Biological Parenthood: Gestational, Not Genetic

ABSTRACT

Common-sense morality and legislations around the world ascribe normative relevance to biological connections between parents and children. Procreators who meet a modest standard of parental competence are believed to have a right to rear the children whom they brought into the world. I explore various attempts to justify this belief, and find most of these attempts lacking. I distinguish between two kinds of biological connection between parents and children: the genetic link and the gestational link. I argue that the second can better justify a right to rear.

KEYWORDS:

• procreation
• gestation genetics
• right to parent
• children's rights

1. Introduction

Legally, biological parents have a presumptive right to the custody of their children. This is universal practice, and it determines an important feature of the institution of parenting—that, by default, children have two custodians. Yet the moral relevance of the biological connection between procreators and children is unclear, and so is the justification of this practice. A better understanding of this connection would help with assessing the institution of parenting and the status quo in the acquisition of parental rights. Moreover, it will help to settle several types of custodial disputes. My paper is a contribution to these endeavours.

Usually, we use the term ‘biological parent’ to mean ‘genetic parent’. Here I distinguish between two ways in which one can be a biological parent, or procreator: by being the genetic procreator, or by being the gestational procreator of the child. I use the gender-neutral term, because transgender men can and do become gestational procreators. Genetic procreators are obviously biological procreators—they contribute biological material that is essential for the future child's existence and identity. I assume that it is unproblematic to see gestational procreators as another type of biological procreator: genes can only express themselves in an environment, and therefore genetic material and environmental conditions together determine phenotypes. Gestational procreators contribute the biological environment that co-determines all of the future child's features [Feldman 1992]. Until recently, it was easy to ignore this distinction between two ways of being a biological procreator: gestational procreators were always also one of the two genetic procreators of the child, so the distinction did not have much practical relevance. Today, a child's gestational procreator need not be her genetic procreator. It used also to be the case that all children had two genetic procreators; today, thanks to the possibility of mitochondrial replacement therapy,¹ there are cases in which three different individuals contribute genetic material to the bringing into existence of a child (providing ova, mitochondrial DNA, and sperm). Therefore, a child can have three genetic procreators and a gestational procreator—altogether, up to four different biological procreators. Unsurprisingly, these technological developments generate challenges to the status quo in acquiring the right to custody.

I argue as follows:

P1. Any account of how we acquire the moral right to parent must be centred, partially or exclusively, in the interest of the child. (Defended in section 3.)

P2. A gestational relationship with a child is a better indicator that the interest of the child will be served than is a genetic relationship with the child. (Defended in sections 4 and 5.)

C. Therefore, more moral relevance should be attached to the gestational, than to the genetic, connection when deciding who has the right to parent a biologically related newborn.

I use ‘right’, unqualified, to refer to moral rights, as most of the present discussion concerns the justification for acquiring the moral right to be a newborn's custodian, which I call ‘a right to parent’. I assume that the existence of a moral right to x ought to be an important basis for a legal right to x, but that the translation may not be always straightforward.

2. Background Clarifications

Consider three types of situation leading to conflicting claims over the custody of a child. They illustrate, schematically, real-life cases. Each raises the question of who has a stronger right to parent:

Case one: Embryo Swap

Two women seek IVF treatment at the same hospital. Only one of them becomes pregnant as a result of the IVF, and at some stage during the pregnancy it turns out that, due to a mistake, the foetus is the other woman's genetic offspring. The foetus is carried to term. Both the gestational and the genetic mother want to raise the baby.

Here the claims of a genetic procreator are pressed against the claims of a gestational procreator. Conflicts in such cases have been settled differently in different jurisdictions around the world, which is indicative of a lack of general agreement concerning the weight of the genetic procreator's claim relative to the weight of the gestational procreator's claim [Bayne and Kolers 2003: 224–5].

Case two: Adoption without the consent of a genetic procreator

A baby is put up for adoption by a woman who is both the baby's genetic and gestational mother without consent from the (estranged or misled) genetic father. The baby settles in well with her adoptive parents. A few months later, the genetic father sues for custody, on the ground that he has never waived his right to parent the child. The adoptive couple wishes to keep custody.

Here, one question is that of whether, in virtue of his genetic connection to the child, the father had a presumptive right to custody: if he did, then putting the baby up for adoption without the consent of the genetic father wronged the latter. Another question concerns the reason in virtue of which the genetic connection with the child gave the father a presumptive right to custody. Even if the genetic father was wronged, it is unclear whether his right to custody endures after adoption. Answering the second question requires understanding what aspect of the genetic connection grounded his presumptive right to custody. In several situations of this kind, USA courts found in favour of genetic fathers, and as a result the children—by that time a few years old—were forcibly removed from the adoptive parents. Judges defended their decision by indicating that a right to custody accrues to genetic procreators, and can only be lost by voluntary alienation or through child abuse or neglect [Richards 2010: 8–10].

Case three: Gestational surrogacy

A woman agrees to serve as a gestational surrogate mother for a couple who provide the zygote. The pregnancy is successful, and at birth the surrogate wants to parent the newborn herself. So does the couple.

This is arguably the hardest of the three cases. The underlying question is, again, that of how to weight the genetic procreators’ and the gestational procreator's respective claims. In some European countries, legislation recognizes the gestational mother as the sole bearer of the right to be the legal mother of the baby and her husband as the legal father [Warnock 1985]. This may also mean that there is no legal venue for the genetic procreators to become the baby's custodians merely in virtue of the gestational mother's wish. In some legislations, once legal parents have voluntarily alienated their right to custody, it is the state's right and duty to identify the child's next custodian. In such places surrogacy contracts are void, because the gestational mother does not have a legal power to transfer her custodial rights to another person. In other jurisdictions surrogacy contracts are valid, but non-enforceable or both valid and enforceable.

For simplicity, assume that the individuals involved in the above cases are similar in every respect that matters to the quality of their parenting, except for the fact that some have a strictly genetic connection to the child while others a (strictly) gestational one. In all of these cases, we need a principled way to decide who has the right to be the child's custodian. This, I will argue, turns on reasons to think that either the genetic or the gestational procreator might be a better parent to their progeny, in virtue of their biological connection. In addition, in the case of surrogacy, the very legitimacy of the practice is in question.

The claim that we ought to attach more moral relevance to the gestational than to the genetic connection in determining the right to parent is limited in two ways. First, neither a genetic nor a gestational connection between parent and child is either necessary or sufficient to justify the acquisition of the right to parent. Neither is necessary, because one can rightfully become the custodian of a child without standing in any biological relationship to the child, as happens in adoption. And each is insufficient, because nobody can acquire or retain the right to parent if they fall below a certain threshold of adequacy. Unfortunately, no kind of biological relationship with the child makes one immune to such failure.

Second, this paper is about the right to parent a child. It does not aim to make a contribution to understanding the rights of parents—that is, the rights that parents may exercise in relation to their children—but some of the analysis will bear on the limits of parents’ right to exclude certain others from forming relationships with the child. Similarly, I do not make any claims concerning the duty to parent—that is, the moral duty to become a particular child's custodian—or the duties that various individuals have in relation to children.

The latter claim is more controversial: a right and a duty to parent are usually acquired together, and some think that this situation is in no need of justification [Bayne and Kolers 2003: 223]. But this is not self-evident. It is possible to have a right to parent a particular child without having a duty to become the parent of that child. According to one group of views, duties owed to children are, fundamentally, collective duties that individual parents then discharge towards individual children [Vallentyne 2002b]: for instance, because children are vulnerable to adults collectively [Goodin 2005]. In this case, if there are enough volunteers for the parenting role it need not be anyone's duty to become a parent. (Or, if there are at least as many children as intending parents, it may be everybody's duty to become a parent in general, but not any particular child's parent.) Some of these individuals may be better qualified to rear a particular child, and, in virtue of this feature, have a right to claim the custody of that child. It is also possible to owe special duties to a child without having the right to be the child's parent: the mainstream view holds that procreators have a duty to parent in virtue of having brought the child into existence, and we collectively acquire a duty to rear children only in cases when their parents fail to discharge their duties. This is compatible with a belief that some procreators, in virtue of having brought the child into existence, acquire extensive duties of provision towards the child but not a right to act as the child's custodian. That is, they would not possess the right to parent the child even though they were duty-bound to provide resources to the child [Austin 2007; Archard 2010]. Therefore, the rights and the duties that adults have in relation to children might have different justifications, and hence it is possible that they are acquired independently of each other.

3. The Right To Parent

There are three types of account of how individuals acquire the right to parent in general and the right to parent a particular child: parent-centred, child-centred, and dual-interest accounts.² Traditional accounts are parent-centred, justifying a right to parent in general entirely by appeal to the interest of would-be parents, and a right to parent a particular child by appeal to the interest of the prospective parent in parenting that particular child. Children have long been seen as appropriate objects of ownership. The most extreme versions of the parent-centred account appeal to the interest of an adult in her or his property; but, as I discuss below, not all of these versions must assume that children can be owned.

The obvious alternative to parent-centred views are child-centred ones, letting children's interests alone play a justificatory role [Archard 1995; Brennan and Noggle 1997; Goodin 2005]. On these views, some people have a right to parent if and only if children in general are better off in the custody of parents than, say, if raised by state-run institutions or left to their own devices. A version of this account says that one has the right to parent a particular child in virtue of making the best parent for that child [Vallentyne 2002a]. Child-centred accounts of the right to parent have been resisted because they deny the existence of a fundamental right to parent,³ and see the right as derivative from the child's interest. Therefore, child-centred accounts may deny a right to parent to individuals who would be very good—but not optimal—parents [Brighouse and Swift 2014: 95].

To avoid this counterintuitive conclusion, a third type of account—the dual-interest account—has been defended, justifying the right by appeal both to the child's interest in being parented and to adults’ interest in playing the role of the parent [MacLeod 1997; Clayton 2006; Brighouse and Swift 2014]. On the Brighouse-Swift view, the right to parent is primarily grounded in the child's interest in receiving adequate care from an individual with whom the child stands in an authoritative, loving, intimate, long-term relationship. Therefore, only individuals who are capable of caring adequately for a child within this kind of relationship have a right to parent in general. Defenders of a dual-interest account believe that the right is held by all individuals who would make adequate parents. According to Brighouse and Swift, the reason is that some adults have a fundamental and very weighty interest in an opportunity to parent—that is, to engage in an authoritative, intimate, and loving relationship with the child. For most individuals, they think, parenting is the source of distinctive and important goods that are essential to their full flourishing. These goods accrue to adequate parents in virtue of the unique kind of relationship between parents and children. To be an adequate parent means to successfully discharge great responsibilities towards an individual who is much less powerful than you, whom you often have to coerce for her own good, whose mind you will inevitably shape as the relationship unfolds, and who loves you in a spontaneous and trustful way unparalleled by other loving relationships. Parenting, so understood, is constitutive of a flourishing life for many of us. On dual-interest theories, the right to parent is not merely derivative from the child's interests; rather, it is sui generis because it is partially grounded in the would-be adequate parents’ fundamental interest in playing a fiduciary role. Brighouse and Swift's account stops short of explaining how one acquires the right to parent a particular child [Gheaus 2012]. But dual-interest views might shed light on this question by indicating how the interests of the would-be parent, as well as the interests of the child, are relevant for establishing the right.

Only child-centred and dual-interest accounts can be successful. Children, I assume, are persons—that is, holders of a right to have their fundamental interests satisfied. They lack full autonomy and hence are owed parents in order for their lives to proceed as well as justice entitles them to do so. Children's general interest in being parented justifies the existence of the family as the default way of rearing children [Vallentyne 2002a; Goodin 2005; Clayton 2006; Brighouse and Swift 2014].

In addition, facts about children's interests are relevant to determining the right type of account of the acquisition of the right to parent a particular child. Parents have a power to exercise, and to exclude others from exercising, a bundle of rights through which they directly and significantly control various aspects of the child's life. Some of these—such as the right to determine the child's daily routine—are uncontroversial, while others—concerning the child's relationships, disciplining, schooling, and values—are more contested. Whatever is the morally justified bundle of parental rights, they give parents the power to manage and direct the child's life to a great extent. Liberals, who believe in the moral equality of persons, usually believe that rights to directly control another person's life must be justified either by appeal to the consent of the person over whom the rights are exercised or, if consent is unavailable, by appeal to her interests. Since children lack full autonomy, they cannot give authoritative consent. Therefore, controlling children's lives—and, more generally, establishing relationships of authority with children—must be justified by appeal to children's interests. This argument applies to the institution of parenting in general, and implies that an individual's claim to parent a particular child cannot be legitimate unless it serves the interests of the child in question.

Because I hold that children are persons who are owed duties of justice, and power over them can be justified only if it serves their interests sufficiently,⁴ I conclude that parent-centred accounts cannot successfully justify the right to parent a particular child. A successful account of how one can acquire the right to parent a particular child must be either child-centred or dual-interest. For the purposes of this paper, I do not commit to either kind of account, nor to any particular view of the fundamental interests of the child that it is the parents’ (rather than other individuals’) duty to fulfil—with one exception. I assume that Brighouse and Swift are right to claim that one of children's most important interests is to establish a loving relationship with their custodian, and, because the object of one's love is not fungible, children have a powerful interest in maintaining such relationships once these are established. This is a weighty reason to protect the child's and adults’ interest in continuity within the parent-child relationship. It is easy to see how agents other than parents could be responsible for fulfilling children's interests in a variety of goods: schools and educators may be best placed to address children's interest in education; health-care providers might bear ultimate responsibility for children's health; and so on. States could, via various distributive mechanisms, take responsibility for ensuring that all children's nutritional needs are met, etc. But it is parents, as children's custodians, who ought to share their lives with children, because they need to acquire knowledge about the child's particularities in order to ensure that the child's fundamental interests are in fact met. Because they should share everyday life with the child, parents are best situated to fulfil the child's interest in continuous loving relationships. Like Brighouse and Swift, I privilege parents’ ability to provide children with loving, intimate, and lasting relationships as one element of the parental function and as a criterion of parental adequacy.⁵ They believe that children's interest is in receiving love and care from the person(s) with whom they stand in an authoritative relationship—that is, from their custodian(s). I make the more modest assumption that children's interest is in a lasting loving relationship with someone, and that parents, as custodians, are best placed to provide this.

4. The Case for the Genetic Procreator

Here are some familiar arguments to the conclusion that procreators have a right to parent their child in virtue of the genetic connection that they share with the child: People sometimes speak as if the mere genetic connection, in itself, could generate the right. But, if meant in earnest, this would be a very mysterious reason since it does not mention anyone's interests, and certainly is a naturalistic fallacy [Feldman 1992]. Moreover, even if the genetic connection alone could justify the right, it would not pick out the two genetic procreators who typically are, under current practice, the child's custodians: one has the same level of connection with, but no parental rights over, one's twin brother [Austin 2007: 20].

Some of the proper arguments for genetic procreators’ right to parent are grounded in self-ownership. One version assumes ownership in one's labour claiming that procreators, who exert themselves to bring the child into existence, have property rights over the child whom they produce [Narveson 1988]. This argument does not allow one to properly distinguish between the claim of the genetic procreator and the claim of the gestational procreator—or, if it does, it seems to favour the gestational procreator. As Barbara Hall [1999] noted, in her own version of an argument grounded in self-ownership, appeal to the procreator's labour cannot do any work unless the labour in this case has been exercised in relation to something that the procreator already rightfully owns. On Hall's view, genetic procreators have the initial right to rear the child because children come from the parents’ bodily parts. Hall's account, too, may be deficient if, as Hillel Steiner [1994: 246–8] argued, we do not own germ line information and hence ownership over children is shared by parents and the larger society. Whatever their merits, these views presuppose child ownership and hence are of the parent-centred type.

More promising arguments do not assume child ownership and point to other parental interests, hence standing a chance to be incorporated into dual-interest accounts. Elizabeth Brake [2015] notes that procreation is, for most individuals, one of the most important (maybe the most important of all) creative activities in which they will ever engage. According to her, individuals who contemplate parenting have a good reason to procreate with their own gametes, since this makes it more likely that the resulting child will have the genetic basis necessary for displaying certain features that the procreator values and wishes to perpetuate. A prospective parent can have reasons—some of which are agent-relative—to value certain features that run in her family and therefore an interest to try and pass them on. Sometimes a desire to have a genetically related child is criticised as narcissistic, because narcissism is likely to detract from parental competence [Overall 2012]. On Brake's account, the valued features need not be one's own—they can be those of a close relative—and therefore this criticism does not apply. Brake's reasoning is compatible with a dual-interest account but, in itself, it cannot be used to extend a dual-interest account into a view on how we acquire the right to parent a particular child: The interest identified by Brake can at most indicate a right to procreate, which is enough for an opportunity to pass on family resemblances. There is no need to supplement this right by a right to parent one's genetically related child—that is, to control the child's life.

Others attempt to derive the right to parent a child from the special duties that procreators are typically believed to incur by voluntarily bringing children into existence. The general idea here is that genetic procreators need the liberties and powers that are encompassed by the right to parent as a means for fulfilling their special duties. Because intentional procreators are causally and morally responsible for the existence of children, and children need the care of others if they are to avoid great misfortune, procreators are under the duty to ensure that children will not come to harm.

Some defenders of a causal account of parental duties argued that would-be adequate procreators have a powerful interest, and hence a right, to discharge their special duties towards children themselves. According to Porter [2015], a procreator can never fully transfer their obligations on to other people so that the procreator becomes free of any responsibility vis-a-vis how their offspring's life is going. For this reason, thinks Porter, the procreator who would be an adequate parent has the right to an opportunity to discharge this duty herself. To see how unconvincing this reasoning is as an account of a right to parent one's genetically related child, consider the following analogy. Assume that, from no fault of mine, I run you over with my car and break your leg, thereby incurring a duty to mend it. I happen to be a medical doctor and I can do a reasonably good job. It is true that, together with the duty, I acquire a right to prevent others from interfering with my mending your leg in case no other, more qualified, person volunteers to take over. Yet this right is entirely grounded in your interest, and therefore I cease having it if a more qualified leg-mender comes along and volunteers to mend your leg herself. If this other person could do the mending more quickly, or in a less painful way for you, or without leaving a scar, I am not permitted to prevent her from doing so by appealing to how important it is for me to fulfil my duty of fixing what I broke. If so, then the right of an intentional genetic procreator is derivative from the child's interest, and depends on who else, if anyone, is willing to parent the child instead of the genetic procreator.

Moschella [2016] argued that between genetic procreators and their children there is a personal relationship generated by the genetic connectedness, and that the genetic parent's love for the child is especially valuable; and therefore that genetic procreators have a non-transferable duty to love the child, which usually implies a non-transferable duty to love and raise the child. This generates a right to raise the child. Should another person take over this role in the life of the child, this would involve discontinuing the relationship between procreator and child. Moschella takes this relationship to personal because our genetic make-up is essential to our identity; this gives it a bodily aspect. But although the relationship is ‘bodily’ (thanks to the genetic connection), it does not necessarily involve physical closeness. This view is implausible. First, it is not clear why a relationship that is not embodied, even with someone who played a major causal role in determining our identity, can be called a personal relationship. After all, the parties may have never been in any kind of contact. For the same reason, it is unclear why discontinuing the relationship should have a negative emotional impact. Moschella indicates the potential worry that the child may feel rejected by her genetic procreators. But how likely this is will depend on background social expectations. Even granting Moschella‘s worry, at least in some cases other people are so much better than the child‘s procreator at entering an intimate relationship with the child that, all things considered, the child will be emotionally better off with them even if the discontinuation of the bodily relationship with the parent has somewhat offset the emotional interests of the child. Moreover, it is unclear that her conclusion follows; a genetic procreator can love the child they procreated and fulfil some of her emotional needs even without having the right to be the child‘s custodian. At most, the child's interest in having a relationship with her genetic procreators indicates that parents do not have a right to exclude genetic procreators for the child's life.

Finally, Olsaretti [2017: 75] thinks that part of the intentional procreators’ duty is to be willing, themselves, to enter the relationship with the child. Her reasons concern respect towards the child:

if someone knowingly and voluntarily causes someone to be needy of a relationship, her refusal to be a party to it can be plausibly viewed as a rejection of the other party as worthy of being in a relationship with.

This argument can be plausibly read as showing that (genetic) procreators who would make adequate parents have a duty to parent their child if no individual who would make a better parent is willing to parent that child. It is a convincing way of accounting for the grounds of a duty to (be willing to) parent a child that one has procreated and (as in the broken leg example, above) it explains a derivative right to prevent others from interfering with one‘s ability to discharge this duty. But it does not explain a right to parent that child, assuming that another individual who would make a better parent is also claiming custody, because in such a case the procreator's failure to parent cannot be interpreted as a rejection of, and hence lack of respect for, the child. To the contrary, the procreator's insistence on having custody in cases when another prospective willing parent would better further the child‘s interests may represent a lack of respect for the child.

Another child-centred argument that might be taken to support the genetic procreators’ presumptive right to custody has been given by David Velleman [2005]. As he notes, many individuals raised in closed adoptions, or whose procreation involved anonymous gamete donation, spend a significant amount of time, energy, and money in search of their genetic procreators. Velleman thinks that the strong desire of which such efforts are indicative is explained by an important interest in self-knowledge that only close acquaintance with one's genetic procreator can satisfy. According to this view, knowledge of your genetic procreators plays a crucial role in identity formation. Close acquaintance to your immediate kin, argues Velleman, gives you a broader context for creating meaning about your life than you could have in the absence of such knowledge. It is like having a unique sort of mirror that lets you understand and explore possible versions of yourself. Velleman aims to show that it is morally wrong to bring into existence a child via, say, anonymous gamete donation, and he concludes that ‘other things being equal, children should be raised by their biological parents’ [ibid.: 362]. Some resist the claim that the putative interests in identity-formation and self-knowledge can be satisfied only by acquaintance with one's genetic procreators [Archard 2005; Haslanger 2009]. But even if the claim was right, it does not seem to support the conclusion that it is in children's interests to be raised by their genetic procreators.⁶ Rather, it entails a child's right to come to know well, in a face-to-face context, the genetic procreator and to have access to family stories. This can surely be achieved by recognizing and enforcing a duty of genetic procreators to make themselves available to their offspring, as well as by requiring gamete donation to be non-anonymous and adoptions to be open. While fully child-centred, and in this sense a good candidate for a proper account of the right to parent a particular child, Velleman's argument cannot entail anything as strong as a right of the genetic procreator to control the life of the child in virtue of the child's interest in close acquaintance with the genetic procreator.

Finally, there is widespread belief that a genetic connection between parent and child is in the child's interest because, in virtue of the genetic connection, genetic procreators are more likely than genetically unconnected adults to bond with their children and to display the deep and selfless love that good parents display for their children.⁷ This is supported by studies finding that children are at significantly higher risk of abuse by their adoptive than by their genetic procreators [McMahan 2002: 376]. The relevance of these studies is unclear. First, there are many confounding factors that could explain why adoptive parents abuse more than non-adoptive parents do [Golombok 2015], and these factors are difficult if not impossible to isolate. Second, the studies do not look separately at genetic and gestational procreators, which makes it hard to conclude whether the genetic or the gestational connection is responsible for the lower levels of abuse in genetically related families. Moreover, other studies conclude that the rates of abuse in adoptive and foster families is very low, at least in the USA and Canada [Vopat 2007: 77–8], and that biologically related parents are more likely to abuse children than adoptive parents are [LaFolette 2010]. In any case, the genetic connection is clearly neither necessary nor sufficient for parental love, as the existence of bonded and loving adoptive parents and unloving genetic ones shows [Archard 2005]. Assuming that this belief is correct, all that it indicates is a relatively high likelihood, not the actual existence, of bonding and love. As David Archard noted [1992: 102], the mere likelihood of bonding and love has much less weight in establishing the right than do actual bonding and loving dispositions. As noted above, I believe that children have a powerful interest in establishing and maintaining loving relationships with the people who rear them. Therefore, if genetic procreators really are more likely than genetically unrelated individuals to bond with and love their child, this is the best reason to believe that the genetic procreator has the right to parent in situations when all of those who claim custody would otherwise make equally good parents of that particular child. But what if the gestational procreators tend to do even better than the genetic procreator on the same count—that is, with respect to fulfilling the child's interest in forming and maintaining a loving relationship?

5. The Case for the Gestational Procreators

Above, I indicated that some of the arguments meant to explain why genetic procreators have a right to parent would support the gestational procreator's claim at least as well. These are Narveson's claims about parental self-exertion (unconvincing on the ground of assuming child ownership) and the popular belief that biological procreators in general are more likely to love their children, empirical support for which comes from cases in which, typically, one of the genetic procreators is also a gestational procreator. But there are also distinctive reasons for supporting a case for the moral right to parent of the gestational procreator.

Two features of gestation have been considered to support the gestational procreator's right. First, gestation is unavoidably burdensome. Second, gestation is typically the context in which the relationship between parent and child starts.

Gestational procreators carry many of the burdens of procreation: financial and health costs (as actual disease/malfunctioning or significant risks), discomfort, emotional upheavals, undesirable lifestyle changes, and opportunity costs (for instance, concerning the gestational procreator's working life). Some of these costs are unavoidable; others could be socially mitigated. One of the costs that cannot be socialised is the experiencing of the ‘difficult’ emotions of pregnancy concerning the future child: worries about the child's well-being and, when the gestational procreator expects to raise the child, about their own abilities to parent well. Mere appeal to the costs of pregnancy, as some have proposed [Narayan 1999], would constitute a parent-centred reason. To make it part of a child-centred or dual-interest account, one would have to show that it is in the child's interest that the procreator who uniquely bears these costs also has the right to rear. Susan Feldman [1992] advanced a consequentialist argument along these lines, noting that gestational procreators can, during pregnancy, greatly influence the development of the foetus and therefore the future child's well-being. Making sure that gestational procreators have a secure right to the custody of the children they bear is, according to Feldman, the best means to motivate gestational procreators to take proper care of the developing baby. This is indeed a child-centred reason and, possibly, it indicates that it would be unwise to jeopardise gestational procreators’ legal right to custody. But it falls short of establishing that gestational procreators have a moral right to parent—that is, to take control of the child's life after birth—merely because, during pregnancy, it is in their power to advance or set back the interests of the future child.

Appeal to the second feature of gestation is more promising. I have argued that, typically, children come into the world being already in an intimate caring relationship with their gestational procreator [Gheaus 2012]. The relationship is based partly on the many emotional responses to the fact of pregnancy: anticipation, planning, hopes, imagination and projection, but also anxiety and doubt. The carrying of some pregnancy-related burdens may be part and parcel of the formation of the relationship. And, partly, the relationship is created through bodily interactions that, at least for the gestational procreators, have meaning and create a history of the relationship with the baby. This intimate relationship is to some degree mutual. While we cannot know what is going on in the minds of newborns, we know that they react positively to the presence of their gestational procreator whose voice and heartbeat they can recognise during the last phase of gestation; and newborns respond preferentially to their gestational procreator, physical contact with whom regulates the baby's hormone levels, temperature, metabolism heartbeat, and antibody production [Sansone 2004]. Because both the child and the parent have an interest in maintaining an intimate relationship, if gestation is the context in which such a relationship starts, this provides a good pro tanto reason to think that the gestational procreator has the right to custody. This is a right to continue a relationship that has already started before birth.

The argument has the following shape:

P3. There is a pro tanto right to maintain intimate relationships.

P4. Gestation usually involves an intimate relationship.

C2. Hence, gestation is a good proxy for the presence of a pro tanto right to parent one's gestational baby.

If successful, this explains the priority of the gestational over the genetic procreator with respect to the right to rear, both on a child-centred and on a dual-interest account.

P4 is contested. Some of my critics wondered whether it is appropriate to describe the relationship between gestational procreators and their babies, at birth, as the kind of relationship to the maintenance of which parties have a right [Brighouse and Swift 2014; Porter 2015]. It is true that, at birth, the intimate relationship between a gestational procreator and their newborns is, in some respects, unlike any intimate relationship between adults or even between a child and an adult (or another child.) The gestational procreator sees the baby for the first time, and may be unable to recognise her. The baby, being pre-verbal, can only show very little about her mental life. On the other hand, gestational procreators do report powerful caring emotions towards and, often, bonding with the baby with whom they have been, for several months, as physically intimate as it is possible to be; and, as noted, babies seem, at birth, to prefer their gestational procreator. The very thought of separating newborns from their gestational procreators (when the latter are unwilling) appears very cruel, and the cruelty charge is best explained, at least in part, by assuming that a significant attachment is already in place.⁸ Finally, miscarriages late in a pregnancy [Mullin 2015] tend to involve mourning, the depth of which is not plausibly explained by the loss of a dear project; more likely, the mourning is explained, at least in part, by the existence of attachment, at the same time physical and emotional, to the baby.

All of this need not imply that the relationship between newborns and their gestational procreators is as developed as intimate relationships between older individuals. But it is the most developed relationship that a newborn can possibly have, and, for this reason, it is worthy of protection. Facts about the complexity and richness of a relationship do not fully determine its relative importance for the individuals in the relationship. Consider a comparison between a rich friendship between two typical adults who are in close relationships with several other people, and a much less complex attachment between a cognitively impaired adult and her caregiver, who is the only person with whom the cognitively impaired adult has a relationship. The second relationship seems at least as worthy of protection, in virtue of its value to the individuals involved in it.⁹

Another worry with my argument is its lack of universality: not all gestational procreators will bond, at birth, to their babies; some will not know that they were pregnant, or are unwillingly pregnant. Perhaps gestational procreators who are not bonded at birth lack the right to parent their newborn—although the newborn's level of attachment may be independent from that of the parent, and that attachment alone, if sufficiently significant, could provide a powerful child-centred reason for the gestational procreator's right to rear. In any case, gestation is an imperfect proxy for the existence of a mutual intimate relationship and hence, possibly, of a moral right to rear; yet it seems good enough to justify the gestational procreators’ presumptive legal right to custody.

I ended the previous section by saying that the best case in favour of the genetic procreators’ right to parent is the higher likelihood that they will bond with the child. But the gestational connection can do even better on the same count: gestation makes it likely that some bonding already exists at birth—unlike the mere genetic connection which, at most, can indicate a likelihood that bonding will occur at some stage after birth. Moreover, in some situations, like those when a surrogate desires to parent the child whom she carried, appeal to gestation provides a plausible explanation of the desire as well as a reasonably good guarantee that bonding is actually present.

6. Implications of the Argument

It has been noted that gestational accounts in general deny the parental parity principle, which states that ‘being a mother doesn't make a person more of a parent than being a father, or vice versa’ [Kolers and Bayne 2001: 280]. Gestational accounts only explain the parental right of gestational procreators, who are most typically mothers. They have been deemed implausible for claiming that men can acquire parental rights only by association with a woman [Bayne and Kolers 2003]. This last criticism does not seem fair, since gestational accounts leave open the possibility that men can acquire parental rights through adoption without any need to associate with a woman.

I accept the implication that my gestational account puts women who are able to gestate in a privileged position. This is not particularly problematic if it is the result of a general account of parental rights grounded (fully or partially) in the child's interests. According to this view, the rearing of children should be (primarily) regulated not by appeal to the parents’, but instead by appeal to the children's, interests. Given biological facts, it is an unfair, hence in one way regrettable, situation that fertile women are better positioned than other individuals to acquire the moral right to custody. However, it may be on the whole desirable that we come into the world through the bodies of other people, with whom we usually are, at birth, already in some kind of relationship. Consider cases in which the gestational procreator shows no emotional investment in their newborns. One has reason to regret, on behalf of the baby, that no bond has been formed in such cases. If so, then the advantage that those able to gestate have in this context is not morally arbitrary, because it is beneficial to all of us.

I also deny that the initial privileging of fertile women that follows from my account needs to be as counter-intuitive as some have claimed [Porter 2015; Ferracioli forthcoming]. My account is compatible with thinking that other individuals acquire the right to parent through establishing intimate relationships with a child. Which individuals other than gestational procreators can acquire the right depends on the details of a full theory of the right to parent and the rights of parents. Such a theory should specify how many parents a child should have and what should be the relationship between co-parents. I cannot even gesture towards the answers here, but some considerations can mitigate the worry that my gestational account privileges fertile women too much. First, third parties, too, can relate to the foetus in such a way that at birth they stand in some kind of intimate relationship with the newborn, albeit one that is even less robust than that of the gestational procreator's [Gheaus 2012]. For instance, pregnant women's supporting partners can see the foetus and hear its heartbeat as early as the bearing mother does: they can feel the baby, talk to it and be heard by it, and can experience the fears, hopes, and fantasies triggered by the growing foetus. Second, I assume that bonding with a newborn can happen in a relatively short time—maybe weeks. It is not the case that, on my account, the gestational parent's privileged relationship with the newborn inevitably lasts for a long time. Further, it is likely that it is in the children's interest to have more than one custodian, and my account can easily and plausibly be supplemented to explain why gestational procreators’ partner(s) can fairly easily acquire the right to parent. Of course, these partner(s) need not be genetic procreator(s).

The last remark brings us to another general implication of my account. As I noted at the outset, by default each child has two parents; but if gestation is the normatively relevant biological connection between parents and children, we seem to lack a principled reason for this practice. How many people can have a right to co-parent a child should turn on grounds other than genetic connection. I find this implication of my views to be a positive feature rather than a bug. There is a long tradition—at least in Western cultures—of seeing children as property [Brennan and Noggle 1997; Goodin 2005; Archard and MacLeod 2010], and the basic structure of the institution of parenthood may continue to reflect this tradition [Brighouse and Swift 2014]. It is a good thing to be critical of this.

7. Conclusions

The overall argument of this paper has direct implications for the kinds of custody disputes presented at the beginning of the paper. In the case of embryo swaps, it indicates that the gestational procreator should be given custody, since—these being wanted pregnancies—the intimate relationship between gestational procreator and baby is likely to start before birth. In cases of adoption with consent from the gestational, but not from the genetic, parent, it is not clear that the genetic procreator had a right to parent in the first place, whereas the gestational procreator most likely did. This will turn on answers to several questions. How many parents should children have? On what criteria, other than an already started relationship with the newborn, can one acquire a (legal) right to parent a particular child? Are genetic procreators really more likely than genetically unrelated individuals to bond with, and love, their children? Even if at birth the genetic procreator had, in virtue of their genetic connection, a right to parent, it is unlikely that this right survived the establishment of an intimate relationship between the child and her adoptive parents. If the right to parent a particular child is justified by appeal to the child's interest, as I argued it must be, and if children have powerful interests in continuity of care, then the courts were wrong to attach moral relevance to the genetic connection. The main moral relevance of the genetic connection, I argued, is increased likelihood of future bonding and love¹⁰ —but in this situation bonding and love have happened with the adoptive parents. The courts ought to have ruled in their favour.

The surrogacy cases are more complicated. If genetic procreators really are more likely to fulfil the child's interest in a loving relationship than genetically unrelated individuals are, it means that, other things being equal, a genetic procreator has a presumptive right to parent. But when a gestational procreator also claims the right, other things are unlikely to be equal. I argued that, at birth, gestational procreators typically had already started to engage in an intimate and caring relationship with their newborns, and a desire to parent the baby on the part of the gestational procreator is likely to indicate actual bonding. A direct implication for gestational surrogacy cases is that the gestational mother ought to have the freedom to decide whether to parent the baby herself (and to return any compensation that she received for her services). But if the gestational procreator has a right to parent the baby only in virtue of the relationship that she is likely to have with the newborn, then she cannot alienate this right. The right exists (partially or entirely) in virtue of how the relationship will serve the interest of the newborn, and therefore severing this relationship, which is a consequence of the alienation of the right, deprives the gestational procreator of the basis for her right. The present account provides the basis of a radical criticism of surrogacy. This might strike some as implausible, since many people tend to distinguish between benign, fully legitimate, forms of surrogacy and morally problematic ones. However, if it turned out that a genetic connection between procreators and their children does not increase the likelihood of future bonding, and therefore is not normatively relevant, then the very practice of assuming that we have a fundamental right to rear children merely because we created them is deeply flawed. The best forms of surrogacy are hostage to this tradition—they cannot have better moral grounding than the practice on which they are parasitical. More generally, it is doubtful that a child-centred account of a right to parent, or even a proper dual-interest account, allows individuals to transfer their right to parent to another individual of their choice.¹¹

Additional information

Funding

This project has received funding from the Ramon y Cajal programme and from the European Research Council (ERC) under the European Union's Horizon 2020 Research and Innovation programme (Grant Agreement Number: 648610).

Notes

¹ For an explanation of what this is, see https://nyscf.org/pdfs2/FAQ_on_Mitochondrial_Replacement_Therapy.pdf.

² I use this classification for the purposes of this paper, and leave to one side the issue of third parties’ interest in ensuring that children are socialized so that they will not violate other individuals’ rights.

³ As well as because they seem to mandate a very different distribution of the right to parent than the status quo. In fact, child-centred accounts can resist most cases of ‘child redistribution’, especially after the parent-child relationship has been established.

⁴ I leave open the question of the relevant threshold.

⁵ See Liao [2015] for children's powerful interest in parental love.

⁶ Although, of course, much depends on what is included in the ‘other things being equal’ claim. Velleman seems to intend merely to exclude neglectful or abusive genetic procreators from parenting. If his claim is interpreted strictly, it can easily be incorporated into a child-centred account: if two individuals are equally in the position to parent well a particular child, and one of these individuals is the genetic procreator, it may be in the child's interest to be parented by this individual.

⁷ For an argument built on a version of this claim, see Ferracioli [forthcoming].

⁸ Caroline Whitbeck noted that the bodily side of this relationship can help to debunk the myth of a mysterious maternal instinct [1984: 191].

⁹ ‘Thank you’ to Jake Earl for this analogy.

¹⁰ The other one—a benefit in terms of self-knowledge—points to an interest that can be satisfied outside a custodial relationship.

¹¹ This paper has benefitted from generous feedback from David Archard, Jake Earl, Tim Fowler, Carolyn McLeod, Amy Mullin, Liam Shields, Hillel Steiner, and audiences in Birmingham, Umea, Louvain la Neuve, Manchester, Munchen, Graz, Linkoping, Montreal, Toronto, Edinburgh, and Barcelona.