Bernard WilliamsBernard Williams wrote an introduction and postscript to a 1960 series of BBC Third Programme shows on the topic "Freedom and the Will." The show's hosts included a number of ordinary language philosophers including Peter F. Strawson, Stuart Hampshire, David F. Pears, and Mary and Geoffrey J. Warnock. Williams' introduction showed an understanding of why Aristotle did not see the "problem of free will".
"Now Aristotle did believe, I think, that if it could be shown that it was a matter of necessity that men acted in one way rather than another, then our ordinary thought about action and responsibility would be undermined. If human actions were necessitated, there would indeed be something radically wrong with our notions of human action; but, he thought, it was quite certainly false that human actions were necessitated. If this is a correct account of Aristotle, we can see one reason why he does not confront the freewill problem. It is because he thinks that it cannot seriously be doubted that human actions are free from necessitation."Williams was mistaken about the origin of the free will problem, which he traced to Christian theology and an omniscient, omnipotent God with divine foreknowledge of the future.
"the problem of freewill makes its first large-scale appearance in a religious context, when men had come to believe that here was one God, omnipotent, omniscient, and concerned with human action. The problem of freewill was first definitely stated as a problem of Christian theology. The problem arose, in fact, from a number of different roots in Christian belief: Christianity asserts on the one hand that man does freely choose his actions, but also asserts on the other hand statements not evidently compatible with this, for instance that God being omniscient knows from all eternity what actions a man will in fact perform."Williams objected strongly to the idea that the natural laws of science could explain everything, that physical determinism required philosophers to accept Humean compatibilism.
Moral LuckIn his 1981 essay, Moral Luck, williams wrote,
I entirely agree with [Thomas Nagel] that the involvement of morality with luck is not something that can simply be accepted without calling our moral conceptions into question. That was part of my original point; I have tried to state it more directly in the present version of this paper. A difference between Nagel and myself is that I am more sceptical about our moral conceptions than he is. "Scepticism about the freedom of morality from luck cannot leave the concept of morality where it was, any more than it can remain undisturbed by scepticism about the very closely related image we have of there being a moral order, within which our actions have a significance which may not be accorded to them by mere social recognition. These forms of scepticism will leave us with a concept of morality."
Freedom and the WillWe can trace back to near the beginning of Western Civilization the notion of individual responsibility: the notion that certain states of affairs in the world can be traced to the actions of human beings, for which those human beings can in various ways be called to account. The actions for which we are in the fullest sense responsible are those for which we can be praised or blamed. This notion of responsibility is to be found already in the moral and legal thought of the Ancient Greeks. The Greeks already worked with the notion which we have today, that the question whether a man is responsible for something that happens is not just the question whether what happens is a consequence of movements of the man's body. For instance, suppose a man's arm moves in such a way as to knock over and break a valuable vase, someone else's property. In a limited sense, we know already in these circumstances what the cause of the damage was — the movement of this man's arm; but we do not yet know whether the man himself is in the full sense responsible for this damage, whether he is to blame. For that, we want to know more about the movement of his arm. In particular, we want to know whether he intended this movement of his arm, or whether perhaps it was just some nervous twitch, out of his control. Again, even if the movement was not out of his control, we still want to know, for instance, whether he realized the vase was there: if not, his breaking of the vase will be unintentional, and to that extent free from blame. There are other sorts of situation, too, that relieve people of responsibility for things that they have done, or at least mitigate it ; being in certain peculiar states, for instance, such as sleep-walking or under the influence of drugs ; or, rather differently, being forced to do things by other persons. Aristotle, in his Ethics, reviews and classifies these sorts of situation that relieve people of responsibility. His account is complex and subtle, but substantially he reduces the types of situation to two classes: those in which the agent is ignorant of relevant matters, and those in which, as he puts it, the originating principle of the action lies outside the agent himself — by which, I think, he principally means cases in which someone is physically forced to do something by someone else. Aristotle's account is of interest for two reasons in particular. The first is a positive reason: that it is interesting to find a Greek philosopher giving an analysis of responsibility and the conditions that relieve people of it that so remarkably corresponds, in its essentials, to some of the considerations that we still employ today, both in the law and in everyday life. The second reason is a more negative one. Aristotle's account is concerned, as I have said, with the conditions that relieve people of responsibility for what they have done. I mean by this that he considers and classifies certain special circumstances that relieve people of responsibility, while taking it for granted that in the usual circumstances people are responsible. He analyses the framework of praise and blame and responsibility as a going concern, as it were. He never, or scarcely ever, considers the notion that we might never really be responsible, that the going concern of praise and blame, our ordinary notions on these matters, might as a whole he founded on an illusion. Aristotle, that is to say, is precisely not concerned with that large-scale philosophical problem or set of philosophical problems which in later times has come to be known as the problem of freedom of the will. For the heart of that problem lies in the fundamental and revolutionary suggestion that our ordinary notions of responsibility may be altogether confused, because based on some false or ultimately unintelligible theory about human beings and their actions. I think we may be able to see, in historical terms, how this fundamental and revolutionary suggestion came to be made, if we ask first why Aristotle does not consider it. One reason is that Aristotle regards it as certain, and indeed makes it a central point of his philosophy, that there is no necessity about human actions — that it is never necessary in any sense that a man should, on a particular occasion, have done this rather than that. For him, necessity is something that applies only to such things as the movements of the heavenly bodies; human actions, on the contrary, are a sort of thing which could always have happened otherwise. Now Aristotle did believe, I think, that if it could be shown that it was a matter of necessity that men acted in one way rather than another, then our ordinary thought about action and responsibility would be undermined. If human actions were necessitated, there would indeed be something radically wrong with our notions of human action; but, he thought, it was quite certainly false that human actions were necessitated. If this is a correct account of Aristotle, we can see one reason why he does not confront the freewill problem. It is because he thinks that it cannot seriously be doubted that human actions are free from necessitation. This is a very important point, because, as we shall see, one reason that the freewill problem did eventually arise was that men did begin to have serious doubts about just this. I shall come back to this point. First, however, it is worth while, for our historical picture, to look briefly at another reason for Aristotle's silence on this subject. This was his lack of belief in any personal god concerned with human affairs. Some earlier Greek writers, in particular the tragedian Aeschylus, do seem to be concerned with problems not far removed from the problem of freewill. Aeschylus's portrayal of Prometheus, or again of Orestes, seems to be in part a dramatic representation of human freedom as against forces set in motion by the gods, or perhaps we should better say, personified in the gods. Plato, nearly a century later, could still, in a poetic passage, write in these terms, and assert human freedom: 'the responsibility is with the soul that chooses its destiny: God is not responsible'. For Aristotle, a little later, there are no such gods, there are no such forces, and the question does not arise. I mentioned earlier that men came eventually to doubt Aristotle's principle that human actions were not necessitated. This they did — or at least, the doubt occurred to them — with the rise of a mechanistic view of the universe, according to which the universe was a closed system, every state of which was determined as a consequence of its earlier states in accordance with natural laws — laws which, in the eighteenth and nineteenth centuries, were believed to have been in essence discovered by Newton. The possibility of such a scientific account of the universe had occurred, indeed, to certain of the Greeks, who realized further that such an account would have consequences for human action. The atomist Epicurus, for instance, was apparently aware of these problems. The speculations of the atomists, however, remained speculations; and it was only with the rise of a successful experimental and predictive mechanical science in the seventeenth century that the problems became acute. How, it was asked, could a human agent intervene in the world, be, in Aristotle's terms, an 'originating principle of action', if every event in the universe was as it was merely in virtue of the earlier states of the universe, however remote? Would not every human action be a case of `the originating principle lying outside the agent himself', so that there would be no action left for which he would be responsible? Descartes, in the first half of the seventeenth century, both held the mechanistic view of the material universe, and saw this problem; he wished further to safeguard responsible action from the realm of natural necessity. This he tried to do by distinguishing two quite separate realms, that of mind and that of matter, to the second of which alone, matter, the natural laws applied. But this was a quite inadequate kind of solution, even in Descartes's terms. Descartes himself realized that a bodily human action is itself, from one point of view, an event in the material world. If all events in the material world are subject to natural law, how does the mind come into it? Descartes' distinction between two realms, designed to insulate responsible human action from mechanical causation, insulated the world of mechanical causation, that is to say, the whole of the external world, from responsible human action. Man would be free only if there was nothing he could do. Once this way of looking at the problem is accepted, there seem to be only two possibilities. Either all physical changes are subject to natural law, in which case those particular physical changes which constitute human actions are subject to it; or human actions are not subject to natural law, so not all physical changes are subject to it. The belief that all physical changes are subject to natural law has attracted, needless to say, a label: the label 'determinism'.1 So here we have in outline an apparently simple opposition: either determinism is true, in which case there is no genuine human intervention in the world, that is, no freewill; or there is freewill, in which case determinism is false. The issue has, of course, been endlessly posed in these terms, and in these terms endlessly discussed. But not all thinkers have accepted these terms, either explicitly by rallying to one side or the other, or implicitly by professing ignorance as to which alternative was true. Right from the seventeenth century there have been philosophers who have criticized this opposition itself, and the terms in which the question was posed. Some have claimed, radically, that when the notions involved are properly understood, the alleged opposition dissolves, and the supposed dilemma can be shown to be a set of muddles and misunderstandings. Hobbes, Hume, and in our own time A. J. Ayer and others have taken this view. Others, less ambitious, have held that although there may be one, or probably more than one, genuine opposition here, nevertheless the terms of the discussion are so imprecise and ill understood that, without much further analysis, it is impossible to understand what the oppositions are, if they exist. Such a view is, I think, held by many linguistic philosophers today. It certainly does seem that a tremendous number of questions have been begged and important distinctions blurred in the simple sort of formulation of the problem that I sketched just now. First of all, the formulation introduced a good deal of metaphor: one spoke of physical changes being 'subject to' natural laws, as though the natural laws exercised some mechanistic tyranny over events. Though an enticing metaphor, it is a very misleading one, and should be eliminated. What then, without benefit of metaphor, will be the formulation of determinism? It would seem to be something like this: that given any total state-description of the universe, it is in principle possible to predict or retrodict correctly any other, however remote, in virtue of a finite set of scientific laws. But there are still difficulties about this formulation. It is still not free from unclarities and ambiguities — great difficulties, for instance, surround that slippery phrase 'in principle'. Under what conditions could we say that we could in principle predict every state of the universe ? Just when we thought that there were laws in virtue of which we could do it, if only we knew them? This would surely not be enough — this would only be a situation of faith in determinism, not a situation in which determinism had been shown to be true. So perhaps we should say that determinism had been shown to be true if we knew all the relevant laws, and could predict all the states of the universe if only we took enough trouble. But then what certainty would we have that we did know all the relevant laws, unless we could actually succeed in making these vast predictions — that is, could not only in principle, but in practice predict? And this even the most fervent determinist would surely agree we could not do. There are other difficulties in the formulation of determinism. What, for instance, are we to make of the phrase 'a total state-description of the universe' ? Even if we made more precise — as we should have to — what terms such a description would have to be in, are there not overwhelming reasons for thinking that no such description could ever be completed? the eighteenth century the astronomer Laplace could perhaps talk glibly in these terms; in the twentieth we certainly cannot. Here someone may say: Good. We now understand that the conditions specified by the determinist could never be satisfied — that is to say, determinism is a false or incoherent doctrine. So the enemy of freewill is out of business, and freewill is all right. Rut such a confident answer would be premature, and this just illustrates the obscurities of the simple opposition we originally set up. For while this extremely grandiose enemy, Laplacean determinism, is perhaps out of business, rather humbler but more effective enemies are certainly still in existence. Here it is worth while to recall the point we started with right at the beginning: the fact that in ordinary life and the law we admit certain conditions as relieving an agent from responsibility for what he has done : conditions such as somnambulism, force majeure, etc. Now it is a most conspicuous feature of our present moral thought that this list of conditions is gradually being extended in the light of advancing psychological and other scientific knowledge. For instance, we now recognize in ordinary life, and perhaps even in the law, the existence of certain compulsive conditions — kleptomania is one sort of example. The proof that an agent was in such a state relieves him of responsibility, or at least mitigates it, for actions of the appropriate type. Now it is not an accident that the discussion of such cases gives rise to constant difficulty and doubt. We feel compelled to admit more and more such conditions into the class of conditions that exonerate, without really being clear on what principle we are doing it. Moreover, there is the lurking feeling that the principle on which we are doing it might be one that eventually might extend to swallow up wide ranges of action now regarded as normal and responsible. For instance, are we to say that a man is not responsible if there is a psychological explanation of what he did ? This by itself is too weak, since the phrase ' psychological explanation' can cover practically anything, including for instance 'he freely chose'. But perhaps some sorts of psychological explanation . . . ? Here we do not know, not just where to draw the line, but how to, and in this ignorance we can feel no a priori confidence how much of our ordinary sphere of action will be left intact by the line when properly drawn. Here we feel the presence perhaps of another type of determinism less grandiose but more pressing than the total physical determinism of Laplace. In this situation, looking for a criterion, we may turn to the other side of the field. We may ask, not what are the general conditions of non-responsibility, but what are the general conditions of responsibility: perhaps there is some sign that an action is, in the appropriate sense, really ours. Here we meet for the first time a concept that has been absent from the discussion so far — the concept of the will itself. It is in virtue of the operation of the will that some philosophers have tried to distinguish this class of responsible actions. But what is the operation of the will, and what are its signs? We indeed speak in ordinary life of 'efforts of will', and it is in the occurrence of these that in reflective moments we perhaps feel most conscious in some sense of our freedom. But here there are many difficulties. First, it is certain that only a very few of the actions for which people are normally held responsible are accompanied by efforts of will, in this psychological sense; and the same goes for any other conscious process that might be suggested here instead, such as explicit decision, formulated intention and so on. Often we just act, without such processes, nor would we regard such actions as any the less responsible of free for that reason. Again, there is a deep difficulty about what an effort of will really is. There is indeed some kind of psychological process in connection with which the term is used — but might it not be just a psychological process which accompanied some actions (perhaps peculiarly difficult ones) and not others? Contrary to what some philosophers have supposed, efforts of will do not wear their metaphysical significance on their face. It may even be that they have none. I have mentioned in the last few pages a good number of problems and difficulties. It is these difficulties that contributors to this book will be considering: in particular, the definition of determinism; the search for general conditions of responsibility; the nature of the will and its connection with what we call efforts of will; the scope and implications of different kinds of psychological explanation. My aim has been to try to link with a few threads these different topics to the history of that set of issues which is over-simply called 'the problem of the freedom of the will'. I have tried to suggest that there is not one simple issue here, but a great network of difficulties. The difficulties are, however, not unrelated to one another — the network has a centre. This centre lies on a line, which is deeply engraved in human thought, the line between man as a conscious, reflective being and man as part of physical nature, conditioned by and acted upon by his environment. The basic nature of this dividing line appears in the constant preoccupation of philosophy with these difficulties! and it is its very depth that makes it necessary for any serious search for understanding in this region now to distrust grandiose theories, and to take the difficulties, so far as possible, one at a time. They are, after all, no less complex than the human situation.