Criticising Strawson’s Compatibilism | Issue 159

Free Will

Nurana Rajabova is wary of an attempt to dismiss determinism to keep free will.

The belief that human beings have moral responsibility is used to judge people based on their actions, then to reward or punish them accordingly. But is this just?

This question becomes unavoidable when the theory of determinism enters the discussion. Determinists claim that every event or occurrence in the world, including human desires, thoughts, and acts, are predetermined by physical laws of cause and effect. In such a world there is no space for free will, since any person’s action at any time could not have been different, if all the physical conditions causing it remain the same. As there is no human free will, say the determinists, there can be no moral responsibility either. At the other end of the axis stand libertarians who also view the two phenomena as incompatible, yet the theory they reject is determinism, as they believe that humans do possess free will. Therefore, assigning moral responsibility is justifiable according to their view. In-between these two positions are the compatibilists, who claim that determinism and moral responsibility are not mutually exclusive after all. Different compatibilists explain this with different arguments. In this article, I will only examine one such argument, made by Peter Strawson in his seminal paper ‘Freedom and Resentment’ (Proceedings of the British Academy, Volume 48, 1962), with the purpose of seeing whether it does resolve the centuries old puzzle.

Peter Strawson (1919-2006) by Gail Campbell, 2023

Strawson’s Argument

Before we get into the specifics of his paper, we should note that Strawson has a slightly idiosyncratic compatibilist position. Unlike other compatibilists, he does not identify as a determinist. In fact, he denies that he even understands the thesis of determinism. Instead he argues that even if determinism were correct, that still would not take away our sense of moral responsibility, since for Strawson the question of the justification of moral responsibility is internal to “the general structure or web of human attitudes and feelings” (‘Freedom and Resentment’).

To demonstrate this point, Strawson puts aside ideas of moral condemnation, approval, and the like, which are commonly used in the debate over free will and determinism, and instead invites his readers to focus on something much simpler and common to all human agents, that he calls ‘reactive attitudes’. These are the emotions people experience when involved in interactions with each other, and include resentment, contempt, sympathy and gratitude, among many others. Strawson speaks of these emotions as a natural part of human psychology. For instance, if somebody does me wrong, it is only natural for me to feel resentment towards them. Or if I am assisted in one way or another, I will naturally develop gratitude towards the person who did me the favor. Strawson further notes that such emotional responses are not only applicable in our relationships with others, but also in our responses to ourselves. The feeling of guilt we have when we recognize our own wrong action is a prime example of this. By drawing attention to this common human psychology, Strawson argues that we all possess these reactive attitudes as a human disposition and as a “fact of human society.” He then argues that this is indeed an indication that ascribing moral responsibility is a justified practice, since doing so is simply one of our natural reactive attitudes. It’s not only justified, it’s inevitable, we might say.

Strawson then inquires about the influence of determinism on such responsibility. He asks rhetorically, “Would or should the acceptance of determinism lead to the decay or repudiation of all such attitudes? Would or should it mean the end of the gratitude, resentment, forgiveness and of all essentially personal antagonisms?” He expects that the answer is ‘no’.

Strawson now mentions some exceptional cases where we modify our reactive attitudes and choose to look at things more objectively, or, to use his own words, where we choose to move away from ‘participatory reactions’ to ‘objective attitudes’. He reduces such attitude changes to two cases: (1) When a person did something by accident, with no ill will; and (2) When we judge that a person isn’t an appropriate target of any reactive attitude because they lack some relevant capacity – say a child, or someone with severe mental illness. Nevertheless, he argues that even “when the suspension of such an attitude or such attitudes occurs in a particular case, it is never the consequence of the belief that the piece of behavior in question was determined… In fact, no such sense of ‘determined’ as would be required for a general thesis of determinism is ever relevant to our actual suspensions of moral reactive attitudes.” In other words, the idea of determinism is irrelevant to the application of our reactive attitudes. Therefore, for Strawson, we cannot seek an answer to the question of moral responsibility in the metaphysical realm. For this reason, he criticizes both determinists and compatibilists equally for over-intellectualizing the subject. He says,

“The optimist’s [compatibilist’s] style of over-intellectualizing the facts is that of a characteristically incomplete empiricism, a one-eyed utilitarianism. He seeks to find an adequate basis for certain social practices in calculated consequences, and loses sight (perhaps wishes to lose sight) of the human attitudes of which these practices are, in part, the expression. The pessimist [determinist] does not lose sight of these attitudes but is unable to accept the fact that it is just these attitudes themselves that fill the gap in the optimist’s account. Because of this, he thinks the gap can be filled only if some general metaphysical proposition [ie free will] is repeatedly verified, verified in all cases where it is appropriate to attribute moral responsibility.”
(‘Freedom and Resentment’, 1962)

We can summarize Strawson’s argument this way:

Point 1: Moral responsibility is real and derives solely from the fact that humans possess reactive attitudes. These attitudes would remain even if determinism were proven true.

Point 2: Therefore, inquiry into the truth of determinism in a metaphysical sense is unnecessary, because the acceptance of determinism in the metaphysical sense would not in any way undermine our practical sense of moral responsibility.

In what follows I will address both these points to see whether they really put the debate about free will and moral responsibility to rest.

Assigning Moral Responsibility

Admittedly, the natural human disposition to react to interactions, to which Strawson draws attention in his first point, is an undeniable reality. It’s a quality humans carry as both embodied and social beings. We have evolved to have bodily reactions and mental processes that ultimately serve the evolutionary purpose of our preservation, and our reactive attitudes are part of that.

Interestingly enough, studies show that human bodily reactions do not occur only in response to physical stimuli; they can also happen in relation to ‘moral things’. The anterior cingulate cortex is a part of the brain that activates to produce disgust when a human is presented with rotten food (a function which evolved to keep people from harmful food). It turns out that the same area of the brain also gets activated when a person witnesses an event they deem morally repugnant. This is just one quick example supporting Strawson’s point that the natural human disposition is to react to moral stimuli. However, the question I want to ask is whether the reactive attitudes (in other words, our emotional responses) are enough to confer moral responsibility. For the sake of brevity and clarity, I will explore this question by analyzing it in only one case, that of death, yet one could explore the same question in almost all types of human emotionally reactive experiences.

In the first scenario I’ll refer to death caused by non-living beings, such as natural events (e.g. earthquakes) or inanimate objects (e.g. water). I will take the case of a child drowned at sea. The sorrow caused as the result of this incident will undoubtedly be beyond description for people related to the child, and may well also raise other emotions, such as anger and hatred. It is possible for the mother to hate the sea for taking her child away from her. But despite the extremity of the pain, and the anger and hatred, no one will be able to hold the sea morally responsible, because the sea has no agency. A lack of agency immediately makes a claim of moral responsibility non-applicable. In fact, thanks to the lack of agency and intention, one’s whole conception of the event changes as we take the act altogether away from the object and instead bring it to the agent that was harmed in the process. Thus, in this specific example, instead of saying the sea drowned the child, we say the child got drowned in the sea. The reason is because the sea has no agency. After all, the sea was not acting in any other way than being a sea. Also due to this lack of agency, we more often relate such experiences to fate or to the flow of the processes rather than blame the particular thing that appeared as the immediate cause. It follows therefore that our sense of moral responsibility is not based solely on our emotional reactions. It is also based on our judgment, which requires some other criteria for moral responsibility to be applicable – the first and foremost being the judgement that agency is present. So despite the pain and negative emotions an event may raise in us, those emotions do not immediately yield the ascription of moral responsibility to the cause of the event.

In our second scenario we’ll consider a death caused by an animal. Let us take a case in which a child dies as the result of a dog attack. The result being the same – the loss of the child – this will raise a similar degree of pain in the mother. This strong pain again may naturally raise feelings of anger or hatred. The mother may even develop a sense of vengeance toward the dog. But again our question is whether the emotion arising as a reaction to this event is enough to hold the dog morally responsible. In this scenario, unlike the first one, there is some agency. A dog is a doer of its own actions. Yet, experience tells us that dogs, perhaps all animals for that matter, cannot be held morally responsible, because they lack sufficient cognitive capacity to understand the moral dimension of their actions. Once again we see that the reactive attitude deriving from the incident does not immediately justify the attribution of moral responsibility. In short, our reactive attitudes do not justify ascribing moral responsibility to beings with insufficient cognitive capacity to see moral distinctions. Before we move on, I want to briefly mention the practice of putting down violent animals. Some might argue that this is the result of ascribing moral responsibility to animals, but I would say that it rather for the purpose of getting rid of the danger than from a sense of justice.

Now, in the third scenario, I refer to a similar incident, but this time caused by a human agent. Once again, the death of the child causes the same pain, and the same emotions or reactive attitudes. Yet here we immediately see the sense of claims of moral responsibility. However, even in such a case, the assigning of moral responsibility is not a straightforward consequence of the emotional reactions, and requires some other criteria to be met. Given that a human being is already viewed as a moral agent, we can say that one of the criteria for moral responsibility is immediately met. However, this is not enough. In fact, our sense of moral responsibility also requires a relevant intention. So if the mother knows that the killer never intended to harm the child, that the incident was purely an awful accident, then at a personal level she may forgive him. At the social and legal levels, intention may not have such a resolving effect, although it may result for instance in a reduced sentence. A real-life example like this happened a few years ago in my home town. A grandfather, while parking, not seeing his little grandchild playing behind the car, accidently hit him, and as a result the child died. This is indeed one example that clearly shows the discontinuity between moral responsibility and our reactive attitudes. The grandfather was held accountable on a legal basis, because he met the two main criteria of legal accountability, which are, a wrong action committed by a moral agent. Therefore, the reactive attitude towards this man happened to be more one of pity than resentment. Because everyone familiar with the story knew that he did not have any intention to harm the child.

Needless to mention the cognitive ability of a human actor will also have a great effect on claims of moral responsibility. For instance, if a killer has a serious cognitive disability, such that they really can’t recognise what they’ve done, their victim’s family may take a more forgiving position in consideration of their lack of understanding. However, in this case too measures will be taken by legal institutions, not so much for punishment as for the purpose of reducing the potential danger.

We can see that despite the fact that the degree of pain and the initial emotional reaction remains the same in all three scenarios, these reactive attitudes do not themselves fully justify the claim of moral responsibility. That claim is based purely on rational judgement, and requires the fulfilment of three criteria: agency, intention, and adequate cognitive capacity. Accordingly, in response to the first point of Strawson’s argument, I conclude that the natural human disposition for reactive emotional attitudes towards events happening to them or around them cannot on its own justify the ascription of moral responsibility. This is because ascribing moral responsibility is not based on emotions. Rather, it is based on judgement.

The Metaphysical Assumption of Free Agency

Now I’d like to turn to Strawson’s second point, where he dismisses the importance of the influence of metaphysical views on our ascription of moral responsibility.

Think of the experience of watching a movie with a strong emotional content, where some great tragedy or injustice unfolds. While watching this movie, sitting indifferently is not usually possible. Just like in real life, we feel emotions, and the emotions we feel towards these movie characters are similar to the emotional reactions we experience in our daily lives. We feel anger towards the bad guys, and sympathize with the good guys. We wish we could help them. However, there is an important factor that needs to be taken into consideration here. We feel strong emotions while watching the movie only when the actors play it so well that they almost make us forget the fact that their scenario is written by a screenwriter. Thus, while watching the movie, we implicitly view the characters as free agents and the direct cause of their own actions. But if the actors fail in their acting, there is no way that we will feel those same strong emotions, even though the scenario is the same. The insistent knowledge that they are only acting based on a script written for them weakens our emotional reaction. In addition, our resentment, sympathy, or any other type of emotional response, stops when the movie ends. It’s possible that for a short while after we may carry a certain emotion. However, the moment we are able to make a distinction between the actors and the roles they played, and realize that they had no choice but to follow a script, then our view changes and becomes emotionally neutral. This again shows the relationship between our judgment-based attitudes and free will. In a similar manner, in real life, our reactions towards people, including our ascriptions of moral responsibility to them, come from our seeing them as free agents who directly cause their own actions. In other words, it’s plausible to argue that our ascription of moral responsibility derives from a metaphysical view of people as free agents with intention and the ability to act otherwise. The moment agency is taken away from them, our moral reaction ceases. Therefore, I would argue that people’s views about metaphysics – specifically whether they believe in free agency – are embedded in our moral judgements.

A good example of this would be people with a strong belief in divine determination, also known as fatalism. In cultures where such beliefs are dominant, it’s not unusual to come across people who are muted in their ascriptions of moral responsibility. Oftentimes, such people can seem rather neutral or passive, both in their reactions to the acts of others, and their own. For instance if they have happened to help someone, they modestly decline praise and gratitude for their action, as they do not see themselves as the cause of the good that happened to others, but only as a mediator. They say they were simply a means in the whole process. Generally speaking, those who believe in fate or some other form of cosmic determinism do not react to events like those who believe in free agent causation. In short, we can see that a belief in determinism absolutely plays a role in defining our reactive attitudes, and consequently our sense of moral responsibility. In the absence of a belief in free agency, our claim for moral responsibility as well as our reactive attitudes may be considerably weakened.

Therefore, unlike Strawson, I think that the metaphysical problem of free will and moral responsibility is still a genuine problem, and does not derive from an over-intellectualization by philosophers. On close examination, Strawson’s dismissal of the significance of determinism proves to be unjustified.

© Nurana Rajabova 2023

Nurana Rajabova studied philosophy and is interested in questions of philosophy of mind and metaphysics.