Consequentialism, Rights Theory

Samaritanism and Duties

Singer’s famous example is taken to establish that some acts of Samaritanism are morally required. He writes, “if it is in our power to prevent something very bad from happening, without sacrificing anything morally significant, we ought, morally, to do it.”

There are three ways to interpret this claim:

1) Strong Duties: There are positive duties. This means it is morally impermissible pursue nonmoral projects instead of assisting others who are very needy. Chris says something like this when he points out that it seems that you would be morally required to throw your new TV in front of a runaway car in order to save a child.

2) Some Duties: Yes, there are positive duties. Samaritanism are morally required and it is impermissible to fail to help people. But we are not required to sacrifice our nonmoral projects to satisfy them in all cases. Sometimes, nonmoral projects can outweigh positive duties because unlike negative duties, positive duties are sensitive to costs.

3) No Duties: People have decisive moral reason to be good Samaritans (we ought, morally, to do it). But Samaritanism is not morally required. That is, it is permissible to pursue nonmoral projects even though you have decisive moral reason not to. This distinction appeals to moral categories like the supererogatory or the suberogatory, which make room for morally better and worse actions that are we are not required to do or avoid. Samaritanism is supererogatory, pursuing nonmoral projects is suberogatory in many cases.

All three of these options are compatible with the claim that you should give some or a lot of your income to charity. You should volunteer to tutor refugee children instead of going to yoga. You should rescue those in need. You should make the world a better place.

The first option seemingly classifies most people’s non-moral pursuits as morally impermissible. The theory seems committed to something like what Williams calls negative responsibility, the idea that we are responsible for the bad things we allow in the same sense that we are responsible for what we do. But we do not think that most people’s ordinary failures of assistance are wrong in the same way that it is wrong to injure or kill someone, even if both choices result in equally bad states of affairs.

The second option seems ad-hoc. If people in need have a claim to our assistance, why should it matter how costly it is to satisfy our duties to them? We do not say the same about negative duties. No matter how much I value a nonmoral project, I am not permitted to injure another person to achieve that project. So why would I be permitted to fail in my duties of assistance for the sake of my nonmoral projects?

The third option is off the table for most people. Matt writes, “The idea that individuals have no positive duties toward strangers is entirely implausible.” But is it? That idea does not deny that a person has decisive moral reason to help strangers. It just denies that helping is a requirement. Such a position can still capture the thought that you should help people, while also explaining why it is morally permissible to pursue nonmoral projects, even if they are not very important to us.

And here is another benefit of the third option. I think we should take accusations of moral impermissibility seriously. It means that someone has a claim, that her right has been violated, that she has been wronged by whomever acted impermissibly. People who act impermissibly are liable to be sanctioned. Blame is warranted (insofar as blame is a sanction), coercive or forceful interference to prevent impermissible conduct is permissible, and punishment can be justified as well.*

But it doesn’t seem like people are liable to be sanctioned at all for going to dinner at a restaurant, buying birthday gifts for their children, taking a yoga class instead of volunteering, or pursuing other morally insignificant projects even when it would not be costly to give them up. I am skeptical that it is morally impermissible (in this sense) for people to pursue non-moral projects, even though there is decisive moral reason to rescue others and donate to charity. For these reasons, I think the third option deserves a second look.



In some cases, a person could be liable to be sanctioned or punished because he did something impermissible, but it is not possible to permissibly sanction him in a way that is proportionate to the wrongful act. In these cases there should be legal ‘rights to do wrong,’ but the fact that an act shouldn’t be sanctioned or punished doesn’t make it permissible.
  • disqus_QZX8ENhLyb

    I think careful definitions are in order here. The words “permissible” and “impermissible” need to be fully explored [along with many others.]

    Both imply that someone or some “higher power” must be asked or requested to grant permission to perform some action. Hence, without that permission, there will be some means of preventing the supplicant from executing that action or should the supplicant execute the action in spite of not having received permission, there will be punitive consequences.

    Just EXACTLY what do “permissible” and “impermissible” mean in the context of this discussion?

    These are just two words that need to be carefully defined. Otherwise GIGO [Garbage In Garbage Out.]

    • Swami Cat

      QZ’s comment above is the only one that makes any sense to me at all. Where are these supposed DUTIES (or their cousins RIGHTS) coming from? You guys are just making them up. Some kind of weird collective fiction. This is why philosophy is such a mess.

      If by duty/right you mean something real like “sacrosanct conventions” (agreed upon intersubjective rules which are so important as to go without questioning), then say so, and we can have a real conversation.

      And for the record, I have a hard time envisioning a modern society flourishing by creating shared conventions that dictates that anyone not putting all their resources into the doing of good deeds for others should be shunned. Feel free to give it a try though.

      Sorry to be so negative, but like the commenter above states, it is critical to define the terms properly. I suspect human nature is such that we have evolved to treat sacrosanct conventions so seriously that most people actually manage to convince themselves that they are objectively something other than important conventions.

      Evolution isn’t just smarter than us, it’s smarter than Kant and Aristotle too.

      • Jessica Flanigan

        Yeah, sometimes I think this too about positive duties, that we are just using the words in different ways… You could just say that there are duties, but that unlike negative duties positive duties are sensitive to cost and weaker for distant people, unenforceable, potentially outweighed by non moral projects, and they do not set conditions of liability and sanction.

        So let’s call them duties* instead to capture the idea that they are very different from negative duties. You have decisive moral reason to satisfy your duties* but you are permitted not to in a range of cases and they are unenforceable. But then how would duties* differ from supererogatory action?

  • I found this discussion confusing, Jessica. Here is my attempt to clarify it.

    Some situations generate rights and duties. The drowning child that you could save easily generates a right in the child which imposes a duty on you. The right is not an enforceable right. It is like the right of a promisee. If you default on the corresponding duty, you do wrong, you behave impermissibly, but because the right is not enforceable, you are exempt from punishment and from claims of compensation. Still you are open to social sanction: you are a sonovabitch and people who know what you did (or did not do) are liable to form an adverse opinion of you and to alter the way they interact with you. You have a right to do wrong only in this sense: no one has the right to stop you (because the duty you are violating is not enforceable).

    So, your 2 is correct. What is wrong with your 1 (as I understand it) is that it requires people to sacrifice their lives for the sake of others. But as W. H. Auden said: if we are here to help others, what are the others here for? Moral duties have to leave people scope to live their own lives. So repeated situations of drowning children cease to generate the rights and duties of the one-off cases. That is why 2 is not ad hoc.

    Your 3 just seems to be incoherent. How can one have a DECISIVE moral reason to do something without it being impermissible not to do it? What could ‘decisive’ mean under such circumstances? It is permissible to sacrifice one’s life for others and, under some circumstances (like the soldier who throws himself on a grenade to save the lives of half-a-dozen comrades) it is admirable; but it is rarely morally required.

    • Jessica Flanigan

      When I say ‘decisive moral reason to A’ I mean that the moral reasons clearly weigh in favor of A’ing more than any alternatives B or C. So there may be things (B or C) that one ought not do, for moral reasons, but they are nevertheless morally permissible. For example, you ought not choose to go to yoga over tutoring refugee children, for moral reasons, but it is morally permissible for you to choose to go to yoga instead.

      Elizabeth Harman’s work has informed my thinking about this moral category. She calls these actions morally permissible moral mistakes. Categories like the supererogatory and suberogatory also capture an idea like this– that moral reasons could come apart from moral permissibility.

      • Okay, I get that. But I find the terminology misleading. ‘Decisive’ seems too strong for a moral reason that is not obligatory: it leaves you moral permission not to do A, so how can it be morally decisive? Similarly, I generally read ‘ought’ or, rather, ‘morally ought,’ as ‘morally required,’ i.e. obligatory. So, to say that one ought to do something which is not obligatory jars with me. But I know that ‘ought’ is used in all sorts of ways in ordinary talk.

    • jdkolassa

      I’ve never heard that Auden quote before but that is great.

  • Hi Jessica, I’m not sure I follow the case for No Duties. It sure seems like one is blameworthy for neglecting the drowning child. If you don’t think we’re at all blameworthy for neglecting more distant needs, doesn’t that suggest the “Some Duties” view?

    • Jessica Flanigan

      Hi Richard! Thanks for commenting.
      Blame can be lots of things, but the way I am thinking of blame here is as a kind of sanction, where the appropriateness conditions for blame are set by what we are responsible for. Blaming is a way of holding people responsible for what they do. And since I think people are not responsible for what they allow, I am reluctant to say that blame is appropriate in this sense.

      But blame can mean other things too. So some people say that blame isn’t about permissibly, it’s about the meaning of an action in the context of a relationship, like the relationship between members of the moral community. Understood in this way, I also feel the force of the idea that one is blameworthy in this case.

  • John Halstead

    Hi, this is an interesting post, but I think there a few problems.

    On ‘some duties’, I don’t think it is true that ‘No matter how much I value a nonmoral project, I am not permitted to injure another person to achieve that project.’. It seems like it depends on the costs involved. I am permitted to drive my car to the beach, even if this imposes some very small costs on others from air pollution. This comes back to Sobel’s arguments about libertarianism’s problems with pollution. This suggests you can’t really challenge the some duties view by appeal to negative and positive responsibility.

    On ‘No duties’. Firstly, many people deny the distinction between a moral requirement and having moral reason. This position cannot just be assumed to be false. Secondly, according to many philosophers, some acts may be impermissible but not blameworthy, and some acts are permissible but blameworthy. e.g. objectivists about rightness believe this. Moreover, it is pretty widely accepted that one may have a valid excuse for doing something impermissible.

    Thirdly, I’m not sure whether you’re saying that, as a conceptual matter, we may coercively prevent people from doing impermissible acts and may punish them (provided the other costs of coercion and punishment are low – from the * point at the bottom). Lots of people (Enoch, Waldron, GA Cohen etc) accept the moral – not just legal – right to do wrong, according to which we may not be prevented from doing something impermissible, even if the costs of prevention are low. I think it is at least a substantive, not conceptual truth, that people lack the moral right not to do their duty.

    If you mean to define morally impermissible acts as those for which people may be coerced and punished, then perhaps you are right. But to some extent you are talking past people who argue that it is morally impermissible not to donate to the poor. And these people seem to have the right end of the stick – it seems to me that they are talking purely about whether people have a duty to donate to the poor, putting to one side the issue of whether they may be forced to give to the poor, or punished for not doing so – this just seems like a different question altogether.

    Fourth, your arguments seem to imply the impermissibility of all redistributive taxation, including all government foreign aid. Redistributive taxation is most plausibly justified by samaritan duties: the reason our governments may tax us for foreign aid is that we have pre-existing duties to help the world’s poorest. So, given that you deny that people have these duties, you presumably deny that there should be any government foreign aid. This doesn’t entail that your position is false, but it does make it extremely controversial and, I suspect to most people, much less plausible.

    Finally, I think calling these things ‘nonmoral projects’, as nonconsequentialists are wont to do, gives many acts which people perform a rhetorically exalted status they do not deserve. The examples you have chosen seem quite wholesome. But consider: the boxer Adrien Broner filmed himself flushing hundreds of dollars down the toilet – It seems to me that he very obviously has an obligation to give this money to the poor instead. Lots of people drive around in very expensive cars that add almost nothing to their welfare. Do I think they have an obligation to use this money to save lives instead? Yes.

    • Jessica Flanigan

      Thanks John,

      Here are my tentative thoughts on your comments:

      I agree that cases of risk and small effects are difficult for a theory that gives negative rights against interference central importance. My read on the pollution case is that it will depend on the extent that people have a right against the risks of pollution, or whether the problems of pollution are over determined. I am also inclined to think that air pollution could be an impermissible violation of negative rights, which is why it would be warranted to expect that drivers compensate those who are harmed by their action. I also agree that one may have a valid excuse for acting impermissibly, but that doesn’t mean they were justified in doing it, or that they needn’t offer an apology or compensation if they did do something impermissible.

      On your third point, all I am committed to is that it is not permissible to do what is impermissible, though it may be more impermissible to prevent it (whether interpersonally or through legal means) A person who acts impermissibly may also be excused (fwiw, I suspect that ‘some duties’ may actually reflect the intuition that not helping is impermissible in all cases but sometimes excused). So I agree that the question of whether force is warranted against A’ing is a different question than whether A is impermissible.

      I addressed my view of redistributive taxation in a different post (Political
      Authority and the Basic Income) and I’m currently writing up some version of it so I don’t want to go too far down this road, but I think that redistributive policies could be justified without an appeal to Samaritan duties.

      And in any case, I would not take this as a reductio of “No Duties” even if “No Duties” did reject the legitimacy of a coercively enforced state-run property system. It seems to me far less controversial that taxation may not be justified than that there are no Samaritan duties. Philosophical anarchists, for example, can accept Samaritan duties and reject the legitimacy of these government programs. So if the admittedly counterintuitive position of “No Duties” affirmed a more widely held position such as “no taxation” such an implication would not necessarily undermine the plausibility of “No Duties.”

      I agree that Adrien Broner’s behavior morally indecent. According to “No Duties” he did something he had decisive moral reason not to do. What is gained by saying he acted impermissibly? Perhaps you think that Broner’s action was intrinsically blameworthy and that it would have been appropriate to sanction him for it, or to have tried to stop him. But interestingly, consequentialists are not committed to this position. On their account, blame is not necessarily warranted in cases where people fail to help when it’s easy. Blame would only be warranted when that instance of blaming (alone or as part of a broader practice) promoted good states of affairs. So while intuition that people like Broner ::just are:: blameworthy for what they do would be a strike against “No Duties” it would also be an objection to consequentialist justifications for Some Duties and Strong Duties.

      • John Halstead

        Hi Jessica, thanks for the response and sorry for the mammoth comment!

        Small risks and negative interference. Doesn’t your response concede the point that people are permitted to injure others for the sake of a nonmoral project? You just hold that people may injure and then compensate. This is different to not injuring at all.

        No duties. In the initial post you say “People who act impermissibly are liable to be sanctioned. Blame is warranted (insofar as blame is a sanction), coercive or forceful interference to prevent impermissible conduct is permissible, and punishment can be justified as well.” So, I think your response takes a different tack to the initial post.

        I agree with you that since no redistributive taxation is less controversial than no duties, pointing out that your position possibly implies no redistributive taxation, doesn’t count much (or probably all) against your view.

        However, I think that the denial of govt redistribution is a reductio and that the acceptance of no personal duties is a reductio. State funded foreign aid has many notable successes, including the eradication of smallpox, the near eradication of guinea worm, the massive reduction in diarrhoeal disease etc. It has saved probably hundreds of millions of lives for not very much money. To me, a theory which says that this ought not to have happened (in the way it did) because it involved coercion of rich people has gone very very badly wrong.

        My point about Adrien Broner lost focus. Really I just wanted to say – this is just as much a ‘nonmoral project’ as the other things nonconsequentialists talk about. For a lot of people I know ‘projects’ apparently includes spending upwards of £1000 on a handbag. I think nonconsequentialists gain a rhetorical advantage from getting to call these things ‘projects’ when in another age, much (perhaps most) first world consumption would have been put in the ‘decadent selfishness’ category.

        Ultimately, if you want to call it a decisive moral reason rather than a duty (which you may tie to blame, coercion and punishment), then I have no substantive moral beef, but only a semantic one (which is nevertheless important). You think that there is a difference between establishing that someone has a decisive moral reason not to do x and establishing blameworthiness, coercability and punishability. Which is fine, by your definitions. But certainly, I think it is important to point out that most philosophers think there is a gap between establishing the lack of a permission to do x, and the right of the state to coerce that person to do x.

        Final note on blame at the end. I think your point elides a person’s blameworthiness with whether we ought to say that someone is blameworthy. On consequentialism Broner is blameworthy for his general dickheadery. But whether we ought to say that he is or ought to punish him is a different issue which depends on the consequences of blaming and punishing (Roy Sorensen’s ‘Unknowable Obligations’ is good on this).

  • geoih

    Define bad.

  • Theresa Klein
  • Pingback: Samaritans and duties - The Indian Economist()

  • Pingback: Samaritanism is an act of morality than an obligation()