I.1. Optionality and voluntariness

First, let us get a better understanding of the ways in which optional and obligatory acts differ.

The language of obligations mirrors that of necessity. Obligatory actions are things that we must do. If we are morally required to do something, then we have to do it. This can be contrasted with the broader notion of ‘ought’. Consider the following sign: ‘After using the bathroom, everybody ought to wash their hands; employees have to.’Footnote ¹³ There is nothing redundant in the second claim.

The fact that we have to do what we are obliged to do means that we are forbidden from failing to perform an obligatory action. Failures to perform obligatory acts constitute serious moral failings. Failing to do what we are obliged to do can say something about us much more generally: if I tell a lie, then I am a liar; if I commit an act of murder, then I am a murderer. All the occasions on which I have not lied or have not murdered do not count equally heavily against those occasions on which I have. In addition, when we fail to do what is obligatory, sanctions of some form are appropriate.Footnote ¹⁴ Which sanctions are appropriate varies, from informal pressure and blame to punishment.Footnote ¹⁵

Obligations therefore stand in contrast to the supererogatory. Like obligatory acts, it is permissible to perform supererogatory acts. However, unlike obligatory acts, it is also permissible to fail to perform supererogatory acts. Thus, supererogatory acts are optional: they are neither morally required nor morally forbidden. Due to the optionality of supererogatory acts, omissions of supererogatory acts do not deserve the sanctions and punishment incurred by failures to perform obligatory acts.Footnote ¹⁶ Performances of supererogatory acts count much more heavily than non-performances. In this sense, moral compulsion (the shutting off of alternative courses of actions that would be morally acceptable) accompanies obligatory acts, but not supererogatory ones. Thus, there seems to be an important similarity between moral and physical compulsion, which grounds a comparison between optionality and voluntariness.

Optionality and voluntariness seem to be related concepts. At some points, Heyd, one of the main proponents of the argument from autonomy, does draw a distinction between these two notions.Footnote ¹⁷ For example, he builds optionality and voluntariness into two separate conditions of his account of supererogation.Footnote ¹⁸ However, in many places he appears to equate these notions. For example, at one point, he claims that the value of supererogatory acts consists in their being ‘totally optional and voluntary’, while in other places saying that these actions are valuable solely because of their optionality.Footnote ¹⁹ Additionally, in several examples he runs these notions together. He describes the system of taxation in Ancient Greece as voluntary.Footnote ²⁰ However, his explanation of why this system had supererogatory elements is better understood as appealing to the optionality of the payments made. In the case of blood donation, Heyd remarks that the particular value of these acts is that they are purely voluntary.Footnote ²¹ However, he then says that people's ‘freedom to give [blood] voluntarily’ is severely curtailed ‘in a system which makes giving obligatory’, that is, when such acts are not optional.Footnote ²²

The running together of these two notions is revealing: as I will show, it is initially tempting to draw an analogy between the voluntary and the optional, and thus between the non-voluntary and the required.

I.2. The Lack of Alternative Possibilities

Imagine that a bank robber puts a gun to your head and demands that you give them the money from your safe. In this situation, there is reason to think that your handing over the money is non-voluntary. The explanation we might offer for this is that there were no acceptable alternatives to handing over the money. There seems to be an analogous situation in the moral case. When we are morally required to keep a promise, there is no morally acceptable alternative to keeping it. That is what it is for an action to be obligatory. Thus, it might be thought that the non-optionality of morally required actions has a disvalue similar to the disvalue of non-voluntary acts: they are both bad from the point of view of autonomy. Thus the argument from autonomy rests on the assumption that a lack of acceptable alternatives does in fact undermine voluntariness. This is to accept the Principle of Alternative Possibilities with respect to voluntariness: a person performs an action voluntarily only if there are acceptable alternatives to performing that act.Footnote ²³ However, I argue that this principle is false and, thus, that the argument from autonomy fails to establish a value unique to the supererogatory.

I.3. Voluntariness without Alternative Possibilities

The Principle of Alternative Possibilities ought to be rejected: a person may freely choose to perform an action even if there are no acceptable alternatives to doing so, because this fact need not play any role in their decision. Serena Olsaretti has developed an account of voluntariness based on this thought. On her account, an act is non-voluntary if and only if it is performed because the alternatives that the chooser believes she faces are not acceptable, where an alternative is ‘acceptable’ if it conforms to some objective standard (e.g. well-being).Footnote ²⁴ A voluntary act is one that is not non-voluntary.

Imagine that you are living in a city in a desert. It is so far from any other city that, if you tried to leave, you would die of thirst before you reached anywhere else. There are, therefore, no acceptable alternatives to staying in the city. If you stay in the city because of this fact then, Olsaretti argues, you do so non-voluntarily.Footnote ²⁵ There are, therefore, two features necessary for an act to be non-voluntary. The first is that there are no acceptable alternatives. The second is that the course of action taken is taken because there are no acceptable alternatives. Thus, you could stay in the city in the desert voluntarily if it were the case that you could leave without dying of thirst. However, it is also possible for you to stay in the city in the desert voluntarily, despite the lack of acceptable alternatives to doing so, provided you have different reasons for staying, such as enjoying the city, the presence of your friends and family or a satisfying job in the city, a lack of desire to see any other parts of the world, and so on, such that the lack of acceptable alternatives plays no part in your motivation.

There is a lesson to be learned here about optionality. It is true that an obligatory act has a lack of morally acceptable alternatives. However, just as we can voluntarily choose to perform an act, despite the lack of prudentially acceptable alternatives, similarly we can autonomously choose to perform an obligatory act, despite the fact that there are no morally acceptable alternatives. This is because, as I will go on to discuss, we can perform an obligatory act for reasons other than the fact that there are no morally acceptable alternatives.

I.4. Two types of motivation to do our duty

Suppose that I promise to give you a lift home after a party we are both attending. The party comes to an end and you ask me for the lift I promised. I could give you the lift as promised because you would be really angry if I refused and would tell other people that I refused to give you a lift despite promising to do so, perhaps leading other friends to consider me a bad person and treat me accordingly. So one possible motivation for giving you a lift would be to avoid all of these negative consequences. Alternatively, I could give you the lift because of the reasons that make it my duty to do so: that I have promised and that I promised because you have no other way of getting home.

Thus, we can do the right thing because of the considerations that make it the right thing to do or we can do the right thing in order to avoid negative consequences that would be appropriate if we didn't.Footnote ²⁶ This distinction is important because, once we have it in mind, we can see that the argument from autonomy would only work if we could only be motivated to do what is best because of the negative repercussions that would ensue if we failed to do so. However, there is no reason to suppose that, conceptually speaking, this is the only possible source of motivation. Moreover, suppose that what is best on a certain occasion would be to give someone a gift. Most people believe this to be supererogatory – and thus that they are not required to do so – and yet many people still give gifts. If it were the case that we could only be motivated to do what is best because of negative repercussions, then we would never perform acts we thought to be supererogatory for which there are no negative repercussions. We must conclude, therefore, that it is in principle possible to be motivated by considerations other than that an act is morally required.

Therefore, just as an action can be voluntary if we are motivated for a reason other than the fact that there are no acceptable alternatives, we can be motivated to do our duty for reasons aside from the lack of morally acceptable alternatives. In recognizing this, we recognize that no threat to our autonomy need result from our doing our duty as a matter of conceptual necessity, even while accepting that justified sanctions would accompany the failure to do so.

II.1. The language of perfectionism

The ‘perfectionism’ of interest here is psychological perfectionism, a separate notion from the ‘perfectionism’ we might find in moral, political or value theory. In a definition that captures the heart of psychological perfectionism, Randy O. Frost et al. describe it as the setting of excessively high standards that are accompanied by overly critical evaluations of one's own behaviour.Footnote ²⁸ Readers may be all too familiar with this phenomenon. There are three main components of perfectionism that reveal important similarities between perfectionism and maximizing theories that demand the best.

First, perfectionists believe themselves to be subject to rules and requirements. Indeed, perfectionism has been characterized as the ‘tyranny of the shoulds’.Footnote ²⁹ Moreover, these standards and rules have the same form of strong necessity as moral obligations: for a perfectionist, it is not just that ‘I ought to be working all the time’, but ‘I have to be working all the time’; it is not just that ‘my talk ought to be perfect’, but ‘my talk must be perfect’. Thus, it feels to perfectionists, just as it does to those who take themselves to be under a moral requirement, that it is unacceptable to do otherwise.

Second, due to the prescriptive force of these standards, perfectionists are preoccupied with what is referred to as ‘concern over mistakes’: a fear of failing to meet the standards in question. Perfectionism thus involves the belief ‘that doing something perfectly (i.e. mistake-free) is not only possible, but also necessary’.Footnote ³⁰ Many studies demonstrate that perfectionism is ‘associated with a fear of failure and an inability to tolerate failure’.Footnote ³¹ In fact, one of the therapeutic techniques used to help those with perfectionism is to ask them to think about the ‘or . . .’ that follows a statement that begins with ‘I must’ or ‘I should’ or ‘I have to’. So ‘I must read absolutely everything on this subject before starting my work’ might be followed by ‘or my colleagues will think I am stupid and lazy’. ‘I have to be working all the time’ might be followed by ‘or I will fail all my exams and no one will ever give me a job’.

This concern over mistakes, central to psychological perfectionism, echoes important features of moral requirements. Just as in the case of failures to meet our obligations, perfectionists believe that blame, criticism and punishment are legitimated by failures to meet the standards set. Moreover, perfectionists believe that even small failures will have – or legitimate – serious negative consequences. The sense of catastrophe that accompanies the prospect of failure is often due to the fact that the person's entire evaluation of their self-worth – as a person, family member or an employee – ‘is contingent on the pursuit and attainment of their goals’.Footnote ³² Thus, evidence shows that perfectionists exhibit a cognitive bias that is caused by concern over mistakes whereby selective attention is paid to failures, whereas successes are discounted.Footnote ³³ This is similar to the way in which violations of moral requirements (such as murder) are viewed as of much more significance than all the times in which moral requirements were met. Thus, the non-moral requirements of perfectionists mirror moral requirements.

Third, psychological perfectionism is more than just the setting of requirements: it is the setting of excessively high requirements, which are difficult to achieve and highly curtail what counts as an acceptable course of action. In this way, perfectionism is the non-moral analogue of maximizing theories in that they both demand our best and consider all else to be unacceptable.

Drawing out these similarities is instructive because, as I go on to show, the well-documented consequences of being a perfectionist strongly suggest that by requiring that we meet an excessively high moral standard, it is made less likely that we actually meet these standards.

II.2. The counter-productivity of perfectionism

Perfectionism is usually considered to be a psychopathology: something that requires addressing in order for the person in question to have a healthy and satisfying life. This is predominantly because of the correlation between perfectionism and depression and anxiety. This correlation arises because of the role that sanctions and self-criticism play when we believe ourselves to be required to meet excessively high standards, which leads to a ‘morbid fear of failure’.Footnote ³⁴ This correlation with anxiety and depression is exacerbated if the requirements are perceived to come from sources outside oneself.Footnote ³⁵ Similarly, the consequences of believing ourselves to be morally obliged to do what is morally best are likely to be depression and anxiety, especially given that moral standards are often viewed as having an external source.

The correlation with anxiety and depression is unsurprising in light of the biases discussed earlier. Whatever we perceive as the source of the requirements, when we are preoccupied with the avoidance of failure and yet hardly notice when we succeed in meeting the standards in question, we are likely to be riddled with fear and guilt. This is true in both the moral and the non-moral spheres.

The anxiety and depression that accompany a fear of failure should be of some concern to moral theorists in and of themselves. However, the real issue is that this fear is counter-productive: by trying to avoid failing to meet the standards, we tend to avoid meeting them.

The evidence for the avoidant behaviour that accompanies perfectionistic thinking, and the counter-productivity that this entails, is well documented and has become central to the definition of perfectionism itself. The pursuit of high standards and the attendant fear of failure become so aversive that perfectionists can (a) engage in procrastination by delaying the beginning of tasks, (b) abandon tasks midway rather than try and fail to complete them or (c) avoid the tasks completely.Footnote ³⁶ In fact, even when perfectionists engage in (rather than avoid) tasks, they can exhibit ‘an inability to know when to quit’ and persist in tasks ‘beyond the point of reason’.Footnote ³⁷ This also constitutes a form of counter-productivity. This lack of responsiveness to the consequences of their actions means that perfectionists typically display an inflexibility that leads to a decreased likelihood of achieving their goals. This is especially true as perfectionists not only fail to change their behaviour in response to failing to meet their standards, they also fail ‘to relinquish standards even when they are not met and result in adverse consequences’.Footnote ³⁸ More worryingly, perhaps, these adverse consequences ‘may not be viewed by the person as aversive since they may be interpreted as evidence of true striving’ even if it means they are less likely to reach the goals they are striving for.Footnote ³⁹ Thus, the literature on psychological perfectionism demonstrates that the exceedingly high standards lead to a fear of failure and this fear in turn leads to counter-productivity.

The literature on perfectionism demonstrates that for perfectionistic thinking to be counter-productive the person in question does not need to consider themselves to be required to do what is best in every area of their life. For example, a university student can engage in counter-productive behaviour when they are perfectionistic about their essays, even if they are not perfectionistic about, say, cooking or sending cards to their siblings on their birthday. Furthermore, it demonstrates that perfectionism entails counter-productivity in a wide range of contexts, from work, to family, to social interactions. Take, for example, the religious context. Those with ‘scrupulosity’ believe themselves to be subject to overly stringent religious requirements and suffer from debilitating guilt and fear about transgressing these requirements: ‘fearing sin where there is none’.Footnote ⁴⁰ The fear generated by taking themselves to be required to meet exceedingly high religious standards causes those with scrupulosity to engage in counterproductive behaviour that actually interferes with ‘social, occupational, and religious functioning’.Footnote ⁴¹ Ironically, patients suffering from severe and long-term scrupulosity, rather than avoiding sins (as is their stated goal), in fact often surrender to ‘the urge to commit the sin’.Footnote ⁴²

On a theory like a maximizing moral theory, on any occasion where we can act, we are required to do what is best and are forbidden from doing otherwise. We are required to do the moral best in every case. And when all morally significant actions are circumscribed by duty, trying to live a life free of moral errors is like trying to walk along a tightrope over a chasm: any small deviation has serious, negative consequences. Thus, the requirement to do what is morally best is likely to lead to what Michael Stocker calls ‘moral schizophrenia’. This involves ‘a split between one's motives and one's reasons [values or justification]’.Footnote ⁴³ This split can come in the form of two different ‘maladies of the spirit’: in the first, we fail to be moved by what we value; in the second, we do not value what moves us.Footnote ⁴⁴ Practically speaking, on theories without the supererogatory, whereby every act is either morally required or morally forbidden, we are likely to exhibit one or both of the maladies Stocker describes. Suffering from either of these ‘maladies’ means that people are unlikely to be motivated by the demands of these theories, and thus are likely to be motivated instead by the fear of sanctions that are legitimated by the failure to meet those requirements. And we can predict that being motivated by fear of sanctions in the moral realm – just as in the cases of perfectionism in the non-moral realm – will lead to fear and anxiety.

We can, therefore, also predict that adherents of maximizing moral theories would engage in counter-productive behaviour, just like psychological perfectionists.Footnote ⁴⁵ Agents may well avoid situations in which they can be called on to do good. This could take many forms, such as wilful ignorance whereby we avoid finding out about the suffering of others; or avoiding situations whereby we are called upon to do good; or refraining from making promises or commitments so that we don't add further requirements to our already heavy load. Just as a perfectionist might spend so much time trying to perfect an email that in the end they never send it, someone who believes it to be required that they give money to the best or most efficient charity may spend so much time working out which one this is that they fail to give any money at all.Footnote ⁴⁶

This counter-productive behaviour – a consequence of fear generated by considering ourselves to be required to do what is best – gives us reason to suppose that if doing good is what we are interested in, then morally requiring what is best will more than likely hinder that pursuit.Footnote ⁴⁷

II.3. Therapeutic solutions: positive perfectionism and supererogation

The therapeutic solution to negative perfectionism is to encourage perfectionists to acknowledge that there is a state of being ‘good enough’ or performing a task ‘well enough’ that is below doing one's utmost or one's best. They must therefore lower the standards they believe themselves required to meet. Doing this opens up the way for exceeding those standards.

Thus, two types of perfectionistic tendencies have been distinguished in the literature: (1) negative perfectionism, which is the type described so far in this article, is characterized by perfectionistic concern; and (2) positive perfectionism, characterized by perfectionistic striving.Footnote ⁴⁸ The therapeutic solution to negative perfectionism is not non-perfectionism but positive perfectionism: to aim to be better without believing oneself to be required to do so.

The solution in the moral case is the same: we must lower the moral standards such that we are permitted to perform what is best, but also permitted to perform less than the best. As in the non-moral case, this opens up the way for exceeding what is required. By reducing what we take to be required, we allow for optional actions, some of which are better than others. In other words, the solution is the supererogatory. Just as negative perfectionism is the analogue of a moral theory that requires us to do the best, positive perfectionism is the analogue of a moral theory that allows for supererogatory acts.

The literature on positive perfectionism demonstrates that when the defining feature of negative perfectionism – concern over mistakes – is removed, as in the case of positive perfectionism, the possibility of healthier individuals is revealed: individuals who are more motivated and more successful. Anxiety and fear is only associated with perfectionistic concern, not perfectionistic striving.Footnote ⁴⁹

Amongst the extensive list given by Stoeber and Otto are the following key findings: perfectionistic striving without perfectionistic concern is correlated with higher levels of conscientiousness,Footnote ⁵⁰ extraversion, endurance, positive affect, satisfaction with life, and achievement in general as well as specifically academic achievement; additionally, it is correlated with lower levels of depression, self-blame, anxiety, procrastination, defensiveness, and maladaptive coping styles.Footnote ⁵¹ These same consequences, we can predict, would also arise from a life following a moral theory that made room for the supererogatory.

An important finding is that achievability has a large part to play in productive behaviour. Problems arise for perfectionists because they feel required to achieve goals that are extremely difficult to achieve. Negative perfectionists ‘establish unreasonably high personal standards’ and this leads to inevitable failure.Footnote ⁵² For positive perfectionism, on the other hand, what is required is in fact attainable: perfectionism only remains adaptive when people avoid ‘the daily experience of achievement setbacks’ as those who ‘experience achievement failures are substantially at risk for depression and dysphoria’.Footnote ⁵³ The standards we are morally required to meet must therefore be achievable in order to avoid being counter-productive.Footnote ⁵⁴

Once we reach our goals, we receive satisfaction and sometimes praise and other positive feedback. This indicates the importance of recognizing both the metaphorical carrot – as well as the stick – as a form of motivation. It is this difference – between positive and negative reinforcement – that marks the distinction between perfectionistic striving and perfectionistic concern.Footnote ⁵⁵

There is also positive and negative reinforcement in the moral case. As established earlier, there are serious negative consequences for failing to meet obligations. This means that performing our duty is negatively reinforced: it enables us to avoid the punishment that would otherwise be legitimated, such as a sense of moral failing, guilt, shame, punishment, and the censure and disappointment of others. However, as established in the non-moral case, it is a mistake to think that negative reinforcement is the only form of motivation. Supererogation provides an alternative mechanism of motivation, involving praise, merit, gratitude and esteem.Footnote ⁵⁶ These consequences encourage the performance of the acts in question more effectively than a fear of failure and the avoidance of punishment.

Thus, the evidence concerning psychological perfectionism I have presented here implies that, practically speaking, there is something to be said in favour of J. O. Urmson's argument that we should not demand what is beyond the capacity of ordinary men.Footnote ⁵⁷ If what is morally required is beyond the capacities of most people, then most people would experience ‘daily setbacks’ and regular failure, which would make the requirements counter-productive. However, Urmson argues that when the law asks too much, as in the case of prohibition of alcohol in America, then ‘as people got used to breaking the law a general lowering of respect of the law naturally followed’.Footnote ⁵⁸ I have demonstrated that the psychological mechanism of counter-productivity actually comes from too great a respect (in a sense) for the moral law: seriously believing that one deserves to be punished and censured for not doing one's best is what leads to the fear and anxiety and thus the counter-productive behaviour.Footnote ⁵⁹ Furthermore, even if the demands of duty are such that it is possible for ordinary people to meet, the mere act of requiring the acts in question leads to the fear of being unable to meet them and this fear is self-fulfilling: people become less likely to meet their duty the more the fear that they cannot.