In what sense or senses may I be said to act freely? Does determinism rule out freedom of choice?

Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 2.5 License

Most people, if asked, would consider themselves to be self directing, responsible agents. Society as a whole takes this for granted, people are held responsible for their actions. If we do something it is assumed we have chosen to do it, even if our efforts are thwarted by circumstances we consider the intention the important thing (attempted murder is considered just as serious as actual murder). The laws of cause and effect, if anything, are even more widely accepted, every event is considered to have a cause and causes are necessarily followed by their effects. Determinists claim that the laws of cause and effect fix the future state of the universe, things can never be other than the way they are. Free will claims there is a possibility of things being other than the way they are. It has been shown that there is a contradiction in these two positions (see Peter van Inwagen), if decisions and actions are events then they have causes, therefore they are necessarily what they are. Either we have to surrender free will or claim the universe is not deterministic, or we have to modify the definitions of one or both.

The notion of free will has much in common with the idea of God, the best arguments for it's existence come not from logic but from personal experience. In contrast with the arguments for and against the existence of God, we all have had the experience of making a choice. Determinists must argue that this experience is delusory, that free will is not necessary for a complete explanation of the universe and so, according to the principal of Occam's razor, shouldn't be incorporated. On the other hand it has been argued, notably by Hume, that there is no logical necessity involved in the law of cause and effect. Just because a particular effect has followed a certain cause in the past there is no reason to expect it to do so in the future.

Determinism has it's roots in religious doctrine. Since God is both omniscient and omnipotent he must know our future actions, the future is thus predetermined. The problem for theologians was how can we be held responsible for our sins since God has predetermined our actions? This was dealt with in a number of ways which are not really applicable to current debates. With the rise of science determinism has a new champion with the position that, given the laws of physics and exact knowledge of the state of the universe at some moment in time, the state of the universe at every other moment in time can be determined. The course of the universe is therefore fixed, and all we need is the right set of laws and sufficiently accurate measuring instruments. Scientific progress has been governed by the search for laws which provide the best predictions of the future behaviour of the universe. In the twentieth century these ideas have come under serious attack.

Quantum theory has turned many of our preconceptions on their heads, among them is the idea that we can measure the universe to an arbitrary degree of accuracy. Heisenberg's Uncertainty Principle has destroyed this particular idea. The principle states that there are several pairs of physical values, the best known example is velocity and position but others include energy and time and magnetic and electric field strength, that are linked in a special way. As the accuracy of our knowledge of one of these quantities becomes greater, the accuracy of our knowledge of the other necessarily decreases. The effect is negligible at the level of people, cars, planets and so on, but at the level of atoms it is insurmountable. If you want to know, to an arbitrary level of accuracy, the position of an atom then you cannot also know it's velocity.

Developments in physics have had their parallels in other fields, notably logic and computing. Godel's theorem states that, within any sufficiently expressive logical system, there are true statements of that system that cannot be proven within that system. Thus any useful logical system, and it would not be stretching a point too far to include the laws of physics in this category, cannot be both consistent and complete. The notion of NP-completeness, really a parallel result to Godel's, is especially interesting. It rests on the notion of algorithmic solutions to problems. If we want to sort things in order of size we can write an algorithm which will perform that task, we can expect the result of the algorithm to be identical for identical sets of input data. It is possible, by analysis of the algorithm, to determine in advance how long the algorithm will take to perform the task and express this time as a polynomial function involving the number of input data as the variable. The performance of the algorithm is then expressed as being 'of the order of' the highest power in this polynomial. I chose this example on purpose as all sorting algorithms have polynomial running time. It turns out that not all problems are so easy, some deceptively simple problems are not solvable by an algorithm which will run in polynomial time (they are Not Polynomial complete). Examples are finding the best packing of differently sized objects into a finite space (the knapsack problem) and finding the shortest route between a number of cities, visiting each city once (the travelling salesman problem). The only way to know for how long a particular algorithm will run is to wait and see. If we consider the laws of physics as constituting an algorithm for determining the future state of the universe from the present state then, if this is an NP-complete problem, there could be no way of predicting the future state of the universe accurately - other than waiting to see what happened. While it is beyond my abilities to demonstrate that the ongoing progress of the universe is equivalent to the performing of an algorithm for an NP- complete problem I don't think the idea is far fetched. Between them Godel's theorem and NP-completeness challenge the ideas that we can have an exhaustive set of physical laws and the notion that we could use those laws to predict the future of the universe in a detailed way. The traditional view, that we would one day be able to predict the future with arbitrary accuracy, is incorrect.

Quantum theory has offered another option: parallel universes. This idea was first proposed by the physicist Hugh Everett and initially shunned by established science. It is now gaining ground as the best explanation of certain quantum mechanical effects. The general idea is familiar from science fiction, there are other universes where certain events happened differently to the way they did in this one. The details are a bit more unsettling, there are an infinite number of other universes, including an infinite number almost identical to our own - differing perhaps in the position of one electron. The determinist accepts that things could be other than what they are, because they are other than what they are, in countless other universes. But this universe cannot be anything other than what it is, a different choice means a different universe. Every possibility of existence is played out, as an infinite number of me choose to do one thing, and infinite number will also do something else. Kierkegaard's proposition, that we make ourselves through our choices, is turned on it's head. We exist in an infinite number of copies and doppelgangers, differentiated by each choice we could have made.

We can accept that the deterministic view of the universe will never be practically tested. This is a problem for determinists, since they are generally of scientific persuasion and would like a falsifiable scientific theory rather than an ideology, but it does not undermine the central proposition that, in principal, the future states of the universe are uniquely determined by the present state. For a refutation of this, the problem of finding an original cause within the human mind still remains. Indeed as determinism has lost ground in the hard sciences it has gained an even stronger position in many eyes with developments in genetics, psychology and sociology. Scientific explanations of behaviour are now given more credence than at any time in history. We consider ourselves to be a complex collection of drives and inhibitions, determined by our genetic makeup and social experiences. These define our goals and, through the application of knowledge and reason, can discern actions which will bring us closer to those goals. The question is, given our goals and our knowledge, could we have chosen to act other than the way we did? Subjectively we feel we are making a choice, but an outside observer can often make accurate predictions of our decisions. The problem is summed up by Sidgwick:

Further, we always explain the voluntary action of all men except ourselves on the principal of causation by character and circumstances. Indeed otherwise social life would be impossible: for the life of man in society involves daily a mass of minute forecasts of the actions of other men ... who are thus necessarily regarded as things having determinate properties, causes whose effects are calculable ... and if our forecast turns out in any case to be erroneous, we do not attribute the discrepancy to the disturbing influence of Free Will, but to our incomplete acquaintance with their character and motives.

The problem with denying that our behaviour is governed, in some way, is that it is hard to see any replacement other than random choice, which is not really an improvement. An alternative perspective is offered by Existentialism. Robert Olson discusses Sartre's position:

Sartre ... denies that either objective situations or subjective motives ever move us to act. The objective situation moves us to act only in so far as we apprehend it, and our apprehension of an objective situation is itself determined by a free choice of goals ... Deliberation, says Sartre, is merely 'an evaluation of means in relation to already existing ends.'

The problem of freedom is at the level of our choice of ourselves, we choose what we are and this defines our goals and desires. This view is echoed by Harry Frankfurt who argues that it is second order desires that are important:

Someone has a desire of the second order either when he wants simply to have a certain desire or when he wants a certain desire to be his will.

He later claims:

It is only because a person has volitions of the second order that he is capable of both enjoying and of lacking freedom of the will.

It is the ability to turn second order desires and volitions into actions that demonstrate free will. In my opinion, this doesn't really solve the problem, just moves it somewhere else. How do we choose our second order desires and volitions? Sartre talks about a free, prereflective choice of oneself, but I find this somewhat vague. The notion that my free choice of myself is somehow outside my deliberative process seems to put the free will where it is of no use to me, but if we put free choice in the realm of reason then, by the very nature of it, there are reasons for our actions. If we can offer an explanation of a person's goal selection in terms of, for instance, external influences in childhood, then we have a causal relationship, and we lose the free will. This perhaps isn't important, Daniel Dennett said in a recent interview:

Let's suppose we are compelled by our beliefs and desires. Isn't that the way we want it to be? Suppose we are very good at getting to the truth about things as a result of these beliefs and desires. It would certainly help if the combination of our beliefs and desires governed our actions. You can't imagine anything better to govern our actions. You'd hope that our beliefs and desires would compel our actions, otherwise we'd be like a loose cannon on the deck.

We want free will to be a choice outside of rules, to be unconstrained by rules of any kind. But if we give up the rules we have nothing to replace them with but random action for no reason, which is hardly more satisfactory.

The precepts of determinism are hard to deny, but also hard to prove. Even with all the evidence which points against a practical demonstration it cannot be disproved that the universe is deterministic. However things are, they are. There is no way of knowing if they could have been different. The problem with free will is that, although we treat each other as if we had a choice, there seems to be no real alternative to the rule based behaviour implied by psychology and genetics. This seems to leave us back where we started which, in such a complex subject, is hardly surprising. On this note, I would like to close with a quote from Dostoevsky's 'The Idiot':

Don't let us forget that the causes of human actions are usually immeasurably more complex and varied than our subsequent explanations of them.


Daniel Dennett, interviewed by Guy Douglas in 'The Philosophers' Magazine' Issue Six, Spring 1999

David Deutsch 'The Fabric of Reality' Penguin 1998

Harry G Frankfurt 'The Freedom of the Will and the Concept of a Person' The Journal of Philosophy, Volume LXVIII, No 1, January, 1971.

Douglas R Hofstadter 'Godel, Escher, Bach: An Eternal Golden Braid' Penguin 1980

Peter van Inwagen 'The Incompatibility of Free Will and Determinism' Philosophical Studies, 27 : 185-99.

Bart Kosko 'Fuzzy Thinking' Flamingo 1994

Robert G Olson 'An Introduction to Existentialism' Dover Publishing Inc 1962

Sidgwick 'The Methods of Ethics' Macmillan, London, 1907.

© Robert Crowther