Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

ALL THE SOVEREIGN'S AGENTS: THE CONSTITUTIONAL CREDENTIALS OF ADMINISTRATION

At a time when citizens must rely on administration to tackle the urgent threats of global warming, a pandemic, and destabilizing inequality, critics both left and right would tear it down. Libertarian opponents denounce agency discretion as obnoxious to the rule of law and its norms of generality and neutrality. They argue that regulatory agencies violate the separation of powers and defy the principle of limited government. But their favored solution, a strict application of the non-delegation doctrine, would create a new, undemocratic separation of powers problem: ascribing to unelected judges responsibility for economic and social regulation. On the other hand, conservatives with a more authoritarian inflection would rescue agencies from their constitutional exile by placing them under more direct executive control. Bootstrapping agencies to the president’s electoral credentials, their solution carries Schmittian implications. Administration fares little better in the hands of the left. They blame it for a variety of ailments: the reification of social differences and the juridification of human nature; regulatory capture; depoliticization of the economy and the subsidization of financial capitalism; and, ultimately, populist politics. This paper will put administration on firm constitutional footing. It argues that critiques of the administrative state rely upon mistaken notions of popular sovereignty and the rule of law. First, they posit the law as the “will” of an organ-body democratic sovereign that is transcribed (however ham-handedly) by elected officials. Yet, as recent theory shows, there is no exogenous democratic will to transcribe. Second, they posit that this law can be uncontroversially applied by neutral judges and administrators. For them, the rule of law protects political equality through the unwavering application of abstract, general legislation. Critical theorists show, however, that the application of abstract law to concrete cases may only exacerbate inequality. Legal realists demonstrate that law is underdeterminate; unelected judges and administrators will inevitably find themselves smuggling outside values into their decision-making. By treating administration as part of representative government, however, agencies can gain constitutional credentials. If democracy aspires not to universal consent and a formalist understanding of law, but instead to democratic autonomy (Urbinati & Warren, 2008), then administration can be legitimate. Indeed, agencies already serve as fora for representative politics as citizens organize around rulemaking. They also act as representatives themselves. By providing points of entry into policymaking, independent agencies help constitutional democracies ensure that power can respond effectively to popular demands while always remaining an empty place.

ALL THE SOVEREIGN’S AGENTS: THE CONSTITUTIONAL CREDENTIALS OF ADMINISTRATION Kate Jackson* Prerogative being nothing, but a Power in the hands of the Prince to provide for the publick good, in such Cases, which depending upon unforeseen and uncertain Occurrences, certain and unalterable Laws could not safely direct, whatsoever shall be done manifestly for the good of the People, and the establishing the Government upon its true Foundations, is, and always well be just Prerogative. Locke, Two Treatises of Government, Ch. XIII ¶ 158 TABLE OF CONTENTS INTRODUCTION ...............................................................................................2 PART I: ADMINISTRATION, POPULAR SOVEREIGNTY AND RIGHTS...................8 A. The Mind and Body of the Democratic Sovereign ..........................10 B. Individual Rights and Rule of Law ..................................................15 C. The Non-Problem of Separating Powers .........................................24 PART II. AGENCIES AND THE REPRESENTATIVE TURN...................................30 A. Defining the Turn .............................................................................31 B. Accommodating Administration ......................................................36 C. 1. 2. 3. Implications ......................................................................................42 Non-Delegation as a Problem of Equality of Representation ......42 Technocracy, Non-Delegation and Representation......................44 Representation and the Chevron Shuffle .....................................48 CONCLUSION .................................................................................................50 * J.D., Ph.D., DeOlazarra Postdoctoral Fellow, University of Virginia Department of Politics, Program in Political Philosophy, Policy & Law. I thank the participants in the Groupe de Recherche Interuniversitaire en Philosophie Politique and ClassCrits for their useful commentary and direction. 2 JACKSON INTRODUCTION We face no less than four urgent crises: an ongoing pandemic1; racial injustice and its consequent civil unrest2; an economic depression approaching the pain inflicted in 19293; and the accumulating, existential threat of climate change.4 Citizens must rely on their state to tackle these burning perils.5 Yet critics both left6 and right7 would tear down its institutional capacity to do so. Some denounce the exercise of administrative power as illiberal, unconstitutional and obnoxious to the rule of law.8 Others impugn it as undemocratic, paternalistic, and corrupt.9 Yet without some kind of agent to carry out collective solutions, these perils may very well proceed unabated. 1 E.g., Exec. Order No. 14002, 86 Fed. Reg. 7229 (January 22, 2021); Exec. Order No. 13995, 86 Fed. Reg. 7193 (January 21, 2021). 2 E.g., Exec. Order No. 13985, 86 Fed. Reg. 7009 (January 20, 2021). 3 E.g., David C. Wheelock, Comparing the COVID-19 Recession with the Great Depression (Aug. 12, 2020), https://research.stlouisfed.org/publications/economicsynopses/2020/08/12/comparing-the-covid-19-recession-with-the-great-depression. 4 E.g., Exec. Order No. 14008, 86 Fed. Reg. 7619 (January 27, 2021). 5 See, e.g., JEDIDIAH BRITTON-PURDY, THIS LAND IS OUR LAND: THE STRUGGLE FOR A NEW COMMONWEALTH (2019); The World We’ve Built, DISSENT, July 3, 2018 (describing the “Leviathan” required to deal with climate change); Adam Tooze, Shockwave, LONDON REV. OF BOOKS, Apr. 16, 2020 (describing the leadership and state capacity required to tackle the pandemic and its disastrous economic implications). 6 See, e.g., Wolfgang Streeck, The Return of the Repressed, 104 NEW LEFT REV. 5, 12 (2017) (agencies’ exclusionary expert elitism contributes to populist blowback); Andrew Arato, How We Got Here: Transition Failures, Their Causes and the Populist Interest in the Constitution, 45 PHILOSOPHY & SOCIAL CRITICISM 1106, 1108 (2019) (executive power contributes to a democratic deficit that drives populism); Albena Azmanova, The Crisis of Europe: Democratic Deficit and Eroding Sovereignty – Not Guilty, 24 L. & CRITIQUE 23 (2013) (technocratic government is insufficiently responsive to the undesirable consequences of its actions); JURGEN HABERMAS, TOWARD A RATIONAL SOCIETY: STUDENT PROTEST, SCIENCE AND POLITICS 81-82, Jeremy J. Shapiro, trans., 1971) (instrumental rationality, when performed by government bureaucracy, becomes a form of domination); THEODORE J. LOWI, THE END OF LIBERALISM: IDEOLOGY, POLICY AND THE CRISIS OF PUBLIC AUTHORITY 233 (1969) (arguing that the welfare state is an example of how private interest groups colonize public lawmaking). 7 See, e.g., Gillian E. Metzger, 1930s Redux: The Administrative State Under Siege, 131 Har. L. Rev. 1, 3-4 (2017) (describing proposed anti-administrative regulation and judicial suspicion of agency power – particularly that of Justice Gorsuch); PHILIP HAMBURGER, IS ADMINISTRATIVE LAW UNLAWFUL? (2014). 8 See ibid. and note 13, infra 9 See supra n. 6. ALL THE SOVEREIGN’S AGENTS 3 Pushing an anti-administravist10 agenda, libertarians continue their “long war”11 against government agencies by insisting that they are an unconstitutional fourth branch of government. For them, administration is a kind of “absolutism”12 that violates the separation of powers and defies the principle of limited government.13 They contend that agencies’ discretionary rulemaking offends the liberal commitment to the rule of law.14 Accordingly, they would punt agencies’ responsibility for social, economic, and environmental problems to courts and legislatures.15 Regulation would thus be placed at the mercy of an undemocratic judiciary who increasingly “weaponizes” the First Amendment in favor of big business16 – or of a Congress whose already inefficient decision-making is crippled by hyperpolarization17 and distorted by the kind of material inequalities that the welfare state is meant to ameliorate.18 10 I borrow this term from Metzger, supra note 7, at 4. Charles H. Koch, Jr., James Landis: The Administrative Process, 48 ADMIN. L. REV. 419, 420 (1996). 12 Hamburger, supra note 7, at 6. 13 E.g., Gary Lawson, The Rise and Rise of the Administrative State, 107 HARV. L. REV. 1231, 1233 (1994); Gary Lawson, Delegation and Original Meaning, 88 VA. L. REV. 327 (2002); RICHARD EPSTEIN, HOW PROGRESSIVES REWROTE THE CONSTITUTION (2006); JOSEPH POSTELL, BUREAUCRACY IN AMERICA: THE ADMINISTRATIVE STATE’S CHALLENGE TO CONSTITUTIONAL GOVERNMENT 4 (2017); Larry Alexander & Saikrishna Prakash, Reports of the Nondelegation Doctrine’s Death are Greatly Exaggerated, 70 CHI. L. REV. 1297 (2003). See generally Jeremy K. Kessler, The Political Economy of ‘Constitutional Political Economy, 94 TEX. L. REV. 1527 (2016). For a recent opinion implementing this view, see State of Florida v. Beccara et al., CA. No. 8:21-cv-839-SDM-AAS, *82-86 (M.D. Fla. June 18, 2021) (J. Merryday). 14 Hamburger, supra note 7, at 35-36, 50, 67-8, 72; ERIC A. POSNER & ADRIAN VERMEULE, THE EXECUTIVE UNBOUND: AFTER THE MADISONIAN REPUBLIC 4 (2010). For theoretical discussions, see FRIEDRICH A. HAYEK, LAW, LEGISLATION AND LIBERTY 27-8 (Vol. II, 1976); PIERRE ROSANVALLON, GOOD GOVERNMENT 26 (2018). 15 See, e.g., H.R. 5, 115th Cong. §§ 103, 202 (2017-2018) (increasing the constraints imposed on agency rulemaking and enhancing de novo judicial review of agency actions); Lowi, supra note 7, at 301. 16 See, e.g., Jedidiah Purdy, Beyond the Bosses Constitution: The First Amendment and Class Entrenchment, 118 COLUM. L. REV. 2161, 2162 (2018) (quoting Janus v. Am. Fed’n of State, Cty., & Mun. Emps., Council 31, 138 S. Ct. 2448, 2501 (2018) (Kagan, J., dissenting)). 17 See, e.g., MATTHEW HOLDEN, JR., CONTINUITY AND DISRUPTIONS: ESSAYS IN PUBLIC ADMINISTRATION 225 (1996); BLAKE EMERSON, THE PUBLIC’S LAW: ORIGINS AND ARCHITECTURE OF PROGRESSIVE DEMOCRACY 17-18 (2019). 18 Peter L. Strauss, Eroding ‘Checks’ on Presidential Authority: Norms, the Civil Service, and the Courts, 94 CHICAGO-KENT L. REV. 581, 585 (2020); Michael A. Livermore and Daniel Richardson, Administrative Law in an Era of Partisan Volatility, 69 EMORY L. J. 1, 45 (2019). 11 4 JACKSON Conservatives with a more authoritarian inflection seek to recall administration from its constitutional exile by subsuming it under Presidential power.19 Such critics would lend administration some democratic credentials by bootstrapping them to the President’s electoral accountability. Yet ridding agencies of their independence by placing them under the discretion of the President grants the President personal control over agency policymaking and adjudication without the checks provided by Congress, the Courts, or an independent civil service.20 It thus, arguably, solves a separation-of-powers problem by introducing a new one.21 More ominously, empowering the President with the patina of democratic legitimacy emits a strong whiff of Schmittian politics.22 The prospect of a largely unbound executive officer claiming a popular mandate to hire and fire civil servants on a whim should alarm any that followed the Trump Administration’s treatment of refugees, civil protestors, polluters, and political cronies. Agency power likely fares poorly in the hands of the left.23 They blame administrative technocracy for a variety of social and political ailments: the reification of social differences and the juridification of human 19 Posner & Vermeule, supra note 13, at 12-15; CARL SCHMITT, LEGALITY AND LEGITIMACY 13 (J.P. Seitzer trans., Duke U. P. 2004)(1932) (“the newly created financial bureaucracy…found the possibility of a new basis in the plebiscitary legitimacy of the German President elected by the entire German people.”); K. Sabeel Rahman, Reconstructing the Administrative State in an Era of Economic and Democratic Crisis, 131 HARV. L. REV. 1671, 1673 (2018) (book review). Recent cases showing this inclination include, e.g., Free Enterprise Fund v. PCAOB, 537 F.3d 667, 685 (D.C. Cir. 2008) (Kavanaugh, J., dissenting) (buffering PCAOB commissioners from presidential oversight is unconstitutional); Seila Law LLC v. Consumer Financial Protection Bureau, 591 U.S. ___, *11-12 (2020); Collins et al. v. Yellen, 594 U.S. ___, *27-28 (2021). 20 See, e.g., ADAM B. COX & CRISTINA M. RODRIGUEZ, THE PRESIDENT AND IMMIGRATION LAW 7-8 (2020); Trump v. Hawaii, 585 U.S. __ (2018). 21 DWIGHT WALDO, THE ADMINISTRATIVE STATE 145 (Routledge 2006) (1948); cf. ADRIAN VERMEULE, LAW’S ABNEGATION (2016) (arguing that no such problem exists because Congress knowingly delegated such policymaking to the executive). 22 Posner & Vermeule, supra note 14, at 53-4, 90-1; Samuel Moyn, The Guardians, THE NATION, Aug. 24, 2020, https://www.thenation.com/article/culture/eric-posnerdemagogues-playbook/ (book review) (reviewing ERIC A. POSNER, THE DEMAGOGUE’S PLAYBOOK (2020); James Chappel, Nudging Towards Theocracy: Adrian Vermeule’s War on Liberalism, DISSENT (Spring 2020). We can observe this dynamic Justice Gorsuch’s dissent in Gundy v. United States, when he states that “to allow the nation’s chief law enforcement officer to write the criminal laws he is charged with enforcing…would...invite the tyranny of the majority.” 139 S. Ct. 2116, 2145 (2019). 23 See Nicholas Bagley, The Procedure Fetish, 118 MICH. L. REV. 345, 348-50 (2019) (arguing that left-wing worries about agency legitimacy opened space for right-wing deregulatory reform). ALL THE SOVEREIGN’S AGENTS 5 nature24; corruption, privatization and regulatory capture25; the depoliticization of economic issues and the subsidization of globalized financial capitalism26 and, ultimately, the constellation of conspiratorial populist politics currently threatening liberal democratic states.27 Their preferred solutions include democratizing agency decision-making28 and constraining Congress’ capacity to delegate its lawmaking function.29 While their interventions are welcome, they may deprive government of the nimble expertise necessary to address environmental and economic crises.30 Moreover, as illustrated by the President’s extraordinary powers to shape national immigration policy despite its “notoriously complex and detailed statutory structure,” increasing the amount of formal legislation may only expand agencies’ enforcement discretion.31 Agency democratization, furthermore, risks reproducing, perhaps under the cover of ostensible public consensus, the same social, economic and political inequalities that distort Congressional lawmaking.32 In this essay, I contend that this multi-pronged anti-administravist attack stands upon shaky conceptual foundations. Each builds atop a theory of constitutionalism that embraces a too-literal conception of popular sovereignty.33 It is a conception that posits that there is, in fact, a “people” with a sovereign “will.” It is a “will” that can be clearly identified (through elections); straightforwardly transcribed (through lawmaking); mechanically 24 Habermas, supra note 6, at 121-2; Michel Foucault, Governmentality, in THE FOUCAULT EFFECT: STUDIES IN GOVERNMENTALITY 87 -104 (Graham Burchell et al. eds., 1991). For a discussion and summary of these critiques, see Emerson, supra note 17, at 7-8, 65. 25 See, e.g., Lowi, supra note 6; JON D. MICHAELS, CONSTITUTIONAL COUP: PRIVATIZATION’S THREAT TO THE AMERICAN REPUBLIC (2017). 26 Streeck, supra note 6; Azmanova, supra note 6. 27 Arato, supra note 6; Rosanvallon, supra note 14, at 162. 28 Rahman, supra note 19, at 1676; Emerson, supra note 17, at 17-18. 29 See, e.g., Robert Goodin, Welfare, Rights and Discretion, 6 OXFORD J. OF LEG. STUDIES 232, 250-3 (1986); Thomas Christiano, Democracy and Bureaucracy, 71 PHIL. & PHENOMENOLOGICAL RESEARCH 211, 212 (2005) (reviewing HENRY O. RICHARDSON, DEMOCRATIC AUTONOMY (2003)); Rosanvallon, supra note 14, at 160. 30 E.g., James Landis, Administrative Policies and the Courts, 47 YALE L. J. 519, 530 (1938). 31 Cox & Rodriguez, supra note 20, at 7-8. 32 Lowi, supra note 6; Emerson, supra note 17, at 76 (agency democratization reproduced racial bigotry); Richard B. Stewart, The Reformation of Administrative Law, 88 HARV. L. REV. 1669, 1764 (1975) (collective action problems and the involvement of unaccountable public interest organizations skew attempts at agency democratization); Theda Skocpol & Kenneth Finegold, State Capacity and Economic Intervention in the Early New Deal, 97 POL. SCI. Q. 255, 265 (1982) (arguing that the NRA, despite its goals of incorporating labor and consumer interests, became captured by business). 33 E.g., Jurgen Habermas, Three Normative Models of Democracy, 1 CONSTELLATIONS 1, 12 (1994); discussed infra, at pt. I(A). 6 JACKSON applied (by administrators) and constrained (by judges).34 But in a country of hundreds of millions, the diverse multiplicity of citizens could never find a common will.35 It is even more impossible that it could ever be accurately expressed through the lawmaking of elected representatives.36 As a result, critics of administration often grant statutory lawmaking more democratic credentials than it deserves.37 The non-delegation doctrine purports to prevent the delegation of something that simply may not exist. Critics commit another mistake when they invoke a theory of constitutionalism that analytically divides functions that cannot, as either a moral or empirical matter, be disentangled. First, they incorrectly posit two separate, autonomous processes: the collective formation of ends (lawmaking) and the implementation (execution) and application (adjudication) of those ends.38 But we cannot presume that judges and administrators can mechanically apply and enforce the law without importing into the process their own value-laden, and therefore political, judgments.39 “They who will the end will the means” is a naïve argument that occludes the power wielded by unelected actors.40 It is also a mistake to presume that the legislative branch concerns itself only with value-laden final ends, and not with the means required to execute them.41 Indeed, most of our most bitter political fights are fights conducted precisely over means: how best to grow 34 Discussed infra, at pts. I(A), I(B). E.g., CLAUDE LEFORT, DEMOCRACY AND POLITICAL THEORY 13-14 (David Macy trans., 1988); Richardson, supra note 29, at 65. 36 See, e.g., Michael Saward, Shape-Shifting Representation, 108 AM. POL. SCI. R. 723 (2014) (describing how politicians stand for different things to different people) and the discussion at part II(A), infra 37 This is called the “legislative primacy” argument. For an example, see Posner & Vermeule, supra note 14, at 8; Jeremy Waldron, POLITICAL POLITICAL THEORY 196 (2016). For a discussion of the historical intellectual roots of legislative primacy and its relation to individual rights, see Aniceto Masferrer & Anna Taitslin, The Ill-Fated Union: Constitutional Entrenchment of Rights and the Will Theory from Rousseau to Waldron, in THE LEGAL DOCTRINES OF THE RULE OF LAW AND THE LEGAL STATE (RECHSSTAAT), 105, 106, 112 (James E. Hickey & James R. Silkenat, eds., 2014) (explaining, inter alia, that Dicey believed in parliamentary supremacy and that the common law, enforced by judges, would adequately protect individual rights against arbitrary executive power and preserve the “rule of law.”). 38 Richardson (supra note 29, at 114) dubs this phenomenon “agency instrumentalism.” 39 E.g., Oliver Wendell Holmes, Jr., The Path of the Law, 110 HARV. L. REV. 991, 997-8 (1997) (1897) (though most well-known for his “bad man” theory of the law, the essay also argues that law cannot be applied using mechanical logic; that “[b]ehind the logical form lies a judgment as to the relative worth and importance of competing legislative grounds, often an inarticulate and unconscious judgment, it is true, and yet the very root and nerve of the whole proceeding”). 40 Richardson, supra note 29, at 116. 41 Ibid., at 119-27. 35 ALL THE SOVEREIGN’S AGENTS 7 the economy; how best to care for the sick; how best to mitigate climate change, etc.42 As a result, the theories overemphasize and distort the purpose of separating powers.43 Critics commit yet another mistake when they divorce the constitutional functions of (1) protecting rights and limiting government power, and (2) providing the decision-making procedures necessary for democratic will-formation.44 They isolate elections and lawmaking from the process of enforcing rights and the rule of law – as if they have nothing to do with one another. Yet quarantining rights from democracy requires reliance on an outsourced moral order external to the political system itself – a reliance inappropriate for contemporary secular polities.45 They therefore lend judges too many liberal credentials while denying any to mechanisms of popular feedback. Rather than critiquing agencies for violating the separation of powers, for their over-reliance on unelected technocrats, or for their indifference to universalizable legal principles, I argue that administration does indeed carry constitutional liberal democratic credentials – credentials borne out by political theory’s “representative turn.”46 By understanding agencies as embedded in a system of representative democracy that aims to set the conditions by which citizens can relate to each other as political equals, we can assess the legitimacy of government agencies without any “idolatrous”47 commitments to a fictitious popular sovereign or legal formalism. I suggest that agency institutions should be measured against the notion that popular sovereignty demands not consensus and consent, but instead institutions that permit citizens to understand themselves as co-equal participants in the collective decision-making process. This essay will proceed as follows. Part I situates administrative agencies in an understanding of liberal democratic constitutionalism that (A) eschews outmoded notions of popular sovereignty and (B) natural law. It will 42 Ibid., at 102. See, e.g., THE FEDERALIST NO. 51 (James Madison). 44 E.g., Habermas, supra note 33, at 8-9. 45 Probably the most seminal argument is found in MAX WEBER, THE PROTESTANT ETHIC AND THE SPIRIT OF CAPITALISM (Routledge 2001) (1904); see also MAX WEBER, ECONOMY AND SOCIETY (Guenther Roth and Claus Wittich eds., 1978) (1921); JURGEN HABERMAS, LEGITIMATION CRISIS 95-102 (Thomas McCarthy trans., 1973); cf. CARL SCHMITT, POLITICAL THEOLOGY 36 (George Schwab trans., 1985) (1922) (“All significant concepts of the modern theory of the state are secularized theological concepts…”). 46 Nadia Urbinati & Mark E. Warren, The Concept of Representation in Contemporary Democratic Theory, 11 ANN. REV. OF POL. SCI. 387 (2008). 47 Vermeule, supra note 21, at 56. 43 8 JACKSON then (C) explain how adequately conceived notions of the separation of powers and the rule of law cannot serve as indefeasible objections to administration. Part II makes a positive case for agency authority by drawing from the insights gained from political theory’s representative turn. It will first (A) define this important intellectual development and then (B) explain how administrative agencies might fit comfortably within a representative system. The essay (C) concludes by showing how theories of representation can inform some enduring debates in administrative law and suggesting some changes that might enhance the legitimacy of agency action. PART I: ADMINISTRATION, POPULAR SOVEREIGNTY AND RIGHTS Democracy promises the rule of “we the people.”48 Democratic citizens, possessing inalienable rights, are to come together, deliberate,49 and jointly create the laws that bind them. The administrative agency, with its unaccountable expert technocrats, policymaking autonomy, and immunity from micromanaging judicial review, looks like an unwelcome uncle at the constitutional dinner table. Intuitively, these knee-jerk objections cannot be quite correct. Agencies carry some obviously democratic credentials. As Adrian Vermeule points out, they are, after all, the creation of statutory lawmaking.50 At least as early as 1798, Congress has delegated coercive rule-making power to Federal bureaucracy on matters as diverse as tax inspections, territorial governance, veterans’ pensions, mail delivery, intellectual property, and the payment of public debts.51 In McCullough v. Maryland,52 the U.S. Supreme Court interpreted the “necessary and proper” clause53 to anticipate Congress’ desire to create such agencies – in this case, a national bank. Bruce Ackerman,54 in his seminal work, argues that our contemporary agencies carry Constitutional credentials. Many were birthed through multiple 48 See generally, e.g., BRUCE ACKERMAN, WE THE PEOPLE: TRANSFORMATIONS 279-314 (VOL. II 1998); U.S. CONST. pmbl. 49 Nathan K. Noh, Note, Non-Delegation as Non-Deliberation, 19 N.Y.U. L. REV. 379, 394 (2016). 50 Vermeule, supra note 21, at 9. 51 See generally, e.g., Nicholas R. Parillo, A Critical Assessment of the Originalist Case Against Administrative Regulatory Power: New Evidence form the Federal Tax on Private Real Estate in the 1790s, YALE L. J. (forthcoming 2021); Julian Davis Mortenson & Nicholas Bagley, Delegation at the Founding, 121 COLUM. L. REV., *76-*96 (forthcoming 2021) (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3512154); Kevin Arlyck, The Founders’ Forfeiture, 119 COLUM. L. REV 1449, 1452 (2019). 52 17 U.S. 316 (1819). 53 U.S. Const. art. I, § 8. 54 Ackerman, supra note 48, at 279-314. ALL THE SOVEREIGN’S AGENTS 9 hyperpolitical elections and constitutional challenges within the courts. Further, from their very inception, agencies struggled internally to accommodate their actions to constitutional requirements.55 The Administrative Procedure Act56 (“APA”), for example, imposes upon agencies principles of due process and the rule of law.57 Regardless, if democratic lawmaking is to shape the community of those that make it, there must be some kind of agent or instrumentality to carry it out.58 A Congressional decision to levy a tax is meaningless without an Internal Revenue Service to collect it.59 Yet it is impossible to imagine that such agencies might operate like mindless, loyal robots. Whether performed by court or administrator, the application of laws will inevitably involve policy judgments that are not included within the text of the law itself.60 Lawmaking is, by its nature, always more abstract than we would like. Such “general propositions do not,” noted Justice Holmes, Jr. in his influential Lochner v. New York61 dissent, “decide concrete cases.” The required elaboration almost always imports values that are not clearly identified in any statutory text.62 The task of accommodating administration to constitutional democracy cannot, therefore, aim at eliminating the agency costs implicit in the application of law. It can only seek to understand how they might comfortably fit within a constitutional order. The next two sections will elaborate upon these intuitions. Many objections to agency power presume antiquated conceptions of sovereignty and rights. They juxtapose the will of a powerful organ-body sovereign63 55 Sophia Z. Lee, Our Administered Constitution: Administrative Constitutionalism from the Founding to the Present, 167 U. PENN. L. REV. 1699, 1710-28 (2019). 56 5 U.S.C. §§ 551 et seq. 57 E.g., Richard B. Stewart, Administrative Law in the Twenty-First Century, 78 N.Y.U. L. REV. 437, 438 (2003). 58 See Woodrow Wilson, The Study of Administration, 2 POL. SCI. Q. 197, 212 (1887) (describing administration as the “Deed” that follows democratic will); JURGEN HABERMAS, BETWEEN FACTS AND NORMS 134 (William Rehg trans., 1996). 59 See Parillo, supra note 51. Theodore Sedgwick also noted, in 1791, that though Article I vested in Congress the right to coin money, “if no part of [Congress’] power be delegable, he did not know but [Congressmembers] might be obliged to turn coiners, and work in the Mint themselves.” 3 Annals of Congress 230-31 (1791) (statement of Rep. Sedgwick). 60 See infra, at pt. I(C); Richardson, supra note 29, at 116; Goodin, supra note29, at 238; Holmes, Jr., supra note 39, at 997-8. 61 198 U.S. 45, 76 (1905). 62 Holmes, Jr., supra note 39, at 997-8. 63 See Wilson, supra note 58, at 205 (juxtaposing democracy’s “corporate, popular will” against the will of an autocrat); Mary E. Guy, Ties that Bind: The Link Between Public Administration and Political Science, J. OF POLITICS 641, 642 (2003) (administration and lawmaking is “the expression of the state will” and “the execution of the state will.”); 10 JACKSON against a governed mass of subjects who hold an array of pre-political liberties that require judicial protection. This all-powerful body is thought to be represented by Congress64 as the commissioned agent (or embodiment?) of the popular sovereign. To preserve citizens’ natural, pre-political liberties, this agent of the popular sovereign is constrained by a separation of powers, checks and balances, a Bill of Rights, etc. – each policed by independent courts capable of identifying and enforcing citizens’ inalienable liberties.65 If this is indeed the rubric of the liberal democratic constitutional state, it is difficult to see how agencies pass constitutional muster. They are not Congress – and so their policymaking cannot be legitimate expressions of the popular will. They often avoid substantial judicial review, and so they might violate natural liberties with impunity. Fortunately, this rubric is wrong. A. The Mind and Body of the Democratic Sovereign True, for much of modern Western history, sovereignty, understood as the supreme, absolute and indivisible power to make law, was thought to be held by a specific body: the one wearing the crown.66 To constitute and justify public power, Hobbes, for example, imagined a state of nature full of individuals authorizing and relinquishing their natural liberties to a “Mortall God,”67 i.e., the modern corporate state, represented (or re-presented) in the flesh-and-blood bodies of the King or legislature.68 During the democratic Beccara, supra note 13, at *85 (the Constitution is “the organic grant of power from the sovereign people and the sovereign states”). 64 E.g., Posner & Vermeule, supra note 14, at 8 (calling this idea “legislative primacy”). Examples of arguments that use the idea of legislative primary include: George I. Lovell, That Sick Chicken Won’t Hunt: The Limits of a Judicially Enforced Non-Delegation Doctrine, 17 CONST. COMMENTARY 79, 85 (2000); Christiano, supra note 29, at 212; Cox & Rodriguez, supra note 20, at 2; JOHN LOCKE, TWO TREATISES OF GOVERNMENT 362 (Peter Laslett ed., Cambridge Univ. Press 1988) (1690) (ch. XI, ¶ 141); Stewart, supra note 32, at 1672 (Legislative primacy “appears ultimately to be bottomed on a contractarian political theory running back to Hobbes and Locke... Since the process of consent is institutionalized in the legislature, that body must authorize any new official imposition of sanctions…The requirement that agencies conform to specific legislative directives … legitimates administrative action by reference to higher authority…”). 65 Posner & Vermeule, supra note 14 (calling this idea “liberal legalism”); P.S. ATIYAH & ROBERT S. SUMMERS, FORM AND SUBSTANCE IN ANGLO-AMERICAN LAW: A COMPARATIVE STUDY OF LEGAL REASONING, LEGAL THEORY, AND LEGAL INSTITUTIONS (1987); see also JUDITH SHKLAR, LEGALISM: LAW, MORALS, AND POLITICAL TRIALS (1986). 66 E.g., JEAN BODIN, ON SOVEREIGNTY: FOUR CHAPTERS FROM THE SIX BOOKS OF THE COMMONWEALTH 1-2 (Julian H. Franklin ed., trans., Cambridge U. P. 1992) (1576); EDMUND S. MORGAN, INVENTING THE PEOPLE: THE RISE OF POPULAR SOVEREIGNTY IN ENGLAND AND AMERICA 18-19 (Rev’d 2013) (1989). 67 THOMAS HOBBES, LEVIATHAN 227 (C.B. MacPherson ed., Penguin Books 1985) (1651). 68 See generally Quentin Skinner, Hobbes and the Purely Artificial Person of the State, 7 J. OF POLITICAL PHILOSOPHY 1 (1999) (for an explanation of the corporate conception of the ALL THE SOVEREIGN’S AGENTS 11 revolutions, radical69 theorists merged the monarch with her subjects.70 They imagined “the people” not only replacing the king as sovereign, but also governing itself as a subject, thereby creating an identity between ruler and ruled. Rousseau’s volonté générale71 serves as a model for this kind of logic.72 Montesquieu, whose thinking influenced the American founders,73 likewise held that the “people as a body have sovereign power” in a republic.74 Even A.V. Dicey, despite his fame as a rule of law scholar, believed that a representative legislature would “produce coincidence between the wishes of the sovereign and the wishes of the subjects.”75 It is a sovereign-subject hat trick: the ruled become the ruler, the democratic “people,” understood as a body, a “unitary macro-subject,”76 come to occupy what was once occupied by the body of the king. Carl Schmitt likewise endorsed a scrupulous identity between governed and governor - with homogenizing and fascist implications.77 For Schmitt, it was impossible to imagine a leader speaking with the voice of the people unless the people themselves first sing in perfect harmony. There are flaws in this equation. The “people,” understood literally, cannot rule. They do not possess a primordial collective will existing outside and independent of their political institutions.78 Moreover, the entire state as both human and corporate (artificial) person); DANIEL LEE, POPULAR SOVEREIGNTY IN EARLY MODERN CONSTITUTIONAL THOUGHT 11 (2016). 69 See Lee, supra note 68, at 4 (the conventional narrative of popular sovereignty couches it as an idea of resistance). 70 E.g., Morgan, supra note 66, at 87-88; MARGARET CANOVAN, THE PEOPLE 24-9 (2005); see generally RICHARD TUCK, THE SLEEPING SOVEREIGN (2016). 71 JEAN-JACQUES ROUSSEAU, THE SOCIAL CONTRACT AND DISCOURSES 192-5 (G.D.H. Cole trans., Everyman 1993); cf. Tuck, supra note 70, at 4-5 (arguing that Rousseau and Bodin drew a sharp distinction between (popular) sovereign and government). 72 Nadia Urbinati, Continuity and Rupture: The Power of Judgment in Democratic Representation, 12 CONSTELLATIONS 194, 203 (2005); Rosanvallon, supra note 13, at 76-7; Habermas, supra note 32, at 12; Nancy Rosenblum, ON THE SIDE OF THE ANGELS: AN APPRECIATION OF PARTIES AND PARTISANSHIP 29 (2008). 73 E.g., PAUL M. SPURLIN, MONTESQUIEU IN AMERICA 1760-1801 (1940). 74 MONTESQUIEU, THE SPIRIT OF THE LAWS 10 (Anne M. Cohler et al. eds., Cambridge U. Press 2011) (1748). 75 Albert Venn Dicey, INTRODUCTION TO THE STUDY OF THE LAW OF THE CONSTITUTION 81 (8th ed. 1931) (1885). 76 Andreas Kalyvas, Popular Sovereignty, Democracy, and the Constituent Power, 12 CONSTELLATIONS 223, 224 (2005). 77 CARL SCHMITT, THE CRISIS OF PARLIAMENTARY DEMOCRACY 25-32 (trans. Ellen Kennedy, 1988) (1923); also, e.g., Canovan, supra note 70, at 74-7; Rosanvallon, supra note 14, at 76; Kalyvas, supra note 76, at 224; HANNAH ARENDT, THE HUMAN CONDITION 57-8 (2d ed. 1998) (1958). 78 E.g., STEPHEN HOLMES, PASSIONS AND CONSTRAINT 164 (1995); p. 164; Jeremy Waldron, LAW AND DISAGREEMENT 277 (1999); Lisa Disch, The End of Representative Politics? In 12 JACKSON population of a diverse community of hundreds of millions cannot be present within those institutions. Nor can that population ever find a unanimous general will, a non-controversial understanding of the common good, no matter how constrained and qualified their public reasoning or how universal and general its aspirations.79 Thus, no coherent popular will can obtain even after undertaking the decision-making processes of political institutions.80 Just as the contractual “meeting of the minds” is a legal fiction of private law,81 a popular “meeting of the minds” is a political fiction of public law. As a result, despite the democratic revolutions, the old gap between ruler and ruled remains.82 In other words, the merger between governed and governor attempted by the democratic revolutions did not remove the danger of heteronomy,83 even if the offices of government might be staffed by elected representatives and even as constitutional systems split powers and limited THE CONSTRUCTIVIST TURN IN POLITICAL REPRESENTATION 10 (Lisa Disch et al. eds. 2019); see, e.g., Saward, supra note 35. 79 E.g., Schmitt, supra note 19, at 7; Franz Neumann, The Concept of Political Freedom, in THE DEMOCRATIC AND THE AUTHORITARIAN STATE 160, 171, 190 (Herbert Marcuse ed., 1957) (1953) (Neumann is worth citing because he engaged contemporaneously with Schmitt’s work); Habermas, supra note 33, at 10); JOSEPH SCHUMPETER, CAPITALISM, SOCIALISM AND DEMOCRACY 251 (Harper 3d ed. 2008) (1942); see also Waldron, supra note 37, at 43 (“It is probably a mistake for any branch of government to assume the mantle of popular sovereign.”), 77 (“I am inclined to say there is no such thing as the people. There is just. Large array of individual persons, millions of them…And there are innumerable ways of mapping onto that array another array, namely the five or six hundred seats that there are in a given legislative assembly.”), 180 (“There is no reason why the theory of democratic accountability should be held hostage to any particular political ontology – ‘the people,’ as a singular entity, the genearl will, the will of the people, the majority, and so on”); Nadia Urbinati, Representation as Advocacy: A Study of Democratic Deliberation, 28 POLITICAL THEORY 758, 774 (2000) (citing J.S Mill, aruges that “[w]ith respect to the ‘identification of interest btween the rulers and the ruled,’ [Mill] assumed that such an identification could hardl exist; in fact, it did not. If it did, representation, and perhaps government itself, would be unnecessary.”). 80 E.g., Tuck, supra note 70, at 3-4. This is perhaps why jurists like Dicey and Austin separated the concepts of legal sovereignty and political sovereignty. Without a “will,” the people cannot speak and act – literally – through their representatives. There is no feasible way to merge the legal sovereignty exercised by rulers and the popular political sovereignty enjoyed by the people. See Mansferrer and Taitslin, supra note 37, at 112 n. 36, 40. 81 See, e.g., Oliver Wendell Holmes, Jr., Privilege, Malice, and Intent, 8 HARV. L. REV. 1 (1894); Robert L. Hale, Coercion and Distribution in a Supposedly Non-Coercive State, 38 POL. SCI. Q. 470 (1923); Hotchkiss v. National City Bank, D.C., 200 F. 287, 293 (S.D.N.Y. 1911) (Hand, J.) (making a strong case for the objective theory of contract). 82 E.g.,Canovan, supra note 70, at 28-9 (showing that the American founders, despite their corporate notion of popular sovereignty, recognized a gap between elected legislators and “the people”); Urbinati, supra note 79, at 759. 83 DIETER GRIMM, SOVEREIGNTY: THE ORIGIN AND FUTURE OF A POLITICAL AND LEGAL CONCEPT 31 (2015); Canovan, supra note 70, at 29; Neumann, supra note 79, at 190. ALL THE SOVEREIGN’S AGENTS 13 legal authority.84 Some (body) would wield public power, and the rest would be subject to its rules. Even Rousseau downgraded the popular sovereign to a silent, passive actor that left the actual business of governing to functionaries.85 Like the client of a travel agent, Rousseau’s democratic citizen was meant only to approve or disapprove the prepackaged plans presented by ministers.86 Lawmaking under constitutional liberal democracy is thus not a question of ascertaining the existence of some non-existent popular “will” to be left in the hands of loyal fiduciaries in government87 to carry out like mindless automatons. Nor is it comprised of the dictates of a caesarist leader purporting to speak with the unified voice of the sovereign people.88 Instead, it a question of developing transparent and accessible collective decisionmaking procedures that ensure that all citizens can understand themselves as equal participants in their collective ordering; that ordinary people are involved in public life and have a say in their collective destiny.89 They do not rule. Rather, they are equal players in the game of representative democracy.90 Thus, although contemporary notions of constitutional liberal democracy ascribe the highest legitimate source of authority to “the people,” they do not understand “the people” as a reified, homogenous whole with an identifiable will that pre-exists whatever governing apparatus might be laid atop it. Though “popular sovereignty” is a political fiction, it is a useful one – at least if it is used as a standard of justification and critique, not as a proper noun. It is an aspirational, regulative idea intended to depersonalize and distribute public power in a way that serves the entire community.91 It is a Kantian “as if” principle.92 Namely, if we try to think like a popular sovereign might think, if such a thing could ever exist, we will orient our public reasoning not towards our individual self-interest alone, but in terms of 84 Morgan, supra note 66, at 53. Tuck, supra note 70, at 4-5; Lee, supra note 68, at 5, 11. 86 Urbinati, supra note 72, at 207. 87 An example of this kind of thinking is found in Locke, supra note 64, at 367 (…[T]he Legislative being only a Fiduciary Power to act for certain ends, there remains still in the People a Supream Power…”)(emphasis in original) and in Rousseau, supra note 71, at 235. 88 ERNESTO LACLAU, ON POPULIST REASON 159-60 (2005) is probably the most influential contemporary statement of this Schmittian notion; also, e.g., Arato, supra note 6. 89 E.g., Grimm, supra note 83, 37, 73-4; Urbinati, supra note 72 at 215; GREGOIRE C.N. WEBBER, THE NEGOTIABLE CONSTITUTION 19-20 (2009); Rosanvallon, supra note 14, at 182; Jurgen Habermas, supra note 57, at 132-3, 169. 90 See Part II, infra. 91 Habermas, supra note 33, at 11-12; Lee, supra note 68, at 14. 92 Urbinati, supra note 72, at 213. 85 14 JACKSON inclusivity, human equality and the public good.93 Because if the sovereign is a “we,” then governing involves more than the interests and preferences of single individuals. We will therefore demand that political institutions remain accountable and accessible to popular complaints. We will adopt a Weberian politics of responsibility, remembering that our decisions might inflict unforeseen costs upon others.94 This figurative idea of popular sovereignty also unlocks the closed doors of power and forces the inclusion of voices previously ignored.95 Whosoever happens to be governing at any given time, that person is not “the people” precisely because “the people” cannot ever be present. As a result, anyone denied an audience can appeal to popular sovereignty as they seek admission to political decision-making. Importantly, popular sovereignty demands, as French philosopher Claude Lefort96 notes, that this place of power remain an empty one – or at least one with a revolving door – where no body at all is permitted to rule permanently. For to fill that void would allow for a part to speak on behalf of the whole. “We the People” might become, as political theorist Nadia Urbinati notes, “Me the People.”97 It would thus force homogeneity upon plural societies as leaders with controversial viewpoints purport to represent everyone as they make and implement policy. Moreover, the usurpation of this space would undermine the depersonalization of power inherent in the idea of a fictional popular sovereign and, importantly, the rule of law and not of men.98 If the place of power remains empty because all citizens contribute in some way to lawmaking, then we can credibly claim that it is law, not our politicians, who rule. As a result, it can be no objection to agency policymaking that it usurps authority from the popular sovereign. Because if we take popular sovereignty literally, so, too, do elected representatives. They likewise cannot logically or credibly speak with the voice of the sovereign people.99 Thus, insofar as theories of non-delegation and legislative primacy rely on an 93 Ibid., at 214. See Max Weber, Politics as a Vocation, in MAX WEBER: ESSAYS IN SOCIOLOGY 41 (H.H. Gerth & C. Wright Mills eds., 1946) (1919) 95 Canovan, supra note 70, at 36. 96 Supra note 35; see also Lee, supra note 68, at 15. 97 See generally NADIA URBINATI, ME THE PEOPLE: HOW POPULISM TRANSFORMS DEMOCRACY (2019). 98 Ibid.; see also Locke, supra note 64, at 136; Rosanvallon, supra note 14, at 26. 99 E.g., Rousseau, supra note 71, at 229 (sovereignty cannot be delegated). 94 ALL THE SOVEREIGN’S AGENTS 15 organ-body theory of popular sovereignty,100 they are misplaced. Attacks against the “technocratic” power wielded by administrative officers may likewise overstate the democratic credentials of the Congressional legislation against which such power is compared – and found wanting. Indeed, it is at least possible that administrative agencies can be made consistent with the requirements of constitutional popular sovereignty.101 Namely, the question is whether and to what extent they operate according to procedures that allow citizens to understand themselves as co-equal participants in shaping agency action. Finally, that independent administration is “headless” is not, as feared by contemporary New Deal critics, fascist or totalitarian.102 It may in fact be a necessary precondition for liberal democracy. A Leviathan with a single head with a single mouth, purporting to speak for all, can be monstrous indeed. B. Individual Rights and Rule of Law Many objections to agency policymaking and adjudication do not emphasize principles of legislative primacy and non-delegation. Instead, they take aim at administrative “particularism”103 and the risk it poses to individual liberty and the rule of law. Though they prioritize the Constitution’s function as a check on power over its function to constitute power,104 these critiques likewise rely on some sovereign authority legibus solutus. This time, though, it is rendered not as voluntas, the will of a sovereign body, but instead as moral truth. Some libertarian critiques rest, 100 See supra n. 37; other examples may include Lawson, Postell and Epstein, supra note 13. According to Waldron, “[i]n liberal political theory, legislative supremacy is often associated with popular self-government…” Waldron, supra note, at 196. 101 As discussed infra, at pt. II, agencies might provide non-electoral forms of democratic representation. See, e.g., Michael Saward, Authorisation and Authenticity: Representation and the Unelected, 17 J. POL. PHIL. 1 (2009). 102 DONALD A. RITCHIE, JAMES M. LANDIS: DEAN OF THE REGULATORS 85 (1980). Lefort, supra note 35, at 19, argues that, indeed, cutting of the head of the king – metaphorically and literally - was necessary for democracy. 103 Objections to executive “particularism” appear in Rousseau, supra note 71, at 229. 104 For example, Hamburger, supra note 7, at 17, argues that “[c]onstitutional law developed in the seventeenth century primarily as a means of defeating the absolute prerogative.” Cf., e.g., Lee, supra note 68, at 14 (explaining that popular sovereignty developed as a concept not just to contest power, but to legitimately constitute it); Waldron, supra note 37m at 23, 34, 284 (constitutions both constitute and limit power; the former is not always emphasized sufficiently); 297 (citing Hannah Arendt, ON REVOLUTION 148, 154 (2006) (1963). See also Gerald E. Caiden, In Search of an Apolitical Science of American Public Administration, in POLITICS AND ADMINISTRATION: WOODROW WILSON AND AMERICAN PUBLIC ADMINISTRATION 51, 55 (Jack Rabin & James Bowman eds., 1984) 16 JACKSON explicitly or implicitly, on the idea of the “unwritten Constitution”105 which protects an array of rationally discoverable pre-political liberties106 that were enshrined into positive law by the Bill of Rights107or guarded by the common law and its courts.108 Rights have an objective existence; therefore, all reasonable people can (or must) consent to them. They accordingly should be identified and protected by impartial, unelected specialist judges.109 Other critiques speak in the language of the rule of law. Lending “the rule of law” a definition endorsed by Dicey and Coke, they charge that because agency rulemaking and adjudication occur outside regular courts (Article III courts) and regular lawmaking bodies (Congress), they risk exercising the same kind of absolute, arbitrary power wielded by unelected Stuart and Tudor monarchs.110 Although couched in terms of black letter constitutional doctrine, the underlying principles motivating this critique, like libertarian rights-based arguments, rely on an idealized moral order that encourages small government. This time, however, that moral order bootstraps itself to the democratic legitimacy provided by Congress.111 105 These arguments come in several forms. Some hold that the Constitution was never meant to supplant pre-existing judicial power to protect fundamental rights, implicitly protected by principles like “the rule of law” and due process. E.g., Hamburger, supra note 7, at 22). They can also come in through “inclusive” legal positivist arguments (e.g., RONALD DWORKIN, LAW’S EMPIRE (1986)) that hold that “outside” moral values should guide judicial discretion. 106 E.g., Richard Schragger & Micah Schwartzman, Some Realism about Corporate Rights, in THE RISE OF CORPORATE RELIGIOUS LIBERTY 345, 353 (C. Flanders & Z. Robinson eds., 2016); Franz Neumann, Types of Natural Law, in THE DEMOCRATIC AND THE AUTHORITARIAN STATE 69, 89 (Herbert Marcuse ed., 1957)(1939); Douglas Sturm, A Prospective View of the bill of Rights: Towards a New Constitutionalism," 13 J. LAW & RELIGION 27, 29-30 (1996). 107 THOMAS MCAFEE, INHERENT RIGHTS, THE WRITTEN CONSTITUTION, AND POPULAR SOVEREIGNTY: THE FOUNDERS’ UNDERSTANDING (2000); Neumann, supra note 79, at 164; Roscoe Pound, Liberty of Contract, 18 YALE L. J. 454, 457 (1909). 108 E.g., Mansferrer and Taitslin, supra note 37, at 112 (“According to Dicey, the common law checked the state’s arbitrary power through the authority of judges.”) The role of the common law in protecting “natural” rights is also observed in cases like Lochner v. New York, 198 U.S. 45 (1905), when “vested” common law property rights were given due process protections under the 14th Amendment. ROBERT DAHL, A PREFACE TO ECONOMIC DEMOCRACY 63 (1985). 109 Christopher Forsyth, Showing the Fly the Way out of the Flybottle, 66 CAMBRIDGE L. J. 325, 332 (2007); Hamburger, supra note 7, at 45 (in his diagnosis of the unlawfulness of administration, Hamburger explicitly relies on norms that transcend Constitutional legal doctrine and could be characterized as a natural law argument. See, e.g., ibid., at 15, 385, 493). See Lawson, supra note 13, at 1529-31; Pound, supra note 107, at 157-8. 110 Hamburger, supra note 7, at 35, 47. 111 Dicey, for example, embraced “parliamentary supremacy.” Mansferrer and Taitslin, supra note 37, at 106. However, his rule of law arguments were not aimed at limiting parliamentary power so much as they were aimed at curtailing executive powers. Ibid. at 112. ALL THE SOVEREIGN’S AGENTS 17 Emphasizing agencies’ irregular rulemaking, they conjure the ghost of the Kantian social contract that, in its impartiality and generality, vindicates citizens’ equal negative liberty while protecting them from personal tyrannical rule.112 According to this view, statutes should arise from a deliberative decision-making procedure that effaces individual interest and consequently produces impersonal rules that permit citizens significant freedoms.113 The law that rules over them should be a “faceless order that aspires to be universal and eternal, after the example of the divine, and equal to it, an order launched into pace and into time, where an anonymous crowd meets invsible generations.”114 In Justice Gorsuch’s langauge, legislation should consist of “generally applicable rules of conduct governing future actions by private persons.”115 If a legislature – Congress – is operating properly, it should promulgate only a small number of abstract, rational and general laws amenable to straightforward application by workmanlike and subordinate judicial and executive bodies.116 It should, as Locke persuaded, “be conformable to the Law of Nature.”117 It would thus preserve freedom 112 Rosanvallon, supra note 17, at 26; Hamburger, supra note 7, at 35-6 (arguing that “Lockean reasons about consent” “lurk[] not far below” the legitimacy of Congressional lawmaking), 50, 72; Franz Neumann, The Change in the Function of Law in Modern Society, in THE DEMOCRATIC AND THE AUTHORITARIAN STATE 22, 26 (Herbert Marcuse ed., 1957) (1937); Waldron, supra note 37, at 153 (general legislation “presents itself in the image of morality…To make law…is not just to exercise power; it is (so to speak) to make a public morality for a particular community”); Laurence Lustgarten, Socialism and the Rule of Law, 15 J. LAW & SOC’Y 25, 25 (1988); Locke, supra note 64, at 357, 363-4. An example of this notion of the rule of (deliberative, congressional law) can be found in Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 654-55 (1952) (J. Jackson, concurring). 113 See, e.g., Gundy, supra note 22, at 2134 (Gorsuch, J., dissenting); Waldron, supra note 37, at 17, 66. 114 Rosanvallon, supra note 14, at 27; see also Neumann, supra note 79, at 170; Noh, supra note 49, at 394 (citing The Federalist’s references to impartial deliberation); Lawson, supra note 13, at 1531; Hayek, supra note 14, at 12. An example of the antithesis of the idea of “general law” is the ratemaking of the Progressive era. For a discussion of the debate between Ernst Freud and Felix Frankfurther on the tension between general, abstract law and “substantive justice,” see DANIEL R. ERNST, TOCQUEVILLE’S NIGHTMARE: THE ADMINISTRATIVE STATE EMERGES IN AMERICA 11-14 (2014). 115 Gundy, supra note 22, at 2133. 116 Neumann, supra note 79, at 169-70 (“the judge [is] merely the mouthpiece of the law,” “applying it through a logical process of subsumption”); see also, e.g., Montesquieu, supra note 74, at 163; Neumann, supra note 106, at 29-38; Rosanvallon, supra note 14, at 29; FRANK GOODNOW, POLITICS AND ADMINISTRATION 24 (1900); M.J.C. VILE, CONSTITUTIONALISM AND THE SEPARATION OF POWERS 26 (1967) (citing Aristotle’s ETHICS V (10) and POLITICS III(16)) 117 Locke, supra note 64, at 369. 18 JACKSON because it, as a real-life variety of the Kantian118 social contract, could only settle on terms that maximized equal (negative) liberty for all. This ideal is juxtaposed to profligate agency decision-making that addresses specific actors in specific situations while addressing specific interests. As under the natural law objection, agencies’ irregular adjudication of this law is likewise objectionable. Only expert judges committed its neutral application will do.119 The rule of law objection therefore often follows the same logic as the natural rights and popular sovereignty objections: there is an unambiguous, pre-constitutional normative framework – the popular sovereign, operating rationally through the social contract to generate objectively good “general legislation” – that should be straightforwardly applied by expert judges in order to constrain executive action. Indeed, for Franz Neumann, a critical contemporary of Carl Schmitt and associated with the Frankfurt School,120 this notion of the rule of law “is really a disguised revival of natural law which is now fulfilling counterrevolutionary functions”121 – despite its ostensibly democratic, secular pedigree. Both the natural rights and rule-of-law variants of this objection to administration conceive the law as something general, simple, self-evident, derived (or theoretically derivable) from the rational consent of the governed, and amenable to non-controversial application to real-life particularities. This conception is mistaken. First, rights, whether legal or natural, are never so clear and unambiguous that they are capable of application without controversy. The Legal Realist critique122 of the late 19th and early 20th centuries shows convincingly that the law is often indeterminate (or under-determinate)123 and that, therefore, judicial elaboration and application of that law commonly amounts to a political choice – a choice that should not be left in the hands of politically unaccountable judges.124 As a result, privileging Article III 118 Neumann, supra note 106, at 34. Vile, supra note 116, at 27 ; Hamburger, supra note 7, at 38, 82. 120 Founded at the Goethe University Frankfurt during the Weimar Republic, this school critiqued capitalism, fascism and Marxism-Leninism. It is a home of the intellectual “New Left.” Members include, e.g., Theodor Adorno, Max Horkheimer, Herbert Marcuse, Raymond Guess, and Jurgen Habermas. 121 Neumann, supra note 112, at 53; Neumann, supra note 106, at 89. 122 This critique is addressed in a little more detail in the next section. 123 CASS SUNSTEIN & ADRIAN VERMEULE, LAW AND LEVIATHAN: REDEEMING THE ADMINISTRATIVE STATE 112 (2020); Brian Leiter, Legal Realism and Legal Positivism Reconsidered, 111 ETHICS 278, 295-6 (2001). 124 Waldron, supra note 37, at 232 (“At most, the abstract terms of the Bill of Rights are popularly selected sites for disputes about [about individual and minority rights]”.). 119 ALL THE SOVEREIGN’S AGENTS 19 judges as the guardians of liberty may simply swap one unpopular sovereign for another. Second, contemporary economic and social conditions frustrate the basic ethical claims of both natural rights and the rule of the law. Each targets the protection of equal liberty. Yet applying both to corporate capitalist society often leads to unfreedom and domination.125 Concurrently with the Legal Realists, for example, Neumann showed that the ideal of liberal parliamentarism – general lawmaking – was suited not for monopolistic capitalism and wage labor, but for the bourgeois, competitive and egalitarian market utopias imagined by Adam Smith and John Locke.126 “The postulate that the state should rule only by general laws,” quips Neumann, “becomes absurd in the economic sphere if the legislator is dealing not with equally strong competitors but with monopolies which reverse the principle of the free market.”127 Because the private power of big business impacts the public in a unique way, “the state can only regulate via individual measures” if it is to achieve substantive social and political equality.128 Meanwhile, critical scholars demonstrated that the law could never cash out its promise of equal liberty if it dealt with citizens as abstract subjects, not as concrete individuals in all their diversity.129 Instead, to vindicate citizens’ equal rights in a complex, unequal and changing society, government must inevitably eschew what Rosanvallon dubs “the twin cults of law and democratic impersonality”130 and instead embrace the particularism liberals once associated with despotism.131 It must treat differences differently. Third, and relatedly, legislation cannot, and usually does not, achieve a politically noncontroversial and equality-protecting generality. According to Neumann, for example, a society riven by intractable class conflict and amalgamations of private economic power could never unanimously consent to any ostensibly rational general law.132 As a result, whatever lawmaking results from Congress, it is not what the liberal parliamentarians hoped it 125 E.g., JOHN DEWEY, LIBERALISM AND SOCIAL ACTION (1999)(1935); Pound, supra note 107; Thomas Hill Green, Lectures on the Principles of Political Obligation, in WORKS OF THOMAS HILL GREEN (Vol. II, 2011); Karl Marx, Capital. Vol. I, in THE MARX-ENGELS READER 294, 343 (Robert C. Tucker ed., 2d ed. 1978) (1867). 126 Neumann, supra note 106, at 39-40. 127 Ibid. 128 Neumann, supra note 79, at 171-2. 129 E.g., William E. Scheuerman, Recent Frankfurt Critical Theory: Down on Law? 24 CONSTELLATIONS 113, 121 (2017) (describing the work of Axel Honneth); Lustgarten, supra note 112, at 29. 130 Rosanvallon, supra note 14, at 41. 131 Ibid., at 65. 132 Neumann, supra note 79, at 169. 20 JACKSON might be: universal rules whose application guarantees equal liberty and the public good.133 Contemporary work on deliberative democracy, along with critiques of Rawlsian notions of justice, support the point.134 True universality requires the incorporation of all marginalized voices – a utopia rarely achieved under conditions of material, social, and cultural inequality.135 Natural rights face a similar fate. As Weber diagnosed, the great problem of secular modernity is our lack of confidence in any outsourced, transcendental moral authority.136 Contemporary liberalism admits the fact that there can be no authoritative notion of moral truth. It therefore quarantines the question, holding instead that the only legitimacy a governing authority can ever hope to claim derives from a Rawlsian ‘overlapping consensus’ regarding the norms according to which diverse people can arrange their collective life peaceably together.137 This overlapping consensus remains, however, more aspirational than real. No longer can we reach to superordinate natural law or social contract theories for instruction. Our social fate is one of ineluctable conflict, dissensus, and plurality. Finally, these critiques of administration envision a constitution whose purpose is primarily the limitation and division of the regrettable-butinexorable presence of state power.138 Rather than empower legitimate decision-making, these critics would hamstring collective action by binding it to an outsourced moral ordering: pre-political natural rights, a primordial common law of “vested” rights, or a rational, epistemic conception of the “general will” that coheres with a deregulatory economic policy. Yet the Constitution is better understood as more than a check on power. It also aims to constitute legitimate power by giving citizens an equal say in forming collective political life, thereby cashing out the promise of popular 133 Ibid. See generally, e.g., Habermas, supra note 57; SEYLA BENHABIB, THE CLAIMS OF CULTURE (2002); Rainer Forst, The Justification of Human Rights and the Basic Right to Justification: A Reflexive Approach, 120 ETHICS 711 (2010); IRIS M. YOUNG, DEMOCRACY AND INCLUSION (2000). 135 Ibid. 136 Max Weber, Objectivity in Social Science and Social Policy, in THE METHODOLOGY OF THE SOCIAL SCIENCES 86 (E.A. Shils & H.A. Finch, eds. & trans., 1949)(1904). 137 E.g., Waldron, supra note 37, at 244-5. For a discussion of this constructivist account of rights in relation to administration, see Emerson, supra note 17, at 13. 138 See, e.g., Hamburger, supra note 7, at 46(“[C]onstitutional law developed in the seventeenth century precisely to bar extralegal power.”). 134 ALL THE SOVEREIGN’S AGENTS 21 sovereignty.139 Focusing exclusively on its power-checking features thus focuses only on one side of the coin. The roadblocks offered by these objections to administration can be overcome if one accepts a democratic constructivist account of rights. It is an account that holds that rights and democracy are co-original. Namely, it is an account that holds that citizens should participate in the creation of the (legal) rights they give themselves.140 The construction of rights depends upon a decision-making process that attempts to cash out the promise of equal human worth – while always remaining subject to critique along the same terms. Rights therefore begin as abstract concepts,141 and, in their most inchoate form, they are a right to have equal rights.142 They are articulated and elaborated by historical democratic polities whose lawmaking institutions are, hopefully, oriented towards cashing out the promise of the equal liberty as citizens themselves understand it.143 Rights result from social movements that emerge within civil society to be provisionally codified within the law through regular,144 constitutional and extra-constitutional procedures.145 Their content remains contingent and always within the grasp of democratic citizens.146 Whatever resolution may temporarily arrive, uncertainty about both future factual circumstances and normative commitments precludes any final answer.147 Thus, the abstract rights that democratic practice concretizes are not indefeasible principles so much as “modes of problematisation” calling for democratic negotiation and 139 See, e.g., Lee, supra note 68; S. Holmes, supra note 78; Habermas, supra note 33; Waldron, supra note 37, at 34; Edward Rubin, Law and Legislation in the Administrative State, 89 COLUM. L. REV. 369, 370 (1989). 140 Habermas, supra note 57, has the seminal statement of co-originality. 141 E.g., Webber, supra note 89, at 1-2; Seyla Benhabib, Democratic Iterations, in ANOTHER COSMOPOLITANISM 45 (2006) (giving an example: how religious liberty and state neutrality were interpreted in France following the headscarf controversy). Democratic discourse thus contains meta-ideals of equality and liberty. In other words, liberal constitutional democracy relies on an abstract a priori concept of equal moral worth. 142 HANNAH ARENDT, THE ORIGINS OF TOTALITARIANISM 296 (1976) (1951). 143 Supra note 141. 144 See, e.g., Ernest A. Young, The Constitution Outside the Constitution, 117 YALE L. J. 408, 412, 424 (2007) (arguing that many important rights are codified via statutory law). 145 JEAN L. COHEN & ANDREW ARATO, CIVIL SOCIETY AND POLITICAL THEORY (1994) (theorizing the creation of rights through social movements); Ackerman, supra note 48 (describing the extra-constitutional creation of the Reconstruction amendments). 146 Nadia Urbinati, DEMOCRACY DISFIGURED 17 (2014); Webber, supra note 89, at 38, 1478. 147 See Dewey, supra note 125, at 21-7, 30-1 (tracing the changing meanings of liberalism in regards to economic life); Pound, supra note 107 (describing the historical and theoretical contingency of the right to contract). 22 JACKSON discourse.148 This uncertainty, and the humility it implies, is no bad thing. For to settle on the truth about their content would mean relinquishing political sovereignty into the hands of something else entirely– namely, someone who claims special expertise about the correct interpretation of rights.149 Meanwhile, the equal liberty promised by an impersonal rule of law150 is perhaps better vindicated through a lawmaking process that affords citizens not equal subjection to the same general law, but instead an equal role in the creation of a legal system.151 Sometimes, the promise of equality and nondomination is best protected not by rational, abstract and general rules but by democratic procedures that allow citizens to speak for themselves about what equal liberty means to them.152 By incorporating universality and reflexivity into the process of public deliberation,153 rather than into the law’s substantive content, democratic polities can ensure the place of power remains empty, a place where law – not men – rules.154 But it is a place that can also respond to diversity and difference with particular and specific laws. They need not be general in substance if they are the co-creation of the general population. There is good reason to believe that American citizens and jurists can and do accept a constructivist account of rights. Indeed, they have demonstrated as much by politicizing and modifying their liberties throughout its history. The right to privacy, ostensibly found under the “penumbra” of the Bill of Rights, is one obvious example.155 The history of 148 James Tully, The Unfreedom of the Moderns in Comparison to Their ideals of Constitutional Democracy, 65 MODERN L. REV. 204, 207 (2002). 149 E.g., Urbinati, supra note 146, at 17. 150 E.g., Scheuermann, supra note 129, at 121; Roger Cotterrell, The Rule of Law in Corporate Society: Neumann, Kirchheimer and the Lessons of Weimar, 51 MODERN L. REV. 126, 130 (1988) (for Neumann, the “ethical minimum” promised by the rule of law is a “promise of equality, if only formal equality before the law”); Vile, supra note 116, at 25 (rule of law means, inter alia, “justice for equals”). 151 For example, Waldron, supra note 37, at 153, speaks of the attractiveness of the rule of law – or general legislation – because it purports to impose rule “for reasons” like impartiality and “treating like cases alike” rather than “arbitrarily or on a whim.” “Generality,” he notes, “connotes reciprocity.” These reasons can be an attribute of the law itself – or of the process by which it is made. One example of this “procedural” understanding of the rule of law is provided by Rubin, supra note 139 at 378-8 and Ernst, supra note 114, at 27-29 (describing how Charles Evans Hughes thought that “rule of law” in agencies could be accomplished if they adopted judicial procedures) . 152 E.g., Young, supra note 134. 153 See, e.g., Forst, supra note 134. 154 The phrase “empire of laws and not of men” is credited to James Harrington. PHILIP PETTIT, REPUBLICANISM: A THEORY OF FREEDOM AND GOVERNMENT 173 (2011). 155 JEAN L. COHEN, REGULATING INTIMACY: A NEW LEGAL PARADIGM (2004). ALL THE SOVEREIGN’S AGENTS 23 economic freedom provides another. After a U.S. Supreme Court majority endorsed “natural” property and contract rights for the first time in Allegeyer v. Louisiana,156 these Lochner157-era, laissez-faire economic freedoms crumbled 50 years later in the face of democratic pressure.158 Couched as “vested” and “unenumerated” protections for Lockean “fruits of labor,”159 natural rights theories underwrote judicial awards of Fourth, Fifth and Fourteenth Amendment protections to business corporations.160 But American Progressives successfully challenged this jealous defense of the market.161 By President Roosevelt’s 1944 State of the Union Address,162 a proposed second bill of social rights provided justification for an expanded social welfare state – the state that anti-administravists now challenge. Thus, agencies cannot be dismissed as illegitimate because courts forsake their duty to expertly police agencies’ treatment of citizens’ natural liberties. Courts hold no monopoly on the interpretation of rights because they are not things that they can “discover” as experts on natural law.163 Nor can agency regulation be dismissed as illegitimate violations of the rule of law, because Congress does not create abstract, general, universal impersonal laws whose formal application promises to vindicate equal liberty and nonarbitrariness. It is at least possible that when the EEOC elaborates on the meaning of the 14th Amendment, as articulated by civil rights statutes and informed by social movements and NGOs, it takes part in a systemic process of democratic rights-construction that does, indeed, permit citizens to understand themselves as a co-equal authors of the rights that protect them.164 156 165 U.S. 578, 589 (1897). Supra note 108. 158 Pound, supra note 107, at 457-64. 159 Justice Stephen Field (dissenting), Slaughterhouse Cases, 83 U.S. 36, 90 (1873). For a discussion of the relation between common law and natural right, see Lee, supra note 68 at 289 (in England, the common law is a legal order without a legislator) 160 Bloomer v. McQuewan, 55 U.S. (14 How.) 539, 553 (1852); Dahl, supra note 108, at 63; ADAM WINKLER, WE THE CORPORATIONS: HOW AMERICAN BUSINESS WON THEIR CIVIL RIGHTS 159 (2018); William J. Novak, Legal Origins of the Modern American State, in LOOKING BACK AT LAW’S CENTURY 255 (A. Sarat et al., eds., 2002). 161 MARC A. EISNER, THE AMERICAN POLITICAL ECONOMY: INSTITUTIONAL EVOLUTION OF MARKET AND STATE 34 (2011) (citing the creation of the Federal Reserve Bank and Federal antitrust regulation). 162 Franklin D. Roosevelt, State of the Union (Jan. 11, 1944), http://www.fdrlibrary.marist.edu/archives/stateoftheunion.html 163 James Landis, supra note 30, at 535. Nor, as Waldron (supra note 37, at 222) argues, are they suited to do so. Many of their debates over rights involve constitutional doctrinal matters that are tangential to the major moral issues at play. 164 See Gillian E. Metzger, Administrative Constitutionalism, 91 TEX. L. REV. 1897, 1923 (2013) (agencies may have a better understanding of how laws impact pregnant citizens and 157 24 JACKSON C. The Non-Problem of Separating Powers If these arguments are correct, they challenge a separation of powers doctrine that (1) juxtaposes a sovereign lawmaking body against (2) an executive that robotically carries out its general will, all subject to (3) an expert judiciary that mechanically applies that will while policing it for rights violations.165 The loss of this doctrinal understanding should not be mourned. Certainly, there are good instrumental reasons to separate powers in a system of checks and balances. But commitment to the separation of powers as an end in itself, rather than as a means, is, as Vermeule points out, a kind of idolatry.166 First, the doctrine is a device whose functions have changed as social circumstances and values have changed.167 Notably, it was not always conceived as a way to hamstring government, but instead to empower it. The concept, according to Adam Smith, “seems originally to have risen from the increasing business of society,”168 not so much to limit power as it efficiently divided the labor of governing. It permits collective decision-making to benefit from specialization and “a more efficient distribution and organization of governmental functions.”169 Second, and more recognizably, it is but one of several useful instruments that can help achieve an end: the obstruction of excessive concentrations of power.170 Madison, for example, admitted that it was a mere “auxiliary” device, an “invention of prudence” can therefore better articulate the non-discrimination principles); Emerson, supra note 17, at 13 (Progressives theorized administration in way that contributed to democratic public reason). 165 Neumann, supra note 79, at 166; see also Mortenson & Bagley, supra note 51, at *55*56 (describing how the Founders and their intellectual contemporaries likewise embraced this stylized tripartite structure and analogized the “body politic” with the individual human body) Urbinati, supra note 79, at 758, notes that one of the reasons why we tend to favor direct, participatory democracy to “indirect” representative democracy is because it “entails a fusion between ‘talking’ and ‘doing’ in political action.” (Ibid., 758). Representation necessarily entails a separation between decision and action. But it is important, she argues, not to overstate this separation. 166 Vermuele, supra note 21, at ch. 2. 167 Vile, supra note 116, at 17-18. 168 ADAM SMITH, AN INQUIRY INTO THE NATURE AND CAUSES OF THE WEALTH OF NATIONS 556 (Management Laboratory Press 2008) (1776). 169 S. Holmes, supra note 78, at 164. 170 Cf. Jeremy Waldron, Separation of Powers in Thought and Practice, 54 B.C. L. REV. 433 (2013); but see Waldron, supra note 37, at 50 (itemizing instrumental reasons for separating powers, including, e.g., providing multiple points of contestation for government action and preventing domination by any single individual) ALL THE SOVEREIGN’S AGENTS 25 that could help “control the abuses of government,” and that though it was helpful, elections “[are], no doubt, the primary control on the government.”171 Those committed to the rule of law generally and the Constitution’s articulation of powers specifically may balk at the contingency implicated in this instrumental conception. But as a matter of logic, the functions of government – lawmaking, judging, and executing – cannot be separated.172 Human reason simply cannot be divided up so cleanly. Decision-making, whether performed by a state or an individual, can never be entirely divorced into executive means and legislative ends. Or, put more philosophically, purposive-rational action (instrumental reasoning) and communicative action (moral reasoning) cannot, pace Habermas,173 be so neatly dissected. When we act, we update our ends as we consider the means required to achieve them. We select some ends precisely because they are means for other ends.174 Administrative agents, like individual agents, will inevitably face the same kind of decisions.175 “In politics as in ordinary life,” political theorist Yves Sintomer notes, “judgments and will are part of action, which encompasses them both.”176Any differentiation between legislative, executive, and the adjudicative reasoning, therefore, is “sociologically naïve” at worst; dialogical and endogenous, at best.177 The conceptual elisions only increase when moving from an individual level analysis to the state. Laws, the ends set by the legislature, necessarily contain some amount of abstraction.178 Such abstraction invites an admixture of government functions. The product of compromise and consensus-building amongst elected representatives, they are full of open, inclusive language.179 Even cumbrous spending statutes, weighed down by 171 THE FEDERALIST NO. 51 (James Madison). E.g., Bowsher v. Synar, 478 U.S. 714, 749-752 (1986) (Stevens, J., dissenting). 173 See supra note 6. 174 Adrian Blau, Defending Instrumental Rationality against Critical Theorists, Political Research Quarterly (forthcoming 2020); Hayek, supra note 14, at, 9; Richardson, supra note 29, at 119-127. 175 Richardson, supra note 29, at 118; Dahl, supra note 108, at 3; Caiden, supra note 104, at 64, 70. 176 Yves Sintomer, Nadia Urbinati’s Democracy Disfigured and the Crisis of Real Existing Democracies, 15 CONTEMP. POL. TH. 230, 232 (2016). 177 Ibid. 178 E.g., Waldron, supra note 37, at 138-40 (citing Nadia Urbinati, REPRESENTATIVE DEMOCRACY: PRINCIPLES AND GENEALOGY (2006)); Frank Lovett, A Republican Theory of Adjudication, 21 RES PUBLICA 1, 4 (2015) (citing Wittgenstein). 179 Ibid. For philosophical explanation of law’s generality, see Hayek, supra note 14, at 1112; Vile, supra note 116, at 26; Richardson, supra note 29, at 119; Shklar, supra note 65; H.L.A. HART, THE CONCEPT OF LAW 124 (1961); Rosanvallon, supra note 14, at 86, 135; 172 26 JACKSON log-rolls and pork-barrels, speak in terms of abstract “interest groups” and not specific individuals.180 “No legislature or legislative body,” observed Frank Goodnow, “can express the will of the state as to all matters of human conduct so clearly that no dispute as to its meaning may arise.181 It was a sentiment he shared with James Madison. In The Federalist, Madison admitted that even when laws are “penned with the greatest technical skill,” their meaning would remain “obscure and equivocal, until their meaning be liquidated and ascertained by a series of particular discussions and adjudications.”182 Indeed, Congress must incorporate some vagueness into its legislation if it is to capture the future behavior of innovative business actors. Antitrust law, for example, cannot effectively target future firm behavior if it specifies beyond the “unreasonable” restraint of trade. Adaptable organizations will find a way to skirt rigid rules.183 When they apply relatively abstract rules to address life’s particularities, courts and agencies will inevitably interpret and elaborate.184 They will likewise confront ambiguity when they find themselves applying old law to new and unforeseen cases.185 In such interpretation and elaboration, value-laden policy choices will be made – choices that at least flirt with the legislative, ends-focused reasoning.186 As a result, state organs will employ conflicting canons of interpretation that are ripe for politicization precisely because they rely on values that are not themselves identified within the relevant statute. True, scholars and courts may evade the separation of powers problem by couching all sorts of rulemaking behavior as “non-legislative” because it derives from an “intelligible principle”187 provided by Congressional statute.188 As a result, it’s hoped that neither executive agencies nor the Courts act as lawmakers as they interpret, elaborate upon, Stewart, supra note 32, at 1684; A.L.A. Schecter Poultry Corp. v. U.S., 295 U.S. 495, 52930 (1935). 180 Waldron, supra note 37, at 138-139. 181 Goodnow, supra note 116, at 72. 182 THE FEDERALIST NO. 37; see also Mortenson & Bagley, supra note 51, at *57 (citing ibid.) 183 Neumann, supra note 79, at 172; see also Pettit, supra note 154, at 175-6 (such discretion need not unduly risk arbitrariness if other constraints are available). 184 Neumann, supra note 79, at 171; see also, e.g., Moyn, supra note 22; Metzger, supra note 164. 185 Sunstein & Vermeule, supra note 123, at 111. 186 See, e.g., Leiter, supra note 123; Holmes, Jr., supra note 39. 187 J.W. Hampton, Jr. & Co. v. United States, 276 U.S. 394, 409 (1928). 188 Thomas W. Merrill, Rethinking Article I, Section 1: From Nondelegation to Exclusive Delegation, 104 COLUMBIA L. REV. 2097, 2103-9 (2004). ALL THE SOVEREIGN’S AGENTS 27 and apply that principle. As Mortenson and Bagley point out, though, this move simply substitute’s a judge’s reasonably contestable opinion on which language is sufficiently certain to be “intelligible.”189 “Almost any statute could conceivably flunk a text that mushy,”190 and a court could then step in as lawgiver. At the same time, the legislative character inherent to the application and execution of law is accepted by progressive and conservative legal scholars alike.191 Samuel Moyn observes, for example, that “[a]ny interpretation of law is a form of rule, and there is no way – contrary to what many of the founders believed – of disentangling ‘judgment’ and ‘will.’”192 For Legal Realists, because the law is rationally under-determinate, the available legal reasons to which a court might appeal when justifying its decisions will at least occasionally fail to offer a uniquely correct outcome.193 Thus, although the Constitution may not “enact Mr. Herbert Spencer’s social statics,”194 outside values inevitably wend their way into caselaw. Vermeule, for his part, points out that “the power to fill in the details [of legislation] is an indispensable element of what ‘executive’ power means.”195 None other than Carl Schmitt recognized the legislative power exercised in the moment of the inevitably discretionary interpretation and application of the law. He recognized it and then weaponized it to legitimize a demagogic Nazi executive.196 If execution is political, it should be made by the political actor that speaks with the voice of “the people.” At the other end of the ideological spectrum, libertarian Philip Hamburger notes that “almost all legislation” “must apply to facts that cannot be known at the time of enactment.”197 Unlike Schmitt, though, he insists (without much conceptual argumentation) that the “discernment” required during the law’s application is not an exercise of legislative will. Instead, the acts of expounding and interpreting are mere functions that ought to be allocated to the judiciary.198 Thus, the judicial branch inevitably finds itself engaging in both execution and legislative policymaking. “Before the rise of administration,” 189 Mortenson & Bagley, supra note 51, at *10. Ibid. 191 Arthur S. Miller, The President and Faithful Execution of the Laws, 40 VANDERBILT L. REV. 389, 399 (1987). 192 Supra note 22. 193 Leiter, supra note 123, at 284 (citing Carl Llewellyn, Remarks on the Theory of Appellate Decision, 3 VANDERBILT L. REV. 395, 399 (1950)); Sunstein & Vermuele, supra note 114, at 110 (noting that this critique of the separation of powers correlated with the Realists’ support for an expanded administrative state); see also Goodin, supra note 28, at 238. 194 Lochner, supra note 108, 198 U.S. at 75. (Holmes, J., dissenting). 195 Vermeule, supra note 21, at 53. 196 Schmitt, supra note 19, at 21. 197 Hamburger, supra note 7, at 97. 198 Ibid., at 52. 190 28 JACKSON notes legal historian Daniel Ernst, “the courts had been the nation’s de facto regulators, and they had left to juries the work of applying vague [legislative] standards.”199 For example, judges went about the piece-meal business of utility rate-setting until the task was given to administrators to execute.200 Likewise, Congressional policymaking cannot extricate itself from execution and particularity.201 Else, the U.S. Constitution’s Necessary and Proper clause,202 empowering Congress to create the laws needed to execute its powers, would have no purpose.203 As exemplified by the 300-page CARES Act204 “[m]aking laws [has] become just another means of governing.”205 There’s a good reason why Congress concerns itself with implementation. For any given end, a variety of means will be available to fulfill it, each implicating different values and preferences.206 Nor can Congress quarantine itself from the adjudication of conflicting demands according to higher principle. As Madison noted long ago, much of lawmaking amounts to the resolution of multiparty disputes amongst citizens alleging conflicting claims.207 Bentham, for example, argued that Parliament, like a common law adversary court, hosted opposing arguments while the people themselves would sit as observing judges.208 The laws passed by legislatures, like the opinions of courts, are often elaborations and interpretations of higher, pre-existing principles.209 Indeed, this is the 199 Ernst, supra note 114, at 13. Landis, supra note 30. 201 Rosanvallon, supra note 14, at 57; Goodnow, supra note 116, at 20; Waldron, supra note 37, at 133; Rubin, supra note 139, at 389-90. 202 U.S. Const. art. 1, § 8. 203 McCullough v. Maryland, 17 U.S. 316 (1819); Schechter, 295 U.S. at 529; Lawson, supra note 13, at 346-50; Merrill, supra note 188, at 2129-30 (suggesting that the blending of means and ends within Congressional powers may underpin the Courts’ indifference regarding whether agencies choose to undertake their activities under either rulemaking or adjudication); see Sunstein & Vermuele, supra note 123, at 55. 204 The Coronavirus Aid, Relief, and Economic Security Act, Pub. L. No. 116-13 (2020). 205 Rosanvallon, supra note 14, at 29. 206 Richardson, supra note 29, at 123. 207 THE FEDERALIST NO. 10 (“With equal, nay with greater reason, a body of men are unfit to be both judges and parties at the same time; yet what are many of the most important acts of legislation, but so many judicial determinations, not indeed concerning the rights of single persons, but concerning the rights of large bodies of citizens? And what are the different classes of legislators but advocates and parties to the causes which they determine? Is a law proposed concerning private debts? It is a question to which the creditors are parties on one side and the debtors on the other.”). 208 Jeremy Bentham, Essay on Political Tactics (1791), in 2 THE WORKS OF JEREMY BENTHAM 310 (John Bowring ed., 1865). 209 Vile, supra note 116, at 27; Hamburger, supra note 7, at 79. 200 ALL THE SOVEREIGN’S AGENTS 29 very model of the deliberative assembly. Political theorist Nadia Urbinati, in an important 2000 essay,210 describes elected representatives as advocates, pleading the case of their constituents before a tribunal for their judgment. To be sure, early scholars of U.S. administration believed that politics and administration – will and its execution – could be cleanly separated. Channeling Rousseau and Condorcet, Wilson and his students hoped that administrators could invoke “nomothetic,” or law-like, principles that would permit them to turn administration into some kind of hard science.211 Scholars of political philosophy are also accustomed to the line drawn between instrumental and substantive reasoning implied by a clean separation of powers.212 Yet their hopes are belied by reality. Indeed, some of the most politically contentious debates are debates over means, not ends. For example: how different executives choose to exercise their “utterly ordinary constitutional duty” to enforce immigration legislation.213 As citizens debate the merits and demerits of border walls, child separations, and the deportation of young undocumented immigrants, Congress may very well find itself considering new legislation delimiting the tools available to the Presidency. No one can reasonably doubt the political contentiousness of tax cuts and deficit spending, yet both are means to achieve the same result: economic growth.214 Electoral politics, like agency politics, likewise finds itself grappling with “the battle of the experts.” Citizens commonly clash not over their final ends, but how they are best achieved: how to grow the economy, lift people from poverty, decrease crime.215 Meanwhile, political associations fail if they fail to find the means that can meet challenges to their security and capacity.216 Elected leaders find themselves negotiating the constraints of international financial markets and capital flows.217 As Dahl notes, democracy, if it is to make intelligent choices 210 Supra note 79, at 773. Caiden, supra note 104, at 51. 212 E.g., Habermas, supra note 6, at 85 (arguing that instrumental reasoning is ill-suited for democratic discourse because it “is immanently [a discourse of] of control.”). 213 Cox & Rodriguez, supra note 20, at 2. 214 STEPHEN TURNER, LIBERAL DEMOCRACY 3.0: CIVIL SOCIETY IN THE AGE OF EXPERTS 4 (2003). 215 Ibid.; Peter Vervosek, Public intellectuals and experts cannot tell citizens what to do, LSEBLOG (JAN. 28, 2020). 216 See ALRED O. HIRSCHMAN, EXIT, VOICE AND LOYALTY: RESPONSES TO DECLINE IN FIRMS, ORGANIZATIONS, AND STATES (1970). 217 Streeck, supra note 6; John Pocock, The Political Limits of Premodern Economics, in THE ECONOMIC LIMITS OF MODERN POLITICS 121 (1990). 211 30 JACKSON in a complex society, “require[s] both technical understanding and sensitivity to the values involved.”218 As a result, those that attack independent administrative agencies as usurping powers properly belonging to the judicial, executive or legislative branch often make a category mistake. None of these branches can possibly enjoy a monopoly on the functions of legislation, adjudication, and execution. Folding administration into any one of these branches will inevitably create another separation-of-powers problem as they will bring with them all the powers they presently wield. Giving the President more control over the Environmental Protection Agency, for example, means giving him power to promulgate pollution regulation and adjudicate superfund claims at the expense of Congress and the Courts. Folding agencies into the Courts by hamstringing their discretion merely swaps some bureaucrats for others.219 It’s just that judges will be making the policy instead of civil servants. Rather, a critique based upon the separation of powers should derive from the important purposes that it is meant to fulfill: non-domination, accountability, efficiency, etc. And, indeed, agencies have responded to such critiques by, for example, adopting the due process and accountability norms set forth in the APA.220 Independent, professionalized and capable agencies both curb arbitrary Presidential power and enable the state’s constitutional obligation to govern.221 Arguably, it is the independence of agencies against Presidential power that curbed many of the Trump Administration’s abuses.222 In the next section, I elaborate on another norm that can be used to assess the legitimacy of agencies: democratic representation. By offering additional fora for representative input and contestation, they help a political regime better vindicate the promise of popular sovereignty: that citizens can play an equal part in forming the laws that govern them. PART II. AGENCIES AND THE REPRESENTATIVE TURN With a better understanding of popular sovereignty, rights and the separation of powers, critics of the administrative state cannot make a failsafe argument against agencies’ legitimacy. Simply, they cannot rely on agencies’ lack of congressional input and immunity from judicial review. Here, I set forth not 218 ROBERT DAHL, SOCIAL REALITY AND FREE MARKETS 227 (1990). Sunstein & Vermuele, supra note 123, at 37; Landis, supra note 30. 220 Supra note 55. 221 Metzger, supra note 8, at 7. 222 Tim Wu, What Really Saved the Republic From Trump? NEW YORK TIMES, Dec. 10, 2020. 219 ALL THE SOVEREIGN’S AGENTS 31 a negative defense of administration. Instead, I make out a positive case. Namely, if agencies operate according to procedures that permit citizens to understand themselves as co-equal political participants, it is at least possible that they are compatible with constitutional liberal democracy. Recent conceptual work in political representation bears out this possibility. It also shows how agencies might improve citizens’ ability to participate equally in collective decision-making. As a result, administration may not be the Constitution’s embarrassing uncle after all. It might instead prove indispensable to the fulfillment of constitutional democracy’s promises. A. Defining the Turn Democratic representation implies difference. If “the people” actually existed as a homogenous whole with an identifiable will, there’d be no need for a multi-member mediating body of partisans.223 A single demagogue would do.224 If “the common good” or “natural right” were easily ascertained, there’d be no need for a legislature to argue about it incessantly. A philosopher king would serve more easily.225 Indeed, organ-body sovereignty, natural right and the “rule of general law” imply a utopianism that can be vindicated in a plural society only through force. Accordingly, if there are any theories of democratic political legitimacy that might avoid these conceptual blunders, it would likely be found within theories of democratic representation. And so there are. The representative, or “constructivist,” turn within political theory226 pushed back against the idea that representative democracy 223 E.g., Rosenblum, supra note 72, at 6. Carl Schmitt, THE CONCEPT OF THE POLITICAL 46-48 (G. Schwab Trans. 2007); Carl Schmitt, supra note 77, at 8-17, 28-32; Carl Schmitt, CONSTITUTIONAL THEORY 264-267 (J. Seitzer trans., 2008). 225 Urbinati, supra note 79, at 774 (citing J.S. Mill, Rationale of Representation, in ESSAYS IN POLITICS AND SOCIETY 16-45, 22-23 (1977)); Rosenblum, supra note 72, at 11 (noting, inter alia, that a “party to end parties is the pose of utopian and revolutionary parties” like Jacobins and Communists), 27-28 (Holism is antipolitical, and holist utopias make this vivid. In every perfectionist community, the absence of political institutions is the ideal…Thus philosophical creators of utopia prescribe anarchy or technical superintendence…Two holist assumptions dictate aversion to politics and therefore to parties: first, that there is an identifiable common good or good of the whole and, second, that no political recognition and arrangement of parts, no dynamic of cooperation, and certainly no dialectic of conflict can illuminate the common good…”). 226 The inception of this turn is generally credited to Jane Mansbridge, Rethinking Representation, 97 AM. POL. SCI. REV. 515 (2003); Andrew Rehfeld, Towards a General Theory of Political Representation, 68 J. OF POL. 1 (2006); Michael Saward, The Representative Claim, 5 CONTEMP. POL. THEORY 297 (2006); and Urbinati, supra note 79. 224 32 JACKSON is merely a watered-down version of direct democracy, a pragmatic or elitist alternative to enlist only when the demos is too unwieldy or too ignorant for more direct institutions.227 For a long time, democratic theorists like James Madison, Joseph Schumpeter and Giovanni Sartori held that elections were a second-best way of ascertaining the “will of the people,” adopted only because incorporating the will of each and every citizen into the lawmaking enterprise is impossible or undesirable (both practically and perhaps normatively).228 Rather than the direct, deliberative, participatory democracy of the ancients, we moderns must sadly settle for a principle-agent model and a political division of labor that will inevitably involve some kind of aristocracy.229 In contrast, the new theories of representation hold that representation is no mediocre substitute for democracy, but instead “an intrinsically modern way of intertwining participation, political judgment, and the constitution of demoi capable of self rule.”230 Eschewing organ-body sovereignty, they do not judge representation according to how well it reproduces a non-existent, pre-political popular will.231 They reject the common-sense notion that representatives take exogenous citizen preferences as the “bedrock for social choice” in a linear bottom-up process that transmits those preferences into law.232 Incorporating empirical research showing that citizens’ opinions are rarely stable outside the context of electoral politics233, they instead hold that representatives help generate constituencies and voter demands through the educational process of partisan advocacy.234 “Parties create, not just reflect, political interests and opinions. They formulate ‘issues’ and given them political relevance.”235 Often, they will distinguish a group, paint it in a 227 E.g., Waldron, supra note 37, at 134-5; Urbinati, supra note 79, at 758-759. Urbinati, supra note 72, at 196; see also, e.g., John Austin, THE PROVINCE OF JURISPRUDENCE DETERMINED 192 (Wilfrid E. Rumble ed. 1995) (1832) (“Where a sovereign body…exercises through representatives the whole of its sovereign powers…It may delegate those its powers to those its representatives, subject to a trust or trusts…”) 229 See generally Bernard Manin, THE PRINCIPLES OF REPRESENTATIVE GOVERNMENT 132 (1997); Benjamin Constant, THE LIBERTY OF THE ANCIENTS COMPARED WITH THAT OF THE MODERNS (1819). 230 Urbinati & Warren, supra note 46, at 402. 231 HANNAH PITKIN, THE CONCEPT OF REPRESENTATION 221 (1967); Saward, supra note 226; Urbinati, supra note 72, at 197 (citing Sieyes, who held that the “sovereign nation was politically mute outside the electoral booth and [that] its will [is] inexpressible without and outside the representative assembly.”). 232 Lisa Disch, Towards a Mobilization Conception of Democratic Representation, 105 AM. POL. SCI. REV. 100, 101 (2011). 233 Disch, supra note 78, at 1, 16. 234 Ibid. (relying on Laclau and Mouffe’s insights about the symbolic construction of the “people”); Urbinati, supra note 79, at 776. 235 Rosenblum, supra note 72, at 7. 228 ALL THE SOVEREIGN’S AGENTS 33 certain light, and then make claims on its behalf. Group members will then either accept or reject those claims based upon their own views236 and in light of competing messages from other representatives in an ongoing practice of judgment-formation and public deliberation.237 Elections thus do not ham-handedly measure any primordial democratic “sovereign” will.238 Nor are they an imperfect, real-life reproduction of an idealized social contract whereby citizens expressly “consent” to the laws that bind them.239 Rather, they structure participation, help constitute political debate, and “make citizens participate in the game of ridding themselves of government.”240 Representatives do not simply hew to past campaign promises. They also anticipate and frame voters’ future judgments, sometimes on brand new issues.241 The laws they pass are the outcome of both formal and informal decision-making procedures that “actually creates a framework in which the nation can for the first time have a will.”242 For example, elections set agendas for public discussion, motivate organizing by providing discrete targets and timelines, institutionalize 236 There is a debate in the literature regarding the exogeneity of voter preferences. Some theorists hold that voters possess pre-political interests that representative politics does not create, but rather simply activates. Some deliberative theorists hold that Rawlsian, consensus-based public discourse motored by, but somewhat independent from, representative politics can take the place of pre-political voter interests with their model. Others, refusing to hitch voter preferences to any outside source, are susceptible to critiques of relativism and the impossibility of distinguishing between autonomous citizen choices and elite manipulation. I am attempting to bracket the issue. If translated through representative claims, objective voter interests and exogenous preferences are captured in the theory. If they are not, they suggest that political legitimacy is based not on democratic autonomy, but on, e.g., objective notions of justice. 237 Saward, supra note 226; Lisa Disch, The Constructivist Turn in Democratic Representation: A Normative Dead-End? 22 CONSTELLATIONS 487 (2015); Urbinati, supra note 79, at 760, 766. 238 Monica B. Vieira, Representing Silence in Politics, 114 AM. POL. SCI. REV. 976 (2020); Urbinati, supra note 79, at 765. 239 See discussion of majority voting, infra. Waldron, supra note 37, describes the vote as not some real-life consent theory of government, but as a fair process that lends legitimacy to outcomes. 240 Urbinati, supra note 72, at 197 (emphasis added). 241 Mansbridge, supra note 226; Urbinati, supra note 79, at 760. 242 S. Holmes, supra note 78, at 164; see also Waldron, supra note 78, at 277; Disch, supra note 78, at 10; Habermas, supra note 57, at 171 (representative legislative decision-making must be “anchored” in the informal streams of communication emerging from public spheres”). 34 JACKSON “countervailing discourses” through political competition243 and thus encourage citizens to exercise thoughtful judgments over matters of public concern.244 They frame continuous, recursive processes of bottom-up demands and representatives’ top-down reactions, permitting citizens to influence lawmaking.245 As a result, electoral politics “activate” citizens beyond the formal act of casting a ballot and prevent public discourse from disintegrating into a cacophony of interventions that remain unchanneled towards any common goal.246 Representatives’ lack of constituent-specific imperative mandate challenges voters to think of themselves as part of a national whole: generally, and in terms of the public good – or, at least, in terms of the abstract interests they purport to represent.247 At the same time, representatives’ accountability to particular constituencies, along with bicameralism, (1) reinforces the notion that there is no permanent, collective “whole” that speaks with the voice of the sovereign “people”248and (2) ensures that “the specific position of the individual citizen is not ignored.”249Meanwhile, elected representatives take these ongoing discourses and demands with them as they undertake their lawmaking responsibilities as advocates, enabling their constituents to be “present” in decision-making through their discourses, if not their physical bodies.250 Juridically and formally, the vote is not a device that, however clumsily, aggregates voter preferences in order to translate them into statutes. Instead, it serves at least two purposes important for political equality. First, at least when compared to acclamation, it encourages individual citizens to, with the aid of public discourse, form their own individual judgments over 243 John S. Dryzek and Simon Niemeyer, Discursive Representation, 102 AM. POL. SCI. REV. 481 (2008); Rosenblum, supra note 72, at 7. 244 E.g., Urbinati, supra note 17, at 184; Waldron, supra note 37, at 141-142 (“structures of representation provide processes for judgment formation and for the deliberative engagement of judgments both among the people and among their representatives”) (citing Urbinati, supra note 79, at 219-220) (comparing elections favorably to judicial decision-making in relation to stimulating meaningful political discourse), 250 (explaining how majority voting is incorporated into deliberative bodies, including those of appellate courts); Urbinati, supra note 79, at 768 (the separation of time and space between election campaigns and actual lawmaking “gives [voters] the chance to reflect by themselves…to defer their judgment” and “enables a critical scrutiny while shielding citizens from the harassment of words and passions” that encourages demagoguery. 245 Jane Mansbridge, Recursive Representation in the Representative System (Harv. Kenney Sch., Working Paper No. RWP17-045, 2017); Urbinati, supra note 79, at 766. 246 Urbinati & Warren, supra note 46, at 391-2; Habermas, supra note 57. 247 Urbinati, supra note 72, at 210; Urbinati, supra note 79, at 761, 764; Waldron, supra note 37, at 136. 248 Urbinati, supra note 72, at 210.; Waldron, supra note 37, at 80. 249 Urbinati, supra note 79, at 770. 250 Ibid., at 767. ALL THE SOVEREIGN’S AGENTS 35 public affairs.251 Each goes alone into the ballot box and each decides which lever to pull. Second, it provides a “temporary resolution of political conflict”252 in a manner that respects citizens as moral and political equals. Majority voting is a decision-making mechanism that promises that even if we agree neither to the outcome of our elections, nor with the lawmaking that follows, we can at least believe that lawmaking was undertaken on fair terms. It is a way, to use Jeremy Waldron’s language, to “give each person the greatest say possible compatible with an equal say for each of the others.”253 Though the universal franchise is important for maintaining a system of political equality and, importantly, legitimacy,254 it is but “one particular moment in a much larger decision-making process.”255 We should not, therefore judge representation and other political institutions according to their congruence with voter preferences and public opinion polls, as if such preferences and polls truly reflect the timeless and enduring “will of the people.” The critical yardstick is not “transcribe the will of the people” by having people, literally, transcribe their will into the statute books. The principle of justification (e.g., popular sovereignty) and the technique of decision (direct, unmediated consensus-based democracy) do not have to overlap.256 Instead, we should judge representative systems according to how well they allow morally equal human beings to affect, in a fair and equal way, the laws that bind them – a fundamental norm that political theorists Nadia Urbinati and Mark Warren call “democratic autonomy.”257 There is, as a result, a lot of room for creativity when it comes to the design of representative institutions.258 Theorists tend to assess these institutions according to whether and what extent they: (1) stimulate public debate and individual judgment, such that a choice to support what is being done in citizens’ names is sufficiently considered to amount to their 251 Ibid., at 765. Urbinati, supra note 72, at 198. 253 Waldron, supra note 37, at 228. 254 Ibid., at 226 (noting that political legitimacy responds to questions of fairness rather than to questions regarding the justice or “rightness” of particular political decisions – perhaps including those that are “right” precisely because they reflect the “general will.”). 255 Sintomer, supra note 176, at 232. 256 Rosanvallon, supra note 14, at 111; cf. Montesquieu, supra note 74, at 11-12 (arguing that the people can be the sovereign “monarch only through their wills. The [popular] sovereign’s will is the sovereign himself. Therefore, the laws establishing the right to vote are fundamental in this government.”). 257 Supra note 46, at 395; see also Waldron, supra note 37, at 228. 258 Waldron, supra note [37], at 78 (Because there is no such unitary thing as ‘the people,’ there can be no single canonical way of representing it.”). 252 36 JACKSON participation in the lawmaking process259; (2) motivate bottom-up demands that are followed by top-down responses (“recursiveness” or “reflexivity”)260 in a manner that orients representatives to the normative – if not ontological – priority of the represented261 by, for example (3) permitting and mobilizing contestation;262 (4) ensuring citizens have some kind of (equal) influence on decision-making, perhaps with some process of electoral authorization that shows citizen-uptake of agency policies,263 or formal registration and/or accommodation of citizen objections. Each element serves democratic autonomy. Each demonstrates the compatibility of this understanding of representation with a systemic, agonistic conception of deliberative democracy.264 Importantly, each can be and often are served by non-electoral forms of representation.265 To provide some obvious examples: lobbyist, nonprofit public advocacy corporations, and NGOs.266 B. Accommodating Administration The same institutional flexibility that accommodates representative politics within a constitutional liberal democracy can likewise accommodate administrative governance. Indeed, those who attack agencies as illegitimate lawmakers mirror the claims that Rousseau levied against democratic representation itself. For Rousseau, the (popular) sovereign could not be represented. Instead, the law must rise directly from the people.267 Representation, therefore, should be confined to a strict principal-agent model, a fiduciary contract whereby the agent is stripped of any policymaking discretion whatsoever.268 Government is to blindly obey the sovereign’s political orders; representatives and other “ministerial,” 259 Disch, supra note 232, at 107 (versus choices that primed out of habit, ignorance, or stereotype); Urbinati, supra note 72; Samuel Hayat, Representation as Proposition: Democratic Representation After the Constructivist Turn, in THE CONSTRUCTIVIST TURN IN POLITICAL REPRESENTATION (Lisa Disch et al. eds., 2019). 260 Mansbridge, supra note 245; Disch, supra note 232, at 111. 261 Vieira, supra note 238; Pitkin, supra note 231, at 140; Disch, supra note 232, at 108; Waldron, supra note 37, at 188, 191; Urbinati & Warren, supra note 46, at 399-400. 262 Disch, supra note 232, at 207; Viera, supra note 238; David Runciman, The Paradox of Political Representation, 15 J. POL. PHIL. 93, 95 (2007); Laura Montanaro, The Democratic Legitimacy of Self-Appointed Representatives, 74 J. OF POL. 1094 (2012). 263 Vieira, supra note 238. 264 Urbinati, supra note 79, at 773-74. 265 Saward, supra note 226. 266 Mansbridge, supra note 226, at 517. 267 Rousseau, supra note 71, at 229-30. 268 Rosanvallon, supra note 14, at 135. ALL THE SOVEREIGN’S AGENTS 37 “fiduciary” state offers could only judge and interpret – presumably in a way that did not usurp the sovereign’s will.269 Echoing Condorcet (administration is a “roi-machine”) and Sieyès (administration is an “intermediary commission of powers,”)270 many contemporary anti-administrativists understand agencies as the incurious fiduciaries of Congress – itself a fiduciary of the people271 – and so should, “like a machine without a mind,”272 avoid playing any role in the formation of law and policy.273 They mimic Woodrow Wilson’s274 pleas to depoliticize bureaucracy and Frank Goodnow’s separation of the “organism” of the state into two distinct operations: those “necessary to the expression of its will” and those “necessary to the execution of its will.” 275 Likewise, the post-war consensus assumed that agencies should serve simply as a “transmission belt” for legislative policy.276 Administration must therefore “appear nonpartisan, scientific, universal, efficient, purposeful.”277 For if administration starts making its own decisions, it can only interfere with the democratic sovereign – and thus, like Rousseau’s representatives, breach their fiduciary contract. As a result, when it became apparent that administration and politics cannot 269 Urbinati, supra note 72, at 203; Sintomer, supra note 176, at 232; Locke, supra note 64, at 367, 369. 270 Rosanvallon, supra note 14, at 29. 271 Mortenson & Bagley, supra note 51, at * 33-*34, 63, argue that the Founders could not have believed that legislative power was non-delegable because it was thought, at the time, that that power was, in the first instance, delegated from “the people” to their elected representatives (understood variously as delegatees, servants, or trustees). 272 Quim Brugue & Raquel A. Gellego, A Democratic Public Administration? 5 PUB. MGMT. REV. 425, 426 (2003). 273 This is the “agency instrumentalist” view (Richardson, supra note 29, at 119) adopted by political theorists including, e.g., Christiano and Goodin, supra note 29; PAUL TUCKER, UNELECTED POWER: THE QUEST FOR LEGITIMACY IN CENTRAL BANKING AND THE REGULATORY STATE (2018); see also, e.g., Mansbridge, supra note 245, at 21; Posner & Vermuele, supra note 14; Turner, supra note 214, at 14; Donald W. Smithburg, Political Theory and Public Administration, 13 J. OF POL. 59, 60 (1951); Cox & Rodriguez, supra note 20, at 11 (summarizing Justice Kennedy’s critique that immigration regulation is “upside down,” subverting a separation of powers doctrine in which the legislative makes laws and the executive, “as a dutiful agent,” enforces them); Youngstown Sheet & Tube Co.. supra note 12, 343 U.S. at 682, 690-91 (C.J. Vinson, dissenting) (arguing that the majority mistakenly treats the executive not as a co-equal branch of government, but as an “automaton” and an “agent” of Congress, which is “enthroned in authority” over the President); Gerry L. Mashaw, Norms, Practices, and the Paradox of Deference: A Preliminary Inquiry into Agency Statutory Interpretation, 567 ADMIN. L. REV. 501, 505 (2005) (agencies are to be the “faithful agents” of the legislature). 274 Supra note 58. 275 Supra, note 116, at 22. 276 Livermore & Richardson, supra note 18, at 28; Stewart, supra note 32, at 1675. 277 Caiden, supra note 104, at 53. 38 JACKSON be cleanly separated in any complex legal regime,278 administration seemed to lose much of its legitimacy. Thus, the attempts to fold agency decisionmaking into a more unified executive or a more active Congress. Yet if representation is not an inferior re-presentation of the will of an embodied popular sovereign, but instead an institutional tool that helps citizens understand themselves as co-equal participants in lawmaking, then there is little reason to exclude agencies ex ante from the democratic process.279 They, like formal electoral representation, can be a part of a system of democratic autonomy, a political order that promises to provide everyone the greatest say possible that is also compatible with the equal inclusion of others. This insight is not unprecedented. French philosopher Pierre Rosanvallon argues280 that independent agencies can and do serve as representative agents (of the trusteeship variety281) and provide the neutrality and impersonality that republicans once sought from deliberative legislative bodies. Richard Stewart, in his seminal 1975 essay,282 described in U.S. administrative law a turn towards pluralist interest representation and away rigid applications of the separation of powers doctrine. The interest-group politics of administrative policymaking can, like the interest-group politics of Congressional lawmaking, provide “opportunities for policy proposals to be criticized from a variety of directions, both before and after their implementation.”283 For Stewart, interest group representation might even serve as a new ground for agencies’ political legitimacy.284 Indeed, the legitimacy ascribed through representation seems to be embraced by agencies themselves. In its own educational literature, the “politically independent” U.S. Federal Reserve System (“The Fed”) asserts that “decentralizing the central bank into twelve districts helped to ensure more voices were represented”285 as it went about its business. 278 See, e.g., Waldo, supra note 21, at 91; Livermore & Richardson, supra note 18, at 3. Smithburg, supra note 273, at 66. 280 PIERRE ROSANVALLON, DEMOCRATIC LEGITIMACY 75-86 (2011). 281 See also Caiden, supra note 104, at 55 (describing Woodrow Wilson’s plea to transform administration into a “public trust”). 282 Supra note 32. 283 Dryzek & Niemeyer, supra note 243, at 482. 284 J. Benton Heath, From the Spirit of the Federalist Papers to the End of Legitimacy: Reflections on Gundy v. United States, 114 NW. U. L. REV. ONLINE 1723, 1732 (2020). 285 Federal Reserve Bank of New York Communications and Outreach, “The Story of the Federal Reserve,” 6 (2018) (available at https://www.newyorkfed.org/outreach-andeducation/comic-books). 279 ALL THE SOVEREIGN’S AGENTS 39 Of course, Theodore Lowi, in his famous 1969 study of interest group representation in administration, did not celebrate it. He lamented its “irrational” incrementalism and the influence it lent to powerful industry actors.286 In contrast, Noh points out that agencies enjoy a better capacity than Congress to achieve ideal representative deliberation, given their smaller constituencies and limited subject matter.287 With the concurrent rise in public interest advocacy and civil rights legislation, agencies accommodated a growing array of individual rights by incorporating rulemaking procedures that accept input from diverse civil society representatives.288 More recently, Emerson,289 Rahman & Gilman290 and Mansbridge291 argue that there is an ongoing, constitutive relationship between administrative agencies and citizens as they collaborate to make regulations. Mary E. Guy, past president of the American Society for Public Administration, characterized the people serving public agencies as “facilitators, interpreters, and mediators of public action”292 – a characterization that could just as easily be attributed to elected representatives. Even Locke, despite his commitment to an organ-body conception of popular sovereignty, theorized a connection between the executive and representation. In his theory of government, it is up to the Prince to create new representative offices if time and social change corrupts the representativeness of extant lawmaking institutions.293 There is a stiff backbone of principle behind these authors’ arguments. As mentioned above, representative systems serve democracy by structuring public contestation, stimulating debate and provoking judgment. They thereby permit disaggregated and diverse citizens to participate in the process of forming collective purposes. One can observe this dynamic within agency decision-making, demonstrated by public reactions following recent U.S. Executive Orders regarding immigration, environmental deregulation, etc. They mobilize objections, enabling citizens to make judgments about agency responses. The public criticism to which agencies are routinely subject are sufficient to limit, at least according to Vermeule and Posner, 286 Supra note 6, at 239-42. Notably, and embarrassingly for Lowi, interest-group bargaining made the welfare state irrational precisely because it brought issues of social justice (like race and gender) into class-based economic policy. Ibid. at 255-45. 287 Supra note 49, at 400. 288 Stewart, supra note 57, at 441-2. 289 Supra note 17, at 65. 290 K. SABEEL RAHMAN & HOLLIE R. GILMAN, CIVIC POWER (2019). 291 Supra note 245. 292 Supra note 63, at 652. 293 Supra note 64, at 372-4. 40 JACKSON executive agencies far more effectively than do the tenets of statutory delegation and the separation of powers.294 Moreover, agencies serve as fora for the kind of non-electoral political representation that supplements political equality by calling forth previously ignored or silent constituencies.295 In the past, minority communities excluded from electoral politics fruitfully engaged with administrators tasked with implementing social welfare rights.296 Administrative experiments in participatory budgeting and collaborative governance encourage citizens’ engagement in the infrastructure shaping their lives.297 They can, in Waldron’s terms, “secur[e] multiple points of access for citizen input” and thus add to the “housing” of the kind of public deliberation that promises to secure democratic autonomy.298 Indeed, administration can, perhaps better than Congress, provide a forum where “conflicts and rendered both comprehensible and solvable” and where the values at stake are sufficiently “specific and understandable to generate opinions and dialogue.”299 Because an elected representative will make a variety of unrelated claims regarding lawmaking in a generalized arena (Congress), the potential exercise of political power can become obscure. In contrast, an agency with a particular subject matter jurisdiction can attract representatives and generate claims that target the specific policies that impact citizens’ lives. They give rise to “affected constituencies,” providing “a point of identification around which [citizens] might coalesce as a ‘people’ or demos defined along some dimensions of common interest.”300 Civil society and public interest representatives, for example, have a better opportunity to exercise judgment on and lodge objections to, e.g., how much pollution is permitted in drinking water (EPA); whether discrimination claims are adequately enforced (Department of Justice); and how policing is executed (“citizen audits” of local departments). Indeed, many social movements target their organizing around agency 294 Supra note 14. Vieira, supra note 238; Rahman & Gilman, supra note 290, at 137; Montanaro, supra note 262; Raul Carillo, Our Money Where Our Mouth Is, DISSENT, Aug. 20, 2020. Notably, Urbinati, supra note 79, at 762-765, notes that even purportedly “direct” Athenian democracy involved non-electoral representation. Though each citizen could come to the forum, inevitably only a few would speak, and they would speak on behalf of interest groups. 296 Karen Tani, States’ Rights, Welfare Rights, and the Indian Problem: Negotiating Citizenship and Sovereignty, 331 LAW & HIST. REV. 1 (2015). 297 Rahman & Gilman, supra note 290, at 134, 172. 298 Waldron, supra note 37, at 34. 299 Brugue & Gellego, supra note 272, at 429. 300 Montanaro, supra note 262, at 1099. 295 ALL THE SOVEREIGN’S AGENTS 41 jurisdictions.301 In other words, agencies provide an opportunity for citizens to form the kind of considered political judgments that new theories of representation value. Indeed, it is agencies that can provide that factual information necessary for citizens seeking to exercise judgment on many matters of public concern.302 One need only consider the usefulness of CDC scientists to democratic deliberation during a pandemic. As a result, agencies, like electoral politics, facilitate the “circuation of judgment and opinion that should unite state institutions and the citizens.”303 Furthermore, agencies themselves serve as representatives of underserved consumer and minority interests.304 For example, the Consumer Financial Protection Bureau (“CFPB”), created in response to the capture and fragmentation of financial regulators in the aftermath of the 2008 crisis, was tasked to protect the consumers of financial products. Typically, consumers are poorly situated to influence banking regulation through normal channels. Accordingly, the CFPB serves not as a neutral bureaucracy, but as consumers’ “proxy advocate” (representative!) that solicits the input of veterans, students and pensioners.305 Meanwhile, financial institutions receive (more than) equal representation in other policy-making locations. The CFPB both educates and solicits the opinions and complaints of those who do not normally possess the capacity to make their voices heard when lawmakers turn their attention to high finance. Indeed, the Progressives who supported New Deal agencies created them precisely to balance public might against the already over-represented “aristocracy of wealth.”306 An agency, serving as a counter-power capable of breaking up monopolies or safeguarding labor rights,307 can speak for the consumers and workers neglected by elected politicians.308 President Franklin D. Roosevelt, in a 1932 301 Rahman & Gilman, supra note 290. E.g., Tucker, supra note 273, at 94. 303 Urbinati, supra note 72, at 198. 304 Stewart, supra note 32, at 1761; see also David B. Truman, THE GOVERNMENTAL PROCESS: POLITICAL INTERESTS AND PUBLIC OPINION 449 (1951); Skocpol & Feingold, supra note 32, at 260 (arguing that the state successfully beat back demands of capitalists in the interests of labor during the New Deal) 305 Rahman & Gilman, supra note 290, at 146. 306 Landis, supra note 30; Joseph Fishkin & William Forbath, The Anti-Oligarchy Constitution, 94 BOS. U. L. REV. 669, 677 (2014). 307 JAMES K. GALBRAITH, AMERICAN CAPITALISM (Routledge, 1st ed. 2017)(1952); K. SABEEL RAHMAN, DEMOCRACY AGAINST DOMIINATION (2016); see also Vermeule, supra note 21, at 62. 308 This departure from procedural equality mirrors much recent work on anti-oligarchy constitutionalism. See, e.g., JEFFREY EDWARD GREEN, THE SHADOW OF UNFAIRNESS: A PLEBIAN THEORY OF LIBERAL DEMOCRACY (2016); JOHN P. MCCORMICK, MACHIAVELLIAN DEMOCRACY (2011). 302 42 JACKSON campaign speech, expressed this idea in colorful language. He dubbed regulators the “Tribune[s] of the people.”309 During Reconstruction, furthermore, federal agencies spoke for those recently freed from slavery when local governments did not.310 C. Implications The lesson offered by the representative turn within political theory is that agencies should be democratized. The goal, however, is not to transform administrative bodies into deliberative mini-publics. Instead, it is to take advantage of the strengths of representative systems: effective decision-making in a manner that respects the normative priority of the represented while remaining sensitive to individuals’ equal political and social dignity. Assessing administration in terms of democratic autonomy can allow us to understand problematic agency decisions in a new and productive light. In this section, I will gesture at how such critiques and reforms might look like. 1. Non-Delegation as a Problem of Equality of Representation Many critiques of administrative discretion are perhaps more aptly described as problems of representativeness than as impermissible delegations of legislative power. In Schecter,311 for example, the U.S. Supreme Court held that the National Industrial Recovery Act improperly delegated legislative power to the Executive. At the same time, the Act, despite goals of incorporating labor and consumer interests, became captured by business leaders through: the direct appointment of industry officers to rulemaking bodies; the non-transparency of business operations; and the state’s reliance on industry organizations to implement any new rules.312 As a result, the Early New Deal is objectionable not (just?) because of broad and vague statutory mandates. It is objectionable because of its lack of democratic representativeness. Many citizens did not enjoy an equal chance to affect regulation because their representatives were shut out of the process. The U.S. Supreme Court’s recent opinion in Seila Law LLC v. Consumer Financial Protection Bureau313 faces similar critique. Giving the 309 https://www.presidency.ucsb.edu/documents/campaign-address-portland-oregon-publicutilities-and-development-hydro-electric-power 310 MATTHEW HOLDEN, JR., CONTINUITY AND DISRUPTIONS: ESSAYS IN PUBLIC ADMINISTRATION 225 (1996) 311 Supra note 179. 312 Skocpol & Finegold, supra note 32, at 265. 313 591 U.S. __ (2020). ALL THE SOVEREIGN’S AGENTS 43 President the power to fire the CFPB’s chief without cause deprives the agency of independence from those already well-positioned to influence financial law, viz., those with access to the Presidential ear. It is a problem likewise echoed during the late 18th century Post Roads Debate.314 Certain congressional representatives balked at assigning the President broad discretion to establish the locations of new postal roads and offices. Such discretion would give the President “a dangerous power of establishing offices and roads in those places only where his interest would be promoted.”315 The concern was not over presidential power as such, but that the President would remain deaf to the concerns of others316 whom his policies might affect. Agency failures to achieve equal representation include not just the lack of formal inclusion of interests in decision-making. Within agency decision-making procedures, inequalities in wealth and experience may reproduce the same distortions of political equality that disfigure electoral politics.317 Agencies are accused of capture for good reason. Given lobbyists’ outsized role in rulemaking and the self-appointment of interest group leaders, it is at best unclear whether agencies remain oriented towards the normative priority of the represented in an equal way.318 It therefore may make sense to, where practicable, decentralize policymaking authority, establish countervailing agencies as points of access, and to populate agency boards with both citizen and public representatives in structures of collaborative governance.319 If constituents are granted formal rights of authorization (e.g., to elect such representatives), they can ensure agencies orient themselves in the right direction.320 Further, staffing offices with representatives drawn from underserved constituencies may further the kind of descriptive representation that encourages mutual education, authorization and uptake.321 They might take on employees whose internal motivations appear consistent with citizens’ judgments through a version of gyroscopic 314 For a summary of this debate, see Mortenson & Bagley, supra note 51, at *96-106. 3 Annals of Congress 235 (1791) (statement of Rep. Vining). 316 Mortenson & Bagley, supra note 51, at *106, point out that these “others” included those that lived at Montpelier and Monticello, who, despite low population numbers, demanded and received a postal road. 317 Emerson, supra note 17; Gilman & Rahman, supra note 290, at 119; Lowi, surpa note 6; Rosanvallon, supra note 14, at 135. 318 Stewart, supra note 32. 319 Rahman & Gilman, supra note 290, at 138-9. 320 Ibid., at 132-35. 321 Young, supra note 134. 315 44 JACKSON representation.322 For example, an agency tasked with consumer financial protection could draw its staff from consumer advocacy groups. Agencies can also play a role in improving democratic representativeness more broadly by helping to create new constituencies from those previously excluded. Electoral politics privileges citizen voice over silence. They therefore privilege those with more cultural and material resources.323 But agencies can be a bit more attentive to those left out of the traditional campaign circus. Recently, for example, the mayoralties of New York City and Boston adopted on-the-ground, campaign style outreach programs that created new, under-served constituencies around enrollment in Pre-K and neighborhood development programs.324 Agencies thus can “generate and mobilize awareness of structural oppression, disruptive claims, and reform.”325 Assembled into newly educated constituencies, excluded citizens might then demand more attention from elected politicians. Further, because many agencies actively distribute resources, they can alter the “background distribution of political power”326 that presently serves to cement political inequality.327 Parents benefitting from free universal Pre-K might, for example, use their newly freed resources to engage more actively in politics. 2. Technocracy, Non-Delegation and Representation Given the potential of capture, it is tempting to inoculate agencies from interest group pressure by granting them political independence.328 Indeed, this “juridical” solution was suggested by the early 20th century Progressives Legal Realists,329 and the technocratic “good governance” reformers of the 1990s and 2000s.330 Under this kind of scheme, agencies can 322 Mansbridge, supra note 226, defines this as the selection of representatives based on voters’ judgment of the whether the representative’s internal motivations and goals are congruent with their needs. 323 Vieira, supra note 238. 324 Rahman & Gilman, supra note 290. 325 Vieira, supra note 238. 326 Rahman & Gilman, supra note 290, at 140. 327 Emerson, supra note 17, helpfully discusses a “bureaucratic vanguard” model proposed by DuBois. 328 Nicholas Lemanm, What Transactions Can’t Do, THE TANNER LECTURES ON HUMAN VALUES 149 (2012); Rosanvallon, supra note 14, at 63; Caiden, supra note 104, at 53. 329 Ernst, supra note 114; Fishkin & Fortbath, supra note 306; Caiden, supra note 104, at 58; Goodnow, supra note 116, at 39; Neumann, supra note 79, at 1912. Lowi (1969) also offerrs a juridical solution to agency capture by big business interests. See Lemann, supra note 328 (2012). 330 Rahman & Gilman, supra note 290, at 118; Stewart, supra note 57, at 443. ALL THE SOVEREIGN’S AGENTS 45 only credibly claim to vindicate democratic sovereignty if (1) there is a clearly identified and consensus-based public interest, or “will” that (2) neutral expert agency officials, as fiduciaries, can be trusted to fulfill.331 Yet finding consensus is often difficult, if not impossible. Moreover, experts commonly disagree and these disagreements may implicate questions of morality and ethics.332 Of all the lessons of the 2008 financial crisis, one of the most important is that not even central banking is a matter of apolitical, non-controversial scientific management.333 Ascribed an open-ended duty to expertly manage employment levels and price stability,334 the Federal Reserve System is perhaps the strongest example of a violation of the nondelegation doctrine.335 What’s more, for the past dozen years, the Fed’s decisions can no longer credibly be described as merely technocratic.336 It blatantly engages in distributional politics as it decides to divert public resources to some actors but not to others.337 Regardless, any choice between inflation (which tends to harm wealthy asset-holders and benefit debtors) and unemployment (benefitting labor) has distributional consequences.338 It thus makes value-laden political choices – choices that, given the complexity and importance of macroeconomic management, many feel uncomfortable placing into the hands of the vagaries of popular lay sentiment.339 The good news is that because representative democracy requires only democratic autonomy and not universal consent, it can tolerate barriers to popular input. It might even include those barriers erected by complex issues that only experts can fully understand. Rather than throwing the technocratic baby out with the non-delegation bathwater, constitutional democracies can achieve both the benefit of expertise and citizen 331 See Tucker, supra note 273, at 101. Richardson, supra note 29, at 118; Elizabeth Anderson, Democracy, Public Policy, and Lay Assessments of Scientific Testimony, 8 EPISTEME 144 (2011); Dahl, supra note 218. 333 See, e.g., supra note 6. 334 The Federal Reserve Act, 12 U.S.C. § 226, Pub. L. No. 63-43 (1913). 335 E.g., Jens van't Klooser, The Ethics of Delegating Monetary Policy, 82 J. OF POL. 587 , 590-91 (2020). 336 See, e.g., ibid., at 587-8; Tucker, supra note 273, at 405; Trevor Jackson, The Sovereign Fed, DISSENT, Apr. 16, 2020; Adam Tooze, The Death of the Central Bank Myth, FOREIGN POLICY, May 13, 2020. 337 Carillo, supra note 295; Nathan Tankus, The Federal Reserve’s Coronavirus Crisis Actions, Explained: Riots, Municipalities and Monetary Policy, NOTES ON THE CRISIS, May 2020. 338 For a good summary of this and other political, distributional consequences of monetary policy, see J. Lawrence Broz, The Domestic Politics of International Monetary Order: The Gold Standard, in INTERNATIONAL POLITICAL ECONOMY 223 (Jeffrey A. Friedan et al. eds., 5th ed. 2010). 339 See e.g., van’t Klooser, supra note 335. 332 46 JACKSON participation through representative politics.340 Recall that democratic autonomy does not require that citizens, literally, transcribe their wills into law as a public “general will.” Instead, it only requires that citizens play an equal role in creating the rules that bind them. They can participate equally by, for example: (1) electing the regulatory experts that set policy; or, more feasibly (2) seeing that their particular interests are represented on the Fed board. Seats might be staffed not only by financial industry players and academic macroeconomists, but also by labor and consumer advocates. Further, experts themselves can serve as competing democratic representatives. They might present citizens with alternative views of what is desirable and possible and thereby stimulate public debate and feedback.341 Elite expertise can be countered by competing expertise to prevent a monopolization of public discourse.342 Moore,343 for example, proposes a theory of critical elitism that would provide ongoing opportunities for citizens to oppose, scrutinize, and protest expert conclusions within a system of deliberative democracy. Rahman344 argues that there can be a way to “embed” expertise within democratic institutions and democratic reason. Administrative specialists can “engage in ethical reasoning, not as a cloistered group of enlightened experts, but rather as partners with affected persons.”345 Indeed, Congress itself specifically seeks expert opinions from agencies because policy goals might be ill-served by reliance on lay opinion.346 Democratic citizens can use representative expert guidance not only as they consider agency actions, but also in engaging in the kind of deliberation and considered judgment that representative theory endorses more generally. To be sure, an expert’s claim to possess truth and knowledge can also amount to a claim for authority that competes with democratic legitimacy.347 340 For examples of how experts can be integrated into individual democratic reasoning, see Anderson, supra note 332; and Coran Stewart, Expertise and Authority, 17 EPISTEME 420 (2020). 341 Brugue & Gellego, supra note 272; see Waldron, supra note 37, at 90-91 (embracing the notion of an elected aristocracy, where voters consider the specific functions to be performed by the official). 342 NORBERTO BOBBIO, THE FUTURE OF DEMOCRACY 60 (Richard Bellamy trans. 1987). 343 Alfred Moore, CRITICAL ELITISM: DELIBERATION, DEMOCRACY, AND THE PROBLEM OF EXPETISE 9-10, Ch. 3 (2017). 344 Supra note 268, at 102 (2011). 345 Emerson, supra note 17, at 64 (2019) 346 Tucker, supra note 273; van’t Klooster, supra note 335. 347 E.g., Hannah Arendt, Truth in Politics, THE NEW YORKER (Feb. 17, 1967); Turner, supra note 214, at 26. For a contemporary example of an argument in favor of epistocracy, or rule ALL THE SOVEREIGN’S AGENTS 47 The truth should govern, no matter other people’s opinions about it. Experts’ claims can therefore lead to an undemocratic concentration of power.348 What’s more, there is a kind of latent totalitarianism implied by claims to comprehend the entire world.349 A belief in one’s own perfect understanding of society will also tempt one to remake it without popular input. Yet we should not allow such worries to paralyze us. Knowledge is also liberating. Vacuum cleaners, washing machines, and other technological improvements allow us more choices in life. As Neumann observed, it is only armed with science and knowledge – however imperfect – that citizens can overcome the anxiety associated with an uncertain and unpredictable world and free themselves of the constraints of necessity.350 “The realization of freedom,” he reminds us, is not always “at the disposal of man’s free will.”351 Only knowledge can expand the zone of such realization. The democratic function of increased knowledge of both nature and humankind is that it can both show us new possibilities and how to achieve them.352 Less abstractly, the importance of agency effectiveness, and thus the salience of technological expertise, only grows once democratic publics append positive social and economic rights to their menu of negative liberties.353 It is one thing to promise full employment; it is quite another to fulfill this promise. Realizing rights requires the use of experts and the collection of data. This is why Italian political theorist Norberto Bobbio was by experts, see Jason Brennan, Does the Democratic Objection to Epistocracy Succeed?, 25 RES PUBLICA 53 (2018). 348 Moore, supra note 343. See Mark Blythe & Matthias Matthijs, When Is It Rational to Learn the Wrong Lessons? Technocratic Authority, Social Learning, and Euro Fragility, 16 PERSPECTIVES ON POLITICS 110-126 (2018) for an application of this point to European monetary policymaking. 349 Andrew Jainchill and Samuel Moyn, French Democracy between Totalitarianism and Solidarity: Pierre Rosanvallon and Revisionist Historiography, 76 THE J. OF MODERN HISTORY 107, 123 (2004). The idea also has an affinity with the marxist “administration of things” that would succeed communist revolution. See Turner, supra note 214, at 21 (citing Friedrich Engels) and, of course, is the target of F.A. Hayek’s THE ROAD TO SERFDOM (1944). 350 Supra note 79, at 180-84. 351 Ibid. 352 Ibid., at 181; Moore, supra note 343, at 5-6. 353 Emerson, supra note 17, at 67 (“rights-focused constitutionalism and federal administrative intervention were often co-original. Reconstruction is a paradigm case for this simultaneous expansion of bureaucratic power and private right.”) (citing Jeremy K. Kessler, The Administrative Origins of Modern Civil Liberties Law, 14 COLUM. L. REV. 1083 (2014); Karen M. Tani, STATES OF DEPENDENCY: WELFARE, RIGHTS, AND AMERICAN GOVERNANCE (2016); William N. Novak, The Legal Origins of the Modern American State, in LOOKING BACK AT LAW’S CENTURY 264-265 (Austin Sarat, Bryant Garther, and Robert A. Kagan eds. 2002)). 48 JACKSON able to argue that “[a]ll states which have become more democratic have simultaneously become more bureaucratic.”354 Administration is, per Jurgen Habermas, “not just a functionally necessary supplement to the system of rights but implications already contained in rights.”355 The protection and vindication of individual rights are perhaps “best addressed by officials who, through repeated encounters with a particular class of disputes, understood their origins.”356 The affinities between expertise and rights is why antiadministrative attacks against bureaucratization at least occasionally conceal attempts to “if not dismantle democratic power, then certainty to reduce it to within clearly circumscribed limits.”357 Without experts, rights are really just parchment promises. Moreover, Emerson suggests that the tension between democratic autonomy and expertocracy can be a productive, constitutive one.358 Citizens (and their representatives) routinely challenge received truths from would-be Guardians. For example, after the 2008 financial crises, many recognized that purportedly neutral, technocratic economic reasoning often rests on debatable values and debatable science. One need only observe the fruitful work on MMT359 and heterodox Keynesianism360 – indeed, Keynesianism itself361 – to know that expertise is fruitfully politicized. The field of inquiry where “reason compels a unique result”362 is smaller than one might expect – and getting smaller. 3. Representation and the Chevron363 Shuffle Theories of representation can also inform an enduring problem of administrative law: whether and to what extent courts should permit agencies 354 Supra note 342, at 38. Supra note 57, at 134. 356 Ernst, supra note 114, at 33 (addressing the work of Charles Evans Hughes). 357 Ibid.; see also Robert Dahl, The Science of Public Administration, 7 PUB. ADMIN. REV. 1, 3 (1947); Wilson, supra note 58, at 200. 358 Supra note 17, at 63. 359 See, e.g., STEPHANIE KELTON, THE DEFICIT MYTH: MODERN MONETARY THEORY AND THE BIRTH OF THE PEOPLE’S ECONOMY (2020). 360 See, e.g., HETERODOX MACROECONOMICS: KEYNES, MARX AND GLOBALIZATION (Jonathan Goldstein & Michael Hillard eds., 2009). 361 For a superb discussion of the politics of Keynes, see GEOFF MANN, IN THE LONG RUN WE ARE ALL DEAD (2019); in Keynes own words, about the politicization of the gold standard, see JOHN MAYNARD KEYNES, ESSAYS IN PERSUASION (Martino Publishing 2012) (1932). 362 WILLIAM GALSTON, LIBERAL PLURALISM 69 (2002). 363 Chevron, U.S.A., Inc. v. Nat. Resources Def. Couns., 467 U.S. 837 (1984). 355 ALL THE SOVEREIGN’S AGENTS 49 to interpret their own implementing statutes.364 Theories of representation, which privilege democratic autonomy, suggest that, at the very least, agencies should not provide an opportunity to relitigate de novo the hard compromises already settled by Congress in regular lawmaking. Whatever democratic autonomy is vindicated through congressional lawmaking, that autonomy should not be undermined by agency decisions taken behind citizens’ backs. As a result, whether an agency interpretation is “reasonable” and whether Congress’ intent is “clear”365 should be informed by whether and what extent that party seeks to undo what Congress has done. Historically, those enjoying an unequal share of resources have exploited agency rulemaking processes to undo laws they find repugnant.366 This is particularly true during periods of divided and polarized party government when agencies are the only fora amenable to new policy initiatives.367 As a result, major policy reversals made exclusively within the ambit of agency decision-making may deprive many of their ability to understand themselves as co-equal political participants.368 Unless agencies are opened to some kind of regular popular feedback mechanism, such relitigation may threaten political equality or amount to a form of corporatism that leads to the privatization of public functions.369 This superficial commitment to statutory formalism is not based on the principles of non-delegation. Congress is not the source of the general will legibus solutus. It is rather to acknowledge that Congressional statutes can reflect the exercise of democratic autonomy370 and that agency action should not contradict them – not unless it has a stronger claim to democratic 364 See generally Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 DUKE L. J. 5101 (1989). 365 467 U.S. at 842-44. 366 E.g., Michael Beckel, Lobbying Push Targeting Federal Regulatory Agencies Continues on Wall Street Reform Fight, CENTER FOR RESPONSIVE POLITICS, June 1, 2011; Tom Holyoke, Theodore Lowi, The End of Liberalism: The Second Republic of the United States, in THE OXFORD HANDBOOK OF CLASSICS IN PUBLIC POLICY AND ADMINISTRATION (M. Lodge et al. eds., 2015). 367 Livermore & Richardson, supra note 18, at 46. 368 See Waldron, supra note 37, at 128-9 (describing the illegitimacy of litigants wanting to “steal a march” on one’s political opponents by pursuing legal reform through the courts rather than through “a forum that everyone knows is the place to go to reach decisions about whether and how the law should be changed,” viz, the legislature.). 369 See AMY C. OFFNER, SORT OUT THE MIXED ECONOMY: THE RISE AND FALL OF WELFARE AND DEVELOPMENTAL STATES IN THE AMERICAS (2019); Neumann, supra note 79, at 192. 370 Waldron, supra note 37, at 126, 128 (legislatures are presented to citizens as where lawmaking is done, and therefore where they should direct their energy when it comes to lawmaking), 154. 50 JACKSON autonomy. And it is quite possible that agency decision-making might at least occasionally better vindicate the principle of democratic autonomy. In a partisan era of omnibus legislation accomplished through the budget reconciliation process, agencies might provide a more accessible point of entry for representative politics.371 Moreover, if the policies at issue impact only a small subset of the population, agencies with representative decisionmaking processes might be better placed to serve democratic autonomy of affected constituents. Finally, it is not unusual for Congress to punt contentious political issues to agencies for resolution (or the avoidance thereof).372 Unless those agencies include some kind of representative decision-making procedure, what policies result may lack any legitimacy whatsoever. Rather than assigning to administration the task of execution and to Congress the task of promulgating “clear” and detailed implementing statutes, it may make more sense to divide responsibilities according to the contentiousness of the issues involved – regardless whether we characterize those issues as “means” or “ends.” When citizens mobilize around an issue, resolving their conflict in a forum that allows each (more) direct influence in a single national conversation may best serve democratic autonomy.373 To illustrate, we might permit representative politics within a highly politicized decision to increase the deficit to stave off depression. The choice involves deeply rooted moral and technical conflicts. On the other hand, we might insulate CDC doctors and IRS accountants unless and until a significant opposition is mobilized. CONCLUSION As the left flank of the Democratic Party spitballs ideas about a Green New Deal, the nationalization of healthcare, anti-monopoly and other business regulation, it is perhaps more urgent than ever to find legitimate ground for the public bodies that will inevitably be tasked to implement them. 371 See WALTER J. OLESZEK ET AL., CONGRESSIONAL PROCEDURES AND THE POLICY PROCESS 61-65 (10th ed. 2016); ALLEN SCHICK, THE FEDERAL BUDGET: POLITICS, POLICY, PROCESS 268 (2007); BARBARA SINCLAIR, UNORTHODOX LAWMAKING 121-28, 138-40 (4th ed. 2012) (noting use of reconciliation to enact controversial policies). 372 George I. Lovell, That Sick Chicken Won’t Hunt: The Limits of a Judicially Enforced Non-Delegation Doctrine, 17 CONST. COMMENT. 79 (2000); Rubin, supra note 139, at 384. 373 This logic is the reverse of that found in Food and Drug Admin v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000) (the question is of such political and economic magnitude that Congress would be unlikely to delegate it). Congress may fail to speak on any number of things for any number of reasons - not just because Congress lacked an “intent” to regulate. ALL THE SOVEREIGN’S AGENTS 51 Vague instructions to “democratize” administration are insufficient. Instead, institutional design should remain sensitive to extant political structures and focus on democratic autonomy, or how agencies can help citizens understand themselves as co-equal participants in making the laws that bind them. This may mean that some agencies should not, in fact, be fully democratized– particularly if they are to act as partial representatives themselves or if they aim to exploit the advantages of expertise. A labor board should not allow itself to be captured by corporate leadership; monetary policy should not take direction from unfiltered social media posts. Hotly contested issues might best be left for resolution in Congress through national elections. Meanwhile, tax authorities and CDC officials might be insulated from political pressure unless and until their actions become politicized. By understanding the lessons of political theory’s representative turn, it is possible to embed agencies within constitutional, democratic decision-making institutions. And it is possible that they can also be embedded in a way that vindicates their virtues of effectiveness and expertise – while also ensuring that citizens enjoy an equal chance to shape their collective goals. It is time to bring administration home from its constitutional exile.