In this essay, I argue that the question of whether humans have free will need not be viewed as important, as it is largely irrelevant to whether one is living a good life, and is even mostly irrelevant to questions surrounding punishment and desert except insomuch as we cast it as—believe it to be, treat it as being, etc.—relevant.
Free will is understandably of concern for those who believe that an all-loving deity would only punish those who deserve it. But for the rest of us, there is no principled reason to view it as relevant, and so it should accordingly be struck from, for example, our core conception of justice, particularly as a vehicle for retribution. Indeed, free will seems to be one of many current questions that turn out to be potentially outmoded holdovers from a theological framework that has been gradually disassembled since the start of the scientific revolution. Where once stood a serious concern about deserving punishment for offending a deity, now stand questions about what counts as being responsible for one’s actions and thus deserving an F in calculus, low socio-economic status, or the death penalty; or, less significantly, about whether one really freely chose to push a button in a psychology experiment.
Depending on one’s definition of free will, what I describe here may imply that humans don’t have it. If so, it’s a trivial kind of lack, due to free will simply not existing any more than a square circle does. This is different from merely saying that humans don’t have free will, because that statement alone doesn’t imply anything about other sorts of beings (deities, alien lifeforms, post-humans, conscious computers…). In other words: If free will turns out to be such a confused or vague concept that the term doesn’t actually denote any phenomenon at all, then no entity of any sort has it or, really, can be conceived of as having it.
On other definitions of free will, what I describe here will seem to imply that people do have free will. That’s fine as well.
My point is that it doesn’t—or at least needn’t—matter either way whether or not one has free will according to this or that definition. What does matter is that we have an understanding of the phenomena that relate to our conceptions of free will—such as justice, praise, blame, and punishment—and how those phenomena relate to our lived lives as beings capable of suffering or thriving (I put more emphasis on the former).
In a longer essay, I might explore my hunch that free will is a general term we put on conditions for moral blameworthiness (as a basis for retribution and, in earlier times perhaps, as an essential feature of religious power) after those conditions have been established—subsequent to which the term has taken on a metaphysical life of its own. What I’m suggesting is that we can just worry about the conditions for blameworthiness, living a good life (however defined), and such without creating the insurmountable difficulties that come along with dragging free will into it.
I’m not dismissing free will lightly. There’s a good reason why smart people disagree on many of the ideas that surround the topic. Also, the default position in society (e.g., in our legal system) is that we do have free will, though that position lacks the nuance that characterizes the finer points of academic debates on the topic. Philosophers, legal theorists, and neuroscientists can have their debates. Society would be better off viewing it as an academic question that may have no answer, and certainly should not build institutions of praise and punishment (especially the latter) on some presumed answer, particularly an unnuanced one.
Finally, free will as a concept or phenomenon per se has, as far as I am aware, has special salience in the West and closely related intellectual lineages. I have done some research into how free will and similar ideas have been viewed in, say, ancient Buddhist thought. I don’t bring this into the discussion here. I don’t view this as primarily a discussion about academic questions, cultural studies, or history, but rather as a serious plea to correct what strike me as dangerous attitudes that persist prominently in the world immediately around me.
I’ve broken the argument into five parts:1
1. You have desires. You didn’t choose them, but, for whatever reasons—genetics, culture, developed tastes, conditioning, neural implants controlled by an alien spy, a computer scientist who built and coded you to believe you’re human—there you are with desires.
2. Some desires are stronger than others. Call these first-order desires.
3. First-order desires are often chased by one or more (let’s call them) second-order desires about whether or not you really desire to see your first-order desire(s) met.2
For example: You desire the jolt of an espresso at 10pm, while you also, though with less relish, desire to be asleep by midnight (for the sake of your Tomorrow Self, who will be mighty grateful to you for the gesture). So you talk yourself down from the desire for espresso.
4. Freedom, as I conceive of it (as distinct from free will) is, in a nutshell, what you have when you feel in control of pursuing (or not) your desires. Feeling free does not require that you feel you have chosen your desires: when you achieve your desire to sleep by skipping the espresso, you don’t consider this a lack of freedom or control merely because you did not initially choose to have the desire get a full night’s sleep (even though you may recognize the restrictions put on your freedom by needing sleep in the first place, and by needing a certain amount of sleep in order to perform well, etc.).
These feelings will be present whatever definition you endorse of free will.
(In this context, it’s often brought up that participants in psychology studies have been shown to act less morally after reading a convincing essay arguing that science has disproven free will’s existence. More on this below.)
4. What’s relevant to your freedom is whether you feel in control of navigating your desires. If you choose to go to bed at 11:30pm without espresso but still cannot sleep due to, say, an upstairs neighbor playing loud music, this isn’t a restriction on your freedom of control in the navigation of your desires. They are not metaphysical impediments to the exercise of your free will. They are, rather, externally imposed restrictions on the range of activities you are able to perform in the world, of the choices available to you as you attempt to exercise your freedom. That attempt is freely made, though it may be obstructed.
5. Even if control is illusory, and even if everything you do has nothing at all to do with any choices you—as if you were some sort of supreme author—have really made, you have a robust feeling of personal control that is either being exercised or obstructed.
6. It’s enough to feel more or less satisfied with life that, on balance, the control you have over yourself feels reasonably unobstructed: You have a desire, you survey the options, you (attempt to) satisfy the desire or not.
7. There are unlikely scenarios where some evil puppeteer is controlling your every thought and movement, yet you still feel completely in control. I view this as a restriction on bodily freedom, rather than on free will, but, either way, learning that you’re in that scenario could be disconcerting (though the puppeteer would control just how disconcerting: it’s every thought and movement).
A more likely scenario is that all conscious agents responsible for making choices do so through a combination of deliberation and intuition, informed and influenced by a complex array of external factors and pressures, in a process that satisfies some, but not all, definitions of free will. My point is that whichever definition you endorse, it all feels the same.
7. Given all this, the question of free will is irrelevant to you in most aspects of your life.
Believing you have or don’t have it might make some small difference in how you feel, but this strikes me as unlikely to be of concern outside of those moments in which you’re contemplating the question. As far as I know, there are no instances of free-will–oriented angst to be found on the order of existential angst (one part of which is the looming and oppressive knowledge that you’re going to die; it is founded, however, on the dread that comes with feeling too much freedom: “Man is condemned to be free; because once thrown into the world, he is responsible for everything he does,” as Jean-Paul Sartre put it in Being and Nothingness).
8. Someone might respond that not being able to choose one’s desires does in fact amount to a conspicuous restriction on freedom, and thus does cause a kind of free will angst once properly understood. This is tricky, but if it’s true, I think we’re stuck with it, at least philosophically. Practically speaking, I suspect the problem could be greatly ameliorated if we could somehow rewire ourselves to have as first-order desires the things that we both believe are good for us and are good for us (e.g., broccoli rather than cocaine). It seems this is primarily an issue for people who have desires they’d rather not have—which is still about external impositions (e.g., addiction to chemicals; an undesired sexual predilection). Let’s briefly explore this from philosophical and practical perspectives.
Consider someone with an all-consuming desire he wishes he didn’t desire. Such as for drugs, harmful sexual practices (e.g., pedophilia), poor diet, or, I don’t know, writing convoluted philosophy essays. Unfortunately, we cannot choose what we desire. To say, “I desire to desire something other,” is itself the expression of a desire. So, we cannot conceive of any sort of free will that doesn’t bottom out at desire. Even with the most robust free will imaginable, you cannot choose what you desire, because that choice itself is an expression of desire; dig deep enough, and you’ll run into a desire you didn’t choose.
If you chose to, say, roll dice to determine what you’ll choose to desire (assuming you could, for example, rewire your brain to make the desire sincere), then this is the expression of the desire to not choose what you desire, and you still were not in direct control of choosing what you now desire, nor is there an unlimited range of possible things a human may desire. And if you do rewire your brain, you do so as an expression of your desire to go along with whatever the dice chose for you. Similar issues arise should you allot the task of rewiring your desires to someone else (imagine a future with Your Best Desires specialists); indeed, that seems the furthest thing from free will.
It would be nice to be able to turn off (chemically or surgically or however) a desire for addictive substances, but one would need a desire to turn it off in in the first place—just as one now needs to make that choice to attend an AA meeting in the first place, or to enter rehab (think of Odysseus having himself tied to the ship’s mast). Thus, again, free will either bottoms out at desire, or you must massage your definition of free will in order to accommodate our lack of control over desire.
Whatever the case, sculpting into existence one definition or another of free will makes little difference on the freedom one actually experiences; so in the end I claim that your quality of life is affected not so much by metaphysical free will (or its absence) but rather by material impositions on your body.
9. You might notice a wavering on my part between talking on the one hand as though we could have free will and it is a task for us to figure out if indeed we do have that thing; and on the other hand talking as though it’s simply a term we use that doesn’t really refer to anything (e.g., as with the term square circle). I mostly believe the latter, but also recognize that if we could pick out some intelligible phenomenon related to personal responsibility, choice, agency, etc.3 and agree to call that thing free will, then we would have free will.
It’s also true that it is difficult for me to shake the feeling that free will, as an independently existing entity, can be imagined, and that there would be something that it’s like to have it. The way I seem to resolve the conflict between this and the square-circle view is to presume that, if we did have free will, it would feel exactly like not having free will. But then I remember that I absolutely cannot conceive of what it would be like to be the supreme (and in essence external and desire-less) author all of my desires, from the bottom up; and then the square-circle view snaps back into focus.
10. Reporting a total lack of desire is a classic symptom of depression, and would be worth exploring in a longer discussion. For example, I wonder whether some people suffering depression would say that they desire not to be depressed, and, if so, how that sort of desire is distinct from the sort of desire they claim to lack.4 I think that to really understand what we mean when discussing free will, we will need to account for the hardest cases, such as depression, addiction, and what it means for someone who has an active mind but little to no bodily control.
Feeling yourself to be at the mercy of addiction strikes me as providing particularly interesting and important dilemmas for any account of free will. These too are best left to a separate discussion. I will concede here, however, that if we could agree on a revised, technical definition of free will, stripped of any metaphysical concerns, it could be a useful concept for discussing phenomena such as addiction and other potential sources of harm towards oneself and others. As the concept stands now, however, I think it more harmful than helpful.5
11. Continuing some thoughts from point 9 in this section: We must take care not to revise the concept of free will arbitrarily. If free will exists as an independent phenomenon, then our goal in describing free will is to describe that phenomenon rather than some other phenomenon, and certainly not so simply come up with some interesting sounding or easy and convenient idea. If free will doesn’t exist, taking care in our attempts to define it should help arrive at that conclusion. Either there is something in the world that we pick out when discussing free will, or there isn’t. I think there isn’t, which is why the concept is so malleable—is why two smart people can have the exact same facts, yet disagree on whether they have free will, or, more precisely perhaps, on whether this or that definition of free will is the correct one.
But if there is some phenomenon out there that could rightly be called free will, and that thing has been more or less the subject of our debates about what we, over the years, have referred to as free will, and we drastically alter our attempts to describe that thing for the sake of, for example, including (non-conscious) computers among those things that have free will (see the Conclusion, below), then we are no longer describing that phenomenon. We would be using the term free will to refer to some other phenomenon that we’ve decided is more interesting, relevant, or useful—as with the case of computers (especially non-conscious ones). This phenomenon may still sufficiently occupy the same discursive space carved out by the idea of being responsible for one’s actions and such, but what would be lacking in that particular case is moral responsibility (a computer, unless it possesses consciousness at least as sophisticated as that of a typical teenage human, can be no more morally to blame for its activities than could be a rock thrown by the wind or a bully; or a tiger who rips into an innocent human child).
Given that I think free will is an incoherent concept, I’m all for either eliminating the concept or revising it for something more useful, but with full awareness that there’s not still some floaty, mysterious force out there that we are now failing to call free will and that continues to make us deserving of retributive recourse and such.
12. A revision might begin with something like: Free will refers to the ability of an intelligent and conscious agent to participate in its own causal influences within a restricted set of physically, metaphysically, and logically possible options; especially when that agent is aware of the limits and general nature of its free will. In simpler terms: When we have free will, we are responsive to reasons and we can contribute to those reasons. (For an example of consciousness affording such participation, see the section on LaPlace in my piece “Free Will Paradox?”)
On this conception, you certainly do not get to choose your desires from the bottom up, though you can attempt to influence your higher-order desires (e.g., through self-conditioning, self-regulating such as when you only buy the small container of ice cream, applying principles of cognitive dissonance by acting as though you desire something, joining AA, seeking hypnosis therapy, etc.; some will be more successful than others, but what’s most important is that you can try).
Note that I do not intend this revision to be an elaboration on what I describe above as freedom. One can feel a lack of freedom (i.e., be bound up with ropes), yet still have free will as I define it here. Also, notice how this suggested revision differs from the conception of free will as the ability to have done (or to have chosen to do) otherwise.
13. Higher-order desires can be good or bad for us, and the same goes for second-order (and third-order, and so on; there may also be competing desires of equal intensity). Indeed, my strongest inclination (my first-order desire) might be to do something that’s best for me, while a small voice (a second-order desire) tempts me towards harm.
1. Free will may be the central reason given (or implied) for why a person should be and deserves to be punished.
2. Though punishment and praise are both important, I tend to be more concerned with the former given that the consequences of punishment tend to be more concrete, dire, and lasting. (I’ll touch on this more in another writing I’m working on in which I respond to James Rachels’ endorsement of punishment on grounds of desert.) At the very least, punishment is meant to make a lasting impression, while praise is meant to be an isolated reward; you get a taste of punishment and your meant to avoid its invitation, but quite the opposite with praise (think of our general What have you done for me lately? ethos).
3. To say someone deserves to be punished is to say that someone is responsible for having committed a morally wrong act, and that the appropriate or natural (or even holy) response to that act is to punish. A key concept here is that of responsibility, which implies free will as that idea is generally understood. (See section III below for more on responsibility.)
4. I claim, however, that punishment need not involve desert or blame. If someone is a persistent danger to others or themselves, that person’s activities must be restricted. When such behavior seems to be coming from a rational expression of that person’s own desires (i.e., when the person seems responsible for that behavior), we tend to think the person should be—or at least deserves to be—punished, or, more precisely, made to suffer; I claim, however, that the person should only be punished insomuch as is a consequence of the minimum restriction that must be put on the person’s activities.
5. The notion of punishment tends to involve retribution. Indeed, to deserve means to make willingly make oneself into an object of retribution. Retribution involves intentionally causing harm to—creating suffering in—someone in exchange for harm that person has caused to someone else.
6. Again, however, we may conceive of punishment as a restriction on behavior. There are of course questionable (or even nefarious) views one may take on that, such as inducing personality-altering chemicals or surgery. I suggest that we should take the most humane approach, which is to say we should not induce more suffering, or blotting out of the subject’s personhood, than is necessary. On this note, however, there are interesting and complicated questions to be asked. For example, suppose one’s personality can be altered through some sort of medical intervention. So altered, the person would be highly unlikely to perform those harmful acts again. Would it be ethical to give the person the choice to undergo the intervention rather than be imprisoned? I think so. I also think, though, that society is far from being willing to accept such an approach—it would probably be seen as letting the person get away with their crime. At any rate, well into the near future I think some sort of imprisonment will be necessary, so this is what we have to work with.
Some have suggested recasting the idea of imprisonment as that of quarantining someone with a communicable disease (notwithstanding, I presume, our cultural tendency to blame people for their diseases)—see, for example, Ferdinand Schoeman’s 1979 article “On Incapacitating the Dangerous,”6; more recently (4/2016), Gregg Caruso discussed the idea on the Philosophy Bites podcast.
I’m sympathetic to that suggestion. What I’d like to emphasize here is that, whatever alternative conception of punishment we land on, that conception need not involve desert or causing suffering (as an end in itself) to the punished.
7. Another conception of punishment and reward (e.g., praise) might involve viewing those things as opposite sides of the same conditioning device (operant or otherwise).
It’s often (controversially) noted that steep punishment of a given individual deters crime in society. Whether that’s true is in part an empirical matter (e.g., How do crime data correlate with punishment data?) and in part a moral and/or jurisprudential matter (e.g., What counts as a crime? As harm? Is it acceptable to harm an individual in order to save a group from harm?). There are also psychological and sociological dimensions to all this (e.g., Why do people need to be deterred from harming others in the first place?).
These difficult questions aside, I’d wager that for most of us, being significantly restricted in our activities would count as unwanted punishment. This seems to me a more humane starting point than the idea that criminals deserve to be made to suffer due to deserving suffering for its own sake and, furthermore, for the benefit of others.
(If there exist those for whom being locked away in a humane correctional facility would be a life improvement or would provide more rather than less freedoms, I blame this on other failures in our society than sufficiently making prisoners suffer for the sake of deterrence.)
1. A typical conception of responsibility is grounded in our general conception of free will: If you chose to perform an act, you are responsible for that choice; and if you then perform the act, you are responsible for the act’s having been committed (barring unusual circumstances). That seems fairly straightforward and obvious. There are other ways of conceiving of responsibility, however.
2. An alternative view: Responsibility means that the behavior is issuing from that person (however personhood is conceived), as a result of that person’s desires, dispositions, system of beliefs (itself influenced by many factors and pressures of which he person is unaware), and so on. This is similar to when we refer to an earthquake as being responsible for the collapse of a building. This is still roughly close to blame, but more in the sense of being the explanation of the source of some activity or event.
3. Another alternative view involves responsibility as a kind of duty rather than as a source of blame. On that view, I might say that I take responsibility for everything in my life. This does not mean I accept the blame for everything in my life (that would be going dangerously too far!). But I do accept that the I am the person who is in the best position to make improvements on any unsatisfactory areas in my life. (I understand that not everyone is in an equally advantageous position in this regard; most of us, for better or worse, are precisely in this position, however, and those of us—particularly if we are adults—who are not in that position are more likely to be worse rather than better off.)
Consider the following example: Suppose you live in a house with several roommates. Each of you has agreed to keep one common area in the house clean. One day, you walk into your assigned area to find that the cat has knocked over a glass of milk left behind by a recently departed guest. Your housemates are also present. Nobody in the area blames you for the broken glass and spilled milk, but all—yourself included—understand that it is your responsibility to clean it up. If you don’t clean it up, you’ll be held accountable—will be blamed—for the glass shards and souring milk, but that is a separate discussion, and is where the notion of duty comes in.
I won’t push a particular view of responsibility, though I do these alternatives to the standard free will take. My point is mainly that there are alternative views that need not imply desert of retributive punishment.
4. These views still require answers to questions about moral responsibility—about how these notions fit within a moral framework. This, again, is partially why I invoke the concept of duty, both to yourself and society (e.g., your housemates). This presents the question, however, of how to assess a person’s degree of responsibility in shirking their duties. Understandably, I suppose, given free will’s traditional connection to theological and jurisprudential concerns, revising its place in society may have significant implications for moral conceptions. I won’t explore that here, but do acknowledge it as yet one more difficulty for lessening free will’s hold on our collective views about justice and desert.
5. Responsibility generally involves notions about rationality. Namely, rational responsiveness to reasons. Humans, conceived of with or without free will, are responsive to reasons. Those reasons may act outside our awareness; I am concerned here with those that we are aware of. If the building I’m in catches on fire and I evacuate, I can reasonably say I ran out of the building because it was on fire. In other scenarios, such as whether to quit one’s job, more deliberation will be required.
An ability to rationally deliberate over reasons, choosing to act on them or not, plays into the extent to which we consider a person to be blameworthy for an act. And we expect the reasons themselves to be reasonable. “I ran out of the building because my cheese sandwich told me to” will be met with suspicion (and, I hope, compassion).
Suppose someone elaborately plans a murder and cites the reason as being: “Hey, I don’t have free will; determinism made me do it.” Claiming determinism doesn’t let one off the hook. Only a disturbed mind would sincerely produce such an abstract concept as justification for planning and performing horrendous acts. At any rate, this is taken care of by recognizing that people who are persistent sources of horrendous acts must be restricted in their ability to perform those acts (as humanely and compassionately as possible, I’d say). We might consider citing bizarre reasons, or bizarre responses to mundane reasons, to be a red flag—or a symptom—for persons who might be in danger of causing harm to others or to themselves.
1. I write about this more in the aforementioned response to Rachels. I’ll give a summary here.
2. One does not choose one’s desires, genetic disposition, physical or circumstantial restrictions, or endowments. One doesn’t even choose the disposition required in order to make the most of meager endowments (I say this in response to the common idea that what counts is not the hand you’re dealt, but how you play it; being good at poker is itself an endowment).
3. The student who studies hard for a C+ does not deserve that grade more than the student who effortlessly gets an A while barely studying at all. As noted above, we might dole out punishment or praise as a kind of conditioning device or as encouragement or discouragement. We might also do it because we like making our friends and loved one’s feel good. Perhaps these can have desirable outcomes on behavior, even when we know they are designed precisely for those outcomes. They can also be taken too far (physical torture, many of us agree these days, is punishment gone too far; and empty praise can lead to false self-esteem, which hurts confidence and is ultimately worse than useless). Again, there are empirical concerns here (e.g. What sort of treatment gets the best results on average?) and moral concerns (e.g., We shouldn’t cause any more suffering than is absolutely necessary in order to get the minimum required results).
At any rate, when we dig deep into who deserves what, the student who got the A didn’t deserve her grade any more than the C+ student deserved his. And so the same must be true if we switch the grades so that she slacked and got a C+ and his hard work earns him the A. (To be clear, I’m not just talking about who has tutors or a loving home environment—I’m referring especially to the distribution of inborn endowments. The students I’m describing here might be a brother and sister in the same home.)
3. In the context of desert, some further brief comments about the moral dimension of punishment and deterrence, and moral reasons for not making punishment any more harmful than we need to. We must recognize limits to what we are willing to do to an individual for the sake of deterring criminal activity in society. We (in the United States, at least) no longer, for example, think it acceptable to burn someone alive, neither as punishment nor as a warning to others. Even though it might be more of a deterrent than, say, death via chemical injection.
If we take the deterrence approach far enough, we might concoct even worse tactics. For example, burning the innocent child of a criminal to deter behavior in others who have children. And equally good results might be had (for less cost) by wrongly accusing someone and falsifying evidence.
Phony desert and genuine desert (if there is such a thing) appear identical to the outside observer; it strikes me as a dangerously malleable and vague category.
4. There is no strict quantifiable connection to be made between the intensity of desert and the intensity of corresponding punishment (i.e., suffering) one thus deserves. For example, suppose a man points a gun at an innocent person’s head, intending to kill that person, then pulls the trigger and hits that person’s leg; the person is mildly wounded. Now suppose another man points a gun at another innocent person’s leg, intending to wound that person, then pulls the trigger and hits the person’s head; the person dies. Whose choices and intentional acts merit greater punishment?
Notice that on my account, where desert doesn’t figure in, the person who persistently accidentally goes around shooting people will need to be restricted in behavior, regardless of intention. As would the poor marksman who goes around accidentally not shooting people.
5. A conception of desert I might be able to endorse, with some qualifications so that it is neutral rather than punishment-oriented, is something like: A consequence or response that follows from some action.
A consequence would be something that seems to happen on its own, organically, naturally, etc. For example: A consequence of a more or less healthy person drinking plenty of water is to be sufficiently hydrated.
A response is something that happens in the actions and thoughts of other people. What counts as an appropriate response to, say, someone stealing a loaf of bread will need to be worked out. It strikes me that there is no absolute answer to the question of what such a person deserves, i.e., what the appropriate response is. I argue that an appropriate response is one that is in line with our our values—i.e., humane, compassionate, and one that is not based on outdated notions about free will and personal responsibility, while also recognizing that human agents are responsive to reasons, etc.
What counts as humane is difficult to work out in this context. My view is that it is that which promotes only the minimum amount of suffering possible in an individual while also meeting some standard for justice. As with anything, there are kinks to be ironed out. Capital punishment may be persuasively argued by some to be humane: the quick and painless death of an adult human counts as a relief from suffering. Well-intended ideas commonly get this sort of treatment, so that the word remains but the underlying concept is distorted—just look at the range of countries that have called themselves democratic; humaneness is no exception. Though, to be clear, I think they are distortable because they are philosophically thorny.
Of course, resentment and such are natural human responses to being wronged. I wouldn’t suggest that they can or even should be avoided. There is a large jump, however, from feeling resentment to structuring institutional policies about justice that are designed to feed resentment. A critical role of institutions is to help avoid the irrational, emotionally driven feelings held by individuals in favor of a well-considered system that is as fair as possible for all involved. The justice system should not be a vehicle for satisfying a collective sense of retribution; at least, such a system strikes me as being designed to satisfy baser instincts, and not the better angels of our nature. Living in a civilized society is, for many of us, a constant exercise in finding ways to dissipate such instincts via socially, morally acceptable activities; our institutions should be designed to make that easier, while also supporting the rights of individuals to function as freely as possible within their society.
V. When does free will matter?
1. If you believe that you have (or in essence are) an eternal soul that will be damned to eternal Heaven or Hell (or some such) by a personal deity, then free will might matter a lot (an exception could be if you think God determined before you were born which place you’ll go). Indeed, this is one of the traditional reasons for the question of free will being so important. And not just for humans, but for supernatural beings as well—e.g., see Saint Anselm’s (1033–1109) On the Fall of the Devil.
2. It might also matter to you personally. Suppose your belief in a lack of free will makes you feel like some kind of finger puppet for external forces. I wouldn’t presume to be able to talk someone out of this feeling (if indeed anyone actually has such a feeling, which I doubt). But one problem I see with this is a potential incoherence arising from the difficulty of distinguishing between the puppet and the puppeteer, given that much of what counts as external forces—which include biochemistry, desires, and so on—accumulate, along with experience of life events (e.g., episodic memories), behavioral tendencies, and so on, to make what we conceive of as the self. There is no clear line, if there even is a line, between where the external forces end and the self, the thing that either does or does not have free will, begins; if there is any pure form of that self, we might be tempted to say it is something like pure consciousness, but that seems sorely misguided (indeed, would seem to be a vague reference to a kind of tabula rasa unhindered by the sorts of baggage that lead to a self, which is precisely what meditation traditions are positing when they equate pure consciousness with detachment from the self).
There’s a lot to unpack here. I’ll simply point out again that it’s unintelligible to say that you could be the author of your own desires, which constitute a key external force and self, because whatever desires you choose will be those desires that you desired, unless you chose them randomly in which case you did not choose them, nor could you have chosen to desire to choose them randomly. There is no intelligible way that you can be exempt from external or unconsciously (non-consciously?) formed internal pressures.
Put simply, if free will means being the supreme author of everything one does, of all unconscious processes and the desires that those manifest—then the concept is so unintelligible that I can’t imagine being bothered by contemplating its nonexistence.
I’ll put this another way. I find this a strange genre of anxiety. Suppose you accept a certain definition of free will. According to that definition, you seem to have free will. No anxiety. But suppose you were to accept some other definition. You would then have anxiety. But this is purely based on the definition you happen to accept. (Or perhaps it goes the other way: You accept the definition that aligns with your already existing anxiety.) Every other aspect of your lived life—your entire phenomenology of choice and action—would be exactly identical aside from the definition you choose, the anxiety this produces (or ameliorates), and whatever justifications you produce or intuitions you experience in favor of accepting that definition and of accepting or rejecting free will more broadly.
You also have the comfort of knowing that this is the same situation all humans are in; indeed, all possible conscious (especially meta-cognitive) agents of any sort, I’d say.
3. It’s tempting to think that accepting one or another definition of free will does amount to quite distinct phenomenological existences; perhaps the amount of control felt by a person, and perhaps even the content of that person’s conscience, can influence which status of free will the person accepts.7 I think that in academic contexts, this is less relevant—there may be many intellectual, professional, philosophical, or intuitive reasons to accept some status of free will. For the guilty person on trial, however, there would be obvious reasons for having a particular view of free will, just as there would be for the attendant lawyers and judge. We might say the same of sex-offenders, addicts, and others who commit the sorts of acts that are typically correlated with failures of the will, as well as of those who are the victims of those actions.
4. Studies suggest that research subjects who read a convincing argument against free will are then more likely to cheat, steal, and lie. It may also inspire empathy. For a critical overview of such studies, see this blog post by Jerry Coyne: “A new paper suggesting that belief in determinism makes you more empathic and less vindictive.”
I’m skeptical about such effects being lasting or far-reaching in normal day-to-day life. I would expect, however, that people who believe that free will is nonexistent or highly restricted by circumstances or metaphysics or what have you, and who dedicate significant portions of time to contemplating such questions, would see a change in their attitudes about related concepts such as desert and punishment; hopefully they will also contemplate the moral significance in their own role as agents who are responsive to reasons (again, this may be especially significant for people who are traditionally considered to have failures or defects of the will, e.g., active alcoholics; and, again, the rational “determinism made me do it” doesn’t generally fly).
I wish to re-emphasize the significance of responsiveness to reasons. Indeed, most who argue that free will is an illusion (e.g., neuroscientists Michael Gazzaniga and Sam Harris) also like to rightly point out that humans are still responsible for their actions in the sense of being responsive to reasons, having moral obligations, and so on, and such thinkers are still happy to, for example, praise their children when they’ve done well. This is how most of us go along through life; it reflects how we feel and shows respect for how others feel, and free will being true or not has little to do with it.
My take on the attitudes of such thinkers is that they don’t see themselves as living in a haze of illusion, but rather live in a kind of clarity about the natural sense of control we humans have, which amounts to the sort of acceptable self-correcting illusion (or delusion, if you prefer) analogous to that of, for example, a color appearing lighter in context than it would in isolation, for the sake of our having a consistent and useful experience of the world. Feeling a sense of control over our actions clearly has some advantage (about which it would be easy to speculate, though I’ll leave that for another discussion).
Interestingly, I recently heard an interview with neuroscientist Chris Frith (Philosophy Bites, 2/20017, “Chris Frith on the Point of Consciousness“) in which he points to studies such as the above, saying, “If you tell people they don’t have free will and they believe you… more compelling to me is that their behavior in reaction time tasks changes. So, normally in reaction time tasks you slow down after you make an error, which is due to some monitoring of your behavior and taking account of this, then you get less slowing down after you’ve been told that free will doesn’t exist, presumably because they have lost their faith in top-down control. And it even changes the amplitude of the readiness potential in the brain, which of course is what Libet was measuring in his famous anti-free-will task. I think this is fascinating because, basically, this is an example of top-down control, what people tell you influencing how your brain works, which is what I think free will is all about. So that telling people they don’t have free will actually demonstrates that we do.”
While I also find this behavior fascinating, I’m not sure I agree. People are responsive to reasons. If you tell me the milk tastes spoiled, I won’t drink it (unless maybe you offer me a thousand-dollar dare); tell me there’s a swarm of bees in the backyard, I’ll avoid the backyard (make that one a million-dollar dare); tell me I don’t have free will, I might get lazier about monitoring my behavior, and I might even do this strictly because I think you think I don’t have free will, even if I in reality continue to believe that I have free will.
An account of being responsive to reasons can be given without invoking free will.
This can play out in subtler ways. Consider the work of Claude Steele and others showing that, for example, when people perform an athletic or intellectual task their performance may be significantly influenced by what they understand the goal of the task to be, and by how their race or gender may be seen as a more or less stereotypical contributing factor to their performance. Steele and colleagues have named this phenomenon stereotype threat. The term does not denote some independently existing thing floating around in the world, but rather is shorthand for a certain set of psychological states that have a certain kind of observable, measurable, predictable, etc. influence on behavior.
It’s imaginable that free will could similarly come to be seen as falling under some class of behavior-influencing psychological states. Indeed, stereotype threat itself could be seen as a sub-class of states that count as a restriction on free will (or at least on freedom).
The more deeply aware people become of such phenomena, the better; particularly if within a sound moral context. For a thought-provoking exploration of the ways in which deep contemplation of such ideas can influence one’s lived life, see Susan Blackmore’s excellent 2007 collection of interviews with leading consciousness researchers (includes scientists and philosophers): Conversations on Consciousness: What the Best Minds Think about the Brain, Free Will, and What It Means to Be Human.
Again, though, while an appropriate revision of the metaphysical concept of free will may facilitate greater empathy among people in general, I don’t think we need the concept in order to achieve that goal, and indeed may be better off without it. It might be easier to promote the idea that someone with a limited range of options, due to whatever factors (including socio-economic), will have fewer available paths than someone with a much broader range of options. This is a materialist, psychological account of restrictions on choices, no complicated or convoluted metaphysical stories involved. (I wrote more about this distinction in my piece, “Free Will Paradox?”)
5. Free will plays a large role in our society. For example, it is a foundational idea in our legal system, which I take to be a formalized implementation of our informal notions about free will. For an engaging discussion about some of today’s most significant connections between law, science, and free will, check out this April 2016 episode of the All in the Mind podcast: “Neurolaw” (featuring guests Jeanette Kennett, who holds joint appointments in Philosophy and Cognitive Science at Macquarie University in Sydney, Australia; and Stephen J. Morse, Professor of Law and Psychiatry, University of Pennsylvania). Professor Morse, a self-proclaimed compatibilist, vehemently supports free will’s place in the foundation of our legal system.
6. Interestingly, while most people believe in—or act as though they believe in—some robust version of free will, there also seems to exist a strong popular belief in the idea that whatever happens is “meant to be.” It would be interesting to study how these attitudes may coexist in individuals, both in secular and religious contexts.
V. Conclusion, Thoughts for the Future, and Desire Revisited
The concepts that conventionally come along with free will—punishment, desert, navigation of desire—are not likely to go away anytime soon, but can be conceived of in ways that are not only free of the long-accumulating metaphysical and supernatural baggage of free will, but are also more in line with our young and still developing humanitarian ethos in which persons are treated with dignity and made to suffer no more than is necessary, and in which the purview of science is increasingly acknowledge (in terms of what it can and cannot tell us, the ways in which it can and cannot guide us, etc.). Such conceptions will of course come with their own practical and conceptual difficulties, but they are better tuned to our new world, and to the world we envisage for tomorrow.
On a final note, speaking of tomorrow, perhaps technology will encourage yet further revisions of our conception of free will.8 Consider, for example: What could it mean to program free will into a computer? There is increasing talk from computer scientists and the like about returning to the once out-dated idea that computers will be at their best when conceptually—and indeed physically—modeled after human cognition and neurobiology. Given that even programs engaged in deep learning must have a starting code, perhaps this will steer us more in the direction of compatibilism about free will. There’s a lot to unpack here. I’ll give it a rudimentary introduction.
First, I’d like to comment that I think it’s important to think about such things. We increasingly see a kind of interdependence between conceptions about the human brain (and mind) and conceptions about computers and computing. We use metaphors, like memory and representation, to talk about computers, but the lines between metaphorical and literal meanings are fading. If our concepts about human attributes are revised, I think it should be for the benefit of conscious entities rather than non-conscious ones, not withstanding how cool it would be to figure out how to program a computer to have free will more or less as that idea is currently understood; my worry is about revising our conceptions about human attributes, including in terms of how they are applied to humans, in order to arbitrarily match the limitations of computing. That said…
We humans don’t have free will as babies. The feelings and capacities associated with free will comes about once we have a system of beliefs and attitudes etc. in place that direct our decisions (which includes navigating and exercising whatever sort of inborn dispositions we may have). That’s my system of beliefs, preferences, desires, and so on that resulted in that decision, whether I made the decision after lengthy deliberation or in a split second. I can participate in a feedback loop with that system (e.g., telling myself daily affirmations in order to enhance my beliefs about myself), but only to a limited degree (the desire to perform those affirmations also arose from the system; and it may be easier to self-instill damaging rather than uplifting beliefs).
We can also imagine multiple instantiations of self-looping software (i.e., self-programming, over time) resulting in cognitively diverse computer systems, even though each started with the same coding—each system may commit to different decisions in the face of the same circumstances due to having even slightly differing experiences and perspectives over time. In other words, you don’t program free will (an oxymoronic coupling of terms), but instead you code a starting point that facilitates the development of a distinct system of preferences.
We might wonder why we would want multiple computers giving different answers to the same questions—to have opinions. I think this could be quite useful; unexpected expertise could develop, or some computers might have stronger and more dependable intuitions than others (something I’m noticing more and more AI specialists calling for), depending on the experiences they’ve had. Some might become smarter—be correct more often—than others. And if multiple computers still gave the same answers to the same question, that would be useful as well; perhaps the average answer given would be better than that from a single computer (as with certain types of questions and human groups).9
Whether computers can be said to have beliefs and attitudes and such will come down to whether a belief is to some degree a cognitive state, or is a purely behavioral disposition. I take belief to be significantly cognitive. But rather than focus on that here, I’ll accept that computers can at least have something much like preferences, and will point out something that strikes me as a deeper worry about the prospect of programming free will, having to do with desire.
Desire is an important component of free will, however currently conceived. By dismissing desire, you could have a non-conscious computer that has free will so long as it has a set of preferences. These preferences may have been developed as in the above-noted process, or given to the computer from the offset. In other words, we would then consider current computers to have free will, despite not being conscious.
Consciousness is relevant here as desire is something that is felt. It’s not just a preference in the sense of what one tends to do when faced with a choice among options (that’s a disposition), but rather is the felt dimension of what one would like to do or have or avoid, etc. It is also relevant given that it seems to be consciousness of a sophisticated (i.e., at least human-like) sort that is what accounts for free will, not only because our belief in its existence comes from a feeling of control, but also given that, for example, while a rock has no choice but to roll down a hill and has no desire either way about it, a human rolling down a hill may wish not to, and even try not to (despite having no choice but to roll down the hill).
Considering what it would mean to program a computer to have—or even to appear to have—free will may inspire new ways of thinking about the question. I explore this a bit more in a forthcoming piece written in response to the show Humans.
- I wrote this after writing a piece called “Free Will Paradox?“—I don’t think either is required reading in order to make sense of the other, though I do reference that piece here when helpful.
- I take it this hierarchy is loosely in line with what Harry Frankfurt describes (with more rigor) in papers such as “Freedom of the Will and the Concept of a Person” (The Journal of Philosophy, Vol. 68, No. 1 (Jan. 14, 1971), pp. 5–20), but my hierarchy is without: accompanying notions about proper accordance within the hierarchy amounting to an expression of free will; distinguishing second-order desires and second-order volitions; self-identifying via second-order volitions (a useful idea, perhaps, for thinking with conceptual clarity about addiction, though as with so many other examples of addict behavior in the free will literature, I’m not sure the account is useful in the case of those with longterm sobriety); et al.
Put simply (as I understand him): Frankfurt’s first-order desire is a desire to perform an action; a second-order desire is a desire to have a particular first-order desire; a second-order volition is a species of second-order desire in which one wants to not only have a particular first-order desire, but for that desire to accord with, or to in fact be, one’s will; that is, what one is effectively, and in the end, moved to do. As Frankfurt puts it in the above article: “A person’s will is free only if he is free to have the will he wants” (page 18). This means that if the person whose will is free has competing first-order desires, then whichever of these he makes his will, “he could have done otherwise than to constitute his will as he did” (page 19). I’m leaving out further nuances (e.g., “It is a mistake, however, to believe that someone … acts of his own free will only if his will is free” [page 19]) and idiosyncrasies (e.g., his notion of a person), but I’ll leave it at this.
- In other words, it wouldn’t do to name your coffee free will, and then claim to have free will in the mornings.
- Perhaps this sort of vagueness is what leads folks like Paul Churchland to argue that desire (as well as beliefs and other such states) are really just objects of folk psychology that will dissolve away as we gain a better understanding of the biological mechanisms responsible for subjective experience.
- Free will is not the only concept of this sort. For example, a technical definition of race—say, one grounded on pharmacological phenotypes rather than vague, unscientific, and unquantifiable notions about skin color or the geographical origin of one’s assumed ancestors—could be more useful than harmful, if universally adopted. I don’t think the time is right for this now, and for several reasons, but maybe in a few more generations it will be.
- Ferdinand D. Schoeman, American Philosophical Quarterly, Vol. 16, No. 1 (Jan., 1979), pp. 27–35
- I’m reminded here of the idea that those who claim there is no hard problem of consciousness, particularly when they do so on the grounds that subjective experience is not a significant feature of consciousness, or even that qualia don’t exist, actually have less vivid visual imaginations. There’s an interesting history of debates about the significance of mental imagery and whether such things even exist. For a brief but interesting discussion that touches on this, check out this August 2016 interview on Philosophy Bites: Jesse Prinze on Thinking with Pictures.
- I’ll note again the care that should be taken in revising concepts and the terms used to refer to them, particularly if they are thought to surround some concrete, independently existing phenomenon in the world (which, again, I do not think free will is such a phenomenon—but either way, it is a significant point to be aware of).
- What sort of question? I don’t mean moral questions. I’m talking about things like what might be the best course of action to take in the development of a drug, for one example of many. Another might have to do with how subjective experience correlates to brain activity. Yet another might be something like having computers that function much in the way service animals do, but at a much more sophisticated level. I’ve been contemplating how computers that have learned about the brains of a particular coma patient might detect whether the patient is to some degree internally aware, and, if so, develop a means by which the patient may communicate with the outside world; that patient’s brain would likely constitute a wildly unique system, and this would be reflected in the computer’s developed system. We might see similar applications for personalized medicine, based on the uniqueness of individual genome sequences.