Permissible Right-Infringement, Part 2

Guest Post by Danny Frederick,

In Part 1 we considered the following example. 

HIKER. A hiker on a back-packing trip in the high mountain country is beset by an unanticipated blizzard which strikes the area with such ferocity that her life is imperilled. She stumbles onto an unoccupied cabin, locked and boarded up for the winter, clearly somebody else’s private property. She smashes a window, enters, and huddles in a corner for three days until the storm abates. During this period she helps herself to her unknown benefactor’s food supply and burns his wooden furniture in the fireplace to keep warm.

It is permissible for the hiker to do as she does, even though she thereby infringes the property rights of the cabin-owner (throughout, ‘permissible’ means morally permissible). However, in virtue of infringing the cabin-owner’s rights, she owes him appropriate amends, where what amends are appropriate depends upon the full circumstances of the infringement. 

This does not entail that it is always permissible to infringe a right so long as one makes appropriate amends. For example, consider someone who burgles for excitement but always fully compensates his victims, even adding a sincere apology to the compensation. His apology is sincere because he is genuinely sorry for the discomfort he causes his victims, but he burgles them anyway because, in his opinion (and perhaps in fact), the benefit to him of the excitement is far greater than the inconvenience, mental and physical, to his victims, and he will compensate them for that in any case. This burglar still does wrong because it is impermissible to infringe a person’s rights for fun even if one makes full amends, and even if the sum total of human happiness is increased thereby. Why should that be so? 

Rights imply moral duties. Our moral duty is what Kant calls a Categorical Imperative: it tells us the thing to be done, no matter what we want. The rights of a person to her home and its contents give the burglar a moral duty not to enter her home and not to remove any of its contents. If this is the burglar’s moral duty, the fact that he enjoys burgling is irrelevant, even if his enjoyment far outweighs the dissatisfaction of his victim, and even if he makes appropriate amends which include full compensation to the victim. Thus, the fact that a permissible right-infringement demands appropriate amends does not imply that it is always permissible to infringe a right if one makes appropriate amends. There are only special circumstances in which it is permissible to infringe a right. 

However, if the moral duty corresponding to a right were always a Categorical Imperative, there could be no circumstances in which it is permissible to infringe a right (because, recall, a Categorical Imperative tells us the thing to be done). Since we have accepted that it is permissible for the hiker to infringe some of the cabin-owner’s rights, it follows that a moral duty of ours that corresponds to another’s right need not be our moral duty. The following example, from Judith Jarvis Thomson (1990, pp. 158-59), can illustrate:

BRAKE FAILURE. Judith is driving, and her brakes have just failed. If she continues straight, she drives into and kills Lewis. Her steering wheel is constricted, so her only alternative to continuing straight is turning to the right. But if she turns to the right, she drives into and kills Michael. 

It is permissible for Judith either to continue straight or to turn right. It is therefore permissible for her to infringe Lewis’s, or Michael’s, right not to be killed. Here, Judith has two duties which conflict. Where duties conflict, one ought to perform that which is most stringent. In BRAKE FAILURE the duties are equally stringent, which is what gives Judith a choice. Because duties can conflict, the Categorical Imperative must be interpreted so that it is always the most stringent duty that is our moral duty or, in case of a tie, one of the tied duties. 

So we can make sense of permissible infringement of a right in cases where there is a conflict of duties. It is permissible for Judith to kill Lewis (Michael) because it is the only way she can fulfil her (equally stringent) duty to Michael (Lewis). In contrast, it is impermissible for the burglar to burgle his victims, because he has duties not to burgle, but no conflicting duties. In his situation, the duties not to burgle are Categorical Imperatives. The fact that it is fun for him to burgle is irrelevant, no matter how much fun it is for him. How does this help us to understand HIKER? 

It is not the case that the hiker raids the cabin because it is fun to do so. Her action is not like the burglar’s. Yet it seems different to Judith’s too. For, it seems as if the hiker’s only duties in the situation are to the cabin-owner. If that is so, then the rights of the cabin-owner imply a compound Categorical Imperative not to break into the cabin and not to consume any of its contents. But, if that is so, it is impermissible for her to enter the cabin even if she afterwards makes full amends to the cabin-owner. How can we avoid this result? 

There seem to be two options. The first is to say that while, in these circumstances, the hiker morally-ought to freeze to death, she rationally-ought to perform a morally wrong action and then make appropriate amends to the victim. However, that seems entirely unsatisfactory because, as we said at the start, virtually everyone agrees that the hiker’s action is morally permissible. 

The second option is to invoke a duty of the hiker to herself, to save her own life, which, in the circumstances, overrides her conflicting duties to the cabin-owner. This might seem to make it obligatory, rather than permissible, for the hiker to raid the cabin. However, we can normally release people from the duties they owe us, so it seems plausible that a person can release herself from her duties to herself. So, it is permissible for the hiker either to fulfil her duty to save herself or to release herself from that duty in order to fulfil her duties to the cabin owner; and hence it is permissible, but not obligatory, for her to raid the cabin. Duties to oneself are not very popular in contemporary moral philosophy; but they seem to have explanatory value. How else are we to explain the permissibility of the hiker’s action?


Thomson, Judith Jarvis. 1990. The Realm of Rights. Cambridge, MA.: HarvardUniversity Press.

This entry was posted in Blog. Bookmark the permalink.

7 Responses to Permissible Right-Infringement, Part 2

  1. Hi Danny,

    I’m very much enjoying your posts here. One small question. You claim that the hiker’s duty to himself is the only way to explain the permissibility of his breaking into the cabin. But might we not seek to explain it not in terms of a self-regard duty but rather in terms of a moral *permission* to save his life at the expense of violating the cabin owner’s rights? Something like a Schefflerian agent-centered prerogative?



  2. Hi Matt,

    It is nice to see you on here.

    The position you suggest seems to be that, in circumstances in which our moral duty makes harsh demands upon us, we are free to ignore it and pursue our own interests. That is an appealing position, no doubt; but hardly a moral one. It is a fact that, human life being what it is, our moral duties will sometimes call upon us to make or risk great sacrifices. But a good person (or, better, a person insofar as he is good) will still do the right thing under those circumstances. Our moral duty is what we morally must do; it is by its nature not such as we can permissibly give it up when it is inconvenient to us. Only a conflicting duty which is not less stringent can allow us to avoid a particular moral duty.

    I don’t know much about Scheffler (so please correct me if I am wrong), but it seems that he was concerned with theories of morality (particularly act-consequentialist ones) which subject moral agents to extremely demanding requirements (always acting so as to maximise the general good, or whatever). These theories seem to give patently incorrect accounts of our duties. Rather than accepting that such theories are refuted, Scheffler proposes his ‘agent-centred prerogative’ to make it permissible for the agent to avoid his duty in some cases when it is too demanding. Which cases? Scheffler gives no rule: sometimes it is permissible, sometimes it isn’t. This seems like an entirely ad hoc fix for a faulty theory of morality. Any theory can be saved from refutation by ad hoc fixes; but a theory amended in such ways (like Marxism) ceases to be intellectually respectable. If we want to improve our knowledge it would be better to accept that the faulty theory is refuted and search for a better one.

    Now, you might say that my resolving the HIKER problem by invoking a duty to oneself is itself an ad hoc fix for the theory of property rights that I espouse. It would indeed be such if I had invented duties to oneself simply in order to get around the problem posed by HIKER. But that is not what I have done. Theories about duties to oneself have been part of moral theorising for centuries. I am making use of a notion that moralists have accepted for a variety of reasons; I am not (as Scheffler seems to be doing) simply introducing a new notion to patch up an inconsistency in my theory.

  3. I want to expand a little. I said:

    “Only a conflicting duty which is not less stringent can allow us to avoid a particular moral duty.”

    This is true only so long as the particular duty remains in force. If the cabin-owner had waived his rights to the cabin and its contents, the hiker would thereby have been released from the duty not to raid the cabin.

    In HIKER, the cabin-owner has not waived his rights, so the hiker has the duty not to raid the cabin. But if we can release ourselves from our duties to ourselves, then the fact that the hiker’s duty not to raid the cabin is less stringent than her duty to herself, does not prevent her from fulfilling the former duty, because she can release herself from her duty to herself.

    You might say that a duty to oneself from which one may release oneself looks a bit like an agent-centred prerogative. What’s the difference? The difference is that duties to oneself belong to traditional moral theory and already have jobs to do in that theory. Scheffler’s notion (unless you can correct me on this score) is introduced to get him out of a fix and does no further work. It is not independently testable. That’s what makes it ad hoc.

    Am I committed to accepting all the baggage that goes with the traditional notion of duties to oneself? No. But, if I am to avoid the charge of making an ad hoc manoeuvre, I must be able to show that the notion of duties to oneself can solve problems in addition to the problem I am using it to solve. That, I have to admit, is a task for my ‘To Do’ list. So, if I am asked the question, ‘Under what conditions would you give up your theory?’, I would say: if the notion of a duty to oneself solves no problems apart from the problem of permissible right-infringement in cases relevantly similar to HIKER.

  4. Hey Danny,

    Good to see you here too!

    Your point about an agent-centered prerogative not being a moral position is interesting, but I’m not sure I agree. It depends, I suppose, on whether one simply assumes at the outset that morality must be essentially agent-neutral in its content. This assumption seems to me to be shared by Kantian and utilitarian theories, but not by eudaimonistic ones.
    You’re right that Scheffler introduced the idea of a preogative as a way of coping with the demandingness objection to consqeuentialism. But it strikes me that (and Scheffler might have even argued this himself) deontological theories can be just as demanding. Certainly a deontological theory that denied that the hiker could permissibly break into the cabin would be making an extremely demanding claim on the hiker. So, one way around that might be the one you suggest. An agent-centered prerogative might be another.
    I’ll have to think more about the ad hoc objection. I think that the agent-centered prerogative reflects something important about the connection between morality and practical reason, and so is theoretically well-founded. But that’s not something I’m prepared to defend here.
    In the end, though, I’m not all that sure that there’s much difference between my claim that the hiker is to *permitted* to prioritize his own survival over the right of the cabin owner, and your claim that he has a *waivable duty* to himself. In deontic terms, these look exactly the same:
    Permission to X = Not wrong to X; Not wrong to ~X
    Waivable duty to ~X = Not wrong to ~X; Not wrong to X
    Or is there a deeper difference that I’m not noticing?


  5. Hi Matt,

    It is true that the agent-centred prerogative and the waivable duty to oneself each imply that it is permissible for the hiker to raid the cabin. But that does not make them equivalent. I have suggested two ways in which they are not equivalent.

    First, the notion of duties to oneself has other jobs to do, so we need it anyway, but the agent-centred prerogative seems to be an ad hoc patch.

    Second, the agent-centred prerogative seems inapproriate for a moral theory. While it is true that a moral theory which is too demanding should be rejected as false, it is also true that a proposed moral theory that is not demanding at all should not count as a moral theory. I assume here that the notion of duty must be included in a moral theory and that our duty will sometimes conflict with what we want to do. This means that the notion of an agent-centred prerogative needs to be beefed up: we need to know in which circumstances it arises and in which it does not; otherwise we have carte blanch to do as we please (which, from what I gather, is where Scheffler leaves it). In contrast, the notion of waivable duties to oneself does not lead to ‘anything goes,’ since many reasons for avoiding a duty would not count as a duty to oneself, and since one may not release oneself from a duty to oneself for just any reason.

    Of course, all this needs much more discussion and, prior to that, a lot more work from me. But it does seem to me that duties to oneself promise a non-ad-hoc and morally acceptable route out of the HIKER dilemma, whereas agent-centred prerogatives do not.

  6. Matt Zwolinski says:

    Thanks, Danny. Food for thought.

    I’d be interested to hear your thoughts on the debate I was having with George Smith, here:

    I don’t take myself to have been saying anything (in that particular thread) with which you would disagree. But I find smith terribly difficult to talk to, and have given up. I’d be quite interested in reading your ideas, though, if you waned to chime in.


    • I agree that debate degenerated, as all too frequently happens. It is a problem with human nature: people find it difficult to distance themselves from their theories, so they take criticism of the latter personally, which then interferes with their ability to make a reasoned, or intellectually appropriate, response. No one is immune: I fall into the trap myself sometimes, despite my Popperian/sceptical philosophy which explicitly warns me against it.

      I will make a small contribution to that debate, though I doubt that I will be able to revive it.

Leave a Reply

Your email address will not be published. Required fields are marked *