Hector C. Parr
Philosophers have long been puzzled by the mystery of free will, whereby we humans seem able to make autonomous decisions for which we alone are responsible, unlike almost everything else in the universe, which must obey the inflexible rules of cause and effect. This paper argues that adopting some modern ideas concerning the nature of time throws new light on the problem.
Throughout our waking lives we are faced with decisions, which we must resolve by choosing one course of action when several are open to us. Such decisions may be of considerable importance, as when we decide what career to pursue or whether to emigrate to a foreign land, or they may be trivial, as when we decide whether to ask for a cup of tea or of coffee, or which of several possible roads to follow to reach our destination. Because we make these choices so frequently one would expect the decision-making process to be fully understood, and yet thinking persons have disagreed over its nature for thousands of years, and today seem no nearer to agreement than they ever were. In recent years it has become clear that the process is associated with activity in the brain rather than the heart or the body as a whole, as was believed by some in earlier times, but there is still disagreement over whether the decision-making process is a purely mechanistic function, with the brain acting as a highly complex computer, or whether its action breaks free from the laws of physics, unlike computers and the whole world of inanimate objects.
The discussion is often portrayed as a conflict between free will and determinism. The physical world outside ourselves is assumed to be deterministic, with every event caused, or uniquely determined, by earlier events, while in contrast we seem to have control of our own lives in a way that transcends the shackles of cause and effect. When we try to discover the nature of our own decision-making process, and assess all the factors on which our choices depend, there is no doubt that chief among them are the influences and previous experiences to which we have been subject in the past, both immediate and remote. But we seem to have an ability to override all these influences, particularly when matters of conscience or principle are involved. We feel that we ourselves are the causes of our actions, and expect to be held responsible for them in a way that no machine or computer can ever be held responsible for its behaviour. We rejoice in the good decisions we have made, and we regret those which turned out less favourably for ourselves or others, and wish we had chosen differently. We treat others also as though they are responsible for their own conduct; we praise them for their noble or generous actions, believing that they were freely chosen, and we blame them for the mistakes they make, believing that they could have performed better. We hold them morally responsible for their behaviour in a manner we cannot possibly apply to mechanical devices, for we know such devices act simply in accordance with the laws of nature, and are incapable of behaving differently from the way they actually do behave. They have not got free will.
BRAINS AND COMPUTERS
Nevertheless there is a close parallel between the way a person reacts with the outside world and the way a computer operates. Every job performed by a computer can be divided into three stages, input, processing and output. Firstly some information must be fed into the computer's memory through one of its input devices, such as the keyboard, the mouse or a scanner. The computer then processes this information in some way, such as by re-calculating a spread sheet, selecting items from a database or changing the characteristics of a photograph. The results are then displayed on the screen, on a printer, or via the loudspeakers or some other output device. In a very similar manner all the knowledge a person has of the world round about must have been input through the five senses of sight, sound, touch, taste or smell. The brain can retain this information much as a computer memory or hard disc stores data, and when we need to "think things out", can process it as can the central processor of a computer. Finally the brain sends output signals to our muscles, and we react with the world by moving, speaking, or performing some other action.
No-one believes that a computer possesses free will, for its behaviour is strictly controlled by the designers who built it and the programmers who supplied its software. An expert who understands its operation in detail could, in theory, follow the whole process through from input to output, and would know at each stage what was going to happen next. The human brain is vastly more complex than any of today's computers, and has on its "hard disc" a far richer store of knowledge, motives, preferences, habits and instincts than has a computer, but is it not possible that in principle it operates in the same way?
Can science help to resolve the question? Neurologists and brain-scientists are constantly learning more about the nervous system, and can trace the transmission of signals from the five senses further and further into the brain. They are also beginning to locate the regions of the brain responsible for sending signals back to our muscles. Both types of signal are found to obey well-understood laws of physics and chemistry. But there remains a gap in our knowledge somewhere between the brain's input and output areas, and it is within this gap that two great mysteries seem to reside. On the one hand is the mystery of consciousness, by which we are actually aware of the world our senses reveal, and by which we experience the sensation of a colour when we look at a flower, or a pain when there is something wrong with a tooth. On the other hand is the mystery of free will, which we all believe we can exercise, but which continues to be very elusive when we ask what exactly is its nature. We know that computers have no freedom in making their decisions. The first of these mysteries, consciousness, is outside our present enquiry, but the following paragraphs are an attempt to explore in greater detail the nature of free will.
Vast areas of paper have been covered in past years and in the present day by the writings of philosophers and others on the free will versus determinism question. While some of the debate has been closely argued, and exposes clearly the difficulties inherent on both sides of the argument, some of it has been of limited value because the authors have not taken sufficient care to define clearly what they mean by "free will" or by "determinism". Indeed some of the disagreements have centred on the trivial question of how these two terms should be defined rather than on the underlying concepts themselves. In an attempt to avoid this weakness we shall give here what we hope to be acceptable and unambiguous definitions of the two terms. Whether or not the reader would give identical definitions is not an issue, but if we endeavour to hold onto exactly these same meanings throughout the rest of this essay perhaps we shall avoid some ambiguities.
A good example of a deterministic system is provided by the major bodies in the solar system. A mathematician knowing only the positions of the sun, the planets and their satellites at any one instant, and their speeds and directions of motion at this instant, can calculate their positions at any future time. Predictions far into the future may be less accurate than those nearer to the present time, but this is just because our knowledge of the present positions is less than perfect, and not because the system itself has a range of possible futures open to it. So we take as our definition of "determinism" the following:
The universe, or any self-contained part thereof, is said to be evolving deterministically if it has only one possible state at time t1 which is consistent with its state at some previous time t0 and with all the laws of nature.
Free will is often described loosely as the belief that when a person has performed a certain action, that person could have performed some other action instead. As it stands this definition is not adequate, for does it mean that this other action could have been performed if the circumstances had been slightly different, or even if all the circumstances had been exactly the same? The first alternative would be much too wide, and would admit the behaviour of mechanical devices such as automobiles and computers, all of which behave differently when they are treated differently, but which we agree do not display free will. So we shall take as our definition the following:
An agent exercises free will in performing some action A if the agent could instead have performed some other action B (or none at all) despite the fact that, right up to the moment that A was performed, every relevant circumstance had been the same.
If everything in the universe behaves deterministically according to our agreed definition, including the brains (or minds) of human beings, then clearly this is inconsistent with humans exercising free will (again according to our definition), for there can be only one possible state of the universe after a person has performed an action, and so he or she could have performed no other. Thinkers who adopt this standpoint, and admit the contradiction, are called incompatibilists. They maintain that universal determinism is incompatible with free will. Most of them do, in fact, believe the world to be deterministic, and assert that free will is an illusion. Our actions, they say, are all determined by mechanistic processes, but we have come to believe wrongly that the workings of our minds which give rise to these actions are under our autonomous control. We shall examine below the possible reasons for this misconception, if misconception it is.
Another group of thinkers would describe themselves as compatibilists. These people also accept the truth of determinism, but they adopt (explicitly or implicitly) a less restrictive definition of free will than the one quoted above. The processes in the brain (or mind) by which decisions are made often involve the weighing up of alternative courses of action, and the choosing of one from among the options which are available. Although this process is purely deterministic (they assert), nonetheless it can be described as the exercise of free will. Most importantly, they maintain that this does not negate the grounds for holding people responsible for their actions, and for praising or blaming them when we find their behaviour particularly worthy or unworthy.
A third class of philosophers call themselves libertarians. (Rather confusingly, this term is also used with an entirely different meaning in political philosophy). Believers in this doctrine do not accept that the whole universe behaves deterministically, for they maintain that the human mind (or the innermost regions of the brain) present an exception. A person's mind can itself cause that person to perform a particular action when he or she chooses so to act, without itself having any antecedent cause. The usual unbroken chain of cause and effect is broken, and when a person exercises his or her free will, a new chain is initiated by the act of will itself.
LAWS OF NATURE
There are certainly good reasons for believing that the whole of the inanimate universe behaves deterministically. Were this not so, life would be intolerable or impossible. The keys we press on a computer keyboard determine which characters appear on the screen or on our printout. Imagine how difficult learning to type would be if this were not the case! The speed and direction with which a cricket ball leaves the bowler's hand determine whether or not the ball hits the stumps. The position of the sun and the moon determine the timings and height of the tides. The exact details of an asteroid's orbit determine whether it will eventually collide with the earth.
In past times it was thought that we could never hope to understand why past events determine present and future ones with this inflexible relationship. David Hume (1711-1776) maintained that our knowledge of causation arises only because we have in the past observed many instances of the same causes being associated with the same effects. We know from past observation that an experienced cricketer is more likely than a novice to hit the stumps, and that high tides always occur when the moon is new or full. But since the time of Newton (1642-1727) we have understood with increasing clarity that the behaviour of the universe can be described by a surprisingly simple set of Laws of Nature, and that these laws often determine uniquely the future behaviour of a system when we have detailed knowledge of the present. Indeed Newton's laws themselves enable us to calculate the trajectory of a cricket ball if we know the speed and direction with which it leaves the bowler's hand, or the future motion of an asteroid if we have details of its present position and motion.
Pierre Laplace (1749-1827) presented this view forcefully by postulating an imagininary superhuman intelligence which could know precisely the state of every particle in the universe at some particular moment, and then could calculate their positions at any time in the future. In this way the whole future of the universe could be predicted. By the beginning of the twentieth century many scientists believed that they understood the laws of nature so well that, on a small scale, they could apply Laplace's method and so foretell the future of sufficiently simple systems. If it proved impossible to do this in practice because of the precision with which the initial data needed to be known, and because of the complexity of the calculations, nevertheless the argument showed, they believed, that the future is totally determined by the present, and that the truth of determinism had been established. Because these Laws of Nature appeared to constrain the whole of the inanimate universe it was very tempting to apply them also to the brain, and belief in a non-deterministic free will became less popular.
Then early in the century this view was challenged by the development of some revolutionary new ideas in what became known as the Quantum Theory. It was discovered that on the tiny scale of atomic and nuclear particles strict deterministic rules did not apply. A simple example is provided by the disintegration of radioactive atoms such as those of the element radium. The nuclei of these atoms spontaneously split into two parts, but there seems to be no cause for this to happen at any one time rather than at any other. A given atom may disintegrate tomorrow, or next year, or not for hundreds or thousands of years. What does seem certain, however, is that all such nuclei have a fixed probability of disintegrating on any particular day, and this has an important corollary. On the larger scale, when we have a piece of radium big enough to observe without a microscope, it will contain such a large number of atoms that we can calculate almost exactly what proportion will have disintegrated by any particular date, just as, if you toss a coin very many times, the proportion of "heads" will be close to one half. On a large scale, therefore, the behaviour of radioactive materials is deterministic; only on ridiculously rare occasions do sizeable quantities of such materials behave in an unexpected manner. If we have one gram of radium we can be virtually certain that one half of it will remain after 1600 years, and one quarter will remain after 3200 years. The macroscopic world, the world that we observe in everyday life, remains essentially deterministic despite the teachings of quantum theory.
DIFFICULTIESAll the standard viewpoints regarding free will are fraught with difficulties. If we believe that the brain acts deterministically we must admit to being deceived when we think "we could have acted differently". The future is uniquely determined by the past, and only one future is open to us. It is significant that we are almost always ready to justify the decisions we have made. Ask anyone "Why did you do that?" and you will usually get such a reply as, "It seemed the best course of action at the time", or "I was not prepared to let anyone stand in my way!". We are thus admitting that we responded to the most powerful of our impulses, a process giving a unique answer to the challenge we faced. This is the hallmark of a deterministic process, and while it does not indicate conclusively that we make all our decisions by means of such a process, there seems nothing to be gained in imagining otherwise, and positing that some non-physical or mystical process is involved. Occasionally we may get the reply, "I had no reason for the way I acted. I just felt that way". But is not this an admission that the choice was purely random? Indeed some philosophers have tried to sieze on the randomness of quantum behaviour to explain the apparent non-deterministic nature of our choices. Perhaps an event like the splitting of a radium atom within the brain can initiate the chain of cause and effect that results in action which we then describe as "free". But surely no-one wants to demote the exercise of free will to random choice, a process devoid of reason.
The libertarian would argue that he himself initiated the causal chain which explains his action, but this raises a fresh set of difficulties. It postulates a type of causation unlike that occurring (so far as we know) anywhere else in the universe. Events are usually caused by other events, and here we are asked to accept an event (the initiation of an action) caused by a thing (the subject himself). Even if we admit that such exceptional causes can exist within a living agent, we can ask again the same question. Was the impulse which the agent delivered to start the causal process the result of previously existing circumstances such as the agent's past experiences, prejudices, preferences, or was it not? If it was, then we have here another chain of cause and effect, and if it was not we have a purely random event. If the agent admits that the process was basically causal, but claims that he or she nevertheless retained a "casting vote" which would have been exercised in exceptional circumstances, then once again those exceptional circumstances must represent another causal chain. The only alternative to our actions being determined by cause and effect appears to be irrational randomness.
However, if we believe our brains to be mere machines, and that we are mistaken when we assert we "could have acted otherwise", what has happened to moral responsibility? When we praise someone, or blame them, we are implying that they could have acted otherwise. We give credit to people for some act of generosity only because we believe they need not have performed it. We blame them for causing an accident by driving carelessly only because we believe they could have been more careful. In everyday life, and in law, we treat people who have done wrong more leniently if they can convince us that they had no choice in what they did rather than when we believe they could have behaved better. A criminal who is mentally deranged, or under the hypnotic influence of another, is not held legally responsible for his acts, for he could not avoid acting as he did. If we are ruled by deterministic cause and effect in the same way as the non-living material world is ruled, if we never really have a choice in the actions we perform, this seems to imply that everyone should be able to plead they were unable to act otherwise, and thereby escape blame. Both the doctrine of determinism and the doctrine of free will seem to lead us to rather ridiculous conclusions.
THE NATURE OF TIME
All these arguments, however, are trumped by a proper consideration of the nature of time. In recent years many philosophers and physicists have come to realise that our intuitive perception of the passage of time is fundamentally flawed, and causes us many conceptual difficulties. Several important books have been written on the subject within the last twenty or thirty years, and a few are quoted in the bibliography at the end of this essay. There is not space here to treat the subject fully, and we can do no more than present a summary of the main conclusions, but we do ask the reader to study carefully one or more of the fuller written accounts of the true nature of time before deciding whether to accept or reject the thesis advanced in the rest of this essay. An account by the present author will be found in another part of this website at The Nature of Time, where an attempt is made to lay out all the necessary ideas. But the reader is asked to bear in mind that the quoted essay is just one chapter from a treatise on some aspects of modern physics, and to study the relevant parts in detail and ignore the parts irrelevant to our present discussion.
The first concept to be discounted is that the particular time which we call "now", or "the present", is applicable throughout the universe. We like to imagine that the time is "now" everywhere, while in fact there is nothing in the outside world to distinguish "now" from any other time, earlier or later. The concept is purely subjective, and has no significance outside the minds of individuals. There are several weighty arguments pointing to this proposition, but the most conclusive becomes immediately evident when one studies Einstein's Special Theory of Relativity, as we try to show in the article referred to above. When two observers are moving relative to each other it is easy to find distant events, such as disturbances on far-away stars, which one observer would describe as being "in the past" and the other as "in the future", showing that the "now" dividing line has a different meaning for the two observers. There is, of course, an additional complication due to the finite speed of light; we can actually witness distant astronomical events only after a delay of many years or centuries. But the effect we are describing remains even after making due allowance for these delays, and shows that the "now" of each observer is just a personal impression, with no real existence in fact.
The second misconception which we must dispel concerns the apparent "flow" of time. We think of the present moment as constantly advancing, the "now" moment as constantly changing. But if the present moment has no existence other than in our own mind, there is nothing in the outside world to change or advance. So the flowing of time must also be totally subjective. There are several other arguments which lead to the same conclusion. When we say something is changing, we mean it changes with respect to time itself; if we say the weather has changed we mean today's weather is different from yesterday's, and if we change gear in a car we mean the gear engaged now is different from that engaged a moment ago. But when we say time itself is changing, or advancing, or flowing, what can this possibly mean?
Perhaps the most significant feature of the "present moment" is the way in which it seems to divide the past from the future. The history of the world appears to be divided into two very different regions. Events in the past have already happened, and there is nothing we, or anyone else, can do to change them, while events in the future are still undecided, we believe, and in some circumstances are dependent on the choices we and others make at the present time. We are convinced that future events have none of the certainty of past events. The dividing line between these two regions is the moment we call "now"; but if we have convinced ourselves that "now" has no objective existence it follows that there can be no intrinsic difference between the nature of events on one side of the line from those on the other. Future events must really possess the same certainty that we ascribe to past ones, and our belief in the uncertainty of the future must be illusory. It is not difficult to understand why we suffer from this illusion, that the future is malleable while the past is set in stone. It clearly results from the directionality of our memory, and of all the other means by which we form records of the past. Because we can remember the past but not the future we believe mistakenly that future events themselves are less real than past ones, while in fact it is just our memories which cannot have access to them.
Convincing ourselves of these temporal misconceptions is easier than purging them from our thinking, but until we do so we are liable to reach wrong conclusions when considering the problems of philosophy, such as the free will question. We must try to view the history of the universe, or of any part of it, not as a three-dimensional world which changes from moment to moment, but as a four-dimensional world in which time is one of the dimensions. On such a representation there can be no change or motion, for time is represented by one of the dimensions, i.e. by one of the directions on a diagram, and there is nothing with respect to which change can occur. The picture is purely static. We write at greater length in the essay quoted above on the advantages and difficulties of looking at the world in this way, and we ask the reader to study it carefully before continuing with the present article.
THE IRRELEVANCE OF DETERMINISM
What has consideration of the nature of time to do with the question of determinism? We defined determinism as the thesis that the future state of the universe, or a part of the universe, is uniquely determined by its state at an earlier time. If this applies to a human agent, including the part of the brain or the mind which makes decisions, then it is inconsistent with the doctrine of free will, for it precludes the possibility of that agent performing an act other than the act actually performed. The feature of determinism which entails this conclusion is that, if the future is determined by the past, then there is only one possible future. But now, if we have shown that the past and future do not differ in their degree of certainty, does it not follow that the future is as unchangeable as the past? Whether or not it is actually determined by the past, there is only one future, and only one possible future. Free will, as prescribed by the definition we have adopted, is impossible, irrespective of whether our minds operate deterministically.
Throughout the ages some philosophers have been suspicious of arguments which rely upon such statements as "So-and-so could have behaved differently". It is certainly very difficult to say exactly what they mean, for there is no way in which they can be tested. The reason for this suspicion has not usually been a consideration of the nature of time, but has rather been related to some theory of language, or even of theology. "If something happens, then it was true in the past that it was going to happen", goes the argument. "So it could not fail to happen". This was very flimsy reasoning. It attempts to derive a fundamental property of the universe from a property of the tense structure of language. Now language is a purely human construct, which has evolved erratically and imperfectly to fulfil the needs of burgeoning civilizations, and provides an unsuitable vehicle for the derivation of the great truths of logic or of the material universe. An alternative to this argument is to claim, "If God is omnipotent and omniscient, then God knows in advance what is going to happen, and so nothing different can happen". But what is meant by God knowing anything at all? There is a general tendency to regard knowledge as one of the fundamental components of the universe, like energy or radiation, and those who advance this argument have not asked themselves what is meant by God having knowledge. Knowledge is the representation of facts by the physical arrangement of some material structure such as a brain, a computer, or ink on paper. What can possibly be meant by God knowing something? It seems likely that those who have advanced these arguments in past years have sensed something suspicious in the generally held view that alternative futures were possible, and that freely acting agents "could have behaved differently". They resorted to these rather frail arguments because they did not understand the nature of time, and wrongly attributed very different natures to the past and the future.
THE ILLUSION OF FREEDOMOur argument appears to have driven us to the conclusion that free will, as we defined it above, is an impossible concept. It seems to depend upon the false assumption that at the present moment, while the past is immutable, the future is still open, is still free, and has alternatives available to it. This, we are maintaining, is impossible, for if the past and the future differ in their nature so fundamentally there must be a clear and universal dividing line between the two, and the only such dividing line lies within the minds of individual living creatures. Because we possess memory, there is such a distinction within the awareness of each person, so that past events are fixed and future events are uncertain in our own minds, but there is nothing in the outside world corresponding to this uncertainty. How, then, can we have we such a strong impression that we are free to "make our own decisions", to overrule our baser instincts or our lazier impulses by the exercise of sheer "will-power"? And why, when we reflect on some of our past decisions, not only do we sometimes wish we had decided differently, but believe that we actually could have done so?
We all understand that most decisions are influenced primarily by what has happened immediately beforehand. We decide to take our umbrella because it is raining, to fill a kettle because a visitor has arrived, we walk to a particular railway platform because the timetable tells us that is where our train arrives. We will also admit that events in the more distant past can have an influence; we drive more carefully along a particular road because of some past frightening experience there. And we know that past events of which we have no conscious awareness can also have an effect; people differ in their responses to similar situations because of their differing natures and upbringing, even though they cannot remember the details of these influences. None of these factors justifies a belief that we act freely. The way a washing machine performs its function depends similarly upon its design, how carefully it was built, whether we have mistreated it in the past, and whether its programme has been correctly set, and yet we do not attribute free will to a washing machine.
There is, however, one additional factor which appears to distinguish a human's decision-making from that of machines and most other animals, and that is our ability to predict the likely outcomes of our decisions. There is a sense in which future events help to determine our behaviour. We fill a kettle because we can foresee ourselves making a cup of tea, and we decide to board one train rather than another because we expect that it will take us to our destination. Perhaps this is the feature of our decision-making which leads us to believe we act freely. But no, it cannot be, for we must remember that a computer can play a very powerful game of chess. It chooses each move not just by reacting to the present position of the pieces and the previous moves, but by working forwards from every possible move it can make, to find out which gives it the greatest advantage several moves ahead. So a computer can take the future into account when making its decisions just as we do, and yet again we do not attribute free will to a computer.
It is this possibility of thinking about the outcomes of several different choices that can make some decision-making such a stressful procedure. Shall we go to work today despite having all the symptoms of influenza? If we decide to go, this might make our illness worse, and could result in our missing many more days in the future. We might also give the disease to our colleagues, and may not be able to perform our work adequately. On the other hand if we do not go, our work will have piled up intolerably when we do return. Further, we may lose a day's pay, and we might be thought of less highly by our superiors if we absent ourselves without sufficient reason. It will also make it more difficult to ask for some extra holiday in the future. To reach a decision not only must we think of all these factors, but we must assess their relative importance to discover on which side the balance of argument lies. In practice we are doing no more than a computer does in planning its next move in a game of chess, but the complexity of the decision-making process, and the stress it can cause, explains why we feel that we ourselves are ultimately responsible for the choice we make. We believe the decision is a free choice which we can make either way, while in fact we are merely witnessing the execution of a computer-like algorithm leading inexorably to the decision we make. Herein must lie the explanation of the illusion of our freedom of will.
MORAL RESPONSIBILITYThis conclusion, that we are not personally responsible for our actions, that we cannot choose any course of action other than the one we do actually choose, seems to lead to some unacceptable results. If we are all mere machines, without controls which we ourselves can manipulate, why do we praise people when they do well and blame them when they do badly? Why do we thank them for their courtesies and criticise them for their selfishness or unpleasantness, if they can not behave differently? If the future is already mapped out for us, why do we strive to improve our own skills, or struggle to overcome difficulties?
In fact it is not difficult to find rational answers to these questions; the trouble is simply that they do not seem to ring true. We can hope that the praise and blame we bestow on people will have some effect on their behaviour; our statements of approval or disapproval become part of the causal chain on which their future behaviour depends. We may thank them partly because our sense of fairness requires us to do something, however small, in return for their kindnesses towards us, but also because we know it will make future kindnesses more likely. The question of free will is not relevant. Making an effort to improve can indeed affect our own future performance because such an effort itself becomes part of the causal chain which determines our own capabilities and behaviour. It is a mistake to argue that the future will be the same whether or not we make an effort; that effort is part of the chain of cause and effect which determines the future, just as the effort itself is the result of past causal chains of which we are probably unaware.
The conclusion must be that our praising, blaming, rewarding and punishing can all be justified rationally, but only on the purely practical grounds that they help to shape other people's behaviour into ways that are socially acceptable. Indeed civilised human society would be impossible were we to stop using these techniques for influencing each other. Such pragmatism may seem singularly barren and soulless, but we have developed ways of disguising it to make it more acceptable. We regard kindness, generosity and altruism as noble and admirable, while meanness, violence and dishonesty are reprehensible or wicked. And just as we have developed a rigourous and fair system for the exchange of goods and money, so civilisations have developed both formal and informal ways of dealing with the various levels of undesirable behaviour by countering it with different levels of deterrence and punishment. On the whole this system is highly effective; deterrence, punishment and the threat of punishment join all the other determining factors which control our behaviour. There are, of course, many people who behave well even without the threat of sanctions, being guided by principles with which they were imbued during their formative years, and while it may seem ungracious to explain away such high minded conduct as a further example of some automatic process of cause and effect, we can at least reward it with due respect, while continuing to recognise it as merely another strand in the causal chain that determines such persons' behaviour.
CRIME AND PUNISHMENTEvery civilised country has an elaborate system of laws which attempt to control the unacceptable behaviour of its less responsible citizens. These laws define different classes of offence, and give guidance to those who administer them on the type and severity of punishment that should be meted out for the various classes and severity of crime. Our legal system is based on the belief that criminals freely choose to behave in unacceptable ways, and so deserve to be penalized. If someone is unable to avoid committing a crime, for example, because he is in a hypnotic trance, we would not expect him to be blamed or punished for the offence, for he had no choice. Likewise, if an offence is committed by someone suffering from a severe mental derangement so that he was unaware he was doing wrong, he would not be treated as a criminal (although he may have to be consigned to a secure hospital for the protection of the public). But it is often very difficult to know where the line should be drawn, for there is a whole spectrum of mental illness, extending continuously into the realm of those who are just a little odd. Indeed it may be argued that anyone who acts in a criminal manner is not normal. And with the rapid growth of understanding in the sciences of psychology and genetics, the day will surely come when scientists can explain crime, and can predict which members of a population are the most likely to become law-breakers. We clearly can not exempt all these classes of people from the processes of law because of their genetic predisposition, for life would then be intolerable for the majority who are law-abiding.
Before considering this problem, let us examine the reasons which lie behind society's insistence that offenders must be deterred. To take a specific example, let us ask why we expect a habitual burglar to be given a period of imprisonment. The following seem to be the chief reasons:
Now what has our present question, the apparent meaninglessness of free will, to say about the problems raised in the two paragraphs above? We have seen that our system of justice tries to distinguish between offenders who choose to break the law, and those who, for one reason or another, are unable to make such a choice. We agreed that there is often difficulty knowing where to draw the line between these two groups. But this paper is claiming that no-one has this choice, for we are deceived in our belief that we could ever "have done otherwise". Does not this mean that no offenders should be disciplined? And what does it imply about the effectiveness of the various reasons we listed for imprisoning burglars? If we examine these five reasons carefully we find that none of them is weakened in its effectiveness or its justification by our view of free will. The first clearly still applies; an offender cannot commit burglaries while in prison. The second continues to provide a deterrent whether or not we believe in free will. The memory of a period in prison can become part of the causal chain that determines an offender's future behaviour, and because of our ability to predict the consequences of our actions, one can hope that the prospect of a term of imprisonment will deter some who might otherwise drift into crime. And the effectiveness of any program of training or rehabilitation is not lessened by an understanding of the way in which it can affect an offender's future conduct. Regarding the fifth reason given in our list above, it has recently become unfashionable to quote our desire for vengeance as a justification for penalising offenders, and this is unrelated to our opinions regarding freedom of the will. We like to think that our laws are just, and the scale of punishments fair, in the same sense that we expect tradesmen to charge prices that are fair. In this way we are acknowledging that one purpose of punishment is to allow society to extract some retribution for the harm an offender has inflicted upon it, but as this is for the satisfaction of those outside prison rather than the fair treatment of those inside it is probably best left unsaid.
The problems raised in the first paragraph of this section are perhaps more easily solved rather than less, when we subscribe to the thesis being advanced in this paper. We can no longer claim that criminals freely choose to offend, for it is not within their power to change the future from what it is going to be. The examples quoted in that paragraph show that it is an offender's intention to commit a crime which we try to assess before deciding a due punishment, but we have agreed that his intentions are not at his command either, and cannot be other than they actually are. It seems that the only rational criteria which the law should adopt in passing sentence are wholly pragmatic. The purpose of the legal system is to reduce crime, to protect as far as possible the law-abiding citizen from the unlawful, and the only question we should ask in dealing with offenders is what penalty will be most likely to achieve this end, in the long term as well as the short. It is surprising, perhaps, that this approach leads to conclusions so close to those reached by conventional tenets of justice. It agrees with established law that those forced to commit a crime should not be punished, for such punishment would not reduce the incidence of this sort of event in the future. For the same reason there is no point in punishing (in the conventional sense) offences committed by someone with severe mental impairement, for such a person is not likely to be dissuaded from future wrong-doing by imprisonment. On the other hand, even those suffering from some measure of inherited abnormality should be penalised for their offences, so long as there is a good chance that their penalty will render future offences less likely. The fact is that our traditional legal system, founded on the belief that we choose freely to behave well or badly, has reached conclusions very similar to those suggested by a belief that free will is an illusion. Does this suggest, perhaps, that many thinking people over the years have had an intuitive suspicion that the concept of an "open" future is flawed in some way?
CONCLUSIONTo the thinkers of ancient Greece the question of free will did not merit much attention. Today we realise that free will, if it exists at all, is possessed only by humans and perhaps some of the higher species of animal. But the ancients not only attributed the gift of freedom to the whole of the animal kingdom, they also assumed that the vagaries of the weather, the passage of the seasons and the motions of the heavenly bodies resulted from their own free will, or at least were controlled by the whimsical impulses of the gods. Free will was just part of the natural order of things, and was possessed by many of the objects in the world, even, in some philosophies, by everything that moved, including rivers and the wind. Since the seventeenth century, however, the methods and discoveries of science have encompassed an ever wider field of human experience, and our increasing understanding of the world has revealed that more and more of it is ruled by strict laws of cause and effect. In recent times such deterministic rules have been found to apply even within the human nervous system and the brain itself, and the possible areas within which any non-causal or non-material freedom of will could reside have progressively diminished.
For this reason, the argument has largely crystallised into a contest between those who believe strict determinism applies to the whole brain, as to everything else in the world, and those who expect that someday we shall discover a region within the brain where the normal rules of physical science do not apply. Each side has had its staunch supporters, but each has had to acknowledge difficulties in the doctrine to which it subscribes. The proponents of free will find it difficult to explain how decisions which are not determined by previous conditions and events can be anything other than random. On the other hand those who maintain that the brain is governed by the same causal laws as everything else, have to explain why we humans behave to each other as if we are responsible for our behaviour. Why do we feel aggrieved when others behave badly towards us, or thank them when they are kind to us, and why do we reward heroes and punish criminals, if their behaviour is uniquely determined by the forces of nature?
We have tried to show that modern beliefs concerning the true nature of time cast a new slant on the argument. The essential feature of determinism, the feature which renders it incompatible with free will, is that the future it defines is uniquely determined by the past, and so leaves no room for the alternative choices which free will demands. Now the view of time which we summarised above (and which we explain much more comprehensively in our essay The Nature of Time), does not necessarily require the future to be uniquely determined by the past; but in fact it is equally restrictive, for it asserts that there is only one possible future, and the future is unique whether or not the past determines it. So the question of determinism is no longer a factor in the argument. However, the discussion can continue to take the same course as it would if we had conclusively established the truth of determinism, for whether or not the future is ruled deterministically, our new approach still maintains that it is not open for us to manipulate.
We saw that the illusion of free will is easily explained. The brain processes which control our decision-making must be exceedingly complex. The data which determine what actions we perform must include our whole genetic make-up, all the knowledge and habits we have acquired throughout our lives, and every relevant experience we have had since birth, as well as the present state of our bodies, and the immediate events to which we are responding. Is it surprising that we feel ourselves to have the onerous task of making the decision, rather than that we are simply observing the working out of a process whose outcome is uniquely determined by all the data?
But what about moral responsibility? In fact taking on board the lessons inherent in this new approach does not affect greatly the way we should react with other people. We can continue to thank our friends for their courtesy, praise them for their achievements and on occasion we may criticise them for their failures. Our system of justice must continue to lay down fair laws and impose penalties on those who break them. We may continue to feel pride in our successes, and regret the occasions on which we did less than our best. And we must continue to strive to reach the goals we set ourselves. All these actions contribute to the chain of cause and effect which determines our own and others' behaviour, and we would all become less efficient creatures if we misread this teaching to imply otherwise.
So an understanding of this philosophy entails very little change in the way we should behave towards each other. But when we sit down to consider the ultimate reasons for our behaviour we must admit that commonly held views are faulty. We may thank our neighbours for their courtesy in order to make some small recompense for their kindness, but not because we believe they could have acted otherwise. If we criticise them for some act of disrespect it should be to discourage them from repeating such slights in future, and not because we think they could have behaved differently in the past. We can hope a criminal will receive a suitable punishment because that should deter him and others from committing further crimes, but not because we want revenge.
We must not allow this new understanding of the relationship between past and future to lead us towards a fatalistic philosophy. "If the future is fixed", you might say, "Why should I bother to do my work to the best of my ability?" Whether you are a joiner, an architect or a playwright you could say, "Why should I care how well I finish off this piece of work, if my efforts have no effect on how it will turn out at the end?" Or if you are a soldier on the field of battle you might say, "Why should I try to conceal myself? If the next bullet 'has my name on it' it will hit me whether or not I try to take precautions." This is false reasoning. Your efforts are themselves part of the causal chain that determines the quality of the piece of work in question, and whether you conceal yourself is one of the factors which determine whether or not the bullet 'has your name on it'. You know from past experience that when you strove for perfection your work was of a higher quality than when you did not. You also know the satisfaction you achieved when you produced a good product, the extent to which your skills will continue to improve over the years if you make a consistent effort, and the advantages you may gain as you are recognised as a good craftsman. All these factors go into the decision-making process and explain why, if it is a part of your nature to work hard, you continue to do so even though you recognise the process as no more than a mechanical balancing of conflicting tendencies, and that you really have no choice in the matter.
Such is the complexity of the deterministic world in which we find ourselves. How fortunate we are to live in it, and to have a part to play in the weaving of its almost unbelievably complex causal web!
BooksDavies, P.C.W., Space and Time in the Modern Universe, CUP (1977)
Davies, P.C.W., About Time, Viking (1995)
Hollis, M., Invitation to Philosophy, Blackwell (1985)
Savitt, S.F. (ed.), Time's Arrows Today, CUP (1995)
Parr, H.C., Time, Science and Philosophy, Lutterworth (1997)
Price, H., Time's Arrow and Archimedes' Point, OUP (1996)
Watson, G. (ed.), Free Will, OUP (1982)
WebsitesHazlitt, H., Free Will and Determinism
Hoefer, C., Causal Determinism
O'Connor, T., Free Will
Parr, H.C., The Nature of Time
Strawson, G., Luck Swallows Everything
Vihvelin, K., Arguments for Incompatibilism