Free will & Moral responsibility


Libertarianism, Determinism & Compatibilism


Libertarianism about free will is the view that we do have the power to make genuine choices and could have done otherwise than what we did. It is the view that free will exists.

Sartre is a libertarian. Sartre claimed that there is no objective purpose, nor anything else determines our actions because “existence precedes essence”, meaning humans exist before they have a defined purpose and so have to subjectively define their purpose for themselves. This suggests that they have the free will to decide their purpose. Sartre’s argument is a psychological one, that people cling to fabricated notions of objective purpose like religion or Aristotle’s ‘final cause/telos’ because they are afraid of not having a purpose, more specifically they are scared of the intensity of the freedom that comes from having to choose their own purpose. Sartre thought that this sense of “radical freedom” led to feelings of abandonment (by God/objective reality), anguish (over the weight of being completely responsible for your actions) and despair (over our inability to act exactly as we’d like due to the constraints of the world). It’s much easier to believe that we don’t have free will than face that existential angst.

As Sartre’s argument is psychological, he does not provide metaphysical grounds for rejecting determinism and so is arguably committing the genetic fallacy. The genetic fallacy is assuming that the way in which someone comes up with a theory is relevant to whether it is true or false. Just because people have a psychological need to believe in objective purpose, that doesn’t mean it doesn’t exist.

Counter-defence: Sartre could respond that this is a misunderstanding of his argument. Sartre’s starting premise is that there is nothing in our experience which suggests we have a telos or are determined. All that we experience is ‘radical freedom’ – a sense that every choice we make is completely up to us because there is nothing in our experience like God or telos which could influence or guide that choice. So Sartre is using an a posteriori approach.

Belief in the mind being non-physical (dualism). If our choices originate from a non-physical self or soul, then they would be unaffected by the deterministic cause and effect of the physical universe.  Descartes’ arguments for the soul. Kant phenomenal vs noumenal realm.

The interaction problem: If the mind isn’t part of this physical universe- then how does it interact with it?

Hard/Philosophical determinism

Barron D’Holbach was one of the first Atheists and observed that if we are not created by God and don’t have a soul, we are just physical things like any other and therefore follow the same laws of cause and effect. Every event is caused by previous events, including human action. If we keep tracing the cause of our action back in time, eventually we will get to before we were born, and could ultimately go all the way back to the big bang. We were not responsible for the big bang, nor our birth, but therefore we cannot be responsible for our actions either. So there is no such thing as free will.

John Locke argued against the idea that the feeling of free will is a reason to believe it exists, by showing how it could be an illusion. Locke asked us to imagine a man in a locked room who wakes up, unaware it is locked, and ‘chooses’ to stay in the room. He felt like he made a choice, when actually reality was such that no choice was in fact available to him. Locke argues this could be the case for every human action. We simply are unable to directly perceive all the causes and effects that determined our action, which leaves us with the illusion that we were not determined, when really we were.

Quantum mechanics tells us that some things happen without a cause. Therefore determinism seems false.

However, if our actions happen because of random quantum mechanics, that hardly seems a better basis for free will than determinism.

Honderich responds to this criticism by arguing that the structures of the brain might be large enough that the laws of quantum mechanics (which only applies to the very small atomic level) might not actually apply to them nor their function. If this is the case, while determinism might not be true at the Quantum level, it could still be true at the macro level.

Libet and Haynes

Experiment suggesting free will does not exist

Compatibilism – also called soft determinism

This is the view that free will and determinism are compatible (can both be true). Hume distinguishes between internal causes (causes that are internal to a person – their beliefs, desires, motivations, intentions) and external causes (causes that are external to a person – someone forcing them to do something). Hume noticed that we only hold people responsible for actions that result from our internal causes. So Hume defined free will as being determined by your internal causes not external causes. Even though our internal causes are just as determined as our external causes, Hume thinks this definition of free will nonetheless gives us the conception of moral responsibility we want.

This is not the definition of free will people want. They want to actually be the uncaused cause of their actions, and to have the ability to have done otherwise. Kant called Hume’s compatibilism ‘wretched subterfuge’ which suggests humans are just like ‘clockwork’.

Compatibilists argue that free will does not exist, however, yet they claim to have found a definition of free will that allows for ethics.

The distinction between internal and external causes is incoherent. Don’t internal causes ultimately trace back, if we go far enough, to before we were born, and therefore to external causes?

The conditions of moral responsibility and its relevance to reward and punishment

Moral responsibility

Moral responsibility is the idea that a person is culpable for their actions in a moral way such that they deserve praise or blame for them.


There are two main types of theory about why punishment is justified:

Retributive: reward/punishment is justified because someone deserves it. This theory of the justification of punishment requires that people are responsible for their actions.

Protection from society: reward/punishment is justified because it protects people and society from harmful actions. This theory of the justification of punishment doesn’t seem to require that people are responsible for their actions. There are three main ways in which punishment can enable protection from society:

Rehabilitation: reward/punishment is justified because it helps to improve a person’s character to a point where it becomes safe enough for them to be released into society again. 

Deterrence: reward/punishment is justified because motivates people to avoid harmful criminal behaviour.

 Incapacitation: punishment like prison is justified because it can completely prevent a person from harming society.

Libertarianism on moral responsibility and punishment

For libertarianism, the conditions of moral responsibility are Libertarian free will, that the person had the ability to avoid doing the action. Intention: the person intended to do the action.

The importance of intention for moral responsibility can be seen in the different levels of punishment given for unlawful killing. A premeditated murder will be punished more severely than manslaughter.

If you kill someone by hitting them with their car, yet they were so drunk that they just jumped in front of the car and there was nothing you could do, you might even be let off for that. However, if in the same situation you were speeding at the time, and you could possibly have stopped before killing them had you not been speeding, then you could be considered responsible.

Libertarianism on punishment. The traditional view of punishment is called retributive punishment; that it was justified because a criminal ‘deserved’ it. This was the justification for medieval forms of punishment involving torture. Although torture is now generally considered barbaric in the west, it is still typically thought that criminals deserve their punishment of prison.

Darrow was a lawyer who used psychological determinism as an argument in the case of two boys, Leopold and Loeb, who had committed murder. Darrow argued the boys were a product of their upbringing, which they did not choose, and therefore should be considered less responsible. Their sentence was reduced from the death penalty to life imprisonment.

Libertarianism seems quite harsh in its ignoring of the consequence of the influence of upbringing on people’s choices, which can be immense.

Determinism on moral responsibility and punishment

Libertarians and determinists typically agree that moral responsibility depends on free will. Without free will, a person could not possibly have acted differently and so we cannot consider them praiseworthy nor blameworthy for their actions. It’s hard to make sense of the idea that a person is responsible for their action if there was no possibly way that they could have avoided it. Since determinism claims that free will does not exist, it looks like moral responsibility cannot exist either and this brings the justification for punishment into question.

We wouldn’t regard a robot as responsible for its actions, because its actions are determined by its programming. Humans do not have programming, but our actions are determined by the state of our brain, which is itself determined by our genetics and environment. So in terms of moral responsibility, it’s hard to see why there would be a difference between us and the robot, on the deterministic view.

Sam Harris gives another analogy which drives the point further. He referenced a true story about Charles Whitman, a man who suddenly went on a murderous rampage, killing his wife and several other random people. He left a note saying that he loves his wife and has no idea why he is behaving that way, requesting that his brain be studied for an explantion. It was discovered that he had a brain tumour pressing on his amygdala gland (the gland connected with emotions like anxiety and anger) which could explain his unexpected violent outburst.

Harris asks you to consider the effect that learning about this brain tumour has on your sense of the Whitman’s moral responsibility. It seems to remove the sense that they were responsible. The murderer had no control over the brain tumour and thus it seems they had no control over their actions. However, Harris then suggests that actually, in terms of moral responsibility, all human actions are like that. The violent actions were predetermined by the neurophysiology of the brain tumour, but if determinism is true, then all human actions are predetermined by neurophysiology. A brain tumour is an obvious striking case of interference in normal functioning, but if normal functioning is just as pre-determined as a brain tumour, then in terms of being a basis for moral responsibility, they are identical. If we are to be logically consistent, we should extend to all humans the same sense of a lack of moral responsibility that we have towards the violent man.

Determinism on punishment

Determinists typically claim that punishment cannot be justified retributively because without free will and moral responsibility, there is no coherent sense in which a criminal can be said to ‘deserve’ the punishment.

The Supreme court of the USA: Free will is a ‘universal and persistent foundation stone in our system of law” and that therefore determinism is “inconsistent with the underlying precepts of our criminal justice system”.

As this statement by SCOTUS shows, libertarians worry that without free will and moral responsibility, the idea of punishment will fail to make sense and we won’t be able to justify punishing criminals, which could make our society fall apart. If a criminal had no choice or control over their actions or even their intentions, as determinism claims, then it’s hard to see how we could justify putting them in prison. Some determinists conclude that punishment cannot be justified.

However, other determinists still think that punishment can be justified, however it needs to be on grounds other than retribution. The defence of society and the reforming of the criminal are justifications for punishment that do not seem to depend on free will. This is the approach taken by northern European countries like Norway. Criminals there typically receive lower sentence durations, live in better conditions in prison and receive greater focus on rehabilitation and re-joinining society.

Instead of falling apart, countries like Norway who practice this rehabilitation approach to punishment have much lower rates of criminals re-offending compared to countries like the USA which still relies on retributive punishment and even the death penalty.

Sam Harris is a determinist who argues for justification for punishment consistent with determinism. Consider the situation of a bear wandering into an area that humans are living in and attacking them. We do not need to think the bear has free will in order to justify tranquilizing it and locking it up if we can or shooting and killing it if we have to. We could regard human beings in this way too. Putting them in prison or even killing them doesn’t have to be justified as a ‘deserved punishment’, it could simply be justified as necessary for the defence of society.

Compatibilism/soft determinism on moral responsibility and punishment

Compatibilists typically attempt to redefine free will to be a human action which is predetermined in a particular way which is consistent with our traditional views on moral responsibility. Hume, for example, claimed that we should define free will as when an action is determined by internal causes, because it is for actions caused in that way that we have traditionally ascribed praise or blame.

So, for Hume, a person has moral responsibility for actions which are determined by their internal causes, such as their intentions, personality, desires, knowledge, beliefs, etc.

A person is not morally responsible for their actions which were determined by external causes, however.

Compatibilism/soft determinism on punishment

Compatibilists believe that criminals are morally responsible for their actions that are determined in the morally relevant way, such as those that are determined by their internal causes.

If a criminal is morally responsible, it follows that they deserve punishment. So, a retributive view of punishment could be justified on compatibilism.

However, compatibilists also believe in determinism, which is difficult to combine with the idea that a criminal deserves punishment. This could cause a compatibilist to adopt a weaker less harsh form of retributive punishment.

Alternatively, compatibilists could simply give up on retributive punishment and adopt a view punishment like Harris does; that punishment for rehabilitation or defence of society could be justified in a way that is compatible with determinism.

Compatibilist free will cannot grant moral responsibility. The motivation behind compatibilism is to find a definition of free will that works with determinism yet also preserves the ethical function of the traditional definition. So they might argue that we are morally responsible for our actions that were determined by our internal causes. However, it’s hard to see how we can we have moral responsibility for something we couldn’t have helped doing.

Psychological determinism would argue that our internal causes are the direct result of conditioning external causes, which also questions the validity of Hume’s conception of moral responsibility.

Compatibilists could abandon the idea of moral responsibility, forgetting about praise and blame, and instead focus on proposing punishment or reward, for those actions that are determined by internal causes.