On Free Will IV: The Nature of Choice

What does it mean to make a choice? This question is at the heart of many of the debates over free will, and justifiably so. It may seem simple initially, but the more deeply one considers it, the knottier it becomes.

The basic dilemma seems to be this: Every event that occurs either had sufficient cause to occur or it did not. If it did, then it seems choice has no part to play: the event happened because of that cause, and could not have failed to happen. But if it did not, then choice seemingly is also excluded: the event “just happened”, not as the result of a choice (or of any other reason). Either way, choice seems to be ruled out. And note, also, that this conclusion seems to hold true regardless of whether one is a theist.

Does this mean that choice does not exist at all? It seems to be like an optical afterimage, something glimpsed only out of the corner of one’s eye that disappears when one tries to look at it directly. Is the very notion of choice a self-contradictory, impossible idea?

There is, of course, no middle ground between caused and not-caused. All events must be one or the other, and I do not intend to argue that there is some mysterious third category into which human actions fall. I intend to argue something very different: that some of our decisions are caused and are also free. This idea may strike some of my readers as strange or self-contradictory. If you are among them, bear with me; in the course of this essay, I will do my best to persuade you otherwise. But before delving into this, it may be worthwhile to limber up the imagination with a few thought experiments.

First, consider the case of a boulder balanced on a cliff edge. One day, it falls and crashes to the ground below. Did it choose to do so? Certainly not: a boulder has no capacity for volitional action. Its behavior is wholly described by the blind workings of cause and effect. If it did not fall, it is because the physical conditions for such an event did not obtain, and as soon as those conditions did obtain, it inevitably had to fall. Perhaps it required a strong gust of wind to give it the necessary push, or the patient working of erosion to alter its size or shape in just the right way, or the seasonal expansion and contraction of the ground that caused it to become unbalanced, or any of these in combination – but once the right causal factors lined up, it had to fall.

Second, consider the case of a virus that infects a cell and hijacks the cell’s reproductive machinery to make more copies of itself. Did it choose to do this? Again, certainly not: a virus is a mindless object, like a guided missile – a bit of genetic information wrapped in a protein coat. Its actions are wholly determined by the chemical and molecular interactions between it and its immediate environment. Its behavior seems slightly more intentional than the falling boulder, in that it has some capacity to control its own destiny and is not simply the passive tool of external causal forces, but nevertheless we feel comfortable in denying it any real ability to choose.

Finally, consider a single-celled organism such as a bacterium or an amoeba. These creatures are considerably more sophisticated than a virus, and better suited still to control their own fate and not merely passively react. But nevertheless, they are still blind, mindless reactors, their behavior bluntly determined by the chemical and physical characteristics of their environment. It is still inappropriate, I think we can agree, to speak of an amoeba “choosing” to do anything.

But now comes the troubling jump. Our brains, the source of all our actions and desires, are made up of interconnected networks of cells, each of which has complexity roughly equivalent to that of, say, an amoeba. Each of our neurons reacts to external stimuli in just as predictable a way as any single-celled organism, and none of them individually has the capability to choose. How, then, does choice arise from their agglomeration? How can we build a free-willed whole out of non-free-willed parts?

It should be obvious that our behavior is not determined by any particular neuron, but by all of them working in concert, giving rise to an enormously more sophisticated ability to sense, conceptualize and react to our surroundings. We are not like a virus or a bacteria, reacting in a mechanically predictable way to a given environmental change, and having only a limited, enumerable number of possible reactions to such changes. On the contrary, human beings are not mechanically predictable (as Part III argued); and we possess an almost infinitely vast array of possible actions open to us. And what is more, our behavior is not merely based on brute physical and chemical facts about our environment, but on information we have gained. We do not have brute dispositions to act, but rather inclinations, an entire subtle and complex web of tendencies and predispositions.

When we make the move from a virus or a bacterium to a human, it is not just a matter of degree, not just a move from lesser to greater complexity of reaction. On the contrary, there is something qualitatively different about this transition. In making it, something altogether new emerges, and that something is this: an ability not just to react blindly, but to anticipate – to build a mental model of the world in one’s head, and to use that model to guide one’s actions. Agents with this capability do not just perform an action that has consequences; they perform that action because it has those consequences. That is what a choice is: an act performed by an agent, in accordance with that agent’s inclinations, as a result of considering the likely outcome.

Although our ability to plan, to anticipate, gives us a degree of freedom that simpler life forms do not possess, the existence of those inclinations seems to be the troubling point. Although we humans are not as simply and mechanically determined as a virus or an amoeba, it is true that each of us comes into existence with a set of inclinations – character traits, if you will – that predispose us to act in certain ways. It is equally true that no one can choose their own inclinations; they are the result of causes that were operating before we were born. But, the question inevitably arises, does this not destroy choice? How can we be freely choosing when our decisions are influenced, if not determined, by factors beyond our control?

To answer this question, try considering it from another angle. If we express the wish to be truly free, free of unchosen inclinations that affect our behavior, the question is: what exactly are we wishing for?

It cannot possibly be the wish to be rid of any inclinations that affect our behavior. A being with no inclinations to act would not have free will; it would not have a will at all. Such a being would simply sit there, doing nothing. Even if a being came into existence with nothing but the bare inclination to choose other inclinations, it would have to do so randomly, and how is choosing one’s own character traits at random an improvement on having those traits chosen for one by someone else?

A personality cannot be built by pulling oneself up by one’s own bootstraps. To be able to take any effective action, an agent must come into existence with traits or inclinations that it did not choose. If that agent is sufficiently intelligent to perceive patterns in its own behavior, it can then recognize which of its own traits are contributing to poor outcomes, and turn this evidence into a new inclination toward changing those traits. Over time, such an agent can end up with a very different, and genuinely self-chosen, set of inclinations than the one they started out with. And even if an agent does not reflect on or change its own behavior, so long as it has the ability to do so, we can still rightly say that its current character is in a sense self-chosen, and that they are therefore responsible for it. Sometimes holding people responsible for their actions is what actually makes them responsible. (This point will be explored in Part V.)

Our characters are the result of a complex interplay between our inclinations, our reflections on those inclinations, and the actions arising from those two in combination. Albert Einstein’s theory of general relativity is sometimes summed up as, “Matter tells space how to curve; space tells matter how to move.” This conception of free will is similar, in a way. We choose what we choose because of who we are; we are who we are because of what we choose. Our choices are caused, it is true; and what causes them is our own characters.

A necessary consequence of all this is that, if you cannot reflect on your own behavior, you don’t have free will. Higher-order thinking – thinking about thinking – is a necessary precondition of this ability. Except for a few pathological cases, humans undoubtedly have this. Some of our relatives in the animal kingdom seem to have it also, albeit in lesser degree than us, and as such they might rightly be described as having a reduced degree of free will compared to what we possess.

We cannot magic our own characters out of the void. The truly “self-made” person, the one who bears ultimate and sole responsibility for every aspect of their own character and therefore bears ultimate and sole responsibility for every act they carry out, is an impossibility. But we can approximate this ideal sufficiently closely that we can, indeed, be said to make genuine choices and to bear the responsibility for those choices. In every way that matters, we are “free enough”.

Coming up: Is there genuine moral responsibility in a materialist world? Whenever we make a decision, could we have decided to do otherwise, or is the future that happens the only one that could have happened?

Other posts in this series:

About Adam Lee

Adam Lee is an atheist writer and speaker living in New York City. His new novel, Broken Ring, is available in paperback and e-book. Read his full bio, or follow him on Twitter.

  • BlackWizardMagus

    My own feelings as to this argument; it seems like you are saying something like “It is because it is”, or some such. I don’t think that’s wrong, of course, and I actually agree that we have choice that is determined by 10000 different factors, but it does LOOK like a rather weak, murky argument, I would think, to those who don’t agree with you. It sounds more like you’ve described your belief instead of proven it.

    Just some constructive criticism, since few seem to have much to say about it.

  • Quath

    Very interesting. I just have a couple of comments to make.

    On the contrary, human beings are not mechanically predictable (as Part III argued); and we possess an almost infinitely vast array of possible actions open to us.

    The bolder on a cliff was not mechanically predictable either. Neither is a pencil on its tip. And the pencil has an almost infinite number of orientations it can fall in. So we don’t seem different in that way.

    In making it, something altogether new emerges, and that something is this: an ability not just to react blindly, but to anticipate – to build a mental model of the world in one’s head, and to use that model to guide one’s actions.

    I am not sure if I agree with this. For example, I play several first person shooter games. The monsters I shoot at “see” the bullets coming at them and they try to dodge them. Do they have free will? After all, they used their “world” knowledge to anticipate and react.

    A necessary consequence of all this is that, if you cannot reflect on your own behavior, you don’t have free will. Higher-order thinking – thinking about thinking – is a necessary precondition of this ability.

    This is something I need to think about. It shows a higher process that we have and does seem to link to what we think about whan we talk about free will.

  • http://www.patheos.com/blogs/daylightatheism/ Ebonmuse

    A few thoughts:

    The bolder on a cliff was not mechanically predictable either. Neither is a pencil on its tip. And the pencil has an almost infinite number of orientations it can fall in. So we don’t seem different in that way.

    I acknowledge the point here; but I still think there’s a relevant difference. After all, even if the specific details vary, the pencil or the boulder’s next “action” is broadly determined in a mechanical sense by features of their immediate environment. Human beings are not like that. Inanimate objects generally have only one “path of action” open to them, in a sense, but we have a comparatively vast array of options, and our choice is determined not just by immediately present environmental causes but also by past information and by our own tendencies and dispositions.

    For example, I play several first person shooter games. The monsters I shoot at “see” the bullets coming at them and they try to dodge them. Do they have free will? After all, they used their “world” knowledge to anticipate and react.

    I don’t think this is the same thing. The NPCs in a video game are not genuinely conscious of anything – they react because they are programmed to do so in response to certain input. Their “actions” are entirely controlled by immediate environmental causes, more like the pencil or the boulder than a human being. To use the philosopher’s term, they are zombies. Only a being that possesses consciously experienced qualia – true sensory impressions of the world, however limited – can assemble them together into an internal world model that decisions can be based on.

  • Philip Thomas

    How can we tell what posesses “consciously experienced qualia”? I’ve always found the notion of zombies somewhat confusing as used by Chalmers and Dennet (who famously said “I am a zombie”.

  • Void

    So your version of free will is the presence of other factors besides immediate environmental ones that determine a person’s actions?

    On a tangent, does a compulsive liar or an addict possess free-will?

  • http://myspace.com/smarterchild06 eruditechild

    Qualitative differences between human and non-human behavior bear no relevance to the question of free will. The question is not of a physical nature. If we’re to ask if will is free, we must look solely at the functioning of the mind, despite it’s materialistic components.

    You’ve fairly demonstrated that humans have the ability to will, but in no way have shown that that will is not deterministic. And I don’t mean deterministic in a neurological sense. Part III, like viruses, bears no relevance to the question. I’m talking about deterministic in the realm of the mind, the abstract whole of a neurological network.

    We are not like a virus or a bacteria, reacting in a mechanically predictable way to a given environmental change, and having only a limited, enumerable number of possible reactions to such changes. On the contrary, human beings are not mechanically predictable (as Part III argued); and we possess an almost infinitely vast array of possible actions open to us.

    Mechanical behavior bears no relevance. An enumerable number of possible reactions is only as enumerable as the actions that precede them, in this case the actions that played out in the mind. But ‘enumerable possiblities’ are useless speculation when we consider the fact that there was one and only one reaction. Something happened in the mind that resulted in that specific reaction, an unmeasurable compilation of instinct, memories, and sensory input in all their subtleties, subtleties that may elude consciousness itself resulting in the illusion of free will.

    If that agent is sufficiently intelligent to perceive patterns in its own behavior, it can then recognize which of its own traits are contributing to poor outcomes, and turn this evidence into a new inclination toward changing those traits. Over time, such an agent can end up with a very different, and genuinely self-chosen, set of inclinations than the one they started out with. And even if an agent does not reflect on or change its own behavior, so long as it has the ability to do so, we can still rightly say that its current character is in a sense self-chosen, and that they are therefore responsible for it.

    Pattern recognition comes in primitive forms that are hardwired into our brain, while other forms are learned through experience. Both of these are hence, deterministic. What prompts an agent to reasess it’s “inclinations” and arrive at a self-chosen character is equally deterministic. If being mean to people results in a pattern of people being mean to me, I will naturally dislike being a societal reject, and modify my behavior accordingly. I can’t think of any instance where a person lacks a prompt to change, not matter how subtle the combination of instinct, suppressed memories, or sensory input.

    A necessary consequence of all this is that, if you cannot reflect on your own behavior, you don’t have free will. Higher-order thinking – thinking about thinking – is a necessary precondition of this ability.

    Are you insinuating that a baby doesn’t have free will because it cannot reflect on it’s own behavior, further implying that free will is learned? ‘Thinking about thinking’ simply boils down to thinking. We think what we think for a reason, no matter how subtle, as I’ve said, the combination of instinct, suppressed memories, or sensory input. Use your imagination, and you will quickly see this point.

    We cannot magic our own characters out of the void. The truly “self-made” person, the one who bears ultimate and sole responsibility for every aspect of their own character and therefore bears ultimate and sole responsibility for every act they carry out, is an impossibility. But we can approximate this ideal sufficiently closely that we can, indeed, be said to make genuine choices and to bear the responsibility for those choices. In every way that matters, we are “free enough”.

    At your fuzziest, I completely disagree with you. You’ve only demonstrated that humans have the ability to will, and with something as vague as “free enough,” have only succeeded in confirming my conviction that free will is an illusion.


CLOSE | X

HIDE | X