Let’s dive head first into one of the internet’s most contentious questions – do we have true free will? This comes up not infrequently whenever I write here about neuroscience, most recently when I wrote about hunger circuitry, because the notion of the brain as a physical machine tends to challenge our illusion of complete free will. Debates tend to become heated, because it is truly challenging to wrap our meat brains around such an abstract question.

I always find the discussion to be enlightening, however. In the most recent discussion I detect that some commenters are using the term “free will” differently than others. Precisely (operationally) defining terms is always critical in such discussions, so I wanted to break down what I feel are the three definitions or levels of free will that we are dealing with. It seems to me that there is a superficial level, a neurological level, and a metaphysical level to free will. Language fails us here because we have only one term to refer to these very different things (at least colloquially – philosophers probably have lots of highly precise technical terms).

At the most superficial level we do make decisions, and some people consider this free will. To be clear, I am not aware of any serious thinker or philosopher who holds that we do not make decisions. There is a deeper discussion about the mechanisms of those decisions, but we do make them, we are consciously aware of them, and we can act on them. From this perspective, people are agents, and are accountable to the choices they make.

But for me, as a neuroscientist, I think the most interesting aspect of the free will question is the neurological underpinnings of our decisions. I also think that objectively this is the most important level, and the one that should (and largely does) guide how we deal with the decisions that people make. It is now pretty well established that our decision-making, as a neurological phenomenon, is only partly conscious. A large component happens subconsciously, even before we are aware that we are making a decision or have made one. We are powerfully influenced by the subconscious circuits in our brains which have their own evolved purpose. But we do have what is called “executive function” – the ability to consider our options, weight the consequences, and make a deliberate decision, even one that countermands what the subconscious circuits want us to do.

However, executive function takes a lot of brain power, depending on the strength of the subconscious motivation we are trying to supersede with pure will power. And, our brains also have wiring that gives us a number of escape hatches – we can rationalize our decisions, going along with what our subconscious brain wants and convincing ourselves that this is what we truly desire. We are good are resolving these conflicts, at tricking ourselves in order to reduce our cognitive dissonance. This includes tricking ourselves into thinking we are not tricking ourselves.

In other words, we seemed to have evolved the capacity to maintain an illusion of free decision-making, even when the reality is we are powerfully influenced by factors largely out of our control. From a neurological perspective, therefore, we have partial free will. We can theoretically make free decisions, but they are not truly free of subconscious influences beyond our control.

Interestingly, I think the world largely operates on this level, with the understanding that people’s choices are only partly free and can be subconsciously or emotionally influenced. Marketing and politics operate largely on this level. The legal system also recognizes this reality. People are held legally accountable for their decisions and actions – but, the law recognizes undue influence as a mitigating factor. There is the “reasonable person” standard, for example. How would the average reasonable person be expected to respond to a certain situation. If most people would have been emotionally manipulated into a bad decision, then perhaps it is not fair to hold someone accountable for such a decision. The law does not treat people like perfect free will agents, but as semi-free agents whose guilt can be mitigated by circumstance or powerful outside influence.

The medical world also treats people as if they had only semi-free will. Public health policy, for example, is premised on the well-established fact that most people do not make perfectly rational fact-based health decisions. In fact, giving people information has only a fairly small effect on their decision-making. If you want to have a sizable effect, you have to emotionally manipulate them, or change their external circumstances. You have to leverage their subconscious circuitry to influence their conscious decisions in a desirable direction. You have to make the healthful decision the easiest decision.

There is also a deeper-still philosophical or metaphysical level to the question of free will – is our decision-making free of the tyranny of cause and effect? Here the answer is also clear – it’s no. From the perspective of physics, our brains are deterministic machines. It may be complex and unpredictable, like the weather, but it still operates within the confines of physics. Quantum mechanics does not rescue us from this fact, because we have no control over the outcome. This is what philosophers mean when they say we do not have true free will.

The real question is – which of these three levels is the most salient to our lives? Should we act “as if” we have unfettered free well, partial free will, or no free will? By this I mean, which level should our laws, customs, and policy reflect? I already argued that I think they reflect the middle level – people have partial free will. I also think this is the correct approach.

We should not pretend, either personally or collectively, that people are fully conscious free agents. This is also what we mean when we call free will an “illusion” – we mean that the 100% unfettered part is an illusion, when in reality we have only partial free will with a huge subconscious component to our decision-making. By default people should be treated as free agents, responsible for the decisions they make, but allow for a host of mitigating factors that recognize people are only partially free agents.

Meanwhile, the philosophical level is interesting, but I don’t think this means we should treat people as if they were robots without any control over the decisions they make. This is sometimes presented as the compassionate view, but I think you get that benefit from the neurological level. I don’t think that “the laws of physics made me do it” defense is of practical use, because it does not distinguish more conscious from less conscious decisions, or external factors vs internal factors. It applies equally to all decisions – so what do we do with that? Not holding people accountable for their decisions also influences their decisions, and might be called a “moral hazard”. Even if philosophically we do not have free will, we need to pretend that we do, or else society cannot function.

I will add one caveat to this – I think the one aspect of the philosophy of free will that is useful is that it can and should temper our anger and wrath at the bad decisions of others, and to remind ourselves that “there but for the vagaries of fate go I.”

In the end I think the neurological approach to free will is the most practical and useful, and is the one that is most used in society. We have partial free will.