Libertarianism and Paternalism

Previously I have argued here (and here) that the Self-Ownership views associated with left and right-libertarianism have difficulties stemming from their failure to adequately differentiate serious from unimportant property rights infringements. The self-ownership libertarian (the only kind of libertarian I am here discussing) tends to conclude that we enjoy very strong protection against paternalism or infringing our property rights for the sake of the greater good of others. They tend to reach these conclusions by supposing that our property rights provide strong (if not absolute) protection even against infringements that involve only small or trivial harm to the person whose rights are infringed. This presupposition is what licenses the inference that such actions are quite generally wrong without an investigation into the size of the harm that would be caused by the infringement.

But such powerful protections would make impermissible most pollution or fires as these things cross the border of other people’s property, e.g. their lungs, without permission. I argued earlier (following Nozick and Railton), that when we see this, we see that the above simple path from self-ownership to a vindication of traditional libertarian conclusions is unpromising. Thus the path from self-ownership to traditional libertarian conclusions needs to become more complicated if it is to be plausible.

One obvious way to respond to the challenge would be to distinguish between important property rights and relatively trivial ones and be willing to sell violations of the less important property rights relatively cheaply for social good. That is, the view might provide a theory of value that explains why some property rights are more significant than others by showing that some protect more valuable things and others protect only trivial things.

So we are supposing our libertarian seeks a theory of value that simultaneously accomplishes 3 things: 1) the distinction in significance of rights stems somehow from the thought that we are self-owners rather than being ad hoc or potentially in tension with that view, 2) it vindicates traditional libertarian conclusions such as the broad impermissibility of anti-paternalism and the near inviolability of our body when it comes to taking hair or blood for others who badly need it, yet 3) it makes room for infringements on our property rights where we think we surely must permit them, such as in the case of some not very toxic pollution, intuitively acceptable risks such as flying planes over people’s heads, and soft-paternalism cases such as pushing people out of the way of busses. Let us call these the three criteria of adequacy for a theory of value that will serve our libertarian’s purposes.

Here I want to try out the start of an argument to the conclusion that there are great difficulties for any such theory of value that would serve our libertarian’s purposes.

Let us consider two different types of such theories of value our libertarian might offer. This theory of value can either, in one of several different ways, defer to the agent’s own point of view in determining what makes an infringement more serious or not do so. Just to have labels, lets call the former subjectivist and the latter objectivist. There will be many different objective and subjective theories. All subjective theories will, in one way or another, defer to the agent’s choices or preferences under certain conditions.  

If the theory of value is objectivist, then although the agent who does not consent to paternalistic action A or B (both of which infringe upon her property rights) but who prefers that A happen rather than that B will, in some cases, nonetheless enjoy less protection from B than from A. There will be cases where, due to the social good involved, paternalistic action B is permissible but A is not. According to our objective theory of value this could either be because B is in fact better for our agent despite her preference to the contrary or it could be because it is thought morally more important that we not infringe on the agent in one way rather than another, despite her explicit preference for the objectively lower ranked option. I want to say that either way, whatever is generating this view about the relative moral significance of different paternalistic actions, it is not stemming from the agent’s self-ownership. To the extent that our ranking of the significance of different paternalistic actions was stemming from the self-ownership of the agent that we are acting paternalistically towards, to that extent the ranking should reflect in some way the agent’s own view of the significance of the infringements upon her. If anything is bad about paternalism on a self-ownership view, it is that others who do not own something are making decisions about what will happen to that thing without gaining the consent of the person whose property it is. Naturally then, what would make such an infringement worse is that it is even less responsive to the point of view of the person who owns the thing. So, I claim, the objective picture is not a good fit for what makes a paternalistic action worse on a self-ownership view. The thing about owning something, at least on the views we are considering, is that this gives me broad authority to have my say determine what may happen to that thing. The objective theory of value we are considering severs the connection between something being mine and being the person with authority over what may happen to it. That is, I am claiming this theory of value fails to capture well the first criteria of adequacy discussed above for a theory of value that serves our libertarian’s purposes. It looks disconnected from and in tension with the idea that we are self-owners.

Alternatively, if our libertarian uses a subjectivist theory of value here, she will have a hard time vindicating the idea that there are powerful considerations against forbidding someone to engage in homosexual sex or being forced to avoid saturated fats. People care about such prohibitions to different extents. Some may not much mind such state requirements while not doing anything that counts as having consented to them. We do not consent to something merely by not minding it. So the problem on the subjective side is that what different people value can differ so widely. As a result, the subjective theory of value will not be able to vindicate the thought that there are classes of actions, such as freedom of conscience, or freedom from interference with self-regarding actions, that we enjoy powerful protections from. On this view under discussion there will be people who do not value or only slightly value this or that traditional libertarian sphere of protection, and then they will not enjoy powerful protections against infringements into that sphere. Further, some may not much mind state paternalism generally, at least when it is objectively correct about what is good for us. Such people will, on the subjectivist theory of value under consideration, enjoy only quite weak protections against a wide range of state paternalism. The traditional libertarian claim that we all enjoy powerful protections against state action that paternalistically infringes upon our property in such ways will not be vindicated. The threatened result would be that the status of our supposed libertarian protections on the less simple, value-responsive libertarian account, will be quite subject to empirical fortune, not unlike consequentialism, in its defense of our traditional Millian liberties. Thus I am claiming such views score poorly in vindicating the second criteria of adequacy for a theory of value that can serve the needs of our libertarian.

The above is sufficient to fuel the suspicion that there is nothing significantly more important about the property rights that libertarians stress than the property rights that protect us from things like pollution. That is, there is nothing about the value of what such rights protect that provides a principled basis for our libertarian’s insistence that the former sort of rights are very stringent while allowing that the latter sort of rights are much more easily made permissibly infringable for the sake of social goods.


19 Replies to “Libertarianism and Paternalism

  1. It is not clear who “self-ownership” libertarians are, but whoever they are they seem not to have a coherent theory of property.
    It is possible to view property rights as rooted in “self-ownership” ( that is in our right to self defense ) and to believe that these rights are “natural” (in the sense that they are not dependent on social convention) without thinking that social convention and accommodation are irrelevant to all property rights.
    That is: I can think that I have a natural right to punch you in the nose for stepping on my foot but not think that my right that you not step on my lawn is natural or that my property rights over my lawn are as unconditional as my rights over my foot.
    You will find such a theory here .

  2. Locke and Nozick are frequently put forward as self-ownership libertarians, but each case is tricky and I myself argue that this is not true of Nozick as he permits a significant range of cases where one’s property rights do not create morally very serious side constraints on other people’s actions. In a range of cases he permits crossing property boundaries provided compensation is paid. I talk about this proposal at length elsewhere but happy to do so here if you like. On the right Rothbard and perhaps Hospers are closer to paradigms. On the left, Peter Vallentyne, Michael Otsuka, and Hillel Steiner are three central figures that strongly stress self-ownership.
    It is an important question how to understand the rights that flow from self-ownership. Two things directly relevant to this discussion should be mentioned. First, Nozick, Hospers, and Rothbard clearly think that pollution is a boundary crossing. Second, to the extent that a principled understanding of how to determine the bundle of rights that flows from self-ownership has taken hold, it is the Cohen-inspired maximal, equal rights over that which one owns; roughly my self-ownership over my body gives me the most maximal set of rights over what I own compatible with your having a similar set of rights over what you own. Vallentyne, Otsuka, and Steiner develop and endorse this picture. That picture would surely have to count pollution as a boundary crossing as well.
    But I certainly did not mean to be assuming that the self-ownership view couldn’t allow that some property infringements are more serious than others. Indeed, I was looking for how they might make such distinctions in a principled way.
    As left and right self-ownership libertarians differ sharply about the extent to which world-ownership is as morally serious as self-ownership, it is not a perfect case to consider cases where one’s lawn is at stake.

  3. I should have also said that I would not myself equate self-ownership with rights of self-defense for a few reasons. First, my self-ownership makes it wrong to threaten me such that my rights of self-defense are activated. My rights of self-defense are, generally, the rights I have against people who fail to respect my self-ownership. If things are going well, other people respecting my self-ownership will prevent me from being in a position to exercise my rights of self-defense. Second, on some right-libertarian pictures the fact that I own myself helps explain how I can come to own parts of the external world by mixing my labor with it. This has nothing to do with self-defense. I think it better to see rights of self-defense as but one important upshot of the fact that I am a self-owner.

  4. You say, “First, Nozick, Hospers, and Rothbard clearly think that pollution is a boundary crossing.” I am not sure exactly what you have in mind here by “boundary crossing,” but if you are attributing to Nozick the idea that ANY pollution, even a trace amount, that falls on a person’s property would constitute a wrong against the owner, I think you are clearly in error. On page 79 of ASU, he says “Since it would exclude too much to forbid all polluting activities, how might a society (socialist or capitalist) decide which polluting activities to forbid and which to permit?” Presumably, it should permit those polluting activities whose benefits are greater than their costs, including within their costs their polluting effects.” Without using the term, he then in the same passage goes on to discuss the virtues of cost/benefit analysis.
    The issue of externalities such as pollution may pose a problem for those whose libertarianism is derived directly from the self-ownership thesis, but Nozick clearly didn’t hold this view. The outline of his ethical argument is found on p.34, “Thus we have a promising sketch of an argument from moral form to moral content…” The substance of this argument has NOTHING to do with self-ownership, nor does he even mention self-ownership in the entire Part I of ASU where he derives libertarian rights. Those interested in greater detail on this may wish to consult my book “Nozick’s Libertarian Project: An Elaboration and Defense” (Continuum International, 2011).

  5. Mark,
    As I read Nozick he starts with a self-ownership view and amends it to overcome problems such as the pollution case. His amendment is to allow crossings onto another person’s property provided compensation is paid that makes the person whose property it is no less satisfied with the combination of cross and compensate than they would have been with neither. That is the upshot of his discussion of pollution. Of course he goes on to limit the appropriate scope of cross and compensate but not in ways that much impact on his discussion of the pollution case. I do think the passage you quote (and surrounding passages) key in showing that that is Nozick’s response.

  6. Perhaps worth adding that a boundary crossing would become an infringement unless compensation is paid. Boundary crossings without consent require compensation (catastrophes aside).

  7. If, as you say, Nozick “starts with” the self-ownership view, and then amends it, I find it fascinating that the first mention of SO is on p. 172 in Part II, after he spends much time discussing rights in Part I. I also find it extremely interesting that he proposes “utilitarianism for animals, Kantianism for people” on p. 39 (before rejecting utilitarianism as too lax even for animals).
    I am sure you agree with me that in discussing the work of other philosophers it is incumbent upon us to give their work the strongest possible interpretation consistent with their core principles. I’m afraid you have not done so here. Nozick’s principle of “border crossing w/ compensation” is directed at a special class of activities that are not risky enough to criminalize, yet the sort of things that people need to do in their everyday lives. Nozick gives the example of an epileptic driver. In cases like this, he advocates that we may ban the epileptic from driving because he presents too grave a risk, but we must compensate him. I don’t think he intends this analysis to apply to pollution cases at all.
    Let me respectfully suggest that after you have had your fun analyzing him your way, you try running the same arguments using an interpretation most favorable to Nozick’s core principles, i.e. rights are based on Kantian principles of respect for people as rational agents, and an approach to externalities that embody this idea. This understanding of Nozick is not just mine, but shared by other philosophers sympathetic to his views, and it at least has the virtue of being consistent with what Nozick actually says in ASU.

  8. Cross and compensate is a different principle from his Principle of Compensation. The latter is, as you say, more limited in scope. The former is limited in a few ways, such as cases where the activity would create widespread fear, but much broader in scope than the Principle of Compensation.
    As tempting as it is to continue discussing Nozick interpretation and descend into trash talking with you, I was hoping to have people talk about the central thought behind the new post.

  9. Many left-leaning philosophers have made a cottage industry out of attributing implausible views to Nozick and then, low and behold, finding his views implausible. I do object to it, but am sorry if you find this “trash talk.” Have fun.

  10. Thank you for your reply, David.
    “Self-ownership theory of property ” is your designation and you are free to use it as you wish but it does seem to me that the Retributive Theory of property ought to be included in that tent.
    The Retributivist holds:

    1. That the right to self defense is (that is, is identical to) self-ownership. Note the Retributive account of rights explicitly rejects the idea that the right to self defense is “activated by” or “explained by” some prior right against harm or self-ownership.

    2. That our property rights in most other things arise from social agreements , that is from promises we make to one another. [cf.


    3. The Retributive theory of Promising analyses promises as conditional relinquishings of our rights to self defense. Cf. Promising

    (2) allows us to negotiate as complex a structure of property rights as we might wish while (3) entails that their moral force always derives from the right of self defense.
    I do not say this is a “Libertarian” theory because I don’t think the label has any useful meaning. But it is a theory of property wholly founded on the right to self defense (= self-ownership).
    I’m not complaining that you aren’t familiar with this stuff, but when you make an argument of the form “any theory of X has these problems” you never know what will come out of the woodwork.

  11. Tomkow,
    I don’t know that I used the phrase “self-ownership theory of property.” The line of thought I mean to be talking about is making a claim about what sort of fundamental rights people have and what grounds them. It is important to that tradition that the rights of self-ownership generate a whole host of things others may not permissibly do to us—indeed I think they think it generates most of the central aspects of enforceable morality. So it is very different from the view you mention. If I call it the self-ownership view of our fundamental, enforceable rights, would that make it clear that I don’t mean to make trouble for the retributivist?

  12. Hi David,
    I’m not sure if this is to the point, and it surely isn’t developed to the point where it even comes close to answering your post, but I think one way a libertarian might proceed is by talking about the various rights that compose ownership rights and taking some as central and others as derivative. My guess is that a libertarian is going to think of rights to control and use as primary and rights to exclude as less important and perhaps even derivative from these. I guess this is on the objective side of your division. And from there a certain sort of libertarian might try to work out what level of pollution would give the most robust right to control and use what one owns. It could be that allowing me to use my stuff in a way that puts others at minimal risk (along with a compensation clause when the risk is realized, or maybe for the risk) generates a more robust right of this sort than one that doesn’t do so. By robust here I mean it protects one’s ability to control and use what one owns better than less robust alternatives.
    I’m not sure how helpful this is . . .
    And, FWIW, a comment that begins, “Let me respectfully suggest that after you have had your fun . . .,” isn’t showing a whole lot of respect so that the trash-talking label does not seem to me to be uninvited.

  13. Mark,
    Thanks for that. It is interesting how to understand the rights of self-ownership even if we accept something like the Cohen-inspired maximal equal rights over oneself. For seemingly maximal equal use rights would differ from maximal equal rights to exclude. So, as you suggest, it seems that the friend of self-ownership must say more than just the Cohen-inspired thought—they have to tell us more directly which are the rights that are central to owning something. I need to read more here but Vallentyne, Steiner, and Otsuka (writing together) and Schmidtz prioritize exclusion-rights over use rights. And this does not surprise me given that they need my rights to use my arm to stop at my neighbor’s nose—here rights to exclude need to take priority over rights to use. And quite generally (especially on the right) other people’s rights to exclude me from access to their money and spare body parts is thought to take strong priority over the rights for me to use my limbs.
    But I definitely see what you are saying and I do think that would provide a principled way to help in pollution (and trivial risk) type cases. It is just difficult to see how the view developed that way will vindicate the powerful side-constraints self-ownership views have tended to be conjoined with.
    More broadly, the general thought behind your suggestion is that perhaps a better conception of the rights that flow from self-ownership could suggest that pollution type cases do not count as an infringement on other people’s bodies. That suggestion seems tempting in part because we do not in fact think of not very toxic pollution as inappropriately usurping control of my body or failing to respect the fact that I get to decide how I will live. Surely, I hear your thought as being, we can find a plausible conception of the rights of self-ownership that vindicate that commonsense thought.
    That strikes me as a good suggestion on their behalf. I have said a bit above about why I would be surprised if such a conception were forthcoming that could make pollution not a problem yet made taking a bit of my money or hair a significant problem, but perhaps the best path is for the self-ownership views to not be so strict with respect to money and hair which would allow them to be more permissive with pollution.
    Need to think more about this tomorrow.

  14. Hello David
    I really like this post, and the dilemma that it poses for libertarians. But I wonder whether the argument against the subjectivist option needs to be made more sharply. It might be argued that standard civil liberties can be protected on a subjectivist basis: it is difficult to know what people will value, so we have strong reasons to provide them with a wide range of options.
    There might be other arguments against a purely subjectivist theory of evaluating side-effect harms in determining when it is permissible to impose them though. Consider the following:
    1) A trolley is heading down a track towards a nuclear device. If I do nothing the world will be destroyed. I can divert the trolley to the left. Harry is on that track and if I do so he will be killed. I can divert it to the right. On that track is a religious idol owned by Harry which will be destroyed if the trolley is diverted towards it. Harry would prefer to lose his life than see the idol destroyed.
    This case seems difficult in the same way as it is difficult to know whether it is wrong to give life-saving blood transfusions to unconscious Jehovah’s witnesses. Suppose that libertarians are committed to killing Harry. His valuing of the idol over his life is decisive. We should treat it as more valuable than his life, given that he treats it in this way.
    2) As 1) except that you are in place of the idol.
    In this case it seems that equal chances ought to be given to Harry and you. Plausibly, I ought to flip a coin.
    3) As 2) except that the idol is in place of Harry.
    Given the judgement in 1) it seems that subjectivists should rank the idol as more valuable than Harry’s life. Given the judgement in 2) it seems that subjectivists should rank Harry’s life as equal in value to your life. This seems to imply that in 3) the idol should be ranked as more valuable than your life. But not only does it seem wrong to kill you rather the idol, it seems as though I must destroy the idol. It would be wrong to flip a coin.
    It might also be worth considering
    4) I have three options: kill Harry, kill you or destroy the idol.
    In 4) it seems to me plausible that I ought to kill Harry. Killing you would be wrong where I have the option of destroying the idol. Respect for Harry might require me to kill him rather than destroy the idol.
    This suggests that subjectivism is plausible when evaluating intrapersonal cases but implausible when evaluating interpersonal cases. It seems that we need a mixed theory. The question is whether such a theory is consistent with libertarian commitments.

  15. Victor,
    Thanks for that! These are interesting suggestions. I have to worry about teaching now but I’ll be replying asap.

  16. Victor,
    Again, all very interesting. I do think it would be a significant change to justify anti-paternalistic policies on epistemic grounds rather than that paternalism certainly violates rights. It would, for example, limit the scope of the cases where paternalism is problematic in ways I have have not heard libertarians accept. If I know your preferences adequately, you may not have strong protections against interference with, for example, religious liberty. Perhaps one could go Rawlsy and say that people need to be given room to change their minds on such important topics. But that would be to be paternalistic about which topics it is important that I get to change my mind about. Additionally, it is not clear the epistemic version will help keep the right options open. We need some reason to think that some pollution, say, is permissible and that taking small amounts of money even from very rich people for redistributive purposes is not ok. I don’t see the epistemic case as helping justify that pattern.
    I very much like your set of cases involving the idol. I like the view that I get to decide what gets weight insofar as I get weight but obviously I do not get to decide how much weight I get compared to how much others get. I do not rule out allowing the idol to outweigh the other person—that may be a price I am willing to pay to permit people full say over what gets weight insofar as they get weight. But I admit that is quite counter-intuitive. (Still thinking about these issues. Thought I had a case that would vindicate the above but I now think it does not work.)
    Thanks for your helpful thoughts.

  17. Mark,
    I had hoped to come up with better things to say in reply to your suggestion but I don’t see those better thoughts on the horizon so for now I’ll have to settle for what I said above.
    Again, thanks for your helpful thoughts.

  18. If a primitive society were made up of two types of people, “smart but weak” and “stupid but strong”, nature would suggest that the stupid people employ their superior strenth as power and the smart people their brains. The libertarian society asks of the less talented and stupid to accept their lot and for them to abstain from combination to politically steal from the fewer and smarter of their fellow citizens. They request of their fellow man not to abide by their nature, and not to use whatever means is available to them to advance their private interests.
    Good luck with that.

Comments are closed.