I. Introduction
Civil societies are nourished not by one but by two freedoms of association. The first is positive freedom of association, a power to combine with some other party or parties in a shared activity or status. Positive freedom of association admits of varying degrees of attractive force. The weakest is an association constituted by the consent of all parties. Stronger is an association one is empowered to join even if some other party or parties prefer not to be so associated, while others desire association. Strongest is where one's membership in the association is guaranteed, not requiring the consent of any other. The second freedom is negative freedom of association, a liberty to dissociate from unwanted relationships. The strongest version of negative freedom of association is the power to withdraw regardless of the assent of others. Less strong forms of negative freedom permit withdrawal subject to the concurrence of certain others. So, for example, in certain traditional marriage practices, a husband enjoys a very strong freedom to dissociate; he is empowered to dissolve the marriage on his own volition. A wife, however, may dissociate only with the concurrence of her husband and therefore enjoys only a weak negative freedom. (If divorce is not permitted under any conditions, then the parties altogether lack the freedom to dissociate.)
I characterize these as two distinct freedoms of association because, although complementary, they stand in potential tension. The former is centripetal, the latter centrifugal, and imperatives to draw together are weakened or negated by permissions to withdraw. This tension is especially acute within a liberal order. Because such an order vests primary authority in the decisions of individuals, it must afford them great latitude to go their own way as they see fit. But because it is indeed a social order, it is charged to uphold a standard of general inclusiveness. The freedom to associate is checked by a freedom to withhold unwanted association, and vice versa. What may be called the Paradox of Association was most memorably stated by Marx when he announced, “I refuse to join any club that would have me as a member.”Footnote 1 The task of this essay is to examine the plausibility of some proposed rules of club membership.
Section II sketches various bases of the negative freedom of association. These are seen to be manifold and indispensable to the liberal enterprise. Section III presents positive association both as liberty and as ideal. Section IV takes up a handful of potential conflicts between the two freedoms of association. Section V sums up with a program for not quite resolving but at least living with the Paradox of Association.
II. Freedom of Dissociation
The absence of liberty to go one's own way is subjugation. Consequently, any regime other than a slave society countenances in some measure that liberty—and even slave societies are unable to extinguish the primordial human desire to live freely. It is the particular distinction of liberalism, however, to put expansion of the domain of protected choice front and center as a regulative political principle. The early liberal campaign was waged along a number of dimensions in which freedom to dissociate is paramount.
Freedom of conscience, especially as related to religious practice, was an early outpost of the struggle. Insofar as it is understood as the struggle of individuals to secure adequate scope to practice their faith alongside others who espouse similar convictions, this is a mode of positive freedom of association. However, these confessional congregations cannot begin to get underway until there is acknowledged a negative freedom to dissociate from the established church. In the early modern period, conscience was understood to be a jealous mistress who does not brook any sharing of favors with competitors. To follow a preferred path to salvation was to eschew incompatible routes. Thus, freedom of religion is, in the first instance, a negative entitlement.
Freedom of speech is similar. Although it is not in any obvious way a mode (either positive or negative) of association, it displays aspects of both. Other than the limiting case of soliloquy, speech is an act of interpersonally shared communication that unites speaker with listener. Sotto voce free speech doesn't count for much. Therefore, the realization of freedom of speech incorporates some form or other of semantic association. But prior to a liberty to speak one's mind is a permission to desist from echoing the words that issue from the commanding heights of pulpit or palace. That is, freedom of speech is no less fundamentally a freedom not to give utterance to objectionable phrases than it is to give declaration to one's own beliefs and attitudes. Speech rights are, then, associational and, in the first instance, negative.Footnote 2
Occupational and residential mobility emerged with the waning of a feudalism that tied individuals to land and labor. Entitlement to take up whatever employment may be on offer presupposes a liberty to divest oneself of prior jobs; moving to a new address presupposes a right to leave the old address. Both liberties were planks of early liberal platforms.Footnote 3 At a somewhat later date, the liberation of women from paternal or spousal control moved to the center of liberal concern.Footnote 4 Classical feminism displayed various interrelated facets, but integral to many was an underlying freedom to dissociate.
Private property rights are not usually thought of in terms of freedom of association. However, insofar as ownership establishes against all others a duty to refrain from access to the object in question absent the owner's assent, it is a strong barrier against unwanted intrusions. Property affords one a say concerning with whom one will associate and on what terms. If a man's home is his castle, that is not in virtue of sumptuous interiors or lofty towers but rather because of a figurative moat around the perimeter that excludes unwanted entry. What one owns is that from which one is at liberty to exclude others for good reasons, bad reasons, or no reasons at all. The other side of the property coin is that it affords individuals means to induce others to undertake positive associational relations. The home/castle is not usually a hermit's cell; more often, it is a site for raising families and hosting friends. Personal property is that with which people truck, barter, and exchange.Footnote 5 If it is only through the ongoing concurrence of others that one holds land or chattels, one's life is only in a diminished sense one's own. Property is the fount of independence.Footnote 6
For early modern political theory, no imperative sounds more emphatically than the need to establish conditions of peace. Thomas Hobbes famously identifies the state of nature with the state of war,Footnote 7 and although John Locke resists that equation, he is under no illusion that people in anarchy will be able to avoid the “inconveniences” consequent on lacking a common judge to resolve disputes.Footnote 8 It is worth noting that freedom to dissociate is a crucial element of the quest for social peace. If individuals' only recourse for avoiding unwanted association is to fight, then fight they will. Far better for purposes of general comity is the extension of a robust liberty to go one's own way. To be obliged to act in concert with those whose ends or attributes one disdains is costly to one's integrity. When that disdain is intense enough, the cost may well be deemed not worth paying. If not altogether eliminated, the cost is at least reduced when the disfavored enterprise can be viewed as theirs rather than ours. Insofar as a freedom to dissociate empowers individuals to avoid involuntary implication in disfavored designs of others, it conspicuously promotes the quest for peace.
If peace is the sine qua non of early liberalism, the leading player in more recent incarnations of the theory is autonomy. An autonomy-centered liberalism holds that conceptions of the good are not to be imposed from above (or below, or sideways) on individuals, but rather are to be self-determined. Autonomy rejects all ideologies of one-size-fits-all. There is, to be sure, a significant role for the state in promoting values, but it is the indirect and derivative role of affording to individuals conditions conducive to independence of thought and deed. Just what those conditions are is the subject of ample philosophical debate. For modern welfare liberals, the extent of these conditions goes well beyond noninterference to include positive provision of items such as education, health care, and wealth redistribution. These are to be afforded not because the state officially ranks them above other goods, but because they enable individuals to achieve autonomy. Whatever one thinks of these additions to the traditional package of state responsibilities, they augment but do not replace the primacy of freedom of dissociation. A person who is constrained to run with the herd may travel in a propitious direction, but it will not be her direction. Only when people are free to detach themselves from other values, modes of life that are not their own, are they able to act autonomously. This is arguably not a sufficient condition for the autonomous life—one may perhaps, with Jean-Jacques Rousseau, hold it necessary from time to time to force individuals to be genuinely free—but it is necessary.
If autonomy was the liberal ideal of the twentieth century, it is conceivable that diversity will assume that status in the twenty-first. There may not be much of Mao's legacy preserved in contemporary China, but his injunction to let a thousand flowers bloom is treasured on every American college campus and by every governmental bureau. To be sure, general declarations of praise for diversity are often best interpreted as coded messages in support of some particular favored pattern of representation, but these invocations could not succeed even as euphemism were it not the case that “diversity” in its primary, uncontaminated sense carries positive connotations. Nothing is more necessary for genuine diversity to flourish than a protected freedom for individuals and groups to separate from and act independently of dominant social entities. A minority culture that is forcibly subsumed within some larger collectivity has its own distinctive identity thereby besmeared. In principle, this applies all the way down to the level of individuals and to novel associational linkages; I do not deny that the practice of the diversity industry often operates otherwise.
Dissociation serves other aspects of a free and prosperous society. Innovation, almost by definition, requires detachment from traditional patterns of personal interaction and productive practices. That is not to maintain that all innovation constitutes improvement, but its absence is economic and cultural stagnation.Footnote 9 Since selection through comparative fitness operates on products of invention much as it does in the realm of biological evolution, there exists an asymmetry in favor of innovation even if its fruits are randomly distributed between useful and useless. Because intention and deliberation guide the process, better than Darwinian outcomes are to be expected. What Joseph Schumpeter calls “creative destruction”Footnote 10 is often unsettling, but it is the antidote to stasis and stultification.
Dissociation also serves to foster the proliferation of positional goods. If there exists a common metric for an entire collectivity, the top will be lonely; only in Lake Wobegon can all the children be above average. However, if people separate themselves out in a multitude of ways, endlessly creating activities to which they can lend their efforts, the likelihood that a representative individual will discover some avenue along which he can be the best (or excel somewhat, or at least be above average) increases. The economist Tyler Cowen notes that there are in America over three thousand Halls of Fame (an improbably large number of them devoted to bowling).Footnote 11 It is arguable whether comparing one's own accomplishments to those of deficient others is an altogether morally savory practice, but it seems clear that individuals' self-esteem is in large measure a function of their positional standing. (Crowds in stadiums declaim “We're Number 1! We're Number 1!” not “We lose a lot of games but play pretty darn hard!”) Enhanced access to positional goods not only massages individual psyches but also serves other political ends noted above, including especially the maintenance of civic peace.
This is not meant to be an exhaustive accounting of the merits of negative freedom of association, the liberty to dissociate. The aim has been to say enough to establish the following two points: (1) freedom of dissociation is a major element of many political desiderata; and (2) positive freedoms of association often presuppose the corresponding negative freedom. It is to the positive freedom that I now turn.
III. Positive Freedom of Association
Should it be concluded that a liberal political order can dispense with any independent concern for a centripetal positive freedom of association? In a word, no. At least one association is mandatory for those within civil society: the universal association that is the political order itself.Footnote 12 For all those located within borders, aliens as well as citizens, obeisance to the social contract is not open for negotiation.
Both Hobbes and Locke officially depict the state as generated by the consent of sovereign individuals, although in Hobbes's case the voluntariness of that consent may be impugned by his insistence that even when extracted at the point of a sword, it remains valid.Footnote 13 For both, however, that consent once given is irrevocable. The only way to withdraw from duties of obedience to the state is physical relocation. Neither secession of a territory over which one holds ownership nor declared intent to resume the status of an independent is permissible. Immanuel Kant's insistence on the necessity of incorporation within the political association is even less compromising. No initial voluntary declaration of allegiance is required:
[T]he first decision the individual is obliged to make, if he does not wish to renounce all concepts of right, will be to adopt the principle that one must abandon the state of nature in which everyone follows his own desires, and unite with everyone else (with whom he cannot avoid having intercourse) in order to submit to external, public and lawful coercion…. In other words, he should at all costs enter into a state of civil society…. Anyone may thus use force to impel the others to abandon this state for a state of right.Footnote 14
Liberals may differ among themselves concerning whether secession is justified in extremis (and how dire circumstances must be to lie on the far side of extremis), but the essential point is that over a very wide range of political practice, there is one association that is not optional, not dispensable.
Mandatory membership promotes peace. It is also conducive to the cultivation of autonomy. Kant argues that autonomy is expressed by directing one's conduct in accord with practical reason's moral law. If individuals are to engage in genuinely moral relations with others as governed by the categorical imperative, there must exist a standard that all parties have reason to acknowledge as rationally authoritative over their wills. That is impossible in a state of nature. Autonomy and the establishment of property rights travel together. In order to respect the distinction between mine and thine, ownership must be interpersonally grounded via impersonal principles of right. Only civil law can provide those principles, and only a state can generate civil law. Positive freedom of association, then, is arguably no less essential for the realization of liberal ideals than is freedom of dissociation.
For many theorists, however, one inescapable association defined simply in terms of law-abidingness is too thin a gruel. They maintain that more is needed to ground an adequately humane society. That criticism of ancestral liberalism is manifest in the French Revolution's replacement of Locke's triad of “life, liberty, and property” (and Thomas Jefferson's “life, liberty, and the pursuit of happiness”) with “Liberty, Equality, Fraternity.” Citizens—and this was by a wide margin the preferred term for co-nationals—should be drawn together by more than an arbitrary shared geography and legal duties. As partners in an exercise of political self-determination,Footnote 15 they are united at a portentous level by a project that unifies and confers meaning on their efforts. According to Rousseau, they transcend the condition of discrete centers of desire and come to partake of a “general will.” As “patriots,” they are the progeny of a fatherland that renders them all brothers and sisters. The conception at work here is premodern, hearkening back to Aristotle's understanding of “civic friendship” and to Plato's Myth of the Metals in the Republic, in which a story of radical political consanguinity is proffered. But although ancient, it surfaces in an important strand of modern liberal democratic thought.
Fraternity, almost by definition, privileges the centripetal over the centrifugal. The good life is not well-lived by the “solitary rights-bearer”;Footnote 16 rather, it is social. Classical liberalism's focus on staying out of each other's way is criticized as generating a desiccated politics, an “atomistic individualism.”Footnote 17 The state is not, of course, the only society that affords sustenance to individuals, but it is preeminent in virtue of being comprehensive in its makeup and requisite for the flourishing of all lesser associations. Therefore, a proper function of government is to provide institutional structures that draw persons together, most especially with regard to activity on behalf of the commonwealth itself. For those who take fraternity seriously, laissez-faire is not nearly good enough.
Opponents of an austere classical liberalism typically fault its excessive devotion to freedom of dissociation. Such is the nub of the communitarian critique.Footnote 18 We do not form conceptions of the good adequate to nourish a meaningful life in a normative vacuum. Rather, says the communitarian, our valuational inclinations are no less the product of a formative social environment than is the language we speak. It is, then, a matter of some urgency for a political order intent on sustaining itself and its members in full civic health to promote meaning-conferring associations. Indiscriminate exercise of the exit option is inimical to the vitality of the body politic. Therefore, communities are to be afforded considerable latitude, including well-aimed applications of coercion, to give effect to their constitutive values and thereby hold themselves together against the corrosive forces of untrammeled dissociation.
Although John Rawls is usually the communitarians' preferred foil in virtue of his emphasis on the importance for moral theory of the separateness of persons, he is not altogether immune to the attractions of extended association. In a striking characterization of his own theory, he declares: “In justice as fairness men agree to share one another's fate.”Footnote 19 Being bound to one another's fate is an extraordinarily powerful form of association. It is a melding of the good of one with the good of all. Rawls does not present sharing of fate as an exclusive attribute of relations of deep friendship or familial solidarity; to the contrary, it is a precept of justice. Thus, it establishes an associative nexus both general and mandatory. Implications within his theory are sweeping, including a strongly egalitarian difference principle from which individuals are not at liberty to dissociate.
Socialism similarly configures the polity as an economic association within which, by default, all individuals are shareholders. Unlike the investors in a capitalistic enterprise, however, their holdings are quantitatively identical and not alienable. If private ownership separates individuals and confers on them independence one from another, public ownership of the means of production renders them mutually interdependent. It would be a mistake to view socialist theory merely as a hypothesis about how to secure efficient and equitable economic production—which, in any case, has never been a locus of conspicuous achievement for the theory. It also, and more centrally, incorporates an ideal of shared fates, perhaps in a sense even stronger than that intended by Rawls.
If social contract is the birth of a state, secession and civil war are its death. Cohesiveness combats dissolution, and strong association promotes cohesiveness. The ideal of the melting pot, although drastically out of favor in an age of enthusiastic paeans to diversity, directly addresses the cohesion problem. Cooperation with those one finds disconcertingly different is problematic. They are instead likely to be perceived as threats eliciting either avoidance or confrontation. But if that which divides potentially antagonistic parties is burned off in the political smelter, then passions that commonly set individuals and groups in opposition will be defused. People who are rendered more similar will dissociate less often and less intensely. That's good for social harmony.
If “melting” is deemed an improbable (or potentially tyrannous) goal, applying layers of “social glue” to hold together disparate segments can present itself as a more moderate means to a similar peace-preserving end. A currently fashionable brand of this glue goes by the name social capital. Footnote 20 It is formulated through associations in which people who know each other on a face-to-face basis come to develop dispositions to trust and to cooperate. Social capital cements a wide range of relationships, including especially those bound up with peaceful coexistence. Even such seemingly apolitical entities as bowling leagues, book discussion groups, and the Loyal Order of Moose have strongly positive effects on general comity. Therefore, governments are well-advised to adopt measures friendly to the formation of social capital. At the very least, governments are most emphatically not to be neutral between civic association and dissociation.
The moral to be drawn from this and the preceding section is that there are powerful currents within modern political philosophy supporting strong freedom of dissociation, and there are comparably powerful currents that favor active promotion through governmental means of positive association. As morals go, this one is feeble, reminiscent of the old joke about the rabbi and the disputing congregants.Footnote 21 The interesting question is not whether there might be something to be said on behalf of both positive and negative association, but rather which ought to take pride of place when they conflict. It is to this inquiry that the discussion now turns.
IV. A Handful of Applications
Consider the following two-part principle: (1) Preservation of the universal association constituted by the political order itself trumps all exercises of freedom of dissociation. (2) Nothing else does. It is as invigoratingly bold as the conclusion of the preceding section was timid. Unfortunately, it is too uncompromising to be supportable. If a regime is sufficiently vile, then its destruction through forceful exit is justified. (The breakup of the Soviet Union presaged by the departure of the Baltic nations is an instance of salutary dissociation.) Also, acts of dissociation which do not seriously threaten the continued existence of the state but which contribute to friction in its internal operations may permissibly be stymied, either through outright prohibition or via more subtle means of discouragement. Therefore, it is necessary to preface the principle with “For the most part.” So qualified, it is defensible but too noncommital to be very helpful.
One way to lend more substance to the (qualified) principle is to apply it to selected cases and observe how they fare as instances of the Paradox of Association. A trade-off between breadth and depth is operative here. Multiplying cases affords more data points against which the theory can be assessed, but each case will be presented more superficially. Because the aim of this section is not so much to adjudicate particular policy disputes by analyzing them all the way down to the ground as it is to exhibit the sorts of considerations that can be brought to bear on a wide range of questions, the superficiality concern is less worrisome. I display in the remainder of this section a handful of illustrative cases, one of mostly historical interest, the others currently disputed. Readers are invited to develop other instances of negative and positive association in tension.
A. “Separate but equal”
In the United States prior to the pivotal 1954 Brown v. Board of Education case,Footnote 22 coercively imposed separation of the races with regard to the enjoyment of state services was countenanced subject to the condition that the quality of provision be equal. There are numerous grounds on which this policy could be (and was) challenged: equal quality was not, in fact, achieved (or even seriously intended); racial discrimination is inherently invidious and thus unequal; the Constitution is color-blind. This is not the occasion for entry into the realm of jurisprudence. Rather, the point of revisiting the distant world of Plessy v. Ferguson (1896)Footnote 23 is to observe that as a matter of pure political philosophy, state enforcement of the “separate but equal” doctrine is dubious. That is because it impugns the integrity of the universal positive association. Forcible exclusion of some citizens from the terrain of others creates the shadow of a state within the state, a house divided against itself. A social contract that systematically differentiates between subpopulations might better be characterized as an antisocial contract.
Suppose, counterfactually, that racial discrimination had been instituted as the expressed preference of majorities within the entire citizenry. Does that afford democratic legitimacy to segregation? Not if the minority reasonably takes itself to be thereby disadvantaged. Suppose (further out on the counterfactual curve) that majorities within both races favored separation. Is it then justifiable policy? The answer is not entirely clear-cut. One theory of peaceful coexistence maintains that good fences make good neighbors. If segregation alleviates frictions that might otherwise be inimical to civic peace, then it could be justified in the name of ensuring the stability of the universal association. In effect, “separate but equal” amounts to a sort of tension-reducing racial federalism. This is, of course, far-fetched. More plausibly, enforced separation is a step in the direction of internal division and disrepair. Unless the avowed interest of all groups is to divide—in which case we would wonder why they wish to remain in political association at all—even scrupulously equal separation policies are contraindicated.
B. Affirmative action
The term “affirmative action” is fuzzy, covering a range of policies from the innocuous to the controversial. I am not concerned here with activities designed to transmit knowledge of employment or admissions openings to classes of people who might otherwise lack awareness of their availability. Instead, I confine my attention to application of differential selection standards predicated on group membership. These include adding some specified weight to applications of favored candidates, affording them an indeterminate preference when qualifications are close, or imposing fixed selection quotas. Although affirmative action so specified can be applied to any (dis)favored group, the paradigmatic context of categorization in the United States is racial. Because the goal is enhanced inclusion of previously excluded parties, affirmative action policies fall under positive freedom of association. Are they justifiable?
If affirmative action is voluntarily practiced by a nongovernmental entity in service of its own values, then it is relatively unproblematic.Footnote 24 Company X or University Y is at liberty to extend its own preferred terms of association to willing others just as you and I are at liberty with regard to with whom we will pass our convivial hours. The two more complicated cases are (i) affirmative action practiced by organs of the state in the conduct of their own operations, and (ii) governmentally-mandated affirmative action requirements that fall on nonstate actors. I take these two cases up in turn.
Considerations adduced in the previous subsection speak against state-practiced affirmative action. Although it aims not to separate races but rather to bring them into more extensive association, the method of doing so is to apply different standards to subpopulations. No one is excluded from the universal association, but terms of membership are to the advantage of some against others. This is inherently inimical to the primary desideratum of the social contract: peace. All else equal, state-sponsored affirmative action is an inferior form of positive association.
All else, however, is far from equal. The great enduring rip in the social fabric of the United States is slavery and its aftermath. Neither the Civil War nor civil rights initiatives a century on succeeded in establishing an adequately inclusive social order. Thus, derivation of recommended principles of political association is a task of non-ideal theory, what the economists call the realm of “second-best.” A history of flawed policy can render acceptable or even mandatory responses that would otherwise be unacceptable. (Compare with the practice of punishing people who have been found guilty of offenses by forcibly depriving them of life, liberty, or property.) Racial frictions are the single ugliest blotch on the canvas of U.S. domestic relations. A significant prospect of substantial melioration might, then, outweigh the presumption against applying differential criteria. This is especially the case if those who bear the burden of these policies are, in other precincts of civic life, beneficiaries of differential standards that work to their own favor. To extend dissimilar treatment to different subpopulations is to infringe standards of association for the sake of promoting association. A separate matter is whether affirmative action as practiced by organs of the state does, in fact, tend to promote harmonious interracial association. That is a question not best answered from the comfort of an armchair. As a matter of theory, though, there is room from a liberal free association perspective for a coherent case on behalf of state-sponsored affirmative action.
Imposition of affirmative action requirements on private parties blocks otherwise available dissociation prerogatives and thus confronts a yet steeper justificatory hurdle. The criteria one employs for undertaking relations with willing others are important components of one's conception of how best to lead one's life. To be subject to external direction in these matters is a significant imposition. That it is within the purview of the liberal state to impose such requirements for broadly utilitarian or egalitarian reasons is highly disputable. If, however, the rationale is enhancement of civic peace, then the case becomes more difficult to adjudicate. In effect, some measure of the negative freedom of dissociation is being traded off for augmentation of positive association. Both freedoms are integral to a decent political order, and, therefore, it is at least an open question whether an exchange of one for the other is well-undertaken. And because the stability of the universal association is a prerequisite for all other associational and dissociational liberties, its maintenance is a trump. Three centuries of racial hostility constitute a prima facie plausible rationale for strong meliorative intervention. Peace is worth some constraint, especially if that constraint is of limited duration such that it will more or less automatically expire when the end for which it was inaugurated is achieved. If duration and scope of application are open-ended, then the onus against restriction is heavier. Again, whether the policy tends more to soothe or to exacerbate conflict cannot be settled as a matter of pure theory.Footnote 25
C. Public schools
Let it be granted for the sake of argument that children possess a welfare right to be educated, and that the duty correlative to this right is lodged against the general public rather than the child's parents. It does not follow that the state is justified in establishing a privileged system of free (to the user, not, of course, to taxpayers) public schools. Governmental support of primary education could instead be decentralized to the household level via cash payments or vouchers for the parents/guardians of school-age children. Nonetheless, in the U.S. each of the fifty states funds a monolithic public school system; not one state offers more than a nugatory amount of decentralization through parental choice.Footnote 26 Why?
This could be interpreted as a question about the balance of political forces, in which case extended reference to the strength of teachers' unions and other battalions making up the educational juggernaut would be to the point. Take it instead as a request for justification. Then answers are not so readily forthcoming. Or, rather, credible answers are elusive. For example, if it is claimed that state schooling is necessary to promote equality, that otherwise the poor would suffer inferior educational services, the obvious retort is that despite—or because of?—state schooling, the poor fare much worse than do the rich. A claim that kids will learn well if and only if the state is directly involved in the provision of schooling will not survive a cursory glance at standardized test scores. The argument that parents in general and uneducated ones in particular will do an inferior job of selection in an open market generates in response the question “Compared to what?” If the relevant comparison class is the schooling currently provided by the public education cartel,Footnote 27 then it is questionable whether parents are liable to do worse. The argument that under a scheme of vouchers the most able and ambitious will exit the public system, leaving only the hard cases behind, is morally disgraceful insofar as it regards holding children hostage as acceptable social policy. The list of defective rationales could easily be extended.
The interesting puzzle for political theory is why, despite such palpably inferior arguments, public schooling remains popular.Footnote 28 Ignorance and differential concern might provide part of the answer: parents estimate more highly the quality of their own children's schools than those attended by other children. In addition, though, associational considerations bear significantly on the continuing strength of affection for public schooling. A Great American Myth presents the public schools as heroic agents of socialization, the indispensable institution that unites children of plenty with children of want, newly arrived immigrant populations with long-time residents, kids of all different ability levels. Whether in the one-room schoolhouse of the prairie or the urban repositories of the children of tenements, public schools were instrumental in turning a diverse population of boys and girls into … Americans. To be sure, the myth never applied very plausibly to substantial segments of the population, most notably those separate but allegedly equal black children and those who assumed the financial burden of funding parochial education. Nonetheless, it is evocative. The ideal it serves is that of a population of diverse citizens united by allegiance to a distinctively public weal. However much they ultimately go their separate ways with regard to religion, economic status, and culture, they nonetheless remain linked by a shared commitment to the universal association. Public schools, on this conception, are validated less by what students learn than by the fact that they are learning together—and that they are thereby learning how to do yet other things together.
I am agnostic concerning how accurately the myth applies to public schooling during its 1850–1950 heyday, and considerably more skeptical with regard to current applicability. If it was/is the case, however, that state schooling supplies public cohesion that would otherwise be absent, and if that cohesion is a significant contributor to civic peace, then restrictions on the liberty of parents to go their own way with regard to educational choices might be defended. I speculate that the continuing popularity of public schooling rests on some such belief. Or if not belief, then pious hope: In an age of fragmentation where along numerous dimensions the center appears not to be holding, it might seem too risky to disestablish the institution with the myth behind it. In the calculations of state schooling cartel advocates, positive association defeats negative. One may have reason to suspect that these calculations incorporate erroneous estimations, but the underlying theme of the need to safeguard crucial positive associations is worthy of respect.
D. Immigration
When individuals who hold foreign passports are interdicted at borders, their freedom of mobility is obviously impaired. Somewhat less obvious is that immigration restrictions thwart association. Very few people cross the Rio Grande or the Pacific into the United States with the intention of establishing themselves as hermits. Rather, they aspire to secure work with willing employers, to obtain residences through mutually acceptable transactions with landlords, and to purchase the necessities and a few of the superfluities of life from eager merchants. They also will tend to establish consensual relationships with their new neighbors in matters religious, cultural, and social. Blocking immigration is not merely an imposition on those left outside of borders looking in; it limits valued associational opportunities for those already here.Footnote 29
There are any number of reasons that can be given on behalf of restricting foreign access. It is my opinion that these mostly fail. Those considerations that do have some merit support less draconian measures than exclusion. I have examined these arguments elsewhere and will not repeat them here.Footnote 30 I here acknowledge, though, that the severe impediments to free association created by tight border control can be countered, not entirely implausibly, with an argument also predicated on the importance of free association.
From the birth of the United States as a land with open borders that seekers of liberty were invited to cross, there have always been nay-sayers who warned that the cohesion of the republic was gravely imperiled by the influx of waves of people ignorant of our traditions and temperamentally unsuited to our way of life.Footnote 31 Although these alarums have always been belied (pace hard-line nativists), this does not clinch the case that they will continue to be inaccurate. Inductions in the service of political prediction are notoriously perilous, especially when salient conditions have undergone transformation. One such condition might be the burgeoning of the welfare state; another is the waning of the melting-pot ideal and the concomitant waxing of cultural pluralism. The nativist contention, then, deserves to be taken seriously, both with regard to the merits of its empirical foundations and also its moral standing. That is, there are possible worlds in which it would be efficacious and morally appropriate to harden borders. Is one of them the actual world?
I do not find plausible the suggestion that even a very large intake of foreign nationals seeking to better their lot and to enjoy the perquisites of residence in an advanced liberal democracy threatens national cohesion. The contrary seems to be more indicated; those who display their values by voting with their feet for this way of life thereby endorse the universal positive association. Their accession can be expected to afford it strength. But immigrants who establish residence in democratic societies while rejecting the liberal values on which those societies are founded, and who seek to form insular oppositional enclaves apart from the host civilization, may indeed be exercising liberties of dissociation inimical to the common good. It is one thing to defend cultural pluralism within the ambit of an overarching universal association that enjoys the support of the various different factions, and quite another to countenance those who would subvert the underpinnings on which that association is sustained. A free society is not obliged to welcome into its midst those who espouse a rejectionist ideology. In the wake of September 11, 2001, the obvious candidate for such designation is revanchist Islam. No great prophetic facility is needed to predict that over the next decade and beyond this will be the locus of a major debate in the United States and, especially, Western Europe about appropriate terms for sustainable civil association.
E. Health-care policy
There are innumerable reasons to believe that the U.S. system of delivering health care is gravely flawed. Many people lack regular access to routine health services; care is inordinately expensive both in absolute terms and as a percentage of national GDP; perverse incentives and cross-subsidies abound. All of these are reasons to support some form of universal health care such as that on offer in Canada or the UK. There are, however, reasons to be very wary of these alternatives, including limitations on choice, shortages of materiel and personnel, and protracted wait times for noncritical (and, occasionally, critical) procedures. Reasonable people can differ concerning what would and would not count as an improvement. Here I confine my attention to one less frequently observed motif underlying the health-care debate: implications for the universal association.
In a society fractured along virtually every conceivable dimension, the cords that hold a citizenry together become increasingly tenuous and frayed. One need not be a communitarian enthusiast to be apprehensive about the staying power of a society that has a shrinking store of shared commitments and experiences. A large part of the attractiveness of universal health care is that it answers to the demand for an important common enterprise in which all have an equal stake. National health care serves something of the same functions that an established church did in premodern times. It addresses itself to our deepest concerns about human fragility and mortality, concerns that are no respecters of wealth or power. Its practice is mostly carried out in imposing edifices served by specially anointed, distinctively garbed professionals who are initiated in mysteries not vouchsafed to the common run of men and who, in keeping with their lofty status, are addressed in deferential terms of respect. Denial of access to their ministrations is an especially cruel form of excommunication. They carry the keys to the kingdom, albeit a thoroughly secular one. In a word, health care in the contemporary world is sacramental. (Or if it isn't, nothing else serves as half so plausible a simulation.) To withhold it from some while affording it in abundance to others may seem indecent, to be in flagrant violation of the terms of a minimally acceptable social contract.Footnote 32
The preceding is admittedly speculative. And it may have less applicability to the United States than to countries that place a higher premium on solidarity. It does, though, go some distance toward explaining why even in the allegedly market-based system of health-care delivery in this country, there exist innumerable mandates governing provision which, taken in toto, render American health care expensive, ponderous, and inflexible. The “free market” is not supposed to operate that way!
Health care is almost universally regarded on both the left and the right as what the economists call a “merit good,” that is, an item that will be underprovided by ordinary market mechanisms even under conditions of perfect competition.Footnote 33 Thus the opposition even in the United States to a truly market-based system. (Republican politicians who made hay with derisive excoriations of “Hillarycare” nonetheless presided over the 2003 Medicare prescription drug benefit legislation that added a trillion dollars, give or take, over the next decade to the already massive socialized component of American health-care spending.) This opposition to a market-based system is not, I believe, a function simply of medicine's role in addressing basic human needs. Food and housing are no less essential, yet they do not seem to generate so strong a demand for uniform provision. If the preceding suggestions are on the mark, health-care delivery is special because of its implications for positive association.
V. Conclusion
The policy examinations of the preceding section yield no algorithm for resolving eruptions of the Paradox of Association but nonetheless may be usefully illustrative. Let me offer a tentative summing-up. A protected freedom to dissociate merits wide scope within a liberal order. Successful pursuit of those personal projects that confer value on individual lives requires that individuals be at liberty to disengage from relationships they take to be unpromising so that they may instead seek elsewhere for meaningful activity. Their choices are not to be held hostage either to ancestral precedents or to other people's uncongenial aspirations. This liberty to disengage has to be of wide scope because a pursuit that is central to one person's conception of a life well-lived may be peripheral to the concerns of others. Free exercise of religion doesn't matter a great deal to me; I don't care very much about which house of worship it is that I'm going to be staying away from tomorrow. But millions of other people who take no interest whatsoever in a liberty to generate philosophical scribblings place a life of fidelity to their god at the pinnacle of their valuations. You may not care to mount a Harley Davidson at all, but others find it insufferable not to be allowed to do so sans helmet. There are, according to some respected accounts, even people devoted to the practice of grass counting. It is not within the legitimate purview of state officials to declare which pursuits are sufficiently meritorious to be protected by associational rights and which are not. Rather, those officials are to take their bearings from the decentralized decisions of the populace they serve. An adequate negative freedom of association must, therefore, be capacious.
Positive freedom of association in all but its weakest version binds individuals even when one or more parties wants out of that relationship. It must, therefore, confront a strong presumption in favor of the right of individuals to dissociate. Serious threats to the maintenance of the political order typically overcome that presumption. The conclusion suggested by the preceding discussion is that, under tolerable conditions of political life, nothing else does. Paternalistic concerns for health or morals do not suffice. Nor do vague aspirations toward community, programmatic models for implementing particular patterns of diversity, or a fondness for income equality. Or rather, none of these suffice until and unless they jeopardize the stability of the universal association. Claims derived from positive association, then, are maximally deep but very narrow.
Advocates of the various positions examined in Section III will argue that this is to be insufficiently generous to positive association. Those who are more enamored of the points raised in Section II will claim, to the contrary, that it concedes too much. They will maintain that would-be constrictors of freedom routinely claim to discern some peril to the integrity of the polity. If no direct chain of causation can be adduced, then some stunningly subtle path to possible social breakdown will be discerned. The goal, of course, is to justify intrusions on individuals' capacity to reject the club that will have them for a member.
I concede the point. Of political inventiveness there is no limit. Has anybody succeeded in keeping count of the many and varied justifications of the Iraq incursion proffered by President George W. Bush (or for the previous Iraq war by the previous President Bush)? Even if most such invocations of necessity are bogus, they could not do useful work as pretexts unless the rationale to which they appeal had genuine instances. It is the task of political wisdom, not political philosophy, to distinguish genuine from bogus.
