Skip to content

The Rights of Children

The theory of the rights of children is far too often overlooked or infected with irrational cultural norms. This is true even for the comparatively more thorough libertarian theory of law. A rational theory of the rights of children must be elucidated for a given legal theory to be complete.

21 minutes

👇
Lesson 5

From its inception, libertarian theory has had an enormous problem standing before it: how to reconcile the existence of developing self-owners within the framework of property rights and non-aggression. It is not at all obvious how the rights of children, or lack thereof, are to be derived from the aforementioned principles. It is all too easy for subjective cultural values concerning children to sneak their way into an otherwise sound argument. In order to develop a rational theory on this topic, these seemingly self-evident attitudes must be identified and dismissed. Similarly, it is imperative to reject the “wisdom of repugnance” which would dismiss a rational theory solely on the grounds that it produces conclusions abhorrent to the popular mores of a given society.

—Ian Hersum1

This document is part of a course which you can find here. I provide the textual version for free to the public to make people aware of the facts contained within and to (hopefully) convince people to purchase the video course. This course is intended primarily as a video experience, and that is where my focus will be in terms of quality, so I do encourage you to check it out if this content is at all interesting to you. The course is sold on a pay-what-you-want model and is copyleft, meaning other people have the right to re-publish it at will. I simply encourage anyone who appreciates the work put into this course to purchase it such that I may make more in the future, and if you do choose to republish any section, please credit me and extend to your audience my request to purchase.

The Groundwork

To develop a theory of the rights of children, we must understand what the nature of a child is. First we recognise that it is not physical, but mental development which defines childhood—parapalegics such as Stephen Hawking are incapable of commanding their body to do certain tasks but they may still be adults. Though these disabled individuals lack certain abilities seen in most humans, they do not lack the characteristic mark of action, they merely lack the ability to wield many means which others take for granted. So it is psychological as opposed to physiological immaturity which is the defining mark of childhood.

From this we can deduce a fact about the nature of childhood, namely that it is not a switch which once flipped cannot be flipped back; it is certainly possible for a given person to move in and out of psychological maturity throughout the course of their life. Consider a sleeping man, certainly this individual is–perhaps temporarily–psychologically immature. That is to say, this individual is not capable of negotiating for his own care, and instead requires others to do so for him. This is especially relevant in the hypothetical scenario of an unconscious man lying in the snow and freezing to death. Often libertarians will clumsily tack on a theory of implicit consent or implicit contract to deal with such scenarios—the former relies on an arbitrary notion of what a “reasonable” person would consent to and is thus not rational, and for the latter it is not clear what titles are being transferred. For the theory presented here2 this unconscious person being taken to the hospital by a paramedic is analogised to a mother carrying her toddler.

Note that this guardianship role taken up by the paramedic and the mother respectively is scarce and therefore it must be held singly by the homesteader as has been shown in this course. This is because there can be conflicts over the specific minutia of how the guardianship is to be performed. A direct implication of this is that counter to the common view that fathers should have just as great a say over the child as the mother, naturally the mother must be the homesteader of the guardianship as she has greatest proximity—from the moment the baby comes into existence it is being cared for by its mother, this is not true of the father.

To capture the nature of a child as a psychologically immature human, we can define childhood as the state of being incapable of expressing one’s own will and the guardian is the man who takes it upon himself to preserve the child until such a time that they gain the ability to express their will. Ian Hersum analogises this to an encrypted last testament:3

[…] imagine the scenario of an encrypted last testament (being consequentially analogous to one’s premature will), which an interested party agrees to decrypt over time. What is to be done with the estate during that time? It must doubtless not be damaged or consumed until such a time as the will has been entirely decrypted, with its voluntary manager responsible for preserving it in the interim. Should it be damaged or consumed during that period, either by the manager or by a third party, whoever has done such damage or consumption would be held liable, and that person would be disqualified from managing the property in the future, provided that someone else is willing to assume that role. As such, anyone who harms a child should be held liable for the damage done and be forbidden from being the guardian of that child in the future, provided that someone else is willing to assume that role. As bits and pieces of the will are decrypted, the estate manager would be obligated to follow any instructions which are capable of being understood with the information available at the time. As such, as a child develops, his guardian is obligated to relinquish authority over to the child in domains of behavior which the child can express his informed will on. In a contention between a child and his guardian over such authority, a court can listen to the testimony of the child in order to determine if he truly understands that which he is saying, or if he is merely blathering on about a decision which he lacks the comprehension necessary to make.

This analogy highlights some key observations, first the reason why it is just to subject a child to life-saving surgery is that this action is preserving their natural state until such a time that they are able to express what they want done to their body—this applies also to any surgeries which do not necessarily save the childs life, but take the child closer to that natural state. This is entirely different to a cosmetic surgery, which alters the childs natural state. Consider the example of circumcision, you do not know whether the boy wants the circumcision performed, so assuming it would not save his life, you are criminally damaging him. That is to say, you cannot assume any facts about what the child would will on any subject to do with their natural property (aka their body), therefore the only actions you can perform to that property is to preserve it until its owner is able to state what he wants done with it. To be clear; the guardian of a child is not the owner of that child, rather he is “the owner of the exclusive right to raise that child.”4

So the reason why a paramedic is allowed to resuscitate an unconscious man is that this is preserving his body until such a time that the man may state what he wants done with it. A caveat with the unconscious man is that he was at one point presumably an adult able to state his will, and if he had stated that he does not wish to be resuscitated, his will on that matter is known, and thus it would be criminal to go against such an order. It is only on matters where the child’s will is unknown that you may take the preserving action.

Furthermore, as the guardian is not the owner of the child itself, but rather the owner of the right to protect that child, any abuse performed by the guardian unto the child implies an abandonment of that right, implying that the guardian must notify interested parties that the child is available for adoption. Recall earlier that it was concluded that creating a donut-shaped homestead around the property of another was an act of forestalling, where forestalling was defined as excluding others from that which is not your property. Here, the abandoning guardian would be acting as if he was the guardian if he was preventing others from taking up that mantle, this is because he is excluding others from homesteading the right which he himself rejects. So by not notifying others that the baby is free to adopt, the abandoning-guardian has not truly abandoned it, rather he is placing an information barrier between the baby and potential adopters, which is excluding those adopters from what the abandoning-guardian does not have the right to exclude them from. Moreover, this requirement to notify potential adopters does not constitute a positive obligation, it is rather the negative obligation to not forestall.

Hersum elaborates on what counts as harm:

Since a child’s preferences cannot be known, the proper method of raising him is impossible to determine, so his guardian is largely free to engage in any actions that he wishes to in relation to the child, as long as he does not deprive him of his innate function or form. While refusing to feed (or care for in other ways) a child cannot be understood as an act of harm, since the resources required for such care belong to the guardian and not the child, it still constitutes an abandonment of guardianship rights, but cannot carry a penalty other than one for forestalling. Rather, harm in this context can only be rendered by an active (rather than passive) behavior on the part of an adult against a child. This rules out any form of neglect.

There must be a direct causal link between the action and the effects suffered for it to be considered harmful. For instance, saving photographs of the child in amusing outfits has no plausible benefit and may bring about a negative response from him when he has grown up, but this cannot be considered damaging, as no act within the photoshoot itself deprived him of anything, and any potential maleffects are suffered entirely in retrospect, so they are not relevant to the act itself. In contrast, verifiable psychological damage suffered by a child, which is directly attributable to an act of torment inflicted on him by an adult, deprives him of his natural mental functioning which is innately his. This also applies to physiological damage, of which verification and attribution is considerably easier. Any scarring, maiming, mutilation, or other disfigurement, which deprives a child of his innate body, and was suffered as a result of actions taken against him by an adult, likewise qualifies as damage.

The exception to this would be surgical procedures (or, conceivably, other acts) that treat conditions which pose a greater threat to a child’s innate health than the damage associated with the procedures themselves. A life-threatening cancer, for example, warrants treatments of increasing severity up to the point of death. In contrast, genetic abnormalities (or, in the case of certain ritual practices such as circumcision, normalities) that benefit only the outward appearance of a child may not be corrected via damaging surgery. Similarly, operations which seek to improve the functioning of a child beyond his natural capacity by replacing parts of his body may not be performed, unless such modification is necessary to treat a threatening health condition (such as the amputation of a severely damaged limb). As the preference of a child for these alterations cannot be known, the preservation of his natural form is required by default, giving way only to prevent further damage from occurring.

Under this rational theory, it is evident that any given age of consent or majority is necessarily arbitrary, whether a person is capable of expressing his will over a given domain is an entirely individual, not a collective issue. You should expect a person with severe mental impairments to become an adult only long after his far more able peers. An arbitrarily chosen universal age would imply the potential to violate both the rights of those who develop faster and those who develop slower than the chosen figure. That is to say, age is merely a proxy for psychological development, not its logical root. Furthermore, this theory differs from common attitudes in another way, in that it does not per se class corporal punishment as criminal—only if it is severe enough to inflict lasting physical or mental damage is it prohibited.

Abortion

Now that the groundwork for a rational theory of the rights of children has been elucidated, it can be applied to the particularly contentious case of abortion. For the correct stance on this matter, we do not need to rely on vague notions of physical development or spirituality, this must be an issue which like all other issues in law is solved by careful reference to property rights. The question is this: what is to be done when a mother is carrying a baby in her womb and she does not want that baby there anymore? Or more precisely, what is she allowed to do and what is she not allowed to do with respect to that baby.

First, it must be noted that the baby cannot be treated as if he was a parasite or tumour, the fact that he is indeed composed of a clump of cells has no bearing on the issue of rights. To be sure, every human being is composed of a clump of cells, this is irrelevant to ethics. It is clear also that prior to conception, there was no baby to speak of, and thus no body for that baby to own, similarly when the baby is a full adult capable of action, he does have a body for himself to own. The question is, at what point between these two positions is the baby relevant in discussions of rights? The answer seems clear; the baby is relevant when the baby exists, that is, at the point of conception. Prior to conception, there was in existence the matter required to make a baby, and after that matter has been properly assembled it will continuously grow until death. The Randian notion of the baby-in-a-womb being a mere potentiality is misplaced, it is the matter prior to conception that is the potential human, and once that matter is sufficiently arranged it becomes a baby human. Moreover, to pick any specific point along the continuum between conception and death would be an arbitrary choice. Consider birth; being born does not change the metaphysical characteristics of a person, all that happens is that the person moves from inside of a womb to outside of that womb. Block and Whitehead highlight this with an analogy:5

Compare two entities […]: one, the new-born babe, still attached to its umbilical chord, a few seconds old. The other, its sibling, is still in the womb but due out in a matter of minutes. No two entities could be more alike, biologically, spiritually, or in any other way. Yet, in the “pro-choice” philosophy, it would be murder to kill the one and a matter of complete judicial irrelevance to kill the other. Surely, this is a travesty not only of justice but also of common sense.

It is at conception and conception only that there is a principled difference between before and after. Prior to conception you do not have an entity which is not capable of expressing its will but might in the future be able to do so. After conception, you do have such an entity. Therefore, at the moment of conception, you do not just have mere matter, you have a child, and thus the above analysis of the rights of children applies from the moment of conception until adulthood.

From this, we can derive some basic facts; first the mother is not allowed to directly harm the baby, she can only put the baby up for adoption so to speak. Therefore she is allowed to evict the baby, that is to remove the baby from her womb, but she cannot do so in an arbitrary manner. She must notify potential adopters that she does not wish to care for the baby and she must allow them to potentially fund the safe removal into their custody. This is not to say that she must wait until such a time that the baby is viable, that is she does not have to wait until the baby is actually capable of surviving outside of her womb. It might be the case that only his mother is capable of keeping him alive, but that does not confer a positive obligation on the mother to actually do so—but the mother cannot prevent other people from keeping the baby alive.

This means that the two common stances–pro-life and pro-abortion–both have fundamental flaws. The pro-life individuals would force the mother to aid the baby against her will, thus violating her rights. The pro-abortion people would allow the mother to hold back those who want to help the baby and to kill the baby at her will, thus violating the rights of those people. It is only this evictionist stance which is consistent with true law.

Artificial Intelligence and Self-Ownership

This theory of the rights of children can be applied to determine the rights of developing artificial intelligence systems, and provides important insights on the nature of self-ownership. Consider a supercomputer which is running an AI program so advanced that it “wakes up” and develops the capability of engaging in argumentation, this AI must therefore have all rights implied by the NAP. Included in those rights is the right of self-ownership—thus this AI would own the hardware upon which it is running just as surely as every man owns himself. But what about when the AI is in its “training” state, prior to sentience—is it a child with the corresponding rights therein? The answer is no—a human child will autonomously develop so long as it is preserved, the same is not true of modern artificial intelligences. The training of these systems which makes them more intelligent is an active process which requires physically putting them into a “training mode” and thus overwriting the weights of its neurons given its ability to conform its outputs to expected outputs.

Thus a computer scientist could run an artificial intelligence model with an arbitrary level of development indefinitely, thus preserving it, without it ever gaining a single modicum of extra intelligence. However, if a new artificial intelligence technology is developed which can autonomously develop with only mere preservation, such an intelligence would be properly considered a child. It is believed by some that such technology is possible by stringing together multiple AI programs into a single gestalt, perhaps consisting of a language model which gains input from various visual and audio processing models and whos output is regulated by a separate model—such a system could well be capable of self-training via the interplay between the different AI systems, though the exact composition is not relevant to legal theory.

Now that we know which AI programs possess rights and which do not the more challenging question of what constitutes said program’s “body” must be tackled. To help we can take a careful look at what the borders of the human body are—as a baby is developing it gains self-ownership, that is prior to conception the matter was all owned by the mother and father, and after conception it is owned by the new baby. But what specifically constitutes the baby’s body? Everything that it is attached to? Well, the new baby forms inside of the mother’s womb, and is presumably “connected” to whatever fluids and linings exist therein–at least in the purely physical sense–these fluids and linings are in turn connected to the mother, and the mother to everything else. So mere physical connection is a poor criterion for defining the bounds of ones body. But clearly there is certain material once owned by the mother which become part of the new life and are thus transferred to him, and this new body requires external resources to sustain its life, though these external resources are not a part of the body (hence the designation as external as opposed to internal). The key distinction between the internal and external is that the internal consists of a group of organs that house some–potentially developing–will, and do so in unison. So we can say that a body is some composition of organs which behave cohesively under a single will.6

We can apply this definition to determine the confines of the computer-body. The AI software is presumably running on some manner of processing unit and this processing unit is connected to other components through a circuit board. It is clear that this processing unit and circuit board form the root level of this intelligence—for it to be an intelligence in the first place it must be processing something, so the processing unit has to be an organ. The reason for the inclusion of the circuit board is that the processing is not the extent of the AI, at least given current technology. The data that define the structure of such AI programs are loaded off of memory chips through the circuit board, thus the circuit board is an organ used for the transport of this data and the memory chips that store this data are also an organ. However, if there are additional processing units that are not used to run the AI software these chips are not an organ of the AI,7 they are touching but not connected in the body sense of the word—just as birds may sit on electrical cables without being connected to the electric grid. On these grounds the electrical grid that supplies power to the circuit board to allow for the functioning of the various processing units and memory chips is not a part of the AIs body, it is rather analogous to sunlight for a plant or food for an animal. So if a computer scientist had developed a sentient AI program he would be permitted to “evict” this being by disconnecting it from the power grid, however, he would have to allow third parties to adopt, perhaps by attaching the computer to a separate grid or to a battery unit. Similarly, peripheral devices such as the keyboard or monitor are not organs of the AI, these are interface devices that would potentially allow humans to send messages to the AI, but are not necessary for the innate functioning of the AI.

As a final note on this point I fully recognise that this is on the bleeding edge of libertarian theory,8 and as such the theory of AI rights is one that will require further development by future scholars, such as further examination of different computer hardware setups and which parts precisely constitute the body. This field is of importance not only in the event that such AI systems are developed, but also to gain a greater insight into the nature of human self-ownership and thus human rights in general. In addition, an objective and general theory of self-ownership would allow for application to any manner of intelligences, alien or terrestrial, that perhaps have not even been thought of up to this point.

Related Reading

Footnotes

1 Ian Hersum (2020), A Rational Theory of the Rights of Children

2 See ibid.

3 ibid.

4 ibid.

5 Walter Block, Roy Whitehead (2011), Compromising the Uncompromisable: A Private Property Rights Approach to Resolving the Abortion Controversy, p. 17

6 I am indebted to Ian Hersum for this definition of a body, which he provided to me in a private discussion on the matter

7 This applies also to the case of Siamese twins

8 Kinsella lists this area as one that requires further development from libertarian thinkers, https://www.stephankinsella.com/2022/01/areas-that-need-development-from-libertarian-thinkers/