Jump to ContentJump to Main Navigation
The Demands of Liberal Education$

Meira Levinson

Print publication date: 2002

Print ISBN-13: 9780199250448

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0199250448.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 29 January 2020

Modifying the Liberal Educational Ideal

Modifying the Liberal Educational Ideal

(p.64) 3 Modifying the Liberal Educational Ideal
The Demands of Liberal Education

Meira Levinson (Contributor Webpage)

Oxford University Press

Abstract and Keywords

Examines four objections stemming from the clash of theory (that liberalism requires the institution of compulsory, autonomy‐driven schools regardless of parental preferences) and intuition (that liberalism entails democratic control over schools and sensitivity to parents’ preferences). It argues that the ‘detached school’ should, with minor modifications, continue to provide the basis for the liberal educational ideal. Section 3.1 addresses concerns about state tyranny, arguing that the detached school both counters the threat of parental tyranny and ensures a substantive pluralism among schools and within society. Section 3.2 shows that detached schools can promote effective parental involvement. Section 3.3 addresses the hidden curriculum of schools, while Section 3.4 discusses the ability of the detached school to help children develop cultural coherence and a stable sense of identity as well as a capacity for choice.

Keywords:   choice, cultural coherence, democratic control, hidden curriculum, identity, liberal education, parental involvement, pluralism, schools, tyranny

In Chapter 1, we saw that the most plausible reading of contemporary liberalism entails a weakly perfectionist commitment to individual autonomy. It is this commitment to the value of autonomy, I argued, that grounds liberalism's commitments to pluralism, the legitimation process, and many of the substantive rights, duties, and institutions commonly associated with contemporary liberal theory. Chapter 2 applied this analysis of liberalism as a weakly perfectionist theory to children, concluding that the liberal commitment to enabling the exercise of autonomy by adults implies a parallel commitment to developing the capacity for autonomy in children, and that this commitment should take the form in part of state‐regulated schooling. After clarifying what the development of autonomy requires, section 2.4 spun out a picture of what the liberal, autonomy‐promoting school would look like—and generated a surprising ideal. The ideal liberal school, I argued there, is an autonomy‐driven, plural community that is explicitly detached from children's families and home communities and is not necessarily designed to reflect the values and commitments held by parents. As a state‐regulated community governed by a determinate, primary aim—that of helping children to develop their capacity for autonomy—and protected from any parental influence that inhibits the development of autonomy, the liberal school is meant to ensure the freedom of all children ultimately to determine their own path in the world just as adults are free under the liberal state to chart theirs.

This ideal of the detached, autonomy‐driven school contrasts radically, even violently, with many liberals' intuitive conceptions of how liberal educational institutions should be shaped. Nowhere to be found are the standard, allegedly liberal notions of democratic control over schools by all members of the community, of the establishment of a wide variety of schools reflecting a range of conceptions of the good, or of the school as responsive to parental, not state, concerns. In this chapter, I examine four objections stemming from this clash of theory and intuition, and argue that the conception of the ‘detached school’ should, with minor modifications, continue to provide the basis for the liberal educational ideal.

(p.65) 3.1 State versus Parent Tyranny: Or, the ‘Liberalism of Fear’ Meets Liberal Education

Modifying the Liberal Educational Ideal State versus Parent Tyranny

Somewhere along the way, many liberals (and readers!) might be motivated to argue, the liberal commitment to autonomy for adults has been manipulated—even betrayed—in its application to children. The principle of freedom embodied in the ideal of autonomy, these liberals would charge, has been replaced with the practice of state tyranny. It is this manifestation of liberal principles, in the form of the ‘liberalism of fear’ and the criticisms of the liberal educational ideal that it inspires, that forms the core of this section. Specifically, I consider two potential forms of oppression: first, the threat of state tyranny in general, and second, the threat of the ‘tyranny of sameness’. My intent is that in the process of countering these criticisms, we will come to see more clearly the internal robustness of the model of liberal education presented in section 2.4. In other words, the very reasons that the liberal educational ideal stands up to the liberalism of fear make the ideal desirable in and of itself.

I suggest that this liberal educational ideal has two major strengths that make it superior to more democratic and/or parent‐centered models of liberal education. First, the detached school balances out a current maldistribution of power that unjustly favors parents. As opposed to imposing state tyranny, the detached school actually establishes a needed counter‐weight to the threat of parental tyranny. Second, the liberal educational ideal ensures a substantive as opposed to formal pluralism both among schools and within society as a whole. Schools can exhibit significant variation within the boundaries of educating for autonomy, thus permitting them to adjust to the specific needs or interests of their student populations. At the same time, I show that their emphasis on common education, mutual toleration, and critical engagement ensures the development and maintenance of an interactive and mutually responsive plural society, as opposed to a society composed of separate, insular, and self‐protective communities which, while formally part of a diverse and multicultural whole, are internally homogeneous and disengaged from other groups in their midst. But first, let us turn to the primary objection concerning the liberalism of fear and the threat of state tyranny.

Liberalism of Fear I—The Threat of State Tyranny

The fear of state tyranny, Judith Shklar reminds us, is most fundamentally a fear of the unchecked physical power of the state: ‘while the sources of (p.66) social oppression are indeed numerous, none has the deadly effect of those who, as the agents of the modern state, have unique resources of physical might and persuasion at their disposal’.1 This statement, however, raises the question of why we are afraid of physical might and persuasion. What sort of ‘oppression’ does the power of the state threaten us with? I suggest that liberals are afraid of two different forms of state oppression that are causally linked but conceptually distinct. The primary fear is of being coerced by the state into leading a bad, unhappy, and/or painful life—the meaning of which will be clarified below. The secondary, more moderate fear (but still forceful, as Mill makes clear) is of being forced to live the same life as everyone else—i.e., to live a life which may be good, and with which one does identify, but whose similarity to others' lives entails an unacceptable loss of plurality. I address the more threatening form of state oppression in this subsection, and turn to the ‘tyranny of sameness’ afterward.

The paradigmatic threat of state tyranny is that of a state which exerts its power to force citizens to lead bad lives. Given this level of generality, all people would agree with this statement; nobody would dispute that being forced to lead a bad life is undesirable, oppressive, and ultimately tyrannical. But of course, what a ‘bad life’ means depends on the context of one's beliefs. From a liberal perspective, as I argued in Chapter 1, a sufficient condition of being forced by the state to live a bad life is to be forced to live a life with which one does not freely identify.2 The principal threat of state tyranny amounts to the threat that, given undue power, the state will force individuals to live in a way which they would not autonomously choose.3 (This statement should not be taken to belittle more fundamental concerns about avoiding torture, fear of state brutality, etc. It is assumed that one would not freely choose to live a physically or mentally brutalized life, and therefore that the free identification condition incorporates these concerns.)

It is important to understand that the state need not intend to force its citizens to lead bad lives in order to be counted as tyrannical. While maliciousness or indifference toward individuals' autonomy may be one cause of state tyranny (because, for example, it is to the advantage of the rulers to coerce the population to lead a certain kind of life), the state may also force individuals to lead bad lives simply through incompetence or myopia. As liberals are well aware, it is not inconceivable in an imperfect world that the state will think that it is establishing the conditions for free, autonomous choice—i.e., that it will operate with the best of liberal intentions—but none the less in practice force individuals into lives with which they do not freely identify. The goal for liberalism, therefore, is to establish a distribution of power among the state and other actors that ensures both that the state has sufficient power to establish and maintain the political conditions (p.67) necessary for the exercise and development of individual freedom (as liberalism requires), and that ‘the people’ (and/or other actors and institutions) have sufficient power to check the state when it shows signs of violating or overstepping these conditions, whether the violation be intentional or unintentional. Of course one would also want to structure the state itself to be self‐checking through means such as judicial review, balance of powers, etc. But one of the points of the liberalism of fear is that no state, no matter how carefully constructed, can be trusted not to be tyrannical. Given the massive power that is available to even the most liberal of states, even those states which are most self‐regulating in form can still inflict massive pain and hardship on their members in practice. Thus, one must establish checks on the state's power from outside government as well as from within.

This goal of avoiding state tyranny through the checks and balances of power seems to be challenged—if not overtly violated—by the ideal of liberal education set forth in section 2.4. By intentionally shielding the state‐regulated school from the influence of parents and local communities, it may be argued, the (so‐called) liberal ideal fails to provide one important check on state power of the sort called for in order to avoid tyranny. Under this analysis, the fact that the school is explicitly conceived of as being a means of promoting children's development of autonomy is not a sufficient defense, because what matters is not what the aim of education is said to be, but who controls what education actually is on the ground. In this sense, the argument about tyranny reverses the hierarchy established in Chapter 2 between children's a priori indifference to who controls their development and their supreme a priori interest in what kind of control is exercised. Chapter 2 argued that from a liberal perspective, children's interests are best fulfilled by providing them an education that aims at developing their capacity for autonomy, and that control over education should therefore be given to whoever best fulfills (or advances) this goal. According to this approach, who controls education is determined by what the aim of education should be (which itself is determined by what children's a priori interests are thought to be). Insofar as the state‐regulated, detached school is in principle best equipped to achieve education for autonomy, therefore, the state and not parents should exercise primary control over schooling. The aim of educating for autonomy dictates that the identity of the educators be the state and not the parents.

If one shifts attention from the demands of inculcating the positive liberty of autonomy, however, to the lexically prior requirements of preserving negative liberty in the form of preventing tyranny, then a very different relationship between who controls schooling and what schooling aims at emerges. For from this perspective, giving the state primary control (p.68) over education is in itself potentially tyrannical, because the power of the state is intrinsically tyrannical if left unchecked. It is both naive and dangerous, this argument goes, to think that one can first dictate the aim of education and then assign the state to implement the aim, for the power of the state is so vast that no pre‐established ‘aim’—even that of educating for autonomy—can check its actions or educative outcomes. Rather, the fact of the state's control over education will inevitably redefine what the aim of education turns out to be.4

This objection carries weight whether one views the state as inevitably malicious and power‐hungry, as Shklar does, or simply as occasionally prone to incompetence and mistakes. A state may begin by being liberal and therefore be accorded primary power over education, but over time reject (if power‐hungry) or misconceive (if myopic) what liberal education is about and thus cease to educate liberally. In either case, Chapter 2's call to detach schools from local and parental control ends up threatening the very preservation of liberal freedoms and institutions, and thus the good of its citizens, because parents, local community members, and children themselves are unable to check the state's power over the education of future generations. Again, the fact that the aim of education—to develop in children their capacity for autonomy—has been carefully spelled out, and even that autonomy itself has been rigorously defined (see sections 1.3 and 2.4), is irrelevant in the face of the power of the state to change definitions, and to alter aims. One‐off definitions are not enough, this argument concludes; one needs assurances of stability, continuity, and continuing checks on state power.5

This argument is right as far as it goes. Liberalism's primary political aim is to remove the shackles of oppression and tyranny from people's lives, and this does amount in part to ensuring that no institution has a monopoly over coercive power. People cannot live autonomously in a state that has unlimited power to direct the shape of their lives. Where this argument goes wrong, however, is in identifying the threat of tyranny with the state alone. By pinpointing the state as the sole coercive power, the liberalism of fear fails to recognize a second group of actors who possess and often exert at least as much coercive control over children as the liberal state ever can—namely, parents. As the power of the state stands in relation to the power of one of its citizens, so can the power of a parent stand in relation to that of his or her child. Parents have the means to coerce their children physically, mentally, and emotionally; they can control the flow of information to and from their children; they possess far greater physical and intellectual might than most children, especially young children; and they often have control over both the long‐range and the day‐to‐day character of their children's lives. From children's perspectives, therefore, parents have the potential to (p.69) be at least as tyrannical as the state—and thus to pervert the course of their education and inhibit their development of autonomy—if not checked by outside institutions (including, for example, the state).

From this perspective, the detached school does not destabilize the balance of power within the liberal state; rather, it is balanced by parental influence over other aspects of the child's life, and thus actually helps to right an imbalance of power that could otherwise permit the exercise of parental tyranny over children. Amy Gutmann sums up this balancing act between the family and the state as follows:

parents command a domain other than schools in which they can—and should—seek to educate their children. . . . The discretionary domain for education—particularly but not only for moral education—within the family has always been and must continue to be vast within a democratic society. And the existence of this domain of parental discretion provides a partial defense against those who claim that public schooling is a form of democratic tyranny over the mind. The risks of democratic and parental tyranny over moral education are reduced (although they can never be eliminated) by providing two substantially separate domains of control over moral education.6

As Gutmann suggests, there is good reason for parents to have primary care of and control over their children; but from the perspective of the liberalism of fear (as well as from children's a priori perspectives) there is equally good reason for the state to place a check on parents' educative power as there is for parents to check the state. This, in brief, was the argument developed in Chapter 2. And it is hard to conceive of a better check on parental power by the state than the liberal educational ideal presented in section 2.4.

Furthermore, the existence of other liberal institutions and freedoms also helps to counterbalance the threat of state tyranny over children's education. The liberal educational ideal, after all, was constructed with the assumption that it would be implemented within a liberal state; it is not and was not intended to be a blueprint for education in all political circumstances. Thus, the ideal in section 2.4 was developed against the background of other substantive liberal institutions and freedoms, such as democracy, freedom of speech and of the press, freedom of association, and so forth. As a result, parents and other concerned community members have mechanisms available not only to find out about what children are learning in school, but also to organize, protest, and help bring about changes in schools, if necessary. Because it is presumed to be embedded within a functioning liberal democracy, therefore, even a corrupt state education system will be open to challenge via other liberal freedoms and institutions. In this context, the detached, autonomy‐driven school plays a successful dual role in counteracting the liberalism of fear and the threat of both state and parental tyranny.

(p.70) Liberalism of Fear II—The Tyranny of Sameness

The notion of the detached school as a check against the potential tyranny of outside actors such as parents also answers the second objection to the liberal educational ideal presented in section 2.4—namely, the objection that autonomy‐promoting schools foster a tyranny of sameness. The argument here, made most forcefully by J. S. Mill in chapter 5 of On Liberty, is that a standardized state education for autonomy will result in standardized people, and thus to a loss of the social and civic conditions of plurality which are essential for autonomy (as well as for other social goods). As Mill declares:

That the whole or any large part of the education of the people should be in State hands, I go as far as anyone in deprecating. All that has been said of the importance of individuality of character, and diversity in opinions and modes of conduct, involves, as of the same unspeakable importance, diversity of education. A general State education is a mere contrivance for moulding people to be exactly like one another. . . . An education established and controlled by the State should only exist, if it exist at all, as one among many competing experiments.7

Although he does believe that the state has the ‘duty of enforcing universal education’,8 Mill argues that it should fulfill this duty primarily by forcing parents to provide schooling for their own children—even when this restriction disables the state from trying directly to advance children's autonomy through education. The state ‘might leave to parents to obtain the education where and how they pleased, and content itself with helping to pay the school fees of the poorer classes of children, and defraying the entire school expenses of those who have no one else to pay for them’.9

How can we make sense of Mill's simultaneous support for autonomy for children and his deprecation of universal, state‐controlled schools? Why does Mill view the imposition of a uniform system of state‐governed schools—even well‐intentioned, freedom‐respecting schools—as a greater threat to individual freedom than the unmediated imposition of parents' views upon their children? I believe that there are two crucial ideas at work in Mill's argument, which depend on an appreciation, first, of the difference in scope between parental and state power, and second, of the peculiarity of education as a practice which helps to create the characters of members of the next generation. First, despite their authority within the family itself, parents have a very limited, localized power. Their power extends over their own children and within their own neighborhoods, but no further (and often, in the case of broken families, rebellious children, or decaying neighborhoods, not even that far). The state, on the other hand, has an effectively universal reach—it is able to exert at least some measure (p.71) of power over all children and in all neighborhoods. (As with parents, the nature and strength of the state's power, of course, varies from family to family and child to child.) Second, education, too, has a long reach, shaping individuals in ways that impose consequences not only on the lives of the educated generation, but also on the lives of future generations to come. The adults that children become are substantially shaped by the education their child‐selves undergo; to educate a generation of children, therefore, is to determine in part the character of the generation of adults they become.

Taking these two insights together, we see that the union of state and educational power generates profound social consequences of a sort which Mill believes parental control over education does not. For allowing the state to monopolize education, as the liberal educational ideal might be accused of doing, is equivalent to allowing it to colonize and homogenize the characters and beliefs of society itself. As Bonna Devora Haberman puts it, ‘Without groups who educate toward different versions of the good life, our societies would veer toward some strange contentless homogeneity. We can conjure images of populations of tolerant, polite people who vote regularly, hope that democracy will continue, but have no sense of personal identification with any culture or tradition, costume or ritual, ethnicity or historical experience. What would be the colour, flavour or aroma of such a society?’10 State control over education, on this reading, inevitably leads to social homogeneity—regardless of the content of the education itself. Parental control over education, on the other hand, results in a heterogeneous population. Although each individual child (and thus every future citizen) may be exposed to only one set of ideas, beliefs, and conceptions of the good, society as a whole will embody a plurality of opinions and ways of life. Members of this second society arguably have a much greater chance of overcoming a ‘tyrannical’ upbringing later in life than do members of the first, state‐educated society, because they will encounter a much greater diversity of ideas and ways of living.11 Thus, while a state‐supervised, autonomy‐driven education may help children develop the individual capacities necessary for autonomy, its universal imposition drastically curtails the social, civic, and cultural conditions of plurality that make the exercise of autonomy possible.12

This is a much more subtle criticism than the first, more general fear of state tyranny. It also resonates strongly with contemporary concerns about multicultural education, forcible assimilation, and the ‘celebration of diversity’. I suggest, however, that the ideal of liberal education presented in section 2.4 can stand up to this critique fairly easily, again for reasons that further prove its desirability as a model for liberal education. Three responses are particularly pertinent.

(p.72) First, it is simply untrue that plurality in and of itself enables individuals to be autonomous. Pluralism is, as Chapter 1 contended, a necessary characteristic of an autonomy‐promoting community, insofar as an individual must have a diverse range of worthwhile options open to her in order to be able to exercise meaningful choice.13 But social diversity is certainly not a sufficient condition for individual autonomy, since an individual may well be confronted by diversity but have no way to respond to it in an autonomous fashion. As well as living in a diverse society, an individual must possess the personal capacities for critical thought, toleration, values pluralism, and so forth if she is to be able to exercise autonomy. Insofar as parent‐governed education would not necessarily help children develop these personal capacities for autonomy, therefore, it cannot be said to advance autonomy. Thus, while Mill is right to remind us that it is counterproductive to sacrifice the social conditions of autonomy to the demands of the personal, he is wrong to emphasize the social to the detriment of the personal. Attention must be paid to both, but putting schooling back in the hands of parents pays attention only to one.14

Second, neither is it true that state‐directed education for autonomy creates uniform individuals, nor, therefore, a uniform society. This is in part because of the diverse forms that education for autonomy can take. Schools under the liberal ideal need not be mirror images of each other. They can be detached from parental control, and be shaped so as to form autonomy‐promoting communities, and none the less specialize in particular academic subjects such as science or creative writing, or emphasize community service instead of sports, or establish a media studies or health science magnet program. On top of this, liberal schools will display a good measure of local variation as they accommodate themselves to the needs and interests of their particular student bodies. One strength of the liberal educational ideal, therefore, is that it permits a wide range of educational structures and practices within its remit of autonomy‐promoting education. (See section 5.2 for an in‐depth discussion of specialization and school choice.)

Even more important than the possible variation among liberal schools, however, is the immense amount of variation that is possible among autonomous individuals, and thus the falsehood of the idea that societal homogeneity along the dimension of autonomy entails homogeneity across the board, or even along any other dimension. For autonomy is not and cannot be itself a conception of the good; to say that one lives an autonomous life is to say almost nothing about the substance or content of that life. Autonomous people will choose to lead very different kinds of lives, among which the only common denominator will be that their lives are not incompatible with autonomous choice—i.e., it is not impossible to (p.73) have chosen them autonomously. The fact that a society is undifferentiated along the axis of autonomy therefore says nothing about the possibility of differentiation along other axes. Society under state‐regulated education for autonomy would likely be at least as plural as society under parent‐directed education, and would be assured of inculcating the personal dispositions for autonomy (such as critical reflection and toleration) as well. Thus, parent‐governed education has no political advantages, and at least a few disadvantages, in comparison to liberal state‐regulated education for autonomy.

Finally, even if liberal schools were more like each other than I have suggested, it is still not true that children who undergo this uniform education (i.e. state‐provided and autonomy‐directed liberal education) would develop into the same kind of people. As I indicated earlier in responding to the initial fear of state tyranny, school is only one influence among many in the life of a child. It is only one instance of what the great historian of education Lawrence Cremin calls ‘the immense diversity in the sources and sites of the education that ultimately makes a difference in people's lives’.15 As Cremin emphasized in a 1989 lecture at Harvard:

profound changes since World War II in the rearing of children and sweeping changes in the nature, uses, and delivery of information have radically transformed the ecology of education in the United States and fundamentally altered the circumstances within which schools and colleges carry on their work. Put otherwise, changes in the education proffered by families, day care centers, peer groups, television broadcasters, and workplaces have drastically altered the overall education being offered to the American people. The result has been a cacophony of teaching, the effects of which have been at best difficult to determine and even more difficult to assess.16

We might add to Cremin's list the educative influences of children's increasing access to computers, summer camps and courses, after‐school lessons and activities, and so forth. In particular, given the recent explosion of sources of information to which children have access (television, computers, e‐mail, the World Wide Web, etc.), it is hardly tenable to claim that schools provide the only, or even the most influential, education that children receive.

In addition, these other educative institutions and activities have an important influence on how and what children learn in school itself. Cremin reminds us that ‘the influence of any particular educative institution is rarely direct and unalloyed; it is almost always mediated—that is, reflected, refracted and interpreted—by other educative institutions. The family is probably the most important of these mediating agencies.’17 In this respect, the fact that state education may everywhere be ‘the same’ does not mean that what children learn is everywhere the same; families (p.74) have a tremendous influence over how their children actually learn and assimilate what is taught in school. James Coleman corroborates this conclusion in his influential report on Equality of Educational Opportunity: ‘within each racial or ethnic group, the factor that showed the clearest relation to a child's achievement was his home background—the educational and economic resources provided within his home’.18 Thus, even universal imposition of the state‐regulated liberal school described in section 2.4 does not portend the imposition of a universal and uniform character on future generations. State‐regulated, autonomy‐promoting education simply does not threaten the plurality and diversity of modern, liberal societies.

In conclusion, I do not find threat of the ‘tyranny of homogeneity’ to be a tenable one. A uniform education does not create uniform people. A society of people who are uniformly autonomous is not uniform in any other way. And state‐regulated education for autonomy enables children to make choices among diverse options in a way that parent‐regulated education does not necessarily do. Liberal schools, meaning the detached, autonomy‐promoting school ideal presented in Chapter 2, can display substantial variation and diversity among themselves, thus enabling them to respond and cater to a diversity of children's needs, as well as promoting an honestly plural society in which people have the critical capacities and personal abilities to engage with each other. In all of these respects, fears of tyranny are unfounded, and the strengths of the liberal educational ideal are evident.

3.2 Parental Involvement and the Dynamics of the Family–School Relationship

Modifying the Liberal Educational Ideal Parental Involvement

Even if schools which are ‘detached’ from parental control are not tyrannical, it may be argued that they are educationally ineffective. Schools can educate successfully, it is often argued, only when they act in partnership with parents, especially by encouraging parent involvement in the school. The detached‐school ideal seems to neglect this important pedagogical point. I contend in this section, however, that while parent involvement is very important in boosting students' achievement (which is the variable used in the literature to measure educational effectiveness), this does not mean that parents must be given greater control over or input into the aims and content of the school. The available research demonstrates that parent involvement programs generally work equally well when there is a gap between the values espoused by the school and by the parents as when both school and parents embrace the same educational values. Thus, I conclude that the liberal school can educate effectively and promote (p.75) parent involvement without abandoning its separation from parochial norms or modifying its commitment to helping children develop the capacity for autonomy.

The following statement effectively sums up the conclusion of a vast literature on parental involvement and student achievement: ‘The single most important determinant of a child's success in school, and ultimately throughout life, is not family status, education level, income, or IQ. It is whether that child's parents are involved in his or her education.’19 As noted sociologists of education James Coleman and Thomas Hoffer concur, ‘one of the most important factors in a child's success in school is the degree to which his or her parents are actively involved in the child's education’.20 It is difficult to find anyone these days who would dispute this conclusion. Furthermore, parental involvement is shown to have a wide range of beneficial effects, of which helping students attain better grades or marks is only one: ‘Involving parents in the education of their children has been found to have an associated link not only with students but also with parents, teachers, schools, and districts. These outcomes include increased student achievement, increased student attendance, lower dropout rates, increased interactions between parents and their children at home and increased positive attitudes by teachers toward parents being involved.’21 Parents' involvement in their children's education thus has significant consequences across the board, from increasing straightforward academic achievement to improving domestic relations and decreasing school‐leaving (drop‐out) rates. In addition, it has been shown to be beneficial for children of all ages, all grade levels, and all socio‐economic levels.22 If anything, parent involvement seems to be especially advantageous for minority and poor students.23 In sum, therefore, the sheer volume of studies and parent‐outreach programs all reaching the same conclusion provides compelling evidence that parent involvement is essential to student achievement and learning.

This fact suggests that the discussion of liberal education in Chapter 2—and particularly the ideal of the detached school—relied on a false and debilitating dichotomy between children's upbringing and education, or between their education in general and their schooling in particular. As section 1 of this chapter reminded us, children learn a tremendous amount outside of school as well as within it, and often the former has a profound influence on how and what they learn in the latter—i.e. within the context of formal schooling. The studies of parental involvement cited above reinforce this conclusion, suggesting that schools and families (let alone schools and other institutions such as the local community) are not as separable from a pedagogical standpoint as liberals might like them to be from a political standpoint.

(p.76) The detachment of the liberal school therefore causes two problems. First, parents are likely to feel alienated from an institution that does not give them a voice. Insofar as parents are explicitly excluded from the process of determining both the overall aim of liberal education (which is defined as helping children develop autonomy) and the values that will shape the day‐to‐day operations of the school, they may feel little loyalty to or inclusion within the liberal detached school. Second, the fact that the development of autonomy in particular is taken as the aim of liberal education is likely to alienate some parents. We saw in Chapter 2 that parents' opposition to teaching autonomy was not a reason in principle for liberal schools not to help children develop their capacities for autonomy. But it does threaten the success of liberal education in practice. Parents are unlikely to support or become involved in a school that has aims to which they are opposed. If parent involvement in and support of their children's schooling is essential for educational achievement, therefore, and if the detached school cannot secure that support, then the liberal educational ideal of the detached school cannot succeed at educating for autonomy.

In response, it is first important to recognize the number of different forms that ‘parental involvement’ can take. It does not mean only serving on the board of governors or chairing the Parent–Teacher Association (PTA), and empirical research shows that much less time‐consuming or interventionist methods of involvement can be equally valuable for boosting student achievement (as well as for advancing the other goals mentioned above). As used in the literature, ‘parent involvement’ means anything from joining the PTA and volunteering in the library, to helping children with their homework, providing a quiet place to study, and asking questions about what happened in school, to baking cupcakes for the choir, chaperoning field trips, and organizing school fundraisers. Joyce Epstein, an influential researcher in this area, separates levels of involvement into five or six categories, all of which have been shown to be effective means of helping children learn.24 Although she suggests that it is important for a school to have parents involved at each level, it is not necessary that all—or even more than a few—participate to the highest degree. Her studies show that indirect parent involvement, such as helping children at home and providing a quiet place to work, is at least as important as volunteering directly in the school. Likewise, parents' efforts to attend their children's school plays and parent–teacher nights can send as strong a message to their children of support for the school and for the importance of learning as would their serving on the PTA.25

I emphasize the number of different ways in which parents can become involved in their children's education because it is significant that few of (p.77) these examples rely on agreement or harmony between parents and the school concerning the aims of education. Attending school concerts, helping with homework, and even volunteering in the library are activities that parents can and likely will support and engage in regardless of their attitude toward the school or its aims. Because these forms of involvement are often not value‐laden, they can be fulfilled even by parents who are opposed, for example, to their children's developing the capacity for autonomy. Thus, the political ideal of the detached, autonomy‐driven school does not necessarily conflict with the pedagogical need for parents to be involved in their children's education.

This conclusion remains true even if one takes into consideration the school's need to encourage parent involvement. For it might be objected that even if school involvement itself does not trigger value conflicts, parents must none the less be encouraged by the school to get involved and take on these responsibilities; in order for the school to speak authoritatively and convincingly, this argument runs, it must appeal to parents by articulating shared values and common concerns. But if one reads the empirical literature on how to boost parent involvement, one discovers that very little attention is paid to establishing shared values or shaping the school to mirror parents' concerns and commitments. Rather, researchers advise schools to make themselves inviting by furnishing a parent room in the school with comfortable couches and a phone,26 or by translating school documents and forms into parents' native languages27—i.e., forms of outreach that are neither reliant on nor related to either the school's educational aims or the parents' values. Other suggestions include: establishing regular teacher–parent contact, with teachers being encouraged to call and even visit the child's home with good news about a child and not just when the child is failing or delinquent;28 inviting parents to visit the school any time during the year and not just on ‘Back‐to‐School’ nights;29 scheduling student performances, parent–teacher conferences, and other school events for nights and weekends in order to accommodate working parents; and setting up fora in which immigrant and non‐English‐speaking parents can learn about the structure and expectations of the school in their own language.30 All of these measures require that schools and teachers commit themselves to reaching out to parents and encouraging parent involvement.31 They do not require, however, that the school adjust its educational aims to meet parents' preferences or opinions, nor that parents come to adopt the school's aims as their own.

The same is true even for parents who might naturally feel disaffected by and/or alienated from the school: e.g., recent immigrants, minorities, and non‐ or limited‐English speakers. Sociologists, educators, and researchers writing about parent involvement among these groups consistently suggest (p.78) that the aims and operations of the school should be made linguistically and/or culturally comprehensible to such parents. But they do not demand that its aims or methods should be radically altered.32 The focus is on giving minority parents knowledge and means (often understood as social and/or cultural capital) through which to engage with the school: ‘Parents who are knowledgeable about the school's expectations and the way in which the school operates are better advocates for their children than parents who lack such skills.’33 Likewise, schools are encouraged to learn about the socio‐cultural backgrounds and needs of their students, and to shape their programs so as to meet and capitalize on these elements.34 Responsiveness to children's needs, however, need not translate into lowering standards or changing one's educational aims. Schools are repeatedly urged to maintain their aim of educating all children to the highest level, regardless of their students' backgrounds.35

Furthermore, it is worth noting that virtually all parents—illiberal and liberal, minority and mainstream alike—in fact share a number of essential educational values both with each other and with the liberal detached school, values that the school can appeal to by way of encouraging parental support and involvement. Virtually all parents, for example, would like to send their children to schools that offer a strong academic curriculum emphasizing reading and writing skills, mathematics, science, history, and other traditional academic subjects. Furthermore, most parents favor schools that nurture their children emotionally and socially as well as academically, and that teach values or habits of character such as honesty, responsibility, courage, and friendship.36 While it is true that some parents also look for other values in their children's schools—values which may (e.g. emphasis on arts education; strong sports teams) or may not (e.g. restriction to children of a certain ethnicity or social class; emphasis on a religious text as the ultimate source of truth) be compatible with the liberal educational ideal—these are generally held to be important in addition to, rather than instead of, the above values. As a result, schools can encourage most parents to involve themselves in their children's education—helping them with their homework, coming to Parent Day, attending awards ceremonies, volunteering in the library or cafeteria, etc.—by appealing to shared values of academic achievement and social and emotional development.37

One more potent objection must be addressed, however, before we can conclude that detached schooling is compatible with parental involvement and thus student achievement. We might call it the ‘Catholic school objection’. In the early 1980s, James Coleman, Thomas Hoffer, and Sally Kilgore set off an extended and heated debate when they claimed that Catholic (p.79) (private) schools in the United States are far superior to state schools in reducing premature school‐leaving rates, ‘adding value’ to poor and minority children's education, and generally increasing student achievement.38 Although their conclusions have been hotly contested,39 they present potentially compelling evidence that Catholic schools benefit students more than state schools do because the former operate within the context of a close‐knit community involving teachers, parents, and children—a community which no longer exists for the average American family, neighborhood, or state school. This would seem to suggest that schools are more effective when they create such communities by reflecting parents' values and commitments, and thus suggest that the ideal of the detached school is pedagogically flawed.

If one reads Coleman and Hoffer's analysis in particular more closely, however, one discovers that Catholic schools do not boast better education because they constitute a values community. Rather, it is because they form a functional community—‘a community in which social norms and sanctions, including those that cross generations, arise out of the social structure itself, and both reinforce and perpetuate that structure’.40 In essence, Catholic schooling is effective, Coleman and Hoffer argue, because it takes place within a community in which parents interact and talk both to each other and to each others' children. This is valuable for at least three reasons.41 First, it enables parents to share information with each other and thus to make better choices on an individual level for their own children:

A functional community augments the resources available to parents in their interactions with school, in their supervision of their children's behavior, and in their supervision of their children's associations. . . . The feedback that a parent receives from friends and associates, either unsolicited or in response to questions, provides extensive additional resources that aid the parent in monitoring the school and the child, and the norms that parents, as part of their everyday activity, are able to establish act as important aids in socializing children.42

Thus, parental interaction provides the ‘social capital’43—i.e. the network of social relationships—that enables parents to work more effectively on behalf of their children's education. Second, this same social capital enables parents to work in tandem in order to press causes of common concern, such as improving the curriculum or replacing an unpopular teacher. Organized parental pressure can be a force for the good, albeit not an unmitigated one. Parents often have privileged insight into their children's needs and desires, and by talking to other parents and comparing their experiences and those of their children, they can discover problems (or identify strengths) within the school of which teachers and school administrators may themselves be unaware.44 Third, parental interaction (p.80) within a functional community often has the effect that initially uninvolved or marginalized parents are motivated to get more involved because of peer pressure from other parents: ‘In a value community, the involvement of families in the school is largely determined on an individualistic basis, while the scope of community involvement is greater in the functional community. Thus, parents who would otherwise avoid the school or be uninvolved in its activities are induced to participate through the social relationships that define the functional community.’45

It is important to recognize, however, that functional communities are not necessarily characterized by shared values: ‘A functional community is neither a necessary nor a sufficient condition for value consistency, though the two have some affinity. Value consistency grows through the interactions found in a functional community, and when it exists it facilitates the norms that grow up in such a community. But there are value cleavages in functional communities, cleavages which may be intense.’46 This is often true even of Catholic schools. Thus, it is neither a necessary nor a sufficient means of increasing parents' social capital to shape schools to reflect parents' values, or to try to attract parents and students whose values accord with those held by the school. For example, independent (non‐religious) private schools often constitute values communities but ‘ordinarily lack . . . a [functional] community, consisting as they do of a collection of parents who have individually chosen a school but who do not constitute a community outside the school’.47 As a result, Coleman and Hoffer argue, such schools' ability to keep students in school and to boost student achievement is statistically much more similar to that of state schools. The Catholic school objection, therefore, emphasizes the importance of encouraging parents to interact with each other and with the school, but it does not suggest that schools need to structure themselves according to values that parents share.

This conclusion is further strengthened, perhaps paradoxically, by an analysis of US Catholic schools that does conceive of them as values communities. Partially in response to Coleman, Hoffer, and Kilgore's work, Anthony Bryk, Valerie Lee, and Peter Holland wrote a highly regarded evaluation of Catholic schools' achievements in which they argued in part that Catholic schools do instantiate values communities.48 What is significant about their analysis is that even though the authors specifically dispute Coleman, Hoffer, and Kilgore's contention that Catholic schools are communities only in a functional sense, they argue that the values that link the community are traditional academic values rather than religious ones. According to Bryk et al., most Catholic schools and teachers believed that all students should ‘take a core of academic courses’ and they emphasized regular homework, testing, and strict discipline as means to achieving these (p.81) goals.49 The core of schools' ‘explicitly moral understanding of the purposes of education’50 is also not centered in Catholic dogma or other explicitly religious wellsprings of morality. Rather, ‘the emphasis was on developing a sense of personal responsibility and a commitment to hard work’.51 A national survey of Catholic high school teachers found overwhelming support for emphasizing ‘mastery of reading, writing, and mathematics skills’, ‘critical thinking skills’, ‘intellectual curiosity’, ‘a healthy self‐concept’, ‘compassion’, ‘tolerance’, and ‘a commitment to justice’. In notable contrast, barely half of these teachers viewed ‘building knowledge of Catholic doctrine’, ‘acceptance of Catholic teachings on moral values’ or ‘clear understanding of the Bible’ as educational priorities.52 Thus, we see that even to the extent that Catholic schools do instantiate values communities (a view that is itself still open to question, since although Bryk et al. show that Catholic school teachers and administrators share many educational values, they do not show that these values are equally or explicitly embraced by students and parents as well), the shared values are ones that are fully compatible with—even essential to—the aims of the detached liberal school: academic achievement, critical thinking skills, social responsibility, toleration, justice, etc. To the extent that these shared values raise parent involvement and student achievement in Catholic schools, they can and should raise parent involvement and student achievement under the liberal educational ideal as well.

In conclusion, while it is true that parents' involvement in their children's education is extremely important, liberal schools need not abandon their detachment from parents' values in order to encourage such involvement. First, research shows that parent involvement relies primarily on the school's engaging in value‐neutral outreach of the sort discussed above—not on its establishing a values community reflective of parental commitments. Second, to the extent that parent involvement is responsive to schools' emphasis on values which they share, schools can appeal to the values of academic achievement and social and emotional development as a means of encouraging parent involvement without abandoning the detached status that ultimately helps children's development of autonomy. In sum, the detached school is fully compatible with parent involvement, and thus with effective schooling and high levels of student achievement in general. This is not to suggest that there is nothing more to say about the problem of creating an ethos of involvement, commitment, and mutuality among the school, parents, and other local community members. Chapter 5 will take up this issue again through the lens of school choice, examining in much greater detail whether market forces in education can and would foster the distribution of social capital among parents or increase student learning in other ways. Until then, however, we can safely conclude that (p.82) liberal education can be effective education, and that students have ample opportunity and encouragement to learn and succeed within the detached school.

3.3 Abandoning the Structure versus Substance Distinction: Coming to Terms With the Hidden Curriculum

Modifying the Liberal Educational Ideal Accommodating the Hidden Curriculum

In this section, I examine the argument that schools are intrinsically structured so as to inculcate heteronomy instead of autonomy. Even if the distribution of political power represented by the autonomy‐driven, detached school is not tyrannical (section 1), and even if it is not pedagogically counterproductive because it forbids certain forms of parental involvement (section 2), it may be that the detached school—in fact, any school—is none the less structurally unable to educate for autonomy. Deschoolers, functionalists, and other radical theorists argue that an ideal of education such as that put forth in section 2.4 is mired in self‐contradiction, because even liberal schools are inherently structured so as to reinforce heteronomy instead of autonomy by rewarding obedience, submission to authority, and intellectual docility. These arguments, and their implications for liberal education, are addressed in the first half of the section. The second half picks up in particular on the substantive moral and cultural messages conveyed through a school's structure. Whether or not schools can foster autonomy by modifying their structures, I suggest, they inevitably transmit cultural norms that may exclude or alienate some minority children. This causes tensions within the detached school whose ramifications will not be fully dealt with until Chapter 5.

Heteronomy, Capitalism, and Conformity: The Threat of the Hidden Curriculum

Schools cannot teach autonomy. This message of pedagogic self‐defeat has been most forcefully articulated by members of the ‘deschooling’ movement headed by Ivan Illich.53 Deschoolers such as Illich claim that the very structure of schooling as a practice suppresses instead of augments freedom, and thus is a priori unsuited for the job of educating for autonomy. As Illich charges in his famous treatise, Deschooling Society:

The claim that a liberal society can be founded on the modern school is paradoxical. The safeguards of individual freedom are all canceled in the dealings of a teacher with his pupil. When the schoolteacher fuses in his person the functions of judge, ideologue, and doctor, the fundamental style of society is perverted by the very process which should prepare for life. A teacher who combines these three (p.83) powers contributes to the warping of the child much more than the laws which establish his legal or economic minority, or restrict his right to free assembly or abode.54

Although it may initially appear that Illich is merely saying that tyrannical or authoritarian teachers infringe the freedom of students, he is actually making a much deeper point. Built into schools, Illich suggests, is an unavoidable structure of power that is antithetical to both the development and the exercise of autonomy, because it establishes a cognitive and psychological relationship of dependence of the student upon the teacher that is impossible to overcome. This relationship necessarily obtains in all schools, regardless of their ‘progressive’ or ‘liberal’ nature, because the teacher's role as evaluator (‘judge’), authority (‘ideologue’), and therapist (‘doctor’) is built into the very essence of schooling. Thus, he concludes, the very conception of an autonomy‐advancing liberal school is incoherent, because as compulsory institutions in which teachers have intellectual, evaluative, and emotional authority over students, schools cannot help but deform children's development of autonomy.

This notion of the self‐defeating structure of liberal education is given a slightly different (and more interesting) twist by the sociologist of knowledge Nell Keddie. In her fascinating account of ‘Classroom Knowledge’,55 Keddie makes the provocative suggestion that a student's success in school is structurally predicated on his being willing to accept the teacher's conception of and approach to the curriculum and the classroom. The most successful students, in other words, are those who display the least autonomy and independence of thought. Again, this is so even in the most autonomy‐driven of schools, as it is not the particular ideology of a particular school but the intrinsic structure of schooling itself that generates the disjunction between individual autonomy and formal educational success. Keddie documents in detail how self‐consciously ‘progressive’ teachers apply regressive and autonomy‐stunting evaluations to those students who challenge their intellectual or disciplinary authority. Such evaluations may include ignoring or ‘render[ing] invalid’56 students' explanations of their motives and objections, and placing these students in lower‐stream classes. As Keddie puts it, ‘It seems probable that the pupils who come to be perceived by teachers as the most able, and who in a streamed school reach the top streams, are those who have access to or are willing to take over the teachers' definition of the situation.’57 As a result, ‘There is between teachers and A [high‐stream] pupils a reciprocity of perspective which allows teachers to define, unchallenged by A pupils, as they may be challenged by C pupils, the nature and boundaries of what is to count as knowledge. It would seem to be the failure of high‐ability pupils to question what they are taught in schools that contributes in large measure to their educational achievement.’58

(p.84) Keddie's analysis, like Illich's, suggests that schools are ill equipped to teach students to be autonomous, because the structure of authority intrinsic to the practice of schooling inadvertently but unavoidably teaches a substantive message of docility and intellectual submissiveness—exactly those qualities which hinder the development of autonomous thought and action. Students learn that in order to be successful in school, they must suppress or alienate their own judgemental powers in favor of the teacher's perspective.59 Thus, the structure of even the most ideal liberal school inadvertently conveys a substantive hidden curriculum60 that teaches students to be heteronomous in even the most fundamental aspects of their lives in school. Students' suspension of judgement concerning what they learn in school might not be so troubling if the teacher's perspective echoed students' own common sense, and thus if students could rely on some measure of overlap between their own judgement and that of their teachers. But because the formal academic curriculum always requires some measure of distance and abstraction from common sense, ‘It would appear that the willingness to take over the teacher's definition of what is to constitute the problem and what is to count as knowledge may require pupils to regard as irrelevant or inappropriate what they might see as problems in a context of everyday meaning. . . . This means that those pupils who are willing to take over the teachers' definitions must often be less rather than more autonomous . . . and accept the teacher's presentation on trust.’61

This is not to say that the overt or formal curriculum of the liberal school is itself necessarily self‐defeating. Children who learn what the liberal school tries to teach through the formal curriculum (e.g. science, history, health, etc.) do presumably become more autonomous. But paradoxically, they can learn what the school is trying to teach only after they learn the lessons taught by the hidden curriculum (i.e., the lessons conveyed by the structure of the school)—namely, obedience and deference to authority. For only if they are obedient and surrender at least part of their judgement to the teacher's authority can children get down to the business of learning the formal curriculum. Thus, the hidden curriculum holds the overt curriculum hostage. Not until children learn from the hidden curriculum to suppress their autonomy can they begin again to develop it through the overt curriculum.

To distinguish between the overt and hidden curricula of the school on the basis of intentionality may itself be misconceived, because according to functionalist theorists such as Samuel Bowles, Herbert Gintis, Philip Jackson, and others, the substantive lessons embedded in the structure of the school are often at least as intentional as those imparted by the overt, formal curriculum.62 Bowles and Gintis put the charge most directly: (p.85)

To capture the economic import of education, we must relate its social structure to the forms of consciousness, interpersonal behavior, and personality it fosters and reinforces in students. . . . we believe the available evidence indicates that the pattern of social relationships fostered in schools is hardly irrational or accidental. Rather, the structure of the educational experience is admirably suited to nurturing attitudes and behavior consonant with participation in the labor force.63

From this perspective, the hidden curriculum of the school is intentionally constructed to teach children the skills, habits, and modes of behavior best suited to perpetuating the capitalist system. Students are taught through the hidden curriculum to be docile and obedient because these are the qualities most needed in workers in the modern capitalist economy. The heteronomy and suspension of critical judgement that Keddie pointed to as accidental (albeit unavoidable) by‐products of contemporary schooling is instead seen under this functionalist analysis as expressing one of the central hidden aims of modern, capitalist‐driven schooling. From the lunch and recess queues one learns to stand in as a primary school student, to the time discipline instilled by the bells, deadlines, and exams imposed in middle school, to the lessons about class rigidity and separation instilled by streamed classes and graduation tracks in secondary school, children are intentionally schooled in the service of capitalism from the start of their formal education. Philip Jackson describes the situation as follows: ‘As he learns to live in school our student learns to subjugate his own desires to the will of the teacher and to subdue his own actions in the interest of the common good. He learns to be passive and to acquiesce to the network of rules, regulations, and routines in which he is embedded.’64 In this way he is turned into ‘the Company Man’.65 Thus, the functionalist argument concludes, by ‘replicat[ing] the relationships of dominance and subordinacy in the economic sphere’, the structure of modern schooling establishes a hidden curriculum that ensures the replication and production of the ‘amenable and fragmented labor force’ needed to perpetuate the capitalist system.66

The liberal educational ideal articulated in section 2.4 must thus confront two powerful sets of objections: first, that schools cannot teach autonomy, because intrinsic to the nature of formal education (schooling) are pedagogic structures that necessarily deform children's development of autonomy; and second, that schools will not teach autonomy, because teachers, school reformers, administrators, and even parents are fatally blinkered by capitalist assumptions and will therefore favor educational structures that diminish autonomy to advance capitalism, rather than vice versa. These objections are linked insofar as they rely on the existence of the so‐called ‘hidden curriculum’—i.e. on the connection between the structure and content of education. Furthermore, these objections are mere subsets (p.86) of a much larger challenge to the liberal educational ideal: namely, that the structure of schools teaches a ‘hidden curriculum’ of substantive norms, values, sources of identification, ways of life, and so forth which themselves diminish or violate children's development of autonomy. It is this fact of the hidden curriculum that poses the greatest challenge to the liberal educational ideal, because it suggests that education is out of our control—that we cannot map aims (educating for autonomy) onto outcomes (liberal schools) in the way that Chapter 2 demands we be able to.

Let us take the latter charge first, that the modern, putatively liberal school has been intentionally structured so as to perpetuate the infelicities and injustices of capitalism. I suggest that even if this is true, it has few if any implications for the liberal educational ideal spelled out in section 2.4. For insofar as the structural threats and subversions identified by functionalists are avoidable given sufficient vigilance, liberal educators must simply learn to be vigilant. It is in fact presumably easier for the liberal school to avoid the pitfalls of capitalist‐dominated education than for other schools. Since the liberal school is conceived of as being partially shielded from the direct influence of parents and the local community, it is presumably also better shielded than most from the influence and interference of the agents of autonomy‐inhibiting, capitalist‐driven education reform.67 Bowles and Gintis's analysis does remind us that the liberal school must remain vigilant in order to ensure that it does not unintentionally socialize children for conformity to the unjust and autonomy‐diminishing aspects of capitalist life and the capitalist system; it would seem that there is little more, however, which the ideal liberal school must do.

Furthermore, I would question whether the character traits identified by functionalist theorists as being required by capitalists of their laborers (and thus taught by schools to their students) really are so deleterious and autonomy‐diminishing as they claim. For example, Philip Jackson points out that the structure of schools requires students to be patient and willing to delay gratification, and that the capitalist workplace requires its employees to display the same qualities.68 He concludes on the basis of this link that schools are directly structured to serve capitalist interests. But, first, as all statisticians know, correlation does not prove causation. Even if the only advantage of learning to delay gratification were that it made the capitalist machinery run more efficiently, that fact by itself would not prove that schools teach children to delay gratification for that reason. Second, and equally importantly, while these character traits are often useful to employers within a capitalist system, it is equally true that patience and the ability to delay gratification of one's desires are also important conditions of living a fully autonomous life. Most goals take time and patience to achieve; someone (p.87) who was unable to delay gratification, therefore, would likely lead a frustrated and/or aimless existence. This is not to deny that one can be autonomous if one is constitutionally impatient, but patience does make full (as opposed to partial) autonomy more likely and easier to achieve. An inability to delay desire‐gratification, on the other hand, does make the achievement of an autonomous life impossible. Similar associations between habits of character and autonomy are true of self‐discipline, time management, deference to others' expertise, and so forth. Thus, education in those qualities of self‐control and even temporary self‐abnegation that functionalist writers point to with such horror as clear signs of the hegemony of capitalist interests in the classroom could be equally well interpreted as being directly intended to advance children's interests in developing their capacities for autonomy—or, for that matter, to advance non‐capitalist ways of life as well. After all, successful planning and implementation of an anticapitalist revolution would also require that its proponents exercise patience and be willing to delay gratification.

Finally, I suggest that insofar as the hidden curriculum can be known and controlled, and insofar as many of its lessons actually further children's development of autonomy rather than diminish it, the hidden curriculum can be turned into a boon rather than a burden. The conception of the liberal school developed in section 2.4 provides an ideal case in point of the hidden curriculum's knowability and appropriability for liberal ends. One of the central strengths of the ideal liberal school as described in section 2.4 is that it explicitly harnesses to its own purposes many of the structural forces that are the primary of focus functionalist and radical critiques of modern schooling. It explicitly acknowledges that the kind of community created by the school is as important an educational force as the overt curriculum, and creates itself accordingly. In this manner, it consciously and openly grapples with the problem of shaping schooling so that the ‘hidden curriculum’ represented by its organizational structure advances the ideology of the school in tandem with the overt or formal curriculum. In the liberal educational context, therefore, the ‘hidden curriculum’ is neither hidden nor deleterious. Rather, it is explicitly acknowledged and appropriated as a central means of advancing children's development of autonomy. In light of this, I propose in the discussion that follows to use the term ‘informal’ as opposed to ‘hidden’ curriculum, as a means of acknowledging its separation from the ‘formal’ curriculum covered in textbooks and lectures, but also of emphasizing its knowable and addressable nature.69 Regardless of functionalism's descriptive validity as applied to current educational practices, therefore, it poses no threat to the liberal educational ideal, which self‐consciously shapes the informal (‘hidden’) curriculum to serve its own just purposes.

(p.88) Although liberal education is able to appropriate some aspects of the informal curriculum for its own purposes, there are none the less still challenges to be met. I have not yet addressed the unmalleable aspects of the informal curriculum, which present a different and more difficult set of challenges to the liberal educational ideal. For example, although one can (and many schools in the 1960s and 1970s did) redistribute power in the school in response to arguments like Keddie's so as to eliminate the hierarchical divide between students and teachers, or even give students greater educational authority than teachers, it has been fairly conclusively demonstrated that this is educationally ineffective. Thus, to reveal the existence of harmful elements within the informal curriculum is not necessarily to overcome them, or to transform them to work toward liberal ends. Structures that are intrinsic to the process of formal education may at the same time be deleterious to children's development of autonomy. Likewise, in altering one feature of the structure of schooling in order better to advance children's autonomy, one may introduce a different feature that poses its own obstacles to the creation of an autonomy‐promoting informal curriculum.

Keddie and Illich introduce what might be thought to be the most devastating aspects of the informal curriculum, insofar as the structural features of schooling they point to as diminishing children's autonomy include some of the most basic features of the formal educational process. But the criticisms that they raise are only part of the story, and not even necessarily the most difficult for liberalism to handle. This is because the conflict that Keddie and Illich illuminate is essentially one‐dimensional: it pits the autonomy‐diminishing aspects of the informal curriculum against other autonomy‐promoting aspects of both the informal and formal curricula. The costs and benefits of liberal schooling raised by Keddie and Illich can thus be plotted and evaluated along a single dimension—the development of children's autonomy—and balanced out against each other accordingly. Assuming that compulsory liberal education on balance advances children's development of autonomy—i.e., that the detrimental aspects of the informal curriculum are outweighed by the autonomy‐advancing elements of the informal and formal curricula—and assuming that no alternative to liberal schooling will achieve those aims as well or better (and thus rejecting Illich's proposals of such substitutes as phone books of people interested in trading knowledge70), then the liberal educational ideal remains on firm foundations. While the complex trade‐off between children's exercise and development of autonomy goes deeper than I had initially acknowledged, insofar as it pervades the very structure of school itself, it threatens not the foundation of liberal education but merely the ease with which its aims can be achieved. So far as Keddie's and Illich's criticisms are concerned, therefore, the liberal educator simply has to make the best of a (p.89) bad bargain, continuing to educate in line with the principles set forth in Chapter 2 but chastened by the knowledge that the task is harder than might initially appear.

Cultural Transmission and the Informal Curriculum

There is a different set of intrinsic structural features of schooling, however, which teach an informal curriculum whose outcomes cannot be measured on the single scale of advancing or retarding children's autonomy, and which pose more fundamental challenges for the liberal educational ideal. These features include the moral and cultural values that inform, shape, and are transmitted by the informal curriculum of the school.71 Unlike the evaluative structures highlighted by Keddie and Illich, a school's cultural affiliation does not intrinsically limit children's development of autonomy. But it does help to determine the shape of children's identities, and thereby both calls into question the possibility of liberal educational ‘detachment’, and potentially causes the school to overstep its bounds as an institution incorporated for the sole purpose of helping children develop their capacities for autonomy. In both of these respects, the moral and cultural aspects of the hidden curriculum challenge the realizability of the liberal educational ideal.

Let me make clear from the start that the problem is not moral education as such. To the contrary, moral education is an essential part of helping children to develop their capacities for autonomy. Learning to tolerate others, to value critical thinking, and to take different points of view seriously are all important moral lessons for developing and exercising autonomy, and are thus important components of the liberal school curriculum. As I briefly mentioned in section 2 of this chapter, powerful liberal cases have also been made for other aspects of ‘character education’, including teaching children the virtues of courage, responsibility, trust, friendship, honesty, and so forth.72 Gutmann gives a helpful example of how schools teach liberal norms through the informal curriculum:

Schools develop moral character at the same time as they try to teach basic cognitive skills, by insisting that students sit in their seats (next to students of different races and religions), raise their hands before speaking, hand in their homework on time, not loiter in the halls, be good sports on the playing field, and abide by many other rules that help define a school's character.73

The problem is that it is misleading to suggest that these are the only norms of character that schools—even liberal schools—will and do teach. Instead, even the detached school will almost inevitably teach a culturally situated informal curriculum that will reflect and reinforce the norms of the dominant culture in society. Consider the testimony of Colin (p.90) Shrosbree, an English working‐class child who was admitted on scholarship to a private school just down the road from where he grew up:

Middle‐class attitudes, middle‐class clothes, middle‐class accents, early and irrevocable selection by middle‐class teachers who made little concession to working‐class ways or children's circumstances all made of education a barrier, not just to academic success but to a sense of owning a place in our own culture. But as children we did not question these social divisions—school was just a place apart, with different rules and alien ways, that had no place in our personal or family life.74

Gutmann herself offers another example by way of comparing American with Japanese school practices and organization. She points out that in Japanese schools, all school members from student to principal (head teacher) help in the day‐to‐day maintenance of the school, thus placing an emphasis on group responsibility for the welfare of the community and the individuals who make it up. Likewise, as Gail Benjamin points out in her excellent book on Japanese primary education, faster students are routinely expected to help slower students master the day's lesson; students work almost exclusively in groups, with praise given to the group but rarely to individual members; and on Sports Day, teams compete enthusiastically with each other for first‐prize, but individual students' first place finishes in individual races are valued only in terms of their contribution to the team—the children themselves are not congratulated or fussed over, nor do they expect to be.75 Responsibility for other members of the group, on the other hand, is not reinforced as an important norm in most American or British schools. Instead, most teachers place a premium on individual achievement, as is demonstrated in their emphasis on the following: ‘doing your own work’ and ‘keep[ing] your eyes on your own paper’; assigning homework and other assessments (e.g. quizzes, papers) to individuals rather than groups; primarily awarding medals, trophies, athletic or academic ‘letters’, and other marks of distinction to individual students; and disciplining students by emphasizing the consequences of their own behavior, regardless of the behavior of the group as a whole. To the extent that teachers in the United States or Britain do use cooperative learning techniques, they usually do so to help students develop group‐work skills, not to teach collective responsibility.

None of these approaches—working‐class versus middle‐class norms, or collective versus individual responsibility—necessarily constitutes a more liberal framework for education; they are all compatible with students' development of autonomy. What is significant is their cultural particularity, and thus their effect on children's development of identity and cultural self‐conception. Children who are raised in a Japanese‐American or Asian‐American community, for example, but are forced to (p.91) attend a putatively ‘detached’ school that reinforces individualistic norms and behavior, may feel alienated from their culture or the school or both, and/or may feel confused about their own identities (e.g., as individual actors versus group members).76 The same can clearly be true of working‐class children thrust into a school structured by middle‐class norms. Furthermore, by emphasizing their supposed cultural ‘detachment’, liberal schools may exacerbate these feelings as students' experiences of cultural incoherence are denied or invalidated.

This is an important challenge to the liberal educational ideal as it has been developed thus far. I do believe that this objection can be overcome, primarily by examining cultural membership within the context of civic membership. Chapter 4 takes on the task of analyzing and comparing cultural socialization with civic education. Because the relationship between civic and cultural membership is a complex one, however, and because the issue of schools as purveyors of cultural identity comes up again in the next section, I propose to leave the problem dangling for the time being. It will finally be resolved in section 5.1.

3.4 Supplementation or Supplantation? Building a Sense of Identity in the Modern World

Modifying the Liberal Educational Ideal Building a Sense of Identity

We could crudely summarize section 1.3 by saying that individuals' capacity for autonomy is composed of two parts. (1) Autonomy requires a sense of cultural embeddedness or coherence, of personal identity, on the basis of which one first begins to judge, weigh, and evaluate the events, opportunities, and ideas that one encounters across a lifetime. (2) Autonomy requires a bundle of personal and mental habits, skills, values, attitudes, structures of belief, etc. that taken together enable one: to recognize and take seriously a variety of ways of life and conceptions of the good; to evaluate critically one's own beliefs and commitments, and to alter them when appropriate; and to make second‐order judgements with which one actively identifies as opposed to fulfilling passively one's unexamined first‐order preferences and desires. As a shorthand, I shall refer to (1) as the sense of embeddedness or cultural coherence, and to (2) as the capacity or conditions for choice. An individual must possess both elements in order to be said to possess the capacity for autonomy.

Liberals typically focus on the capacity for choice as being harder both to achieve and to preserve, and thus as demanding attention from the state to a degree that individuals' sense of cultural coherence is thought not to require. This elevation of developing the capacity for choice over maintaining the (p.92) conditions of identification implicitly structured my own analysis of liberal politics in previous chapters. It generated the legitimation conditions and substantive institutions discussed in Chapter 1, the conception of autonomy developed in Chapters 1–2, and the form of liberal education for autonomy derived in section 2.4. In evaluating how schools should help children develop the capacity for autonomy, for example, I focused on how to get children to become autonomous choosers, not on strengthening their personal or cultural values and commitments. The ideal liberal school, it was judged, would provide a community that was separated from, not connected to, the conceptions of the good that children were meant to be learning at home. The development of a capacity for choice was seen to be the liberal school's most important aim; the inculcation of a sense of identity in children, on the other hand, was left to parents as a task which, it was assumed in section 2.3, they could naturally (even inevitably) achieve on their own.

Many critics, however, have suggested that liberalism's obsession with establishing and protecting the conditions of choice, and its relative disdain for protecting individuals' sense of cultural coherence, is woefully inadequate in the (post‐)modern or contemporary world. (When this world began is naturally a matter of dispute; estimates range anywhere from 1789 to the mid‐1960s.) Protagonists as diverse as back‐to‐basics Tories, African‐American separatists, the Christian Coalition, Blairite ‘New Labour’ supporters, Michael Lerner, the Archbishop of Canterbury, Allan Bloom, and the most avant‐garde multiculturalists have all argued that it is the explosion of the capacity for choice, and the accompanying desiccation of cultural identification and embeddedness, that now pose the greatest threats to social stability and personal autonomy.77 As William Galston solemnly warns, ‘The greatest threat to children in modern liberal societies is not that they will believe in something too deeply, but that they will believe in nothing very deeply at all.’78 According to this view, modern individuals lack a clear identity and purpose in life because they lack cultural coherence. This lack of purpose in turn occasions more general feelings of fear and anomie, which on their part pose the most serious modern threat to the maintenance of both political and personal freedom. Deprived of a standard by which to evaluate their and others' choices and afflicted by low self‐esteem, modern anomic individuals either move aggressively to impose their values on others (as a means of stoking up their failing confidence), and thus violate the demands of political stability, or they follow others' leads instead of relying on their own judgements, in violation of the demands of personal autonomy. In either case, freedom is sacrificed as a result of the ascendance of choice over cultural coherence.79

Three arguments are typically put forward to justify the claim that the relative strengths, and therefore the relative political and educational (p.93) relevance, of the capacities for choice and personal identity have shifted position. First, it is argued that there has been an explosion in the number and range of choices available to children in modern times. The very forces that I argued in section 1 counteracted the threat of state tyranny, such as children's exposure to television, the World Wide Web, and other forms of mass media, also counteract parents' attempts to provide their children an integrated cultural upbringing. As Galston remarks, ‘the basic features of liberal society make it virtually impossible for parents to seal their children off from knowledge of other ways of life’.80 Children have become increasingly sophisticated about the range of different lives people lead, and thus about the range of options potentially available to them in their own lives. Schools are therefore not needed to help children achieve a greater sense of choice than they already have, since children's capacities for choice are well developed in contemporary times.

Second, this increase in children's range and knowledge of choices is said to lead to a decrease in their identification with any one conception of the good. The capacity for choice and a sense of embeddedness are opposed attributes, according to this view, in that an increase in one must always be matched by a decrease in the other. This counterbalance is especially true for children, for whom a sense of options from a very early age can translate into a sense of interchangeability or indifference among choices. The prevalence of children's and teenagers' attitudes that ‘all values are relative’ is the most obvious instance of this phenomenon. Children and adolescents perceive standards to be arbitrary or even dangerous in the face of the multiplicity and supposed equality of available options, and they consequently either lose or abandon their ability to evaluate choices on the basis of a coherent set of values and cultural commitments. As a result, this argument concludes, the prevalence of choice in the modern world has led to a drastic and ultimately devastating loss of cultural coherence among children.

Third, critics argue that this movement toward the abandonment of standards of evaluation and the ascendance of the fetishism of choice is further exacerbated by a breakdown in families, neighborhoods, institutions, and community structures that would otherwise work in concert to reinforce children's initiation into an integrated way of life. Families move more often and over longer distances; parents divorce and remarry, or never marry at all; life is no longer centered around the church or other stabilizing community institutions; neighborhoods change character as gentrification occurs or public council housing is built; single and working parents have neither the time nor the personal resources to inculcate in their children a stable sense of community and identity; small, home‐grown businesses and traditions (p.94) are replaced by faceless conglomerates and mass‐produced products; one can generate multiple examples. For all of these reasons, and many more besides, this argument concludes that children's senses of identity, like the character of human lives in general in Western liberal democracies, have become fractured and disjointed. The capacity for choice exercises its own kind of tyranny over children's lives, depriving them of the sense of embeddedness which is so central not just to personal autonomy but also to the capacities for identity and agency themselves.

If these arguments are right, their implications for liberal education are manifold. At the very least, they would suggest that the ideal of the detached school needs to be substantially rethought. The concept of the detached school was developed as a vision of an educational community in which students are able to learn the habits, dispositions, and skills that ground the capacity for choice. It was designed to promote toleration of other points of view and ways of life, to promote critical thinking skills, and to provide a space in which children could step back and question the (supposedly) ingrained values in which they had (supposedly) been inculcated by their parents. But according to the arguments presented above, students do not need to become members of yet another community that presses them to develop the capacity for choice; instead, they need a community that develops their sense of cultural coherence and reinforces the values and commitments that are being taught, however ineffectually, at home. Schools are needed to build up parents' attempts to inculcate a sense of cultural coherence in their children, not to tear them down. From the ages of 5 or 6 to 16 or 18, children live nearly as much of their waking life in school as at home. Parents must be able to rely on the school to help reinforce their values, to be ‘connected’ instead of detached, because otherwise they have no hope of battling the other social factors that contend for the minds and allegiances of their children.

I suggest that the liberal educational ideal can generally stand up to the first two objections. First, the fact that there has been an explosion in children's exposure to choices and to different ways of life does not mean that their capacities for choice, in the senses relevant to autonomy, have also increased. It is important that children become aware of and receptive to a variety of ways of life, values, ideals, etc. But as I discussed in section 1, such awareness is not sufficient in and of itself to ground the capacity to make autonomous choices. In order to develop autonomy, children must also be able to evaluate critically the choices that are available, to imagine options that do not immediately present themselves, to assess the short‐, medium‐, and long‐term outcomes of the choices they make, to integrate the choices they make in one sphere of life with their goals and interests in (p.95) other spheres of life, and so on. Thus, the liberal, autonomy‐promoting school must still teach the skills, habits, beliefs, and knowledge that ground the capacity for choice, despite children's growing awareness of the availability of different ways of life and conceptions of the good.

Second, I reject the idea that there is a necessary trade‐off between the capacity for choice and a sense of embeddedness, so long as the liberal school educates in an age‐sensitive and age‐appropriate manner. Very young children, it is true, may well experience confusion and distress if confronted with a plethora of choices too early, or with teachers who tell them that their way of life embodies only one possibility among many. In this context, age matters. The early grades are an appropriate place to teach patience, listening skills, how to express oneself to others, and how to find one's way around the neighborhood, all of which are important for the development of autonomy; they are not the appropriate context in which to teach radical doubt or critical thinking skills. Studies in child development and psychology support this conclusion: children are simply not cognitively equipped before a certain age to engage in the sort of critical thought and analysis that ultimately provides the foundation for the capacity for choice.81 As Benjamin Barber puts it, ‘Teach the child what she needs to know, but also teach her what she is ready to learn.’82

The fact, however, that very young children are not equipped to think critically does not mean that children can never do so successfully, or that there will always be a trade‐off between choice and cultural coherence. As children develop they certainly become able, like adults, to learn about other cultures and conceptions of the good while remaining confident in their own way of life. It can be argued, in fact, that being in a school that teaches respect for other traditions may actually strengthen children's commitments to their own traditions.83 When a student sees other students learning respect for his way of life, he feels proud and learns to see his culture as something worthy of respect. Also, in contrasting his own traditions with others', he learns what is distinctive and noteworthy about his own. Finally, in a school that teaches about a diversity of traditions, children see their own traditions taught to other people from an external perspective, and thus see how and why others are led to understand and respect their own ways of life. In all of these ways, the detached liberal school may actually advance children's senses of embeddedness at the same time as helping them to develop the capacity for choice.

The third contention—that the breakdown of family, neighborhood, social, and cultural institutions has deprived children of important sources of identification and cultural membership—poses the most difficult challenge to liberalism and the liberal educational ideal. This is in large part because the loss of ‘social capital’ in modern societies is closely connected (p.96) to the development of ‘modern culture’. Many Western democracies, such as Great Britain and the United States, have both spawned and been partially taken over by a culture of modernity—by a way of life that derives from, plays off of, and perpetuates the distinctively ‘modern’ elements of contemporary society. It is this fact, that liberal democracy tends to breed a culture of modernity and discourages traditionalism, and not the fear that liberal democracy breeds no cultural identification, that I believe really lies at the heart of some of the criticisms articulated above. The claim that the conditions of modern life have produced cultural incoherence and identity confusion, in other words, should properly be understood as a different claim in disguise: namely, that the conditions of modern life have fostered cultural identification, but it is identification with an intrinsically distasteful (because valueless, unworthy, secular, reason‐obsessed, modern, Enlightenment‐driven, anti‐traditional, capitalist, etc.) culture of modernity that should be rejected. Under this interpretation, the detached school is seen as yet another agent of liberal cultural imperialism, one which not merely withdraws from the task of reinforcing parents' attempts at cultural reproduction, but actively substitutes its own cosmopolitan, culturally opportunistic, and choice‐ridden set of values in place of parents' conceptions of the good.

It is hard to know how seriously to take this assessment. On the one hand, I have made clear my belief that liberal principles do not privilege parents' chances of passing on their cultural traditions over other groups' opportunities to do the same. Section 2.3 acknowledged that children do in general have a better chance of developing a stable identity if it is done within the context of the home, and if this attempt by their parents is respected by outsiders. Thus, liberalism encourages parents to provide their children with such an upbringing, and respects the integrity of the family in part for this reason. In addition, liberalism may even perceive it to be a loss for aesthetic or other reasons if too many children abandon a particular way of life and that culture goes defunct. Many liberals do view the fact of pluralism as an active good, rather than merely as a value‐neutral condition of the world. But it is not an injustice toward parents or their children if children do not end up identifying with their parents' way of life, so long as they end up turning toward some other set of substantive cultural forms and practices instead. As I have argued before, liberalism does not have the responsibility to rescue or artificially preserve cultures that would otherwise die out or change character, so long as a sufficient range remains available to promote future autonomy. Autonomy is achievable despite not identifying with one's parents' way of life, and ultimately it is the development of children's autonomy that matters to liberal theory in this context.

(p.97) In addition, as Jeremy Waldron cogently points out, it may be both historically and normatively misrepresentative to suggest that anyone growing up in a multicultural society is truly grounded in one culture—including adults who grew up decades ago: ‘in the modern world—in the wake of imperialism, global communication, world war, mass migration, and frequent flying—people are not able to absorb their cultures whole and pure. . . . In what seems like chaos or anarchy . . . each of us makes a discovery about herself. We are not so lost or timid that we cannot build an identity for ourselves without the guiding hand of a pre‐established framework.’84 From this perspective, it does not make sense to speak of preserving ‘a culture’, or to speak of children growing up in and adopting ‘the same culture’ as their parents. Every person's identity is a mélange, Waldron argues, of disparate cultural influences. Although modernity and especially self‐conscious cosmopolitanism may represent two particularly distinctive forms of mélange, it is wrong to contrast it with other, more unified constructions of individual identity.

That said, on the other hand, modern culture itself really may not be valuable. Such a judgement is not necessarily illiberal: insofar as liberals (and the liberal state) would fight against a child's coming to identify constitutively with a culture of criminality even if it fostered her cultural coherence, liberalism must evaluate the worth of cultures, at least in a very basic way. Thus, if modern ‘culture’ really is as shallow and superficial as many of its detractors claim, then liberal schools should not self‐consciously create the conditions that promote it.

Furthermore, even if modernity can provide a valuable source of cultural identification, there will none the less be cases in which children will not develop the sense of embeddedness and identity that they need to be intentional, directed, and ultimately autonomous agents. First, it is possible that because of the fragmentation of families in certain areas, especially when combined with a lack of cultural, social, and economic resources in the community as a whole, some children will be deprived of the attention and cultural participation that are necessary to develop a sense of cultural coherence and embeddedness. Whether or not identity is a mélange, after all, children need something (or many things) from which to build an identity. If few stories, visions of history, rituals, rites of passage, etc. are shared and practiced as a child grows up—whether these stories and practices derive from one culture or many cultures—then it will be hard for a child to feel embedded in any discernible way (or ways) of life. Such children therefore require an alternative source of participation and identification if they are to develop the embeddedness that is a prerequisite for autonomy. Insofar as the ideal of the detached school is not designed to acculturate children in this way, it fails to meet these children's needs. Such (p.98) children may not benefit from the ideal of liberal education as it now stands.

The second group of children who may have difficulty forming a stable, healthy sense of identity are those who come from cultures and traditions that are in a minority or are in some way discriminated or selected against in the larger society. Children who suffer discrimination, of course, often experience lowered self‐esteem, lack of confidence, and shame about their culture or background. Although many children rightly fight back and overcome this disadvantage, these feelings are nonetheless at the time deleterious to the construction of a healthy and stable self.85 Even when discrimination is not a factor, children from minority backgrounds can still feel excluded from or on the outskirts of mainstream society. This feeling would not be problematic from the perspective of identity formation if minority children felt fully comfortable in their ‘home’ culture, i.e. in the culture of their parents. But such membership is often equally elusive, as children evaluate their parents' practices and customs according to the standards of the self‐styled ‘normal’ world beyond. Able to judge each culture with the eyes of an outsider, such children fail to become full ‘insiders’ in either, torn between the conflicting demands and practices of their overlapping yet competing worlds. As I suggested in the conclusion to section 3, this problem is exacerbated by the informal curriculum of so‐called ‘detached’ schools which, via their structure and language, in reality (and inevitably) promote the norms of the dominant culture. Although many people successfully adopt composite identities, taking and rejecting elements from both cultures at will, many others may fail ever to develop a sense of personal strength, identity, and belonging in which they feel confident and from which they can act.86 The detached, liberal school would seem to fail these children as well.

Thus, although I am not sympathetic to some of the objections raised above about cultural identification—I believe that children need to develop a capacity for authentic or autonomous choice, that age‐appropriate education need not trade off choice for membership (or vice versa), that many children even in this day and age do have a stable sense of identity, and that modernism can sometimes provide an adequate replacement for traditionalism—it must be acknowledged that some children do and will lack a sense of embeddedness or cultural connectedness. How must the liberal educational ideal be altered to take this problem into account? I would reiterate that it should not retreat from developing children's capacities for choice. The liberal school must still teach children toleration, critical thinking skills, respect for other points of view, etc. The qualities of the detached school are still important in this respect, as are the skills, habits, and beliefs that were listed and discussed in section 2.4. But it is (p.99) possible that, in deference to people's need for cultural embeddedness and membership, and in recognition of the pre‐eminence of identity over autonomy (insofar as the former is a precondition of the latter), the liberal school should be more accommodating to particular cultural traditions. It might be argued that if education for choice (via the detached school) and education for cultural coherence (via a culturally ‘connected’ school) conflict, then the latter should take precedence over the former. If it is possible, especially, to teach children the knowledge, skills, and dispositions necessary for civic membership in a liberal polity, it could be suggested that schools should not attempt to educate for autonomy as well. Chapter 4 addresses this possibility by examining whether a state can educate for citizenship alone while respecting students' cultural convictions and safeguarding their cultural membership. The arguments examined there will point to a final conclusion about the connections among autonomy, choice, culture, and citizenship in section 5.1.


(1.) Judith Shklar, ‘The Liberalism of Fear’, in Nancy Rosenblum, (ed.), Liberalism and the Moral Life (Cambridge, Mass.: Harvard University Press, 1989), 21.

(2.) It is a sufficient and not a necessary condition because lives with which one does identify may also be bad. But the liberal state must remain agnostic about this; it may not judge the relative merits of competing conceptions of the good which are autonomously chosen. Thus, this section ignores other conditions of bad lives as lying outside the purview of the liberal state.

(3.) It is worth noting that this formulation is compatible with the weakly perfectionist liberal stance because it allows individuals to lead non‐autonomous lives—the state simply cannot force individuals to lead such lives.

(4.) As Karl Popper agrees: ‘A certain amount of state control in education . . . is necessary, if the young are to be protected from a neglect which would make them unable to defend their freedom, and the state should see that all educational facilities are open to everybody. But too much state control in educational matters is a fatal danger to freedom since it must lead to indoctrination’ (Karl Popper, The Open Society and Its Enemies, vol. 1 (London: Routledge and Kegan Paul, 1952), 117).

(5.) Paradoxically, the act of positing autonomy as the aim of education may be especially dangerous. Autonomy has a dangerous propensity to ‘drift’ politically toward authoritarianism, as Isaiah Berlin outlines in Four Essays on Liberty: ‘My thesis is that historically the notion of “positive” liberty—in answer to the question “Who is master?”—diverged from that of “negative” liberty, designed to answer “Over what area am I master?”; and that this gulf widened as the notion of the self suffered a metaphysical fission into, on the one hand, a “higher”, or a “real”, or an “ideal” self, set up to rule “lower”, “empirical”, “psychological” self or nature, on the other. . . . in the course of (p.186) this process, what began as a doctrine of freedom turned into a doctrine of authority and, at times, of oppression, and became the favourite weapon of despotism, a phenomenon all too familiar in our own day’ (Isaiah Berlin, Four Essays on Liberty (Oxford: Oxford University Press, 1969), p. xliv). If liberal education is structured as sect. 2.4 suggests, then it would be quite easy for the state to veer over time to a more rigid and demanding notion of “autonomy” that would rightly fall afoul of Berlin's criticisms concerning positive liberty and tyranny. The point is not merely that the state has undue power to interpret the aims of education by its own lights, but that the drift toward tyranny is more likely precisely because it is justified in terms of autonomy—i.e. because citizens are less likely to be vigilant in judging acts (such as education) conducted under the name of freedom than under the banner of other educational goals. Thus, it is not only the imbalance of power between the educative state and the parents that raises the specter of state tyranny, but also the fact that autonomy—which according to Berlin has historically been a byword for authoritarianism—is the justificatory aim.

(6.) Amy Gutmann, Democratic Education (Princeton: Princeton University Press, 1987), 69.

(7.) John Stuart Mill, On Liberty (Harmondsworth: Penguin, 1974; 1st pub. 1859), 177. John Locke responds similarly to William Penn's Frame of Government order to ‘Erect and order all public schools’, commenting that it imposes ‘The surest check upon liberty of conscience, suppressing all displeasing opinions in the bud’ (quoted in Maurice Cranston, John Locke (Oxford: Oxford University Press, 1985), 262).

(8.) Mill, On Liberty, 176.

(9.) Mill, On Liberty, 176.

(10.) Bonna Devora Haberman, ‘What is the Content of Education in a Democratic Society?’, Journal of Philosophy of Education, 28/2 (1994), 185.

(11.) William Galston uses this same reasoning to reject the positing of autonomy as such as the aim of the liberal state: ‘To place an ideal of autonomous choice—let alone cosmopolitan bricolage—at the core of liberalism is in fact to narrow the range of possibilities available within liberal societies. In the guise of protecting the capacity for diversity, the autonomy principle in fact represents a kind of uniformity that exerts pressure on ways of life that do not embrace autonomy’ (William Galston, ‘Two Concepts of Liberalism’, Ethics, 105/3 (Apr. 1995), 523).

(12.) Although this argument has treated pluralism solely as an instrumental good, it is in no way harmed—and is in fact strengthened—if pluralism is taken to be an intrinsic good as well. Likewise, the fact that I have focused on the instrumental benefits of pluralism (as promoting autonomy) does not mean that I reject, or even take a position on, the notion of pluralism as an intrinsic good.

(13.) See Joseph Raz, The Morality of Freedom (Oxford: Oxford University Press, 1986), esp. ch. 14.

(14.) ‘The “pluralism” commonly identified with the state of families is superficial (p.187) because its internal variety serves as little more than an ornament for onlookers. Pluralism is an important political value insofar as social diversity enriches our lives by expanding our understanding of differing ways of life. To reap the benefits of social diversity, children must be exposed to ways of life different from their parents and—in the course of their exposure—must embrace certain values, such as mutual respect among persons, that make social diversity both possible and desirable’ (Gutmann, Democratic Education, 33).

(15.) Lawrence A. Cremin, Popular Education and its Discontents (New York: Harper and Row, 1990), 64.

(16.) Cremin, Popular Education and Its Discontents, 59.

(17.) Cremin, Popular Education and Its Discontents, 61.

(18.) James S. Coleman, Equality and Achievement in Education (London: Westview Press, 1990), 128. The role of parent involvement in mediating children's education is addressed at much greater length in sect. 2 of this chapter.

(19.) Wisconsin Department of Public Instruction, Families and Education: An Educator's Resource for Family Involvement (Madison, Wisc.: Wisconsin Department of Public Instruction, 1991), 6. For just a small sampling of American studies demonstrating a causal link between parent involvement and student achievement, see the following: Association for Supervision and Curriculum Development, Parent Involvement (Alexandria, Va.: Association for Supervision and Curriculum Development, 1995); Joyce L. Epstein, ‘Effects on Student Achievement of Teachers' Practices of Parental Involvement’, in Advances in Reading/Language Research, vol. 5 (Greenwich, Conn.: JAI Press, 1991), 261–76; Paul G. Fehrmann, Timothy Z. Keith, and Thomas M. Reimers, ‘Home Influence on School Learning: Direct and Indirect Effects of Parental Involvement on High School Grades’, Journal of Educational Research, 80/6 (Aug. 1987), 330–7; Thomas Kallaghan et al., The Home Environment and School Learning: Promoting Parental Involvement in the Education of Children (San Francisco: Jossey‐Bass, Inc., 1993); Janine Bempechat, ‘The Role of Parent Involvement in Children's Academic Achievement: A Review of the Literature’, Trends and Issues, no. 14 (New York: ERIC Clearinghouse on Urban Education, June 1994); Anne T. Henderson, Carl L. Marburger, and Theodora Ooms, Beyond the Bake Sale: An Educator's Guide to Working with Parents (New York: National Committee for Citizens in Education, 1986); and Nancy Feyl Chavkin, (ed.), Families and Schools in a Pluralistic Society (Albany, NY: State University of New York Press, 1993).

Far fewer British studies exist, in part because ‘parent involvement’ has been understood in the recent past to mean parental choice about schools, as opposed to involvement in their children's actual learning. Such studies as do exist, however, include: Barbara Tizard et al., Young Children at School in the Inner City (London: Lawrence Erlbaum, 1988); Central Advisory Committee for Education, Children and their Primary Schools (Plowden Report) (London: HMSO, 1967); Jack Tizard, W. N. Schofield, and J. Hewison, ‘Symposium: Reading—Collaboration between Teachers and Parents in Assisting Children's (p.188) Reading’, British Journal of Educational Psychology, 52/1 (1982), 1–15; Paul Widlake and Flora Macleod, Raising Standards: Parental Involvement Programmes and the Language Performance of Children (Coventry: Community Education Development Centre, 1984); Sandra Jowett and Mary Baginsky with Morag MacDonald MacNeil, Building Bridges: Parental Involvement in Schools (Windsor: NFER‐Nelson, 1991); and Miriam E. David, Parents, Gender, and Education Reform (Cambridge: Polity Press, 1993).

(20.) James Coleman and Thomas Hoffer, Public and Private High Schools: The Impact of Communities (New York: Basic Books, 1987), 52.

(21.) Barry Rutherford, (ed.), Creating Family/School Partnerships (Columbus, Ohio: National Middle School Association, 1995), 54.

(22.) A representative sample of studies would include: Joyce Epstein and Susan Dauber's discussion of ‘School Programs and Practices of Parent Involvement in Inner‐city Elementary and Middle Schools’, Elementary School Journal, 91/3 (1988), 289–305; Anne T. Henderson, (ed.), The Evidence Continues to Grow: Parent Involvement Improves Student Achievement (Columbia, Md: National Committee for Citizens in Education, 1987); David Armor et al., Analysis of the School Preferred Reading Program in Selected Los Angeles Minority Schools (Santa Monica, Calif.: Rand Corporation, 1976); Council of Chief State School Officers, A Guide for State Action: Early Childhood and Family Education (Washington, DC: Council of Chief State School Officers, 1988); and James P. Comer, ‘Educating Poor Minority Children’, Scientific American, 259/5 (Nov. 1988), 2–8. Dozens if not hundreds of other studies exist which prove the same point.

(23.) Coleman and Hoffer, Public and Private High Schools; James Coleman, Thomas Hoffer, and Sally Kilgore, High School Achievement: Public, Catholic and Other Private Schools Compared (New York: Basic Books, 1982).

(24.) Her five levels of involvement are identified as: basic parental obligations, basic school obligations, parent involvement at school, parent involvement in learning at home, and parent involvement in decision‐making, governance, and advocacy. More recently, she has added a sixth form of involvement at the level of community collaboration. See Joyce L. Epstein, ‘School/Family/ Community Partnerships: Caring for the Children We Share’, Phi Delta Kappan, 76/9 (May 1995), 701–12; School and Family Partnerships: Report No. 6 (Baltimore, Md.: Center on Families, Communities, Schools, and Children's Learning, Johns Hopkins University, 1992); ‘How Do We Improve Programs of Parent Involvement?’, Educational Horizons, 66/2 (1988), 58–9.

(25.) Anne T. Henderson, ‘Parents Are a School's Best Friends’, Phi Delta Kappan, 78 (Oct. 1988), 150.

(26.) Vivian R. Johnson, ‘Parent Centers Send a Clear Message: Come Be a Partner In Educating Your Children’, Equity and Choice, 10/2 (Winter 1994), 42–4; Leon Lynn, ‘Building Parent Involvement’, Practitioner, 20/5 (May 1994), 2.

(27.) Barbara Tizard, Jo Mortimore, and Bebb Burchell, ‘Involving Parents from Minority Groups’, in John Bastiani, (ed.), Parents and Teachers 2: From Policy to Practice (Windsor: NFER‐Nelson, 1988), 77, 83.

(28.) Daphne Johnson and Joyce Ransom, ‘Family and School—the Relationship Reassessed’, in Bastiani (ed.), Parents and Teachers 2, 58–71; Hazel Joucks, ‘Increasing Parent/Family Involvement: Ten Ideas That Work’, NASSP Bulletin (April 1992), 19–23.

(29.) Margaret Finders and Cynthia Lewis, ‘Why Some Parents Don't Come to School’, Educational Leadership, 51 (May 1994), 53.

(30.) See Concha Delgado‐Gaitan, ‘Involving Parents in the Schools: A Process of Empowerment’, American Journal of Education, 100/1 (Nov. 1991), 20–46; Tizard, Mortimore, and Burchell, ‘Involving Parents from Minority Groups’, 83.

(31.) As a result, they are also likely to be successful only when teachers are given the resources and support they need to contact parents regularly. Such support might include smaller classes, extra planning time built into the school day specifically for meeting with parents, compensation for frequent evening or weekend PTA or coaching duties, and even simply a telephone in or close to the classroom (a luxury afforded to few teachers) so that teachers can easily and quickly make phone calls to parents between classes.

(32.) Norm Fruchter, Anne Galletta, and J. Lynne White, ‘New Directions in Parent Involvement’, Equity and Choice, 9/3 (Spring 1993), 38–9.

(33.) Delgado‐Gaitan, ‘Involving Parents in the Schools’, 21.

(34.) See Tizard, Mortimore, and Burchell, ‘Involving Parents from Minority Groups’.

(35.) Norris M. Haynes and James P. Comer, ‘The Yale School Development Program: Process, Outcomes, and Policy Implications’, Urban Education, 28/2 (July 1993), 167.

(36.) ‘Character education’ has inspired a great deal of interest and writing in America, as well as less but still significant interest in Britain, where it is often linked with citizenship education. See Patricia White, Civic Virtues and Public Schooling: Educating Citizens for a Democratic Society (London: Teachers College Press, 1996); Thomas Lickona, Educating for Character: How Our Schools Can Teach Respect and Responsibility (New York: Bantam, 1992); Thomas Lickona, What Is Good Character? And How Can We Develop It In Our Children? (Bloomington, Ind.: Poynter Center, 1991).

(37.) To the extent that parents or children value additional educational experiences that are compatible with the liberal educational ideal, schools should be varied and flexible enough to be able to provide such experiences, and to be chosen by interested families. I discuss school choice and diversity in detail in sect. 5.2.

(38.) James Coleman, Thomas Hoffer, and Sally Kilgore, Public and Private Schools (Washington, DC: National Center for Education Statistics, 1981), and Coleman, Hoffer, and Kilgore, High School Achievement. Coleman and Hoffer refined their argument in Public and Private High Schools: The Impact of Communities, to which I primarily refer in the rest of this section. All studies controlled, of course, for student's initial ability, family background, etc.

(39.) See two special issues of the journal Sociology of Education: 55/2 and 3 (Apr./July 1982) and 58/2 (Apr. 1985). See especially Karl Alexander and (p.190) Aaron M. Pallas, ‘School Sector and Cognitive Performance: When is a Little a Little?’, Sociology of Education, 58/2 (Apr. 1985), 115–28, and J. Douglas Willms, ‘Catholic‐School Effects on Academic Achievement: New Evidence from the High School and Beyond Follow‐up Study’, 98–114 of same issue.

(40.) Coleman and Hoffer, Public and Private High Schools, 7.

(41.) Coleman and Hoffer also mention a fourth advantage of functional communities for schools, that being the reinforced authority that schools and teachers enjoy in such communities. ‘The absence of value dominance in a school, as exists in a value‐heterogeneous residential area once the functional community has vanished, means that standards and demands can no longer be imposed by teachers, as is true where there is a functional community or where the teachers' values are shared by students. Rather, the demands, the efforts put forth, and the sets of standards must arise as a negotiated compromise between teacher and students, a compromise that in some settings may be very different from that which would be imposed by the same teacher if that teacher had the authority, legitimate in the students' eyes, to do so’ (Coleman and Hoffer, Public and Private High Schools, 62). Perpetuation of authority based on homogeneity of values and replication of traditional hierarchies, however, is not desirable within a liberal, autonomy‐driven system of education.

(42.) Coleman and Hoffer, Public and Private High Schools, 7.

(43.) Coleman and Hoffer, Public and Private High Schools, 221 ff. and passim; Claire Smrekar, The Impact of School Choice and Community (Albany, NY: State University of New York, 1996), 3–4.

(44.) ‘[S]chool administrators can initiate events and activities designed specifically to bring together parents of children in the school. Many administrators know that, by creating collective strength among parents, they create a force that can be a nuisance; less often do they recognize that this collective strength can be a resource that both eases their task of governing a school and benefits the children who attend it’ (Coleman, Equality and Achievement in Education, 323).

(45.) Coleman and Hoffer, Public and Private High Schools, 55. See also Delgado‐Gaitan, ‘Involving Parents in the Schools’.

(46.) Coleman and Hoffer, Public and Private High Schools, 8.

(47.) Coleman and Hoffer, Public and Private High Schools, 214.

(48.) Anthony S. Bryk, Valerie E. Lee, and Peter B. Holland, Catholic Schools and the Common Good (Cambridge, Mass.: Harvard University Press, 1993), especially chs. 5 and 11.

(49.) Bryk et al., Catholic Schools, 132–3.

(50.) Bryk et al., Catholic Schools, 133.

(51.) Bryk et al., Catholic Schools, 133.

(52.) Bryk et al., Catholic Schools, 134–5.

(53.) See Ivan Illich, Deschooling Society (London: Calder and Boyars, 1971); Ian Lister, (ed.), Deschooling (Cambridge: Cambridge University Press, 1974); Everett Reimer, School is Dead (Harmondsworth: Penguin Books, 1972).

(54.) Illich, Deschooling Society, 31.

(55.) Nell Keddie, ‘Classroom Knowledge’, in Michael F. D. Young, (ed.), Knowledge and Control (London: Collier‐Macmillan Publishers, 1971), 133–60.

(56.) Keddie, ‘Classroom Knowledge’, 139.

(57.) Keddie, ‘Classroom Knowledge’, 150.

(58.) Keddie, ‘Classroom Knowledge’, 155–6. Keddie gives evidence for this claim earlier in the article, citing the following interview:

‘A pupils are generally more sensitive to what they have been told about the course. Thus when I asked them what they thought of the course, typical responses were:

  • “It's very good; you can disagree with the teacher.”

  • “You can link up subjects.”

  • “You can think out things for yourself.”

  • “It's good for learning how to work at university.”

It seems likely they had accepted definitions received from teachers, because when I asked these pupils to tell me about a time they had disagreed with the teacher or about a time when they had been able to link up between subjects, they could recall no instances of either. There appears to be a discrepancy between their definition and their experience of the course of which they were not aware’ (Keddie, ‘Classroom Knowledge’, 150).

(59.) Eamonn Callan makes a similar point: ‘Teaching presupposes that the student defer to the teacher's judgment with regard to the material being taught, and since learning is necessarily directed through teaching in the school, it follows that this attitude will be a pervasive feature of students' participation in any form of that institution. The necessity for this pervasive attitude lends some plausibility to the deschoolers' claims about the psychological impotence any form of schooling is liable to induce’ (Eamonn Callan, Autonomy and Schooling (Montreal: McGill‐Queen's University Press, 1988), 94).

(60.) The term ‘hidden curriculum’ was coined by Philip W. Jackson, Life in Classrooms (New York: Holt, Rinehart and Winston, 1968).

(61.) Keddie, ‘Classroom Knowledge’, 151.

(62.) See Jackson, Life in Classrooms, Samuel Bowles and Herbert Gintis, Schooling in Capitalist America (London: Routledge and Kegan Paul, 1976); Martin Cornoy and Henry M. Levin, Schooling and Work in the Democratic State (Stanford: Stanford University Press, 1985); Joel Spring, Education and the Rise of the Corporate State (Boston: Beacon Press, 1971).

(63.) Bowles and Gintis, Schooling in Capitalist America, 9.

(64.) Jackson, Life in Classrooms, 36. Powell, Farrar, and Cohen corroborate this idea with the following striking depiction of the ‘unspecial’: ‘The allotted role of the unspecial often seems to be to call attention to the specialness of others. They made up, for example, an important part of the audience for Friday night football. Spectatorship was their lot, and most played the role happily. . . . One teacher even thought that spectatorship, in the stands and in the classroom, was sound preparation for life. Such students would play roles in life requiring a “passive position.” While in school they were “already orienting themselves (p.192) to the environment in which they're going to be.” It was sound pedagogy to know that “they respond better if you don't try to force them to take over” ’ (Arthur G. Powell, Eleanor Farrar, and David K. Cohen, The Shopping Mall High School (Boston: Houghton Mifflin, 1985), 176).

(65.) Jackson, Life in Classrooms, 36. Or again: ‘Yet the habits of obedience and docility engendered in the classroom have a high pay‐off value in other settings. So far as their power structure is concerned classrooms are not too dissimilar from factories or offices, those ubiquitous organizations in which so much of our adult life is spent. Thus, school might really be called a preparation for life, but not in the usual sense in which educators employ that slogan’ (Jackson, Life in Classrooms, 33).

(66.) Bowles and Gintis, Schooling in Capitalist America, 125.

(67.) Who these agents are is, of course, a pertinent question, but one that seems to go unanswered by Bowles and Gintis. While they offer compelling a posteriori examples of how school reform has tended to help capitalism, they are much less good at identifying who ensures that this happens—i.e. who the agents of schools' subservience to capitalism are. As a result, there is an air of conspiratorial fantasy surrounding their analysis (as is often the case with functionalism). I am doubtful about how seriously we must take such scenarios in developing our own theory of liberal education.

(68.) ‘Returning to the situation in our schools, we can see that if students are to face the demands of classroom life with equanimity they must learn to be patient. This means that they must be able to disengage, at least temporarily, their feelings from their actions. It also means, of course, that they must be able to re‐engage feelings and actions when conditions are appropriate. In other words, students must wait patiently for their turn to come, but when it does they must still be capable of zestful participation. They must accept the fact of not being called on during a group discussion, but they must continue to volunteer’ (Jackson, Life in Classrooms, 18).

(69.) As David Paris comments, ‘One should be cautious about ascribing too much—for good or ill—to the hidden curriculum. It is not clear exactly what it is, and it is odd that something hidden is often talked about as if it is well understood’ (David C. Paris, ‘Moral Education and the “Tie That Binds” in Liberal Political Theory’, American Political Science Review, 85/3 (1991), 886). I might add that it is odd that something which is so well understood is often talked about as if it is hidden.

(70.) Illich, Deschooling Society, 81. His proposals might seem more plausible in an updated form encompassing use of the Internet and the World Wide Web. I suspect that Illich would reject this move, however, because computers and high technology are inaccessible to poor people. He rejects the use of television (1970s high technology) in favor of tape recorders on the same grounds.

(71.) ‘There is in effect really no point in debating whether there should be moral education in the schools. What needs to be debated is what form this education should take since we believe that moral education, in fact, “comes with the territory” ’ (David Purpel and Kevin Ryan, ‘It Comes With The Territory’, in (p.193) David Purpel and Kevin Ryan, (eds.), Moral Education . . . It Comes With The Territory (Berkeley, Calif.: McCutchan, 1976), 44).

(72.) White, Civic Virtues and Public Schooling; Betty Sichel, Moral Education: Character, Community, and Ideals (Philadelphia: Temple University Press, 1988).

(73.) Gutmann, Democratic Education, 53.

(74.) Colin Shrosbree, Public Schools and Private Education (Manchester: Manchester University Press, 1988), p. vi.

(75.) Gail R. Benjamin, Japanese Lessons (New York: New York University Press, 1997).

(76.) See Robert D. Morrow, ‘The Challenge of Southeast Asian Parental Involvement,’ Principal (Jan. 1991), 20–2.

(77.) I am grateful to Stuart White for impressing upon me the importance of this argument, and for convincing me to devote space to this problem.

(78.) William Galston, Liberal Purposes (Cambridge: Cambridge University Press, 1991), 255.

(79.) For clarity's sake, it should be noted that this argument is distinct from the commonly articulated (and stereotyped) communitarian argument that liberals do not value cultural coherence and membership. This argument acknowledges that liberalism does value cultural embeddedness, but suggests that liberal policies inadvertently fail to protect it.

(80.) Galston, Liberal Purposes, 255. See also Shelley Burtt, ‘In Defense of Yoder: Parental Authority and the Public Schools’, in Ian Shapiro and Russell Hardin, (eds.), Nomos XXXVIII: Political Order (New York: New York University Press, 1996), 426.

(81.) Jean Piaget, Judgement and Reasoning in the Child, trans. Marjorie Warden (New York: Harcourt, Brace, and Co., 1928); Jean Piaget, The Psychology of Intelligence, trans. Malcolm Piercy and D. E. Berlyne (London: Routledge and Kegan Paul, 1950), ch. 5; Victor Lee and Prajna Das Gupta, (eds.), Children's Cognitive and Language Development (Milton Keynes: The Open University and Oxford: Blackwell, 1995); Patricia H. Miller, Theories of Developmental Psychology, 3rd edn. (New York: W. H. Freeman, 1993).

(82.) Benjamin Barber, An Aristocracy of Everyone: The Politics of Education and the Future of America (New York: Ballantine Books, 1992), 142.

(83.) I am indebted to Joseph Raz for pointing this out to me.

(84.) Jeremy Waldron, ‘Multiculturalism and Mélange’, in Robert K. Fullinwider, (ed.), Public Education in a Multicultural Society (Cambridge: Cambridge University Press, 1996), 106.

(85.) ‘Where social group differences exist, and some groups are privileged while others are oppressed, this propensity to universalize the particular reinforces that oppression. The standpoint of the privileged, their particular experience and standards, is constructed as normal and neutral. If some groups' experience differs from this neutral experience, or they do not measure up to those standards, their difference is constructed as deviance and inferiority’ (Iris Marion Young, Justice and the Politics of Difference (Princeton: Princeton University Press, 1990), 116).

(86.) See Salman Rushdie, Midnight's Children (New York: Knopf, 1981) for a magnificent exploration of self‐construction in a culturally divided India.