Welcome to another NDPR Forum, this one on Joseph Millum’s The Moral Foundations of Parenthood (OUP 2018), recently reviewed in NDPR by Liezl van Zyl. As always, all are welcome to join in on the discussion.
From the OUP blurb: “Most people believe that parents have moral rights and responsibilities regarding their children. These rights and responsibilities undergird the nuclear family and are essential to the flourishing of its members. However, their basis and contents are hotly contested. Do a child’s genetic parents have a right to parent her? The importance of genetic ties is affirmed by many people’s gut responses, everyday talk, and many court decisions, but the moral justification for tying parenthood rights to genetics is unclear. Parents are routinely permitted to make far-reaching decisions about their children’s medical care, education, religious practice, and even how to punish them. When can parental rights be limited by the interests of the child or society?
…
“In The Moral Foundations of Parenthood, Joseph Millum provides a philosophical account of moral parenthood. He explains how parental rights and responsibilities are acquired, what those rights and responsibilities consist in, and how parents should go about making decisions on behalf of their children. In doing so, he provides a set of frameworks to help solve pressing ethical dilemmas relating to parents and children.”
From the NDPR Review: “An alternative strategy, and one that I find more reasonable, is to give up the attempt to base parental rights and duties in a single concept or principle.
“This is the strategy Millum adopts. His theory of moral parenthood has four parts. (1) Parental rights are based in desert: they have their origin in parental work, that is, work directed at the flourishing of a child. (2) The content of some of these rights is grounded in the interests of children, whereas the content of others is grounded in the interests of parents. (3) Parental responsibilities arise through certain voluntary actions that gain their significance from social convention. (4) Their content is determined by children’s justice-based claims. Millum faces two important challenges when it comes to defending his theory. The first is to show that it is more successful than competing theories in accommodating our considered intuitions about parenthood. The second is to show that the theory is unified and coherent, despite drawing on a number of different values and principles. In my opinion, Millum meets both of these challenges. In what follows I highlight some of the central aspects of the theory, and indicate the advantages it has over existing theories.
…
“Millum’s theory gives what I consider to be plausible answers to a number of controversial cases. [I]t explains why accidental fathers can have parental responsibilities without having parental rights. It also explains why absentee biological fathers who were deliberately excluded from a child’s life, but then shows up at some later point, can have an interest in forming a relationship with a child and yet not have a moral right or claim to do so. The theory has some interesting implications for non-traditional families created through artificial reproductive technology, divorce and remarriage, and same-sex partnerships, most notably that a child can have more than two real parents. Although his theory doesn’t automatically resolve all disagreements, it offers a framework in terms of which such issues can be discussed. Consider, for example, the case of surrogate motherhood. Surrogacy is often viewed as a form of adoption: the surrogate is the real mother, but promises to transfer parental rights and duties to the intending parents. Against this, others argue that the intended parents are the real parents from the outset, and that the surrogate is merely gestating the baby on their behalf. As we’ve seen, Millum argues that some parental work can be contracted out, which is what happens in the case of hired caregivers. But whether the same is true in surrogacy depends on whether gestational care is the kind of parental work that can be contracted out. Millum points out that there is a limit to the amount and type of parenting work that others can do on behalf of parents. Generic work, such as clothing, feeding, and comforting a child, can be fulfilled by proxy parents, but responsibilities to provide filial goods, such as an intimate relationship or loving bond with parents, cannot be contracted out. In the case of surrogate motherhood, then, the question is whether gestational care is best described as providing generic goods or filial goods. Research on maternal-fetal attachment and, in particular, surrogates’ own perspectives on their pregnancies, appear to support Millum’s view: gestational care can be contracted out, which means that one can give birth to a baby without becoming a parent.”
First, many thanks to Liezl van Zyl for reading the book so carefully and writing such a sympathetic review! The review already did an excellent job of outlining the main themes of the book. I’ll just very briefly summarize what I think are the key features of my theory and highlight a couple of implications that I think are interesting or controversial.
The idea of the book is to present a theory of moral parenthood, which at root I take to comprise an account of how we acquire parental rights and what they are and an account of how we acquire parental responsibilities and what they are. I develop an investment theory of the acquisition of parental rights, according to which we acquire parental rights as the result of putting in parental work, where that is morally deserving work aimed at the flourishing of the child. Thus, who has the right to parent a particular child depends on who deserves it in virtue of the parenting they’ve already done. The investment theory is the entitlement theory of parental rights—it tells us how people get to have parental rights. We also need to determine what the contents of parental rights are; that is, which parental rights exist. I argue that parents have fundamental rights to realize the goods that they have made possible through their work in raising the child. These include rights against other parties not to interfere with (good) parenting, which allow parents to obtain goods like the intimate relationships that Brighouse and Swift describe. Crucially, these fundamental parental rights are constrained by parental responsibilities and the existing claims of other parties (e.g. the rights of children). Other parental rights are derivative, mostly from the interests of the child.
Turning from parental rights to parental responsibilities, I think that parental responsibilities do not have the same foundation as parental rights. I argue for what I call a conventional acts account of how we acquire parental responsibilities. Basically, we acquire them through the performance of speech acts that signify taking on responsibilities. One of these speech acts in modern Anglo-American societies is having (heterosexual, voluntary) sex, since having sex is widely understood in our society as leading to parental responsibilities when it results in a child. What is the content of parental responsibilities? I argue that parental responsibilities are a subset of what is owed to children as a matter of justice. Parents, as it were, take on responsibilities on behalf of society. Depending on how a society is set up this can vary—there are different ways to divide up responsibilities for providing children with what they are owed. But parents must at least have duties to provide filial goods—those goods that can only be provided by constant caregivers.
The views I argue for, then, are intended to give an account of the foundations of moral parenthood. They also have some specific implications for questions about the ethics of parenthood. Here are three to start with.
1. The investment theory implies that genetic parenthood does not in itself have any special weight when it comes to claims to parent a child. Perhaps more controversially, as van Zyl notes, it implies that if a child’s genetic father has not put in parental work, then he does not have a claim to parent the child. So, for example, suppose a man did not realize that he had “fathered” a child until the child was a toddler. Provided he was not doing any parental work up until that point, he would have no claim now to start parenting the child.
2. The view that the content of fundamental parental rights is constrained by the existing claims of others means that parents do not have fundamental rights to favor their own children in ways that they would not be morally permitted to favor others. That is, parental rights do not include what I dub nepotistic liberties. This is contrary to what I take to be the widespread view that I am permitted to favor family members in ways that I’m not permitted to favor others.
3. The conventional acts theory of how parental responsibilities are acquired implies, as I mentioned, that accidental fathers—men who got their partners pregnant despite taking every precaution against doing so—still acquire parental responsibilities. On the other hand, contrary to what some theorists have argued, it implies that gamete donors have no parental responsibilities to their progeny. The fact that, for example, donating sperm is not generally understood to entail taking on parental responsibilities means that it is not a speech act that has this effect.
Thanks again to Professor van Zyl for her review! I’m looking forward to the discussion.
Thanks to David and to PEA Soup for hosting this discussion, to Liezl for her thoughtful review of Joe’s book, and to Joe for his pathbreaking work on this important set of issues!
Joe, I have two questions for you, one about your account of parental duties and another about your account of parental rights.
First, how does the conventional-acts account of parental duties deal with the common objections to other forms of moral conventionalism? One difficulty with conventionalism is that it ascribes moral authority to seemingly capricious and arbitrary social norms. For example, conventional norms might not ascribe parental duties to men who donate their sperm to single women seeking to procreate under the auspices of a fertility clinic, but yet ascribe those duties to men who donate sperm non-coitally outside of a clinic, or coitally with a well-documented contract establishing that the sperm donor accepts no parental responsibilities. There are surely differences between these sorts of cases, but conventionalism commits us to saying that those differences make a moral difference.
Another difficulty with conventionalism that you seem to have engaged with in your book is that it ascribes moral authority to seemingly unjust and oppressive social norms. For example, in various times and places in history, conventions have held that married men have no parental duties for children conceived out of wedlock. Those social conventions are certainly unjust—it is unfair to those vulnerable children to be deprived the resources given to other children, and unfair to ascribe burdensome parental duties to single men who conceive out of wedlock but not married men. But it also just seems false that whether a man is single or married makes any difference to whether he has parental duties to the children he has helped create. It is one thing to criticize the justice of a convention, but another thing to criticize its view of the moral facts, and there seem to be ample resources for making the second kind of criticism of historical (and contemporary?) conventions for assigning parental duties.
Second, how does the investment theory of the acquisition of parental rights yield the conclusion that procreators are entitled to their own biological children, rather than some child or other, or some other kind of fair compensation? Labor-based proprietarian theories of parental rights must explain why putting one’s labor into an object gives one an entitlement to that object, rather than to fair compensation for one’s labor given the object’s value. For example, if I build a house and the government needs to demolish it for legitimate reasons (these might be justice or promotion of the collective good), I am certainly entitled for fair compensation given the labor I put into it, but it is not clear that I am entitled to a house, let alone to that house. (Anca Gheaus has raised similar concerns in her 2012 paper, “The Right to Parent One’s Biological Baby.”) How does the investment theory deal with this problem, and could you say more about the ways in which it differs from labor-based proprietarian accounts of parental rights acquisition?
Hi Jake,
Thanks for joining in and for raising these important issues!
Regarding the role of social convention, we agree that it is possible to criticize the conventions regarding which acts entail taking on parental responsibilities. The locus of our disagreement, I think, is whether alternative ways to set up the conventions really alter who acquires parental responsibilities. Can it really be the case that (e.g.) in a society where a man is understood to take on parental responsibilities through sex only if married he really does only take them on if married? I would say, provided that everyone involved acts voluntarily and the significance of marital status is understood by all, then yes, in this context marriage is necessary for the speech act of taking on parental responsibilities to be successful. This might seem counter-intuitive, but I think that’s because we (correctly) regard the conventions as unjust and because we are used to our own conventions.
Note, further, that the man in this case might have responsibilities as a result of being a procreator, that is, he might be responsible for the burdens placed on others as a result of knowingly creating a child. What I deny is just that he would have parental responsibilities.
Regarding your second question, I think labor based theories are very well-placed to say why the people who put in parental work are entitled to parental rights over the child that they have parented, not some other compensation for the work they have done. First, this is intuitive for many cases in which we create something, e.g. if I make a sculpture out of found objects, then you are not permitted to take it from me by paying me the value of my labor – I have to agree to the transaction. Exceptions to that rule, like eminent domain, require special justification and many people find them objectionable. Second, there are good reasons why when you labor on something you should get the rights to that thing, not something else. It allows us to carry out projects that we regard as valuable. If I make the sculpture, it is likely because I regard doing so to be valuable (given my artistic skills, I’m definitely not in it for the money). If others could just interrupt our projects, then a lot of meaningful human activity would be much harder. Likewise, with parenting. Third, the people who have carried out parental work have made certain goods – the goods of parenting – possible. Those goods, as it were, are now up for grabs. Other goods, ones that already exist, are ones to which other people already have claims. Given that the parents deserve compensation for their work, what can they claim as their right? Clearly, the goods of parenting, which are the goods that others do not have prior claims to.
Joe, your responses are very helpful, thank you!
I appreciate your point that it can be difficult to distinguish objections about moral matters of fact from objections related to justice. It also would be unfair to expect you to defend moral conventionalism broadly construed, when your task in this book is to defend its plausibility relative to other views about the acquisition of parental responsibilities.
I am wondering, though, the extent to which you think people can take steps to prevent their actions from falling under the authority of conventions ascribing parental duties. For instance, it might be socially normative for people to acknowledge their co-workers’ birthdays, but one can permissibly avoid being subject to this convention by providing adequate warning that one will never acknowledge any birthdays or participate in office birthday celebrations. Could I similarly avoid acquiring parental responsibilities from procreating by warning my sex partners, my government, and my community in clear and unwavering terms that my engaging in sex is not to be understood as falling under the conventions ascribing parental duties? If not, why not?
Turning to the investment theory, I suppose I would object to your claim that it is intuitive that our labor gives us natural moral claims to objects rather than to fair value or something like that. Even in the case of the sculpture, it seems equally intuitive to me that you are entitled to fair value for your efforts, which will usually entail that you are entitled to keep the chair unless you decide to sell it. If, however, your community assesses taxes based on some share of the value of your total holdings, and the chair is necessary to cover the unpaid remainder of your share at the end of the assessment period, then you are not entitled to it, precisely because you were only ever entitled to some fair share of its value.
Intuitive quibbles aside, I think part of our disagreement here is that I favor a more social account of property or investment claims, according to which whether you are entitled to X or Y is determined by local social rules, though those rules might be more or less just. It is interesting that you adopt a conventionalist account of parental duty acquisition, but a naturalist account of parental right acquisition. Asymmetries in philosophy are interesting, so maybe you could say something about why we should expect different kinds of explanation of how we acquire parental duties and rights?
Finally, I see your point in favor of a general rule favoring allowing people to keep that in which they have invested, but the distributive objection raised by Brighouse and Swift, Gheaus, and others is that children, unlike other objects of investment, have their own interests which would sometimes be better served by redistributing parental claims. Could you say a bit about when and why you think desert claims arising from parental investment should outweigh children’s claims to have their best interests advanced?
(The above response was mine, in case that was unclear!)
Hi Jake, thanks for these follow-up points. Apologies in advance for my selective replies…
The issue of when people can avoid responsibilities is key. After all, it would be problematic if individuals could just say “I’m not taking on responsibilities by doing this!” in order to get out of their acts having moral significance. In a case in which someone was mimicking a speech act like promising, this might be possible. For example, I might be able to avoid promissory duties resulting from pretend-promising by indicating to the apparent promisee that I didn’t mean what I was about to say. But in other cases, I don’t think this is possible. For example, if you want to avoid getting married, you should avoid going through the marriage ceremony with an ordained minister—simply saying earlier in the day that you won’t mean it when you say “I do” is not sufficient to render the speech act unsuccessful. I think taking on parental responsibilities is more like the latter than the former. One possible explanation is that the promise is made between two people and it is straightforward for them to re-negotiate the terms of the promise, including what will signify promising. The marriage, like taking on parental responsibilities, does not just involve two parties – it is in fact a public act. I think of it as analogous to the would-be parents promising society that they will parent a child. They don’t get to re-negotiate how that speech act works for an individual case (rather, the social conventions would have to be changed in order to change the meaning of the speech act).
On the investment theory, you say that the sculptor would only have entitlement to keep his works provided that (e.g.) he didn’t owe taxes levied on his holdings. That seems fine to me. But it doesn’t speak against the view that the sculptor has a special claim to own his sculpture, rather than something equal to it in value; it just shows that his claim to property can be overridden by the justice-based claims of others. Moreover, I assume, in this case, that it would be acceptable for him to pay his taxes in cash, rather than in sculptures. So, it’s not that he lacks a claim to own the things he makes. He just also ought to pay his taxes.
Finally, on the interests of children and why they shouldn’t be redistributed to the best possible parents. First, fundamental parental rights are grounded in the interests of the would-be parents. This is true on my view, as it is for Brighouse & Swift, and for Ghaeus. If children had the right to optimal parenting then, on my view, that right would trump parental rights. But I assume, like most other writers, that children do not have a right to the best possible parents, but only to sufficiently good parenting. Provided that parents are sufficiently good, the child’s rights are satisfied. The redistribution problem, then, is not about why children don’t have a right to be redistributed, but why any particular person gets parental rights with respect to any particular child. That problem is dealt with (I hope) by the investment theory. Note that Brighouse & Swift don’t have an answer to this problem (as Ghaeus also notes): they have an account of why people have a rights-grounding interest in being parents, but not an account of why specific people have rights over specific children.
I’m sorry to be so late in joining this discussion.
The main reason I found the conventionalist view of parental responsibilities plausible is that naturalistic or essentialist accounts (based on, for example, genetics, gestation, intention, the best interests of the child) all have obvious shortcomings. At the end of the day, someone has to take responsibility for the child (or rather, society has to assign responsibility to someone), and so the best reason for assigning it to a particular person (or persons) might well be something like: ‘This just is the convention in our society: if you voluntarily have heterosexual sex with someone, and it results in a child, then you’re responsible for parenting that child’. And there are good moral and practical reasons to support the view that the convention includes a stipulation that one cannot opt out of it, for example by entering an agreement before having sex.
A few questions and concerns I have with this view:
1. Is this in fact the convention we follow in our society? As far as I know, where the child is the result of involuntary sex (rape), we still assign responsibility to the genetic parent. The most straightforward example here is where the man is the victim of a rape that results in a child. I suspect he will be assigned parental responsibility regardless, so the convention seems to be something more like: genetic parents are responsible for the child, except in the case of gamete donation that occurred in a clinical setting.
2. I take the point that we can give moral reasons against or in support of whatever convention we have, for instance by appealing to fairness, the interests of children, of women and men, etc. To take Jake’s example of the society where married men are not held responsible for a child they father out of wedlock, this is certainly an unjust convention, and its injustice is a reason for changing the convention, but according to the conventionalist account, a particular married man is not responsible for the child he fathered out of wedlock. So my concern is whether the convention has any independent moral force: can it serve as a basis for moral parentage? I take it that part of the point of providing a moral theory of parenthood is to support a particular legal theory, and so it seems odd if the moral theory itself is in need of moral support.
3. Joe has been careful to distinguish between moral and legal parenthood early on in the book, but in thinking about the conventionalist account of parental responsibilities I find it difficult to keep these separate. For instance, when we want to find out what our society’s conventions regarding parental responsibility are, we look at what the laws say. Again, we might criticise these laws on moral grounds, but they still apply. And someone who wonders whether a particular law applies to them, given that it is unfair or immoral in some way, can be told: ‘This just is the law in our society. If you do X, and it results in a child, then you are legally responsible for parenting that child’.
Hi Liezl,
Thanks for these interesting points. I’d like to add something regarding how I see the conventions relating to individuals taking on moral responsibilities. The conventions tell us which acts constitute the speech act of agreeing to be a parent. They also tell us something about what is being agreed to, i.e. the exact content of parental responsibilities within a particular society. But I don’t think that the existence of a societal convention about who has responsibility is sufficient for someone to actually have a moral responsibility. There are other conditions for responsibilities to be acquired. For example, I think that the speech act must be performed voluntarily. This is true for promising, too: linguistic and social convention dictate exactly what must be said in order to acquire a responsibility by promising (e.g. “I promise that I will X”) but the promise still has to be voluntarily made in order to be binding.
This point is relevant to your first question. I agree that social conventions sometimes assign parental responsibilities to someone who was the victim of rape. But since the act was not voluntary, I think that this is not a way that someone can acquire responsibilities. And to your third point, I don’t think the law (or social convention) directly dictates what someone’s moral responsibilities are. But, for those responsibilities that are voluntarily taken on, social convention can dictate which acts constitute taking on those responsibilities.
Hope this makes sense and doesn’t seem circular!
I’m glad to see these issues discussed here, and I’ve enjoyed reading the comments so far! Sorry I’m late in joining.
I had two questions related to some issues that have already come up, both focusing on the account of parental responsibilities. One potential problem for a conventional acts account, as Jake noted above, is that someone could just try to avoid the convention (e.g., say “I know that the convention says that doing such-and-such amounts to promising or consenting, etc., but I do not promise or consent”). If I understand Joe’s response correctly, he appealed to the marriage example and suggested that it is as though the child’s parents make a promise *to society* to care for the child. But I wonder whether there are difficulties in viewing the parents as, in effect, making a promise to people other than the child (or each other). Part of the concern is that to the extent that society has a stake in the matter, it is probably only concerned that someone parent the child; it may not much care who parents the child. Suppose, for instance, that an “accidental father” leaves his biological child to be raised entirely by its mother, refrains from paying child support, and so forth. But suppose that the mother is an exceptionally good and relatively affluent parent. It might be that the rest of society does not particularly care that the father has abandoned his family, since (we can suppose) the child will still receive an adequate, or even exceptionally good, upbringing. And many people would probably think that the father acts wrongly because he wrongs his child, not because he breaks a promise to the rest of society.
A second issue is whether an act that is performed under an unjust convention really has the significance ascribed to it by the convention. Suppose that some society has a convention according to which women who go out in public wearing revealing clothing consent to sex with any men they happen to meet on the street. Even if a woman voluntarily chooses to wear revealing clothing (because it’s hot out, say) while fully aware of this convention, she hasn’t consented. This sort of example can probably also be varied so it better matches the marriage example from above. Even if social conventions require women in heterosexual marriages to do all the dishes, it doesn’t seem to me that a woman’s voluntary choice to get married gives her an obligation to do more than her fair share of dishes. So I guess I’m just wondering how the conventional-acts account would deal with these cases, in light of the way it deals with accidental fathers.
Thanks again for the great comments so far!
Hi Jason, thanks for joining us! Great questions. Let me hazard a response.
1. I agree that society’s stake in the matter is primarily that children receive adequate parenting from someone (I argue in the book that children are owed parenting as a matter of justice). The conventions are set up so that certain voluntary acts constitute taking on parental responsibilities. Typically, they result in two people taking on parental responsibilities. Now take the case you describe: the father performed acts that signify taking on parental responsibilities, but the mother would in any case give the child all the parenting to which the child had a claim. Here, it seems clear to me that the father wrongs the mother, by leaving her with more than 50% of the caretaking. But, if it’s really the case that the child is not owed any more parenting than the mother will provide, the father may not be wronging the child. Seems odd, I agree! But then I think we’re implicitly imagining that there are goods that the child is owed that are best provided by the father. (Suppose the mother has a partner who is willing to co-parent the child instead of the biological father; do we now think that the biological father is wronging the child?)
2. I agree with you on these examples. In the first, there is an issue of cost—if it is too demanding to avoid some act, then that act will not signify taking on a responsibility, since it will not be sufficiently voluntary. But there is also the fact that consent to sex is given to the individual with whom one has sex, not to society. So, it would be possible for a woman to state to the person who took her to have tokened consent that she was not in fact doing so. In any case, not a great convention to have—we don’t want people to constantly have to explain that they don’t mean what others take them to mean! My response to cases like your second one also relies on the possibility of renegotiating duties between parties. The woman in this case could decline to get married except on condition that the man do his share of the dishes (or whatever joint tasks need to be done). Of course, given the difficulties in practice of defying conventions, this is not a great convention to have, but I don’t see any conceptual problem with the possibility of renegotiating divisions of labor.
Interested to know if these responses help or not.
(I should note, also, that part of the reason for my adopting the conventional acts view is that the alternatives all seemed inadequate. So, I agree that there are some counter-intuitive implications of the view, but think it nevertheless does better than its rivals and so I’m happy to bite some bullets and follow the implications of the theory where it takes me.)
Thanks for the reply, Joe. Your comments are helpful. (I should add: I’ve read some of your papers on these issues, but haven’t yet read the book–though I will now!)
Your response to the concern about “accidental fathers” seems plausible. Part of the reason I focused on that case is because it seems to separate your position from Elizabeth Brake’s, which seems to let accidental fathers off the hook. I think many people would take that to be a problem with Brake’s view. It seems that the conventional-acts account may also *sometimes* let accidental fathers off the hook, as in the example you gave. Though I imagine some people will take that to be a problem for the view, I’m not sure it is. (Alternatively, it might be that the accidental father in the case you describe does incur parental obligations and then discharges them by ensuring that the child’s mother, or the mother and her new partner, provide adequate care.)
I think we agree that the alternatives to the conventional-acts account face some pretty serious difficulties. I do still have a lingering concern, again probably related to Jake’s above, about how a convention gains authority over a person, and in particular over a person who opposes the convention. I guess the concern is that it seems society can’t just say “Anyone who voluntarily does X will be understood as having taken on parental responsibilities,” if doing X does not generate “natural” (as opposed to artificial) duties and there are a lot of people who would do X despite not wanting to take on the relevant parental responsibilities. I think, though this is really only a hunch, that my concern might be analogous to a concern commonly raised about consent accounts of political obligation—e.g., that the government can’t just say that anyone who does such-and-such will be understood to promise or agree to obey the law, unless it is *already* thought to have authority over these people.
Thanks again for the comments. Looking forward to reading the book!
Hi Jason, I didn’t mean to let accidental fathers off the hook there! I just think that in the case you describe they do not wrong their child (because the responsibilities are all fulfilled by someone else). They do, however, breach a commitment that they made to society and they do wrong the mother.
And I’m in sympathy with the concern you raise. Perhaps it can be ameliorated somewhat by emphasizing that the significance of an act is not typically decided by a government. For example, the government doesn’t get to say what words mean, it doesn’t get to say what acts constitute an apology or an insult, and so on. The significance of many speech acts is established much more organically than that. Of course, governments and other bodies can affect the social norms that determine what constitutes a speech act. It’s just that they don’t get to dictate what they are. Likewise, I believe, with taking on responsibilities; for example, whether shaking hands entails taking on moral responsibility to do what was just negotiated depends on social conventions, not on what the government says. However, those social conventions are liable to be heavily influenced by what the law says (and the law often tracks those social conventions).