-- The Determinism and Freedom Philosophy Website --
Some say there is no progress in philosophy, and certainly there is one sense in which they are wrong. There are at least significant developments in philosophical doctrines that have been persistently advocated in the past. With confidence I leave you to arrive at a satisfactory understanding of 'significant'. There is no doubt that Robert Kane has made some progress, probably more than any other contemporary philosopher, in the laying out and defending of the doctrine that an understandable freedom is importantly inconsistent with determinism, and that we do have this freedom. If the past is any guide to the present, I myself, with the aid of further study, will come to disagree. But certainly this summation of Kane's views, put together for the Determinism and Freedom Philosophy Website, is strongly commended to you.

1. The Compatibility Question

In a number of writings over the past two decades, I have sought to answer four questions about free will: (1) Is it compatible (or incompatible) with determinism? (2) Why do we want it? (3) Can we make sense of a free will that is incompatible with determinism? (4) Can such a free will be reconciled with modern images of human beings in the natural and social sciences? On all four questions, I have tried to point current debates about free will in new directions. I consider a few of these new directions in what follows.

Consider question (1)--the so-called Compatibility Question--which has received most of the recent attention in free will debates. I am an incompatibilist about free will and so I take a clear position on this question. But I have made a point of noting in my writings that if we formulate the Compatibility Question as in most textbook discussions of free will--"Is freedom compatible with determinism?"--the question is too simple and ill-formed. The reason is that there are many meanings of "freedom" and many of them are compatible with determinism. Even in a determined world, we would want to distinguish persons who are free from such things as physical restraint, addiction or neurosis, coercion, compulsion, covert control or political oppression from persons who are not free from these things; and we could allow that these freedoms would be preferable to their opposites even in a determined world.

I think those of us who believe in a free will that is incompatible with determinism--we incompatibilists and libertarians so-called--should simply concede this point to our compatibilist opponents. Many kinds of freedom worth wanting are indeed compatible with determinism. What we incompatibilists should be insisting upon instead is that there is at least one kind of freedom worth wanting that is incompatible with determinism. This significant further freedom, as I view it, is "free will," which I define as "the power to be the ultimate creator and sustainer of one's own ends or purposes." To say this further freedom is important is not to deny the importance of everyday compatibilist freedoms from coercion, compulsion, political oppression, and the like; it is only to say that human longings transcend them.

This is one shift in direction for the Compatibility Question that I insist upon in my writings. But there is another, more important one. Most philosophical debate about the incompatibility of free will and determinism has focused on the question of whether determinism is compatible with "the condition of alternative possibilities" (which I call AP)--the requirement that the free agent "could have done otherwise." Most arguments for incompatibilism, such as the "Consequence Argument" of van Inwagen and others, appeal to AP. Critics of such arguments either deny that AP conflicts with determinism or deny that alternative possibilities are required for moral responsibility or free will in the first place. There is a longstanding tendency for these contentious debates about AP and incompatibilism to stalemate over differing interpretations of "can," "power," "ability" and "could have done otherwise," though people differ about why the stalemates occur and what they mean. I have argued in my writings that there are good reasons for these stalemates having to do with the different meanings of freedom just mentioned. I argue that what the stalemates suggest is that we need to look in new directions. AP alone provides too thin a basis on which to rest the case for incompatibilism: the Compatibility Question cannot be resolved by focusing on alternative possibilities alone.

Fortunately, there is another place to look. In the long history of free will debate one can find another criterion fueling incompatibilist intuitions that is even more important than AP, though comparatively neglected. I call it the condition of ultimate responsibility or UR. The basic idea is this: to be ultimately responsible for an action, an agent must be responsible for anything that is a sufficient reason (condition, cause or motive) for the action's occurring. If, for example, a choice issues from, and can be sufficiently explained by, an agent's character and motives (together with background conditions), then to be ultimately responsible for the choice, the agent must be at least in part responsible by virtue of choices or actions voluntarily performed in the past for having the character and motives he or she now has. Compare Aristotle's claim that if a man is responsible for wicked acts that flow from his character, he must at some time in the past have been responsible for forming the wicked character from which these acts flow.

This UR condition accounts for the "ultimate" in the original definition of free will: "the power of agents to be the ultimate creators and sustainers of their own ends or purposes." Now UR does not require that we could have done otherwise (AP) for every act done of our own free wills--thus vindicating philosophers such as Frankfurt, Fischer, and Dennett, who insist that we can be held morally responsible for many acts even when we could not have done otherwise. But the vindication is only partial. For UR does require that we could have done otherwise with respect to some acts in our past life histories by which we formed our present characters. I call these "self-forming actions," or SFAs. Consider Dennett's well-known example of Martin Luther. When Luther finally broke with the Church at Rome, he said "Here I stand, I can do no other." Suppose, says Dennett, at that moment Luther was literally right. Given his character and motives, he could not then and there have done otherwise. Does this mean he was not morally responsible, not subject to praise or blame, for his act, or that he was not acting of his own free will? Dennett says "not at all." In saying "I can do not other," Luther was not disowning responsibility for his act, but taking full responsibility for acting of his own free will. So "could have done otherwise," or AP, says Dennett, is not required for moral responsibility or free will.

My response to Dennett is to grant that Luther could have been responsible for this act, even ultimately responsible in the sense of UR, though he could not have done otherwise then and there and even if his act was determined. But this would be so to the extent that he was responsible for his present motives and character by virtue of many earlier struggles and self-forming choices (SFAs) that brought him to this point where he could do no other. Often we act from a will already formed, but it is "our own free will" by virtue of the fact that we formed it by other choices or actions in the past (SFAs) for which we could have done otherwise. If this were not so, there is nothing we could have ever done to make ourselves different than we are--a consequence, I believe, that is incompatible with our being (at least to some degree) ultimately responsible for what we are. So SFAs are only a subset of those acts in life for which we are ultimately responsible and which are done "of our own free will." But if none of our acts were self-forming in this way, we would not be ultimately responsible for anything we did.

If the case for incompatibility cannot be made on AP alone, it can be made if UR is added; and thus, I suggest, the too-often neglected UR should be moved to center stage in free will debates. If agents must be responsible to some degree for anything that is a sufficient cause or motive for their actions, an impossible infinite regress of past actions would be required unless some actions in the agent's life history (SFAs) did not have either sufficient causes or motives (and hence were undetermined). I develop this line of argument in my book The Significance of Free Will 1996 (chapter 5) and in other writings, including "The Dual Regress of Free Will and the Role of Alternative Possibilities" (Philosophical Perspectives vol. 14, 2000). But this new route to incompatibility raises a host of further questions, including how actions lacking both sufficient causes and motives could themselves be free and responsible actions, and how, if at all, such actions could exist in the natural order where we humans live and have our being. These are versions of questions (3) and (4) listed above, which I call the Intelligibility and Existence questions for free will, to which I now turn.


2. The Intelligibility Question

The problem of intelligibility is an ancient one: if free will is not compatible with determinism, it does not seem to be compatible with indeterminism either. The arguments here are familiar and have been made since ancient times. An undetermined or chance event, it is said, occurs spontaneously and is not controlled by anything, hence not controlled by the agent. If, for example, a choice occurred by virtue of a quantum jump or other undetermined event in one's brain it would seem a fluke or accident rather than a responsible choice. Or look at the problem in another way that goes a little deeper. If my choice is really undetermined, that means I could have made a different choice given exactly the same past right up to the moment when I did choose. That is what indeterminism and probability mean: exactly the same past, different possible outcomes. Imagine, for example, that I had been deliberating about where to spend my vacation, in Hawaii or Colorado, and after much thought and deliberation had decided I prefered Hawaii and chose it. If the choice was undetermined, then exactly the same deliberation, the same thought processes, the same beliefs, desires and other motives--not a sliver of difference--that led up to my favoring and choosing Hawaii over Colorado, might by chance have issued in my choosing Colorado instead. That is very strange. If such a thing happened it would seem a fluke or accident, like that quantum jump in the brain just mentioned, not a rational choice. Since I had come to favor Hawaii and was about to choose it, when by chance I chose Colorado, I would wonder what went wrong and perhaps consult a neurologist. For reasons such as these, people have argued through the centuries that undetermined free choices would be "arbitrary," "capricious," "random," "irrational," "uncontrolled," and "inexplicable," not really free and responsible choices at all.

Defenders of an incompatibilist or libertarian free will have a dismal record of answering these familiar charges. Realizing that free will cannot merely be indeterminism or chance, they have appealed to various obscure or mysterious forms of agency or causation to make up the difference. Immanuel Kant said we can't explain free will in scientific and psychological terms, even though we require it for belief in morality. To account for it we have to appeal to the agency of what he called a "noumenal self" outside space and time that could not be studied in scientific terms. Many other respectable philosophers continue to believe that only some sort of mind/body dualism can make sense of free will. Science might tell us there was indeterminacy or a place for causal gaps in the brain, but a non-material self, or what Nobel physiologist John Eccles calls a "transempirical power center," would have to fill the causal gaps left by physical causes by intervening in the natural order. The most popular appeal among philosophers today is to a special kind of agent- or immanent causation that cannot be explained in terms of the ordinary modes of causation in terms of events familiar to the sciences. Free and responsible actions are not determined by prior events, but neither do they occur merely by chance. They are caused by agents in a way that transcends and cannot be explained in terms of ordinary modes of causation by events involving the agents.

I call these familiar libertarian strategies for making sense of free will "extra factor" strategies. The idea behind them is that, since indeterminism leaves it open which way an agent will choose or act, some "extra" kind of causation or agency must be postulated over and above the natural flow of events to account for the agent's going one way or another. I have been disenchanted with all these traditonal and modern extra factor strategies since my first encounters with free will issues. I agree with virtually all the objections made to them by dozens of able critics, from A. J. Ayer and C. D. Broad to present day critics such as Gary Watson, Ted Honderich, Bernard Berofsky, John Bishop, Susan Wolf and endless numbers of others. I agree with these critics that extra factor strategies--including agent-causal theories which are now so popular--do not solve the problems about indeterminism they are suppose to solve and they create further mysteries of their own. My strategy in writings on free will over the past two decades has been to see what can be done to make sense of incompatibilist free will without appealing to extra factor strategies of any kind, including special forms of agent-causation. If we are going to make progress on the Intelligibility and Existence questions about incompatibilist free will, we have to strike out in entirely new directions. To do this means rethinking issues about indeterminism and responsibility from the ground up, a topic to which I now turn. (The arguments to follow are developed in greater detail in the works listed at the end of this essay.)


3. Indeterminism and Responsibility

The first step is to note that indeterminism does not have to be involved in all acts done "of our own free wills" for which we are ultimately responsible, as argued earlier. Not all such acts have to be undetermined, but only those by which we made ourselves into the kinds of persons we are, namely "self-forming actions" or SFAs. Now I believe these undetermined self-forming actions or SFAs occur at those difficult times of life when we are torn between competing visions of what we should do or become. Perhaps we are torn between doing the moral thing or acting from ambition, or between powerful present desires and long term goals, or we are faced with a difficult tasks for which we have aversions. In all such cases, we are faced with competing motivations and have to make an effort to overcome temptation to do something else we also strongly want. There is tension and uncertainty in our minds about what to do at such times, I suggest, that is reflected in appropriate regions of our brains by movement away from thermodynamic equilibrium--in short, a kind of "stirring up of chaos" in the brain that makes it sensitive to micro-indeterminacies at the neuronal level. The uncertainty and inner tension we feel at such soul-searching moments of self-formation is thus reflected in the indeterminacy of our neural processes themselves. What is experienced internally as uncertainty then corresponds physically to the opening of a window of opportunity that temporarily screens off complete determination by influences of the past. (By contrast, when we act from predominant motives or settled dispositions, the uncertainty or indeterminacy is muted. If it did play a role in such cases, it would be a mere nuisance or fluke, as critics of indeterminism contend.)

When we do decide under such conditions of uncertainty, the outcome is not determined because of the preceding indeterminacy--and yet it can be willed (and hence rational and voluntary) either way owing to the fact that in such self-formation, the agents' prior wills are divided by conflicting motives. Consider a businesswoman who faces such a conflict. She is on her way to an important meeting when she observes an assault taking place in an alley. An inner struggle ensues between her conscience, to stop and call for help, and her career ambitions which tell her she cannot miss this meeting. She has to make an effort of will to overcome the temptation to go on. If she overcomes this temptation, it will be the result of her effort, but if she fails, it will be because she did not allow her effort to succeed. And this is due to the fact that, while she willed to overcome temptation, she also willed to fail, for quite different and incommensurable reasons. When we, like the woman, decide in such circumstances, and the indeterminate efforts we are making become determinate choices, we make one set of competing reasons or motives prevail over the others then and there by deciding.

Now let us add a further piece to the puzzle. Just as indeterminism need not undermine rationality and voluntariness, so indeterminism in and of itself need not undermine control and responsibility. Suppose you are trying to think through a difficult problem, say a mathematical problem, and there is some indeterminacy in your neural processes complicating the task--a kind of chaotic background. It would be like trying to concentrate and solve a problem, say a mathematical problem, with background noise or distraction. Whether you are going to succeed in solving the problem is uncertain and undetermined because of the distracting neural noise. Yet, if you concentrate and solve the problem none the less, we have reason to say you did it and are responsible for it even though it was undetermined whether you would succeed. The indeterministic noise would have been an obstacle that you overcame by your effort.

There are numerous examples supporting this point, where indeterminism functions as an obstacle to success without precluding responsibility. Consider an assassin who is trying to shoot the prime minister, but might miss because of some undetermined events in his nervous system that may lead to a jerking or wavering of his arm. If the assassin does succeed in hitting his target, despite the indeterminism, can he be held responsible? The answer is clearly yes because he intentionally and voluntarily succeeded in doing what he was trying to do--kill the prime minister. Yet his action, killing the prime minister, was undetermined. Or, here is another example: a husband, while arguing with his wife, in a fit of rage swings his arm down on her favorite glass-top table top intending to break it. Again, we suppose that some indeterminism in his outgoing neural pathways makes the momentum of his arm indeterminate so that it is undetermined whether the table will break right up to the moment when it is struck. Whether the husband breaks the table or not is undetermined and yet he is clearly responsible if he does break it. (It would be a poor excuse for him to say to his wife: "chance did it, not me." Even though indeterminism was involved, chance didn't do it, he did.)

Now these examples--of the mathematical problem, the assassin and the husband--are not all we want since they do not amount to genuine exercises of (self-forming) free will in SFAs, like the businesswoman's, where the will is divided between conflicting motives. The woman wants to help the victim, but she also wants to go on to her meeting. By contrast, the assassin's will is not equally divided. He wants to kill the prime minister, but does not also want to fail. (If he fails therefore, it will be merely by chance.) Yet these examples of the assassin, the husband and the like, do provide some clues. To go further, we have to add some new twists.

Imagine in cases of inner conflict characteristic of SFAs, like the businesswoman's, that the indeterministic noise which is providing an obstacle to her overcoming temptation is not coming from an external source, but is coming from her own will, since she also deeply desires to do the opposite. Imagine that two crossing (recurrent) neural networks are involved, each influencing the other, and representing her conflicting motivations. (These are complex networks of interconnected neurons in the brain circulating impulses in feedback loops that are generally involved in higher-level cognitive processing.) The input of one of these neural networks consists in the woman's reasons for acting morally and stopping to help the victim; the input of the other, her ambitious motives for going on to her meeting. The two networks are connected so that the indeterministic noise which is an obstacle to her making one of the choices is coming from her desire to make the other, and vice versa--the indeterminism thus arising from a tension-creating conflict in the will, as we said. In these circumstances, when either of the pathways "wins" (i.e. reaches an activation threshold, which amounts to choice), it will be like your solving the mathematical problem by overcoming the background noise produced by the other. And just as when you solved the mathematical problem by overcoming the distracting noise, one can say you did it and are responsible for it, so one can say this as well, I argue, in the present case, whichever one is chosen. The pathway through which the woman succeeds in reaching a choice threshold will have overcome the obstacle in the form of indeterministic noise generated by the other.

Note that, under such conditions, the choices either way will not be "inadvertent," "accidental," "capricious," or "merely random," (as critics of indeterminism say) because they will be willed by the agents either way when they are made, and done for reasons either way--reasons that the agents then and there endorse . But these are the conditions usually required to say something is done "on purpose," rather than accidentally, capriciously or merely by chance. Moreover, these conditions taken together, I argue, rule out each of the reasons we have for saying that agents act, but do not have control over their actions, (compulsion, coercion, constraint, inadvertence, accident, control by others, etc.). Of course, for undetermined SFAs, agents do not control or determine which choice outcome will occur before it occurs; but it does not follow, because one does control or determine which of a set of outcomes is going to occur before it occurs, that one does not control or determine which of them occurs, when it occurs. When the above conditions for SFAs are satisfied, agents exercise control over their future lives then and there by deciding. Indeed, they have what I call "plural voluntary control" over the options in the following sense: they are able to bring about whichever of the options they will, when they will to do so, for the reasons they will to do so, on purpose rather than accidentally or by mistake, without being coerced or compelled in doing so or willing to do so, or otherwise controlled in doing or willing to do so by any other agents or mechanisms. Each of these conditions can be satisfied for SFAs as conceived above. The conditions can be summed up by saying, as we sometimes do, that the agents can choose either way, at will.

Note also that this account of self-forming choices amounts to a kind of "doubling" of the mathematical problem. It is as if an agent faced with such a choice is trying or making an effort to solve two cognitive problems at once, or to complete two competing (deliberative) tasks at once--in our example, to make a moral choice and to make a conflicting self-interested choice (corresponding to the two competing neural networks involved). Each task is being thwarted by the indeterminism coming from the other, so it might fail. But if it succeeds, then the agents can be held responsible because, as in the case of solving the mathematical problem, they will have succeeded in doing what they were knowingly and willingly trying to do. Recall the assassin and the husband. Owing to indeterminacies in their neural pathways, the assassin might miss his target or the husband fail to break the table. But if they succeed, despite the probability of failure, they are responsible, because they will have succeeded in doing what they were trying to do.

And so it is, I suggest, with self-forming choices, except that in the case of self-forming choices, whichever way the agents choose they will have succeeded in doing what they were trying to do because they were simultaneously trying to make both choices, and one is going to succeed. Their failure to do one thing is not a mere failure, but a voluntary succeeding in doing the other. Does it make sense to talk about the agent's trying to do two competing things at once in this way, or to solve two cognitive problems at once? Well, we now know that the brain is a parallel processor; it can simultaneously process different kinds of information relevant to tasks such as perception or recognition through different neural pathways. Such a capacity, I believe, is essential to the exercise of free will. In cases of self-formation (SFAs), agents are simultaneously trying to resolve plural and competing cognitive tasks. They are, as we say, of two minds. Yet they are not two separate persons. They are not dissociated from either task. The businesswoman who wants to go back to help the victim is the same ambitious woman who wants to go to her meeting and make a sale. She is torn inside by different visions of who she is and what she wants to be, as we all are from time to time. But this is the kind of complexity needed for genuine self-formation and free will. And when she succeeds in doing one of the things she is trying to do, she will endorse that as her resolution of the conflict in her will, voluntarily and intentionally, not by accident or mistake.


4. Responsibility, Luck and Chance

You may find all this interesting and yet still find it hard to shake the intuition that if choices are undetermined, they must happen merely by chance--and so must be "random," "capricious," "uncontrolled," "irrational," and all the other things usually charged. Such intuitions are deeply ingrained. But if we are ever going to understand free will, I think will have to break old habits of thought that support such intuitions and learn to think in new ways. The first step in doing this is to question the intuitive connection in most people's minds between "indeterminism's being involved in something" and "its happening merely as a matter of chance or luck." "Chance" and "luck" are terms of ordinary language that carry the connotation of "its being out of my control." So using them already begs certain questions, whereas "indeterminism" is a technical term that merely precludes deterministic causation, though not causation altogether. Indeterminism is consistent with nondeterministic or probabilistic causation, where the outcome is not inevitable. It is therefore a mistake (alas, one of the most common in debates about free will) to assume that "undetermined" means "uncaused."

Here is another source of misunderstanding. Since the outcome of the businesswoman's effort (the choice) is undetermined up to the last minute, we may have the image of her first making an effort to overcome the temptation to go on to her meeting and then at the last instant "chance takes over" and decides the issue for her. But this is misleading. On the view I proposed, one cannot separate the indeterminism and the effort of will, so that first the effort occurs followed by chance or luck (or vice versa). One must think of the effort and the indeterminism as fused; the effort is indeterminate and the indeterminism is a property of the effort, not something separate that occurs after or before the effort. The fact that the effort has this property of being indeterminate does not make it any less the woman's effort. The complex recurrent neural network that realizes the effort in the brain is circulating impulses in feedback loops and there is some indeterminacy in these circulating impulses. But the whole process is her effort of will and it persists right up to the moment when the choice is made. There is no point at which the effort stops and chance "takes over." She chooses as a result of the effort, even though she might have failed. Similarly, the husband breaks the table as a result of his effort, even though he might have failed because of the indeterminacy. (That is why his excuse, "chance broke the table, not me" is so lame.)

Just as expressions like "she chose by chance" can mislead us in such contexts, so can expressions like "she got lucky." Recall that, with the assassin and husband, one might say "they got lucky" in killing the prime minister and breaking the table because their actions were undetermined. Yet they were responsible. So ask yourself this question: why does the inference "he got lucky, so he was not responsible ?" fail in the cases of the husband and the assassin? The first part of an answer has to do with the point made earlier that "luck," like "chance," has question-begging implications in ordinary language that are not necessarily implications of "indeterminism" (which implies only the absence of deterministic causation). The core meaning of "he got lucky" in the assassin and husband cases, which is implied by indeterminism, I suggest, is that "he succeeded despite the probability or chance of failure"; and this core meaning does not imply lack of responsibility, if he succeeds.

If "he got lucky" had other meanings in these cases that are often associated with "luck" and "chance" in ordinary usage (for example, the outcome was not his doing, or occurred by mere chance, or he was not responsible for it), the inference would not fail for the husband and assassin, as it clearly does. But the point is that these further meanings of "luck" and "chance" do not follow from the mere presence of indeterminism. The second reason why the inference "he got lucky, so he was not responsible" fails for the assassin and the husband is that what they succeeded in doing was what they were trying and wanting to do all along (kill the minister and break the table respectively). The third reason is that when they succeeded, their reaction was not "oh dear, that was a mistake, an accident--something that happened to me, not something I did." Rather they endorsed the outcomes as something they were trying and wanting to do all along, that is to say, knowingly and purposefully, not by mistake or accident.

But these conditions are satisfied in the businesswoman's case as well, either way she chooses. If she succeeds in choosing to return to help the victim (or in choosing to go on to her meeting) (i) she will have "succeeded despite the probability or chance of failure," (ii) she will have succeeded in doing what she was trying and wanting to do all along (she wanted both outcomes very much, but for different reasons, and was trying to make those reasons prevail in both cases), and (iii) when she succeeded (in choosing to return to help) her reaction was not "oh dear, that was a mistake, an accident--something that happened to me, not something I did." Rather she endorsed the outcome as something she was trying and wanting to do all along; she recognized it as her resolution of the conflict in her will. And if she had chosen to go on to her meeting she would have endorsed that outcome, recognizing it as her resolution of the conflict in her will.

Perhaps the problem is that we are begging the question by assuming the outcomes of the woman's efforts are choices to begin with, if they are undetermined. One might argue this on the grounds that "if an event is undetermined, it must be something that merely happens and cannot be somebody's choice or action." But to see how question-begging such a claim would be, one has only to note what it implies: if something is a choice or action, it must be determined--that is, "all choices and actions are determined." Is this suppose to be true of necessity or by definition? If so, the free will issue would be solved by fiat. But beyond that, there is no reason to assume such a claim is true at all. Was the husband's breaking the table not something he did simply because the outcome was not determined? Recall that "undetermined" does not mean "uncaused." The breaking of the table was caused by the swing of his arm and, though the outcome was not inevitable, that was good enough for saying he did it and was responsible. Turning to choices, a choice is the formation of an intention or purpose to do something. It resolves uncertainty and indecision in the mind about what to do. Nothing in such a description implies that there could not be some indeterminism in the deliberation and neural processes of an agent preceding choice corresponding to the agent's prior uncertainty about what to do. Recall from preceding arguments that the presence of indeterminism does not mean the outcome happened merely by chance and not by the agent's effort. Self-forming choices are undetermined, but not uncaused. They are caused by the agent's efforts.

Well, perhaps indeterminism does not undermine the idea that something is a choice simply, but rather that it is the agent's choice. But again, why must it do that? What makes the woman's choice her own on the above account is that it results from her efforts and deliberation which in turn are causally influenced by her reasons and her intentions (for example, her intention to resolve indecision in one way or another). And what makes these efforts, deliberation, reasons and intentions hers is that they are embedded in a larger motivational system realized in her brain in terms of which she defines herself as a practical reasoner and actor. A choice is the agent's when it is produced intentionally by efforts, deliberation and reasons that are part of this self-defining motivational system and when, in addition, the agent endorses the new intention or purpose created by the choice into that motivational system as a further purpose to guide future practical reasoning and action.

Well, then, perhaps the issue is not whether the undetermined SFA is a choice, or even whether it is the agent's choice, but rather how much control she had over it. It may be true, as I argued earlier (in the discussion of plural voluntary control), that the presence of indeterminism need not eliminate control altogether. But would not the presence of indeterminism at least diminish the control persons have over their choices and other actions? Is it not the case that the assassin's control over whether the prime minister is killed (his ability to realize his purposes or what he is trying to do) is lessened by the undetermined impulses in his arm--and so also for the husband and his breaking the table? And this limitation seems to be connected with another problem often noted by critics of libertarian freedom--the problem that indeterminism, wherever it occurs, seems to be a hindrance or obstacle to our realizing our purposes and hence an obstacle to (rather than an enhancement of) our freedom.

There is something to these claims, but I think what is true in them reveals something important about free will. We should concede that indeterminism, wherever it occurs, does diminish control over what we are trying to do and is a hindrance or obstacle to the realization of our purposes. But recall that in the case of the businesswoman (and SFAs generally), the indeterminism that is admittedly diminishing her control over one thing she is trying to do (the moral act of helping the victim) is coming from her own will--from her desire and effort to do the opposite (go to her business meeting). And the indeterminism that is diminishing her control over the other thing she is trying to do (act selfishly and go to her meeting) is coming from her desire and effort to do the opposite (to be a moral person and act on moral reasons). So, in each case, the indeterminism is functioning as a hindrance or obstacle to her realizing one of her purposes--a hindrance or obstacle in the form of resistance within her will which has to be overcome by effort.

If there were no such hindrance--if there were no resistance in her will--she would indeed in a sense have "complete control" over one of her options. There would no competing motives that would stand in the way of her choosing it. But then also she would not be free to rationally and voluntarily choose the other purpose because she would have no good competing reasons to do so. Thus, by being a hindrance to the realization of some of our purposes, indeterminism paradoxically opens up the genuine possibility of pursuing other purposes--of choosing or doing otherwise in accordance with, rather than against, our wills (voluntarily) and reasons (rationally). To be genuinely self-forming agents (creators of ourselves)--to have free will--there must at times in life be obstacles and hindrances in our wills of this sort that we must overcome.

Let me conclude with one final objection that is perhaps the most telling and has not yet been discussed. Even if one granted that persons, such as the businesswoman, could make genuine self-forming choices that were undetermined, isn't their something to the charge that such choices would be arbitrary? A residual arbitrariness seems to remain in all self-forming choices since the agents cannot in principle have sufficient or overriding prior reasons for making one option and one set of reasons prevail over the other. There is some truth to this charge as well, but again I think it is a truth that tells us something important about free will. It tells us that every undetermined self-forming free choice is the initiation of what I have elsewhere called a "value experiment" whose justification lies in the future and is not fully explained by past reasons. In making such a choice we say, in effect, "Let's try this. It is not required by my past, but it is consistent with my past and is one branching pathway my life can now meaningfully take. Whether it is the right choice, only time will tell. Meanwhile, I am willing to take responsibility for it one way or the other."

It is worth noting that the term "arbitrary" comes from the Latin arbitrium, which means "judgment"--as in liberum arbitrium voluntatis, "free judgment of the will" (the medieval philosophers' designation for free will). Imagine a writer in the middle of a novel. The novel's heroine faces a crisis and the writer has not yet developed her character in sufficient detail to say exactly how she will act. The author makes a "judgment" about this that is not determined by the heroine's already formed past which does not give unique direction. In this sense, the judgment ( arbitrium) of how she will react is "arbitrary," but not entirely so. It had input from the heroine's fictional past and in turn gave input to her projected future. In a similar way, agents who exercise free will are both author's of and characters in their own stories all at once. By virtue of "self-forming" judgments of the will (arbitria voluntatis) (SFAs), they are "arbiters" of their own lives, "making themselves" out of past that, if they are truly free, does not limit their future pathways to one.

Suppose we were to say to such actors "But look, you didn't have sufficient or conclusive prior reasons for choosing as you did since you also had viable reasons for choosing the other way." They might reply. "True enough. But I did have good reasons for choosing as I did, which I'm willing to stand by and take responsibility for. If they were not sufficient or conclusive reasons, that's because, like the heroine of the novel, I was not a fully formed person before I chose (and still am not, for that matter). Like the author of the novel, I am in the process of writing an unfinished story and forming an unfinished character who, in my case, is myself.



When I began discussing the Intelligibility Question several sections ago, I said I would avoid appealing to any "extra factors" to account for libertarian free agency, such as noumenal selves, transempirical power centers, or special forms of agent- or nonevent causation, that libertarians usually invoke. The preceding account makes no such appeals. It does appeal to the fact that free choices and actions can be caused by efforts, deliberations, beliefs, desires, intentions and other reasons or motives of agents. But this is causation by events or states of affairs involving agents. It is not the special causation of agent-causal theories that cannot be spelled out in terms of events or states of affairs involving agents, either physical or psychological. Moreover, causation by efforts, beliefs, desires, intentions and the like is something that compatibilists also appeal to in their accounts of free actions and choices; and it is hard to see how they could give accounts of free agency without doing so. The case is otherwise with such things as noumenal selves, transempirical power centers or nonevent causation, which are invoked specifically to salvage libertarian intuitions about free will and are not needed by non-libertarians. This is what I mean by not invoking "extra" factors. My account of free will postulates no additional ontological entities or relations that non-libertarian acounts of free agency do not also need. The only added assumption I have made to account for libertarian free agency is just what you would expect--that some of the mental events or processes involved must be undetermined, so that the causation by mental events may be nondeterministic or probabilistic as well as deterministic.

Of course, if any such theory is to succeed, there must be some indeterminism in the brain where undetermined efforts and choices occur. But such a requirement holds for any libertarian theory. If free choices are undetermined, as libertarians suppose, there must be some indeterminacy in the natural world to make room for them; and it is an empirical question whether the indeterminism is there. This is true even if one postulates special kinds of agent-causes or a non-material self to intervene in the brain. My suggestion about how indeterminism might enter the picture, if it were available in the physical world, was that conflicts in the wills of agents associated with self-forming choices "stir up chaos" in the brain sensitizing it to quantum indeterminacies at the neuronal level, which would then be magnified to effect neural networks as a whole. The brain would thus be stirred up by such conflict for the task of creative problem solving. This is speculative to be sure. Others have suggested different ways in which indeterminacy might be involved in the brain and free will. But such speculations are not entirely idle either. There is growing evidence that chaos may play a role in human cognitive processing, as it does in many complex physical systems, providing some of the flexibility that the nervous system needs to adapt creatively to an ever-changing environment. Of course, chaotic behavior, though unpredictable, is usually deterministic and does not of itself imply indeterminism. But chaos does involve "sensitivity to initial conditions." Minute differences in the initial conditions of chaotic systems, including living things, may be magnified giving rise to large-scale undetermined effects. If the brain does "make chaos to understand the world" (as one recent research paper puts it ), its sensitivity to initial conditions may magnify quantum indeterminacies in neural networks whose outputs can depend on minute differences in the timing of firings of individual neurons. The general idea is that some combination of quantum physics and the new sciences of chaos and complexity in self-organizing systems may provide sufficient indeterminacy in nature for free will. But it is only an idea. The question is ultimately an empirical one, to be decided by future research.

What I have tried to do in this paper is answer a different, but equally daunting, question: what could we do with the indeterminism to make sense of free will, supposing it were there in the brain? Some people say we couldn't do anything without adding some extra factor in the form say of a special kind of nonevent or nonoccurrent agent-causation. What would such a factor add? Some say it would add the idea that the agent caused the free action, something one cannot capture by event causation alone. But this is mistaken. The theory presented here does postulate agent causation (though not of the nonevent or nonoccurrent kind). Agents cause or bring about their undetermined self-forming choices (SFAs) by making efforts to do so, voluntarily and intentionally; and agents cause or bring about many other things as well by making efforts to do so, such as deaths of prime ministers, broken tables, messes, accidents, fires, pains, and so on. Whether there is agent causation in general is not the issue here. What is at issue is agent-causation (hyphenated)--a sui generis form of causation postulated by agent-causal theorists that cannot be spelled out in terms of events and states of affairs involving the agents. The fact is that both sides believe in agent causation. The issue is how it is to be spelled out.

Some say we need a special kind of nonevent agent-causation to account for agent control and production of action. But I have given an account of control over actions--indeed plural voluntary control--earlier without invoking such a notion. As for production, just as, on the theory presented, agents can be said to cause their self-forming choices (SFAs) and many other things, so it can be said on this theory that agents produce or bring about their self-forming choices by making efforts to do so and they produce many other things by their choices, efforts and other actions. Is there a residual fear functioning here that the "agent" will somehow disappear from the scene if we describe its capacities and their exercise, including free will, in terms of states and events? Such a fear would be misguided. A continuing substance (such as an agent) does not absent the ontological stage because we describe its continuing existence--its life, if it is a living thing--including its capacities and their exercise, in terms of states of affairs, events and processes involving it. One needs more reason than this to think that there are no continuing things or substances, or no agents, but only events, or to think that agents do not cause things, only events cause things. There is nothing inconsistent in saying that agents are continuing substances and holding that the lives of agents, their capacities and the exercise of those capacities, including free will, must be spelled out in terms of states, processes and events involving them.


Readings: More detailed discussion and defense of the ideas of this paper may be found in two books, Free Will and Values (1985) and especially The Significance of Free Will (Oxford, 1996; paperback, 1998) and numerous articles, including "Responsibility, Luck and Chance: Reflections on Free Will and Indeterminism" (Journal of Philosophy, May, 1999) and "The Dual Regress of Free Will and the Role of Alternative Possibilities" (Philosophical Perspectives 14 (December 2000).

There is some surprising science  that may be in accord with Kane's outlook, although he does not depend on it. For an unpersuaded account of it, see Is the Mind Ahead of the Brain -- Benjamin Libet's Evidence Examined and also Is the Mind Ahead of the Brain -- Rejoinder to Benjamin Libet.

BACK : Determinism & Freedom Philosophy
BACK: T.H. front page