Dividing the Body Politic

I. Introduction

No human society understands itself as a single, undifferentiated whole. Federal states divide themselves into self-governing provinces or regions. Unitary states divide themselves into administrative subunits. Democracies divide themselves into election districts. Even the smallest traditional societies divide themselves for many purposes into clans or family groups.1

Assuming, for the moment, that the successful practice of democracy requires, or is at least greatly facilitated by, division of the body politic into election districts, the question I wish to address here is: how ought these lines to be drawn? How should a society divide itself for the purpose of practicing democratic politics?

This is a question that has long been asked, and vigorously debated, by lawyers, political scientists, legislatures, courts, districting commissions, and citizens themselves. The reasoning by which the question is addressed, however, invariably proceeds along a well-worn path characteristic of the practice of institutional design. It begins from the premise that how we divide ourselves is entirely a matter of choice. Choices, in the domain of institutional design, should be rational,2 and the rational way to narrow down a wide range of options is to choose some set of preferred policy goals, and then identify the design path most likely to achieve them. This way of thinking about how to divide ourselves has spawned a vast literature evaluating, critiquing, and proposing criteria and methods of legislative districting.

I want to proceed differently here, by posing a question that most of the literature simply assumes away: Do we, in fact, havechoices—or at least legitimate choices—about how to divide ourselves for the purpose of practicing democratic politics? If so, under what constraints? Answering these questions in turn requires addressing important antecedent ones: when we divide ourselves, what exactly are we dividing? Is that entity, which I shall call here the body politic, reasonably viewed as divisible, and if so, along what dimensions, for what legitimate purposes, and with what potential consequences?

Until very recently—2016, say, or perhaps 2020—any attempt to answer these questions could rely confidently on a premise that seemed incontestable at the time: the United States is a liberal democracy, meaning a society committed to the philosophical liberalism of John Locke and the Declaration of Independence. As I explain below, since roughly the mid-twentieth century, our assumptions about how to divide ourselves have been based mainly on aggressively liberal conceptions of who and what we are, how and why we divide ourselves, what it means to do so, and the consequences of self-division.

Those premises have lately been thrown deeply in doubt by an immense resurgence of populist illiberalism.3 How we understand division of the polity is highly influenced by whether we approach the subject from the point of view of liberalism or illiberalism, particularly from the point of view of the right-wing populism that has been sweeping the globe, and which is now embraced by a substantial portion of the American electorate.4 Liberals and populists disagree profoundly on the most basic features and understandings of politics. They disagree on what I will call “political ontology”—who and what we are as a political society. They disagree on “political metaphysics”—the basic features of the political world and what in consequence is possible and desirable within it. And they disagree on “political epistemology”—how we know what we know politically.

Reexamining our practices of political self-division through the lens of this liberal-illiberal divide helps illuminate some vexing and deeply felt disagreements now emerging in our political discourse. It also helps to explain—though not to resolve—longstanding and extremely challenging anomalies and tensions in the practice and jurisprudence of legislative districting.

The Article proceeds as follows. Part I sets out some basic terminology by defining liberalism and illiberalism. Part II provides a brief overview of practices of political self-division in the premodern world, a world illiberal by definition. Part III highlights some complications arising in Europe from the emergence of liberalism in the seventeenth and eighteenth centuries. Part IV turns to a historical overview of practices of political self-division in the United States, tracing the gradual transition from premodern, illiberal practices to a firm embrace of a robust form of liberalism in the mid-twentieth century. Part V examines the current situation, in which revived illiberal views associated with right-wing populism have aggressively challenged the post-War liberal consensus. Part VI explores some possible escape routes from the present confrontation, which might either defuse it or attempt a modus vivendi by making concessions to illiberal commitments that would not require abandoning liberal ones.

II. The Ideological Landscape

A.    Liberalism

There is no single universally accepted definition of “liberalism,” a term whose meaning has from time to time been deeply contested even by its own adherents.5 Nevertheless, among contemporary writers, a rough consensus has emerged concerning several of its core principles. On most accounts, a foundational commitment of liberalism is popular sovereignty—the idea that the people are entitled to rule themselves, and that governments “deriv[e] their just powers from the consent of the governed.”6 Often implicit in the idea of popular sovereignty is a commitment to majority rule.7

The principle of popular sovereignty, however, itself rests on a prior commitment to the fundamental political equality of citizens, the axiomatic point of departure for both Thomas Hobbes and John Locke.8 Similarly, most contemporary accounts of liberalism require a “lively civil society”9 in which citizens enjoy the freedom necessary to develop their own beliefs about the nature of the good life and how the polity ought best to pursue it.

Commentators also generally agree that an essential attribute of liberalism is the rule of law, a condition without which popular sovereignty is unlikely to produce meaningful self-rule.10 Many contemporary versions of liberalism also understand it to include certain basic human rights, such as the rights of free speech and free association, that are thought necessary to the effectuation of popular self-rule.11

Apart from these basics, consensus tends to diminish. Some versions of liberalism include features such as limited government, dispersion of official power, a wider array of dignitary rights such as due process and criminal procedure rights, or free markets.12 For present purposes, when I speak of liberalism, I refer to the characteristics that command the greatest degree of consensus: the equality of citizens, popular sovereignty, a free civil society, the rule of law, and some package of basic human rights necessary to operationalize these commitments.

B.    Illiberalism

Illiberalism comes in a considerably wider variety of regime types than liberalism, ranging from dictatorship, to totalitarianism, to theocracy, and many others.13 For purposes of clarity, I will describe a regime as illiberal if it is characterized by denial of any of the core commitments of liberalism set out above. However, because at this historical moment the dominant version of illiberalism around the globe and in the United States is populism—and, overwhelmingly, a right-wing, authoritarian-leaning version of populism14 —I will refer more specifically to some of the key features of this particular strain of illiberalism.

In right-wing populist accounts, individuals are not equal politically, and consequently there is no general right to popular self-rule or rule by a majority of citizens. 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.15 Similarly, civil society in populist accounts is not free but is controlled strictly by higher laws of religious or traditional origin that establish customary and presumptively unalterable ways of life.16

Populists nearly always agree that the basic problem with their own government is that it has been captured by a small “corrupt elite” who run the show entirely for their own benefit and at the expense of the true people.17 Their preferred solution is to vest leadership in a strong individual who rules in the name of, and for the benefit of, the people, rightly understood.18 The exercise of governmental power, moreover, is not a contingent, and thus revocable, delegation of authority but an entitlement held by those properly destined to wield it. Consequently, its exercise ought to be neither restrained, impeded, nor questioned.19 On this account, 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 true leader’s pursuit of the good of society, properly understood.20

III. Political Self-Division in the Premodern World

The vast majority of human communities in the premodern West were, by contemporary standards, small, isolated, and largely self-contained.21 Typically, each had its own customs, laws, and language or dialect, and was ruled by its own minor prince.22 These communities were universally understood to possess a basic kind of integrity and distinctiveness: “A kingdom was never thought of merely as the territory which happened to be ruled by a king. It comprised and corresponded to a ‘people’ (gens, natio, populus), which was assumed to be a natural, inherited community of tradition, custom, law, and descent.”23

Things began to change with the arrival of the early modern “state.” Ambitious rulers, first in France, and then in other parts of Europe, began to expand their realms by acquiring and incorporating previously independent jurisdictions.24 These larger states often were cobbled together by any available means—alliance, marriage, inheritance, or when necessary, conquest. The result was what historians of empire call a “composite monarchy”—one composed of many previously independent constituent subunits.25

A significant premise of this kind of state-building was that the incorporation of new territories into the central state did not in any way imply homogenization of the populace. To the contrary, the prevailing practice in both ancient and early modern empires was to respect, and to leave intact, the laws, customs, institutions, and systems of authority prevailing in newly acquired territories. In this system, the head of the acquiring state simply stepped into the shoes of the previously ruling dynast and governed the acquired territory according to prior practice.26

For example, when Philip II reigned as head of the Spanish Empire in the mid-sixteenth century, he was King of Spain (Castile), but simultaneously King of Portugal, King of Naples and Sicily, Duke of Milan, and Lord of the Provinces of the Netherlands, among others. These titles were not honorific—they were literal. A Castilian king presiding over his empire ruled subject kingdoms as their monarch, meaning he did so under their laws and customs.27 Often this caused great inconvenience for monarchs when they wanted to mobilize the full resources of their empires because they had to comply with different sets of laws and customary restrictions on royal authority that acquired populations had in earlier times extracted from prior rulers.28 In this system, territorial distinctions were deeply naturalized: constituent communities were simply understood to be organically distinct, and there was a considerable degree of truth to that conception, as the various subunits often shared little except a common apex ruler.29

The belief that long-established geographical boundaries marked the limits of organically distinct societies fit neatly into a broader premodern conceptual framework for understanding political relations, the model of the body politic. In this model, each community was composed internally of distinct parts. In a highly literal analogy to the human body, which is sustained only by the cooperation of numerous organs with very different functions, the body politic was understood to be composed of distinct segments, each performing a different but necessary function. Thus, “the actual stratification of medieval society . . . led to the concept of society as an elaborately articulated structure in which each member had his own appropriate rights and duties . . . [which were] not necessarily equal to the rights and duties of others.”30

This belief system drastically, and apparently quite effectively, limited the scope of what could be considered to fall within the realm of politics. Because political society comprised a profoundly interconnected whole, deviations from the performance of a segment’s assigned role threatened the health of the entire organism, making resistance a profound political offense.31 Moreover, the partnership of church and state in that era meant that any attempt to deviate from traditional, inherited social roles was not only a secular offense, but also a violation of the religiously ordained order of things32 —a failure by the individual “to do his duty in that state of life . . . to which it shall please God to call him”33 —and thus in principle subject to surveillance and sanction down to the parish level. As Paul makes clear in 1 Corinthians, “The eye cannot say to the hand, ‘I have no need of you,’ nor again the head to the feet, ‘I have no need of you.’ . . . [T]here may be no dissension within the body . . . .”34 Thus, to the extent politics existed at all, in the contemporary sense of an arena in which to debate and decide how to live, it was exclusively for and among rulers.35 Even among monarchs the possibility of altering received laws and customs was limited.36

There was, however, one pragmatic exception to this general rule: premodern practices of representation. The institution of some kind of council whose consent was required for certain kinds of monarchical action was surprisingly common in the premodern world.37 These councils were not democratic, in the sense we use the term today, but they were representative in the sense that they were understood to represent and defend the interests of some segment of the community.38 The principal purpose of such councils was to provide communal consent to furnish support or cooperation to the monarch, typically in the form of funds and military manpower, often in exchange for the right to present grievances to the monarch and to demand the imposition of royal justice at the local level.39

Although historians disagree about the precise timing and sequence of development of the English Parliament, it seems reasonably settled that by roughly 1400, it had become a representative council of this type. Because fulfillment of Parliament’s purpose required raising and transferring revenue, and wealth in medieval England was linked inextricably to land, an arrangement was eventually devised in which the requisite kind of consent was aggregated through representation based on territory. Thus, although members of the earliest parliaments may have consisted mainly of barons or other great men of the realm without regard to territory, by the late fourteenth century, Members of Parliament generally consisted of two knights from each county,40 the county being the jurisdiction of a local lord who was a vassal of the king. When sources of wealth unconnected to land began to appear, representation was extended to cities and boroughs in order to gain access to their resources.41

My point in relating this history is twofold. First, the public practice of politics in the premodern world was both episodic and highly limited in scope. Representative politics was confined largely to the exchange of public political support and cash for voice. This was, it bears emphasizing, a wholly illiberal (or if you prefer, preliberal) system: there was no equality of citizens, no human rights, and no free domain of civil society. Indeed, civil society, under the body politic model, was distinctly unfree—quite differently from our own times, it “combined acceptance of inequality and subordination with a high degree of voluntary co-operation.”42

Second, territory was an integral aspect of the structure of states, and its significance was deeply naturalized. Territorial communities were much more than places on a map; the communities inhabiting such territories were understood as distinct and organic, with their own traditions, customs, and interests.

IV. The Complicating Impact of Liberalism

The arrival of liberalism in the late seventeenth and early eighteenth centuries was, well, liberating. The most immediate goal of early liberal theorists was to justify revolt against monarchs, and in the English case, to justify the transfer of state power to a legislature. The foundational assumptions of liberal theory, however, have much more far-reaching consequences. Most importantly, following the advent of liberalism, longstanding, inherited practices of governance could no longer be justified as dictated by God or tradition. In liberal thought, a body politic is no longer understood as an organic yet internally differentiated whole in which each part is assigned its role by God and nature. On the contrary, “[a] body politic, on this view, is a group of persons related by a system of [voluntary] agreements; to be a member of a body politic is to be a party to the system of agreements.”43

Liberalism, moreover, demands liberal justifications. If a society is a system of voluntary agreements, this clearly means that a society needs a reason to do things in one way or another, and preferably a good reason. As Locke wrote, “no rational creature” choosing to leave the state of nature for organized society “can be supposed to change his condition with an intention to be worse.”44 Rejustification of inherited illiberal practices in persuasively liberal terms, however, was not always easy, if such justifications even existed.45 It was in many cases less disruptive simply to carry forward familiar practices and worry about their justifications later, if at all.46

Another important effect of liberalism was the lifting of the constraints that had previously confined the domain of politics. In a liberal society, politics can be about anything; the nature of the good life itself can be up for grabs.47 This is particularly true where liberalism is linked to democracy, which was the case for most societies that moved toward liberalism.48 By broadening the base of political participation to include, at least eventually, substantially all members of society, the array of interests and opinions eligible for inclusion in political discourse became limited only by the concerns and imagination of the citizenry.

Moreover, the arrival of liberal democracy thoroughly transformed the nature of political representation. In the post-liberal world, a representative council was no longer merely a mechanism for the instrumental trading of narrowly defined favors between a monarch and his subjects. Instead, the legislature was almost immediately reimagined as a stand-in for the collective will of the sovereign people themselves. As John Adams later maintained, the legislature “should be, in Miniature, an exact Portrait of the People at large. It should think, feel, reason, and act like them.”49 This conceptual repurposing created endless complications that simply did not exist in the premodern analogue. How does a legislature represent? Whom does it represent? What obligations do representatives owe to their constituents, to other parts of society, or to the common good? What kind and degree of accountability is necessary to maintain a correspondence between the popular will and the sentiment of the representative body, and how close must that correspondence be? These are questions that to this day remain subjects of considerable disagreement.

Finally, liberalism did not triumph in a day. Wherever it appeared, a bitter struggle ensued between proponents of liberalizing reforms and an ancien régime that stood to lose its status and privileges.50 As a result, in Europe liberal ideals had to be fought for, bitterly, at great length, and often through deeply disruptive revolutionary means, of which the French Revolution, with its attendant brutal violence and internal social upheaval, was the paradigm.

V. Practices of Political Self-Division in the United States

Events in the United States followed a similar pattern. An aspirationally liberal state born in a premodern era, it inherited practices and customs whose origins and justifications lay in illiberal beliefs about politics and society. As liberalism gained increasing influence on American political thought, inherited practices came under stress and required either rejustification or alteration.

A.    The Colonial Period and the Early Republic

As in England, legislative representation in colonial America was based universally on territory. Individual communities were conceived as homogeneous, distinct, organic, and internally unified: “[t]he local community [of the colonial period] had a closed quality . . . . [It] provided within itself a focus for the economic, political, social, and religious lives of the townspeople . . . .”51 As a result, “[t]he town was essentially homogeneous: It had one religious belief, a unified political vision, even a fairly even distribution of wealth and a narrow range of occupations. It was a remarkably undifferentiated society, and it was difficult to draw the line between family and community, private and public.”52 Sometimes, of course, local communities were not unified. But internal dissent, when it appeared, was often vigorously and effectively suppressed for the sake of maintaining the public appearance of organic unity:

So long as the maintenance of an effective community required widespread consent of the governed, the unit of government had to be one in which such united public opinion could be obtained . . . . When harmony and homogeneity were broken, the territorial integrity of the town itself often had to be ruptured accordingly.53

Consistent with these beliefs, representatives to American legislative bodies were elected during the colonial period, and on through the first half-century of the republic, from geographical subunits: counties in the mid-Atlantic region, towns in New England, and counties, parishes, plantations, or hundreds in the South.54 Even when population was taken into account in the allocation of legislative seats, it most commonly manifested itself in alteration of the number of representatives assigned to a community, rather than in alteration of community boundaries, which were conceived as fixed, meaningful, and organic.55

Unfamiliarity with this earlier belief system routinely leads modern observers to misunderstand the significance of the well-known eponymous Gerrymander of 1812. What shocked contemporaries was not so much the manipulation of electoral rules for partisan gain, though that was surely offensive. What was shocking was that the Massachusetts legislature took Essex County—a county that had existed since the earliest days of the Massachusetts Bay Colony—and ruptured it into two senatorial election districts.56 Manipulating electoral rules for partisan gain was nothing new, even in the early nineteenth century;57 what was new and shamefully brazen was pursuing partisan gain by shattering the unity of what was generally assumed to be an organic local community.

As time passed, however, and liberal ideals began to circulate more widely and exert more influence on American political thought,58 Americans began to adapt and rejustify inherited practices. One such adaptation was the increased use after about 1850 of the more flexible “election district”—a territorial unit of representation not corresponding to the boundaries of an existing county or town—so as to reduce the most extreme population disparities.59 The most common approach, however, especially for state senates, was to balance two competing concerns—a liberal concern with equality of voting power and a preliberal concern with the integrity of local communities—by limiting population adjustments to the combining of small, adjacent counties, fully intact, into single election districts.60

B.    The Growing Influence of Liberalism

As the nineteenth century gave way to the twentieth, inherited beliefs in the organic distinctiveness of each individual town, parish, and county came under ever greater stress. An increasingly mobile population, revolutions in transportation and communication, industrialization, the emergence of a nationwide commercial marketplace, and sentiments of nationalism encouraged by the U.S. entry into World War I all tended to undermine traditional, place-based ways of life.61 The erosion of traditional ways of life in turn tended both to undermine the idea of localities as distinct communities, and correspondingly to provide support for the core liberal proposition that how one lives one’s life—and where one lives it—is a matter of rational, largely unconstrained, individual choice.62

If inherited conceptions of the organic distinctiveness of localities were becoming increasingly implausible, then the inherited justification for allocating legislative representation to local communities was simultaneously becoming obsolete. State courts adjudicating disputes arising from the districting process responded by providing an updated justification for the traditional mode of dividing the body politic, a justification that sounded in somewhat more liberal terms. Because the practice of allocating representatives to localities, and the law guiding it, remained constant, courts were not free to deny the significance of local community. Instead, they began to describe that significance in thinner and less comprehensive terms. One common move was to offer an updated version of the body politic model in which the distinctive contribution of each locality to the health of the larger organism was reconceived as economic rather than social or political:

Anciently, and still today, the counties reflect different economic interests . . . . So, certain counties have a dominant concern with manufacturing and commerce; others have a large stake in agriculture; still others lean heavily upon the resort industry; and finally a few counties have a special interest in the products of the sea.63

On this view, what is represented in a state legislature is not distinct self-contained communities, but distinct interests, and indeed interests of a narrow and highly specific type: regional economic interests. One of the consequences of this model is that politics in the state legislature is itself reconceived as consisting mainly of a clash of local economic interests, each of which advances the health of the larger organism in different ways.64 The principal job of the legislature, then, would be to allocate central resources in a manner that will best ensure the health of the larger organism—in this case the state economy—a function that the legislature presumably would discharge by deciding where and in which local economic activities to invest.

C.    The Jurisprudential Revolution of 1964

The model of the “economic body politic” enjoyed a period of dominance that lasted well into the twentieth century, until it was displaced at a stroke by the Supreme Court’s dramatic 1964 intervention in Reynolds v. Sims.65 Reynolds posed a direct challenge to the inherited practice of political self-division, which, as we have seen, had since colonial times been based, with only minor variations, on the idea of one community, one vote.66 Because decades of internal migration from farms to cities had in many cases produced substantial population variation among counties and municipalities within states, residents of comparatively crowded jurisdictions challenged this malapportionment as a violation of equal protection.

In extremely strong and assured language, the Court categorically rejected the inherited model, thereby invalidating the composition of every state legislature in the nation. “[T]he fundamental principle of representative government in this country,” the Court announced, “is one of equal representation for equal numbers of people, without regard to . . . place of residence within a State.” “[A]ll voters,” it said, “stand in the same relation regardless of where they live.” “Legislators,” the Court continued, “represent people, not trees or acres. Legislators are elected by voters, not farms or cities or economic interests.”67 The Court went on, moreover, to deny the relevance to state legislatures of the model furnished by the U.S. Senate, in which each state is separately and equally represented. Consequently, it held, members of both chambers of a state legislature must be elected from equipopulous districts rather than from historically recognized and legally incorporated local communities.68 In announcing a new one person, one vote standard to replace the prior community-based model, the Court presided over the complete destruction of what had been the dominant theory of legislative representation in America since the arrival of the first British settlers.

In so doing, the Court discarded conceptions of the body politic inherited from premodern times in favor of a view that relies on principles much more closely associated with liberalism, and that indeed can plausibly be characterized as aggressively liberal. At the root of the one person, one vote principle, as the Court elaborated it in Reynolds, is the view that all citizens, for political purposes, are essentially individual atoms without pre-political attachment to any other citizens or groups of citizens. At the very least, the Court’s position implies that if any such attachments exist, they are categorically without salience to the practice of democratic politics. That is to say, in the Court’s view, no politically salient distinctions are rooted in place; place simply is not, for the Court, a politically salient source of identity, or even a politically salient source of experience which might in turn shape beliefs or identity. It follows that individual citizens or groups of citizens can be moved in and out of election districts to equalize their populations without any consequence for the way politics will be practiced, either on the ground or in the legislature.

A corollary of this view is that if differences in political belief are not influenced by inhabiting a particular place in common with a particular group of people, where that group has a particular history and a body of shared experiences and inherited customs, then differences in political belief can result only from different uses of reason by the post-political citizen. This is obviously a premise that resonates strongly with basic principles of philosophical liberalism: the way any group of people lives is purely a matter of choice—choice that ideally is rational, substantively sound, and the result of aggregating individual opinions arrived at independently through reason.

One consequence of the viewpoint undergirding its decision in Reynolds is the Court’s utterly assured commitment to the optimistic belief that citizens’ views are not fixed, or even significantly influenced, by any pre-political identity. From this commitment of political metaphysics, others readily follow. One such commitment, which the Court later cashed out fully in its First Amendment jurisprudences of campaign speech and spending, is that voters are by hypothesis permanently open to persuasion based on appeals to their reason.69 Another is that democratic politics is, in consequence, a process of eternally fluid coalition formation and reformation based on choices that are by hypothesis temporary and contingent.70 These presumptions, founded firmly on the belief that voters are fully rational in their consumption and processing of political information, have unfortunately been disproved over and over by social science research, which consistently shows American voters to be ignorant, irrational, biased, and increasingly tribal in their allegiances.71

D.   Unresolved Tensions Following Reynolds

Reynolds and the one person, one vote cases that followed resulted in an uneasy reimagination of inherited practices of territorial representation. The logic of Reynolds pushes strongly against territoriality altogether, or at best renders it a purely administrative practice rather than a normative one. That is, we need to split up the body politic, not because it is composed of distinct and incommensurable communities each of which is entitled to voice through representation, but only because it is just too large to administer as a single unit. And because the essence of sound administration is simply efficiency, the only justification required to support any particular division of the body politic is that it is efficient; no other constraint applies.

Nevertheless, Americans have, since Reynolds, clung tightly to territoriality as the basis for legislative representation. The Court’s ruling produced no great rush in the states to rethink or restructure inherited practices of territorial representation. Indeed, the immediate reaction was resistance—thirty-two state legislatures called for a constitutional convention to overturn the decision, and constitutional amendments to that effect were introduced in both chambers of Congress, in some cases receiving a majority of votes, though short of the required two-thirds.72 When states did comply, they did the minimum; there was no sudden turn, for example, to proportional representation, a way of structuring legislative politics far more consistent with the liberal premises of Reynolds or to any other system in which territory plays no role or only an incidental one.73

The Court’s post hoc rejustification—and significant modification—of practices of self-division that originated in preliberal times has thus produced a jurisprudence of redistricting based on a naked compromise between fundamentally incommensurable views, rendering it both inherently unstable and theoretically unsatisfying. The compromise works like this. Under the one person, one vote doctrine, jurisdictions must divide themselves into election districts that are of precisely equal population, with two exceptions: strict adherence to equal population can be relaxed for the purpose of keeping intact (1) local government units and (2) “communities of interest.”74

1.     The local government exception

Under the Court’s compromise, the premodern disposition against dividing local communities was retained, but for very different and considerably less pressing reasons. In the past, the justification for keeping cities, towns, and counties intact was premodern and thus illiberal: such communities were understood to be organically distinct from one another, and consequently to have distinct voices entitled to be heard distinctly. This justification was normatively thick—it was based on a conception of identity as not only deeply rooted in local communal membership, but as highly politically salient.

After Reynolds, a preference for keeping local communities intact required rejustification in liberal terms. The justification that emerged is based on a reimagination of local governments as “administrative communities”—that is, as communities constituted by little more than the practice of decentralized governance itself. As the North Carolina Supreme Court has explained, counties “serve as the State’s agents in administering statewide programs, while also functioning as local governments that devise rules and provide essential services to their citizens.”75 It follows, the court said, that

people identify themselves as residents of their counties and customarily interact most frequently with their government at the county level. Based on the clear identity and common interests that counties provide, the impetus for the preservation of county lines [under the state constitution] is easily understood within the redistricting context.76

On this view, the usual arrow of causality is reversed. Community is not something that arises spontaneously and organically from shared experiences rooted in lives spent in close physical proximity, the existence of which might later be ratified by official recognition of the community as a corporate entity. Instead, community arises from the common experience of being administered by a certain set of state bureaucrats; in the words of the Vermont Supreme Court, a “sense of community [is] derived from established governmental units . . . .”77 Community thus does not precede but instead follows from the central practice of creating local governments for administrative convenience.78

2.     The “communities of interest” exception

The strict population equality requirement of one person, one vote may also be relaxed for the purpose of avoiding division of a so-called “community of interest” (COI). Communities of interest do not track boundaries of local government units. They are not based on identity, clan, or shared history or traditions. As the name implies, a COI is simply any group of people who happen to share a politically salient interest.79 These interests can be just as thin, or even thinner, than the interests shared by people who inhabit a common local government unit, and they are highly contingent because relevant interests can wax and wane in political salience with the shifting winds of politics.

In general, there is no particular reason why a group of people who share a politically salient interest should live in close geographical proximity; environmentalists, isolationists, pacifists, balanced-budget enthusiasts, or free marketeers can live anywhere. To the extent that COIs are territorial, they tend to be territorial only adventitiously. One situation in which COIs can coincide with territory is when people who share an interest happen to cluster together, the way that people who like the beach prefer to live near one. Another is when people who live in a place share an interest only in the most literally local of issues—local roads, local zoning, local parks, and so on. These kinds of interests form a “community” only in the most tautological sense; as Chandran Kukathas has observed, “a group of people standing at a bus stop may share an interest in the bus arriving on time, and in forming an orderly queue to board quickly—but this does not make it a community.”80

Predictably, then, the COI exception to the one person, one vote principle has resulted in legislative and judicial recognition of COIs on the thinnest imaginable bases. In actual practice, those responsible for districting tend to identify COIs mainly on the basis of shared participation in highly local commercial markets—shopping at the same mall, working for the same regionally dominant employer, utilizing the same local hospital or medical providers, or attending the same venues for sports, entertainment, or recreation. COIs have also been identified on the basis of common access to roads, airports, and other transportation networks.81

To say that people who shop at the same mall or use the same interstate highway comprise a “community” is to define the term with such extraordinary breadth as to deprive it of any real meaning, and courts have understandably struggled to apply the COI exception in ways that make sense on the ground. The Maryland Court of Appeals has gone so far as to condemn the idea of COIs as “nebulous and unworkable” on the ground that “the number of such communities is virtually unlimited and no reasonable standard could possibly be devised to afford them recognition in the formulation of districts . . . .”82

3.     Jurisprudential anomalies

The post-Reynolds doctrine of territorial self-division also leaves significant anomalies in the jurisprudence. In fact, it has resulted in two parallel regimes, one robustly liberal and adhering closely to the one person, one vote doctrine, and the other ignoring that doctrine and clinging instead to inherited premodern conceptions of the meaning of local boundaries.

Consider the City of Chicago. It is overwhelmingly Democratic; a Republican has not won a mayoral election in nearly a century. Chicago is consequently one of the most partisanly lopsided and least competitive jurisdictions in the nation. Yet nobody says that the boundaries of the city should be moved to take in western suburbs for the purpose of making elections for Chicago city offices “more competitive.” Yet that is precisely what people say all the time about congressional and legislative districts, including those lying within or athwart city lines,83 and the boundaries of such districts are routinely shifted on exactly these grounds. Why the different treatment? Are the boundaries of the City of Chicago somehow “real,” or “meaningful,” or “organic” in a way that the boundaries of election districts are not, or should not or could never be? What makes the boundaries of a city or town conceptually prior to the partisanship of its residents, yet treats partisanship as prior to the boundaries of a legislative district?

Or consider this. Erie County, New York, where I live, with a population of nearly one million, is led by an elected County Executive. Ulster County, in the Hudson Valley, with a population of just over 180,000, also has an elected County Executive.84 This means that the vote of an Ulster County resident is about five times more influential in determining the identity of that county’s chief executive than my vote in Erie County. Much smaller disparities between county populations—indeed, trivial ones—were sufficient to cause invalidation in Reynolds under the one person, one vote doctrine, yet that doctrine is simply deemed inapplicable to disparities in voting power across counties, cities, or towns. Why are those disparities not only tolerable, but beneath constitutional notice, while population disparities of a handful of voters brings down the constitutional hammer in the case of legislative districts? This seems especially odd given that the law of most states makes it perfectly clear that counties are for legal purposes little more than administrative arms of the state itself, the means by which the state implements its programs on the ground.85 Why is strict equality of influence demanded in constituting the state’s legislature, but not in constituting the democratically accountable administrative apparatus by which the legislature’s programs are actually carried out? The only answer seems to be that a county is somehow “organic” or “real” in way that election districts are not.

I don’t mean by this discussion to take a position on the integrity or meaningfulness of local community boundaries. My point here is simply that the enterprise of justifying in liberal terms practices inherited from an illiberal era can present great difficulties. Indeed, it is quite possible that institutions and practices that had perfectly sound, if illiberal, justifications in an earlier era will simply have no good justification on liberal grounds, meaning in turn that any suitably liberal solution might well require complete abandonment of prior practices.

VI. The Current Situation

A.    Complicating Trends

I have argued thus far that the American jurisprudence and practice of political self-division suffer from confusion and incoherence caused by a gradual divergence between the inherited practice and the available set of ideologically acceptable justifications. Or to put it another way, Americans have strongly resisted discarding or even substantially altering inherited practices so as to bring them into conformity with contemporary ideological commitments, which until recently have been overwhelmingly liberal.

In most circumstances, a society’s inability to justify in acceptable terms a practice it wishes to utilize might amount simply to a knotty problem that, with enough study and creativity, might eventually be resolved. This has been the premise of virtually all scholars who have proposed reforms to contemporary practices of legislative districting. Things have recently become considerably more complicated, however, and the underlying tensions far less tractable, due to the confluence of four recent trends in American politics.

First is the dramatic and wholly unexpected resurgence in the United States of a wide variety of long quiescent illiberalisms. Since Donald Trump’s 2016 campaign for the presidency, incidents of religious, ethnic, and racial bigotry and violence have increased dramatically.86 More disturbing, however, is that many on the far right now seem much more willing to affirm publicly, and without qualification, beliefs in long-discredited ascriptive hierarchies, such as Christian or white supremacy, that are deeply illiberal in their denial of the political equality of all citizens.87 The dominant manifestation of this wave of illiberalism has been the startlingly sudden rise of right-wing populism, with its characteristic suspicion of democratic processes and fondness for strong-man authoritarianism.

Second, a decades-long trend of partisan sorting along ideological lines has recently consolidated in a way that aligns cleavages of partisanship with those dividing liberals from illiberals. For many years, those who share the commitments of the Democratic and Republican Parties have been sorting themselves more accurately into partisan identities and voting patterns, leading the parties to diverge ideologically in fundamentally coherent ways, and in general offering voters a clearer choice than at any time in the last three-quarters of a century.88 What has made this polarization much more challenging is that the Trump wing of the Republican Party has pushed it so far to the right that it has fallen off the liberal spectrum altogether.89 So long as the Trump wing continues to dominate, the Republican Party is best understood as a party of illiberalism. This situation is unprecedented. Party competition in the United States is no longer between two parties committed to liberalism but disagreeing about how best to implement it; the competition today pits a party committed to liberalism against an “anti-system” party that would be content to tear down the entire edifice.90

Third, partisanship in the United States has become increasingly associated with personal identity. As Achen and Bartels report,

[V]oters choose political parties, first and foremost, in order to align themselves with the appropriate coalition of social groups. Most citizens     support a party not because they have carefully calculated that its policy         positions are closer to their own, but rather because “their kind” of person   belongs to that party.

Thus, “partisan identities are . . . frequently bound up in [voters’] social identities . . . .”91 It follows, on this account, that policy commitments are not antecedent to party identification, but instead follow from it—voters choose a party for identitarian reasons and then adopt its ideological and policy commitments. This phenomenon greatly increases the difficulty of deviating from the positions of one’s party because to do so is tantamount to undermining the integrity of one’s own identity.

Fourth, as Americans have increasingly sorted themselves by partisanship, and in consequence by their commitment to liberalism, they have also been sorting themselves geographically. As political scientist Jonathan Rodden writes, “[V]oting behavior is spatially dependent: the probability that two randomly drawn individuals . . . exhibit similar voting behavior is a function of the distance between their locations.”92 This means that voters inhabiting the same locality exhibit increasingly similar partisan preferences;93 in particular, Republicans have tended to self-concentrate in rural and ex-urban areas while Democrats have congregated in urban ones. Whether this has occurred because of deliberate relocation decisions or is due to social forces operating among permanent residents is unclear,94 but the reasons hardly matter; increasingly, co-partisans, and thus liberals and illiberals, cluster together on the map.

Alas, these trends function as mutual intensifiers. As a result, partisanship in U.S. politics aligns increasingly with ideology, ideology aligns increasingly with liberalism or illiberalism, and these factors in turn align increasingly with both personal identity and territory. These circumstances place enormous stress on the assumptions underlying Reynolds, in two ways. First, the emerging political landscape is beginning to bear an uncomfortable resemblance to that of premodernity, in which territorial boundaries were understood to demarcate meaningfully distinct—and thus non-fungible and presumptively indivisible—peoples. Second, even if the actual differences of identity and ideology that divide groups of Americans on the landscape are less profound than those that divided premodern peoples and kingdoms, there is no reason to expect consensus going forward on questions of how properly to divide the body politic, for liberals and illiberals tend to disagree starkly about both the significance of place and its political salience.

B.    Place and Politics

1.     Liberal accounts

Liberalism comes in many varieties, but a central tenet of all contemporary versions is that large, modern polities are inevitably diverse and pluralistic.95 Liberals, to be sure, disagree about the nature of that diversity and how deep it runs, from a folk belief in the basic similarity and brotherhood of all human beings, to the Rawlsian view that citizens of a liberal polity may be expected to embrace a variety of fundamentally incommensurable “comprehensive doctrines.”96 But whatever their conception of social diversity, all versions of liberalism share the belief that such diversity poses no obstacle to a satisfying, meaningful, and fully inclusive politics, conducted in a universally shared public square. This politics is characterized not only by persuasion and compromise, but by the possibility of reaching very broad consensus on important questions so long as citizens approach democratic politics with an appropriate degree of mutual tolerance and a realistic and appropriately limited set of expectations.97 In consequence, liberalism tends to downplay the significance of place, especially as compared to reason, and as we have seen, the version of liberalism adopted by the Supreme Court in Reynolds denies altogether that place holds any real significance for the formation of political beliefs or of politically salient personal identities.

An intriguing corollary of the liberal account is that deliberate institutional design has the capacity to influence the way political opinion is formed, and consequently, its content. Because liberal citizens, regardless of their private attachments and commitments, are by hypothesis open to influence, those whom they encounter in the public sphere, and the context in which such encounters occur, can make a difference in how individuals formulate their views, and consequently in the content of their ultimate political commitments.98 It follows that political community can to some degree be engineered by combining people for political purposes in this way or that, thereby influencing the course of their reasoning, and leading probabilistically to certain kinds of preferred political outcomes. As democratic theorist Andrew Rehfeld has cogently observed, “The localness of politics is . . . epiphenomenal, nothing more than a byproduct of territorial electoral constituencies. If electoral constituencies were defined by profession . . . , all politics would be ‘vocational.’”99

2.     Illiberal accounts

In contrast, most premodern thought, and many versions of contemporary illiberalism, hold place in much higher regard, viewing it as highly significant for, and even constitutive of, social and political relations, and consequently, of personal identity. Aristotle viewed place as “prior to all other things.”100 In medieval political thought, “[l]and and people were assumed to be one,”101 a view echoed much later in Hegelian forms of nationalism, which posit an organic congruence between a people and their territorially defined nation,102 a claim in turn often taken up by twentieth-century fascists.103 Theologians sometimes understand place to have spiritual significance, in which “a yearning for a place is a decision to enter history with an identifiable people in an identifiable pilgrimage.”104 Place, after all, provides “a principle of intelligibility for the person who observes it,”105 and that principle is necessarily, first and foremost, one that “marks a differentiation between an ‘inside’ and an ‘outside.’”106 A distinction between those who belong and those who do not is a first principle common to many kinds of illiberalisms; indeed, what is in some ways distinctive about liberalism is its tendency to universalize human experience and to abstract it away from any kind of rootedness in place, or in the community inhabiting that place.107

Illiberal accounts of place therefore tend to deny the possibility of a liberal politics, a politics in which any and all topics are open for discussion, and whose participants are endlessly open to persuasion. For illiberals, many such decisions have already been made, decisively and permanently, by God or tradition. Liberals, on this view,

have failed to appreciate that territory is both a symbolic expression and a concrete condition for the possibility of (or aspirations to) a distinct way of life, and that in the modern epoch it gathers together many of the associations through which individuals come to see themselves as members of a political society.108

On this view, “[s]pace affects how individuals and groups perceive their place in the order of things. Spatial configurations naturalize social relations by transforming contingent forms into a permanent landscape that appears as immutable rather than open to contestation.”109

This kind of account suggests strongly that place not only exercises a significant role in the formation of political opinion but tends to fix and naturalize those opinions in a way that discourages rather than invites open-minded appreciation of alternative views. This in turn greatly inhibits the possibility of persuasion, compromise, and consensus. In these circumstances, the kind of social engineering that liberalism invites, in which individuals may be grouped or regrouped for the purpose of facilitating useful, opinion-shaping political encounters, cannot possibly have the desired effect. Dividing the members of territorial communities will simply throw people into encounters with strangers, whose views they find alien, and with whom the possibility of reaching common ground is essentially nil.110 In these circumstances, a forced political encounter can scarcely reveal new, unexpected arenas for cooperative collective action; it is far more likely to produce a distinctly non-discursive politics of naked power, in which each group co-habiting an artificially created political space seeks whatever degree of domination available institutions permit.111

C.    The Populist Account

Populist versions of illiberalism present something of a paradox in their conceptions of place. On one hand, populism’s emphatic rejection of core tenets of liberalism might suggest that, like other illiberalisms, it conceives of place as highly significant. On the other hand, populism’s close association with nationalism casts doubt on whether populists invest significance in any territories more local than the nation.

One of the signal characteristics of right-wing populism is its anti-pluralism.112 For populists, the body politic, rightly understood, is not diverse and plural. On the contrary, populists view any people who comprise a “popular sovereign” to be homogeneous by definition; that is what makes them, in any kind of meaningful sense, “a people.”113 For this reason, populists tend to be deeply skeptical of the very aspects of democratic politics that liberals value: deliberation and debate. Political conflict, on populist assumptions, is not a sign of normal, healthy disagreement within the body politic; it is, to the contrary, a sign of the influence of the corrupt elite, who first foment and then exploit disagreement to advance their own welfare at the expense of the true people.114

This constellation of beliefs makes populists skeptical of legislatures tout court.115 When the answers to political questions are obvious, there is no need for a deliberative body. That is why, for populists, power is best vested in the executive, who serves as the sole symbol of a unified people, and who operates ideally without institutional checks that retard implementation of the popular will.116 The very purpose of democratic politics, on this model, is not to deliberate on and choose from among policy alternatives; it is, rather, to affirm through acclamation the authority of the true leader, the identity of the sovereign volk, and a societal commitment to the liberation of the raw power of the true and united people.117 This, no doubt, is why populists tend to express a clear preference for plebiscitary measures, such as referenda of nationwide scope which bypass what they view as the obstructionism inherent in mediating institutions such as legislatures.118 Populism thus seeks “a post-place politics in which there is unmediated connection between people, leader, and government.”119

On populist assumptions, then, there seems to be no real need to divide the body politic; it is unified by hypothesis. Dividing the sovereign volk, on these premises, may well look like some kind of trick to subvert the unified expression of the popular will. Indeed, if we examine the actual behavior of American populists confronted with a legal obligation to draw lines of division—that is to say, recent Republican-controlled state legislatures engaged in redistricting120 —we generally observe two strategies, both of pure instrumental domination. One strategy is to draw lines so as to isolate those who do not qualify for membership in the sovereign volk (nonwhites, non-Christians, immigrants, Democrats), ceding them territory, but only for the purpose of quarantining them in their own ghettos where the harm they inflict can be confined to their own kind. The other strategy, when the first is unavailable, is to divide non-members finely enough to render them small minorities in numerous jurisdictions, where they can be properly dominated by local majorities of the true people.121

I recognize, of course, that these two tactics—“packing” and “cracking”—are among the standard tools of gerrymandering practiced from time to time by both Democrats and Republicans. The difference is that liberals recognize these practices as illiberal and condemn them, whereas populists do not.122 The fact of the matter is that, whatever the tactics, in no version of populist thought is creation of an election district cohabited by people of diverse views a means of bridging divides; it is strictly an invitation to use whatever tools of domination happen to be available, to be deployed instrumentally wherever possible.123

In sum, then, in a polity in which liberals and populists compete democratically for power, such as the United States today, there is no reason to expect any agreement whatsoever about how to divide the body politic because these groups disagree profoundly on what the body politic is.

VII. Escape Routes?

Is there, then, some way to defuse this confrontation, some modus vivendi, perhaps, that might preserve liberal democracy in the face of the present right-wing populist challenge? This might mean nudging American politics in the direction of liberalism by rewarding a liberal style of politics that includes coalition-building, deliberation, and compromise; or it might mean, at the very least, impeding the ability of the illiberal right to dominate liberals through the single-minded deployment of raw power. I’ll mention briefly three possible institutional modifications.

A.    Proportional Representation

If dividing the body politic into seemingly arbitrary election districts is the problem, one obvious solution is simply to eliminate districting altogether, or at least to greatly minimize it, by adopting proportional representation (PR). In PR systems, which award parties legislative seats in proportion to their overall vote totals, legislators are elected either from a single, undivided, jurisdiction-wide district, or from a small number of very large, multi-member districts that, depending on the jurisdiction, generally elect somewhere between five and fifteen legislators each.124 PR minimizes the problems of dividing the body politic associated with American-style first-past-the-post, single-member-districted systems (FPTP) because the small number, large size, and unavoidable internal diversity of PR districts makes partisan manipulation of district lines much harder to achieve.125

The problem with PR for present purposes, however, is that it would involve the imposition of an overtly liberal solution as a cure for a conflict between liberal and illiberal outlooks. The design premises of PR lie firmly in utilitarianism, a quintessentially liberal theory of politics. As I have argued elsewhere, PR not only presupposes the internal pluralism of the body politic—exactly what populists deny—but postulates that the politically relevant kind of pluralism is based on the pursuit of individual self-interest. PR accommodates these inputs by institutionalizing a mechanism of negotiation and compromise in the process of legislative formation of a governing coalition and in the coalition-building necessary to enact legislation.126 For these reasons, I assume illiberals, including populists, would be likely to resist it, unless they found themselves performing so poorly in an FPTP system that PR would offer them a way to avoid the kind of complete submersion that is often the fate of political minorities under FPTP.127

B.    Cross-Cutting District Lines

A second kind of institutional modification, often advanced in the international setting as a possible solution to bitter ethnic conflict, is to draw jurisdictional lines in a way that deliberately divides and commingles conflicting groups.128 The idea is to create an institutional arrangement in which political control of a jurisdiction or district can be obtained only if otherwise hostile groups cooperate to form a governing coalition. In this way, the institutional structure “will provide rational political actors with incentives towards cooperation, moderation and accommodation between themselves and their rivals . . . .”129 In the U.S. setting, this would mean drawing district lines whenever possible so as to take in roughly equal numbers of Democrats (liberals) and Republicans (illiberals) in the hope that such districts could be carried only by candidates capable of appealing across the ideological divide.130 Adoption of an alternative voting system, such as ranked-choice voting, could also help, insofar as such systems encourage candidates to campaign for second-place votes, providing them with incentives to take more moderate and centrist positions.131

A pragmatic difficulty with this suggestion is that it might require the relaxation of district compactness requirements, which make it more difficult to draw the kind of meandering districts that might be necessary in some places to take in competitive numbers of partisans in as many districts as possible. A more substantial obstacle, however, may be that the current liberal-illiberal divide is qualitatively different from the former Republican-Democrat one, which spanned policy disagreements withinthe bounds of liberalism. For today’s anti-system illiberals and populists, politics is not about choosing among policies; as a recent study concluded, “rank-and-file Democrats reliably prioritize policy preferences over symbolic attachments, but rank-and-file Republicans tend to reconcile the conflict in favor of their symbolic attachments to their ideological identity.”132 What this seems to mean is that Democrats and Republicans participate in politics for entirely different reasons—for Democrats, in typical liberal fashion, the point is to adopt good policies, whereas for Republicans the point of politics is to affirm their identity. It is difficult to imagine a position from which a candidate, much less an ideologically stable series of candidates, could appeal to both groups, other than through the adventitious and deeply unlikely congruence of liberal policy preferences with illiberal identitarian requirements—when, for example, Democrats come to view the open carry of guns and the elimination of public education as desirable policy goals.

C.    Fixed Election Districts

A third option, and one that strikes me as more promising, would be to revert to a method of legislative districting widely used in the states during the early nineteenth century. In this system, the boundaries of election districts were permanently fixed, and corresponded to established units of local government. Population disparities above a certain threshold were handled by adjusting the number of representatives elected by each unit, and in turn allowing the size of the legislature to expand or contract to accommodate the adjusted number of representatives.133 This system had the benefit of taking seriously both the premodern idea of localities as distinct organic communities and the liberal idea that individual citizens are entitled to roughly equal voice in the formulation of legislative policy regardless of where they happen to reside. It thus suggests the possibility of a modus vivendi in which important liberal goals could be achieved in circumstances that would not require illiberals to abandon some of their own views of community or the proper nature and subjects of legislative politics.

Ideologically, this position bears a strong resemblance to communitarianism, an intermediate theory of politics that is skeptical of extreme forms of both liberalism and illiberalism.134 On the communitarian view, the body politic is indeed internally plural and diverse, as liberals maintain. Unlike liberals, however, communitarians conceive of political pluralism as attributable to much more than different uses of reason. For communitarians, political pluralism is, to a significant degree, rooted in identity, as illiberals often maintain, which in turn is rooted in meaningfully distinct experiences of community.135 These experience-shaping communities are often geographical and extremely local—a town, a neighborhood, a local parish or congregation, and so forth—though not all such communities need be geographical.136 What is crucial to the communitarian view is the belief that all human lives are “situated”—that they are lived embedded in communities, and that these communities play a formative role in constituting their members’ identities, belief systems, and political commitments.137

Unlike most forms of illiberalism, however, communitarianism does not view these formative experiences and predispositions, and the identities they generate, as immutable. Initially acquired identities may be sticky and difficult to dislodge, but they are nevertheless defeasible—by reason, by new experiences, or by re-embedment in different communities or new institutional settings characterized by different sets of commitments.138 However, like many illiberalisms, communitarianism seems sympathetic to the idea that local communities are meaningful and organic and that dividing them for exogenous purposes may do them considerable harm.

Implementing the kind of fixed-district solution I have just described would require some relaxation of the presently acceptable limits on deviations from one person, one vote because the disparities among local community populations are too great to achieve mathematical precision without enlarging the legislature to an unmanageable size. The Canadian approach, however, provides a feasible model. According to the Supreme Court of Canada, district population variances of up to twenty-five percent, and possibly more, are acceptable when motivated by a belief that equipopulous collections of individuals are not the only thing entitled to representation in a provincial legislature.139

VIII. Conclusion

The way in which the American polity divides itself for the purpose of democratic politics comprises a messy and often internally contradictory mélange of liberal and illiberal practices and justifications. The possibility of sorting things out in a rational way—long the object of legal and political science scholarship in the field—is greatly complicated today due to the recent resurgence of various forms of illiberalism, to the point where the central cleavage of U.S. politics is now one that divides liberals from illiberals. This political alignment is complicating because liberals and illiberals disagree profoundly about the nature of the body politic, its susceptibility to division, and the significance and goals of such division. There may be ways to nudge American politics in the direction of liberal democracy notwithstanding resistance to it in certain quarters, but the options for doing so are few and may be exceedingly difficult to implement.

• 1The classic anthropological work on the seemingly basic human urge to create social and political divisions along various and shifting cleavages is E.E. Evans-Pritchard, The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People (1940).
• 2See, e.g., Philip Pettit, Institutional Design and Rational Choice, in The Theory of Institutional Design 54 (Robert E. Goodin ed., 1996).
• 3Seegenerally Pippa Norris & Ronald Inglehart, Cultural Backlash: Trump, Brexit, and Authoritarian Populism (2019); Yascha Mounk, The People vs. Democracy: Why Our Freedom Is in Danger and How to Save It (2018).
• 4For example, a recent poll found that forty-two percent of Republicans agreed with the following statement: “Strong, unelected leaders are better than weak elected ones.” David Nather & Margaret Talev, Two Americas Index: Democracy Deniers, Axios (Sept. 12, 2022), https://www.axios.com/2022/09/12/two-americas-index-democracy [https://perma.cc/RRK8-54AJ].
• 5See Helena Rosenblatt, The Lost History of Liberalism: From Ancient Rome to the Twenty-First Century 1–3 (2018).
• 6The Declaration of Independence para. 2 (U.S. 1776).
• 7See, e.g., John Locke, Second Treatise of Government §§ 95–99, at 52–53 (C. B. Macpherson ed., 1980) (1690).
• 8Thomas Hobbes, Leviathan141–45 (John Plamenatz ed., 1963) (1651); Locke, supra note 7, §§ 4–8, at 8–10.
• 9Larry Diamond, Ill Winds: Saving Democracy from Russian Rage, Chinese Ambition, and American Complacency 19 (2019).
• 10See, e.g., Tom Ginsburg & Aziz Z. Huq, How to Save a Constitutional Democracy 9 (2018).
• 11Id.; Juan J. Linz, Totalitarian and Authoritarian Regimes 20, 58 (2000).
• 12See, e.g., Stephen Holmes, The Anatomy of Antiliberalism 3–4 (1993) (separation of powers); James Traub, What Was Liberalism? The Past, Present, and Promise of a Noble Idea 1 (2019) (free markets).
• 13See, e.g., Nenad Dimitrijevic, Illiberal Regime Types, in Routledge Handbook of Illiberalism 121, 121 (András Sajó et al. eds. 2022); Ran Hirschl, Theocracy, in Routledge Handbook of Illiberalism, supra, at 152.
• 14Norris & Inglehart, supra note 3, at 3–4; Mounk, supra note 3, at 2–3.
• 15Jan-Werner Müller, What is Populism? 17, 19–22 (2016); Diamond, supra note 9, at 62; Norris & Inglehart, supra note 3, at 66–75.
• 16One example is the path of Christian nationalism taken in democratically backsliding states like Hungary and Poland. Diamond, supra note 9, at 61–62; Traub, supra note 12, at 205–09.
• 17Müller, supranote 15, at 16; William A. Galston, Anti-Pluralism: The Populist Threat to Liberal Democracy 36 (2018).
• 18Müller, supra note 15, at 18–21; Mounk, supra note 3, at 41–43.
• 19Günter Frankenberg, Authoritarian Constitutionalism: Coming to Terms with Modernity’s Nightmares, in Authoritarian Constitutionalism: Comparative Analysis and Critique 1, 18–20 (Helena Alviar García & Günter Frankenberg eds., 2019).
• 20See Robert C. Lieberman et al., The Trump Presidency and American Democracy: A Historical and Comparative Analysis, 17 Persps. on Pols. 470, 470–71 (2019).
• 21Especially in western Europe, by the tenth century, “rulership passed often to individual castle lords governing a few square kilometres with the aid of some knights.” D.E. Luscombe, Introduction: The Formation of Political Thought in the West, in The Cambridge History of Medieval Political Thought, c. 350 – c. 1450, 157, 157 (J.H. Burns, ed., 1988).
• 22Susan Reynolds, Kingdoms and Communities in Western Europe, 900–1300, at 8–9 (2d ed., 1997); J.H. Elliott, A Europe of Composite Monarchies, 137 Past & Present 48, 50–54 (1992).
• 23Reynolds, supra note 22, at 250.
• 24Sheri Berman, Democracy and Dictatorship in Europe: From the Ancien Régime to the Present Day 20–21 (2019); H.G. Koenigsberger, Composite States, Representative Institutions and the American Revolution, 62 Historical Research 135, 135 (1989).
• 25Koenigsberger, supra note 24, at 135; Elliot, supra note 22, at 51–54.
• 26Elliott, supra note 22, at 52–53.
• 27J.H. Elliott, The Revolt of the Catalans: A Study in the Decline of Spain (1598–1640), 7–8, 45 (1963).
• 28Elliott, supra note 22, at 54, 63.
• 29Elliott, supra note 22, at 7–8; Reynolds, supranote 22, at 8, 156, 220–221.
• 30Ewart Lewis, Organic Tendencies in Medieval Political Thought, 32 Am. Pol. Sci. Rev. 849, 864–65 (1938).
• 31To give one example, before the late sixteenth century in England, military service was a function of the gentry. The bearing of arms and military participation by other segments of society was prohibited out of a concern that these segments would be empowered by carrying weapons and would cease to be subordinate. Christopher Hill, The World Turned Upside Down: Radical Ideas During the English Revolution 19 (1974).
• 32See Hirschl, supra note 13, at 152–53 (discussing, for example, the doctrine of the “divine right of kings”).
• 33Arthur O. Lovejoy, The Great Chain of Being 206 (22d prtg. 2001) (quoting the Anglican catechism).
• 341 Corinthians 12:21–26 (New Revised Standard Version).
• 35Lewis, supra note 30, at 864–67.
• 36See, e.g., John Fortescue, Commendation of the Laws of England 17–18 (Francis Grigor trans., Sweet & Maxwell 1917) (c. 1537) (describing the obligation of the king to rule subject to the received laws of the realm).
• 37David Stasavage, The Decline and Rise of Democracy: A Global History from Antiquity to Today 4–5 (2020).
• 38As Reynolds observes, during the Middle Ages “it was assumed that the leading men in any community represented the whole of it, whether or not they had been elected.” Reynolds, supra note 22, at 251. See also Stasavage, supra note 37, at 4–8.
• 39Deborah Boucoyannis, Kings as Judges: Power, Justice, and the Origins of Parliaments 18 (2021); G.L. Harriss, The Formation of Parliament, 1272–1377, in The English Parliament in the Middle Ages 40–42 (R.G. Davies & J.H. Denton eds., 1981).
• 40A.L. Brown, Parliament, c. 1377–1422, in The English Parliament in the Middle Ages, supra note 39, at 117–18.
• 41A.F. Pollard, The Evolution of Parliament 50–55 (1920).
• 42Reynolds, supra note 22, at 332.
• 43Joseph Tussman, Obligation and the Body Politic 7 (1960).
• 44Locke, supra note 7, § 131, at 68.
• 45Slavery, of course, is the quintessential example of an illiberal practice carried forward into an aspirationally liberal society without the possibility of liberal justification. Or to take another example, many Britons struggle to understand why the country still has a king.
• 46As Ian Shapiro observes, “Humans inevitably make political choices at the margins, redesigning inherited institutions as they reproduce them into the future. We rebuild our ships at sea.” Ian Shapiro, Politics against Domination 4–5 (2016).
• 47The culmination of this tradition is surely John Rawls, Political Liberalism (1993).
• 48Kim Lane Scheppele, Autocratic Legalism, 85 U. Chi. L. Rev. 545, 548 (2018) (“[D]emocracy, constitutionalism, and liberalism [have] marched arm in arm through history . . . .”).
• 49Letter from John Adams to William Hooper (Mar. 27, 1776), in 4 Papers of John Adams 75 (Robert J. Taylor et al. eds., 1979) (1776).
• 50Berman, supra note 24, passim.
• 51Thomas Bender, Community and Social Change in America 65 (1978) (footnote omitted).
• 52Id. at 68.
• 53Michael Zuckerman, Peaceable Kingdoms: New England Towns in the Eighteenth Century 138–39 (1970).
• 54Edmund S. Morgan, Inventing the People: The Rise of Popular Sovereignty in England and America 41 (1988).
• 55Rosemarie Zagarri, The Politics of Size: Representation in the United States, 1776–1850, at 37–38 (1987); James A. Gardner, Foreword: Representation Without Party: Lessons from State Constitutional Attempts to Control Gerrymandering, 37 Rutgers L.J. 881, 900–08 (2006).
• 56See Elmer C. Griffith, The Rise and Development of the Gerrymander 19–20, 64–74 (1907); George Athan Billias, Elbridge Gerry: Founding Father and Republican Statesman 316–17 (1976).
• 57Gardner, supra note 55, at 892–94.
• 58See, e.g., Gordon S. Wood, The Radicalism of the American Revolution chs. 13–19 (1992) (documenting the eclipse of a preliberal patronage society to one more egalitarian and democratic).
• 59Gardner, supra note 55, at 901–02.
• 60Id. at 902.
• 61Edward S. Casey, Getting Back into Place: Toward a Renewed Understanding of the Place-World, at xv (1993) (“By late modern time, this world had become increasingly placeless, a matter of mere sites instead of lived places . . . .”). Augé refers to this phenomenon as the rise of the “non-place.” Marc Augé, Non-places: Introduction to an Anthropology of Supermodernity 77–78 (John Howe trans., Verso 1995).
• 62Francis Fukuyama, Liberalism and Its Discontents 71 (2022) (discussing “[t]he liberal tendency to believe that all group memberships are voluntary”).
• 63Jackman v. Bodine, 205 A.2d 713, 718 (N.J. 1964). This view still holds today. In a decision issued in April 2023, the Alaska Supreme Court affirmed that under the state constitution, counties (boroughs in Alaska) are deemed by law to be “socioeconomically integrated.” In re 2021 Redistricting Cases, 528 P.3d 40, 74 & n.132, 79–80 (Alaska 2023).
• 64Gardner, supra note 55, at 962–63.
• 65377 U.S. 533 (1964).
• 66James A. Gardner, One Person, One Vote and the Possibility of Political Community, 80 N.C. L. Rev. 1237, 1243–46 (2002).
• 67Reynolds, 377 U.S. at 560–61, 565, 562.
• 68Id. at 571–82.
• 69James A. Gardner, Neutralizing the Incompetent Voter: A Comment on Cook v. Gralike, 1 Election L.J. 49, 50–51 (2002); Daniel R. Ortiz, The Engaged and the Inert: Theorizing Political Personality Under the First Amendment, 81 Va. L. Rev. 1, 3–4 (1995).
• 70There is a strong echo here of mid-twentieth century political science, especially Dahl’s highly influential book, Robert A. Dahl, Who Governs?: Democracy and Power in an American City (1961).
• 71The classic work is Angus Campbell et al., The American Voter(1960). Campbell’s results have been repeatedly confirmed. See, e.g., Eric R.A.N. Smith, The Unchanging American Voter(1989); Michael X. Delli Carpini & Scott Keeter, What Americans Know about Politics and Why It Matters (1996).
• 72Johnny H. Killian, Libr. of Cong. Legis. Reference Serv, State Legislative Apportionment: An Analysis of Proposed Constitutional Amendments 1 (1965).
• 73James A. Gardner, Autonomy and Isomorphism: The Unfulfilled Promise of Structural Autonomy in American State Constitutions, 60 Wayne. L. Rev. 31, 46–50 (2014).
• 74Mahan v. Howell, 410 U.S. 315, 324–26 (1973) (local government units); Bush v. Vera, 517 U.S. 952, 964 (1996) (plurality opinion) (communities of interest).
• 75Stephenson v. Bartlett, 562 S.E.2d 377, 385 (N.C. 2002).
• 76Id. at 386 (internal citation omitted).
• 77In re Reapportionment of Towns of Hartland, Windsor & W. Windsor, 624 A.2d 323, 330 (Vt. 1993) (quoting Carstens v. Lamm, 543 F. Supp. 68, 88 (D. Colo. 1982)).
• 78This position seems to ratify Benedict Anderson’s contention that arbitrarily established geographical boundaries can acquire real meaning over time. Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism 52 (rev. ed., 2006).
• 79See Miller v. Johnson, 515 U.S. 900, 919–20 (1995).
• 80Chandran Kukathas, The Liberal Archipelago: A Theory of Diversity and Freedom 171 (2003).
• 81For an overview, see Gardner, Representation Without Party, supra note 55, at 937–38.
• 82In re Legislative Districting of State, 475 A.2d 428, 445 (Md. 1982). The court seems to have since softened its views somewhat, viewing the recognition of COIs as a custom that is harmless, if not particularly meaningful. In re Legislative Districting of State, 805 A.2d 292, 297 (Md. 2002).
• 83See, e.g., Comm. for a Fair and Balanced Map v. Illinois State Bd. of Elections, 835 F. Supp. 2d 563, 574–80 (N.D. Ill. 2011) (adjudicating partisan gerrymandering claims regarding two state legislative districts (11 and 13) lying within the City of Chicago).
• 84New York: 2020 Census, U.S. Census Bureau (Aug. 25, 2021), https://www.census.gov/library/stories/state-by-state/new-york-population-change-between-census-decade.html [https://perma.cc/WGY6-4CUH].
• 85Richard Briffault and Laurie Reynolds, Cases and Materials on State and Local Government Law 8 (6th ed., 2004).
• 86For just one example, see ADL Audit Finds Antisemitic Incidents in United States Reached All-Time High in 2021, Anti-Defamation League (Apr. 25, 2022), https://www.adl.org/resources/press-release/adl-audit-finds-antisemitic-incidents-united-states-reached-all-time-high [https://perma.cc/6EFL-LKXN] (reporting incidents in 2021 have reached “an all-time high”).
• 87Thomas J. Main, The Rise of Illiberalism1–9 (2021).
• 88James E. Campbell, Polarized: Making Sense of a Divided America 1–4 (2016); Jacob M. Grumbach, Laboratories against Democracy: How National Parties Transformed State Politics 54–69 (2022).
• 89See Lieberman et al.,supra note 20, at 470–71; Norris & Inglehart, supra note 3, at 245–46; Karyn Amira et al., Adversaries or Allies? Donald Trump’s Republican Support in Congress, 17 Persps. on Pols. 756, 767 (2019); Jacob S. Hacker & Paul Pierson, Restoring Healthy Party Competition, in Democracy Unchained: How to Rebuild Government for the People 40, 43–44 (David W. Orr et al. eds., 2020); Paul Pierson & Eric Schickler, Madison’s Constitution Under Stress: A Developmental Analysis of Political Polarization, 23 Ann. Rev. Pol. Sci. 37, 53–54 (2020). For more thorough discussion, see James A. Gardner, Illiberalism and Authoritarianism in the American States, 70 Am. U. L. Rev. 829, 863–68 (2021).
• 90An “anti-system” party is one that has “played the democratic game with the objective of destroying democracy itself.” Giovanni Capoccia, Anti-System Parties: A Conceptual Reassessment, 14 J. Theoretical Pols. 9, 13 (2002).
• 91Christopher H. Achen & Larry M. Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government 307 (2016) (footnote omitted).
• 92Jonathan Rodden, The Geographic Distribution of Political Preferences, 13 Ann. Rev. Pol. Sci. 321, 322 (2010) (internal citation omitted).
• 93Bill Bishop, The Big Sort: Why the Clustering of Like-Minded America Is Tearing Us Apart 5–8 (2008). Some have expressed skepticism of Bishop’s account. See, e.g., Samuel J. Abrams & Morris P. Fiorina, “The Big Sort” That Wasn’t: A Skeptical Reexamination, 45 PS: Pol. Sci. & Pols. 203, 208 (2012).
• 94Corey Lang & Shanna Pearson-Merkowitz, Partisan Sorting in the United States, 1972–2012: New Evidence from a Dynamic Analysis, 48 Pol. Geog. 119 (2015), suggests the latter is the more important factor.
• 95See Galston, supra note 17, at 5 (“[P]luralism . . . characterizes all free societies in modernity”); Rawls, supra note 47, at xviii (stating that “political liberalism takes for granted . . . the fact of reasonable pluralism”).
• 96Rawls, supra note 47, at xvi, 12.
• 97Thomas Christiano, The Rule of the Many: Fundamental Issues in Democratic Theory 116–27 (1996); John S. Dryzek, Deliberative Democracy and Beyond: Liberals, Critics, Contestations 1–7 (2002).
• 98Bruce Ackerman & James S. Fishkin, Deliberation Day 3–16 (2004).
• 99Andrew Rehfeld, The Concept of Constituency: Political Representation, Democratic Legitimacy, and Institutional Design 8 (2005).
• 100Aristotle, Physics bk. IV, 208b34–35 (Edward Hussey trans., Clarendon 1993) (c. 350 B.C.E.).
• 101Reynolds, supra note 22, at 259.
• 102Liah Greenfeld, Nationalism: Five Roads to Modernity 481 (1992); Elie Kedourie, Nationalism 38 (3d ed., 1985).
• 103Roger Eatwell & Matthew Goodwin, National Populism: The Revolt against Liberal Democracy 59–60 (2018).
• 104Walter Brueggemann, The Land: Place as Gift, Promise, and Challenge in Biblical Faith 5 (1977).
• 105Augé, supra note 61, at 52.
• 106David Delaney, Territory: A Short Introduction 14 (2005).
• 107This is a key element of the communitarian critique of liberalism. See, e.g., Daniel Bell, Communitarianism and Its Critics 6–8 (1993).
• 108Uday Singh Mehta, Liberalism and Empire: A Study in Nineteenth-Century British Liberal Thought 119 (1999) (footnote omitted).
• 109Margaret Kohn, Radical Space: Building the House of the People 3–4 (2003) (footnote omitted).
• 110Fukuyama, supra note 62, at 97–98.
• 111This is, for example, largely the account that emerged out of the South following the Civil War to justify secession on grounds of the impossibility of Northern and Southern co-habitation, an account according to which the war, and subsequent attempts by the central government to weaken illiberal institutions, were simply a form of domination by conquerors. See especially, Frank Lawrence Owsley, The Irrepressible Conflict, in I’ll Take My Stand: The South and the Agrarian Tradition 61, 63–65 (Harper 1962).
• 112Galston, supra note 17, at 5–6.
• 113Jane Mansbridge & Stephen Macedo, Populism and Democratic Theory, 15 Ann. Rev. L. & Soc. Sci. 59, 62 (2019).
• 114See Margaret Canovan, Trust the People! Populism and the Two Faces of Democracy, 47 Pol. Stud. 2, 4–5 (1999).
• 115Nadia Urbinati, Democracy Disfigured: Opinion, Truth, and the People 135–37 (2014).
• 116Carlos de la Torre, Introduction, in The Promise and Perils of Populism: Global Perspectives 19 (Carlos de la Torre ed., 2015) (stating that the leader is “the incarnation of the people,” an idea that rests on “[t]he fantasy of the unity of the people and of their merging with the leader”).
• 117Luigi Corrias, Populism in a Constitutional Key: Constituent Power, Popular Sovereignty and Constitutional Identity, 12 Eur. Const. L. Rev. 6, 19 (2016); Geneviève Rousselière, Can Popular Sovereignty Be Represented? Jacobinism from Radical Democracy to Populism, 65 Am. J. Pol. Sci. 670, 670 (2021).
• 118de la Torre, supra note 116, at 19. See Müller, supra note 15, at 29.
• 119John Agnew & Michael Shin, Mapping Populism: Taking Politics to the People 16 (2020).
• 120See generally Zachary Roth, The Great Suppression: Voting Rights, Corporate Cash, and the Conservative Assault on Democracy 98–99 (2016); Richard L. Engstrom, Partisan Gerrymandering: Weeds in the Political Thicket, 101 Soc. Sci. Q. 23, 23–24 (2020).
• 121For a recent example, see South Carolina State Conf. of NAACP v. Alexander, No. 321CV03302, 2023 WL 118775, at *7 (D.S.C. Jan. 6, 2023) (three-judge court) (finding intentional racial gerrymandering described as “bleaching” a district). For an overview of “egregious” Republican gerrymandering following the 2010 census, see generally Engstrom, supra note 120.
• 122To give but one example, H.R. 1, the For the People Act, contains several provisions barring partisan gerrymandering. The Act, introduced in 2021, had 222 co-sponsors, all Democrats. Cosponsors H.R.1–For the People Act of 2021, Congress.gov, https:‌//www.congress.gov‌/bill‌/117th-congress‌/house-bill/1/cosponsors [https://perma.cc/82L2-M93Q].The bill passed the House on a party-line vote. Brian Slodysko, House Passes Sweeping Voting Rights Bill over GOP Opposition, AP (Mar. 4, 2021, 7:41 AM), https://apnews.com/article/house-passes-sweeping-voting-rights-bill-88088175552f13a8e3f6f25d7bc45f6c [https://perma.cc/JD83-QFD5]. The Republican-controlled Senate declined to take it up. Jacob Pramuk, Senate Republicans Block Democrats’ Sweeping Voting, Ethics Bill, CNBC (June 22, 2021, 7:45 PM), https:‌//www.cnbc.com‌/2021‌/06/22/senate-to-vote-on-s1-for-the-people-act-bill.html [https://perma.cc/WRU2-NFJ7].
• 123For populists, as Müller observes, “[t]he constitution ceases to be a framework for politics and instead is treated as a purely partisan instrument to capture the polity.” Müller, supra note 15, at 40. See also Urbinati, supra note 115, at 7 (stating that politics for populists is “a terrain of conquest”).
• 124James A. Gardner, Electoral Systems and Conceptions of Politics, in Comparative Election Law 140, 146 (James A. Gardner, ed., 2022).
• 125See Ferran Martínez i Coma & Ignacio Lago, Gerrymandering in Comparative Perspective, 24 Party Pols. 99, 102 (2018) (showing that “majoritarian systems are more prone to gerrymandering than . . . PR systems”).
• 126Gardner,supra note 124, at 150–153.
• 127PR has clearly been beneficial to the parliamentary presence of numerous populist parties in Europe. Norris & Inglehart, supra note 3, 318–19.
• 128Donald L. Horowitz, Ethnic Groups in Conflict 617–19 (2d ed., 2000) (discussing scenarios in which the creation of heterogeneous states can reduce ethnic conflict).
• 129Benjamin Reilly, Democracy in Divided Societies: Electoral Engineering for Conflict Management 6 (2001).
• 130This is the position I took in an article published a decade ago. James A. Gardner, How to Do Things with Boundaries: Redistricting and the Construction of Politics, 11 Election L.J. 399 (2012).
• 131Katrina vanden Heuvel, Opinion, Ranked-Choice Voting is Already Changing Politics for the Better, Wash. Post (May 4, 2021, 8:00 AM), https://www.washingtonpost.com/opinions/2021/05/04/ranked-choice-voting-is-already-changing-politics-better/ [https://perma.cc/6F27-54ND].
• 132Adam Cayton & Ryan Dawkins, Incongruent Voting or Symbolic Representation? Asymmetrical Representation in Congress, 2008–2014, 20 Persps. Pols. 916, 917 (2022).
• 133James A. Gardner, What Is “Fair” Partisan Representation, and How Can It Be Constitutionalized? The Case for a Return to Fixed Election Districts, 90 Marq. L. Rev. 555, 583–84 (2007). A representative example is one established by the Missouri Constitution of 1875, art. IV, § 2. Under that provision, the population of the state was divided by 200 to establish a “ratio of representation.” Mo. Const. of 1875, art. IV, § 2. Each county having a population less than or equal to one ratio of representation was entitled to elect one representative to the state house. Id. Counties with populations equal to or greater than two-and-one-half ratios earned a second representative. Id. A third representative was allocated to counties with four ratios, and so on. Id.
• 134Bell, supra note 107, at 9–10; Alasdair MacIntyre, After Virtue: A Study in Moral Theory 220–21 (3d ed., 2014).
• 135Michael Sandel, Liberalism and the Limits of Justice 150 (2d ed., 1998).
• 136Kukathas, supra note 80, at 169 (describing trade unions and professional organizations as types of community).
• 137Sandel, supra note 135, at 150–52.
• 138See, e.g., Bell, supra note 107, at 10 (“[W]e are indeed able to re-examine some attachments, [though there may be] others so fundamental to our identity that they cannot be set aside . . . .”); MacIntyre, supra note 134, at 222 (describing individuals as being embedded in inherited traditions, but traditions that are “living,” and thus subject to alteration).
• 139Reference re Prov. Electoral Boundaries (Sask.), [1991] 2 S.C.R. 158.