The term ‘supererogatory’ is a technical term that one can use however one wants. But I'll use it such that an act counts as supererogatory if and only if, in performing it, the agent exceeds some morally required minimum. As such, my use of the term coincides with such colloquial expressions as ‘going the extra mile’, ‘doing more than one has to’, and, most illuminatingly, ‘going beyond the call of duty’.Footnote 1
In this article, I'll consider two competing accounts of supererogatory acts so conceived: the standard account and the maximalist account. I'll argue that the maximalist account is the more plausible of the two, because it avoids two serious problems associated with the standard account: the latitude problem and the transitivity problem.
I take the standard account to be as follows.
The Standard Account: (1) For any subject S and any option X, S's X-ing is supererogatory if and only if there is an alternative option Y such that: (a) she is both morally permitted to X and morally permitted to Y, and (b) she has more moral reason to X than to Y. (2) For any subject S and any two alternative options X and Y, S has more moral reason to X than to Y if and only if she morally ought to prefer how things would be if she were to X to how things would be if she were to Y.Footnote 2
One problem with the standard account (call it the latitude problem) is that it has counterintuitive implications in cases where the duty to be exceeded is one that allows for significant latitude in how to comply with it. To illustrate this sort of duty and the problem it generates for the standard account, consider the following two examples.
The Miser: Mr Scrooge lived from 1920 to 1995. In all that time, he never did a thing for anyone except on Christmas Eve 1938, which is when he found a five-dollar bill on the ground and donated it to the Salvation Army. Moreover, had he not done this, he would have ended up never having done anything for anyone.
The Visit: Carlos's grandmother was in the hospital last week. And although he had promised that he would visit her at least twice during the week, he ended up visiting her only once: on Tuesday. Moreover, had he not visited her then, he would not have visited her at all, for things only got busier for him as the week progressed (although it never got so busy that he was unable to visit his grandmother).
In both cases, the agent has significant latitude in how to comply with the relevant duty. In The Miser, the relevant duty is the imperfect duty of beneficence. This duty requires each of us ‘to make the happiness of others a serious, major, continually relevant, life-shaping end’.Footnote 3 And this in turn requires each of us to act throughout our adult lives so as to exhibit a sufficient propensity to promote the happiness of others – sufficient, that is, to count as having adopted the happiness of others as ‘a serious, major, continually relevant, life-shaping end’. For having such a propensity is constitutive of having adopted the happiness of others as such an end. Thus, if someone with the abilities and opportunities that are typical of those living in the First World did little or nothing to promote the happiness of others, this would show that she had not truly adopted the happiness of others as such an end.Footnote 4
But as this is an imperfect duty, it gives us a lot of leeway as to how and when to promote this end. For having a propensity to promote the happiness of others does not necessitate that we take advantage of every favourable opportunity to do so. It requires only that we do so to some sufficient extent.Footnote 5 Thus, Mr Scrooge was not obligated to donate that five-dollar bill to charity on Christmas Eve 1938. He was permitted instead to spend it on himself. After all, he was only eighteen years old at the time, and there were to be plenty of future opportunities for him to promote the happiness of others. That said, it was good that he donated it, as he did have more moral reason to donate it to charity than to spend it on himself. After all, he morally ought to prefer the way the things were as a result of his donating it to how they would have been had he spent it on himself and never did anything for anyone. Still, he was permitted to forego this opportunity to promote the happiness of others and to choose instead to do so to a greater extent in the future. So, in donating that five-dollar bill, Mr Scrooge performed a morally optional act that he had more moral reason to perform than to perform the morally permissible alternative of spending that five-dollar bill on himself. Thus, on the standard account, his donating that five-dollar bill counts as supererogatory.
Now, consider The Visit. Here, the relevant duty was for Carlos to visit his grandmother at least twice during the week. And this duty allowed for significant latitude in how he was to comply with it. For instance, he wasn't required to visit his grandmother on any particular day that week. He was just required to visit her at least twice. Thus, he was not obligated to visit her when he did – that is, on Tuesday. For he could have fulfilled his duty by visiting her on, say, Wednesday and Thursday, Thursday and Friday, or Wednesday and Friday. Nevertheless, it was good that he visited her on Tuesday. For he had more moral reason to visit her on Tuesday than not to do so. That is, he morally ought to have preferred the way things were as a result of his visiting her on Tuesday to the way things would have been had he not done so and, consequently, ended up not visiting her at all. So, in visiting his grandmother on Tuesday, Carlos performed a morally optional act that he had more moral reason to perform than to perform the morally permissible alternative of not visiting her on that day. And, so, the standard account implies that his visiting his grandmother that Tuesday was supererogatory.
The problem, then, is that this is implausible. The standard account implies both that (1) in donating that five-dollar bill to charity on Christmas Eve 1938, Mr Scrooge went beyond the call of duty and performed a supererogatory act and that (2) in visiting his grandmother on Tuesday, Carlos went beyond the call of duty and performed a supererogatory act. Yet, to the contrary, it seems that not only did they each fail to exceed the minimum that morality required of them, but that they each fell well short of that required minimum. In other words, the problem with the standard account is that one can meet its conditions 1a and 1b without exceeding the minimum that is demanded by the relevant duty.Footnote 6 Take the duty of beneficence. Since there is almost always more moral reason to act beneficently than to act self-interestedly, and since this duty almost never requires one to perform any specific beneficent act, an agent can easily meet conditions 1a and 1b without exceeding the minimum required by the duty of beneficence. To illustrate, suppose that for someone with Mr Scrooge's abilities and opportunities to exceed the minimum that the duty of beneficence requires, he would have to dedicate more than 20% of his time and resources to promoting the happiness of others over the course of his adult life. The problem, then, is that Mr Scrooge was able to meet conditions 1a and 1b in donating that five-dollar bill without coming anywhere close to meeting this moral minimum.
So if we're going to avoid the latitude problem, we'll need to adopt an account that, unlike the standard account, acknowledges that one can meet or exceed the call of such a duty only by performing a certain series of actions over time. Moreover, we'll need to adopt an account that acknowledges that in order for such a series of actions to count as supererogatory, it must more than just partially or minimally fulfil the relevant duty. Thus, we'll need to adopt what I call the maximalist account of supererogation.
On the maximalist account, we are to distinguish between two types of options: maximally specific options and non-maximally specific options. An option X is a maximally specific option if and only if there is no other option that entails it – that is, if and only if there is no option Y such that Y-ing entails X-ing but not vice versa.Footnote 7 Otherwise, it's a non-maximally specific option. And an option X entails another Y if and only if X-ing without Y-ing isn't an option. To illustrate, I have both the option of walking and the option of walking in the park, and my walking in the park entails my walking, for I don't have the option of walking in the park without walking Footnote 8 But walking in the park at t1 won't be a maximally specific option if walking fast in the park at t1 is an option, because walking fast in the park at t1 entails walking in the park at t1 . Likewise, walking fast in the park at t1 won't be a maximally specific option if walking fast in the park at t1 and then resting at t2 is an option. And walking fast in the park at t1 and then resting at t2 won't be a maximally specific option if walking fast in the park at t1 and then resting on a bench at t2 is an option. And so on and so forth until we get to an option that is not entailed by any other option – that is, a maximally specific option.Footnote 9
We must distinguish between these two types of options, because, on the maximalist account, they are to be assessed differently.
The Maximalist Account: (1) For any subject S and maximally specific option M x , S's M x -ing is supererogatory if and only if there is an alternative maximally specific option M y such that: (a) she is morally permitted both to M x and to M y , and (b) she has more moral reason to M x than to M y . (2) For any subject S and any non-maximally specific option N x , S's N x -ing is supererogatory if and only if both: (a) in N x -ing, she does not merely minimally or merely partially fulfil some positive duty D; and (b) there is an alternative non-maximally specific option N y such that: (i) she is morally permitted both to N x and to N y , and (ii) she has more moral reason to N x than to N y . (3) For any subject S and any two maximally specific options M x and M y , S has more moral reason to M x than to M y if and only if she morally ought to prefer how things would be if she were to M x to how things would be if she were to M y . (4) For any subject S and any two non-maximally specific options N x and N y , S has more moral reason to N x than to N y if and only if she morally ought to prefer how things would be if she were to perform an optimal maximally specific option that entails her N x -ing to how things would be if she were to perform an optimal maximally specific option that entails her N y -ing.Footnote 10
Now, when it comes to maximally specific options, the maximalist account and the standard account converge on the same verdicts. The two diverge only when it comes to non-maximally specific options. To illustrate, take Carlos's non-maximally specific option of visiting his grandmother on Tuesday. Whereas the standard account holds that this option is supererogatory given that he had more moral reason to perform it than to perform the permissible alternative of not visiting her that Tuesday, the maximalist account holds that this was not supererogatory given that, in performing this option, he merely partially fulfilled his duty to visit her at least twice that week. Let me explain.
In X-ing, S merely partially fulfils a positive duty D if and only if her X-ing is a proper subset of some set of actions by which she merely minimally fulfils D. And, in X-ing, she merely minimally fulfils a positive duty D if and only if, although her X-ing fulfils D, there is no more moral reason for her to X than to perform any other option that would fulfil D. Now, Carlos has several ways of merely minimally fulfilling his duty to visit his grandmother at least twice this week: (W1) visit her on Monday and Tuesday, (W2) visit her on Tuesday and Wednesday, (W3) visit her on Monday and Wednesday, (W4) visit her on Wednesday and Thursday, etc. These are all ways of merely minimally fulfilling his duty, because, although they each fulfil his duty, he has no more moral reason to perform any one of them than to perform any other alternative that would fulfil his duty. Thus, visiting his grandmother on Tuesday counts only as partially fulfilling his duty to visit her at least twice this week. For, in visiting her on Tuesday, he performs only a proper subset of a set of acts that would merely minimally fulfil his duty, such as the set consisting of his visiting her only on Tuesday and Wednesday.
To exceed the minimum required by this duty, Carlos would have to do something that he has more moral reason to do than to just visit her twice that week – e.g. visit her three or more times. Thus, on the maximalist account, visiting her only on Tuesday would not count as supererogatory, but visiting her on Tuesday, Wednesday and Thursday would. The latter would count as a supererogatory, because Carlos is morally permitted both to (N1) visit her on Tuesday, Wednesday and Thursday and to (N2) visit her on only Tuesday and Wednesday, and Carlos has more moral reason to N1 than to N2. He has more moral reason to N1 than to N2, because he morally ought to prefer how things would be if he were to perform an optimal maximally specific option that entails his N1-ing to how things would be if he were to perform an optimal maximally specific option that entails her N2-ing, where M x is an optimal maximally specific option for S if and only if M x is a maximally specific option for S and there is no alternative maximally specific option for her, M y , such that she has more moral reason to M y than to M x . The idea, then, is that an optimal maximally specific option that entails his N2-ing isn't going to be as morally good as an optimal maximally specific option that entails his N1-ing. And, therefore, he has more moral reason to N1, making his N1-ing supererogatory.
So it seems that, unlike the standard account, the maximalist account is immune to the latitude problem. For it has plausible verdicts in cases such as The Visit, where the duty to be exceeded is one that allows for significant latitude in how to comply with it. This gives us at least one reason to prefer the maximalist account. But there is another reason to prefer the maximalist account: it is immune to another problem facing the standard account – viz. the transitivity problem.Footnote 11 This problem has been presented by both Frances Kamm and Dale Dorsey, but I'll focus on Dorsey's formulation of the problem.Footnote 12 The problem arises with respect to the following sort of triad of cases.
‘Gus: Gus finds himself the recipient of an inheritance from a wealthy relative. This inheritance will allow Gus to buy a new car, which Gus desires to do, and which will allow Gus to see much more of his significant other, who lives in a distant town. Alternatively, Gus could [use this inheritance to] . . . save ten people from death’.Footnote 13
Roy: Roy could save ten people from death, but only by assaulting Jerry.
‘Stan: Stan knows that Jerry has just inherited a substantial amount of money from a wealthy relative. Were Stan to intimidate Jerry into giving him the money as a result of beating Jerry up, this would be a prudential benefit to Stan, given that this would allow him to buy a new car, from which he will derive pleasure, and which will allow Stan to see much more of his significant other, who lives in a distant town. (Assume that Stan would avoid punishment.)’Footnote 14
The problem is that the following four claims concerning this triad each seem individually plausible but are jointly inconsistent.
(C1) The self-interested reason there is to buy a new car for oneself morally justifies acting against the moral reason that there is to save ten people from death. Thus, Gus is permitted to buy a new car for himself, and saving the ten is supererogatory, not obligatory.
(C2) The moral reason there is to save ten people from death morally justifies acting against the moral reason there is to refrain from assaulting someone. Thus, Roy is permitted to assault Jerry.
(C3) The self-interested reason there is to buy a new car for oneself does not morally justify acting against the moral reason there is to refrain from assaulting someone. Thus, Stan is not permitted to assault Jerry.
(C4) The one-reason-morally-justifies-acting-against-another relation is transitive.
C4 is difficult to deny.Footnote 15 And yet each of C1–C3 also seems plausible, and they jointly entail that C4 is false. So we have a paradox. I believe the solution to it lies in realizing that, depending on how we spell out the details of the three cases, we should reject either C1 or C2, which we can do if, and only if, we accept the maximalist account of when there is more moral reason to perform one non-maximally specific option than to perform another. Thus, the solution lies both with further specifying the details of the three cases and with adopting the maximalist as opposed to the standard account of supererogation.
There are at least two ways of further specifying the details in the three cases. One way (I'll call it the strangers way) is to substitute ‘save ten distant strangers from death’ for ‘save ten people from death’ throughout the descriptions of the three cases. The other way (I'll call it the relatives way) is to substitute ‘save ten close relatives from death’ for ‘save ten people from death’ throughout the descriptions of the three cases. I'll consider them in turn.
To start, then, let's take the strangers way of further specifying the details. When we do so, we get the following versions of the first two cases. (The third case remains the same, since it makes no mention of saving ten people from death.)
‘Gus[v1]: Gus finds himself the recipient of an inheritance from a wealthy relative. This inheritance will allow Gus to buy a new car, which Gus desires to do, and which will allow Gus to see much more of his significant other, who lives in a distant town. Alternatively, Gus could donate his inheritance to Oxfam International, which would save ten [distant strangers] . . . from death’.Footnote 16
Roy(v1): Roy knows that Jerry has just inherited a substantial amount of money from a wealthy relative. Were Roy to intimidate Jerry into giving him the money as a result of beating him up, Roy could donate the ill-gotten money to Oxfam International, which would save ten distant strangers from death. If Roy doesn't beat Jerry up, Jerry is going to use it to buy himself a new car, and those ten people that Roy could have saved will die. Everything else, though, will be the same. Thus, we are to assume that Roy would avoid punishment even if he were to assault Jerry.
If this is how we are to think of the three cases, then we should, I believe, reject C2, which says: ‘The moral reason there is to save ten people from death morally justifies acting against the moral reason there is to refrain from assaulting someone. Thus, Roy is permitted to assault Jerry.’Footnote 17 Admittedly, if we were to adopt the standard account, it would be difficult, if not impossible, to deny C2. For, on the standard account, it would seem that Roy has more moral reason to assault Jerry than to refrain from assaulting Jerry. After all, it seems that he morally ought to prefer, other things being equal (and they are stipulated to be equal), the world in which he assaults Jerry and thereby saves ten distant strangers to the world in which he refrains from assaulting Jerry and fails to save those ten distant strangers.
But if we accept the maximalist account, we can, and should, deny that Roy has more moral reason to assault Jerry than to refrain from assaulting Jerry. To see this, compare how things would be if Roy were to perform an optimal maximally specific option that entails his assaulting Jerry to how things would be if he were to perform an optimal maximally specific option that entails his refraining from assaulting Jerry. And let's suppose (plausibly, I believe) that any optimal maximally specific option will strike a balance between the use of one's limited time and resources to promote the happiness of others and the use of one's limited time and resources to promote (or respect) various other moral goals, such as justice, beneficence, self-improvement, and caring for one's family. That is, let's assume that an optimal maximally specific option will not be one where one dedicates all of one's time and resources just to promoting the happiness of others and, so, at the expense of neglecting other important moral ends, such as justice, self-improvement, and the well-being of one's family. Given this, Roy will not be morally required to save as many distant strangers as he could save if he were to set aside these other important moral goals. After all, we could all save more distant strangers if we were to disregard the well-being of our families, disregard our own self-improvement, and disregard justice by, say, robbing the rich to give to the poor. Thus, it seems that Roy will be morally required to save the lives of no more than n distant strangers over the course of his life, where n is significantly less than the number of distant strangers that he could save were he to disregard all other moral goals.
Admittedly, then, Roy could save more than n strangers if he were to some extent to neglect these other moral goals. But it would not be morally better for him to do so. After all, to exceed n is, by definition, to fail to strike an acceptable balance between the goal of saving more distant strangers and the goal of doing more to promote (or respect) various other moral ends. Moreover, if Roy were to insist on saving, say, n + 10 strangers, the best way for him to do so would not be by his setting aside the moral goal of respecting justice and assaulting Jerry, but instead by his setting aside some prudential goal (e.g. the goal of taking more overseas vacations) or some less important moral goal (e.g. the goal of improving himself a bit more). It seems, then, that if Roy were to insist on saving n + 10, it would be better for him to do so by using less of his time and resources on either his self-improvement or his overseas vacations than by obtaining additional resources through the assault of Jerry.Footnote 18 And, if all this is right, we should not think that Roy morally ought to prefer how things would be if he were to perform an optimal maximally specific option that entails his assaulting Jerry (one in which he saves the lives of n + 10 strangers over the course of his life and assaults Jerry) to how things would be if he were to perform an optimal maximally specific option that entails his refraining from assaulting Jerry (one in which he saves the lives of only n strangers over the course of his life and assaults no one). And, so, we should think that Roy does not have more moral reason to assault Jerry than to refrain from assaulting Jerry. Thus, we should deny that the moral reason that Roy has to save these ten strangers morally justifies his acting against the moral reason he has to refrain from assaulting Jerry. And, so, we should deny C2 – or, at least, we should so long as we adopt the maximalist account and take the strangers way of further specifying the details in these three cases.
But what if we were instead to take the relatives way of further specifying the details? In that case, we would get the following different versions of the first two cases.
‘Gus[v2]: Gus finds himself the recipient of an inheritance from a wealthy relative. This inheritance will allow Gus to buy a new car, which Gus desires to do, and which will allow Gus to see much more of his significant other, who lives in a distant town’.Footnote 19 But ten of Gus's close relatives have just been infected with a rare virus. And, unfortunately, these ten cannot afford the antiviral drugs that they need to survive. Of course, if Gus were to act altruistically, Gus would use his inheritance to purchase the needed antiviral drugs and save his ten relatives. But if he doesn't, they'll die.
Roy(v2): Ten of Roy's close relatives have just been infected with a rare virus. Unfortunately, neither Roy nor these ten can afford the antiviral drugs they need to survive. But Roy knows that Jerry has just inherited a substantial amount of money from a wealthy relative. Were Roy to intimidate Jerry into giving him the money as a result of beating Jerry up, Roy would use the money to purchase the needed antiviral drugs and save his ten relatives. If Roy doesn't beat Jerry up, Jerry is going to use it to buy himself a new car and Roy's ten relatives will die. Everything else, though, will be the same. Thus, we are to assume that Roy would avoid punishment even if he were to assault Jerry.
And if this is how we are to think of the three cases, then we should, I believe, reject C1, which says: ‘The self-interested reason there is to buy a new car for oneself morally justifies acting against the moral reason that there is to save ten people [in this case, ten close relatives] from death. Thus, Gus is permitted to buy a new car for himself, and saving the ten is supererogatory, not obligatory.’ It seems to me that, in Gus[v2], the relevant duty is the duty to take any decent opportunity to save the life of a close relative. And there is no way for Gus to fulfil this duty without using his inheritance to save his ten relatives. Thus, Gus's saving the ten, in Gus[v2], is not supererogatory, but obligatory. This contrasts with Gus[v1], where the relevant duty seems to be the imperfect duty of beneficence – the duty to adopt the happiness of others as an end and, thus, to act throughout one's adult life so as to exhibit a sufficient propensity to promote this end. And Gus needn't use his inheritance to save these ten particular strangers to fulfil this duty. For it doesn't matter, morally speaking, whether he forgoes this opportunity to benefit himself and saves these ten strangers or takes this opportunity to benefit himself but forgoes some later opportunity to benefit himself so as to save some other ten strangers. Either way, he will exhibit the same propensity to promote the happiness of others. But because, in Gus[v2], the relevant duty is not the imperfect duty of beneficence but the duty to take any decent opportunity to save the life a close relative, we must deny that Gus is permitted to use his inheritance to buy a new car for himself rather than use it to save his ten relatives. Thus, we should reject C1 if we take the relatives way of further specifying the details of these three cases.
Either way, then, we have a solution to the transitivity problem. For no matter how we further specify the details of these three cases, saving the ten will either be an instance of taking one of several alternative paths to partially fulfilling an imperfect duty or it won't. If it is, as in the case of the strangers way of filling in the details, it will be plausible to deny C2 – or, at least, it will so long as we adopt the maximalist account of supererogation. And, if it isn't, as in the case of the relatives way of filling in the details, it will be plausible to deny C1. So, either way, we resolve the transitivity problem if we adopt the maximalist account.
To sum up, then, we seem to have two good reasons for favouring the maximalist account over the standard account, for only it is immune to both the latitude problem and the transitivity problem.Footnote 20
