Causation and Free Will by T L Hurst - Revised 17th June 2022


Abstract

This paper looks at the main philosophic positions on free will. It concludes that neither compatibilism nor soft determinism are tenable. This is because both require that events are determined by prior events in the distant past, together with the laws of nature. I.e. By factors that are wholly outside our control. Whilst free will requires that we have input to the choices that are made, and hence to the events that transpire.

Similarly libertarianism is discarded as it fails to explain how it is possible for you to choose other than the option that you do choose.

An alternative position, called "freewillism", is introduced. This allows that a choice may be determinate at the time that it is made (i.e.that we can only choose the option that we do choose), but may be considered to be freely willed because, as sentient beings, we have input to the factors that determine the choice that is made.

//A mapping of the philosophic positions on free will to the types of causation is derived, and the evidence supporting the types of causation considered.

Introduction

This paper deals with the relationship between determinism and causal free will. Compatibilism is treated as being allied to soft determinism. I.e. Soft determinism says that both causal determinism and free will are true, and are compatible. Whereas compatibilism merely says that causal determinism and free will are compatible, without making any assertion as to whether either are true.

It has been suggested[1] that, put simplistically, the basic philosophical positions on the problem of causal free will can be distinguished by the answers to two questions:

  1. Is determinism true?
  2. Does free will exist?

The positions are:

That categorization will be used as a starting point for this discussion.

Terminology

Free Will - The ability of sentient beings to make choices of their own volition.ss
Causation - The view that events happen because of specific factors, including random chance.
Determined, determinate or deterministic - A choice where there is only one genuinely realizable option, or an event that will happen or has happened.
Causal (or hard) determinism - The view that the state of the universe is wholly determined by past states together with the laws of nature.
Soft determinism - The view that deternminism is compatible with free will, and that both are true.
Compatibilism - The view that free will is compatible with determinism.
Determinism - When used alone, this term normally refers to hard determinism.
Indetermined, indeterminate or indeterministic - A choice where there is more than one genuinely realizable option, or a random event or one that may or may not happen.
Indeterminism - The view that events are uncaused, and happen by random chance.
Freewillism - The view that events can be caused by determinate freely willed choices.
Libertarianism - The view that events can be caused by indeterminate freely willed choices.
Incompatibilism - The view that free will is not compatible with determinism.

Premises

  1. There are three types of causation: Note: The types of causation are not necessarily exclusive.
  2. Event-event and random causation are not willed.

Note: It follows from this that, to be compatible with free will as a causal factor a philosophical view must allow that agent causation is real.

The Basic Positions

Causal determinism is a global view that the state of the universe at any given time is wholly determined by past states together with the laws of nature. This means that at any given time there is only one future that is possible. Individual events, including your choices and actions, are entailed by the previous events to form a causal chain. Each event in the chain is causally effective in determining subsequent events, therefore your will is causally effective. I.e. Barring other constraints, you wilfully choose your actions.

Incompatibilism

Incompatibilism is the view that causal determinism and free will are not compatible, Incompatibilists comprise:

Libertarians define free will as "the ability to do otherwise", or, more formally, the ability to choose between multiple genuinely realizable alternatives.

Soft Determinism

Soft determinists use a variety of arguments to support the claim that causal determinism is compatible with free will:

Free Will as a Moral Issue

The claim that causal determinism is irrelevant is based on the view that free will relates to moral responsibility, not causal factors. However, although being deemed to have acted freely is crucial for moral responsibility, it does not follow that you did in fact act freely. Indeed that is a logical fallacy:

Although the first two statements are true, this is known as "affirming the consequent", because the conclusion does not necessarily follow from them.

Freedom from Coercion and Other Constraints

Freedom from coercion and physical, psychological and social constraints is necessary for free will, but is not sufficient,. This is because, although your will is causally effective, causal determinism requires that your will is wholly entailed by past events that are outside your control. So causally effective will is not synonymous with free will.

The Liberty of Spontaneity

Compatibilists focus on the events immediately prior to an action, and claim that because your will is causally effective, you have the "liberty of spontaneity" and act freely. However, although the liberty of spontaneity is sufficient for free will, it is not compatible with causal determinism. The reason is that, as we have already observed, causal determinism requires that your will is wholly entailed by prior events outside your control. Therefore causal determinism requires that your will, although causally effective, is not free. Hence causal determinism and the liberty of spontaneity are not compatible.

Libertarianism

The above considerations have established that in order to be termed "free" your will must not be wholly entailed by factors outside your control. Libertarianism is a scenario that meets this criterion, however, it has its own issues. Libertarian free will is based on the idea of "could do otherwise". This can been expressed more formally as a choice between multiple genuinely realizable alternatives. However, it remains to be explained how you can willfully choose an option other than the one that accords with your nature. Such a choice would, arguably, be random, and, if so, would not support free will.

Freewillism

To avoid the flaws in both soft determinism and libertarianism a scenario must allow that:

The term "freewillism" is used for such a scenario. Freewillism is the view that events can be caused by determinate willed choices, that are, at least in part, entailed by our natures rather than external factors. The philosophical position that free will is true neither requires nor denies that event-event and random causation are true.

Under freewillism, choices are considered to be free because, as sentient beings, we can reflect on past choices, actions and events, and resolve to do otherwise in future. Thus, over a period of time, we have input to our natures, and hence to the choices that we make.

The basic questions relevant to freewillism are:

  1. Does free will exist?
  2. Is free will restricted to one realizable alternative (determinate)?

The positions then are:

Furthermore, we can include indeterminism (and some other options) by expanding the questions to:

  1. Is determinism true?
  2. Is indeterminism true?
  3. Does free will exist?
  4. Is free will restricted to one realizable alternative (determinate)?

Note: The term "indeterminism" specifically relates to uncaused random events, to distinguish it from libertarianism.

Expanding the questions allows correlation to the causal questions below...

Analysis

The questions that classify the types of causation are:

  1. Is event-event causation real?
  2. Is random causation real?
  3. Is agent causation real?
  4. Is agent causation restricted to one realizable alternative?

Stage 1

There being four yes/no questions, the total number of combinations is 24 i.e. sixteen, which are shown in stage 1 of the table. However, we can simplify this table...

Stage 2

Stage 2 now shows that:

Stage 3

Stage 3 shows:

  1. The answers to q1 to q3 "NNN" support no causal relationships, which equates to Eternalism.
  2. The answers to q1 to q3 "YNN" support only event-event causation, which equates to Hard Determinism.
  3. The answers to q1 to q3 "NYN" support only random causation, which equates to Indeterminism.
  4. The answers to q1 to q3 "YYN" support event-event and random causation but not agent causation. I call this "Non-Agentism".
  5. The answers to q3 and q4 "YN" support agent causation without the outcome being restricted to one realisable outcome. This equates to Libertarianism.
  6. The answers to q3 and q4 "YY" support agent causation that is restricted to one realisable outcome. I call this "Freewillism".

Types of Causation

We have taken as a premise that there are three types of causation, and have linked them to the philosophic positions on free will, but is there evidence that they actually relate to reality?

Event-Event Causation

There is plenty of evidence of event-event causation. Indeed the "laws of physics" would be incoherent if the universe did not behave, on a macro scale, in a determinate manner. So although this does not prove that event-event causation is ontologically real, it is a strong indication. It underlies scientific realism, which is the view that the universe exists outside our perception of it, and that the "laws of physics" describe that reality.

Random Chance

There are examples, like the roll of dice, which are thought to be random, but it is questionable whether they are truly so. The reason is that determinate systems can give unpredictable results if all the variables are not known. Hence if all the variables were known for a particular roll of dice, the outcome could, in theory, be predicted.

Also, it is possible to define statistical rules as to how truly random systems behave, but it is known that pseudo-random (determinate) systems can emulate such behaviour. So the evidence that a system may behave statistically in a random manner is not conclusive.

Perhaps the best example of random chance is the way that the universe behaves at the atomic scale. Quantum Mechanics suggests that if a single unstable atom is placed in a controlled environment, it cannot be predicted how long it will take for the atom to decay. Only the probability of decay within a given time can be calculated. However, although there is agreement that QM is probabilistic rather than deterministic, there is no general consensus as to how this should be interpreted philosophically.

Agent Causation

In general, we intuitively believe that we have free will and act as causal agents, but is that true? After all, the universe around us behaves in a determinate way, and we, including our brains, are physically part of that universe. So how can we be causal agents if our brains are determinate systems? Furthermore, even allowing that events at the atomic scale are essentially probabilistic rather than determinate does not offer any support for free will, as neither determinism nor indeterminism allow free will. Because of this, many philosophers think that free will, as a causal agent, is illusory. So what evidence is there?

The strongest evidence of free will as a causal agent may be technology. This is because the creation of technological artifacts usually involves intent. So the artifacts are intentionally designed in a sense that natural objects are not. Plants are not intentionally designed to metabolize carbon dioxide into oxygen to produce an atmosphere that is breathable by us. Evolution is not intentionally designed to produce human beings.

It could be argued that the real reason why it is almost certain that many man-made artifacts could not arise in absence of an intelligent designer is simply because there is no easy evolutionary path whereby such artifacts could arise by chance. It has nothing necessarily to do with the intended purpose. But whether this is true or not does not change the fact that many artifacts are intentionally designed, and have an intended purpose, in a way that no natural objects do.

For example, if the first person to land on a newly discovered planet found dense jungle, empty beaches, and untilled soil, would she conclude that was natural? Yes. But if she found a nuclear power station, cars and a metropolis, would she conclude that was natural? No. She would conclude that she was in the presence of alien technology. The products of sentient beings. Why? Because the artifacts are intentionally designed.

Furthermore it is clear that such design is not possible under determinism, or indeterminism, because free will is not possible.in either of those scenarios.

Freewillism vs Libertarianism

There remains the question of freewillism and libertarianism. In both cases intentional design is theoretically possible, as the cause is our free will. Also, in both cases there is only one outcome. I.e. All choices are determinate once they are made. The difference is that under freewillism the choice is determinate, whilst, under libertarianism, it is indeterminate. This raises some issues:

So, although we believe intuitively that we have the capacity to dynamically choose between multiple genuinely realizable alternatives, it is conceivable that, if we do have free will, it may operate in an entirely different manner.

Example - Choosing Breakfast

It may seem paradoxical that choices under freewillism are causally determinate yet freely willed. So an example may help to explain it. Let's say I have a computerized machine for making my breakfast. It is in my nature to always have egg on toast or a bacon sandwich. So I write a little program for the machine:

Function Breakfast (egg_days)
!---------
If egg_days < 1
  make egg_on_toast
  increment egg_days
Else
  make bacon_sandwich
  egg_days = 0
End If
Return egg_days

I should explain that "egg_days" is a variable that is passed to the procedure each time it is run. It reflects the number of times I have had egg on toast on sequential days. So:

I think you can agree that this code is determinate, but it accords with my will. Then, on reflection, I decide that I want to have egg on toast two days in a row. So I change one number in the code (from 1 to 2):

If egg_days < 2
etc...

The code is still determinate and so is the choice, but because, on reflection, I decided to change the code, the behaviour is different from what it was before. I now have egg two days in a row, then bacon once.

Hence a choice can be determinate when made, as it is determined by my nature. However it reflects my free will as my nature can change over time, and I have input to that process.

References

[1] Wikipedia - Free Will: here (since amended).
Graphics based on a design courtesy of Wikimedia.