The grenade-jumper

Urmson (Reference Urmson and Melden1958 p. 202) offers the drastic example of a soldier who throws himself on a live grenade, thereby fatally absorbing the blast with his body, but saving the lives of his nearby comrades. One's first reaction to such an example would have to be wonder at the soldier's sheer spontaneous goodness and selfless concern. Certainly the quantity of sacrifice involved tempts one to accord it the highest praise. However, upon further reflection there are a number of factors that complicate the example, and while such factors may not reduce the goodness of the act, they may undermine its claim to supererogatory status.

In one sense, clearly the soldier has paid the highest cost. In another sense, however, it is not clear that what he has incurred is a cost, given that he does not live to experience it as a cost. Real costs always involve some sort of comparison, both before and after choice. If I am asked to coach the girls’ football team, then I weigh the effort and time of my weekends against the other activities I might need or like to do on those same weekends. If I agree to the coaching, then I will go on to ‘feel’ the opportunity costs whenever I lack the time or energy to fix the roof or head out to the pub. Essential to the revealed felt cost of the coaching is the possibility of regretting the choice, and a new deliberation about whether to reduce or reject my commitment.

Now the grenade-jumper story described by Urmson is ambiguous; it is not clear whether the jumper believes that he would be killed by the grenade anyway. If he does believe this, then he also believes he has nothing to lose and he may as well try to save his comrades. Without a thought to costs, it is no longer clear that the act was supererogatory. But let's assume that the soldier believed he could save himself, and so he really is sacrificing his life for his comrades. This would seem to involve the notion of cost in comparing alternative futures, but of course it would not allow for post hoc felt costs, let alone the possibility of regret. In addition to that, there is the old philosophical question of whether death is a sufficiently meaningful alternative to be even contemplated, since it is unlike anything we have ever experienced. In some rare situations of extraordinary pain or despair, anything might be preferable to continued life; but the jumper did not have that problem. As such it is not clear that he could make a meaningful enough comparison between the two alternative futures for his act to be admirably supererogatory in the full sense.

In fact there probably was no time for comparison or for deliberation. The grenade-jumping example is thus also potentially misleading in comprising a spontaneous act, and this starts to undermine the freedom which I took to be necessary for the act to be supererogatory and admirable. We can still admire the man for doing it, but here we are admiring him for being the type of person to react in this particular spontaneous way, rather than admiring the chosen act. All in all, I am not sure the example is as useful as Urmson thinks it is. As it turns out, I will soon be challenging the freedom criterion even in cases with enough time for deliberation.

The Good Samaritan

According to Luke (10:29–37), a Jew had been robbed and lay wounded at the side of the road. Both a priest and a Levite passed by on the other side, but a Samaritan stopped to help him. Not only did he tend his wounds and accompany him to the nearby town, he then paid a hotelkeeper to allow the Jew to stay there and recover. I have distinguished these two phases of the Samaritan's actions because, as Heyd points out (Reference Heyd1982 p. 17), one could plausibly see the provision of emergency assistance to a proximate and obviously needy victim at manageable cost to oneself as a matter of obligation rather than supererogation, and so the real supererogatory act was the subsequent payment for the hotel room. (On this understanding, the priest's and the Levite's omission would therefore become blameworthy neglect of an objective ‘duty of rescue’.)

One might avoid Heyd's redescription by adjusting the example so that the cost and risk of intervention become higher, and the victim becomes less proximate. Let's say, for example, that the Jew is crying for help in a turbulent river. (The Priest and the Levite would then be blameless for not jumping in – although they would still have a duty to call for help.) With this adjustment, however, it is still philosophically interesting to distinguish two new phases: the Samaritan's spontaneous supererogatory decision to jump in the river, followed by his deliberative supererogatory decision to then take the soaking wet, penniless Samaritan to a hotel.

Now we have the same problem with spontaneity as we did with the grenade-jumper. The Samaritan's spontaneous jump might seem like an unfree gut response to the situation, and therefore less praiseworthy – at least as an act; (but the Samaritan would remain praiseworthy as an individual with such good dispositions.) Unlike the grenade-jumper, however, the spontaneous water-jumping Samaritan survives to experience the cost to his own comfort and projects.

In addition, it is also revealing to imagine how he would answer a question from a CNN reporter about his heroic jump in the river. One plausible answer would be “I couldn't just leave him to drown.” I think it is philosophically important to take this self-description seriously, that is, not as a matter of false modesty. From within the agent's perspective, this is a statement of a genuine moral incapacity (“I couldn't”) – I am using Bernard Williams's termFootnote ² – and this would lend further support to the thought that this was indeed not a free decision. And so we must conclude that either freedom is not a necessary condition for supererogatory action, or that the action in question is supererogatory in a different, perhaps lesser way. As Archer points out in his contribution to this volume (Ch. 6), such an incapacity undermines the claim that the Samaritan was actually sacrificing something. For a sacrifice, in order to be my sacrifice rather than an unchosen experience of hardship, has to involve a choice, motivated by the good, between alternatives with a view to perceived relative costs. Yet the Samaritan had no appreciation of the costs either before or after the rescue.

What is interesting about the reporter's interview with the Samaritan, however, is that he might very well refer to the same incapacity when describing the second phase of action as well. Recall that this second phase involved deliberation and decision, and thus seems like a better candidate for an essentially free supererogatory action. But the Samaritan might say: “Once I'd rescued him I couldn't just abandon him.” Again, I claim that these words can be taken at their face value. We have to distinguish between physical and available options. When I am standing at the top of a cliff, jumping to my death is a physical option but, unless I'm suicidal, it is not an available option – that is, available to me, within my perspective. You might point out that I have the physical option, and I wouldn't deny it, but that by itself would not persuade me to contemplate the physical option as an available option. So even if the Samaritan had the physical option to abandon the Jew after rescuing him, he did not contemplate such an option seriously, and so effectively he did not have that option. On the other hand, we could imagine another adjustment to the example whereby the Jew, once on land, says “Thanks very much, I'll be fine now, I've got a friend who lives just down the road,” and only then does the Samaritan become aware of the physical options he had all along.

So I would conclude that the element of choice is not essential to a supererogatory act. There is one last comment worth making about the Good Samaritan. Let us assume that he rescued the Jew from the turbulent waters, and handed him over to a waiting ambulance at the side of the river. Now if he explains to the CNN reporter that “he couldn't just leave him there,” he is being careful to make a statement only about himself, with no implications for what anybody else ought to have done. However, another kind of Samaritan might have said something pointing in a different direction: “I did what anyone else would do.” Again, assuming that he is speaking sincerely, what are we to make of this? On one level, he must know that other people would not do what he did – after all, he saw the priest and the Levite walk by without jumping in. But this does not necessarily falsify his modesty. It might be that, linked to his heroically good disposition, is an equally heroic good disposition to think well of others. In this case, he might assume that the priest and the Levite did not jump in because they had very good reasons for not doing so (they couldn't swim, they were on their way to something much more important etc.); or he might be assuming that in the future, most people would in fact jump in to save the Jew in just such circumstances, regardless of their performance to date. I will have more to say later on about such supererogatory attitudes.

The firefighter

Another stock character in the supererogation debates is the firefighter. He rushes into the burning building, saves the baby, and is celebrated as the hero. Some firefighters relish the adrenaline, the machismo, and have no problem with boasting about it afterwards. Such boasting would not undermine the admirability or praiseworthiness of the acts. So far, so supererogatory. But consider the firefighter who, when asked by the same CNN reporter to comment on the events, merely says “I just did my job.” Once again, these words could be ignored as false modesty, but I think they could be taken seriously. Surely it is possible that he sees himself as having volunteered for a role, a role with certain explicit and implicit obligations attached to it – including an obligation to take certain life-threatening risks for the sake of others – and until he either resigns or is fired, then he “just” (an important word) has a job to do. This idea of a role distinguishes him from the Good Samaritan, although the Samaritan might have seen himself as occupying the wider ‘role’ of neighbour.

Now maybe the firefighter's job description is vague when it comes to the precise amount (i.e. probability or quantity) of life-threatening risk that ought to be taken, according to established norms. Presumably there are some risks that would be stupid for any firefighter to take, given their understanding of the situation. However, within the maximum and minimum permissible risks, there would be room for some variation: one firefighter might take a greater risk than another. So while it remains true that both of them are “just doing their job,” some firefighters could in fact be more heroic than others. Usually, however, we are inclined to admire all firefighters, whether or not they merely meet or exceed the minimum standards of the job. Certainly the job is demanding, and only certain people will have the physical abilities and judgement and courage to meet the standards. So it is perhaps easy to see why I, very much a non-firefighter cowering in safety outside the burning building, would admire all firefighters for their courage. However, our man continues to deflect the praise, repeating “I just did my job;” in other words he did what he was obligated to do by his employment contract and the inherent norms of risk-taking. But this is odd, because it seems to shift the locus of admiration: instead of the firefighter's actions here and now, what we praise is effectively his past supererogatory decision to apply for the job in the first place, rather than for a less heroic job. But then there is the problem of mixed motivation: in applying for the job, he might well have been more interested in the machismo and prestige of the job than in the welfare of the future beneficiaries. (There is something unseemly about longing to be a hero.)

While the firefighters’ heroism has a lot to do with the physical risks they take, it is important to recognise a whole host of jobs that we would describe as heroic despite very low physical risks: the ‘caring professions’ of doctor, nurse, social worker, psychotherapist, primary school teacher. Obviously there will be rewarding success stories in these jobs, but our admiration focuses more on their capacity to deal, day in day out, with the grinding, messy unrelievable physical and mental suffering of their patients and clients and pupils, and to deal with the limits of their own skills and knowledge. Perhaps a secondary admiration focuses on those who are able to maintain a heroic good humour throughout, and who manage to resist sliding into despair on the one hand, or into compassion fatigue on the other. And teachers have to deal with the additional burden of their pupils’ open contempt and the state's reluctance to resource the sector properly.

The saint

Of Urmson's two eponymous categories, the saint raises special complications not only because of her religious beliefs, but also of her conception of morality in religious terms; it is not clear that the saint can distinguish the dutiful from the admirably optional if both kinds of action ultimately serve a perfectly good God. Perhaps she might even say that she, too, is “just doing her job.”

There are at least two famous discussions of saints in the Anglophone philosophical literature. First, there is Susan Wolf's ‘Moral Saints’ (Reference Wolf1982), which defined the secularized ideal as “a person whose every action is as morally good as possible, a person, that is, who is as morally worthy as can be” (p. 419). In addition, the saint has to be “very, very nice. It is important that he not be offensive” (p. 422). Notoriously, she argues that moral perfection, in the sense of moral saintliness, does not constitute a model of personal well-being toward which it would be particularly rational or good or desirable for a human being to strive. (p. 419)

While such sainthood might remain supererogatory in going beyond the call of duty, it would not be admirable. The gist of Wolf's argument is that such sainthood would not be an appealing property in a friend; what I look for in a friend is a person with the same sorts of imperfections as I have, and therefore with the capacity to understand and sympathise with me. In addition, I want my friend to give me priority and preference in her life, rather than to apportion her efforts to the most morally deserving causes.

Her argument was thus not targeted against the great saints from history, but rather to a certain kind of self-righteous moral perfectionism, allied with a stoic ideal of detachment from ‘corrupting’ relationships. But I don't quite share Wolf's intuitions here. Part of what I admire in my friend is precisely her effort to do good in the world, however modestly or ambitiously; and my admiration helps me improve as well, in order to earn her friendship. In this respect I admire her supererogatory achievements rather than her merely dutiful ones. And I do not at all agree that a saint has to be very, very nice: part of the saint's job is to fight evil and vice and moral weakness in all its guises, and this will often involve being quite forceful – again, however, my friend's willingness to go beyond her duty to fight for goodness is precisely something I admire about her.

A more serious discussion of secular sainthood is to be found in Raimond Gaita's book A Common Humanity (Reference Gaita2000) (p. 150 ff).Footnote ³ Gaita relates a story from Primo Levi's If This Is A Man (Reference Levi1987). While Levi was working at the ‘hospital barracks’ in Auschwitz, he witnessed the remarkable actions of a man called Lakmaker in attending to the suffering of a fellow inmate. There was little that Lakmaker could do in a practical sense, not only because the inmate had advanced typhus, but mainly because of his own physical weakness and the terrible conditions in all areas of the camp, including these barracks. Gaita's point, however, was Lakmaker's saintly capacity for tenderness, a tenderness that had the power to illuminate the inmate's humanity in these darkest of conditions. While the traditional discussion of supererogation focuses on action that is beyond the call of duty, here were circumstances in which action was highly limited and ineffective; what was far more important was the attitude of tenderness that Lakmaker showed, far beyond anything that could be reasonably expected of him in the circumstances.

It might be objected that attitude is a passive phenomenon, and too far away from the choice that Heyd and others see as essential to supererogation. And yet surely an attitude can be chosen indirectly: one chooses to attend to certain circumstances of the scene, one chooses to remind oneself of certain ideals (religious or other), one may even invoke the Golden Rule. Importantly, the Lakmaker example differs from that of the Good Samaritan in that the former's circumstances greatly impeded the possibility of effective action, and Lakmaker the individual had to overcome many more internal difficulties in deciding to get out of his own bed to attend his fellow.

The forgiver

The final example I want to consider in this section is that of the victim who forgives her wrongdoer. This is one of Heyd's “paradigm” examples (Reference Heyd1982 p. 154), and is also discussed by Hamilton, Stout and Levy in this volume. Forgiveness can be loosely defined as the victim's elective forswearing of justified resentment against the wrongdoer for a wrongdoing;Footnote ⁴ insofar as the resentment is justified, then the elective forswearing is supererogatory. Indeed, the very etymology of forgiveness is to “give in abundance.” As an example of supererogation it is interesting because it involves not an act so much as a forbearance, as Heyd puts it.

Forgiveness has seen a sustained recent interest in Anglo-American philosophy, and the most systematic recent treatment has been offered by Charles Griswold in Reference Griswold2007. There is a debate about the conditionality of the forgiveness; while one person can forgive another, it only becomes admirable, writes Griswold, when the victim waits for the wrongdoer to fulfil certain conditions (above all repentance, but also apology, compensation, and commitment to non-recidivism). Without such conditions, and without a sensitivity to antecedent power structures between wrongdoer and victim, there is too much of a risk that forgiveness will express the victim's self-denigration and the tacit condonation of the wrong: one should consider, first, the example of the battered wife who ‘forgives’ her husband over and over again, and second, certain monumental crimes that have been plausibly described as ‘unforgivable’.

On the other hand, conditionality is also seen as removing the electivity of loving forgiveness and thereby as bringing it too close to a transaction, such that the ‘properly’ repentant wrongdoer can then reasonably expect, even demand the victim's non-supererogatory forgiveness (Gerrard and McNaughton Reference Gerrard and McNaughton2003).Footnote ⁵

A separate debate concerns the role of understanding in forgiveness. On the one hand, we have the spirit of ‘Tout comprendre, ç'est tout pardonner’. However, understanding a wrongdoing would attenuate resentment through a process of appropriate (non-supererogatory) excusing rather than forgiving: the question of whether to forgive (supererogatorily) only comes into play when the victim cannot understand the wrongdoer, even after the latter's sincere attempts to explain the origins of the wrong. Importantly, a different response to the failure to understand would simply be to ignore, and to try to forget; perhaps the great proportion of minor wrongs are processed in just this fashion among people who otherwise have strong reasons to remain in one another's company. However, ignoring is not forgiving because it does not involve the admirable and elective change of heart.

The worry is then that unconditional, uncomprehending forgiveness, because insufficiently anchored in the behaviour of the wrongdoer, becomes effectively unilateral, as if the victim were simply coming to terms with a natural disaster rather than forgiving the wrongdoer for the wrongdoing. Certainly the tone of a lot of self-help literature is unilateral in this sense: what they call ‘forgiveness’ often amounts to no more than the attempt to let go of the resentment, as far as possible, for self-oriented reasons of psychic health. The second worry about unconditional, uncomprehending forgiveness, insofar as it can be sufficiently aimed at the wrongdoer, is that it does not seem to admit of reasons to forgive; and without reasons, forgiveness seems capricious in a way that would take away from its moral admirability.

Dealing with these two problems requires thinking of forgiveness not as an act (an outer object, as it were) like any other act that can be contemplated by weighing reasons for and against; nor is it a matter of deciding on the best mental strategy for manipulating a corrosive resentment (an inner object). Instead, forgiveness begins as an attempt to see the wrongdoer in a new light. Such a new light is not a matter of adverting to the fulfilment of behavioural or situational conditions, but rather of coming to see something – perhaps the humanity – in the wrongdoer. Such a coming-to-see might best be described as love, and I will consider this in the next section. But the important point about forgiveness for any discussion of supererogation is that we have moved from acts to attitudes.

Trust and hope

In a great number of ordinary situations, of course, we have to trust others, and insofar as we have to, then such an attitude cannot be supererogatory. We trust the neighbour, the bus driver, the waitress, the bank teller, the university lecturer. Here it is only the deeply paranoid person who would show a failure of reason in insisting on bulletproof evidence before making herself vulnerable to the others whose goods and services she needs.

However, there are more interesting questions of trust. Consider a famous example from the philosophical literature on self-deception. A “gullible” husband who “deceives himself” about the significance of the growing evidence of his wife's infidelity: she comes home from work later and later, her clothes are dishevelled, there are flirty messages on the answering machine etc.. Nevertheless he refuses to think she might be unfaithful, refuses to challenge her, and this is allegedly because of an understandable but ultimately reprehensible self-deception driven by fear and cowardice.

However, I think another reading of the gullible husband is available, one in which the husband trusts his wife – indeed, he resolves to trust his wife (rather than finding himself, willy-nilly, in a situation of trusting his wife). Such a trust can of course be rationally imprudent, and his close friends have warned him about it; but nevertheless it can also be courageous and ethically admirable. When the husband married the wife he made a vow, and here his vow is being tested: the evidence presents itself, evidence that would incline the impartial enquirer toward certain conclusions that at the very least merit further investigation. And he refuses to entertain either the conclusions or the need for investigation. If he found himself suspecting her, he would consider this to amount to a betrayal on his part, as if it were he who was abandoning the marriage, not she. His imprudence might mean that he is utterly unprepared for her leaving him one day – but so be it. He will tell himself that he is entirely committed to the marriage right up until she leaves, and not a moment earlier.

It might be objected that there is nothing heroic about misplaced or unearned trust, and that it comes too close to servility and self-denigration. (In both respects it would be like unearned forgiveness under the ‘conditional’ account.) We might also might feel differently if the roles were reversed, and it was the wife who was trusting ‘too much’ – but part of this, as I suggested above, would be due to the implied power imbalance. In the unfaithful wife example, the husband has a more-or-less dignified ‘exit option’ that he refuses to use.

Hope is similar to trust in drawing positive conclusions about the future that are not rationally warranted by available evidence. Again, a normal inductive confidence that the future will resemble the past will not be supererogatory if such a confidence is required to lead a life of any sort of normal complexity. However, there are some people who might be justified in considering their lives to be blighted by misfortune, and yet they retain an admirable capacity to hope, not only that fortune will eventually smile on them, but also that other people will be good to them when so many have been so cruel (one thinks of Hardy's Tess, for example). It is tempting to reduce such unwarranted hope to the status of an adaptive survival mechanism, but this would not do justice to two essential parts of the phenomenology: first, the capacity to resolve to hope and to resist the temptation to despair; second, to the way we observers can be moved by such hope (even when it does not avert further misfortune).

Love

I'm using the word ‘love’ as a fairly broad place-holder, and I do not want to get bogged down in definitions. Whether the Samaritan showed love or compassion or care or charity is not important for the point I want to make. Love is interesting in that, unlike the examples described above, it can be meaningfully shown toward various different kinds of objects: individual people, but also abstractly-defined groups (e.g. ethnic or national), institutions, geographical areas, and even ideas or ideals. As before, there is a question of whether proper love has to be ‘earned’ or not; but I will avoid this by pointing out that there was no question whether the wounded Jew deserved the Samaritan's love.

Of course we could just say that insofar as love motivates one to act beyond the call of duty, then love is a supererogatory attitude. But not only do I believe that love can be supererogatory even when it does not motivate actions beyond duty, I also believe that the relationship between love and duty (or rather: between the experience of love and the experience of duty) is not at all straightforward. Consider the example of a father of a young infant. At three in the morning, the father is woken by the infant's piercing cry from the room next door. Since he is ‘on duty’ while his partner sleeps, he rises and attends to the infant until it falls asleep again. One question here is whether such an act is supererogatory, and the answer may well invoke many of the discussions above. But before that we have to ask: why exactly does the father get up at all?

At the basic level, it is clear that he does not want to get up: his greatest preference at that moment would be to continue sleeping. It is also true that the father has no loving feelings for the “little brat” (his words, not mine!) at that moment: quite the opposite. So it is tempting to think that he rises out of parental duty. But, as Harry Frankfurt has argued (Reference Frankfurt1998 p. 8), this would be to interpose an object, as it were, in between the father and the infant: as if the father has to first ask himself what his duty is and then rise in adherence of that duty, with no explicit thought of the beneficiary. And yet this is false to the phenomenology: the father rises because of the infant. It is a direct response to a perceived claim of need. The father's perspective is ‘primed’ to hear the infant, to hear his infant, and to respond directly. This is a similar point as that made in my discussion of the Good Samaritan, above; but in the father-infant context it is obscured by the widespread thought that a parental duty is meant to pick up the slack when the spontaneous loving feelings are absent. However, it is better to understand the father's experience of the infant's claim of need as an expression of his love for her – a love that is not essentially about feelings, but about a whole long-term orientation toward the infant, informed by his own resolve to be a good father in accordance with his understanding of an ideal of fatherhood he has inherited from his family and from wider society. Such a love goes very far beyond the call of duty; one can imagine a hired nanny performing her duties immaculately but without the orientation of love.

While love is often manifest in action, it is also important to consider those cases involving no action, or no change of standing behavioural dispositions. Iris Murdoch (Reference Murdoch1970 p. 17) offers the famous example of M and D to show that an admirable loving attitude – beyond the call of familial duty – can be cultivated without any noticeable change in action. M's son has just married D, and M's spontaneous attitude is one of disapproval, and yet she behaves impeccably to D throughout. With imaginative effort and love, M brings herself, partly by acknowledging her own jealousy, to see D in a better light; and such a change in seeing is genuine, and not merely a matter of ‘making the best’.