Everybody seems to have their own meaning of Free Will. There are two extreme senses about which I’m guessing there shouldn’t be any argument. At one extreme there is the naively-optimistic interpretation that Free Will means you can do anything you want....
I’m not considering that kind of Free Will, I’m just talking about the ability to choose between options that are consistent with reality.
At the other extreme we have “naive fatalism”, the view that there are no options because everything has been predetermined from the beginning of time and that this means there’s no point in making a choice....
Naive Fatalism about Free Will
I hope we’d all agree that that is a fallacy as well. Even if the world looked at from outside was fully predetermined in this sense, that wouldn’t affect the internal logic of the situation that says that if you go out into the rain without an umbrella, you are likely to get wet. An ‘if’ statement like that is equally true whether it is actually raining or not. To put it formally, there are truths that are invariant with respect to the prior situation, and are therefore unaffected by the fact that the prior situation is in fact unique. These are the truths that we need to evaluate potential choices, and there is no conflict between them and determinismi.
Between the definition that says “believing in Free Will means believing that any choice is possible” and the definition that says that “denying Free Will means denying that we ever have any choice”, it’s clear that whether we “believe” in Free Will or not depends on where we draw the line between these extremesii.
It might be useful to note Albritton’s distinction between:
freedom of action,
efficacy of will,
strength of will and
freedom of will.
The last he defines as “the capacity to form intentions or to decide in favour of one course of action or another”. This seems to be the most useful thing to focus on, as without that capacity the other conditions would be of no value.
The old argument was that freedom of choice is impossible because everything is predetermined by the laws of Newtonian physics applied to every particle in the universe. But even in this case, we could still decide in favour of some course of action - if we wanted to. Whether we wanted to or not would, of course, be
Who we are ‘determines’ what choices we make, so ......
Shock! horror! We enslave ourselves with every choice!
predetermined! But it would still be our own choice: unless you are a Cartesian dualist, you can’t identify a separate will wishing to do otherwise, so it is hard to see why this should be regarded as a “restriction on freedom”, as alleged.
In any case, this argument no longer applies, because with the development of quantum uncertainty, Physics no longer believes in a single, predetermined sequence of events. So now philosophical critics of Free Will use a different argument. They say that even without total predetermination, we still can’t really make choices because everything that happens is really governed by random events at the sub-atomic level that are outside “our” control. But although this is a widely held view among philosophers I think this is completely missing the point, which I shall try to show by a simple example.
If a ball is thrown into a conical cup then its centre will finish up in the same place despite wide variations in the initial speed and angle of throw. The structure of the circumstances ensures that a particular result is liable to be arrived at whatever the details of the previous conditions. I think of this as “structural causation” as distinct from “initial condition causation”.
Quantum uncertainty ensures that there are a myriad of different possible sequences of events at the sub-atomic level, but the structure of circumstances often ensures that a great many of those different singular sequences converge to give the same effective outcome, so in fact in a quantum world structural causation is often the only kind of cause that can be identified. By contrast, in other circumstances the sequences diverge, and in this case the outcome really is random.
The nature of a particular person’s knowledge, experience, temperament, repertoire of metaphors and so on is one example of a structure that can impose convergence on otherwise randomly influenced sequences, so we see that structural causation makes the idea of outcomes being influenced by personal choices perfectly consistent with modern physics.
“Ah yes”, the critic of free will might say, “but even if we accept that choices influence outcomes, I still claim you are ignoring the fact that the choices we make are themselves the consequence of previous events”. But am I really ignoring that?
No-one believes they have knowledge through no cause
No-one believes they exist through no cause
Internalised effects are recognized as caused, but from the point of view of the self, such causes cease to count
When I make a choice, I know very well that I am only standing here as a consequence of previous events. I also know that all the information and skills that I use in making a choice are the result of previous events. The fact that I am me, with my particular DNA, my outlook and so on, are all the result of previous events. Nobody is unaware of causes like this – they just don’t think of them as relevant when they make a choice - which is absolutely right. When a computer program carries out a piece of logic, the circumstances that led to the program being coded in that particular way are irrelevant. To put it formally, once you have fully specified the nature of the causal agent making a decision, it is “double-accounting” to include any causes that have affected that nature.
Perhaps we could summarise my argument so far by saying that although there is no need to believe in a Will that is an “uncaused cause”, I can see no problem in seeing a Willed choice as a “causal process that is conscious of itself”.
At this point we have to stop and consider the sense of Free Will that distinguishes it from Coercion, because even somebody acting under coercion is strictly speaking “submitting to coercion of their own free will” in the sense I have been talking about.
Let us define internalised causes more carefully as causes whose effects have been incorporated in the internal picture of the Self whose Will we are talking about.
My decision to include Albritton’s definition of Free Will in this presentation is in part a consequence of my having recently read this definition and having formed a positive judgement of it. I have internalised these effects upon my knowledge and outlook, so that I know myself to be someone who knows this definition and finds it useful. In making the decision to quote it, I felt no kind of conflict with my sense of myself. Then we can define “Psychological Free Will” as freedom from the influence of non-internalised causes. If I thought Albritton’s definition was rubbish but I had been bribed to include it, I might perhaps decide to include it, but my explanation to myself of how I had reached that decision would start from the Will of the briber rather than from my own. Coercion on this account is coercive because it is understood, by the person on the receiving end, as being coercion.
There are two important things missing from this account so far. The first is that almost everything I have said so far could apply equally well to a thermostat. It is made to have two possible output states, and in that sense has the possibility of choice as Albritton requires. It internalises the room temperature, and then chooses an output according to its nature. Isn’t that what I’ve been saying? What is it about decision-making by a conscious being that is different from this, and so warrants it being given a special name?
Well, the first thing to ask is what we mean by a “conscious decision”, because it could mean two things with a very important difference. It could be a thought process in which every single step is conscious, or it could mean a thought process in which consciousness plays an essential role.
In this metaphor, the surface of the water represents consciousness, but the wave is a process involving the whole body of water, although is only possible because the water has a free surface. They way I see it, a conscious decision is like the surface of a wave, not like a glint on the surface. That is, what we call a conscious decision is a decision chosen in a way that consciousness makes possible, not one in which every step takes place in the light of conscious attention. To take an example, even to choose a word like “wave” using only conscious steps would be utterly impossible – I would have to hold the entire dictionary in my consciousness simultaneously. Whereas what I did was to hold the idea of a wave in the “spotlight” of my conscious attention for a moment (to change the metaphor) and the word “wave” appeared like a moth from the darkness, furnished by unconscious steps quite unknown to me. So if Free Will means the ability to decide using conscious steps only, then of course we should reject it, as we couldn’t have such an ability. But if it means a decision in which consciousness plays an essential role like the free surface in my metaphoriii, then I think that still allows it to be something special.
My argument for this starts from a common fault in many accounts of causality, the assumption that a single effect can be attributed to a single cause.
Now it is certainly true, for example, that a person from a particular socio-economic background is traditionally more likely to vote for a particular party, but all of us know very well that such connections are not absolute and inevitable, because that is not how the mind works. We never know when a thought might not strike us that leads us to act in a previously unexpected way, although we recognize the thought as our own.
Functionalism has the idea of a specified list of causes generating a mental state with specified effects, but even that is not enough.
To be absolutely certain what the outcome of a particular conscious thought process will be, we would need to know every genetic and environmental input to that thinker’s brain from the moment of conception onwards, because every input affects connection strengths in the enormous unique mesh of neural pathways in the brain that is potentially involved in our thoughts at any time.
Consciousness: The extraordinarily rapid and complete integration of an extraordinarily diversified range of processes. [Tononi]
So from this point of view, when we speak of Free Will, we mean a will that is free from any pre-ordainedrestriction on the list of previously internalised causes that may enter any particular decision. It is this open-endedness that distinguishes conscious decisionsiv. Physiologically, Tononi and othersv have suggested that consciousness is dependent on the extraordinarily rapid and complete integration of an extraordinarily diversified range of processes, and that ability to integrate fits in with this picture of conscious decision-making that I have been giving.
The final thing that needs saying about the exercise of Free Willvi, is that proving that it is logically possible is not the same as saying that it is easy.
Somebody – was it Spinoza? - came up with a very good metaphor for this. He said that exercising free will is like defending a narrow pinnacle that is beset with enemies on all sides. It is only possible with the utmost vigilance, courage and determination. But to do that we first have to forget those philosophers who claim that this pinnacle we are trying to defend doesn’t even exist!
Roger Haines, April 2015
i: This is a considerable simplification of the problems with this view. The root problem, I suspect, is that the person expressing the view in the “bubble” is also a part of the world (s)he is talking about, which leads to a vicious circle of self-reference reminiscent of the proverbial Cretan liar.
ii The reason for this ambiguity is that free will means the possibility of doing otherwise than as we actually do, and that begs the question as to how much is allowed to be changed when we say “otherwise possible”. If we mean “possible in a world with different physical laws” we are using the sense that is required by the naive optimist. If we mean “possible in exactly this world, with exactly the sequence of events that occurs in it” then we are using the sense that is required by the naive fatalist and strict incompatibilist. The usefulness of the debate comes in identifying the most relevant intermediate senses and their implications, not in dogmatically defending either of the extreme senses, which are equally sterile. The simplest intermediate sense takes only the circumstances facing a chooser as given, and insists that the ability of different people to make different choices in the same circumstances is the kind of Free Will that matters. I think this sets the bar too low because it ascribes Free Will to a thermostat, so I have tried to add more restrictive conditions, especially freedom from coercion by another’s Will and freedom to go beyond a predetermined list of relevant causes.
iii Perhaps the connection to the metaphor needs spelling out. The word “wave” didn’t just become conscious for no reason. It appeared because that was the idea I was conscious of wanting to express – and, of course, because at some previous point in time I learnt that word and that sense by conscious attention. What we call “attention”, the essential element of consciousness in this context, is what shapes and responds to the unconscious activity just as the surface of a wave reflects and constrains all the movement beneath. In fact it is misleading to describe either element as “causing” the behaviour of the other. The surface and the body process are integral aspects of the same phenomenon.
iv Only by this means can we explain creativity. Sometimes somebody has a thought that nobody has ever had before – how could that be explained by a mind modelled as a functional list of inputs and outputs? But an open-ended mind can construct new analogies from previous thoughts.
v E.g: Edelman & Tononi. How Matter Makes Imagination. Although such a process may well be what makes consciousness possible, that doesn’t mean, of course, that everything in such a process is conscious – in fact, the bit we consider ourselves to be conscious of must just be the summarised outcome of that process that is held for a moment in some kind of “working memory” – it becomes the new focus of attention and triggers & shapes the next stage of unconscious processing. This also ties up with Daniel Dennett’s “multiple drafts” theory of consciousness, in which many different versions of the same thought compete for priority, and only the winner is recognized as conscious – what Dennett calls competition is what Tononi calls integration.
vi I have left out the moral implications in this presentation due to limits of time. There are perhaps two key points I could have made:-
A moral decision requires a degree of competence, maturity and knowledge, all of which are the outcome of previous causes. On the other hand, certain causes, such as those resulting in some kinds of insanity, certainly have to be precluded. This refutes any notion that moral decisions require freedom from all effects of previous causes – only some causes have to be ruled out, while others are necessarily active.
Most accounts of morality – the Golden Rule, Kant’s Categorical Imperatives, most formulations of Utilitarianism – explicitly require that we take ourselves out of the equation, to be replaced by a typical or universal ‘other’, exactly as considered in the intermediate sense of Free Will defined in end-note (ii). Individual differences may explain why one person may find it harder than another to identify or act upon a particular decision, but the “right action” is what a non-specific person should do in the circumstances concerned. Thus the possibility of moral behaviour requires only that the right action is possible in those circumstances.