substantial weight to the agency’s view.”).

      [119].     Aditya Bamzai, The Origins of Judicial Deference to Executive Interpretation, 126 Yale L.J. 908 (2017) (arguing that Chevron deference is inconsistent with the APA and prior precedent).

      [120].     Loper Bright Enters. v. Raimondo, 603 U.S. 369, 394–412 (2024); see also City of Arlington v. FCC, 569 U.S. 290, 312 (2013) (Roberts, C.J., dissenting) (“A court should not defer to an agency until the court decides, on its own, that the agency is entitled to deference.”); Michigan v. EPA, 576 U.S. 743, 761 (2015) (Thomas, J., concurring) (“Chevron deference raises serious separation-of-powers questions” because it “precludes judges from exercising [independent] judgment” and “forc[es] them to abandon what they believe is the best reading of an ambiguous statute in favor of an agency’s construction.” (internal quotations omitted)).

      [121].     See, e.g., Brief of Amicus Curiae America First Policy Institute in Support of Petitioners at 4, Loper Bright, 603 U.S. 369 (No. 22-451).

      [122].     Loper Bright, 603 U.S. at 410 (internal citations omitted).

      [123].     Gerard C. Henderson, Railway Valuation and the Courts, 33 Harv. L. Rev. 902, 913 (1920).

      [124].     See, e.g., Smyth v. Ames, 169 U.S. 466, 526 (1898); Chi., Milwaukee & Saint Paul Ry. Co. v. Minnesota, 134 U.S. 418 (1890); Ex parte Young, 209 U.S. 123 (1908). As Stephen Siegel has pointed out, utilities argued both that administrative rate setting was a “confiscation” and that it violated “the allocation of power among the branches of government.” Siegel, supra note 7, at 202, 210.

      [125].     Henderson, supra note 123, at 904.

      [126].     Id.

      [127].     Siegel, supra note 7, at 210.

      [128].     See Henderson, supra note 123, at 906 (“The argument which finally prevailed, rested on the analogy of the law of eminent domain. If the federal government were to take physical possession of a railroad, obviously it would be necessary under the Fifth Amendment to pay just compensation. If instead of taking possession it issued an order compelling the railroad to give the use of its property to the public free of charge, this would virtually be taking the property for public use without just compensation. If it allowed the railroad to receive compensation for its services, there would still be a question for the court whether, the compensation was just. It required, perhaps, a slight wrench to make a doctrine which required the government to pay just compensation, serve the purpose of requiring the government to permit the railroads to collect just compensation from their patrons; but the matter was never very minutely inquired into.”).

      [129].     Stone v. Farmers’ Loan & Tr. Co., 116 U.S. 307, 331 (1886).

      [130].     Id.

      [131].     Id. at 324.

      [132].     Id. at 335.

      [133].     See id.

      [134].     Id. at 339 (Harlan, J., dissenting). Harlan did, however, agree that the state had the power “to establish a railroad commission or to enforce regulations, not inconsistent with the essential charter rights of the companies, in reference to the general conduct of their merely local business.” Id. at 342.

      [135].     Id. at 347 (Field, J., dissenting) (“[T]he argument which supports the statute of Mississippi seems to proceed upon the ground that such is the legitimate outcome of the decisions of this [C]ourt with respect to the control which the legislature may exercise over such corporations . . . .”).

      [136].     Id.

      [137].     134 U.S. 418, 458 (1890); see also Reagan v. Farmers’ Loan & Tr. Co., 154 U.S. 362, 398–99 (1894) (clarifying that Chicago, Milwaukee & Saint Paul held that rate review was constitutionally required). Until at least 1898, there was considerable uncertainty about whether courts would insist on a specific standard of review or defer to reasonable interpretations. For example, in 1892, in Budd v. New York, the Supreme Court held that the rate regulation was sustained on the ground that “the records do not show that the [rates] fixed by the statute are unreasonable, or that property has been taken without due process of law.” 143 U.S. 517, 548 (1892).

      [138].     Reagan, 154 U.S. at 399.

      [139].     Chi., Milwaukee & Saint Paul Ry. Co., 134 U.S. at 461 (Bradley, J., dissenting).

      [140].     See Siegel, supra note 7, at 223 (“[I]n the mid-1890’s the prudent investment and the reproduction cost theories [of determining rates], along with Brewer’s original [cost] analysis, competed for acceptance as the framework to distinguish rate regulation from confiscation.”).

      [141].     169 U.S. 466, 546–47 (1898).

      [142].     Id.

      [143].     See Chi., Milwaukee & Saint Paul Ry. Co., 134 U.S. at 456 (“Under pretense of regulating fares and freights, the state cannot require a railroad corporation to carry persons or property without reward; neither can it do that which in law amounts to a taking of private property for public use without just compensation, or without due process of law.” (citation omitted)); Smyth, 169 U.S. at 547.

      [144].     In Smyth, the Court conceded that additional factors could influence rate decisions but insisted that rate determinations look at original and reproduction costs. 169 U.S. at 547 (“We do not say that there may not be other matters to be regarded in estimating the value of the property. What the company is entitled to ask is a fair return upon the value of that which it employs for the public convenience. On the other hand, what the public is entitled to demand is that no more be exacted from it for the use of a public highway than the services rendered by it are reasonably worth.”). See generally Henderson, supra note 123, at 913 (“To any one familiar with the intricacy and refinement of detail in any business contract involving large sums and complex relations it must be apparent that the general language of the rule in Smyth v. Ames could only be regarded as a preliminary formulation, and that the task of giving it precision of meaning and fullness of detail still lay in the future.”).

      [145].     See San Diego Land & Town Co. v. City of Nat’l City, 174 U.S. 739, 757 (1899). Stanislaus Cnty. v. San Joaquin & King’s River Canal & Irrigation Co., 192 U.S. 201, 215 (1904) (Peckham, J.); San Diego Land & Town Co. v. Jasper, 189 U.S. 439, 442 (1903) (Holmes, J.); see also Willcox v. Consol. Gas Co. of N.Y., 212 U.S. 19, 52 (1909); City of Knoxville v. Knoxville Water Co., 212 U.S. 1, 9, 13 (1909); Cotting v. Godard, 183 U.S. 79, 91 (1901) (Brewer, J.) (“[The Court] has declared that the present value of the property is the basis by which the test of reasonableness is to be determined, although the actual cost is to be considered . . . .”).

      [146].     A three-justice minority of the Fuller Court would have made utility commissioners’ rate-making decisions final and largely unreviewable by courts. See Chi., Milwaukee & Saint Paul Ry. Co., 134 U.S. at 464–65 (Bradley, J., dissenting) (“It is complained that the decisions of the board are final and without appeal. So are the decisions of the courts in matters within their jurisdiction . . . . The important question always is, what is the lawful tribunal for the particular case? In my judgment, in the present case, the proper tribunal was the legislature, or the board of commissioners which it created for the purpose.”).

      [147].     Hepburn Act, ch. 3591, § 4, 34 Stat. 584, 589 (1906) (amending Interstate Commerce Act, ch. 104, § 15, 24 Stat. 379, 384 (1887)).

      [148].     Id. For a description of the legislative history of the Hepburn Act and the role it played in establishing the appellate model of judicial review, see Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum. L. Rev. 939, 953–64 (2011).

      [149].     Merrill, supra note 148. For a discussion of the evolution in judicial deference to agencies in this period, see Blake Emerson, Vindicating Public Rights, 26 U. Pa. J. Const. L. 1424 (2024).

      [150].     Merrill, supra note 148, at 953–64; Ill. Cent. R.R. Co. v. Interstate Com. Comm’n, 206 U.S. 441 (1907); Interstate Com. Comm’n v. Ill. Cent. R.R. Co., 215 U.S. 452, 470 (1910).

      [151].     Great N. Ry. Co. v. Merch.’s Elevator Co., 259 U.S. 285, 291 (1922).

      [152].     Id.

      [153].     Id. at 291–92.

      [154].     See id. at 293–94 (describing Loomis v. Lehigh Valley R.R. Co., 240 U.S. 43 (1916), involving a tariff that was silent about who should bear the costs of ensuring certain safety devices were used, and deferring to the commission’s judgment because the issue called for the uniform exercise of administrative discretion).

      [155].     St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 51 (1936).

      [156].     See id. at 51–53 (describing the duty of the judiciary to make an “independent judgment” on the takings question); see also id. at 53 (“The established principle which guides the court in the exercise of its judgment on the entire case is that the complaining party carries the burden of making a convincing showing and that the court will not interfere with the exercise of the rate-making power unless confiscation is clearly established.”).

      [157].     The definition of the general law is a scholarly riddle in its own right. We take this definition from the latest explication of that theory. William Baude, Jud Campbell & Stephen E. Sachs, General Law and the Fourteenth Amendment, 76 Stan. L. Rev. 1185, 1185 (2024). But there have been a number of modern efforts to define this area of the law, principally regarding the question of incorporation of public international law. See Curtis A. Bradley & Jack L. Goldsmith, Customary International Law as Federal Common Law: A Critique of the Modern Position, 110 Harv. L. Rev. 815, 822–24 (1997); see also Caleb Nelson, The Persistence of General Law, 106 Colum. L. Rev. 503, 518–19 (2006); William A. Fletcher, The General Common Law and Section 34 of the Judiciary Act of 1789: The Example of Marine Insurance, 97 Harv. L. Rev. 1513, 1517–27 (1984).

      [158].     See, e.g., Jud Campbell, General Citizenship Rights, 132 Yale L.J. 611, 616 (2023); cf. Joshua C. Macey, Ketan Ramakrishnan & Brian M. Richardson, Against General Law Constitutionalism, 93 U. Chi. L. Rev. (forthcoming 2026).

      [159].     See Ernest A. Young, Sorting Out the Debate Over Customary International Law, 42 Va. J. Int’l L. 365, 369 (2002) (describing customary international law as “general law”).

      [160].     See Jonathan Gienapp, The Foreign Founding: Rights, Fixity, and the Original Constitution, 97 Tex. L. Rev. Online 115, 120 (2019) (“Carrying forward longstanding constitutional habits, [the founding generation] readily presupposed that constitutions were comprised of a seamless field of written and unwritten content. In some instances, constitutional text was the source of fundamental authority, but in others it was merely a reminder of a more fundamental authority outside of it. In some instances, text was understood to have enacted certain institutions or legal standards, but in others it was assumed that this text easily blended with other kinds of fundamental authority such as natural law, Magna Carta, ‘right reason,’ general law, the law of nations, or just the ‘Laws of the Land.’”).

      [161].     Caleb Nelson, The Persistence of General Law, 106 Colum. L. Rev. 503, 505 (2006).

      [162].     Id.

      [163].     Whitman, supra note 9, at 1350–51 (quoting 2 Samuel Pufendorf, Elementorum Jurisprudentiae Universalis Libri Duo 161 (1931)).

      [164].     Id.

      [165].     Id. at 1324–25.

      [166].     See id. at 1325–29 (describing the falling out of fashion of first “natural” law and then “custom” in twentieth-century histories of the founding and noting the problem of explaining the “mingling” of both modes in American legal thought).

      [167].     Id. at 1348.

      [168].     See id. at 1362–65 (describing the migration of these ideas into Blackstone’s Commentaries); see also Jud Campbell, The Emergence of Neutrality, 131 Yale L.J. 861, 887–88 (2022) (arguing that the distinction between common law and constitutional rights is an Erie anachronism and that “jurists understood the common law as a variant of general law and viewed many (but not all) common-law rules as fundamental in status”); Stephen E. Sachs, Pennoyer Was Right, 95 Tex. L. Rev. 1249, 1261–65 (2017).

      [169].     See Baude, Campbell & Sachs, supra note 157, at 1251–52 (entertaining the possibility that the general law could be “rediscov[ered]” or “translated”).

      [170].     See Sachs, supra note 168, at 1267 (“[F]all[] through the holes of one type of law to be answered by another—and sometimes slipping all the way through, falling outside the laws of any one state to be caught at the bottom by general law.”); William Baude & Stephen E. Sachs, The Misunderstood Eleventh Amendment, 169 U. Pa. L. Rev. 609, 618–19 (2021) (extending this intuition to the Eleventh Amendment). Professor Sachs’s description of the trajectory of constitutional interpretation—that legal puzzles should descend through various possible sources before being decided by custom—is strikingly similar to what Jim Whitman has called the common lawyer’s “confusion” of custom and reason. See Whitman, supra note 9, at 1350–51 (quoting Pufendorf’s similar turn of phrase).

      [171].     See Baude, Campbell & Sachs, supra note 157, at 1248 (“[G]eneral law is shaped by legally recognized custom and practice; its contours can change as those practices change.”).

      [172].     See Whitman, supra note 9, at 1326–27.

      [173].     See Baude, Campbell & Sachs, supra note 157, at 1191 (contending that the general law included the “idea that the Constitution can secure rights without conferring them”).

      [174].     William Baude & Robert Leider, The General-Law Right to Bear Arms, 99 Notre Dame L. Rev. 1467, 1471–72, 1493 (2024).

      [175].     See, e.g., Sachs, supra note 168, at 1249; Caleb Nelson, The Legitimacy of (Some) Federal Common Law, 101 Va. L. Rev. 1, 3 (2015); Campbell, supra note 158, at 611 (describing a set of rights that are “[p]remised on social-contractarian assumptions and a common jurisprudential heritage”); see also Jack Goldsmith, Erie and Contemporary Federal Courts Doctrine, Harv. J.L. & Pub. Pol’y Per Curiam (2023), https://journals.law.harvard.edu/jlpp/erie-and-contemporary-federal-courts-doctrine-jack-goldsmith/ [https://perma.cc/YP43-8LWY] (“Erie is a challenge to originalism and related historically-minded constitutional theories of interpretation because so many constitutional and subconstitutional law doctrines at the founding rested on a conception of general law . . . that the Court rejected in Erie . . . .”).

      [176].     Baude & Sachs, supra note 170, at 613.

      [177].     Sachs, supra note 168, at 1249.

      [178].     Black & White Taxicab & Transfer Co. v. Brown & Yellow Taxicab & Transfer Co., 276 U.S. 518, 533 (1928) (Holmes, J., dissenting).

      [179].     See id. at 533–34 (majority opinion) (rejecting this idea and arguing that “the law of that [s]tate existing by the authority of that [s]tate without regard to what it may have been in England or anywhere else”); see also Digging a Hole: The Legal Theory Podcast, Episode 55: Robert Post (Feb. 20, 2024), https://www.diggingaholepodcast.com/episodes/post [https://perma.cc/BBY7-GJNS] (describing the Holmesian critique that the general law is an anti-democratic idea).

      [180].     See, e.g., Balt. & O.R. Co. v. Baugh, 149 U.S. 368, 397 (1893) (Field, J., dissenting) (“For those courts to disregard the law of the state as thus expressed upon any theory that there is a general law of the country on the subject at variance with it . . . would be nothing less than an attempt to control the state . . . by the opinions of individual federal judges at the time as to what the general law ought to be,—a jurisdiction which they never possessed, and which, in my judgment, should never be conceded to them.”).

      [181].     See, e.g., Spokeo, Inc. v. Robins, 578 U.S. 330, 345 (2016) (Thomas, J., concurring) (canvassing Blackstone’s distinction between public and private rights and contending that “differences between legal claims brought by private plaintiffs for the violation of public and private rights underlie modern standing doctrine and explain the Court’s description of the injury-in-fact requirement.”) (citing Caleb Nelson & Ann Woolhandler, Does History Defeat Standing Doctrine?, 102 Mich. L. Rev. 689, 689–733 (2004)).

      [182].     See infra Part II.D.

      [183].     S. Rep. No. 28-399, at 93 (1844) (“‘[A]ll executive power is vested in the President;’ and then, turning to the pages of Vattel, Grotius, Puffendorf, or Blackstone . . . he . . . appropriates . . . them to himself.”).

        [184].     Id. at 118.

        [185].     Id.

        [186].     Id.

        [187].     Id.

      [188].     Id. at 93.

      [189].     Cf. Phillip Quincy Wright, Enforcement of International Law through Municipal Law in the United States 223–26 (1916) (summarizing, before Erie, the predominant judicial view that “American courts from the earliest time have given voice to the doctrine that international law is law in the United States and must be applied by the courts in appropriate cases. The philosophy basing law on natural rights, so prominent among the founders of the Republic, found expression throughout its constitutional system”); see also id. at 225 n.10 (noting three reasons given by courts for treating international law as part of our law: (1) “International law was part of the common law and was accepted with it”; (2) “International law was impliedly received by the terms of the constitution”; and (3) “International law itself and the privilege of membership in the family of nations, put the courts of the United States under an obligation to apply international law in appropriate cases”).

      [190].     See David M. Golove, Treaty-Making and the Nation: The Historical Foundations of the Nationalist Conception of the Treaty Power, 98 Mich. L. Rev. 1075, 1120 (2000).

      [191].     Treaty of Amity, Commerce, and Navigation, Gr. Brit.-U.S., Nov. 19, 1794, 8 Stat. 116, 122. Daniel Hulsebosch has discussed the treaty and its connection to the United States’ interest in recognition and doux commerce in several works. See, e.g., Daniel Hulsebosch, Magna Carta for the World? The Merchants’ Chapter and Foreign Capital in the Early American Republic, 94 N.C. L. Rev. 1599, 1630–32 (2016) (describing Hamilton’s contracts-clause and law-of-nations arguments).

      [192].     See Alexander Hamilton, From Alexander Hamilton to Defence No. XX [23 and 24 October 1795], in 19 The Papers of Alexander Hamilton, July 1795–December 1795 329–47 (Harold C. Syrett ed., 1973) (canvassing law-of-nations “Writers” to prove there is no right of confiscation); see also Alexander Hamilton, From Alexander Hamilton to The Defence No. XVIII, [6 October 1795], in 19 The Papers of Alexander Hamilton, July 1795–December 1795 299–305 (Harold C. Syrett ed., 1973) (quoting Bynkershoek’s Questiones Juris Publici for the proposition that the right of confiscation “presupposes as the condition of its exercise an actual state of War and the relation of enemy to enemy”).

      [193].     Hamilton, Letter to Defence No. XX [23 and 24 October 1795], supra note 192.

      [194].     See Hulsebosch, supra note 191, at 1632.

      [195].     Fairfax’s Devisee v. Hunter’s Lessee, 11 U.S. 603, 620 (1812); see generally Golove, supra note 190, at 1193–1202 (describing this history, and of related treaty provisions about confiscation that were upheld in Fairfax’s Devisee).

      [196].     The seminal work was Bradley and Goldsmith, supra note 157, at 816. As some defenders and critics of general law acknowledge, the connection between Erie’s reallocation of power between state and federal governments did not require any special theory of positive law and needn’t have swept too broadly. See, e.g., Jack Goldsmith & Steven Walt, Erie and the Irrelevance of Legal Positivism, 84 Va. L. Rev. 673, 675 (1998) (noting that Erie did not require the Court to elaborate a general theory of law); see also Stephen E. Sachs, Finding Law, 107 Calif. L. Rev. 527, 571–72 (2019). Still, in 1938, Erie announced a more sweeping erasure of the general-law method.

      [197].     See Georg Schwarzenberger, The Inductive Approach to International Law, 60 Harv. L. Rev. 539, 544 (1947).

      [198].     See, e.g., The Case of the S.S. “Lotus” (Fr. v. Turk.), Judgment, 1927 P.C.I.J. (ser. A) No. 10, at 18 (Sept. 7) (deriving a customary rule about territorial jurisdiction from a survey of state practice).

      [199].     Schwarzenberger, supra note 197, at 544.

      [200].     Id. at 545.

      [201].     See Anthea Elizabeth Roberts, Traditional and Modern Approaches to Customary International Law: A Reconciliation, 95 Am. J. Int’l L. 757, 758 (2001) (describing this as the “traditional” view of customary international law).

      [202].     Schwarzenberger, supra note 197, at 570.

      [203].     See Anthony A. D’amato, The Concept of Custom International Law 49 (1971).

      [204].     Baude, Campbell & Sachs, supra note 157, at 1234–52 (suggesting that the “general law” basis of the Fourteenth Amendment may be “inherently backward-looking” and that the Court’s more recent focus on rights that are “deeply rooted in this Nation’s history and tradition” in the context of substantive due process better translates the general-law theory).

      [205].     In international law, the general turn from the “traditional” account of custom to the more “modern” approach that focuses on “opinio juris,” is that the “traditional” account is bound to change too slowly to accommodate rapid changes in state behavior. See Anthony D’Amato, The Inductive Approach Revisited, 6 Indian J. Int’l L. 509, 509–10 (1966) (arguing that insistence on the inductive method will exhibit a “psychological bias toward conservatism that may very well be implicit in the insistence upon ‘confirmation’ by ‘inductive methods’”); Roberts, supra note 201, at 759 (arguing that the modern approach allows for the emergence of new custom “rapidly” and noting that “modern custom” can be “embrace[d] as a progressive source of law that can respond to moral issues and global challenges”).

      [206].     Kim Lane Scheppele, “Looking Over the Crowd and Picking Your Friends”: The Social World of Legal Cases, U. Md. Const. Law Schmooze (Feb. 24, 2012), at 7 https://digitalcommons.law.umaryland.edu/schmooze_papers/144/ [https://perma.cc/Z58Z-F9UW]. See generally Adam M. Samaha, Looking Over a Crowd—Do More Interpretive Sources Mean More Discretion?, 92 N.Y.U. L. Rev. 554 (2017) (describing the origins of the critique of customary law and testing the proposition that more interpretive sources increase arbitrariness).

      [207].     See, e.g., Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 I.C.J. 226, at ¶¶ 36–40 (Higgins, J., dissenting) (critiquing the Court’s resort to non liquet because “[t]he corpus of international law is frequently made up of norms that, taken in isolation, appear to pull in different directions . . . . It is the role of the judge to resolve, in context, and on grounds that should be articulated, why the application of one norm rather than another is to be preferred in the particular case”).

      [208].     Compare Roberts, supra note 201, with Sachs, supra note 196 (tackling the same problems).

      [209].     Cf. Baude, Campbell & Sachs, supra note 157, at 1199.

      [210].     Id. at 1196.

      [211].     See id. at 1206.

      [212].     Schwarzenberger, supra note 197, at 549.

      [213].     See Maureen E. Brady, The Illusory Promise of General Property Law, 132 Yale L.J. F. (2023).

      [214].     See Jonathan Gienapp, The Second Creation: Fixing the American Constitution in the Founding Era 125–62 (2018).

      [215].     See, e.g., Seila Law LLC v. Consumer Fin. Prot. Bureau, 591 U.S. 197, 203, 213 (2020) (noting that Article II vests the executive power in the President and citing the first Congress for the proposition that the President’s appointment power “generally includes the ability to remove executive officials”); Collins v. Yellen, 594 U.S. 220 (2021).

      [216].     See Myers v. United States, 272 U.S. 52, 151 (1926) (quoting a Daniel Webster speech that refers to the “decision of 1789”); Saikrishna Prakash, New Light on the Decision of 1789, 91 Cornell L. Rev. 1021, 1026 (2006) (“In passing three acts in 1789 that assumed the President enjoyed a preexisting removal power, majorities in the House and Senate affirmed the executive-power theory on three separate occasions.”).

      [217].     272 U.S. at 151.

      [218].     See Humphrey’s Ex’r v. United States, 295 U.S. 602 (1935).

      [219].     See 487 U.S. 654 (1988). For a small sampling of the recent wave of scholarship, see generally Jed Handelsman Shugerman, Venality: A Strangely Practical History of Unremovable Offices and Limited Executive Power, 100 Notre Dame L. Rev. 213 (2024); Jonathan Gienapp, Removal and the Changing Debate over Executive Power at the Founding, 63 Am. J. Legal Hist. 229 (2023); Jed H. Shugerman, Movement on Removal: An Emerging Consensus about the First Congress and Presidential Power, 63 Am. J. Legal Hist. 258 (2023); Christine Kexel Chabot, Interring the Unitary Executive, 98 Notre Dame L. Rev. 129 (2022); Jed H. Shugerman, The Indecisions of 1789: Inconstant Originalism and Strategic Ambiguity, 171 U. Pa. L. Rev. 753 (2023); Bamzai & Prakash, supra note 32, at 1761 & n.23 (defending the view that “the Constitution grant[s] Presidents the power to remove executive officers at pleasure” and describing Professors Bamzai and Prakash’s several contributions to the removal debate); Andrea Scoseria Katz & Noah A. Rosenblum, Removal Rehashed, 136 Harv. L. Rev. F. 404, 406–15 (2023) (critiquing Bamzai and Prakash’s historical claims); Jane Manners & Lev Menand, The Three Permissions: Presidential Removal and the Statutory Limits of Agency Independence, 121 Colum. L. Rev. 1, 5 (2021); Peter M. Shane, The Originalist Myth of the Unitary Executive, 19 U. Pa. J. Const. L. 323 (2016).

      [220].     See generally Bamzai & Prakash, supra note 32; Prakash, supra note 216.

      [221].     See Macey & Richardson, supra note 5, at 141–53.

      [222].     Our focus here is on the constitutional arguments advanced in Congress, not the Supreme Court. By the late nineteenth century, the question whether the debate of 1789 had settled the removal question had become, as Andrea Katz and Noah Rosenblum have recently explained, “hoary.” Katz & Rosenblum, supra note 219, at 411. In their view, the Supreme Court had repeatedly “refused entreaties to turn questions of statutory construction [about removal] into problems of constitutional law.” Andrea Scoseria Katz & Noah A. Rosenblum, Becoming the Administrator-in-Chief: Myers and the Progressive Presidency, 123 Colum. L. Rev. 2153, 2202 (2023).

      [223].     We have called these separation-of-powers arguments “political science” theories because they reason through the separation-of-powers doctrine by asking which kinds of governance devices would most improve electoral politics, the functioning of state technocracies, and which would mitigate agency-cost and capture problems experienced by legislators. While we think the label is apt, we note that it is unusual to find evidence of a political science approach in this era of constitutional law. The period is often cast as having “dissolved into a number of tactical political positions with little coherence or consistency.” M.J.C. Vile, Constitutionalism and the Separation of Powers 289 (2d ed. 1998). And, indeed, this period is seen as the prelude to an account of the rise of political-science approaches in the Progressive era. See, e.g., id. at 289–322 (describing the rise of political science in separation-of-powers theory as following “a demand for reform that built up through the Granger, Greenback, and Populist movements to its climax in Progressivism” and describing the emergence of administrative governance and a functionalist theory of separated powers as accepting the “multifunctionality of political structures”). The preoccupation of the actors in our story with politics is no less true of the Progressive Era, and focusing on what they took themselves to be doing suggests that they thought the separation of powers should be mobilized to pursue better government.

      [224].     Statement of Henry Clay, reprinted in S. Rep. No. 28-399, at 142 (1844) (“The basis of this overshadowing superstructure of executive power is the power of dismission . . . . Of what avail is it that the Senate shall have passed upon a nomination, if the President, at any time thereafter . . . may dismiss the incumbent?”).

      [225].     Id. at 13 (“The question then resolved itself into this: Is the power of displacing an executive power?”); see also id. at 143 (reprinting Henry Clay’s statement: “The power of removal, as now exercised, is nowhere in the Constitution expressly recognised. The only mode of displacing a public officer for which it does provide, is by impeachment”).

                 The question whether “displacement” and “removal” are synonyms has generated controversy in recent removal disputes, since Federalist 77 (often a guide to originalist modes of constitutional interpretation), suggests that Congress shares both the appointment and removal powers. Compare Seila Law LLC v. Consumer Fin. Prot. Bureau, 591 U.S. 197, 270 (2020) (Kagan, J., concurring in part) (“Hamilton presumed that . . . ‘the consent of the Senate would be necessary to displace as well as to appoint’ officers of the United States” (cleaned up)), with Josh Blackman, Justice Kagan on Hamilton in Federalist No. 77, Reason: Volokh Conspiracy (July 1, 2020), https://reason.com/volokh/2020/07/01/justice-kagan-on-hamilton-in-federalist-no-77/ [https://perma.cc/TRG4-3V7A] (arguing that Justice Kagan misreads Federalist 77). Kagan likely has the better argument. Joseph Story’s Commentaries interpreted Federalist 77’s reference to “dismissal” to refer plainly to removal. After quoting verbatim from the work, Story added the sentence: “No man can fail to perceive the entire safety of the power of removal, if it must thus be exercised in conjunction with the senate.” 3 Story, supra note 67, at 392. Notably, Chancellor Kent, in a letter, attributed Story’s interpretation to the original text of Federalist 77, an error which was repeated both in nineteenth-century histories of removal and by Chief Justice Taft’s law clerk in connection with Taft’s decision in Myers. See 10 Robert C. Post, The Taft Court: Making Law for a Divided Nation, 1921–1930, at 426 n.58 (Maeva Marcus ed., 2023) (Kent’s rendition of Federalist 77, which misquotes Story, is sent to Taft); Lucy M. Salmon, History of the Appointing Power of the President, in 1 Papers of the American Historical Association 299, 312–13 (New York, G.P. Putnam’s Sons 1886); see also Robert Post, Tension in the Unitary Executive: How Taft Constructed the Epochal Opinion of Myers v. United States, 45 J. Sup. Ct. Hist. 167, 173 (2020) (describing Taft’s correspondence with his law clerk).

      [226].     See generally Shugerman, supra note 219, at 258; Shugerman, supra note 219; Bamzai & Prakash, supra note 32; Prakash, supra note 216. But see Katz & Rosenblum, supra note 219.

      [227].     S. Rep. No. 28-399, at 2 (1844). In fact, the now-quaint but then-notorious removal controversy of this period—the removal of William Gardner and Joseph Whipple from their custom-office posts in Portsmouth—came at the request of their local congressmen, not executive fiat. See Joseph Whipple, To Thomas Jefferson from Joseph Whipple, 9 April 1801, in 33 The Papers of Thomas Jefferson 559 (Barbara B. Oberg ed., 2006), https://founders.archives.gov/documents/Jefferson/01-33-02-0487 [https://perma.cc/KHC5-A965]; John Adams, Review of James Hillhouse, Propositions, 12 April 1808, Nat’l Archives: Founders Online, https://founders.archives.gov/documents/Adams/99-02-02-5237 [https://perma.cc/Z3ZK-FART] (“The Power of Removal was never abused in the first twelve years, [except in the two Cases of Gardner and Whipple at Portsmouth]; except perhaps in 2 Instances and those Removals were made at the earnest and repeated Solicitations of all the Members of the House and one of the Members of the Senate . . . .” (text in brackets marked as deleted text in online version of the document)). After some controversy over these early removals, the two were restored to office by Thomas Jefferson.

      [228].     S. Rep. No. 28-399, at 6.

      [229].     Id.

      [230].     Id. at 2.

      [231].     Id. at 321.

      [232].     Id.

      [233].     Id. at 323.

      [234].     Id. at 324.

      [235].     Id. at 4.

      [236].     Underscoring the fragility of the precedent of 1789, several Senate reports emphasized how the vote was exceedingly close and required the Vice President’s vote to break the deadlock. See id. at 5. (“[T]he reports of two committees, now sustained by a decisive vote of the Senate itself, disclaimed the authoritativeness of the precedent of 1789, and the practice of the Government in virtue of it since that period, and by the same decisive vote the Senate distinctly announced the opinion that the question had not been put to rest.”).

      [237].     See Edward Stiglitz, The Reasoning State 2–9 (2022) (defending reason-giving requirements as a means to restore trust between legislatures and the electorate).

      [238].     S. Rep. No. 28-399, at 5.

      [239].     Id.

      [240].     Id. at 99.

      [241].     Id. at 113.

      [242].     Id. at 114.

      [243].     Id. at 116.

      [244].     Id.

      [245].     Id. at 113. On the founding conceit that officeholders would possess civic republican virtue, see Nicholas R. Parrillo, Against The Profit Motive: The Salary Revolution in American Government, 1780–1940, at 9 (2013).

      [246].     S. Rep. No. 28-399, at 116.

      [247].     Id. at 117.

      [248].     Id. at 18.

      [249].     Id. at 15.

      [250].     Id. at 27.

      [251].     Id. at 27–28.

      [252].     Id. at 28.

      [253].     Id. at 24.

      [254].     Id. at 31. As a concrete example, the report noted that a “vast quantity of land to which the Indian title has . . . been extinguished” has led to a “a great increase of the number and influence of surveyors, receivers, registers, and others employed in the branch of the administration connected with the public lands; all of which have greatly increased the influence of executive patronage over an extensive region.” Id. at 57.

      [255].     Id. at 58.

      [256].     Id.

      [257].     Id. at 58–59.

      [258].     Id. at 59.

      [259].     Id. at 65.

      [260].     On the exceptional character of foreign-affairs concerns in separation-of-powers disputes, see generally Curtis Bradley & Jack Goldsmith, Foreign Affairs, Nondelegation, and the Major Questions Doctrine, 172 U. Pa. L. Rev. 1743 (2024).

      [261].     See Consular Convention Between the United States of America and his Majesty the Emperor of the French, Fr.-U.S., art. II, Feb. 23, 1853, 10 Stat. 992.

      [262].     U.S. Dep’t of State, United States Consular Regulations: A Practical Guide for Consular Officers 172–73 (3rd ed. Washington, French & Richardson 1868).

      [263].     Henry Brooks Adams, Civil-Service Reform, 109 N. Am. Rev. 443, 462 (1869).

      [264].     Id. at 463.

      [265].     Id. A statute authorizing twenty-five pupils to be appointed after satisfactory evidence by examination was passed in 1856 but never funded. See id. at 462–72.

      [266].     Id. at 465.

      [267].     See Cong. Globe, 38th Cong., 1st Sess. 1131 (1864).

      [268].     Id. at 1132.

      [269].     Id.

      [270].     Id.

      [271].     Id. at 1133.

      [272].     Id. at 1134 (statement of Sen. Fessenden). The threat of self-dealing was paramount during this period, and some consular officials were removed for committing large frauds against the treasury. See Government Officials, Home and Foreign, N.Y. Times (Apr. 4, 1864), https://timesmachine.nytimes.com/timesmachine/1864/04/04/78720943.html?pageNumber=4 [https://perma.cc/M4X8-4UMF].

      [273].     Cong. Globe, 38th Cong., 1st Sess. 2229 (1864).

      [274].     Id. at 2231.

      [275].     Id.

      [276].     Id. at 2232.

      [277].     Id. at 2789.

      [278].     Id.

      [279].     Act of June 20, 1864, ch. 136, 13 Stat. 137 (1864).

      [280].     Government Officials, Home and Foreign, supra note 272.

      [281].     See id.

      [282].     Id. The editorial also proposed civil-service exam or apprenticeship requirements, pensions for civil service, and internal promotion. Id. To other contemporary observers, tenure protection for such a small group of clerks did not go nearly far enough. “[T]he original purpose of creating this corps of officers failed because the lack of permanency of tenure in the higher offices of the service made consular clerks unwilling to accept promotion.” Wilbur J. Carr, The American Consular Service, 1 Am. J. Int’l L. 891, 902 (1907). It would be decades before such a system would be enacted and even longer before the diplomatic service was consolidated and subject to foreign-service examination requirements. See Katherine Shaw, Partisanship Creep, 118 Nw. U. L. Rev. 1563, 1573–77 (2024) (describing the history of the Pendleton Act).

      [283].     S. Rep. No. 28-399, at 6–17 (1844).

      [284].     Id. at 12.

      [285].     Id. at 6.

      [286].     Charles Francis Adams, An Appeal from the New to the Old Whigs 5 (Boston, Russel, Odiorne & Company 1835).

      [287].     Adams, supra note 263, at 451; see also S.W. Kellogg, The Beginnings of Civil Service Reform, 8 Yale L.J. 134, 134 (1898) (“For the first forty years of our history the evils of partisan appointments in the civil service were wholly unknown.”).

      [288].     See Curtis A. Bradley, Historical Gloss and Foreign Affairs 3 (2024); Curtis A. Bradley & Trevor W. Morrison, Historical Gloss and the Separation of Powers, 126 Harv. L. Rev. 411, 432 (2012) (“According to many accounts of how historical practice relates to the separation of powers, a practice by one branch of government that implicates the prerogatives of another branch gains constitutional legitimacy only if the other branch can be deemed to have ‘acquiesced’ in the practice over time.”); see also Michael J. Glennon, The Use of Custom in Resolving Separation of Powers Disputes, 64 B.U. L. Rev. 109, 134 (1984) (explaining that the “custom in question must consist of acts” and that “if a coordinate branch has performed the act, the other branch must have been on notice of its occurrence” and “acquiesced in the custom”).

      [289].     David A. Strauss, Common Law Constitutional Interpretation, 63 U. Chi. L. Rev. 877, 879 (1996).

      [290].     Id.

      [291].     See Aziz Rana, The Constitutional Bind: How Americans Came to Idolize a Document That Fails Them 5–9 (2024).

      [292].     See, e.g., id. at 58–77, 83–89 (describing non-creedal movements that arose in the wake of the failure of reconstruction and a pervasive turn-of-the century uncertainty about whether the Constitution could meet the many challenges of post-reconstruction national politics).

      [293].     Id. at 10–11, 19–21, 31–32.

      [294].     Id. at 19–20; 104–09.

      [295].     See, e.g., id. at 96–101 (describing the late-nineteenth-century populist attack on judicial review, which rose as the judiciary acquired new salience in American politics); see also id. at 104–09 (describing the movement of the constitutional critique into establishment circles).

      [296].     See H.S. Thayer, Meaning And Action: A Critical History of Pragmatism 6 (1968); Felix S. Cohen, Transcendental Nonsense and the Functional Approach, 35 Colum. L. Rev. 809, 812 (1935); Louis L. Jaffe, Law Making by Private Groups, 51 Harv. L. Rev. 201, 201–02, 206–07 (1937).

      [297].     Cf. Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. (18 How.) 272, 281 (1856).

      [298].     David A. Strauss, Does the Constitution Mean What It Says?, 129 Harv. L. Rev. 1, 18–21 (2015) (arguing that the Constitution’s text and structure are no more “fundamental” sources of constitutional meaning than concern about precedents, fairness, and good policy).

      [299].     Aziz Z. Huq & Jon D. Michaels, The Cycles of Separation-of-Powers Jurisprudence, 126 Yale L.J. 346, 354 (2016).

      [300].     Id. at 354, 380–90.

      [301].     See id. at 354 (quoting Neil MacCormick, Reconstruction After Deconstruction: A Response to CLS, 10 Oxford J. Legal Stud. 539, 556 (1990)).

      [302].     Cf. Martin S. Flaherty, The Most Dangerous Branch, 105 Yale. L.J. 1725, 1740–41 (1996) (describing “accountability and energy” as “root values” driving some modern separation-of-powers theories); Gundy v. United States, 588 U.S. 128, 155, 159–69 (2019) (Gorsuch, J., dissenting) (arguing the nondelegation doctrine follows from the separation-of-powers ideal that “[t]he sovereign people would know, without ambiguity, whom to hold accountable for the laws they would have to follow”).

      [303].     For example, in Nebbia v. New York and West Coast Hotel v. Parrish, two cases that are thought to have overturned Lochner, the Supreme Court cited Munn and the Charles Wolff Packing case for the proposition that legislatures could regulate private ordering. Nebbia v. New York, 291 U.S. 502, 532 (1934); W. Coast Hotel Co. v. Parrish, 300 U.S. 379, 392 (1937) (citing Munn, among other cases, for the proposition that the “power under the Constitution to restrict freedom of contract has had many illustrations”); Townsend v. Yeomans, 301 U.S. 441, 450 (1937) (“So far as the present controversy turns upon the power of the state to give this sort of protection to this industry, provided its regulation is not arbitrary or confiscatory and in the absence of conflict with the federal power over commerce, our rulings are decisive in support of the state action.” (citing Munn v. Illinois, 94 U.S. 113 (1876))).

      [304].     A Westlaw search of Charles Wolff Packing, for example, reveals that the Supreme Court has not cited the case since the 1950s. Westlaw, http://www.westlaw.com [https://perma.cc/TDX5-9WQM] (type “262 U.S. 522” in the search bar; click on the case; hover over the “Citing References” dropdown and click “Cases”; click “Filters”; click “Jurisdiction” to expand; click “Federal” to expand; click the box next to “Supreme Court”; click “Apply”; click “Date” to sort from most recent). In the twenty-first century, the Court has cited Munn, but not to define legislative or administrative powers. It has instead used the case to describe (1) public utility regulation, (2) regulatory takings, and, curiously, (3) gay marriage. See Verizon Commc’ns Inc. v. FCC, 535 U.S. 467, 477 (2002) (public utility regulation); FCC v. Fla. Power Corp., 480 U.S. 245, 253 (1987) (regulatory takings); Obergefell v. Hodges, 576 U.S. 644, 725–26 (2015) (Thomas, J., dissenting) (gay marriage).

      [305].     See Loper Bright Enters. v. Raimondo, 603 U.S. 369, 387 (2024) (citing St. Joseph Stock Yards Co. v. United States, 298 U.S. 38, 51 (1936)); id. at 435 (Gorsuch, J., concurring) (citing Kisor v. Wilkie, 588 U.S. 558, 607 (2019) (Gorsuch, J., concurring in judgment)).

      [306].     See sources cited supra note 219.

      [307].     See supra Part II.D.

      [308].     See id.

      [309].     Indeed, some scholars have seen the absence of stare decisis as inevitable and perhaps desirable. See, e.g., Huq & Michaels, supra note 299, at 431 (“Stability in the form of stare decisis would do little to promote the full spectrum of relevant normative values given continuing developments, both good and bad, within the thick political surround.”). But see Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 Colum. L. Rev. 723, 739–48 (1988) (endeavoring to rescue constitutional stare decisis from its critics).

      [310].     See, e.g., Baude, supra note 94, at 1522-23; Caleb Nelson, Vested Rights, “Franchises,” and the Separation of Powers, 169 U. Pa. L. Rev. 1429, 1432 (2021); Gregory Ablavsky, Getting Public Rights Wrong: The Lost History of the Private Land Claims, 74 Stan. L. Rev. 277, 277 (2022).

      [311].     Much of the current debate tries to parse the meaning of the Court’s statement that “there are matters, involving public rights, which may be presented in such form that the judicial power is capable of acting on them, and which are susceptible of judicial determination, but which [C]ongress may or may not bring within the cognizance of the courts of the United States, as it may deem proper.” Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. (18 How.) 272, 284 (1856). For discussions of the case and its relevance to agency adjudication, see James E. Pfander & Andrew G. Borrasso, Public Rights and Article III: Judicial Oversight of Agency Action, 82 Ohio St. L.J. 493, 498 (2021) (discussing the historical context of Murray’s Lessee to “offer a new account of the public rights doctrine in Murray’s Lessee and a new synthesis of the public rights cases” based on a typology of how statutory delegations to agencies interact with the judicial power).

      [312].     See Michael C. Dorf, The Persistence of Public Rights Doctrine (Jarkesy Oral Argument Edition), Dorf on Law (Nov. 30, 2023), https://www.dorfonlaw.org/2023/11/the-persistence-of-public-rights.html [https://perma.cc/3F9D-7VAW].

      [313].     See supra Part II.B.

      [314].     Even in the public utility period, some matters had to go before courts. But the special place for Article III adjudication, or at least adjudication before a judicial tribunal, appears to have been reserved for questions of constitutional right. See Smyth v. Ames, 169 U.S. 466, 526 (1898) (“While rates for the transportation of persons and property within the limits of a state are primarily for its determination, the question whether they are so unreasonably low as to deprive the carrier of its property without such compensation as the constitution secures, and therefore without due process of law, cannot be so conclusively determined by the legislature of the state, or by regulations adopted under its authority, that the matter may not become the subject of judicial inquiry.”). Sometimes, publicness existed because the regulated entity was given a privilege, as occurred when it was given power to exercise eminent domain or a certificate of public convenience and necessity. But the grant of a government privilege was not dispositive. At other times, however, courts found that the entity had become affected with the public interest because it possessed market power or was important to the economy.

      [315].     See Munn v. Illinois, 94 U.S. 113, 121 (1876).

      [316].     See supra Part II.B.

      [317].     See supra Part II.B.

      [318].     See Oceanic Steam Navigation Co. v. Stranahan, 214 U.S. 320, 339–40 (1909).

      [319].     See Gundy v. United States, 588 U.S. 128, 167–68 (2019) (Gorsuch, J., dissenting).

      [320].     See Nebbia v. New York, 291 U.S. 502, 512 (1934) (allowing legislatures to determine the publicity of industries).

      [321].     Development of separation-of-powers doctrine over time could be cyclical and perhaps progressive, see, e.g., Huq & Michaels, supra note 299, at 416–33 (describing a recurring cycle between rules and standards), or it need not be. See Jeremy K. Kessler & David E. Pozen, Working Themselves Impure: A Life Cycle Theory of Legal Theories, 83 U. Chi. L. Rev. 1819, 1880 (2016) (contending that most interpretive theories inevitably “work themselves impure” during a process of inevitable compromise over constitutional time).

      [322].     David A. Strauss, The Living Constitution 37–38 (2010).

      [323].     Baude, supra note 27, at 9.