I. Some quick preliminaries
I believe there is free will. I am a compatibilist.
In this essay I will defend this view. I’ve taken this line ever since college, but have never taken the time to elaborate it since then. Especially with neuroscientists and other scientists asserting that free will doesn’t exist, and others buying their incompatibilism, I felt moved to respond.
So I decided to do a blog post, which is a lot easier than doing research and writing a journal article, which frankly I don’t have time to do these days. To the philosophers out there: I know I’m probably just reinventing the wheel.
Let me admit up front that while I was trained as a philosopher, this isn’t my area of expertise. While I have given this argument some thought over many years, this isn’t a researched academic paper. I don’t claim to have the final word. That would take a book or two.
2. Some observations about the concept of free will
Free will exists and is compatible with determinism. The way to defend this is by explaining what “free will” means—and, more importantly, by arguing against an assumption about “free will” that incompatibilists make.
The way to motivate an argument for a certain definition is to make some observations about free will, and then hunt about for a concept that answers to those observations.
(1) Free will gives us moral responsibility.
It is the freedom of our will that accounts for our being morally responsible for our actions. If we are guilty of doing something wrong, or praiseworthy for doing something right, then we must have acted freely; if we had no control and are not answerable for our actions, neither can we be praised or blamed.
(2) Free will gives us our human dignity.
Whatever else we might want to say about it, it is our freedom that gives us our dignity qua human beings. I don’t merely mean our political dignity as citizens, though that is surely related; I mean our dignity as unique, valuable individuals who deserve the regard of other individuals and institutions. This is a common assumption about free will and dignity. This is why, for example, B.F. Skinner titled one of his books Beyond Freedom and Dignity.
(3) Sometimes we act with free will, and sometimes we don’t.
If the concept of free will is to have some value, it should have instances where it applies and instances where it doesn’t apply.
(4) The law gives us a clue as to when we lack free will.
The law finds people lacking in free will, and hence (in keeping with proposition (1)) less culpable or sometimes completely innocent, when people are insane, under the influence of drugs or alcohol, hypnosis, brainwashing, and of course under duress.
Now let’s see if we can find a concept of free will that makes these four observations come out true.
3. Why does free will require a lack of causal determination?
The assumption incompatibilists make is that freedom of will requires a lack of causal determination. But why make that assumption?
A common way to formulate the problem motivates an argument. If we are free, philosophers often point out, then we have the sense that we “could have done otherwise.” But science increasingly shows us that we couldn’t have done otherwise, say determinists; there was just one causal path we could have taken. And if we couldn’t have done otherwise, then we weren’t free, even if it seemed that we were. Freedom is just an illusion.
The problem with this lies in the operative phrase, “could have done otherwise”: it isn’t obvious what this means. “Could” makes the phrase modal, one about possibility, and as such it is deeply freighted; in fact, it carries a lot of theoretical freight. Sometimes “possible” means logically possible, sometimes physically possible, and sometimes it simply means consistent-with-some-assumptions.
Now, when we act freely, and could have done otherwise, which sense of “possibility” do we actually mean? The incompatibilist says, “physical possibility.” But why think that? Why think that, when we assert that we could have done otherwise, we mean it was physically possible for us to do otherwise? That seems far too strong to me.
Similarly, I see no reason to think that free will requires a lack of causal determination. This is a common assumption, and it seems to be commonsensical; but on examination it becomes clear that it is not.
The problem is that it does not square with my four observations. Let’s go through them in turn.
(1) Does lack of causal determination give us moral responsibility? One might say so, because if we aren’t determined, then we are the original source of our actions, and nothing and nobody else can be credited or blamed for our actions.
But on reflection, this doesn’t make very much sense, at least not to me. If indeterminism were true, then our decisions would have no cause whatsoever (or at least an inadequately determining cause) and it is the lack of a cause that would make us responsible for them. But why should we be saddled with the responsibility of decisions that lack causes? On the one hand, sure, it makes sense that since the decisions come from us, they have to do with us; but it is their failure to have a cause that makes us responsible. Wouldn’t it make more sense, in that case, to say that our decisions can’t be blamed on us precisely because they lack any causes? Why would the causelessness of our decisions make us responsible for them?
As far as explanations of free will go, we can do better (as I’ll argue further down).
(2) Does lack of causal determination give us our dignity? This makes even less sense. Death penalty critics find a basic level of human dignity even in murderers. What does our will’s lacking a cause have to do with that? Imagine a death row inmate’s defense making the following argument: “Your honor,” the defense says, “we cannot put John Doe to death. After all, like all of us, he has dignity. His decisions are uncaused.” Surely that’s a non sequitur.
(3) Is it possible that we sometimes have free will, and sometimes we don’t, on the view that free will is uncaused? It seems it would be more of an all-or-nothing affair. If determinism is generally correct, then the “free will is uncaused” view would entail that we never have free will. I think that philosophy’s main subject is concepts that are found and applied in natural language. If, by some proposed definition of a word, the word does not apply where we ordinarily think it does, that is excellent reason to think that that definition is incorrect. This is all the more true if a definition is available that does justice to the assumptions we ordinarily make about the concept.
(4) The law makes assumptions about free will. As I said above, legal concepts of freedom, which I suppose are grounded in the ordinary notion of free will, are used to distinguish between culpable and non-culpable (or less-culpable) states of mind. Does the “free will is uncaused” view do justice to these legal concepts?
Surely not. After all, when a court asserts that a defendant lacks free will, and is found not guilty by reason of insanity, are they saying that the defendant’s will was causally determined, whereas normally it is not? Of course not. Abstruse questions of causal determinism don’t even enter legal deliberations. Questions about whether the defendant was functioning normally, unimpaired, not under duress, etc., however, do.
As with moral responsibility and dignity, lack of causal determination doesn’t appear to explain such legal concepts—which are eminently useful and consequential—at all.
4. A commonsense notion of free will
We are discussing free will; we may take a clue from the word free.
Sometimes the will is free, and sometimes it isn’t. So when it is free, of what is it free? Free will lacks something; it lacks constraints of some sort. What sort?
We merely have to look at the four observations about free will to find an answer suggested. Let’s take observation (4) first. Notice what circumstances lead us to say that a defendant in a legal case did not act freely: such things as duress, severe alcohol- or drug-induced impairment, hypnosis, brainwashing, sleep (as in action taken during sleepwalking), senility, insanity, and let’s not forget childhood. If those circumstances prevent, remove, or impair our freedom of will, then what can we say, in general, about what freedom of will is?
I propose the following account:
The will is free, if it is an operation of a normally-functioning brain that is not impaired by anything that would prevent normal, rational, adult deliberation on our actions, whether or not we actually perform such deliberation.
If I act without thought, I might still act freely, because I could have stopped, taken stock, and restrained myself, even if I didn’t; I “could have” in the sense that nothing was stopping me.
Next let’s see how this account makes sense of my four observations about freedom.
As to (4), I listed circumstances that I said do, or probably should, prevent legal culpability. Each circumstance is rather handily explained by this account. Duress, of course, can cause us to make decisions we would not make but for the duress. Drug- and alcohol-induced impairment, as well as hypnosis, brainwashing, sleep, senility, and insanity, do in varying degrees impair our ability to deliberate about our actions normally and rationally. Under such influences we might act out of emotion, delusion, or incomprehension where we would not so act if we were functioning normally. As for childhood, we say children, especially small children, are not fully free agents simply because their ability to deliberate maturely, and regulate their decisions based on such deliberations, is not fully developed.
In this way, as you can see, my account of free will satisfies (3): there are instances of free will and instances of non-free will. Adults are normally free agents with free will, because we are functioning normally and not under duress. But when for example a person completely breaks with reality and starts fighting with things that aren’t there, he is no longer properly considered a free agent with free will, precisely because he’s no longer functioning normally.
Perhaps the most interesting argument for my account of free will is how it satisfies (1), i.e., how it explains why free will makes us morally responsible. This has nothing whatsoever to do with a will that lacks causal determination—again, why would that matter? But what clearly does matter is that our ability to deliberate, and thus our ability as adults to take full moral stock of our actions and compare them both to our personal principles and our legal and social constraints, be functioning normally. We are normally regarded as having free will, and are normally accorded respect qua fully adult, free agents, because we are normally unimpaired and not under any duress. We aren’t laboring under severe delusions, drug-induced hallucinations, insane rage, fear of harm from criminal influences, and so forth.
It is precisely our ability to deliberate that gives us our culpability. Regardless of the causal influences and streams, we (in my view, quite rightly) expect each other to own our free choices, regardless of the amount or type of thought we put into them. That’s because we could have given them more thought, we could have considered how they stack up against our duties, and we could have chosen the difficult but right way over the easy but wrong way. And if a person’s scruples were not sufficient to restrain him from the wrong thing—or, by the same token, if a person’s principles are so finely developed that they impel him to do the difficult but outstanding thing—then we blame or praise the person precisely because of the influence of his own deliberations on his course of action.
Similarly, my account of free will explains (2), or why free will is sometimes thought to give us our dignity qua human beings. It is, in short, our capacity for reason and rational action, our ability to guide ourselves as rational agents, that gives us our dignity. This also explains and comports very well both with the tradition of human rights and, for better or worse, with the view that animals lack the sort of dignity that humans have.
I’ll explain the latter points a bit further in the next section.
5. The functions of the concept of free will
My arguments so far go some distance to showing that, as it is normally used in ordinary language, “free will” is more of a moral concept than a metaphysical one. To assert that we have free will is to hold ourselves responsible for what we do, and in particular to highlight the importance of our rational, deliberative capacity in guiding our action. Those who lack such a capacity are not free. That’s why we do not treat drugged-out wastoids, the severely mentally ill, sleepwalkers, animals, small children, zombies (if they existed), and so forth, as free, rational agents as they move about in the world. They lack free will. That’s why they require institutionalization or special care of other kinds. (Or in the case of zombies, shotguns.) In this way, the notion of free will has real-world use.
This use is important and, indeed, indispensible—decidedly not something we will ever be able to discard, some determinists notwithstanding. This is not just because of how we treat those who lack free will, but much more because of how we treat those who have free will, and so are free agents. Let me explain.
To be treated as a free agent is to be treated as an autonomous individual, capable of making and being held responsible for important decisions. To operationalize this in the realm of the law: an autonomous, free individual answers for himself before the courts and in legal agreements; neither another person nor society as a whole (in the form of the government) may take on these responsibilities for him, unless he is found legally incompetent.
In the moral realm, to be treated as a free agent is, simply, to be treated as morally responsible. How we apportion praise and blame depends deeply on questions of a person’s maturity, basic mental competence, and further on how well the persons exercises his judgment. If someone gives $1,000 to charity for “pure” motives, we find it much more praiseworthy than if the person was forced or incentivized to give. And if a person kills in self-defense, the decision to kill is one made under duress and not nearly as heinous as one taken without duress (or insanity, etc.). In this way, how a person exercises his free will is crucial to how we evaluate the merit of his actions.
It also means something to be treated as a free agent in the political realm. In short, it means to be treated with the sort of respect that adults expect, and that children crave more and more as they get older. Adults give a certain sort of credit to the decisions of other adults; whether wise or foolish, our decisions are our own. The notion of tolerance owes something to the notion of freedom: it is not contradictory to tolerate things we find morally objectionable, because when we exercise tolerance, we respect not the action (or the speech) but the free decisions of persons out of respect for the varieties of human reason and experience. Rights in general are often regarded as being based on human dignity, which in turn is, I think, a function of the normal human capacity to deliberate and take action on our deliberations.
If we jettison the concept of free agency, we are in effect treating rational, responsible adults as if they were zombies, or children, or the enemy “other.”
Without the notion of free agency, we have much less reason to treat individuals as legally autonomous and uniquely competent to speak for themselves. In such a case, the law can disregard the wishes of individuals and replace them with somebody else’s notion of what serves the individual’s (or the public) good. Without the notion of free agency, we are neither to be praised nor blamed for our actions, or if we are, our good actions are praised not as our own achievements but as the community’s, and our bad actions are criticized not as our faults but the community’s. And without a robust notion of free will and free agency to inform concepts of political freedom, we are treated as children, and governments tend to act paternalistically, as “Big Brother.” Treating their citizens like children, governments do not credit citizens with the free agency, and therefore the basic levels of tolerance and rights, that citizens normally feel they deserve.
In general, it is the freedom of will that forms the basis for political freedom and human rights.
Those who claim we don’t have free will seem to be completely unaware of these deeply important functions that free agency play in our legal, moral, and political life. If one accords freedom of contract little respect, thinks culpability is always collective and morality generally outdated, and has contempt for tolerance, rights, and dignity of one’s political enemies, then one will also probably be comfortable chucking the notion of free will.
6. What about determinism?
The account advocated here may be especially difficult to swallow if you have drawn a strong equation between “free will” and “uncaused will.” Some people who discuss free will and determinism ad nauseam appear simply to assume there is such an equation; and so incompatibilism seems obvious to them.
Nevertheless, I think the arguments above make it clear that “free will” doesn’t mean “uncaused will.”
There remains a difficult problem. Many people do, for whatever reason, have the overwhelming sense that, if our decisions are completely determined by forces ultimately outside of us, then those decisions simply are not free. And perhaps in that case we don’t have moral responsibility for our actions. It might not be a matter of definition, but it’s a strong intuition nevertheless. So, some might find what I’ve said about the meaning of “free will” plausible, but still find themselves unable to agree that we have free will for the simple reason that it seems incompatible with determinism. Call this the “incompatibilist intuition.”
So there’s one part of my case in defense of free will, namely, I need to explain where the incompatibilist intuition comes from—and then explain, further, why this intuition is wrong.
Presumably, the intuition originates approximately like this: if I lack control over all the inputs to my decision, then I lack ultimate control over my decision as well. It is one thing to be able to deliberate about whether I’ll have soup or salad, and perhaps that gives us the illusion of control; but unless we can go back in time and control our own parents and teachers, we don’t really have control over whatever habits and principles we bring to bear in our deliberations.
This, I want to argue, is irrelevant to free will.
When we are deciding whether someone acted freely and is responsible for, say, committing an assault, we do not really care whether their parents taught them to be mean. Maybe they did; too bad, if so. If we accord them respect as free agents then we credit them with the ability to change.
But more to the point, we credit adults with their principles and habits, and the decisions that flow from them, as all coming from themselves. The point can be made more forcefully and relevantly: such principles and habits are part of who we are. Insofar as the person’s moral identity or selfhood partly consists of such things as internalized principles and habits—insofar as things like selfless honesty or self-serving dishonesty are part of who a person is—then in acting out of such principles and habits we are expressing ourselves. This is why our actions are free. When we credit a decision or action as free, and a person as a free agent, we are honoring the choice as flowing from “who the person is,” from the person’s identity or selfhood.
It is true that “the people we are” is made by our upbringing, environment, and genes, and that our decisions are influenced by various ephemera. And all of that might indeed mean that we are determined by those inputs. Nevertheless, whatever I—bundles of principles and habits that I am—happen to decide is credited as my free choice precisely because it comes from me. So the fact, supposing it is a fact, that I am causally determined to be as I am is perfectly irrelevant to my freedom and agency.
Consider an example of something that removes or attenuates our responsibility. Suppose someone holds a gun up to my head and tells me to give him my wallet. I am not acting freely when I hand over the wallet, because a relevant input to my decision comes from elsewhere. Except for the mugger’s threat, I wouldn’t hand it over. Another example. After ingesting hallucinogenic mushrooms, someone decides to climb a tree, because he hallucinates a bag of money up in the tree. There are elements of free action in the drug-induced state, but the tree-climber is not fully free because we do not credit the hallucination as coming from the drug user but from the mushrooms.
To say that a decision or action is free is not to say it is uncaused, but that it comes from habits, principles, etc., that we ascribe to the person himself and not to outside influences. To ascribe a principle to a person is not to say the person adopted the principle for no reason whatsoever; it is simply to say that he has in fact adopted it—that he owns it, so to speak.
I’m not sure many incompatibilists will be able to take this line very seriously, because the “uncaused will” view of free will is very much ingrained in our discourse about free will and determinism. Old habits die hard. But I think I’ve made an excellent case that free will means something along the lines of what I’ve explained.
In that case, we might resign ourselves to talking about two concepts of free will. Then I would agree that we have no free will, on the metaphysical, incompatibilist concept. But I would hasten to add that that concept simply does not matter. The ethical, compatibilist concept matters because we rely on it in our legal, moral, and political discourse.
We do have free will—in the sense that actually matters.