alincucu.de | Is determinism compatible with free will?
This is the philosophical homepage of Alin Christoph Cucu.
philosophy, alin cucu, soul, mind, brain, philosophy of religion, philosophy of science, christianity, faith
15845
post-template-default,single,single-post,postid-15845,single-format-standard,op-plugin,ajax_fade,page_not_loaded,,vertical_menu_enabled,side_area_uncovered_from_content,qode-theme-ver-16.6,qode-theme-bridge,disabled_footer_top,wpb-js-composer js-comp-ver-5.5.1,vc_responsive

Is determinism compatible with free will?

Free Will Determinism

Is determinism compatible with free will?

Abstract

I will argue in this essay that free will is incompatible with determinism. My argument focuses on the causation of mental events in contexts of temptation. A temptation occurs when a strong impulse in a person occurs that runs counter to the moral convictions of that person. If determinism is true, then it is impossible to resist the temptation. However, if free will is to be understood as constitutive for moral accountability, it depends on the possibility of resisting temptation. I will also consider some objections against my construal of determinism and my construal of free will, finding them all wanting.

I

The question whether determinism and free will are compatible touches on the very core of our self-concept as human beings. It is no coincidence that people have fancied that, if determinism is true, we are to be likened to billard balls, marionettes or swimmers in a stream that irresistibly sweeps us away. Kadri Vihvelin (Vihvelin 2017, p. 1) writes:

We believe that we have free will and this belief is so firmly entrenched in our daily lives that it is almost impossible to take seriously the thought that it might be mistaken.

A quick glance at the definitions of determinism and free will suffices to understand why such incredulity is so widespread. To be sure, there are different definitions of both terms in use by philosophers, but they all do not differ fundamentally from each other and from mine. By determinism (D) I understand the doctrine that

(D) Every event is conditionally necessitated by prior events and circumstances.

In other words, if certain events and circumstances are conducive to the occurrence of an event E, it is impossible that E not occur. By free will I mean

(FW) Free will = df: The kind of control over mental events necessary for moral accountability.

It is worth noting that my definition focuses on mental events rather than actions. I take this to be an advantage, because it avoids the tedious discussion about “Frankfurt cases”, i.e. whether “being able to do otherwise” is necessary for moral responsibility. Still, my approach naturally extends to actions, because intentional actions follow from mental events: if one has control over one’s mental events and no external impediment or manipulation occurs, it follows that one also has control over one’s actions. Why is it adequate to factor out external impediments and manipulations from the definition of free will? It is because it is adequate to consider a person who is tied to a chair and gagged (so that she can neither move nor talk) possessing free will, as long as she still has control over her mental events. I cannot see how tying and gagging a person robs her of free will. On the other hand, it is reasonable to consider a person heavily manipulated (e.g. by a mad brain scientist who can arbitrarily bring about thoughts by manipulating the brain) to be devoid of free will as long as the manipulation persists.

By “control over mental events” I mean the ability to see to it that a mental event is brought about, even in the presence of counteracting forces. To make this clear, consider the following situation. Imagine a burglar breaks into your house. You surprise him, but he knocks you unconscious. When you awake, you’re tied to a chair, gagged, watching the burglar ransacking your valuables. Obviously, you can neither stop him nor talk to him nor call the police. You are left alone with your thoughts. Let’s assume you believe that hating another person is wrong, no matter what that person does. You are strongly tempted to give in to hate. But you can also control your thoughts, and, while of course disapproving of his deeds, refrain from hating him and instead pray for him to realize his folly. That is what seems to me free will boils down to. No (physical) action is required for it, but actions can of course issue from it.

For what I’ve been using the rather bulky term “counteracting forces” we might borrow a well-established term from religion: temptation. Temptation is, in my view, the occurrence of a state of affairs conducive to bring about a morally bad mental state in a person. By a morally bad mental state I mean one that runs counter to one’s moral beliefs, not necessarily one that contradicts objective moral standards. Thus, even moral anti-realists can accommodate this model.

With all this in mind, I want to suggest the following, honed definition of free will:

(FWP) = Human beings have free will if in at least some cases some person (P) intentionally brought it about that, given that P was confronted with a temptation T leading to MEB, in P’s mind MEB was not realized but instead a morally good mental event MEG.

Or, a little more illustrative:

This definition is inspired by Daniel von Wachter’s directedness theory of causation (Von Wachter 2009, ch. 5). The directedness theory is an alternative to determinism as depicted by (D). Its heart is the assumption that an entity S (physical or mental) and its circumstances C together form a base B that possesses certain directednesses to behave in certain ways. For example, a book (S) held above the ground has the directedness to fall, the base (B) of which is grounded in the book having a mass and gravity acting upon it (C). In space, it wouldn’t have that tendency, because gravity is much weaker (S is equal, but C is different). When the book is supported by my hand, it still possesses B and hence its directedness to fall, but my hand holding it constitutes another base B’ counteracting B and thus preventing B from being realized.

Von Wachter’s theory allows for predictable causality, but also for interventions of, for example, free agents. Thus, a cause only brings about its effect so long as nothing interferes with it. And interference is exactly what we’re looking for when constructing a theory of how people can resist temptation.

An advantage of my case-oriented definition is that it avoids arguments about what “ability” exactly means. It instead aims at concrete cases in which a person manages to resist temptation, or, in other words, mentally manages to prevent a directedness from being realized and instead realizes another directedness. The occurrence of such cases suffices to refute determinism, because determinism requires that a directedness necessarily be realized.

                                                       II

In what follows I will argue that (D) and (FWP) are incompatible. I call my argument the Temptation Resisting Argument for incompatibilism (TRA):

[TRA]

  1. If (D) is true, any temptation T leading to MEB in circumstances C causally necessitates a morally bad mental event (MEB).

[(T & C ® MEB)]

  1. (FWP) = Human beings have free will if in at least some cases some person (P) intentionally brought it about that, given that P was confronted with T in circumstances C, in P’s mind MEB was not realized but instead a morally good mental event (MEG).

[(T & C (Ø ® MEB)) & (P ® MEG)]

  1. Therefore, if (D) is true, we have no free will. (From 1, 2)
  2. Therefore, (D) and (FW) are incompatible. (From 3)

Premise 1 is a straightforward derivation from (D). If indeed some prior events E and circumstances C causally necessitate an event E’, it is unavoidable that some temptation event T cause a morally bad event MEB. We only need one uncontroversial assumption to derive premise 1 from (D), namely that there is some causal connection between T & C and MEB (and not just a correlation). If in one case T & C bring about MEB because of a causal connection, then, according to determinism, T always brings about MEB. But can it be true that there is a causal connection between a physical state of affairs (T) and a mental event (MEB), given that T and MEB belong to different categories of events?

I think it can. First of all, it is phenomenologically what we experience. I see a wallet full of money on a park bench (T); no one is around. Suddenly the thought of snatching it pops up in my mind. It seems clear that the event of the wallet lying on the bench together with the circumstances C (e.g. no one is watching) is somehow causing my belief (MEB) that I should keep it. Second, the tracking of the causal chain renders it plausible that T & C cause MEB. Sense perception yields information about the world (the lone wallet, no one around) which in the brain triggers other belief states to be caused or come to awareness, i.e. having entered the realm of the mental. You do not have to be a substance dualist to believe that. Note that I am here not claiming a causality pointing from the mental to the physical, but the other way round. I think this is quite uncontroversial.

Now, according to (D), if T & C are causing MEB once, they cause it every time. Maybe you do not feel tempted to commit theft at the sight of a lone wallet; substitute whatever situation fits you. The point is that it is reasonable to assume a causal connection between any T & C and a corresponding MEB, and determinism leverages this singular connection to a necessary one.

What are we to do with cases in which T & C seemingly do not bring about MEB? David Hume took up this question. In his Enquiry Concerning Human Understanding, he defends the truth of determinism and explicitly extends it to human behavior. However, he allows for deviations:

We must not, however, expect that this uniformity of human actions should be carried to such a length as that all men, in the same circumstances, will always act precisely in the same manner, without making any allowance for the diversity of characters, prejudices, and opinions. (Hume 1975)

Translated into my terminology, Hume accounts for different actions (and, presumably, precedent mental events causing those actions) by assuming different C. In other words, T1 together with C1 (among which is a person with bad character) causes MEB, but T1 together with C2 (among which is a person with righteous character) causes MEG. There is no exception to determinism, quite in accordance with what Hume wanted to show.

Premise 2 reflects the construal of free will as depicted in section I, so I shall not bother to expound on it anymore.

Premise 3 is a valid conclusion from 1 and 2. To repeat, determinism does not allow any exceptions; if T & C cause MEB at one instance, they must necessarily cause it in all other instances.

III

I will now consider some possible objections and reply to them.

Against premise 1

Against premise 1, it could be objected that modern physics has shown that there are indeterministic quantum events. If those happen in our brain in a causal chain pertinent to a mental event experienced as the expression of free will, our brain events and thus our mental events would not be fully deterministic. Imagine at least one neuron in a causal chain from T leading to MEB fires indeterministically, i.e. even though it is excited by a precedent neuron, it fires with a probability p < 1. This means that in some cases, the neuron fires, whereas in other cases with the same T and C, it does not fire. Thus, it would not be true that a certain tandem of T & C necessarily brings about a certain MEB.

It must be noted first of all that this approach breaks ranks with classical determinism. This might be taken as a welcome opportunity to align determinism with modern physics. Indeed it seems that quantum physics has made the assumption of indeterministic events inevitable. We might therefore mend (D) to a version of determinism updated with quantum mechanics:

(DQ)   Every event is either conditionally necessitated by prior events and circumstances or the prior events and circumstances necessitate its occurrence with a probability p of 0 < p < 1.

However, deterministic views of quantum events persist (see e.g. Goldstein 2017) . So it is not even clear that we can take the existence of indeterministic events for granted.

But let us arguendo grant their existence. Some philosophers (e.g. Swinburne 2013) have argued that the occurrence of indeterministic processes in the brain opens up the possibility for free will in an otherwise deterministic world. But even if we allow indeterministic events in the brain, this does not solve the problem. Let’s reformulate premise 1 accordingly:

1’. If (DQ) is true, a temptation T in circumstances C causally necessitates a morally bad event or necessitates a morally bad mental event with 0 < p < 1.

In connection with premise 2, this still yields an incompatibility of determinism and free will. Indeterministic events in the brain do no entail a possibility for free will, but the opposite. A brain event with an uncertain outcome does not give its bearer more control over his mental events, but rather divests him of all control at this point!

There is another way of trying to reconcile free will with determinism via indeterminism, but as it affects premise 2, I shall discuss it there.

Against premise 2

Several objections can be raised against premise 2. Naturally, they all aim at construing free will differently.

Indeterministic readings of free will

The first alternative construal is to bring indeterminism into the picture. It might be suggested that quantum events in the brain could be the point at which a free agent takes influence on a stream of events the rest of which is fully deterministic in the sense of (D). We might formulate it as a modified version of premise 2:

(FWPi) = Human beings have free will if they can bring it about by use of their willful capacities that indeterministic quantum states in the brain are being “pushed” in one direction, thereby overriding the inherent probabilities.

This idea is similar to Alvin Plantinga’s Divine Collapse Causation (Plantinga 2011, Kindle location 1784). The hope is to salvage determinism while allowing for free will in the sense of “origination” (cf. Honderich n.d.), according to which a person is the ultimate source of its actions and thoughts. But it does not attain a compatibility between determinism and free will. Apart from the fact that it denies (D), (FWPi) amounts to a more spelled out version of (FWP); in fact, it renders a possible mechanism through which P might bring about MEG instead of MEB! Whether or not I bring about the thoughts I wish via collapse of some wave function or directly by influencing the matter of my brain, I am still free in the sense of (FWP). It is, however, in neither case a freedom compatible with (D), not even with (DQ). (D) obviously precludes any indeterministic events, and (DQ) denies the person any control necessary to fulfill the requirements of (FWP). I take it, therefore, that no indeterministic approach can provide compatibility between determinism and free will.

Free will as doing what one wants

The second alternative is the classical “desires”-approach to compatibilism as proposed by Thomas Hobbes (1655) and David Hume (Hume 1975). According to Hobbes and Hume, one has free will if one does what one wishes or desires to do. On this picture, resisting temptation is irrelevant to free will. If the temptation coincides with what one desires, one is expressing one’s free will in giving in to the temptation. To fit Hobbes’s and Hume’s action-oriented approach to my mental event-oriented one, I assume that Hobbes and Hume would have agreed that in order to perform an action, one has a corresponding mental event first.

The problem with this approach is, of course, that many people report to not want what they desire. Take the desire to eat junk food. Many people say they feel such a desire or rather craving, but they would rather not follow it, because they know that it’s unhealthy or because they feel physically bad after having eaten such food or both. And some report they mentally manage to push back the unhealthy desire and replace it with a more positive thought like “My body deserves healthy food”. But this is exactly what (FWP) expresses. On the desires theory, such a mental overcoming is not possible. Given the fact that it exists, the desires approach fails in providing an adequate concept of free will.

Free will as having been able to do otherwise

As a third way out of the free will dilemma, a more recent compatibilist theory says that free will consists in an agent having been able to do otherwise. This, the proponents of this approach claim, is compatible with determinism. Assume person P does action X (or in my terms, has the mental event MEX). The crucial point is, had P wanted to bring about MEY instead of MEX, she would have brought about MEY. Of course, as the determined history of the past was such as to bring about P’s wish to bring about MEX, she brought that about; but given another past, MEY would have been equally possible. Again, this definition of free will makes resisting temptations in the above sense obsolete; it suffices that the past could have been such as to make a person bring about MEG instead of MEB.

To me, this sort of “freedom” rings hollow. What freedom is it if for the moment of the occurrence of a mental event, it was determined by the past that exactly this mental event had to happen? The person as controller of her thoughts is completely out. Furthermore, it undermines moral responsibility. If it depends solely on the past which mental events (and hence, which actions) I perform, there is no way of holding me accountable for any thought or action, which is tantamount to denying the very essence of free will. Hence, this approach does not provide any compatibility between free will and determinism either.

Free will as the will one wants

Fourth, one might try making free will compatible with determinism by claiming that

(FWH) one acts of one’s own free will if and only if one’s action issues from the will one wants.

That is, in summary, Harry Frankfurt’s hierarchical theory. Frankfurt distinguishes between first-order desires (like eating food) and second-order desires (desires about desires, e.g. desiring to have more desire to eat healthier food). The second-order desires also serve to adjudicate between conflicting first-order desires. Now, in Frankfurt’s theory, the desires a person has are part of her psychological “mesh”. For Frankfurt, the person need only act in accordance with this “mesh”, i.e. in accordance with her second-order (or third-order, fourth-order, etc.) desires in order to exercise free will.
Several objections can be raised against this. First, an infinite regress is impending. Why? Because if second-order desires are in conflict, the person needs to shift to a higher level (i.e. third-order desires) to resolve the conflict. But what if the desires at that level are also in conflict? Then the next level is due, and then the next, ad infinitum. Second, Frankfurt’s theory does not take into account how the person came to have her mesh. It is possible that her psychology was heavily manipulated beforehand, e.g. by a mad scientist who implanted into her brain a chip through which he can determine her desires on whatever level. On Frankfurt’s theory, such a person must still be considered as having free will, although this analysis runs counter to our intuitions. But this objection goes even deeper, thereby sounding the death knell to Frankfurt’s theory. The manipulation case is, in my view, an apt description of how desires are caused in a deterministic world. In a deterministic world, our innate psychological mesh is determined by precedent events and circumstances, and so are the initial decisions that flow from it. Even if those decisions change the mesh, they are themselves produced by the mesh, and so there is no room for freedom. However, I believe Frankfurt was on to something. His theory requires a “self” that (even though he does not say how) determines what one’s will is, determines that desire#1 (ME1) not be realized and instead desire#2 (ME2). So, in way, it seems one might take his theory further by claiming something like the possibility of interaction depicted by the directedness theory of causation. This, however, means breaking ranks with determinism altogether, because then the occurrence of ME1 cannot be causally necessitated. But even if Frankfurt’s theory could be converted into an incompatibilist theory, it would still be wanting, because it does not leave space for the consideration of rational reasons. In fact, why should we call a person morally accountable who can only (freely) adjudicate between desires, but without reference to reasons? The choice of desires would seem arbitrary, but if that’s all there is to our free choices, I cannot see how we can blame someone for having chosen MEB rather than MEG.

Free will as responsiveness to reasons

The fifth and last alternative account of free will I will look at sets out to tackle precisely the problems depicted in the last paragraph. This account has come to be known as the reasons-responsive view of compatibilism. Its core contentment is that a person has free will if, in a given situation, she reacts to an appropriately broad range of reasons. For example, if Ernie is in his office writing an article for his boss, and Bert comes in, asking him to help him out with a computer problem, Ernie certainly will interrupt his work and help Bert out. This is because helping Bert is a good reason (R) and therefore, if he follows it, Ernie is acting of his own free will, even if determinism is true. Or, to align the case to my focus on mental events, Ernie brings about the mental event MEG of adopting R, MEG having the content that Ernie wants to help Bert and interrupt his work.
This approach adds the important ingredient of reasons and gets somewhat closer to what I consider moral accountability. Although I haven’t mentioned it explicitly, reasons figure importantly in the background of my free will account. It is a (moral) reason that motivates me to bring about MEG instead of MEB. At first glance, this might sound very similar to what the reasons-responsive view says. But I think it is not. It seems that on the reasons-responsive view, I am causally determined by a reason to think accordingly (leaving aside the debate about whether or not reasons have causal powers). Without this claim, the view could not be compatible with determinism. Hence, I take it that the agent and his reasons-responsiveness are determined. On my account, however, Ernie is not determined to follow R resulting in the bringing-about of MEG; he could as well bring about MEB (content: rejecting to help his friend although he could do it without suffering loss) thereby ignoring R.

My definition of free will allows for moral accountability, whereas the reasons-responsive view does not (and therefore is not a good theory of free will). In order to see this, I need to work out the differences in more detail. The reasons-responsive view is to be understood in such a way that the occurrence of a reason necessitates a mental event:

RG ☐→ MEG
RB ☐→ MEB

A morally good reason RG (like the belief that it is good to help a friend) leads to a good thought (wanting to help him), while a morally bad reason RB (like believing it is better to always prefer one’s convenience) leads to a bad thought (rejecting to help him). Now this seems to completely undermine moral responsibility. How can I be blamed for a morally reprehensible thought (and resulting act) if a reason triggered by outward circumstances determined me to have that thought? And how can I be praised for a mental event that is just determined by the occurrence of a reason?
A compatibilist holding the reasons-responsive view might protest here that I am positing an unreasonably wide notion of reasons. He might contend that RB is not a reason at all, but rather a non-cognitive desire. But this leads to the implausible view that everyone who does not respond to a morally good reason does not act of his free will. More still, with this view we end up with acquitting all people who do (and hence think) morally reprehensible things, because we are committed to believing they did not act out of free will. Of course, a compatibilist unwilling to bite this bullet might adduce the amendment that the lack of responsiveness to reasons is due to former bad decisions. But this only shifts the problem to a prior decision: there, too, we must ask how it can be possible that the person chose something contrary to reason. One answer is that prior bad decisions altered her in such a way that she became unresponsive to reasons, those decisions in turn being due to an unresponsiveness which can be traced back to bad decisions, and so on: this is an infinite regress. An alternative answer is that she chose to bring about MEB of free will, without being prompted by the desire represented by RB; but this is, of course, the sort of free will I have in mind, a free will incompatible with determinism, because it denies causal necessities.

In summary, it seems that the reasons-responsive view renders a view of free will that is incompatible with moral accountability, no matter how we turn it. If we take (good and bad reasons) to determine our mental events, moral accountability is out; if we take only good reasons to determine good mental events (and take that to be free will), we would have to say that people adopting morally bad mental events are not free, thereby again making moral accountability impossible. As none of above objections succeed against either premise 1 or premise 2, I take it to have shown that my argument is sound and that therefore, free will is incompatible with determinism.

References

Goldstein, Sheldon. 2017. “Bohmian Mechanics.” In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, Summer 2017. Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/sum2017/entries/qm-bohm/.

Hobbes, Thomas. 1655. Elementa Philosophica: De Corpore.

Honderich, Ted. n.d. “DETERMINISM AND FREEDOM PHILOSOPHY — ITS TERMINOLOGY.” The Determinism and Freedom Philosophy Website (blog). Accessed January 29, 2018. http://www.ucl.ac.uk/~uctytho/dfwTerminology.html.

Hume, David. 1975. “An Enquiry Concerning Human Understanding.” In Enquiries Concern- Ing Human Understanding and Concerning the Principles o FMorals, edited by L.A. Selby-Bigge. Oxford: Clarendon.

Plantinga, Alvin. 2011. Where the Conflict Really Lies: Science, Religion, and Naturalism. 1st ed. Oxford University Press.

Swinburne, Richard. 2013. Mind, Brain, and Free Will. Oxford: Oxford University Press, Usa.

Vihvelin, Kadri. 2017. “Arguments for Incompatibilism.” Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/archives/fall2017/entries/incompatibilism-arguments/.

Von Wachter, Daniel. 2009. Die Kausale Struktur Der Welt. Alber.


No Comments

Post A Comment

*