Child-rearing With Minimal Domination: A Republican Account

Abstract

Parenting involves an extraordinary degree of power over children. Republicans are concerned about domination, which, on one view, is the holding of power that fails to track the interests of those over whom it is exercised. On this account, parenting as we know it is dominating due to the low standards necessary for acquiring and retaining parental rights and the extent of parental power. Domination cannot be fully eliminated from child-rearing without unacceptable loss of value. Most likely, republicanism requires that we minimise children’s domination. I examine alternative models of child-rearing that are immune to republican criticism.

Introduction

Contemporary civic republicanism is an influential theory about the legitimate exercise of power. Republicans are concerned about domination, which they deem objectionable and define as a ‘condition suffered by persons or groups whenever they are dependent on a social relationship in which some other person or group wields arbitrary power over them’ (Lovett, 2010: 20). Relationships of this kind are dominating whether or not such power is actually exercised to setback the interests of the subordinated party. Child-rearing is a state-sanctioned, pervasive social relationship involving an extraordinary degree of power of some human beings over others. Yet, republicans have had surprisingly little to say about parenting. Here, I aim to start filling this gap. I claim that republicanism cannot accommodate the institution of parenthood in its current form because parents hold arbitrary power over children to a degree unnecessary to the fulfilment of children’s interests.

Instead, republicans can endorse only the minimal degree of arbitrary power that rearers need to exercise over children in order to protect their interests, though not necessarily to maximise their well-being. More than any other theory of legitimate power, republicanism requires a profound reform of child-rearing. Republicanism can criticise parenting in its current form while staying clear of the counterintuitive assumptions and conclusions of other theories that have revolutionary implications for child-rearing, such as Child-liberationism. ¹

A republican-friendly theory of child-rearing – as I explain in this article – is also different from liberal accounts; the main difference is that republicans are not only concerned about actual or likely abuses of power, but also about potential ones. For this reason, republican child-rearing is, in one sense, more demanding than liberal child-rearing. Since republicans, unlike liberals, assume that non-domination is a constraint on the permissible holding of power, they must require minimally dominating forms of child-rearing. Specifically, I argue that republican child-rearing requires a division of power over children as the least dominating child-rearing arrangement. Liberals, too, may endorse a division of power over children, if and when such arrangement best serves children’s interests on the whole ² but, unlike republicans, they do not identify an interest in non-domination. Given the importance they place on non-domination, republicans should be willing to sacrifice, to some extent, the fulfilment of children’s other interests, if this is needed to ensure less dominating child-rearing practices. These important differences aside, liberalism may have the resources to criticise many of the same elements of the status quo in child-rearing as republicanism, and to support reforms similar to those that, I shall argue, republicans should support. ³

Current forms of parenting involve domination in two different ways. First, existing standards for acquiring and retaining parental status are too low because they track, in part, adults’ interests in parenting; instead, they should only track children’s and some third parties’ interests. ⁴ Second, parents have rights to control their children’s lives that are more extensive than what is necessary to serve children’s interests. Yet, it is possible to restructure child-rearing in ways that minimise rearers’ arbitrary power by denying them some powers which can setback their children’s interests. For instance, republicans ought to reject parents’ power to oppose medical procedures (like vaccinations) that are beneficial for the child, and parents’ right to use their child in order to express their own values in cases when this is detrimental to the child’s interests (e.g. circumcision). More radically, because republicans oppose monopolies of power, they should deem parents’ power to exclude others from forming beneficial close relationships with the child particularly objectionable. I briefly sketch institutional arrangements that embody republican child-rearing, such as universal mandatory early day care and a state-mandated, universal secular, institution of ‘god-parenting’ whereby every child establishes long-term caring relationships with individuals other than her biological or adoptive parents.

Some preliminary caveats: ‘Right’, unqualified, refers to legal rights. ‘Parenting’ and ‘parent’ refer to the current organisation of child-rearing, in which one or two closely related custodians have a number of rights in relation with their children, to the exclusion of all other individuals. ‘Child-rearing’ and ‘rearer’ refer to practices and legal institutions that can, in principle, be legitimate. Readers are free to decide whether republican child-rearing represents merely a radical reform in parenting or a revolution large enough to warrant a different name.

The next section explains why republicans ought to pay attention to child-rearing and why parenting as we have it is dominating. The ‘Domination is not fully eliminable from child-rearing’ section is about the principled limits to the elimination of all domination from child-rearing. The ‘Republicans on child-rearing’ section explains why existing republican accounts of non-dominating child-rearing are unconvincing. The ‘Legitimate republican child-rearing’ section briefly examines alternative models of child-rearing that are immune to republican criticism.

Parenting and Domination

Traditionally, republicans have been interested in relationships of non-domination between adults. Like liberals, they are critical of coercive non-consensual relationships such as slavery. Republicans can also explain why various consensual relationships between adults, such as those with exploitative employers or abusive spouses, can involve wronging. But they have very rarely concerned themselves with non-consensual relationships between adults and children, although prominent republicans make it explicit that child-rearing can be dominating (Lovett, 2018; Pettit, 1996). This is surprising, given widespread beliefs about children’s moral status and the power aspect of any child-rearing relationship.

On the first count, children, I assume, are persons; as such, adults’ exercise of authority over them is a matter of moral concern. On the second count, power is a prominent and pervasive aspect of child-rearing. Parents hold considerable and asymmetric power over their children, and, to some extent, beneficially so. Children cannot exit all relationships with rearers without imposing enormous costs upon themselves. For this reason, some adult power over children is legitimate. In addition, parenting is a continuous occupation, hence more pervasive than most other power relationships. Even when parents delegate child-rearing decisions for limited periods of time, they never step out of their parental role – they never take responsibility breaks. Because parents control children’s lives and share their lives with their children continuously and intimately, a lot of parental power is discretionary.

It is congenial to republicanism to worry about relationships in which one side has discretionary power over a dependent person because such relationships can be arbitrary, and hence of dominating. The distinctive mark of republicanism is that it is objectionable for a power-holder to have the ability to setback, with impunity, the interests of individuals over whom power is exercised. It can be wrong to have power over somebody even when the stronger party is in a benign mood, or likes the subordinated party, or even, according to some republicans such as Philip Pettit (1999), is a morally good individual. You enjoy non-domination only if others lack the power to interfere in your life arbitrarily. So domination is an obvious threat in any relationship where parties are unequal in their dependency on each other. The relationship between children and rearers is a paradigmatic example of asymmetrical power. As I explain below, the power asymmetry in child-rearing is required for both republican and non-republican reasons. Yet, the asymmetry in power needed to meet children’s legitimate claims to adequate care introduces the possibility of arbitrary uses of that power.

What Is Arbitrary Power Over Children?

Republicans’ own remarks on legitimate child-rearing, albeit occasional, indicate that children are within the scope of their concern (Ferejohn, 2001; Lovett, 2010; Pettit, 1996). Pettit (1996: 584), for instance, notes that the parent-child relationship can involve a very high degree of domination, and welcomes ‘a culture of children’s rights and appropriate guards against child abuse’, in the absence of which ‘parents individually or jointly will enjoy subjugating power over their children’. Yet, the application of republican theory to child-rearing is not straightforward. Children lack full autonomy and so they do not have the same authority as adults to make choices about themselves. In what sense, then, can children be subject to domination?

Arbitrary, hence dominating, power, has been interpreted in different ways, the most prominent of which are procedural and democratic (Lovett, 2018). On the procedural view, arbitrariness consists in failing to constrain power by legitimate procedures that are common knowledge to all those over whom power is exercised (Lovett, 2010; Pettit, 1996). On the democratic view, power is arbitrary to the extent to which it is uncontrolled by those subject to it (Pettit, 2012). To the extent to which children lack the moral authority to decide over public procedures, or to control the exercise of power over them, these interpretations are unfit to apply to children (Macleod, 2015a).

But a third interpretation of arbitrary power, initially defended by Pettit, and recently by others (Arnold and Harris, 2017), appeals to the interests of those over whom power is exercised: ‘On the substantial view, x dominates y if, and only if, x yields power over y in a way that fails to track y’s well-being’ (Pettit, 1999). This sense of domination seems most adequate for understanding relationships with children – or, indeed, with any individuals who lack the normative authority to give or withhold consent to how power may be exercised over them, like people whose autonomy is temporarily or permanently compromised mental illness. Since the justification for the republican concern with child-rearing should be that children’s interests are as morally important as adults’, the relevant understanding of domination in relationships with children is primarily substantial. ⁵

Children’s Interests

In the case of children, domination is an ability to interfere with children’s lives in arbitrary ways that affect their interests. I draw a distinction between children’s well-being interests and their respect interests.

Well-being interests protect children’s well-being, and comprise the widely acknowledged interests in security, shelter, food, healthcare, intimate and trustful relationships and various other goods that are necessary for good childhoods – including goods that have developmental value such as education. Moreover, I assume that children have moral rights to adequate, although not maximal, levels of material, emotional and intellectual well-being. One complication that arises when looking at child-rearing through a republican lens is that the fulfilment of their interests requires resources, for which children themselves cannot be held responsible. Yet, republicanism does not require the promotion of individuals’ well-being interests; it merely bans arbitrary harming or interference. Therefore, republicanism alone cannot generate a full theory of just child-rearing. Like republicans such as Lovett (2018), I assume that non-domination is only one good among others.

Children’s respect interests protect their status as individuals who are capable of becoming (fully) autonomous. Their first respect interest is to be provided with the means for acquiring (full) autonomy. Their second respect interest is in the freedom to decide for themselves in accordance to their level of autonomy. Because autonomy develops gradually, children’s choices become increasingly worthy of protection; hence, republicans ought to also be concerned about threats to those choices over which children at different stages of development have already acquired moral authority. Here I do not engage at length with the question of how exactly should republicans protect this second respect interest – that is which authoritative choices should children of different ages have. But one pillar of this article’s proposal is that rearers’ power over children must be exercised such as to allow children to gradually expand their freedom of choice.

A third, more contested, type of respect interest, concerns the fact that children who reach full autonomy are capable to retrospectively object to the ways in which their rearers had treated them. This view, developed at length by Matthew Clayton (2006), points to a respect interest in not having been used by one’s rearer as a means to advance the rearer’s interests – for instance, to express their values through their child. On Clayton’s (2006) account, rearers have a duty to teach their children about moral rights and duties – that is, to instil in them a sense of justice – but lack a permission to intentionally enrol children in controversial conceptions of the good, such as a particular set of religious beliefs. As I do not aim to provide a full account of justice in child-rearing here, I remain agnostic with respect to the existence of this type of respect interest. If it exists, then children are dominated by rearers who coerce, or manipulate them to conform to particular ideas of a good life – at least if doing so is unnecessary to advance children’s well-being interests to the level required by justice. It is an interesting question, which a full republican theory of justice in child-rearing should answer, whether it would be dominating to require a child to conform to a particular idea of a good life if this was necessary in order to respect her moral right in adequate well-being.

Authoritative hierarchical relations involving children do not raise the exact same problems as in the case of adults: More than adults, children are owed positive duties to fulfil their interests – duties referred to, here, as ‘adequate care’. One of the children’s interests is to be subject to paternalistic power exercised by adults who take continuous responsibility for their well-being and development. Yet, the delivery of adequate care involves, itself, the risk of domination. While typical adults have the authority to accept or decline whatever resources they are owed, children lack this normative power. Therefore, rearers always exercise power over children even when they abstain from coercing children and when they provide children with the resources owed to them. The threat of domination is more pervasive in children’s than in adults’ lives because the delivery of non-republican duties towards children is possible only within a coercive, non-consensual relationship.

It is not enough that the exercise of power over children advances their interests relative to an alternative in which they have no rearers. Even very bad parenting serves children’s interests better than growing up left to their own devices; yet, some republicans correctly criticised as illegitimate, for instance, Roman fathers’ power of life and death over their children. According to the substantial view on domination, adults’ power over children is dominating if it fails to be organised in ways that are fully justifiable by appeal to children’s own well-being interests and respect interests.

Therefore, republican child-rearing would, ideally (1) assign rearing status only to rearers who are most likely to advance children’s interests ⁶ – which, by stipulation, I call ‘optimal rearers’ and (2) otherwise minimise the arbitrary power of rearers, by denying them a right to setback children’s interests for the sake of their own interests. Rearers do not have a duty to optimise the satisfaction of children’s interests, but on republican grounds, they lack a right to prevent others, or the child herself, from benefitting the child. Current child-rearing fails on both counts.

Not Only Optimal Rearers

Generally speaking, parental status is in first instance allocated automatically to procreators; the mere fact that another individual, who is better able to advance a child’s interests, is willing to take over the role of parenting that child never constitutes a ground for denying parental status to procreators. Prospective procreators, unlike prospective adopting parents, are not subjected to any tests meant to check their rearing competence. Therefore, parental status is usually not allocated strictly according to the interest of the child and this allows sub-optimal rearers to exercise power over children.

Many contemporary societies have child-protection policies. Parents who seriously risk the lives or health of their children can, and sometimes do, lose their parental status. But these are safeguards only against the very worst forms of bad care, and they are highly fallible safeguards, with a significant amount of bad parenting going unsanctioned (LaFollette, 2010). Parental rights can only be involuntarily lost on grounds of detected abuse or neglect.

This status quo is criticisable by liberal standards too. Most contemporary liberal theorists of justice in child-rearing believe that only individuals who would make an adequate rearer have a moral right to parent (Brighouse and Swift, 2014; Clayton, 2006; Macleod, 2015b) and assume that adequacy entails (significantly) more than avoiding abuse and neglect. The same liberals, however, also claim that the allocation of parental status should track both children’s and adults’ interests, thus allowing for trade-offs between the two sets of interests. Republicanism is distinct in that republicans ought to oppose such trade-offs. Any asymmetric power over particularly vulnerable individuals, ⁷ the holding of which cannot be justified by appeal to that individuals’ own interests, is dominating.

The one exception to the dominant liberal view is Peter Vallentyne’s (2003) account according to which only the ‘best available parents’ have the moral right to rear; he defines the best available parent as the person for whom possession of the right is in the child’s best interest, and who is willing to parent the child. But even Vallentyne’s account is (subtly) different from what I deem to be the republican position on this matter. In cases when more than one individual qualifies as ‘the best available parent’, Vallentyne allows the possibility that the right be allocated to the individual who would get the greatest benefit from parenting (Vallentyne, 2003: 997). This is consistent with the general liberal view: so far, liberals have not argued that exercising power over children is in any way objectionable as long as children’s rights are being met. By contrast, republicans must acknowledge that, in practice, rearers’ power over children is always dominating to some degree, as I argue below. This makes it objectionable to give any weight to the prospective parent’s interest in parental power in the allocation of the right to rear.

To sum up, the discussion of principled republican grounds for allocating the right to parent, those who advance the children’s interests the most exercise power over the child in the least arbitrary way. If a rearer actually maximised the interests of the child, that would be ideal on republican grounds, not because the child has an independent entitlement to maximisation but because only in that case would the exercise of power over her be fully justified. It is logically possible that a single rearer maximises the child’s interests. But, in practice, the satisfaction – let alone the maximisation – of some of children’s interests require several rearers. Or so I argue in the ‘Legitimate republican child-rearing’ section.

In practice, republicanism offers robust principled support for parental licencing schemes meant to identify optimal rearers. Yet, such schemes face serious epistemic and implementational difficulties. Standards of good parenting are highly disputed, as the instability of experts’ opinions about parenting styles illustrate. If we are unlikely to reach stable agreement on optimal child-rearing, we are even less likely to design tests that reliably identify the adults who meet such standards. Furthermore, restricting parental status to optimal rearers could entail forced separation of children from their current parents. Such separation, I argued in previous work (Gheaus, 2018a), could destroy valuable relationships between children and birth parents, thus also setting back one of the children’s central well-being interests. Even if the child-rearing abilities of the parents in question would not qualify them as optimal rearers in the absence of an already existing relationship, it is far from clear what is best for the child in such cases. Finally, much of parenting happens in people’s private homes, making close monitoring impossible without disrupting the intimacy between parents and children – intimacy in which children have a morally weighty interest. Assuming it was possible to legitimately overcome all these barriers, it is difficult to see how we could enforce widespread compliance with parental licences.

Together, these considerations suggest serious limitations on realising the first desiderata of republican child-rearing. I do not know whether they are conclusive; perhaps sound parental licencing schemes can be found and implemented. In any case, in section ‘Domination is not fully eliminable from child-rearing’ I argue that, if parents lacked the right to exclude others from their children, the effects of low thresholds of quality for acquiring parental status could be largely mitigated.

Parents Can Setback Their Children’s Interests With Impunity

Here is a rough characterisation of the juridical status quo: Parents are entitled to exercise, to the exclusion of all others, a number of powers which entail significant control over almost every aspect of children’s lives and development: Where children reside, what and when they eat, how they are clothed, when and where they rest, and how, where and with whom they spend their time. Parents also have the right to set demands on how children behave at home and in society (as long as the child does not break the law), decide how they are to be disciplined and make medical and health-related decisions on their behalf. Some laws give parents final say on the latter matters. Parents also have the right to decide with which institutions (such as day care centres, kindergartens, schools, churches and clubs), if any, their children become affiliated. The limits of these rights vary across legislations, and many states require some promotion of children’s well-being interests: compulsory school attendance, curriculum requirements, and bans on corporal punishment.

Yet, certain existing parental powers are hard to defend entirely without appeal to parents’ interests. For instance, many legislatures allow parents to refuse their children’s vaccination even when the children are not at high risk of negative side-effects, or to circumcise their children as a means to preserve their culture; and, to the best of my knowledge, there are no legal prohibitions against homophobic parents inflicting harmful judgemental attitudes on their gay children. To the extent to which existing parental rights setback children’s well-being interests in order to protect parental liberties, many liberals can and do criticise them, ⁸ arguing that vaccination ought to be mandatory (Pierik, 2018), that children’s circumcision is impermissible (Sarajlic, 2014) and that homophobic parents lack a moral right to rear (Brennan and Macleod, 2017). Many liberals believe that children ought to be raised in ways that promote their respect interest in future autonomy. As already mentioned, some liberals (Clayton, 2006) go further, denying that parents have the moral right to intentionally shape children’s values beyond cultivating in them a sense of justice, even if they do not thereby violate their children’s future autonomy: Qua future adult, the child may have a complaint that she had to live according to a conception of the good life that she came to reject once she acquired full autonomy.

Yet, republicanism generates a deeper criticism of the parents’ power over their children. Liberals object to actual uses of parental power. By contrast, republicans should worry about parental power even when it merely could be used against the children’s interests. For instance, republicans must object to parental power (whether or not exercised) to shape children’s values to the extent to which such power can also be used to setback children’s well-being interests, and to parental powers (whether or not exercised) to setback children’s respect interests in exercising choices for which they are sufficiently mature.

One feature of parenting that is especially objectionable on republican grounds is the exclusive nature of parental rights. Parents can ensure that particular other individuals are unable to influence, for instance, their child’s whereabouts, dress, company, activities, interests, values, or character. And, they have the power to exclude any particular individuals from developing robust, long-term relationships with their child. Of course, not many parents can, in practice, exclude all others from influencing their children. Regular social interaction relaxes parents’ control over how other people influence their children and in many places there are laws that prevent children’s complete social isolation. Nonetheless, parents have the power to at least try to exclude any ⁹ other particular individual from playing an important role in their child’s life, and definitely from becoming a person on whom the child can rely for the provision of physical or emotional care. Parents can require their children not to form friendships with particular others. They can uproot their children from their communities. They can interrupt their children’s beneficial relationships: for example, they have the power to fire nannies at will. Wherever home-schooling is allowed, parents can exclude others from their child’s education. And parents can exclude all others from their pre-school children because day care and kindergarten attendance is universally optional. Generally, parents can avoid particular individuals and institutions by moving home.

It is implausible that a general justification of the exclusionary nature of existing parental powers can entirely rely on children’s interests in adequate care. In some instances, it is likely that rearers need such right for the child’s sake. For instance, it would be hard to ensure that a baby is well-fed and settled into a good feeding rhythm if others were free to interfere at will. As Robert Goodin (1985) noted, it is efficient to allocate the duties to care for each particular child to particular individuals; to discharge some of these duties, rearers need the right to prevent others’ interference. But this justification does not generalise; to the contrary, in many cases children benefit from exposure to multiple educational and formative influences, and in particular from multiple caring relationships.

Epistemic difficulties certainly complicate judgement on this issue, too. Sometimes, parents exclude other adults from close relationships with children because they deem the influence of these adults detrimental – and they can be are right. However, most people whose company a parent may judge detrimental to her own children’s well-being have the right to be, and often are, other children’s parents. The fact that we as a society do not prevent those individuals from raising children of their own is – or should be – an indication that we as a society do not publicly deem their influence detrimental to children’s well-being. This is consistent with recognising that some individuals’ influence on children is nefarious and that they should be excluded from close relationships with all children.

The exclusionary nature of parental rights is therefore dominating also because it confers to parents a (conditional) monopoly of care in relationship with their children: Either parents are the only providers of reliable long-term care, or they can control the conditions under which children receive care from other parties. A general parental right to exclude others is part and parcel of this monopoly. This is particularly problematic with respect to parents’ power to control their children’s relationships with others, especially caring relationships.

Parents’ right to prevent or sever children’s relationships with other potential care-givers – a former nanny, educator, teacher, neighbour or friend – is, as I explain in the ‘Republicans on child-rearing’ section, the parental right most inimical to minimising children’s domination. Above and beyond the way in which monopolies of care are actually detrimental to children’s interests, republicans (unlike liberals) must deem them illegitimate as forms of arbitrary power over children.

Domination Is Not Fully Eliminable From Child-Rearing

Could, and should, rearers be divested of all their powers that fail to track children’s interests? The case of parents’ rights to shape their children’s values illustrates a more general claim that domination is not fully eliminable from child-rearing. Child-rearing without domination would require the elimination of the possibility to use, with impunity, power over children in ways that do not track their interests – a goal that is unattainable without the sacrifice of other, more important (non-republican) goals: children’s general interest in adequate care, including their shared interest in intimacy.

Rearers require some discretionary power in order to meet children’s interests. To flourish, and, presumably, also to develop their autonomy, children need attentive and affectionate human beings, rather than robots, to provide them with resources and to make decisions on their behalf. Even if an impersonal agent, such as the state, could decide what it is best for them to eat, when they should go to bed and what games they should play, it is good for children that those in charge with hands-on child-rearing make judgements about, and therefore retain some degree of discretion over, the details of care provision. This fact raises republican worries.

First, continuously sharing one’s life with a child involves myriads of decisions, and in this context, it seems impossible for rearers to take all children’s interests into account, let alone to give them priority, when reasoning about every decision. Nor, indeed, is it possible for rearers to worry about reasons in every situation: some of the decisions they must make do not allow time for deliberation and may unavoidably end up advancing trivial interests of the rearer at the expense of trivial interests of the child. This is because rearing is an unusually intensive and demanding job: rearers cannot take long daily breaks from it, in order to pursue their own legitimate interests in ways that cannot conflict with those of their children’s. Moreover, rearers often control their children’s lives through the same decisions through which they (must) control their own lives.

Even the best rearers will at times inadvertently decide, on the spur of the moment, to engage in an activity for their own sake in spite of their child having required another activity that would have been better for her. In the case of adults, such failures of consideration need not entail domination because relationships between adults usually do not involve asymmetric authority, and because adults are free to exit most relationships. Yet, children lack a normative power to exit all relationships with caring adults – although, as I argue below, they should have more exit freedom than they currently have, for the sake of minimising domination. Even if desirable, it would be unfeasible to require that every rearer’s decision which impacts the child tracks exclusively the child’s interest, and to hold rearers accountable for failing to live up to this ideal.

However, it would not be desirable for every rearer’s decision which impacts the child to track exclusively the child’s interest. Some exercise of parental power that tracks parental interests is instrumental to creating the long-term intimacy in which children (and adults) have a powerful interest. A degree of life-sharing between children and their rearers is necessary for long-term intimacy – sharing, for example, a language, meals or recreational activities. Rearers must have some discretion to decide for their children on common lifestyle issues. Yet, as separate individuals, children’s interests will inevitably be in occasional conflict with the legitimate interests of their rearers. To maintain intimacy, decisions on common activities must track both parties’ interests and, in cases of conflict, individuals take turns in letting their own interest take a back seat. Adults can make compromises in order to maintain intimacy; but children, at least the very young, lack the moral power to compromise. Therefore, to make intimacy possible, rearers must sometimes exercise power over children that do not fully track children’s interests (other than in intimacy). For instance, they must be free to chose family pastimes according to their own tastes, even when this means overriding their child’s preference for an activity that would benefit the child more – and hence, exercising coercive power that does not track the child’s interest. This, very limited, degree of parental discretion, might be non-dominating in case it does not represent an all-things-considered setback of the child’s interest. (This will depend on the extent to which the child’s interests must sometimes come second in order to enjoy intimacy, and in exactly how weighty children’s interest in intimacy is relative to other interests they have.)

However, to allow children to have many of their interests met within personal relationships, parents must enjoy a much higher degree of discretion that what is strictly needed to create intimacy; they must enjoy the power to make decisions that setback the all-things-considered interests of their children, and do so with impunity. Intimacy itself could not be preserved if uses of parental power were constantly monitored, or if parents themselves were to engage in constant self-censorship (Brighouse and Swift, 2014). Non-domination excludes any power that rearers could exercise, with impunity, in ways that fail to track children’s interests. Because it involves such easy-to-abuse powers, intimacy in relationships with children is dominating; and yet, eliminating it from child-rearing would require too large a sacrifice of children’s interests. In other words, children have an interest in being raised by people who have some dominating power over them – although, of course, they do not have an interest in the actual use of this power. If so, than some of the justified parental power is dominating. At the same time, it is regrettable: if the interest of children and their rearers could always be simultaneously satisfied, then domination would be avoidable with no loss of value. I assume this would be a serendipitously happy case.

Therefore, the republican ideal of non-domination in child-rearing cannot be fully realised without setting back important interests of children: Rearers could not be functional and continuous decision-makers for themselves and their children, or maintain intimacy with their children, if they were entirely stripped of the power to setback their children’s interests with impunity. Just like the problem of non-optimal parents, this problem, too, I will argue, can be partly mitigated by restructuring power over children.

Republicans must find such domination prima facie objectionable, although ultimately outweighed by other considerations. This conclusion may extend beyond child-rearing contexts and indicate, more generally, the limits of the republican criticism of dependency. The nature of children’s dependency on their care-givers is to some extent similar to most adults’ all-things considered desirable dependencies. We are all dependent on one person or another for love and intimacy, which we can only experience if we make ourselves vulnerable to our beloved – that is, when they have some power to setback our interests and even interfere with our choices, for instance, concerning our emotional life, with impunity. Unlike children, adults can decide to leave particular, or all, such relationships – yet, at a cost that makes it look permissible, although regrettable, to accept some degree of domination. The ideal of non-domination is aspirational; in practice, we should seek ways of minimising, rather than eradicating, domination.

Perhaps dependency as exposure to some arbitrary power cannot be eliminated, without great loss, from human life in general. But in child-rearing, this fact is especially problematic. Adults’ choice of whether or not to put up with some domination for the sake of intimacy makes domination less objectionable. But children should lack the choice of entirely withdrawing from any and all rearing relationships. However, children could be given (temporary) exit options from particular relationships with rearers, without the same sacrifice of their interests. Ensuring that all children have caring relationships with a multitude of adults who cannot veto each other would lower each rearer’s arbitrary power and foster children’s exercise of their developing autonomy – a point which I explore below.

Republicans on Child-Rearing

Domination comes in degrees: it can be more or less intense and more or less extended (Pettit, 1996: 581). Republican child-rearing would minimise domination without sacrificing children’s interests in adequate care. Existing parenting practices are not such an arrangement, due to the low threshold for holding parental rights and the extent of parental rights – in particular, their right to exclude others from the child. But republicans have not, so far, taken issue with parental monopolies of power over children.

Instead, some republicans, like Pettit, seem to assume that the current recognition of children’s rights and policies, protecting against parental abuse and neglect, may be all that republican child-rearing requires, if properly implemented. I argue that republican child-rearing is more demanding: it requires a child-centred regulation of rearers’ rights, limiting the degree to which they can exercise, with impunity, power that sets back children’s interests. For instance, parents could be denied rights to impose educational and medical decisions on their children that fail to serve the children’s interests.

Nevertheless, since much parenting happens in people’s private homes, away from the public eye, there is only so much that legal and institutional measures can accomplish without undermining intimacy in child-rearing. Other scholars of republican child-rearing, such as John Ferejohn (2001: 82), acknowledge this fact and suggest we need to rely, instead, on parental virtue. His solution depends on highly idealised assumptions about parents being responsive to moral demands and motivated by benevolence. Yet, this is at odds with the typical republican insistence that it is not enough for people to voluntarily refrain from using their power arbitrarily. Ferejohn’s position deserves consideration because it may capture a widespread view on the legitimacy of the extent of parental power. However, his reliance on parental virtue seems overly optimistic even under the assumption that current child-rearing practices are permissible. Data on child abuse, child neglect and runaway children suggest that parental virtue is a very fallible safeguard against mistreating children (LaFollette, 2010). On the more demanding view of the content of parental rights presented here, it is even more unlikely that the morality strategy is adequate.

This conclusion is not surprising. There is a long tradition of not relying on individual morality in order to prevent the evils of concentrated political power and of devising mechanisms to limit power. Echoing Montesquieu’s idea that separating powers can buttress citizens’ freedom, Pettit (1996) argues that freedom is promoted by opposing anti-power to power. A standard republican way of making the exercise of state power legitimate is to divide it between several agents who have structural incentives to remain independent from each other. Similarly, republicans should endorse a change in the structure of power over children, by denying rearers a right to exclude, at will, others from access to their children.

There are multiple reasons for doing this, some of which are instrumental. Multiple legally protected caring relationships would better advance children’s well-being interests, including in avoiding abuse and neglect, and their respect-interests, by letting them exercise, in age-appropriate ways, their budding autonomy. This would make power over children better track their interests.

But republicans should also deem parental monopolies of care and power over children intrinsically objectionable. This is in line with Pettit’s (1996: 583) own, and plausible, identification of one feature that generates domination, that is ‘the resource of being . . . say, the only doctor or police officer around whose help and goodwill the other may need in various possible emergencies’. Parents are, in fact, in such a position, by virtue of having a right to exclude others from having caring relationships with the child.

We are not used to think that the power relationship between children and people who rear them is analogous to the power relationship between citizens and the state. Yet, the analogy dominated the imagination of ancient philosophers like Aristotle (1999) as well as early modern philosophers like Robert Filmer (1991) and, presumably, their contemporaries. More recently, Clayton (2006) argued that relationships between parents and children are like relationships between state institutions and citizens in three important respects: they are not voluntary at entry, involve coercion, and deeply influence the lives of the less powerful party. I pointed out that the parent-child relationship displays an additional feature that we consider illegitimate in other contexts such as political or economic life: monopoly of power, ensured by the fact that children have no moral exit right from the relationship with their parents, unlike citizens who have a moral right to leave their country.

Neither legal arrangements restricted to detecting child abuse and neglect, nor reliance on individual morality is sufficient. In addition, we should – as a strategy to minimise domination – to eliminate monopoly of essential care for children. Once children have access to several independent sources of essential care, they also acquire the means to effectively contest their parents’ decisions – when care at home fails, they can turn to another adult whom they know well and who cares for them. This anti-power mechanism would render children’s general dependency on their care-givers acceptable to republicans.

Legitimate Republican Child-Rearing

At the very least, republicans must insist that rearers are not permitted to exclude others from access to the children they raise, unless such exclusion serves the children’s interest. They should insist that we find ways of establishing who the adults are, with whom children may form additional caring relationships, and in what circumstances they may turn to these individuals for care. For instance, we may licence the adults who are willing to serve this function, without encountering the same difficulties raised by procreative parental licencing for optimal parents; all we need to establish is who are the adults unlikely to harm children. The implementation of such a scheme does not threaten to destroy pre-existing relationships between children and rearers. Furthermore, we should determine what are age-appropriate choices that children can make with respect to the sources from which they receive care.

More likely, republicans will support a stronger claim: That societies ought to be organised such that all children have robust opportunities to develop intimate, long-lasting, caring relationships with several individuals, who should be unable to exclude each other from the children’s life other than for the children’s own sake.

Before seeing how these solutions – the less and the more demanding – could look in practice, it is worth addressing the most obvious reasons to oppose them.

Objections to Abolishing Parents’ Right to Exclude Others

The envisaged reform represents a serious challenge to parenting as we know it. A fundamental question is why parenting – that is, the rearing of children by one or two individuals who associate freely and hold exclusionary rights in relation to their children – is legitimate in the first place (Brighouse and Swift, 2014). It is plausible that the duties to care for particular children should be allocated to particular individuals if they are to be properly discharged. But why should these duties be assigned to an individual or couple who has the right to choose with whom to associate for the purpose of raising children, and why should this individual or couple thereby, earn a right to exclude others from access to the children they parent? Why not assign child-rearing duties towards each child, together with authoritative power over that child, to a larger number of individuals, and why not give prima facie protection to all caring relationships that children forge with individuals other than their rearers?

One principled case for the institution of parenting, that is for allocating the hands-on care of children in this way, stems from the belief that procreative parents bear the duty to rear the children of whose existence they are responsible. Children typically have two procreative parents; hence, the institution of child-rearing recognises two parents as the default number of rearers for any particular child. Yet, even if procreators bear special duties in virtue of being responsible for the existence of their offspring, it is not necessary that they thereby acquire a right to rear the children they procreated. As David Archard (2010) argues, procreative parents’ causal role in bringing children into the world may be the reason to require them to support some or all of the costs of rearing children. Yet, the fact that they are responsible for children’s existence does not entail that procreative parents ought to have rights to control their offspring’s lives.

Several widespread empirical assumptions might be involved in legitimising parenthood in a pragmatic, rather than principled, way. ¹⁰ Of these, the most salient is probably that natural parents are significantly more likely to love their offspring and attend to their well-being than biologically unrelated individuals (Macleod, 2007). A second empirical assumption may be that the continuous attachment necessary for children can only happen with one person or, at most, with two. ¹¹ A third likely assumption is that having more than two parents invites too much conflict between parents and hence fragments their authority to the detriment of the child.

I cannot discuss these empirical claims in detail. Even if correct, it is unclear that they provide unambiguous support for existing child-rearing arrangements. For instance, even if procreators made the best main rearers to their offspring, it is possible that children’s well-being would be further advanced if others had legally protected access to children, and opportunities to forge caring relationships with them.

Moreover, these assumptions are not beyond doubt. If parents – procreative or not – are a lot more likely than strangers to be motivated to benefit their offspring, it is surprising that children are safer, on average, in formal child care settings than they are at home (Waldfogel, 2006). The experience of those growing up in extended families with more than two parental figures show that children can easily form intimate, and beneficial, attachments with several people. Perhaps some rearers could be disinclined to become emotionally attached to a child, should they know that other adults compete with them for the child’s time and attention, and that there are other adults that could step in should they fail to exercise their child-rearing responsibilities. But I assume this to be the exception rather than the rule, since family love and intimacy can usually flourish when there is no assumption of exclusivity – as we know from cases of friendship, or love between numerous siblings. ¹² In addition, the current status quo is discouraging adults other than parents to become emotionally attached to children, since the relationship with them depends on their parents’ will. Finally, we as a society allow – and sometimes encourage – divorced couples to continue to co-parent. Assuming that divorced parents are, other things equal, more prone to disagreements and conflict, this fact may indicate a belied that conflicts between rearers are less detrimental to children than having one parent less. Even if having more than two parents introduced additional conflict in child-rearing, this would not be a conclusive argument against giving children more than two rearers.

Multiple Care-Givers and Children’s Interests

The abolition of rearers’ right to exclude others from forming and maintaining caring relationships with the children they rear must be shown not to be generally detrimental to their well-being interests and respect interests, and it is an important virtue of such an arrangement if it is more likely to result in serving children’s interests better than existing forms of child-rearing. Indeed, research on institutional non-parental care shows it is beneficial to children when a certain quality threshold is met (Waldfogel, 2006).

First, children’s continuous exposure to several caring adults is likely to deter all of the adults in a child’s life from engaging in child neglect and abuse that is quickly detectable by the others. Even when undeterred, abuses of power over children are much easier to detect when children have regular contact with adults other than their parents and immediate family. This is a better mechanism to contain child abuse and neglect than intrusive universal regular home visits by social workers, which require sweeping information-gathering about all families and much intrusion in the intimacy between children and their rearers.

Furthermore, when children suffer from failures of care – including but not restricted to neglect or abuse – a diversification of rearers who lack a right to exclude each other from children’s lives can mitigate the results. It would give more children exposure to at least one good rearer, which is likely to be highly beneficial. For example, some research (Lamb, 1996; Rishel et al., 2005) indicates that, in order to develop resilience and learn how to relate to others in healthy ways, children need one good adult who is caring and dependable.

Third, a plurality of caring relationships advance children’s respect interest in future autonomy, by exposing them to several individuals from whom they can learn firsthand about various conceptions of the good and who can therefore help them make informed, and emotionally less costly, choices about particular values and lifestyles. ¹³

Fourth, such an arrangement would obviously promote children’s respect interest in exercising their developing autonomy. In general, the older children are, the more dominating it is for them to be under the authority of a single care-giver. While not all children’s choices carry the same moral weight as choices made by adults, their choices do carry some moral weight. Children who start forming their own ideas about a good life are owed opportunities to experiment with various possibilities. Older children still need some guidance, and some harmful options should be ruled out for them; yet, it is increasingly dominating to impose on them the conception of the good of any particular individual or couple who happens to be their sole rearer. One way of minimising child domination is to ensure that, before they acquire the moral authority to fully follow their own conceptions of the good, children have choices on a gradually expanding range of matters.

Finally, following Clayton (2006), consider the possibility that children have a further respect interest that rearers refrain from intentionally shaping their values, as explained earlier. Assuming that most rearers are unlikely to spontaneously refrain from intentional shaping of their children, and that it is either impermissible, or hopeless, to enforce this requirement, then ensuring multiple normative influences in the life of each child looks like a second best arrangement.

Note that a republican requirement that we minimise child domination is distinct from – although compatible with – the aim of maximising children’s well-being interests. But this is not problematic: children are only entitled to adequate, rather than optimal care. Eliminating monopolies of care minimises each rearer’s dominating power, and hence, results in a minimally dominating child-rearing; for this reason, republicans must deem such arrangements morally required. Therefore, it is enough to show that the elimination of a right to exclude others from the child is compatible with adequate, rather than optimal, care for all children. But the conflict should not be overstated: Perhaps in a world of ideally virtuous procreators a right to exclude others from the child would do little to maximise children’s well-being. But, given empirical realities about the levels of child abuse and neglect, and the benefits of non-parental care from vetted care-givers, it is indeed plausible that children’s well-being would on average be increased if they had access to multiple caring relationships. Moreover, in the absence of monopolies of care, children’s respect interests can obviously be better protected.

Practical Arrangements

Minimally, republicanism ought to defend children’s freedom to cultivate relationships with various benign adults – such as neighbours, teachers, or relatives – and seek support from these individuals, independently from the arbitrary will of their primary rearers. That is, children’s relationships with adults deemed fit to care for them ought to be protected. This could mean that a school-age child should, given the choice, have the freedom to spend her Sunday morning visiting a museum with her teacher, or going to church with her neighbour rather than playing sports with her parents. Or that a teenager whose relationship with her parents is in dire straits may seek emotional support, lodging or help with mediation from others with whom she has a caring relationship.

More likely, republican child-rearing should include robust provisions for caring relationships between children and adults other than their primary rearers. Examples include mandatory, universal, highly regulated and very substantial institutional child care to supplement parental care – as I have argued in previous work (Gheaus, 2011, 2018b). Another possible reform is a formalised institution of god-parenting – stripped of religious significance – whereby, each child is assigned non-residential rearers whose duty is to serve as an alternative source of care to the child.

In addition, or as an alternative, to god-parenting, republican child-rearing could be realised through daily attendance at well-staffed day care centres, kindergartens and schools with non-academic functions. Such measures would enable children to depend on more people than their own parents for the provision of the material, emotional and intellectual resources they need. For this solution to work, the non-parental care would have to be very robust: continuous, reliable and meeting a wide range of children’s needs. Together, these conditions could ensure that no child’s care is overly controlled by one individual or a group of closely related individuals; they would also ensure that, should parental care fail, the child could safely turn, or be turned, to the non-parental carer for help. In practice, this arrangement requires that child care institutions be ready to give children a default alternative to family care. This would lower, as much as possible, the practical and psychological costs that children have to pay in order to (temporarily) exit relationships that involve inadequate care, and to seek remedies for less significant abuses of power.

Such reforms do not require a huge leap of imagination: In many countries, children already receive significant non-parental care, and many children have substantial relationships with their ‘godparents’. The difference, however, is that existing day care, kindergarten and god-parenting are optional and there is no guarantee that they are continuous sources of robust care, since parents have the right to sever these relationships at will. To dismantle the monopoly of care, it is necessary to ensure that all children can receive regular and substantial hands-on care from individuals who have a right – and possibly a moral duty – to remain in the lives of the children in question.

Children have an equal right to minimally dominating rearing; the reforms sketched above must be universal. This requires sufficient people willing to co-raise children, as god-parents, or workers in child-rearing institutions. Republican duties of child-rearing are unusually demanding also with respect to finding enough individuals willing to ensure minimally dominating child-rearing arrangements. This, however, is only a (perhaps extreme) instantiation of the general problem that duties towards children are very demanding. Even on the view that procreators are primary duty-bearers towards the children they procreate, non-procreators have weighty duties to provide for children in cases in which procreators fail to do so – because they die, abandon, neglect or abuse their children – or when the content of the duty requires institutional, rather than individual, action – for instance, providing children with schooling. I cannot properly address here the interesting issue of finding sufficient people willing and able to rear children in arrangements that preclude monopolies of care. I merely note that it is not unthinkable that enough people would be, or could be nudged into being, happy to commit to long-lasting caring relationships with children, less resource-intensive than current parenting but, nevertheless, legally protected. An adequate discussion of how to ensure minimally dominating child-rearing to all children – a project for another paper – would have to examine the question of whether, and how, it is possible to find legitimate means of implementing it.

Embedding children in networks of caring relationships with people from outside their residential families mean that relocation to a different city or country is likely to negatively affect children’s well-being interests and their respect interests, to an even larger extent than it does in present circumstances. If so, then more of the current parental rights may have to go – such as the right to arbitrarily decide to disembed children from their social environment. Republican parenting comes at a high prudential price for rearers in terms of control over their children’s lives (at the same time as bringing significant gains for them in terms of security for their children and, likely, time resources from themselves). It is not surprising that voluntarily taking control over the well-being and, partly, the future, of another human being requires significant sacrifices. This conclusion seems to me a desirable feature, rather than a shortcoming, of republican child-rearing.