Free Will
This piece originated in a public letter exchange between the philosophers Benjamin Burgis and Jon Rosen. The letters are no longer public, but I quote enough of them here to make it easy to follow their main arguments, into which I interject my own views on free will. I also talked to Ben and Jon extensively on the topic on my monthly podcast, Two for Tea, here. A version of this piece appeared in Areo Magazine: you can read the original here.
Compatibilism and Determinism
There are three main philosophical approaches to the question of free will: libertarianism, which emphasizes personal freedom and has nothing to do with the eponymous political movement; compatibilism, which includes a version known as semi-compatibilism; and incompatibilism (also called determinism). A full account of these schools of thought is above my pay grade. But here are some jumping-off points for further exploration.
Burgis begins his exchange with Rosen by arguing that we have at least enough free will to be held morally responsible for our actions. He correctly characterises Rosen’s stance as two-way incompatibilism (free will determinism): i.e., independently of what quantum mechanics may reveal as to whether the actions of subatomic particles are random or predetermined, Rosen believes that we do not possess free will. Burgis explains,
I’m a compatibilist. I believe that even if we live in an entirely deterministic world—one where every single event, including the ones that happen in the human brain, is something that could not have not happened given previous events and the laws of physics—we have the kind of control over our actions that matters for moral responsibility … Incompatibilism is, of course, the view that free will and determinism are incompatible … Two-way incompatibilists believe that free will is incompatible with both determinism and indeterminism.
Burgis also invokes John Martin Fischer, who argues that, even if the universe is completely deterministic, we still bear moral responsibility for our actions. Centrally at stake is the degree to which we control our actions and therefore can be regarded as morally praiseworthy or culpable. Despite our tendency to anthropomorphise—“who among us has not told a self-assembly wardrobe to fuck off?” as Will Storr has quipped—we don’t seriously blame inanimate objects, weather systems or animals for the effects they cause. But should we treat humans differently?
The God Problem
The most extreme way to hold a person responsible is to punish them with eternal torment or reward them with infinite bliss. Rosen relates,
I grew up in a fairly religious Jewish household and as a child I believed in God—an all-seeing, critical Mind that looked down on human beings, observed their thoughts and actions, and maintained a ledger of good and evil deeds, according to which God would determine appropriate compensations … So intensely did I believe this, that I posted the 10 Commandments above my bed, and every night would review the day’s behavior, acknowledge failure, and vow to improve.
My faith in a fair-minded God came under strain when my young friend—an irreproachably sweet-natured kid—developed leukemia. I tried to imagine some secret, heinous deed he’d done to deserve this punishment. But I couldn’t come up with anything that could justify the exhaustion, the shaved head, the deceptively hopeful periods of improvement, the brutal grind to death. “Why did this have to happen to me?” he had asked shortly before dying. I could fathom no plausible rationale, and so, lacking the patience of Job, I rejected God.
Religious people have always wrestled with the problem of free will. Given God’s omnipotence and foreknowledge and our own frailty—with which he, presumably, endowed us—it seems perverse that he should allow people to sin and be condemned for it. If it’s true that “as flies to wanton boys are we to the gods,” moral judgements are both meaningless and unfair. Hence Milton’s description of Adam in Paradise Lost (1608–74) as “sufficient to have stood, though free to fall.” “If I foreknew,” Milton’s God explains, sounding a tad defensive,
Foreknowledge had no influence on their fault,
Which had no less proved certain unforeknown.
Without free will, we cannot allocate praise or blame:
Not free, what proof could they have given sincere
Of true allegiance, constant faith and love,
Where only what they needs must do appeared,
Not what they would? what praise could they receive?
If life is a test, it must not be rigged. Yet how can it be fair, when it was devised by an omnipotent being, who already knows the outcome? No wonder William Blake considered Milton’s God a tyrant and believed that the poet was “of the Devil’s party without knowing it.”
Some Arguments in Favour of Determinism
Rosen’s recognition of the fundamental unfairness of much human suffering soon extended from the religious to the social domain:
Beyond that, when I observed the struggles of others I knew—and the homeless and the insane and the addicts in the streets of San Francisco—I could imagine no acceptable account whereby such people “deserved” their sufferings. There were plenty of conventional explanations … they … had made bad choices, or hadn’t lived responsibly, and, as such, brought their sufferings upon themselves. After all, it was commonly understood that intelligent, moral, hard-working people would never wind up in the streets, become drug addicts, or go insane.
Burgis is also disinclined to blame people for their own hardship. Like many leftists, he places more emphasis on caring for the unfortunate than on personal responsibility. He differentiates, however, between deserved and undeserved unhappiness: “A non-believer who believes in free will can say, ‘Some people freely commit wrongs, and they deserve blame, but no one deserves to live in poverty. No one deserves leukaemia.’” Here, he separates actions stemming from character from those resulting from external factors: “an unjust social system” and the amoral “natural world.” Rosen doubts that these are valid distinctions: “the more deeply I looked into the subject, the harder it became for me to identify a locus of personal agency that could render an individual genuinely responsible for their character or their circumstances.”
As a radical non-believer in free will, Rosen extends this beyond people suffering from misfortunes to encompass even those who have committed crimes. Criminals must be punished, Rosen stresses, in order to protect others and as a social deterrent. But he is reluctant to ascribe moral blame: “I am not arguing that humans shouldn’t be held accountable for their actions. If Jack robs a bank, Jack … needs to be held legally accountable. I am arguing that Jack is not, in a moral sense, blameworthy … he is not deserving of our vindictive wrath.”
The relevant question, Burgis proposes, is “When we say our will must be free in order for us to be responsible for our actions, what are we saying that it must be free from?” Clearly, as both philosophers agree, a number of external factors—of wildly varying degrees of likelihood—could exculpate bank robber Jack: he could have been hypnotised, mind-controlled by a Cartesian demon or a mad neuroscientist, blackmailed, etc. A neurological illness—such as Huntington’s disease—or a brain injury—think of Phineas Gage—might radically alter Jack’s personality. The Texas tower shooter, Charles Whitman, begged to be autopsied after his death, as he believed a neurological abnormality had led him to kill sixteen people, including his wife and mother. And, indeed, the pathologist discovered a pecan-sized tumour pressing against one of his amygdalae, a brain region associated with the regulation of emotions.
However, as Burgis acknowledges, there are also conditions “‘internal’ to his psyche,” such as “an addiction or compulsion,” that would render Jack morally incapacitated. These are murkier waters. It is often unclear where the dividing line lies between a personality quirk and a psychopathology. How can we tell disease from character, illness from evil? Determinists would argue that we no more choose our personalities than we decide whether or not to have a brain tumour.
Sam Harris pursues this line of thinking to its logical conclusion in his book Free Will. Harris describes the horrifying actions of two men who, between them, raped and strangled a mother and then set her house on fire, with her two daughters tied up inside. “If I were to trade places with one of these men, atom for atom,” Harris asserts, “I would be him: There is no extra part of me that could decide to see the world differently or to resist the impulse to victimise other people.” That Harris is not a mass murderer is a matter of pure luck: “I cannot take credit,” he tells us, “for the fact that I do not have the soul of a psychopath.”
Determinism is, as Rosen acknowledges, deeply dispiriting. It suggests that we are conscious automata, mega nanobots, meat hardware encasing neurological programs that run as automatically as our hearts beat and our mitochondria produce ATP—that we have only the illusion of control and therefore only the illusion of identity. “The possibility that I did not possess such authorial power,” he writes, “seemed to contradict my most fundamental sense of who I was.” As fellow determinist Ted Honderich puts it, “There can be no … hope if all the future is just effects of effects,” hence determinism is a “black thing,” an “incubus,” that has caused him depression and anguish. Yet this gloomy hypothesis is supported by some powerfully suggestive scientific evidence.
Some famous experiments with split brain patients have shown how capable we are of confabulation: we often do not know the true reason for our actions—and are skilled at inventing false motives and fooling ourselves. This is true even of people with an intact corpus callosum. As David Kestenbaum puts it, “We are machines that don’t know it.”
In his Enquiry Concerning Human Understanding (1748), David Hume writes:
We feel that our actions are subject to our will on most occasions; and we imagine we feel that the will itself is not subject to anything … For example, I play with the question of whether to raise my right hand or my left, and raise my left, but I have the feeling that in doing this I performed a kind of image or shadow of a decision to raise my right. We persuade ourselves that this image or faint motion could at that time have been completed into the thing itself.
There is indeed a “faint motion,” known as the readiness potential, which precedes physical actions like lifting the hand or flexing the wrist. Benjamin Libet’s controversial experiments showed that an MRI scanner can detect this precursor to movement before subjects make a conscious decision to move. What we experience as the intention to flex the right wrist, for example, may be simply our awareness of the wrist already preparing to flex. Libet himself later speculated that the readiness potential might signify not a decision but an urge, which “bubbles up” inside the brain unbidden and that there might be a window of time within which the subject can veto the action, exercising what he called free won’t. Critics have, however, pointed out that the veto impulse might also be merely an unbidden urge, bubbling up inside the brain. In Free Will, Harris takes this further: current fMRI techniques, he argues, allow us to predict movements and choices with a high degree of accuracy as much as four seconds before the subject is aware of her own intentions. This could provide neuroscientists with the right equipment with freakish short-term clairvoyance.
Some Arguments in Favour of Compatibilism
The arguments for free will are more difficult for a layperson to disentangle. Intuition and common sense suggest its existence. But science has shown that both those guides to reality are deeply fallible. The pro-free will side of this debate is therefore crammed with abstruse thought experiments and ruminations surely only a philosophy PhD could competently grapple with. We’re dealing here not with libertarian free will, but with compatibilism: the idea that people have sufficient free will to be reasonably held morally responsible for some of their actions.
Awareness is a prerequisite of responsibility, as Walter Sinnott-Armstrong elegantly illustrates here. If someone has—without my knowledge—filled the sugar bowl with arsenic and I poison you by brewing you a sweetened coffee, I am not a murderer. I am, however, also aware of many things over which I have no control, including thoughts, feelings and urges (anyone who has meditated knows how recalcitrant and unruly the monkey mind is).
Only certain kinds of knowledge are important here. Burgis cites a sumptuous piece on time travel, by David Lewis. Lewis poses a classic paradox of the kind featured in dozens of Star Trek episodes: his hypothetical traveller, Tim, journeys back to a time before his own birth and lurks in wait, loaded shotgun in hand, ready to kill, at point blank range, the grandfather without whom his own existence would be impossible. Is he free to shoot? In a sense, we are all time travellers with one-way tickets. We cannot know the future, but it will happen and it will happen in only one specific way. To some omniscient deity or all-powerful alien being, who did not experience linear time, we might not seem free at all. As Lewis explains,
You cannot change a present or future event from what it was originally to what it is after you change it. What you can do is change the present or the future from the unactualized way they would have been without some action of yours to the way they actually are. But that is not an actual change: not a difference between two successive actualities.
Despite this, Lewis rejects fatalism. Facts about the future, he writes, are irrelevant to the question of whether or not we are free at each present moment:
I am not going to vote Republican next fall. The fatalist argues that, strange to say, I not only won’t but can’t; for my voting Republican is not compossible with the fact that it was true already in the year 1548 that I was not going to vote Republican 428 years later. This is a fact sure enough; however, it is an irrelevant fact about the future masquerading as a relevant fact about the past, and so should be left out of account in saying what, in any ordinary sense, I can do.
But, as Rosen points out, citing Schopenhauer, the question is “Is it possible to demonstrate human free will from self-consciousness?” That is, does self-consciousness afford human beings a unique capacity to assume an autonomous or impartial stance with respect to their own character and fate?” What’s important here is the stance. Our ethical judgements stem from an evaluation of character. We judge a person to be good or evil on the basis of his actions—but only when we feel that those actions originate from his personality. That’s why we wouldn’t hold someone culpable in the same way if they killed someone by accident, while sleepwalking or hypnotised or during a psychotic break. According to Seneca, “man is affected not by events but by the view he takes of them.” Man is judged, too, not by events, but by the view he takes of them. As Hume puts it,
Actions are by their very nature temporary and perishing; and when they don’t come from some cause in the character and disposition of the person who performed them, they can neither bring him credit (if they are good) or discredit (if they are bad) … because they didn’t come from anything in him that is durable and constant as his character is, and they leave nothing durable and constant behind them in him.
Even a person who has no actual choice as to how to act, can be held morally responsible for her actions if she believes herself free to choose and is happy with the choices she has made. John Martin Fischer gives many such examples. Imagine, he suggests, that a man called Green has had a device implanted in his brain through which neuroscientists can monitor his brain activity and intervene to change his intentions, if necessary. Green is walking past a pond when he spots a drowning child and dives in to save her. However, if he had not done so, the neuroscientists observing him would have flipped a switch that made him decide to go to her rescue. Should Green still be praised for having saved the child? Yes, Fischer answers. Harry Frankfurt applies this same line of reasoning to people in the grip of addictions and compulsions. Maya Seddon and Peter Singer have summarised Frankfurt’s views:
having free will is a matter of identifying with one’s desires in a certain sense. Suppose that Jack is a drug addict who wants to reform. He has a first-order desire for a certain drug, but he also has a second-order desire not to desire the drug. Although Jack does not want his first-order desire to be effective, he acts on it all the same. Because of this inner conflict, Jack is not a free agent. Now consider Jack’s friend Jill, who is also a drug addict. Unlike Jack, Jill has no desire to reform. She has a first-order desire for a certain drug and a second-order desire that her first-order desire be effective … not only does she want the drug, but she also wants to want the drug. Jill identifies with her first-order desire in a way that Jack does not, and therein lies her freedom.
Sam Harris summarises this stance as “A puppet is free as long as he loves his strings.” Rosen cites Lynne Rudder Baker, who puts it much less dismissively: “If I can say, ‘These desires reflect who I am, and this is the kind of person that I want to be,’ then (surely!) I am morally responsible for acting on those desires.”
If we have any degree of free will, we must have at least the freedom to evaluate our wishes and actions and to decide if we identify with them. I can surely be held accountable, at least, for the actions I am proud of—though why not also for those I regret, which may stem from impulses and habits just as intrinsic to my character? But whether we can freely choose to make such judgements, whether we are in control of deciding what we celebrate and deplore about ourselves and our histories is itself an open question.
Cognitive Dissonance
I do not feel as though, at a deeper level, I have any real control.
My non-belief in free will inspires me to feel compassionate towards others, who did not choose their characters, proclivities, habits or anything that follows from them. And—most difficult of all—it helps me to forgive myself. Harris’ words resonate: “You are not controlling the storm, and you are not lost in it. You are the storm.”
Yet, I also feel as though I have free agency to follow the Stoic injunction to focus on my own words and actions as the only things under my control, as though it’s within my power—with an effort—to respect the Zoroastrian teaching of good thoughts, good words, good deeds. My mind seems, confusingly, to harbour both notions simultaneously and perhaps this is inevitable. Jerry Coyne—though himself a determinist—argues that, even if we know that free will does not exist, “nihilism is not an option: We humans are so constituted, through evolution or otherwise, to believe that we can choose.” Hume fittingly ends his musings on free will with the recognition that “the natural sentiments of the human mind … can’t be controlled or altered by any philosophical theory or speculation whatsoever.” Or, as Christopher Hitchens put it three centuries later: “Yes I have free will; I have no choice but to have it.”