Paternity testing: a poor test of fatherhood

Abstract

In this paper we argue that there are few morally acceptable justifications for a man to seek a paternity test. The relationship between material responsibility and the burdens generated by the creation of a needy child, means it is not unreasonable for a man to establish that he is responsible for the creation of a child before accepting this financial responsibility. Some men embrace fatherhood irrespective of their involvement in the children's creation and in the absence of a genetic relationship. Others feel differently. Given that parenting is a life-long commitment it is reasonable for those for whom genetics matters to request a paternity test prior to making this commitment.

Once a man has become a (moral) father there is no justification for initiating paternity testing, save in the interests of the child. Where the child's interest in knowing his/her genetic origins is unclear we argue this is a decision that can be left for the child to make for him/herself when he/she is sufficiently mature. There is no justification for performing infidelity testing on a child. Because fatherhood is not contingent on genetic relatedness, suspicion of misattributed paternity is no justification for paternity testing.

The justifications for, and problems with, paternity testing suggest that it is something men should seek antenatally, at birth or not at all.

Keywords:

• paternity testing
• fatherhood
• moral interest
• genetic relatedness
• moral fatherhood

Introduction

Men are becoming increasingly concerned about misattributed paternity. On the one hand, this concern is related to the financial implications of paternity in state systems that pursue so-called ‘absent fathers’ for child support. Few men – particularly those not in a stable relationship with the mother of the child – are willing to pay without some evidence that they are indeed the absent father. On the other hand, some previously willing and active fathers have discovered misattributed paternity with devastating effect on their general well-being, as well as on their relationship with their children and partners. Some men have sought financial redress for misattributed paternity arguing that it amounts to ‘paternity fraud’ (Draper 2007).

At the same time, in the UK, those who are conceived as a result of gamete donation after April 2005 (once they become adults) will be able to access the identity of ‘their’ gamete donor (Human Fertilisation and Embryology Act 1990, modified June 2004). This change, from previously assured donor anonymity, sought to give donor-conceived people the same right to information about their genetic origins as those who had been adopted (Johnson 2004). This move was not without its critics. There were those who thought that the change didn't go far enough, because the removal of donor anonymity was not accompanied by legislation requiring the disclosure of information about the circumstances of their birth to donor-conceived children (British Association for Adoption and Fostering 1995). Such critics have argued that the birth certificates of donor-conceived individuals should either state that they were donor-conceived or include the identity of the donor(s). Anything less and donor-conceived individuals might never realise that they were not genetically related to their parents.

Not knowing one's genetic parents, and consequently not knowing one's genetic history, is often regarded as a significant harm, which can impact negatively upon an individual's sense of identity and self-worth and can leave them feeling disconnected and alone (Sants 1964, McWhinnie 1996, Turner and Coyle 2000, Ryburn 2005) – although whether or not this ‘genealogical bewilderment’ (Sants 1964) leads to tangible mental health problems, as some suggest, is open to challenge (Humphrey and Humphrey 1986). This significant harm has been cited as a foundation on which to ground a ‘right to know’ one's genetic parents (McWhinnie 2001), although, as Frith (2001) notes, the evidence for harm is mostly drawn from research with adopted children, and may not be applicable to donor conception.

This supposed right to information about genetic origins for donor-conceived individuals also calls attention to the fact that many of us might never learn the truth about our genetic origins. Estimates for misattributed paternity vary quite widely from 0.8% to 30% (Bellis et al. 2005). If, however, the harm caused by not knowing one's genetic history is sufficient to ground a right to know, then this right should be available to all of us – however we were conceived (Draper 2005).

In this paper we explore some of the ethical and conceptual problems associated with paternity testing. We explore two triggers for testing – the interests ¹ the man has as a father/potential father and the interests of the child. We argue that both of these triggers highlight difficulties with the ways in which paternity testing is sometimes justified, which are especially apparent when a man seeks a paternity test after a paternal relationship has already been formed. We conclude that specifically paternity testing should be performed at birth, antenatally, or not at all. ²

We will begin by discussing, in the context of some recent empirical research, what we mean by ‘father’.

What is a father?

This is an important question because a test for fatherhood may be quite a different kind of test to a paternity test. A fatherhood test cannot even be devised until we know what it is that we are testing for; or, put another way, until we know what a father is. If ‘father’ merely refers to a particular kind of genetic relationship, then a paternity test may be sufficient. However, at best, this produces an exceedingly thin view of what it means to be a father, which can be met by a man giving a drop of sperm to help in the creation of a child who he has nothing more to do with. If this is all that being a father means a paternity test would be an appropriate test for fatherhood. But being a father denotes much more than this, which is clear both from our examination of the term as a philosophical concept and from our empirical work (Ives 2007, Ives et al. 2008). We are not alone in questioning the primacy of biological relatedness in familial relationships – either on sociological or moral grounds (see, for example, Smart and Neale 1999, Weston 1991, Dowd 2005, Kaebnick 2004, Fuscaldo 2006, Bayne and Kolers 2003). We are also not alone in noting a fragmentation in the concept of fatherhood (for example see Sheldon 2005, Collier and Sheldon 2009). Collier and Sheldon, however, discuss the fragmentation of fatherhood from the perspective of observers of socio-legal discourse, whereas we are discussing fragmentation in normative terms – where fragmentation is seen as morally necessary in order to unpick and accommodate the wide-ranging rights and duties accompanying the various aspect of modern fatherhood.

We suggest that the meaning of ‘father’ has three components that may be found in the person of one father, or spread amongst different men. These are ‘causal father’, ‘material father’ and ‘moral father’. ‘Causal father’ refers to the man responsible for bringing a child into existence, with one such causal route being the provision of the sperm in natural reproduction. ‘Material father’ refers to a man who provides materially for a child (in terms of providing physical care, or the means of providing physical care, such as food, shelther, clothing, etc.). On the other hand, ‘moral father’ refers to a man who forms a loving parental relationship with a child. Moral fatherhood seems to be that which is valued most highly, both philosophically and by the men who contributed to our, and others', data (Ives et al. 2008, Henwood and Procter 2003, Smeaton 2003, Cohen 1993, Lloyd 2003). The distinction between moral and material fatherhood can be best understood by considering an absent father who pays child support but has no social parental relationship with the child. Such a man fulfils the criteria for material fatherhood because he is providing the means of physical support (through providing financial support to the mother) but is not a moral father because he does not engage in any kind of parental relationship (there is no love, no day-to-day care, no sense that the protection of the interests of this particular child are vital to the protection of his own interests and happiness, and no sense that the interest of the child are put before his own).

Being a ‘progenitor’ – the one who provides the sperm – may be significant in terms of providing a moral reason for becoming either a material or moral father but it lacks any normative value. Rather, what is important to ‘fatherhood’ is the formation and continuation of a particular kind of relationship, which is distinct from biological kinship. (Ives et al. 2008, p. 82) Causing a child to exist carries, under most circumstances ³ , an obligation to ensure that the material costs of raising the child are met, and this is premised on the notion that a person is morally responsible for the foreseeable consequences of their voluntary actions. ⁴ Whilst we acknowledge that causal fatherhood incurs moral responsibilities, our claim that it nonetheless lacks normative value is based on our empirical work that found progenation itself is not valued as a moral ideal, and where it is valued it is valued instrumentally – for what it is thought to achieve socially (Ives et al. 2008). Genetic relationships, in this sense, can generate moral responsibilities without being intrinsically valuable in themselves. Our argument is that even if a man has no desire to be a moral father, and even if the conception was unintended, there is a duty of justice to ensure that the other party to the procreation is not left to carry the financial burden of caring for the resulting child alone. Child support is, we argue, therefore owed not to the child but to whoever has to bear this burden, and most often this will be the mother. This material relationship is, therefore, primarily one with the mother (or whoever has the financial burden of caring for the child).

In many paternal relationships, however, fatherhood is not fragmented along these lines and the ‘moral father’ not only cares for the child practically and emotionally, but is also the causal father, and shares financial/material responsibility for her upbringing with the mother. Through being a moral father, we argue, a man acquires ‘paternal rights’, including the right to make decisions that affect the child's life. Only a man who demonstrates he is able to act responsibly as a father and look after the interests of a child should have parental rights (including, for example, the right to determine residency, to take decisions about schooling and healthcare, etc.). This claim is premised on the notion that parental rights are, in essence, significant powers over a child, and that such power is only legitimately held, and wielded, by a person who has demonstrated a loving commitment to a child. A sufficient commitment, we believe, is demonstrated by a man taking on the role of ‘moral father’, and actively taking on and discharging parental responsibility.

But not all moral fathers are also progenitors. Social changes (like relationships between adults that are less enduring than traditional life-long marriage, adoption and assisted conception techniques using donated gametes) have meant that many men now live with, and are fathers to, children to whom they are not genetically related.

We disagree on what this ‘fragmentation’ means for the definition of fatherhood. One of us (Ives) thinks that the term ‘father’ now has different meanings and that we simply need to be clear – using appropriate prefixes (i.e. causal/progenitor, material, moral) to specify which meaning we are referring to. The other (Draper) thinks that there should be only one use of the word ‘father’ – that encompassed by moral fatherhood – and anything less than this should not be referred to as ‘fatherhood’. This disagreement, however, is not over whether these different kinds of relationship exist, or the value we should attach to them, but over what terms should be used to describe them. We are in agreement, for example, that the ‘fragments’ (or components) we have described accurately account for the different kinds of relationship between man and child – we simply disagree over whether or not all of these relationships can be denoted properly by the term ‘father’.

We also agree that there can be more than one (moral) father for a given child, although obviously only one progenitor. (Moral) fatherhood is not exclusive and many fathers in combined families, for instance, share paternal burdens, duties and pleasures with at least one other man. These kinds of arrangements, however, may well be informal, with only one of the men possessing legally recognised parental responsibility.

Tests for fatherhood

Once we are clear about what it is we mean by ‘father’ we can see what kind of tests might be able to tell us which men are fathers, and in what sense, and how markedly these tests differ from a standard paternity test.

If Draper is correct, and the only kind of father is a ‘moral father’, then the test for fatherhood would depend upon the answers to a series of questions about the nature of a man's relationship with a child. These questions would elicit objectively verifiable answers about the provision of care, feelings of responsibility for a child, nurture, moral guidance and love. The answers would suggest a voluntary relationship, motivated by paternal love to protect the interests of the child concerned. The nature of the love might be what distinguishes being a father from other kinds of relationship, e.g. nephew, professional carer or friend (in the case of older children).

If Ives is correct, we may require two further kinds of tests. A test to determine causal fatherhood would focus on establishing which man (if any) stands in a morally relevant causal relationship to the child. In cases of natural reproduction, excluding occurrences of ‘sperm banditry’ (Sheldon 2006) or rape, a paternity test, because it would reveal the progenitor, is an appropriate test of causal fatherhood. Material fatherhood, on the other hand, would be demonstrated by means of having offered and usually also having delivered financial and physical support. ⁵

Triggers for paternity testing

Given this understanding of fatherhood and means of determining it, when might paternity testing be appropriate? The answer to this question depends in part on what motivates or triggers the test.

Triggers related to the interests of the man

i. A preference for a genetically related child

A man may want to ensure that he is the progenitor before taking on financial and other responsibilities for a child. This preference could be the result of any number of factors: for instance, a general unwillingness to be a parent, a reluctance to increase financial out-goings unnecessarily, the feeling that he is unable to make an emotional investment in a child to whom he is not genetically related or a strong desire not to be cuckolded. Assuming, as we do, that (moral) fatherhood is a voluntary activity, all of these reasons are morally defensible, although not equally laudable. There is no general personal obligation to take responsibility for a newborn whose existence does not originate from some action or negligence of our own. If genetic relatedness matters to whether a man is willing to be a father to a child (material, moral or both), then he is justified in instigating a paternity test before he enters into any kind of paternal relationship. Given that antenatal testing of foetuses using the mother's blood is not yet completely reliable (and given that the test would also require the mother's consent), the first practical opportunity for testing would be at birth.

If, on the other hand, the presence or absence of genetic relatedness is not important to his decision to become a father, he has no reason to request a paternity test. Further, nor he is necessarily obliged to participate in testing before taking on parental responsibilities since (moral) fatherhood is not dependant on being the progenitor.

ii. The possibility of misattributed paternity

Given the current legal association in the UK between genetic relatedness and financial obligation (in the case of natural reproduction), a man might believe that he is justified in avoiding further financial (and perhaps other) responsibilities for a child whom he is already fathering if he can demonstrate that he and the child are not genetically related. Even if, however, the result demonstrated the absence of genetic relatedness this fact alone would not negate future responsibilities to the child because the paternity test is not a test of (moral) fatherhood – but merely a test of causal fatherhood by progenation. Whilst the paternity test may demonstrate that the man did not have the initial personal obligation to take responsibility for the child as a causal father, by virtue of taking on the mantle of (moral) fatherhood he has acquired obligations to the child that cannot now be lightly put aside. Although the financial obligations that are generated by causing a child to exist are owed in the first instance to others (usually the mother) who would otherwise have to shoulder them alone, once (moral) fatherhood has been acquired this obligation passes to the child as part and parcel of protecting his/her interests. Thus, just as the presence of a genetic connection does not make a man a (moral) father, so its absence does not put a stop to (moral) fatherhood.

But what of the man who genuinely believed himself to be the progenitor and, only because of this belief (reinforced by a sense of financial obligation borne from being partially responsible for causing the child to exist), became a (moral) father? Indeed, let us suppose that this man was actually an unwilling father in the sense that his only motivation for becoming one was his sense of obligation to his genetic offspring. Arguably, it would be unfair to impose a continuing obligation because had he known that he was not the progenitor he would not have taken on the responsibility in the first place. We are sympathetic to the view that, because (moral) fatherhood should be acquired voluntarily and because misinformation (deliberate or otherwise) affects voluntariness, there is some degree of unfairness in operation here. This is, however, unfairness that operates independently of the relationship with the child. The unfairness arises from his relationship with the child's mother, as Draper (2007) has argued elsewhere. The lesson to be learned from a case such as this one is not that we should alter our views about fatherhood to accommodate uncomfortable truths about the way adults conduct their relationships. Rather, the lesson is that if it matters to a man that he is genetically related to his child, he should seek a paternity test before becoming a (moral) father. This speaks to the wisdom of seeking the test at the earliest possible opportunity, which in most cases will be at or around the time of the child's birth.

Finally, a distinction must be drawn between paternity testing and infidelity testing. Where a relationship is based on an assumption of monogamy, misattributed paternity is evidence (mistakes in infertility clinics and accidentally switched babies in maternity units aside) of infidelity. Paternity testing could, therefore, be carried out not to establish absence of paternity but to establish infidelity. As we have discussed elsewhere (Ives and Draper 2005), there is a danger that a negative paternity test could result in the loss of legal rights and responsibilities for a child. It could, therefore, be imprudent for a (moral) father to use paternity testing to prove infidelity. Indeed, it could be argued that it would be a failure of paternal virtue for a (moral) father to put in jeopardy his relationship with his child for this reason. This is another reason to keep a clear distinction between relationships between parents and between parents and their child, and between tests for infidelity and paternity testing.

iii. To exclude the claims to fatherhood of another man

As we have just discussed, infidelity (suspected or proven) should not impact on a (moral) father's relationship with his children. Some cases of infidelity, however, may involve a second man who wishes to assert his own claim to be the father of any resulting child. We can distinguish two sets of circumstances here. The first is where the disputed paternity occurs before or near to the birth of the child. The second is when a dispute arises sometime after the birth of the child and when there is already an established (moral) father.

The first type of case is covered by i. above. How matters finally resolve will depend in part on how the relationships between the adults concerned resolve themselves. If the mother and the non-progenitor form a family in which the child will be raised, then it is likely that the non-progenitor will become a (moral) father. However, the progenitor still has a prima facie material obligation, which he might be unwilling to abdicate, and exercising this obligation would, on our model, pave the way to his becoming a (moral) father. He should at least not be prevented from doing so unless this is demonstrably contrary to the interests of the child. In this case, the child will have two (moral) fathers and the onus will be on all the parents to exercise sufficient parental virtue to ensure that their behaviour towards each other does not damage the child's interests. If the mother and the progenitor form a relationship, then there is little scope for the non-progenitor to acquire (moral) fatherhood without the active blessing of the other adults.

With regard to the second kind of circumstance, given the nature of (moral) fatherhood as we have described it, no third party, progenitor or not, can undermine a (moral) father's claim to fatherhood. A paternity test is not a test for (moral) fatherhood, and even if a negative test result was forthcoming this would not affect a (moral) father's status as a (moral) father. It is, therefore, unnecessary – ineffective, even – for a (moral) father to seek a paternity test as a means of excluding another man as a potential father to his child in order to protect his own interests in being a father. Moreover, as we pointed out above, nor is a (moral) father obliged to participate in a test. On the other hand, as we will shortly be discussing, it may be argued that it is in the interests of the child for the matter of her genetic origins to be settled. If this is the case, then the (moral) father may be obliged to capitulate for the sake of his child. If the results show that he is not the progenitor, this does not affect whether or not he is the (moral) father. But, given that the progenitor (as a causal father) may have a material obligation, given that he might not chose to abdicate this obligation to the existing (moral) father, and given that in exercising this obligation he sows the seeds for acquiring (moral) fatherhood, more needs to be said about what follows from the result.

The first point to make is that excluding a third party is not necessary in order to maintain and protect the paternal interests of the existing (moral) father. Next we need to recall that (moral) fatherhood is not necessarily exclusive. This means that an existing (moral) father has no a-priori reason to exclude another potential father. What matters, then, is the a-posteriori fact of whether or not maintaining an exclusive fathering relationship with the child is in the child's interest. The child's interest, here, will of course be closely related to those of the existing parents, and it is likely that if the existing (moral) father was unwilling to admit a second moral father, then doing so would impact negatively on the child to the extent that it would be preferable to maintain the original, exclusive, family unit. On the other hand, given that the justification for our hypothetical test was that it was in the interests of the child to know his/her genetic origins, the (moral) father may have to do his best (parental virtue comes into play again here) to accommodate some kind of relationship with the progenitor.

This brings us to a discussion of whether paternity testing can be justified in the interests of the child.

Triggers related to the interests of the child

Modern paternity testing is a very minor procedure requiring only that cells are taken by a swab on the inside of the cheek. This is hardly a serious intervention and is probably less distressing to even the most reluctant minor than having her teeth cleaned against her will. It is not, therefore, the intervention itself which generates questions about the child's interests but the potential impact of the test results, or the significance of information that will remain unknown if the test is not conducted.

It is often simply asserted that a child has an interest in knowing her genetic origins, although whether this interest should have the status of a right is open to question (Draper 2005). Moreover, it may be unclear what precisely this interest amounts to, especially when the results are unlikely to alter the immediate relationships she has with the adults concerned, or can only alter them in a way that is likely to be detrimental (as, e.g., in the case of misattributed paternity when an existing father severs his relationship with the child concerned). An interest in knowing one's genetic origins is sometimes regarded as the inverse of the harm caused by not knowing where, or from whom, one originates. Whether one's identity is actually dependent on knowing one's genetic history, and whether or not knowing leads to any kind of tangible harm, can probably not be demonstrated a-priori; but many people do feel this way (Whipp 1998, Adie 2005, Turner and Coyle 2000). This implies that there is not a necessary connection between knowledge of genetic history and a satisfactory sense of identity but, rather, that this connection is contingent upon the perspective of the particular individual. It is, therefore, too simplistic to say that all children need to know their genetic history – and more accurate to say that some people come to feel it is vitally important to them to have this information, whilst others may not.

As Draper (2005) notes, however, the interest described in the previous paragraph in knowing genetic origins is often shorthand for something more substantial. In the case of people conceived as a result of gamete donation, or of those who were adopted as infants, knowing one's genetic origins is about being able to meet and form relationships with the adults concerned, and can lead to a sense of being entitled to know about and have the opportunity to meet and form relationships with hitherto unknown genetic siblings. Simply identifying genetic parents or siblings may not be sufficient to satiate ‘genealogical bewilderment’. Rather, there may be an expectation that the identified individuals should also be willing to form a relationship, and this requires at least some degree of cooperation. No-one can be forced to have a meaningful relationship with either an adult or a child, and thus simply identifying a genetic connection is not sufficient to establish the substantial relationship that donor-conceived or adopted people often seek with their genetic relatives. It is somewhat ironic that what people may want, and expect, to flow from this kind of genetic information is some form of social relationship. Underlying this may be the assumption that genetic relationships act as natural facilitator of social relationships and, as we (Ives et al. 2008) have observed in fathers, this suggests that in some cases genetic relationships are not valued for themselves but for what they are thought to represent socially. Genetics are, in this sense, instrumentally, rather than intrinsically, valuable.

The fact that the wish to know one's genetic history is so often coupled with the wish to form substantial relationships means that knowledge comes with the risk of the harm of rejection if genetic parents or siblings do not wish to form a relationship. This is especially problematic if the decision to perform a test is based upon the oblique assumption that having the knowledge about identity will lead to having a relationship with those identified. It is impossible to determine, in advance, if the harm from not knowing is greater than the harm of knowing and being rejected. These potential harms seem insufficient to warrant banning paternity testing, but neither are the potential benefits sufficient to warrant making it compulsory. Both are, however, sufficiently serious not lightly to assume that it is always in a child's interest to know.

Given that the harms of not knowing one's genetic history and the benefits of knowing are equally contingent and none are guaranteed, it may be appropriate to apply the same criteria to paternity testing as are applied to genetic testing for late-onset genetic conditions (Clarke 1994). The offer of testing, and the decision to test, should (immediate and urgent need for medical treatment aside) be deferred until adulthood, or until the child reaches a level of maturity when she could make her own informed choice based upon the significance she attaches to her genetic origins.

Arguably, things are different when the child is very young and the outcome of the test may have the effect of turning a reluctant father into a willing one: minors do have a clear interest in having willing and good parents. Accordingly, it may be in the interest of an infant for a paternity test to be performed to persuade a man to take a financial and emotional interest in her upbringing. It is less clear that it is in the interests of a minor for the test to be performed to justify taking punitive action against an unwilling man that will force him to provide financial support that would otherwise have been provided by the state. It makes no difference to the infant what the source of the financial support is if the amounts are roughly equal and if the results do not cause the man to reconsider his reluctance to fulfil other roles associated with being a (moral) father. Unless the amount of financial support being sought will make a difference to the life of the child, the issue of child support alone would not seem to pass the best interest (of the child) test, given that we, in the UK, live in welfare state. It may be in the interest of the state to instigate paternity testing if doing so would reduce the welfare burden, but the interests of the state do not always coincide with the interests of an individual child, and the two sets of interests should not be conflated.

Courts in the UK are obliged to make the interests of the child their paramount concern. Whilst it is often assumed that parents will act in the best interests of their children, they cannot be similarly obliged to regard the interests of their child as paramount in all circumstances. Parents may have obligations to more than one child and may have to prioritise the interests of one child over those of another. Similarly, especially in the case of more trivial decisions, it is usually morally sufficient for parents to act in a manner that does not seriously compromise the interests of their child. At other times, the interests of parent and child may be in tension; for example, when the parent has to choose between a much longed for promotion or personal relationship that, to be successful, requires moving many miles away from the child's current school and friends, perhaps disturbing his study for significant adolescent exams. Parents generally make these kinds of choices without interference from the courts. Paternity testing can present similar kinds of tensions between the interests of parents and children.

Obtaining certainty in the case of misattributed paternity is one such example. Misattributed paternity often emerges in the context of the breakdown of relationships. It may provide concrete evidence of infidelity and spousal deception. A man who believes in honesty and fidelity in relationships may have a strong interest in the certainty that a paternity test may provide. But is this sufficient reason to consent on behalf of a minor to a paternity test? Another example is where a man values biological information for its own sake – perhaps along the lines described by Waldby (2002). In cases such as these, the value of this information to the man must be weighed against the potential for harm the information might bring to the child.

These kinds of scenario test the significance of both the interests of the minor in such decisions, and also the significance of genetic origins per se. If a minor has a longstanding relationship with her (moral) father, she clearly has no interest in this relationship being unsettled. Given our definition of (moral) fatherhood, there is no ‘truth’ to learn about who her father is. A paternity test may only demonstrate who her progenitor is not. Even if a potential material provider and aspiring (moral) father is waiting in the wings it is no easy matter to exchange one (moral) father for another (Kaebnick 2004), though the child could benefit from having two (moral) fathers.

We close with two observations about paternity testing instigated by (moral) fathers. In the case of minors, parental consent or a decision of the court is required before testing can take place. ⁶ Any decision made should reflect the interests of the child concerned. We have argued that the harm of not knowing one's genetic origins is a contingent harm and it should not, therefore, be assumed that all children will benefit from knowing their genetic history in all circumstances. For this reason we have argued that acquiring knowledge about genetic origins should, all things being equal, be treated like knowledge of late onset genetic disorders – something best left for the child to decide for him/herself once he/she is sufficiently mature to do so. Our first observation is that for this reason there will be few circumstances where a (moral) father ought legitimately to consent to paternity testing. Our second is that if the purpose of the test is to establish the absence of genetic relatedness with a view to presenting this as justification for ceasing to behave as a (moral) father, not only is this an erroneous justification (as we argued above) but his motivation undermines his moral authority to make the decision. Given that the authority for consenting on the child's behalf is derived from his status as the child's (moral) father, and given that his intention in instigating the test is not to serve the child's interests but rather to demonstrate that he is not after all her father, it is difficult to see how this man has the moral authority to consent for the test on behalf of the child, regardless of whether the law regards him as a person with parental responsibility.

Conclusions

We have argued that there are few morally acceptable justifications for a man to seek a paternity test. Given the relationship between material responsibility and the burdens generated by the creation of a needy child, it is not unreasonable for a man to seek to establish that he has been party to the creation of a child before accepting this financial responsibility. Some men are willing to become (material and moral) fathers to children irrespective of their involvement in the children's creation and whether or not there is a genetic relationship. Others feel differently. Given that parenting is a life-long commitment it is not unreasonable for those for whom genetic relatedness matters to request a paternity test to establish genetic relatedness prior to making it.

Once a man has become a (moral) father, however, we see no justification for initiating paternity testing, save in the interests of the child. Where it is not clearly in the interests of the child for her genetic origins to be established (urgent medical needs might be one such justification) we have argued that this is a decision that can be left for the child to make for herself when she is sufficiently mature. There is no justification for performing infidelity testing on a child. Because fatherhood is not contingent on genetic relatedness, suspicion of misattributed paternity is no justification for paternity testing.

On the other hand, a (moral) father ought not to feel that his position as a parent can be undermined by another man seeking a paternity test, even though this may result in the child having two (moral) fathers. In cases like this, however, both parties should consider the interests of the child, particularly as what is at stake here is the right to parent, which ought properly to be located in the exercise of the desire to be a responsible and loving parent.

The justifications for, and problems with, paternity testing suggest that it is something best sought by men antenatally, ⁷ at birth or not at all.

Notes

^1. Here we use ‘interest’ to mean a moral interest. This can be distinguished from interest as in a hobby or the like. Interests are not always to be equated with desires. Many people desire that which they have an interest in – continuing to live might be an example here. But not everything that we desire can be regarded as an interest simply because it is desired. Having an interest provides others with a moral reason to take this interest seriously, but it is not necessarily the same as asserting a right to it. That someone has an interest does not necessarily mean that this interest cannot be outweighed by the interests of someone else.

2. In making this argument, we recognise that people may desire to know the results of all kinds of genetic tests on other people for a variety of reasons. Waldby (2002), for instance, observes the value that people place on genetic information – its ‘biovalue’. Likewise, the information may be used to inform medical decisions or other major life choices (for instance, whether or not to have children or whether to save for the future or enjoy the present). And some people are enormously curious about their actual and potential genetic relationships with other people. In this paper we cannot evaluate every potential reason for seeking a test that provides genetic information, or even all kinds of genetic tests. We recognise that individuals might have a multitude of reasons to acquire a specific piece of genetic information or may value certain kinds of genetic relationships – perhaps without having reflected on whether this value is warranted. Rather, we are targeting particular justifications for a particular kind of test.

3. We should note that ‘under most circumstances’ is intended to exclude ‘thin’ causality – that which is insufficient to generate a moral responsibility for the child's existence. So, for example, both the fertility specialist and a gamete donor may play some causal role in the creation of a child, but this causal role is too thin to generate a moral responsibility for the child's existence. In this case, responsibility lies with the parents who initiated the parenting project since without their instigation neither the skills in infertility treatment nor the donated gametes would have been required or put to use.

4. See Fulscaldo (2006), Austin (2007) and Weinberg (2008) for good discussions of causal arguments of moral responsibility in relation to fatherhood.

5. Material fatherhood would usually be established on basis of providing physical support (or the means to it), but it may be sufficient to have shown willingness to do so – for example where disability prevents a man from being able to provide it.

6. In the UK, the consent of all parties to the test is required; The Human Tissue Act 2004 criminalised non-consensual DNA testing. Our point here is not related to the consent required from the adults concerned for the comparison with their own genetic material, but rather centres on consent for the child's tissue to be tested.

7. Assuming that this can be done using maternal blood sample.