Putting Educational Equality in its Place
University of Wisconsin, Madison
Balliol College, University of Oxford
Education, in its broadest sense, mediates between the wider society and our individual prospects. How well a child will do, even what sort of person she will become, in any given society, will be significantly shaped by the kind of education she receives. That education is, in its turn, profoundly influenced by the character of the social order. Whether or not the government makes schooling compulsory and free at the point of delivery makes a difference to everyone’s prospects. If it uses schooling to indoctrinate children in a favoured religious view, they will live very different lives from those they would live if it used schooling to promote personal autonomy, or to promote an ethic of hedonistic enjoyment. So the issues of how to distribute education and what kind of education to distribute are pressing to the theorist of justice. We shall put aside the issue of what kind of education to distribute, and concentrate on the distributive rule.1
How should education be distributed? It is common to invoke the principle of educational equality - the idea that everyone should have an equally good education. But it is not at all clear what it means to say this. Some accounts of educational equality demand equal educational resources or inputs, for which per-student spending is often considered a proxy. But, of course, students are different in terms of what they need to reach any particular level of achievement; whether because they come from a disadvantaged social environment or because they have special educational needs, some students will achieve much less at a given input level than others. So an alternative account calls for equal educational outcomes or achievement; achieving this demand would require that more resources be spent on less advantaged students than on more advantaged students. A further complication is that governments only directly control expenditure on educational resources in schools and other public institutions; aiming for equal outcomes by expending compensatory resources on disadvantaged students might trigger an arms race as advantaged parents supplement with private resources to improve their own children’s achievement. Short of measures severely curtailing parental power, equal achievement would be impossible to achieve.2 These, and other problems with a principle of educational equality, have led some to endorse the recent movement in the United States preferring the rhetoric of adequacy to that of equality. Rather than demand that educational opportunities be equal, all should have an adequate education (Liu 2006; Anderson 2007; Satz, 2007, 2008).
In part this shift has been inspired by the realities of political strategy. While equality might be an appealing political ideal, adequacy is likely to get more traction in litigation that attempts to appeal to the provisions of State constitutions. But which is the better fundamental moral standard: equality or adequacy? The view in favour of adequacy gains support from the fact that few egalitarians object to all educational inequalities, even in principle, and that it seems impossible to produce strict equality of educational outcomes even if that were desirable.
The burden of this paper is to defend an ideal of educational equality. We start by providing what we take to be the intuitive argument for that ideal and elaborate the very demanding conception of educational equality which that argument supports. The bulk of the paper is taken up with exploration of two other values – benefiting the less advantaged members of society and parental liberty - which we, in common with most opponents of educational equality, believe properly constrain the pursuit of that ideal. That these other values are more important does not warrant wholesale rejection of the egalitarian principle, though it does constrain what may permissibly be done in its pursuit. We look at the practical consequences of these constraints and suggest that in fact a good deal can be done to pursue the proposed conception of educational equality while observing these constraints. As a test case, we look at the recent proposal for Weighted Student Funding, arguing that educational equality requires it and that the other values permit it. Finally, we return to the principle of educational adequacy and argue that, when properly understood, in the nested sense that we set out, a principle of educational equality is superior.
1. The Fairness Case for Educational Equality.
The intuitive case for educational equality rests on an intuition about what it takes for a competition to be fair. Modern industrial societies are structured so that socially produced rewards – income, wealth, status, positions in the occupational structure and the opportunities for self-exploration and fulfilment that come with them – are distributed unequally. Education is a crucial gateway to these rewards; a person’s level and kind of educational achievement typically has a major influence on where she will end up in the distribution of those potentially life-enhancing goods. It is unfair, then, if some get a worse education than others, because, through no fault of their own, this puts them at a disadvantage in the competition for these unequally distributed goods.
So the intuitive case for educational equality is fairness-based; more specifically, it depends on the idea that, in order to be legitimate, inequalities should result from fair procedures. The dominant understanding of educational equality in contemporary Anglo-American political discourse is meritocratic. Think of the call, in the US, to eliminate the achievement gap which, if understood strictly, demands that there should be no difference in achievement between children born into lower or higher socio-economic classes.3 In the UK, which has a quite different education system from the US but is similar in having a high degree of economic inequality relative to other wealthy democracies, successive Secretaries of State for Education have called more explicitly for the elimination of any influence of social class on educational achievement (Clarke 2003; Miliband 2004). The broad principle is what we shall dub the meritocratic conception of educational equality.
The Meritocratic Conception: An individual’s prospects for educational achievement may be a function of that individual’s talent and effort, but it should not be influenced by her social class background.
This is very demanding. Given what we know about the influence of social class on achievement, for example, it seems to require that considerably more resources be spent on educating children from lower socio-economic backgrounds than on children from more advantaged backgrounds, and that these resources be spent effectively. In other words, it appears to imply some form of weighted student funding, in which effective spending is inverse to advantage. It also strongly suggests that measures going beyond the education system should be adopted. If it is not known how to educate large numbers of children who are raised in relative poverty to the levels that can be achieved by more advantaged children in the same society, for example, the principle demands the elimination of child poverty.4 If, as some researchers argue, aspirations to educational achievement are strongly influenced by the educational level of the neighbourhood in which a child is raised, then the principle suggests measures to integrate neighbourhoods by educational level (Ainsworth 2002).
Demanding as it is, to some the meritocratic conception of educational equality may nonetheless seem insufficiently egalitarian. It is concerned to eliminate unfair inequalities in prospects for achievement between children of different class backgrounds but it is entirely silent about inequalities in prospects for achievement between children with different levels of effort and talent. If it is unfair for a child’s prospects for achievement to be influenced by her social origins, why it is fair for them to be influenced by her natural talent (which is entirely beyond her control) or level of effort (which is itself heavily influenced by familial and neighbourhood factors)? Thoughts along these lines may exert pressure in the direction of a more complete, and radical, conception of educational equality, but for present purposes we put that more radical conception to one side.
It is also important to note that, standing alone, the meritocratic conception permits, although it does not require, considerable inequality of both educational resources and educational achievement, as long as those inequalities do not track social class. For example, it is consistent with concentrating resources on those who have high levels of talent and motivation, with the aim of producing very high levels of achievement for them, while leaving those with lower levels of talent and motivation to fend for themselves with, presumably, low levels of achievement. It would be equally consistent with this conception to concentrate resources on those with very low levels of talent and motivation, in order to produce more equal levels of achievement across the board. The conception simply doesn’t tell us. We dub the conception meritocratic, because it meshes well with the demands of supporters of meritocracy to reward talent but not class background; we describe it as a conception of educational equality because it is closely connected to Rawls’s principle of fair equality of opportunity. But, alone, it rejects only one source of inequality. However, as we shall go on to show, when it is put in its proper place, together with other principles it guides us more precisely.
It is sometimes objected to the intuitive argument from the importance of fair competitions that society is not a race: there is not “one Grand Racetrack on which we are all bidden to run” (Lomasky 1987, pp.180-181). Society is indeed not a race. But, as we shall show later, it is relevantly like a race. The distribution of the benefits of social cooperation is structured to reward those who do well and penalise those who do badly in competitions they have no feasible alternative to participating in.
Notice two things about our conception of educational equality. First, it does not support a principle of equal educational resources, if that principle is understood to mean that the government spends equally on each child in school. As we have said, it seems clearly to require that the government spends more resources on children disadvantaged by their class background than on children advantaged by theirs. Of course, there is another sense in which the government, in spending additional resources on those disadvantaged by class, is attempting to achieve equality of educational resources; it is simply compensating for the inequality of educational resources provided by the family and neighbourhood. But there is no support in this conception for the idea of equal government spending per child. Second, even having put more radical ideas aside, the barriers to achieving educational equality are enormous. Its demands are such that it is hard to see how to achieve it in the US today, for example. In the subsequent sections we shall show that, when put in its proper place and weighed against other values, the meritocratic conception appears somewhat less demanding, and some of its apparently implausible implications are muted.
2. Putting Educational Equality in its Place
All we have so far, though, is a prima facie case for acting to implement the proposed conception of educational equality. Even if there are good reasons to value something, that does not mean that we should implement it wholesale, only that we should implement it as far as possible without undermining other more important values. So it is important to consider whether the conception might be susceptible to the following two objections:
The Harming the Less Advantaged Objection.
The harming the less advantaged objection says that an unfair distribution of education can work ultimately to the benefit of the less advantaged in society. Of course, they don’t get more positional advantage than they would under an equal distribution of education, but positional advantage is not all that matters: what matters ultimately is that people get to live rewarding and flourishing lives. And these are not distributed in a zero-sum game. The opportunities of the less advantaged for rewarding and flourishing lives can be enhanced by distributing education in ways that violate the meritocratic conception of educational equality. Perhaps wealthy parents could be permitted to buy unfairly unequal educational opportunity for their children, say by paying for them to attend elite private schools, or by paying for extensive private tuition. As a result, those children have a better chance of getting the college places, jobs, and status, to which all are aspiring, than other (similarly talented and hardworking) children do. But because parents can invest in their children they do so, and so the total stock of human capital in society is enhanced; the economy can then harness the productivity gains, due to that enhanced human capital, to the benefit of the less advantaged. Notice, moreover, that an unequal distribution of education might, in time, lead to the worse off having more or better education than they would otherwise have, and that this could itself yield important benefits. Education is partly a positional good, but only partly so. It also yields benefits that are nothing to do with people’s competitive position, relative to others, but accrue directly to the educated person: the enjoyment of being able to read literature, or appreciate movies, or write poetry, or read a musical instrument, or understand scientific issues. This objection might concede that educational equality is immune to the leveling down objection when that objection focuses on the positional aspect of education – its role as a competitive means to other goods. But it is susceptible to a ‘bigger picture’ version of the leveling down objection -- one that says “equal education results in the less advantaged having fewer opportunities to lead flourishing lives than they otherwise might”.
B. The Parental Liberty/Family Values objection
The second objection is expressed by Nathan Glazer (Glazer 2005, p.13) in his review of Jonathan Kozol’s recent book (Kozol 2005):
To be sure, the case for both [racial] integration and equality of expenditure is powerful. But the chief obstacle to achieving these goals does not seem to be the indifference of whites and the non-poor to the education of white and the poor…. Rather, other values, which are not simply shields for racism, stand in the way: the value of the neighborhood school; the value of local control of education and, above all, the value of freedom from state imposition when it affects matters so personal as the future of one’s children.
Parental liberty is extremely important, and if it has moral priority over our conception of educational equality, then we should not do anything in pursuit of the latter (desegregate public schools, abolish elite private schools, enforce neighbourhood diversity) that violates it.5
Before evaluating these objections, we want to make a comment about the form they take, and why we have rendered them in this form. It is commonplace in everyday political discussion to observe of distinct values that they conflict, and then to conclude that because one value should take priority the other does not matter. It is often observed, for example, that equality conflicts with freedom; some argument is then made that freedom is more important, and the inference drawn that equality should be abandoned. But the inference is simply invalid. Even if freedom did matter more than equality, still equality might be sufficiently important that we should non-freedom-violating measures that promote it. Just because one value matters more, and so should win out in cases of conflict, it does not follow that the other value matters not at all.
So we have posed the objections not as objections to the meritocratic conception of educational equality, but as objections specifically to policies designed to foster that conception. In our rendering each objection breaks down into two claims
Policy X violates/impedes/jeopardizes value Y
Value Y is more important than educational equality
Rendering them this way makes for clearer philosophical discussion. But we think it also makes it easier to map the philosophical terrain onto actual policy debates. For example, no-one uses the family values/parental liberty argument against equalized state funding, because it is hard to see how that could violate parental liberties.6 But they do use it against abolishing private schools, and against bussing for integration purposes, because it is much clearer how those policies interfere with the liberties of parents.
3. Evaluating the Harming the Less Advantaged Objection.
Let’s consider the harming the less advantaged objection. This asks us to consider the possibility, for example, that abolishing elite private schools would lead to diminished investment in the human capital of those children of the rich who have fewer than normal marketable talents, and that instead parents would spend the money they are prohibited from investing in their children bidding up the prices of naturally scarce goods; or that the resources needed to combat their powerful campaign against egalitarian policies and to prevent them from evading their effects are so great that it would have been better, for the less advantaged, to expend those resources directly on programs designed to enhance their prospects for flourishing. Adopting these measures to pursue meritocratic educational equality might, then, come into conflict with benefiting the least advantaged.
We shall not provide an argument in any detail, but we believe that in the realm of values the medium to long term prospects for all-things-considered flourishing of those who flourish least in a society is the most urgent consideration of justice, and the prospects for flourishing of the less advantaged are more urgent than those of the more advantaged, even in wealthy societies (Brighouse and Swift 2006a). In other words we believe that the second part of the standard two-part objection is correct in this case: concern for benefiting the less advantaged all things considered should, indeed, trump our principle of educational equality. So, if it were the case that prohibiting elite private schools harmed the prospects of the less advantaged in general and the least advantaged in particular, then we would oppose prohibition.
But we are somewhat skeptical that the policy in question, prohibiting elite private schools, does damage the prospects for flourishing of the less advantaged. Whether it does so depends entirely on the facts in the particular social context.
The objector has to specify the mechanism whereby an equalizing reform makes things worse for the less advantaged. In the scenarios we have laid out, the mechanisms are the incentives and motives (and hence actions) of the parents of advantaged children. and the subsequent actions of their children. Whether permitting the parents to pay for private schooling actually benefits the less advantaged depends on several factors. For example:
* Whether the gain in human capital to the privileged children yielded by private schooling exceeds the loss in human capital to those who attend government schools caused by the absence of the advantaged children from those schools and (a) the attendant detrimental impact on the quality of the learning experience in the school and (b) the loss of lobbying power and support for the schools from their parents.
* How unequal the distribution is of income and other benefits attached to competitive positions.
* How the tax-transfer system is devised or, more broadly, the way that the economy as a whole distributes the additional product between the advantaged and the less advantaged.
* How that additional product contributes to prospects for flourishing at different places in the distribution.
* The relative contribution of the competitive labor market and the non-competitive non-labor market benefits of education to prospects for flourishing.
The objections to prohibiting private schooling that we have outlined are more often presented in terms of income and wealth, rather than flourishing, because enhancing the total pool of human capital more directly increases the production of consumable resources than the production of human flourishing. The assumption behind the objection is that increased production can be readily transformed into increased human flourishing, and turned to the benefit of the less advantaged.
We are somewhat suspicious about this assumption, at least in wealthy economies structured like those of the wealthy capitalist countries in the OECD. In such economies, education influences access to positions in the occupational structure that have unequal status, income, and levels of control over work attached to them. Broadly speaking, the better your education the better your prospects for income, status, and high levels of control over your work. But inequality of status, income and control over work (in societies likes ours) itself has a detrimental impact on some people’s health states, longevity, and subjective wellbeing. People with less status and control over their work have worse health states and longevity than those with more, and the direction of explanation seems to be from the lower status and control to the lower health and longevity (Marmot 2004). Similarly, controlling for absolute level of income, those with less income than others report lower levels of subjective wellbeing; again, the direction of explanation seems to be from the income inequality to the lesser wellbeing (Layard 2005). Subjective wellbeing has a non-linear relationship to economic growth: in wealthy countries people seem to have continued reporting increased wellbeing in line with economic growth up to a certain point, after which economic growth continues but subjective wellbeing reaches a plateau (Frank 2000; Lane 1998). Good health and longevity are both components of flourishing, on any plausible account, and indicators of it and subjective welfare reports are indicators of flourishing. We suspect that they are quite good indicators of flourishing (but there is certainly room for disagreement about how good). To the extent that this is true, relative position is more important than absolute wealth in determining one’s prospects for flourishing.
It is obvious that the putative increased production yielded by, for example, the unfairly superior education of the privileged in elite private schools will only improve the prospects for flourishing of the less advantaged if some of that product finds its way into their hands (because the tax-benefit system or the design of the economy facilitates this). For example, a story can be told on which the increases in life-expectancy across the income profile which have continued long after subjective wellbeing reports ceased to improve with growth is connected to the growth that inequality has wrought, and these increases have, presumably, improved prospects for flourishing in one respect. This observation would constitute a partial defense of economic growth, and whatever educational inequality was necessary to generate it, though not of the educational inequality that just happened to accompany it. But even if the increased production does redound to the benefit of those whose prospects are worse in some respects, the negative effects, on their prospects for flourishing, of the inequalities they suffer might off-set those benefits. The challenge to the defender of particular mechanisms that violate the meritocratic conception of educational equality is to show precisely why it is that they are necessary for improving the all things considered prospects for flourishing of the less advantaged. We suspect that, in most cases in wealthy industrial countries, it will be hard to make a plausible case for educational inequality on these grounds.
The story might be very different in poor societies, of course. In societies in which it is clear that increased economic growth yields improved prospects for flourishing throughout the population, perhaps because basic material, medical, and educational needs have not yet been met, economic growth is a tremendously urgent imperative. Most of the societies in the history of the world so far fit this description -- material resources have an enormously important impact on flourishing. In those societies there may be a great deal to be said for the idea that an unfair distribution of education could, over time, benefit the least advantaged all things considered (Tooley 2005, 2007). In such circumstances, a society would be justified in tolerating and implementing such educational inequalities as have the best prospects of benefiting the less advantaged in terms of their prospects for flourishing. Education matters a great deal, but it is not all that matters in assessing the quality of a life, and all-things-considered-flourishing is (by definition) all that matters. The claim that an inequality really benefits the least advantaged relative to the alternative feasible arrangement that is the next best for them is, in our view, a successful justification of that inequality to them.
Which inequalities those are – how big, who benefits from them - will depend on the circumstances, for it is the circumstances that determine which arrangements are feasible. In some contexts, it might be necessary, for the sake of economic growth, to tolerate the nepotistic endowment of competitive advantage by relatively advantaged parents on their children, but it is important to see that in such cases it is the motivations of parents that are being taken as given, as determining the feasible set, and hence as grounding the case for that particular kind and extent of educational inequality. On its own, the growth case for educational inequality lends no support to the idea that parents should be free to invest in their own children. Only in combination with a further claim about how differential investment in children will be more efficient if carried out by parents than by the state, which might choose rather to direct more-than-equal educational resources towards those children most likely to yield productive return to the extra investment, does this become an argument for that idea.
So we doubt that in wealthy societies this is a successful objection to measures such as prohibiting elite private schools, even though the objection correctly identifies a value that is more important than educational equality. We may be wrong, of course; it depends on the facts. But two kinds of argument warrant particular suspicion. In the first, the person pressing the objection is simply insincerely invoking a correct value in objecting to an egalitarian policy. Consider, for example, someone who makes this objection to a policy of taxing elite private schooling, but simultaneously supports a political party that shifts the relative tax burden away from high earners and toward middle- and low earners. Such a concatenation of arguments is not uncommon, and certainly it is possible coherently to favor such a distribution of the tax burden on the grounds that it will do better even by the less advantaged in the very long run, Still, it can be hard to take seriously the harming the less advantaged objection to educational equality when it comes from someone who otherwise expresses hostility to policies that benefit the less advantaged. The objector simply seems to be arguing in bad faith, unless they can accompany their argument with a powerful case for expecting the benefits to reach the worse off without any action by the state. The second case is even more widespread. This is the case in which the person making the argument is simply mistaken about the effects of policy in question. We have already suggested that in wealthy countries it is usually the case that raising the relative educational levels of the less advantaged just is the best way to benefit them. But even if that is not true, people are often mistaken about the effects of policies. Perhaps in some circumstances prohibiting elite schooling would indeed diminish the overall pool of human capital. Still, in other circumstances the elite schools themselves might be inhibiting the development of human capital, by depriving children in non-elite schools of the benefit of the political clout of the elite parents, or the beneficial peer effects of sharing schools with their children, without, in fact, doing much to enhance the productivity of the elite children in them. This might be the case if, for example, the elite parents are motivated by snobbery and are ill-informed about the relative effects, in terms of the enhancement of human capital, of elite and non-elite schools.
Although we have focused on the fairness argument for educational equality, the preceding sentence suggests a circumstance in which a second, rather different, argument supports the meritocratic principle of educational equality. This is an efficiency argument -- it is, in many circumstances, actually inefficient for wealthy parents to be able to buy educational advantages for their children, especially when those educational advantages are positional rather than substantive in character. Suppose, for example, that instead of actually developing the human capital of the children attending them, elite private schools provide them with training in examination techniques or support in negotiating the application procedures to elite Universities. Suppose further, that those universities are gateways to occupations that have more beneficial effects for society if they are carried out better rather than worse. Allowing wealthy parents to invest in pushing their less talented children up the queue in those circumstances may be socially inefficient.
In section 1 we noted that the meritocratic conception taken alone was consistent with a wide range of educational inequalities, because it prohibits just one source of inequality, but we promised that more guidance would be provided once we put it in its proper place vis-à-vis other principles. The objection under consideration in this section depends on a deeper principle, normatively prior to the meritocratic conception of educational equality, that inequalities are justified when they redound to the benefit of the less advantaged.7 So consider the choice between two conflicting policies that are consistent with the meritocratic conception; concentrating resources on students above the median level of talent, and concentrating them on those below the median level of talent. The principle that inequalities should redound to the benefit of the less advantaged commands the policymaker to make a judgment about which of these policies is more likely to benefit the least advantaged in the context. In other words, putting the principle of educational equality in its place provides more precise guidance than the principle does on its own.
4. Evaluating the Parental Liberty Objection.
What about the parental liberty objection? Nathan Glazer says that it would be a violation of individual freedom to prevent people from spending their money on their children’s education. Similarly, he suggests, it would be a violation of their freedom to prevent them from buying houses in neighborhoods whose composition came about as a result of the voluntary choices of individuals in the housing market. In order to prevent segregation or inequality, policymakers would have engage in this kind of prevention. Blocking that kind of gift or choice inhibits freedom, and it does so in an apparently peculiar way; it singles out for prohibition the provision of something widely recognized to be of great value, while allowing the provision of more frivolous, less valuable, goods (expensive cars are fine, expensive educations are not).
Egalitarians might be tempted to respond by saying that there is no freedom at stake here, but that seems a mistake to us. Freedom really is restricted; some action or actions are specified which the parent is not free to take. The interesting question is whether she has a right to perform the action that she is prevented from taking. Many measures infringe freedom and are none the worse for that. We are barred from bribing trial judges even on behalf of our own children; candidates for political office in most countries are restricted as to how much of their own money they can contribute to their own campaigns; taxation restricts the individual’s freedom to use all of her market-earned income as she might like.
Simply saying that some measure restricts someone’s freedom does not show that it is wrong. The answer to the question ‘Why shouldn’t I be allowed to spend my money on trying to save my child from being convicted of a crime she committed?’ is that fairness requires the criminal justice system to be insulated from background inequalities of wealth. In this arena, fairness trumps freedom. The answer to the question ‘Why shouldn’t I be allowed to spend my money buying my child a superior education to that which others get?’ is that in order for it to be fair the competition for socially licensed benefits must be similarly insulated. The burden of proof is on the opponent of the measure supporting equality. Mere demonstration that some measure inhibits freedom is insufficient to impugn it. The objector must show that the measure violates some basic liberty: some freedom to which we are entitled as a matter of justice (Dworkin 1985, chapter 9; Rawls 1995). Establishing that we are entitled to a particular freedom requires one to show that it is necessary to some basic human interest in a way that gives others a duty to respect it. So the objection from parental liberty has to shift to offering grounds for the parental liberty, and showing that it can bear the weight that the objection places on it.
The best way of doing this, we think, is to posit an interest in maintaining the value of the family, and to argue that mechanisms designed to equalize or desegregate violate that interest. How powerful this move is depends on what is included in ‘the value of the family’. A plausible account will allow parents to spend a good deal of time with their children, and to express partiality toward their children in a range of ways. We surely think that reading bedtime stories to one’s own children (and not, if one doesn’t want to, to other people’s) is something one has a right to do, even at some cost to educational equality (Swift 2005). Why? If we were prevented from doing that sort of thing with our children we would be deprived of the opportunity to create and maintain a valuable familial and loving relationship with them. Similarly, it seems obvious that parents must have distinctive rights to share their values and enthusiasms with their children. They have the right to take their child to their church, and to serve them food that reflects their cultural background, as long as they are not thereby harming their children (e.g. by indoctrinating or poisoning them), and no-one else has that right. Both parent and child get something distinctively valuable from being able to share themselves with each other, and for this the parent needs a space of prerogatives with respect to her child.
In a short paper it is impossible to offer a full theory of the value of the family of the kind that would answer all the difficult questions about which equalizing policies do and do not conflict with that value. Our own account (Brighouse and Swift 2006b, 2006c) focuses on the specific value to parents and children of their enjoying an intimate relationship, such that parents share their lives with their children on a day-to-day basis and play a fiduciary role in their children’s lives. The idea of parents as fiduciaries is not at all new, but we believe that parents have a non-fiduciary (self-regarding) interest in acting as fiduciaries for their children. This does not support an interest in being able to control every aspect of a child’s upbringing and education, but it does rule out certain kinds of measure. Here is the rule of thumb for deciding whether a policy is justified:
When an activity
conflicts with some other important value like fairness/equality of opportunity AND
is not itself essential to realizing/expressing the value of the family AND
is such that removing/prohibiting it would, with appropriate other institutional measures, leave ample (perhaps just as much) space for activity realizing family values