Moral demands

Most of what Norcross says is aimed at rejecting the use of cut-off points as exemplified in the all or nothing view. But Norcross suggests that he has a further and more radical thesis. He says in various places (including the title of the book) that he rejects the idea that morality issues demands at all. My aim in this essay is to elucidate and then reject this more radical proposal. It is not always clear that Norcross is aiming for a more radical claim. He often reverts to discussion of the all or nothing view. But in various places, he states a view that is independent of the all or nothing view.Footnote ² The radical view that I am aiming to elucidate is stated on p. 36: “Utilitarianism should not be seen as giving an account of right action, in the sense of an action demanded by morality, but only as giving an account of what states of affairs are good and which actions are better than which other possible alternatives and by how much. The fundamental moral fact about an action is how good it is relative to other available alternatives. Once a range of options has been evaluated in terms of goodness, all the morally relevant facts about those options have been discovered. There is no further fact of the form ‘x is right,’ ‘x is to-be-done,’ or ‘x is demanded by morality’” (p. 36).

How then, to further characterize the radical view? What does it mean to give up the notion of rightness as a demand? We can start with the thought that the distinction between the concepts of goodness and rightness is a matter of ‘deontic force’. A description of goodness has no deontic force at all. In that way, moral goodness is like any other kind of goodness. It can be observed and measured (at least roughly) with no implication for action. I might see a good racehorse, or a really good racehorse. I might see a good thief, or a good agave, a good camera, a good painting, a good dancer, a good dance, and so on. All of that talk of goodness makes sense, but none of it has any implications for how I should respond to that goodness: it does not have any implication for practical reasons. For that, we need a moral theory – a theory of what to do as opposed to a theory of value. Different moral theories give different stories about how we should respond to the values we encounter. Consequentialists think we should maximize value. Pluralists think that different sorts of value call for different responses. Deontologists think that there are things we ought to do regardless of consequences.

Norcross can accept what is important in this distinction between theories of goodness and theories of morality. According to Norcross, the essential claim for consequentialism, is that moral reasons to act correspond to the amount of value produced. The more value produced, the more moral reason there is do that act. On this understanding, a moral theory is primarily a theory of moral reasons, and all the deontic force we need is in the reasons.

What Norcross has in mind, I think, can be illustrated as follows. Imagine a doctor says to a patient, “…you have two options. You can opt for the treatment, which has various advantages. It will cure you of pain at least in the short term, it may even cure you permanently. It may have some other good effects. But it may not do much at all, and it will be hard on your body, and sometimes with this sort of thing there are knock on effects. It is hard to predict.” So, obviously, the patient says, “well what should I do?”. And the doctor says, “I cannot tell you that, you must make the decision.” Now, in this case, it is difficult to make a decision because it is not clear which option is better. But we can imagine that the doctor always says this. So, in a different case, she describes two options to her patient. She says, “…you can have the operation, which will remove the tumour. We have a 90% long term survival rate with this sort of operation. Or, you can opt for no treatment. In that case you will almost certainly die within a year. It is up to you.” In that case the doctor is taking the same stance – she is describing the reasons, but she is not telling the patient what to do. I think Norcross imagines morality as being like the doctor. It lays out the reasons, but does not tell us what to do. On Norcross's view, nothing is lost by seeing morality as being like the non-committal doctor.Footnote ³

It is worth stressing that one could give up the all or nothing view without giving up the imperatival model. One could embrace a grey area where it is not clear that an act is obligatory. One could also say that rightness in the imperatival sense comes in degrees. So, there are things that must be done, things that must not be done, and degrees in between, where the force of the imperative is weaker. This is not the way that philosophers generally use terms like ‘must’, but in every day life we do so frequently. Think for example of what I might say to my children – ‘you must do your homework tonight’. I mean it. But I intend it to have a lot less force than when I say to them, ‘you must not accept friend requests from strangers on the internet’. Or, to take an example of conscience: a person may say to themselves, ‘I must try harder to be kind to my difficult mother-in-law’. They mean it. But there is less deontic force there than when they say to themselves, ‘I must tell the truth about my colleague who I know to be sexually harassing students’. Some ‘musts’ are more serious than others.

Arguably, Mill's view, which Norcross quotes several times, exemplifies a scalar approach that retains the imperatival model. “The creed which accepts as the foundation of morals, Utility, or the Greatest Happiness Principle, holds that actions are right in proportion as they tend to promote happiness, wrong as they tend to produce the reverse of happiness.”Footnote ⁴

Norcross's defence of the no demands view

In section 2.7, Norcross addresses the worry that a morality that does not make demands cannot guide action, and in this section he characterizes and criticizes the ‘imperatival model’ directly. Norcross tries to explain away the attraction of the centrality of rightness and wrongness by saying that, first, we often think of very easy cases, where the choice is between a very bad action and good one. When the options are as described, it is easy to think that one option is right, and the other is not, in the sense that our attention is not drawn to the scale, to the possibility of a grey area. So the all or nothing view seems unproblematic. I agree with Norcross that this sort of example is misleading. But we can also see how a distinct issue, of whether a demand is being made, arises. When one option is bad and one good, it is easy to think that you ought to do one and not the other. And here, I think, the easy thought is correct. I do not agree that there is anything misleading going on: you really should do the good thing, and you really should not do the bad thing. Morality demands that you pick the better option. So Norcross's ‘explaining away’ of the all or nothing view succeeds, but he does not explain away the imperatival model.

Norcross goes on to focus more directly on the imperatival model, and the worry that his ‘reasons only’ account fails to be action guiding. “Morality is commonly thought of as some sort of guide to life. People look to morality to tell them what to do in various circumstances, and so they see it as issuing commands. When they obey these, they do the right thing, and when they disobey, they do a wrong thing. This is the form of some simple versions of divine command ethics and some other forms of deontology” (p. 41). Norcross rejects this model. Part of what he rejects, of course, is that it is all or nothing. But he also rejects the idea that there are demands at all.

Norcross first considers two rather weak arguments for the imperatival model. He suggests that one motivation for the imperatival model is that it is simple and easy to follow. Obviously, morality is not simple, and so this is not a good reason for adopting the imperatival model. I will not dwell on this point, as I do not think that simplicity is one of the reasons to adopt the imperatival model. Next, Norcross argues that although it may be useful to think of morality as making demands (he speculates that a society that works with the illusion that there are moral demands may be happier and more functional than one that does not), but he insists that morality does not make demands. Even if this is true, it would not show that morality actually makes demands.

A thornier issue is the question of action guidance. One may worry that a morality with no demands does not guide action. Norcross's view is that morality (utilitarianism) gives us reasons for action, and so it does guide action in the appropriate sense, but it does not tell us what to do. The notion of action guidance is tricky. If what we expect from a moral theory is that it is action guiding in the sense that it gives us a detailed instruction manual on how to do the right thing, we are certainly asking too much. But that is not the claim Norcross is defending, when he says that his view is action guiding. Rather, he is contrasting a theory that lays out the reasons with one that tells you to follow the strongest (or strong enough) reason. I am happy to concede to Norcross that there is no important difference that could be captured in terms of action guidance. I do think there is an important difference between merely laying out the reasons and telling the agent what to do. As Norcross himself thinks, it has something to do with blameworthiness.

I say more about blameworthiness below, but let me first criticize Norcross's own account of the role of blameworthiness in thinking about rightness and wrongness. Norcross considers two implausible accounts of the relationship between wrongness and blameworthiness. First, the idea that an action is wrong if it would be appropriate to impose sanctions on the agent. Norcross analyses this as, ‘we ought to impose sanctions’. This is obviously problematic as it uses the very notion (ought/obligatoriness) that we are trying to define. Next, Norcross considers the possibility that an action is wrong if it would be optimific to impose sanctions. Of course this is implausible: as Norcross, points out, it can be optimal to punish someone as a scapegoat, or as an example. The lesson here is that it is obvious that the notion of wrongness is more closely tied to the agent than the optimality of punishment is. We do not say that something is wrong unless the agent was connected to it in the right sort of way.

Norcross expresses this thought slightly differently. On p. 31 he offers two principles that ‘constrain our concept of wrongness’. His first principle of wrongness is as follows: “If action x is wrong, then an action y done by someone in exactly similar circumstances, with the same intention and the same consequences, is also wrong.” Norcross describes this as a ‘principle of universalizability’. But what this principle really captures is the thought that the intention is a necessary part of the wrongness.

For non-consequentialists, or for those who think that wrongness is ‘subjective’ (i.e. entirely dependent on the agent's point of view), the third part of this principle could be dropped. Any action done in (relevantly similar but not identical) circumstances with the same intention would be wrong, but we could vary the consequences. This is a familiar sort of example from discussions of moral luck. An agent who drives carelessly and does not kill a pedestrian has the same problematic state of mind as one who drives carelessly and does kill a pedestrian. Some people think that the agent acts equally wrongly in either scenario. Others argue that what happens as a lucky or unlucky result of an action affects the moral status of that action, and even the agent's blameworthiness. Interestingly, this argument is not usually made from a consequentialist standpoint, but from somewhere more sceptical of neat moral theories.Footnote ⁵ But we do not need to get into the discussion of whether we should include consequences in a definition of wrongness.

The point I want to focus on here is that the notion of wrongness is clearly related to the agent's state of mind. The optimality of blame or punishment may have nothing to do with the agent's state of mind, so clearly, Norcross is correct that we cannot analyse wrongness in terms of the optimality of blame or punishment. Norcross is also correct that intention plays a central role in the notion of wrongness. And of course, it is intention that gives us the link to blameworthiness. Looking at the optimality of blame or sanctions take us in the wrong direction: what we need to look for is an account of wrongness that connects it to blameworthiness – the aptness, or fittingness of blame – where the agent's intentions – the quality of her will – is what makes blame apt or fitting.

So if we are to give up on rightness and wrongness, we are also giving up on this moral appraisal that is focused on the agent's state of mind, and the praise- or blameworthiness that ensues. Norcross does not pursue this angle. In fact, having given us the first principle of wrongness that connects it to intention, he backs away from that in his summary, saying that, for a consequentialist, rightness and wrongness depend only on consequences: “It is at least part of the essence of any consequentialist view that the moral status of anything (whether an action, a character trait, an institution, or anything else) is entirely determined by the consequences of that thing itself. In particular, an action's moral status, compared with the status of alternative possible actions, is determined by comparing the consequences of that action with the consequences of the relevant alternatives” (p. 34).

To put it in the terms that are commonly used, this is a fully objective sense of rightness, according to which rightness has no connection to the agency of an agent. The contrast between objective and subjective forms of rightness is well illustrated by an example of Frank Jackson's:Footnote ⁶ Imagine a doctor has a choice between three drugs she could prescribe. The first drug will ameliorate the symptoms, but will not cure the patient. The doctor knows that one of the other two drugs will cure the patient completely, and the other one will kill the patient. Unfortunately, she does not know which is which. According to an objective sense of rightness, the right one to prescribe is the one that will actually cure the patient. According to a more subjective form of consequentialism, the right one to prescribe is the safe drug, even though the doctor knows that that would not be the best possible option. The most common argument for subjectivism about obligation is that rightness must be connected to responsibility.Footnote ⁷ The option that completely cures the agent is best – of course – but we need a connection to the agent. We need to take into account the agent's position, and to think about her intentions.

Despite that compelling argument, there are many defenders of an objective account of rightness and wrongness. Let us assume then, that Norcross is focusing on such an account when he says that we can do without the notion of rightness. So, on this interpretation of Norcross's radical claim, it is the claim that we do not need the objective senses of rightness and wrongness. But theories that use objective senses of rightness and wrongness are not making much of the notions of rightness and wrongness in the first place. In objective consequentialism, goodness and bestness are doing all the work. When we say that an act is right, we are simply saying that it is the best (or good enough, on a satisficing account). We are not genuinely saying that the agent ought to do it. Or at least, when we say that, we are saying it in a very attenuated sense: we are saying, it would be good if the agent did that. The doctor who is faced with a choice between a drug she knows will ameliorate the symptoms and a drug that has a 50% chance of killing the patient, should not choose the one that may kill the patient. In any ordinary sense of ought, the agent ought not to do what is objectively right. This, of course, is all familiar from debates about objective and subjective senses of right. My point here is that it is easy to get rid of the notion of ‘rightness’ in this sense. Objective theories of rightness and wrongness have already eliminated the sense in which morality makes demands. But this does not help us evaluate whether or not it is a good idea to eliminate the notions of rightness and wrongness.

Let us try again. What exactly is at stake in the disagreement over whether morality issues demands? Norcross gives an argument against a claim by Sidgwick here. Norcross argues that Sidgwick defends the claim that morality must ‘tell us what to do’ in the following passage: “Further, when I speak of the cognition or judgement that ‘X ought to be done’ – in the stricter ethical sense of the term ought – as a ‘dictate’ or ‘precept’ of reason to the persons to whom it relates, I imply that in rational beings as such this cognition gives an impulse or motive to action: though in human beings, of course, this is only one motive among others which are liable to conflict with it, and is not always – perhaps not usually – a predominant motive.”Footnote ⁸

As Norcross points out, Sidgwick is expressing a form of motivational internalism in this passage, i.e. Sidgwick is saying that that if you sincerely believe that something is right, you will be (if you are rational) motivated to do it. Norcross argues that whether or not we accept motivational internalism, Sidgwick's point here does not show that morality issues demands. Norcross argues that if we accept motivational internalism, then we should say that the judgment that a state of affairs is good is enough to activate motivation (we do not need the further judgment that it is right). And if motivational internalism is true, the utilitarian should simply say that the reasons are there, and we may or may not care about them. If we do care about those reasons, we will be motivated.

But this argument is beside the point. Norcross accepts that morality gives us reasons, and so of course he can accept motivational internalism or externalism about reasons and apply whichever view is independently more plausible to his own view. But the question was not, ‘is morality motivating?’; the question was ‘does morality issue demands?’. And that is a different question. A demand is a demand even if it is not motivating. Compare the law: a law is a sort of demand, not because of how it attaches to our motivation, but because of something inherent in its structure. Sidgwick's paragraph is not a good statement or defence of the view that morality makes demands.

Norcross considers and rejects two more reasons for preferring the imperatival model. I will not address the second reason – which is that we need to compare deontology to consequentialism in terms familiar to deontology – I agree with Norcross that that is not a valid consideration. However, the first worry Norcross describes seems to me to be important. This is the worry that his account collapses everything into a theory of the good, so there is nothing distinctive about a moral theory beyond the theory of value. Norcross attempts to resist that by appealing to reasons. On his view, utilitarianism tells us there is more reason to do some acts than there is to do others. But, Norcross insists, that does not mean that the better act is right. It does not even mean that we ought to do it! “The fact that there is a moral reason to perform some action, even that there is more moral reason to perform it than any other action, doesn't mean that one ought to perform it” (p. 44). The objection that eliminating rightness collapses everything into a theory of value is close, but not quite there. As Norcross argues, he can still talk about reasons. Again, we need to hear more about what we might have lost by saying that it is not true that you ought to perform the better act.

In defence of moral demands

In order to see what might be lost by jettisoning moral demands, let us look again at the non-committal doctor. Doctors do speak like this. They leave these choices up to the patient. But why is that? It is because the doctor does not have the right sort of authority to tell the patient what to do. If someone wants to die rather than undergo treatment, it is up to them. The patient has authority over her own life. So, the question is, does morality have the authority to tell us what to do? It obviously does not in any simple sense – it is not a person. Norcross's mention of divine command theory suggests that this is part of his thinking here: if God was telling us what to do, that would validate the imperatival model. But there is no-one to issue these imperatives, so there is no imperative. But that argument is too quick. There may be other ways that demands work.

Imagine that the doctor is talking to a parent, whose child is unconscious. The doctor explains that she can operate, and that that will almost certainly save the child's life (and give her a good quality of life thereafter), but if she does not operate, the child will almost certainly die. In this case, as in the previous case, the doctor does not have authority over the patient's life. In many medical contexts, parents do have that authority. But usually, when it is a life or death matter, the medical establishment can overrule the parent to save the child's life. And in a real case, the doctor in this situation will certainly overrule any objections the parent has and save the child's life. What is going on there? The doctor has legal authority. But the doctor's legal authority is not the end of the story. Rather, the explanation for her legal authority in this sort of case is that saving the child is the right thing to do. Morality demands it.

The point is that this goes beyond there being good reasons to save the child's life. We can see this by contrasting the case where there is only an adult patient. An adult patient in full possession of her faculties can refuse treatment, even if the reasons point to treatment. In Joshua Gert's terms, there may be justifying reasons here that are not requiring reasons.Footnote ⁹ But in the case where the patient is a child, we think that the reasons do more than point, they demand. The doctor must do what she can to save the life. The parent must not interfere. We do not think of the law that allows the medical establishment to overrule a parent on this as merely convenient or expedient. It is the morally right law to have.

In the case of an adult patient, the doctor can be non-committal because it is not her moral business. And notice that the adult patient may also lack any sense of demand when she examines her own reasons, even though she is the place the buck stops. The reasons in play here are prudential reasons, and assuming for the sake of argument that they have a different form to moral reasons, they do not command her. If she chooses to do what is in her own best interest, there may be no sense of ‘mustness’, of obligation, about it. It is just the choice she makes. But other choices she makes are not optional in this way. Some reasons command.

So we can see that the example of the non-committal doctor is very different to the case of morality. It is fine for the doctor to merely present the reasons and avoid telling her patients what to do. But, on the standard conception, there is at least a part of morality that does not work like that. There are cases where morality presents the reasons, and demands that we act in accordance with them. ‘Rightness’ is thus a genuinely ‘further fact’ here, and if Norcross wants to leave that fact out of his view, he is indeed saying something radical.

I take it that this establishes a sense for the idea that morality makes demands that are independent of the cut-off point issue in the all or nothing view. But it remains to be seen whether demands play an important role. Perhaps we can just abandon the idea, as Norcross suggests. I will briefly argue for an account of demands that links them to blameworthiness, and argue that this link is something we do not want to jettison.

What sort of thing is a moral demand? Legal demands come with a structure of enforcement. Enforcement may not always be applied, but it is nonetheless essential to the idea of law. According to a common and very plausible account of moral demands, they similarly come with a structure of enforcement. This is what Mill is saying when he says: “We do not call anything wrong, unless we mean to imply that a person ought to be punished in some way or other for doing it; if not by law, by the opinion of his fellow-creatures; if not by opinion, by the reproaches of his own conscience.”Footnote ¹⁰ The point here is not that something is wrong if we ought to punish it, or it would be optimific to punish it (as Norcross imagines the options). It is that there are demands such that failure to comply makes punishment fitting.

On this picture, moral reasons sometimes command. This is not inconsistent with a scalar account. For example: helping an old person cross the road when they are clearly struggling is obviously right. But, as Norcross points out, there may be a slightly less good option that it would be absurd to say was wrong – help them most of the way over and then let them make the last few yards on their own. And of course, as we go down the scale from goodness to badness it is clear that there comes a point where minimal praiseworthiness would shade into blameworthiness. It makes sense to say that there is a grey area here. And we can also say, that the action gets less right as we go down that scale – it gets less praiseworthy. And on the other side of the grey area the action gets more blameworthy as it gets more bad. This all needs more work, it is just a rough sketch.

The crucial point is that to be blameworthy, is to do something that you should not have done. Not just something you had reason not to do (that may nor may not be blameworthy depending on the context). Blameworthiness attaches to ignoring the commands of morality. Ignoring commands involves two ideas: the ignoring – which I talked about earlier: blame attaches to intentional actions – and the commands. The idea is that a command is defined by the structure of enforcement, just as legal commands are. Commands (or demands) are a special sort of presentation of reasons – they point out that these have a special status. Which is why it is blameworthy to fail to abide by these reasons. This, of course, is just to state our concept of obligation. And we can contrast reasons that do not command. So long as reasons do not command, one is not blameworthy for ignoring them. An adult who chooses not to have treatment is not blameworthy (aside from any obligations to others she may have).Footnote ¹¹

Can we do without that? If we do, we seem to have lost what is distinctive about morality. It is not so much that we have collapsed everything into a theory of the good – as Norcross points out, he can still talk about reasons. But collapsing everything into reasons loses something very important. We lose blameworthiness. An agent is only blameworthy for failing to abide by reasons if those reasons command. Assume, for example, that prudential reasons do not command. Then to say that someone is imprudent is to apply a merely descriptive term, and that term cannot properly be used as castigation. Not following a reason that we are not commanded to follow cannot be something we are criticizable for. Being imprudent (or irrational, or immoral on Norcross's view) is just like failing to abide by trivial reasons – for example, reasons related to culinary value. It is like eating cornflakes over a gourmet meal – if that is what someone choose to do, we can only shrug.Footnote ¹² But being immoral attracts blame.

I admit that in our everyday discourse we often use terms like ‘imprudent’, ‘irrational’ and so on as if they attach to disapprobation. There are various possible explanations. Maybe we are simply being nasty. Or perhaps we do think that reasons other than moral reasons sometimes command, or perhaps we think that there is an overall ought that commands, and the local judgments coincide with a global judgment of an all-things-considered ought, so that to be irrational on some occasion is to do the thing that was overall commanded against.

If, however, we are clear that there are no demands, we can only point out that a choice was immoral, and that term has no inherent censure. This is what Norcross takes to be the case at the fundamental level. However, he allows that we can use the notion of rightness in a contextualist way, as he explains in chapter 5. Again, it is sometimes hard to see if the contextualism applies to rightness in the sense of a cut-off point, or if it is supposed to apply to the imperatival use of rightness. It is easy to see how contextualism can make sense of reintroducing a non-fundamental sense of rightness as a cut-off point – the context affects where the cut-off point between permissibility and non-permissibility is. But that takes for granted that we are thinking in terms of demands. Norcross does not argue directly for the view that even though there are no demands at the fundamental level, it can make sense to talk about demands. This does not seem to be a matter of contextualism (it is not the case that whether there is a demand or not varies with context). Rather, this is a matter of speaking as if something is the case when it is in fact not the case. This is more like a form of error theory (one that endorses the error).

Take, for example, the following passage:

In this passage, Norcross clearly intends to refer to the notion of rightness in the imperatival sense. Furthermore, the implication of this passage is that he thinks this imperatival form can legitimately be used, even if it is not fundamental. Indeed, he expresses the view that it is obvious that it is required (i.e. ‘right’) to press a button to prevent a serious pain to others. Furthermore, it is very plausible that what he is saying here when he uses the term ‘required’ is that it would be blameworthy not to press the button. And that seems correct. It is obviously something that we find useful and sensible in our ordinary moral discourse.

So what advantage does Norcross gain from eliminating the imperatival sense of rightness from the view at the fundamental level? He is clearly comfortable with the idea that we use demands and blameworthiness in a healthy moral discourse. Norcross may point out that he can have whatever reactions he wants to people who act immorally. He can disapprove of them all he wants. He does not need an unwieldy Kantian idea of a categorical demand to license his blame. But the idea of rightness as a demand need not bring anything more ambitious than the idea of a reason. Norcross accepts that there are reasons. He does not seem very worried by the idea that some considerations count in favour of some actions. Surely that is the major hurdle. So why not accept our useful and ordinary conception of obligation and rightness, according to which some reasons correlate with demands and blameworthiness? This can be understood in a scalar way: we can say that there are degrees of demands.