I. The Public Sphere, the Common Good, and Democratic Education

Public schooling systems across the developed democracies have a decidedly mixed record. Large numbers of children leave school with minimal competency in areas such as basic numeracy and literacy, and the suggestion that public education reduces crime and other social maladies looks doubtful in the face of high levels of youth alienation which have coincided with massive increases in public expenditure on schools. This questionable record has led to increasing calls for an expansion of parental choice and market-like structures in primary and secondary education. The influence of these market liberal ideas has, however, largely met with disapproval from educational philosophers. From the critics’ perspective, reducing democratic control over educational institutions is an inappropriate response to the problems of public school systems that mischaracterizes educational goods and the mechanisms through which they should be delivered.

Arguments of this nature are not attributable to any one philosophical school of thought but have been made by authors endorsing a variety of substantive political doctrines. Most theorists, however, draw on a fundamental critique of the assumptions thought to underlie the case for parental choice. Following John Dewey’s influential account, the suggestion is that citizens need to develop an “enlarged outlook” moving beyond an individualistic understanding of relationships between teacher and pupil, or between teacher and parent, to recognize that what the best and wisest parents want for their own children, the community should also want for all of its children. ^{Footnote 1}

Although these Deweyan arguments have a communitarian flavor, they are also evident in the civic republican and liberal egalitarian traditions. Amy Gutmann, for example, criticizes a model she describes as the “state of families” — a model defended by advocates of parental choice on both natural rights and consequentialist grounds. ^{Footnote 2} The former justification claims that parents have a natural right to educational authority because they have a personal interest in passing on their own values to their offspring, and that a state committed to individual freedom must respect parental choice. ^{Footnote 3} The consequentialist justification meanwhile suggests that parents know best what is in their children’s interests and can be relied upon to make beneficial educational choices. ^{Footnote 4} Gutmann rejects both of these arguments. She suggests that while parents may have a liberal right to educate their children as members of a family, this does not imply a right to insulate children from the broader society and exposure to ways of thinking that conflict with their own. The natural rights argument for parental choice suggests that freedom of the individual is the desired end of education, yet it is precisely this freedom that may be threatened by parents’ educational choices. With respect to the consequentialist argument, Gutmann contends that it is implausible to suggest that parents can be counted on to equip their children with the skills necessary for rational deliberation. Parents may, for example, teach values based on racial or religious intolerance. Thus, “[h]istory suggests that without state provision or regulation of education, children will be taught neither mutual respect among persons nor rational deliberation among ways of life.” ^{Footnote 5}

Harry Brighouse offers some related arguments noting that because children cannot be assumed to be competent choosers, second parties must of necessity make educational choices on their behalf. ^{Footnote 6} To make an effective case for granting exclusive authority to parents in this capacity, he suggests, would require either an assumption that parents are always good proxy consumers for their children’s interests or an argument that parents have “moral ownership” of their offspring. The latter view is incompatible with an emphasis on individual rights because children have rights distinguishable from those of their parents. The former view meanwhile is implausible because many adults are not even effective consumers when acting for themselves. While it may be reasonable to hold adults responsible for their own bad choices,

Seen in this light, citizens have moral obligations to other people’s children to ensure that they receive a decent education, and state payment for education via taxation, and regulation of its provision are the mechanisms by which this collective obligation is fulfilled.

A second set of arguments maintains that the common good requires an active public sphere that enables processes of critical reasoning. The latter contention is witnessed in Elizabeth Anderson’s suggestion that education should not be conceived as a marketable commodity but as a good that reflects “reasoned ideals rather than merely private tastes.” ^{Footnote 8} On this understanding, it is inappropriate to make decisions about moral ideals according to the criterion of willingness to pay. Where incommensurable ideals are concerned, it is impossible to aggregate such values into a utilitarian “social welfare function.” Decisions over educational values should be based instead on publicly reasoned judgments about the relative merits of the goods concerned. A market system would allow parents to exit from schools that do not concur with their private tastes; but the advantage of a publicly controlled system is that it requires people to persuade others that the reasons underlying their educational values are commensurate with the common good. ^{Footnote 9} At the same time, by exposing parents and their children to other people who may exhibit different values and perspectives, the democratic sphere provides the space where preferences can be transformed and where norms that exclude minorities of various kinds can be challenged.

Distributive arguments focused on equality of opportunity constitute the final and perhaps most influential set of complaints against educational markets. On the one hand, the use of the exit option is thought to enable better informed and more energetic parents to leave the children of more apathetic parents in institutions deprived of input from the most sophisticated of their fellow citizens. ^{Footnote 10} On the other hand, such cultural inequalities may be magnified by money power. Children with similar talents and aptitudes may receive unequal levels of provision owing to differences in parental income. Financial disparities also limit the extent to which poorer people can compensate those of their children who are less talented by spending more on their education. ^{Footnote 11} Such inequalities are deemed unacceptable both from the perspective of the disadvantaged and from that of the community at large. For the individual, it is unjust that access to goods that have a determinative effect on life chances should be affected by the caprice of genetics, family background, and income. From the societal perspective meanwhile, the common good requires that those who have talents and abilities that can best secure broader objectives such as economic growth or cultural advancement have access to the resources that enable them to fulfill such potential. ^{Footnote 12}

The above arguments, it should be recognized, do not necessarily preclude some role for markets and parental choice in the provision of educational services. While many of these criticisms are made by authors sympathetic to the “common school” or “comprehensive ideal” — where public authorities allocate children to schools within their local neighborhood ^{Footnote 13} — other supporters of democratic education such as Gutmann and Brighouse are willing to concede a case for a limited form of parental choice provided that the funding of education and the curriculum it delivers is subject to democratic control. What is rejected by all supporters of “democratic education,” however, is the case for “unfettered” markets where the supply, regulation, and the funding of education is left to the decisions of private actors in a context of only minimal regulation focused on contract enforcement and the prevention of fraud.

If unfettered markets are deemed inappropriate to the allocation of educational services, there are two separate though interrelated claims in favor of social democratic alternatives. The first is the instrumental or epistemic argument that democratic control of education produces better decisions. “Better” refers to an enlarged process of public reasoning as the perspectives of children, parents, teachers, and the broader community are reflected in the decisions made. ^{Footnote 14} Public reasoning in this sense refers to the process by which people learn to and help transform one another’s preferences and values through engagement in democratic dialogue. Democratic decisions may also ensure a more equitable distribution of educational opportunities. Though democracy itself may take place in conditions of background inequality, the public nature of these decisions makes it more likely that unjustifiable inequalities will be brought to public attention and subjected to correction. ^{Footnote 15}

The second argument for democratic control is a procedural claim from moral or political legitimacy. This suggests that subjecting one’s decisions to public scrutiny is a requirement of showing the mutual respect toward fellow citizens that is needed to sustain the stability of the social order. Democratic deliberation promotes public legitimacy of social outcomes because all people, regardless of class, cultural background, or ethnic origin can have their say in the decisions that will influence education for the community as a whole. ^{Footnote 16} Institutions having these characteristics are thought to meet a higher-level test of “public reasonability” — because the rules chosen have the potential to work in everyone’s interests and do not unduly reflect the interests of those with greater wealth or social status. Rules of this nature may be accepted as legitimate even in the eyes of those who disagree with or lose out from a particular decision. ^{Footnote 17}

Within this context, most advocates of democratic education are committed to the enforcement of a curriculum that grounds students in the values necessary to sustain the conditions of civility on which an open democratic order depends. This may require both exposing children to a range of political views and cultural and religious traditions, and equipping them with the skills necessary to engage in rational deliberation. Whatever element of choice is permitted, therefore, parents should not be able to deprive their children of the skills necessary both for their own well-being and for their capacity to contribute to democratic deliberation. ^{Footnote 18}

II. The Private Sphere, Public Reasoning, and Social Order

Some of the criticisms sketched above proceed from the assumption that the case for parental choice is neglectful of the social aspect of children’s education and the importance of public reasoning about educational values. Others challenge the case for markets on distributional grounds. It will be my contention, however, that all of these arguments are misguided. Far from limiting public reasoning, the case for parental choice maintains that open educational markets are a precondition of an extended form of public learning that transcends verbal discourse. Distributional criticisms are mistaken because while markets cannot secure “equality of opportunity” they increase the quality of opportunities available to all people in a manner that does not privilege the better-off. In this section I sketch an account of the relationship between the private and the public spheres that forms the background to these claims. The subsequent four sections develop the argument for privatizing primary and secondary educational services and challenge the case for “democratic education.”

The case for private education developed in this paper is concerned with specifying the minimum moral and institutional requirements that enable people who disagree about what ends to pursue in life, and/or the means to achieve such ends, to coexist without permanent and potentially destructive conflict. The basic moral and social rules identified here include toleration, secure possession of property, voluntary contract, and the prevention of violence and fraud. The private sphere secured by these rules enables people to make decisions without the consent of countless others. These negative rights of “noninterference” are judged more likely to secure the minimum moral support needed to sustain the social order because they can be supplied for a modest cost and because the capacity for any actor to pursue personal projects depends on the observance of a private sphere by other actors. ^{Footnote 19} Positive rights to assistance or a particular share of wealth, by contrast, are more costly and may disrupt the conditions for peaceful and mutually beneficial cooperation. Rights of this nature can only be supplied at the direct expense of others — in effect by using some people to secure ends of which they may disapprove. While some positive claims to assistance might be justified in “emergency situations” to preserve social peace, or to care for the indigent, if the conditions for mutual benefit are to be preserved, then it is essential that such claims be minimized. ^{Footnote 20} A distinction may thus be drawn between the minimum social rules that constitute enforceable obligations of justice without which society could not function and “imperfect rights” which should be governed by persuasion and voluntary exchange. ^{Footnote 21} Whereas “perfect rights” to justice must be backed by coercion, in the case of imperfect rights such as economic well-being, charity, and generosity, the use of compulsion may, to use Adam Smith’s words, be judged as “. . . the highest degree of insolence and presumption.” ^{Footnote 22}

It is important to emphasize at this juncture that the case for protecting the private sphere does not seek to hermetically seal people off from social influences but rather to entrench the conditions for public reasoning and social cooperation — properly understood. The observation of human disagreement reflects that people are fallible agents who operate in a world of chronic ignorance. This recognition does not deny the possibility of objective “truths,” but in a context of tightly bounded rationality emphasizes ongoing uncertainty whether an understanding of such truths has in fact been achieved. ^{Footnote 23} Toleration and the protection of the private sphere are thus important because they act as a check against erroneous claims to omniscience. When there is doubt about the rightness of one’s own or others’ beliefs, it is important not to impose any view of the appropriate course to pursue. The existence of an extensive private sphere reduces the risk of large-scale error and increases the possibility for actors in the wider public to learn from different experiences. Within this context, the argument advanced here mirrors its social democratic rival in having both an instrumental or epistemic element and one grounded in the language of public justification and political legitimacy.

With respect to the instrumental/epistemic argument, the thrust of this essay differs from theories that equate public reasoning with the making of collective decisions and that emphasize the importance of conscious deliberation and explicit dialogue in the transmission of values. In a complex and advanced society, it is impossible for people to be consciously aware of the many diverse factors that contribute to the evolution of different ends and means. When people engage with one another in social situations they often produce outcomes that are unintended because they cannot fully anticipate how others will react to the reasons they give and the decisions they make. Properly understood, the public sphere is an “ordered anarchy” whose outcomes are “public” because they emerge as the product of no one’s intention. It is ordered, because it depends on the enforcement of private rights and produces observable patterns of coordination. It is anarchic, however, because the patterns concerned are not the intentional outcomes of the participants and, indeed, these participants lack the cognitive capacity to produce the relevant patterns via conscious or deliberative coordination.

Moreover, the process of public reasoning is not limited to acts of articulate persuasion. One of the most important ways that people can influence the values of others is by lived example. An important advantage of a secure private sphere is the space it creates for people to experiment with modes of living that do not conform to prevailing opinions. It is only when private projects are put into practice that a wider learning process may be set in motion as previously unanticipated benefits can be witnessed by other actors and thus become the subject of emulation. ^{Footnote 24}

This conception of the public sphere as an “ordered anarchy” is best encapsulated by the workings of the competitive market economy. The advantage of such an economy resides not in the rationality of individual actors but in the “super-individual” processes that facilitate order; these processes also enable people to learn, often “unconsciously” from the practices of countless others, while reducing the significance of human error. Markets on this view are fundamentally a part of the public sphere and perform several functions that cannot be replicated by systems that do not rely on persuasion and voluntary exchange.

First, the market economy performs a coordinative role where countless acts of exchange enable both producers and consumers to bid for resources on the basis of knowledge about their own local circumstances. The price signals that emerge from such exchanges enable people to adjust to the shifting choices of unknown agents and to the changing scarcity of goods. Prices, profits, and losses do not provide all the necessary data needed to inform individual actions but they reflect more information and enable more adjustments than would be possible if an organized group were required to determine synoptically which goods and services to produce and in what quantities, for society as a whole. ^{Footnote 25}

Second, markets perform a public learning function where the factors determining which goods are in fact valued, are shaped endogenously by the process of competition. The polycentric nature of decision-making in markets allows opposing ideas in many different production and consumption lines to be tested simultaneously without approval from any one center of control. Markets thus reduce the risk of systemic failure and generate more knowledge about rival ends and means than monocentric alternatives. To perform in this manner there need be no requirement of so-called “perfect competition” where large numbers of price takers adjust their decisions instantaneously. Rather, what is required is that incumbents remain open to challenge from new entrants offering better opportunities than currently available. ^{Footnote 26} The principle obstacle to effective competition on this view is often the existence of price controls and regulatory barriers that stifle the signaling function of prices and raise the costs of market entry to the benefit of incumbents.

The processes described above generate unequal rewards reflecting combinations of skill, luck, and effort, and different starting points for people born into circumstances not of their own making. Markets, therefore, do not generate “equality of opportunity,” but the dynamics of competitive emulation work nonetheless to generate more and better opportunities than might otherwise be available. Redistributive and regulatory measures, by contrast, run the risk of undermining the conditions for social cooperation by requiring that the goals and projects of some are sacrificed in the interests of others. In the worst case scenario this may result in a scramble for control of the political apparatus and an undue focus on transferring resources rather than on the generation of more opportunities — a scramble that is by no means certain to work in the interests of the worst-off. ^{Footnote 27} Even in the most “idealized” scenario, however, where there is agreement on the desirability of redistribution as a goal, there is little reason to suppose that public authorities can access the knowledge to secure an allocation that achieves the desired pattern of opportunities. The knowledge of “who should give what to whom” in order to secure a particular distribution of opportunity can no more be centralized in the hands of a public bureaucracy than can knowledge of the complex and diverse factors that determine the content of prices in a market. ^{Footnote 28}

Markets do, of course, require a background of institutions that secure property rights, and on the view advanced here it is the primary purpose of political authority to maintain such institutions. Disputes over property rights and what the principle of “noninterference” implies in specific cases will always arise and political authorities must make judgments on such cases. Neither does this minimalist account rule out the possible involvement of public authorities in the supply of some collective or public goods which may not for logistical reasons be organized adequately in the private domain. Given the epistemic difficulties facing such authorities in deciding how and in what quantities to produce such goods, however, and the fact that large-scale collective action problems may also plague public provision, ^{Footnote 29} it does suggest that the burden of proof should rest with those who maintain that private supply is impossible or massively inadequate. The resolution of such issues cannot be decided solely on the basis of abstract philosophical principles, but will be based on conventions that reflect historical accidents and time- and place-specific judgments. Moreover, it is important that even where there is a need for coercive authority this authority remains open to challenge. The institutional mechanism emphasized here is that of interjurisdictional competition or political federalism which enables people within the same state to resolve disputes in different ways.

The ethical underpinning of the argument thus far is based on a form of “rule consequentialism” highlighting the long-run benefits to all individuals, irrespective of wealth or social status, from respecting a clear private domain. ^{Footnote 30} The latter increases the possibility for people to learn what their interests might be and how best to achieve them. These arguments are, of course, dependent on certain factual claims about the social world that can be refuted by an accumulation of evidence. The case advanced here, however, goes beyond the merely consequentialist claim that toleration increases the possibility of discovering what may prove to be better courses of action. The private sphere is not just to be favored because it “maximizes the use of reason” but because to abandon toleration is to forsake reason itself. ^{Footnote 31} It is a necessary, though not a sufficient, condition for the development of reason that people have the freedom to think, to speak, and where commensurate with social order, to act in accordance with their beliefs — and that no one’s convictions can be overturned because they are held to be in error by a majority, or those who proffer special claims to authority. ^{Footnote 32} Thus, even where the consequentialist/evidential case is less than clear-cut there remains a general presumption against the use of coercive measures. To abandon persuasion is to show disrespect for the individuals concerned, and measures that are imposed by force rather than accepted and internalized by people are likely to be resisted and thus disruptive of social order. ^{Footnote 33} Only where there is clear evidence that private action risks undermining social cooperation should a resort to coercive measures be made. ^{Footnote 34}

It should be apparent from this account that where there is ongoing disagreement over ends and means, the allocation of goods and services should not be seen as “perfect rights” or as enforceable claims to justice, and that the widest possible scope should be granted to individuals and civil associations to engage in acts of persuasion and voluntary exchange. In the specific context of primary and secondary education, this implies a strong case for leaving the provision, regulation, and funding of educational services to the choices of parents to enter into or exit from relationships with a diversity of suppliers. These may include conventional “schools” providing a set curriculum on a specialist educational site. Alternatively, they may include a diversity of specialist teachers or institutes providing tuition in a particular subject from which parents select their own customized bundle. Education need not therefore consist of “school attendance,” but may involve students in a mix of cross-cutting settings including home-schooling, private tuition, membership of civil associations, and employment-based learning. ^{Footnote 35}

The privileged position granted to parental choice in this model rests on the proposition that no authority or majority has sufficient knowledge to coordinate educational provision on a society-wide basis and to warrant imposing its particular view of how and what children should be taught. Within this context, four themes constitute my critique of “democratic education”: (i) the superiority of market prices in facilitating the coordination of educational means and ends; (ii) the rejection of democratic deliberation and majority rule as the primary mechanism for the transmission of educational ideals and practices; (iii) the incompatibility of equality of opportunity with ethical diversity and educational progress; and (iv) the suggestion that maintaining the background conditions for social order does not depend on the state shaping the character of its citizens. The subsequent sections develop and defend these arguments in greater detail.

III. Market Pricing and the Coordination of Educational Services

Advocates of democratic education maintain that citizens should reason consciously and collectively about the desired form of education for the community at large. From the perspective defended here, however, these requirements are incompatible with public reasoning about educational values because they impose cognitive burdens that are impossible to overcome. Parents and educators, either individually or collectively, cannot be aware of all the factors concerning different children’s talents, personalities, and cultural backgrounds, or the availability of tutoring skills, and the suitability of different teaching methods and curricula. It is difficult enough for fallible parents to be aware of their own children’s interests, let alone those of large numbers of other people. Attempts to secure the “conscious social reproduction” of which Amy Gutmann speaks ^{Footnote 36} are thus likely to fail, not because parents and educators are “too individualistic” or “too selfish,” but because the central problem is to enable adaptation to circumstances and interests of which people cannot be synoptically aware.

Now, this claim might be thought to overstate the case against democratic education since most advocates do not envisage a monocentric regime, arguing for localized participation within universal standards developed by higher tiers of government. Elizabeth Anderson, for example, argues that local democratic procedures need not result in a “bland uniformity of provision” but would allow variation across communities and adaptation to local circumstances and knowledge. ^{Footnote 37} To defend democratic education on these grounds, however, is somewhat to miss the point. Even the “planned economies” of the former Soviet Bloc and pre-1979 communist China devolved responsibility for the implementation of “plans” to individual factories or localized tiers of administration, just as many educational decisions in contemporary Britain and the United States are devolved to schools or elected local authorities. None of this apparent decentralization, however, undermines the critique of such systems advanced here. If decentralization down to localities is to be encouraged on informational grounds, then it is not clear why a more radical decentralization down to individual parents should not be encouraged as well. Moreover, when theorists who prioritize the “common good” over decentralized choice focus on the virtues of localized control, they beg the question of how in the absence of market prices, the educational policies of many different localities are to be adjusted in accordance with that good. It was the lack of a functioning price system that was responsible for the inability of the “planned economies” to secure the common good by coordinating their disparate parts.

A defining feature of democratic education systems is the suppression or elimination of market prices. Instead of the “ordered anarchy” that may emerge when decentralized producers and consumers respond to shifting relative prices in an open market — a process that accounts for the vast increase in modern living standards — public education systems are often characterized by a relatively “disordered anarchy” as people respond to their own experiences in a manner that fails to prompt suitable adjustments from others. As long as people perceive differences in the quality of education across localities and as long as freedom of relocation exists, parents can be expected to move between institutions and jurisdictions. Even “community-minded” parents may do so when they perceive that moving offers a greater chance of securing their desired educational ideals. Yet, without the freedom to bid the price of education up or down, and without profits and losses enabling successful models to expand or to be imitated, and unsuccessful ones to contract and be avoided, imbalances between demand and supply will result from these decentralized choices. One would expect, therefore, the emergence of alternative coordination mechanisms such as queuing systems and shadow markets that lack the more nuanced informational properties of market prices. This is not an inaccurate description of the processes that govern educational access in many advanced democracies where it is common to find waiting lists for oversubscribed schools, and the existence of pronounced real estate market effects where the quality of education is signaled not by the profits of educational suppliers but by the price of homes within close proximity to what are perceived as good schools. ^{Footnote 38} Signals of this kind fail to prompt adjustments from educational suppliers because they do not work to shift resources into the hands of those who have fulfilled a particular market demand. Neither do they provide the same quality of information to consumers because in real estate markets the value of the education offered is bundled up with many other attributes of the neighborhood concerned, relative to a scenario in which parents pay a specific price to a specific educational institution.

Criticisms of this nature also apply to those programs that have moved beyond local democracy as the primary form of decentralization; these programs allow a greater element of parental choice, but forbid or restrict the emergence of market prices. “Open enrollment” procedures where the “money follows the pupil,” for example, do not keep money in the hands of parents, who remain unaware of the price of the tax-financed education they are choosing. Without access to comparative price information, parents cannot determine whether the education their children receive represents value for money. Resources are rarely if ever freed up to expand more successful outlets since incumbent suppliers can use political influence to block the closure of failing institutions. As long as politicians and public officials retain the power to determine which institutions will be eligible for tax funding, the “creative destruction” that results from open competition and the “free entry” of rival educational models is blunted. Similar problems, though less pronounced, arise with voucher schemes where parents are prevented from “topping up” their allotted voucher to bid more for educational services, and where eligibility to be in receipt of vouchers is determined by a process of bureaucratic regulation. The latter constraints are evident even in those voucher programs that do allow “topping up” but where eligibility for public payments is centrally determined. ^{Footnote 39}

Within this context, there is a broad body of evidence to corroborate the view that market pricing and open competition are key factors affecting educational performance across a range of basic indicators such as numeracy, literacy, and language proficiency. Although the absence of a “pure” market education system in any major country makes extrapolating these results conjectural, there is sufficient work to suggest that those models that come closest to the market ideal tend to generate better outcomes. In an analysis of over fifty econometric studies conducted in twenty different nations, Coulson notes that for every finding showing an advantage for public provision there are a further eight findings indicating an advantage for private provision. More significantly for present purposes, in those cases where parents pay part or all of the school fees directly and where there are no or only limited price controls, for every finding that suggests an advantage for public provision there are seventeen findings that show an advantage for private schools. ^{Footnote 40}

Several notable objections to market pricing in education should be addressed at this juncture. The first is that prices fail to reflect the full “social value” of educational services owing to the existence of positive externalities such as greater economic growth as well as the creation of a more engaged and democratically aware population that cannot be captured adequately by the decisions of individuals and families. From Adam Smith to Milton Friedman, this “public good” argument has led not only those who are broadly critical of markets but some who fall within the classical liberal tradition, to advocate at least some minimal role for the state in subsidizing education and/or mandating its provision. ^{Footnote 41} Although it is frequently raised, this particular argument neglects Buchanan and Stubblebine’s crucial distinction between policy-relevant and policy-irrelevant external effects, a distinction that hinges on the issue of infra-marginality. ^{Footnote 42} Infra-marginality applies when externalities exist but where interventions would not produce marginally valued additions to supply. The privately motivated decision to attend fitness classes to maintain the interest of one’s spouse may, for example, generate positive externalities for others who appreciate the beauty of the human form, but it is unlikely that a more satisfactory level of supply could be induced by subsidizing gyms or mandating a specified number of workout hours. Indeed, to intervene in such a way might be to transfer resources away from more marginally-valued uses. ^{Footnote 43} Similarly, it is not unreasonable to suppose that the desire of parents to prepare their offspring for a successful life will generate positive externalities without government inducement and that funding additional units of education may fail to satisfy the interests of both children and parents alike. Parents may, of course, be mistaken in their choice of educational ends and means, but in a context of generally bounded rationality, the possibility of error does not constitute a sufficient case for overriding parental choices and the market prices they might give rise to.

Evidence from nineteenth-century Britain and the United States indicates that in the absence of compulsion and with little or no state funding, a combination of private and philanthropic provision produced close to universal levels of literacy and the basic educational skills relevant to the society of the day. ^{Footnote 44} Indeed, it was exposure to this evidence that led Milton Friedman to abandon his previous support for state funding and compulsion. ^{Footnote 45} Adjusted for the higher income levels of today, there is no evidence that states are uniquely placed to secure the necessary educational standards, and indeed there is some evidence that they perform considerably worse than privately organized alternatives. ^{Footnote 46}

A further objection to unfettered markets suggests that the opacity of information may prevent parents from making effective judgments on the quality of suppliers. In the market for music tuition for example — which approximates the largely unregulated model advocated here — it is easy for educators to stop teaching students they decide will not succeed; it can be hard for parents, therefore, to judge whether the prices charged are reflective of the quality of teaching itself. This is part of a wider complaint suggesting that many attributes of a good education are intangible and unlikely to be captured directly by price. Without independent audit or regulation, the market may not communicate how well students are progressing. ^{Footnote 47} Informational objections of this type are hard to sustain, however, because if discerning educational value is as intractable a problem as these critics suggest, there is no reason to suppose that bureaucratic regulation will perform any better. If parents cannot discern who is best placed to educate their child, then why should educational planners be thought capable of judging the education for whole swathes of children — and why should parents be thought to have the information needed to vote for the politicians who will appoint such regulators in the first place? It cannot be emphasized enough that the case for markets and parental choice does not presuppose anything like “perfect information.” Market prices do not convey all the information that parents need, but they may help nonetheless to inform these decisions in conjunction with other knowledge that cannot easily be made available to a central regulator. Parents may, for example, evaluate a price by talking to their child about the teachers’ attitudes, observing the behavior and attitudes of other students, looking at the type of work that their child completes and how the child is progressing over time — and by talking to other parents directly about their experiences or, increasingly, joining online discussions. This context-specific knowledge can help parents decide whether they are receiving good value or whether to exit to another provider — decisions which will themselves feed back into prices. Parental information may well be imperfect but may still be a more reliable guide than the crude central auditing techniques and “league tables” of exam results that governments regularly use in their assessment of schools.

A somewhat different complaint against laissez faire questions the value of the exit principle itself. According to this view, exit may undermine the relationships between teachers, children, and their parents that are integral to a quality educational experience. While it may be of trivial concern, if parents chop and change where their children get drumming lessons it may be detrimental to a child’s overall education if he or she lacks the long-term relationships with teachers and fellow students derived from a stable learning environment — preferably in close proximity to the child’s home. It is for this reason that local schools will often be in a semi-monopolistic position, since parents may be reluctant to move their children, and economies of scale may prevent more than one or two options in a particular locality. The problem with this criticism however, is that many, if not most, relationships both personal and in the world of employment and commerce may work best if carried out over a longer time scale. While a loving marriage and stable employment matter enormously in many people’s lives, it hardly follows that the option of escaping from the misfortune of having an unloving or tyrannical spouse, or an employer who fails to pay good wages is overvalued. Similarly, while it may be undesirable to have to switch educational institutions, parents and their children are surely better off having this option than when exit is prohibited or severely constrained.

Within this context, there is no reason to believe that an open educational market would result in the monopoly power currently exercised by public schools for the simple reason that compulsory “school attendance” is no part of the model proposed here. While some parents may continue to send their children to “schools,” all manner of alternative nonschool models are likely to proliferate both within and outside the localities concerned. Even in the context of schools, the large-scale institutions that dominate today’s educational landscape are unlikely to be the only models on offer. Research on private education in the developing world reveals multitudes of small-scale suppliers competing for clients at the neighborhood level. ^{Footnote 48} Whether there are economies of scale and how extensive these are cannot be known in advance of open competition. To secure such a process, however, it would also be necessary to eliminate other government interventions — such as land use regulations that might (by accident or design) block new entrants to the market.

A final objection to consider here is the claim that the commodification of education misses the essence of a good, the value of which cannot be measured against a common monetary scale. The case for market-pricing, however, does not imply that the value of education can be so measured. There is no suggestion that “educational enlightenment” is or should be a marketable commodity, but that the means to acquire such enlightenment are not unlimited resources that would be immune from pricing. Money, moreover, is a medium of exchange, not a measure of “social value.” What matters is that decisions on educational expenditures are not issues of scientific truth that can be judged “right” or “wrong” on the basis of reasoned argument. People may be unable to put a price on educational enlightenment, but they are likely nonetheless to disagree about how much to spend on education relative to other ends, and it is these differences that may give rise to prices reflecting what the marginal buyer is willing to give up to secure the good concerned. Markets do not aggregate preferences into a “social welfare function,” for no such function can exist. Rather, they enable mutual adaptations between actors pursuing a diversity of possibly incommensurable goals. Allocation via price, therefore, is fully compatible with the common good where the latter requires a reduction of imbalances between demand and supply to increase the possibility that any one of a range of goals might successfully be achieved. ^{Footnote 49} Indeed, on the view advanced here, a process that restricts market prices is properly seen as contrary to the common good. It is hard to see how depriving parents and educators of information that conveys the opportunity costs of different educational models advances public reasoning. ^{Footnote 50}

IV. Educational Values, Majority Rule, and Nondeliberative Public Reason

The foregoing analysis has emphasized the coordinative role of market prices and the consequences of suppressing them. It is my further contention, however, that by unduly privileging the majoritarian principle democratic education stifles public reasoning about educational ends and means. Within this context, the case for educational privatization does not assume that parents have natural rights of ownership over their children or that they necessarily know what is best for them. Rather, it maintains that under bounded rationality, parental authority results in a greater diversification and dispersal of risk in comparison to authoritarian or majoritarian regimes. Toleration of minority educational practices, moreover, is an essential requirement of a public reasoning process that is not confined to verbal discourse.

Whether at the local or national scale, democratic control of education may afford both too little and too much power to majorities. It may give too little power because beyond decisions made on a highly localized scale (such as a local school board meeting) the vast bulk of the public has very little scope to determine the content of the educational agenda, most of which may be determined by political elites. It may give too much power, however, because once that agenda has been determined, the majority position will be imposed on all against the wishes of the dissenting minority. A first consequence of majoritarianism, therefore, is an increased risk of systemic failure. ^{Footnote 51} If the majority, whether of educational experts or of the electorate at large, errs in its choice of curricula or teacher training methods, then all children will be affected accordingly. In an open educational market by contrast, if a minority of parents prove a better judge of educational ends and means, then their capacity to exit erroneous models may ensure that at least some children receive appropriate skills — and that there are successful educational models that can then be copied by the majority. Correspondingly, if parents in the minority prove mistaken then the consequences, though unfortunate, will be confined to a relatively small number of children.

It is not as though theorists of democratic education are unaware of the value of minority opinions. John Dewey, for example, rejected a populist view of democracy stressing that “. . . all valuable as well as new ideas begin with minorities . . .” and “even a minority of one. . . .” He went on to emphasize that, “opportunity [should] be given to that idea to spread and to become the possession of the multitude.” ^{Footnote 52} Amy Gutmann, meanwhile, does not place so much faith in democratic education as to suggest that all democratic decisions are justified. ^{Footnote 53} She is concerned that democratic majorities once in power will control what is taught in a manner that might unduly repress those in the minority. From Gutmann’s perspective, therefore, it is important that educators have sufficient autonomy from populist pressures to resist majoritarian demands when these threaten the open public space that democracy is supposed to uphold. The mechanisms she proposes include nationally or regionally enforced requirements that schools teach religious toleration, racial equality, and sexual equality, all of which reflect the principles of nonrepression and nondiscrimination.

From the perspective advanced here, however, Dewey and Gutmann’s insistence on democratic decision-making in a context of predominantly state schools deprives minorities of the most important mechanism that can contribute to public reasoning — the freedom to act out their educational opinions — and not just to talk about them. Restricting parental exit in favor of majoritarian institutions limits the total number of decision points and thus the range of educational experiences from which other parents and their children may learn. The latter contention is of particular significance because much of the knowledge pertinent to the communication of educational values may be tacit. Such knowledge cannot be communicated via explicit debate but is only revealed in practice and transmitted via results — even when the reasons underlying these results cannot be articulated verbally. ^{Footnote 54} This process of inarticulate transmission is particularly important in the context of educational markets where less well-informed parents may simply copy the decisions of those who have secured better outcomes for their children in a manner that raises the overall standard. It is insufficient, therefore, to suggest with Dewey that the process of democratic deliberation affords the best prospect for minority opinions to be disseminated by keeping open the possibility of convincing other citizens. ^{Footnote 55} If no new educational models are allowed to proceed without first persuading a majority of other parents and nonparents of their merits — whether at the local or at the national level ^{Footnote 56} — then innovation and the discovery of new and potentially better educational methods and values may be limited.

Now, Amy Gutmann seems partially aware of this argument in conceding a potential role for private educational institutions to serve as “exemplars of new educational practices,” and it is for this reason that she does not favor the abolition of private schools. ^{Footnote 57} If, however, private education performs this public learning function, it is not clear why this potential should not be enhanced by placing all educational decisions in the private sector. Instead of allowing a minority of parents to be involved in educational innovation, a fully private system would enable “multiple minorities” the space to try out educational models including those that do not involve “schooling” in the conventional sense. Gutmann’s argument implies that the need to challenge dominant modes of reasoning justifies only a selective tolerance of private education. Yet, public reasoning cannot be extended to the fullest extent when any one authority or majority has the ability to decide “how much” and “what sort” of innovation to tolerate.

The principle objection to this argument is, of course, that it does indeed give “too much” latitude for parents to experiment in ways that may be inimical to their children’s interests. Parents committed to certain religious or traditional beliefs may, for example, use market freedoms to choose a model that teaches that women and gay people are unequals, and may isolate themselves from contact with the broader community, thus undermining the openness of the public sphere. Others may not even bother to educate their children. The proper role of the democratic state, therefore, is to ensure that the interests of children are secured. Harry Brighouse, for example, supports intervention in the curriculum to secure an “autonomy facilitating” education — an education he believes would not be guaranteed in a free market because groups such as Christian fundamentalists and the Amish deliberately refrain from teaching these skills in order to sustain their religious practices. ^{Footnote 58}

From the perspective defended here, it cannot be denied that the exercise of parental authority will result in the interests of some children being blighted by their parent’s choices. It hardly follows from this, however, that the democratic state should be trusted to secure children’s interests; for error is an inescapable aspect of the human condition. Few people deny that children have interests that should be protected. The difficulty is precisely that in conditions of constrained rationality there is often no agreement on what those interests are or how they are best secured. ^{Footnote 59} On the one hand, it is not obvious that an autonomy-enhancing education should in fact be pursued; some religious groups, for example, do not believe that the pursuit of autonomy advances children’s interests, and prefer instead to emphasize obedience to the word of God. As long as such groupings are not able to monopolize decision-making at the societal level, then educational models based on such beliefs must be seen as contributing to the wider process of public reasoning about educational values.

On the other hand, to invoke notions such as autonomy is to give insufficient attention to the problems that arise when there is disagreement on what an autonomy-facilitating education requires. For Brighouse, all children are owed such an education as a matter of basic justice. Many religious groups, however, do not seek to isolate themselves in order to stifle autonomy but because, in their experience at least, the practices of the surrounding consumerist/permissive culture are antagonistic to autonomy. In effect, Brighouse’s position endorses the authority of those who control the state to impose their own view of what autonomy requires. The assumption here must be that educational planners have privileged knowledge of how best to advance autonomy. Imposing such a view, however, effectively terminates public reasoning by preventing competition between rival views of what is required for an autonomy-enhancing education — or education for any other substantive goal.

Amy Gutmann’s position does not imply the same faith in an omniscient state that Brighouse seems to have since she views participation in democratic argumentation as the mechanism whereby rival interpretations are heard, with the state limited to preventing discrimination against those in the minority while ensuring that the latter do not detach themselves from the public sphere. On this view, it is an illegitimate claim to educational authority on anyone’s part to deprive children of the skills of rational deliberation necessary to choose among good lives. According to Gutmann, in circumstances of democratic participation the political authorities have a “compelling response” to those who might object to the content of a publicly determined curriculum on the grounds that it subverts or conflicts with their own moral values. Thus:

This response, however, is anything but compelling, for it assumes that public reasoning and the ability to “collectively shape” society are dependent on the making of democratically enforceable decisions. On the view advanced here, public reasoning takes its most advanced form not necessarily by discussion, but where there is maximum scope to lead by example. Even where parents seek to isolate themselves from the wider society, they have not departed from the public sphere per se. As long as others remain aware of their existence then they can act as role models or antimodels for educational practices that can be copied or avoided. The purpose of allowing such freedom of exit, therefore, is not simply to let parents impart their own values to their children but to allow the broader community of children and parents alike to benefit from whatever wisdom may be contained therein — even if this cannot be demonstrated via rational argument. In order for this process to occur, it is essential that parents are not forced to live out an overly compromised version of their educational ideals; this is because, for example, having to teach secular values or sexual mores of which they disapprove would stop them from demonstrating the virtue of their own particular standards. ^{Footnote 61} It must be emphasized that this approach does not seek to justify the sacrifice of children’s interests on the altar of social experimentation. Rather, the point is that when there is uncertainty over what any particular child’s interests actually are, it is for the potential benefit of all children that the process of educational discovery be allowed to proceed.

Within this context, strong rights of parental exit may afford those in the minority the greatest scope to influence public reasoning while not being immune to majoritarian influence. Just as majority awareness of the practices of those in the minority preserves the scope for the majority to learn from others, so knowledge of majority practices preserves the scope for those in the minority to learn accordingly. In a society with modern communications it is virtually impossible to seal children off from alternative ways of life and from defecting to such ways at a later date. That influence, however, will be considerably less all-consuming than in a context where nonconformist education is limited by majoritarian legislation. Research on political psychology suggests that even when there are legal prohibitions on discrimination, the necessity to voice what may be unpopular opinions publicly results in “information cascades” that reinforce majority views. ^{Footnote 62} People may fail to reveal their own private standpoints in a public forum because they fear censure or lose confidence in their own ideas. Seen in this light, what theorists such as Gutmann give to those in the minority by prohibiting discrimination against them, they take away by legally limiting the capacity of those in the minority to live out preferred educational ideals and by requiring that they publicly justify these ideals.

One of the strongest arguments for educational markets in the above regard is that they may reduce, though not eliminate, the effect of cultural “lock-ins” because educational entrepreneurs, whether commercially or in civil associations, need not voice their opinions in a democratic forum before starting a new venture. ^{Footnote 63} All that is required is that they persuade sufficient investors, donors, and parents such that the enterprise covers its costs. With lower persuasion costs and the scope for many different minority opinions to be implemented, the possibility to escape from and to challenge prejudices — whether of majorities or those in the minority — via demonstration effects, is thus enhanced. In a free educational market, therefore, while those in the conservative minority may educate their children according to sexist and similar discriminatory values, there would be correspondingly greater scope for other parents who are in the minority to opt for curricula that challenge, say, conventional gender roles, on a far more radical basis than might be approved by any overarching majority.

To restrict rights of exit because some parents may not give sufficient weight to the skills required for rational deliberation is to beg the question of who should have authority to determine what counts as being sufficiently rational. Gutmann, for example, is careful to point out that not all religious education is inimical to the deliberative values she is keen to promote, highlighting in particular the excellent record of private Catholic schools in the United States. ^{Footnote 64} She is willing therefore to allow these schools to have curricular freedoms that enable them to teach values that depart from the majority culture. This is not, however, a privilege she would extend to groups such as the Amish and Christian fundamentalists whose desire to keep their children apart from mainstream social influence is deemed incompatible with the values of democratic deliberation. ^{Footnote 65} What Gutmann thinks counts as being sufficiently open to rational deliberation, remains unclear. More important, it is not obvious why promoting rational deliberation should be afforded privileged status given that most people are in fact nondeliberative learners. Empirical research in cognitive psychology suggests that people are “cognitive misers” who make most of their decisions on the basis of imitation, cultural conventions, and rules of thumb. Given the fallibility of human reason there is always the possibility that “too much” reflection may lead to worse outcomes. Gutmann’s insistence that every child should have the opportunity to “choose rationally and freely among the widest range of lives” ^{Footnote 66} is neglectful of the role that any culture or tradition provides in steering people toward some choices and away from others and that parental attempts to inculcate their traditions are one way of coping with uncertainty. In a wider context of cultural contestation where public reasoning takes place between conservative and nonconservative traditions, although there are no grounds to constrain educational values that emphasize rational criticism and deliberation, neither it would seem is there a strong case for privileging such values.

The argument of this section suggests a strong presumption in favor of parental authority owing to the contribution it can make to a nondeliberative form of public reasoning. This is not to suggest, however, that parents can treat their children in any way they see fit. Parents do not “own” their children, but are better thought of as “stewards” who have fiduciary responsibilities of care toward those in their charge. The abusive treatment of children, therefore, requires action from political authorities in the same way that such authorities are required to resolve disputes and punish transgressions between adults. ^{Footnote 67} What counts as abusive conduct, however, may itself be subject to disagreement. In the same way that countries differ in specifying an age of sexual consent, so there are likely to be legitimate differences as to whether parents should be able, for example, to force ten-year-old children to participate in combat sports such as mixed martial arts. Where no clear-cut answers are available, the argument here implies a strong case that the authority to decide such questions should be fragmented.

V. Inequality, Ethical Diversity, and Educational Progress

With its emphasis on disagreements over educational ends and means, this essay does not endorse the pursuit of egalitarian objectives as a responsibility of basic justice. The first point to emphasize here is that there is potentially no limit to what could be spent in time and resources ensuring that all children — at whatever level of ability or from whatever social background — maximize their educational potential. As Amy Gutmann points out, however, to pursue such limits would deprive parents and children alike of the capacity to follow other worthy ends such as leisure, economic well-being, or the building up of friendships and other attachments. ^{Footnote 68} It is important, therefore, to specify what parents do in fact have an obligation to provide.

On the view endorsed here, parents’ fiduciary responsibilities involve the supply of food, clothing, and housing because the very young cannot secure these goods themselves. Parents should also prepare their children to become self-sustaining actors — though what counts as satisfying this requirement will vary depending on the cultural/economic context concerned. Failure to meet these obligations constitutes an abuse and the political authorities must intervene in these cases. In view of the arguments set out previously, it is essential that such intervention avoids specifying how parents meet any minimum standards such as those pertaining to, for instance, basic literacy and numeracy. Parents in the “no schooling” movement, for example, believe that children’s psychological interests are best served when they do not receive a structured curriculum (even in the home), preferring instead to rely on the child’s natural inquisition and experiences of “self-learning.” In whatever manner they are fulfilled, beyond these fiduciary responsibilities parents need not make sacrifices in terms of time and resources to educate their children if this conflicts with other valuable ends.

The primary responsibility for welfare and educational provision resides with a child’s parents rather than with society at large because the existence of the child in question results from the pursuit of the parents’ ends. ^{Footnote 69} Beyond ensuring that basic obligations of parents to their children are met, there is no requirement that members of the wider public support the education of other peoples’ children. The latter would demand sacrifices of personal ends, whether educational or otherwise, to indefinite numbers of other actors. Mandatory sacrifices of this type are incompatible with the formation of deep attachments to particular persons and particular ends whether cultural or economic, in the context of families and other voluntary groupings.

In light of these comments, the attempt to pursue equality of opportunity via public authority is likely to promote conflict between the potential interests of different children, different parents, and between parents and children. Eliminating the effect of genetic inequalities by concentrating resources on those with the least natural ability ^{Footnote 70} may, for example, require the education of gifted children to be neglected. Indeed, it may be that in order to improve the prospects of the least talented, those with natural ability should not be educated at all — an approach that would likely damage relationships between parents and their more gifted offspring. Similar problems face “meritocratic” approaches to equality ^{Footnote 71} which require that the most talented receive the same education irrespective of parental income or cultural background. As well as stopping wealthier parents from focusing on the educational excellences of their more able progeny, such approaches would also limit the scope of parents to compensate their less able children if public authorities deem that resources would be better spent on gifted children in poorer homes. The meritocratic approach might also involve stopping highly educated parents from imparting knowledge in the home because such advantages might not be available to those in less cultured surroundings. In view of the central role that family background plays in shaping children’s lives, it would be necessary to minimize the time that children spend with their parents in a manner that may directly conflict with the interests that both children and their parents may have in maintaining intimate family relationships. There may be some children in poorer families (and some in richer families) who would benefit from such a move, but outside cases of child abuse it is difficult to see how public authorities are to determine in any particular case whether the gains to the child in terms of improved educational opportunity (if there are any) would outweigh the psychological and other losses that might follow separation from their parents.

While inequalities resulting from differences in talents and parental income/cultural background are not “deserved,” on the view advanced here there is no basis for agreement on how these inequalities can be “corrected” in a manner that balances the conflicting interests of different children and different parents. There is every reason to endorse the diverse and decentralized efforts of parents and civil associations in making their own attempts to improve the lot of those who are disadvantaged, by, for example, providing educational bursaries or spending time with the worst-off; but these attempts cannot be required as a matter of basic justice.

Instead of seeking to equalize opportunities, the framework supported in this essay aims to improve the total number and quality of opportunities available across different income groups and cultural traditions. In a competitive educational setting, one need not fear that opportunities for the worst-off will be confined to the low-quality offerings equivalent to what is served up at, say, the cheapest fast-food joints. The dynamic effects of open competition would see a proliferation of educational models expanding the range of options across the income distribution in a manner that can reduce the significance, though not the existence, of inequalities. There are, for example, many price levels for refrigerators catering to different income groups; but an expensive refrigerator today offers little more in terms of functionality than an “ordinary” model — compared to a time when the gulf between the refrigerators of the rich and the ice boxes of the poor was huge. ^{Footnote 72} Even in a “service centered” context such as education, one would expect similar competitive dynamics and technological innovations to spread opportunities that improve the functionality of education for children from lower-income backgrounds at whatever level of ability. These innovations might expand the supply of people who are able to perform an educational role beyond a narrow band of professionalized “teachers.” Moreover, while lower-income parents might not be able to afford direct personalized access to the most sought after educators, new innovations such as online learning may enable greater access to the work of such tutors at a price that the vast majority of parents can in fact afford.

The above arguments are, of course, unlikely to cut any ice with radical egalitarians. Harry Brighouse, for example, suggests that “bottomless pit” objections to strict equality of opportunity allow too much room for children’s prospects of achieving materially rewarding and status-conferring jobs to be affected by factors beyond their control. ^{Footnote 73} Similarly, against Loren Lomasky’s suggestion that educational equality elevates collectivist goals above those of the family, Brighouse maintains that fair equality of opportunity is a principle that identifies goods that are valuable whatever one’s ends and to which all people should have equal access. ^{Footnote 74} Yet privileging equality as Brighouse does requires agreement on a moral end where no agreement is likely to exist and, thus, requires interference with the ends of individuals. It may, for example, imply that parents who value hard work and enterprise should not encourage their children to follow their example for fear that this might disadvantage those from less industrious homes in the competition for jobs. Similarly, it may demand that parents who believe that an antimaterialist education best serves their children’s interests must submit to a regime that places a premium on economic success. Although it is a somewhat extreme case, the forced attendance of Native American children at boarding schools which forbade the speaking of native languages and sought to eliminate the attachments of children to their parents and to tribal cultures, illustrates the destructive interference in family life that the institutionalized pursuit of equal opportunity may lead to. ^{Footnote 75}

More important, in the event that there is in fact agreement on educational ends, the pursuit of equal opportunity may be in direct conflict with securing children the resources Brighouse believes they are due. Providing “too many” resources to the disabled or untalented to compensate for their lack of ability, or to those from the poorest families to compensate for their poverty may, for example, lower economic growth and hence result in a lesser absolute standard of opportunity for the disadvantaged. In such circumstances, Brighouse fears that equality of opportunity might not have privileged status in a theory of justice. This is not a position he is willing to accept, but neither is it one he is able to reject. ^{Footnote 76} He maintains instead that there are few circumstances where such trade-offs might be relevant. Inequalities can only be justified if they raise the absolute level of opportunity for the least advantaged by, for example, developing the potential of those with better opportunities in ways that contribute to the lives of others. In Brighouse’s opinion, there is no way educational inequalities could so operate, and even if they could, there would be no association between such a process and differences in parental wealth.

There is, however, good reason to question Brighouse’s assertions on this matter. Richer parents may, for example, afford an education that utilizes the latest learning technologies, technologies that may be far too expensive to provide for all equally talented children simultaneously because the first units are always the most expensive to produce. The actions of such parents may benefit less advantaged children nonetheless. They may benefit them, first, by covering development costs and, thus, increasing subsequent availability, and second, by giving their own offspring a chance to secure jobs making use of this experience — jobs which may increase opportunities available to others over time. If no children are allowed access to the latest technology until all are able to do so, then educational progress will be halted. ^{Footnote 77} Brighouse, therefore, may need to choose between “leveling down” in the name of equal opportunity, and abandoning the concept in favor of unequal opportunities that work to everyone’s interest.

There is, though, a much deeper objection to the idea that a state-regulated and funded system can secure equality of opportunity. The assumption that children with equal talents should receive roughly the same level of educational resources or that those lacking in talents should be compensated in some way presumes that knowledge of what the relevant talents are can be “given” to educational planners who are to distribute the relevant opportunities in advance of the process of social interaction. In a context of tightly bounded rationality, however, this assumption is invalid. Although talents may be genetically determined, their existence is in many instances ^{Footnote 78} only revealed as children are exposed to different parenting methods and educational models. If parents are deprived of the option of experimenting with alternative models, by for example spending more on their children’s education, then this may prevent them from revealing “unknown talents” that their children might in fact have —discoveries that could improve the opportunities available to others and that could prompt the search for similar abilities across a wider pool of children. Any process that allows for new discoveries and the transmission of knowledge about educational possibilities in this way requires that some have advanced “unfairly” ahead of others.

Insofar as he addresses objections of this kind, Brighouse suggests that the primary difference between egalitarianism and the view advanced here centers on their relative degrees of optimism. For Brighouse, the democratic state is:

Libertarians must, he suggests, exhibit at least some such optimism about the possibility of enforcing their own institutional ideal of a state largely confined to the enforcement of private property relations. This statement does indeed reveal a crucial difference between egalitarianism and the argument of this essay, reflected in the distinction between perfect and imperfect rights. It is precisely because the democratic state lacks the knowledge necessary to determine what educational justice consists of that access to educational goods cannot be guaranteed as a “perfect right.” Discerning whether people are abiding by private contracts or committing acts of violence, by contrast, is not only a necessary condition for any kind of social order but is a condition that requires nowhere near the same “information-gathering, processing and enforcement powers.” In a context of limited knowledge and constrained rationality where there are differences of view about the moral status of educational ends, and where knowledge of what is to be equalized and in what manner cannot be “given” to a center of political control, there is no way that such a principle can be “guaranteed” as a matter of basic justice.

If strict equality of opportunity is unworkable, theorists such as Gutmann opt instead for a weaker egalitarianism, allowing for inequalities owing to cultural or economic background and natural abilities, provided that all children are guaranteed a sufficient standard. Above this standard it should be possible for some localities to spend more on talented children or less-talented children than elsewhere; and provided it is approved democratically, richer parents should be able to spend more accordingly. With respect to the standard itself, Gutmann argues that all children must receive an education enabling them to participate effectively in democratic decisions. This is because it is the democratic process (both locally and nationally) that must determine the trade-offs between educational spending and conflicting objectives. In order to participate effectively in this process, the state must ensure that everyone is guaranteed the relevant threshold.

From the standpoint of this essay, what is surprising about these suggestions is: (a) the emphasis Gutmann places on the capacity for effective democratic participation as the basis for her sufficiency threshold; (b) her suggestion that only state-funded or regulated education can secure the relevant threshold; and (c) the notion that democratic processes rather than private decisions should determine trade-offs over and above whatever threshold is appropriate.

On the first point, Gutmann sets the bar at levels that are, to put it mildly, difficult for the vast majority of people to reach. On her view of democracy, it is not enough that voters have sufficient knowledge to remove leaders they consider unsatisfactory — people must also be able to discuss policy issues in a rigorous way and meet appropriate standards of rational deliberation. ^{Footnote 80} It is not clear what is egalitarian about these requirements; for to emphasize deliberative skills may disadvantage those who lack the aptitudes necessary to engage in the articulate persuasion of majorities. The informational requirements of Gutmann’s threshold, moreover, are well beyond what most people — including the well educated — are ever likely to meet. Even a cursory glance at voters’ knowledge of democratic processes shows they come nowhere near meeting these standards. Evidence from Europe and the United States suggests that most voters do not know the name of their representatives let alone the fine print of legislation. In 2002, over sixty percent of U.S. voters thought Marx’s dictum “from each according to his ability to each according to his needs” was in the U.S. Constitution or were not sure if it was. In 2006, meanwhile, a survey revealed that only twenty-eight percent of U.S. voters could name two or more of the five rights guaranteed by the Constitution. More recently, only twenty-four percent of Americans realized that “cap and trade” legislation was addressed toward climate change! ^{Footnote 81}

This lack of “democratic knowledge” has been a persistent characteristic of voters in democratic states notwithstanding massive increases in expenditure on public education programs. On the second of the above points, therefore, state-funded education clearly offers no guarantee that children will secure the skills that Gutmann deems necessary. This is, however, an unfair indictment of state education, for there are grounds to believe that no education system, either private or public, could ever meet these standards. The reason is that democracy itself provides few incentives to become adequately informed about the content of democratic decisions. In a large number electorate, the chance that any voter’s decision to acquire more information will have an impact on the outcome for himself or his fellow citizens is vanishingly small. And, even in relatively smaller jurisdictions, the odds are not sufficiently better. Ignorance of democratic processes and policies is not necessarily the result of apathy, stupidity or lack of public spending, but a reflection of the basic incentives with which all democracies confront their citizens. ^{Footnote 82} Moreover, as shown earlier, even if people did have incentives to acquire more information, in a complex society there is a large body of knowledge pertinent to socioeconomic decisions that is beyond the cognitive grasp of even the most sophisticated actors. ^{Footnote 83} It is questionable, therefore, whether the skills that Gutmann values are of significant relevance to the way that most people live their lives.

On the third of the above points, many of the trade-offs Gutmann recognizes between educational equality and other objectives may be best left to the diverse decisions of individuals and voluntary associations. When people disagree on how much to compensate different children for lack of talent or family background, there is no warrant to impose the view of the majority. As emphasized throughout this essay, majoritarian decisions stifle the processes of public learning that occur when people can observe the results of a polycentric process. To abandon this principle, moreover, is to move away from a position where both rich and poor exercise a decisive effect over the allocation of educational resources—although the rich have a wider range of options—to a situation in which the vast majority of people exercise little or no influence over the choices made in their name. Under democratic education, greater equality between rich and poor goes hand in hand with greater inequality between people in general and a limited number of political representatives who wield more power in shaping children’s opportunities than would otherwise be exercised by anybody.

It might be argued, of course, that even if Gutmann’s sufficiency standard is unworkable, this does not undermine the case for an alternative minimum, such as one ensuring universal access to basic literacy and numeracy that needs to be guaranteed by the state. On the view advanced here, if parents are unable to fulfill their fiduciary responsibilities to their children, and if philanthropic support fails to fill this gap, then the possibility of public assistance cannot be ruled out. Insofar as such a case can be made, however, it is likely to depend on time- and place-specific judgments rather than on any generalized “institutional guarantee,” for it may be that even basic literacy and numeracy cannot be secured for all children as a “perfect right.” Securing such a right for children on a global scale, for example, may entail opportunity costs in terms of lost economic growth in both richer and poorer countries ^{Footnote 84} and potentially unwelcome interventions in the cultural practices of people in the developing world that could damage the prospects of the very children who have the least opportunities. Similarly, if anything approaching a regime that enables migrants from less advantaged countries to access economic opportunities in the developed world is to be maintained, there may be little scope for agreeing on how many children are to be eligible to receive the relevant threshold. If the effect of having a threshold is that some of the most disadvantaged children in the world are excluded from a country in order to make the threshold affordable, then it may be to the benefit of these children that no such threshold is provided.

VI. Toleration, Markets, and Civic Education

Against all of these arguments, it might still be maintained that radical privatization is untenable because it lacks any formal mechanism to ensure that people will in fact exhibit toleration. Gutmann makes this argument as part of her case for a compulsory civic education. The case against exclusive parental authority, she suggests, depends neither upon parental ignorance nor upon irrationality but on the claim that toleration is a public good that will not be supplied voluntarily. From the perspective of individual parents seeking to maintain their own religion, it may be rational to teach disrespect of other religions. As long as one’s own faith is respected by others, one need not worry about teaching one’s children the value of respect for rival beliefs. And, even if one’s own religion is disrespected, parents have insufficient reason to address this problem by teaching respect for the disrespectful. Citing nineteenth-century U.S. public schools where the dominance of the Protestant majority resulted in the brutal mistreatment of Catholic children, Gutmann invites her readers to ponder what would have happened had schooling then been privatized. Protestants, she suggests, would have continued to send their children to Protestant schools, which would have continued to heap scorn on Catholicism, whereas Catholics would have escaped to their own ghetto schools. As a consequence,

For Gutmann, the fact that Catholics in America today need not worry about such intolerance is testament to the “moral capital” built up by the U.S. public school system and its democratic ethos over the last hundred years. This is part of the broader argument made by supporters of democratic education which suggests that the making of collective, democratic decisions is a way for people both to acquire the values of respect and toleration needed to sustain an open public sphere and to show such respect by having decisions over matters such as educational expenditure determined democratically. Democracy on this view encourages development and internalization of the appropriate civil norms. By contrast, a privatized model of education is a modus vivendi which at best represents a mere compromise between advocates of rival standpoints, a compromise likely to break down as soon as one side or another achieves the power and resources to impose its own will.

In a context of radically unequal bargaining power, there can be little doubt that the interests of minorities and others lacking such power are liable to be abused. There are, however, few reasons to believe that democracy provides the best defense against such abuses, or that democratic processes can generate the norms needed to avoid abuses where power imbalances arise. On the view advanced here, toleration does not require that one offer respect toward religions or practices that one disapproves of, but merely that dissenters are allowed to “go their own way.” Gutmann’s account of the horrors inflicted upon Catholic children in U.S. public schools is instructive in this regard. Educational privatization here would have at least enabled Catholic parents to avoid having their children “whipped for refusing to read the King James Bible,” ^{Footnote 86} by choosing a Catholic school, or withdrawing from formal education entirely. It is strange to say the least that Gutmann, who elsewhere puts considerable store on the need for the state to protect children from their parents, would in this instance have been willing to sacrifice Catholic children’s physical security in the hope that the exercise of democratic voice might generate “moral capital” in the longer term. In these circumstances the value of the exit principle is surely clear—relative to democratic alternatives, it provides weaker parties with at least some scope to avoid unwanted decisions being imposed upon them.

Now, of course, it might be argued that the exit option offers only a temporary “solution” here and that it fails to resolve the “public good” issue in building the conditions for an open public sphere. It is, however, far from clear that norms of toleration are in fact “public goods” and that insofar as they might be, democratizing decisions may not be an appropriate route to securing them.

Gutmann assumes that the changing attitude to Catholics in U.S. public life has been an effect of democratic public schooling. There is, however, an alternative explanation for this shift that points toward the socializing forces arising from the dominant role played by commercial relationships in U.S. society. In such a society, where most people must acquire their living by voluntary exchange, there are considerable personal benefits to people from exercising the values of critical open-mindedness that supporters of democratic education wish to instill. Put simply, holding to prejudicial attitudes in markets has a direct cost to the actor concerned because it reduces the supply of potential partners to exchange. The employer who refuses to hire people owing to their race, religion, or sexual orientation, and the consumer who refuses to buy products made with “minority labor” will typically have to pay higher prices relative to more open-minded market participants. It is precisely because relationships in markets are “weak” and do not depend on a wider agreement on ethical or religious norms beyond a willingness to observe contracts, that people are more likely to tolerate those from other cultural and religious backgrounds in search for a better deal. Anthropological research suggests that those from societies with higher levels of market integration are more likely to exhibit cooperative social norms than those with relatively lesser contacts to the commercial world. ^{Footnote 87} Though tolerance may be adopted initially for prudential reasons, in a market-driven economy where these incentives are omnipresent there may be a pronounced tendency for people to internalize the relevant norms and to act in accordance with them, even where such commercial imperatives are lacking. It is, therefore, at least as plausible to suggest that the greater tolerance of religious and other minorities witnessed today has arisen owing to the “civilizing powers of commerce” as it is to attribute these advances to democratic education.

People are not fully rational actors so the tendency highlighted above will undoubtedly be an imperfect one. In an uncertain world, holding to habits and traditions that have “worked” for people in the past has an important function, and these habits may be slow to change. The educational model endorsed here would allow those who wish to isolate themselves and their children from the world of commerce the liberty to do so, and would enable those holding to traditions that have exclusionary tendencies to find niches within markets where they can maintain their practices. Nonetheless, wherever commerce is the primary ordering force, one would expect large numbers of people to incline toward liberal norms and to educate their children accordingly, just as market imperatives tend to encourage a common language of trade to reduce transaction costs. Indeed, far from inhibiting this process, the commercialization of education may reinforce such values. Where parents must pay directly for educational services, those minorities who wish to educate their children in isolation from others must bear the full price for doing so, either by teaching their children themselves, or by paying higher fees for what may be a relatively limited supply of educators who share their particular prejudices.

Of course, in any society where the majority are themselves intolerant, the reception granted to “education for tolerance” may be a hostile one, and the fate of parents and children in the minority may be bleak. Yet there are reasons to believe that democratic processes provide fewer incentives to undermine such intolerance than do markets. In the latter, provided that the political authorities secure the protection of person and property and that they prohibit incitements to violence, whether in the classroom or elsewhere, then minorities of more open-minded entrepreneurs and consumers can profit personally from challenging majority prejudices because doing so can impact decisively their own well-being by improving the quality of a service they receive or reducing its price. In a collective choice context, however, the incentive to challenge dominant views may be relatively weaker because the costs of holding on to the majority’s prejudices are spread across society as a whole, and the chance that any personal contribution to public debate may determine its result is vanishingly small. ^{Footnote 88} Insofar as tolerance has public-good attributes, contra Gutmann’s assertions, these may be more pronounced in a democratic setting than in markets.

There may, therefore, be nothing about a specifically deliberative democratic form of decision-making that better equips it to reduce intolerance. Indeed, requiring that parents justify their educational values in a public forum may be as likely to promote disrespect between people as to promote fellow-feeling. The more parents come to learn about the value systems of their children’s classmates and their parents, the more they may disrespect them, and the more they may seek to control the curriculum in order to impose their own particular vision. “Agreeing to disagree” and not deciding educational matters on a collective, democratic basis is an alternative way of promoting greater openness. Whether this modus vivendi is a viable way of maintaining an open public sphere is an empirical matter that cannot be determined here. Nonetheless, there are important findings that suggest that this stance is not entirely implausible. Recent studies of democratic participation indicate that the tolerance of others necessary to sustain public reasoning is more prevalent among those who avoid active politics. The more tolerant people are of other viewpoints the less likely they are to want to engage in activity that results in coercive decisions. By contrast, the high levels of political knowledge and active participation deemed essential by enthusiasts for democratic education tend to correlate with the most highly partisan and intolerant, and those who report minimal social contact with people from alternative standpoints. ^{Footnote 89} If this analysis is accurate, then it implies that limiting the scope of democratic decision-making may be necessary to constrain the intolerant. Whatever the doubts about the self-sustaining potential of educational laissez faire, there seem comparable concerns about social democratic alternatives.

VII. Conclusion

I have sought in this essay to make a case for the privatization of primary and secondary educational services. I have not suggested that parents have natural ownership rights over their offspring; nor have I claimed that parents know what is in their children’s interests. Rather, I have argued that where educational ends and means are in dispute the exercise of parental authority may offer the greatest scope for an extended form of public reasoning. To those interested in the enforcement of “educational guarantees,” this framework will doubtless fail to inspire because it allows significant scope for parents to make choices that may damage their children’s interests. In a world of constrained rationality, however, where error is inevitable, and where parents, educators, and voters alike are often “slow learners,” there are few grounds to suppose that “democratic education” will limit such failures, and good reasons to believe that it will stifle the discovery of what may prove to be better educational ideals and practices than those of which we are currently aware.