Are we able to do otherwise?

In a nearby possible world, where free will— that is, the ability to do otherwise — is considered necessary for moral responsibility, physicists manage to undeniably prove that the thesis of determinism is true. (Don’t ask me how) In this world, Maxim de Winter kills his wife. Surprisingly or perhaps not, he does not deny it. He calmly calls the police and waits patiently beside the body. At trial, his lawyer does not dispute the facts either. He confidently agrees that Maxim de Winter did in fact kill his wife, but insists he shouldn’t be held responsible. After all, determinism is true, and consequently his client could not have done otherwise. He turns to the judge: “Your honor, as you know, it has now been established that determinism is true. And if that’s the case, then “a complete description of the state of the world at any time together with a complete specification of the laws entails a complete description of the state of the world at any other time.“ Given the state of the world prior to the murder, together with the laws of nature, there was exactly one possible state it could be in at the next moment. My client’s decision to kill his wife was therefore entirely determined by the prior state of his brain — which was itself determined by states of the world long before he was born, in conjunction with the laws of nature. It follows that he could not have done otherwise. If determinism is true, then he was not able to refrain from killing his wife; and if he could not refrain, he was not free. For he had no control over the laws of nature and no control over the past.” He turns to the jury: “Since my client did not act freely, he cannot be held morally responsible for what he did. And if that is true of Maxim de Winter, it is true of every person who has ever stood in this court.”
To formulate the lawyer’s argument (LA):
1) Determinism is true.
2) If determinism is true then Maxim wasn’t able to do anything other than kill his wife.
C1) Maxim was not able to do otherwise.
3) If Maxim was not able to do otherwise, then he wasn’t morally responsible for killing his wife.
C2) Maxim wasn’t morally responsible for killing his wife.

According to premise (2) of the LA, if determinism is true then Maxim’s action was entailed by the past and the laws of nature — neither of which he had any control over — he was never genuinely able to act differently. The main support for premise (2) comes from Van Inwagen’s influential Consequence Argument. He writes: “If determinism is true, then our acts are the consequence of laws of nature and events in the remote past. But it’s not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us.”
So if determinism is true, we never have the ability to do otherwise. As Van Inwagen notes: “And this implies that if determinism is true (if the past and the laws of nature determine a unique future), [...] then free will simply does not exist: no one is ever able to do anything other than just exactly those things that he or she does.” ¹
In what follows, I will argue that the conclusion does not follow from the premises. Even in a deterministic world, we are still able to do otherwise.

Free will is one of those commonsense beliefs that seem hard — absurd, even — to deny the reality of. One of that bundle of Moorean facts. We also seem to exercise this ability constantly: we deliberate over different courses of action, weigh our options, and act on the basis of reasons. And we take for granted a whole range of other abilities besides, such as the ability to speak French or the ability to play the piano or the ability to decide. So, the fact that we have abilities is uncontroversially compatible with determinism. A deterministic world does nothing to undermine that I am able to play the piano or that I am able to jump.
Suppose that at the moment, I am not playing the piano. Nonetheless, there is a clear sense in which I retain this ability. I endured years of practice as a child, sat through countless hours with a piano teacher, and in doing so added it to my set of already existing abilities. In other words, I have what it takes to play the piano even when I am not currently doing so. In the same way, Maxim had the ability to refrain from killing his wife. He retained the ability to walk away and do otherwise, even if he did not exercise it.
A plausible objection here is that Maxim did not have the ability to do otherwise and refrain from killing his wife — after all, the laws and the past entailed that he would decide as he did. But notice that a person can retain an ability even when they do not, and perhaps cannot, manifest it. Consider the following example:
suppose that I am locked in a room with no piano in sight. Again, there is a clear sense in which I retain the ability to play. What I lack is not the ability itself but the opportunity to exercise it. My surroundings prevent me from manifesting one of the abilities in the set. The ability remains—however, the circumstances simply do not permit its manifestation.
It seems then that Maxim is in the same position. He had the ability to do otherwise and hence refrain from killing his wife; he simply lacked the opportunity to manifest it. His circumstances, the facts about the past and the laws of nature, prevented him from doing so. And if this is right, then it seems we have succeeded in showing that Maxim was able to do otherwise after all.
It’s clear that the ability to not kill her and also the ability to form that intention were in his set of abilities. He was in good health, had no relevant mental disorder or compulsive condition, and was able to form intentions and act on reasons. He had what it takes to refrain from committing the murder.
As you may have noticed this is too quick. If having the ability to play the piano in a locked room is sufficient to say that I was free to play it, then freedom requires nothing more than the mere possession of an ability. However, I was obviously not free to play the piano in a room with no piano. The opportunity was entirely absent. And so the ability, on its own, is not enough. The same point applies to Maxim—possessing the ability to refrain from killing his wife is not, by itself, sufficient for freedom.

If we take a step back, it is now evident that there are two senses of ability.
When I am locked in the room, I possess what Kadri Vihvelin calls a narrow ability — the relevant skills, competence, and know-how required to play the piano. But my surroundings deprived me of the opportunity to exercise this narrow ability. To have a narrow ability to X together with the opportunity to X is what Vihvelin calls, the wide ability. What I lack in the locked room is the wide ability: the narrow ability is present, but the conditions necessary for its expression are not. Acting freely, it now seems, requires both the narrow and the wide ability to do otherwise.

So the sense in which the lawyer claims Maxim could not do otherwise must mean that he lacked the wide ability — for it is obvious that he possessed the narrow one. He had what it takes to decide otherwise. The more salient question is whether his circumstances, specifically the laws of nature and the facts of the past, deprived him of the opportunity to exercise that capacity, just as the locked room deprived me of the opportunity to play.

Now, it appears that there is a clear asymmetry between the locked room and Maxim’s case. I lacked the wide ability in the room because something external blocked the expression of my narrow ability. Maxim, on the other hand, lacked nothing of the kind. He had the narrow ability to decide otherwise, and nothing in his surroundings prevented him from exercising it: he wasn’t held at gun point, no one was forcing his hand to commit the murder. He, therefore, had the wide ability too. The opportunity to do otherwise was available to him all along. He simply did not take it. If this is right then, he acted freely—he had both the narrow and wide ability.

Drawing on this distinction, we can now offer the following account of ability². An agent is able to X (or is free to X) if and only if two conditions are met:
(i) The narrow condition: both the ability to X and the ability to form the intention to X are within the agent’s set of abilities. That is, an agent has the narrow ability to X and the narrow ability to form the relevant intention.
(ii) The wide condition: if the agent forms the intention to X, he would X. That is, his circumstances are sufficiently favorable for the intention to issue in action — nothing in his surroundings impede the move from intention to action.

For example, in the locked room scenario. I satisfy condition (i): playing the piano and forming the intention to play are both within my set of abilities. I fail condition (ii): even if I form the intention, I would not play — the piano is absent and the room is locked. Conversely, Maxim’s case is different. He satisfies condition (i): the ability to refrain from killing his wife and the ability to form the intention to refrain are both within his set of abilities — further, he is in good health, has no relevant mental disorder or compulsive condition, and is capable of deliberating and acting on reasons. He also satisfies condition (ii): if he forms the intention to refrain he would do so, nothing in his circumstances would have prevented him from doing so. Both conditions are met. If this is what it mean to be able to do otherwise, then Maxim had the ability to refrain from killing his wife — and therefore acted freely.

Addressing some potential counterexamples:
One might object that this is just a reiteration of the classical compatibilist conditional analysis of free will, according to which, to say that an agent is able to X is to say that if she intended/decided to X, then she would have done so.
Her ability to have done otherwise is constituted by the truth of this counterfactual. But as you may know, the classical analysis faces various counterexamples:
“Suppose that Danielle is psychologically incapable of wanting to touch a blond haired dog. Imagine that, on her sixteenth birthday, unaware of her condition, her father brings her two puppies to choose between, one being a blond haired Lab, the other a black haired Lab. He tells Danielle just to pick up whichever of the two she pleases and that he will return the other puppy to the pet store. Danielle happily, and unencumbered, does what she wants and picks up the black Lab.”

In this example, Danielle is unable to pick the blonde Lab but the classical compatibilist analysis mistakenly grants her this ability. Since the counterfactual “if she wanted to pick up the blond Lab, then she would have done so”, is true. So, compatibilists were right to abandon it. However, the account I provided does not seem to fall to this counterexample. Danielle, at the moment of action, lacks the ability to form the intention to pet the dog. That is, at the time, the ability to form this intention is not in her set of abilities. From the outset she fails to meet condition (i).

Let’s consider another example by Randolph Clarke (2019):
“Ordinarily Sam can raise his right arm as well as any of us. Not so today. A key nerve bundle leading from his brain to his shoulder, though connected, is loose. Were Sam to try to raise his right arm, the nerves would disconnect, and he would fail to move his arm.”

Here, Sam meets the narrow condition: both the ability to raise his arm and the ability to form the relevant intention are in his set of abilities. But he fails the wide condition, if Sam forms the intention to raise his arm at this moment, he would not raise it — the nerve bundle would disconnect and the action would fail. The conditional in condition (ii) is evaluated against Sam’s actual circumstances at this moment — including the loose nerve — and it comes out false.

The lawyer’s argument, therefore, fails not because determinism is false — we have granted it freely (pun intended). It fails because the inference from determinism to not being able to do otherwise does not go through. There are two senses in which a person might lack the ability to do otherwise: they might lack the relevant skills, and capacities — the narrow ability — or they might possess all of those and yet be blocked, externally, from exercising them — lacking the wide ability. The lawyer’s argument needs Maxim to lack the wide ability. But he does not. His surroundings were favorable. His abilities were intact.
If the proposed account of ability is correct then Maxim de Winter was actually able to do otherwise and not kill his poor wife. The jury, I submit, should convict.

Sources:
Vihvelin, K. 2013. Causes, Laws, and Free Will: Why Determinism Doesn’t Matter.New York: Oxford University Press.
van Inwagen, Peter (1991). The Consequence Argument. In Peter van Inwagen & Dean W. Zimmerman, Metaphysics: The Big Questions. Oxford: Wiley-Blackwell.
Vihvelin, Kadri, “Arguments for Incompatibilism”, The Stanford Encyclopedia of Philosophy (Fall 2022 Edition), Edward N. Zalta & Uri Nodelman (eds.),
van Inwagen, P. 1975. ‘The Incompatibility of Free Will and Determinism’,Philosophical Studies, 27: 185-99
Finch, Alicia and Ted Warfield, 1998, “The MIND Argument and Libertarianism”, Mind, 107(427): 515–528. doi:10.1093/mind/107.427.515
McKay, Thomas J. and David Johnson, 1996, “A Reconsideration of an Argument against Compatibilism”, Philosophical Topics, 24(2): 113–122. doi:10.5840/philtopics199624219
McKenna, Michael & Pereboom, Derk (2014). Free Will: A Contemporary Introduction. New York: Routledge. Edited by Derk Pereboom.
McKenna, Michael and D. Justin Coates, “Compatibilism”, The Stanford Encyclopedia of Philosophy (Summer 2024 Edition), Edward N. Zalta & Uri Nodelman (eds.),
Clarke, Randolph (2019). Free Will and Abilities to Act. In Martin Breul, Aaron Langenfeld, Saskia Wendel & Klaus von Stoch, Streit um die Freiheit – Philosophische und Theologische Perspektiven. Schöningh. pp. 41-62.

1. Below is a more precise formulation of the argument and it avoids the old rule Beta which entails Agglomeration (McKay and Johnson 1996).
-P expresses a proposition about the total state of the word at a given time t in the remote past.
-L expresses a proposition about the laws of nature.
-□((P&L)→ Q): it is metaphysically necessary that Q is implied by the conjunction of P and L.
-Np: no one is able to render p false.
-Inference rule: β-box: from Np, □(p → q) ˫ Nq

1. N(P&L)

2. □((P&L)→ Q)

3. NQ from 1, 2 by β-box

2. Read this for suggested corrections by Tower of Babble: