Some Of Our Books


« Attributability and Daddy Issues | Main | Purdue Conference on Challenges to Religious and Moral Belief »

April 26, 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi Alec,

I assume you think there are moral requirements and permissions that govern personal relationships.

If so, I wonder what you would say about this "Anscombian" pair of parental punishment cases:

(Reluctant Randy) Randy punishes Randy Jr. because that will help junior learn to do better.

(Gleeful Glenn) Glenn punishes Glenn Jr. because that will help junior learn to do better and because he will enjoy inflicting pain on junior.

I assume it is not objectionably "illiberal" to claim that, if we specify the case a bit more, Glenn acts impermissibly but Randy does not.

I wonder how you will handle this case -- given your optimism about our ability to determine the scope of our intentions, I am thinking Glenn can avoid acting on an impermissible intention, as you understand those, and he seems to be acting impermissibly because he is acting on an "immoral reason".

I think this is an objection, but it might just show I have failed to absorb the details.

My general worry is that your overall claim that, "the victim may claim that it adds insult to injury to be acted upon by one who thinks so hatefully or instrumentally about him...but we just aren’t entitled to demand that others not have insulting thoughts about us," sound plausible when we think only of relations between strangers but will break down in various more personal cases.

Nice set of cases. I am torn between 2 reactions. One says that Gleeful Glenn does not ACT impermissibly. Rather, he lives in a really morally bad state, that of being a parent who enjoys inflicting pain on his child. He has a strong moral reason to fix that about himself, but insofar as he is constraining himself to doing what is permissible, he acts permissibly. After all, what is he to do? Not inflict the deserved and helpful punishment? And even if he can avoid acting On his sadistic feelings, he may not be able to avoid Having them while he administers the punishment. And perhaps its the Having of them that is really objectionable.

My other reaction is to say, no, you're right, there is something Extra objectionable about acting On the sadistic reason. But then it's not really that hard for me to take this on board. It's a special relationship point. Parents have special duties towards their children that in many ways are like the duties one has towards those with whom one makes voluntary commitments to treat them in special ways. And with those special duties comes a special responsiveness to the needs, both emotional and physical, of the others. So this could really be categorized as a kind of expression and expectation case: it's an expectation case where the special relationship grounds a special right to expect respectful reasoning.

In other words, I am more or less happy to accept your last line, that my point about my liberal, anti-perfectionism, namely that it may apply mostly between strangers (thought also, perhaps, in certain business contexts, and other contexts of a less than intimate nature).

I am grateful to you, Alec, for this excellent post. I learned a lot from our discussion. There is a wide range of issues in the post, and we might be best focusing on a smaller range at first.

One thing that you could clarify is your idea of the right to do wrong – doesn’t this give the game away, in that where a person has a right to do wrong, although that person may not be interfered to prevent her from acting wrongly, she nevertheless acts wrongly. If she acts wrongly, she doesn’t act permissibly. Hence, on your own presentation, it seems that intentions are relevant to permissibility – it’s just that others may not interfere with a person on that basis.

Secondly, let me respond, first, to your way of dealing with Gang (the case of the post office robbery) and Poisoned Pipe.

One difficulty with your account is that it indicates that members of the gang in Gang and the poisoner in Poisoned Pipe would act permissibly if they form an intention only to act in the same way that a person with good intentions would act. This doesn’t seem plausible to me. Suppose that A in Poisoned Pipe poisons in order to receive a reward that he has been offered by Boss for the killing of victim. However, he does this only on condition that B also acts. He does this only because that will allow him, if he is caught, to offer the inevitability of B’s action as a justification for his action. On the DII he has done nothing wrong. That is hard to believe.

Another difficulty is this: it seems that A and B, if they act in order to kill Victim simply out of hatred or for the money, not only act wrongly, but are liable to preventive, compensatory and punitive harm. These liabilities that they have are importantly related to the magnitude of harm that Victim is threatened with or suffers. We can focus on compensation, as that is the simplest case. It is highly intuitive that A and B may be harmed in order to provide compensation to Victim. The intuitive way to calculate the magnitude of compensation owed, compares Victim’s circumstances with the circumstances that she would have been in had neither A nor B acted. Furthermore, if A acts only in order to benefit Victim and B acts in order to kill Victim, B alone owes Victim compensation, calculated in the same way. These judgements seem compelling. They suggest at least that liability to pay compensation depends on intentions. But liability to pay compensation is at least importantly related to permissibility.

Here is one final idea. You acknowledge that your account is counterintuitive in that it implies that there is no murder in Poisoned Pipe. You then claim that my account is counterintuitive in another way. I agree that if we cannot encourage A to act with good intentions in Poisoned Pipe we ought to encourage him to act with bad intentions, for doing so will benefit Victim. This is even clearer in cases like Bad Trolley Driver – if we cannot encourage the bad driver to turn the trolley in order to save the five, we are justified in encouraging him to turn it to kill the one.

This, you suggest, has the counterintuitive implication that we are justified in encouraging wrongdoing. I argue in The Ends of Harm that there is nothing counterintuitive about that idea – it is familiar in other circumstances. I aim to demonstrate this with the following disgusting case (invented by Matthew Clayton for a slightly different purpose):

Claw Hammer. I am standing on the shore of a lake and a boy is drowning out of reach. There is a boat on the shore and only one person can get in the boat. Unfortunately, though, I’m no sailor and if I try to rescue him I will almost certainly fail. You, an experienced sailor, are standing next to me. I encourage you to save the boy, but you are unwilling to do so. You are wearing a shiny new suit, and once you get out to the boy your suit will almost certainly get wet. I have a claw hammer and I could give it to you. That would allow you to save the boy without damaging the suit. But unfortunately you would have to jam it into his eye to pull him out.

It is wrong for you to save the boy with the claw hammer. Yet I am clearly justified, perhaps required, to provide you with it – better that the boy is saved with the clawhammer than not saved at all.

The moral structure suggested by the case is as follows:

You have these three options:

1) save the boy at the cost of your suit
2) save the boy with the clawhammer
3) don’t save the boy

I, in contrast, cannot motivate 1). Unlike you, I cannot bring it about that the boy is saved without him losing an eye. The best that I can do for the boy is ensure that you choose 2) rather than 3). The same thing is true about Poisoned Pipe. A, in that case, could act in order to save Victim (we agree about that). He will not do that – he will only act for the money. He has an option that I cannot get him to take. I have two remaining options – let Victim suffer a slow and painful death or encourage A to act wrongly, improving Victim’s condition. As in Clawhammer, I am justified in encouraging A to act wrongly for the sake of Victim.

I’m so glad to see your reply Victor, and I'm glad you started your reply with the right to do wrong.

No, my position that the act is wrong doesn’t give the game away. You’re failing to distinguish should from must. There’s ambiguity in the concept of an act that is morally wrong. It can be one that the agent should not perform, or one that she may not perform. Put third person interference to the side. Just from the first person point of view, there is a difference. One can say to oneself, I know I shouldn’t do X, but screw it, I want to and I will. Of course one can also say to oneself, I know I must not do X, but screw it, I want to and I will. The difference is that one can see oneself confronting different kinds of norms, different levels of stringency in the moral address. The “should” address takes the form, essentially, of a request. It’s like morality saying to you: It would please me if you did not do X. But it leaves you the privilege to say back: I know, but I am sorry, it matters more to me to do X than to please you. The “must” address takes the form, essentially, of a command or a demand. It’s like morality saying to you: Stop. Given that you move in the space of rights, and make demands of others, so you must respect this demand now: you have no privilege to do X.

We learn this difference as children. Parents use it all the time when they say, for some things, “I wish you wouldn’t do X” and for others “You may not do X.” And as we internalize the voice of authority and turn it into the voice of morality, we retain the sense—I think—that there is a difference between the requests of morality and the demands or morality. We can take morality seriously and turn down its requests, but not its demands.

One more point: don’t confuse this point with the supererogatory. The supererogatory is not a matter of request; it is simply a matter of praise for going above and beyond any request. There’s more to be said there, but all I need now is to ward off the confusion.

Turning now to the Poisoned Pipe cases, I simply disagree with your intuitions. But let's be real clear about the cases.

First, I know you know, but to be clear: I can easily avoid approving of A and B conspiring to poison V, arranging to put poison in the pipe only if the other does, and then only because they would get the reward. If that is the situation, there is no inevitability of the other putting poison in. The DII can handle that by saying that the choice to conspire was wrong.

But if we hold constant that A will put poison in, then surely what you say about B being liable to preventive harm is wrong. The whole premise of your example—which I granted you for the sake of argument, though I actually think it’s wrong—is that we should want B to put the poison in. And once I have that, it’s hardly so clear as you make out that B should be liable for compensatory or punitive harm.

In fact, I think part of the problem with your intuitions is that your example is actually a bad one. We do not think that B should be able to poison A, even to speed A’s death, because, as I said in my original post, it is like poisoning a person dying a slow painful death due to cancer: you still need his permission.

So let’s change the example to someone who would turn a trolley from 5 onto 1 but only for the sake of killing the one, her enemy. She wouldn’t kill him if it weren’t permissible for someone in her position to do so, but she’s a moral loopholer: she kills not for the sake of the five, but because of her hatred of the side-track man. Does she owe compensation that another wouldn’t owe to the side-track man’s family? May she be punished? I think the answer to each is clearly no.

Finally, I HATE your clawhammer example, but OK, you brought it back up again, and to good purpose. I agree that we may encourage wrongdoing. But, as I argued in a paper called “Permissibly Encouraging the Impermissible,” Journal of Value Inquiry 38 (2004): 341-354, there are scales of difficulty in justifying the performance of acts that are more or less extreme on the impermissible spectrum. In that paper I considered three types of encouraging cases (in all of which it is important that the thing be done): (a) the person knows how to do the thing in a permissible and in an impermissible way, will take us only to be encouraging her to do the thing, and yet we believe she’ll choose to do it in an impermissible way; (b) the person doesn’t know that there is a permissible way to do the thing, though there is, and thus will take us to be encouraging her to do the permissible thing in an impermissible way; (c) the person is demanding that we encourage her to do it in the impermissible way or she won’t do it (in your case: she demands that you chant at her: stick him in the eye, go ahead, do it!).

Well, just as I think that spectrum moves from easy to justify to hard to justify—for reasons I go over in the paper—so I think that it’s easier to encourage someone to attempt to do something wrongful (attempting is itself an impermissible act, so one is still encouraging the impermissible), than to do an act that is actually a crime. So in that regard, I think my account still comes out ahead in this way.

Thanks Alec

so, you are better of not to say 'the right to do wrong' but rather 'the right to do what one shouldn't' (given that what is not wrong is permitted and what is not permitted is wrong, which I take it is the standard view)

you disagree about whether it is permissible for A to put poison in the pipe even though this will give victim a swift rather than a slow and painful death. We can adjust the case to make it more compelling (though I think it clearly compelling enough). Just make Victim's death instantaneous if both put poison in and really slow and really painful. If you are worried that Victim lives longer if she has a slow and painful death than if she has a swift death (which might drive your judgement about the cancer case), correct for that by making the time of death identical in each case, but if only B puts poison in the pipe, the lead-up to death will be painful, but not if both A and B do so.

I have given Poisoned Pipe to many audiences, and almost no one shares your intuition that A and B are not liable if they act on bad intentions (off the top of my head I can't think of another person). But perhaps other Pea Soupers will jump in to your defence here. Some people respond by claiming that liability tracks blameworthiness in this case rather than permissibility. That has other problems.

I think the explanation of cases like clawhammer is much simpler than you. What we are permitted to do depends on our other options. You cannot rescue the child with the clawhammer because you could rescue him in a way that harms him much less at a tiny cost to yourself. The option of the child being rescued at this tiny cost is not available for me. For me, it's either the child gets saved at the cost of an eye, or he dies. So I am permitted to assist in him being rescued at the cost of an eye. This explanation applies equally to cases like poisoned pipe (given that, as we agree, A could act on an intention to benefit Victim). So I don't see the advantage that you claim for your view.

Hey Victor,

I think you are simply mistaken about the standard view of what is meant by wrong in the right to do wrong. The classic source for this issue is Waldron's piece, and his examples all concern vices such as stupidity, cowardice, tastlessness, etc. These seem more like "should" cases than "must" cases. Granted, he dismisses the idea that what is important in the right to do wrong is the first-person privilege to do so. But I think that was a deep mistake on his part.

I don't know what to say about your intuition polling. I suspect it would change if I presented the case to audiences with a persuasive background theory, allowing us to say that the persons acting on bad reasons deserve censure, but nonetheless act permissibly. In saying that I'm just taking the Scanlon & Thomson line, and I guess they've presented their position to many audiences as well.

Lastly, I get the set up of clawhammer. I just say that one should be more uncomfortable with this class of cases than with cases in which that act performed is itself permissible.

Well, we may have run out of steam between us. Perhaps others will pick up a thread or two. Or perhaps by posting the longest post in Pea Soup history I've doomed the discussion to oblivion. I hope not.


Hi Alec and Victor,

Thanks for this absorbing discussion. I’m not familiar with this debate at all, but I’ve found myself puzzled by and sucked into it by your arguments. Thanks! (I think.)

I have two questions. I hope that they’re more interesting than an invitation to rehearse Intentions & Permissibility 101. If not, please feel free to direct me elsewhere.

The first is simple: is it an assumption of the debate between you that intentions are, at the fundamental level, going to be either relevant for all judgments of moral permissibility or irrelevant for all judgments of moral permissibility? I ask because it seems that one might suppose that judgments of permissibility could be intention-sensitive for certain acts (e.g. helping someone carry his shopping into his house) and intention-insensitive for others (e.g. not stealing from someone), and that there might be a systematic explanation of the distinction between acts of one type and acts of the other. Is that possibility ruled out or accommodated by everyone in some way (or have I misunderstood something)?

The second question is for Alec. If I understand you aright, you want to explain a right (here, a privilege) that we have to do wrong in certain ways by appeal to the more fundamental right (which must itself be a privilege, in order to do the explanatory work) to “lead our own lives, and to decide how and when to work on perfecting ourselves”. And you want to explain the existence of the more fundamental privilege and the fact that it makes sense of the privilege to do wrong by appeal to the idea that the duty to work on perfecting oneself morally is imperfect, which, as you interpret it, makes room for a permission to be vicious.

Could you say more about how the imperfection of the duty to perfect oneself morally makes room for or is a permission to be vicious? On the face of it, since you might think that one can on a given occasion be failing to work on perfecting oneself morally without thereby being vicious, it doesn’t look like the imperfection of the duty to work on perfecting oneself morally must make space for permissible viciousness.

Suppose that things are as they appear here—that it’s not the case that the imperfect duty to work on perfecting oneself morally must make room for permissible viciousness. Then the existence of any fundamental privilege to lead our own lives that is explained by the imperfect duty isn’t necessarily a reason to think that we have the right to do wrong in your sense. And in that case, insofar as the liberal anti-perfectionist view is the view that there is such a fundamental privilege, it doesn’t ground any case against intuitions of intention-sensitivity by itself. We need some reason to think that the imperfect duty to work on oneself morally implies a permission to be vicious. Do you have such a reason that amounts to something different from and so capable of explaining the view that it’s sometimes permissible to be vicious in your sense?

Again, apologies if these questions betray misunderstandings that it's boring to correct, and do feel free to direct me elsewhere if so.

The comments to this entry are closed.

Ethics at PEA Soup

PPE at PEA Soup

Like PEA Soup

Search PEA Soup


  • Unless otherwise indicated, the views expressed in any given post reflect the opinion of only that individual who posted the particular entry or comment.