Can Compatibilism Save Free-Will?

Ryan Hubbard, PhD
Apr 11, 2020 · 14 min read

I used to live next to a small duck pond. Sometimes I would stroll around the pond and see if I could get a duck to eat a piece of bread out of my hand. Usually, one of them would timidly walk up to my hand and snatch the bread. The duck performed an action. Back at the college where I work, a colleague down the hall would sometimes offer me a chocolate candy. He would offer it up and I would timidly walk up and take it. I performed an action.

Is there any difference between my action and the duck’s? Most would say that the duck acted out of instinct and I acted out of free-will. Despite my sweet tooth, I freely chose to eat the chocolate. According to many philosophers, my action was free because I could have acted otherwise. This is what it means for an action to be free. I could have chosen differently, if, say, I was trying to cut my sugar intake. The duck didn’t really have a choice since it was driven by nature. The duck lacks free-will. Indeed, one of the things that makes people special and distinguishes us from all — or most — other animals is that we have free-will. At least that’s what it seems.

The belief in free-will is so ingrained that it’s difficult to imagine how we could get by without it.

In philosophy, any idea is up for investigation, even ideas that we take for granted, whose truth seems obvious. Free-will is one of these ideas. In fact, the belief in free-will is so ingrained that it’s difficult to imagine how we could get by without it. It may even be impossible to participate in life while believing none of our actions are free. Nevertheless, we can still question free-will’s existence.

Whether we have free-will depends on what makes an action free. There are two different accounts of what a free action is. According to the uncaused account, a free action is an action that is not caused: nothing determines the action. According to this account, if an action is determined by something, then it cannot be free. The other account is the self-caused view. According to this account, an action is free if it’s caused by the self, the will, our desire, or some other state of mind. This is different from the uncaused account because an action can be free even if it is determined. It just has to be determined by the right kind of thing.

With these two views of what counts as a free action in mind, I’ll start with a theory that denies we have free will.


Determinism is the view that free-will does not exist. To motivate this view, consider all the factors that influence our behavior.

One factor is human nature and biology. Humans are born with basic instincts that have a strong influence on our behavior. If I don’t eat enough food I get hungry. There’s no way around that. And this causes me to have certain experiences like feeling hunger and discomfort, which, in turn, cause me to seek food.

The social environment also is a factor that influences our behavior. People are shaped and conditioned by their environment, which influences the kind of people we turn out to be, the views we adopt, our tastes, and so on. For the most part, we cannot determine the environment in which we’re born, so its influence on us is largely beyond our control.

Psychological forces also exert a strong influence on our behavior. For example, if someone has a horrific near-death experience of almost drowning as a child, they may develop a powerful fear of large bodies of water. This fear may sink into the unconscious as he grows older, but, nevertheless, exerts an influence on his behavior. He may, for example, avoid going to water parks. He may think this is his own decision, but it is really his unconscious fear that is driving him.

There are innumerable things that influence our behavior. When we begin to take them all together, it starts to look like much of what we do isn’t really up to us. Rather, all our behavior, and even our mental processes, are caused by things we have no control over. It may turn out that none of our actions are truly up to us. Consider this example:

I like to listen to music when I work. I usually listen to ambient electronic music with slightly melancholic melodies played with vintage style synthesizers. It certainly seems like I choose to play this kind of music. But there are thousands of things about my life that nudge me to pick out this genre. As a kid, I just happened to be drawn to my dad’s record collection which just happened to have a lot of heavy classic rock, like Black Sabbath. This could have influenced my taste in melancholic melodies. I have happy childhood memories which formed in the 1980s, just the time when a lot of synthesizers — now considered vintage — were being used. This could explain why I’m drawn to vintage synths. I could list more influences, but you get the idea. My taste in music, what I’m drawn to, and what I ‘choose’ to listen to at any given moment, is so highly influenced by all these things that my ‘choice’ isn’t really a choice at all, because it’s determined by things I had no control over. What I like to listen to isn’t up to me.

In d’Holbach’s work, determinists take the scientific model of nature and apply it to human behavior.

Once we wrap our heads around the billions of small things that influence our behavior, we’ve moved beyond mere behavioral influence and into the territory of behavioral determination. This lends support to determinism.

We can find an early argument for determinism in Baron d’Holbach’s book The System of Nature. In d’Holbach’s work, determinists take the scientific model of nature and apply it to human behavior. According to science, the universe and everything in it are ultimately governed by natural laws, such as the laws of physics. These laws help us predict the behavior of objects since they explain the causal relations between things in nature. If we place a bucket of water in an atmosphere that is below zero degrees Celsius we can predict that the atmosphere will cause the water to solidify. If Bobert accidentally places his hand on a hot stove — and his nervous system is functioning properly — I can predict that he will jerk his hand away.

Everything in nature is subject to causal laws. The human body is made up of natural, material stuff, so how are we any different than any other natural, material stuff? The determinist claims that, ultimately, we are not. We are subject to the laws of the universe just like everything else. If this is true, then all our actions are determined. We have no free-will.

Here’s the determinist’s argument:

  1. All events in the natural world are caused.
  2. Humans are part of the natural world.
  3. Therefore, all human actions are caused.
  4. An action is free if and only if it is uncaused.
  5. Therefore, no human action is free.

The argument is valid, but is it sound? Are any of the premises false? The first is probably true. Everything that happens, happens because it’s caused by something else. The second premise is true just so long as we accept the scientific model of the universe. For most philosophers, this is uncontroversial. Given that (1) and (2) are true, (3) must be true: all actions are caused. Assuming that (4) is also true — that an action is free if and only if it is uncaused — will get us determinism.

But is it true that a free action must be uncaused? This premise is controversial and challenging it may be the ticket to saving free-will. W. T. Stace undertakes this challenge by developing compatibilism.


Determinism has gained traction today. This makes sense because it accords well with our scientific understanding of human beings. If all behavior can be explained based on neurochemical interactions in our nervous system, there leaves little room for free-will. However, while a lack of free-will accords with science, it doesn’t fit well with our common, everyday experiences and practices. W. T. Stace begins with this idea in his critique of determinism.

Stace begins with a common-sense observation: Determinists believe that there is no free-will, but they certainly act as if they do! In fact, believing in free-will seems inescapable. This is because there are at least two things we all do that require believing in free-will: (1) we make normative claims; (2) we hold each other morally responsible. Let’s start with the first.

We make a normative claim when we say that someone ought to do or not do something. We make normative claims all the time. If I tell my friend she shouldn’t spend so much time playing video games, I’m making a normative claim. Notice that for my suggestion to make sense, my friend must have the ability to choose not to play video games. It makes little sense to tell someone that they should or shouldn’t do something if he or she couldn’t have done otherwise, that is, if they don’t have free-will. It would be like saying you shouldn’t have jumped back after stepping on hot coal barefoot or you shouldn’t have coughed when you got something stuck in your throat.

Stace is saying that in order to give normative reasons — claiming you ought to do something — we have to assume that it’s possible to act otherwise. Therefore, our practice of giving normative reasons for action presupposes free-will. If normative reasons exist, then this suggests that free-will must exist.

The second practice we all participate in is holding each other morally responsible. This also seems to require free-will. Generally, it makes little sense to hold someone morally responsible for something they could not have done otherwise. Imagine, for example, I step on your toe because I accidentally tripped on the sidewalk. Even if I caused you a lot of pain, it wouldn’t make sense to hold me responsible or blame me, since tripping was not of my doing. My tripping was caused by something external to my self. If determinism is true, then all my actions would be like tripping on the sidewalk. Therefore, it would not be appropriate to hold me morally responsible for anything I do. But, surely this is absurd!

Here’s the argument from (2) above in step-by-step form:

1. If determinism is true, then we do not have free-will.

2. For it to be possible to hold others (and ourselves) morally responsible, then we must have free-will.

3. It is possible to hold others (and ourselves) morally responsible.

4. Therefore, we have free-will (from 2 & 3).

5. Therefore, determinism is false (from 1 & 4).

I think this is a strong argument against determinism. The first premise is true by definition. Common-sense suggests that premise three is true. This leaves premise two up for debate. Is there any way to challenge this premise? I leave it to you to decide (pun intended)!

Piggybacking off Stace, I’ve tried to show that the practices of holding responsible and giving reasons for action require free-will. It may be possible for the determinist to get around these objections by claiming that these practices are like games that we shouldn’t play since they require us to believe something that is false, namely, that we have free-will. Therefore, we should simply opt-out of these practices.

We simply cannot remove ourselves from the games of holding others responsible or giving normative reasons. If this is true, then the determinist must live in a kind of schizophrenic state.

There are two problems with this response. First, notice that the recommendation is a normative claim about what we should do. As we saw, if the determinist is correct, then, by his own lights, it’s incoherent for him to suggest this.

Another reason why it makes little sense to say we should opt-out of these practices is that they are inescapable. We simply cannot remove ourselves from the games of holding others responsible or giving normative reasons. If this is true, then the determinist must live in a kind of schizophrenic state: he believes that free-will does not exist, while simultaneously behaving as if free-will was real. His actions of holding responsible and giving normative reasons contradict his belief that free-will doesn’t exist. He falls into a performative contradiction.

From the standpoint of science, however, determinism still exerts a strong pull. If you believe in science — which we all should — we can’t help but see ourselves as part of nature and so subject to the same causal laws as everything else in nature. Indeed, it must be true that all events, including human actions, have a cause. At the same time, it really really seems like we have free-will! Otherwise, we couldn’t make sense of normative reasoning and holding responsible. How can we make compatible these two ideas? To repeat, here are the two claims that we want to hold onto:

(1) All events have a cause (the scientific worldview)

(2) Free-will exists

Determinists claim that (1) and (2) are incompatible and opt to maintain the first since it accords with science. Stace, however, argues that these two claims are compatible. We can have our cake and eat it too.

The mistake the determinist makes for thinking that (1) and (2) cannot both be true is that he assumes an action is free if and only if it’s uncaused. The determinist makes a semantic mistake about the meaning of free-will. According to Stace, if we examine our common use of the term ‘free,’ we will see that the determinist’s definition of a free action is wrong and that, contrary to the determinist, free actions can be caused.

To identify what it means for an action to be free, Stace presents us with two lists of actions. One list consists of actions that we would commonly regard as free. The other consists of actions that we would commonly regard as unfree. Here are the lists.

If our everyday use of language is any indication, then the meaning of ‘free action’ is whatever characteristic is common to the FREE ACTION column that is absent from the UNFREE ACTION column. If the determinist is correct that a free action means an uncaused action, then the action being uncaused would be the characteristic common to the FREE ACTION column that is absent from the UNFREE ACTION column. However, this is clearly not the case. The actions in both columns are caused.

In the FREE ACTION column, Gandhi fasting is caused his desire to fast. Stealing the bread was caused by one’s mental state of hunger. Signing the confession occurred because it was caused by wanting to do so. In the UNFREE ACTION column, the man fasting was caused by the state of affairs in which there’s no food. Stealing the bread was caused by a threat. And so on. According to Stace, this suggests that the determinist’s definition of a free action is mistaken. It’s not the case that an action is free if and only if it’s uncaused.

If the actions in both columns are caused, then what explains the difference between what we commonly refer to as free actions and unfree actions? According to Stace, the difference is that free actions have a different kind of cause.

If you look at the two columns, what is common to the FREE ACTION column is that the cause of the action is internal to the person performing the action. The cause of the actions in the UNFREE ACTION column is external to the person. This suggests that an action is free if and only if it is immediately caused by the person’s internal state and the person could have done otherwise. An action is not free if and only if the action’s immediate cause is a state of affair external to the person and the person could not have done otherwise.

Are our unimpeded, rational desires really up to us?

The internal or psychological states that cause free actions are internal constraints. My desire for a slice of cheese pizza, for example, is an internal constraint on my actions. States of affairs external to the agent that influences her behavior are external constraints. If I eat pizza because you threaten to beat me if I don’t, your threat is an external constraint. According to Stace’s compatibilism, free actions are compatible with internal constraints.

To sum up: Compatibilism makes compatible the fact that all events (including human actions) are caused with the claim we have free-will, by defining a free action as one that is caused by internal states. Just what these internal states are requires further investigation. A potential problem for Stace’s compatibilism points us in this direction.

To show that Stace’s account of free-will is problematic, it would help to come up with counterexamples. Are there any examples of actions being caused by internal constraints (or internal states) that we would commonly regard as not free? There are plenty! Here’s one:

I once read a case in which a person with a psychiatric disorder covered his windows with aluminum foil because he believed that the birds outside were sent by the government to spy on him. His action of covering the windows was caused by an internal state consisting in a belief about the birds and a desire to cover the windows. This is consistent with Stace’s account of free action, but surely this person’s action wasn’t free. The action was a product of paranoia and delusion. It doesn’t seem like it was up to him.

Compatibilism’s Burden

Moritz Schlick, in his critique of Stace’s compatibilism, makes this point[1]. Schlick claims that actions aren’t free when one’s internal constraints — or mental states — are the result of mental illness or neurosis. This certainly seems true. If my action is based on irrational desires, delusions, or psychosis, it hardly seems that the action was free. In these kinds of cases the action isn’t up to me. Rather, it’s the result of me being bound by psychosis.

Photo by Gerd Altmann

Schlick’s insight suggests that we need to amend Stace’s account of free action. There are many ways we could amend it to address Schlick’s criticism. We could say that an action is free when it is caused by our unimpeded rational desires that originate in our authentic selves. It seems that these kinds of desires are part of who we are and so actions stemming from them are caused by our self.

However, we can press against this amended version of free action. Are our unimpeded, rational desires really up to us? If we take a step back and observe them, it looks like desires just seem to pop in and out of our minds. At least at some level, what we want doesn’t seem to be up to us.

To save compatibilism, we need to identify an internal cause of action that makes the action genuinely up to the agent. This requires an account of self-causation. So, the compatibilist faces the burden of explaining how the self can bring about actions in a way that is consistent with science. This, I would say, is a heavy burden indeed.

[1] See his essay, “When is a Man Responsible?”

A Philosopher’s Stone

A place for a discussion of the ideas all around us in…

Sign up for A Philosopher's Stone Newsletter

By A Philosopher’s Stone

A place for more deep, more exclusive, and more experimental thoughts (Also...dream journal, psychedelic journal, exclusive podcast and more)  Take a look.

By signing up, you will create a Medium account if you don’t already have one. Review our Privacy Policy for more information about our privacy practices.

Check your inbox
Medium sent you an email at to complete your subscription.

Ryan Hubbard, PhD

Written by

A philosophy professor who works in practical ethics. @ryankhubbard

A Philosopher’s Stone

A place for a discussion of the ideas all around us in society, culture, philosophy, and more.

Ryan Hubbard, PhD

Written by

A philosophy professor who works in practical ethics. @ryankhubbard

A Philosopher’s Stone

A place for a discussion of the ideas all around us in society, culture, philosophy, and more.

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app