Keywords

At school, there is a football tournament between classes from all the different year levels. This is a highlight of the school year. The pupils look forward to participating in dramatic matches out on the grass. There will be uneven matchups when pupils from different age levels meet each other. The pupils from the seventh grade are bigger and more powerful than those in the fourth grade. The oldest can tackle and shoot the ball harder than the youngest. The younger pupils enter the field in joy and horror when they face sixth and seventh graders. Here they get to test themselves against the school’s biggest and toughest pupils. These are boys and girls that they usually follow with admiration in their eyes during recess. Now they get to experience close contact with them out on the playing field. If they are lucky, they can get in one tackle or shoot the ball towards their goal. Scoring against them seems implausible. Imagine dribbling past the older pupils by sending the ball between their legs!

All the pupils are engaged and give everything in such matches. The teachers have discussed whether it is safe for pupils from different age levels to meet. The risk of some of the younger ones being injured is higher than usual. Normally, they play football with each other and are more or less at the same physical level. In this school tournament, they come up against bigger and stronger players. Nevertheless, the teachers have concluded that the risk of harm is acceptable. They consider it healthy for the youngest to encounter physical superiority. It is joyful, risky play, which is good for the children’s development. Then they have also asked the older pupils to show consideration for the younger ones. This is good training in adaptation for them as well.

Max is in goal for his class in the fourth grade. He has made a heroic effort in several matches already. He has fearlessly thrown himself after the ball and knocked it away from the danger zone. It is both wonderful and scary to stretch your arms to catch the ball in matches against older players. The ball comes towards him with greater speed than he is used to. Older pupils have also noticed Max’s efforts and praised him for good saves.

Then comes the match against the best team in the seventh grade. Max finds his place in the goal. His heart beats extra fast. The referee blows the whistle, and the match is underway. After ten minutes, a seventh grader gets the ball just a few metres from the goal. He hits the ball as hard as he can, and Max throws himself after it to make a save. The young goalkeeper feels intense pain as the ball hits his forearm. He screams. The referee stops the match, and several adults bend over Max to check his condition. They understand that this is serious. A teacher takes Max to the emergency room at the local hospital, where they find arm fractures. Max has to be plastered and cannot stand in goal for a while.

Afterwards, teachers and parents reflected on the incident. What lessons could be learned from it? Did it show that it is unreasonable to have a football tournament where pupils meet across grade levels? Was Max’s arm fracture proof that the pupils should rather play football in a safer environment, with opponents of the same age? Or should his injury be seen as an example of what one must reckon with and tolerate in children’s development towards becoming robust and strong individuals? When the teachers allowed the football tournament to take place, they gave priority to do-good-ethics over avoid-harm-ethics. Did Max’s injury document that they were wrong in doing so?

Morality and Luck

The philosophers Bernard Williams (1981) and Thomas Nagel (1979) introduced the concept of moral luck to point out how the moral assessment of actions is often coloured by actual outcomes. This happens even if the outcome is affected by conditions that are beyond the person’s control. A good outcome is used as evidence to say that a person has done a good deed, while a bad outcome supports a conclusion that what the person did was wrong. Attempts to save people from a burning building are often considered more morally commendable if they succeed, than if they do not. We can imagine two people storming into the building and each making their own rescue attempt. Only one comes out again carrying a person from the house. The other comes out empty-handed, because a burning beam got in her way. Both have risked their own lives to save others, and only one of them succeeded. We can assume that the difference in outcome is due to coincidences. Initially, both have made a great moral effort, but the one who saves a life is likely to receive significantly more moral praise.

In the same way, we differentiate between people who have endangered the lives and health of others. Two people drive home from a party in their own car, after consuming a considerable amount of alcohol. One of them hits and injures a pedestrian, while the other gets all the way home without causing any accident. The first driver is often blamed more than the one who has had luck and not run down anyone. The risk they have exposed others to is exactly the same, but coincidences lead to only one causing an accident, and on that basis receive much sharper moral criticism than the other. One thing is that the accident makes others aware of the drunk driving, which does not happen in the case of the person driving home without injuring anyone. This form of luck concerns what others learn about a person’s reprehensible actions. Moral luck is something else. It affects the moral judgement of what has been done. The assumption here is that when we hear both stories, we direct sharper moral criticism at the person who caused the damage.

Moral luck is a paradox. On the one hand, we think that moral praise and criticism should be directed at what people have control over. What was their motivation, and what did the risk picture look like? What was the probability that they would succeed in doing good or avoid inflicting harm? On the other hand, the moral judgements are coloured by the outcome, even though coincidences have played a crucial role.

Professionals who work with children can be exposed to moral luck. The football tournament at Max’s school has been considered a success for years, since it has created a lot of joy and happiness, and no one has been seriously injured. Then Max’s arm fracture happens, and people immediately reconsider the activity where old and young pupils meet for pretty rough matches on the football field. The assessments of the teachers who are responsible are seen in a different light. Is the arm fracture proof that it is morally irresponsible to let fourth graders play football against seventh graders? Have the teachers had moral good luck until now and suddenly experienced moral bad luck?

It is possible to reject that luck has anything to do with morality and say that the moral judgement must put the spotlight on intent and risk, regardless of whether someone breaks their arm or not. Risk is determined, as we have seen, by a combination of probability and severity. Teachers who say yes or no to a football tournament across grade levels must assess the positive consequences of the activity against the risk of injury. Is this worth taking a chance on? It should be possible to make a factual moral assessment in advance and build on the knowledge that is available when the decision is made. It may well be morally justifiable to take a chance now that later turns out to lead to a negative outcome.

Research on moral luck shows that people’s immediate moral judgements are often coloured by actual outcomes (Martin & Cushman, 2016). In fact, we have a tendency to be morally stricter towards those who cause harm than those who do not, even though the risk they expose others to is identical. In that sense, moral bad luck is a real phenomenon, but it is also documented that the effect weakens when people have time to think about. When we gain knowledge about the course of events and reflect on what has happened, the moral luck disappears (Kneer & Machery, 2019).

In the previous chapter, we mentioned the distinction that Kahneman (2013) draws between quick and intuitive decision-making (system 1) and slow and analytical decision-making (system 2). Now we can see that this distinction is also relevant in terms of moral luck. It is only in the quick and thoughtless assessment that we find moral luck. In thinking where the pace is slower, chance does not affect the conclusion about how morally right or wrong a particular action is.

It is natural that a dramatic and sudden outcome of a decision leads to immediate moral criticism of those responsible, especially from people directly affected. Goalkeeper Max’s parents had reason to be morally upset and angry that the teachers let their son be a goalkeeper against seventh graders. They are close to the situation and react impulsively to it based on their system 1 thinking and will then understandably point a critical finger at the teachers and the school. Even their assessment may be different when they think more closely about it. Then they can conclude that the risk that Max was exposed to through a football match against older pupils was, after all, morally acceptable.

A natural human reaction after an accident is to look for someone to blame, a scapegoat. Who has failed in the situation? This is a reaction that is typical of system 1. Professional accident investigators say that they always try to shift attention from blame to causes (Kvalnes, 2017). How could this happen? The inquiry activates System 2 and a process geared towards learning from the situation. These investigators interview people who have been active in the situation or witnessed it all. If the focus is on blame, people will respond defensively and cautiously, so as not to put themselves or others in a bad light. When attention is shifted to causes, the answers immediately become more honest and concrete. This makes it easier to capture how the risk assessment has been and whether it was correct or deficient.

School and kindergarten teachers are constantly in situations where they have to weigh what is an acceptable risk in play and activities for children. Based on previous experiences with how parents, management, authorities, and the media react when children are injured during play, they may have more or less good reason to fear that they can suffer from moral bad luck. Patterns and habits can exist for how accidents and injuries are handled. If it is system 1 that prevails in their work environment, then it generates caution. If, on the other hand, there is also room for system 2—a way of thinking that has no room for moral luck—then it provides professionals with sufficient safety to implement risky play. Teachers depend on some form of moral protection for in their work. The next section is about how this protection can come in different degrees, affecting how much risky play professionals dare to allow children to engage in.

Moral Protection

The Greek philosopher Aristotle (385–323 BC) claimed that good human qualities lie on the golden mean between extremes. For example, being brave lies on the middle ground between being cowardly and being overconfident. Aristotle proposes an alternative to thinking in opposites and pairs of concepts. Instead of the division between brave and cowardly, kind and naughty, open and closed, caring and indifferent to others, we can have a division that shows that good qualities can be exaggerated and turn into something negative:

Insufficient

Golden mean

Excess

Coward

Brave

Overconfident

Naughty

Kind

Naive

Indifferent

Caring

Self-destructing

Secretive

Transparent

Loose mouthed

Villainish

Honourable

Fanatic

Suspicious

Realistic

Naïve

We can also use this way of thinking in an attempt to define what is reasonable moral protection for adults who have an influence on children’s opportunities to engage in risky play. Here we are talking about risk on two levels:

  • What risk does the adult allow the children to face?

  • What risk does the adult expose himself to by opening up to risky play?

The last form of risk can be assessed on the basis of the same pattern that we have presented previously, by comparing probability and negative outcome. By allowing children to engage in risky play, the school or kindergarten teacher exposes himself to the possibility of being affected by negative consequences. What will be the reaction from your own leader, parents, and other affected parties if a child injures himself? How likely is it that this will have negative consequences in the form of accusations and criticism, and career repercussions?

Teachers and others who work with children need some form of moral protection when making decisions that concern children’s scope of action for risky play. We can define the reasonable level of protection by thinking in terms of the golden mean. Professionals can get both too much and not enough moral protection. If they feel morally under-protected, a natural reaction would be to be very restrictive about what kind of activities children are allowed to participate in. In the worst case, it can lead to moral paralysis, where professionals do not dare to release children in play and activity, even if the risk of injury is minimal. The other extreme is when professionals experience moral overprotection, where they assume that they will always escape negative consequences for themselves, even though they have allowed high risk and children are harmed.

The three levels of moral protection can also be linked to the degree of personal and systemic responsibility. The middle ground is one where both are present. When the ten-year-old breaks his arm in a football match at school, it is the responsibility of both the teachers who arrange the tournament and the school, the principal, and leader at the school. A main reason for perceived moral under-protection may be that the professional feels alone in having to bear the burden and become a scapegoat if things should go wrong, and an absence of systemic support. Conversely, the experience of moral overprotection may be due to the actors assuming that they can always blame the system if someone is harmed and thus can abdicate any personal responsibility.

The balance between do-good-ethics and avoid-harm-ethics can also be linked to the perceived level of moral protection. Extra low levels nourish avoid-harm-ethics. Then the practitioner will be nervous to allow any activities that may lead to injury, in the first instance for the child, and in the second instance for the practitioner himself, although this process is very unlikely. Extra high levels of protection can lead to the blossoming of an unrestrained do-good-ethics, which neglects the ways in which things can go wrong. With this addition we get an overall picture that looks like this:

Level

Mindset

Consequence

Responsibility

Ethical orientation

Moral under-protection

If something goes wrong, I have to take it completely on my mantle. Fear of moral bad luck

The agents become morally paralysed. They choose the cautious solution every time

Only personal responsibility

Avoid-harm-ethics dominates

Reasonable moral protection

If something goes wrong, we get a factual assessment of whether it was justifiable to take that risk

The agents become morally active and engaged, creating reasonable scope for risky play

Both personal and systemic responsibility

Avoid-harm-ethics and do-good-ethics are in balance

Moral overprotection

If something goes wrong, then I do not have to take any responsibility. It is the system that takes the brunt. No fear of moral bad luck

The agents become partly morally hyperactive and can also partly neglect long-term negative consequences that result from passivation

Only systemic responsibility

Do-good-ethics dominates

The test of where a kindergarten or school is located in this landscape often comes when a dramatic event occurs. How is the situation handled by the kindergarten director or principal? In the event of serious injury, it can be tempting for management to shift the responsibility onto their employees and make them scapegoats. Then they go free themselves. Such a reaction will reinforce the impression among their own employees that they are morally under-protected. The other extreme will be that there is an injury and no one meets affected parties face to face to convey compassion or personal responsibility. The system is to blame, and no individuals take responsibility.

There may be ambiguity in the moral overprotection of people who work with children. Firstly, it can lead to the release of dangerous activities, since any sudden and immediate negative consequences are something that the individual does not have to take responsibility for. It is very unlikely that harm to children will have any negative consequences for the professional’s own life and career. Secondly, overprotection can provide fertile ground for a reduction in opportunities for risky play. There can be laziness among the actors, since they do not have to take responsibility for the long-term negative consequences of children not getting enough physical and mental challenges. No one is going to point to them and ask for an explanation of why they have kept the children indoors for long periods of time, so that they have had limited motor development.

Moral protection of employees in a kindergarten or school is to a large extent a managerial responsibility. It is the managers who need to find the right balance between doing good and avoiding harm and who should stand by the employees’ side when things go wrong. An employee may have allowed children to engage in unreasonably risky pla, and thus deserve some form of moral criticism. The situation may also be that the risk has been within the acceptable range, and then it has nevertheless led to harm. A leader should in principle be dismissive of moral luck and put the spotlight on what one knew and how one assessed probabilities and severity in advance. Parents and other upset parties may be dazzled by the drama of the situation, but a principal or kindergarten leader needs to keep a cool head and give a sober assessment of the situation at hand.

Agent Regret

Max had to go to the hospital and have his arm plastered after his heroic save of a hard shot from a seventh grader. Some adults have decided that it is acceptable for teenagers to play football against younger children, even though there is a considerable difference in physical strength between them. From time to time, the younger pupils will be injured in clashes with the older ones. Teachers and their leaders are aware of risk, have decided that it is reasonable, and also have a responsibility to cope with the situation when pupils are harmed. Therefore, it is natural that Max and his parents get to meet one or more teachers who can show compassion and regret. These teachers can do this without necessarily claiming or admitting that they consider the risk of such unequal football matchups to be morally unacceptable. They can explain, apologize, and show responsibility, without it having to involve a retreat from a positive attitude to risky play of this kind.

When things go wrong, the active party who directly or indirectly caused the injury are involved in the matter in a way that is different from an ordinary spectator, even if what they have done is not morally unacceptable. The term agent regret has emerged from the discussion of moral luck and can be used to account for this relationship (Sussman, 2018; Williams, 1981). This is a form of regret that is linked to being the agent who caused or allowed something to happen that had a negative outcome. The agent has a special connection to the harm that has occurred, even though there is no reason whatsoever to blame him. Nagel uses the example of a driver who keeps the speed limit and has full attention to the traffic in front of him and still hits a child who runs straight onto the road in front of the car. There is no time to slow down or turn away. Everything happens so fast, and hitting the child is inevitable.

An assessment of this situation may be that the driver can consider himself a spectator to the incident. There is no rational basis for criticism or blame. Therefore, the driver has no stronger reason to apologize than any other person who has witnessed the accident.

Williams (1981) has offered a different view, one that takes into account the special relation of the person who through no fault of his own has caused harm. He has launched the concept of agent regret and uses it to explain that anyone who has had their hands on the steering wheel and their foot on the accelerator can actually feel a deeper connection to the event than someone who happened to be a passive witness. Even if the driver has not done anything morally wrong, and so deserves no moral blame, it is this person’s actions that have led to harm. The agent has a unique relation to the events at hand and may thus have a particular kind of responsibility for expressing regret and concern in relation to family and other people affected by the negative outcome.

In 2002, the Norwegian ice hockey player Espen Knutsen was a professional in the club Columbus Blue Jackets in the United States. During a game against the Calgary Flames, he sent off a shot where the puck changed direction when it hit the stick of an opponent. The puck disappeared into the crowd and hit a thirteen-year-old girl on the forehead. She was sent to hospital and died the following week from a blood clot that had occurred when her head was thrown back. There was no reason to blame Knutsen for the death. Here it was rather a systemic error that allowed a hard puck to sail over the fence and hit a person in the audience. After the incident, protective netting was introduced in all arenas in the league.

After this tragic event, Knutsen has met the family of the girl who died. They have conveyed that they are obviously not critical of him in any way. By meeting them, Knutsen expresses a form of agent regret. “It’s clear. I was the one who shot the puck. If I had not done that, it would not have happened” (VG 19.12.10). The ice hockey player admits that he cannot consider himself a spectator to the event. It was he who set in motion a causal chain that led to the loss of a life.

Teachers and others with influence on children’s scope for risky play can also adopt an attitude of agent regret when an accident occurs in a school or kindergarten. They are more than mere spectators to the events, although they have not done anything morally wrong. It is reasonable to expect that they show particular care and compassion for those who have been directly or indirectly harmed, similarly to how the ice hockey players Knutsen took time to meet the family in the example above. He does not distance himself from the tragic event but admits to having a particular causal relation to what has happened. This would not have happened without his action. Teachers who have allowed football to be played between pupils of different ages can also tell the pupil who gets injured and his parents that they set the stage and thought that this was acceptable. Without their approval, there would have been no match and no harm. They can admit this, without taking moral blame for the outcome, since they still believe that the risk they allowed was reasonable.

Summary

In this chapter we have seen that the outcome of our actions tends to influence the moral assessment of them. Anyone who opens for risky play for children can have moral bad luck in circumstances where a child is injured while partaking in such play. It is first and foremost the quick and intuitive moral judgements that are influenced by what happens as a result of a particular action. When we have time to think about it, the element of moral luck tends to disappear. Then we understand that the moral assessment of actions should focus on the knowledge that was available when the decision was made. We can assess whether something is a reasonable risk, given the information people had at the time of the decision. It is understandable that people who are directly affected by a negative outcome react with moral anger and outrage. Even they have the opportunity to adjust this reaction when they consider the incident from a distance and can decide whether it was actually morally unacceptable to let the child participate in the risky play.

We have also seen that it is crucial in schools and kindergartens to put in place a reasonable form of moral protection for employees. The fear of moral bad luck can lead to professional people restricting the scope for risky play. They can become morally paralysed by the fear that they alone will have to take the burden if risky play should lead to harm. The opposite situation is where employees are morally overprotected, and experience that will escape any form of responsibility or blame if an activity should lead to harm. The golden mean is a state in which both personal and systemic responsibilities are recognized. The employees dare to allow children to develop within the framework of what is perceived as an acceptable risk.

The last concept under discussion in the chapter was that of agent regret. We consider it to be a useful term to acknowledge and reflect on how those who are opening for risky activities can end up with a special responsibility to express regret when things go wrong. The risk involved may have been reasonable, but the outcome was nevertheless negative. We can imagine that Max’s teacher felt an extra responsibility for the brave goalkeeper’s arm fracture. This is a teacher who has not done anything morally blameworthy or wrong, but who is still involved in the situation differently than a passive spectator.