70 Am. U. L. Rev. 829 (2021).

* Bridget and Thomas Black SUNY Distinguished Professor of Law and Research Professor of Political Science, University at Buffalo School of Law, The State University of New York. Thanks to Jessica Bulman-Pozen, Jack Chin, Neal Devins, John Dinan, Lise Gelernter, David Landau, and Rich Schragger for comments on a prior draft, and to Daniel Caves and Kyla Markowitz for valuable research assistance. Special thanks to Nina Cascio of the University at Buffalo Law Library for extraordinary assistance obtaining sources during the pandemic.

Federalism contemplates subnational variation, but in the United States, the nature and significance of that variation has long been contested. In light of the recent turn, globally and nationally, toward authoritarianism, and the concurrent sharp decline in public support not merely for democracy but also for the philosophical liberalism on which democracy rests, it is necessary to discard or to substantially revise prior accounts of the nature of state-to-state variation in the United States. All such accounts implicitly presuppose a common commitment, across the political spectrum, to the core tenets of democratic liberalism, and consequently assume that subnational variations in policy preferences and modes of self-governance reflect nothing more than disagreements within the shared American liberal tradition. That assumption, if it was ever valid, may be no longer. Like other federal states in which subnational “illiberal enclaves” have persisted over time, the United States may be witnessing a replication at the subnational level of what appears to be happening at the national level: a growing chasm along a cleavage between democratic liberalism and illiberal authoritarianism, in which some states remain committed to inherited forms of democratic liberalism while others cling to (or develop, or resurrect) patterns of illiberal authoritarianism.

This Article examines both the “large-C” formal constitutions of the states and their “small-c” informal constitutions and behavior for evidence of such a development. The evidence shows that North Carolina and Wisconsin have advanced the farthest down the road to subnational authoritarianism, with Florida, Texas, Kansas, Arizona, and Alabama not far behind. Disturbing initial signs of democratic backsliding have appeared as well in Georgia, Iowa, Michigan, Ohio, Pennsylvania, and Tennessee. Although subnational authoritarianism in the United States is at this point far less severe than that found elsewhere in the world, the evidence suggests considerable cause for concern.


Federalism contemplates subnational variation. By establishing and then dividing power among national and subnational governments, federalism generates the potential for subnational units to adopt disparate policies in a range of domains commensurate with the scope of decentralization. Indeed, according to some theories of federalism, the principal point of federalization is precisely to allow the expression of regionally distinct policy preferences through processes of subnational self-governance.1See Thomas O. Hueglin & Alan Fenna, Comparative Federalism: A Systematic Inquiry 12, 85–86 (2006); Michael W. McConnell, Federalism: Evaluating the Founders’ Design, 54 U. Chi. L. Rev. 1484, 1493–94 (1987).

There is no question that the United States displays, to at least some degree and in at least some domains, the kind of subnational variation one would expect in a highly decentralized federal state. Such variation may be found in the habits and customs of state populations, in their policy choices, in the form and substance of legal regimes, and even in the states’ foundational charters—their constitutions. Yet there is longstanding disagreement among American scholars about the nature and significance of this variation. Some claim that state-to-state variations in policy choices reflect profound underlying differences in the cultures and deeply held values of state polities.2See Ernest A. Young, The Volk of New Jersey? State Identity, Distinctiveness, and Political Culture in the American Federal System 1, 6, 47 (Feb. 24, 2015) (Working Paper); Daniel J. Elazar, Foreword: The Moral Compass of State Constitutionalism, 30 Rutgers L.J. 849, 850–51, 862, 868 (1999) [hereinafter Elazar, Moral Compass]. Others contend that policy variation in the American states reflects little more than normal variation occurring within a largely bounded domain of deeply shared national values and policy commitments.3See, e.g., Jacob T. Levy, “States of the Same Nature”: Bounded Variation in Subfederal Constitutionalism, in New Frontiers of State Constitutional Law: Dual Enforcement of Norms 26 (James A. Gardner & Jim Rossi eds., 2011); James A. Gardner, Interpreting State Constitutions: A Jurisprudence of Function in a Federal System (2005) [hereinafter Gardner, Interpreting State Constitutions]. Still others argue that state-to-state differences in policy choices reflect variations not in the values of state polities so much as differences in the prepackaged policy commitments of whichever national political party happens to control the levers of subnational governance in any particular state at any particular moment.4See Jessica Bulman-Pozen, Partisan Federalism, 127 Harv. L. Rev. 1077, 1081–82, 1119, 1122, 1144–45 (2014); Heather K. Gerken, Foreword: Federalism All the Way down, 124 Harv. L. Rev. 4, 17 (2010).

Recent events, however, both globally and nationally, put these subnational variations in a significantly new light. Around the world, liberal democracy—the dominant form of political self-organization since the conclusion of the Second World War—is in decline. In longstanding liberal democracies such as France, Germany, Austria, and Italy, populist parties that reject the core tenets of twentieth-century liberalism have not only appeared but also become influential players in national politics.5Pippa Norris & Ronald Inglehart, Cultural Backlash: Trump, Brexit, and Authoritarian Populism 9–12 (2019); see also Morris P. Fiorina, Unstable Majorities: Polarization, Party Sorting and Political Stalemate 154 (2017) [hereinafter Fiorina, Unstable Majorities]. In newer but seemingly settled liberal democracies such as Hungary, Poland, and Turkey, authoritarians who openly reject liberal democracy have risen to power.6See Steven Levitsky & Daniel Ziblatt, How Democracies Die 3, 5–6 (2018); Tom Ginsburg & Aziz Z. Huq, How to save a Constitutional Democracy 21, 46–47, 70, 73, 150, (2018); Larry Diamond, Ill Winds: Saving Democracy from Russian Rage, Chinese Ambition, and American Complacency 54 (2019). In the United States, the Republican Party pursues a systematic program of subverting democratic processes through vote suppression and the manipulation of electoral processes in order to maintain itself in power.7See infra Section III.C. An individual with “clear authoritarian tendencies”8Levitsky & Ziblatt, supra note 6, at 2; see also Robert R. Kaufman & Stephan Haggard, Democratic Decline in the United States: What Can We Learn from Middle-Income Backsliding?, 17 Persp. on Pol. 417, 417 (2019) (“Trump has exhibited many autocratic traits.”). occupied the White House from 2016 to 2020, with the full-throated support of his party, from its leadership to its rank and file.9See infra Section III.C. At the same time, a growing body of academic literature has brought into focus a previously overlooked phenomenon in federal states in which subnational “enclaves” sometimes maintain illiberal or overtly authoritarian forms of government in defiant opposition to powerful trends of democratization and liberalization on the national level.10Edward L. Gibson, Boundary Control: Subnational Authoritarianism in Democratic Countries, 58 World Pol. 101, 105, 107 (2005). See generally Robert Mickey, Paths out of Dixie: The Democratization of Authoritarian Enclaves in America’s Deep South, 1944–1972 (2015).

The hypothesis I wish to explore here is whether subnational variation in the United States can now best be explained as reflecting a deeply fraught realignment of state polities along a cleavage—a chasm, really—dividing those that continue to support established, long-dominant forms of liberal democracy from those that simply do not—from those that favor illiberal and authoritarian arrangements. Indeed, this possibility raises antecedent questions of whether Americans ever truly embraced liberalism, and if they did, how deeply such commitments actually ran, and how any such commitments are distributed.

Part I of this Article discusses subnational variation in federal theory and reviews the major accounts of subnational variation in the United States. I argue that we must now view all of these accounts as subject to revision on the ground that they share an unstated premise that no longer holds—that all subnational variation within the United States occurs, by definition, within the bounds of a universally shared American commitment to liberal democracy. Part II describes the current crisis of liberal democracy around the world.

Part III turns to the history and scope of liberalism and illiberalism in the United States, reviewing a variety of different kinds of evidence suggesting that the American commitment to liberalism has been thin and fragile and that American politics has been intertwined since the Founding with a strong counter-tradition of illiberalism. As is typically the case, however, illiberalism in America is not distributed haphazardly, but tends to be concentrated, either in a political party or in a region. Part III reviews the evidence showing that, in the United States, illiberalism and authoritarianism are concentrated in the Republican Party.

Part IV examines the regional concentration of illiberalism in the United States. It begins with a brief overview of the phenomenon in federal states of authoritarian “enclaves”—persistent pockets of illiberal and authoritarian rule in states otherwise committed to liberal democracy at the national level. It then closely examines the constitutional regimes of the American states for evidence of subnational authoritarianism, either formally, embedded directly in state constitutions; or informally, in the form of patterns or norms of official behavior. I find the evidence to be inconclusive, but suggestive and deeply troubling. The Article concludes with some observations on the relationship between federalism, subnational illiberalism, and authoritarianism.

I. Substantial Variation in the United States

A. Federal Theory

Federalism is a form of governmental organization in which official power is divided between national and subnational governments;11See K.C. Wheare, Federal Government 10 (1946); William H. Riker, Federalism: Origin, Operation, Significance 11 (1964). in Daniel Elazar’s famous phrase, a federal system is one in which each level of government enjoys a combination of “self-rule and shared rule.”12Daniel J. Elazar, Exploring Federalism 5 (1987). The main consequence of this division of authority is the opportunity it creates for public policies to vary within a federal system, and in particular among the subnational units comprising the federation.

In some theories of federalism, the opportunity that federalism offers for geographically defined subnational populations to adopt the public policies they prefer furnishes the principal justification for the adoption of a federal form of organization.13The classic articulation of this view is Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J. Pol. Econ. 416 (1956). This justification for federalism is often associated with multinational states, in which a guarantee of some measure of autonomy to an ethnically, linguistically, or religiously distinct subnational unit may be a critical term of the “federal bargain” that makes formation of the state possible in the first instance.14Riker, supra note 11, at 12; Mikhail Fillippov & Olga Shvetsova, Federalism, Democracy, and Democratization in Arthur Benz & Jörg Broschek, Federal Dynamics: Continuity, Change, and the Varieties of Federalism 167, 168–70 (2013). However, although the kind of subnational variation associated with ethnonationalism is a sufficient condition for the creation of a federal state, it is not a necessary one. Subnational policy variation may be equally desirable for other reasons: to maximize overall social utility;15See Thomas R. Dye, American Federalism: Competition Among Governments 12–15 (1990). to allow Brandeisian policy experimentation;16See New State Ice Co. v. Liebmann, 285 U.S. 262, 310–11 (1932) (Brandeis, J., dissenting) (“It is one of the happy incidents of the federal system that a single courageous State may, if its citizen choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”); Francesco Palermo & Karl Kössler, Comparative Federalism: Constitutional Arrangements and Case Law 318–19 (2017). or to establish a system of intergovernmental competition and policy contestation for the purpose of protecting liberty—a rationale generally taken to provide the principal justification for federalism in the United States.17See, e.g., The Federalist Nos. 46, 47, 48, 49, 50, 51 (James Madison) (Jacob E. Cooke ed., 1961); see also James A. Gardner, The Theory and Practice of Contestatory Federalism, 60 Wm. & Mary L. Rev. 507, 508, 511 (2018) [hereinafter Gardner, Contestatory Federalism].

The degree to which U.S. federalism in practice provides the benefits that federalism is in theory capable of delivering is deeply contested. Most obviously, public policy varies from state to state to some degree, and in some domains, often highly salient ones. Some states, for example, punish certain crimes by death while others have abandoned the death penalty.18See States and Capital Punishment, Nat’l Conf. of State Legislatures (Mar. 24, 2020), https://www.ncsl.org/research/civil-and-criminal-justice/death-penalty.aspx [https://perma.cc/H8WY-KCD5]. Some permit physician-assisted suicide while others condemn it as murder.19See Death with Dignity Acts, Death with Dignity, https://www.deathwithdignity.org/learn/death-with-dignity-acts [https://perma.cc/6QKH-6R5C]. Some have legalized the recreational use of marijuana while others prohibit it.20See Marijuana Overview, Nat’l Conf. of State Legislatures (Oct. 17, 2019), https://www.ncsl.org/research/civil-and-criminal-justice/marijuana-overview.aspx [https://perma.cc/BWA6-ZYEQ]. States tax their citizens at different rates, and offer dramatically different levels of public services.21See Erica MacKellar, State Tax Reliance, Nat’l Conf. of State Legislatures (July 2015), https://www.ncsl.org/research/fiscal-policy/state-tax-reliance.aspx [https://perma.cc/LH6C-VYD9].

Yet at the same time, policy choices in the American states often exhibit an enormous amount of convergence and uniformity. Huge swaths of private law, including tort law, contract law, and property law, are fundamentally similar from state to state, and even across common law countries around the world.22See, e.g., Yun-Chien Chang & Richard A. Epstein, Introduction to the Symposium on Convergence and Divergence in Private Law, 92 S. Cal. L. Rev. 741, 741 (2019). Powerful forces beyond the control of individual jurisdictions, such as markets and jurisdictional competition, lead frequently to “uniformity in the rules of different legal systems.”23Id. at 744. The apparatus of federal states typically includes not only mechanisms that facilitate conflict and diversity, but also mechanisms designed to facilitate coordination and legal uniformity.24Unification: Comparing Methods, Results, and Explanations Across 20 Systems, in Federalism and Legal Unification: A Comparative Empirical Investigation of Twenty Systems 3–4 (2014).

Potent forces of conformity and isomorphism frequently channel state policies into the same narrow paths even at the foundational level of constitutional law.25James A. Gardner, Autonomy and Isomorphism: The Unfulfilled Promise of Structural Autonomy in American State Constitutions, 60 Wayne L. Rev. 31, 34 (2014) [hereinafter Gardner, Autonomy and Isomorphism]. Every state, for example, maintains a presidential form of government even though the parliamentary model is the dominant one throughout the world.26Id. at 48–49. All but one state has a bicameral legislature.27The lone exception is Nebraska, which has a unicameral legislature. Neb. Const. art. III, § 1 (1934). Thirty-two state constitutions contain due process clauses identical, word-for-word, to each other and to the Due Process Clause of the Fifth Amendment to the U.S. Constitution, and thirty-seven contain language identical to the Speedy Trial Clause of the federal Sixth Amendment.281 James A. Gardner, State Expansion of Federal Constitutional Liberties: Individual Rights in a Dual Constitutional System xi (1999). Modes of cooperative federalism, moreover, facilitated by the extraconstitutional discipline imposed by national political parties,29See Larry D. Kramer, Putting the Politics Back into the Political Safeguards of Federalism, 100 Colum. L. Rev. 215, 218–19, 275–76, 282 (2000) (suggesting that political parties “forg[e] links between officials at the state and federal level”). can produce policy uniformity across states by subordinating regional preferences to national ones.30James A. Gardner, The Myth of State Autonomy: Federalism, Political Parties, and the National Colonization of State Politics, 29 J.L. & Pol. 1, 18, 53 (2013) [hereinafter Gardner, Myth of State Autonomy]; Bulman-Pozen, supra note 4, at 1086. Subnational autonomy, in other words, need not produce differentiation; it can just as easily yield policy convergence, depending upon the circumstances.

The seemingly haphazard and largely unpredictable nature and dimensions of policy variation among the American states thus pose difficult interpretational questions. What produces state-to-state policy variation? In what policy domains, and in what circumstances? How do we explain it?31There is a large literature on this topic. For a recent overview, see R. Kent Weaver, Policy Dynamics in Federal Systems: A Framework for Analysis, 50 Publius: J. Federalism 157 (2020).

B. Three Accounts of Subnational Variation in the United States

American federalism scholarship has produced, roughly speaking, three major schools of thought about how best to interpret and explain policy variations among the American states. I shall call these the traditional, the normal variation, and the partisan accounts.

1. The traditional account

The traditional account, the oldest of the three, holds that variations in policy choices made at the state level generally reflect underlying differences in the basic values held by state polities.32See James A. Gardner, Southern Character, Confederate Nationalism, and the Interpretation of State Constitutions: A Case Study in Constitutional Argument, 76 Tex. L. Rev. 1219, 1221–22 (1998) [hereinafter Gardner, Southern Character]. Some versions rest on an even stronger, ontological claim to the effect that state-to-state differences in values themselves reflect, in Hegelian fashion, even more profound underlying differences in the fundamental human character—in Hegel’s terminology, the “[s]pirit”33See, e.g., Georg Wilhelm Friedrich Hegel, The Philosophy of History 45–47 (J. Sibree trans.) (2001).—of state demoi.34E.g., Elazar, Moral Compass, supra note 2, at 860, 863–64.

As I have elsewhere argued, the traditional account traces its origins to antebellum efforts by Southern elites to build support for secession by supplementing political, economic, and legal justifications with what we would today call an “ethnonational” one, which claimed that deep-seated cultural and even ethnic differences distinguished Northern and Southern American populations.35See, e.g., Gardner, Southern Character, supra note 32, at 1220, 1285, 1287. In this sense, the justifications offered for Southern secession in the 1850s differ little from contemporary justifications sometimes offered by Scots or Catalans for secession from Britain and Spain. Later, early in the twentieth century, the traditional account was elevated to academic respectability by the influential historian of the American West, Frederick Jackson Turner.36Gardner, Interpreting State Constitutions, supra note 3, at 62–64. Turner argued that the American colonies had been settled in the seventeenth century by three distinct sub-populations—New England by “restrain[ed], . . . intellectual[]” Calvinists; Virginia by “the American version of the English country gentry”; and South Carolina by “a West Indian planting society” that was “self-centered, haughty, and fiery-tempered.”37Frederick Jackson Turner, The United States, 1830–1850: The Nation and Its Sections 41, 147 (1935). These three original populations, according to Turner, subsequently migrated west in clear, identifiable patterns, carrying with them—and retaining—their distinct cultures and values,38Id. at 6, 31. See generally Frederick Jackson Turner, The Rise of the New West (1906). distinctions which, in turn, Turner argued, fully explained then-observable differences in political preferences and behavior.39See generally Frederick Jackson Turner, The Significance of Sections in American History (1932).

Although Turner’s account received a thorough academic pummeling after his death,40Gardner, Interpreting State Constitutions, supra note 3, at 63–64. it nevertheless continued to gain adherents among both political scientists41Most notably the distinguished federalism scholar Daniel Elazar. See generally Daniel J. Elazar, Cities of the Prairie: The Metropolitan Frontier and American Politics (1970); Daniel J. Elazar, American Federalism: A View from the States (2d ed. 1972). and historians.42David Hackett Fischer, Albion’s Seed: Four British Folkways in America viii, 5 (1989); Nicole Etcheson, The Emerging Midwest: Upland Southerners and the Political Culture of the Old Northwest, 1787–1861, xi–xii (1996). The traditional account eventually made the leap to law in the early 1980s primarily through the work of political scientist Daniel Elazar, who invoked it to argue that differences among state constitutions are best explained by the internal migration patterns of distinct American sub-groups holding distinct values and inhabiting different political cultures.43See Daniel J. Elazar, The Principles and Traditions Underlying State Constitutions, 12 Publius: J. Federalism 11, 18 (1982).

Within the legal academy today, Professor Ernest Young has mounted by far the most sophisticated defense of the traditional account. In a working paper that will eventually form the core of a book-length treatment,44Young, supra note 2. Young approaches the issue from the premise that the central purpose of federalism is to permit the state and national governments mutually to check one another. Moreover, Young argues that the capacity of states to exercise this checking function in a meaningful way depends heavily on whether Americans “identify” sufficiently strongly with, or are adequately “loyal” to, their states.45Id. at 4–6, 34–35.

It is not strictly necessary, Young explains, for this loyalty to be rooted in antecedently distinctive state identities or characters; it is possible, he concedes, that subnational autonomy, enjoyed over long periods of collective self-governance, can by itself generate a political community that commands the loyalty of its members.46Id. at 69, 87. But, he continues, an antecedently distinctive state identity, although not necessary, is nevertheless sufficient to support the kind of subnational loyalty on which the proper operation of federalism depends, and as an empirical matter, he claims, such identities exist.47Id. at 37–38. Thus, he argues, differences of geography, demographics, religion, forms of economic organization, and political culture48Id. at 46–66. contribute to the construction of a subnational network of “meaningful”—and meaningfully distinct—“communities of value.”49Id. at 6.

2. The normal variation account

The normal variation account—the position I have long taken in my own work—arose in part as a reaction against what its proponents saw as the implausibility of the traditional account. On this view, the traditional account approaches with undue credulity what are largely folkloric beliefs about American identity that are fundamentally implausible in a society long knitted together by dense networks of transportation, communications, and internal migration.50See Gardner, Interpreting State Constitutions, supra note 3, at 68–71; Robert A. Schapiro, Polyphonic Federalism: Toward the Protection of Fundamental Rights 16–26 (2009). Moreover, these folkloric beliefs are, in the end, self-discrediting as excessively “essentialist”51Bulman-Pozen, supra note 4, at 1112. and “deterministic”52George Wilson Pierson, The Frontier and American Institutions: A Criticism of the Turner Theory, in Turner and the Sociology of the Frontier 36 (Richard Hofstadter & Seymour Martin Lipset eds., 1968). in their adoption of an account in which “first causes are made to lie in real estate, not state of mind.”53Id. at 38.

The normal variation account, in contrast, begins from the premise that contemporary Americans share what Malcom Feeley and Edward Rubin call “a unified political identity” that is overwhelmingly national rather than subnational.54Malcolm M. Feeley & Edward Rubin, Federalism: Political Identity and Tragic Compromise 115 (2008). Consequently, observable state-to-state variations result not from significant underlying differences in culture and values, but merely from normal disagreement within a shared—and consequently bounded—domain of political beliefs and commitments. Such disagreements, that is to say, are intrafamilial, rather than cross-cultural, and their range is for the most part correspondingly narrow.

On this account, the conditions, practices, and beliefs of early American settlers are irrelevant, for even if early Americans were at first distinctive, the population has long since been thoroughly mixed through liberal internal migration, the passage of time, and, since at least the mid-twentieth century, the growing homogeneity of channels of information and cultural transmission.55See Gardner, Interpreting State Constitutions, supra note 3, at 69; Schapiro, supra note 50, at 25–26. Even more fundamentally, American identity has from the beginning been founded not on the kind of ethnic, linguistic, or cultural distinctiveness found in modern multinational federations like Spain or Switzerland, but on a commitment to a certain set of ideas. American nationalism was born in the revolutionary struggle against Britain, a struggle in which “Americans sought independence to vindicate rights they held not on account of their status as a distinct people but on account of being English.”56Gardner, Southern Character, supra note 32, at 1285. These rights, moreover, were understood to be held not merely by British subjects, but universally, by all mankind.57Id. Thus, as Hans Kohn has argued, American nationalism “was not founded on the common attributes of nationhood—language, cultural tradition, historical territory or common descent—but on an idea . . . . To become an American has always meant to identify oneself with the idea.”58Hans Kohn, American Nationalism: An Interpretive Essay 20 (1957).

Moreover, the most salient aspect of American identity—the one that counts for nearly any significant purpose—is political rather than cultural. It is beyond doubt that the American population displays a good deal of variety in culture and folkways; Professor Young may be correct in observing a difference between “clean-living Utah and gambling-mecca Nevada.”59Young, supra note 2, at 45. But it is erroneous to assume that these kinds of differences translate predictably and systematically into distinctions in the political and legal regimes by which state polities govern themselves. If I blindfolded you and gave you a plate of North Carolina barbecue and a plate of Texas barbecue, you might be able to tell them apart, just as, blindfolded, you could likely distinguish the speech of a New Englander from that of a Southerner. But I am quite confident that if I showed you the text of two state constitutions—the fundamental law of those states—you would not be able to tell them apart.60E.g., Gardner, Southern Character, supra note 32, at 1260–82; see also Daniel J. Hopkins, The Increasingly United States: How and Why American Political Behavior Nationalized 193 (2018) (describing non-political content of contemporary state identity).

Finally, according to this account, the primary locus of Americans’ identity—their “loyalty,” as the traditional account would have it—is not fixed, much less attached permanently or even predominantly to their states, but rather is fluid and changeable, and consequently attaches itself to the state or to the nation opportunistically, as circumstances change. All humans belong to many overlapping groups and communities simultaneously, and the dominant aspect of their identity at any given moment moves continually as one group or the other becomes more salient to the achievement of their immediate objectives.61The classic anthropological work on this topic is E.E. Evans-Pritchard, The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People (1940). In a federation, all citizens belong simultaneously to at least two political communities, state and national, and it is to be expected that their allegiance will vary between the two levels of government depending upon which level appears more likely to satisfy their most immediate, pressing political demands.62See James A. Gardner & Antoni Abad I Ninet, Sustainable Decentralization: Power, Extraconstitutional Influence, and Subnational Symmetry in the United States and Spain, 59 Am. J. Compar. L. 491, 494–95 (2011).

In sum, according to the normal variation account, state-to-state variations are much more persuasively explained by normal variation and geographical clumping among a population that shares a common ideological and political tradition, and disagreement occurs within this shared tradition, not across its boundaries.63See id. at 513–14, 523. Although federalism may, in some circumstances, serve as a vehicle for accommodating significant differences among subnational populations, under American conditions it does not; it serves instead as a vehicle for protecting liberty by offering alternative fora for political contestation among people of fundamentally similar views.64Gardner, Contestatory Federalism, supra note 17, at 518–19.

3. The partisan account

If the normal variation account offers a thinner conception of state identity than the traditional one, an even thinner version is offered by the partisan account. In this conception, advanced most thoroughly and rigorously by Jessica Bulman-Pozen, state-to-state variation in policy choices is best explained by the accident of whichever national political party happens to control the governing apparatus of any given state.

In Bulman-Pozen’s account, politics in the United States has become so thoroughly nationalized that, if not in design then in actual current practice, the structural apparatus of federalism now serves primarily as a forum not for competition among state and national polities or their governments, but as a forum for competition by the major national political parties.65Bulman-Pozen, supra note 4, at 1090–91, 1096. In this account, federalism’s main structural function today is to provide an important opportunity for the political party out of power at the national level to maintain its national competitiveness by taking control of as many state governments as it can and, in those venues, to pursue and refine its national policy agenda until it regains power at the national level.66Id. at 1092. In this model, possessing the levers of governance in a state is not a means by which a culturally distinct subnational polity devises and implements policies best suited to its own values and preferences.67Id. at 1122–23. Quite to the contrary, in this model, “[s]tate governance is . . . itself a means of participating in national partisan politics.”68Id. at 1123 (emphasis added). Federalism, that is to say, has become a mechanism serving the foundational democratic principle of alternation in power.

Bulman-Pozen does not deny that distinctive state identities of the kind associated with the traditional account are sufficient to support a well-functioning federal system in which state power is capable of providing an effective check on national power.69Id. at 1089–90. Rather, she argues, national party affiliation is itself a sufficiently meaningful form of collective identity to support a well-functioning federal system in which states can fulfill the checking function with which federalism charges them.70Id. at 1113–16. I have expressed my skepticism of this contention in James A. Gardner, Federalism and Subnational Political Community, 127 Harv. L. Rev. F. 153 (2014). In this case, however, the nature and grounds of subnational checking are identical to those that characterize the kind of mutual checking performed in the national political arena when the major parties compete for national office.

In this model, then, state-to-state variation in public policies is best understood as driven by the adventitious control of the levers of state government by national political parties as they pursue their aspirations for national office.71Bulman-Pozen, supra note 4, at 1123–24. State policies differ not because states “represent something essentially different from the nation, but rather because they represent competing Democratic and Republican visions of the national will.”72Id. at 1081.

C. Why All Three Accounts May Be Wrong

Although these three accounts of subnational variation in the United States differ dramatically from one another, all three may well be wrong, and for reasons different from any their proponents might have imagined. Despite their diversity, all three accounts rest implicitly on the same basic premise: that variation in subnational policy preferences, no matter how wide it may seem, occurs by definition within a domain bounded by the limits of a profound, collective, nationwide American commitment to liberal democracy. I will describe these commitments in greater detail in the next part, but in brief, the commitments to which I refer include popular sovereignty, the rule of law, the political equality of citizens, a free civil society, and a conception of human beings as clothed with certain dignitary rights which governments and popular majorities must respect. The possibility I wish to raise is that (1) a collective American commitment to these principles, if it ever existed, may be fast eroding; if so, (2) this erosion is not occurring uniformly throughout the states; and (3) variations in policy preferences among the states consequently may now be realigning along a highly consequential cleavage between liberalism and illiberalism, and between democracy and some form of populist authoritarianism.

None of the three accounts of subnational variation described in the previous part can survive an ideological realignment of such tectonic scale. When a traditionalist like Young refers to state “communities of value [that] . . . remain quite distinctive in key ways,”73Young, supra note 2, at 6. he is not, I assume, suggesting the possibility that what might be distinctive in some states is the polity’s full-throated embrace of National Socialism, say, or Stalinist communism. The picture of the United States that traditionalists tend to paint is a Norman Rockwell-esque one of charming local quirks, none of which threaten the basic unity of the nation. It is a conception of the nation in which positions on the issues that most strongly divide Americans—abortion, for example, or gay marriage—range from liberal progressivism to libertarianism, but all of which fall firmly within the bounds of democratic liberalism.

The normal variation account is even more vulnerable to the type of developments I have described, for it expressly presumes as axiomatic the existence of a deeply shared set of American political commitments, and those commitments are clearly liberal.

The partisan account is somewhat more difficult to dismiss, though in the end I think it, too, stumbles. As we shall see, state-level drift in the United States away from liberal democracy and toward authoritarianism are closely associated with political control of states by the Republican Party.74See infra Section III.C (detailing the drift away from liberal democracy). Thus, to the extent that the Republican Party’s policy commitments have become increasingly authoritarian—a proposition demonstrated below in Section III.C—it could be Republican policy commitments rather than distinctive popular attitudes and sentiments that are driving democratic liberalism over a cliff in some states. I do not think this, however, an adequate account of what is going on. The partisan account, it seems to me, takes as its observational and descriptive baseline a permanent alternation in power at the national level of two major political parties each committed to some variety of democratic liberalism. It does not contemplate conditions in which one party embraces democratic liberalism and the other supports authoritarianism—conditions in which the illiberal party, should it gain power, is highly unlikely to relinquish it, or even to tolerate the continued existence of ideological dissent.75See Diamond, supra note 6, at 18–19; Ginsburg & Huq, supra note 6, at 23. See generally Steven Levitsky & Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes After the Cold War (2010). Its explanatory power thus seems to me limited mainly to situations other than the one now emerging in the United States.

Consequently, with none of the existing models adequate to describe the situation that may now be developing, it is necessary to propose, albeit tentatively and with due caution, a new account of subnational variation. The next Part begins that task.

II. The Crisis of Liberal Democracy

A. Liberalism and Illiberalism

1. Liberalism

There is no single accepted definition of “liberalism”; its meaning has evolved considerably since its earliest appearance in the ancient world, and has from time to time been deeply contested even by its own adherents.76Helena Rosenblatt, The Lost History of Liberalism: From Ancient Rome to the Twenty-First Century 1, 2 (2018). Although contemporary liberalism is typically thought to have deep roots in the thinking of Enlightenment political philosophers such as John Locke,77See Rosenblatt, supra note 76, at 1–2 (stating that some scholars trace liberalism to John Locke). See generally James Traub, What Was Liberalism? The Past, Present, and Promise of a Noble Idea 7 (2019); Louis Hartz, The Liberal Tradition in America: An Interpretation of American Political Thought Since the Revolution (1955). according to some accounts, liberalism did not fully acquire its contemporary signification and associations until the period immediately following the Second World War,78Traub, supra note 77, at 257–64; Sheri Berman, Democracy and Dictatorship in Europe: From the Ancien Régime to the Present Day 9 (2019). or even later, following the fall of the Berlin Wall in 198979See Diamond, supra note 6, at 18–19, 50–51; Robert Kagan, The Jungle Grows Back: America and Our Imperiled World 61, 83–89 (2018); see also Rosenblatt, supra note 76, at 268.—periods during which the astonishingly rapid spread of liberal democracy made it seem like the culmination of some kind of “natural human evolution.”80Kagan, supra note 79, at 61; see Rosenblatt, supra note 76, at 271. This argument was most notably made in Francis Fukuyama, The End of History and the Last Man (1992).

Nevertheless, contemporary writers seem to agree on several core features of liberal thought. First and foremost, perhaps, is popular sovereignty—the doctrine that the people of a society are entitled to rule themselves,81Popular sovereignty, at least early on, did not necessarily imply democracy: “Endorsing popular sovereignty did not mean endorsing universal suffrage.” Rosenblatt, supra note 76, at 83. and that in consequence, as Thomas Jefferson put it, governments “deriv[e] their just powers from the consent of the governed.”82The Declaration of Independence para. 2 (U.S. 1776). Adherence to the principle of popular sovereignty in turn tends strongly to pull along with it certain other commitments that are either (1) conceptually antecedent to popular sovereignty and upon which it relies; or (2) widely considered essential to the success and maintenance of meaningful popular self-rule. In the first category, for example, typically lies a commitment to the fundamental political equality of citizens,83Ginsburg & Huq, supra note 6, at 24. the axiomatic point of departure for both Locke and Thomas Hobbes.84Thomas Hobbes, Leviathan ch. 13 (John Plamenatz ed., 1963) (1651); John Locke, The Second Treatise of Government ch. II, §§ 4–8 (C. B. Macpherson ed., 1980) (1690). Similarly, most contemporary accounts of liberalism require a “lively civil society,”85Diamond, supra note 6, at 19. in which citizens enjoy the freedom necessary to develop their own ideas about the nature of the good life and how the polity ought best to pursue it.

In the second category, commentators agree that an essential attribute of liberalism is the rule of law,86Ginsburg & Huq, supra note 6, at 9. a condition that must be satisfied if popular sovereignty is to issue in meaningful self-rule. The concept of the rule of law is often said to include constitutionalism, a specific kind of rule of law in which a constitution is understood to contain a polity’s fundamental law, binding on all citizens, including government officials.87See Stephen M. Griffin, American Constitutionalism: From Theory to Politics 12 (1996). This category also is typically said to include basic human rights, such as the rights of free speech and free association, that are thought necessary to the effectuation of popular self-rule.88Ginsburg & Huq, supra note 6, at 9; Juan J. Linz, Totalitarian and Authoritarian Regimes 20, 58 (2000).

From this point, consensus tends to diminish. Some varieties of liberalism include among their requirements features such as limited government, or dispersion of official power,89Norris & Inglehart, supra note 5, at 78 (“[I]nstitutional checks and balances with the legislature and states limiting the powers of the federal executive.”); Yascha Mounk, The People vs. Democracy: Why Our Freedom Is in Danger and How to save It 26 (2018) (“[S]eparation of powers.”); Stephen Holmes, The Anatomy of Antiliberalism 4 (1993) (“[C]onstitutional government based on a separation of powers.”). while others view such arrangements as contingently desirable but not necessary for a system of social organization to count as liberal.90See, e.g., Ginsburg & Huq, supra note 6, at 10 (offering a minimalist definition of liberal democracy as consisting of democratic elections, rights to speech and association, and the rule of law). More progressive conceptions of liberalism might include a wider array of dignitary rights, such as due process and strong protections for those accused of crime, as well as strong rights of political participation and a robust commitment to majoritarianism as a form of political organization.91Holmes, supra note 89, at 3–4; Norris & Inglehart, supra note 5, at 78. More conservative versions of liberalism, in contrast, might include free markets as a foundational condition.92See Traub, supra note 77, at 1; Patrick J. Deneen, Why Liberalism Failed 1 (2018); see also Ronald Dworkin, A Matter of Principle 195, 199 (1985).

For present purposes, when I speak of liberalism, I refer to the characteristics that command the greatest degree of consensus: popular sovereignty, the equality of citizens, a free civil society, the rule of law, and some package of basic human rights necessary to operationalize these other commitments.

2. Illiberalism

There are at least as many varieties of illiberalism as liberalism,93See Günter Frankenberg, Authoritarian Constitutionalism: Coming to Terms with Modernity’s Nightmares, in Authoritarian Constitutionalism: Comparative Analysis and Critique 3–4 (Helena Alviar García & Günter Frankenberg eds., 2019). and probably many more. It is useful to remember, in this regard, that liberalism itself arose largely as a reaction against various systems of political thought rooted mainly in theology or religiously informed traditionalism—the European ancien régimes.94Berman, supra note 78, at 15–28. In those systems, the people did not rule themselves, but were ruled by God and His appointed earthly agents—kings, nobles, and clergy—according to laws of divine rather than human origin.95A key text in this tradition is Robert Filmer, Patriarcha (1680), which prompted Locke to develop foundational conceptions of liberalism in his two treatises. All people were born into rigidly assigned and unalterable stations established either by divine decree or through the sheer weight of inherited tradition; the order of things, in this system of belief, was “absolutely rigid and static,” and “inconsistent with any belief in progress, or, indeed, in any sort of significant change in the universe as a whole.”96Arthur O. Lovejoy, The Great Chain of Being: A Study of the History of an Idea 242 (1964). Attempts to deviate from social assignments of station were understood by leaders as serious challenges to their authority and to the stability of the entire system.97For Rosenblatt’s account of the reaction against the French Revolution, see Rosenblatt, supra note 76, at 41–87. Or, as Robert Kagan puts it, “[l]iberalism is all that keeps us, and has ever kept us, from being burned at the stake for what we believe.” Robert Kagan, The Strongmen Strike Back, Wash. Post (Mar. 14, 2019), https://www.washingtonpost.com/news/opinions/wp/2019/03/14/feature/the-strongmen-strike-back. In Europe, the Church was long seen as the exemplar of illiberalism, and was a chief and continuing source of opposition to liberal thought through the early twentieth century.98Rosenblatt, supra note 76, at 48, 191–92. The Church was also an ardent opponent of liberalism in the United States. Id. at 214–15, 268–71. Indeed, the Church favored the South in the U.S. Civil War, seeing it as “a more traditional and aristocratic society compared to the North”; a papal communication of 1866 pronounced slavery consistent with natural and divine law. Id. at 173. For an appraisal of contemporary varieties of religiously founded antiliberalism, see Richard Schragger & Micah Schwartzman, Religious Antiliberalism and the First Amendment, 104 Minn. L. Rev. 1341, 1413 (2020).

For present purposes, I understand illiberalism as a system of social organization characterized by any of the following clusters of beliefs, which are defined mainly in opposition to the commitments of liberalism. I have weighted these descriptions heavily toward populist versions of illiberalism because they appear to be the most prevalent among nations now undergoing democratic backsliding.99Norris & Inglehart, supra note 5, at 67–68; Mounk, supra note 89, at 2–3.

  1. Individuals are not equal, and consequently there is no general right to popular self-rule or rule by a majority of citizens.100See Jan-Werner Müller, What Is Populism? 21–22 (2016). Instead, only the “right” people, whether a majority or a minority, are entitled to rule, to vote if the system so provides, and to make laws for the governance of society.101Id. at 21–23; Diamond, supra note 6, at 62; Norris & Inglehart, supra note 5, at 66–75; see Mounk, supra note 89, at 43. See generally Stephen Gardbaum, The Counter-Playbook: Resisting the Populist Assault on Separation of Powers, 59 Colum. J. Transnat’l L. (forthcoming 2020).
  2. Civil society is not free, but is controlled strictly by higher laws of religious or traditional origin that establish customary and presumptively unalterable ways of life.102This is the path of Christian nationalism taken in democratically backsliding states like Hungary and Poland. Diamond, supra note 6, at 61–62; Traub, supra note 77, at 205–09.
  3. Leadership is best vested in a strong individual who rules in the name of, and for the benefit of, the people, rightly understood.103See Müller, supra note 100, at 34. Leaders establish law for others, but themselves are constrained only by the obligation to promote the interests of the society as the leaders may, in their wisdom and insight—itself derived from a profound, organic bond with the people—understand those interests.104See id. at 32–38; Mounk, supra note 89, at 41–43.
  4. The exercise of governmental power is not a contingent delegation but an entitlement held by those properly destined to wield it.105Frankenberg, supra note 93, at 18–20. Its exercise thus ought to be neither restrained, impeded, nor questioned.106See id. at 19 (“Convinced that they own the authority they exercise, rulers are likely to assume they are the law and therefore need not answer for their politics pursuant to principles and procedures of accountability.” (emphasis omitted)).
  5. Democracy is unnecessary to societal well-being, and may in fact be inimical to it to the extent that it acts as an impediment to the leader’s pursuit of the good of society, properly understood.107See Robert C. Lieberman et al., The Trump Presidency and American Democracy: A Historical and Comparative Analysis, 17 Persps. on Pol. 470, 471 (2018).

Some flavor for the nature of modern illiberalism may be gleaned from some well-known assertions of modern autocrats. For example, Óscar Benavides, a military dictator of Peru in the early twentieth century, is reputed to have remarked: “For my friend, everything; for my enemies, the law.”108Quoted in Ginsburg & Huq, supra note 6, at 13, but misattributed to the Brazilian dictator Getulio Dornelles Vargas. Clearly, it makes no difference which dictator actually said it. Recep Erdoğan, the present ruler of Turkey, has said: “Democracy is like a streetcar: You ride it until you arrive at your destination, then you step off.”109Id. at 68. Hungary’s autocratic ruler Viktor Órban, recently asserted: “we have to abandon liberal methods and principles of organizing a society, as well as the liberal way to look at the world.”110Csaba Tóth, Full Text of Viktor Orbán’s Speech at Băile Tuşnad (Tusnádfürdő) of 26 July 2014, Budapest Beacon (July 29, 2014), https://budapestbeacon.com/full-text-of-viktor-orbans-speech-at-baile-tusnad-tusnadfurdo-of-26-july-2014 [https://perma.cc/DP42-3YPS]. “Checks and balances,” he remarked a few months later, “is a U.S. invention that for some reason of intellectual mediocrity Europe decided to adopt and use in European politics.”111Gábor Halmai, Illiberalism in East-Central Europe 4 (Eur. Univ. Inst., Working Paper No. 2019/05, 2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3486420. To these well-known capsule summaries of illiberalism we might perhaps add another, a recent assertion of U.S. President Donald J. Trump: “When somebody is the president of the United States, the authority is total.”112See, e.g., James Fallows, 2020 Time Capsule #14: ‘The Authority Is Total, The Atl. (Apr. 16, 2020, 12:08 PM), https://www.theatlantic.com/notes/2020/04/2020-time-capsule-14-the-authority-is-total/609976 [https://perma.cc/GY9C-ELSD].

3. Forms of government

Liberalism and illiberalism are at bottom political philosophies, and their contrasting commitments thus raise the question of whether each requires any particular institutional arrangement or form of government for its successful implementation. Specifically, a lively debate has arisen among political theorists over whether democracy is the only form of government fully compatible with liberalism, and conversely, whether authoritarianism is intrinsically illiberal.113For present purposes, I understand “democracy” to be any system that meets Schumpeter’s minimalist criteria, i.e., a system in which “individuals acquire the power to decide by means of a competitive struggle for the people’s vote.” Joseph A. Schumpeter, Capitalism, Socialism and Democracy 241 (2010). I understand “authoritarianism” to be, in contrast, a system “characterized by the complete absence of effective political competition.” Ginsburg & Huq, supra note 6, at 22; accord Linz, supra note 88, at 159; Mark Tushnet, Authoritarian Constitutionalism: Some Conceptual Issues, in Constitutions in Authoritarian Regimes 45 (Tom Ginsburg & Alberto Simpser eds., 2014) (“[A] dominant party[] makes all relevant public policy decisions, and there is no basis in law for challenging whatever choices the regime makes.”).

As an empirical matter, as Kim Scheppele has observed, “democracy, constitutionalism, and liberalism [have] marched arm in arm through history.”114Kim Lane Scheppele, Autocratic Legalism, 85 U. Chi. L. Rev. 545, 548 (2018). But it is unclear whether that connection is intrinsic or contingent. Some theorists contend that liberalism and democracy are distinct phenomena, and that the existence of one neither requires nor implies the existence of the other.115Fareed Zakaria, The Rise of Illiberal Democracy, 76 Foreign Affs. 22, 22–23 (1997). Fareed Zakaria, for example, has described certain contemporary states as “illiberal democrac[ies]” in which elections are more or less free and fair, and thus count as “democratic,” but in which “what happens after the elections” flouts liberal principles, sometimes egregiously.116Id. at 23. Yascha Mounk similarly sees liberalism and democracy as two distinct phenomena, and argues that they are in many places “[d]econsolidating,” producing either “[d]emocracy without [r]ights” or “[r]ights without [d]emocracy.”117Mounk, supra note 89, chs. 1–3.

Others insist that any kind of meaningful liberalism is impossible without democracy. Gábor Halmai, for example, argues “not only [that] democracy presupposes liberalism, but [that] there is no liberalism without democracy.”118Halmai, supra note 111, at 10. Jan-Werner Müller claims that “illiberal governments are inherently undemocratic: ‘if opposition parties have been hindered in making their case to the electorate, and journalists do not dare to report on the government’s failures, the ballot boxes have already been stuffed.’”119Mounk, supra note 89, at 51 (quoting Jan-Werner Müller, The Problem with ‘Illiberal Democracy,’ Soc. Eur. (Jan. 27, 2016), https://www.socialeurope.eu/the-problem-with-illiberal-democracy [https://perma.cc/58SK-Q9SE]). Mounk, of course, disagrees with this assessment. Id. at 52. Similarly, according to Larry Diamond, “all the political systems that protect liberty are democracies.”120Diamond, supra note 6, at 6; see also Kim Quaile Hill, Democracy in the Fifty States 128 (1994) (undertaking a study of the degree of democratization in the American states and concluding that “[s]tates that are more democratic ensure a greater range of individual civil rights”).

Whatever their positions, however, no political theorists seem to deny that liberalism and democracy most commonly travel together, as do illiberalism and authoritarianism; nor has any theorist to my knowledge gone so far as to deny that liberal democracy and illiberal authoritarianism remain the modal—and polar—arrangements. Thus, at the very least, it is reasonable to assume that where we find authoritarianism, we are overwhelmingly likely, if perhaps not certain, to find illiberalism, and I shall proceed on that assumption throughout the balance of this Article.

B. The Recent Turn to Authoritarianism

For most of human history, democracy was exceedingly rare. Following the American and French revolutions, it began to attract a following around the world, and over the course of the nineteenth century, possession of the franchise gradually became more common and widespread.121Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century 17 (1991). This democratizing trend reversed during the 1930s, when illiberal and authoritarian forms of government enjoyed a robust efflorescence,122Diamond, supra note 6, at 42–43. but reversed again following the Second World War.123Id. at 43. The defeat of powerful fascist states by an alliance of liberal democracies badly discredited illiberal regimes, initiating a so-called “second wave” of democratic expansion.124Joshua Kurlantzick, Democracy in Retreat: The Revolt of the Middle Class and the Worldwide Decline of Representative Government 35 (2013). A third wave of new democracies appeared after 1989, when it seemed to much of the world as though liberal democracy had, after a vigorous struggle with communism, proved itself the more successful form of government.125See Diamond, supra note 6, at 43­–53; Kurlantzick, supra note 124, at 53–54 (counting this as a “fourth wave”). By the late 1990s, “most of the countries on Earth were democracies . . . and a third of all states were fairly liberal.”126Diamond, supra note 6, at 53.

Today, however, there has been a precipitous turn globally away from liberal democracy and toward authoritarian forms of governance. In Hungary, Poland, Turkey, Venezuela, and many other places, young but seemingly stable democracies have turned to authoritarian leaders who have dismantled liberal institutions.127The stories of these nations’ fall into authoritarianism has been told frequently. For a particularly detailed account of the Hungarian experience, see Petra Bárd & Laurent Pech, How to Build and Consolidate a Partly Free Pseudo Democracy by Constitutional Means in Three Steps: The ‘Hungarian Model,’ (Reconnect, Working Paper No. 4, 2019). The Polish case is related in, among many other places, Traub, supra note 77, at 201–09. Brief overviews of the progressions in Venezuela and Turkey may be found in Ginsburg & Huq, supra note 6, at 45–48. Pro-democratic uprisings in the Arab world and Eastern Europe sputtered and resulted mainly in the further consolidation of illiberal autocracy.128See Diamond, supra note 6, at 54. Even more troubling is the new phenomenon of “democratic backsliding,” in which longstanding, stable democracies have suffered slow, incremental degradation of previously reliable democratic institutions.129Nancy Bermeo, On Democratic Backsliding, 27 J. Democracy 5, 5, 14 (2016); see Ginsburg and Huq, supra note 6, at 43–47; Diamond, supra note 6, at 55. According to the Bertelsmann Foundation’s “Transformation Index,” the number of “highly defective democracies” around the world doubled between 2006 and 2010.130Kurlantzick, supra note 124, at 8–9. In 2016, The Economist’s “Democracy Index” downgraded the United States to a “flawed democracy.”131Democracy Index 2016, Economist Intel. Unit (2016), https://www.eiu.com/public/topical_report.aspx?campaignid=DemocracyIndex2016 [https://perma.cc/5ZZJ-3F8V]. Thirty-five percent of the world’s population now lives in “autocratizing nations.”132Seraphine F. Maerz et al., State of the World 2019: Autocratization Surges—Resistance Grows, 27 Democratization 909, 909 (2020).

Most worrisome of all, however, is that democratic decline, far from being something imposed on an unwilling public, appears not only to bear the public’s approval where it has occurred, but to be responsive to a significant shift in public opinion.133See Bermeo, supra note 129, at 14. Unlike its twentieth-century predecessors, today’s authoritarian takeover typically is initiated not by a military coup or naked electoral fraud on a massive scale, but following an open democratic process in which an illiberal party wins power under reasonably fair rules of democratic competition before entrenching itself in power through anti-democratic means.134See id. at 10–11. The modern movement to authoritarianism, in other words, is in many cases majoritarian and largely populist in nature.135See Norris & Inglehart, supra note 5, at 9–11; see also Müller, supra note 100, at 36 (“[P]opulist parties are particularly prone to internal authoritarianism.”).

So common has this form of incremental democratic backsliding become that comparativists have been able to identify a typical series of moves that authoritarian leaders gradually make to consolidate power by undermining liberal democratic institutions—an “authoritarian’s playbook,” as it were.136Diamond, supra note 6, at 64–65. Ginsburg and Huq, for example, list “five specific mechanisms by which democratic erosion unfolds”137Ginsburg & Huq, supra note 6, at 72.:

[1] the use of constitutional amendments to alter basic governance arrangements; [2] the elimination of checks that operate between different branches; [3] the centralization and politicization of executive power as exercised through the bureaucracy; [4] the contraction or distortion of a shared public sphere in which liberal rights of speech and association can be exercised; and [5] the elimination or suppression of effective partisan political competition and the related prospect of rotation out of elected office.138Id. at 72–73.

Similarly, András Jakab’s “handbook for dictators” lists the following conventionally pursued tactics:

(1) [E]mploy populist rhetoric against institutions of rule of law and old elites . . .

(2) [Dismantle the capacity of the] civil service [to] . . . provid[e] neutral information and [to follow] regulated, pre-determined, work processes . . .

(3) [E]limination of judicial review . . .

(4) The intentional and conscious erosion of the public sphere . . . as an epistemic and discursive basis for democracy . . .

(5) [C]odify authoritarian centralization attempts at the highest level of the law [i.e., the constitution] . . .

(6) Attacks against NGOs . . .

(7) [E]limination or suppression of an effective political competition . . .

(8) [A]cademic freedom and the autonomy of universities and research institutes shall also be restricted . . . .139András Jakab, What Can Constitutional Law Do Against the Erosion of Democracy and the Rule of Law? 7–8 (Max Planck Inst. Compar. Pub. L. & Int’l L., No. 15, 2019), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3454649.

These tactics, in short, effectuate the degree of “social control” necessary to achieve the authoritarian’s goal of “the possession and preservation of power of and by dominant individuals and groups.”140Lynette J. Chua, Legal Mobilization and Authoritarianism, 15 Ann. Rev. L. & Soc. Sci. 355, 357 (2019).

To say that the rise of authoritarianism in many places has come with popular approval is not, of course, to say that such approval—and the illiberalism that underlies it—is universal. Far from it. Even in a recent sham Russian plebiscite to ratify the lifting of term limits applicable to Vladimir Putin—a decision that had previously been taken by the government and at the time of the plebiscite already had the force of law—more than 20% of those who bothered to show up voted no.141Andrew Higgins, The Theatrical Method in Putin’s Vote Madness, N.Y. Times (July 1, 2020), https://www.nytimes.com/2020/07/01/world/europe/putin-referendum-vote-russia.html. In assessing the impact of illiberalism on official action, the critical factor is not the presence of illiberal opinion within the population, but its distribution.

In any given population, it is of course possible, though unlikely, that illiberalism is distributed haphazardly. But whatever its distribution, the translation of illiberal public opinion into illiberal or authoritarian action by definition requires political mobilization. Whenever illiberalism has organized or mobilized itself for political action, it has been concentrated in one of two ways.

First, illiberalism can be concentrated in a political party. Thus, in France, for example, a strongly liberal state, support for right-wing populism is mobilized mainly through the National Front, which in recent years has earned about 14% of the vote nationwide.142Fiorina, Unstable Majorities, supra note 5, at 154. In Denmark, another thoroughly liberal state, political illiberalism is mobilized and concentrated mainly in the Danish People’s Party, which earns a similar percentage of the vote.143Id. In Hungary and Poland, where illiberal sentiment is more widespread, it is nevertheless concentrated in the governing Fidesz Party and the governing Law and Justice Party, respectively.144Norris & Inglehart, supra note 5, at 242–45.

Second, illiberalism can be concentrated in a region. In unitary states, this kind of concentration may be of little importance for a politics that is conducted predominantly at the national level. In a federal state, however, geographically concentrated illiberalism can have significant ramifications, not only for governance in the affected subnational units, but for the nation as a whole.145See infra Part III. The next Part provides a brief overview of the history and scope of American illiberalism, and demonstrates its present concentration in the Republican Party. Part IV turns to the issue of regional concentration of illiberalism in particular states.

III. Illiberalism in the United States

A. The American Tradition of Liberalism

The idea that Americans share a profound, longstanding, universal commitment to liberalism is traceable in large part to Louis Hartz’s 1955 classic, The Liberal Tradition in America.146Hartz, supra note 77. Widely recognized as “one of the most influential books on American history published in the twentieth century,”147Mark Hulliung, Louis Hartz, His Day and Ours, in The American Liberal Tradition Reconsidered: The Contested Legacy of Louis Hartz 11 (Mark Hulliung ed., 2010) [hereinafter Contested Legacy]. and “the dominant interpretative text in American political thought for a generation,”148Philip Abbott, Still Louis Hartz After All These Years: A Defense of the Liberal Society Thesis, 3 Persp. on Pol. 93, 93 (2005); accord, e.g., Brian J. Glenn, Louis Hartz’s Liberal Tradition in America as Method, 19 Stud. Am. Pol. Dev. 234, 234 (2005) (stating that the book “deserves to rank among the canon of political science”). the book argued that the liberal philosophy of John Locke had not merely influenced the founding generation, but had provided the template for an enduring national consensus on political first principles—in Hartz’s words, “[a] Lockian creed.”149Hartz, supra note 77, at 9. This creed, Hartz argued, follows Locke in taking as its centerpiece a “basic social norm, the concept of free individuals in a state of nature.”150Id. at 60. So completely did Lockean thought infuse American political beliefs, Hartz argued, that what is called the “American Way of Life” is better conceived as “a nationalist articulation of Locke.”151Id. at 11.

Most importantly for present purposes, a consequence of the Hartzian thesis is that “domestic struggles” occurring in the forum of American politics “have all been projected with the setting of Western liberal alignments.”152Id. at 15. That is to say, it follows from the depth of the American consensus on liberal first principles that political disagreements among Americans necessarily take place within the domain of liberal thought.153Hartz is of course not the only thinker to take this position. See, e.g., Michael J. Sandel, Democracy’s Discontent: America in Search of a Public Philosophy 4–5 (1996) (“[L]iberalism describes a tradition of thought . . . that runs from John Locke, Immanuel Kant, and John Stuart Mill to John Rawls. The public philosophy of contemporary American politics is a version of this liberal tradition of thought, and most of our debates proceed within its terms.”). Disagreements thus are internal to liberalism, not between liberalism and other, competing belief systems.154Id.

Hartz’s analysis has had a profound impact. It “has had no rivals as the most influential book in the study of American political thought.”155Alan Gibson, Louis Hartz and Study of the American Founding: The Search for New Fundamental Categories, in Contested Legacy, supra note 147, at 149. It still finds supporters today,156See, e.g., Sanford Lakoff, Liberalism in America: Hartz and His Critics, 8 Critical Rev. Int’l Soc. & Pol. Phil. 5, 5 (2005). often among political scientists,157For arguments against the existence of any fundamental polarization of values among the American electorate, see, e.g., Morris P. Fiorina, Culture War? The Myth of a Polarized America (2005); Fiorina, Unstable Majorities, supra note 5. But see James E. Campbell, Polarized: Making Sense of a Divided America 1 (2016). and clearly furnishes the unifying assumption for the three schools of thought, identified earlier, concerning the nature of subnational variation in the United States—i.e., that all subnational variation occurs within a domain bounded by a shared American commitment to liberalism.158See supra Section I.B.

Nevertheless, The Liberal Tradition in America has also attracted a good deal of criticism.159Hartz was a political scientist and political theorist, and much of the critique has been mounted by historians, who have called the book “dogmatic,” Contested Legacy, supra note 147, at 11, 13; “deficient in important respects,” Rogers M. Smith, Civic Ideals: Conflicting Visions of Citizenship in U.S. History 17 (1997); and even “mistaken in its general argument about American consensus and in almost every one of its particular arguments as well,” James T. Kloppenberg, The Virtues of Liberalism 11 (1998). Hartz’s critics tend to tell a very different story about the development of political thought in the United States. On their account, although liberalism without question figured prominently in American thought, it was only one of “a number of strands in American public life,”160James T. Kloppenberg, Requiescat in Pacem: The Liberal Tradition of Louis Hartz, in Contested Legacy, supra note 147, at 92. and it existed side by side with other, deeply illiberal strands of political thought.

At its strongest, the anti-Hartzian account identifies persistent, deeply embedded strands of American thought that are exclusionary,161Smith, supra note 159, at 17; Judith N. Shklar, American Citizenship: The Quest for Inclusion 13 (1991). nativist,162Smith, supra note 159, at 139–40. and even feudal in origin,163Karen Orren, Belated Feudalism: Labor, the Law, and Liberal Development in the United States 1–3 (1991). and that lie at such a remove from liberalism as to generate recurring disputes within American politics that are best characterized as deep-seated “disagreements over forms of life” and “ultimate ends” rooted in “rival political cultures.”164Richard J. Ellis, American Political Cultures 151 (1993) [hereinafter Ellis, American Political Cultures]. Such disputes, that is to say, spill over the boundaries of democratic liberalism, pitting it against entirely different systems of thought. But even Hartz’s less extreme critics agree that a persistent, socially and politically significant illiberal theme in many aspects of American life has been hierarchy—hierarchies of race, gender, ethnicity, and religion.165Smith, supra note 159, at 17; Kloppenberg, supra note 160, at 99–100; Richard J. Ellis, The Liberal Tradition in an Age of Conservative Power and Partisan Polarization, in Contested Legacy, supra note 147, at 214 [hereinafter Ellis, Liberal Tradition]; James T. Kloppenberg, In Retrospect: Louis Hartz’s The Liberal Tradition in America, 29 Revs. Am. Hist. 460, 463 (2001). Such hierarchies are illiberal not only in their denial of the foundational liberal tenet of human equality, but in their embrace of the concept of permanent, inherited social stations from which little or no deviation is possible except at profound risk to the legitimate social order.166Smith, supra note 159, at 17.

None of this, to reiterate, means that liberalism has not been deeply influential in American thought. But, if true, it means that liberalism has in fact not been universally embraced, and indeed has been opposed periodically, sometimes successfully, by adherents of illiberal belief systems that have always existed within a broader and more diverse sphere of American political thought than Hartz imagined, sometimes gaining power and at other times losing it. In the words of Judith Shklar: “America has not marched single file down a single straight liberal highway . . . . What has been continuous is a series of conflicts arising from enduring anti-liberal dispositions that have regularly asserted themselves, often very successfully, against the promise of equal political rights.”167Shklar, supra note 161, at 13.

B. How Liberal Are Americans?

It is beyond doubt that illiberalism has from time to time appeared on the American political stage—not just informally, in private living rooms or around water coolers, but publicly, in organized fashion. There is, after all, an American Nazi Party.168See American Nazi Party, https://www.americannaziparty.com [https://www.americannaziparty.com]. The Communist Party USA fielded a presidential candidate in every election between 1972 and 1984.169See U.S. Presidential Elections: Leftist Votes, Marxists, https://www.marxists.org/history/usa/government/elections/president/timeline.htm [https://perma.cc/BT9J-GM4U]. Overt segregationists ran for President in 1948 and 1968, and carried, respectively, four and five states.170South Carolina Governor Strom Thurmond (1948), running as the candidate of the segregationist “Dixiecrat” wing of the Democratic Party, received more than one million votes and carried Alabama, Louisiana, Mississippi, and South Carolina. Kari Frederickson, The Dixiecrat Revolt and the End of the Solid South, 1932–1968 184, 191 (2001). Alabama Governor George Wallace (1968), ran for President on a segregationist platform as the candidate of the American Independence Party, carrying Alabama, Arkansas, Georgia, Louisiana, and Mississippi. Leada Gore, George Wallace 1968 Presidential Run: ‘Most Influential Loser’ in Political History, Al.com (July 25, 2019), https://www.al.com/news/erry-2018/08/937bc749e09952/george-wallace-1968-presidenti.html [https://perma.cc/SV6A-TL7D]. However, whether due to the pervasive influence of the Hartzian thesis or simply a persistently Whiggish sense of their own history, Americans seem to have a tendency to think of domestic illiberalism as something fundamentally confined—confined to the past, confined to a region, confined to the political margins—to “the lunatic fringe,” as the New York Times labeled the John Birch Society in 1961.171Editorial, John Birch Fantasies, N.Y. Times, Apr. 22, 1961, at 24. The term apparently was coined by Theodore Roosevelt. Theodore Roosevelt, History as Literature ch. 11 (1913). A mass of social science evidence, however, suggests otherwise.

In 1954, at the height of the Cold War, Samuel Stouffer, a Harvard sociologist, conducted a large and wide-ranging survey aimed at determining the degree to which Americans supported the application of core civil liberties to unpopular groups—socialists, communists, and atheists. In a book published the following year, Stouffer revealed some shocking conclusions.172Samuel A. Stouffer, Communism, Conformity, and Civil Liberties: A Cross-Section of the Nation Speaks Its Mind (1955). Nearly one-third of respondents thought a socialist advocating nationalization of industry should not be allowed to speak in their community.173Id. at 29. Sixty percent believed that an atheist should not be permitted to speak, 60% believed a book written by an atheist should be removed from public libraries, and 84% said that an atheist should not be permitted to teach in a college or university.174Id. at 32–33. Two-thirds said a book by a communist should be removed from a library, and 91% said that a communist high-school teacher should be fired.175Id. at 40. Seventy-seven percent thought the citizenship of an admitted communist should be revoked, and more than half thought he should be jailed.176Id. at 43–44. Nearly two-thirds thought that the government should be able to eavesdrop on private telephone conversations to gather evidence of communism.177Id. at 44.

Other researchers obtained similar results in numerous follow-up studies.178See, e.g., James W. Prothro & Charles M. Grigg, Fundamental Principles of Democracy: Bases of Agreement and Disagreement, 22 J. Pol. 276, 293 (1960). See generally Clyde Z. Nunn et al., Tolerance for Nonconformity (1978); John L. Sullivan et al., Political Tolerance and American Democracy (1982). A full thirty years later, for example, Herbert McCloskey and John Zaller found that 70% still believed a community should not allow an atheist group to use the civic auditorium; 79% supported prayer and religious observances in public schools; and 54% believed censorship of book publication an appropriate strategy to protect public morality.179Herbert McCloskey & John Zaller, The American Ethos: Public Attitudes toward Capitalism and Democracy 25, 31, 39 (1984). Forty-nine percent—half of all respondents—agreed that “[w]hen it comes to the things that count most, all races certainly are not equal.”180Id. at 70. Although 97% affirmed their belief in freedom of speech, 76% said that fascists and communists should not be allowed to hold meetings or to express their views in the community.181Id. at 38.

Even today, more than three decades after McCloskey and Zaller’s follow-up study, investigators continue to document similar public attitudes. In a 2015 poll, for example, researchers asked the following question: “Thomas Jefferson wrote [in the Declaration of Independence] that ‘[A]ll men are created equal.’ Do you believe this statement is true or false?”18260 Minutes/Vanity Fair Poll: Inequality, CBS News (Aug. 3, 2015), https://www.cbsnews.com/news/60-minutes-vanity-fair-poll-inequality-america [https://perma.cc/S3PZ-S3MH]. Twenty-two percent of Americans answered “false.”183Id. In a 2017 survey of college-educated Americans, about half agreed that speakers advancing various kinds of unpopular views—ranging from the contention that whites are racist to the view that Muslims should be barred from immigrating to the United States, and from assertions of the backwardness of Christians to claims about the superiority of men over women at math—should not be allowed to speak on college campuses.184Emily Ekins, The State of Free Speech and Tolerance in America, Cato Inst. (Oct. 31, 2017), https://www.cato.org/survey-reports/state-free-speech-tolerance-america [https://perma.cc/MR26-7V9P]. In another recent poll, one in three expressed opposition to the separation of church and state.185See Daniel Cox et al., What It Means to Be American: Attitudes in an Increasingly Diverse America Ten Years After 9/11, Brookings Inst. (Sept. 6, 2011), https://www.brookings.edu/research/what-it-means-to-be-an-american-attitudes-in-an-increasingly-diverse-america-ten-years-after-911 [https://perma.cc/A9ZB-ZH9N]. Forty-one percent expressed discomfort with a Muslim teaching elementary school.186Id. Nearly one-third believed that the aim of Muslims in the United States is to establish Shari’a law as the governing legal regime for all Americans.187Id.

These studies generally support two kinds of conclusions. One is that Americans by and large support civil liberty in the abstract, but are willing to tolerate the exercise of such liberties only selectively, and in particular, not by “groups and ideologies they [find] most objectionable.”188J.L. Sullivan & J.E. Transue, The Psychological Underpinnings of Democracy: A Selective Review of Research on Political Tolerance, Interpersonal Trust, and Social Capital, 50 Ann. Rev. Psych. 625, 633 (1999). In other words, Americans support the exercise of civil liberties except in the only situations in which the protection afforded by such liberties would matter.189These results seem consistent with more recent, unrelated research demonstrating that Americans dislike political conflict, and that the reason they dislike it is that they tend to think that all problems have clear and obvious solutions, and consequently that disagreement can only be explained only by bad faith or corruption. John R. Hibbing & Elizabeth Theiss-Morse, Stealth Democracy: Americans’ Beliefs About How Government Should Work 33 (2002). Civil liberties, evidently, are for me and my group, not for you and yours.

Relatedly, these studies suggest that a latent illiberal intolerance can be readily activated, even in citizens otherwise predisposed to tolerance, by a perception of threat,190Sullivan & Transue, supra note 188, at 632–33. or by priming associated with some group perceived as an “enemy.”191James L. Gibson, Intolerance and Political Repression in the United States: A Half-Century After McCarthyism, 52 Am. J. Pol. Sci. 96, 98 (2008). Thus, for example, studies of intolerance find American intolerance more or less stable over time, but focused on different groups in different periods—communists in the 1950s, but “Radical Muslims” today (with atheists and communists close behind, even after fifty years).192Id. at 101–03.

Second, these studies have consistently shown that elites—community leaders, business leaders, leaders of trade and bar associations, party and elected officials—as a group display greater commitment to civil liberties and greater tolerance for political deviance than ordinary people.193E.g., Stouffer, supra note 172, ch. 2; Sullivan et al., supra note 178, at 194–202. For a somewhat more skeptical view, see Mark Peffley & Robert Rohrschneider, Elite Beliefs and the Theory of Democratic Elitism, in The Oxford Handbook of Political Behavior 66, 71–72 (Russell J. Dalton & Hans-Dieter Klingemann eds., 2007). This in turn suggests of course that the comparatively strong record of the United States on civil liberties since the late twentieth century results principally from the efforts of governing elites rather than of the citizenry, and that more robust and responsive democracy—of, say, the populist kind now emerging—is likely to produce less rather than more liberal results. Indeed, one factor commonly accepted as contributing to Donald Trump’s political rise is the transfer of control over presidential nomination, beginning in the 1970s, from party elites to the party rank and file.194See Elaine C. Kamarck, Returning Peer Review to the American Presidential Nomination Process, 93 N.Y.U. L. Rev. 709, 710, 712 (2018).

If a substantial minority of the American public holds views on the merits of public policy that are illiberal—or that, if liberal, are readily abandoned under minimal pressure—a similarly substantial minority also demonstrates a declining commitment to democracy. In a recent study, Yascha Mounk found that the percentage of Americans who thought it “essential” to live in a democracy declined from 71% among those born in the 1930s to only 29% among those born in the 1980s.195Mounk, supra note 89, at 105. One in four millennials believes that democracy is a “bad” or “very bad” system of government.196Id. at 107. In 1995, Mounk reports, “34 percent of young Americans aged 18–24 felt that a political system with a strong leader who does not have to bother with Congress or elections was either good or very good. By 2011, 44% of young Americans felt the same way.”197Id. at 109. Support in the United States for direct military rule rose from 8% of 18-to-24-year-olds in 1995 to 24% in 2011, and for all ages rose from 7 to 16 percent.198Id. at 110.

In a similar study, Larry Diamond obtained less dire results, but concluded that “we still have serious cause for concern.”199Diamond, supra note 6, at 149. Diamond’s study found overall American support for democracy still strong—somewhere between about 75% and 85% for most formulations of the question200Id. at 149–50.—but found that the proportion of Americans who support “a strong leader who does not have to bother with Congress and elections” was 24%, much higher than in Canada, France, and Germany.201Id. at 149. Support among Americans for military rule in the United States rose from 8% to 18% between 1995 and 2017.202Id. Only 54%, Diamond found, “consistently [hold] a prodemocracy position,” and 28% “gave a nondemocratic response” on two of five items on which he surveyed.203Id. at 151; see also Matthew H. Graham & Milan W. Svolik, Democracy in America? Partisanship, Polarization, and the Robustness of Support for Democracy in the United States, 114 Am. Pol. Sci. Rev. 392, 393 (2020) (finding that Americans overwhelmingly will not cross party lines to punish a co-partisan candidate who adopts undemocratic positions).

Let us suppose, then, that some substantial minority of Americans holds views that are either outright inconsistent with the nation’s historical, publicly professed commitment to liberal democracy, or, if nominally liberal, can be easily moved to an illiberal position. What kind of risk, if any, to longstanding, national, liberal democratic commitments and institutions might be posed by such a condition? The answer depends in great measure on how the holders of such views are distributed. If they are dispersed randomly, or at least unsystematically, then, as Madison observed, “relief is supplied by the republican principle”204The Federalist No. 10 (James Madison).—they will be everywhere a minority, and everywhere outvoted.

Illiberalism in the United States, however, is not distributed randomly. As is the case elsewhere, it tends to be concentrated in a particular political party, in this case the Republican Party.205See infra Section III.C. It is also concentrated geographically. The next Section reviews briefly the growing illiberalism and authoritarianism of the Republican Party. The Part immediately following examines in some detail the concentration of illiberalism geographically, in particular states. As will be seen, they are states governed, logically enough, by Republicans.

C. Illiberalism in the Republican Party

Americans have for some time now been wringing their hands about the “polarization” of the two main political parties.206See, e.g., Campbell, supra note 157, at 2–3. The concept of “polarization,” however, by definition presupposes a single spectrum along which two polarizing things may both be placed. In the common conceptual apparatus of political science, the relevant spectrum is one running from conventional notions of political liberalism on the left to conventional notions of political conservatism on the right. It is rarely remarked, however—and for most of the last six decades it would have been unnecessary to remark—that this spectrum is located entirely within the bounds of philosophical liberalism. The present question, in contrast, is altogether different: does the concept of “polarization” any longer apply? Has the Republican Party merely moved to the right on a scale that still counts as liberal? Or has it fallen off the liberal spectrum altogether—has it become, in Paul Krugman’s characterization, “an authoritarian regime in waiting?”207Paul Krugman, Arguing with Zombies: Economics, Politics, and the Fight for a Better Future 346 (2020).

The boundary between democracy and authoritarianism is hazy,208Ginsburg & Huq, supra note 6, at 21–22; Ozan O. Varol, Stealth Authoritarianism, 100 Iowa L. Rev. 1673, 1681 (2015); Norman W. Spaulding, States of Authoritarianism in Liberal Democratic Regimes, in García & Frankenburg, supra note 93, at 267. and that between liberalism and illiberalism even more so, and it can therefore be difficult to identify the precise point at which a party or leader slides definitively, in the most common contemporary pathway, “from hardball democrat to legalistic autocrat.”209Scheppele, supra note 114, at 581. Nevertheless, especially since 2016, the evidence has mounted to the point where it is possible for serious political scientists to proclaim that, in the United States, there has been a “grafting of authoritarian behavior onto one of the two major parties,”210Lieberman et al., supra note 107, at 471. and that “the dangers of [democratic] backsliding are closely related to developments at work in the contemporary Republican Party.”211Paul Pierson & Eric Schickler, Madison’s Constitution Under Stress: A Developmental Analysis of Political Polarization, 23 Ann. Rev. Pol. Sci. 37, 53 (2020)./mfn]

Any consideration of this question must begin with the basic fact that Donald Trump now heads the Republican Party, and that his behavior, first as a candidate and then as President of the United States, was breathtaking not only in his profound rejection of long-established norms of liberal democracy, but in the ferocity of his assault on their legal and institutional foundations. Others have exhaustively documented Mr. Trump’s offenses against liberal democracy; here is just one such account:

Does President Trump fit the authoritarian-populist classification? We think so. His speeches feature a mélange of xenophobic fear-mongering and Islamophobia, narcissism, misogyny and racism, conspiracy theories (“millions of fraudulent votes”), and isolationist “America First” policies. It is in his legitimation of authoritarian values that Trump represents the gravest threat to American democracy with his equivocal treatment of neo-Nazi and white supremacist hate groups, his open approval of some of the world’s most repressive regimes, attacking the press and using Twitter to slam “fake news,” seeking border limits on migrants from Muslim-majority countries and promising to build a wall to keep out Mexicans, casting doubts on the integrity of American elections and the independence of the judiciary, prioritizing military security and American jobs over defending democracy and human rights abroad, weakening multilateral cooperation and international conventions, and disparaging the rule of law. Like other authoritarian rulers, he shows a casual disregard for the truth and a willingness to challenge the legal constraints on his powers. Many of the tensions observed in his administration arise from his rejection of democratic restraints.211Norris & Inglehart, supra note 5, at 245–46 (footnote omitted).

And this summary does not even mention his attacks on the loyalty and integrity of the federal civil service,212See Lieberman et al., supra note 107, at 470; Neil S. Siegel, Political Norms, Constitutional Conventions, and President Donald Trump, 93 Ind. L.J. 177, 195 (2018). his interference with law enforcement activities directed against himself and his associates,213Levitsky & Ziblatt, supra note 6, at 177–78. his nepotism in political appointments and attempts to profit personally from official activities,214Timothy K. Kuhner, American Kleptocracy: How to Categorise Trump and His Government, 28 King’s L.J. 201, 204–05 (2017); David Frum, Trumpocracy: The Corruption of the American Republic 53 (2018). his denial of the authority of Congress to oversee or even to question his activities,215Trump v. Mazars, 140 S. Ct. 2019, 2026 (2020). or his attempts to manipulate foreign powers into interfering with American democratic processes for his own advantage.216Impeaching Donald John Trump, President of the United States, for high crimes and misdemeanors, H.R. Res. 755, 116th Cong. (2019).

None of this is in issue. The real concern for present purposes is the response of the Republican Party. A “conservative” party in the traditional American sense—one still committed to liberal democracy217Ellis, Liberal Tradition, supra note 165, at 207.—might have resisted Mr. Trump’s behavior by pushing back from its bases in Congress and the judiciary, or at least complaining publicly. The Republican Party did nothing of the sort: it fell compliantly into line.218See Lieberman et al., supra note 107, at 476. By the time it was clear that Trump would be the nominee, leading Republicans, including the President pro tem of the Senate and the Speaker of the House, endorsed and then “closed ranks behind” Donald Trump, which “normalized” his election.219Levitsky & Ziblatt, supra note 6, at 70. Once Mr. Trump took office, Republicans responded with loyalty; the result has been that “Trump’s deviance has been tolerated by the Republican Party, which has helped make it acceptable to much of the Republican electorate.”220Id. at 201. “Acceptable” is an understatement: for virtually every week of his presidency, Trump’s job approval among rank-and-file Republicans was in the upper eighty- to lower ninety-percent range.221Presidential Approval Ratings—Donald Trump, Gallup, https://news.gallup.com/poll/203198/presidential-approval-ratings-donald-trump.aspx [https://perma.cc/TG3Z-6M9A]. At the same time, Republican office-holders were unwilling to use congressional power or other institutional levers to control even Mr. Trump’s worst excesses.222Lieberman et al., supra note 107, at 473, 476. Only Mr. Trump’s incitement of a mob to disrupt congressional ratification of the results of Electoral College voting seems to have crossed some kind of behavioral red line for the great majority of Republicans. See Brian Bennett, After President Trump Incited a Riotous Mob, Will He Face Any Consequences?, TIME (Jan. 7, 2021, 7:41 AM), https://time.com/5927361/donald-trump-incited-capitol-consequences.

Historically, the Republican Party has been “conservative” in the sense of supporting policies that lie at the politically conservative end of the liberal policy scale—policies such as free markets, free trade, limited government, and so forth. Today, however, “the fact that conservative and establishment Republicans represent the president’s strongest supporters lends indirect evidence of a realignment.”223Karyn Amira et al., Adversaries or Allies? Donald Trump’s Republican Support in Congress, 17 Persp. on Pol. 756, 767 (2019). This realignment is clearly one that takes the party “away from its conservative roots,”224Id. and the final destination appears to be some form of illiberal authoritarianism.

This is not necessarily a sudden development. As historian Joseph Ellis observed a decade ago, “[t]o reduce the Republican party to its antitax, antistate rhetoric and policies is to miss the authentically nonliberal and antiliberal elements in the party’s creed.”225Ellis, Liberal Tradition, supra note 165, at 209. Political scientists have for some time been documenting the asymmetry of the divergence between the two major parties.226Pierson & Schickler, supra note 211, at 51. The growing chasm in American politics does not, however, result from each of the parties purifying itself ideologically and consequently moving further, in tandem, from the political center.227Id. In fact, “[t]he Republican Party has made a much more pronounced shift toward extremism.”228Id.; see also Jacob S. Hacker & Paul Pierson, Off Center: The Republican Revolution and the Erosion of American Democracy 6 (2005); Joseph Fishkin & David E. Pozen, Asymmetric Constitutional Hardball, 118 Colum. L. Rev. 915, 940–43 (2018). The reasons for this shift need not concern us here, though, as Richard Schragger and Micah Schwartzmann observe, “[t]he evangelicalization of the Republican Party” is doubtless an important factor.229Schragger & Schwartzman, supra note 98, at 1413. According to James Traub, “Republicans released the germ of illiberalism into the national political bloodstream” by courting conservative evangelicals. Traub, supra note 77, at 9. Indeed, Rachel Marie Blum and Christopher Sebastian Parker argue that the fundamental project of the Republican Party has evolved to become something “very different from traditional conservatism”; the central mission of Republicans now focuses principally on “efforts to maintain the prestige of their group”230Rachel Marie Blum & Christopher Sebastian Parker, Trump-ing Foreign Affairs: Status Threat and Foreign Policy Preferences on the Right, 17 Persp. on Pol. 737, 739 (2019).—a fundamentally inegalitarian, and thus thoroughly illiberal, political telos.

Pursuit of an illiberal political telos may naturally enough demand illiberal methods, and thus “[t]he GOP’s efforts to cling to power by coercion instead of persuasion”231Yoni Appelbaum, How America Ends, The Atl. (Dec. 2019), https://www.theatlantic.com/magazine/archive/2019/12/how-america-ends/600757 [https://perma.cc/5752-8WNK]. should not be entirely surprising. As David Frum observed in 2018, “[i]f conservatives [i.e., Republicans] become convinced that they cannot win democratically, they will not abandon conservatism. They will reject democracy.”232Frum, supra note 215, at 206. That prediction has been accurate. As we will see in the next Part, the principal, though by no means the only, manifestation of Republican Party illiberalism has been its persistent efforts to subvert democratic processes so as to avoid sharing power with—much less being ruled by—the opposition party.233In a recent article, Larry Bartels identifies racial and ethnic antagonism as strong motivating factors in Republicans’ abandonment of the party’s historical commitment to democracy. Larry M. Bartels, Ethnic Antagonism Erodes Republicans’ Commitment to Democracy, 117 Proc. Nat’l Acad. Scis. 22752, 22752, 22754, 22756 (2020).

Leading Republicans have been increasingly—and shockingly—vocal and transparent about this goal. Nearly twenty years ago, Paul Weyrich, a leading Republican activist and founder of both the Moral Majority and the Heritage Foundation, proclaimed: “‘I don’t want everybody to vote.’ The Republican Party’s ‘leverage in the elections quite candidly goes up as the voting populace goes down.’”234Carol Anderson, One Person, No Vote: How Voter Suppression Is Destroying Our Democracy 48 (2018). More recently, a sitting Republican United States Senator, addressing a small crowd of supporters on the topic of voting by college students, told them: “[T]here’s a lot of liberal folks in those other schools who maybe we don’t want to vote. Maybe we want to make it just a little more difficult. And I think that’s a great idea.”235Paul Waldman, GOP Senator: Voting Should Be ‘a Little More Difficult’ . . . If You’re a Democrat, Wash. Post (Nov. 16, 2018, 10:04 AM), https://www.washingtonpost.com/blogs/plum-line/wp/2018/11/16/gop-senator-voting-should-be-a-little-more-difficult-if-youre-a-democrat. A Republican member of Congress openly praised property qualifications for voting.236Jed Handelsman Shugerman, Hardball vs. Beanball: Identifying Fundamentally Antidemocratic Tactics, 119 Colum. L. Rev. Online 85, 111 (2019). Trump himself, speaking recently of a federal House bill designed to make voting easier, remarked: “The things they had in there were crazy. They had . . . levels of voting that, if you ever agreed to it, you’d never have a Republican elected in this country again.”237Aaron Blake, Trump Just Comes out and Says It: The GOP Is Hurt when It’s Easier to Vote, Wash. Post (Mar. 30, 2020, 1:47 PM), https://www.washingtonpost.com/politics/2020/03/30/trump-voting-republicans. Later, Mr. Trump acknowledged that his opposition to funding for the U.S. Postal Service is rooted in his opposition to Democratic efforts to make voting by mail easier during the pandemic. Jennifer Rubin, Trump Confesses to Voter Suppression, Wash. Post (Aug. 13, 2020, 3:58 PM), https://www.washingtonpost.com/opinions/2020/08/13/trump-confesses-voter-supression.

These views are fundamentally illiberal. They deny the equality of citizens for purposes of self-governance—the most basic root of a liberal order of any kind, not just a democratic one.238Ginsburg & Huq, supra note 6, at 24. They are also authoritarian. They deny the legitimacy not merely of political competition, but of political pluralism itself.239Id. at 22. Those who are not members of the group destined to lead, or who do not at the very least profess the beliefs approved by that group, lack any claim to agency in the political sphere. Their role is to accept, not to compete for, leadership.240It must be acknowledged that a similar attitude occasionally crops up on the (very small) extreme left flank of the Democratic Party. However, unlike the Republican leadership, which has endorsed, validated, and mainstreamed such behavior as a normal aspect of Republican politics, the Democratic leadership has made significant efforts to squelch these behaviors, and to steer Democratic legislators with possibly illiberal instincts toward a more conciliatory and mutual mode of politics. E.g., John Parkinson, Pelosi Works to Rein in Freshman Progressives like Ocasio-Cortez amid Intraparty Squabbling, ABC News (July 10, 2019, 5:05 PM), https://abcnews.go.com/Politics/pelosi-works-rein-freshman-progressives-ocasio-cortez-amid/story?id=64244999 [https://perma.cc/R2FM-XJCR].

IV. Authoritarianism in the States

As indicated above, in addition to being concentrated in a political party, illiberalism can be concentrated in a region. In a federal system, the key question is whether illiberalism is sufficiently concentrated in any subnational unit to justify concluding that the state or province is itself illiberal. With that objective in mind, this Part first examines generally the problem of regional concentration of illiberalism in federal states, and then, turning to the United States, collects relevant evidence of the growth of subnational authoritarianism here at home.

A. Illiberalism in Federal States: The Problem of Authoritarian Enclaves

Neither liberalism nor democracy arrives in a place suddenly and spreads uniformly, in linear fashion.241See Berman, supra note 78, 8–10; Alexander Keyssar, The Right to Vote: The Contested History of Democracy in the United States 316–17 (2000). Rather, liberalism and democracy tend to gain small initial footholds from which they then gain and lose ground repeatedly, often over the course of lengthy, vigorous, and repeated struggles against beneficiaries and defenders of the illiberal regimes they replace.242Keyssar, supra note 242, at 317–18. Full consolidation of liberal democracy thus often takes a very long time. Even in the United States, a place unusually receptive from the beginning to liberal democracy, democratic consolidation occurred in fits and starts over the space of more than a century-and-a-half, and in the view of many, did not arrive fully until enactment of major national civil rights legislation during the 1960s,243According to Robert Mickey, the “death blows” to authoritarian “enclave rule” in the American South were delivered by the Civil Rights Act of 1964 and the Voting Rights Act of 1965. Mickey, supra note 10, at 260. or even later.244See, e.g., id. at 32 (“[T]he United States was not a fully democratic polity until preparations were well under way for the celebration of the bicentennial of the Declaration of Independence.”).

A spate of recent research has added greatly to our understanding of the way democracy spreads in federal states.245Edward L. Gibson, Boundary Control: Subnational Authoritarianism in Federal Democracies 5 (2013); Mickey, supra note 10, at xi. See generally Agustina Giraudy, Democrats and Autocrats: Pathways of Subnational Undemocratic Regime Continuity Within Democratic Countries 43–44 (2015); Illiberal Practices: Territorial Variance Within Large Federal Democracies 1 (Jacqueline Behrend & Laurence Whitehead eds., 2016) [hereinafter Illiberal Practices]. Federalism is of course deliberately designed to enable subnational units to adopt their own preferred policies, but the range of policies from which subnational units may choose is bounded in principle by shared policy commitments established at the national level and typically entrenched in a federal constitution.246Levy, supra note 3, at 28. Nevertheless, the same federal constitutional structures that enable subnational units to deviate from national policy commitments within a permitted range often can be manipulated to resist the imposition subnationally even of foundational principles of governmental organization to which the entire nation has by hypothesis committed itself.247Id. at 28, 30. In consequence, as Edward Gibson writes in his leading study, “local authoritarian enclaves . . . dot the landscapes of democratic countries around the world.”248Gibson, supra note 246, at 2. That is to say, democratizing states do not democratize throughout their territory uniformly at the same rate, and subnational units previously under authoritarian rule may hold out locally against national democratizing trends with considerable tenacity—perhaps even indefinitely—using tactics of what Gibson calls “boundary control.”249Id. at 5.

In the most common pattern, boundary control involves a struggle between, on one hand, an authoritarian or illiberal250See Jacqueline Behrend & Laurence Whitehead, The Struggle for Subnational Democracy, 27 J. Democracy 155, 155 (2016) (arguing that the label “subnational authoritarianism” is “too narrow” because the real concern is “illiberal structures and practices,” which “can hang on at the subnational level” even after local authoritarianism has been formally subdued). subnational unit seeking to seal off local institutions from national forces of democratization; and on the other hand, a democratic or democratizing national government seeking to reform subnational institutions by conforming them to nationally adopted norms of liberal democracy.251Gibson, supra note 246, at 24–25. Tactics deployed by subnational units to maintain the autonomy necessary to resist local democratization commonly include the manipulation of local electoral rules to maintain the ruling party in power; clientelistic control by subnational leaders over the distribution of fiscal resources received from the national government; and the exercise of direct influence over national politics through control over local nominations to national office and the ability to deliver local votes on demand to national leaders.252Id. at 26–30. While these tactics can be effective even for a single authoritarian enclave, when several such enclaves “band together in defense of a set of illiberal structures and practices, their combined leverage at the federal level becomes sufficient to block or distort nationwide democratic responsiveness.”253Jacqueline Behrend & Laurence Whitehead, Setting the Comparative Agenda: Territorially Uneven Democratization Processes in Large Federations, in Illiberal Practices, supra note 246, at 9.

Studies of subnational illiberal enclaves have identified them mainly in contemporary federal states ordinarily conceived as democratically fragile, such as Argentina, Mexico, Brazil, and Russia.254See Gibson, supra note 246, chs. 4–5; Illiberal Practices, supra note 246, chs. 4–9. But the problem of subnational authoritarianism is not confined to foreign countries with young and weak democratic institutions. In fact, the most successful, powerful, and longstanding enclave of subnational illiberalism in the history of democratization is home grown: from their founding through the late twentieth century, the states of the American South “are best understood as eleven enclaves of authoritarian rule.”255Mickey, supra note 10, at 33.

The illiberalism of the slaveholding states prior to the Civil War is scarcely debatable,256Ellis, Liberal Tradition, supra note 165, at 213 (“[T]he starting point for understanding the South is its illiberalism. Blacks and whites were not born equal, and few whites regarded blacks as their equals.”). but it is commonplace to view the South between Reconstruction and roughly the 1970s as something far better, if still imperfect—as a region “democratically governed [but] racist.”257Mickey, supra note 10, at 13. Recent scholarship, however, persuasively debunks this view. Even as the United States, following the Civil War, experienced “the most extensive case of territorial democratization in history,”258Gibson, supra note 246, at 35. leaders of a consolidating Solid South slowly constructed

a single regime, and equipped it with institutions to demobilize white electorates, extrude blacks from electoral politics, and forestall workers’ challenges to state institutions and policies. Enclave rulers carefully protected their polity’s conditional autonomy and skillfully deployed federal officeholders to block potential interferences, especially concerning voting rights and state-sponsored violence.259Mickey, supra note 10, at 34.

Suffrage restrictions that disenfranchised Blacks and “dampened poor whites’ political participation” were supplemented by measures that “weakened local governments, converted many elective offices into appointive positions controlled by the governor or state legislature, and further insulated state judiciaries from popular input.”260Id. at 43. Before long, these and similar measures “eliminat[ed] the possibility of building opposition parties”261Id.—the hallmark of authoritarian rule.

Visiting the United States at the crest of a Jacksonian movement of democratic egalitarianism,262See Keyssar, supra note 242, at 33–52. See generally Glyndon G. Van Deusen, The Jacksonian Era, 1828–1848 (1959). Alexis de Tocqueville wrote that the expansion of voting rights is inherently irreversible, and has no stopping point short of universal suffrage.263Alexis de Tocqueville, 1 Democracy in America 57 (Knopf 1945) (orig. 1835). The ingenuity of Southern authoritarians stands as the only significant counterexample to Tocqueville’s dictum: “The enfranchisement of southern blacks and poor whites in the 1860s and 1870s and their wholesale disenfranchisement shortly thereafter took place under a national democratic regime. Never before or since in a national democracy (absent a coup or regime change) has this feat been matched.”264Gibson, supra note 246, at 38. That is to say, democratization imposed on the South by force and entrenched in national law eventually was deconsolidated so thoroughly as to convert the South into an illiberal subnational enclave not all that different from the kind of quasi-feudal subnational regimes researchers have found lingering in places like Argentina or Mexico.265Id. chs. 3–5. Indeed, it is thought that the tactics employed by the American South to retain local authoritarian rule have for more than a century provided a reliable template for authoritarians around the world.266Id. at 38–39.

In light of the recent American tilt toward authoritarianism at the national level, described above, this analysis raises profound questions about the nature of contemporary subnational variation in the United States. If the South was illiberal for so much of U.S. history, did it ever actually become liberal? Was illiberalism confined to southern enclaves, or did it take root elsewhere? Was the long, national struggle to consolidate a liberal democracy ever won? Just how common, in other words, is illiberalism in the United States, and how is it distributed?

B. What Counts as Evidence?

In assessing the degree to which state-to-state variation tracks, or has begun to track, or threatens to track a cleavage between liberalism and illiberalism, a preliminary question concerns what counts as evidence one way or the other. Rather than get bogged down in definitional questions about whether specific states’ policies and actions count as liberal or illiberal, I shall instead focus on whether state actions and policies can be characterized as anti-democratic and authoritarian. Unlike illiberalism generally, the meaning of which is contested, the characteristics of authoritarianism are less so, and scholars have identified concrete, empirical indicia that mark out the path from democracy to authoritarianism—a path which, to be sure, is not necessarily identical to the path from liberalism to illiberalism, but is sufficiently closely related to it to provide a highly useful proxy measurement.

For the purpose of determining whether and to what degree national trends toward authoritarianism may be manifesting themselves on the state level, I will draw on well-established patterns that have emerged in nations around the world in which democratic backsliding has occurred. For example, the hallmark of democratic backsliding globally is a steady accretion of power in the chief executive.267Bermeo, supra note 129, at 10 (discussing “[e]xecutive aggrandizement”); see also Ginsburg & Huq, supra note 6, at 150–54, 178–79. Authoritarian leaders frequently consolidate their power by engineering the replacement of the existing constitution with one granting additional executive power,268This strategy was pursued by Viktor Orbán in Hungary, Hugo Chávez in Venezuela, and Rafael Correa in Ecuador, among others. Scheppele, supra note 114, at 549, 554; David Landau, Abusive Constitutionalism, 47 U.C. Davis L. Rev. 189, 203–10 (2013). or by ramming through amendments that consolidate executive control.269This was the strategy of choice in Colombia. Landau, supra note 269, at 200–03. Another common tactic is gaining control of the judiciary and then eliciting from it decisions that interpret the scope of existing executive power much more broadly than had previously been thought.270As, for example, in Hungary, Turkey, and Poland. Scheppele, supra note 114, at 550–53. We seem to observe some of this behavior at the national level; do we observe it as well subnationally, in the states?

On the other hand, constitutions in authoritarian regimes do not always reflect the actual dynamics of power; sometimes such documents retain the trappings of democratic structure as a kind of “window dressing . . . to obfuscate actual political practice.”271Tom Ginsburg & Alberto Simpser, Introduction to Constitutions in Authoritarian Regimes 7 (Tom Ginsburg & Alberto Simpser eds., 2014) (emphasis omitted). Moreover, as events at the national level have vividly revealed, much of what we think of as the constitutional order rests not on formally enacted constitutional rules, but on informal norms of official behavior which, although unwritten, are nevertheless widely understood and obeyed by those committed to the existing order.272Levitsky & Ziblatt, supra note 6, at 7; Daphna Renan, Presidential Norms and Article II, 131 Harv. L. Rev. 2187, 2190 (2018); Siegel, supra note 213, at 179–80; Jamal Greene, Trump as a Constitutional Failure, 93 Ind. L.J. 93, 96 (2018).

It follows that the better place to look for authoritarian tendencies may not be in the realm of formal law—in the polity’s “large-C” constitution—but in the actual behavior of the regime—its “small-c” constitution.273This increasingly common distinction refers to the difference between a formal constitutional text (large-C) and the informal and largely unwritten conventions, settlements, and understandings that typically undergird or accrete around the formal document (small-c). For a good definition, see Davis S. Law, Constitutions, in The Oxford Handbook of Empirical Legal Research 376–77 (Peter Cane & Herbert M. Kritzer eds., 2010). To detect such behavior, I will use as my guide the “authoritarian’s playbook” developed by political scientist Larry Diamond, which focuses on informal behaviors such as the rhetoric employed by authoritarian leaders; attacks on the independence of courts, the media, and the civil service; cronyism; and, significantly, manipulation of the electoral process.274Diamond, supra note 6, at 64–65. Here is Diamond’s formulation, which is worth quoting at length:

  1. Begin to demonize the opposition as illegitimate and unpatriotic, part of the discredited or disloyal establishment, hopelessly out of touch with the real people.
  2. Undermine the independence of the courts. . . by purging judges and replacing them with political loyalists, or by restructuring the judiciary so it can be packed and placed under partisan control.
  3. Attack the independence of the media, by denouncing them as partisan fabulists, mobilizing public fervor against them, . . . regulating them[, and] finally taking over their ownership . . . .
  4. Gain control of any public broadcasting, politicize it, and make it an instrument of ruling party propaganda.
  5. Impose stricter control of the internet, in the name of morality, security, or counterterrorism, thus further chilling free speech and the freedom to organize.
  6. Subdue other elements of civil society—civic associations, universities, and especially anticorruption and human rights groups—by painting them as part of the arrogant, effete, selfish elite that have betrayed the people and the country. Make university professors afraid to criticize the government in their writings and classrooms. Render student groups liable to prosecution for peaceful protest. Create new, fake civic organizations that will be faithful to the populist leader and party.
  7. Intimidate the business community into ending its support for political opposition. Threaten to unleash tax and regulatory retribution on businesses that fund opposition parties and candidates . . . .
  8. Enrich a new class of crony capitalists by steering state contracts, credit flows, licenses, and other lucre to the family, friends, and allies of the ruler and his clique.
  9. Assert political control over the civil service and the security apparatus. Start referring to professional civil servants and military officers loyal to the democratic constitution as members of a “deep state.” Purge them. . . .
  10. Gerrymander districts and rig the electoral rules to make it nearly impossible for opposition parties to win the next election. Ensure that the ruling party can retain its grip on power even if it fails to win most of the vote.
  11. Gain control over the body that runs the elections, to further tilt the electoral playing field and institutionalize de facto authoritarian rule.
  12. Repeat steps 1 to 11, ever more vigorously, deepening citizens’ fear of opposing or criticizing the new political order and silencing all forms of resistance.275Id.

In the next Section, I examine the formal constitutional record—the large-C constitutions of the states—for evidence of practices listed by Diamond. In the subsequent Section, I turn to the informal behavior that makes up the small-c constitutions of the states.

C. The Formal Constitutional Record

The United States is one of the most decentralized federations in the world, and given the degree of autonomy that American states enjoy, especially concerning the structure of their own political processes,276Gardner, Autonomy and Isomorphism, supra note 25, at 33–34. it is sometimes surprising to observers how little variation actually exists among state constitutions.277Id. One of the most common—and most frequently criticized—habits of state courts is to interpret state constitutions to have meanings identical to that of the federal constitution, a practice known as “lockstep” interpretation. See generally Robert F. Williams, The Law of American State Constitutions 193–232 (2009). In particular, consistent with powerful currents in the American constitutional tradition, all state constitutions are formally democratic and liberal.278Williams, supra note 278, at 17–20. They uniformly affirm liberal principles of popular sovereignty; provide that all major offices are to be filled by direct popular election based on essentially universal suffrage; adhere to inherited conceptions of constitutionalism and the rule of law; and provide—formally, at least—substantial protection for citizen equality and individual dignitary rights.279See generally id.

Nothing that has occurred recently—over the last five years, say—has altered these basic facts. No state constitution, for example, has been replaced with a more authoritarian version; in fact, no state has replaced its constitution since the mid-1980s,280See R.I. Const. (1986); Ga. Const. (1983). and indeed, no state has held a full-scale constitutional convention since 1986.281John Dinan, State Constitutional Politics: Governing by Amendment in the American States 30–31 (2018). On the other hand, some states provide for mandatory periodic ballot measures asking the people whether a constitutional convention should be called.282Id. at 12. In the two most recent such votes—in New York (2017) and Rhode Island (2014)—convention calls were rejected,283Jesse McKinley, New York Voters Reject a Constitutional Convention, N.Y. Times (Nov. 7, 2017), https://www.nytimes.com/2017/11/07/nyregion/new-york-state-constitutional-convention.html; Linda Borg, R.I. Voters Reject Constitutional Convention, Providence J. (Nov. 4, 2014, 9:45 PM), https://www.providencejournal.com/article/20141104/NEWS/311049874. in large part over fears that a convention would result in illiberalization of the state constitution. In New York, for example, public employee unions opposed a convention call over fears that a convention “would open up protections for labor, education and social services to attacks from right wing groups.”284Ari Paul, Unions Set to Oppose a Constitutional Convention PSC CUNY (Apr., 2017), https://www.psc-cuny.org/clarion/april-2017/unions-set-oppose-constitutional-convention [https://perma.cc/K57V-XRUH]. The New York Civil Liberties Union opposed a convention because it would place at risk existing state constitutional protections for “free speech, public education, separation of church and state, the environment, aid and care to the needy, workers’ right to organize and pension rights.”285Donna Lieberman, A Constitutional Convention Gambles with New Yorkers’ Rights, NYCLU (Oct. 18, 2017, 10:15 AM), https://www.nyclu.org/en/news/constitutional-convention-gambles-new-yorkers-rights. New York environmental groups argued against a convention on the ground that opening up the state constitution to change would put at risk the prized “forever wild” provision limiting development in the Adirondack Park.286Editorial: Adirondack Compromise, Times Union (May 12, 2016, 7:13 PM), https://www.timesunion.com/tuplus-opinion/article/Editorial-Adirondack-compromise-7465765.php [https://perma.cc/C42P-3F5V]. In Rhode Island, Planned Parenthood opposed a constitutional convention on the ground that it would endanger constitutionalized women’s reproductive rights, and “could send women back to the 1950s.”287J.H. Snider, Does the World Really Belong to the Living? The Decline of the Constitutional Convention in New York and Other U.S. States, 1776–2015, 6 Am. Pol. Thought 256, 286 (2017). It is noteworthy that these fears were so pronounced in what are probably two of the most politically liberal polities in the nation.288See id. at 286 (noting the paradox in Rhode Island of opposition to a convention for fear of weakening of a woman’s right to choose in a state in which support for that right was overwhelming).

Similarly, no state constitution has been amended in recent years in a way that suggests movement toward significant new concentrations of power in the hands of the governor. Amendments to structural aspects of state constitutions are in general far less common than amendments dealing with individual rights or substantive policy matters,289See generally Dinan, supra note 282. and in the last few years, structural amendments have been few and marginal in impact.290See, e.g., Fla. Const. art. IX, § 7 (2018) (altering the structure of higher education governance); Ga. Const. amend. 2 (2018) (establishing a new specialized business court); Ala. Const. amend. 6 (2016) (raising the threshold for a legislative impeachment vote). About as many have worked slight reductions in executive power as have increased it. For example, a 2018 Florida amendment prohibited state courts from deferring to administrative agencies’ interpretations of state statutes or rules, transferring power from the executive to the judicial branch;291Fla. Const. amend. 6 (2018). and a 2016 Idaho amendment constitutionalized the authority of the state legislature to review, approve, and reject administrative rules, transferring power from the executive to the legislature.292Idaho Const. amend. HJR5 (2016). In addition, some amendments that would have increased gubernatorial power were defeated. See, e.g., Wis. question 1 (2018), which would have eliminated the independently elected position of state treasurer.

Nor have state high courts recently issued major decisions altering the separation of powers in favor of the executive. Indeed, in what is probably the most significant state separation of powers decision issued in the last five years, the North Carolina Supreme Court went the other way, upholding a diminishment of executive power. That case, Cooper v. Berger,293822 S.E.2d 286 (N.C. 2018). concerned a statute enacted by the Republican-controlled state legislature in December 2016, two weeks before newly elected Democratic governor Roy Cooper was to take office.294Conflicts Between Gov. Roy Cooper and the General Assembly of North Carolina, Ballotpedia, https://ballotpedia.org/Conflicts_between_Gov._Roy_Cooper_and_the_General_Assembly_of_North_Carolina [https://perma.cc/D5WN-4MSJ]. The new statute for the first time subjected the governor’s cabinet appointments to senatorial confirmation, reducing the governor’s power over his own executive branch appointments.295Cooper, 822 S.E.2d at 286. The court upheld the statute against Cooper’s separation-of-powers challenge.296Id. at 300–01. In a case decided a month earlier, the North Carolina Supreme Court invalidated another measure pushed through in the waning days of the prior administration, this one restructuring the existing elections and ethics boards so as to place them under significant legislative control and concomitantly diminish the governor’s control over the relevant officials. Cooper v. Berger, 809 S.E.2d 98, 115 (N.C. 2018). In so doing, the court defended the governor’s existing constitutional authority, but did not expand it. Id.

This information, however, must be understood in a broader context. First, American governors tend to be very powerful state actors, and any with authoritarian aspirations may not need formal enhancement of their power through the kinds of constitutional changes most often sought by authoritarians elsewhere.297See Saladin M. Ambar, How Governors Built the Modern American Presidency 25, 29–30 (2012) (discussing early efforts of American governors to expand the boundaries of their constitutionally granted power). Although originally meant to be weak and subordinate to state legislatures,298See id. at 22. governors emerged in the late nineteenth and early twentieth centuries as the most powerful players in state government, a consequence of an American turn toward “executive leadership” as the most efficient and democratically legitimate way to implement the popular will.299See id. at 21, 23, 41, 45, 51. Indeed, as Saladin Ambar argues, the powerful, modern presidency was to a great extent shaped by governors who, upon their elevation to the presidency, inhabited the office in much the same way as they had previously inhabited their governorships.300Id. at ix–x, 8.

Second, the extent of gubernatorial power—and in particular the degree to which it exceeds legislative power—cannot typically be read off the face of the formal constitutional document. For example, in many states the legislature is a part-time and largely amateur assembly.301See, e.g., Mont. Const. art. V, § 6 (“The legislature shall meet each odd-numbered year in regular session of not more than 90 legislative days.”). Although this practice is rooted in a benign and charmingly republican wish to ensure that legislators are drawn from the general citizenry rather than a distinct political class, and that they remain in close touch with their constituents, the actual effect is to transfer considerable power to the governor and the executive branch bureaucracy, which are permanent, full-time, professional, and well-funded.302John. M. Carey et al., The Effects of Term Limits on State Legislatures: A New Survey of the 50 States, 31 Legis. Stud. Q. 105, 124 (2006); Thad Kousser, Term Limits and the Dismantling of State Legislative Professionalism 154–56 (2005). Governors tend also to dominate the budgetary process.303Thad Kousser & Justin H. Phillips, The Power of American Governors: Winning on Budgets and Losing on Policy 5 (2012). Many have an extremely powerful and flexible item veto that permits them to shape legislation, and even, in some cases, to alter it significantly.304See, e.g., State ex rel. Wis. Senate v. Thompson, 424 N.W.2d 385, 399 (Wis. 1988) (holding that the governor may use the item veto in a way that changes the meaning of legislation); Welsh v. Branstad, 470 N.W.2d 644, 650 (Iowa 1991) (same). The reach of the Wisconsin case was later constricted somewhat by a constitutional amendment, now contained in Wis. Const. Art. V, § 10(1)(c). As Jonathan Zasloff has perceptively argued, the governor of California is not best thought of primarily as an executive officer, but as a “super-legislator, who wields his power through the lawmaking and budgetary processes”305Jonathan Zasloff, Taking Politics Seriously: A Theory of California’s Separation of Powers, 51 UCLA L. Rev. 1079, 1109 (2004).—a power highly valued by authoritarian leaders around the world.306To give just one example, Argentine Presidents have worked hard to maintain control over central budgetary transfers to the provinces in order to maintain and nurture clientelistic relationships with provincial governors that are crucial to their continuation in office at the national level. Martín Ardanaz, Marcelo Leiras & Mariano Tommasi, The Politics of Federalism in Argentina and Its Implications for Governance and Accountability, 53 World Dev. 26, 27–30 (2014).

Third, the de facto power that governors exercise does not necessarily track what we might suspect to be the divide, if one exists, between liberal-democratic and illiberal-authoritarian U.S. states. According to rankings of gubernatorial power produced periodically by the late political scientist Thad Beyle, some of the most powerful governors have been those in strongly liberal states like New York, Massachusetts, and Maryland, and some of the weakest in states of the Old Confederacy such as Alabama and Mississippi.307The relevant documents were recovered through searches of the Internet Archive, Wayback Machine. See Thad Beyle, Gubernatorial Power: The Institutional Power Ratings for 50 Governors of the United States, Univ. N.C. (July 2008), https://web.archive.org/web/20130721000517/http:/www.unc.edu/~beyle/gubnewpwr.html. See generally Tyler Lang Reisinger, Beyond Beyle: Assessing the Measurement of Institutional and Informal Gubernatorial Powers (Aug. 7, 2008) (Masters dissertation, Va. Polytechnic Inst.), https://vtechworks.lib.vt.edu/handle/10919/34495 [https://perma.cc/5LJJ-HGH3].

Beyle’s analysis, moreover, is generally consistent with the state constitutional jurisprudence. In states like Alabama, Florida, Kentucky, Missouri, and West Virginia, apparently heavily influenced by what Jim Rossi calls “[a]ntifederalist separation of powers ideals,”308Jim Rossi, Institutional Design and the Lingering Legacy of Antifederalist Separation of Powers Ideals in the States, 52 Vand. L. Rev. 1167, 1172 (1999). the separation of powers and the concomitant scope of executive power tend to be “strictly construed.”309Riley v. Joint Fiscal Comm. of Ala. Legislature, 26 So. 3d 1150, 1155 (Ala. 2009); e.g., Corcoran v. Geffin, 250 So. 3d 779, 783 (Fla. Dist. Ct. App. 2018); Scheer v. Zeigler, 21 S.W.3d 807, 813 (Ky. Ct. App. 2000); State ex rel. Jones v. Atterbury, 300 S.W.2d 806, 817 (Mo. 1957) (en banc); State v. Bostic, 729 S.E.2d 835, 843 (W. Va. 2012). In Massachusetts and New York, in contrast, courts take a more generous view of the scope of executive power.310See, e.g., Op. of the Justs. to the Council, 334 N.E.2d 604, 609 (Mass. 1975); Bourquin v. Cuomo, 652 N.E.2d 171, 173 (N.Y. 1995) (taking a very broad view of the gubernatorial power to govern by executive order in both cases). Thus, if there is at this moment a strongman ruling an American state, it might well be Andrew Cuomo of New York or Gavin Newsom of California, governors apparently as strongly committed as any in the United States to democratic liberalism.311If dynastic control by ruling families is relevant, both Andrew Cuomo and Jerry Brown, who stepped down as governor of California in 2019, are sons of prior governors. Dominic-Madori Davis, Meet the Cuomo family, the New York Political Dynasty that’s Become the Face of America’s Response to the Coronavirus Pandemic, BUS. INSIDER (Apr. 22, 2020, 1:04 PM), https://www.businessinsider.com/meet-the-cuomo-family-andrew-mario-new-york-politics-2020-3; George Skelton, A Brown–Father or Son–Has Been California Governor for 40% of the Last Six Decades, L.A. TIMES (Jan. 29, 2018, 12:05 PM), https://www.latimes.com/politics/la-pol-sac-skelton-jerry-brown-state-of-the-state-speech-20180129-story.html.

Finally, some of the most severe constitutional norm-bashing occurring in the states over the last few years has been the work not of governors, but of state legislatures attempting to enhance their own powers at the expense of governors. The most prominent examples come from North Carolina and Wisconsin, in which Republican-controlled state legislatures enacted measures to strip power from newly elected Democratic governors before they took office, discussed further below.312See infra notes 325–27 and accompanying text. These actions constitute a gross breach of longstanding, universal norms of democratic contestation, which require fair alternation of power among the parties and gracious acceptance of electoral defeat.313In a well-known formulation, Robert Dahl’s minimal conditions for democracy include “free, fair, and competitive elections.” Steven Levitsky & Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes After the Cold War 5–6 (2010).

D. Informal Behavior

When we turn our attention from the large-C constitution of formal text and jurisprudence to the small-c constitution of informal norms and actual behavior of government officials, the picture looks considerably different. Below, I work through the most pertinent items from Diamond’s authoritarian’s playbook.

1. Demonization of the opposition

Donald Trump is without peer in his persistent demonization of those who oppose him, demolishing, with evident glee, a critical, longstanding informal norm of civility in American democratic politics.314Siegel, supra note 213, at 191–98; see Levitsky & Ziblatt, supra note 6, at 8, 102–12. During his first campaign he repeatedly called for his Democratic opponent in the presidential election, Hillary Clinton, to be “locked up,” and continued tacitly to encourage the idea during his reelection campaign.315Peter W. Stevenson, A Brief History of the ‘Lock Her Up!’ Chant by Trump Supporters Against Clinton, Wash. Post (Nov. 22, 2016, 4:56 PM), https://www.washingtonpost.com/news/the-fix/wp/2016/11/22/a-brief-history-of-the-lock-her-up-chant-as-it-looks-like-trump-might-not-even-try; Chris Cillizza, How ‘Lock Her Up’ Just Blew Up, CNN (Jan. 10, 2020, 10:52 AM), https://www.cnn.com/2020/01/10/politics/hillary-clinton-donald-trump-justice-department/index.html [https://perma.cc/J385-WLS8]. He routinely insults and denigrates political adversaries, often, as in the case of judges, investigators, and prosecutors, merely for performing the tasks entrusted to them by the Constitution or laws.316For example, Trump has repeatedly denigrated judges who rule against him and his administration, and he bitterly and constantly denounced the special prosecutor’s investigation into possible collaboration with Russia by his campaign. Levitsky & Ziblatt, supra note 6, at 177–81. Some members of Congress have emulated the President’s behavior.317E.g., Dana Milbank, Could Lindsey Graham Be Any More Obsequious?, Wash. Post (Oct. 24, 2019, 6:51 PM), https://www.washingtonpost.com/opinions/2019/10/24/lickspittle-lindseys-week-walk-backs. Do we observe similar behavior from state officials?

To be sure, some state officials have occasionally adopted similar kinds of rhetoric. A Republican Oklahoma state representative called for Hillary Clinton’s execution by firing squad.318State Rep. Comments ‘Firing Squad’ Against Hillary Clinton on Post, News On 6 (Nov. 2, 2016, 3:23 PM), https://www.newson6.com/story/5e3604082f69d76f62032be3/state-rep-comments-firing-squad-against-hillary-clinton-on-post [https://perma.cc/WN5B-236G]. A Republican county chair in Kansas compared a mandatory masking order issued by Democratic governor Laura Kelly to Nazi orders rounding up Jews for transportation to death camps.319Michael Levenson, Kansas G.O.P. Official Removes Cartoon Comparing Mask Order to Holocaust, N.Y. Times (July 4, 2020), https://www.nytimes.com/2020/07/04/us/anderson-county-review-laura-kelly-holocaust-cartoon.html. Republicans in Texas, Oklahoma, and elsewhere have made hateful, demeaning comments about immigrants or Muslims,320E.g., Jessica Campisi, Latino Texas Lawmakers Condemn Governor for ‘Dangerous’ Tweet About Immigrants, Hill (Aug. 24, 2019, 8:16 AM), https://thehill.com/latino/458656-latino-lawmakers-condemn-texas-governor-for-dangerous-tweets-that-demonized-immigrants [https://perma.cc/7XTM-9G5U]; Oklahoma Muslim Leaders Concerned After Lawmaker Calls Islam “A Cancer in Our Nation,” Week (Sept. 17, 2014), https://theweek.com/speedreads/446092/oklahoma-muslim-leaders-concerned-after-lawmaker-calls-islam-cancer-nation; Ralph Ellis & Sheena Jones, Oklahoma Legislator’s Questionnaire Asks Muslims: ‘Do You Beat Your Wife?’, CNN (Mar. 5, 2017, 9:06 PM), https://www.cnn.com/2017/03/05/us/oklahoma-muslim-legislator-questionnaire/index.html [https://perma.cc/7KSM-YRFP]. both groups that might be thought to lean politically toward Democrats. In general, though, in what we shall find to be a fairly consistent pattern, activity at the state level seems to involve less rhetoric and more action. In these instances, the states are indeed serving as laboratories—not laboratories of democracy, as Justice Brandeis proposed,321New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting) (“It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”). but as “laboratories of authoritarianism.”322Levitsky & Ziblatt, supra note 6, at 2.

To insult or demonize another person is, literally, depersonalizing—it proceeds by definition from the view that the insulted are not fully human, and thus need not be treated with the decency to which humans are inherently entitled. In the political context, to insult or otherwise demonize others for holding contrary beliefs is to treat such individuals as political inferiors who do not fully count in any appropriate calculus of public opinion, and whose participation in politics is thus fundamentally illegitimate. This belief, however, need not manifest itself solely in rhetorical attacks like insults; it may take in addition the form of justification—justification for disenfranchising and otherwise ignoring, blocking, or overturning political participation by adversaries, so that the relevant jurisdiction can be ruled through the popular sovereignty of the “true” people, those legitimately entitled to rule.323Diamond, supra note 6, at 64–65.

In what is surely the most remarkable example of this phenomenon, Republican-controlled governments in North Carolina, Wisconsin, and Michigan reacted to recent elections of Democratic governors by enacting legislation, in the waning days of lame duck sessions, to strip from the executive branch substantial amounts of power for the sole purpose of diminishing the political agency of incoming Democratic officials and the voters who put them in office. In North Carolina, immediately following the 2016 gubernatorial election in which Democrat Roy Cooper defeated the incumbent Republican Pat McCrory, the legislature enacted and Governor McCrory signed a package of bills that transferred control of the Board of Elections from the governor to the legislature; subjected gubernatorial cabinet appointments to senatorial confirmation; stripped the governor of power to appoint trustees of the University of North Carolina system; and reduced the authority of the governor to hire and fire government employees.324Amber Phillips, Amid Outcry, N.C. GOP Passes Law to Curb Democratic Governor’s Power, Wash. Post (Dec. 16, 2016, 3:15 PM), https://www.washingtonpost.com/news/the-fix/wp/2016/12/16/amid-growing-outcry-nc-gop-pushes-forward-with-plan-to-curb-democratic-govs-power. Some aspects of the law were later overturned by the North Carolina Supreme Court on separation of powers grounds. Cooper v. Berger, 809 S.E.2d 98, 112–14, 116 (N.C. 2018). The legislature, still controlled by Republicans, then proposed a sweeping constitutional amendment that would have drastically reduced the authority of the governor to make hundreds of executive branch appointments, but that measure was invalidated by North Carolina courts on the grounds that its language was misleading. Lynn Bonner, Republicans Advance New Versions of NC Constitutional Amendments, News & Observer (Aug. 24, 2018, 11:00 AM), https://www.newsobserver.com/news/politics-government/article217265280.html. The legislature then rewrote the amendment more narrowly, but voters rejected it. North Carolina Legislative Appointments to Elections Board Amendment, Ballotpedia (2018), https://ballotpedia.org/North_Carolina_Legislative_Appointments_to_Elections_Board_Amendment_(2018) [https://perma.cc/R988-RFA7].

In 2018 in Wisconsin, following wins by Democratic candidates for governor and attorney general, the outgoing Republican-controlled legislature enacted, and the lame-duck Republican governor signed, legislation that deprived the governor of control over significant public programs and transferred from the attorney general to the legislature authority to withdraw from various kinds of litigation.325Sophie Tatum, Wisconsin Gov. Scott Walker Signs Legislation Stripping Incoming Democrats’ Power, CNN (Dec. 14, 2018, 6:25 PM), https://www.cnn.com/2018/12/14/politics/scott-walker-wisconsin/index.html [https://perma.cc/W6J2-ZGR8]. The legislation was apparently intended in great measure to cripple the ability of the incoming Democrats to deliver on their campaign pledges—the very basis, presumably, of their election to office326Id.—thereby subverting the only reason to hold elections in the first place. Similarly, in 2018, following a sweep of statewide offices by Democrats, the Republican-controlled Michigan legislature enacted and the lame-duck Republican governor signed legislation limiting executive authority to oversee campaign finance laws.327Dan Simmons & Felicia Sonmez, Republicans in Wisconsin and Michigan, Defeated in November, Seek Payback by Limiting the Powers of Incoming Democrats, Wash. Post (Dec. 4, 2018, 7:37 PM), https://www.washingtonpost.com/politics/republicans-in-wisconsin-and-michigan-defeated-in-november-seek-payback-by-limiting-the-powers-of-incoming-democrats/2018/12/03/393f478e-f72a-11e8-8c9a-860ce2a8148f_story.html.

2. Undermine judicial independence

As Stephen Levitsky and Daniel Ziblatt observe, “[i]t always helps to have the referees on your side.”328Levitsky & Ziblatt, supra note 6, at 78. For this reason, capture of independent judicial bodies is high on the wish list of would-be authoritarians; it is a tactic that autocrats have successfully employed in Hungary, Turkey, Russia, Peru, Egypt, Venezuela, Ecuador, and elsewhere.329Id. at 78–79; Varol, supra note 208, at 1687–93; Tom Ginsburg, International Courts and Democratic Backsliding, 37 Berkeley. J. Int’l L. 265, 279–83 (2019); Kriszta Kovács & Kim Lane Scheppele, The Fragility of an Independent Judiciary: Lessons from Hungary and Poland—and the European Union, 51 Communist & Post-Communist Stud. 189, 189–92 (2018). It is also a tactic that has been deployed in American states controlled by Republicans.

Control over the ideological commitments of judges has long been a strategic objective of the Republican Party.330Jason Zengerle, How the Trump Administration Is Remaking the Courts, N.Y. Times Mag. (Aug. 22, 2018), https://www.nytimes.com/2018/08/22/magazine/trump-remaking-courts-judiciary.html. That goal has been pursued single-mindedly on the federal level since the Reagan Administration through careful, highly selective ideological vetting of judicial appointees.331Steven M. Teles, The Rise of the Conservative Legal Movement: The Battle for Control of the Law 142 (2008). The task is more complicated on the state level: most state judges are elected rather than appointed, and in states where the governor has the power to appoint high court judges, the most common arrangement limits gubernatorial choice by requiring nomination from a list forwarded by an independent nominating body.332Thirty-eight states elect all or nearly all of their judges. The most common arrangement is the “Missouri Plan,” in which judges are initially appointed by the governor and then subject to periodic “retention” elections. James A. Gardner & Guy-Uriel Charles, Election Law in the American Political System 502–03 (2d ed. 2018). Different tactics are thus required.

One such tactic has been court-packing—altering the size of the court. In 2016, for example, the Arizona legislature expanded the size of the Arizona Supreme Court from five to seven to permit the Republican governor to appoint two new justices.333Marin K. Levy, Packing and Unpacking State Courts, 61 Wm. & Mary L. Rev. 1121, 1136–38 (2020). That same year, the Georgia legislature enacted legislation to expand the Georgia Supreme Court from seven to nine.334Id. at 1139. At the time of the enactment, four of the justices were Democratic appointees and three were Republican appointees, allowing the Republican governor to alter the partisan balance on the court.335Id. Unsuccessful attempts at court-packing were made in Florida (2007) and in Iowa (2009), the latter following a decision by the Iowa Supreme Court striking down a state law prohibiting same-sex marriage.336Id. at 1139–42. Other unsuccessful attempts were made in South Carolina (initially by Democrats but then, in 2013, by Republicans) and Louisiana (2017).337Id. at 1142–43. Conversely, attempts have been made by Republican-controlled legislatures to shrink the size of the state supreme court to deny Democratic governors new appointments, most notably in Montana.338Id. at 1145–46. Additional attempts to shrink the size of state supreme courts have been made for reasons less easy to fathom, except in one case (Washington 2013) a bill to do so appears to have been introduced to punish the court for a disliked ruling. Id. at 1146–49. Relatedly, in 2017, the Republican-controlled North Carolina legislature enacted a law shrinking the number of intermediate appeals court judges from fifteen to twelve, apparently for partisan reasons. Melissa Boughton, An In-Depth Look at N.C. Lawmakers’ Attempt to Shrink the Court of Appeals, NC Pol’y Watch (Mar. 16, 2017), http://www.ncpolicywatch.com/2017/03/16/depth-look-n-c-lawmakers-attempt-shrink-court-appeals [https://perma.cc/QQ6G-BNAD]. When it then appeared that the North Carolina Supreme Court, controlled by Democrats, would invalidate the law, the legislature restored the additional judgeships. Melissa Boughton, Cooper Signs Bill Restoring State Court of Appeals Bench to 15 Judges, NC Pol’y Watch (Feb. 28, 2019), http://pulse.ncpolicywatch.org/2019/02/28/cooper-signs-bill-restoring-state-court-of-appeals-bench-to-15-judges [https://perma.cc/BS96-9ZZ2].

A more conventional tactic to control the judiciary in states where high court judges are elected has been to pour resources into contested judicial races, as well as to increase the brutality of the campaigns, which by longstanding prior custom had been low-key, genteel affairs of extremely low political salience.339Herbert M. Kritzer, Justices on the Ballot: Continuity and Change in State Supreme Court Elections 238–39 (2015). Spending on judicial races more than doubled between the 1990s and 2000s,340James Sample et al., Brennan Ctr. for Just., The New Politics of Judicial Elections: 2000–2009 1 (2010). and previously unheard-of amounts were poured into state supreme court races by outside interest groups.341Douglas Keith et al., Brennan Ctr. for Just., The Politics of Judicial Elections, 2017–18 1–2 (2019). Although the sources of these funds are nearly impossible to determine,342Id. at 2. it is highly likely that much of the funding was generated out-of-state. In 2017–2018, the highest levels of expenditures in judicial races by far were spent in eleven states: Alabama, Wisconsin, Michigan, West Virginia, Arkansas, North Carolina, Pennsylvania, Texas, Ohio, Nevada, and Georgia.343Id. at 3. During this period, Republicans already controlled the governor’s office and both chambers of the legislature in seven of these states (Alabama, Wisconsin, Michigan, Arkansas, Texas, Ohio, and Georgia), and controlled both chambers of the legislature in three (West Virginia, Pennsylvania, and North Carolina).3442017 State & Legislative Partisan Composition, Nat’l Conf. of State Legislatures (Mar. 1, 2017, 9:00 AM), https://www.ncsl.org/Portals/1/Documents/Elections/Legis_Control_2017_March_1_9%20am.pdf [https://perma.cc/9FZG-9GHA]; 2018 State & Legislative Partisan Composition, Nat’l Conf. of State Legislatures (Jan. 10, 2018), https://www.ncsl.org/Portals/1/Documents/Elections/Legis_Control_011018_26973.pdf. Nevada is the only state in this group in which Democrats controlled the legislature, although the governor was a Republican.345Annual data on partisan control of state government can be found at Ballotpedia. Gubernatorial and Legislative Party Control of State Government, Ballotpedia https://ballotpedia.org/Gubernatorial_and_legislative_party_control_of_state_government [https://perma.cc/3WY8-KRY8]. North Carolina had actually eliminated partisan judicial elections in 2004, but reinstated them in 2016 in what was widely seen as a nakedly partisan attempt by Republicans in the state legislature to maintain partisan control of the state bench.346Chris W. Bonneau & Melinda Gann Hall, In Defense of Judicial Elections 108–09 (2009); Anne Blythe, NC Lawmakers Create Partisan Election Process for Courts that Review Their Laws, News & Observer (Dec. 17, 2016, 7:48 AM), https://www.newsobserver.com/news/politics-government/state-politics/article121449157.html.

Efforts to control the state judiciary have not, however, been limited to attempts to win judicial seats in ordinary elections; they have also included attempts to inflict electoral punishment on judges who make rulings that either on the merits might be characterized as philosophically liberal, or which impede the ability of Republicans to maintain political control of the state.347See Neal Devins, State Constitutionalism in the Age of Party Polarization, 71 Rutgers U. L. Rev. 1129, 1166 (2019). Sometimes, where electoral punishment is not available, judges issuing such rulings have been threatened with impeachment.348Id.

For example, in 2010 three justices of the Iowa Supreme Court were ousted in retention elections—the first time any Iowa justice had not been retained—in retaliation for a 2009 decision legalizing same-sex marriage in the state.349Id. at 1160. The outcome of this election was heavily influenced by a surge of spending by out-of-state groups that routinely support Republican causes.350A. G. Sulzberger, Ouster of Iowa Judges Sends Signal to Bench, N.Y. Times (Nov. 3, 2010), https://www.nytimes.com/2010/11/04/us/politics/04judges.html; David Pozen, What Happened in Iowa?, 111 Colum. L. Rev. Sidebar 90, 98 (2011). Similar, though unsuccessful, efforts were mounted in Alaska, Colorado, Kansas, Illinois, and Florida “targeting supreme court justices for rulings on abortion, taxes, tort reform and health care.”351Sulzberger, supra note 351. In Pennsylvania, following a historic 2018 ruling by the state supreme court invalidating on state constitutional grounds a grotesque partisan gerrymander favoring Republicans, Republican leaders in the state legislature introduced a motion to impeach the four Democrats on the court who had ruled to invalidate the gerrymander.352Liz Navratil, Pa. Republican State Legislator Moves to Impeach Four State Supreme Court Justices, Phila. Inquirer (Mar. 20, 2018), https://www.inquirer.com/philly/news/politics/state/pennsylvania-congressional-map-impeachment-supreme-court-justices-20180320.html [https://perma.cc/SLR6-NHW7]. Although most of these efforts have been unsuccessful, even the unsuccessful ones subject theoretically independent courts to discipline imposed by the other branches because of their sheer in terrorem effect.353Devins, supra note 350, at 1158–62.

It is difficult, of course, to gauge whether and to what degree these efforts might be affecting the independence of state courts. Nevertheless, in a number of recent cases state high courts have issued surprising decisions upholding measures taken by the legislative or executive branches of state government that have been taken straight out of the authoritarian’s playbook—most notably, measures to suppress voting. For example, following passage in 2018 of an initiative amendment to the Florida Constitution eliminating permanent disenfranchisement of convicted felons, the Florida legislature enacted a law, clearly intended to thwart the voters’ intent, that prohibited restoration of voting rights to convicted felons until they had paid off all fees, costs, and other financial obligations associated with their convictions.354See S.B. 7066, 2019 Reg. Sess. (Fla. 2019). The Florida Supreme Court upheld the law.355Advisory Op. to the Governor Re: Implementation of Amend. 4, 288 So. 3d 1070, 1084 (Fla. 2020) (per curiam). The law was eventually struck down by federal courts. Jones v. Governor of Fla., 950 F.3d 795, 800, 807, 827 (11th Cir. 2020) (per curiam); Jones v. DeSantis, 462 F. Supp. 3d 1196, 1203, 1210 (N.D. Fla. 2020), rev’d & vacated sub nom., Jones v. Governor of Fla., 975 F.3d 1016 (11th Cir. 2020). However, the Eleventh Circuit has at this writing stayed the earlier opinions, and U.S. Supreme Court refused to lift the stay.

In Wisconsin, a pitched battle erupted between the Democratic governor and the Republican-controlled legislature over whether to take measures to make voting safer during the coronavirus pandemic.356Nick Corasaniti & Stephanie Saul, ‘Your Health or the Right to Vote’: A Battle in Wisconsin as Its Primary Nears, N.Y. Times (Mar. 27, 2020), https://www.nytimes.com/2020/03/27/us/politics/wisconsin-primary-coronavirus.html. The legislature repeatedly refused the governor’s requests to postpone the state’s spring election, including the Democratic presidential primary, and to require the Secretary of State to mail absentee ballots to all voters.357Alex Seitz-Wald & Shaquille Brewster, GOP Lawmakers Reject Wisconsin Governor’s Call for Delay in Election Deadline, NBC News (Apr. 4, 2020, 10:45 AM), https://www.nbcnews.com/politics/2020-election/wisconsin-governor-calls-special-legislative-session-consider-delaying-election-deadline-n1176186 [https://perma.cc/ZJ7Z-P6UD]. Eventually, the governor invoked his emergency powers and issued an executive order suspending in-person voting.358Wis. Exec. Order No. 74 (Apr. 6, 2020), https://elections.wi.gov/sites/elections.wi.gov/files/2020-04/EO074-SuspendingInPersonVotingAndSpecialSession.pdf [https://perma.cc/AR2L-5QVV]. The Wisconsin Supreme Court—one of the most bitterly partisan in the nation359Partisan disagreement among the justices was so severe that one justice accused another of assaulting her physically. Monica Davey, Wisconsin Judge Said to Have Attacked Colleague, N.Y. Times (June 25, 2011), https://www.nytimes.com/2011/06/26/us/26judge.html.—invalidated the governor’s order.360Wisconsin Legislature v. Evers, No. 2020AP608, at 3–4 (Wis. Apr. 6, 2020). In so doing, it left voters no option but either to risk their health by voting in person, or to take their chances with applying for an absentee ballot and hope it arrived in time. As it turned out, some 13,000 such ballots were never received by voters.361Henry Redman, Picture Still Murky One Day After Election, Wis. Examiner (Apr. 8, 2020), https://wisconsinexaminer.com/2020/04/08/picture-still-murky-one-day-after-election [https://perma.cc/3Q4T-PXVM]. Later, the court went on to rule that the legislature, and not the governor, had constitutional authority to deal with the pandemic and so invalidated the governor’s stay-at-home orders altogether.362Evers, No. 2020AP608, at 3–4.

Meanwhile, the Texas Supreme Court affirmed the position of the governor and attorney general overruling an interpretation of state law issued by various county clerks and election administrators that fear of contracting COVID-19 during a pandemic constituted a “disability” under state law entitling those who claimed such a fear to vote by absentee ballot.363In re State of Texas, 602 S.W.3d 549, 550, 567 (Tex. 2020). The Tennessee Supreme Court also affirmed the state’s position that fear of COVID-19 does not constitute an excuse for absentee voting, and overturned a lower court ruling that such a restriction during a pandemic unconstitutionally infringes the right to vote. Fisher v. Hargett, 604 S.W.3d 381, 391, 405 (Tenn. 2020). And in North Carolina, the state high court upheld a measure enacted by the Republican-controlled legislature diminishing the power of the incoming Democratic governor by for the first time subjecting gubernatorial cabinet appointments to Senate confirmation.364Cooper v. Berger, 822 S.E.2d 286, 290, 295, 300–01 (N.C. 2018).

It is worth mentioning that in reaching these kinds of rulings, state supreme courts are to some degree following a path marked out by the U.S. Supreme Court, which for decades has been controlled by a majority of Republican appointees.365The precise moment was 1970, when Nixon appointee Harry Blackmun succeeded Lyndon Johnson appointee Abe Fortas. Supreme Court Nominations (1789–Present), U.S. Senate (Oct. 26, 2020), https://www.senate.gov/legislative/nominations/SupremeCourtNominations1789present.htm. However, appointments to the Court were not nearly as politicized then as they became later, so probably a more meaningful tipping point from the perspective of self-consciously ideological appointments was reached in 1986, when Ronald Reagan elevated William Rehnquist to the Chief Justiceship and simultaneously appointed Antonin Scalia to replace him as Associate Justice. Id. Although state supreme courts are free to interpret state constitutions independently from the interpretations federal courts give to the U.S. Constitution, in practice they often take their lead from the U.S. Supreme Court, proceeding in “lockstep” with it366See generally Williams, supra note 278 (describing a lockstep approach in jurisprudence of individual rights); James A. Gardner, The Positivist Revolution that Wasn’t: Constitutional Universalism in the States, 4 Roger Williams U. L. Rev. 109, 111, 117, 129–30 (1998).—a practice for which they have often been criticized.367William J. Brennan, Jr., State Constitutions and the Protection of Individual Rights, 90 Harv. L. Rev. 489, 491, 502 (1977); Hans A. Linde, First Things First: Rediscovering the States’ Bills of Rights, 9 U. Balt. L. Rev. 379, 379–80, 383 (1980); Robert F. Williams, In the Supreme Court’s Shadow: Legitimacy of State Rejection of Supreme Court Reasoning and Result, 35 S.C. L. Rev. 353, 361, 389 (1984). In cases involving voting, the Supreme Court has over the years sided repeatedly with Republican initiatives, upholding burdensome photo ID requirements,368Crawford v. Marion Cnty. Election Bd., 553 U.S. 181, 198, 204 (2008). striking down the preclearance provisions of the Voting Rights Act,369Shelby Cnty. v. Holder, 570 U.S. 529, 556–57 (2013). and most recently, overturning a district court order directing the state of Wisconsin to accept late-arriving ballots during a primary election scheduled to be held in the midst of a pandemic.370Republican Nat’l Comm. v. Democratic Nat’l Comm., 140 S. Ct. 1205, 1206–08 (2020) (per curiam). See Garrett Epps, The Supreme Court Has Chosen Reckless Partisanship, The Atl. (Apr. 8, 2020, 12:20 PM), https://www.theatlantic.com/ideas/archive/2020/04/supreme-court-wisconsin/609631 [https://perma.cc/SN4G-A5NM].

3. Subdue civil society institutions

If a “lively civil society” is a critical requirement of a successful liberal democracy,371Diamond, supra note 6, at 19. then universities, as centers of inquiry committed to freedom of thought, speech, and learning, play an important role in a liberal state. This is precisely what makes them tempting targets in states experiencing democratic backsliding. Perhaps the paradigm here is Viktor Orbán’s persistent and ultimately successful efforts to push the Central European University, one of Europe’s most prestigious centers of independent learning and scholarship, out of Hungary altogether.372Franklin Foer, Viktor Orbán’s War on Intellect, The Atl. (June 2019), https://www.theatlantic.com/magazine/archive/2019/06/george-soros-viktor-orban-ceu/588070 [https://perma.cc/U9ZB-J875]. Orbán has attacked other universities as well. Benjamin Novak, Student Blockade Protests Viktor Orban’s Reach at a Top Arts University, N.Y. Times (Sept. 6, 2020), https://www.nytimes.com/2020/09/06/world/europe/hungary-students-blockade-orban.html.

Although the United States is probably too large, with too many universities, for aspiring authoritarians to exile or control them all, several Republican-controlled states have nevertheless waged what has been called a “[w]ar on [p]ublic [u]niversities.”373David Sarasohn, The Republican War on Public Universities, New Republic (Aug. 10, 2016), https://newrepublic.com/article/135972/republican-war-public-universities [https://perma.cc/N9K2-NVCF]; Derek Thompson, The Republican War on College, The Atl. (Nov. 20, 2017), https://www.theatlantic.com/business/archive/2017/11/republican-college/546308. One method by which this war has been waged is through sharp attacks on public employee unions that represent university employees, discussed further below. But the principal weapon in the war has been the state’s higher education budget, which in Wisconsin, Louisiana, Kansas, North Carolina, Arizona, Alaska, and Iowa, among others, has been cut severely.374Sarasohn, supra note 374; Bloomberg, Alaska’s Governor Makes Unprecedented Cuts in State University Spending, L.A. Times (July 6, 2019, 11:21 AM), https://www.latimes.com/nation/la-na-alaska-university-financial-brink-20190706-story.html; Sarah Watson et al., A New Front in America’s Pandemic: College Towns, N.Y. Times (Sept. 10, 2020), https://www.nytimes.com/2020/09/06/us/colleges-coronavirus-students.html.

Funding for higher education suffered around the country following the 2008 recession, as states cut deeply into all spending sectors to balance their budgets.375Tracy Gordon, State and Local Budgets and the Great Recession, Brookings (Dec. 31, 2012), https://www.brookings.edu/articles/state-and-local-budgets-and-the-great-recession [https://perma.cc/WT9R-BQ6V]; see Kim Clark, The Great Recession’s Toll on Higher Education, U.S. News (Sept. 10, 2010, 12:00 AM), https://www.usnews.com/education/articles/2010/09/10/the-great-recessions-toll-on-higher-education. But while spending climbed back steadily in most states following economic recovery, in some states it has not, and reductions have even continued.376‘Lost Decade’ Casts a Post-Recession Shadow on State Finances, Pew Charitable Trs. (June 4, 2019), https://www.pewtrusts.org/en/research-and-analysis/issue-briefs/2019/06/lost-decade-casts-a-post-recession-shadow-on-state-finances. In Louisiana, for example, per-student spending on higher education dropped 38% between 2008 and 2012377State Higher Educ. Exec. Officers Ass’n (SHEEO), State Higher Education Finance: FY 2017 27 tbl.4 (2018). and has continued to drop ever since, falling a total of 55% since the recession.378Sarasohn, supra note 374. Alabama’s spending on higher education fell by 36% following the recession but has fallen further since then.379SHEEO, supra note 378, at 27 tbl.4. The Republican governor of Alaska in 2019 vetoed $130 million of higher education spending resulting in a cut of 41% in state funding.380Bloomberg, supra note 375. Spending on higher education in Arizona fell 55% between 2008 and 2018.381Jim Small, Arizona Higher Ed: Deepest Cuts, Biggest Tuition Increases Since 2008, AZ Mirror (Oct. 24, 2019, 11:46 AM), https://www.azmirror.com/blog/arizona-higher-ed-deepest-cuts-biggest-tuition-increases-since-2008 [https://perma.cc/9MJ4-9C2N].

As political scientist Larry Sabato observed, these kinds of cuts were predictable: “Most GOP elected officials believe that universities are hotbeds of Democratic support—and the voting patterns in most college precincts support this.”382Sarasohn, supra note 374. Or, as a Republican Tennessee legislator remarked in a moment of candor, higher education in the state should be eliminated because it serves as a “liberal breeding ground.”383Tennessee Lawmaker Says He Supports Getting Rid of Higher Education, Later Calls Comments ‘Hyperbole’, Tennessean (Sept. 10, 2019, 10:09 AM), https://www.tennessean.com/story/news/politics/2019/09/09/tennessee-lawmaker-kerry-roberts-calls-removal-higher-education/2269207001 [https://perma.cc/B2WE-GHFW].

In Florida, just to make sure that the state university’s board of trustees did not attempt to mitigate the impact of reductions in funding, a 2018 amendment to the Florida Constitution changed the voting rules for the state university’s board of trustees on a single matter—raising fees charged to students—by requiring an affirmative vote of nine of the thirteen trustees, a 70% supermajority.384Fla. Const. art. IX, § 7 (2018). This change guaranteed that the six members of the board appointed by the governor could effectively block any attempt by public universities to collect additional operating revenue.385See Florida Amendment 7, First Responder and Military Member Survivor Benefits, Supermajority Board Votes for College Fees, and State College System Amendment (2018), Ballotpedia, https://ballotpedia.org/Florida_Amendment_7,_First_Responder_and_Military_Member_Survivor_Benefits,_Supermajority_Board_Votes_for_College_Fees,_and_State_College_System_Amendment_(2018). The amendment also constitutionalized the existing statutory structure of the state university system so that future administrations would be unable to alter it.386Id.

4. Attacks on economically powerful actors that support the opposition

In other countries experiencing democratic backsliding, attacks on disfavored holders of economic power tend to follow the model pursued by Turkish President Recep Erdoğan, who, as a tactic to consolidate power, “threatened businesses that might dare to fund opposition parties.”387Diamond, supra note 6, at 56. In Russia, Vladimir Putin went much further, neutralizing potential financial backers of opposition parties through a program of outright state confiscation of the business holdings of uncooperative Russian oligarchs,388See id. (describing how state asset control ensured loyalty to the new regime); Ginsburg & Huq, supra note 6, at 107 (describing how Putin used selective enforcement of tax laws to control opposition). and even conducting criminal prosecutions charging tax evasion, fraud, money laundering, and embezzlement.389Varol, supra note 208, at 1708. In the United States, the equivalent behavior consists of a constant, systematic barrage aimed at labor unions, which historically have been among the most consistent and best-resourced supporters of Democratic candidates and their policies.390Taylor E. Dark, The Unions and the Democrats: An Enduring Alliance 1–3, 5–10 (1999). See generally Daniel Schlozman, When Movements Anchor Parties: Electoral Alignments in American History ch. 3 (2015). For recent data, see Labor: Long-Term Contribution Tends, OpenSecrets, https://www.opensecrets.org/industries/totals.php?cycle=2020&ind=P [https://perma.cc/87V9-2266]. Some have suggested that Republican attacks on trial lawyers also fit this pattern. Mark Tushnet, Taking Back the Constitution 59 (2020); Michael J. Klarman, Foreword: The Degradation of American Democracy–and the Court, 134 Harv. L. Rev. 4, 39, 87, 111, 145 (2020).

Some American states have long attempted to weaken unions through such measures as “right-to-work” laws.391Ross Eisenbrey, Right-To-Work-Laws: Designed to Hurt Unions and Lower Wages, Econ. Pol’y Inst. (Mar. 2, 2015, 3:45 PM), https://www.epi.org/blog/right-to-work-laws-designed-to-hurt-unions-and-lower-wages [https://perma.cc/67VS-8LPY]. These laws, which prohibit mandatory union membership or mandatory payment of the equivalent of union dues as a condition of employment, weaken unions by allowing non-union employees to free-ride on the organizational power of dues-paying union members, thereby undermining the incentives of workers to join and financially support unions.392David T. Ellwood & Glenn Fine, The Impact of Right-to-Work Laws on Union Organizing, 95 J. Pol. Econ. 250, 250–52, 266, 271 (1987). The loss of membership and dues reduces union resources, thus limiting the capacity of unions to contribute resources to political candidates and causes.393See id. at 266, 270–71; Maurice C. Benewitz, Nature and Effect of State Right-to-Work Laws, 1 Wayne L. Rev. 165, 181, 185 (1955); Raymond L. Hogler, How Right to Work Is Destroying the American Labor Movement: From the Ku Klux Klan to the Tea Party, 23 Emp. Resp. & Rts. J. 295, 300–01 (2011).

Such measures continue to the present day, even more aggressively in the form of attempts to constitutionalize right-to-work principles so as to insulate them from alteration or repeal should partisan control of state government change hands.394Dinan, supra note 282, at 248–50. Constitutionalization of what are ordinarily considered routine policy choices is a common tactic of state constitutional politics. Id. at 249. In 2016, for example, Alabama voters approved Amendment 8, a legislatively referred amendment that constitutionalized existing state policy prohibiting any requirement of union membership as a condition of employment.395Alabama Right to Work, Amendment 8 (2016), Ballotpedia, https://ballotpedia.org/Alabama_Right_to_Work,_Amendment_8_(2016) [https://perma.cc/3QWH-VZ67]. A similar measure, also referred to voters by the state legislature, appeared on the 2016 Virginia ballot but was rejected.396Virginia Right to Work Amendment, Question 1 (2016), Ballotpedia, https://ballotpedia.org/Virginia_Right_to_Work_Amendment,_Question_1_(2016) [https://perma.cc/63Y8-YCMF].

A more recently adopted strategy involves legislatively prohibiting state agencies bidding out contracts from requiring bidders to use union employees. A 2017 Wisconsin law, for example, not only prohibits state agencies from requiring a labor agreement as a condition of submitting a bid, but forbids even any consideration of the existence of such an agreement as a factor in awarding contracts.397S.B. 3 (Wis. 2017). Republican-controlled states have also taken measures to prevent Democrat-controlled localities within the state—typically large urban areas—from using otherwise available powers of local autonomy to insist on contract terms that contain or mimic terms negotiated by labor unions. A 2016 Alabama law, for example, withdraws local authority to regulate terms of employment, and prohibits local governments from requiring a minimum wage, minimum leave, or any other employee benefits, whether for their own employees or for the employees of contractors.398Alabama Uniform Minimum Wage and Right-to-Work Act, H.B. 174 (Ala. 2016).

The scope of these attacks seems in recent years to have expanded to include generalized attacks on workers themselves, whether unionized or not. For example, during the current coronavirus pandemic, Iowa and Oklahoma announced that fear of contracting the virus would not count, for purposes of eligibility for unemployment insurance, as a valid excuse for turning down an offer to return to work.399Tony Romm, Republican-Led States Signal They Could Strip Workers’ Unemployment Benefits if They Don’t Return to Work, Sparking Fresh Safety Fears, Wash. Post (Apr. 30, 2020, 8:08 AM), https://www.washingtonpost.com/business/2020/04/30/republican-states-unemployment-benefits. Some states are presently considering legislation to relieve employers of liability should returning workers contract COVID-19, measures that would weaken incentives employers might otherwise have to take stringent measures to protect their own workforce.400Ana Swanson & Alan Rappeport, Businesses Want Virus Legal Protection. Workers Are Worried, N.Y. Times (June 12, 2020), https://www.nytimes.com/2020/06/12/business/economy/coronavirus-liability-shield.html. In view of longstanding attacks on unionized workers, it is difficult to escape the conclusion that the best explanation for this behavior is a desire to keep all workers in a terrorized condition of compliance with official policy.

5. Attacks on the civil service

At the federal level, attacks on civil servants have been fierce, though mainly rhetorical. Former President Trump notoriously labeled the federal civil service a “deep state,” which he characterized as opposing, and through bureaucratic means undermining, his various programs and initiatives.401Evan Osnos, Trump vs. the “Deep State,” New Yorker (May 14, 2018), https://www.newyorker.com/magazine/2018/05/21/trump-vs-the-deep-state [https://perma.cc/6AVL-H5LR]. Although Trump waged this campaign mainly verbally rather than through action, early in his term he issued several executive orders meant to weaken the federal civil service by restricting the role of federal public employee unions in the workplace and increase the flexibility available to agencies in disciplining civil servants.402Lisa Rein, Trump Takes Aim at Federal Bureaucracy with New Executive Orders Rolling Back Civil-Service Protections, Wash. Post (May 25, 2018, 9:00 PM), https://www.washingtonpost.com/politics/trump-takes-aim-at-federal-bureaucracy-with-new-executive-orders-altering-civil-service-protections/2018/05/25/3ed8bf84-6055-11e8-9ee3-49d6d4814c4c_story.html.

At the state level, there has been less bluster but more substance. These attacks, initiated by Republican administrations, appear to have two motives. One motive is to weaken legal protections long meant to professionalize and de-politicize the civil service, thereby providing governors with significantly more power to hire and fire, to engage in patronage employment practices, and to tighten their control over and ability to discipline state bureaucracies.403See, e.g., Howard Fischer, Brewer Signs Bill for Arizona At-Will Workforce, Tuscon.com (May 11, 2012), https://tucson.com/news/local/govt-and-politics/brewer-signs-bill-for-arizona-at-will-workforce/article_945fc45d-5592-5b68-a4ba-07cbc1b35f87.html [https://perma.cc/ES7A-SLNW]. A second motivation appears to be to punish unions generally, and public employee unions in particular, for their tendency to provide financial support to the Democratic opposition, and to weaken them by drying up their resources, thereby reducing their political efficacy.404Rick Ungar, ‘Right-to-Work’ Laws Explained, Debunked and Demystified, Forbes (Dec. 11, 2012, 2:37 PM), https://www.forbes.com/sites/rickungar/2012/12/11/right-to-work-laws-explained-debunked-demystified/?sh=172f79b3480b.

By far the most significant effort to de-professionalize a state civil service is an Arizona statute enacted in 2012 with strong backing from then-Governor Jan Brewer. That law ended civil service protections for all state employees hired after September 29, 2012, making their employment expressly at-will.405Ariz. Rev. Stat. § 41-742(A)(1) (2020). It also provided that any state employee who voluntarily accepts a new job assignment within Arizona government—a promotion, for example—also becomes at-will.406Id. § 41-472(A)(3). To an employee population that had been subject to a wage freeze for four years,407Melissa Maynard, Civil Service Reform Passes in 3 States, Stateline (June 12, 2012, 2:20 PM), https://www.governing.com/blogs/view/civil-service-reform-passes.html. the law additionally offered a one-time 5% pay bonus to any employee who agreed to move from protected to at-will status,408Fischer, supra note 404. but provided expressly that any change in status from protected to at-will would be irrevocable.

A more common strategy deployed against civil servants has been to attack public employee unions. The first major broadside occurred in 2011 when, with the energetic backing of Governor Scott Walker, the Wisconsin legislature enacted Act 10, a collection of measures meant to badly weaken the state’s public employee unions.409Steven Greenhouse, Wisconsin’s Legacy for Unions, N.Y. Times (Feb. 22, 2014), https://www.nytimes.com/2014/02/23/business/wisconsins-legacy-for-unions.html. Among its provisions, Act 10 eliminated all collective bargaining rights for public university employees and health care workers, and “effectively abolished collective bargaining for all [other] public employees,” except law enforcement and fire personnel,410Martin H. Malin, The Legislative Upheaval in Public-Sector Labor Law: A Search for Common Elements, 27 A.B.A. J. Lab. & Emp. L. 149, 156 (2012). by stripping public employee unions of the authority to bargain over any terms of employment other than base wages, and then only within a range capped by the annual rate of inflation.411Paul M. Secunda, The Wisconsin Public-Sector Labor Dispute of 2011, 27 A.B.A. J. Lab. & Emp. L. 293, 296 (2012). Act 10 also added “clearly punitive provisions” unrelated to the Governor’s ostensibly budgetary justifications, including a burdensome requirement of annual recertification of public employee unions by vote of an absolute majority of employees (rather than just a majority of those voting), and a prohibition on paycheck deductions for union dues, requiring unions to divert resources to collection of dues.412Id. at 297.

Oklahoma and Tennessee also repealed collective bargaining rights of public employees in 2011.413Malin, supra note 411, at 154. Nevada and Ohio restricted the availability of collective bargaining to certain classes of public employees, though Ohio’s measures were subsequently reversed by referendum.414Id. at 150, 156. Many similar measures have been adopted since then. For example, 2017 legislation adopted in Iowa and Kentucky prohibits collective bargaining by public employee unions on dues checkoffs and payroll deductions.415Tom Loftus, Labor Unions File Suit to Block Right-to-Work Law, Courier J. (May 25, 2017, 5:55 PM), https://www.courier-journal.com/story/news/politics/2017/05/25/labor-unions-file-suit-block-right-to-work-law/344955001 [https://perma.cc/EY27-P4YF]; Jason Noble & Brianne Pfannenstiel, Here Are the 5 Key Changes in Iowa’s Collective Bargaining Bill, Des Moines Reg. (Feb. 8, 2017, 7:25 PM), https://www.desmoinesregister.com/story/news/politics/2017/02/08/here-5-key-changes-iowas-collective-bargaining-bill/97658446 [https://perma.cc/4BP4-7XKG]. The Iowa measure also requires mandatory union retention elections before the expiration of any public employee contract, and the Kentucky law prohibits strikes by public employees.416Iowa Code § 20.15(2)(a) (2020); Ky. Rev. Stat. Ann. § 336.130 (West 2020). These and other measures are usefully collected at the website of the National Conference of State Legislatures. See Union Legislation and Collective Bargaining Law, Legislation, Nat’l Conf. of State Legislatures (Sept. 9, 2020), https://www.ncsl.org/research/labor-and-employment/collective-bargaining-legislation-database.aspx [https://perma.cc/GNU4-VJZS]. In 2017, Illinois Republican Governor Bruce Rauner vetoed the Collective Bargaining Freedom Act,417Collective Bargaining Freedom Act, S. 1905, 100th Gen. Assemb. (Ill. 2017). a measure enacted by the Democrat-controlled state legislature to establish the right of all employees, including public employees, to bargain collectively.418Id.; Kim Geiger, Rauner Vetoes Bill that Would Prevent Local ‘Right-to-Work’ Zones, Chi. Trib. (Sept. 29, 2017, 5:38 PM), https://www.chicagotribune.com/politics/ct-bruce-rauner-collective-bargaining-met-0930-20170929-story.html.

6. Manipulation of electoral rules and processes

By far the most significant steps by Republican-controlled states down the path toward authoritarianism are an extensive suite of measures designed to manipulate electoral rules and practices to favor the interests of the Republican Party, and indeed to do so to such an extent as to ensure as far as possible that the party and its adherents retain power, even in the face of future erosion of public support, should it occur.

The baseline assumptions in modern democracies are (1) that suffrage should be universal and (2) that all who wish to participate in democratic self-rule should have a reasonable opportunity to do so.419See International Covenant on Civil and Political Rights art. 25, Dec. 19, 1966, 999 U.N.T.S. 171 (creating a universal human right to democracy); Thomas M. Franck, The Emerging Right to Democratic Governance, 86 Am. J. Int’l L. 46, 64 (1992). But see Amnon Rubinstein & Yaniv Roznai, The Right to a Genuine Electoral Democracy, 27 Minn. J. Int’l L. 143, 147 (2018) (calling for more clarity in international law of requirements for free elections). Republican-controlled states have launched sustained and forceful attacks against both of these premises of popular sovereignty. These measures have by now been so well-documented that it will suffice to review them only briefly.

a. Attacks on voter eligibility

Under the constitutional scheme, states control eligibility to vote in all elections,420U.S. Const. art. I, § 2; id. art. II, § 1. States control eligibility for their own elections. See, e.g., Lassiter v. Northampton Cnty. Bd. of Elections, 360 U.S. 45, 50 (1959) (“The States have long been held to have broad powers to determine the conditions under which the right of suffrage may be exercised.”). Of course, the Constitution places limits on how states manage this task. See U.S. Const. art. I, § 4 (“Congress may at any time by Law make or alter [state election] Regulations . . . .”); id. amends. I, XIV (free speech, equal protection, and due process clauses). and for more than a century registration has been a nearly universal prerequisite to voting.421See Keyssar, supra note 242, at 65 (detailing initial spread of registration requirements in the 1830s). North Dakota is the only state that does not require registration. Elections Unit, N.D. Sec’y of State, North Dakota . . . The Only State Without Voter Registration (2017). Moreover, unlike in many other countries where voter registration is considered to be the responsibility of the state, in the United States the burden of registering to vote has long been laid upon citizens.422Jennifer S. Rosenberg with Margaret Chen, Brennan Ctr. for Just., Expanding Democracy: Voter Registration Around the World 1 (2009). This procedural quirk has created an opening for vote suppression through state regulatory measures designed to increase the difficulty and burden of registering to vote. Below is a partial listing of some of the measures Republican-controlled states have taken to restrict voter eligibility.

  • Proof of citizenship. The “most restrictive” of these measures, enacted in Kansas, Arizona, Alabama, and Georgia, requires those seeking to register to produce documentary proof of U.S. citizenship.423Ben Merriman, Conservative Innovators: How States Are Challenging Federal Power 101 (2019). Although U.S. citizenship may legitimately be required as a precondition for voting,424Sugarman v. Dougall, 413 U.S. 634, 649 (1973). A citizenship requirement is not, however, mandatory. Jamin B. Raskin, Legal Aliens, Local Citizens: The Historical, Constitutional and Theoretical Meanings of Alien Suffrage, 141 U. Pa. L. Rev. 1391, 1393–94 (1993). and is in fact everywhere required as a condition of eligibility to vote in federal and state elections, in the past a registrant’s affirmation of citizenship under penalty of perjury has been deemed sufficient evidence to satisfy the requirement.425This is, for example, the only federal requirement. See 42 U.S.C. § 1973gg–7(b)(2). On the state level, a typical example is New York’s voter registration form, which requires a voter to check a box inquiring about citizenship and then swear or affirm to the truth of the statement. N.Y. State Voter Registration Form, https://www.elections.ny.gov/NYSBOE/download/voting/voteregform-eng-fillable.pdf [https://perma.cc/B5Z2-U44F]. These new provisions require production of a birth certificate, naturalization papers, or a U.S. passport—documents that not all prospective voters possess or can easily obtain. Where implemented, these policies have had a considerable impact: “In Kansas, nearly 20 percent of new registrations were suspended for lack of proof of citizenship,” and in the 2014 general election, a number of voters equal to 2.5% of total turnout were affected.426Merriman, supra note 424, at 103. Moreover, “Democrats were 1.55 times more likely than Republicans to have their registrations suspended,” and “[u]naffiliated voters were 3.8 times more likely than Republicans” to face suspension of registration.427Id.
  • Registration purges. Under the federal National Voter Registration Act of 199342842 U.S.C. § 1973gg. (NVRA or “Motor Voter” law), states are required to take steps to maintain the accuracy of their voter registration rolls on the theory that inaccuracies create opportunities for fraudulent voting.429See id. § 1973gg-6(a)(4), (b) (requiring states to implement programs to remove ineligible voters from eligible voter lists because maintain “accurate and current voter registration roll[s]” will “protect the integrity of the electoral process”). Republican-controlled states have exploited this mandate to conduct massive purges of voter registration rolls.430See id. § 1973gg-6(b)(2) (allowing states to remove unresponsive voters and voters who have not voted or appeared to vote in two or more consecutive general election under the purported goal of maintaining accurate and current voter registration rolls). These purges have been badly inaccurate and have disproportionately stripped from voter registration rolls “the poor, minorities, and the young”—groups that tend to support Democrats.431Anderson, supra note 235, at 80. In a five-year period, for example, Ohio Secretary of State Jon Husted purged two million names from the state’s voter rolls by using a method that purported to identify obsolete listings through a combination of non-response to mailed inquiries and subsequent non-voting for a period of four years.432Id. at 75; see also Husted v. A. Philip Randolph Inst., 138 S. Ct. 1833, 1846 (2018) (upholding Ohio’s method as consistent with the NVRA). Under this system, voters in neighborhoods that supported Barack Obama for President in 2012 saw more than twice as many registration purges as neighborhoods where Obama earned less than 40% of the vote.433Anderson, supra note 235, at 77. Similarly, Kansas Secretary of State Kris Kobach tirelessly promoted the “Crosscheck” system, which purported to identify obsolete voter registrations by comparing databases across states; the system, however, has “an error rate of more than 99 percent.”434Id. at 82, 87.
  • Restrictions on voter registration drives. A standard tactic of campaign mobilization has long been drives to register new voters. Research finds, moreover, that “minority voters are nearly twice as likely as white voters to register through a drive.”435Merriman, supra note 424, at 104. Several states have taken aim at these drives in an effort to suppress participation. A 2012 Florida law, for example, imposed onerous fines on civic groups conducting voter registration drives that fail to turn in the forms within forty-eight hours or maintain copious records, causing the League of Women Voters and Rock the Vote, two major, nonpartisan supporters of voter registration, to suspend efforts in the state and later to sue.436Michael Cooper & Jo Craven McGinty, Florida’s New Election Law Blunts Voter Drives, N.Y. Times (Mar. 28, 2012), https://archive.nytimes.com/query.nytimes.com/
    gst/fullpage-9F00E4DF133CF93BA15750C0A9649D8B63.html. A federal district court in Florida enjoined the law in League of Women Voters v. Browning, 863 F. Supp. 2d 1155, 1157, 1167 (N.D. Fla. 2012).
    Under a 2011 Texas law, registration drives could be conducted only by officially appointed “volunteer deputy registrars” who had to be Texas residents and eligible voters, and were required to complete a training course offered only intermittently.437Ari Berman, Texas’s Voter-Registration Laws Are Straight out of the Jim Crow Playbook, Nation (Oct. 6, 2016), https://www.thenation.com/article/archive/texass-voter-registration-laws-are-straight-out-of-the-jim-crow-playbook. The law also criminalized failure to turn in new registration forms within five days.438Diana Kasdan, Brennan Ctr. for Just., State Restrictions on Voter Registration Drives 46 (2012); Berman, supra note 438.
  • Legislative subversion of loosening of registration requirements. In 2018, Florida voters approved an initiative amending the state constitution to restore the right to vote to convicted felons.439Tim Mak, Over 1 Million Florida Felons Win Right to Vote with Amendment 4, Nat’l Pub. Radio (Nov. 7, 2018, 2:46 AM), https://www.npr.org/2018/11/07/665031366/over-a-million-florida-ex-felons-win-right-to-vote-with-amendment-4. In so doing, Florida voters reversed a voting disqualification so stringent that it disqualified nearly one-third of adult Black males in the state.440Gardner & Charles, supra note 333, at 180. Immediately, in the guise of implementing the amendment, the Republican-controlled legislature enacted a law instead subverting it by establishing as a prerequisite for restoration of voting rights the repayment in full of all outstanding fines, fees, and restitution—a requirement impossible to meet for most convicted felons upon the conclusion of their prison terms.441Corey Goldstone, One Year After Historic Amendment 4: Can Floridians with past Felony Convictions Exercise Their Right to Vote?, Campaign Legal Ctr. (Jan. 8, 2020), https://campaignlegal.org/update/one-year-after-historic-amendment-4-can-floridians-past-felony-convictions-exercise-their [https://perma.cc/W7F6-P744]. Although a federal district court enjoined the law, the U.S. Court of Appeals for the 11th Circuit upheld the law as constitutional. Jones v. Governor of Fla., 975 F.3d 1016, 1025 (11th Cir. 2020).

b. Suppression of voting by eligible voters 

These measures aimed at curtailing eligibility have in many Republican-controlled states been supplemented by measures designed to reduce voter turnout among those who are already eligible, and in ways that “have the most marked effects on the poor and members of racial and ethnic minority groups,”442Merriman, supra note 424, at 91. the natural and likely constituency of the Democratic Party. Again, below is a partial listing of such measures.

  • Voter identification requirements. For most of American history, registered voters appearing at a polling place to vote were required to identify themselves either by simply stating their identity, or by signing a poll book, which constituted a legally binding affirmation of their identity.443With the exceptions of Texas and Florida, states did not begin requiring voters to produce any form of identification, photographic or otherwise, until 1980, and no state required photographic identification before 1995. See Rene R. Rocha & Tetsuya Matsubayashi, The Politics of Race and Voter ID Laws in the States: The Return of Jim Crow?, 67 Pol. Rsch. Q. 666, 669, 670 tbl.1 (2014). Voter impersonation is in every state a felony, and as a result, the incidence of fraudulent voting through impersonation is approximately zero.444Lorraine C. Minnite, The Myth of Voter Fraud 29 (2010); see Anderson, supra note 235, at 50–51, 65. Nevertheless, a nationwide, coordinated Republican effort over the last fifteen years has led to the enactment of photo ID requirements as a prerequisite to voting in numerous states.445Richard L. Hasen, The Voting Wars: From Florida 2000 to the Next Election Meltdown 43 (2012); Keesha Gaskins & Sundeep Iyer, Brennan Ctr. for Just., The Challenge of Obtaining Voter Identification 2 (2012). Ten states—Alabama, Georgia, Indiana, Kansas, Mississippi, Pennsylvania, South Carolina, Tennessee, Texas, and Wisconsin—have enacted photo ID requirements that are both highly restrictive and highly selective in the kind of documentation that will be accepted.446Gaskins & Iyer, supra note 446, at 1. For example, in Texas and Tennessee, acceptable forms of identification include gun licenses—documents more likely to be possessed by whites than Blacks447Anderson, supra note 235, at 69.—but not student ID cards.448Shugerman, supra note 237, at 112–13. In invalidating a North Carolina law imposing strict photo ID requirements, the U.S. Court of Appeals for the Fourth Circuit noted that the law “retained only those types of photo ID disproportionately held by whites and excluded those disproportionately held by African Americans.”449N.C. State Conf. of the NAACP v. McCrory, 831 F.3d 204, 215, 227 (4th Cir. 2016).
  • Making voting inconvenient. Although states are under no obligation to make voting easy, many states by the early 2000s had introduced measures to ease the burden of voting. These measures included adding polling places, extending polling place hours, and introducing early voting before Election Day.450See Leslie Wayne, The 2000 Campaign: Alternative Voting; Popularity Is Increasing for Balloting Outside the Box, N.Y. Times (Nov. 4, 2000), https://www.nytimes.com/2000/11/04/us/2000-campaign-alternative-voting-popularity-increasing-for-balloting-outside-box.html. These kinds of measures proved extremely popular with African-American voters; according to some estimates, up to 70% of Black voters in some jurisdictions voted during the early voting period.451Wendy Weiser & Erik Opsal, Brennan Ctr. for Just., The State of Voting in 2014 3 (2014). Between 2010 and 2014, however, eight Republican-controlled states—Florida, Georgia, Nebraska, North Carolina, Ohio, Tennessee, West Virginia, and Wisconsin—cut back on these measures.452Id. at 1, 3 & n.16; see also Zachary Roth, The Great Suppression: Voting Rights, Corporate Cash, and the Conservative Assault on Democracy 42 (2016) (covering Republican efforts in Ohio to cut evening and weekend voting and other measures). For example, a 2013 North Carolina law cut back the number of days of early voting by nearly half; eliminated an existing county option to offer early voting hours between 1 p.m. and 5 p.m.; and required early voting hours be uniform throughout the state without regard to actual utilization patterns.453See Complaint at 6–7, United States v. North Carolina, No. 1:13-cv-861, 2014 WL 494911 (M.D.N.C. Feb. 6, 2014). One consequence of these cutbacks has been to increase lines at polling places significantly in some neighborhoods, disproportionately Democratic-leaning ones.454See Justin Levitt, “Fixing that”: Lines at the Polling Place, 28 J.L. & Pol. 465, 468 (2013).

A newly developed tactic that greatly increases the stakes is a 2019 Indiana law that strips voters of standing to petition state courts to extend polling place hours to accommodate voters waiting in long lines and hoping to vote before the polls close.455Ind. Code § 3-11.7-7-2(a) (2020), enjoined by Common Cause Ind. v. Lawson, No. 1:20-cv-01825-RLY-TAB, 2020 WL 5671506, at *1, *8 (S.D. Ind. Sept. 22, 2020), rev’d, 978 F.3d 1036 (7th Cir. 2020). Most recently, Republican-controlled states have resisted calls to facilitate voting in the 2020 cycle, which is occurring during an unprecedented pandemic, by expanding the availability of absentee voting, or by mailing absentee ballots preemptively to all voters, whether or not they have requested a ballot.456Scott Bauer, GOP in Power Grab to Rein in Democrat Governors on Virus Response, Associated Press (May 9, 2020), https://apnews.com/article/308ae16f3260dfe026f3b050302db5bf; Laura A. Bischoff, Coronavirus: Elections Reform Bill Sparks Controversy, Dayton Daily News (June 4, 2020), https://www.daytondailynews.com/news/state–regional-govt–politics/coronavirus-elections-reform-bill-sparks-controversy/q9ENhcYhfIO2sZAvKQXz5O [https://perma.cc/86G5-U67Z]; Editorial, Republicans Are Engineering an Electoral Disaster This Fall, Wash. Post (June 10, 2020, 2:41 PM), https://www.washingtonpost.com/opinions/republicans-are-engineering-an-electoral-disaster-this-fall/2020/06/10/15afd038-ab3b-11ea-9063-e69bd6520940_story.html. The only conceivable explanation for this behavior is to suppress voting by forcing voters to choose between risking their health and casting a vote. The Republican Governor of Missouri, Mike Parsons, has essentially admitted as much, encouraging voters who “don’t feel safe” to “stay home.”457Editorial, Missouri Gov. Mike Parson: If You Don’t feel Safe, Just Don’t Vote. That’s Democracy?, Kan. City Star (May 29, 2020, 9:23 AM), https://www.kansascity.com/opinion/editorials/article243081151.html.

c. Combined impact of attacks on voters

Taken individually, none of these measures necessarily has a significant impact on the results of most elections. Such restrictions, however, often are packaged together. Alabama, for example, recently adopted a proof-of-citizenship requirement, a strict photo ID law, and other measures that makes voting less convenient.458Merriman, supra note 424, at 102 tbl.4.1. See generally Maggie Astor, Seven Ways Alabama Has Made It Harder to Vote, N.Y. Times (June 23, 2018), https://www.nytimes.com/2018/06/23/us/politics/voting-rights-alabama.html (describing Alabama’s voting restrictions). Kansas, Texas, and Virginia also have recently adopted three different kinds of measures aimed at suppressing voting.459Examples include, for Kansas, requiring “[d]ocumentary proof of citizenship”; for Texas, “[c]ut[ting] back use of mobile early voting sites”; and for Virginia, requiring “photo ID . . . to vote and limit[ing] third-party voter registration.” See Brennan Ctr. for Just., New Voting Restrictions in America 4–5, 8 (2019). After losing before the court of appeals, Kansas filed a petition before the Supreme Court for its documentary-proof requirement. See Fish v. Schwab, 957 F.3d 1105, 1111 (10th Cir. 2020). Most remarkable of all, perhaps, is North Carolina, which in 2013, following almost immediately on the heels of the Supreme Court’s evisceration of the preclearance provisions of the Voting Rights Act in Shelby County v. Holder,460570 U.S. 529 (2013). enacted a law that

included a strict voter identification provision[,] . . . cut a week off early voting, . . . barred local election boards from keeping the polls open after 1 p.m. on the final Saturday before the election[,] . . . eliminated same-day voter registration[,] . . . eliminated preregistration of sixteen- and seventeen-year-olds in high schools[,] . . . [and] provided that a voter who votes in the wrong precinct . . . would have her or his whole ballot thrown out . . . .461Richard L. Hasen, Election Meltdown: Dirty Tricks, Distrust, and the Threat to American Democracy 34–35 (2020).

It is difficult to imagine a justification capable of supporting a measure so thoroughly hostile to democratic processes other than a desire to suppress voting.462It could well be argued that another tactic of this type is manipulation of the scope of the demos itself through state withdrawal of powers of self-governance from large cities, which are typically controlled by Democrats. See, e.g., Richard C. Schragger, The Attack on American Cities, 96 Tex. L. Rev. 1163, 1165 (2018).

7. Gain control over body that runs elections

Unlike many other countries, the United States does not have a centralized, professional agency that regulates and administers elections.463Richard L. Hasen, Beyond the Margin of Litigation: Reforming U.S. Election Administration to Avoid Electoral Meltdown, 62 Wash. & Lee L. Rev. 937, 984 (2005). Instead, the power to make the electoral laws that govern elections for federal, state, and local offices, and to direct the manner of their administration, is lodged in the hands of state legislatures.464U.S. Const. art. I, § 4. In the United States, then, to control the body that runs elections means controlling state legislatures. The main route to control of a state legislature is the achievement of partisan dominance, and the main technique for obtaining such control—in the absence of confidence in actual popular support—is partisan gerrymandering. Although both parties have engaged in partisan gerrymandering from time to time,465For an historical overview of state efforts to confront partisan gerrymandering, see James A. Gardner, Foreword, Representation Without Party: Lessons from State Constitutional Attempts to Control Gerrymandering, 37 Rutgers L.J. 881 (2006) [hereinafter Gardner, Representation Without Party]. since the 2010 redistricting cycle the Republican Party has made it a top priority, one that has been implemented aggressively and successfully by Republican-controlled state legislatures.466Vann R. Newkirk II, How Redistricting Became a Technological Arms Race, The Atl. (Oct. 28, 2017), https://www.theatlantic.com/politics/archive/2017/10/gerrymandering-technology-redmap-2020/543888.

Knowing that state legislatures elected in 2010 would control redistricting following the 2010 Census, Republicans invested heavily in state legislative elections in 2010.467See Republicans Exceed Expectations in 2010 State Legislative Elections, Nat’l Conf. of State Legislatures (Nov. 3, 2010), https://www.ncsl.org/press-room/republicans-exceed-expectations-in-2010.aspx (describing how Republicans fielded more candidates for state legislature races than Democrats were able to put forward); Communications, Election 2010 to Shatter Spending Records as Republicans Benefit from Late Cash Surge, OpenSecrets.org (Oct. 27, 2010, 4:00 PM), https://www.opensecrets.org/news/2010/10/election-2010-to-shatter-spending-r (providing an overview of how Republicans outraised and received more outside money than Democrats in 2010). The results were spectacular: Republicans picked up 675 state legislative seats around the nation, flipped control of 21 state legislative chambers, and gained control of both chambers in 25 states; Democrats managed to control only 13.468Roth, supra note 453, at 98–99. Among the states Republicans then dominated were the nationally important battleground states of Florida, Pennsylvania, Ohio, Michigan, North Carolina, and Virginia.469Id. at 99.

Under a plan coordinated by the national Republican Party, known as REDMAP, redistricting experts, using software far more sophisticated than had previously been used, redrew legislative maps in Republican-controlled states to maximize Republican seats and minimize Democratic voting power.470Newkirk, supra note 467. These efforts were highly effective. In the 2012 congressional elections, Democratic House candidates received more votes in Michigan, North Carolina, Pennsylvania, and Wisconsin, yet Republican candidates won more seats.471Roth, supra note 453, at 101; see also Richard L. Engstrom, Partisan Gerrymandering: Weeds in the Political Thicket, 101 Soc. Sci. Q. 23, 27, 28 tbl.1 (2020). For example, in Pennsylvania, 49% of the vote handed Republicans 72% of the House delegation, and in Ohio, 55% of the vote gave Republicans 75% of the congressional delegation.472Roth, supra note 453, at 101. As a result of the gerrymandering undertaken by state legislatures, Republicans earned a thirty-three-seat advantage in the House of Representatives, flipping it to Republican control.473Id. Republican gerrymanders in Wisconsin and North Carolina enacted pursuant to the REDMAP policy, as well as a retaliatory Democratic gerrymander in Maryland, were so egregious that lower federal courts, for the first time in three decades, invalidated them as unconstitutional partisan gerrymanders.474Common Cause v. Rucho, 279 F. Supp. 3d 587, 597–98 (M.D.N.C. 2018), vacated, 138 S. Ct. 2679 (2018) (remanding for further consideration in light of Gill v. Whitford, 138 S. Ct. 1916 (2018)); Whitford v. Gill, 218 F. Supp. 3d 837, 910-11 (W.D. Wis. 2016), vacated, 138 S. Ct. 1916 (2018) (holding that partisan effect of a Wisconsin partisan gerrymandering act was not justified by legitimate state concerns); Benisek v. Lamone, 266 F. Supp. 3d 799, 816 (D. Md. 2017), aff’d, 138 S. Ct. 1942 (2018) (staying proceedings while the Supreme Court decided Gill v. Whitford, but labeling “political gerrymandering [as] a noxious and destructive practice” that is “repugnant to representative democracy”). In Gill, the Supreme Court had ruled that Democratic voters in Wisconsin had not demonstrated standing to challenge the state’s legislative districts. Gill, 138 S. Ct. at 1922–23. In a companion case from Maryland, the Court affirmed a district court’s denial of a preliminary injunction to Republican voters challenging a state congressional district. Benisek v. Lamone, 138 S. Ct. 1942, 1943, 1945 (2018) (per curiam). The U.S. Supreme Court, however, overruling thirty years of precedent, reversed, holding that partisan gerrymandering presents a nonjusticiable political question.475Rucho v. Common Cause, 139 S. Ct. 2484, 2497, 2506–08 (2019).

The aggressive Republican approach did not stop there. During the previous redistricting cycle, in a dramatic breach of longstanding norms of democratic contestation, Republican-controlled state legislatures engaged in additional rounds of redistricting between censuses.476Gardner, Representation Without Party, supra note 466, at 885–86. In Texas, Democrats controlled the lower legislative chamber through 2002.477Texas House of Representatives, Ballotpedia, https://ballotpedia.org/Texas_House_of_Representatives (click “Historical party control” on right sidebar or scroll down to view). When Republicans unified control over both chambers in 2003, the legislature enacted a new redistricting plan to increase Republican advantage.478League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 409, 412 (2006) (plurality opinion). The U.S. Supreme Court rejected a challenge to the constitutionality of mid-decade redistricting.479Id. at 423. In Colorado, the state legislature, which had been split in 2001, achieved Republican unity in 2003, and enacted a new redistricting law that year.480Carl Hulse, Colorado Court Rejects Redistricting Plan, N.Y. Times (Dec. 2, 2003), https://www.nytimes.com/2003/12/02/us/colorado-court-rejects-redistricting-plan.html. However, the Colorado Supreme Court ruled that mid-decade redistricting was unconstitutional under the state constitution.481Colorado ex rel. Salazar v. Davidson, 79 P.3d 1221, 1226 (Colo. 2003).

Partisan gerrymandering is extremely unpopular with voters,482According to a recent poll, 71% of voters “want the Supreme Court to place limits on lawmakers’ ability to manipulate voting maps.” See Kylee Groft, The Results Are In: Most Americans Want Limits on Gerrymandering, Campaign Legal Ctr. (Sept. 11, 2017), https://campaignlegal.org/update/results-are-most-americans-want-limits-gerrymandering [https://perma.cc/HE9L-HMNN]. and one increasingly common reaction has been for voters in initiative states to bypass the legislature to establish independent redistricting commissions. Nineteen states now have taken the power to redistrict away from partisan legislatures and lodged it in such commissions.483Gardner & Charles, supra note 333, at 316. Republicans have bitterly resisted these initiatives, repeatedly bringing lawsuits—thus far, unsuccessfully—to challenge virtually every aspect of the makeup and activity of independent redistricting commissions.484See, e.g., Ariz. State Legislature v. Ariz. Indep. Redistricting Comm’n, 576 U.S. 787, 793 (2015) (rejecting, by a five-four vote, the contention that the U.S. Constitution requires redistricting to be performed by state legislatures); Amy Biolchini, Michigan Republicans Sue to Stop Redistricting Commission, MLive (July 30, 2019), https://www.mlive.com/news/2019/07/michigan-republicans-sue-to-stop-redistricting-commission.html [https://perma.cc/3PDU-7NHE].

Finally, in the most blatant attempt to gain effective control of the body that regulates elections, the Republican-controlled North Carolina legislature in 2016 enacted a law during a lame-duck session, immediately preceding the installation in office of a newly-elected Democratic governor, that would have consolidated into one body the state boards of elections and ethics, given the legislature half the appointments, and prohibited the governor from removing any member of the board except for cause.485Cooper v. Berger, 809 S.E.2d 98, 100–01 (N.C. 2018). The North Carolina Supreme Court invalidated the law under state constitutional separation of powers principles as an excessive invasion of the executive powers of the governor.486Id. at 116.

Conclusions: The Nature of Subnational Variation in the United States 

It is clear from this overview that any movement in the American states toward authoritarianism remains to this point far removed from the worst excesses observed abroad. North Carolina is not Hungary or Turkey: no new state constitution has been rammed through; the powers of the governor have not been rapidly increased; members of the opposition party have not been arrested; journalists have not been detained and prosecuted; the government has not imposed massive fines on independent news organizations; the University of North Carolina at Chapel Hill has not been forced out of the state.487Cf. Diamond, supra note 6, at 56–57; Levitsky & Ziblatt, supra note 6, at 84. In a recent article, Kurt Weyland has argued that the circumstances leading to successful right-wing populist takeovers have been quite narrow and specific. Kurt Weyland, Populism’s Threat to Democracy: Comparative Lessons for the United States, 18 Persps. on Pol. 389, 390, 402 (2020). Despite the erection of significant electoral obstacles and aggressive vote suppression, a candidate of the opposition party was able to win the governorship.488David A. Graham, What’s Really Going on in North Carolina’s Gubernatorial Race?, The Atl. (Nov. 23, 2016), https://www.theatlantic.com/politics/archive/2016/11/whats-really-going-on-in-north-carolinas-gubernatorial-race/508520.

At the same time, it is equally clear that a number of American states have taken many small steps, with a potentially significant cumulative impact, on a path toward authoritarianism that has been well marked out by foreign leaders such as Viktor Orbán in Hungary, Recep Erdoğan in Turkey, Hugo Chávez in Venezuela, and Jaroslaw Kaczynski in Poland. The worst behavior at the state level in some domains lags well behind what is occurring domestically at the national level, and in other domains far exceeds it. State leaders typically do not exhibit the lack of verbal self-restraint displayed by Mr. Trump, yet in many cases they have acted where the President merely ranted on Twitter.

In the preceding review of authoritarian-style actions, summarized in the table below, certain states show up far more frequently than others.

Table 1: Creeping Subnational Authoritarianism

Measure from the Authoritarian’s Playbook States Adopting or Attempting
Demonization of the opposition MI, NC, WI
Undermine judicial independence AL, AK, AZ, AR, CO, FL, GA, IL, IA, KA, LA, MI, MT, NV, NC, OH, PA, SC, TX, WI, WV
Subdue civil society institutions AK, AZ, FL, IA, KA, LA, NC, WI
Attacks on actors supporting opposition AL, IA, OK, VA, WI
Attacks on civil service AZ, IA, IL, KY, OH, OK, NV, TN, WI
Manipulation of electoral rules and processes AL, AZ, FL, GA, IN, KA, MS, NE, NC, OH, PA, SC, TN, TX, VA, WI, WV
Gain control over body that runs elections CO, FL, MD, MI, NC, PA, OH, TX, VA, WI

Based on this analysis, it seems plausible to conclude that North Carolina and Wisconsin have advanced the farthest down the road to subnational authoritarianism, with Florida, Texas, Kansas, Arizona, and Alabama not far behind. Disturbing initial signs of democratic backsliding may be seen as well in Georgia, Iowa, Michigan, Ohio, Pennsylvania, and Tennessee. Meanwhile, New York and California, whose governors openly, repeatedly, and sometimes loudly opposed Mr. Trump,489To give just one example, at a recent press conference on the pandemic, Governor Cuomo demanded Mr. Trump “lead by example” and “put a mask on.” Governor Cuomo to President Trump: ‘Lead by Example,’ Daily Scrum (June 29, 2020), https://www.thedailyscrum.ca/2020/06/29/governor-cuomo-to-president-trump [https://perma.cc/CLW5-7NUV]. may have deepened their states’ commitment to liberal democracy, though how well those polities actually implement that commitment is an entirely different question.

This brings us back to the question with which we began—the question of how to describe subnational variation in the United States. What conclusions does the evidence support? First and foremost, it is probably too early to say anything definitive. No American state resembles a paradigmatic enclave of illiberal authoritarianism such as the Argentine province of Santiago del Estero, in which a long-serving governor and his wife enjoyed a “cult of personality” and ruled through personally directed violence.490Gibson, supra note 246, at 2.

Nevertheless, the evidence does tend to pile up, to the point where it is plausible to consider an account quite different from the three prevailing models of state-to-state variation described at the outset. That alternative amounts to this: several American states are well along the road to becoming—or, in the cases of North Carolina, Florida, Texas, Alabama, and Georgia, reverting to—illiberal, authoritarian enclaves. Contrary to the traditional account, they differ from other states not in their embrace of values at one end or the other of the liberal scale, but in their rejection of the basic tenets of liberal democracy. Contrary to the normal variation account, they are not liberal. And contrary to the partisan account, their public policies do not reflect the commitments of an out-party at the national level seeking to regain national power; they reflect instead a rejection of the democratic commitment to alternation in power itself. In these respects, the Southern states in particular bear a close resemblance to countries like Hungary and Turkey, which seemed for a while to have made the transition to liberal democracy, but in which democratic backsliding has been so severe as to call into question the degree to which liberalism ever really took root. These American states, that is to say, may be in the process of severing their relatively recent—and it turns out, shallow—ties to the family of liberal democracies.

To be sure, authoritarianism in the American states, insofar as it has to this point manifested itself, looks different from the kind of authoritarianism that presently prevails elsewhere. It does not, for example, conform to the model of excessive concentration of power in the office of the chief executive, nor to the classic populist model of a highly charismatic strong, individual leader. In some states, such as Wisconsin and North Carolina, the principal driver of authoritarian measures most recently has been the legislature rather than the executive. Within the apparatus of state government, party seems to be a much more important factor associated with authoritarianism than office.

State-level authoritarianism in the United States may be different from national-level authoritarianism in other ways as well. The public might well be more tolerant of authoritarian behavior on the subnational level than on the national level because of the generally lower salience of state politics,491Gardner, Myth of State Autonomy, supra note 30, at 44, 47. and the concomitant possibility that the norms of state politics have historically demanded a lesser level of refinement, civility, and even rationality than have the norms of national democratic politics. States are of course also subject to many more legal and constitutional constraints than is the national government, and these constraints not only confine their discretion, but open opportunities for the national government to rein in state excesses in ways that have no counterpart at the national level.

On the other hand, whereas illiberal populist parties elsewhere tend to be third parties,492See Norris & Inglehart, supra note 5, ch. 7. the structure of the American two-party system means that when illiberalism and authoritarianism do find a home in a major political party, that party will immediately be in contention for public office, and in circumstances that may not require it to compromise for the purpose of negotiating its way into power, as would be the case in a parliamentary system. And although the range of available state action may be for the moment restrained by the operation of normal checks and balances, what seems unrestrained in at least some states is the sheer bellicosity of the anti-democratic measures officials have conceived and implemented.

In sum, these developments provide legitimate cause for concern. Committed, liberal Americans around the country have worked hard for half a century to make good on the nation’s founding promise by pushing and hauling the country ever closer to the achievement of a truly liberal, truly democratic state. These efforts are now being undermined by a turn to authoritarianism not only at the national level, but at the state level as well, where states seem to be clustering into two groups, one still committed to democratic liberalism and the other willing to entertain a radical and highly risky experiment in authoritarianism.

If this account is accurate, the really important question will be whether and to what extent the nation’s federal structure both permits illiberal states to experiment with authoritarian forms of self-governance, and even more importantly, the extent to which federalism is capable simultaneously of providing security and protection for state polities that want no part of it. Abraham Lincoln famously proclaimed: “A house divided against itself cannot stand.”493Abraham Lincoln, The Historic “House-Divided-Against-Itself” Speech, at Springfield, ILL, June 16, 1858, in The Speeches of Abraham Lincoln 52 (Lincoln Centenary Ass’n, 1908). He was referring at the time to a nation “half slave and half free.”494Id. We may then ask, similarly: is it possible for the United States to survive if federalism provides a path to a nation half democratic and half authoritarian? The answer is at this point unclear, and perhaps the best response is simply to hope that the question need never be answered.

Share this post