Determined: A Science of Life Without Free Will by Robert M. Sapolsky
[review by S.F.]
Let me set your bearings. This is a review of Robert M. Sapolsky’s 2023 book Determined: A Science of Life Without Free Will, divided into three sections. The middle section is a straight up, no-nonsense overview of what’s in the book, delivered with me trying to keep my opinions mostly to myself. The first and third sections provide what I believe to be the book’s missing on and off ramps. To me, Determined is a book about religion that doesn’t realize it is. But despite that (perceived) shortcoming, I loved the book.
Section I starts with me defining religion in a way that should be new to you. It’s a definition with a simple, “mathematical” heart. You don’t need to be convinced of my definition to get through the full review, but you will have to be able to work with it. That definition goes on for quite a while, so if you can’t work with it, or don’t want to, but you’re still interested in hearing about a book on free will by a scientist (not a philosopher), just read Section II, which won’t win me any prizes, but should get the job done.
I.
Stranger and Enemy Used To Be the Same Word
British anthropologist Robin Dunbar accomplished a rare feat. He had a number named after him in his own lifetime: 150. Dunbar’s number is the approximate number of people any single person can know and remember well enough to assign a trustworthy trustworthiness score to. Dunbar was the first to notice that 150 shows up all over the human place. But for our purposes here I want to focus in on just one of them: certain types of groups split when membership hits 150. They can’t grow past it.
Humans lived in groups much smaller than 150 until relatively recently. The small sizes were probably due to frequently recurring ice ages. Humanity just couldn’t bloom until the next ice age failed to come along (and it still hasn’t arrived). In the extended temperate climate, the overall human population started to grow, and then, triggered by the subsequent crowding, the size of groups started to grow. Somewhere about 12,000 years ago the size hit 150. And then (relatively) shortly after that, moralistic religions appeared.
Non-moralistic religions had existed prior to that point, in the under-150 groups, but those religions never bothered themselves with defining right and wrong; they limited themselves to explaining the natural world. So it looks like morality was required for groups of 151+. Why?
The problem was strangers. Back then, all humans were born with a deep, fight-or-flight instinct for unrecognized humans. Unrecognized humans were like packs of wolves, but worse—because they could reason. “Run away!” was the instinctive response to an encounter with unknown humans, unless you were trapped or had an immediate and overwhelming gut feeling that you and yours could win the fight.
Stuck at 150
It’s easy to see why 150 was such a hard ceiling. When adult-151 came on the scene, anyone who met them forgot some other adult member. And then later, when that forgotten member was happened across, flight-or-flight triggered. (An intruder!!!) Soon enough, all hell broke loose across the group as the membership grew further and further past 150. Internal strife split the group.
This splitting-at-150 phenomenon went on for quite some time. It was an era of wildly proliferating societies, cheek by jowl, each with 150 adults, and each (eventually) with its own language (not just dialect). It mimicked cell division, but with very little communication occurring between the cells (think about why). And it was pointless to conquer your neighbors (think about why). Today, in the 21st century, the echoes of this stuck-at-150 stage still reach us. New Guinea has ~800 languages; Nigeria, ~500; India, ~450; China, ~300; and so on.[43]
The Secret to 151+
In order to break through the 150 threshold on group size–and there was most certainly selective pressure to do so–the fight-or-flight instinct for strangers needed to be replaced with a new, wait-and-see behavior. From our vantage point today, that seems like an easy problem to solve. But in fact it was the greatest problem humans ever solved. Instincts are not easy to override. Try walking through a roomful of large poisonous snakes after I tell you they don’t bite. And if you can do that (All hail the ophiophilists!) then come play with my pet hyena.
The solution to the problem was, of course, religion. Its sole job was to convince (the highly impressionable) young children in the group that humans were a tame species, that every person—all the way up to and including the stranger—was “essentially” kind. The stranger, even though you feared him most of all, was to be given a chance. As the children grew older and witnessed how the adults actually treated strangers, they would outgrow the belief, for the most part. They wouldn’t be able to fully dispatch it because it had come in too early, and was too deeply embedded. And over the generations the belief would take deeper and deeper hold until eventually everyone, adults included, came to embrace the lie about the tameness of humans. And then an amazing property of mass delusion kicked in: if everyone in a group believes something, it becomes true, in the only sense that matters to the group most of the time.
The All-Important Sales Pitch
Now, while children may be highly impressionable, they won’t believe just anything you tell them, especially when life and death are involved. So, the children had to be sold to, and sold to hard. The pitch that ended up working best involved the promise of an eternal, carefree afterlife featuring a scorekeeper who tracked the behavior of everyone on earth. It was a simple game: score well, and you go to heaven. And the scorekeeper could be schmoozed. He could hear people’s thoughts. He understood the mistakes people made and would forgive their bad behavior (if they were truly contrite, of course) by revising their scores upward. Although his voice could not normally be heard, the scorekeeper was always there and his presence could always be felt. (The marketing department later added a fiery temper, large muscles, long hair, and a flowing beard, but the scorekeeper started out as a simple, friendly and caring statistician with a crew cut.)
Variations of that pitch, incredibly (in retrospect), carried the day, allowing humans to live in groups of unlimited size—for thousands of years. And then it was fatally infected by science. True belief in the afterlife—and true belief is required for all (living) religions—could no longer be passed reliably across the generations. A new story was needed. Without one, large societies would collapse. Lucky for us, Religion 2.0 emerged.
The New Sales Pitch
The new 2.0 story dumped the gods (and their holy books) because they were the source of the problem. The 1.0 religions had all unwittingly taken on the baggage of the earlier “nature-explaining” religions, so their holy books foolishly contained claims which could someday be disproven—a religion conjurer’s rookie mistake, though I’m sure it must have been hard to conceive of modern science. So, in the design of religion 2.0, the gods and the holy books were replaced with science and a brand new, undisprovable white lie: human life is sacred. (Spoiler alert: human life is not sacred.) Stamping human life sacred not only causes each human life to be infinitely precious (providing the required tameness delusion), but also infinitely equal. The infinitely equal part turned out to be much truer than they knew at the time. It didn’t apply to just rich white men, as they had originally contrived it.
But the infinitely precious part was a stroke of genius. I myself was taught as a child that human life is sacred, and I believe it, deep, deep down. Even though I know there’s absolutely no evidence for it (we’re just smart animals), I couldn’t purge that belief in a million years. I am a humanist, you are a humanist, and we will die humanists. Religion works.
Morality is Implanted, Not Awakened
The morality that extends beyond the inborn morality for the 150, the morality that reaches all the way to the very last human, is not awakened in the individual, it’s implanted there. We are told that the love-of-the-150 (only) is a stunted, insufficient thing that was always meant to be extended to all humans. But there’s no evidence for that claim—and can’t be. However, does convincing ourselves of its “truth” help us survive? (Yes.)
We were brainwashed, pure and simple. We were bombarded with morality tales from the moment we could speak. And the moral instruction continues throughout our lives via story based entertainment. All our stories are just morality tales dressed up in action, comedy, mystery, romance and sex. We are desperate to be reminded that human life is sacred, that being kind to strangers is good behavior—because the brainwashing atrophies over time, and without it, group cohesion erodes. We start seeing ourselves as the dangerous animals we truly are. Even today, all humans are still being born with a deeply instinctual fear of strangers that must be suppressed. When it can’t be, we fight wars.
The Current Dominant Religion is Humanism
Now don’t get me wrong. I’m not using ‘brainwashed’ pejoratively here. It was all for the good! Humanism—Religion 2.0—has worked amazingly well with its sacred life gambit. There are two clear signals of its success (in addition to its emergence having prevented the collapse of civilization). The first is that all the 1.0 religions are upgrading themselves to 2.0 religions. Slavery, forced clitoridectomies, homophobia—and many other sordid cruelties—were once all totally in, and are now almost all totally out. The 1.0 holy books, which were literally their law books in the beginning, are now to be largely ignored, or “taken metaphorically”. The gods themselves have been forced into anger management classes, from which they emerge as humanists, as they must if their societies are to survive.
They wouldn’t be able to survive because the other clear signal that humanism has worked well is that humanists have the bigger bombs, and that’s the bottom line. People who are kinder to each other, it turns out, cooperate better and survive more effectively.
Humanism is the Way, But Not Forever
As humanists, we think we’ve got it all figured out. Like all true believers of all religions, we believe that morality has been solved! We believe that we don’t havea religion, we have the truth. We’re secular. Other people have religions. We believe that humanism is the end game; that there is no better core belief around which to organize a society; that before humanism, humans simply didn’t understand that just treating each other better was the answer to everything! Forever, and for all time. (Redundancy intended.) It’s just the way the universe works.
But it’s not. Think about where the never-ending reduction in human suffering leads when pushed to the extreme. Not to nirvana. To bubble babies. There’s only so much gas in the kindness tank, and there’s no gauge to tell us how much is left. If we go on forever like this we’ll eventually get caught in an infinite politeness loop of “No, you go first.” Do we escape that loop with the flipping of coins inside it? Where would that leave us? With randomness ensuring equality? Is that a good way to organize a society? I don’t know, but I doubt it. Have we run the game theory simulations on this?
It’s not sufficient to just swat down bubble babies and the politeness loop as ludicrous, extreme examples. Why are we always in the business of drawing moral lines? Why isn’t it always perfectly obvious to us what the correct way to reduce human suffering is? Is abortion ok? What constitutes a hate crime? What knowledge is harmful to children? Should children be able to command their own sex changes before puberty? And on and on and on. There are no answers built into the universe that we can discover. Like all other living things, humans are antifragile; we require some smacking around. Or are we to exist outside of the grip of natural selection?
We’re trying to define what constitutes human suffering as we go. We pretend like we know what we’re doing, but we always have to wait and see if what we chose worked out or not. And then we still don’t know. Not for sure. There are just too many variables. Many very smart people still believe, for example, that despite its repeated, consistent face plants, on both large and small scales, that communism is the answer. Is it?
Religion is All the Rules We Make Up for Behavior in a Society
A religion starts as a touchstone, as something like “there’s an afterlife”, or “human life is sacred” (aka, “all men are created equal”). Then from that touchstone all culture and law grow. Religion is not just the touchstone coupled with a book containing some stories that we consult in our spare time for spiritual guidance. Religion is the entire compendium of all the agreed-upon limits of acceptable behavior, and the punishments for breaking them, in a mind bogglingly complex world, with all of it justified against the touchstone.
Natural selection provided humans with a well-honed instinct for intuiting correct behavior in groups where everyone is known to everyone else. But once we strolled past 150 and introduced strangers, we had to define group behavior on our own, using rationality. And rationality has no idea what it’s doing. It’s still guessing with “everybody just be nice to everybody else”, and then trying to define exactly what that means as problems crop up.
Religion Helps Us Survive, Not Please The Universe
As pathetic as our religions are at defining large group behavior, they’re good enough for the job if they can win (or draw) wars against any other bumbling rational attempts at trying to define behavior for groups with strangers.
Can wars be won in the 21st century without women in the workforce? Probably not. Can a slave society defeat a mechanized society? Definitely not. And it’s not because slave owners are bad people and the universe is rigged to let the good people win. It’s because societies with slaves are inefficient. They waste human minds. It’s not about good and evil, it’s about winning wars. In fact, our definitions of what is good and evil are constantly being reworked to suit that purpose.
Consider this. Slave societies are gone now because they couldn’t survive against mechanized societies, but if things change and we find ourselves once again in a world where slavery is required for winning wars, it will surely come back in force. And we’ll have a lock-solid moral justification for it, which will be just what it was back then: get slaves or be slaves.
Last Shots
Still not convinced humanism is a religion? Think about this. Canceling someone is a profoundly religious act; it’s excommunication. How much more obvious could it be?
Or how about this? Freedom of religion in the United States is the freedom to practice exactly those parts of your religion which do not conflict with the actual American religion, our humanist laws and our humanist culture. So, by all means, go to town! Build your buildings, dance your dances, sing your songs, speak your words. Just don’t break the law or annoy other (reasonably respectful) Americans with your legacy-religion ornamentation.
Or, why are happy endings required for children’s stories? We tell ourselves that it’s because the children are too fragile, that they need to be toughened up before they’re ready for accurate depictions of reality. But wouldn’t toughening them up be best accomplished by gradually weaning them off happy endings? But we don’t do that. We provide them a steady supply of happy endings their whole lives through. Why? Because the unconscious can’t distinguish between story and reality! (This is why advertisements work.) It’s all reality to the unconscious—it determines truth by counting experiences. It doesn’t understand the context; that’s a logical construct. To the unconscious, the more something happens the truer it is, period. The unconscious mind is quite literally a dog’s mind. And it’s running the show! We’re continually counterbalancing actual human behavior with depictions of model human behavior so that the unconscious becomes deluded as to the state of ongoing human behavior in the outside world and acts accordingly.
You think you don’t have a religion because you don’t go to mosque/temple/church or read the Bible? They only did that stuff because they didn’t have Netflix. You get your good behavior supplements over the Internet. They had to go somewhere somebody could read theirs to them.
To you, the concept of universal love is an infinite, magical thing that tickles you deep down at the bottom of your soul. That’s your God you feel down there. You fail to see love for what it truly is: a high-valued group-cohesion coefficient. Why are all the gods so darned concerned with FEALTY TO THE GROUP??? You would swear they were proxies for the group if you didn’t know better.
OK, we’ve finally reached the end of my endless take on religion. Thank you for bearing with it. I hope I didn’t bore, exasperate or offend you, but I fear all three. The book I’ll now review in the next section is about Religion 3.0, which is just starting to peek over the horizon. Our humanism, religion 2.0, like religion 1.0, is going to fall victim to science.
II.
Sapolsky Fever
Determined was written by Stanford Professor Emeritus, Robert M. Sapolsky, one of my very favorite people in the whole world—in the fanboy sense. If you’ve seen any of his Stanford biology lectures from about 15 years ago, you’re probably in the club, too.[44] My first instinct is to say I like Sapolsky so much because he’s a deeply caring person, but that doesn’t get to the core of it. It’s the caring combined with a rare skill that you have to experience to understand. Let me put it this way: if speaking fluently meant speaking mellifluously for long stretches with no mistakes or hesitation, then we would say that Sapolsky speaks fluently. It’s uncanny to watch.
His previous book, his magnum opus, Behave, is—according to me—required reading for all humans. It’s a beautiful compendium of what we know about human behavior today. (Though it unfortunately appeared just prior to the replication crisis.) You simply don’t know how mis- and under-informed you are about humans until you read it. And the delivery! It’s a joy to spend time in Sapolsky’s company, even via the indirection of the written word. He’s organized, he’s interesting, he’s clear, he’s smart, and he’s funny. What else is there? So, if you revel in nonfiction tomes, it’s definitely worth checking out. (But do read the appendix carefully first. At least once.)
Determined
Determined, the book I’m reviewing, is his latest, and is still “new” as I write this in 2024, having come out in 2023. It’s very much a follow up to Behave (2017). The through-line in Behave is that we can’t explain anything without pointing to what came before it, and caused it. And if we further assume that there’s zero randomness popping up between cause and effect, then in order to get our final answer we have to walk cause and effect all the way back to the Big Bang. This is a painting of a strictly deterministic reality with no room for free will. Although Sapolsky does dip his toes into the free will waters at the end of Behave, it’s in Determined that he takes the full plunge.
Sapolsky is not a philosopher, he’s a scientist. The value he brings to the free will table is the uniqueness of that scientific view. Because he’s very close to the science, and to lots of different branches of the science, he’s in a position to effectively scout the universe for places where free will might be implemented, as it must be, somewhere, if it exists and is worth investigating.
Sapolsky is not trying to convince the reader that the world is deterministic. He’s assuming it. To most scientists and philosophers today, the world is so obviously deterministic that he doesn’t need to do that work. Instead, he’s addressing those people who agree that the world is deterministic but also insist that there’s a secret back door that allows free will to slip in. These are the compatibilists; they believe free will is compatible with determinism. According to Sapolsky’s estimate, their numbers make up about 90% of philosophers and legal scholars. Sapolsky himself is in the small minority of (hard) incompatibilists: for him, determinism unequivocally rules out free will.
Sapolsky is also not trying to convince the reader of no free will: “Again, my goal isn’t to convince you that there’s no free will; it will suffice if you merely conclude that there’s so much less free will than you thought that you have to change your thinking about some truly important things.” (p. 12)
The Approach
The book is organized into two halves. The first half is Sapolsky saying, OK, let’s have a good hard look at those secret back doors where free will is supposedly slipping in. He starts off by setting the table, much as I just have, and then, after dismissing the central, roiling debate in the free will space (Libet[45]) as inconclusive and irrelevant, he walks us through a quick review of what’s in Behave (minus the replication crisis problems). That brings us to the heart of the first half: the in-depth takedowns of the three main (remaining) compatibilist back doors. All three, as you might guess, are places where the science is currently stymied or misunderstood. Sapolsky proceeds to explain away each of the back doors. All three are fascinating to learn about independent of the whole free will fight. One of the back doors was mostly new to me, but I enjoyed Sapolsky’s walk-throughs of all of them. I’ll give them very brief introductions in just a bit.
The second half of the book is very different, and less dense. It’s essentially a history of free will itself, in the context of human ignorance. In this second half, we get to witness how our sense of free will is steadily weakening over time with respect to physical and mental illnesses. We once (truly) believed that people suffering from epilepsy were witches to be burned, and that mothers caused schizophrenia with frigid child care. But today those ideas strike us as stupid and cruel. So if we’re walking down a path on which we’re continually wringing a bit more of the free will out of misbehavior, where are we headed, and will we arrive there, and what would it mean if we did? Sapolsky’s histories are gripping reading.
My deep disappointment with the book happens in this second half. But before we get to that, let's get back to the three back door takedowns, so you can get a quick glance at those first half subtopics.
The Back Doors
One of the back doors is the randomness in quantum mechanics. This is one of the ones I was familiar with, and suspect you might be as well. All scientists agree that when an elementary particle “collapses” from a wave to a particle, the place in the world where that particle appears can only be probabilistically determined. It’s somewhat random. So can this randomness be the source of free will? No, according to Sapolsky, because (1) it’s confined to such a microscopic distance with no way to bubble up to where humans make decisions, and (2) because even if it could bubble up, it would cause random will, not free will.
The other two back doors are not so easy to summarize, so I’ll just breeze over them. One is chaos theory, which I also knew quite a bit about going in because Sapolsky speaks so fondly of James Gleick’s book Chaos (1987) in one of his online lectures, so I had read that. It’s great. The essence of chaos theory is that lots of stuff scientists had long assumed could never be explained (because it was “just too chaotic”) actually could be explained; it just needed a different mathematics. It’s quite a collection of stuff, from the weather, to fluid dynamics, to arrhythmias, to fractals. If you’re new to this topic, fascinating is an understatement.
The third of the back doors is emergent complexity. I’ve understood that all complexity must somehow be emergent ever since reading Dawkins’ The Selfish Gene (stupendously good), but to be introduced to such deep complexity growing from such simple rules in so many unrelated places was, well, only fascinating.
Where Sapolsky Leaves Us
So, here’s the high-level logic of Sapolsky’s case.
• Since there’s no place for free will to reside, inside or outside the human, we should try to proceed as if it’s not there, despite what it feels like.
• If it’s not there, then people are not acting intentionally.
• If people are not acting intentionally, then praise and blame make no sense—we simply do what we do.
• Therefore, we should try, even though it’s very hard, to praise and blame less. Misbehavior is not a choice, nor is good behavior. So, let’s do our best to reason our way through the counter-intuitiveness to a more humane world.
There’s an important clarification that goes with that final bullet point. Preventing bad behavior and blaming someone for it are separate things. So, for the safety of the group, it’s perfectly ok to confine someone (in a jail, say) but we shouldn’t be beating them up or berating them as a bad person who intentionally chose not to follow the rules. Chances are that in the future things will change, and we’ll appear as heartless ignoramuses to those looking back at us. So let’s at least be kind in our stupidity.
And the same goes for praise. No one is deserving of anything in a world without free will.
III.
What I Don’t Like About Sapolsky’s Conclusion
Here’s a specific example of what stains the book for me. In the second half Sapolsky relates how American mothers got shackled with the blame for their schizophrenic children through a twist of historical fate. The Freudian revolution was largely a European thing at first, but many of Freud’s followers were driven out of Europe by WWII. They wound up occupying most of the leading psychology positions across America. This cohort of analysts had convinced themselves that people were turned schizophrenic by “refrigerator mothering”. The history is fascinating and no doubt accurate, but Sapolsky lacerates the psychologists (“the psychoanalytic scumbags”, p. 329) and deifies those who took said scumbags down.
Sapolsky’s good-guy/bad-guy framing is engaging, heartbreaking and heartwarming—for a humanist who sees the world through a lens of good and evil! But the whole first half of the book describes and defends a deterministic view of the world! As we just saw, in such a world, everyone, scumbags and heroes alike, have no free will. There are no good guys and no bad guys in a determinist world; everyone is just doing what comes next, all the way back to the Big Bang. Sapolsky appears to forget this, or doesn’t really understand its implications, or thinks we won’t notice. He wants us to see the underdogs as victims of their deterministic biological fates, but not the privileged. They get to keep their free will so that Sapolsky can beat up on them.
The hardest thing to swallow about hard determinism is that it must apply to everyone, all the way to Hitler. Reality is just unfolding. Humans are not agents, they’re automatons. They’re like robots running on computer code, always following the next instruction, which may be a reaction to the environment they’re experiencing, but they still don’t ever have the power to jump the logic. They’re completely locked in.
According to the hard determinist, human history is a horror show of ignorance and suffering not because people are choosing poorly, or selfishly, or fiendishly; they’re not choosing at all! Human behavior is exactly like the weather. Sometimes it’s sunny and breezy and sometimes it’s a tornado. The weather isn’t deciding anything, and neither are we, despite how it feels.
Determinism is a hard concept to hold in one’s head, never mind defend, but that’s Sapolsky’s job. He fails at it because he fails to model it. He wants to have his good and evil and eat them, too.
His conclusion should have been, “The steady progression of science is going to make it clearer and clearer that free will—the keystone of morality—is a delusion, so what are we going to replace it with?” Good and evil are going to be outed as the bogey men they are. At some point, our children are going to start rejecting our happy-enders, just like some of us started rejecting Bible stories. What do we do then?
What We Do Then
Let’s review. For 12,000 years humans have been able to live in groups of unlimited size because we’ve managed to convince ourselves that cause and effect applies everywhere in the universe except inside humans. But, those groups of unlimited size compete against one another. And in order to win those competitions—those wars—we’re being forced to learn more and more about how reality actually works, including how the humans insideit actually work: deterministically. The free will delusion is on a collision course with science, and it will lose.
The free will delusion has survived against the onslaught of science so far by shape-shifting from communities of gods, to a single god, to a distributed god. Sapolsky’s book is telling us that the last bastions of the distributed god are now crumbling. And people like Sapolsky are beginning to struggle to maintain the delusion. This is a contagion that will spread, just like atheism did. I’m spreading it now.
While my purpose in writing this book review is to get you, the reader, to start thinking about what the next religious touchstone will be, I have my own guess. We’re going to replace sacred life with transparency.
Transparency is the New Freedom
Sacred life works by getting people to self-police on a false premise. The false premise is: treating any human life as less than sacred makes you an irredeemably bad person. We need the false premise because so much of our behavior happens when no one’s watching—when we could get away with it! As technology advances and surveillance increases, less and less of our behavior is going to require the false premise. Because if more and more of our behavior is recorded, and filtered by a (community-trained) artificial intelligence for violations, we won’t need the false premise. We’ll have a real realtime omniscience to use for that purpose.
That might sound like an Orwellian dystopia, but it won’t be one if we make the surveillance data public, and we will. We’ll create safe spaces (much like we create college campuses today) that are fully monitored by AI gods. As long as any of us can check and challenge the work of the AIs, we’ll be perfectly fine with the monitoring. It’s when only the government has(or only the corporations have) the surveillance data (and we don’t) that it’s dangerous.
These safe spaces will be required, in the survival sense, for all societies, just like colleges are today. Without them a society’s military and economy will fall far behind. Because inside the safe spaces, where all behavior is publicly transparent and duly and timely punished, misbehavior will diminish, cooperation will surge, and better weapons will be produced.
Think of it like this. Open source software was a step in the right direction. But what we need is open runtimes. How do we build systems that are fully transparent, and thus fully trusted? I think it can be done, and I think it will be done.
Our big problem living in big groups has always been that we don’t trust each other. Survival concerns will continue to lead us in directions where technology increases trust, as it has over the past several hundred years. Trust wins wars. Global, public, surveillance will increase trust. Therefore, it will emerge and survive.
Last Thoughts on Free Will
If your inclination, like Sapolsky’s, is to think that free will—even if it is a delusion—is just too deeply embedded in human consciousness to ever be dispensed with, let me try to disabuse you of that position.
Like good and evil, I believe free will is “taught” to us (forced down our throats). Young children experience their bodies moving like we experience our bodies twitching. With surprise. From the child’s perspective, her arms and legs are moving, but she’s not moving them; she’s watching them move. She has to be convinced otherwise so that she can function in a society where everyone takes responsibility for their own behavior. Free will is part of the morality package, and only 12,000 years old.
Sam Harris has convinced me that I actually have no idea what I’m going to think next, or say next, or do next. But I’ve been trained to take responsibility for all of it. Since so much of it is repetitive, I typically know what’s coming next, so believing that I’m willfully orchestrating it all is not so hard.
When the time comes, I predict, we can and will drop the delusion of free will relatively quickly. We’ll stop teaching it to our children.
And just because you can’t imagine how things could possibly work without free will doesn’t mean they can’t. People used to be befuddled contemplating a world without God. Some people still are.
Determined: A Science of Life WIthout Free Will by Robert Sapolsky
[review by G. H. T]
In sci-fi author Ted Chiang’s short story What’s Expected Of Us, the narrator describes a device called the Predictor. This device has a button and a light, and the light always, without fail, flashes a second before you press the button. No matter what you do, no matter what clever trick you try to pull, you’ll never be able to press the button before the light flashes, and if the light flashes you’ll always press the button.
The existence of this device definitively disproves free will, and it’s such a profound demonstration of this fact that a third of people who interact with it end up becoming catatonic. The narrator transmits a warning to the past, urging everyone to keep pretending as if free will exists anyways; “Civilization now depends on self-deception”, he says. “Perhaps it always has.”
Robert Sapolsky, who references this story in his book, ignores the narrator’s warning and charges forward anyways, in an attempt to introduce the Predictor to our world in the form of a hefty, interdisciplinary tome entitled Determined: A Science of Life Without Free Will. Unlike the narrator of that story, Sapolsky doesn’t want us to shy away from this uncomfortable truth: he wants us to confront it head on, and internalize it as much as possible. Is he right about free will being an illusion? And if he is, is he right that we should try our best to make ourselves really, truly believe it?
We’re Just Not Compatible
To talk about free will, we first have to define what free will even means. For Sapolsky it is as follows:
Show me a neuron (or brain) whose generation of a behavior is independent of the sum of its biological past, and for the purposes of this book, you’ve demonstrated free will. The point of the first half of this book is to establish that this can’t be shown.
The free-will supporters that Sapolsky mainly addresses in this book are compatibilists: people who believe free will and determinism are compatible. So what exactly is determinism? Sapolsky uses an example provided by eighteenth-century French polymath Pierre Simon Laplace:
If you had a superhuman who knew the location of every particle in the universe at this moment, they’d be able to accurately predict every moment in the future. Moreover, if this superhuman (eventually termed “Laplace’s demon”) could re-create the exact location of every particle at any point in the past, it would lead to a present identical to our current one. The past and future of the universe are already determined.
Hmm. Kind of sounds like Sapolsky’s criteria for free will - a neuron that exhibits behavior that can not be derived from anything preceding - is mutually exclusive with his definition of determinism, which states that all things can be derived from what precedes.
Is this a fair definition of free will? Well, it does indeed show that nothing could have happened other than how it happened. But what exactly would the alternative to that even look like? That, if you made an exact replica of the universe and ran it parallel, there would be two different outcomes? That you would make one choice in one universe and another choice in the other, despite your physical and mental state being exactly the same in both? That sounds less like free will and more like a dice roll.
Which Sapolsky agrees with: in the ninth and tenth chapters, Sapolsky talks about quantum indeterminacy, which is a concept that describes the inherently probabilistic nature of quantum mechanics. Essentially, at the quantum level, things are actually random, not fully deterministic. So this introduces the possibility that we could, in fact, act nondeterministically! Except Sapolsky argues that the actual effects of this quantum indeterminacy are so small that it’s almost impossible for it to result in anything of material consequence:
As summarized by one philosopher, “The law of large numbers, combined with the sheer number of quantum events occurring in any macro-level object, assure us that the effects of random quantum-level fluctuations are entirely predictable at the macro level, much the way that the profits of casinos are predictable, even though based on millions of ‘purely chance’ events.”
And if quantum indeterminacy really was somehow strong enough to introduce randomness into our decisions? Well, Sapolsky says, for proponents of free will, that’s even worse:
In his influential 2001 essay “Free Will as a Problem in Neurobiology,” philosopher John Searle wrote, “Quantum indeterminism gives us no help with the free will problem because that indeterminism introduces randomness into the basic structure of the universe, and the hypothesis that some of our acts occur freely is not at all the same as the hypothesis that some of our acts occur at random… How do we get from randomness to rationality?”
So, by Sapolsky’s definition, counter-intuitively, agency is dependent on determinism, and therefore dependent on there not being free will.
Deserve’s Got Nothing To Do With It
Sapolsky’s definition of free will isn’t the only one, of course. As Sapolsky says: “Many focus on agency, whether a person can control their actions, act with intent. Other definitions concern whether, when a behavior occurs, the person knows that there are alternatives available. Others are less concerned with what you do than with vetoing what you don’t want to do.” Essentially, these people use the term “free will” the same way Sapolsky uses the term “agency”. At the end of the day, what does it matter what definition we use?
What Sapolsky really cares about when he talks about free will, though, is what free will implies: moral responsibility. Because if people choose their actions, then they deserve to be praised for them, to be blamed for them, and to be punished for them. And that is the primary material consequence of belief in free will that Sapolsky is concerned about.
If Sapolsky is one of free will’s greatest haters, Daniel Dennett (who sadly passed away during the course of the writing of this review) is one of free will’s greatest lovers. Sapolsky addresses Dennett directly several times in the book, and they’ve even debated each other directly on the topic following the book’s release. In spite of the fact that everyone has a fundamentally different starting point in life, Dennett has this to say in defense of free will:
For example, in Elbow Room: The Varieties of Free Will Worth Wanting, he asks us to imagine a footrace where one person starts off way behind the rest at the starting line. Would this be unfair? “Yes, if the race is a hundred-yard dash.” But it is fair if this is a marathon, because “in a marathon, such a relatively small initial advantage would count for nothing, since one can reliably expect other fortuitous breaks to have even greater effects.” As a succinct summary of this view, he writes, “After all, luck averages out in the long run.”
This view seems laughably incorrect to me. For one, even within the confines of his metaphor, it’s wrong; a starting disadvantage in a marathon is still unfair, even if it’s not as unfair as it would be in a sprint! And with regards to the reality that Dennet is pointing at with this metaphor, it’s even less applicable. I could make my case, but Sapolsky’s already done it for me:
No, it doesn’t. Suppose you’re born a crack baby. In order to counterbalance this bad luck, does society rush in to ensure that you’ll be raised in relative affluence and with various therapies to overcome your neurodevelopmental problems? No, you are overwhelmingly likely to be born into poverty and stay there. Well then, says society, at least let’s make sure your mother is loving, is stable, has lots of free time to nurture you with books and museum visits. Yeah, right; as we know, your mother is likely to be drowning in the pathological consequences of her own miserable luck in life, with a good chance of leaving you neglected, abused, shuttled through foster homes.
So in a post-free will world with no responsibility, what does Sapolsky suggest we do about crime? The approach to justice that Sapolsky advocates for is something called the quarantine model:
As outlined by the hard incompatibilist philosopher Derk Pereboom of Cornell University, it’s straight out of the medical quarantine model’s four tenets: (A) It is possible for someone to have a medical malady that makes them infectious, contagious, dangerous, or damaging to those around them. (B) It is not their fault. (C) To protect everyone else from them, as something akin to an act of collective self-defense, it is okay to harm them by constraining their freedom. (D) We should constrain the person the absolute minimal amount needed to protect everyone, and not an inch more.
On the surface level, it seems like the quarantine model isn’t necessarily that different from what we currently have; this isn’t some radical abolish-prisons kind of proposal. We’d still incarcerate people for committing crimes, removing them from the rest of society, roughly in proportion to the severity of their crime. The difference is fundamentally philosophical in nature: rather than coming at it from a place of retribution, or even rehabilitation-qua-rehabilitation, the primary motivating factor would simply be public safety. We would hold people only for as long as necessary, avoid any unnecessarily harsh or cruel treatment, and focus on rehabilitating offenders to the point that they are able to rejoin society without presenting a danger to anyone else.
On the specific, practical details of such a system, Sapolsky has little to say. What would minimally harsh treatment look like? How would we know whether someone has been reformed to the point that they could safely be reintroduced to the public? Does this mean mitigating factors revolving around intent (e.g. the difference between first and second-degree murder) would no longer matter?
But I think it’s wise of Sapolsky not to venture too far out of his area of expertise. This is a book about science and philosophy, not policy, and I think enough people would disagree with even his basic assertion of what should motivate our idea of justice that it’s a non-trivial argument. And such a drastic change to the very foundation of our justice system would certainly result in very different outcomes if we could truly stick to such a philosophy while designing it, even if we couldn't predict exactly what they'd be. So I think it’s an admirable mission - certainly better than the retributive one we currently embrace.
(Un)True Grit
Nonetheless, I still find it really, really hard to shed the idea of moral responsibility entirely. Our idea of moral responsibility is built on intent, and that does exist; even Sapolsky wouldn’t deny that. What he would instead say is that you do not choose your intent:
You wish to do something, intend to do it, and then successfully do so. But no matter how fervent, even desperate, you are, you can’t successfully wish to wish for a different intent. And you can’t meta your way out— you can’t successfully wish for the tools (say, more self-discipline) that will make you better at successfully wishing what you wish for. None of us can.
As Schopenhaur puts it, you can do what you will, but you can’t will what you will. But… what would it even mean to “will what you will”? That implies there’s something, some higher-order “you”, that chooses your will, and then that must have been chosen by something that chooses the choice of your choice of will… the very idea of being able to entirely decide who you are and what you intend, independent of who are you and what you intend, seems inherently paradoxical.
In describing the infinite chain of causality (what you do is influenced by what happened a second before, which is influenced by what happened minutes before, which is influenced by what happened days before, which is what influenced by what happened years before…), Sapolsky uses the analogy of an infinite series of turtles stacked on top of each other in outer space; “turtles all the way down”, as the classic expression goes. To him, absurd as it is, infinite turtles are still more plausible than there being a final turtle: effects always have a cause, and those causes must have causes as well, and on and on and on until the beginning of time. But in society, we consider intent to be that “final turtle”, and it’s primarily how we determine for which actions someone should be held morally responsible.
But where is the line? The following is a table from the book that outlines what Sapolsky believes to be the false dichotomy we make between behaviors which are “biological” (i.e. out of our control) and the result of “grit” (i.e. within our control):
“Biological Stuff”
Do you have grit?
Having destructive sexual urges
Do you resist acting upon them?
Being a natural marathoner
Do you fight through the pain?
Not being all that bright
Do you triumph by studying extra hard?
Having a proclivity towards alcoholism
Do you order ginger ale instead?
Having a beautiful face
Do you resist concluding that you’re entitled to people being nice to you because of it?
Sapolsky argues, however, that both sides of the chart “are equally the outcome of uncontrollable biology interacting with uncontrollable environment,” and provides a few examples of behaviors - namely, epilepsy and schizophrenia - that have shifted from the left to right as our understanding of the conditions have evolved. We did it before, he says, so we can do it again.
But those particular examples are, crucially, about the lack of intent; and once we discovered the lack of intent inherent in those behaviors, we stopped placing blame on people for having them. We once thought epilepsy was caused by demonic possession, and specifically demonic possession that the epileptic individual had willingly accepted, which gave us a reason to blame them for having seizures. But now that we know it's simply a neurological condition that cannot be consciously controlled, we no longer blame people for it. We didn’t start becoming more forgiving of people for choosing to have seizures; we simply realized that they weren’t choosing to have them at all.
Case in point is another example from the book: Dorothy Bruns was a woman who struck and killed two children and injured several other pedestrians when she had a seizure while she was driving. Even though we don’t blame someone for having a seizure, she was still charged with involuntary manslaughter and criminally negligent homicide. Why? Because she was diagnosed with epilepsy and doctors told her not to drive, but she did anyway. We don’t blame her for having epilepsy, but we do blame her for acting irresponsibly in light of her condition.
Yet Sapolsky would say that this blame is misplaced:
How many neurons are there in your frontal cortex and how well do they work? What do the underlying disease and the drugs taken for it do to your judgment and frontal function? Is your frontal cortex a little light-headed and sluggish because you skipped breakfast and now your blood sugar levels are low? Have you had a sufficiently lucky upbringing and education to have a brain that has learned about the effects of blood sugar on decision-making and frontal function, and a frontal cortex functional enough to make you have decided to eat breakfast? What are your gonadal steroid hormone levels that morning? Has stress in the previous weeks to months neuroplastically impaired your frontal function? Do you have a Toxoplasma infection latent in your brain? At one point in adolescence were your meds working well enough that you could finally do the single thing that made you feel normal in the face of a shattering disease, namely driving a car? What were your adverse childhood experiences and ridiculously lucky childhood experiences? Did your mother drink a lot when you were a fetus? What sort of dopamine D4 receptor gene variant do you have? Did the culture that your ancestors developed glorify following rules, or thinking of others, or taking risks? On and on. We’re back to the table on page 104 in chapter 4 —“ having seizures” and “deciding to drive even though you haven’t taken your meds” are equally biological, equally the product of a nervous system sculpted by factors over which you had no control.
Within the quarantine model, it would be perfectly consistent to say that Brunns would still need to be incarcerated. Because even though her epilepsy was not her fault, nor was her choice to ignore her epilepsy to drive, she still displayed poor judgment and a lack of self-control that makes her a danger to herself and others, so it would benefit society to keep her from enacting more harm in the future. But in the quarantine model we would do this as a purely pragmatic decision, and in Sapolsky’s ideal society, we would view this tragedy the same way we view a hurricane: sad, but there’s no use in getting mad at the weather.
Deserve’s Got Something To Do With It
Yet if I was put in a classic trolley problem situation, and on one track was tied Dorothy Brunns and the other was a person with epilepsy who hadn’t had a seizure while driving and killed someone, I know who I’d pick. And it’s not just because I made a logical calculation about who’s a greater danger to society (though it’s certainly part of it); at its core, I’m making a moral judgment from a place of emotion. It’s about who deserves it.
Our desire to see wrongdoers punished serves a greater purpose - to foster cooperation by discouraging cheating. But the development of evolutionarily beneficial behaviors is a messy process. Sex, for example, has the biological purpose of passing down one’s genes. So why aren’t men satisfied with dropping a deposit off at the sperm bank every week instead of having sex? Well, evolution didn’t account for the existence of sperm banks, so it just gave us a libido as a proxy for the desire to procreate. So even when we have the means to achieve reproduction without going through the sexual process itself, we still find ourselves wanting to have sex.
Similarly, our desire for enacting punishment is a proxy for the ultimate goal of fostering cooperation. A quarantine model may lead to the same desired outcome of separating harm-doers from the rest of society, and reducing harmful behavior overall through deterrence and rehabilitation; it may even do a better job of it. But it won’t satisfy that innate urge in us to see bad people punished.
Still, a standard of impartiality that rises above human folly is, in fact, the goal of the justice system. In the United States, we have a principle of innocence until proven guilty. And though it clearly doesn’t always work entirely as intended, to put it mildly, the intent is there. As a jury member, even if you personally believe someone is guilty, you’re supposed to acquit as long as there’s some reasonable doubt. So that’s how a quarantine model of justice could survive contact with reality - we may personally never free ourselves from a sense of moral judgment in our day-to-day lives. But we can try, to the best of our ability, to free it from our criminal and judicial procedures.
Sapolsky brings up the classic liberal example of the Norwegian justice system, which is downright utopian compared to the United States’:
“...in Norway’s “open prison” system, criminals, even those under maximal security, have rooms rather than cells, computers and TVs in each, freedom of movement, kitchens to cook in communally, workshops for hobbies, music studios filled with instruments, art on the walls, trees on the campus-like grounds, a chance to ski in the winter and go to the beach in the summer.”
And contrary to concerns that such a system would be an ineffective deterrent, or even an incentive for crime, we see that Norway has a far lower recidivism rate. This means far fewer prisoners, savings from a smaller police force, and the obvious direct benefits of there just generally being less crime in society. It seems like a win-win: the only ‘loss’ is the satisfaction of our bloodlust.
So as social policy, I’m broadly in favor of the quarantine system as Sapolsky describes it, and the example of Norway seems like proof to me that, slowly but surely, we can as a society reduce our desire to see bad people punished to the maximum extent possible. We’ll probably never get it down to zero, but as they say, perfect is the enemy of good.
I’ve Got No Choice
In the last chapter of the book, hidden in a footnote, Sapolsky reveals that he suffers from clinical depression. The depression, he says, is actually a big factor in why he so easily accepts the earth-shattering notion of there being no free will:
Bummed out by the scientific evidence that there’s no free will? Try looking at your children, your perfect, beautiful children, playing and laughing, and somehow this seems so sad that your chest constricts enough to make you whimper for an instant. After that, dealing with the fact that our microtubules don’t set us free is a piece of cake.
My personal experience with the condition, however, left me with a more complicated view on the matter. Last year, I experienced depression for the first time. Both during and afterwards, I’ve wracked my brain trying to figure out what might’ve triggered it - I know that depression is a biological disorder, but from what I’ve seen, for most people it’s still related to something going on in their life, even something minor. Maybe it’s the ending of a relationship, maybe it’s trouble with work, maybe it’s loneliness, maybe it’s just a general malaise and lack of direction. But I’ve come up empty handed. At the time, I found myself worrying and feeling bad about many things: work, my social life, my relationship, whether I was a good person - but looking back on it, it feels like it was the depression that caused those negative feelings, rather than the other way around.
And it arrived very suddenly: one night, I went to bed perfectly content, and the next morning I woke up with the worst feeling of mental anguish I’d ever experienced. During this time, I felt like I’d essentially lost all control of my mind and body. At a very visceral level, all I wanted to do was lie in bed all day because I hated myself and thought of myself as completely worthless, and because I was so incredibly tired, physically and mentally and emotionally. For a few months I tried to power through it, hoping it was just a temporary phase, a slightly stronger version of the blues that I’d felt before and that I was sure everyone went through sometimes. But it only got worse and worse, and finally I decided to take antidepressants, which my psychiatrist had prescribed to me weeks before but which I’d been deeply hesitant about. They took some time to take effect, but when they did I started to feel a little better. But just a little bit. And then, about a month after I started taking them, I was cured: one night I went to bed in total mental anguish, and the next morning I woke up feeling perfectly content.
This extreme depressive episode left me with a set of contradictions in my head regarding free will. On the one hand, it strengthened my belief in it: while I spent a lot of time lying around doing nothing and feeling sorry for myself, some small part of my brain was constantly trying as hard as it could to fight that in a way that did make me feel like I had a sense of agency and self beyond my surface-level actions, and even my surface-level thoughts and feelings. You may not be able to choose your own intent, but during that time I felt what it was like to have two competing intents simultaneously. My first-order intent, the one closest to the surface, was to wallow and rot; my second-order intent was to do anything but that. And even though that second-order intent rarely won in the war raging in my mind, it felt a lot like that one was the “real me”, and that the depression was just an external threat that had invaded my mind like a parasite and implanted this first-order, stronger intent into my brain, taking away my free will - which only makes sense if I had a free will to take away in the first place. This stark contrast between how I felt when I was in the thick of it and when I was out of the woods really illustrated for me what not having agency feels like - and conversely, what having agency feels like.
But it also really hammered home for me how brutally unfair life is, to be born into this world and be forced to exist against your will. Because sometimes, the “real me” seemed to fade away entirely, and the first-order intent was all that was left. And I think that experience - of something wholly superseding and rewiring my intent - is the closest I’ll come to really internalizing Sapolsky’s radical claim, and how arbitrary the line we draw between what happens to us and who we are really is.
In the book, Sapolsky brings up the classic Psych 101 story about Phineas Gage, a man who survived having an iron rod driven through his head but whose entire personality shifted after the incident due to severe brain damage, turning him from a generally well-liked guy into a profane, impatient, and obstinate jerk. Most people, I think, would say they feel sorry for Gage, that what happened to him wasn’t his fault and that his post-accident behavior was understandable, perhaps even forgivable, in light of his unfortunate circumstances. But what if, instead of suffering an accident, he was simply born like that? That, whatever state his brain was in after the rod had injured him, that was just the state of his brain the moment he was born? Such that he was naturally just an ill-tempered dick? In both cases the cause of his affliction would be entirely out of his control in exactly the same way, neurologically speaking. But would people feel sorry for him then?
For all the elaborate scientific explanations about our behaviors and radical claims about moral responsibility that Sapolsky makes, the most important message of Determined isn’t any more complicated than the lesson at the end of a sappy children’s cartoon:
There is no justifiable “deserve.” The only possible moral conclusion is that you are no more entitled to have your needs and desires met than is any other human. That there is no human who is less worthy than you to have their well-being considered. You may think otherwise, because you can’t conceive of the threads of causality beneath the surface that made you you, because you have the luxury of deciding that effort and self-discipline aren’t made of biology, because you have surrounded yourself with people who think the same. But this is where the science has taken us.
In the end, I don’t think the free will debate will ever have a definitive answer, because it’s not a question of fact but of feeling. There are times when I think the framework of free will is useful, times when I think it’s not, and times when I think it’s simply inescapable. Sapolsky is right that everything that happens to us is, in a sense, predetermined; and people like Dennet are right to say that we still have agency anyways, and that agency in and of itself gives us some level of moral responsibility, in a practical, societal sense.
So, having faced the Predictor and lived to tell the tale, I’ve come out more on Chiang’s side than Sapolsky’s: the illusion of free will, if it is an illusion, is a helpful one, and moreover I don’t think our brains are wired to ever be able to reject it entirely. Even Sapolsky admits that he rarely succeeds at holding the idea in his head consistently. But his notion that we should try and understand the complicated, multifaceted reasons why everyone is the way they are? To understand that we are simply the sum of our parts and our past, that the line between you and someone else is much thinner than you think, and that we should therefore treat each other with a little more forgiveness, a little more understanding, a little more grace?
I’ve got no choice but to agree with that.