For all the debate surrounding the freedom of will, surprisingly little attention has been given to analysis of the concept of "free will", independent of its truth or falsity. It is a vital issue, for without such a definition all arguments which use the phrase run the risk of incorporating a meaningless term, and thus lacking sense. Further, by simple analysis of the phrase "free will" we come across a contradiction which is potentially fatal to the libertarian position, and acts as a road sign for the kind of compatibilist theories which should be pursued.
Traditionally, "free will" has been used by libertarians in the following fashion: to will freely is to will without constraint from forces or causes outside one's control. This usage needs some small further clarification. Firstly we are discussing mental will, not action. No serious philosopher proposes that humans are omnipotent. Secondly, "will" is taken to be something like the process of making a decision or choice. Thus in the simple case of making a choice between an apple and an orange, a libertarian will say that for the choice to be free (and they claim it is), it must be made without being influenced to the point of constraint. This is to say that they may be influenced, but only in a trivial way: if the influence is so strong that the decision can be said to follow causally from it, then it is a constraint, and the choice is not free.
Determinists appear quite happy to take this definition and run with it. It is easy to see why: the position practically walks itself to the gallows. In the case of the apple and the orange, it is simple enough to see that the decision will be based on taste (excluding scenarios where the individual has reason to pick against his or her taste). If I like oranges better than apples, I will pick the orange. Yet I never decided to like oranges better than apples independently of how I found their taste. I just prefer them. It is a fact about me that is outside of my control.
This counter-position often held by determinists is entirely unnecessary. Convincing though it is, it is not logically impossible that a counter argument may be constructed. A stronger argument against the libertarian position can be achieved, one which not so much argues against libertarian free will, but under libertarian free will. By going straight to the foundations of the concept of free will we can display the concept to not just be false, but a logical contradiction in terms.
At the risk of redundancy, the phrase "free will" contains two parts: "free" and "will". "Free will" clearly refers to a certain kind of willing: willing that is free. Let us first then investigate the broader class: what it is to will, and from there move on to the subset: what kind of willing "free will" would be.
Pinning down such a hazy notion as "will" is a tricky business. It is related to desire, but not identical. One main difference between the two is that a desire may be subconscious, but to will something has an element of deliberation to it. I cannot will something without my own knowledge as I can desire something without knowing it. Another difference is that willing something is an event; a desire can last an indeterminate length of time and thus is more akin to a state.
Will also possesses an aspect of intention to act. While I may possess a desire to write this essay for some months without ever writing it, I could not will myself to write it and then not do so. If I will myself to write then I am enacting something - or at least attempting to. It is of course the case that any number of physical limitations may prevent me from making my will actual. For example, someone may physically restrain my hand, yet in this situation I am still willing that I write the essay, for I am trying to do so - I am making a serious attempt to write, unlike when I held only an unfulfilled desire. From this we must take care to stress that, though willing has an element of deciding to enact something, the actual successful enactment is not necessary (in fact it is irrelevant). Will is a mental phenomena, not physical.
With these considerations in mind we may formulate a definition of will. To will X is to consciously and deliberately attempt to enact X. At this point one key observation must be made. Will, as outlined here, is conscious and deliberate. If this is the case, then it is part of a "train of thought" of which we can tell a kind of causal story or narrative. I chose the orange because I liked it better. Or, I chose the apple because I wanted to show that I can make choices contrary to taste. One corollary of this is that it excludes instinctual action from being counted as will. If I jerk my hand away from a hot pan I did not will that I do so. Rather it is a reflex action. Biology confirms this fact: actions such as withdrawing a hand from something hot is coordinated by the spinal column, not the brain.
Armed with this conception of will, we can move forwards and look at what kind of willing libertarians (and consequently determinists) speak of when they discuss freedom of will. Recall the libertarian usage of free will mentioned above. Under this lose definition, a freely willed choice was one which was not constrained by anything which was itself not a free choice made by that individual. This is a way of saying that a freely willed action is one that is uncaused, except by the individual's will. If my decision to pick the orange is caused by my taste then it is no longer my free choice: it is, as it were, my taste's choice. Thus, for the choice to be mine, there cannot be any other cause to "steal" it from me. Freedom comes to mean a lack of causality. Yet what is something uncaused? Something without a causal chain behind it is something random, and not even the randomness of a hypothetical fair die, which is governed by probability. The random nature of a free choice would be more akin to the randomness of a spontaneously existing universe. To use the religious terminology, free choice is choice ex nihilo.
It is here that the contradiction emerges. Will is the end of a chain of thought: a causal process. "Free will" would have to be will without cause - will ex nihilo. Such a thing is a logical contradiction, for will is by nature caused and non-random. A free choice would have to be a choice both with and without a cause, which is clearly contradictory. It is clear, then, from this analysis that "free will" means nothing: it is nonsense.
It is important here to close off one line of escape. It could be claimed that if will could be the cause of itself then the causality of will and the unconstrained nature of freedom could both be satisfied. However, such a proposal would require an infinite regress of willing, which I take most will be unwilling to accept - especially given that an individual's life does not continue back for eternity.
This contradiction draws startling light upon the debates surrounding “free will”. Firstly, it is clear that the libertarian idea of free will is untenable and thus no credible theory may be built using it. To base an argument on a contradiction is to have an argument of nonsense (or to be a Hegelian!), for you are asserting the existence of something which makes no sense. However, it is also a consequence that the determinist arguments, designed to refute libertarianism and therefore utilising that conception of free will, suffer the same fate. Determinism attempts to deny the existence of free will. However, this makes as much sense as one attempting to deny the existence of “gaga”. It is not a thing which we can discuss as existing or not, for it is not a thing at all. The sentence “gaga does not exist” means nothing, just as “gaga exists” means nothing, because “gaga” denotes nothing.
Readers familiar with Russellmay feel the need here to object, claiming that it is not the case that these phrases mean nothing, but rather that the affirmation of the existence of “gaga” is simply false (1905). However, this is not a “present King of France” situation, for it is not that “gaga” or “free will” are definite descriptions which sensibly describe something which is simply non-existent. Rather, they are nonsense themselves, thus rendering the sentence referring to them as nonsense. We understand what "the present King of France is bald" is trying to say - it makes sense - it is merely the case that there is no present King of France. "Free will", meanwhile, is more like the phrase "white blackness". We understand the words but cannot conceive of what they mean when conjoined.
It could perhaps be proposed that the nonsense of “free will” and the nonsense of “gaga” are not the same kind of nonsense. One might be tempted to bring in some kind of concept of “illuminating nonsense” (such as in Hacker, 2001). However, it seems clear to me that “illuminating nonsense” is as much of a contradiction as “free will” is. Nonsense, by definition, means nothing, and if it means nothing then there is no way it can communicate anything meaningful. We can only mistakenly believe that it communicates something: it is this mistaken belief (probably brought about by the fact that the individual elements of “free will” do make sense, before they are brought together) which makes us believe that “free will” and “gaga” could be different kinds of nonsense.
The only discussions on free will which still make sense, therefore, are those which do not utilise the idea of a completely unconstrained freedom of will. The only theories which survive this culling are those of compatibilism. There exists here something of a linguistic misfortune. Compatibilism is named such as it is supposed to be a half way point, combining determinism and libertarianism. It is important to note, however, that despite this naming convention, the theories of compatibilism do not revolve around the affirmation or denial of libertarian “free will” and concordantly do not fall into the trap of discussing a contradiction.
A new debate on the nature of will is needed. This debate would overlap with much of what compatibilism is today, but a certain measure of distancing is needed. The compatibilist theories, while they attempt to construct new definitions of “free will”, are misguided in their motivation. We should not be trying to “rescue” free will. The concept should be left to die, and the debates around will must move on. Using the phrase “free will” to denote something that is quite different to libertarian free will betrays the extent to which compatibilists are still trying to salvage a defunct concept. Instead, a study into the nature of constrained willing must take place, drawing from compatibilist theories, but with a completely different aim: not to maintain or deny freedom of will, but to understand how it is we will and act as part of a larger causal narrative.
