I previously discussed my opinions of some papers working towards answering this question; in this post, I intend to discuss a theory I have been developing, and discuss the very serious problems with its current state.
I start with the assumption that it is wrong for a child to be brought into life if they can be expected to have a life not worth living. Furthermore, this is not merely wrong but it is a violation of the child's rights. I integrate this assumption into a kind-of-Nozickian position and emerge that children should be treated in a way they would consent to in the hypothetical situation where they - or a rational agent representing them - and their parents signed a contract regarding how the child should be brought up.
As it stands at this, without working out what it implies, there are already several serious issues with the theory.
Some rights theorists have argued that having rights requires the ability to enforce them. I don't think I would necessarily go that far, but it is certainly fair to say that rights suffer in the absence of an enforcement mechanism. And, practically speaking, it is difficult to see how this theory would be enforced.
The starting point would be that, if a child's upbringing fails to meet whatever is decided to be just, then the child would have a right to sue their parents. There are problems with this, at least one of which I see no way of resolving.
Suppose the child is so badly mistreated that they die before reaching the age of emancipation. Then, presumably the right to sue the parents would return to the commons, and could be homesteaded by someone who prosecuted the parents. (This would not be much comfort to the child, but since no theory can raise the dead this is hardly a problem unique to my theory). But what if the homesteader of this right is a confederate of the parents, who does a deliberately bad job of prosecution? Even supposing this is solved, then suppose there is a competitive market of lawyers who will take up such cases. Suppose also that there is an inverse relationship T (between the time spent accumulating evidence before taking a case to court) and P (the probability of a successful prosecution, and hence a profit). If the right to take the case to court can only be homesteaded once, then clearly the market equilibrium is for cases to be homesteaded as soon as they appear and have a positive P.
This might be resolved by having an organisation which is automatically assumed to gain the right to prosecute a case, which might either prosecute cases itself or sell the rights on to lawyers for a fee. This might fund (for example) an orphanage. Such a system would be far from perfect, but does not seem completely unworkable.
The problem is actually greater when the child is still alive. Most people would be unwilling to sue their parents; even if the right to sue the parents were somehow homesteaded by one of our crusading lawyers, the case would be unlikely to succeed without the co-operation of the key witness. So parents would be able to get away with many abuses.
What does it even mean to "hypothetically consent"?
It is in many ways strange that one can be morally bound by a promise that one has not made. How does the fact that in a particular hypothetical scenario I would have agreed to take on a certain obligation bind me to it in the real world where I have not?
The best answer, so far as I can tell, is that it doesn't; rather, it is in one's best interests to act as though it is. Suppose that I would like to see a certain band live in concert, but am unwilling to pay the £50 it costs to buy a ticket - the most I would be willing to pay is £40. In order to prevent ticket touting, all tickets to see the band have the name of their owner printed on them and require proof of ID. An acquaintance of mine, B, has an opportunity to buy a ticket to see the band for £20, and so buys the ticket in my name. (B has no interest in seeing the band herself). While I would receive the ticket from B whether or not I paid her for it, if I wish B and other people I know to do similar things for me in the future then I would be well-advised to pay B at least the £20 it cost her to buy the ticket.
But this fails to solve the issue of exactly how much I should pay - it should be at least £20 and no more than £40, but could be anywhere in-between. In the scenario above we might well say £20 and call it quits (or alternatively £20 and either a box of chocolates, a bouquet of flowers or a bottle of wine) but suppose that rather than as a friendly gesture, B bought the ticket because this is how she made her living. Quite clearly, then, I would pay more than £20.
Taking this to the case of childrens' rights, it seems that children should have positive rights going beyond "a life worth living"; however, we have no idea how extensive these rights should be, except that they should not cause it to cease to be worthwhile to have children.
Parental influence on the child's values
As good liberal neutrals, we should not wish to assume that there is a particular, uniquely and universally justified measure for how well a child was raised. Rather, we should allow a different metric in every case, dependent largely upon what the child ends up developing as their conception of the good.
The problem here is that parents have a fantastic opportunity to essentially brainwash their children. A child could be brought up in a cult, and so long as the child continues to believe that the cult is virtuous and that being part of it is beneficial, it is hard to see how we can object.
A basic theory of children's rights based upon hypothetical consent runs into several problems, which all tend in the direction of allowing parents far too much license in the way they raise their children.