It’s about paying attention, not making choices.
‘Free will’ is, to most people, synonymous with making choices. To exercise one’s free will is to make choices deliberately and consciously. To many people, a belief in free will is essential for success. I believed this growing up, until I started to learn about physics, the idea of physical determinism, and the structure of the brain. I stopped believing in free will for a few years, and the quality of my choice making — and thus my life experience — deteriorated rapidly.
I could tell that I was very unhappy as a result of making poor choices, and I desperately wanted to understand how I could make better choices. This meant that I needed an understanding of free will that was compatible with the materialist understanding of the world. Answering a famous unsolved question in philosophy became a matter of survival.
The implicit model I had picked up as a kid went like this: the whole brain is mechanical, and there’s a small box inside — the “choice making machine” — which is how my mind interacts with my body to direct its choices. The workings of this box were never explicitly laid out in my thoughts. Once I started to understand the idea that neurons follow simple mathematical rules, I realized I was believing in this silly idea. Without an alternative explanation for how free will worked, I stopped believing in free will. It was depressing.
I understood I’d be happier if I could say I believed in free will, but I felt compelled to believe in the truth as I understood it, even if it hurt. My inability to choose to believe something I knew was false convinced me further that I didn’t have free will.
If you aren’t like me, and you’re capable of operating your day to day life without an explicitly stated, coherent philosophy of free will and its interaction with mechanical determinism — if you just accept that you make choices and that’s that — then this may not make much sense to you. If, however, you are like me and wonder how choice making could work in a universe which is largely mechanical, in this essay I will explain how I think about free will now, give you some practical ways to use this theory, and then argue that it is compatible with our understanding of science.
To show you the way I think about free will, it will help to directly contrast it with the old model I used — one which I think many people use without realizing it.
The ‘Free Will As Choice Making’ Model
Consider this short story: Alice wakes up in the morning and brushes her teeth, as usual. She gets dressed, selecting a green blouse with a white top. She has eggs and rye toast for breakfast — which she does every day. She gets in her car. While selecting music for the drive, she sees there is heavy traffic on her normal route to the 101 and decides to take an alternative route.
In the past, I would have told you that the choices Alice made in this story were what to wear, what to listen to, and what route to take to work. I wouldn’t count ‘brushing her teeth’ as a choice, because this is a habit of hers; it is as if her body made the decision on ‘auto pilot’. I would have applied the same logic to her breakfast — if she eats the same thing every day, her “choice making machine” would not have come into play there.
Notice that after the story, I started labeling “choices” with an assumption: I considered all of the actions Alice took, and decided which of the actions involved “making a choice” and which of the actions were simply “autopilot.” I was assuming that free will applies only to choice making. That assumption — that free will only applies when actions are carried out — that is an assumption I no longer make.
I believe now that free will is not ‘the mechanism by which we make choices’ — which I think is almost entirely autopilot behavior, determined by the laws of physics operating on our mechanical neurology. I believe we exercise our free will when we choose what to pay attention to.
If you want to make better choices, you must change what you pay attention to before the choice is made.
Making Better Choices Through Selective Attention
This belief — that free will is involved in my selection of what I pay attention to — changes entirely the way I look at making choices. For example, suppose I go to a party and plan to drive home. This means I need to limit my alcohol intake. The ‘old model’ suggests that the way to do this is to not drink after a certain point, and say ‘no’ to any offers of drinking after that point. The old model doesn’t say how to do this, though. The “Free Will as Choice-Making model” just says “If I wish to do X, I must do X” — which is an empty tautology and useless for practical action.
The new model I have — Free Will as Selective Attention — says, “If I wish to do X, I must consciously focus my attention on my reasons for wanting to do X. Why do I want it? What will happen if I do X? What will happen if I don’t do X? What won’t happen if do X? What won’t happen if I don’t do X?” This new model provides a recipe for action — an algorithm.
My old way of thinking said free will only came into play when I was faced with a choice. My new way of thinking says “Free Will must be active at all times If I wish to accomplish my goals.” The new model says you can constantly exercise your free will mechanism by directing your attention willingly.
Exercising your attention selectively strengthens the neural connections between concepts and experiences in your mind — when you think a certain way, over and over, you make it easier and more automatic for your brain to think that way in the future — you reprogram the autopilot. This ability to reshape our neural pathways is called neuroplasticity. Because the way you pay attention to your internal thoughts over long periods of time changes the hardware of your brain, the next time you are faced with the same choice, you will respond differently — because the mechanical system responding to the choice the second time is different — you are different — and so it responds in a different, predetermined way.
The only choice you have is to select which possible future version of yourself is the one you will experience.
I still spend a lot of my time with my attention responding to my environment — which is to say, reacting instead of acting. What I have worked on doing lately, to some success, is to to notice unpleasant emotional states, and direct my attention to some positive aspect of the situation.
Crying babies still annoy me. They are loud and frustrating. Now, however, I am sometimes able to remind myself that the baby is crying because it wants attention, and by thinking about someone else — by paying attention to my internal mental simulation of a different person, rather than paying attention to my internal simulation of myself — I am able to reduce the unpleasant aspects of the situation.
My grandfather always uses the phrase “Positive Mental Attitude”, and I understand now what he means by this. The attitude you take towards a situation is a choice — and I believe it is really the only choice a person has.
If I go to jail for being angry on the road and rear-ending someone at a stoplight, the ‘old analysis’ would say that I had a “failure of willpower” at the moment I was on the road. The error I made was to “not be in control of my emotions.” The ‘new analysis’ says that I failed, over weeks or months or years, to practice reflexively diffusing anger. I didn’t fail at the stoplight — that was an automatic response. I failed in the weeks, months and years leading up to that stoplight, in which I failed to cultivate a peaceful state of mind as my brain’s default tendency. The decision to ram the car ahead of me would still be my responsibility — but it would be outside my control at the moment it happened.
Losing control doesn’t mean you’ve failed in that moment — it means you did a bad job navigating the space of possible configurations of reality, to get yourself to a place where you would lose control.
Is this model based upon Empirical reality? I believe so and present my argument in this section.
Feedback Loops — Even In Deterministic Systems — Lead to Unpredictable Behavior
We know the universe is not deterministic thanks to experiments in quantum mechanics. There are many people who have tried to use quantum theory to explain aspects of a ‘spiritual matter’ and I don’t want to join their ranks. The question at hand, to be answered, is why a certain object with a given mass and volume moves the way it does. That the object in question is “my body” shouldn’t invalidate the laws of physics as we understand them- nor should it remove the object from being considered as acting under the influence of physical laws.
The standard argument against free will goes as follows: “You aren’t making choices. There are physical processes in your brain doing all of this, and the idea that you are in control is an illusion. If I, the adversary, understood exactly where all of the atoms in your body were, what energy level of each of the electrons is at, the local magnetic and gravitational fields, I could predict what choices you would make.”
The argument is by contradiction.
Let us grant the adversary’s assumption — that my actions can be predicted with perfect accuracy from my physical structure. Because we have granted the nihilist’s assumption — that he could predict what choice I will make — I must also be allowed to have information about the sate of the nihilist’s brain, its atoms, electric fields, energy levels, etc — so I can predict the nihilist’s prediction, and counteract it.
The nihilist claims he will use the standard model of physics to compute the actions taken by my body. If this is possible, then I can do the same, and predict which claim the nihilist will make about my actions. Once I know how the nihilist will predict my actions, I can go and do something different, proving the nihilist wrong.
The nihilist cannot deviate from the prediction I make — because as long as I make the prediction using the same procedure the nihilist claims is valid, if he deviates from the prediction I made, then he’s proving himself wrong.
Granting the nihilist’s assumption — that he can predict my behavior with perfect accuracy — allows me to make a choice which contradicts the nihilist’s assumption — the nihilist’s assumption must be wrong.
Wouldn’t the nihilist predict my response and change his prediction as a result? There is clearly a feedback loop at play here — my knowledge of the nihilist’s prediction allows me to invalidate the prediction. If the nihilist factors this knowledge into his prediction, the nihilist must use a computer to simulate my prediction of the nihilist’s prediction — and then i’ll be able to simulate the nihilists’s prediction of my prediction of his prediction, and so on.
Readers who are familiar with Gödel, Escher, Bach will notice the pattern emerging here: a twisted feedback loop. This is the same pattern that Alan Turing used in his proof that the halting problem is uncomputable.
The Role of Quantum Indeterminacy In Cognition
I believe that feedback loop in cognition is how tiny quantum mechanical events play a role in determining the motion of my body — because my brain is arranged in ever larger loops of cognition which feed back upon themselves. The precise role of quantum effects is irrelevant to this claim — our knowledge of their existence alone is evidence that quantum mechanical effects play a role in human action.
It would be preposterous to make the claim “no macroscopic body has ever changed its motion because of quantum mechanical events” — or else all the physicists who fly to quantum mechanics conferences would have to be claimed to be “destined” to fly to a building with a sign saying “Quantum Mechanics Conference” raised above it, regardless of whether quantum mechanics had been discovered.
If there is a conference to discuss “Why Quantum Mechanics does not Influence Cognition”, the people at the conference are flying there because of the observed indeterminacy of quantum mechanical events; it cannot be denied. The only argument to be made at this point is in the relative proximity of the cause (the indeterminacy of quantum mechanical outcomes) to the effect (the motion of educated bodies of meat and bone discussing precisely how they are influenced by those outcomes).
Do electron wave-functions collapsing in in my brain play a direct causal role in the motion of my body — or is it merely that the outside world changes, changing the input presented to my brain, which processes that input deterministically? The question is, to me, irrelevant to free will. Of course, I’d like to know the answer — but even if the brain is an entirely deterministic piece of hardware, and the ‘external environment’ is the only thing which changes nondeterministically — I can still show that feedback loops in cognition lead quantum mechanical events to determine the motion of substantially larger objects, because you are reading this article, which you wouldn’t have done if nobody had discovered that quantum mechanics is non-deterministic.
If my observations of the world reflect my position in it, which reflects my observations of it in the past, which in turn reflect my position of it … then we get the same sort of feedback loop that Douglas Hofstadter (author of Gödel, Escher, Bach) has proposed is the basis of consciousness — a twisted loop of selfhood. If the closest proximal cause for my body’s motion is “the feedback loop created by my neurological structure’s mirroring of its own mirroring of the external world” — then it is entirely reasonable to say that I am the primary cause of my actions.
The clouds cause the rain. It would be silly to argue that “it is not the clouds, but the precise arrangement of atoms in the clouds, as well as their pressure” — It’s like saying you didn’t read an article, but merely scanned your eyes across a grid of glowing dots. It isn’t even an argument, it’s a series of sounds with no meaning.
I am the cause of my actions — but only when I pay attention to my environment according to my own wishes, rather than merely responding to the loudest signal in my environment.
Of course, if I go repeating the same mistakes I’ve made a hundred times before, because I’m not paying attention to what i’m doing or why it failed — then I’m not really the cause of my actions; I’m just a meat robot like the nihilist suggests we all are.
Empathy Makes The Difference
The trick which lets the free will mechanism escape the nihilist’s predictions is empathy. If the nihilist claims, “I can predict the behavior of all meat robots which are not allowed to simulate the behavior of twisted loops” — then the trick I used (predicting the nihilist’s prediction) becomes invalid.
In other words, an aspect of this model is the idea that you cannot truly have free will if you can’t think from someone else’s perspective. Even if the “someone else” is a “a future version of you who’d rather you didn’t bet everything on black” — you are still thinking from the perspective of something which is not you — and that ability is what allows you to escape the feedback loop of self-directed thought.
The difference between a meat robot and a self-aware person is that when you tell a person, ‘you are hurting me’, the person apologizes, while the meat robot offers a defense of its actions as being in accordance with Principle X, which all Good Persons know to be a valid justification for harming non-good Persons.
The choice to empathize with someone who is upset at you is very difficult — but it’s rewarding and ultimately has improved my life in a huge way.
I think our minds have a crazy amount of computational power at their disposal, and when you work on simulating your copy of yourself — when your default mode network is active and your mind wanders — you tend to get unhappy pretty quickly. When you think about other people, you can’t get nearly as far down in the recursive call stack, because you don’t know nearly as much about others as you do about yourself. This is why people who routinely think about others are much happier — thinking about yourself all the time is a surefire way to make yourself miserable.
I’d say I do this type of thinking — about myself — maybe 80% of the day. That’s a long drop from the 100% it used to be, and I’m substantially happier for it. I’ve got a long way to go, though. I’m still mostly a meat robot, who occasionally experiences the joy of thinking of others.
Descartes said, “cogito, ergo sum” — I think, therefore I am. I believe he was talking about the same thing I’m talking about — the choice to pay attention selectively is what drives the existence of the self as a causal agent.