What is the first animal to come to your mind when I ask this question? I bet at least one animal comes to mind. But why did the animal you chose come to mind? Why did, say, a blue spotted salamander or box turtle or naked mole rat come to mind when I asked the question? You might reply that it is because it was the animal you chose to think about, using your own free will. But was it, though?

Sure, we can say that you chose to think of the salamander (pretending for now that the salamander was the animal that first came to your mind). But why did you choose the salamander? You might then answer that it was because you wanted to. That is, after all, how we make most decisions: we do the thing we want to do, even if it is because of some future reason (i.e., you may not want to go to work in any immediate sense, but you want to go to work because you know it is how you make money which you will want for paying bills or buying things you like in the future; more on future planning later).

But why did you want to think first about the salamander? You might think that somewhere in your mind all the possible animals you know about were presented to you and then, using your free will, you chose the salamander, but that still does not answer the question of why you chose the salamander instead of Taenia solium – why is it that you decided to choose the salamander out of the memory-warehouse of all the animals you know of? Did you choose that you wanted to choose the salamander? You almost certainly did not choose to want to think of the salamander, but even if you contend that you did, that only pushes the question back: why did you want to want to choose the salamander? As Schopenhauer says: “Man can do what he wills but he cannot will what he wills.”

Now you might think “sure, okay, maybe I wasn’t in control of which animal I thought of in this instance. But this was a snap decision. I am still in control of my decisions when reasoning and while thinking and planning long-term.” But reasoning and long-term planning are both just a series of thoughts that you think in the moment, none of which you had any control over.

Say you have applied for two different jobs and both of them have given you an acceptance offer. Which one will you take? Well, you will probably take a bunch of things into consideration: which one pays more? Which one has better benefits? Which one will you enjoy more (or at least hate less)? Which one has the better commute? Which one has better hours? Which one has more opportunity for promotions? And so on.

This long-term planning, about which career you want, is about answering those questions. There are two aspects to each of those questions: the facts presented to you, and the choices you make. We are all aware (sometimes painfully so) that we have no control over the facts presented to us. That job A is offering X amount of money while job B is offering Y amount of money is not in your control; where the job takes place (i.e., what your commute will be) is not in your control. We can therefore say that these facts presented to you are safely out of the realm of free will. But your choices, given these facts, are still under your voluntary control, are they not?

Say, for instance, you choose job A because it pays more than job B and has more room for promotion than job B; and even though job A has a longer commute and the work doesn’t sound as interesting, you would really like that money. Why is it that the money and room for promotion is more important to you than the commute and the kind of work you will do? We could answer this in one of two ways: first, there are extenuating circumstances, such as you having a lot of student debt or something. These would be external factors and therefore not under your direct control, i.e., no free will is really at play (maybe). The second way would be to say that you value money and perhaps status more than you devalue your commute time and enjoyment. But again, the question is: why? Why do you value money over enjoyment? Is that something you chose? And even if you say you did choose to have those values, then why did you value having the values that you do?

You might object to what I said about the first way we answered, that something like needing the money swayed you to job A and therefore the decision was not part of your free will. You might say that, sure, this is an extenuating circumstance, but you still have a choice, there just might be consequences. In other words, you could still choose job B over job A, but then simply have to live on a tighter budget, or even be alright with simply reneging on your student loans and deal with those consequences (or something). Just because the consequences are undesirable does not mean you didn’t still have a choice. The point being here that free will consists of there being counterfactuals: a choice between A and B is freely chosen when, having chose A, it could have been the case that you chose B. Put another way: there is a possible world where you chose B, even though in the actual world you chose A.

The question still remains: if you did choose job B over job A, then what made you value enjoyment over money (and value it so much you were willing to suffer the consequences)? To say that there is a counterfactual world still does not answer why, in the actual world, you have the values and desires that you do. If you chose job B over A in the actual world, then it simply would mean that some other values or desires not chosen by you were at play when you made the decision.

We can also think about it this way (this thought experiment is adapted from “Free Will Remains a Mystery” by Peter van Inwagen (2000)). Say at some time t₁ you make a decision that will lead to either outcome S or outcome W at time t₂. Now lets say God, or whoever is running our universe simulation, or through some weird time loop, that time rewinds back to t₀ prior to the decision time t₁. This rewinding-and-replaying then reoccurs many hundreds or thousands of times, and with each rewind you do not retain any memory of anything that occurred after t₀ for any of the multiple replays. For the sake of argument, we will say that t₀ is sufficiently long enough prior to t₁ for the decision not to be made based on mere whim or due to some sort of decision-making inertia (e.g., you cannot choose to not be in your car at t₁ if you were already in your car at t₀ if t₀ is not sufficiently long enough prior to t₁), but not so long that any sort of intrinsic randomness (e.g., quantum effects) could significantly or unduly alter the decision made at t₁. We will also stipulate that any time t_n with n < 0 and n ∊ ℝ^– is exactly the same, i.e., that nothing prior to t₀ is altered. We then end up with the following probabilities:

P(choose S at t₁) = α

P(choose W at t₁) = β

Such that 0 ≤ α ≤ 1 and 0 ≤ β ≤ 1 and α + β = 1 (the two probabilities are mutually exclusive and jointly exhaustive). We could add any number of other possible decision-outcome pairs at t₁, but doing so will not change our analysis in any substantive way (or, as the mathematicians might say, we will not lose generality by focusing on a binary decision). After, let us say, 10,000 rewind-and-replays, one of the following three things will occur:

1.      α = 1 and β = 0
2.      α = 0 and β = 1
3.      0 < α < 1 and 0 < β < 1 (still with α + β = 1)

The first two cases would be a very hard determinism in which you could not have chosen to do otherwise. The third case says there is some non-zero but not absolutely determined probability α of doing S and likewise a probability β for doing W. You might think that this means the third case is an example of free will – had the rewinding thing never happened, it really was the case that you could have chosen to do otherwise than what you had actually done, i.e., there is a possible world (represented by the rewind trials in this thought experiment) where you had done otherwise than what you did in the actual world.

There are two issues with this interpretation. The first is that this seems probabilistic (i.e., it is chance). Lets say, for the sake of argument, that our probabilities converged on α = 0.70 and β = 0.30; what accounts for this distribution? Why is it that you had a 70% chance to choose S and a 30% chance to choose W? That there is such a distribution means that there is some cause for why choosing S was more likely than choosing W. Your propensity for S was greater than your propensity for W, which cannot be explained by the free will hypothesis. More than that, being probabilistic simply does not entail free will, lest we attribute free will to all probabilistic phenomena in the world (e.g., that, in the northern latitudes, there is a higher probability that it snows on any given day in January than in March is not due to free will, but the tilt of the earth’s axis and its revolving around the sun, which are very deterministic phenomena).

Then, you might argue, if there is free will, then perhaps what we would observe is that the probabilities converge on α = 0.50 and β = 0.50, i.e., a 50/50 chance of choosing S or W. But this brings me to the second issue: this seems random. Randomness, just as much as probability, certainly does not entail free will: it is not you choosing to do S or W because the decision is just random chance. It is not under your control nor the control of any other causes. A choice made at t₁ is either determined by something – something has occurred that caused you to make a choice at t₁ – or it is mere chance or randomness. Meaning that the choice made at t₁ is either due to determinism or indeterminism, neither of which are free will.

Thus, the best a proponent of free will can say is that a possible counterfactual, as in case 3 from above, may be a necessary condition for free will, but it is certainly not sufficient.

Now maybe you think: could religion get us out of this predicament? This stuff about determinism and indeterminism seems to presuppose a naturalist or materialist metaphysics. Couldn’t something of an immaterial substance, like the soul, be the agent of causation? Maybe there is some third way aside from just determinism and indeterminism?

There are two issues with this, one of which is theological and the other scientific. The theological problem has to do with God’s omniscience: if God already knows everything that everyone who will ever exist will do in the future, then what I am going to do is determined. Determinism entails that, in principle, everything that will, in fact, happen is predictable before it will ever actually happen, given enough information about the present. A being that is omniscient is, by definition, a being that has all of the information about the present (and the past, and possibly the future). Thus, if a being with all of the information can predict with 100% accuracy everything that will occur in the future, then the future is determined. We could put it syllogistically like this:

P1: if free will exists, then the future cannot in principle be predicted with arbitrary accuracy
P2: an omniscient being can in principle (and by definition) predict the future with arbitrary accuracy
C: therefore, free will does not exist (if an omniscient being does exist) (modus tollens)

Indeed, the favored argument of theists, the Kalam cosmological argument, takes as a premise that everything that begins to exist has a cause. Any decision I make begins to exist (there was a time before that decision existed and then a time after which I have made the decision, thus the decision (the act of my will) began to exist) and so, based on this premise, it must be that my decision is determined by some kind of cause.

I am aware that there is more nuance to God’s omniscience, but that is beyond the scope of this post. If you want to read more on that, check out this post of mine (I also discuss the Kalam cosmological argument in that post, among other things).

The scientific issue is that, if free will exists, then that means something is breaking causality in the physical world. This is something that could, at least in principle, be measured. It would appear like an uncaused force acting on the material world, some ghostly F = ma acting upon physical particles that could be measured. Even if we could not directly access the immaterial or spiritual substance (the agent causation) doing the uncaused causing, we could measure its effects in the material world. (You might object to my use of the term uncaused causing here, since this is a power usually ascribed only to God, but the alternative would be that your will was still caused, thus determined, just by something non-physical; thus, if we have free will, our will would need to be an uncaused agent causation). And so, the best case scenario for the proponent of free will is that free will is a falsifiable hypothesis that is, so far, unverified.

Perhaps at this point you will throw up your hands and say “so, if we don’t have free will, then how come it feels like I have free will? What possible reason, whether theological or scientific, could there be that accounts for the incorrigible feeling of having free will?”

My response would be this: you might be convinced that you have free will, but you never feel like you have free will. Going back to how this post began, I asked you to think about an animal, and one came to your mind. What did it feel like when that happened? It felt like the thought just arose within your mind, as if it just came out of nowhere. Again, even if you disagree with this, then the alternative would be that you, in some way, browsed through your memory of all possible animals and then, based on some criteria, chose the one you did. But what are those criteria? And why do you have them? Why did you choose those criteria over any other criteria? Most likely you did not choose those criteria, but even if you did, your choosing of those criteria was done based on some other criteria, your adherence of which you had no choice. The point is, there comes a certain point in the regress where there are things about yourself – desires, preferences, tendencies, values, beliefs – that you did not freely choose. Not only that, but it never even felt like you chose them.

Finally, you might then ask: “okay, even if I concede that it does not feel like I have free will, then why am I convinced that I do?” This is an interesting question that can be viewed through multiple lenses. I’ll look at it in two ways.

The first lens is anthropological: is being convinced of having free will a cultural artifact? That is to say, are there now, or has there ever been, cultures that did not believe themselves to have free will? I actually do not know the answer to that. Yet there do seem to be a lot of cultures that posit various spirits, ghosts, daemons, djinn, witches, hexes, jinxes, spells, prayers, vibes, ancestors, gods, karma, fates, destinies, prophecies, “everything happens for a reason”, astrologies, synchronicities, collective consciousnesses, subconsciousnesses, and so on that have powerful (perhaps even deterministic?) influences throughout our lives. Even many in the Judeo-Christian tradition posit spiritual forces – angels, demons, noetic effects of sin, predestination and preordination, and even interventions by God Himself – influencing our lives in various ways. Secular ideologies, such as Marxism and modern Woke ideology, also posit forces beyond agent causation as determining our thoughts and behaviors (e.g., historical materialism, structural racism, and so on).

The second lens to look at this question – why are we humans convinced that we have free will if it is in fact that case that we do not – is evolutionary. If we assume that being convinced of our free will is something humans have inherited through biology (as opposed to culture), I would hypothesize that it has to do with unpredictability. I’ve said in other posts on this blog that free will is a useful fiction since no single person or committee of people will ever be able to predict the future with 100% accuracy (often we are lucky to do better than mere chance). Thus, free will is sort of allowing for error bars on our ability to predict the future. The human brain evolved primarily for prediction, and so convincing ourselves that we can act freely is conceding that we do not know how we (or the world around us) will behave in the future. Positing some kind of “will” or “freedom to act” is an account (or perhaps personification) of this unpredictability.

Concluding Remarks

What might a world with free will even look like? We probably wouldn’t be able to talk about personalities, since all that is is a tendency to behave and think in certain ways. If we had free will, such a thing would be meaningless, because every new decision I am faced with I would be free to choose, not constrained or determined by my personality (whether personalities are nature, nurture, or some mixture of the two). Indeed, a personality attests to something about oneself that is not freely chosen but that has an influence on the kinds of decisions I will predictably make.

Proponents of determinism are often asked how things like justice could be carried out if all our choices are determined. I think the question could be turned back on the proponents of free will: if we are able to freely make choices, why not simply ask criminals not to commit a crime in the future and let them go free? They have the freedom to make the choice not to commit crimes, so why are people who have committed crimes in the past statistically more likely to commit them in the future? If it really was all freely made choices, we should not lock anyone up because we lose any sense of persistent identity on the free will hypothesis, because what could we say is preserved through time if not for all the constraints on our ability to freely choose (e.g., our personality)? Put another way, if person A commits a crime at time t₁, the fact that they committed that crime at time t₁ does not determine their choices at any future time t₂, since they are completely free to make a different decision at time t₂ (the person at time t₂ could simply make the choice not to be the person they were at time t₁). If we think about it like Roderick Chisholm, we are looking at the ens per se and ens per alio:

The collection of parts that exists at a moment is an “ens per se” — it exists in its own right; but the ship [of Theseus] is an “ens per alio” — it exists in virtue of the fact that some other things exist.

If person A has free will, then their will exists ens per se, since their will does not exist in virtue of the fact that something else exists (e.g., their personality, or upbringing, or genetics, or whatever other causal thing determines their decisions). As such, just like Chisholm’s analysis of the ship of Theseus, switching one thing out makes it something different – person A making a decisions at t₂ makes them a different person, because the decision was separate from past choices due to not being caused by them. Each free choice a person makes is essentially like switching out a part of themself.

Indeed, the fact that person A committed a crime at t₁ at all makes no sense under the free will hypothesis, because nothing could have compelled them to make the decision to commit the crime, since it was a completely free choice – the only way to explain why person A committed the crime would require bringing in non-free-will variables, such as the person’s circumstances (physical, cultural, economic, etc.), their personality (and propensity for committing crime), mental/emotional issues (e.g., psychopathy) or physiological imbalances in the brain (e.g., caused by a tumor), their sinful nature or being influenced by demons, and so on.

What it comes down to is that free will has no explanatory power. If someone asks why a person did something, the proponent of free will could only answer that the person made a free choice. When asked what that means, the proponent of free will has no answer – it is just a brute fact that the person chose to do something by exercising their free will. Any explanation beyond that requires putting constraints on the person’s free will – if the person has a propensity for doing something, this attests to some causal influence (their upbringing, genetics, neurobiology, circumstances, etc.) that was not freely chosen. The better the explanation we want for some behavior, the more constraints we will need to put on that person’s free will. Determinism is simply saying that, as we try to get more and more accurate in our explanation (or prediction), those constraints will keep adding up until free will vanishes.