1. Objections and Responses (I): Indeterminism, Causation and Chance
You might find the preceding view of free will 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 rooted in our thinking and difficult to shake. But if we are going to understand free will, I think we have to break old habits of thought supporting such intuitions and learn to think in new ways.
The first step in responding to these common charges is to question the intuitive connection in people’s minds between “indeterminism’s being involved in something” and “it’s happening merely as a matter of chance or luck.”
Step 12: Chance and luck are terms of ordinary language that carry the meaning of “it’s being out of my control.” So using them already begs certain questions. Whereas “indeterminism” is a technical term that merely rules out deterministic causation, though not causation altogether. Indeterminism is consistent with nondeterministic or probabilistic causation, where the outcome is caused, but is not inevitable. It is thus a mistake (in fact, one of the most common mistakes in debates about free will to assume that “undetermined” means “uncaused.”
A related mistake (which also occurs in debates about free will and in some textbook discussions of it) is to define determinism as the view that “every event has a cause” or “every event is caused.” This has led many to suppose that if free choices must be undetermined, as incompatibilists and libertarians about free will affirm, then libertarian free choices must also be uncaused. But this conclusion again does not follow, as step 12 makes clear. Indeterminism merely rules out deterministic causation, where the outcome is inevitable, it does not rule out causation altogether.
Here is another source of misunderstanding. Since the outcome of the businesswoman’s effort (her self-forming choice to stop and help the assault victim) is undetermined up to the last minute, we may imagine her first making an effort to overcome temptation (to go on to her meeting) and then at the last instant “chance takes over” and decides the issue for her. But this image is misleading.
Step 13: On the view presented, one cannot separate the indeterminism and the effort of will the agent is making, so that first the effort occurs followed by chance or luck (or vice versa). Rather, 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 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 because of the indeterminacy.
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.
2. Objections and Responses (II): Responsibility, Luck, Choice and Action
Just as expressions like “she chose by chance” can mislead in such contexts, so can expressions like “she got lucky.” Recall that in the cases of the assassin and the husband, one might say “they got lucky” in killing the prime minister and breaking the table because their actions were undetermined and might have failed. Yet, as noted, it does not follow that they were not responsible.
So ask yourself this question: why does the inference “he got lucky, so he was not responsible” fail when it does 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).
Step 14: The core meaning of “he got lucky” in the assassin and husband cases, which is implied by indeterminism 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 further meanings in the husband and assassin cases that are often associated with “luck” and “chance” in ordinary usage (e.g., the outcome was not his doing, or occurred by mere chance, or he was not responsible for it), then the inference (“he got lucky, so he was not responsible“) would not fail for the husband and assassin, as it clearly does. What the failure of the inference shows 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, knowingly and purposefully, not by mistake or accident.
Step 15: But note that these reasons why the inference (“he got lucky, so he was not responsible”) fails 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 acknowledged the outcome as something she was trying and wanting to do all along, recognizing it as her resolution of the conflict in her will. And if she had chosen to go on to her meeting she would have thereby endorsed that outcome, recognizing it as her resolution of the conflict in her will.
3. Objections and Responses (III): Efforts, Introspection and Free Will
Another frequently-made objection is that we are not introspectively or consciously aware of making dual efforts and performing multiple cognitive tasks in self-forming choice situations. But I am not claiming that agents are introspectively aware of making dual efforts. What persons are introspectively aware of in SFA situations is that they are trying to decide about which of two options to choose and that either choice is a difficult one because there are resistant motives pulling them in different directions that will have to be overcome, whichever choice is made. In such introspective conditions, I am theorizing that what is going on underneath is a kind of distributed processing in the brain that involves separate attempts or endeavorings (efforts) to resolve competing cognitive tasks.
There is a larger point here that I have often emphasized:
Step 16: Introspective evidence cannot give us the whole story about free will. Stay on the introspective surface and libertarian free will is likely to appear obscure or mysterious, as it so often has in history. What is needed is a theory about what might be going on behind the scenes when we exercise such a free will, not merely a description of what we immediately experience; and in this regard new scientific ideas can be a help rather than a hindrance to making sense of free will.
It is now widely believed that parallel processing takes place in the brain in such cognitive phenomena as visual perception. The theory is that the brain separately processes different features of the visual scene, such as object and background, through distributed and parallel, though interacting, neural pathways or streams.[1]
Suppose someone objected that we are not introspectively aware of such distributed processing in ordinary cases of perception. That would hardly be a decisive objection against this new theory of vision. For the claim is that this is what we are doing in visual perception, not necessarily that we are introspectively aware of doing it. And I am making a similar claim about free will. If parallel distributed processing takes place on the input side of the cognitive ledger (in perception), then why not consider that it also takes place on the output side (in practical reasoning, choice and action)? That is what I am suggesting we should suppose if we are to make sense of libertarian free will.
4. Objections and Responses (IV): Efforts, Will-setting and Rationality
It has also been commonly objected that it is irrational to make efforts to do incompatible things. I concede that in most ordinary situations it is. But I contend that
Step 17: There are special circumstances in which it is not irrational to make competing efforts. These include circumstances in which (i) we are deliberating between competing options; (ii) we intend to choose one or the other, but cannot choose both; (iii) we have powerful motives for wanting to choose each of the options for different and competing reasons; (iv) there is a consequent resistance in our will to either choice, so that (v) if either choice is to have a chance of being made, effort will have to be made to overcome the temptation to make the other choice; and most importantly, (vi) we want to give each choice a fighting chance of being made because the motives for each choice are important to us. And of being these are the conditions of “will-setting” or “self-forming” actions (SFAs).
It is critical here to recognize the uniqueness of such “will-setting” situations. For our normal intuitions about efforts are formed in everyday situations in which our will is already “set one way” on doing something, where obstacles and resistance have to be overcome if we are to succeed in doing it. We want to open a door, which is jammed, so we have to make an effort to open it. In such everyday situations, it would be irrational to make incompatible efforts because our wills are already set on doing what we are trying or endeavoring to do.
There are, in other words, “rationality constraints” on making efforts in will-settled situations because it is irrational to attempt to do contrary things when one’s will is already set on doing one of them (and even more irrational if one’s will were set on doing both). But will-setting situations, such as those of SFAs, represent a third alternative in which one’s will is not yet set on doing either of the things one is trying to do, but where one has strong reasons for doing each (e.g., deciding to A and deciding to B), and neither set of reasons is as yet decisive.
Step 18: Because most efforts in everyday life are made in will-settled situations, where our will is already set on doing what we are trying to do, we tend to assimilate all effort-making to such situations, thereby failing to consider the uniqueness of will-setting, which occurs in self-forming choice situations, and is of a piece, in my view, with the uniqueness of free will.
It is also important to note that effort is not the only term that could be used to describe these competing cognitive activities preceding self-forming choices. One could as well use other terms (as I have done above), such as tryings or attempts or endeavorings.[2] Or, one could give these distributed cognitive activities a distinctive name, as I have also sometimes done, and call them volitional streams. What is important, however, whatever one calls them, is that these efforts or volitional streams be conceived as goal-directed cognitive activities of agents whose goals are deciding or “setting one’s will” in a certain way.[3] They our not mere events that happen to agents.
5. Objections and Responses (V): Luck and Chance Revisited
Return now to further objections about luck and chance, which have taken many forms in contemporary debates about free will. Here is another form such objections have frequently taken:
If the occurrence of a choice depends on the occurrence of some undetermined or chance events (say, quantum events) in the brain over which the agent lacks control, then whether or not the choice occurs would appear to be a matter of luck or chance, rather than something the agent brought about and was responsible for. Such thoughts, as noted earlier, send us scurrying around looking for extra factors, other than prior events or happenings, to “tip the balance” to one choice or the other, such as an immaterial agent or (non-event) agent-cause.
But there is an alternative way to think about how indeterminism might be involved in free choice that first occurred to me thirty years ago, a way that avoids such familiar strategems and requires a transformation of perspective. It was stated earlier as Step 6 (of the preceding Part 3), which I repeat here for convenience.
Step 6: Think, instead, of the indeterminism involved in free choice, not as a cause acting on its own, but as an ingredient in larger goal-directed or teleological activities of the agent, in which the indeterminism functions as a hindrance or interfering element in the attainment of the goal.
Such is the role of indeterminism in the “efforts,” or “volitional streams” preceding undetermined SFAs. Each of these efforts is a temporally extended goal-directed activity, whose goal is a particular choice and whose input consists in the reasons or motives for making that particular choice, in which indeterminism is a hindering or interfering element. The choices or SFAs that result from these temporally extended activities, thus do not pop up out of nowhere, even though undetermined. They are achievements of goal-directed activities of the agent that might have failed due to the indeterminism, but did not.
Step 19: Note a further important consequence of viewing the indeterminism involved in free choice in this way. If indeterminism or chance does play this kind of hindering or interfering role in larger goal-directed processes (the efforts) leading to choice, the indeterminism or chance will not be the cause of the choice that occurs.
This follows from a general point about probabilistic causation. A vaccination may hinder or lower the probability that I will get a certain disease, so it is causally relevant to the outcome. But if I get the disease despite it, the vaccination is a not the cause of my getting the disease, though it was causally relevant, because its role was to hinder that effect and thus to lower the probability of my getting the disease. By contrast, the causes of my getting the disease are those causally relevant factors (such as the infecting virus) that significantly raised or enhanced the probability of its occurrence. Similarly,
Step 20: In the businesswoman’s case (and SFAs generally), the causes of the choice she does make (the moral choice or the ambitious choice) are those causally relevant factors that significantly raised the probability of making that choice from what it would have been if those factors had not been present. These factors include her reasons and motives for making that choice rather than the other, her conscious awareness of these reasons and, importantly, her deliberative efforts to overcome the temptations to make the contrary choice. By contrast, the presence of indeterminism lowers the probability that the choice will result from these reasons, motives and efforts from what that probability would have been if there had been no competing motives or efforts and hence no interfering indeterminism. The indeterminism is therefore not the cause of the choice that is actually made, though it is causally relevant, since that choice is made despite its presence.
Step 21: Moreover, since those causally relevant features of the agent, which can be counted among the causes of the woman’s choice, are her reasons or motives, her conscious awareness and her deliberative activity including her effort, we can also say that she, the agent, is the cause of the choice. The indeterminism or chance (like the vaccination) was causally relevant to the outcome, but it was not the cause. This explains why the husband’s excuse was so lame when he said “Chance broke the table, not me.” The chance was a hindering factor, not the cause.
6. Objections and Responses (VI): Purposive Action, Neurons, and Macro-control
But isn’t it the case, one might ask, that whether agents in SFA situations succeed in choosing A rather than B (or vice versa) (i) depends on whether certain neurons involved in their cognitive processing fire or not (perhaps within a certain time frame). And isn’t it the case that (ii) whether or not these neurons fire is undetermined, and hence a matter of chance, so that (iii) the agent does not have control over whether or not they fire?
And if these claims are true, it would seem the choice merely “happened” as a result of these chance firings and so (iv) the agent did not make the choice of A rather than B (or vice versa) and (v) hence was not responsible for making it. To many persons, this line of reasoning clinches the matter. It looks like the outcome of choices that are not determined must be merely a matter of chance or luck after all.
Step 22: But this reasoning is too hasty. For the really astonishing thing is that, even if the first three of these claims ((i)-(iii)) are true, the conclusions that (iv) the agent did not make the choice and (v) was not responsible for making it, do not follow, when the following three further conditions obtain: (a) The choosing of A rather than B (or B rather than A, whichever occurs) was something the agents were endeavoring or trying to bring about, (b) the indeterminism in the neuron firings was a hindrance or obstacle to the achievement of that goal and (c) the agents nonetheless succeeded in achieving the goal despite the hindering effects of the indeterminism.
For, consider the husband swinging his arm down on the table. It is also true in his case that (i) whether or not his endeavoring or trying to break the table succeeds “depends” on whether certain neurons in his nervous system fire or do not fire. And it is also true in his case that (ii) whether these neurons fire or not is undetermined and hence a matter of chance and therefore (iii) not under his control. Yet, even though we can say all this, it does not follow that (iv) the husband did not break the table and that (v) he is not responsible for breaking it, if his endeavoring or trying to do so succeeds.
And this will be true in the businesswoman’s case as well, whichever way she chooses, since the three conditions, (a), (b), and (c), will also obtain in her case whichever choice is made. Whichever choice she makes (a) will be the result of her endeavoring or trying to bring about that choice against resistance in her will, (b) the indeterminism in the neuron firings (stirred up by the conflict in her will) will have been a hindrance or obstacle to the achievement of that goal and (c) she will nonetheless have succeeded in making that choice, despite the effects of the indeterminism.
Astonishing indeed! But this is the kind of surprising result one gets when indeterminism or chance plays an interfering or hindering role in larger goal-directed activities of agents that may succeed or fail.[4]
Step 23: It is well to meditate on this: We tend to reason that if an action (whether an overt action of breaking a table or a mental action of making a choice) depends on whether certain neurons fire or not (in the arm or in the brain), then the agent must be able to make those neurons fire or not, if the agent is to be responsible for the action. In other words, we think we have to crawl down to the place where the indeterminism originates (in the individual neurons) and make them go one way or the other, if we (and not chance) are to be responsible for the outcome. And we realize, of course, that we can’t do that. But we don’t have to. It is the wrong place to look. We don’t have to micro-manage our individual neurons one by one to perform purposive actions and we do not have such micro-control over our neurons even when we perform ordinary actions such as swinging an arm down on a table.
Step 24: What we need when we perform purposive activities, mental or physical, is macro-control of processes involving many neurons—processes that may succeed in achieving their goals despite the interfering or hindering effects of some recalcitrant neurons. We do not micro-manage our actions by controlling each individual neuron or muscle that might be involved. But that does not prevent us from macro-managing our purposive activities (whether they be mental activities such as practical reasoning, or physical activities, such as arm-swingings) and being responsible when those purposive activities attain their goals.
7. Objections and Responses (VII): Agency, Complexity, The Disappearing Agent Objection
But don’t we have to postulate an additional kind of “agent-causation” over and above causation by states and events involving the agent to fully capture libertarian free will? This is another one of those questions that has had a hypnotic effect on free will debates, reflecting deeply-rooted intuitions. And it has led many to believe that in order to make sense of libertarian free will, one must postulate a special, metaphysically primitive, kind of causation by an agent, or substance (often designated “agent-causation,” with a hyphen) that cannot be spelled out in terms ordinary kinds of causation by states and events familiar to the psychological and physical sciences.
There is 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.[5] This fear is sometimes formalized by philosophers as the “disappearing agent” objection to event-causal or causal indeterminist accounts of free will. But such a fear is misguided at best.
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 substances, or no agents, but only events, or that agents do not cause things, only events cause things. I agree with Aristotle when it comes to thinking about the nature of living things and the relation of mind to body. Human agents are continuing substances with both mental and physical properties. But there is nothing inconsistent in saying this and thinking 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.
Step 25: In contemporary debates about free will, the view I have been developing is often referred to as an “event-causal” view of free will and contrasted with “agent-causal” views (and with noncausalist views). This designation of the view as “event-causal” is unfortunate in many respects because it has suggested to some persons that those who believe in event-causal views of free will do not believe in agent causation and that somehow one has to choose between event causation and agent causation in giving an account of free will.
This is a mistake. What those who believe in event-causal views of free will reject is not agent causation in general, but agent-causation (hyphenated), a special metaphysically primitive kind of causation by an agent or substance that cannot be spelled out in terms of states and events involving the agent. In sum,
Step 26: One does not have to choose between agent (or substance) causation and event causation in describing freedom of choice and action. You can affirm both, as I do. In the case of self-forming choices, for example, it is true to say both that “the agent’s deliberative activity, including her effort, caused or brought about the choice,” and to say that “the agent caused or brought about the choice.” Indeed the first claim entails the second.
Step 27: As noted, “a continuing substance (such as an agent) does not absent the ontological stage (“disappear”) 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.” Such event descriptions are not meant to deny that agents cause their free choices and actions. Rather they spell out in more detail (thus providing further information about) how and why the agents do so.
Step 28: Of relevance here, as I’ve emphasized in many writings, is a peculiarly modern scientific way of understanding human agency and causation by agents that has roots in the ancient view of Aristotle just mentioned. Agents, according to this modern conception with ancient roots, are conceived as information-responsive complex dynamical systems.
Complex dynamical systems are the subject of “dynamical systems theory” and also of what is sometimes popularly called “complexity theory.” They are systems (which are now known to be ubiquitous in nature and include living things) in which new emergent capacities arise as a result of greater complexity or as the result of movement away from thermodynamic equilibrium toward the edge of chaos. When these emergent capacities arise in complex dynamical systems, the systems as a whole impose novel constraints on the behavior of their parts that did not constrain the parts before the new complexity or disequilibrium was achieved. In such complex dynamical systems there is thus a reciprocal causal influence of wholes to parts and parts to wholes.[6]
Thus, in the account of free will I’ve proposed, it is a conflict in the larger motivational system of the agent taken as a whole—the self-network, as I have elsewhere called it—that stirs up chaos and amplifies indeterminism at the neuronal and synaptic levels; and the resulting amplified indeterminism in turn interferes with the goal-directed activities of the larger network. There is thus a mutual influence of wholes to parts and parts to wholes characteristic of complex dynamical systems. And “emergent” capacities are also involved.
Step 29: Only when creatures attain the kind of inner complexity capable of giving rise to conflicts in their wills, or motivational systems, does the capacity for self-formation characteristic of free will arise. So we are talking about a special kind of living complex dynamical system that is information-responsive in highly complex ways, not seen in non-rational animals. The businesswoman, as I said, is torn inside by different visions of who she is and what she wants to be, as we all are from time to time; and this is the kind of complexity needed for the novel capacity of genuine self-formation or free will to “emerge.”
There is more to be said about how viewing human agents in this way—as dynamical systems that are information-responsive in highly complex ways—can help to explain the kinds of control that agents can exercise over the deliberative activities and efforts themselves that lead to self-forming choices. These additional steps emerge as we consider further objections.
8. Objections and Responses (VIII): Control and Responsibility: Three Assassins
One such further objection concerns control and responsibility. Does not the presence of indeterminism or chance, one might ask, at least diminish the control persons have over their choices or actions? And would that not affect their responsibility? Is it not the case, for example, 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?
The answer to this last question is yes: Their control is diminished by the presence of the indeterminism. But a further surprising point worth noting—a point that I believe is often missed—is that diminished control in such circumstances does not entail diminished responsibility when the agents succeed in doing what they are trying to do. Ask yourself this question: Is the assassin any less guilty of killing the prime minister, if he did not have complete control over whether he would succeed because of the indeterminism in his neural processes?
Suppose there were three assassins, each of whom killed a prime minister. Suppose one of them had a 50% chance of succeeding because of the indeterministic wavering of his arm. Another had an 80% chance, and the third near a 100% chance. (With this third assassin there was no wavering at all; he was a young stud assassin.) Is one of these assassins less guilty than the other, if they all succeed? Should we say that one assassin deserves a hundred years in jail, the other eighty years and the third fifty years? Absurd. They are all equally guilty if they succeed. The diminished control in the assassins who had an 80% or 50% chance does not translate into diminished responsibility when they succeed.
Imagine a lawyer for the 50% assassin arguing that his client was not guilty because the prime minister’s dying as a result of what his client did was a “matter of chance.” Therefore chance was the cause of the prime minister’s death, not his client. That would make the notorious “Twinkie Defense” look brilliant by comparison. (This was the defense offered by a lawyer in California that his client was not responsible for a crime because the client’s blood sugar was so high from having eaten too many Twinkies that he could not control his actions.)
Step 30: There is an important further lesson here about free will in general. 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).
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 be no competing motives standing in the way of her choosing it and therefore no interfering indeterminism. 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.
Step 31: 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.
Libertarians about free will have traditionally tended to play down this aspect of indeterminism. They believed indeterminism was required for free will on their view, but they assumed the indeterminism could be entirely circumvented by special agencies. But hindrances and obstacles and resistance in the will are precisely what are needed for self-formation and free will, which, like life itself, involves struggle.
Some notable historical figures, who discussed free will from a religious perspective, such as St. Augustine, related this fact to the problem of evil. There must be hindrances and obstacles to our choices and resistance in our own wills to be overcome, if we are to be capable of genuine self-formation and free will. Compare Evodius’s question to Augustine (in Augustine’s classic work On the Free Choice of the Will) of why God gave us free will since it brings so much conflict, struggle and suffering into the world. It does bring struggle, hindrances and resistance in our wills. But such things are necessary for genuine responsibility.
Of interest here also is Kant’s image, which I have used before, of the bird that is upset by the resistance of the air and the wind to its flight and so imagines that it could fly better if there were no air at all to resist it. But of course, as Kant points out, the bird would not fly better if there were no air. It would cease to fly at all. So it is with indeterminism in relation to free will. It provides resistance to our choices, but a resistance that is necessary if we are to be capable of true self-formation.
9. Objections and Responses (IX): The First SFAs and Character Development
Turn now to a different set of objections often made against this view and other views of free will requiring ultimate responsibility or UR. According to UR, if a choice issues from, and can be explained by, an agent’s present will (character, motives and purposes), 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 (SFAs) for having the will he or she now has.
But it has often been argued that this suggests a potentially vicious regress. For in order to be ultimately responsible for these earlier SFAs by which we formed our present wills, wouldn’t we have to be responsible in turn for the characters, motives and purposes from which these earlier SFAs issued? And would this not require still earlier SFAs by which we formed these characters, motives and purposes?
We would thus be led backwards to the earliest choices of childhood when the wills from which we chose were not formed by us at all but were entirely the product of influences outside ourselves, parents, social conditioning, heredity, genetic dispositions, and so on? It may thus appear that all responsibility for later choices in life would go back to the earliest choices of childhood when we seem to have far less freedom and responsibility than we have later in life, which is absurd. (The most well-known modern version of this sort of regress argument against libertarian free will is made by Galen Strawson (1986, 1994), who calls it the “Basic Argument.”)[7]
Step 32: The first response to make to this familiar worry about a regress of responsibility is to note that the ultimate responsibility for choices in later life need not have its source entirely in choices of childhood. This would be true only if we made no subsequent SFAs in later life, which is certainly not what is being assumed here. To the contrary, on the above account, we make SFAs throughout out lives and more so as we mature and life becomes more complex.[8] And in doing so, we are constantly forming and reforming our existing characters, motives and purposes as we go along in ways that, while influenced by our prior characters, motives and purposes, are not determined by our prior characters and motives.
Thus ultimate responsibility does not have to completely backtrack to earlier and still earlier SFAs. Rather we add to, reform and change our characters, motives and purposes, as we go along with each SFA that we make.
Step 33: I believe therefore (in agreement with philosophers, such as Aristotle, who talk about the development of “character”) that responsibility for our wills (characters, motives and purposes) accumulates over time (see Kane 2008). It is by making many SFAs through a lifetime that we gradually form and reform our characters, motives and purposes in ways not determined by our past. With regard to most of the SFAs we make, as a consequence, our responsibility has a two-fold source: First, in the choices we make now between our conflicting motives and purposes and, second, in the conflicting motives and purposes themselves for which the choices are made, many of which had their source in earlier self-forming choices of our making by which we gradually formed our present wills over time.
The only exceptions, of course, are the very earliest SFAs of childhood when it is normally true, if we go back far enough, that the motivations among which we choose all come from sources outside ourselves, parents, society, genetic inheritance, etc. I have discussed these first SFAs of childhood in a number of writings (Kane 2008, 2009) and have a distinctive view about them.
Step 34: In the earliest SFAs of childhood, our responsibility, so far from being the source of all later responsibility, is very limited, precisely because there is as yet no backlog of self-formed character. That is why we hold children less responsible the younger they are. I further argue that the earliest SFAs of childhood have a probative (or probing or learning) character to them.[9] Young children are often testing what they can get away with (the limits) and what consequences their behavior will have on them and others (one of many reasons why child-rearing is so exhausting). Their character is thus slowly built up by how they respond to the responses to these earliest probes. Character and purposes to which they commit themselves accumulate and they become more responsible for subsequent acts that flow not just from present efforts but from past formed character and purposes as well.[10]
If a three-year old is told not to take more than his share of cookies, but tries to do so anyway the next time, then the child is responsible, but not as responsible as when he does it a second, third or fourth time and it becomes a pattern of behavior. The wise parent will not punish him severely the first time, but may do so mildly, by withholding something he wants. But the wise parent will also know that it is a mistake never to hold the child responsible at all for these earliest probes; for it is only by being so held responsible in however limited ways in our earliest years that we gradually become self-forming beings with wills of our own making.
10. Objections and Responses (X): Kinds of Control and Another Potential Regress
A further potentially vicious regress in this account of freedom of choice concerns the efforts or cognitive activities that precede and give rise to self-forming choices or SFAs.[11] If SFAs are brought about by the efforts of agents, then if the SFAs are to be ultimately free and responsible actions, it would appear that the efforts of the agents that bring them about must also be free and responsible actions. But if these efforts which give rise to self-forming choices are themselves to be free and responsible actions, do we not need to postulate further free SFAs to initiate each of the efforts, and so on backwards ad infinitum?
The answer to this question is no. We do not need to postulate further SFAs to initiate each of the efforts that give rise to SFAs. The plural efforts preceding SFAs might have been initiated by further SFAs in certain cases. Agents may sometimes be conflicted about whether even to begin to deliberate about a difficult choice they have an aversion to thinking about. But this need not always be the case and often will not be the case.
Step 35: The plural efforts preceding self-forming choices will normally be initiated by the confluence of the agent’s conflicted will plus the agent’s recognition of the situation he or she is in. Consider the businesswoman, torn between her moral motives to stop and help the assault victim and her ambitious motives for ignoring the situation and going on to her meeting. When she sees the assault and sees that no one else is currently present to help, that information is filtered through the present state of her will, including her conflicting desires. Deliberation thereby commences, causally initiated by the recognition of her situation and the awareness of a conflict in her will; and in such conditions, the competing efforts commence as well, for they are a part of the larger deliberative process.
Step 36: Agents must of course exercise some kind of control over each of the efforts or goal-directed cognitive activities that are thereby initiated in SFA situations. But the control agents must have over each of these efforts or activities is not plural voluntary control (PVC) of the kind they must have over the SFAs themselves that may result from the efforts. That would indeed lead to a vicious regress. Rather the control required over each effort is what I call, following neuroscientist Marius Usher (2006), teleological guidance control (TGC). Such control, Usher argues, is necessary for any voluntary activity; and he interprets it in terms of dynamic systems theory (cited in earlier steps 28 and 29). The behavior of complex dynamical systems (of which living things are prime examples), Usher argues, exhibits teleological guidance control when it tends through feedback loops and error correction to converge on a goal in the face of perturbations.
Step 37: It is important to emphasize (as Usher does) that such teleological guidance control is compatible with determinism (unlike the plural voluntary control or PVC agents exercise over SFAs). For it is a significant feature of my view that the kind of control (TGC) agents exercise over the efforts or volitional streams that may lead to SFAs—in contrast to the kind of control (PVC) they exercise over the SFAs themselves—is a compatibilist kind of control.
The control exercised over the efforts (TGC) is akin to what John Fischer and Mark Ravizza call “guidance control.” The efforts or volitional streams leading to SFAs exhibit such guidance control when they are “reasons-responsive” (in something very much like Fischer and Ravizza’s sense) to the specific reasons that motivate them—e.g., moral reasons for the effort to make a moral choice, ambitious reasons for a self-interested choice, etc.
Step 38: But it is equally important to note, as Usher also does, that such teleological guidance control, TGC, is also compatible with indeterminism. Dynamical systems can exhibit TGC, tending through feedback and error correction to converge on a goal, even when, due to the presence of indeterminism, it is uncertain whether the goal will be achieved. Indeed, this is the condition of the husband and the assassin. It is because they exercise such teleological guidance control over their efforts and succeed in attaining the goals they were aiming at, despite the probability of failure, that they can be responsible when they succeed. And so it is with the businesswoman when she succeeds in her effort to make, say, the moral choice, despite the possibility of failure.
Step 39: Note where we thus arrive: Multiple parallel goal-directed cognitive processes (efforts or volitional streams) simultaneously exercised by the agent, over each of which the agent has “one-way” or singular voluntary control (TGC), when occurring together, make possible “more -than-one-way” or plural voluntary control (PVC), since the agent might succeed in attaining the goal of either process at a given time and will do so voluntarily, on purpose and for reasons, whichever choice is made.
Stated differently, two cognitive processes over each of which an agent has what Fischer and Ravizza call guidance control, exercised simultaneously and in parallel, give rise to what they call regulative control, that is, the dual power at a time to bring about a choice by attempting to bring it about and the power at that same time to bring about an alternative choice by attempting to bring it about.
Step 40: This argument illustrates something I have often maintained, that incompatibilist freedom and control presuppose compatibilist freedom and control. We cannot get to incompatibilist freedom in one fell swoop in the real world. That is one leap too far. We must get there step-wise, by exercising compatibilist guidance control (TGC) over cognitive processes aimed at making choices, and from their, through parallel distributed processing, to incompatibilist regulative control (PVC) over the choices (SFAs) that result from these cognitive processes.[12]
11. Objections and Responses (XI): Luck, Explanation and Agency
I conclude with perhaps the most powerful version of the argument from luck, or the “luck objection,” against this and other incompatibilist or libertarian views of free will. This objection has been stated in the following way by Alfred Mele (1998), one of its most astute and persistent defenders:
“If different free choices could emerge from the same past of an agent [which seems to be required if they are undetermined] there would seem to be no explanation for why one choice was made rather than another in terms of the total prior character, motives and purposes of the agent. The difference in choice, i.e., the agent’s choosing one thing rather than another, would therefore be just a matter of luck.”
This objection is now so widely cited and affirmed by critics of libertarian views of free will that it is often referred to as “the” luck objection in the literature. And many philosophers assume it is decisive.
I think they are mistaken. But I also think this objection has the power it has because it teaches us something important about free will. The first question to ask about the objection is this: What is supposed to follow from the premise of this luck objection? That is, what is supposed to follow from the fact that (P) “if different free choices could emerge from the same past of an agent, there would seem to be no explanation for why one choice was made rather than another in terms of the total prior character, motives and purposes of the agent”?
Is it suppose to follow from this premise, for example, that (1) the agent did not cause or bring about the choice that was actually made? This does not follow on the view I have described. On that view, the agent causes or brings about the choice that is made by engaging in a goal-directed process of trying or attempting to bring about that choice (for good reasons, though not conclusive or decisive reasons) and by succeeding in attaining that goal, whichever choice is made.
Is it suppose to follow from this premise P that (2) the agent did not have control over the occurrence or non-occurrence of the choice that was made when it was made? This also does not follow on the view I have described. On that view,
Step 41: The agent has control over the SFA choice that is made in a very important sense of control. In this sense, for an agent to have control at a time t over the being on not being of some event (e.g. a choice) is for the agent to have ability or power at the time t to make that event be at t and the ability of power to make it not be at t. And in an SFA, one exercises just this kind of control over the choice one makes (e.g., the choice of A rather than B) at the time one makes it. For, one not only had the ability or power at that time to make that choice be, one also had the ability or power at that time to make it not be, by making the competing choice (of B rather than A) be.
One has both these powers because either of the efforts or volitional streams in which one was engaged might have succeeded in attaining its goal (choosing A or choosing B) at that time, despite the probability or chance of failure because of the interfering effects of the indeterminism; and if either effort or volitional stream, which is a goal-directed activity of the agent, did succeed in attaining its goal, one would have brought about the choice thereby made by endeavoring are attempting to bring it about. To have the power at a time to make some event be and the power to make it not be is what it commonly means to have control over the event (over whether it will be or not be) at the time.
Perhaps it is suppose to follow from the premise P of this luck objection that (3) the choice that was made would have been irrational in the sense that it was not made for good reasons that the agent wanted to act upon, or that (4) the agent did not make the choice (rather than some other) voluntarily (without being coerced against her will) or that (5) the choice was made by mistake or accident rather than being made on purpose?
None of these things follows as well for SFA choices, as we have seen from earlier discussions. The agents have good, satisficing, reasons for whichever choice is made and they choose it for those reasons. They are not coerced into making the choice, but make it voluntarily and could have made the alternative choice voluntarily. And they make the choice that was made intentionally or on purpose as a result of making an effort to make that particular choice for those reasons, and not by accident or mistake.
It follows from all this that the control agents have over the choice that is made is what I have been calling plural voluntary control (PVC). For they not only have the power to make the choice that is made be or not be at the time they choose , they have the power to make it be rationally (for reasons), voluntarily (in accordance with their will without being coerced) and intentionally (on purpose rather than by mistake) and the power to make it not be also rationally, voluntarily and intentionally.
Finally, is it suppose to follow from the premise P of this luck objection that (6) the agent was not responsible for the choice that was made? This does not follow either, if all the above five conditions, including the conditions of plural voluntary control are in place, as the above arguments show they can be in cases of SFAs, whichever way the agents choose.
Step 42: The result is that none of the conclusions (1)-(6) follows from the premise P of the luck objection on the view presented. It does not follow that the agent did not cause or bring about the choice that was made, nor that the agent did not have control over the choice that was made, nor that the agent did not make the choice rationally, voluntarily and intentionally, nor that the agent could not have made the alternative choice rationally, voluntarily and intentionally. So, if the conclusion of the argument—that the agent’s making one choice rather than the other “is just a matter of luck”—is meant to imply any of these things, it is the wrong conclusion to draw from the argument.
And, of course, if one were to say that “just a matter of luck” is meant to be consistent with all of these things, the argument would seem to lose all traction.
12. Conclusion: Liberum Arbitrium Voluntatis
Well, not quite all traction. And this is where things get interesting. With powerful arguments in philosophy, it is not enough to show their conclusions do not necessarily follow from their premises. One needs also to show why they seem to have such power and seem irrefutable.
The luck objection in this popular form does not show that libertarian free choices must be “just a matter of luck,” if that means any of the claims (1)-(6) are true. But it does show that there is something to the oft-repeated charge that such choices must be arbitrary in a certain sense. A residual arbitrariness seems to remain in all self-forming choices since the agents cannot in principle have sufficient or overriding (“conclusive” or “decisive”) prior reasons for making one option and one set of reasons prevail over the other.
Therein lies the truth in this luck objection (a free choice cannot be completely explained by the entire past, including past causes or reasons). And I think it is a truth that reveals something important about free will. I have argued elsewhere (Kane: 1996: 145-6) that such arbitrariness relative to prior reasons tells us that
Step 43: Every undetermined self-forming choice is the initiation of novel pathway into the future, whose justification lies in that future and is not fully explained by the past. In making such a choice we say, in effect, “I am opting for this pathway. It is not required by my past reasons, but 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.”
Step 44: Of special interest here, is that the term “arbitrary” comes from the Latin arbitrium, which means “judgment”—as in liberum arbitrium voluntatis, “free judgment of the will” (which is the medieval 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.
Step 45: In a similar way, agents who exercise free will are both author’s of and characters in their own stories 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.
If we should charge them with not having sufficient or conclusive prior reasons for choosing as they did, they might reply: “True enough. But I did have adequate reasons for choosing as I did, which I’m willing to endorse 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.”
[1] For an overview of research supporting such views about parallel distributed processing in vision see Bechtel ed. 2001.
[2] Chisholm 1971, 1976.
[3] Unlike distributed perceptual pathways or streams in the theory of vision, distributed volitional streams in deliberation would be such that only one of them could attain its goal because the goals are competing, not complementary, as in perception. That is the distinctive nature of deliberation.
[4] Some philosophers (e.g. Balaguer 2010, Ekstrom 2003), while agreeing with me about event-causal (EC) libertarianism, argue that my view has features, such as dual efforts, parallel processing, that are not necessary to defend EC libertarianism. I strongly disagree. Careful readers will note, for example, the importance of appeals to multiple volitional streams with competing goals or attractors to the arguments of this essay. I’ve often emphasized that, while is a mistake to overestimate the force of luck and chance objections to libertarian free will, as these EC libertarians rightly argue, it is equally a mistake to underestimate their force and to think objections from luck and chance are more easily answered than in fact they are.
[5] The assumption is therefore often made that to explain an incompatibilist free will, one has to postulate a special kind of causation by an agent or substance (often designated “agent-causation,” with a hyphen) that is a metaphysically primitive relation that cannot in principle be spelled out in terms of causation by states and events familiar to the sciences.
[6] There are a number of important recent works explaining how such complex systems may be used to account for human agency and action, including among others, McKay 1991, Juarrero 1999, Usher 2006 and Murphy and Brown 2007. Juarrero emphasizes the Aristotelian roots of these modern theories. Usher and Murphy/Brown (though they do not endorse my approach to free will), note its consistency with such theories. Murphy/Brown is an excellent overview of the resources and sources for such a dynamic systems approach to agency and action.
[7] The most well-known statement of this argument against my view and all libertarian views that emphaisize UR is that of Galen Strawson 1986, 1994 (the latter specifically directed against my view. Another persistent and insightful critic of libertarian accounts of moral responsibility is Carlos Moya (2006). Others who have made objections of similar kinds to my view include E. J. Coffman 2004, Robert Allen 1995, 2007 and John Lemos 2007.
[8] If this were not the case, or if, for example, mental capacities never developed beyond those of a child, as happens in some cases, then ultimate responsibility (and liability for punishment and blame) would be severely limited and minimal, as with young children of similar capacities.
[9] Kane 2007, 2008. Kane 2007 is a response to insightful critical papers by Robert Allen (2007) and Katherin Rogers (2007).
[10] Neil Levy (2008a) and Manuel Vargas (2005b) have raised important objections to my view and others which require that responsibility for later actions be dependent in this way on earlier actions. I have attempted to respond to Levy and implicitly to Vargas in Kane 2008.
[11] Objections of this kind are raised in slightly different ways by Mele, Ginet, Fischer, Clarke and Pereboom.
[12] Another objection involving agential power or control often made against event-causal views of free will like this one is that they give us no more power to determine what one does than compatibilists give us. (See e.g., Pereboom 2001 and Clarke 2003.) And (the argument continues) some further power to determine what one does than compatibilists give us is required for true libertarian freedom. I call this the no-more-power objection. It is, in my view, misguided for the following reason: If agents in SFA situations can exercise plural voluntary control over undetermined alternatives, as I have argued they can, then I submit that they do have more power than compatibilists can give us in a determined world. For the most compatibilists can say of agents in a determined world who act voluntarily, intentionally and rationally is that they may have acted otherwise voluntarily, intentionally and rationally, if the past and the laws of nature had been different then they actually are. Compatibilists cannot say that agents have the power to have acted otherwise in these ways, given the actual laws of nature and the past as it actually was at the moment of action. Not only is such a plural power more power than compatibilists can give us in a determined world, but it is just the kind of power that libertarians have always demanded for free will and moral responsibility. That is, it is a power than can be voluntarily (non-coercively), intentionally (purposefully) and rationally exercised in more than one way here and now, in the actual world as it is, not in some hypothetical world that might have been, but never was.