Genetic parenthood and causation: An objection to Douglas and Devolder’s modified direct proportionate genetic descent account

Abstract In a recent publication Tom Douglas and Katrien Devolder have proposed a new account of genetic parenthood, building on the work of Heidi Mertes. Douglas and Devolder’s account aims to solve, among other things, the question of who are the genetic parents of an individual created through somatic cell nuclear transfer (i.e. cloning): (a) the nuclear DNA provider or (b) the progenitors of the nuclear DNA provider. Such a question cannot be answered by simply appealing to the folk account of genetic parenthood, according to which the genetic parents of an individual are those individuals who produced the egg and sperm, respectively, which fused to create the embryo. It cannot be so as in cloning there is no fertilization as such. In this article I critically examine Douglas and Devolder’s new account of genetic parenthood and demonstrate that it is vulnerable to counterexamples that exploit the lack of a condition specifying that genetic parents should cause a child’s coming into existence.

gle-parent" offspring, is a radical departure from the natural human way, confounding all normal understandings of father, mother, sibling, grandparent, etc., and all moral relations tied thereto.' 6 In a recent publication Tom Douglas and Katrien Devolder have proposed a new account of genetic parenthood. Their account, which builds on the work of Heidi Mertes, 7 aims to solve the question of who are the genetic parents of an individual created through cloning. They defend the view that the progenitors of the nuclear DNA provider are the clone's genetic parents. In this article I critically examine Douglas and Devolder's new account of genetic parenthood. Firstly, I show that Douglas and Devolder draw an incorrect conclusion when they apply their new account of genetic parenthood to answer the question of whether an egg donor is a genetic parent, in cases where the egg used in the reproductive procedure has been enucleated (e.g. maternal spindle transfer, cloning).
Secondly, I demonstrate their account is vulnerable to counterexamples that exploit the absence of a condition specifying that genetic parents should cause the child's coming into existence.
The article proceeds as follows. In the second section I briefly present and explain why previous accounts of genetic parenthood are flawed. In the third section I present Douglas and Devolder's new account of genetic parenthood and show that they draw the wrong conclusion when faced with cases where an egg has been enucleated for a reproductive purpose. In the fourth, and final, section I present a case that shows that their account of genetic parenthood is found wanting.

| ACCOUNTS OF G ENE TI C REPRODUC TION
In this section I review the accounts of genetic parenthood that Heidi Mertes investigated in her article 'Gamete Derivation from Stem Cells: Revisiting the Concept of Genetic Parenthood'; and explain why they are found wanting. I begin with Mertes, as Douglas and Devolder do likewise. The first account that she presents can be named the informational account of genetic parenthood: [A] child is my genetic child when it has 50% of my DNA Furthermore, it can also be the case that I share 50% of my nuclear DNA, in the informational sense, with someone who is not a close relative of mine, and that would entail that she is my genetic child or parent.
A more promising account, which Mertes discusses and which was first presented by Avery Kolers, 9 can be named the Direct Derivation Account of genetic parenthood: 'X is a genetic child of Y if X is directly derived from Y's genes '. 10 The direct derivation condition rules out the possibility of my siblings being my genetic parents, as I was not directly derived from them, and vice versa. As Kolers asserts: '[d]erivation is fundamentally a causal relationship; the offspring is as it is because of its relationship to its parents, whereas the inverse is not true '. 11 Mertes contends that the problem with this account is that it leads to counterintuitive conclusions when we examine two reproductive cloning cases. 12 Let us, following Douglas and Devolder, call the first case the Cloned Child: Stanford, CA: Stanford University. https ://plato.stanf ord.edu/entri es/cloni ng/ | 1087

PALACIOS-GONZÁLEZ
[A] couple (Mr and Mrs X) may become infertile after already having conceived one genetically related child (Y). As they long for a second child, they opt to clone their existing child Y, which results in the birth of Z. 13 According to the Direct Derivation Account, Mr and Mrs X are not the genetic parents of Z. They are not so as Z was not directly derived from their genes. On the other hand, Y is the genetic parent of Z, as Z was directly derived from Y's genes. Nevertheless, according to Mertes, in this scenario 'it is most likely that Mr and Mrs X will "feel" like the genetic parents of Z, whereas Y is unlikely to think of herself as Z's mother.' Let us now consider the second case that Mertes presents, that I will call Cloned Parent, also fol- The overall issue here, according to Mertes, is that we have contradictory intuitions on similar cloning scenarios. On the one hand, in Cloned Child we seem to support the idea that the progenitors of the nuclear DNA provider are the genetic parents of the clone. But on the other hand, in Cloned Parent we seem to support the idea that the nuclear DNA provider is the genetic parent of the clone. Of course, holding both statements as true at the same time is contradictory.
After presenting these two cloning cases Mertes does not bite the bullet and accept that, in Cloned Child, Z's genetic parent is Y.
And she also does not try to revise the Direct Derivation Account in order to solve this purported contradiction in our intuitions. What she does is assert that these two cases show that ' [t]here is no fixed, scientific, everlasting criterion of genetic parenthood that everyone can agree upon' and that 'the term genetic parenthood is not valuefree, but dependent on personal intuitions, intentions or judgements'. 16 Let us now move to Douglas and Devolder's account of genetic parenthood.

| THE MOD IFIED D IREC T PROP ORTI ONATE G ENE TI C DE SCENT ACCOU NT
Douglas and Devolder contend that Mertes's conclusion about genetic parenthood is premature, in the sense that she has not shown that genetic parenthood is 'a subjective concept that depends on the views of people about what sorts of genetic relation matter '. 17 Rather than falling for this subjectivist account of genetic parenthood they propose a new account, one that can deal with the Cloned Child and the Cloned Parent cases. But importantly, their account is intended to be one 'that captures the concept of genetic parenthood implicit in everyday usage'. 18 Before engaging with Douglas and Devolder's account, it is important to notice that they do not start from Mertes's presentation of the Direct Derivation Account, Suppose a sperm from P1 is used to fertilize an egg from P2. The resulting zygote then has its DNA removed and replaced by DNA from some other individual T. This zygote is then carried to term and eventually a child, C, is born. There is a sense in which C derives directly from P1 and P2's genes; those genes governed the development of gametes, which created a zygote from which C developed. But C's genes do not derive from P1 and P2's genes, they derive instead from T's genes, and this surely prevents P1 and P2 from qualifying as C's genetic parents. 20 Douglas and Devolder's Direct Genetic Descent, which is a revised version of Mertes's presentation of the Direct Derivation Account, runs into the same issues mentioned above when we consider the Douglas and Devolder here introduce the condition that the genetic derivation must be via a gamete produced by the intending genetic parent. Thus it follows from this account that the nuclear DNA provider, in cloning cases where the enucleated egg does not come from said nuclear DNA provider, is not a genetic parent of the clone.
In such cases he or she is not a genetic parent as the clone was not derived from one of their gametes. In Cloned Child, Y, who is the nuclear DNA provider, is not the genetic parent of clone Z. And in Cloned Parent, Mr X, who is the nuclear DNA provider, is not the genetic parent of Q. Bracketing the previous issue regarding egg donors, Douglas and Devolder assert that Direct Gametic Genetic Descent is not only problematic because the clones would not have genetic parents, which they do. They contend that Direct Gametic Genetic Descent is flawed because there are other cases that are 'closer to those of normal human reproduction' that also create problems for such an account. As a point in case, imagine that the following is possible: we enucleate a zygote, and we replace its nuclear material with the genetic material obtained from two somatic cells taken from individuals A and B. In this case each individual contributes 50% of the nuclear DNA material. 25 If this were ever to happen then, under direct gametic genetic descent, we would have to accept that any resulting child from this biotechnology, let us call it (following Douglas and Devolder) Two-Donor Genome Transplantation, would not have genetic parents. Yet it seems that in this case A and B would be the genetic parents.
In order to solve the problem posed by Two-Donor Genome Transplantation, Douglas and Devolder propose to replace the condition of genetic inheritance via gametes with a condition of genetic inheritance of some determined proportion from parent to child. If proportion X, which they do not specify, were to be <100% of the nuclear DNA, then under this account the nuclear DNA provider is not the clone's genetic parent. And if proportion X were to be > 0.1% of the whole DNA then the egg donor, in cases where we employ maternal spindle transfer, for example, would not be a genetic parent. Now, the upshot of the Direct Proportionate Genetic Descent account is that it does not entail the absurd conclusion that in Two-Donor Genome Transplantation the resulting child has no genetic parents. According to this account A and B would be the child's genetic parents. This would be the case if the proportion of genetic material established by this account (i.e. X) is set to include the proportion that obtains in cases of sexual reproduction. It is important to note that this conclusion does not necessarily follow from their account. It does not do so as proportion X could be a proportion of genetic material in the range between 50% and 100%. The fact that the nuclear DNA provider passes 'too much' of her genetic material does not rule her out from being a genetic parent in principle. Regardless of how we solve the previous point, Douglas and Devolder accept that 'the proportion of genes' will most probably not be a set figure but rather a range, and that this range could have fuzzy boundaries. Before I show why this account of genetic parenthood is found wanting, it is relevant to mention that according to it egg donors, both for cloning and mitochondrial replacement procedures, are not in principle ruled out as genetic parents. Why? Because, as stated above, the authors do not specify the proportion of genes that must be transmitted for someone to classify as a genetic parent. If the proportion were ≥ 0.1% of the whole DNA content of the human organism then the egg donor would be the genetic parent of the resulting child. This would be so as the mitochondrial genes of any resulting child would have been derived from the egg donor's genes, and not at all through some third intervening individual M. 30

| THE C A S E AG AIN S T THE MOD IFIED D IREC T PROP ORTI ONATE G ENE TI C DE SCENT ACCOUNT
Even if Douglas and Devolder's account avoids what they consider to be a counterintuitive conclusion (i.e. that Mr X is Q's genetic parent, and that Y is Z's genetic parent), they run into a more severe problem for not having a causation condition built into their account. In order to appreciate this issue consider the following case, which I will call 'Genome Editing '. Alfred and Betty want to have a child. They have been unsuccessful in achieving their goal through sexual intercourse, as both of Betty's fallopian tubes are completely blocked. Alfred and Betty decide to resort to Charles, a fertility expert. Charles prescribes Betty some fertility drugs and then proceeds to surgically retrieve her eggs. Once the eggs have been retrieved he uses Alfred's sperm in order to carry out in vitro fertilization, and zygote E is produced. After the IVF procedure Charles does not transfer the zygote back to Betty, but rather he lets it grow in his lab for 3 days.
On the fourth day Charles uses a Genome Editing technique in order to modify E's cells. In this case he inserts 10% of his own nuclear genes into E's cells. Finally, Charles transfers the genetically modified embryo to Betty, and after some months E is born.
According to Modified Direct Proportionate Genetic Descent both Alfred and Betty are E's genetic parents, if 'proportion X' is set to include a range between 5% and 50% of the total nuclear material. They are so in that 45% of E's nuclear genes were derived from each one of them, and not through deriving from the genes of some third intervening individual. The question to answer now is whether Charle, the fertility doctor, is E's genetic parent. When E is first created Charles is not his genetic parent, as he does not satisfy condition (a). But, Charles becomes E's genetic parent per means of a Genome Editing technology when E is 4 days old. This is the case as 10% of E's genes were derived from Charles' genes, and not through deriving them from the genes of some third intervening individual.
The fact that Charles becomes E's genetic parent shows that there is something wrong with Modified Direct Proportionate Genetic Descent. The problem is that this account allows for an adult to become the genetic parent of an already existing individual; and this cuts against one of the necessary conditions of genetic parenthood: that one is one of the material causes of an individual coming into existence. Adopting a stance that maintains that such a causation condition is not necessary entails a radical revisionism of the concept of genetic parenthood, a revisionist position that needs to be defended. Now, it could be the case that all our DNA is necessarily required for the capacity to regulate and coordinate metabolic and life processes to be there; and thus, if some of it were to be replaced by similar DNA, but from a different origin, then the original capacity to regulate and coordinate would be destroyed and a new capacity would be created. Even when it is intuitively appealing the former is incorrect. According to recent research 75% of our DNA is non-coding DNA. 35 This means that Charles could edit 10% of E's non-coding DNA, and this would not affect E's capacity to regulate and coordinate metabolic and life processes thus, also, not affecting E's numerical identity. Furthermore, my response to this second objection holds true even if it were to be the case that the findings of the Encyclopedia of DNA Elements project, which were able to assign biochemical functions to 80% of the genome, were true. 36 Let me finish by saying that Genome Editing shows that Douglas and Devolder's Modified Direct Proportionate Genetic Descent account of genetic parenthood is presently found wanting, and that they need to do more work in order for it to be a plausible account of genetic parenthood. 33 I thank an anonymous reviewer for raising this objection.