Whenever I mention the title to people, they ask if it’s non-fiction. It sounds like a self-help book. And one of the things I love about this book is that it blurs the line between fact and fiction. Non-fiction authors (think: Malcolm Gladwell, Charles Duhigg, Adam Grant) have gotten increasingly savvy about presenting their arguments in the guise of a story. The natural extension of this trend—the only way for academic arguments to become more storified—is to write a novel that incorporates scientific research. That’s pretty much what Hill has done.

But the book isn’t just about clever presentation of ideas; it’s also a great read. The plot centers on a married couple, Jack and Elizabeth, living in Chicago. Jack is an artist (though he’s not quite achieved the success everyone once believed he would) who now teaches at a local university. Elizabeth is a psychologist. She runs a clinic—called Wellness—which was spun out of a research lab in which she worked. The story charts their relationship from the intense romance of falling in love, when their entire future was ahead of them, to the chronic condition of being middle-aged, trying to keep the flame alive in mature relationship, and having already failed to convert a large portion of one’s potential into actual success.

Throughout the book, the tacit question behind the story is: What does it mean to be well? Whose theory of wellness is the right one? Is ‘more well’ something we can ever really be?

And Hill’s characters really do have theories of wellness (either explicitly rendered or implicitly lived). For example, the treatments offered by Elizabeth’s clinic are based on what she calls the “meaning effect”—a kind of rebranding of the placebo effect. Each character is exploring, in their own way, what story they need to tell to make sense of the life they are living. It’s a book for an age that offers a lot of solutions, without grappling with the deeper questions they’re meant to answer.

Subscribe now

[I asked Claude to prompt me with a series of questions about the book. The AI’s questions have been lightly edited; the answers are my own.]

From your description, both Jack and Elizabeth are involved in forms of meaning-making - him through art, her through psychology. How do their professional approaches to creating meaning interact or conflict in the story?​​​​​​​​​​​​​​​​

Both Jack and Elizabeth are interested in exploring how far we can stretch storytelling. How much of a good story is reality? And how much of it is just the way you present it?

Elizabeth’s exploration is much more direct, through her clinic. The premise of the clinic is that we’ve been misinterpreting the placebo effect. The way it is normally framed, placebos are meant to show whether the effects of an experimental drug are real, or whether they’re due only to psychological factors. For example, if I’m testing a new pill, I would design a trial where I give half the people the real pill and half the people a fake sugar pill. No one knows who gets the real one versus the fake one (even the experimenter). Often times what happens is that both groups show improvement, simply due to the benefits of the patient’s belief that they are receiving a cutting-edge treatment. The real pill out must outperform the placebo control, otherwise the only effect is due only to the story patients are telling themselves about treatment.

But, according to Elizabeth, that framing misses the point. The point is that patients can tell themselves a story and physically heal themselves as a result. This is the “meaning effect”: it is a treatment that, as Elizabeth explains, is “elicited by a the strong sense of significance and substance surrounding the placebo itself: the context, story, ritual, metaphor, and beliefs associated with the placebo.” The placebo effect is, in other words, “the brain’s response to finding meaning.”

And so Elizabeth’s Wellness clinic is designed around telling the most effective story possible about the meaning of the treatment, while providing patients only with sugar pills. The effect depends on getting the details right—pills of a certain size, weight, and color; adjusting the branding and presentation of the clinic depending on the alleged treatment being offered (e.g., natural vs clinical); and a compelling lie about what biological mechanisms the treatment targets. It is narrative as treatment.

Jack’s interest in storytelling is more abstract. He’s a photographer. But his photographs are of… well, he doesn’t actually take photographs. Rather, he creates abstract images by manipulating the photographic development process. It’s an artistic counterpoint to Elizabeth’s take on the placebo effect. Jack’s thesis is less compelling than Elizabeth’s, since no one buys his art. But we ultimately learn—no spoilers here—that this practice is meaningful to himself personally, for reasons having to do with his sister.​​​​​​​​​​​​​​​​

Would you say the novel takes a position on whether Elizabeth's more calculated, outcome-oriented approach to meaning-making is more or less "valid" than Jack's more personal, intrinsic one? Or does it maintain an ambiguity about this tension?​​​​​​​​​​​​​​​​

One of the things that I love about Hill’s book is that it doesn’t offer easy answers. Everyone’s argument is presented as the strongest version of itself. No one is caricatured. Everyone is very clearly right—in certain ways and at certain times. Everyone is also very clearly wrong. No one has the answers in full.

This is one of my favorite effects in literary fiction: we get to see the same event viewed from the perspective of multiple characters. Often what happens is that the first time around we’re sympathetic to the character whose point of view we’ve had privileged access to. We wonder why the other character is so staunchly committed to being a bastard. Then we see the event from the other character’s perspective. We see why they are the way they are—what it is in their past they’re reacting against that makes their actions seem like a legitimate, even necessary, way of operating in the world. We see that it’s not that they are a bastard in some intrinsic sense, but that the situation has conspired to give them no way out.

Hill’s novel does this really well. And it’s especially important when it comes to the section on Jack and Elizabeth’s relationship in middle age. It’s obviously screwed up. It’s obviously suboptimal. What can they do to get themselves out of the predicament? So often, in non-fiction, we’re sold the idea that quality decision-making can get us out of whatever life situations we don’t want to be in. But with Jack and Elizabeth, we see there is no easy way out, at least not something simple. They’re not missing something. Getting separate bedrooms isn’t going to fix things. Trying to reignite the passion of their youth won’t make them young again. They’re in a genuinely difficult situation. And the fact is, individual decision-making is not a strong enough force to affect the tsunami wall of reality.

To me, this is one of the great psychological facts which can be revealed only in literature, not experiments. In experimental psychology, the force of human-decision making is all powerful. There’s no history to contend with, no weight of meaning of an action taken. There’s only the four white walls of the sterile lab. In narratives, as in real life, our decisions are a thin strand in a vast web of more significant forces.

Let’s connect this back to Elizabeth's work at the Wellness clinic. While she's professionally invested in the idea that the right story can heal, her own life seems to resist such neat narrative solutions. This seems to create an ironic tension in the book - between her professional theory about the power of meaning-making and her personal experience of meaning's limits. How do you see that playing out in the novel’s plot?

One of Hill’s major themes in Wellness is the precarity of the story. It’s fair to say in the book that no one is actually “well.” But some people think they are.

For example, Elizabeth gets invited to join a discussion group, led by the queen mom of her neighborhood. This mom, Brandie, is able to draw on a seemingly infinite reserve of energy; she bakes the best cookies and can instantaneously get Elizabeth’s kid to stop crying, even when Elizabeth can’t. Elizabeth admires Brandie and seeks her validation. So when Brandie invites her to the discussion group, Elizabeth wants to make a good impression. What she finds is that the group is dedicated to unrelenting positivity. As Brandie explains it: “If you want your life to change, you have to believe it will change, you have to envision the change, and then it will change. Speak of things as if they’re true, and they will become true.”

One of the guys in the group claims that after joining, he and his wife got back together. Elizabeth asks him when she moved back in. The guy clarifies that his wife is still technically—he says it with air quotes—“living with Chad.” But, according to Brandie, if he just believes in the story hard enough, reality will come to reflect it.

Hill’s point is that this kind of naive positivity is obviously stupid.

Brandie’s discussion group is a kind of foil to Elizabeth’s clinic. Hill uses it to convey that relying on stories alone isn’t enough. It’s insufficient as a life strategy to just tell a story about how you want things to be then cross your fingers that reality will conform to it. The hard thing, though, is that Brandie seems to be getting results in areas where Elizabeth is not. This makes the novel’s central question more nuanced: Just how much do you have to affect reality to make your story work?

Part of Hill’s broader thesis is that we go to great pains to exist in the tiny sliver of reality in which our story actually works. This is something that our decisions can really affect (of course, until something big and unexpected happens—which is always possible). Towards the end of the book, one of the characters has an extended discourse about the Leaning Tower of Pisa. In trying to tell a story about our lives, we’re like the tourists trying to take pictures where it looks like they’re holding the tower up with their hand. It only works from one specific angle. From every other point of view you see how ridiculous it really is.​​​​​​​​​​​​​​​​

Share

The contrast between Brandie's group and Elizabeth's clinic seems particularly telling. Both are selling stories, but Elizabeth seems to understand that you need more sophisticated machinery around the story - the right pill size, the clinical aesthetic, the scientific-sounding explanations. It's not enough to just declare "speak it into existence" like Brandie's group. You need to create the conditions that make the story plausible.

One of Hill's insights seems to be that modern wellness culture often tries to skip straight to the "meaning effect" without doing the hard work of building the contextual framework that makes meaning possible. We want the benefits of a good story without investing in making it believable. What do you make of that?

It’s definitely the case that some characters’ strategies for storytelling are more effective than others. There are better and worse ways of doing it. But no one in the novel is so good at storytelling that it ameliorates the actual problems of their life. The only person who is that good at storytelling is Brandie. She’s evidently a great mom. But she’s also a fool. And eventually we learn that perhaps she’s not as great as Elizabeth first thought. I’ll let you read the book and make the judgment for yourself. Hill’s main point, I think, is that the right contextual framework can make a stronger or flimsier foundation for meaning. On the whole, it’s better to have a strong one than a flimsy one.

But Hill makes it clear that a strong contextual framework for one’s story is not an unmitigated good. For example, a strong foundation for meaning doesn’t yield Brandie’s kind of unrelenting positivity. It yields self-doubt, skepticism, and an insistence on digging into life’s toughest parts. So in this sense a “good story” doesn’t just yield benefits. It also forces the characters to confront a lot of negativity—both in their present lives and the route that got them there.

Ultimately, for Hill, a story isn’t just measured in terms of its believability, but also in terms of what he seems to think of as humanity. Holding one’s beliefs lightly and compassionately: this is clearly important to Hill. Part of the deconstruction of the Brandie illusion is how she has a death grip on her beliefs and is unwilling to give them up even in the face of overwhelming contradictory evidence. Elizabeth’s academic mentor has a great line about how if you want the god to really laugh you, then by all means—responding to Elizabeth’s insistent labelling of her and Jack’s fancy new apartment—call it your “forever home.”

But I think the line that captures it best is this: “that between ourselves and the world are a million stories, and if we don’t know which among them are true, we might as well try out those that are most humane, most generous, most beautiful, most loving.”​​​​​​​​​​​​​​​​

There's something deeply human about both wanting to believe in permanence and knowing better - and perhaps wisdom lies not in resolving that tension but in holding it with grace. Would you say the novel suggests that true "wellness" might have more to do with how we hold our stories than with what those stories actually are?​​​​​​​​​​​​​​​​

Elizabeth’s clinic does not “solve” the problem of being well. In the exposition on the theory behind the “meaning effect, Elizabeth makes it very clear that placebos are most effective on chronic illness. They usually can’t help with acute problems, like a broken arm, but often can help with longer-term issues like inexplicable back pain, “ambiguous inflammation, endless emotional anguish, spiritual hopelessness” or (possibly) being middle-aged. So there’s a distinction between what meaning can do for you in life’s long-term versus the short-term.

At any given moment, in a specific time-slice of your life overall, meaning can’t solve your problems. It can’t actually make a difficult situation easier or a sharp pain feel better. But over the broader course of living, it can help you make sense of things. It can give you a frame for explaining the past, using it to organize the present and orient yourself toward the future. In that past, you may have made bad decisions. You can’t do anything about that now. But you are still on the hook to make good decision in the present.

The reason why the effectiveness of a story is important is that it is answers the individual question of what constitutes a good outcome. As Elizabeth explains, the basis of her clinic’s interventions was that patients “created a story that explained themselves to themselves, and then they believed their made-up story was the actual objective truth.”

The reason why the humanity of a story is important is because it acknowledges the Pisa problem. Just because the story you’re telling works for you doesn’t mean it also works for everyone around you. It doesn’t guarantee that you’re making the world a better place. You aren’t actually holding up the tower. In the end, you have to remember a story isn’t reality. It is just a story.​​​​​​​​​​​​​​​​

Subscribe now

Their argument is that wherever you look in psychology, people are more sensitive to bad things than good ones. For example, a classic finding in psychology is what’s known as propinquity: the biggest predictor of whether we make friends with someone is how close we live to them. They point out that a subsequent study showed that, on average, living close to someone was more likely to turn them into an enemy than a friend.

In studies of marriages, where researchers record an interaction between a couple then see which factors predict whether or not that couple eventually divorces, the positive aspects of the interaction have almost no predictive power. The negative aspects do.

Baumeister and his colleagues surveyed versions of this effect across emotion (there are more words for negative emotions than positive ones), learning (people only need one negative example to learn not to do something—the so-called hot stove effect—and that lesson seems to never fully go away), child development (the good things parents do in raising kids seem to be a lot less impactful than the bad ones), memory (we remember bad events longer and in greater detail than good ones), stereotypes (they’re usually bad), and so on.

Baumeister and his colleagues write:

This is not to say that bad will always triumph over good, spelling doom and misery for the human race. Rather, good may prevail over bad by superior force of numbers: Many good events overcome the psychological effects of a single bad one. When equal measures of good and bad are present, however, the psychological effects of bad ones outweigh those of the good ones.

Since I finished my PhD, I let go of writing on Meaning Lab. I’ve now settled in a new job (researcher in humanities and social sciences for AI at the Alan Turing Institute in London). I’m ready to get back into writing.

This Baumeister paper felt like the right entree for a return to Substack. It has a certain resonance, I think, with our societal moment.

Subscribe now

[P.S., I’m experimenting with ways of incorporating AI into my writing process; below I asked Claude to interview me about the paper and its themes. The prompts come from the AI, but the responses are mine. I’ve edited its text, but only a bit.]

What drew you to make this connection between Baumeister’s paper and the role of negative events in storytelling?​​​​​​​​​​​​​​​​

The concept of positive valence is dramatically overvalued in psychological research. Whenever we talk about meaning in psychology, we tend to talk about positive events. We assume that meaning is like happiness, that you can just gather up all the things or experience which give you that warm fuzzy feeling and that’s the stuff of meaning in life. But when you look at the way people talk about meaning, the things they’re building on more often are negative.

For example, a classic text on this topic is Victor Frankl’s Man’s Search for Meaning. What’s that book about? Is it about Frankl’s 7 tips for success in the boardroom? No, it’s about his time in a concentration camp—the absolute epitome of human darkness. Frankl’s account centers around negative events (to put it lightly), not positive ones. If our theory of meaning only accounts for light and fluffy “human flourishing,” it seems like we’re going to miss out on a lot of stuff that really matters.​​​​​​​​​​​

Can you think of a specific negative event in your own life that, looking back, was particularly meaningful or transformative? What made it meaningful rather than just painful?​​​​​​​​​​​​​​​​

My own experiences aren’t as traumatic as Frankl’s. But I think there’s a pervasive message in the modern world, which is: negativity can be eliminated. This is the whole ethos of Silicon Valley. Any minor annoyance or suboptimality in life can be fixed by bolstering it with a superior technology. It’s all about eliminating friction. So the ideal world in the tech imagination—the one whose promise they’re selling you on at least—is one with no pain points.

But when I look at my own life, I see pain points. I see them in my relationships with friends and with my partner; I see them in my job; I see them in the million little ways the world just tries to get you done. If I’m looking at this through the Silicon Valley lens, I see a kind of failure to optimize. The implication is that if I still have those pain points, it’s simply because I haven’t done the work of finding a better solution.

Things look a lot different through the lens of meaning. From this perspective, the presence of bad things is not a negative signal. It is not a signal that you’re doing something wrong. It is not a failure to optimize or discover an elegant solution. The presence of bad things is a signal that you’re human. Bad things are part of the fabric of human experience. It’s about what you do with them, and how you reckon with them in your own internal narrative that matters.​​​​​​​​​​​​​​​​

Here’s a bit more of a speculative question: If we imagine a world where technology actually succeeded in eliminating all friction and negative experiences - a kind of “frictionless utopia” - do you think we would still be recognizably human? What would happen to our capacity for meaning-making in such a world?​​​​​​​​​​​​​​​​

It would be a world without meaning as we know it. The thing about the human capacity for meaning is that it’s incredibly flexible. And we get better at telling meaningful stories for a diversity of experiences. A couple millennia ago, you maybe had only a few broad brushes to paint with. There was tragedy, comedy, the hero’s journey—these were recognizable templates to the audiences who were engaging with these works and the authors who were composing them. The point is not that works from this era lack nuance or are without value. It’s that we’ve now found ways to tell stories—novels, plays, TV shows, comic strips—about everything from two guys waiting for someone who never shows up to a single day in the life of some Irish guy. If we ever found ourselves in the frictionless utopia of the future, there’s no doubt we’d find ways of wringing meaning from it.

That said, there’s a big difference between whether it’s possible to squeeze meaning out of an experience via literary expertise versus whether it’s easy or natural to do so. In that case, the answer would be no: we’d be more adrift than ever. One of the refrains I reach for most often is that humans are natural storytellers but not necessarily good ones. Sure, Samuel Backett or James Joyce or Han Kang can come up with these sophisticated ways of spinning stories out of nothing, but what about the rest of us?

The truth is that this is one of the problems of modernity. The stories we naturally tell are those of a previous world. They’re simple stories about good and evil, right and wrong, gods and devils, heroes and villains, knights and damsels. That’s not the world we live in. As our lives become more comfortable, it becomes harder to tell the kind of stories to imbue our experiences with a similar level of meaning or intrigue or purpose we would get when the plot was about fighting off bad guys or struggling to meet a basic need. In the modern world, there isn’t a beautiful woman trapped in the castle waiting for you to save her. There’s a thousand young professionals who have their own thing going on, and you have to tell from a 10 second snapshot of their life whether you think their thing could align to yours. That’s a much harder story to tell in a way that’s compelling and coherent.​​​​​​​​​​​​​​​​

Bringing this back to the paper, Baumeister argues that bad is stronger than good across psychological phenomena. In storytelling, do you think this “negative dominance” might actually be adaptive - like a cognitive shortcut that helps us create meaning more easily? Put another way: Is the outsize impact of negative events perhaps a feature rather than a bug of our meaning-making system?​​​​​​​​​​​​​​​​

You can definitely look at meaning in terms evolutionary adaptation. I think the story is this: meaning comes from plot. Meaning-making is your ability to say why a person or event moves a given plot forward. It’s the answer to the question ‘Why does this matter?’ in the context of a specific narrative. In humanity’s environment of evolutionary adaptiveness, the key plot line was survival and procreation. As our linguistic abilities became more sophisticated, we developed better ways of convincing others of our own centrality within the larger enterprise of the group. So the more ‘meaningful’ you could make your own contributions out to be, the more the group would value you. The better your selection of mates. The more likely you were to pass on your genes. The more likely our species was to produce individuals with a drive for meaning-making and storytelling.

Now, I’m not an evolutionary biologist. So I wouldn’t be surprised if some specific aspect of that is inconsistent with the historical record of human evolution. But the general point I think is solid: the evolutionary function of what we think of as meaning was to convince others in our group of our own crucial role in group efforts. The people who were better at that were more likely to pass on their genes.

All that said, Baumeister’s evolutionary argument is broader than meaning. It’s that what kills you isn’t the good stuff; it’s the bad stuff. Those individuals who paid more attention to stuff that could kill them had more kids. And so we ended up with a species which is, on the whole, more sensitive to bad than good.​​​​​​​​​​​​​​​​

Share

That's interesting - you’re drawing a distinction between two different evolutionary pressures: one that shaped our general negativity bias (avoiding death) and another that shaped our meaning-making abilities (social coordination and status).

Here’s a potentially provocative question: Do you think modern meaning-making has become partially decoupled from its evolutionary origins? It seems like we’ve taken these cognitive tools that evolved for survival and group cohesion and started using them in much more abstract ways - like finding meaning in art, or philosophy, or even in failure itself. Is this perhaps why modern meaning-making feels more difficult - because we're using these tools in ways they weren’t “designed” for?​​​​​​​​​​​​​​​​

There’s a difference between our cognitive tools for finding meaning and the kind of meaning we know how to find. Up until quite recently these were aligned. People may not have discussed their lives in terms of purpose or meaning or self-actualization. But they knew where they fit in the plot.

I’m an American living in London, and this is something I realized about the British class system a while back. As an American, I grew up believing (and in most instances in my adult life still believe) that it’s important for a person to aspire to be better than they currently are. In America, the default story is that you should spend your life attempting to raise your station in society. If you’re not doing that, then you’re just sort of wasting space in America—that’s the implication, at least. But in a society with more explicit class boundaries like Britain, the difference is that it’s okay to be who you are. There’s a place for you in society, even if that role isn’t necessarily the top one. There’s recognized value in that, even if the role is a minor one. You know where you fit in the plot. (Literally: a Dickens novel is essentially a stream of encounters with minor Britons.) Whereas in America, if you’re a tech worker who isn’t Elon Musk or Jeff Bezos, you really need to keep at it until you’ve reached that level. Now, I’m not endorsing classes or castes, but I do think we in America often miss how daunting that implication is: that whoever you currently are, it’s not enough.

The point is that while technology gives us more opportunity, it also burdens us. It puts the pressure for meaning-making on the individual. Historically, this is not where the pressure goes. In fact, the concept of meaning itself implies a kind of individualism. This is why the modern usage of meaning wasn’t common until relatively recently. It was something given to you by the community. It didn’t come with individual autonomy. That was incredibly limiting. But it also gave you a ready answer, rather than shoving you out the door and telling you to go look for one on your own.

As with everything that technology touches, it puts more power in the hands of the individual person. But it’s also made the effective use of that power far more burdensome as well.

Let’s return more directly to ‘ thesis.

You mentioned earlier that negative events are crucial for plot development in novels. In your view, is this literary pattern just reflecting the psychological reality Baumeister identifies - that bad is stronger than good in how we process experiences? Or is there something specific about narrative structure itself that requires negative events to function?​​​​​​​​​​​​​​​​

Tolstoy wrote that all happy families are alike, whereas unhappy families are unhappy in the different ways. He took this as all the justification he needed to talk about an unhappy family for the next 800 pages.

From a narrative perspective, contentment is boring. It’s placid. It’s an emotion of stasis. Most positive emotions, like contentment make you want to stay where you are. Most negative emotions, like anger, make you want to go somewhere new. Stories are about going somewhere new. In order to progress a narrative, someone somewhere has to want something they don’t currently have.

In that sense, bad is definitely stronger than good when it comes to moving forward the plot. But I also think in a way it doesn’t have to be, at least not the extent we intuitively assume. In a story with mass market appeal like Mission: Impossible or James Bond or a Marvel movie, the bad thing driving the plot has to be really obviously bad, like the world getting blown up. Everyone can relate to why that’s bad. But one of the things we get better at over time is to tell stories that are a lot more nuanced, that carve up the core elements of the human experience in a more subtle way or in a way that wouldn’t be obvious to everyone.

For example, in A House for Mr Biswas by V.S. Naipaul, he took the heft of a Victorian novel (where the plot is essentially: domestic life, but do it with gravitas) and set it in Trinidad. The crucial implication there is that you don’t actually need all the fancy British upper class stuff to make that kind of story feel important. That’s a really difficult thing to show when no one has given you a template for how to do it. Naipaul unlocked a whole new way of showing how you can take different kinds of experiences and take them really seriously. You can give them the same weight of meaning.

All that said, what creates narrative motion is friction. There has to be, in the nomenclature of Silicon Valley, a pain point. Is it any wonder that in our effort to use technology for the elimination of friction that we’ve made it harder than ever to grasp a sense a meaning?

Final question: You've drawn this parallel between Silicon Valley’s “frictionless” ideal and our struggle for meaning in modern life. Given everything we've discussed about negative events, friction, and meaning-making - what would you say to someone who's actively trying to optimize away all the “pain points” in their life? Is there a middle ground between accepting necessary friction and being needlessly masochistic?​​​​​​​​​​​​​​​​

For me, it’s not about completely setting aside any attempt to improve. I do think there’s a role for optimization, no doubt. The difference is we need to be more critical of how self-improvement is being sold. The underlying message is that there’s a series of decision you can make to get yourself to some sort of ideal, frictionless state, where you only ever eat healthy foods, never miss a session at the gym, and only have positive, edifying conversations with your spouse. That state doesn’t exist. There’s no set of steps you can take to get there. Friction is intrinsic to human life.

But that isn’t a bad thing. It’s in grappling with the causes of friction in our life—even when that’s something that we’d rather turn away from—that we make our own personal sense of meaning. The dark side may naturally be more powerful than the light side. But overcoming that disadvantage is what makes the light side’s story worth telling​.​​​​​​

Subscribe now

To the Land of Happiness

To land in Bhutan is to feel, mostly because you are told repeatedly, that you are part of something special. The experience of the foreign tourist is carefully and intentionally designed to convince you that this country is singular—that no matter where else you’ve been on this planet, you’ve never been anywhere quite like this.

My indoctrination to this idea came earlier this year, in January 2023, when I traveled from Tan Son Nhat International Airport in Ho Chi Minh City, Vietnam, to Paro, the only international airport in the remote Himalayan outpost of Bhutan. This is not a normal flight to take. It is not a bustling leg of the global commerce circuit. To even get from Vietnam to Bhutan, a special flight must be requisitioned. The only reason that I found myself embarking under these conditions was the invitation of a family friend whose tourist company specializes in taking Vietnamese people to obscure Central Asian destinations. Bhutan is the most highly sought, and I had just so happened to be in country when the longstanding invitation to make the pilgrimage was finally able to be accepted. And so it was that I found myself en route to Bhutan with myself, my mother, and the giddy energy of about eighty Vietnamese people anticipating an opportunity to see a country few people ever get to see. It was quite possibly the most delighted I’ve ever been.

The first sign that Bhutan is not your usual tourist destination is the flight. Not only does it have to be specially chartered, it leaves at an hour of ungodly earliness, before the morning dew has yet to settle on the grass outside the airport. Given that the flight is specially scheduled, it might seem advisable for it to take off at a more reasonable time. However, Bhutan operates according to a special logic. For the foreign tourist, it costs money simply to exist there. For every day a tourist spends in the country, they pay a fee. This fee is designed in essence to pay for other tourists not to be there. The Bhutanese government is stingy with its visas and under the auspices of their “High value, low volume” policy opens a small number of spots to outsiders every year. Prior to the pandemic, it was on the order of a couple hundred thousand visitors per year. Compare this to Nepal, a country of similar cultural and geographical makeup, which welcomes well over a million tourists each year. As part of the same policy, tourists in Bhutan must be accompanied by a tour guide. The Bhutanese government doesn’t trust outsiders to wander the country unfettered, so they dispatch a local to oversee every group of tourists in their exploration of the nation’s landmarks. This is meant both to increase the quality of the visitor’s experience, as well as make sure they don’t do any of the things that tourists so often do to screw with one’s national heritage, like discarding a Snickers wrapper on the grounds of an important historical monument.

The other thing about flying into Bhutan is that only Bhutanese people are allowed to do it. The only airlines allowed to fly in and out of the country are operated by the Bhutanese government. The international airport at Paro is not a normal airport. It requires a pilot to maneuver swooping in and out of mountainous terrain at an altitude so low a passenger could reach out the window and brush the treetops below. Not only is run-up treacherous, but the runway is short. It is one of the most dangerous runways in the world. Landing at it requires a set of skills that would almost certainly qualify someone to be a stunt pilot on the set of Top Gun 2. At some point in the flight, the pilot addresses the passengers that the Himalayan range, including Mount Everest, is viewable from the left window, at which point the forty people not on that side of the plane scramble over to get a view. Boarding the plane in Vietnam, the temperature was 31 degrees centigrade (88°F). Landing in Bhutan, it was a decidedly more brisk -1°C.

The air in Bhutan is thin but highly oxygenated. Paro stands at 2,400 meters above sea level, about a kilometer higher than Nepal’s capital, Kathmandu. Yet somewhere between 50% and 75% of the country’s entire area is forested. But once one acclimatizes to this novel sensation of respiration, there’s an even more striking observation to be made. There are no other planes on the runway. Not a single one. Whereas most airports carefully ferry passengers to and from the tarmac in front of an airplane, or dispense them directly from the terminal to the fuselage, freshly disembarked tourists in Bhutan are allowed to sort of wander off the plane, snap some photos, and have a look around. They can’t wander very far. It’s not like there are any other groups for them to get mixed up with. Nor will they be reduced to a smudge on the runway by an incoming plane. There’s only one building nearby, a monastery-looking structure where the terminal should be. The structure is a single white block, pristine as a new stack of printer paper, with uniform rows of dark red windows. In the center is the unmistakably Himalayan pattern of repetitive, ornate design. The building is painstakingly well-kept, in a way that gives it an air of spiritual significance (a visitor to Bhutan quickly learns that this is the preferred architectural style for any building of national importance, monastic chic). Eventually, the group of new arrivals has nothing else to do but wander toward it.

The current of excitement running through that mountain air is palpable. There’s an immediate sense of solidarity in collective anticipation of what’s about to happen. You are about to do something few people get to do. You are about to enter Bhutan.

Subscribe now

Dzong at the Top of the World

To say that the country of Bhutan is remote is to engage in a kind of droll understatement. As far back as the country’s historical record stretches, Bhutan has never been colonized or conquered by an outside power. This is a rather more significant distinction than it might at first seem, as the British had the opportunity to do so in the heyday of their empire. Bhutan hovers like a balloon above two of Britain’s former colonial strongholds, India and Myanmar (colonial alias: Burma). Bhutan’s entire southern border runs along the part of India wedged above Bangladesh. One way of reading Bhutan’s history during the colonial period is that the British, having firmly established themselves to the south in India, briefly considered proceeding north into Bhutan, then decided it wasn’t worth the trouble. They couldn’t be bothered to climb the requisite hills.

Not only is Bhutan isolated from the rest of the world but the country also sought to separate itself from its neighbors. For instance, Bhutan and Nepal are essentially brother states with fundamentally different theories about how those states should be run. They are similar in relative size (at least to their behemoth neighbors, China and India), geography, and economy, but whereas Nepal has had a policy of coaxing as many foreigners into its borders as possible Bhutan has played its hand a little closer to the vest. Bhutan’s policy concerning tourism has largely been to watch Nepal make a decision, see how it plays out over twenty or so years, then consider how best to implement a comparable policy based on what went wrong in Nepal’s experiment. For example, everyone knows that Mountain Everest is the world’s tallest mountain and most people will be familiar with its south base camp, in Nepal, which is the. most popular way to approach it. But most people won’t be aware that the world’s tallest unclimbed mountain is Gangkhar Puensum in Bhutan. The reason it is unclimbed is not that it is unclimbable, but that Bhutan doesn’t want people on it. In 1994, the Bhutanese government declared that the mountain was sacred—a common theme in the Bhutanese relationship to the land—and prohibited outsiders from clambering over it for recreational purposes.

Since its formal incorporation as a nation-state in 1907, Bhutan has prioritized cultural preservation to a degree that is likely unmatched by any other country on the planet. Bhutan has consistently chosen to make less money from tourism, discouraged its citizens from courting outside influence, and supported its cultural traditions and institutions in a way that from a Western perspective seems almost backward. As a result of its isolationist policies, Bhutan has maintained what makes the country and its people special in a way that the rest of the globalized world has not. One symbol of this effort is the Bhutanese national obsession with the dzong—the fortified monastery that characterizes all major buildings from the airport terminal to the country’s largest and most prominent sites of government and religion. The Bhutanese call their language Dzongkha, literally: the language of the fortress. And it is this sense of fortification that pervades the way they insulate themselves from the outside world.

One of the reasons that Bhutan has been able to shun globalization and still maintain its status as a sovereign nation is its dependence on India. Bhutan is a net importer, with the majority of its food (such as rice) coming across the Indian border. Its main source of income, likewise, is selling hydroelectric power to the Indian government. This exporting of energy is actually a foundation stone of Bhutan’s economy. Bhutan is the only nation in South Asia that has a surplus of energy available for export, and so India, densely populated as it is, has few options to turn to if it wants to supplement its own energy resources. Hydroelectric power constitutes about 20% of Bhutan’s economy, and almost 50% of its national revenue comes from taxes imposed on hydroelectric companies. While the Bhutanese people as a whole aren’t especially rich, the country’s monarchy is actually rather wealthy. Think: hydroelectric energy is to Bhutan as oil is to Saudi Arabia.

It is worth noting that whereas usually this kind of wealth inequality—a rich aristocracy ruling over a population consisting mostly of peasant farmers—would stir resentment, the Bhutanese people have an immensely positive disposition toward their king. On the whole, they were rather reluctant to engage in democratic processes when they were first implemented in Bhutan in 2008, mostly because they would take power away from the beloved monarchy. This isn’t a Uzbekistani approval-is-the-correct-answer public opinion pole, either; the positivity isn’t an artifact of a hereditary king coercing his population into a forced smile for the rest of the world to see. For the past century, Bhutan has been blessed with a uniquely good succession of five kings. Perhaps uniquely good in all of modern history.

In short, Bhutan in its modern history has only had to deal with one country, India—and by extension of India’s history, Britain—and so despite its strong sense of cultural identity, the nation remains a relative unknown to outsiders. There is no Starbucks in Bhutan, no Pret a Manger. The King, the Clown, and the Colonel are nowhere to be found here. Bhutan maintains an embassy in only six countries, and only three countries—India, Bangladesh, and Kuwait—maintain embassies in Bhutan. Its national language, Dzongkha, is fortress-speak. Spoken by only half a million people, it isn’t even featured on Google Translate. Yet no app is necessary to see that the theme of the fortress runs through every aspect of the nation’s affairs: dzongkhag (the local municipalities), dzongda (the district attorneys), dzongdag (royal appointees to each municipality). Bhutan, in every sense, is the fortress at the peak of the world.

Share

The Log Cabin of Wangchuck

After arriving at the airport, and making the acquaintance of our government-approved tour guides, our group was shuttled into Paro—Bhutan’s second fastest-growing city, after its capital, Thimphu—for lunch. The central business district of Paro, if you want to call it that, was essentially a single street. The structures are mostly wooden. Though the city’s population is growing, there isn’t any obvious presence of construction downtown. The street is lined with shops, the kind tourists might like to shop at, if there were in fact any tourists. As far as I could tell, it was just us.

We were escorted to a restaurant, what seemed to be the only game in town for serving food to a large-ish group of foreigners. The restaurant featured two wooden stories and had the bright piney finish of a cabin in rural America. On the bottom floor was a Western-style coffee shop, complete with an espresso machine and pastry case. The restaurant was on the top floor and stylistically would have been at home and unremarkable as a fixture in any anonymous small town on the globe. The menu was filled mostly with food customarily suited to a generic cosmopolitan palate—a stereotypical if somewhat incoherent mix of Western specialties: pizza, burgers, various Panda Express-style Chinese dishes, spaghetti, and tomato soup.

This being my first meal in Bhutan, I was eager to try a local specialty. I lodged this request with the waitress, and she directed me to the appropriate page of the menu. I had never heard of any of it. The English descriptions weren’t especially illustrative. I asked the young woman what she recommended. She asked if I liked spicy. I told her yes. It is a matter of identity to me to be the kind of white tourist whose tolerance for spicy food becomes a subject of local lore, ideally passed down from generation to generation like a family heirloom. She recommended ema datshi, a kind of curry with cheese she told me.

“Cheese like paneer?” I asked her.

“No, not really,” she said.

She asked how spicy I wanted it. I told her to make it the way she’d normally eat it herself. The rest of the table opted to order an eclectic set of more familiar dishes.

Ema datshi is one of the main national dishes of Bhutan. In Dzongkha, ema means chili and datshi means cheese, most commonly from a yak. This is a rather euphemistic label for what the dish really is. To get a sense of it, imagine a standard-issue Thai green curry. Now, replace the coconut milk base with a kind of thin cottage cheese. It’s not identifiably cheese, per se, but rather a dairy-based template for the curry. So far, so good. Supposing it’s a basic Western version of Thai curry you’ve got in your mind’s eye, let’s say it includes the following: bamboo shoots, carrots, onions, bell peppers, and chicken. For ema datshi, replace the bamboo shoots with chilis. Spicy, skinny, eyebrow-scorching chilis. They look, like a kind of evolutionarily adapted subterfuge, like green beans. Then replace the carrots with chilis. After this replace the bell pepper and onions with chilis. Finally, substitute the chicken for, you guessed it—chilis. With these substitutes, you have pretty well grasped the concept behind the dish. The most appropriate English translation would be something along the lines of “Satan’s cheese curry.” And you know what? All things considered, it was pretty tasty. I think even now, my stomach lining has, at the time of this writing, mostly regenerated. The body is an amazing thing.

After lunch, we were shuttled off to a history museum. Specifically the National Museum of Bhutan, a tall, cylindrical building in the traditional style built into the side of a steep hill. Just outside the city of Paro, it was originally built as a dzong, erected in 1651, and repurposed in 1968 to house a selection of the nation’s culturally important artifacts. The museum’s floor plan was laid out rather in the manner of an Ikea, with a single one-way lane for pedestrian traffic winding its way through the many items on display over the building’s seven floors. The majority of the public collection featured items of anthropological interest: elaborately designed wicker baskets, ceremonial clothing, assorted weapons, agricultural implements made of iron, religious symbols, rows of large urns, used in centuries past for holding wine or the remains of ancestors or, perhaps, both. There was a lot to take in.

The trek through the museum was lightly narrated by our guides. For our sub-group of six people, two guides were assigned. The tall one, named Tenzin, was the charismatic, wide-smiling historical authority and leader of our group. He was accompanied by Pulpo, who acted as a kind of secret service agent. He was stocky, perpetually clad in black aviators (even indoors), and not given to offering verbal communication unless addressed directly. The exceptions to his taciturn disposition came mainly when he identified an opening to deploy a one-liner from the stock he’d accumulated over years of working with tourists. He had feel for perfectly timed mischief. At a later opportunity in the trip, while on a long hike, I asked Tenzin what he knew about Pulpo. He reported that Puplo had played some role in Bhutan’s military. Other than that, he knew nothing. Later on the same hike, I floated the question of background casually in Pulpo’s direction. The only thing he’d say was that he preferred his current job to his old one.

The nation of Bhutan, Tenzin told us, was unified in the 17th century by a monk named Ngawang Namgyal. It was this monk, whom Tenzin referred to by the title Zhabdrung, which gave Bhutan its endonym, Drukyal: land of the thunder dragon. It was also Zhabdrung who instilled in Bhutan’s people and landscape a flair for the high-walled, white fortress. These early Dzongs were designed to keep out Tibetan invaders, the very same people Zhabdrung had been fleeing when he settled in Bhutan. Tenzin told us a story concerning this founding father, which I wrote down at the time but had a hard time confirming in third-party sources. Allegedly, on his way out of Tibet, Zhabdrung had used his escape as an opportunity to lift some sacred artifacts. This perhaps goes some way toward explaining why the Tibetans insisted on invading the nascent Bhutanese settlement. At length, emerging from his newly erected dzong, Zhabdrung presented himself to the Tibetan invaders. In one hand, he held the sacred artifact they sought. In the other, an orange. Zhabdrung made as if to insert the artifact in the orange. With a great heave, he chucked the orange into the nearby river. Such was the depth and rapidity of the river, that the orange—and thus the artifact—was now irretrievably lost to eternity. Evidently, the Tibetans were taken in by this act of prestidigitation and with a collective shrug shuffled on home.

By far the most arresting exhibition at the museum was a row of portraits and significant objects delineating the Bhutan’s five successive kings. The Bhutanese head of state is known as Druk Gyalpo, or the Dragon King. As I mentioned, the Bhutanese monarch has produced a line of the kind of wise and gracious rulers who would make Plato stand back and take note. It seems rather unlikely that an unbroken line of five kings is each a more level-headed and selfless leader than the previous one. But that has been the case in Bhutan for the past century with the House of Wangchuck.

Each of Bhutan’s kings has his own signature accomplishment, marking off his epoch in the country’s history. The first king, Ugyen Wangchuck, is credited with founding the modern Bhutanese state. In November 1908, he was unanimously elected to head the government by a collection of worthy elites—clergy, aristocratic families, local officials—who recognized his hereditary claim to the Bhutanese thrown. His signal achievement was the signing of the Treaty of Punakha, in 1910, which codified Bhutan’s independent status from India and colonial Britain. In return for the guarantee of independence, as well as some amount of financial backing, Bhutan agreed to cede all control of its foreign relations to the British. In the style of many national founders, Ugyan Wangchuck was considered by his contemporaries to be a master statesman.

The second king, Jigme Wangchuck, built on his father’s legacy. Taking over after the first king’s death in 1926, the second Druk Gyalpo continued his father’s isolationist policies for thirty years. During this time he made two contributions which would pay dividends for his country into this century: the Treaty of Friendship with India (providing the country with enough economic autonomy to ignore the rest of the world) and to begin building modern—well, relatively modern—infrastructure.

Upon his death, the third Druk Gyalpo ascended the throne at age 16, having already studied abroad in India and the UK. It was the third king who broke from his predecessor’s strictly isolationist policies and began to open Bhutan to the outside world. Legal codes were reformed. Legislative procedures were modernized. For the first time, foreigners were allowed to enter. Then, after twenty years at the helm, the third king died in 1972. The throne was passed to his son, who was also only 16 years of age. It was the 4th Druk Gyalpo who instituted the policies of social and sustainable development for which Bhutan is famous. He was the one who came up with the idea of Gross National Happiness. Bhutan is the world’s first carbon-negative country largely because of reforms which he set in motion.

The fourth king, perhaps being surprised at his own ability to live past his son’s 16th birthday, shocked the country in 2005 by announcing his intent to abdicate the throne. His son, Jigme Khesar Namgyel Wangchuck, assumed the throne in 2006, though the public coronation did not take place until November 2008—to mark the 100th anniversary of Bhutan’s monarchy.

The fifth king is by far the most modern of the Druk Gyalpo lineage. His first act as king was to democratize the country. He oversaw the drafting of the nation’s constitution and traveled throughout the country in order to explain it in person to his constituents. He is known for his encouragement of young people to seek education and to set higher standards for themselves in business and civil service. In 2021, he decriminalized same-sex marriage in Bhutan. While this may seem somewhat late to the game, there’s no rule that says small, autonomous, highly religious nation-states have to adopt this policy at all.

Perhaps most importantly, he was the first king of Bhutan whose royal wedding was an international media event. It helps, surely, that King Jigme has the suave, rounded looks of a young Elvis Presley. Upon the announcement of engagement to his selected Dragon Queen, King Jigme put out a press release stating: “Now, many will have their own idea of what a Queen should be like—that she should be uniquely beautiful, intelligent and graceful... For the Queen, what is most important is that at all times, as an individual she must be a good human being, and as Queen, she must be unwavering in her commitment to serve the People and Country. As my queen, I have found such a person and her name is Jetsun Pema.” Their son, Bhutan’s heir apparent, was born in February of 2016. Pictures of the current royal family can be seen in nearly every room—public, private, religious, commercial, significant, mundane—around the country.

One fact about the current king which I found particularly compelling was his educational background. After beginning his education in Bhutan, he enrolled at Phillips Academy in Massachusetts, Wheaton College (also in Massachusetts), and then the Diplomatic Studies Programme at Magdalen College, Oxford. I wonder what his peers in rural Massachusetts made of him. Can’t you just imagine it? “So Jigme, remind me. What did you say you were going to do after graduation?” Oh, you know. Same as everyone else I suppose. A bit of this, bit of that. Head back home, move in with my parents, take over the family business, get married, bring democracy to my people. You know. The usual.

Leave a comment

Gross National Happiness: Not necessarily a measure of happiness

The concept of Gross National Happiness (GNH), instituted in 1972 by Jigme Singye Wangchuck, the fourth Dragon King of Bhutan, was an attempt to recognize that a country is more than its economy. What is notable about the idea is not that it measures happiness per se—there’s a lot more to it than that—but that it is something besides GDP to gauge the nation’s trajectory.

In recent years, the rationale behind Gross Domestic Product’s hold on the imaginations of governments around the world has come into question. For example, in 2015 the Duke University economic historian Dirk Philipsen published a book called The Big Little Number, a critique of how GDP has shaped the global economy. In an interview on the book, he said that GDP is “the only measure that has ever existed in history, I think, that has been followed by regimes around the world, and that people from the right to the left subscribe to. It is what scholars call a ‘truly hegemonic’ metric. That’s amazing, because it is a metric that essentially measures the well-being of capital—not of people, and not of the environment.”

GDP is meant to measure the total value of goods and services produced by a country. But it leaves a lot out. In terms of capital, that includes unpaid work, the trading or purchase of used goods, off-the-books transactions, and large swathes of informal economic activity (which potentially includes more than 50% of workers globally). Given that the point of a healthy economy is to improve the lives of its denizens, the use of GDP as a hegemonic metric assumes that human well-being can be assessed in entirely material terms. It is this assumption that the country of Bhutan rejects.

At the center of Bhutan’s vision for GNH is sustainable development. The country’s economic growth should first and foremost support the enrichment of its environment and culture, in addition to the psychological well-being of its citizens. As such, direct measures of happiness actually make up only a small portion of the overall GNH figure. The final number reflects nine categories, which can be seen in the graphic below (taken from Bhutan’s official GNH website). Each of these categories—such as education, health, community vitality, psychological well-being, and ecological diversity—is given equal weight. Within each of those categories are yet finer distinctions, which are given greater or lesser weight within the larger category. For example, the category of Cultural Diversity and Resilience is divided into ten points. Of those ten points, two are given to speaking the native language, two are given to adhering to a code of proper etiquette known as driglam namzha, three are given to “cultural participation,” and three are given to the cultivation of artisan skills (called zorig chusum). Another example is “Time Use.” This category is given the same weight as the other eight, but it is divided into just two equal parts: work and sleep. (Can you imagine, if 5% of the governing metric of the US were based on whether its citizens got a good night’s sleep?) In this way, the country’s overall GNH figure is comprised of 33 individual metrics which in some way or another contribute to the overall well-being of Bhutanese society and its citizens.

The core vision of Bhutan’s measure of GNH is sustainability not just in the sense of positive environmental impact, but in the sense of being far, far less likely to do something that will have a long-term detrimental effect on the country’s citizens in the name of stimulating the economy. When a country relies on GDP alone to measure its national success, all sorts of nuances can be overlooked—not the least of which is the health of its people and environment. Bhutan was able to become the world’s first carbon-neutral country not just because it has a lot of trees, but because it values its forests as more than just the dollar amount its timber would fetch on the international market. To the Bhutanese, their forests have intrinsic environmental value, as well as spiritual value. Many of the forests in Bhutan, such as a particularly vast one called Nyalalum, are believed to be inhabited by evil spirits. No economic activity of any kind, including building a residence anywhere nearby, happens on any appreciable scale. While this sentiment does not factor directly into the GNH calculation, it is representative of the fact that the leaders of Bhutan govern in a way which reflects this multifaceted, highly nuanced way of assessing value.

Over time, Bhutan’s GNH index has gone up—implying that the country is happier, on the path to truly sustainable development. In 2010, the country’s overall GNH score was 0.743 (on a 0-1 range). In 2022, that number increased by almost four points to 0.781. Another way Bhutan draws a verdict from the GNH figure is to assign a happiness label to every individual citizen. For those who are considered Happy, they can either be Deeply Happy (77% or above) or Extensively Happy (66%-76%). Bhutan’s goal is to get every citizen to a 66% score or above on the index. Those who are below this figure are deemed “Not-Yet-Happy.” Citizens between 50% and 65% are considered Narrowly Happy, and those scoring less than 50% are labeled as “Unhappy.” In 2015, the overall Happiness ratio was 43% Happy to 57% Not-Yet-Happy. That figure tilted slightly in favor of the population urban, with 55% of those living in cities surpassing the ideal happiness threshold while the same was true for only 38% of the rural population.

By 2022, the overall number of people above the 66% happiness threshold in the country had risen to 48.1%. This breaks down into approximately 10% of the country being Deeply Happy, 39% Extensively Happy, and 46% Narrowly Happy, leaving only 6% of the country below the 50% mark of the happiness index threshold.

From a psychological perspective, what Bhutan’s GNH index gets right is that an individual’s overall happiness is not a simple function of self-reported happiness alone. In the West, our primary means of measuring happiness is to ask people whether or not they feel happy (Daniel Gilbert gave a strong defense of this methodological approach, based on his decades of research, in his 2006 book, Stumbling on Happiness). However, one well-documented feature of happiness—both in empirical psychological research, and long before it—is that the more one asks the question “But am I really happy?” the less one tends to think the answer is yes. As John Stuart Mill once wrote: “Ask yourself whether you are happy, and you cease to be so. The only chance is for you to have as your purpose in life not happiness but something external to it.” A while back, I wrote a piece arguing that the only way for an individual to maximize their happiness is to maximize for the root causes of happiness—particularly, as Mill suggested, a purpose greater than oneself. This is the same kind of core belief underlying Bhutan’s happiness philosophy.

The crucial innovation in Bhutan’s happiness index is that it defines happiness externally to an individual. Whether or not an individual claims to be happy is considered, but only as a single aspect of a larger story. If the individual: is a member of a community with a thriving cultural tradition and an increasing material standard of living, is governed by leadership who takes their best interest to heart, is able to get enough sleep every night, has a job which they enjoy but does not claim the entirety of their time, invests in their education, holds themselves to a standard of appropriate community conduct, invests in their own spiritual and religious enrichment, and lives in a clean, beautiful, and well-maintained area—then it is assumed that that person will be, on the whole, happy. It’s hard to argue with that.

My interactions with my Bhutanese guides and other locals seemed to support this conception of happiness. They are undoubtedly proud to be a member of a nation with GNH at the core of its development goals. But they spoke of happiness more as an abstract concept, more like a religious identity such as Buddhism, rather than constantly trying to convince outsiders that everyone in Bhutan smiles all the time no matter what the circumstances might be. Feeling happy and telling other people about it is a small part of the Bhutanese national theory of happiness. Both in the way it is measured in GNH, and in the way it is expressed by its citizens. I liked this a lot.

One of our guides, Tenzin, reported that since the pandemic, the country’s conception of GNH has started to shift. He described it as a “reality check.” Covid put pressure on the entirety of the global economy. But places like Bhutan were especially vulnerable. While the country is modest in its economic ambitions, it nonetheless leans heavily on a few key corridors of trade. For example, selling power to India and in return importing crucial staple crops such as rice. Also, tourism. Even though Bhutan doesn’t bring that many people in, it still relies on the influx of foreign money, particularly the high per diem tax for tourists to enter the country. Any disruption to these can cause a major fissure in the foundations of national happiness. This is another thing Bhutan is beginning to realize about psychological well-being: economic precarity does not make a good basis for individual happiness.

Happiness is a function of expectations. This is what it means for happiness to be relative: two individuals might have the same material standard of living but feel very differently about it depending on their expectations about what that level is supposed to be. Emerging from centuries of isolation, the expectations of Bhutanese citizens are changing. They’re engaging more with the outside world; more of them have studied abroad. They’re reluctant to continue traditional means of employment, such as yak herding. They see what everyone beyond the borders of their country has in terms of material possessions, and many of them are wondering—very reasonably—whether the maintenance of their nation’s cultural institutions is really worth trading for the allures of modernity. How far can the Bhutanese push the project of modernizing their country while still preserving their treasured national identity?

Whatever the answer is, we’re about to find out. Bhutan recently lowered its tourist tax. That means that it’s going to be increasingly possible for people like you and me—outsiders—to get our hands on the Bhutanese brand of happiness. It’s a turning point for a country that has for a long time built more fences than bridges. They’re eager to bring more people into the country, and with them foreign money—a move that will certainly have a positive impact on their GDP. Whether the same will be true of GNH, I guess we’ll just have to wait and see.

Subscribe now

Share

A brief history of ideas leading to the modern computer

The history of the modern computer begins with mathematical logic. This was the discipline, booming as it was in the mid-19th century, which first took seriously the idea that human reason could be distilled into a non-human mechanism. The flagship work in this line of inquiry was George Boole’s 1854 treatise, The Laws of Thought. Though Boole’s manuscript dealt mainly with mathematical formalisms, his main goal was, as he wrote, “to investigate the fundamental laws of those operations of the mind by which reasoning is performed.”

The idea that the processes of human reasoning could be codified in math was in the air in Boole’s day, and we remember Boole largely because he was the first logician to bring his ideas to market in a useful form. His contemporary was a man named Augustus De Morgan. Boole and De Morgan rivaled one another in much the same way as Newton and Gottfried Leibniz, or Darwin and Alfred Wallace. In the independent discovery of calculus, Newton’s and Leibniz’s work served different ends. Newton understood what calculus could do; he was essentially an applied mathematician. But Leibniz understood the primacy of notation (the use of the term dx to denote a derivative was his, not Newton’s, invention). The work produced by Boole and De Morgan featured a similar tradeoff, though in the case of logic, it was Boole’s superior notation that won out. Their rivalry was representative of an abiding interest in mathematical logic that carried on deep into the later years of that century and the beginning of the next.

The culmination of this tradition was Russell and Whitehead’s Principia Mathematica. The result of ten years of diligent mathematical labor, Russell and Whitehead’s work was meant to present the world with a logical system to end all logical systems. Russell and Whitehead were trying to solve a specific problem, one which plagued the previous work in logic. The preceding work, such as Boole’s and De Morgan’s, was based on assumptions. If you took certain things for granted, their systems worked. But why take those things for granted? How do you know they’re really true? Answering this question was the holy grail of mathematical logic, and in their desperation to sip from the blessed chalice spent seven hundred pages attempting to prove that 1+1 does indeed equal 2. Their goal was to create an axiomless mathematical foundation for logic, an instantiation of reason in its most unassailable form, where nothing needed to be taken for granted. They failed to achieve it.

Ultimately, Russell and Whitehead’s system had elements that were either unprovable or inconsistent. If you made everything provable, inconsistencies arose. If you made everything consistent, it required assumptions. For example, you can make a mathematical foundation for logic using set theory. But then you’re subject to Russell’s paradox. Russell and Whitehead never found a way to have it all.

Then Kurt Gödel came along, with his famous incompleteness theorem. Gödel showed that it wasn’t Russell and Whitehead’s fault. Not only was the provability-consistency tradeoff an ineradicable part of their system; it was an ineradicable part of any system. The kind of logical system Russell and Whitehead wanted to create wasn’t possible. That promise of completeness could never be achieved.

Gödel’s theorem broke the great chain of achievements in the tradition of logic stretching from Boole to Russell. As a consequence, the attention of the best and brightest mathematicians (which is what Boole and De Morgan and Russell and Gödel undoubtedly were in their day) was diverted to working on other problems.

Enter Alan Turing and Alonzo Church—yet another intellectual duopoly—who independently developed what is known as the Church-Turing thesis. I’m not a logician by trade so the nuance of it is lost on me. But in essence it says to hell with provability. All that matters is consistency. Specifically, what matters is that you can create a system where you have a guaranteed method for determining an effective solution to any calculation you might plug in to your system. This project, as Turing and Church, showed was possible. It turned logic from a metaphysical problem into a practical one.

The most famous iteration of this insight is what we now know as the Turing Machine. This was not actually a machine but an abstract mathematical model. Benedict Cumberbatch movies aside, an actual Turing Machine as such was never built. It had, in theory, an infinite tape on which could be printed symbols (such as ones and zeros), which could in turn be read or rewritten by the machine. A Turing Machine was “complete” in the sense that Turing’s theoretical machine was capable of simulating, given enough time and space, any other Turing Machine. That is, it was a mathematical blueprint for hardware that could run programmable software. It approached Russell and Whitehead’s goal of the ultimate logical system—but from an entirely different angle. Unlike Russell and Whitehead, Turing succeeded.

The Turing Machine became the theoretical basis for the modern computer. It is the central idea behind the CPU. But Turing didn’t actually construct the machine himself. It was a Hungarian mathematician, named John von Neumann, who finally gave form to Turing’s function. Turing described a computer. Von Neumann actually built one. And it is von Neumann, as one of the 20th century’s most central intellectuals, who is the subject of an imaginative work which straddles the line between historical fiction and intellectual history: Benjamín Labatut’s latest novel, The Maniac.

Subscribe now

Labatut’s remarkable formal innovations

Labatut’s novel is about the man behind the machine. More specifically, it is about the men and women behind that man, and how for the most part they found him exasperating. The project of Labatut’s novel is to draw a line between von Neumann and modern AI’s achievement of AlphaGo, specifically the victory of Google DeepMind’s system over the generationally talented South Korean Go champion, Lee Sedol. It is a how-the-sausage-is-made story, centering on the development of modern computers around the time of the Second World War.

Like the great logicians of the past, Labatut’s innovation is one of formalism. It is a way of constructing a novel that no one—no one whose reviews I’ve read anyway—has quite seen before. The novel is constructed as a series of texts, dozens of them: some formal essays, some inner monologues, some journal entries. Each one is authored by a different person who was connected to von Neumann in some way, personally or professionally. The book draws on promises inherent to both fiction (no one actually said these things exactly as they are written) and non-fiction (the sentiments expressed and the experiences transcribed belonged to real people). It makes the book hard to place on the shelf with the standard divisions of genre.

This aspect of the book is dealt with in different ways by those who have reviewed it. The Guardian labels the book “semi-fictional.” Becca Rothfeld, in The Washington Post, poses the question: “Is The MANIAC a work of fiction? Or do we call it fiction because we lack a better word for its creative conquest of fact?” And Tom McCarthy, writing for the New York Times, makes the claim that although the book does “assume the guise of fiction,” it could plausibly be recategorized with a “(minor and essentially rhetorical) tweak into long-form journalism”. A big reason to read the book, and to be excited about it, is for the intriguing novelty of Labatut’s formal innovations.

So let’s take for granted that Labatut achieves structural excellence in his novel. What does that formalism reveal about the novel’s subject?

One less charitable review summarizes the book, not entirely inaccurately, as “a repetitious, linear account of von Neumann’s life from the point of view of his acquaintances.” Labatut presents us with a novel of a hundred characters (none of whom, by the way, is actually the alleged protagonist, John von Neumann, who despite his prominent place on stage was cast in a non-speaking role). However, the characters don’t recur. They enter stage right, waiting in a single-file queue just off stage, say their bit, then exit stage left never to be heard from, or often even referenced, ever again. It has the effect of keeping the story moving. But it also makes it hard to have any sort of cumulative pay-off in the course of that progress.

Labatut’s writing itself is excellent. But almost too excellent. He tasked himself with differentiating the voices of dozens of 20th century mathematicians, the vast majority of whom come from the eastern border of central Europe. The writing is accomplished, but not in a way that seems representative of what 20th century mathematicians actually sounded like. Labatut gives them an ability of articulation which bursts the bubble of realism.

Another issue is that the stylistic differences between voices are somewhat muddled. The attempts to establish stylistic markers to differentiate speakers are a bit clumsy. For example, most characters—it felt like a large majority—write in looping, polysyndetic sentences. Others talk in clipped, staccato sentences in which the subject is habitually omitted. In the book’s first passage, the writing comes across as a work of immense virtuosity, rendering the mind of the troubled intellectual (though not von Neumann’s) in a fresh and profound way. But it loses its edge when the same idiosyncratic flourishes are used to articulate the perspectives of so many different people. (That being said, my current standard for an author’s ability to render multiple voices is Hernan Diaz’s Pulitzer-winning Trust. So it’s possible I’m being overly harsh.)

The Guardian’s reviewer, Sam Byers, poses the slightly uncomfortable question at the end of this line of criticism: “All that a brilliant novel requires, then – talent, ambition, skill, intelligence – is present in abundance. And yet, somehow, a brilliant novel is not quite what we end up with. It’s a thermodynamic conundrum. With this much creative energy invested, why does the result feel underpowered?”

Byers’s answer is: “diffusion.” The author has bitten himself off too much. He has given himself too many limitations, too many characters, and perhaps too few pages, to fully chew through it. And I agree, the result is diffuse: it accomplishes very little in the way of world-building. This, after all, is the standard payoff associated with historical fiction, intellectual history, and the novel of a hundred characters when done well. It allows you to privileged insight into a world you would be otherwise unable to inhabit.

But yet, I don’t think that’s the core issue. I’m not sure that world-building was what the author set out to do. I think the author’s intention was to reveal some important insight about technology, artificial intelligence, the nature of genius, or the nature of John von Neumann. My problem is that I came away uncertain about what exactly that insight was.

Share

What exactly is this book about?

So what does Labatut’s story reveal about John von Neumann? About technology? About the past, present, or future of artificial intelligence?

Both Tom McCarthy (whose Making of Incarnation is next up on my fiction shelf) in the NYT, and Becca Rothfeld, my favorite literary influencer, in WaPo, come across as unflaggingly enthusiastic about the book. But they both seem happy to leave this question unaddressed: What exactly is this book about?

For example: over the course of the book far, far too much is made of von Neumann’s illness. At the end of his life, von Neumann goes mad from the treatment of his terminal cancer. The book’s characters are simply obsessed with this fate, seeing it uniformly as the inevitable culmination of von Neumann’s intellectual pursuits. It is a Sontagonal illness-as-metaphor taken to an uncritical extreme. I appreciate that Labatut, by his own rules, must remain within the historical facts of von Neumann’s life. But as a narrative device, it’s incredibly flimsy. Certainly, I appreciate the tempting narrative symmetry between the heights of von Neumann’s intellectual achievements with the depths of his illness. But dwelling on this trajectory, as the characters often do, has the unfortunate effect of weakening the supporting cast’s observations about the protagonist.

All the accounts provided by the various characters agree: John von Neumann was the most brilliant man of his generation, then suffered a debilitating illness and lapsed into insanity at the end of his life. They all seem to read this demise as dictum on his life—how else was a life of such immense genius supposed to end? Well, I mean… lots of ways! You could go like Einstein, with quiet elegance. Like Russell, you could die with the dignity of old age and vast wisdom. Like Turing, you could die in unjust tragedy. Like Marie Curie, you could be killed by that to which you had been most devoted. If the book’s novel insight is that the mad scientist is fated to topple from genius into madness, that is neither novel nor an insight.

Here are some other possible theses the author builds toward:

(1) The possibility of building technology—am I required to call it Promethean?—so advanced that it actually imperils rather than augments humanity. Particularly in the context of the atomic bomb and AI, I don’t think this theme is lost on anyone these days.

(2) The interpersonal costs of success. Von Neumann’s unparalleled intellectual acumen clearly comes at the cost of his ability to maintain normal human relationships. Yet this is a story as old as success itself.

(3) The inability of rational minds to conceive of all possible outcomes. This is a possibility. But I think the only people who will be surprised by this conclusion are the so-called rational minds in question.

(4) Technological process as a human enterprise and not one that is carried out by the especially normal or functional members of our species. This seems to me the plausible thesis of the book. No one has the full story: there is no single narrator to technological progress. Omniscience is not possible. The effect of Labatut’s formalism, in this reading, would be to show that technological progress is undertaken by the proverbial blind men (afflicted by mania, in this case), feeling the different parts of the elephant and coming to separate conclusions about what kind of thing the overall beast might be.

All that said, the book’s best and most accomplished sections are its opening and its closing. These sections—one man’s rational descent into madness, and AlphaGo’s recent victory against Lee Sedol—land most effectively. They are also the only ones that have pretty much nothing to do with John von Neumann. Labatut’s book does draw a line between von Neumann and the contemporary success of AlphaGo. But it is one that is immensely squiggly and convoluted around the life of John von Neumann, then dashed between that period and the modern one, denoting that something happened in that interim, but what exactly it might be is not specified. I kinda wish it were.

Leave a comment

The future of non-fiction: fiction

Though I’ve been critical of Labatut’s work, let me say that the criticality derives from an ardent, near-religious belief in the project he set for himself. I believe that in this book, Labatut has put his finger on the future of non-fiction: fiction.

Over the next decade, I predict we will see more projects like this: fiction dressed up as non-fiction, or non-fiction dressed up as fiction. Around twenty years ago, with the publication of Malcolm Gladwell’s Tipping Point, there was an innovation in non-fiction. The use of the story to illustrate serious academic ideas. That paradigm has now reached a point of saturation, of exhausting that medium’s capability—to the point where ChatGPT could probably write Adam Grant’s next book for him, and the reading public would not know the difference. The only narrative frontier for the exploration of ideas to go toward is fiction. We see this done in Labatut’s work. We also see it in Nathan Hill’s recent Wellness (which I plan to write about soon). The most interesting and influential books of ideas in the coming years will, like Labatut’s, test out different literary devices for how best to convey this currently enigmatic blend of fact and fiction.

On this front, it will be exciting to see Labatut’s continued contributions to this venture. He clearly has the right stuff. For me, The Maniac is in the realm of work that competes for the cutting intellectual edge—but ultimately, is slightly off. It is what Boole wrote before The Laws of Thought. It is De Morgan’s work to rival Boole’s. (Both are quite excellent company.) I look forward to seeing what he comes up with next.

Subscribe now

Thanks for reading! If you liked this piece, you might also like my piece—call it a treatise of sorts—about how to read a novel as a theory of behavior. It’s very AI x literature crossover, which will probably be of interest to you if you made it this far.

Science, fast and slow

For most scientists, a big part of doing science is the pressure to publish. Scientists feel this pressure because the alternative to publishing is perishing, which is generally considered an undesirable outcome. Unlike other human pursuits, science isn’t supposed to be this way. It should be about revealing profound insights about the nature of the world, the kind that reverberate through the ages. Or at least something close to that. So when a scientist, in her off time, holds the pressure-cooker reality of the enterprise up to the leisurely tempo of this romantic ideal, many scientists come to the same conclusion. Science moves too fast.

At its core, I’m sympathetic to this concern. The scientific enterprise has deep structural issues. It puts pressure on researchers to churn out lots of papers, many of them lower quality than they would like. It rewards scientists almost exclusively on whether they have a lot of highly cited papers in high-impact journals. The culmination of this incentive structure is that it can make science feel like a process of flimsy, short-term optimization rather than the robust, long-term project that it ought to be.

But while I agree with the core of the concern, I think the diagnosis is wrong. The problem is not that science moves too fast. It’s that it doesn’t move fast enough.

Specifically, the problem is the timeline of feedback. The vast majority of academic research projects do not receive meaningful third-party feedback until well after an experiment has been designed and implemented, the data collected and analyzed, and the paper formatted and submitted for publication.

In any other field, it would be ludicrous to expect a product be finalized before it hits the shelves. From big corporations to small startups, there’s a pipeline for testing out a small thing before having to make the full investment in the big version of it. Feedback from an external set of eyes other than the people developing the product is a crucial part of the process.

Scientists also need a pipeline for testing their ideas before they have to turn them into finished products. As it currently stands, scientists (specifically graduate students) can spend years working on a project and not even know whether anyone will even care about what they’ve found even if an experiment turns out exactly as they hoped. It’s not a good way to develop anything, whether a scientific or consumer product.

The issue is not that publication come too soon. It is that feedback comes too late.

Subscribe now

Proposal: The Scientific MVP

My proposal is that science needs a conceptual equivalent to what is known in the startup world as a Minimum Viable Product. An MVP is a stripped-down version of the ultimate product a company plans to build. It is designed to empirically test the underlying assumptions of that product and validate whether or not it will, if built in full, find a market eager to use it.

Though it predates the concept of the MVP, a classic example would be Facebook. It didn’t start out as a behemoth social media platform attempting to connect everyone on the planet; it started off trying to connect students on a single college campus, then a select network of campuses. It validated the assumption that the style of connection offered by Facebook’s nascent product was a kind of connection that people were interested in. With an MVP, the point is build the smallest version of your ultimate vision. By testing whether the light-weight version is received in the way you hope, you get useful feedback about how to improve the core of your vision before building out the rest of it.

The process of developing an MVP is often compared to the scientific method—but applied to startups rather than traditional academic disciplines. The point is to identify the untested assumptions of a product, then create an experiment that will show whether the assumption holds up. It is a process of empirical experimentation.

This scientific inspiration may seem somewhat ironic in the context of my proposal that science should take a page from the startup playbook. But the practices of science are often appropriated by Silicon Valley, where they are refined and redeployed with increased effectiveness. This refinement is possible in the resource-rich environment of startups in a way that it is not in the resource-scare one of academia. For example, in How Google Works, Eric Schmidt’s book on the company’s cultural, innovation, and management practices, Schmidt describes the company’s organizational structural was originally designed to mimic the kind of small working groups found in scientific laboratories. After all, Google’s cofounders, Larry Page and Sergey Brin, met as PhD students. But while they were inspired by academic precedent, the management practices development at Google and other modern tech companies have gone far beyond the haphazard, informal practices of academia. The same is true with the notion of empirically-informed feedback in the minimum viable product.

The process of the MVP looks like this: First, identify the riskiest, highest-impact assumption of your product. Then design the smallest experiment necessary to meaningfully test this assumption. Once those data have been analyzed, refine the assumptions and continue testing them as necessary. The objective is to minimize the amount of time it takes to complete this cycle.

Typically, in a startup the assumption being tested is whether there is a market for a given product. The founders think they’ve got an interesting idea. Initially, they’re assuming that others will find it interesting as well. They need to test that assumption. In a study of more than one hundred startups, 35% failed because of “no market need” for their product. This was the second most common cause of failure, with the most common (38%) being “ran out of cash.” Both of these problems are designed to be addressed by the iterative experimental process of the MVP. If your riskiest assumption is that people will actually want to buy the product you’ve built once you’ve built it, then that assumption needs to be tested. And if you don’t test it quickly enough, you’re going to run out of money before you can successfully refine your assumptions about market interest. Of the 111 failed startups surveyed, three quarters died because of a problem the MVP was designed to fix.

Share

“No market need” science

A lot of scientific work suffers from the same problem: a lack of interest once it is released into the relevant market. A scientist puts their work out into the world. And it turns out that even though the work is solid, the finding isn’t something that other scientists find especially interesting.

How big is this problem? What percentage of science suffers from the equivalent of “no market need”? It is difficult to estimate what percentage of scientific papers go uncited. And evidently a lot of the numbers that thrown around are highly exaggerated. But according to a 2018 analysis published in Nature, the number depends a lot on which discipline and which journal one is looking at. The overall percentage in the natural sciences could be around 21%. Of course, there are a lot of scientific papers that have a non-zero but exceedingly modest number of citations. This would expand that figure of “no market need” science quite a bit. Either way, I think a lot of scientists would anecdotally support the claim that there’s a lot of fluff out there.

The scientific community needs an equivalent concept of a minimum viable product for the same reason as the startup community. You need to show that on the other side of this big investment of time, money, and energy, someone will actually care about what you’ve done. As it stands in science, it’s possible to get through an entire PhD without actually having to haul your work before someone who isn’t you or your advisor and see how your colleagues who aren’t emotionally invested in the work react to it. In science, no market need means no citation. No citations means no career.

As much as scientists may despise this fact, having a career as scientist isn’t just about doing good science; it’s about getting other people to recognize how good that work is. As a negative example, there is the recent controversy about Francesca Gino’s alleged fake research on dishonesty. That’s a case of low quality science with high quality marketing. Another more entrenched version of this would be the replicability crisis—many headline-grabbing findings (not just in psychology, but other disciplines as well) turn out to be more intriguing than true. Getting people to appreciate one’s research is not the sole goal of the scientist. But it is one of them.

Here’s a more positive example. In an interview from a couple years ago, I asked Mark Granovetter about his paper, “The Strength of Weak Ties.” He told me that he had initially titled the paper “Alienation reconsidered.” He sent that paper to a journal and the reviewers hated it. Granovetter was discouraged, but didn’t give up. He reworked the paper a bit, keeping the same central themes and ideas while adjusting how they were framed, then presenting it as “The Strength of Weak Ties.” Today, this paper has over 70,000 citations. It is one of the most cited papers in all of sociology.

Granovetter’s example is a genuinely foundational paper in its field, but one needed some reworking before his colleagues were able to make sense of it. How many other great insights languish in obscurity because they didn’t get the adjustment needed to make them understood by the widest possible audience?

Again, my point is not that there is no feedback loop in science. It is that the feedback loop is way too long. It shouldn’t be the case that a researcher conducts a whole paper’s work of experiments, then submits it to the scrutiny of peer review with their fingers crossed hoping that their colleagues don’t hate the work they’ve done. Granovetter’s example shows that even good science can benefit from tweaking its framing. Peer reviewed feedback comes late in the process and often isn’t that helpful. Instead, we need to construct a scientific pipeline where this sort of reworking comes early in the process, and sharpens an idea into its best, most useful possible version.

A lot of scientists will recoil at this proposal because it turns scientific inquiry into something vulgar: a product. While I’m as drawn to the romantic vision of science as anyone, there’s also a pragmatic side to the way science is produced. It can be more or less effective. And from this pragmatic view, science is already a product whether you like it or not.

The product is scientific papers. This is the sole currency of the scientific community. A researcher’s work is only as valuable as the papers they publish; it is the way they are judged by their peers, as well as their departmental employer. Scientific papers themselves are not judged in terms of the amount of money they make for a company. But they are judged in a similarly quantitive way: based on the number of citations they receive. The prestige of the journal may also matter, but mainly because it is perceived as a predictive measure of the paper’s long-term potential to rack up a large citation count. The product’s “users” are other members of the scientific community, who are themselves engaged in the same process. This side of science is itself a capitalistic enterprise. The capital is not monetary, but measured in units of prestige.

Another concern would be that this proposal discourages long-term innovative thinking. It incentivizes researchers only to work on ideas that will succeed in the short-term, ones that are easily digestible to a broad audience. Who knows what will turn out to be valuable in the long-run despite being initially under-appreciated? The entire point of scientific research is to take chances on finding out something that may not be fully explained by current thinking.

In general, I agree with this concern: Science should be about big, long-term impact and not easy, short-term success. However, this concerns dramatically overestimates the extent to which scientific practices are currently set up to do this. Most scientific projects are aimed at incrementally adding minor contributions to existing frameworks. They’re just not doing so as effectively as they could be.

In fact, the MVP approach would likely open scientists up to take greater risks. Under the current scheme, scientists have to be reasonably certain that a given project will pass peer review scrutiny. And because only fully completed projects make it to the peer review threshold, scientists don’t invest in a project unless they’re sure it will pass peer review scrutiny. So they are incentivized to make cautious choices about which projects to take on. If instead they could validate a small part of a larger, more ambitious vision, then they might be more willing to take a risk.

The process of getting feedback does not make the scientist beholden to a committee. It does not subject one’s work to the homogenizing forces of short-term optimization. Certainly not any more than the current feedback loop of peer review. Instead, the MVP process I’m proposing gives scientists further information to work with, sooner and in greater depth.

I mean, after all, we’re talking about scientists. If anyone should be comfortable with the idea that more information is better, it’s them.

Leave a comment

But what about… peer review? lab meetings? departmental talks? conferences?

But don’t we already have this process in place? Isn’t this what peer review is all about? In my own example, I cite how Mark Granovetter successfully reframed his paper via… peer review. And for that matter isn’t the iterative process of finding a solid framework, then publishing lots and lots of papers making small improvements to that initial idea the kind of process I’m describing?

All good questions. But no, I don’t think so. Peer review is not the same thing as what I’m describing. There are two reasons:

(1) Timescale matters a lot.

The problem I’m trying to solve is that researchers, particularly PhD students, can spend a very long time working on a project before getting any sort of validation or constructive criticism from someone who doesn’t have an emotional investment in the project’s success. Imagine what a PhD would look like if instead of investing years into a project before getting critical feedback, the system was set up to provide actionable feedback in a matter of weeks or months. My sense is that it would make PhD students feel less isolated, more confident in their work, more effective in finding a niche to fill, as well as serving to make their initial good ideas even stronger, disabuse them of their bad ideas, and decrease the probability they’d spend their whole PhD on a project that doesn’t end up working. I don’t think anyone suspects peer review is an effective cure for any of these issues.

(2) Peer review isn’t that effective.

In a post on his Experimental History substack, Adam Mastroianni argued that peer review is ineffective system, and one that hasn’t adequately been tested by scientific scrutiny. In other words, we have no empirical evidence that our primary means of evaluating empirical evidence is any good. I agree with that argument overall, but specifically for the purposes of a minimal viable product the problem is that peer review doesn’t give you feedback about the fundamental assumptions of your experiment—whether any one will care that you did it.

Peer review is designed to poke holes in your experimental methodology (which in practice often means getting feedback on whether it accords with the reviewer’s personal opinions about how best to study a topic). But it is not designed to answer the question: “If I run a scientifically sound version of this study, will anyone care about it?” As suggested by the Francesca Gino example, lots of scientifically questionable work gets through the porous filter of peer review. And lots of scientifically sound work gets through peer review, only to languish in academic obscurity. So no, I don’t think peer review is a good solution to the problem I am interested in solving.

The scientific community also has other, less formal methods for providing feedback to nascent research. For example: lab meetings, departmental brown bag talks, and poster presentations at conferences. These can be useful. But ultimately their utility depends on the culture and practices of the lab, department, or conference. It isn’t a standardized part of the scientific process, certainly not to the extent that peer review is.

In The Art of Computer Programming, Donald Knuth wrote that “premature optimization is the root of all evil.” It seems to me that the incentives of scientific production require scientists to optimize their studies before validating their core assumptions. As much as the idea of treating early-stage scientific research as a minimum viable product might be unsavory to many scientists, it is an unavoidable observation that there are many commonalities in the developmental processes of science and startups. The unfortunate fact is that the startup people have lots of money and the academics have comparatively little. Based on this alone, startup people have invested a lot more in figuring out the best-practices of their process. Because science as an enterprise has invested comparatively little into figuring out the best practices of its own production process, I think this should give scientists pause to gather up enough intellectually humble to be willing to learn from the example of a field which may be less rarefied but is far better funded.

I read Michael Lewis’s new book, Going Infinite, last week. I wasn’t necessarily planning to, as I’m not hugely invested in the Sam Bankman-Fried saga, and even though I admire Lewis I don’t read every one of his books. But I wanted to read this one because it was universally panned by the critical media. I dove into the book rubbing my hands together, eager to see what brought down this giant of journalism in the eyes of the critical public. And yet… I couldn’t help but come away feeling that Lewis got it right.

Moneyball Author Swings and Misses; Big Short Author Comes Up Short; Blind Side Author Turns Blind Eye; Liar’s Poker Author Gets Bluffed; Undoing Project Author Leaves Latest Project Undone… you get the idea.

Michael Lewis is one of the most successful and widely admired journalists of his generation. Even if you haven’t read his books, you’ve probably watched the movies they’ve been turned into: The Blind Side, Moneyball, The Big Short. Every rock Lewis turns over seems to be hiding the undervalued narrative material of a best-seller in the making. His debut, Liar’s Poker, was recently released for its thirty-fifth anniversary, and in that three and a half decades Lewis has established himself as a master of his craft.

In his last two books, Lewis detailed the decimation of important but publicly unappreciated government employees during the Trump administration and the scientists who saw the pandemic coming. Both were well-received. And so it’s a bit of an eyebrow-raiser whenever Lewis’s newest offering is met with negative critical response—mostly because I’ve never actually seen that happen before. But it is exactly what has happened to his latest book on former crypto-billionaire, former soon-to-be-trillionaire, and current alleged fraudster Sam Bankman-Fried.

Pretty much everyone who reviewed this book agreed on two things: (a) Lewis is generationally great at narrative journalism, but (b) this time he got it wrong. Jennifer Szalai, of the New York Times, called Lewis’s book “strange”—in her opening sentence. A number of reviews said that the book on crypto you really want to read is Zeke Faux’s Number Go Up, including a compare-and-contrast in a very helpful Economist article, which notes that Lewis’s book “reveals little about the inner workings of crypto.” Even money people (as opposed to, like, literary people) really didn’t like it: Fortune declared that Lewis had unprecedented access to SBF but failed to reveal any new information. The original title of Helen Lewis’s piece for The Atlantic was “Michael Lewis is Buying What Sam Bankman-Fried is Selling.” Ouch.

All that is bad. But you know it’s really bad when Noah Smith weighs in without having even read the book. Usually the paragon of level-headed, evidence-based reasoning, Smith wrote the following of Lewis’s book:

I haven’t read Lewis’ SBF book yet, but in recent interviews, Lewis certainly says things about SBF that are less negative than anything else I’ve read. For example, talking to 60 Minutes, Lewis said that he doesn’t believe that SBF knowingly stole his customers’ money. And in a recent Time interview, he asserted that the money was lost due to SBF’s “catastrophic sloppiness” rather than to deliberate malfeasance. When asked if SBF is a serial liar, Lewis hedges, calling him a “serial withholder” instead.

… I don’t have enough information to say Lewis is wrong here, but I expect when all is said and done, there will be solid evidence that SBF knowingly stole from his customers.

Wow! I mean, what does Lewis say in this book to make people so mad? Is it some kind of love letter to Mr Bankman-Fried? Is the book an argument for SBF’s innocence? When I picked up this book, I was curious to read it so I could see for myself what about this book made it a near-universal target of ire. I came into it with a sincere eagerness to see how exactly the mighty have fallen.

But then I read it, and well… I didn’t agree with the critical backlash. I don’t think Lewis got the story wrong. He may not have told the story people wanted him to tell. But that’s not at all the same thing as getting it wrong. Lewis is very clear about the book’s mission at the beginning, and in my reading he achieves it. I don’t think he missed the bigger story. I don’t think he went easy on SBF. To claim otherwise seems like a pretty serious misreading of the book. Because this deviates from what so many others thought, I want to try to explain why.

Subscribe now

In Lewis’s telling, what is the SBF story actually about?

Let me say up front that in my personal theory of non-fiction criticism, a book should be judged mainly on whether it lives up to the promise made by the authors at the book’s outset. This is the fundamental compact between writer and reader. The writer makes a promise about the story that they are going to tell you. And if the value of what’s promised seems sufficiently high, you read the book. The title, subtitle, cover, and author’s brand recognition are part of this promise. And the introduction is part of this promise. It is possible for an author to miss what’s important about a story, absolutely—and it is your right as a reader to come to that conclusion. But the author’s core responsibility is to make a promise, then fulfill it. So what does Lewis promise?

At the outset of Going Infinite, Lewis poses a very specific question. In the final paragraph of the book’s introduction, Lewis presents the reader with the culmination of his first meeting with SBF. An investor friend had asked him to meet with Bankman-Fried to form his own impression, to figure out who this guy is and whether it would be wise to invest money with him. Lewis writes:

By the end of this walk [with Bankman-Fried], I was totally sold. I called my friend and said something like: Go for it! Swap shares with Sam Bankman-Fried! Do whatever he wants to do! What could possibly go wrong? It was only later that I realized I hadn’t even begun to answer his original question: Who was this guy?

The part that most critics quote in their pieces was what Lewis said to his investor friend: Go for it! Swap shares with Sam Bankman-Fried! Do whatever he wants to do! But the important part, the part that sets up the promise of the book, is the last sentence. This is what the book is about. Lewis, like everyone else in the finance world, was taken in by Sam Bankman-Fried. But no one—not then, not even now—could answer this very basic question. Who was this guy? The story Lewis sets himself up to tell is one that attempts to answer this question.

In trying to understand Lewis’s aim, it’s helpful to look at his previous books. I remember many years ago reading an interview by Malcolm Gladwell where Gladwell claimed, essentially, that his own style wasn’t all that unique because he was essentially doing what Michael Lewis does but across ten stories in a book instead of one. (I can’t find that interview now, but in looking for it I did come across a piece in WSJ magazine, called A Unified Theory of Michael Lewis, which makes all the relevant assertions.) I remember having read this because I thought it was totally off. The whole point of Gladwell’s approach of stitching stories together is to pull out their underlying dynamics—making a specific claim about the generalizable principle which the story embodies in concrete form. Lewis by contrast doesn’t offer this kind of critical insight. He never has.

For example, when he wrote Liar’s Poker, his goal, which he explains in an interview with Tim Ferris circa 2020, was to write what he saw happening in his first-hand experience as a trader in the 1980s as clearly and directly as possible. As Lewis says:

I had only one kind of moralistic thought in mind when I wrote it, because I really just thought, my models that I had in my head when I wrote it were Education Of Henry Adams and Rousseau’s Confessions. The model was: just tell the world what happened exactly as you remember it and that’s enough. You don’t need to layer on an interpretation of what happened. What happens is good enough.

Nonetheless, readers wrote in to say: Dear Mr Lewis, thank you for writing a book on how to get rich on Wall Street. Are there any other tips that you did not include in the book? Michael Lewis has always treaded that line between buying into the story he’s telling and trying to give you the unvarnished facts of what he’s seen. For as long as there have been Michael Lewis books, there have been misreadings of Michael Lewis books.

This is what you are getting when you buy a book with the name Michael Lewis on the front. This is what Lewis promises to do. Present the facts as he sees them. To let the reader draw their own conclusion. This has always been his formula: fall in love with a character, gain privileged access to their perspective, and present the reader with the facts of what this character is seeing that the rest of us aren’t.

That said, I admit it’s totally the case that the way you tell a story reveals how you think about it. Several reviewers have noted Lewis’s M.O. is to fall in love with his characters. In Going Infinite, this was the first time that he fell in love with a character, started in on the story, and then mid-reporting the character turned into a villain in the eyes of the public. Therefore, it is reasonable to suppose (as Helen Lewis does in her Atlantic piece) that Lewis was unable to take the rosy glasses off when new information came to light.

In this reading, Lewis’s latest book is a kind of Kahneman & Tversky-style cognitive bias experiment. In the classic work, detailed both in Kahneman’s Thinking, Fast and Slow and Lewis’s story of their collaboration, The Undoing Project, our heuristic-based cognitive biases—such as confirmation bias, anchoring and adjustment, etc—work well 95% but in clever experimental setups can be shown to be wildly incorrect. Maybe the circumstances of the SBF story contrived to reveal the shortcomings of Lewis’s which have up until now remained invisible. This time around the situation was arranged to pull the curtain back on just how poorly this strategy can play out—specifically when you’re used to writing about unsung heroes and then end up writing about well-known crooks.

But I don’t buy it. I think yes, Michael Lewis is still enamored of his subject. But not overly so. In my reading what other people are mistaking for admiration is that, in short, Lewis finds the story amusing. It is not a story about who is right and who is wrong. It is just kind of nuts—and Lewis found it fun to watch it all go down. One of the first set piece scenes of the book is when SBF is on a Zoom call with Anna Wintour, and Bankman-Fried is playing video games counting the seconds until the call is over. It is a unique moment in human history. Lewis was there to document it.

It is not until the last chapter that he really gives the reader any insight into his own verdict on the matter. If you read all but that final chapter, you’d probably be a bit confused why everyone is so up in arms about it. And it is these conclusions which he talks about disproportionately in interviews (and then quoted by Noah Smith et al) because, well, that’s what people want to know. So Mr Lewis… do you think he’s guilty? Frankly, I’m surprised at how desperate people are for Lewis to come to a very simple, straightforward answer that yes, SBF is unequivocally guilty of knowingly defrauding his investors.

Ultimately, the question “Who was this guy” has two functions in the story. The first has to do with giving an answer; the second has to do with a kind of non-answer. Lewis’s investigative reporting is meant to give his best answer possible, and that’s the content of the story. But part of the conclusion, Lewis’s own take on what he’s uncovered, is that he isn’t really able to give a simple answer. It’s not just “He’s a crook” or “He’s unbelievably naive.” In Lewis’s reading, any who tries to give a straightforward answer like that misses some important part of the SBF account. And anyone who actually seems to know him—that is, who has spent some large amount of time working closely with him—has no idea, even still, who he is. That’s the whole point! When you look at all the facts, it’s really hard to tell who this guy is, what motivates him, why he’s done what he’s done, or even what he did.

Lewis at no point promises to reveal whether or not SBF is guilty. He doesn’t promise to tell you everything you want to know about cryptocurrency or Effective Altruism. He doesn’t promise to tell you whether investors were knowingly defrauded. He simply doesn’t. And if your criticism of him is that he doesn’t reveal enough on this topic—which seems to be the main criticism leveled against him—then you have a right to feel that way. But it’s not the author’s responsibility to write the book you want them to. Doing otherwise is like going to your local Persian restaurant and criticizing them for not serving Korean food. If you want Korean food, pick a different restaurant! It is the author’s responsibility to give you a sense of what the book is trying to do and do it to the best of their ability. This is the final critical measure of whether a book succeeds. On this metric, Lewis hits the mark.

Share

What specifically does Lewis claim about SBF’s guilt/innocence?

So let’s talk about that final chapter. As it stands, I find Lewis’s argument in it compelling. I’m not an expert on the SBF story (though I did follow it prior to Lewis’s book) or crypto as a field. I’m evaluating the argument Lewis makes at face value.

First of all, the question that everyone is taking Lewis to task on is a psychological one. It is about intent. It is about getting into Bankman-Fried’s head and deciphering his past beliefs and goals. It is not about whether investors were defrauded; they definitely were. It is about whether they were knowingly defrauded. Does the evidence, as presented by Lewis, point to witting malfeasance or unwitting incompetence?

Here’s my summary of the specific claims Lewis actually makes: His overall conclusion to the story is that he’s a smart kid who is misunderstood by pretty much everyone, not unlike Michael Burry in The Big Short, or Billy Beane in Moneyball. But unlike those characters, Bankman-Fried gets in over his head in a way that flips the script. Instead of continuing to be misunderstood, people start to believe in him. They buy into this ineffable quality he has which is something like mystique. Lewis’s story is structured to show that SBF’s well-documented incompetence at the larger scale (misplacing billions worth of investor’s money) is more or less identical to the incompetence that he exhibited at the smaller scale (which is what the first third or so of the book is about, particularly misplacing millions of his own money). When things were going well for SBF and his companies, people said he was going to be the first trillionaire. And then it all goes wrong, and everyone is just as quick to believe that he was the world’s greatest schmuck. In Lewis’s view, these opposing beliefs were based largely on intuition rather than evidence but nonetheless maintained with equal fervor by public consensus.

As far as I can tell, people who are mad about Lewis’s conclusions are either overextending his claims into something stronger than he actually says or feel that he has cherry-picked all of the evidence leading up to this conclusion. It seems like the first one is pretty well explained by people not reading the book with enough nuance and being quick to anger when the story doesn’t share their own prejudices. However, the second one is definitely a plausible issue, and I would very much be willing to update my position based on additional evidence.

Yet in rebuttal to this second concern—that told the cherry-picked story in a way that made SBF look good—my main problem with that concern is that Lewis didn’t do that. The story isn’t meant to get us to fall in love with SBF or believe him or think he’s a good guy. It’s hardly a flattering portrait at all. Lewis isn’t exactly making fun of Bankman-Fried, but more making fun of people who were taken in by himself (including the author himself): Look! This is the guy you thought was going to be the world’s first trillionaire! In reviews, people write as if Lewis himself thinks Bankman-Fried was truly destined to save the world then it was society who unfairly decided to throw him under the bus. Not at all! Lewis may need to fall in love with his subjects to write his books. But the reader doesn’t need to fall in love with them to read ‘em.

Anyway, there are also a lot of specific claims Lewis makes which I think are strong. For the most part, I see people disagreeing with the general points of his conclusion because it doesn’t fit with the story they want to tell about the obviousness of SBF’s guilt. But I don’t see a lot of evidence to combat the specific holes in the “witting malfeasance” story Lewis points out. Here are the specific claims Lewis makes supporting why is skeptical that SBF knowingly defrauded investors:

• If he wanted to knowingly defraud, he could have done it better. He could have covered his tracks. Lewis goes into the details, but in short it would have been very easy for him to make slightly less money but dramatically decrease the likelihood of getting caught. SBF was all about calculating “expected value” and was particularly sensitive to Black Swan events that are low probability but catastrophic. So if he was actually trying to defraud, he would’ve done a lot better job of it.

• There were lots of people who should have noticed that something was amiss (i.e., had the incentive to identify the issues and should have had access to the relevant knowledge, if that knowledge was in fact available). This is true both internally for the company and for its outside investors. For the internal team, the question is why pretty much all of the insiders acted in a way that went against their best interests. Lewis writes:

None of the characters in this financial drama had behaved as financial characters are expected to behave. Gary had owned a piece of Alameda Research, but his stake in FTX was far more valuable. Nishad owned a big chunk of FTX and none of Alameda Research. Ditto Caroline, who ran Alameda Research but owned shares only in FTX. None of these people had any interest in moving money out of FTX into Alameda Research [which is the main thing at issue in the fraud allegation] in a way that put FTX in jeopardy. Just the reverse: it might as well have been their money that was being moved. And yet at least up until late spring of 2022, when crypto prices began to plunge, and possibly much later than that, none of them expressed disapproval of the risk being taken with their fortunes. Why not?

• Lewis’s interpretation also leans on a version of Hanlon’s razor: “Never attribute to malice that which is adequately explained by incompetence.” In Lewis’s reading of events, incompetence is the simplest explanation. The reason he doesn’t present any interpretation or call into question other people’s readings of the facts is that he’s showing you all the little things at which Bankman-Fried was incompetent are pretty much the same as the big thing of which he is being accused of malicious intent. Lewis provides evidence that all the little stuff was due to incompetence, albeit incompetence on an epic scale. In the absence of direct evidence, it’s difficult to leverage the evidence in the little stuff into supporting a totally different psychological interpretation of the big stuff.

• Lewis is also skeptical of the prosecutor appointed to the case. According to Lewis, the prosecutor doesn’t know anything about crypto. A lot of things that look like classical fraud in the pre-crypto paradigm are perfectly explicable by common knowledge facts in the crypto world. Lewis isn’t saying evidence of knowing fraud will never be revealed. He is saying that a lot of the assumptions that we (the public, the prosecutor, etc) are used to making need to be reexamined.

• What best explains the public certainty that SBF is guilty is the same thing that explained the public certainty that he would one day be a trillionaire. People just don’t know what they’re talking about and are speculating with an undue degree of confidence. Lewis’s argument, in short, is that if we seriously ask the question “Who was this guy?” the answer is that we still don’t really know.

Leave a comment

Same level of niceness + same level of criticality + different evidence = different conclusion

Here’s a quote from Noah Smith that gets to the heart of why I’m defending Lewis against the criticism his book has received:

So my bet is that Michael Lewis is just being way way too nice to SBF here. But why? Lewis didn’t hold back when criticizing the bond traders at Salomon Bros. or the creators of mortgage-backed CDOs; why would he lob such a softball at a guy whose finance empire was infinitely more scammy than those others?

I think there’s a simple answer to why. It is this: Same level of niceness + same level of criticality + different evidence = different conclusion. Michael Lewis is just as nice as he always was. Michael Lewis is just as critical as he always was. Michael Lewis may fall in love with his characters, but that doesn’t mean he has no critical distance. If he was really just a run-of-the-mill finance bro, he would’ve just stayed on Wall Street instead of becoming a writer. The reason that he told the story the way he did is not because he changed. It is because the facts changed.

In summary, what Lewis set his book up to do was tell the backstory about who SBF was and why that question was so hard to answer—even now and even when asking those closest to him. Lewis’s general claim is that SBF’s psychology is difficult to decipher and that makes the psychological question of intent difficult to answer. His specific claims succeed in creating uncertainty about people’s belief that SBF is guilty of knowingly defrauding investors. In his own defense, Lewis said:

Lord knows the very easy, lazy thing to do right now would be to throw Sam under every bus you could, and try to make him seem as bad as possible, because people respond to that. But it just wouldn’t be true. It’s not the character I knew. It’s not the situation I knew…There’s a long history of people getting in trouble for books that are right, and the books end up being extremely valuable...And in this case, again, it felt like there’s a tribe waiting that were natively hostile to the truth of the story I experienced.

From the evidence and arguments I’ve seen, I buy what Lewis is selling. No one else has convinced me otherwise, even though I came in wanting to agree with them. Most articles have said something along the lines of SBF is clearly guilty but Lewis doesn’t say so, therefore Lewis has screwed something up in his story. Whether SBF is guilty, I don’t know. I’m not an expert on financial crime, and frankly what will become of this young man is not a concern that keeps me up at night. I’d very much be willing to flip the position I’ve described here based on additional evidence, and I look forward to reading the other crypto book by Zeke Faux to see if that gives an insight into something Lewis overlooked.

But I do think that we need to judge books, if not by their cover, then by the fulfillment of promises made in their introduction. Michael Lewis kept his promises. He’s still got it.

Subscribe now

Share

For most of educational history, essays have been a big part of evaluating student learning. But now, with technologies like ChatGPT, for the first time those essays can be manufactured automatically at little or no cost to the student. This has thrown the entire system of writing and grading into chaos. Yet, the question most educators and writers seem to be asking is how we can preserve the original system in the face of this new technology. Another way to look at it is that ChatGPT has revealed the flaws in the system. For years, we’ve been asking students to write essays that elicit generic, unreflective responses. No one seems to think ChatGPT produces the kind of writing anyone wants to read. And yet that seems to be the standard we ask students to meet. Maybe the problem isn’t the new tech; it’s the old assignments.

Mere coherence as the sole measure of competent writing

I went to a high school that prided itself on teaching its students to write well. At the very least, they would tell students and parents, when you graduate from here you’ll know how to write. I always found this claim amusing. When I graduated high school what I had learned was not how to write an essay. It was how to write an essay that teachers would give an A. These were not at all the same thing.

It was clear to me that these were different things because I never did the assigned readings. I would skim the relevant SparkNotes, pull a few key quotes, and make connections between plot points, quotes, and general human themes. Teachers seemed satisfied with this approach, and for the most part I got As. But it was clear to me I wasn’t learning to write. I was learning to bullshit.

If a resource like ChatGPT had been available when I was a student, I don’t honestly know if I would have used it. Maybe. But regardless of what my personal choices would have been, the fact is that ChatGPT has made the cost of composing bullshit—the kind of externally coherent but intellectual hollow work I specialized in as a youth—fall to zero.

And as a result, the educational pipeline is now facing a system failure. Teachers and professors are concerned they can no longer assign essays in the way they used to. For example, in a December 2022 article from The Atlantic, the author presents a ChatGPT passage which she had been sent by a professor friend. The professor considered the passage “graduate level.” The author herself gave it a B+; she writes, “The passage reads like filler, but so do most student essays.”

The author goes on to cite another professor, one who teaches at a business school: “The OpenAI chat is frankly better than the average MBA at this point.”

The truth is, most student essays do read like filler. And yet, when teachers encounter a computer program capable of producing filler, their instinct is to question the technology and not what they’ve been asking their students to do this whole time. I don’t blame anyone personally for feeling this as a gut reaction. This is exactly the kind of dramatic revision of the status quo which is hardest for people to deal with. But at some point we need to pose the uncomfortable question. If the prompts we’ve been assigning produce writing that’s indistinguishable from filler, doesn’t that mean we should be rethinking the prompts?

As it currently stands (and this is definitely subject to change), what defines the prose of ChatGPT and its peers is that it is generic. It gives generic answers. It writes in a generic style. It is a statistical model of baseline, generic human writing—the idealized Wikipedia voice from nowhere in particular. When we say that writing sounds like “filler” what we mean is that it meets a baseline standard of coherence without going beyond these kind of generic, exactly-middle-of-the-road insights.

This is just a more technologically sophisticated version of what I was doing in high school. And reflecting on my own experience, it makes me suspect that, perhaps, the system isn’t worth preserving. Certainly not at all costs. Instead of asking how we can perpetuate the existing system for assigning and evaluating essays in light of this new technology, we should use this as an opportunity rethink what exactly it is we are trying to teach when we teach writing.

Subscribe now

But students have to walk before they can run—don’t they?

The number one argument I can imagine people having against this perspective is that the current writing system is about teaching the fundamentals of writing. Sure, a lot of what students produce is generic and asks them to do little more than produce a grammatical, logically coherent essay. But students need to get that under their belt before their can tackle more advanced aspects of writing. You need to walk before you can run.

But while I agree that some fundamentals are important, I don’t agree that they’re the ones we’re currently teaching.

The point we have to grapple with is that we’re not currently minting strong writers. Academics are a case in point. Anyone with a PhD has gone through pretty much the full educational pipeline. They have successfully completed uncountably many writing assignments. But when is the last time you tried to read an article written by a PhD for a group of their peers? In most cases, it would be a blessing if it had been written by ChatGPT.

And it’s not just academics. Written communication in government or business is rarely as effective as it should be. This is understandable. It is genuinely hard to say what you want to say in written form. It doesn’t come naturally to us as humans. But I think we need to be honest about that and admit that there’s probably a lot of room for improvement.

So what are the fundamentals we’re currently attempting to teach? In light of ChatGPT is it worth teaching them in the same way we’ve taught them so far?

One of the most obvious fundamental skills is grammar and basic sentence composition. I would agree that a baseline ability to use words and form sentences is important for writing. But I don’t see technology having any impact on this, any more than SparkNotes did back when I was a student. First of all, a lot of this gets taught in elementary and middle school, before students are going to be tempted to turn in counterfeit assignments. Even if that’s not entirely true, the technology to aid student’s grammar and spelling has been around for a long time. Students are still able to form sentences.

Second of all, grammar is more of a social signal than the objective, right-or-wrong skill we credit it to be. The ability to use “correct” grammar reflects whether someone has received an elite education—not whether they can reason logically or write prose worth reading. As a tool of logic, people naturally learn syntactic structures which allow them to get their point across to other members of their community. For example, in one of my favorite ever linguistics papers, William Labov (the founder of sociolinguistics, and one of the first linguistics to truly mount a countermovement to Chomsky) argues that sophisticated grammar can actually obscure the speaker’s point, and that what we think of as “non-standard” grammatical structures can actually be more direct.1

Another fundamental skill is how to structure an argument. The conventional wisdom is that students can’t learn to compose more sophisticated forms of essays until they have mastered the basic format. So we teach students to write the ubiquitous, hyper-standardized five paragraph essay: an introduction, three body paragraphs supporting the claim made in the introduction, and a conclusion.

This assumption more reflects what is easier for teachers to grade than what is meaningful for students to write. The inescapable fact is that almost no writing in the real world conforms to the five paragraph template. I’m not necessarily against the five paragraph essay, as I do think it’s a good baseline structure for making an argument. But relying on this as a kind of gold standard of coherent writing seems that it encourages students to learn the rules of the five paragraph essay while excusing them from having to do the more difficult work of thinking through the best way to say what they’re trying to say.

I think it’s worth questioning whether we’re doing a good enough job teaching the principles of argument structure as it currently stands. It seems to me a little like calculus: we tell students that this is the standard to which they must aspire, and then once they’ve learned it we set them free into a world which will pretty much never ask them to deploy it. Is technology really what stands in the way of students learning how to make an argument?

Overall, my response to the walk-before-run argument is that we’re just okay at teaching students to walk and to be frank most of them never end up running all that fast anyway.

Ultimately, in order to figure out what fundamentals we want to teach, we need to figure out what we want to students to be able to do in the long-term. At the very least, one consideration should be matching what we’re requiring of students to the problems they are going to face after school. This is not the only point of education, but it is an important one. And there’s no rule in real life that says you can’t use technological solutions. For most students, they need to be learning to write with the aid of technologies like ChatGPT, not to learn to be scared of them because using them is tantamount to cheating.

In cognitive science, an individual’s ability to perform an action is a function of two things: their capacity to perform the action, and their motivation to do so. A lot of how we think about education is about in terms of increasing students’ capacity. The need to be better at whatever it is we are trying to teach them—in this case, writing. But they also need to want to be better. And to me, this is the problem that more often than not gets overlooked. Therefore, the bigger gains to be made are not in the effectiveness of our increasing student capacity, but in increasing student motivation.

And here’s the thing. What we’ve been asking students to learn is the most boring form of writing possible, the kind that ChatGPT can now produce ad infinitum. Is it any wonder students don’t like writing? From this perspective, the introduction of ChatGPT is great. We just automated the most boring part of writing. Now students don’t have to bother with writing filler to please their teacher! If we play our cards right, they might discover that no, they do not in fact despise writing. They might actually love it.

Honestly. This is great news.

Share

Good writing starts with good reading

My proposal is that the way to improve how we teach writing is to improve how we teach reading. As Stephen King once wrote, “If you want to be a writer, you must do two things above all others: read a lot and write a lot.”

The problem is that we start off by giving students literary masterworks before they’re ready, and because they don’t have the skills to grapple with them they conclude that so-called great books are a lot less great than teachers say they are.

Imagine if we taught math this way. Students would walk into the classroom in ninth grade, and the teacher would write out the fundamental theorem of calculus on the blackboard (or, you know, the digital screen thing). “This is one of the greatest inventions of human intellect,” the teacher would say. “If you really dig into it, you’ll uncover the secret wisdom of the ages. Trust me.” But no one teaches math this way. It would be madness. Obviously, we all know that you have to start with the basics and build up to the most advanced, most sophisticated concepts. This is even how we try to teach basic writing skills, as discussed above. But somehow, this logic is completely lost on us when we teach reading.

We give students the best stuff, the highest achievements of humanity’s literary minds, way too soon. Students don’t have the skills or experience necessary to engage with these masterworks, just as they wouldn’t have the ability to handle calculus if they didn’t have a grasp of algebra first. As a consequence, they don’t get the point of reading. And as a consequence of that, they don’t get the point of writing.

So why not? Why not get ChatGPT to write your essay for you? What’s the use of reading the book when you can just get a decent grade on the assignment and move on with your life?

The reason this strikes me as a big problem is largely because there’s an easy and obvious solution. Simply put, there is SO much outstanding work written to appeal to young adults. And instead we insist on teaching… Shakespeare?

Shakespeare is great and all, but it kinda seems to me like the calculus of literature—an advanced skill rather than an entry-level one. There is just too much excellent writing out there specifically for middle school and high school kids for us not to teach them. The fact that something is a “literary classic” is not a good enough reason to teach it students without giving them the necessary preparation to appreciate why it is such a profound literary achievement.

For example, a couple months ago I read a work of Young Adult fiction called More Than This by Patrick Ness. I picked it up because the cover featured a blurb from John Green (another YA author whose work is incredibly good) which said, “Just read it.” So I did. It’s about a high school boy who wakes up one day to find that he’s in hell. For this young man, hell is a post-apocalyptic version of his childhood hometown, but with no one in it. He has to figure out what is actually going on. Is he really, like, in hell? Why? What’s he supposed to do here? Is there a way to get out—or is this it?

The plot of the book is incredibly well-structured. The imagery of this grayscale, ash-covered hellscape is as vivid and worthy as any literary work. It deals with big themes, but ones that teens care about. They are not dissimilar from the themes adults care about—family, purpose, trauma, death, friendship—but are presented in the sharp relief in which teenagers experience them, rather than the oblique, layered way that adults do. And, most important of all, the book is constructed to achieve this in simple, straightforward language. You get all the upside of a literary classic, without any of the barriers to entry. And for the emotional circumstances that most teenagers find themselves, the number of students who are going to get something out of an author like Patrick Ness versus Charles Dickens or Emily Brontë… it’s huge!

For that matter, what about graphic novels? We’re not talking about comic books here. We’re talking about serious literary works that make use of a visual medium in addition to a verbal one. For example, Persepolis by Marjane Satrapi. Or Maus by Art Spiegelman. Or my personal favorite: Logicomix, a graphical based on the life and work of Bertrand Russell, by Apostolos Doxiadis and Christos Papadimitriou.

My own educational philosophy is that you must meet a student where they’re at. You have to ask them to go slightly beyond what they’re currently comfortable with. You can’t undershoot it, or they get bored. And you can’t overshoot it, or they get discouraged. There’s no use in trying to force them to do something they’re not yet ready to do. It’s better to be patient, to play the long game; the teacher’s job is to big the student tools with which they can build their own future. In most cases, it’s really difficult to make this work at scale: how do you know where each student is at individually? To me, choosing the right books seems like the lowest-hanging fruit of meeting a student where they’re at.

The logic is simple. If they’re a young adult, ask them to read a book for a young adult. As much as you want them to be able to read Shakespeare, you can’t force them to do it now. They have to build up to it. And you have to trust that if you can get them hooked on Patrick Ness and Marjane Satrapi, and give them the tools to continue challenging themselves as a reader, that they’ll get to Shakespeare and Jane Austen and Henry James eventually.

As it currently stands, we structure our students’ literary curricula as if they are all going to be avid readers of literary fiction from a diverse group of international authors by the end of high school. And do you know what the result is? The vast majority of adults, even educated ones, grow up to think they either have no interest in literary fiction or wouldn’t get much out of it even if they tried.

Sure, a tiny fraction of students will find that kind of work—the Brontës and Dickens of this world—engaging before they start college. And those kids? They’re going to be alright. They’re going to find their way. They’re not going to be leaning on ChatGPT to complete their assignments.

The problem we need to solve is how to get students to care about reading. This is one big difference between math and literature. No one thinks that we’re going to educate a populous of life-long mathematicians. People aren’t going to sit around in their forties or fifties in an armchair doing problem sets. But there’s a different expectation for literature. In a very real sense, an educated populous should be sitting down, at least every once in a while, to read a book. The extent to which we’re able to instill that motivation early matters—even more for books than for mathematics.

We do a decent job teaching students to have the capacity to read and write. Could we do better? Absolutely. And we’re going to figure out how ChatGPT fits into that. But where we really fall short is equipping students with the motivation to care about reading and writing. Just because something is part of the “literary canon” isn’t a good enough justification for teaching it; in fact, it might have a diminishing effect on student motivation. And if we can get that right—creating students who want to be life-long readers and therefore have an intuitive sense of the value of good writing—then we’ll have gone along way toward improving our educational system for reading and writing. Motivation can’t be automated.

Leave a comment

The field of artificial intelligence has often used what we know about human intelligence as a template for designing its systems. But for the most part, AI has always drawn on a view of the mind in social isolation, disconnected from the community of other minds. Yet one of the big insights of modern psychology—for example in work on embodied cognition—is that thinking doesn’t just take place between an individual’s ears. Human cognition is shaped by social and culture forces. What might it look like for AI models to take these forces into account?

AI as the search for fundamental principles of intelligence

In general, there are two ways to think about the development of modern artificial intelligence. The first is that AI is an exercise in engineering. A given algorithm either works or it doesn’t, and whether it works isn’t necessarily indicative of any deeper underlying principles in any scientific sense. The guiding metaphor for this view is aviation. We didn’t invent airplanes by studying birds. We started from scratch, without looking to evolution for inspiration, and simply tried different things until some of them started to work. But this isn’t the main sentiment among researchers in AI. Throughout the history of AI, its practitioners have always suspected that what they’re engaged in something deeper, that their work is uncovering something more profound. In this view, AI is not just engineering problem. It is the search for the fundamental principles of intelligence.

As a consequence, the influence between what we know about AI and what we know about human cognition is bidirectional. By formalizing their theories with models from AI, cognitive scientists can go beyond the box-and-arrow diagrams of their field and actually show whether what they know about cognition can be used to build a system with human-like thinking capabilities.

In the other direction, insights about human intelligence have tended to trickle down into AI—having sat around for a while with unappreciated utility and used at length by people in only in their vaguest, most general forms. This began long before the invention of modern computers. For example, George Boole tried to formalize human reason in his 1854 treatise on mathematical logic, The Laws of Thought. This work was based on the oldest insight in the psychological book—that man is a rational animal—and, as there was not yet a formal field of psychology in Boole’s time, he built on the Enlightenment’s renewed enthusiasm in reasoning and rationality.

The cycle of insight to application has turned with increasing speed over the years, given that the transfer of knowledge moves at an exponential clip. But one way to interpret the ideas in AI in the middle of the 20th century are as a response to ideas from psychology and neuroscience at the beginning of that century. For example, in his 1950 paper, Computing Machinery and Intelligence, Alan Turing posed the question, “Can machines think?” The dominant insights in psychology in the early 1900s, prior to the establishment of Behaviorism in 1913 by John Watson, came from William James: the stream of consciousness. Turing was, in a sense, asking whether it made sense to apply the same concept to computers. He called his proposed method for answering this question the imitation game; we know it as the Turing Test.

Perhaps slightly less well-known, but no less central, was John von Neumann. Besides the Turing Test, Turing’s other enduring invention was the Turing Machine. This was not actually a machine, but a theoretical construct showing what it would look like to create a “universal computer.” Essentially: a program that could designed to run other programs. Von Neumann used the theoretical work by Turing as a basis for the architectural design of the modern computer. Turing showed a computer was possible in theory; von Neumann actually made one. In his 1958 monograph, The Computer and The Brain, von Neumann speculates that, some important differences aside, the machine he built with digital circuits appeared to work in a similar way to the neural circuits of the brain. What were his insights about the workings of neurons based on? Little more than what could be inferred from the pioneering work of Santiago Ramón y Cajal.

Likewise, the first chess playing programs were designed by cognitive scientists intent on modeling the strategies of expert chess players (Herb Simon had to write the programs out by hand, because contemporary computers weren’t sophisticated enough to run them). Artificial neural networks—the precursor for today’s deep learning—were designed to address some of the differences pointed out by von Neumann. And reinforcement learning, through which an AI program interacts with an environment and learn from feedback, was based on the stimulus-response theories of Behaviorist psychology.

The overall pattern is that a lot of the central paradigms in AI are based on psychological insights, but only very generally. They are based on a true story—but with the caveat that the adapter is importing the story from another language, one she doesn’t really speak, and at any rate is way more focused on the kernel of wisdom that makes the story compelling rather than maintaining fidelity to the details of the original work.

In much the same way, the concept of Artificial General Intelligence—AI’s holy grail—is only loosely based on an understanding of what it means to be an agent with the full breadth of human cognitive abilities. Specifically, flexibility. What AGI is after is flexible cognition. Because a single human mind can learn to use language, perform complex reasoning, write computer programs, write poetry, tell you that a pound of feathers weighs the same as a pound of steel, and solve problems from architecting skyscrapers to navigating across open sea—then it must be possible to design an AI that can do so as well.

AI’s biggest and best paradigms come from looking at human cognition, but from very far away and only with any eye for general patterns that might be well-suited to implementation in computer programs.

So it makes sense to ask the question: What is the most general insight from psychology and cognitive science that is under-appreciated by people in AI?

Many cognitive scientists are happy to provide answers to this question. For example, Gary Marcus has been on a decades-long quest to disbelieve that progress in AI is really happening. But what I take to be the orthodoxy is a 2017 paper by researchers at Harvard, NYU, and MIT (for the record, I’m biased on what counts as the central beliefs of the field; I was working in one of their labs at the time). But paper focuses on specific things that need to be built into AI systems. Their goal is to say, if you really understood how the mind works and wanted to construct your AI accordingly, here is what you’d do.

I want to try to address a different question. What if you didn’t really understand the cutting edge of what was happening in cognitive science? What if you were, say, 20 years behind what was happening in the field, with only a rudimentary ability to appreciate what these insights mean? What if you stood really far away from the sum total knowledge of the field and just sort of squinted at it? After all, this is what the lineage of Boole and Turing and von Neumann and Simon and all the other paradigm-makers of AI seemed to have done.

Suppose you did that. What would you see?

Subscribe now

The mind: a view from far away

What you would see, I think, is that human cognition is a more social phenomenon than we previously gave it credit for. What AI people see when they look at intelligence is a lone individual—probably a computer programmer or a mathematician; definitely a genius—sitting alone in a room, with the full resources of human knowledge at their finger tips, synthesizing and applying this knowledge in ways no one had ever thought to do before. And, according to this lone-genius view of intelligence, the fact that human beings are social creatures is entirely separate from the fact that they are smart ones.

To be fair to AI people, this belief goes beyond the walls of computer science departments to the heart of Western philosophy. This is what Boole, inspired by Enlightenment thought, has in common with modern AI: the belief that we are the fundamental processes that make up our mind. It is for this reason that we named our own species Homo sapiens (wise man), rather than Homo cooperans (cooperative man), and endorse statements as I Think, Therefore I Am. In this view, the fundamental unit of intelligence has always been the individual. Not the neuron. Not the species. Not the tribe. But the lone-genuis sitting at her desk, typing away in the dark. It is easy to forget that this is not a fact, but a framework.

As the anthropologist Clifford Geertz once wrote:

The Western conception of the person as a bounded, unique, more or less integrated motivational and cognitive universe, a dynamic center of awareness, emotion, judgment, and action organized into a distinctive whole and set contrastively both against other such wholes and against its social and natural background, is, however incorrigible it may seem to us, a rather peculiar idea within the context of the world’s cultures.

The point is that people in AI think they will develop AGI by recreating a single human-like mind: one deep neural network to rule them all. But I don’t think that’s right. The problem with trying to follow the formula laid out by leading cognitive scientists I cited above, eminently reasonable and defensible though it is, is that AI researchers will hit the same ceiling they did when they tried to code systems for natural language processing by hand. There are just too many things to include in the system to get it to work. How do you know what to include, when to stop?

Here’s my from-a-distance version of what I think cognitive science tells us to do instead: Humans aren’t born terribly smart. They become smart by interacting with other humans. To be sure, humans are definitely born with some innate cognitive structure, that much is obvious and well-documented in developmental psychology. But the inescapable fact is that newborn human babies are helpless, physically and cognitively, in just about every possible respect. They have to do a lot of learning—and even though they get to from here to there eventually, they start from a point of pretty blatant incompetence.

The field of AI has already learned a version of this once, when they realized the importance of data. In the early years of AI, researchers concerned themselves largely with internal structure. This was the impulse, for example, of so-called expert system approaches, which tried to bake in as much concrete human knowledge to their architecture they could pack into it. But the approach didn’t scale. The biggest breakthrough in modern AI—going from AI being almost entirely useless to actually quite useful—was the development of deep learning. Yet this paradigm largely resembles the technical specifications for the neural nets dreamt up by cognitive scientists like David Rumelhart and Jay McLelland in the 1980s. It was only when it became possible for these systems to process a lot more data that they achieved the impressive results of modern deep learning. It wasn’t the internal structure of the model that mattered. It was what was going on outside of it.

The big insight of modern cognitive science is that the same is true of the human mind. Internal structure is definitely important. But the mind is more than just its fundamental processes. How that mind works depends largely on the world that mind inhabits. In particular, its social and cultural milieu.

To use the computer metaphor of the mind favored by many cognitive scientists and people in AI: trying to understand a mind in the absence of a larger social context is like trying to understand a laptop without realizing it is supposed to be connected to the internet. The laptop can do an awful lot without having to be connected to the cloud. You can create a document in a word processor; you can record a song in an audio editor; you can search through the files on the hard drive; you can even write a new program for the machine to run. But you will also miss a lot of the functionality that defines how people typically use the device. You will open the web browser and conclude this program is basically useless. You will think the video calling app is designed mainly to check your own appearance in the camera. You will suppose that the number of songs that can be played by this machine is quite limited. And you will have a hard time imagining what to do next with that document you’ve written or the song you’ve recorded besides turning the screen around and showing it to the colleague sitting across from you. In short, you will be able to understand a great deal about how that system works, and you might even be impressed by its capabilities. But you will fall dramatically short of appreciating the full breadth of what it can do.

AI models don’t need to be terribly smart. They need to become smart by interacting with other AI models. In a word, they need a mechanism for learning via culture.

Share

The social foundations of human intelligence

Specifically, there is a prominent theory from a cognitive scientist named Michael Tomasello which makes a prediction about what exactly it is about culture that is crucial for the development of human intelligence.

Tomasello’s academic advisor was a man named Jerome Bruner, who was an influential early cognitive scientist but skeptical of computer metaphor of the mind. Bruner’s research focused primarily on developmental psychology, and so Tomasello started off studying infant behavior. But Tomasello’s personal interest was in apes. In particular, he wanted to figure out what it was about human infants that allowed them to grow into fully-fledged adults, capable of learning human languages and participating in society, in the way that chimps, for all their sophisticated cognitive and social skills, could never do.

Tomasello presents his most influential theory based on two decades of work, published in 2000, in his book The Cultural Origins of Human Cognition. Tomasello opens the book with a puzzle, one based on the archeological record of human evolution.

According the best available estimates, humans split from the other great apes about six million years ago. Two million years after that, the genus Homo emerged. Initially, there were multiple species of Homo. But then, suddenly, around 200,000 years ago, one population of Homo began outcompeting all the others. Their descendants became Homo sapiens: cognitively modern humans.

What happened during this course of evolution that allowed the human side of the branch to build cities and farms and Facebook, while chimpanzees and bonobos remained in the trees? Tomasello points out that this adaptation happened in an incredibly short amount of time by evolutionary standards. After all, humans and chimps share the same amount of genetic material as rats and mice, lions and tigers, and horses and zebra. But the difference in outcomes, civilization building-wise, between humans and chimps is far greater than those between rats and mice.

“Our problem,” as Tomasello writes, “is thus one of time.”

The fact is, there simply has not been enough time for normal processes of biological evolution involving genetic variation and natural selection to have created, one by one, each of the cognitive skills necessary for modern humans to invent and maintain complex tool-use industries and technologies, complex forms of symbolic communication and representation, and complex social organizations and institutions.

In other words, Tomasello’s argument is that, overall, the fundamental processes of human cognition must be very similar to that of our great ape ancestors. There wasn’t enough time for evolution to patch together an entirely new set of cognitive processes. Instead there must have been a small number of cognitive innovations—say, one or two—which allows humans to leverage the capacities of their otherwise ape-like minds in new ways. This cognitive innovation, then, is a kind of cognitive missing link—leading from our great ape ancestors to cognitively modern humans—which allowed early Homo sapiens to accumulate, generation over generation, the cultural knowledge that modern civilization is built on.

Tomasello describes the mechanism behind this cognitive missing link as the “cultural ratchet.” Decades of research had already shown that non-human primates were capable of innovative tool use, crucially in work by Jane Goodall and her colleagues in the 1980s. The problem was slippage. A chimp might come up with a new way of using a stick to get food. But other chimps wouldn’t notice the effectiveness of the new behavior and appropriate it for themselves, as would be natural for humans. In order for a species to accumulate cultural knowledge over time, they need both the creativity component and the stabilizing ratchet component.

In other words, human intelligence isn’t just about the processes inherent in a single mind. It reflects our species’ unique ability to, as the quotation goes, stand on the shoulders of giants. What we think of as the modern human mind is a single cross-sectional moment in the upward trajectory of human intelligence. It is the accumulation of thousands of generations of our ancestors, each turning the cultural ratchet just a bit, and passing on the best of their insights and innovations to the next generation. This, for example, is one explanation for the Flynn Effect, the empirical observation that our species’ average score on intelligence tests tends to go up over time. Without the ability to accumulate cultural knowledge, there would be no cultural ratchet. And without the turning of the cultural ratchet, we would not have human cognition as we understand it today.

So what might this mean for pursuit of AGI, an artificially intelligent agent with human-like cognitive capabilities? In short, it means that even if we did replicate the fundamental processes of human cognition, it wouldn’t be enough. It would be like creating a laptop without the ability to connect to the internet. This machine would have a lot of useful functionality. But it would fall short what we know that technology is potentially capable of doing.

If Tomasello’s theory holds water, then human-like intelligence has two ingredients. The first would be the fundamental processes that underlie human cognition—including the cognitive ability to turn the cultural ratchet. The second ingredient is an opportunity to actually turn it. In other words, to engage in the kind of generation-over-generation learning that led to contemporary human intelligence.

As Tomasello wrote in a 2014 update to his theory, called A Natural History of Human Thinking:

Human adults are cleverer than other apes at everything not because they possess an adaptation for greater general intelligence but, rather, because they grew up as children using their special skills of social cognition to cooperate, communicate, and socially learn all kinds of new things from others in their culture, including the use of all of their various artifacts and symbols.

We typically think that what allows humans to be social is their general intelligence. But what if we have it backwards? Typically, we think that our species’ increased intelligence caused us to become social creatures. If this is the case, then the fact that we are social creatures really is divorced from our being smart ones. There is no reason that sociality is a necessary ingredient in intelligence. But according to Tomasello’s theory, we have the causality wrong: we are intelligence creatures specifically because we are social ones. What we perceive as human intelligence—yes, even the lone genius alone in her room—is the result not just of the fundamental processes of the mind, but thousands of generational turns of the cultural ratchet.

As the developmental psychologist Jean Piaget once wrote, “Only cooperation constitutes a process that can produce reason.” AI is a search for the beating heart of reason. But, as it stands, it doesn’t think it needs cooperation to get there.

Leave a comment

Artificial Collaborative Intelligence

If this is true, that human-like cognition is only possible via social mechanisms, then what the field of AI should be aiming for is less like AGI and more like ACI: Artificial Collaborative Intelligence.

For example, the current approaches in AI spawn ChatGPT, a chatbot which can produce language at a human level. But imagine: What if ChatGPT had the ability to talk with other ChatGPTs? What if it could learn from these conversations, the same way that one human can learn from talking to another? Unlike human conversation, this dialogue wouldn’t be limited by the speed of human reading or speaking or even thought. These two ChatGPTs would be able to analyze the entirety of human knowledge, going back and forth on what to make of it, with essentially no bounds on the depth of that scrutiny.

Now imagine that kind of conversational ability on the scale of thousands of ChatGPTs. They would all be engaged in discussion simultaneously and sharing their insights centrally, like a condensed version of all the world’s universities, think-tanks, and institutes combining their collective intellectual power.

This version of Artificial Collaborative Intelligence would replicate the way human learning works across historical time. New data comes to light, and new ideas and insights are proposed; these insights are debated and refined. The best of them are taken up by the next generation, and the ones that don’t work are replaced by new ideas. This evolutionary process would yield the kind of intelligence we think of truly human in a way that emulating the mind’s fundamental processes never could. But with ACI, it would occur on an unprecedented scale: each agent with its own ability to consult, analyze, and communicate its perspective on the sum total of human knowledge. At lightning speed, this perspective gets checked by the perspective of a potentially unlimited number of other agents engaged in the same process. It is difficult not to see this as the AI breakthrough of the future.

It is also worth noting that in the cases where a preliminary version of this approach has been tried, it is has been a huge success.

The most obvious example is AlphaGo playing against itself. In the original version of the program, developed by Google Deep Mind, the AI program was trained on a data set including thousands of previously played matches. This was enough to go toe-to-toe with the leading human Go player, Lee Sedol. However, the real breakthrough came with AlphaGo playing games not against human players—but against itself. It was then able to play a functionally infinite number of games, turning the cultural ratchet on its own ideas. As another Go champion said of these matches: they were like “nothing I’ve ever seen before - they’re how I imagine games from far in the future.”

A more subtle example is AlphaFold. This is Deep Mind’s program for predicting the structure of unknown protein molecules. The key innovation was creating a program that could not just learn from examples of known proteins, but generate predictions of what unknown proteins might be discovered. These hypothetical proteins were used as examples for another program to learn from, the way a teacher might come up with example problems to give to their students. It was only with this approach that the program was able to predict novel protein structures down to an atomic level of accuracy.

Personally, I think the reason that AI hasn’t spent much time considering an ACI framework is that it runs against the field’s most deeply held philosophical assumptions. If the mind is its own bounded cognitive universe, there is no principled reason why multiple minds would be needed to create intelligence. With Tomasello’s theory, it becomes clear why multiple minds and a mechanism for cultural learning is necessary. If people want AI to have more intelligence, what they need is not to build AI systems with a greater capacity for intelligence. It is to build them with a greater capacity to engage with one another. AI needs a mechanism for culture.

I mean, letting AI build its own culture. What could go wrong?

Subscribe now

Share

Here’s one way to think about the arc of social development in a society: Over time, the range of possible stories one can tell about oneself increases.

Think about who you are now, and what it would have been like for you to be that person a hundred years ago—or even before the rise of the internet. The set of stories you could have chosen to tell about yourself would have been far more limited.

Subscribe now

Sexual identity is a case in point. If you were gay in 1960s America or Victorian England or even today in a country with antiquated laws around sexuality and identity, then you would face severe limitations on how you could express that side of yourself. You could admit it and be pigeonholed by it (or worse). Or you could maintain a public fiction of straightness. Now, there are more options. Even in the last couple years, we’ve seen the widely-held norms around what counts as a culturally-accepted story around transgender identity change dramatically.

Today, for most of us, we have an incredibly large, if not functionally infinite, number of permutations to choose from to piece together a story about who we are and what we’re doing here.

The trade-off with this newfound flexibility is that it’s now up to us to construct our own story. Previously, you didn’t have to be a good storyteller to tell the story of yourself and tell it well. It was given to you, set out on a platter by your culture and your family and your role in society. Now it isn’t. You have to fend for yourself. That’s not something we’re equipped to do.

This is what makes the contemporary version of the paradox of choice so daunting. The places you could live, the partners you could choose, the career paths you could take—well, they’re basically unlimited. The problem is not just determining the value of these choices in a sense of personal utility. It is about determining the way they work together to form a coherent story about your identity and how it fits into the world more broadly. When faced with the possibility of being any one we want to be, we end up not knowing who we are.

It is a truism about humans that we are storytellers. What we often overlook is the fact that just as not all stories are created equal, neither are all storytellers. Sure, we’re all storytellers. But not necessarily good ones.

Share

One recent study which gives a nice illustration of this is called “Seeing your life story as a Hero’s Journey increases meaning in life” led by Ben Rogers, Kurt Gray, and their colleagues. The thrust of their research is that getting people to emphasize the heroic drama of their own achievements changes the way they feel about them. They call it a “re-storying” intervention.

The idea of the “hero’s journey” was first popularized by Joseph Campbell in his 1949 book The Hero with a Thousand Faces. (Though the term did not enjoy the ubiquity it does today until the 1990s). Campbell calls it the “monomyth”—the core foundation of the important myths and stories from cultures across the globe. Campbell’s account of the hero’s journey is actually rather complicated. There are the three acts of departure, initiation, and return—which is simple enough—but within each of those is five or six sub-stages, such as “The Meeting with the Goddess,” “Belly of the Whale,” and “Rescue from Without.”

Rogers and his colleagues streamlined Campbell’s hero’s journey down to seven elements: protagonist, shift, quest, challenge, allies, transformation, legacy. In essence, the protagonist is beckoned from the comfort of their home by a call to action. They’re helped along the way in their quest by an ally or mentor before encountering a Challenge, one which is just about the edge of their zone of proximal development. At length they overcome the Challenge, but just barely. Then they return home to find that not only is their own life better, the lives of their compatriots have substantively improved as well.

The nice thing about the hero’s journey is that it is flexible. You can pretty much shoehorn any set of life events to fit into it: “There was something I wanted; I went to go get it but something got in the way; then when I got it I went home again.” That’s Lord of the Rings in a nutshell. It’s also your daily trip to the grocery store.

The first set of experiments run by Rogers and his colleagues were correlational. In the first study, they developed a questionnaire to determine how hero-like an individual’s personal narrative is. They had people fill out this questionnaire, as well as previously validated ones associated with the perception of meaning in life. They found a nice, tidy correlation between the two.

They showed the same trend in two other data sets, where they had people tell their life stories as they naturally would. Then a separate set of people analyzed how many elements of the hero’s journey were featured in those stories. The more likely people were to naturally tell their story in a heroic manner, the higher they scored on the questionnaires about life satisfaction.

But what Rogers and his colleagues are clearly most excited about in the paper—as they should be!—is a set of experiments showing that if you get people to tell their own narratives more heroically, then they also are willing to endorse the experiences they’re describing as more meaningful.

Here’s what that looked like:

First they asked participants to tell their life story in two or three paragraphs—with no prompting on how exactly to do it. After they’d put down that first draft, participants were presented with seven prompts (one for each element of the hero’s journey) and then asked to consider how that theme might be brought out in their own life events. Afterwards they were asked to describe a version of their story in which they were like a hero on a quest.

The experiments also had a control condition, where participants reflected on their life—but without trying to shape it in any way to resemble the hero’s journey. The point of this was to get them to think about aspects of their life related to their identity, but without putting a narrative frame around them.

Across the board, participants who did the re-storying invention (versus the control) rated their experiences as more meaningful after the invention.

That said, the effect was pretty small. For example, one group of participants in the intervention condition rated their life’s “meaningfulness” just north of 5 (on a 7 point scale), whereas the control participants rated it just south of 5. So the effect was clearly there—but it didn’t result in a complete re-evaluation of the participant’s worldview or anything like that.

Still, the study shows that people can come to view their own life as more meaningful if they’re given tools to improve how they tell their own story.

Leave a comment

The authors of this paper call their intervention “re-storying” but in the literary craft there’s actually a term of art for this: revision.

In these studies, people are revising—if only temporarily—their life stories. This particular strategy for revision made people more likely to endorse the events of their life or career as meaningful. This, more than anything else, is how our default mode of storytelling differs from what the best storytellers do. The best storytellers revise.

The utility of editorial feedback has recently been something of a debate in my content feeds. A few weeks back, Erik Hoel ranked Substacks by famous authors. His conclusion was that without editors, even the best writers look a lot more like the rest of us mortals. On the other hand was a tweet from VC and essayist Paul Graham in which he said that editors are useless (specifically: “editors are not intrinsic to good writing.”) But here’s the thing—even in every one of Graham’s essays, he thanks at least three or more people who have read over his work and given him critical feedback. One way or another, good stories require revision. As Malcolm Gladwell once said: “I am not the greatest writer but I am the greatest rewriter; my superpower is rewriting.”

And yet, revising our own story of ourselves is surprisingly difficult to do.

There’s an Ezra Klein quote I really like on this front, from his recent podcast episode with Dan Savage (which, by the way, is probably my most-shared podcast episode of the year to date). From minute 36 or so, Klein says:

A lot of life is having a story that you believe of yourself and that other people believe enough about you that you can move through the world in a way that you recognize who you are and recognize how you’re seen and you’re okay with what you see in that recognition.

He calls this a “skill,” the ability to alter or modify that story—to revise it. This is something we as humans can all do, but not necessarily with the same level of acumen.

And I think this is what the hero’s journey paper shows us: it is an exercise in providing editorial feedback (albeit indirectly) for the participants’ life stories. The experiment says, “That’s a good first draft. Let’s try to tighten up the hero’s arc on the next one.” It’s a skill we could all probably work to shape up—whether we’re writers or not.

Get 50% off for 1 year

“What I cannot create, I do not understand.” — Richard Feynman,
the de facto mission statement of the cognitive scientist

When the field of cognitive science was founded in the 1950s, its architects had a clear goal. It was to develop a theory of the human mind so powerfully insightful, so comprehensively accurate that it could be implemented in a computer.

Unlike previous approaches to psychology, the point was not to understand the mind by drawing sophisticated diagrams of boxes with vague labels connected by arrows, nor to convey one’s theory in an incisive essay. No, the point was to show (not tell) that your theory was correct because you could construct something that, when presented with the same set of stimuli, would respond in the same way as a human.

The foundational belief of the early cognitive scientists was that whatever the mind happened to be, not only could we know about it: we could build one.

This is the mission of cognitive science I fell in love with as an undergraduate. I was captivated by the idea that if you really figured out how the mind works—if your theory was sufficiently true—then you should be able to demonstrate that knowledge by recreating a mind in the laboratory.

And it is in this vein that pretty much every cognitive scientist reads Feynman’s claim that engineering is the ultimate proof of understanding. What counts as “creating” a mind is to replicate its fundamental processes via computational models. If you get that right, then everything else can simply be added on top. Having a sophisticated computational model, in this sense, is like having a good soup stock on hand: one can always add in vegetables, seasoning, and other accoutrements later on, all of which are undoubtedly present in any given soup but are not the soup itself.

And yet a stock is not a soup: a model of the mind’s fundamental processes is not a mind itself. The pursuit of the computational cognitive scientist is to build a model of the mind in general. It is to create a paired down version of the mind’s most basic structures of thinking, remembering, reasoning, and understanding that are common to all of us. And don’t get me wrong. These are things worth knowing about.

But there is another way of reading Feynman’s call to action.

It is not to synthesize a version of the mind in its most general and abstract form, a kind of lowest common denominator of what we all share. It is to capture the mind in its specificity, to account for what makes the act of thinking—given one’s particular life experience and circumstances and personality and social milieu—so irrepressibly unique to a given individual.

An individual’s psychology is not made up of sterile algorithmic processing. It is inextricable from the experiences, the significant life events, the relationships with other human beings, the sources of meaning which give form to its function. Cognitive science, as it is current practiced, is happy to forfeit its claim to modeling this kind of cognition. Everyone gets everything they want. And for their sins, cognitive scientists will get a facsimile of the mind divorced from what makes our minds not just information processors—but us.

Luckily, there is a group of people whose goal is also to construct a facsimile of human behavior, but not the paired down one of underlying, fundamental computational principles. It is closer to what an individual human life looks like in all of its complicated glory. This kind of model cannot process novel stimuli, or tell you which images feature cats and which don’t, or compose insipid college application essays (all instances of a soup insufficiently seasoned). But it is nonetheless, as Feynman’s dictum requires, a way of showing one’s understanding of something by attempting to create it for yourself. The specific goal of this approach is to reconstruct the human experience from the point of view of the person experiencing it. And the people who create this kind of model of the mind and its consequences are not computational cognitive scientists. They are novelists.

Subscribe now

At first glance, novels and computational models couldn’t be more different. But they have more in common than you might initially think.

For instance, both require their creators to confine behavior to a highly-controlled realm, where all the levers and dials sit laid out for fine tuning by the person behind the curtain. For the novelist, this is a world that is made up, composed at the author’s god-like discretion. It is one which may resemble ours in some ways, though it will inevitably deviate from it in others. But the cognitive scientist shares the same requirement: the laboratory is itself a kind of fictional world.

To take in-lab behavior seriously requires suspension of disbelief. We must believe, even temporarily, that the only thing which matters is one specific influence on behavior. This is whatever is being experimentally manipulated. If that weren’t the case, if there were any number of things which could be influencing a participant’s behavior, then it wouldn’t be a very well-controlled experiment.

And though we know that’s not how things actually work, that behavior is always determined by far more than just one variable, we agree to set aside that knowledge for the sake of learning what the experiment (and the model designed to predict people’s behavior in it) can show us.

The fictional world of the novelist is also a laboratory, one of a different sort. Perhaps some of it is overtly unrealistic. There might be spaceships, or US presidents who never existed, or streets in London which aren’t actually there. No one actually did the things that are being described, at least not exactly as the events have been depicted by the author. But keep in mind: in a controlled lab experiment, no one would be taking those actions either if the experimenter didn’t arrange things just so. The job of the author, like the scientist, is to fix the parameters of the world at hand, then within those constraints create a realistic simulation of what, all else being as it is, might actually occur.

This is the implicit contract between the author and the reader:

If the reader accepts the fictionalized circumstances—the controlled variables which can be set up in a fictional environment but are not guaranteed, or even possible, in real life—then the rest of it will be real.

This is also the contract between the author and reader of a scientific paper, as well.

For both a well-designed experiment or novel, the promise of its architect is to create a scenario in which the behavior under scrutiny is uniquely revealed by the circumstances. If the reader accepts the limitations of the fictional world, the author promises to give them a privileged vantage into the topic at hand.

And, in a way, this is already how people naturally read novels.

They don’t view them as purely “fictional” in the sense of being wholly unrelated from reality. They view them as hypothetical versions of what really exists, the consequence of the author’s own simulation of how the world works. A cognitive scientist simulates behavior based on an understanding of the mind in terms of its most fundamental, decontextualized processes. And when a novelist simulates behavior, they do based on an understanding of a specific mind in the full-breadth of its social context.

When people read novels, they are implicitly trying to reverse engineer the generative model the author used to create these events. This is why people get worked up and emotionally-moved (or, alternatively, worked up and offended) by stories. The events are made up. But the stakes of the underlying generative model are real.

In a fictional world where anything can happen, what the author chooses to have happen says a lot about what they think does, in fact, happen.

Share

In my final episode of Cognitive Revolution, I presented a rough sketch of this idea to Stephen Kosslyn. Kosslyn is an eminent cognitive scientist, the kind whose name permeates any introductory cognitive psychology course. He’s been in the game a long time; he was Steven Pinker’s PhD advisor. We were talking about the nature of computational models, and what they can (or can’t) tell us about the mind, when I made this claim about novel being a kind of models of mind. “So let me ask you a question,” he said. “Let’s say that a novel is a construction of an understanding of psychology, writ large, in all of its ‘complicated glory.’”

“How do you know whether the author got it right?”

With a computational model, there’s a straightforward answer. The model is a good one if it predicts the behavior of human participants in an experiment. And if the model bares no resemblance to the behavior of your participants, then you’ve failed to provide evidence that it’s a good model. What the equivalent mechanism for evaluating a novel as a model of behavior?

When he first posed this question, I didn’t know how to answer it.

It seemed like an obviously worthwhile follow-up question. After all, it’s the most direct violation of the scientific method committed on a regular basis by novelists—lack of experimental rigor!

But I’ve come to think that it’s actually not the right question to ask about novels, because it’s not the right question to ask about models.

What changed my mind was another Cognitive Revolution interview, one with Sam Gershman, whose lab I worked in for two years. At one point in the conversation, I asked him about his general approach to modeling.

What he told me was that models aren’t about creating an unimpeachably accurate representation of the world. Rather, they are tools for thinking.

“Particularly when it comes to models, I don’t see models as literal descriptive statements of how the world works, or how cognition works, or whatever. The models, really—when they’re good—are tools for thinking. And they’re part of an intellectual dialectic that will eventually produce better models, which are better tools for thinking about some problem domain. And, of course, along the way you’ll produce models that are better descriptively. But I don’t really see that descriptive goal as really the primary goal of modeling, at least for me...

“In general, we have to approach our theoretical arguments with a sense of humility and not believe them too literally—treat them more as a conversation piece rather than a literal statement about the world.”

The measure of a model is not its ability to simulate the world as it really is. Instead, it is how well it articulates a way of how things might possibly be. Computational models are a means of outlining the space of possible worlds, any one of which we may happen to find ourselves in—or not. A model is good to the extent that it can help you think more clearly about the topic at hand [1].

This changes the frame for evaluating a novel as a model of human behavior. The relevant question, then, becomes not “Did the author construct the most accurate picture of reality?” but “Does the author give the reader a potentially-useful tool for thinking about reality?”

If the author has done her job—and we as the reader have the skills to figure out what we’re looking at—then this rendering should be a useful tool for helping us think through a particular aspect of reality in a way we would not have otherwise been able to see as clearly.

Think about it this way: If every painting had to be realism, there would be no impressionism, surrealism, cubism, expressionism. Our “pictures” of the world would be more accurate, strictly speaking, but our set of tools for representing and thinking through reality would be smaller.

“All models are wrong, but some are useful”.

This is a quote originally from George Box, and a refrain that’s nearly as common in cognitive science circles as Feynman’s.

My goal in making this argument is not to suggest that computational models from cognitive science are any more wrong or not-useful than anyone within the field might personally believe. Rather, in acknowledging their wrongness, as well as the limits of their utility, I would like to indicate beyond the usual scope of what is considered within the space of legitimate models of the mind. In particular, novels are good for modeling the one thing that computation never can: experience.

Leave a comment

There is a reason that novels, as models of experience, are not verifiable. It is that experience itself is not verifiable.

All any of us has is our own fiction. What is “true” for any of us, ultimately, is not the events themselves. It is that each of us lives our own version of them. Novels give us a template for how construct our own narrative, our own way of telling the story of who we are. Whether or not that way of telling it is generalizable in some pan-human sense of fundamental cognitive processes is beside the point. The only universal human fact is that we are all living our own complicated human experience, and it’s up to us to make sense of it.

Still, I understand why scientists don’t like novels, why they’re not an obvious object to incorporate into scientific protocols. They lack generalizability. They are not verifiable in the positivist’s sense, and for the most part they are immune to falsification. They do not lend themselves to replication; unlike the sleek generative models of the cognitive scientist, the model of the novelist is a single simulation affair. The novel offers little in the way of an ability to sample from a large number of probabilistic iterations in order to determine what happens on average [2].

But nonetheless, I can’t help but feel the endeavor of cognitive science will remain incomplete until it has a means of dealing with experience.

My proposal is not for cognitive scientists to forfeit their computational models and devote themselves to literary criticism full-time. But rather my claim is that the scientific method alone isn’t enough to reckon with the entire scope of mind, meaning, and experience. Those who are interested in studying the mind need not abandon the scientific enterprise. But I do think we need to take seriously the methods of interpretation, traditionally the purview of the humanists, if we want to deal with questions of human meaning. Without this, cognitive science will one day find itself in possession of a bowl of something warm and hearty, filled to the brim with a base of undoubted quality, as well as a lot of the ingredients you’d hope to find. But it’ll be missing something. The soup will forever remain slightly off.

Get 50% off for 1 year

[1-2] See comments for footnotes

“There are no facts, only interpretations”—also me,
because my interpretation is that I said this, not Freddy Nietzsche.

Being a scientist, as a matter of identity, is not about what you know. It is about the questions you ask, and the methodology you use to answer them. Talk to any scientist about what they do—this is probably what they’ll bring up first. Why? Because what you know depends in large part not only on what you’re interested in figuring out but also on what constitutes a legitimate attempt at doing so. With any methodology, there are things it can tell you and things it can’t. This also applies to the scientific method as a whole. Scientists, for example, are usually comfortable making the claim that science cannot give you answers to moral questions. Science can’t tell you about meaning, either. Meaning cannot be discerned through empirical experiments. Meaning requires interpretation.

This makes meaning almost fully intractable from a scientific standpoint, because this approach is very different from empiricism—which is how scientists are used to thinking. Empiricism, in the way that I’m using it here, means that whatever you want to know is best figured out through running experiments. Experiments can’t tell you anything about meaning. When it comes to meaning, the only thing experiments can detect are empirical trends in how people tend to report on meaningfulness. But that’s not studying meaning itself; that’s just outsourcing the labor of interpretation to someone else. If, unlike scientists, you are in the business of trying to understand what things means, then interpretation is the tool you must use.

Many people will be familiar with the famous distinction made by C.P. Snow about the two cultures—the sciences and humanities—which is often invoked in lamentation of how nice it’d be if we all just got along (or at least read one another’s papers once in a while). But asking scientists and humanists to share a common culture is like wishing that Argentina shared a common culture with Uzbekistan. They are separate cultures because they cover separate territories. They are different cultures by virtue of their interest in fundamentally different parts of the world—with different concerns, different ways of looking at things, and different stories how they came to do whatever it is they’re currently doing. It is in the nature of Argentina and Uzbekistan to be separate jurisdictions. The separation of the two cultures, likewise, is not at heart a distinction between two “cultures.” It is a distinction between two methodologies.

I think it would be okay if the two cultures/methodologies were simply distinct entities. (There’s no rule that says Argentinians can’t just travel to Uzbekistan if they want to see what the Uzbeks are up to these days.) But that’s not the case. Scientists get really, really offended by even the mere presence of interpretation. It’s as if they feel that once they let anyone with a different way of doing things into their territory, they’re suddenly going to be overrun with intellectual immigrants—and things will just never be like they used to! The scientists just can’t stand the idea of dealing with problems of meaning. It’s a real shame.

Subscribe now

Interpretation is really cool, and I don’t know why you don’t want to do it.

More specifically: the people who use interpretative methods to get at questions of meaning are mostly individuals who scientists A) can’t stand, B) have never heard of, or C) think maybe are okay, but have absolutely no bearing on their professional interests.

The standard-bearer of Category A is Freud. There is no one who is more associated with psychology yet absolutely despised by those who practice it (the experimental version of it, anyway) than Vienna’s most famous fondler of the mind. But I think that’s a loss for the field. The entire premise of Sigmund Freud’s enterprise—hairbrained though it was (and which, to enumerate its hairbrainedness in full, would require whipping out a Borges-size map with the same dimensions as the territory itself)—was to use interpretation to address questions of meaning. What was the meaning of a dream? A slip of the tongue? A particular cultural institution? These are all questions Freud felt, deep in his tingly, Oedipal loins, that he could answer with the aid of interpretation.

My favorite candidate for Category B (the Ignored) is the anthropologist Clifford Geertz. Anyone familiar with the behavioral sciences will know the work of Danny Kahneman and Amos Tversky. They are among most influential researchers in the last half century of empirically-informed cognitive, social, and behavioral sciences. Their most influential paper (on Prospect Theory) has, at the time of this writing, around 75k citations. That is an F load of citations. But what about the Interpretation of Cultures, published at almost exactly the same time, by Clifford Geertz? More than 140,000 citations—almost twice as many as K&T! (And in a field where people publish papers at a MUCH slower rate.) Geertz is unequivocally foundational in the qualitative social sciences (e.g., anthropology, sociology, any field that does interviews, etc; if you’re not doing what he said you should, you at least need to explain why). He literally wrote the book on interpretation, how it might be used in a scientific context, and doing so might allow you to discover. Yet most cognitive scientists have never heard of him.

Category C (the Irrelevant) probably has some overlap with Category B, depending on whose work you have come across or happen to admire. But pretty much any literary critic or narrativist historian—one interprets books, the other events on the historical timeline—of good repute could be included here. For example, the critic Harold Bloom is pretty well-known as a scholar, and I know more than a few scientists who have read his work. Then there’s the Metaphysical Club by Louis Menand, a narrative-based history of ideas in American and perhaps my single favorite book I’ve ever read. That’s also a book that’s gotten good reviews from every psych-adjacent person I’ve convinced to read it. And maybe you’ve come across Jill Lapore in the pages of the New Yorker or on her Pushkin podcast. (Okay, now that I think about it basically anyone who writes for the NYer is in the biz of interpretation.) These kinds of people tend to be respected by science-y types. But you know what? So is Dr Suess! (Who is not actually a PhD.) All of the above are great at what they do; it’s just not science.

I would also submit that this is why another instance of Category A (the Reviled) are popularizers of science. Whenever I tell people something to the effect that I’m more interested in telling stories about science than doing science myself, many will respond, “Okay, so you want to do science communication.” Actually, no. Not at all! You know who is a great science communicator? Ed Yong. He is potentially the greatest science communicator in the game at the moment, and his goal is to communicate scientific findings as they are understood by the people who found them in a clearer way than the scientists themselves ever could. In other words, he’s not trying to go beyond the science—for example, to extrapolate the results of a study into a point about how you should apply it in your own life. He’s all about the facts. What most pop science writers do—the ones who many scientists really seem to despise, like Malcolm Gladwell and Steven Pinker—is not science communication, but science interpretation. They are not just here to tell you the facts. They’re here to tell you what they think the facts mean. And that to me is way more interesting!

But because meaning is not part of the scientific purview, no matter how these popular authors construe what they think of the facts—scientists will take issue with them. This doesn’t mean the authors are wrong. What it means is that they’re doing something that’s not allowed within the rules of the scientific game: interpretation. And so when what they’re talking about is represented as “science” the scientists are like, “Woah! Hold on! That’s not what we do at all!” Exactly. You’re interested in the facts. And that’s cool, we appreciate that. But the rest of us are also interested in something else: we want to know what the facts mean. For that, you need someone who is willing to wade into the murky, empirically-unverifiable waters of interpretation.

Share

All this rests on the question “What is meaning?” — which I really don’t want to answer.

Okay, so I’ve made the claim that experiments can’t tell you anything about meaning—only interpretation can. Why? First of all, let me say this is not an easy question to answer. The humanities (and the qualitative social sciences) have been grappling with it for, like, a century. And honestly, I don’t think it’s gone uniformly awesome. Scientists have stayed away from problems of meaning. Their project has had some issues as well (many of which, by the way, are problems of underappreciating the importance of solid interpretations of their data). But with science, it’s easier to point to a trend that looks like progress!

The basic answer rests on my previous argument about how meaning is contextual. In short: there’s no such thing as objective, de-perspectivized meaning. And that’s precisely what science can’t handle that. It can’t study things in fun-sized, individual packaging. This one-and-done approach is labeled as “idiographic.” But given that that’s a term that even I, in my love of obscure and academic-sounding terminology, think feels ham-fisted and overwrought, we can call it N=1. This, of course, is scientific shorthand for something which doesn’t count as science. A study with only one participant isn’t a study. It’s an anecdote!

Let me try to express it this way: because meaning is always based on an individual perspective, all you have to do is change the person looking at it and the meaning changes as well. That’s, like, the main thing that you can’t have happen in a solid empirical experiment. Let’s say I construct an experiment designed to “detect” meaning (whatever that would entail). In science, you would want it so that I can show the results of that experiment to you, a different person, and you, too, can see the same results for yourself. Even if you had a meaning detector, it would buzz for different things depending on who was holding it.

By contrast, science—as typified by the grand, hold-the-universe-in-your-palm-sized-equation school of physics—is “nomothetic.” In the best case scenario it is N = all of the above. That’s what the scientific method strives for. You’d have your measurement, and even if everyone in the world had a look at it they’d all see the same thing. I know I’m not giving a very nuanced picture of the scientific enterprise here, but people write entire books on the basic concept of objectivity. I’m here to make extravagant claims now, ask questions later. So my point is this: the theory of everything would be a theory of everything but meaning. (Fine. Morality can come, too.)

Another important thing here is another thing I already said: meaning is post-hoc. There is no way to predict ahead of time what the meaning of something will be. It may seem like there is. But there’s not. We can detect empirical trends between what someone does and what they report as meaningful, but that’s not because it’s reflecting how meaning works. That trend, as I argued, reflects how we all have a propensity to reach for a fairly limited number of culturally-endorsed stories about what constitutes meaningful activity.

One of the things I tried to convey in that post was that the creation of meaning resists statistical regularities in a way that’s different from other activities. Because meaning is an interpretation—it depends on perspective; it is not objective; etc—the role that an event or experience has can change instantaneously. So don’t get me wrong: there are definitely things that we’re told via our culture that are important and that we should construct meaning from them—like a degree, or a kid, or a book that really changed the way we see things. But whether or not we actually do that is up to us. As a consequence, there are statistical facts about what people endorse as meaningful or not, just like there are statistical facts about anything that occurs in the world. It’s just that these statistical facts don’t necessarily have any bearing on what’s happening for any specific individual considering their own specific experiences. (Actually, I dunno. That kinda sounds like how statistical facts work in general? Maybe that isn’t so different. Still trying to figure that out, I guess.)

Anyway, what I should be really excited to argue for, but for some reason am not, is that experiments can’t detect meaning, fundamentally, because of what meaning is. Or rather, where it lives. Meaning lives in symbols.

This is the explanation given my favorite account, the one from Geertz. So what exactly is a symbol? Well, err... it’s a thing that contains MEANING.

These words—the ones I’m using here!—are symbols, within the symbol system “English.” A crucifix is a symbol within the symbol system “Christianity.” A Birkin Bag is a symbol within the symbol system “socioeconomic status” for some, while for others it resides in the system of “Huh?” A Balinese funeral rite is a symbol within the symbol system “Balinese culture.” One’s middle finger is a symbol within the symbol system “American diplomacy.”

Each of these symbols is meant to be interpreted in terms of its role in the larger system it’s a part of. When you leave the system—for example, when you don’t know anything about designer handbags or Balinese funerals—then the symbol’s meaning evaporates. (Oh, hey look! Meaning is still contextual!) In a way, this is the inverse of experimentalist’s ideal. Interpretations are only N=1. There are no N=all interpretations. You have to look at the specific symbol within the specific symbol system from a specific point of view within that system. So, yeah. The reason that I should be excited to make this argument is that I love symbols. The reason I’m apprehensive about it is that I’m not sure that I’m not sure I’ve explained anything or made things any clearer. I’ve been studying symbols—first as a cognitive scientist, then as a Geertz disciple; and the symbol “symbol” has slightly different significance depending on which system you’re in—for like a decade now. I’m STILL not sure what they are.

This is, perhaps, part of a larger MO. The thing is, I really hate definitions. I know it sounds ludicrous. But I really don’t think they’re helpful. I feel like when people starting arguing about definitions, then everything becomes about definitions and it’s like, for fuck’s actual sake. Who cares? And “meaning” is the worst offender. “What is the meaning of meaning?” is the only question more grating to me than “What is the meaning of it all?” It’s just not a game I’m interested in playing. So here. You want a definition of meaning? Fine. I’ll leave you with this.

Meaning... IS... interpretation...!!!

Leave a comment

Everyone loves a third culture but no one actually wants to be part of one.

Louis Menand once wrote—in an essay I really spent a while trying to find but couldn’t and now have to move on because I have other things to do with my life… but here’s one of his NYer pieces on Freud instead!—something to the effect that there are really three cultures: the scientists/experimentalists, who are interest in what there is; the humanists/interpretators, who are interested in what things mean; and the social/behavioral scientists, who are interested in a bit of both.

But I actually think he’s wrong! I don’t think this “third culture” based on the behavioral sciences is actually really a thing in any substantive capacity! But even more than that, I think it should be! I think the scientists shouldn’t get angry when people try, in good faith, to overlay interpretation on top of empirical findings. I get why scientists feel it’s not their purview to do so themselves, but I think it should be part of the scientific pipeline: eventually, we combine our knowledge about what there is with an attempt at grappling with what it means.

I think maybe—ultimately, eventually, in the fullness of time—the overlap here has to do at the intersection of everyone’s mutual interest in symbols. Cognitive scientists, as I alluded to above, love symbols. It’s like their thing—what distinguishes the way they think about the mind from the way other thinkers, with similar interests, think about it. Gary Marcus has an entire Substack about how collections of neurons can create symbols but collections of neural networks can’t! (Okay, that’s not the Substack exactly, but it’s kinda the thrust of his book, The Algebraic Mind, from way back.) And computer science also has a definition, perhaps the most precise one of all, of a symbol. Computers can deal with meaning. Not, like, the full scope of it, in terms of crucifixes and Balinese funerals. But in terms of taking fundamentally meaningless, arbitrary components and combining them in such a way to create a system where symbols can be manipulated. It may not be the most flexible sense of the word, but commands in a computer programming language still mean something—at least within the context of the larger system they’re a part of. Even more exciting: You know whose job it is the do this interpreting? To put together meaningful arrangements from meaningless components? The homunculus eating popcorn in the Cartesian theatre of your MacBook? It’s called an “interpreter” :D

(Strictly speaking it’s not that simple... But it’s also not not that simple, either!)

Anyway, I do think this suggests a way forward for the third/culture. Meaning is interpretation. But what is interpretation? In a word, it’s an algorithm. Now THAT’s the kind of thing scientists can wrap their heads around.

Get 50% off for 1 year

Leave a comment

“Life: lived forward, understood backward”—Me,
paraphrasing my recent paraphrase of Kierkegaard

Happiness is a problem of prospection. The trick of being happy is to figure out what you can do now in order to be happier later on. Is ordering that dessert going to make you happy while eating it? What thirty minutes afterward? What about two days? Would you be happier putting aside some extra money for retirement, or taking a trip you really want to go on? These are decisions faced by present-you for which future-you will bear the consequences. When it comes to happiness, it’s possible to perform a kind of empirically-informed psychological calculus to determine which course of action is most likely to lead to the greatest long-term accrual of dopamine. Meaning is different. It is not prospective. Meaning can only be evaluated in retrospect.

It may not feel like this is true. “But wait!” you might say. “My powers of prospection totally work for meaning, maybe even more than for happiness. I know what will be meaningful, and what won’t, before I do it.”

As it turns out: false. This is an illusion. It’s not that there is no empirical connection between what people do and what they claim to find meaningful. This, after all, is the kind of thing the psychologist’s tool kit is well-suited to uncover: have someone do X and ask them if it makes them feel Y. No, my claim is something more specific. You can’t discern the meaning of future events because that’s not how meaning works.

Asking someone to tell you the meaning of something—an action, an experience, a relationship—is like asking them to tell you what a book is about without actually having read it. And you know what? That’s totally something they can do. It is how I got through AP English in high school. I never read a single book. I read the SparkNotes summary (or at least part of it). Then I discussed the “big themes” of human nature. Your English teacher isn’t evaluating whether you’ve understood the meaning of the book, because meaning isn’t something that can be right or wrong (like a math problem), or present or absent (like an increase in dopaminergic activity), or even externally observable (like what you can find with a good science experiment). They’re looking to see if you can successfully approximate something that resembles our culturally-endorsed stories about meaning. From the outside, all we can see is whether something looks like meaning. We can’t actually tell if it’s there. This makes it really tricky to study.

So yes! It’s possible to say “here’s what I think this means” and draw on easy, forecastable answers. But if we’re gonna take meaning seriously—and on this Substack we are—it turns out a legit attempt at meaning-making can only happen, as Kierkegaard points out, by looking in the reverse direction of the temporal arrow in which it originally took place.

Subscribe now

This seems wrong. Can you please give an example?

Before getting into my argument about the mechanisms of meaning, I want to start with a couple examples. When we look at generic instances of potentially meaningful or non-meaningful behavior—e.g., having kids vs sitting at home smoking weed and playing videos—it seems like a no brainer. There are some things you can do which are just fundamentally more meaningful than others. But that starts to change when we look at specific instances.

On my podcast, I end each interview by asking my guests which three books have most influenced their thinking. By far and away, the most common answer is Victor Frankl’s Man’s Search for Meaning. So let’s take that at face value, and assume that Frankl gives some useful or important insight into the way meaning works or impacts our lives. What I find strange about people’s adoration of the book is that I’ve never heard anyone actually talk, on my podcast or in any context, about the core argument of the book: a school of thought Frankl called logotherapy. What people appear to be interested in is how he came to relate to the meaning of his experiences. And what was the epicentral experience he relates in that book? Producing offspring? Creating a lot of value for his shareholders? No. The book is about his time… in a concentration camp.

So go ahead. Play out that argument. Please explain to Dr Frankl about how meaning is eminently forecastable. “Look, Dr Frankl, this concentration camp thing—it’s going to be great for your literary career. Fantastic fodder for a book. I think you’ll find it very inspiration, you know, in terms of intellectual engagement.” To me, it seems like three things are true simultaneously: Nazi camps were an atrocity and shouldn’t have to be experienced by anyone; going through it was also a profoundly meaningful experience for Viktor Frankl, which he was able to render in a way that meaningful to many others; and it wouldn’t be possible to predict ahead of time whether the former would causally lead to the latter.

You can also see the retrospective nature of meaning by look at great works which went unappreciated during the artist’s life but were determined to be of tremendous value after it. One instance you probably have heard of is Vincent van Gogh, who sold only one painting in his life. Evidently, his father had the sale arranged, in order to make his son feel better. (Being a painter and not selling you work is bad enough. Not only was that the case for van Gogh, but his brother was also an art dealer. Like, how bad would you feel if your brother was an art dealer and you still couldn’t sell any of your work?)

One you may not be familiar with, though, is Zora Neale Hurston, the anthropologist, folklorist, and author of Their Eyes Were Watching God. She was a participant in the Harlem Renaissance in the 1920s, friends with Langston Hughes (who experienced a large amount of success very early on in life), and was pretty accomplished in terms of her published work. But then, after her death, she was mostly forgotten about. It wasn’t until Alice Walker rediscovered Hurston’s work in the 1970s that she became the staple of the high school literary canon she is today. (Thank god I wasn’t asked to read that book in high school, so I didn’t develop an aversion to it; I didn’t read it until my mid-twenties, and I frequently cite it as my favorite novel.)

Obviously, you could pick any number of “great works” which only became great works after the artist’s death. What can we learn from that? We don’t know how meaningful a work is before it’s created, when it’s created, or even potentially for some time after. There’s always a post-hoc story about why the work is intrinsically deserving of its place in the canon. But that’s the point. You can’t tell the story until the end of it has taken place—where usually in this context “end” refers to the conclusion of the artistic or social period.

Final example: drug addicts. The logic of this example is similar to the Frankl one above, so if you found that one grating you’re not going to like this either. But if you listen to how recovering addicts talk about their life and their experience, often it’s being on the other side of addiction (or the unterminating pursuit of recovery) is what gives their life meaning. They’re not looking forward to escaping from the experience entirely so they can move onto something else. A lot of them like to talk about it, over and over again. Take Dax Shepard, for example. If you’ve ever listened to an episode of his podcast Armchair Expert, then you know he’s been in recovery for, like, 15 years. Why do you know this? Because he talks about it on pretty much every episode. Why does he do this? Because clearly it’s one of the most significant sources of meaning in his life. But this, like Frankl’s, is a story it’s only possible to tell in retrospect. You’d never ask someone who is beginning to dabble in heroin why they decided to give it a shot and have them say something about how they were really hoping to increase the meaningfulness of their life, long-term. That’s because meaning-making is a retrospective act. It can only be made post-hoc.

Share

Now I’m ready to hear about the mechanism. Tell me why this post-hoc business is the case.

When we look at generic examples of meaning, it seems that there’s a fairly small and reliable set of activities we can engage in which will bring us greater meaning than others. But the examples above suggest that that changes when we look at specific instances. The reason for this is that the generic examples reflect our stories about meaning at a culture level, while the specific ones reflect our stories at an individual level. These are shaped by one another (especially individual stories by the larger culture they’re a part of). But there’s nothing that says they have to be the same. And in fact one reason they’re different is that often a great source of meaning for people is precisely where their personal story deviates from that of the broader culture in which they’re embedded.

This is one of our standard strategies for meaning-making: appealing to the collective sociological imagination. There are meaningful stories we know how to tell because our culture gives us templates for telling them. It is precisely because the above examples aren’t part of this culturally prescribed set of approved endeavors that they can be so powerful. There’s no template in our culture which suggests that the route to a meaningful life is via an extended stay in a death camp or drug addiction or spending your entire life creating something with no immediate recognition of its value. These specific stories derive their power from this deviation from the generic cultural templates.

But, okay… Well, that’s not entirely true. They do all conform to a template, at least kinda. That’s the hero’s journey, popularized by Joseph Campbell’s The Hero with a Thousand Faces. Each one of them set off from the comfort of “home,” went through an extended series of trials in the wildness, then emerged on the other side victorious. As always, though, the problem with universals—and I do think it’s fair to say that the hero’s journey is at least a quasi-universal when it comes to storytelling—is that it is so abstract as to not be particularly instructive. It’s like the search for cross-culture universals in music. They totally exist! But what are they? Things like: okay, all music has sound... and some of that sound is pitch... and some of that sound is rhythm... Great, thanks. You just described absolutely nothing that’s really interesting or important about any specific piece of music. Likewise: we can’t use the hero’s journey as a template for our own story... until we’ve already undergone our hero’s journey!!

The other way to see this is to look at people who are really invested in being more intentional about meaning, e.g., good writers of novels, criticism, memoir, etc. Because they’re less likely to reach for the generic templates of meaning, they can show just how wide a range our conclusions about what’s meaningful can be. For example, my favorite literary influencer, Becca Rothfeld, has written extensively about the cultural story we tell about how having kids is the most meaningful thing you can do, and anyone who doesn’t subscribe to that belief just doesn’t understand the prompt. She argues convincingly that, sure, some people find reproduction really meaningful. But beyond the (general) biological inclination to engage in it, there’s nothing that says a life not given to the pursuit of bearing offspring isn’t meaningful. To say otherwise is incorrect. It’s also condescending to people who make different choices than the child-bearers.

Also in terms of jobs: A lot of us thinkitty-think type people who want to jobs that create something of value—like blog posts; ha ha ha—look at more blue-collar, manual labor jobs and think: “crucial to society, no doubt! But meaningful? Probably not.” But it turns out that if you ask people who have those jobs about whether they find them meaningful, the answer is way more complicated. It’s complicated because they still recognize that their job is low-status within our society and therefore not likely to be viewed as meaningful by other people. Nonetheless they find in their work not an absence of meaning but a huge amount of meaningfulness in the way their job contributes to their society, their community, and their family. This significance is primarily grasped through the concept of “dignity,” if you are to believe the argument of sociologist Michelle Lamont in her excellent book Dignity of Working Men, which I very much do.

Overall, the point is that, despite what it may seem, there’s no such thing as activities that are intrinsically meaningful. You may think there are. But that’s only because most of us tend to draw on a culturally-prescribed set of predetermined story templates. When it comes to our own specific experience, there are only activities to which we’ve retrospectively assigned meaning.

Get 50% off for 1 year

I don’t think the endowment effect is a bad thing, and I think we should be less effort averse.

In behavioral economics, the “endowment effect” is when an individual assigns a higher value to an object simply because it’s theirs. Mostly, researchers view this as an example of human irrationality. Whether or not you’re currently in position of an item does not, in an objective sense, alter its value. If someone offers you an exchange for a better item, you should take it—regardless of which one was originally yours!

But I think there’s another way of viewing the endowment effect. When we come into possession of something—a car, a relationship, a job, a towel, a cup of coffee, a small dog—we naturally, if subliminally, begin to tell a story about how we came into possession of it and why. We begin to assign meaning to it. Were we to come into possession of something else, we’d also with equal earnestness assign meaning to it as well. But we can’t begin to assign meaning to that thing (or at the very least it doesn’t occur to us to do so) until after it has become “ours.”

Another way to think of this is through the recent line of work from the Toronto-based lab of Mickey Inzlicht. A few of his recent papers have looked at the connection between effort and meaning: essentially, the more effort we put into something, the more meaningful we are likely to find it. This is in line with the hero’s journey template. The hero usually has to, among other things, put in quite a bit more effort than she would have if she just decided to stay at home. Inzlicht and his co-authors point out that this contradicts our usual interpretation of effort, which is that people want to avoid it. Given the choice between two paths to equivalent outcomes, people will tend to choose the easier one. It turns out that if you make people take the harder one, they’ll report that they found it more meaningful to do so.

This goes beyond Inzlicht’s empirical findings, but here’s my interpretation of what’s happening here. We humans tend to avoid hard work. So in prospect, what sounds good is the path of least resistance. But, as we have established as an unassailable truism throughout this post, meaning is post-hoc. Hence, in retrospect, what we’re most proud of are the tribulations we’ve overcome.

I think this gives us a clue about where to look for meaning: in hard things. This doesn’t guarantee the presence of meaning, and it certainly doesn’t guarantee what that meaning might be. But I do think it suggests that when we’re at a crossroads, the road that we know will be harder will also be the one which is most meaningful. Does that suggest you should go and get yourself addicted to heroin, just so you can talk on your podcast about kicking the habit later on? I don’t think so. But it does give you a lot to work with. It does not change the fact that the way meaning works is fundamentally post-hoc. It’s a good place to start, though.

Leave a comment

"Life isn’t about the answers you give. It’s about the questions you ask."—my mother

This is a claim my mother makes from time to time, about the supremacy of inquiry over conviction. I’m not sure she’s entirely right. I think some answers are pretty important. But there’s no doubt that the kind of answers you reach depend a lot on the kind of questions you ask. And among the most unhelpful, most misleading questions that we humans are inclined to ask is: "What is the meaning of it all?"

This question is often posed in a way that juxtaposes the smallness of our own actions against the largeness of the grand void of space. It is usually paired with an illustrative metaphor. We are but an atom in a mote of dust on desk of God’s rarely-used home office in the guest annex of his vacation universe etc... So how could any of this thing we call existence have any meaning?!

But this kind of bottoming-out of meaning—a definition of meaning in the ultimate context of material existence—represents a fundamental misunderstanding of how meaning works. It’s a valid question about human meaning in the same way Noam Chomsky’s famous “colorless green ideas sleep furiously” is a valid sentence in the English language. Sure, you’re allowed to put those words together in that order. But doing so isn’t going to get you very far in the way of conveying anything especially worthwhile to yourself or anyone around you.

What this question misses is the fact that meaning is contextual. When you leave the context, the meaning dissipates. There is no legitimate definition of meaning where this is not the case. Therefore, when the premise of your question is about meaning in a scope where, by definition, you’ve transcended the context of everything that exists... of course there’s no meaning there!! You might as well be surprised there’s no oxygen for you to breathe out there, either. Both would be a really big problem, if you ever found yourself on the outer reaches of our galaxy.

And let me be clear. This doesn’t imply the meaning is an illusion. It’s not a nihilistic argument. It’s a fact about the way meaning is constructed. Meaning isn’t something you can put in a bag and weigh up, like all the atoms in the dust on God’s guest-room desk. It is something that is created in the mind. Specifically, it is something that requires both a figure and ground, an object and backdrop. Changing the background changes the object in fundamental ways. Let me give you a few examples.

Subscribe now

Exhibit A: Language

I already invoked the name of Noam Chomsky. I recently did a podcast episode with one of my favorite cognitive science of language researchers, Gary Lupyan, and the title of that episode was “How words get their meaning.” If you were going to put the answer in a single word, what would it be? Context!

As Michael Tomasello notes in the opening passage of Constructing a Language, among the simplest utterances we have in human language are the partial phrases used by children, such as “more juice” or “doggie gone.” And yet it is an astounding fact that even these phrases cannot be understood by 80% of our conspecifics. Most people don’t speak English. These words, simple and expressive though they may be to us, don’t mean anything to them. The most basic fact we have about language is that there are different ones. And it’s clear to anyone who has ever traveled abroad that there is no way to import meaning from your own language into someone else’s if you don’t share the same linguistic background.

More than that, a shocking amount of meaning in language—complex meaning, not just more doggie and juice gone—can be understood in terms of context. Specifically, which words appear in close proximity to other words. This is literally how ChatGPT works (at a super basic, overly but usefully simplistic level; for a non-specialist’s overview, see the second half of this Ezra Klein interview). It processes a huge amount of text, and learns which words tend to show up next to which other ones. It turns out that if you do that really, really well for a hugeee amount of data, then you get a system that really seems like it understands language.

And this doesn’t just go for natural languages as well! It’s the same thing for computer languages. If you have a line of code that does something useful in one language, then simply copying and pasting in an effort to recruit its algorithmic utility in another language is almost certain to fail.

Exhibit B: Religion

Okay, you might say, all that linguistic stuff is fine. But mostly when people talk about “the meaning of it all” they’re talking about God. Religion is what, for many people, defines the ultimate context of all meaningful endeavor. In a universe with God, you get a grounded, unwavering definition of the meaningful (that which contributes to or is in accord with God’s plan). In a universe without God, you don’t have that. You have just atoms and dust—and it’s all just a haphazard, inert arrangement of matter. Right?

Wrong. The thing that no one ever talks about is what happens when you ask about what makes God meaningful? Like, what happens when you take a step back and ask what the point of God’s plan in an even grander scheme? Why does the whim of one deity—allegedly omniscient though he may be in our physical universe—suddenly suffuse everything with value? The answer is that when you begin to go down this line of questioning, you’re back to meaninglessness. God only provides you ultimate meaning if you define Him as the arbiter of ultimate meaning. If you continue to play the “But what does it really mean in the grander scheme?!” game, you’ve still got the same problem on your hands. Meaning is contextual, even when the context is God.

Exhibit C: Poetry

My favorite poetry book is a work called Love, an Index by Rebecca Lindenberg. It’s hard to convey just how fantastic, accessible (even to someone, like me, who knows nothing of poetry), and perfectly crafted this collection of poems is. It’s meant to be read front to back, like a book but in verse rather than prose. The first thing you learn is that it’s dedicated to someone—presumably a lover; you don’t have to be Sylvia Plath to notice the word ‘love’ is in the book’s title—who has died. From there, the poems seek to capture the texture of falling in love, from the first moments of attraction to the well-worn grooves of an entrenched relationship, where the form of the poem is meant to match the form of the feeling. Soon, you begin to get glimpses of what happened to this lover and the circumstances of his death. You can get through it in about a two, or two-and-a-half hour sitting. It’s perfect. Anyway, the reason I mention it now is that one of the poems (specifically the ‘index’ of the title) gives Lindenberg’s definition of poetry. Poetry is “how thought feels.”

All of this is to say, poetry is about as densely meaningful as any human endeavor can be. And how does meaning work in poetry? Well, in one of the most famous critical essays of all time, Tradition and the Individual Talent, T.S. Eliot gives an answer. Poetry is meaningful only in the context of other poems. “No poet, no artist of any art, has his complete meaning alone,” writes Eliot. “His significance, his appreciation is the appreciation of his relation to the dead poets and artists. You cannot value him alone; you must set him, for contrast and comparison, among the dead.”

So, based on that theory: is poetry meaningful because there’s something intrinsically meaningful about that particular arrangement of those particular words? No. It’s really just that the more “meaningful” a poem is—i.e., the more it’s able to convey, if you know what you’re looking for—the more it needs to be considered in the context of what’s preceded it. That’s why, for most of us, poetry is so hard to understand. We don’t know anything about poetry! And according to T.S. Eliot, what’s the one thing you need to know about in order to “get” a poem? Other poems!!

It’s context, all the way down.

Exhibit D: Music

For our final example let’s turn to music. There’s a book from the 1950s called Emotion and Meaning in Music by an author named Leonard B. Meyer. I returned to it recently, and it wasn’t great as a book because pretty much all of the details in it feel a little antiquated now. But it’s one of the main sources that psychologically-inclined musicologists cite when trying to make a connection between music theory (the structural study of how notes fit together, like pieces in a puzzle) and the perception of meaning (how those theoretical structures manifest in feelings and reactions from individual listeners). But the main argument, I think, still holds quite well: no individual note has meaning in and of itself. A note is only meaningful in the context of other notes.

In a way, this is just a rephrasing of the definition of harmony. Harmony is the sonic context of a song. In short, it’s the chords that are being played underneath the melody. And the reason the melody sounds a certain way—happy, sad, exciting, etc—is because the series of notes being played or sung as the melody has a particular relationship to the underlying harmony.

There’s no doubt that in another strand of the multiverse I became an ethnomusicologist/music cognition researcher. I studied this sort of thing a lot in undergrad, and even wrote my thesis on “computational models of jazz improvisation inspired by language.” So I could get really nerdy here. But just to take a (relatively) simple example: If you the melody is playing a C, and the harmony is playing a C major chord. Then that note is going to sound really stable, perhaps even have a certain finality to it depending on the chord that proceeded it. But if you play the same note, a C, and move the chord up a half step, to a C# major chord—that same note now plays a completely different role in harmony. It’s now the major seventh of the underlying chord: it’s dissonant, but in (potentially) a very pretty way. In general, it’s a musical signal that something nice, but not final, is happening. The way meaning works in music is not exactly the same as how it works in language. (That’s why we use words instead of music notes to communicate; but if you’ve ever read the book Hail Mary Project, which I keep recommending over and over again, then you know that there’s no principled reason why that has to be the case.) But nonetheless—where do musical notes get whatever kind of meaning they do have? Context.

Share

The general principle here is that meaning is what you might think of as a problem of “local knowledge.” There are no universal, scientific principles that can tell you what something means in the way that you can discover principles about, say, the motion of planetary bodies. The concept of local knowledge was introduced by the economist Friedrich Hayek in his super interesting 1945 essay The Use of Knowledge in Society. He writes:

Today it is almost heresy to suggest that scientific knowledge is not the sum of all knowledge. But a little reflection will show that there is beyond question a body of very important but unorganized knowledge which cannot possibly be called scientific in the sense of knowledge of general rules: the knowledge of the particular circumstances of time and place.

Hayek goes on to argue that some things just don’t admit to scientific scrutiny. For example, which of the nearby groceries sells the best produce? Or which one has the largest selection of imported foodstuffs? That’s just something you kinda have to live on that street or in that area to know.

Hayek says that “practically every individual has some advantage over all others because he possesses unique information of which beneficial use might be made”. It’s an essay about economics, and specifically the economics of central planning, which has a hard time making use of local knowledge. It’s not explicitly about meaning. But I think meaning works in the same way. (Not just my opinion: see Local Knowledge by my favorite anthropologist Clifford Geertz for a full theory of meaning.) No one can look from the outside, from some God’s-eye-view central planning agency in the sky, and say what about your life and your activities makes them meaningful. Only you can do that.

And more specifically, you can only do that… if you get the context right.

To put it concretely: One of the mistakes we make is to look at too big a picture when evaluating questions of meaning. We ask what makes our work meaningful or impactful on a “global” level. But more global you go, the less meaning you’re going to find. Meaning, in this sense, is about an activity or endeavor’s impact within a system. (That’s not the only way to define it, but it is a useful one.) The more impactful something is, the more meaningful it is. But when you look beyond that system and continue to ask about impact—of course you’re not going to find it!

The smaller our circle of meaning, the more likely we are to feel our actions are impactful.

One of my core life mottos is Bertrand Russell’s statement: “The good life is one inspired by love and guided by knowledge.” It is through love that we are driven to want certain outcomes in the world. And it is through knowledge we discover how best to achieve them. This also gives us an idea of what it means to ask the right questions, and how to begin looking for answers. Love, in the best case, is what motivates us to ask the questions that we do. It helps us to define our context. Knowledge is what helps us answer them.

When the scope of our question exceeds that which we truly care about then we’re no longer inspired by love. We’ve lost the thread of meaning. Love is what defines our context of meaningful engagement. Knowledge is what tells us how best to act within it.

The boundedness of meaning to a particular context is not something to shy away from, or to be disappointed by. It is something to embrace. It is also something to be cognizant of when evaluating how meaningful you feel your current activities are. Meaningful in what context? Get the context right. The meaning will follow.

Leave a comment

In the first half of the sixth century BCE, a Greek man named Anaximander, born in Turkey, sketched the world in a way no one had previously thought to do.

It featured a circle, divided into three equal parts. He labeled those parts Europe, Asia, and Libya, and separated them by the great waterways of the Nile, the Phasis river, and the Mediterranean. To call it a map would perhaps be a bit overgenerous. It was really more of a schematic. But it nonetheless represented a crucial innovation. Anaximander had rendered the world in a way that no person had ever seen it before: from above.

Anaximander’s sketch wasn’t especially useful as a cartographic instrument, and it also came with some peculiar conceptual baggage. He believed the earth sat atop a column, following the architectural sensibilities of the day. It was Plato, several hundred years later, who proposed the idea of a spherical earth. He had no strong, principled reason to do this; the guy just liked spheres. But though they were fascinated by the shape of the earth, neither Plato nor Anaximander, nor even Aristotle, can count among their accolades credit for creating the first scientific map of the world. That distinction falls to a North African man by the name of Eratosthenes.

Eratosthenes is best known today as the founder of the field of geography. He coined the term in his magnum opus, Geographika, published during his tenure as the head librarian at Alexandria, sometime between 240 and 220 BCE. It was three books in length and covered topics ranging from climate zones to the geological history of planet to the customs of different populations. What Eratosthenes offered in this book was a perspective informed by data. Any prior works that touched on these topics had been conceived in much the same way as Anaximander’s schematic: a vague suspicion about what was out there, unconstrained by any commitment to describing it in actuality.

We don’t actually have a copy of Eratosthenes’ original map (it probably burned down with the Library of Alexandria). But the Roman geographer Strabo, in the first century BCE, constructed his own map, which we do have, based off of Eratosthenes’ original—a sort of tidying-up, rather than a total rewrite. He quoted at length from Eratosthenes’ text, including detailed descriptions of the original rendering. So historians today have a pretty good idea of what Eratosthenes’ map probably contained.

In Eratosthenes’ map, the world’s landmass looks like a parallelogram. It is divided into four quadrants. They are, roughly, Africa, Europe, Russia, and India—or at least that’s what we’d call them today. As far as maps of the world go, it features—to make the point gently—some peculiar omissions. The Americas, Australia, and East Asia come to mind. Even in what Eratosthenes does try to depict, he fails to grasp how expansive the earth’s landmasses really are.

From our perspective, it’s clear that Eratosthenes didn’t know what the world looked like. But what’s a little trickier to explain is that it didn’t really occur to him that his map might have been off. In fact, he thought he had pretty much nailed it. He claimed, for instance, that the only reason no one had yet sailed around the world was because of unfavorable tides, and not for more consequential reasons—like the existence of the rest of Africa, China, and Russia.

In the famous first line of his Tractatus Logico-Philosophicus, the philosopher Ludwig Wittgenstein presented his readers with a proposition: “The world is all that is the case.” Though Wittgenstein meant this as a claim about the bounds of logic, it can also be interpreted psychologically. Our conception of the world is based on what we think is out there. In fact, that’s all we think the world is—whatever’s out there.

What has changed a lot since Eratosthenes’ time is our technology. We now have real images that depict what the world actually looks like. What hasn’t changed are the strategies our brain uses to make sense of our surroundings. The story of Eratosthenes exemplifies the kinds of biases that those strategies can lead to.

In this sense, the map of Eratosthenes resembles something we all have in our heads—a reconciliation of the terrain with which we have immediate familiarity with that which we have not yet encountered. This means of organizing information in the brain is well studied by psychologists and neuroscientists.

They call it a “cognitive map.”

Subscribe now

The idea that an animal’s brain creates a map of its environment as it runs around in it was first proposed by Edward Tolman, in his influential 1948 paper, “Cognitive Maps in Rats and Men.” Tolman’s initial proposal was simply that the brain uses some sort of spatial representation to organize knowledge about the layout of the world.

In his experiments, Tolman would set a rat at a random point in the maze, then measure how long it took the rat to find her way to the cheese. As expected, the rats got quicker each time Tolman put them in the maze. Tolman then compared this with a second condition. He would put a rat in the maze, but without any cheese. The rat would wander aimlessly, and whenever she got to the place where the cheese would have been, Tolman would take her out and set her back in the cage. The rats in this condition each did five of these non-cheese trials before moving on to the normal cheese-motivated trials.

What Tolman’s contemporaries would have expected to see was that the non-cheese trials didn’t matter. The accepted theory of the day was that learning was motivated by reward. And without the cheese there was no reward, and therefore no learning. But that’s not what Tolman found. What he discovered was that as soon as he began putting the cheese in the maze, the rats who had merely been wandering were able to solve the maze just as quickly as those who had been going after the cheese the entire time. The reason for this, Tolman explained, was that the rats were creating a cognitive map of their environment the whole time. Once they knew where in that map the cheese was, they could go straight there.

In this sense, Eratosthenes’ map of the world can be thought of as a sort of cognitive map, worked out in pen and paper. Like Tolman’s rats, he developed a picture of what he thought was out there. Except, unlike the rats, Eratosthenes couldn’t just traipse around the world to get a feel for it. He made up for this with a particular strength of his: working out new conclusions from existing data.

Eratosthenes had little respect for disciplinary silos. He wrote quite a bit of philosophy, and his contemporaries considered his work adequate but derivative. He was a decent poet, but by no means the most renown of his generation. The same could be said of his status as a mathematician. His friends, picking up on this pattern, took to calling him “Beta” because he was, according to them, second best in everything.

During his younger years Eratosthenes had a particular interest in prime numbers. His most celebrated mathematical invention is known today as the Sieve of Eratosthenes. The Sieve—which rhymes with “give”—is an algorithm for generating a list of all the primes up to a chosen number. To this day, it’s still the most efficient way for a computer to generate all the primes up through about one hundred thousand. The payoff for this idea would come far after Eratosthenes’ own lifetime, with relevance in contemporary fields like number theory and cryptography.

The intellectual feat that garnered Eratosthenes the most fame in his own lifetime, though, was his estimate of the earth’s circumference. He managed to get within about a tenth of a percent of the modern calculation using only Euclidean geometry and a few simple measurements. It was well known in the ancient world that the sun was directly over the city of Syene during the summer solstice, because of a deep well that cast no shadow. The length of the Nile from Syene to Alexandria was also known, because it had recently been surveyed by Ptolemy II. So one day on the summer solstice, Eratosthenes measured the angle of a tall building in Alexandria. He got a reading of 7.2 degrees. Doing the math, this meant that this section of the Nile was exactly one-fiftieth of earth’s circumference (if you cut a pie into fifty equal pieces, the interior angle at the point of each slice would be 7.2 degrees). His calculation was off by about the distance between John F. Kennedy airport in New York and Newark airport in New Jersey along I-95; in other words, you could walk Eratosthenes’ margin of error in about nine hours. One historian calls Eratosthenes’ estimate “a feat so profound yet so simple that it remains today one of the most amazing pieces of ancient scholarship, treated as such since antiquity.”

There’s a Greek word, philologos. Literally it means “lover of reason,” but at its heart it meant a learned person, an academic. It was a somewhat pejorative term. An ancient Greek might use it in the same way we might say today that someone is a thinker, as opposed to a doer. Eratosthenes, allegedly, was the first to use the word to describe himself. That’s sort of like someone today being declared the first “thought leader.” Not necessarily by anyone else, but according to their own estimation.

Of course, he was probably correct.

Share

In 1976, the psychologist Stanley Milgram had just come off a decade-long research program on obedience to authority, in which he induced participants to apply extravagant levels of electric current to one another in a dingy basement on the campus of Yale University. Milgram had recently moved to Paris, where he was keen to employ an experimental paradigm that was decidedly more benign. In particular, he was interested in what he called people’s “psychological maps” of the city. “It is not,” he wrote, “an examination of Paris as a geographic reality, but rather of the way that reality is mirrored in the minds of its inhabitants.” How, in other words, do people’s psychological maps differ from the way Paris actually looks? “The first principle,” as Milgram put it, “is that reality and image are imperfectly linked.” 

What you might have expected Milgram to find is that his participants’ maps looked like impoverished versions of a tourist map. After all, each of Milgram’s participants would have seen the canonical map of Paris—featuring the Eiffel Tower, Notre Dame, the Arc de Triomphe—hundreds of times. To be sure, their hand-drawn maps wouldn’t look exactly like the actual map; these people are not professional cartographers. But there is no doubt that they would have been familiar with the real thing represented as it actually is. They could also rely on their quotidian interactions with Paris, wandering its streets and traversing the cityscape as they go about their business. But that was exactly the problem.

What Milgram found is that people’s maps were more strongly influenced by their own idiosyncratic experiences than a general knowledge of the city’s layout. One map was drawn by a fifty-year-old woman who had lived for fifteen years in the city’s fourth arrondissement. She mapped that area, just north of the Seine, in scrupulous detail, down to the direction of the one way streets. But she included almost no landmarks south of the Seine, except for Montparnasse—no Eiffel Tower, no Musée d’Orsay, no Jardin du Luxembourg.

In this respect, her map is representative of what Milgram found overall. Almost half of Milgram’s participants failed to include the Eiffel Tower and other major landmarks. “She centers her map not on Paris as a whole,” wrote Milgram, “but on a segment of it that has special meaning to her.”

When you look a little closer, the maps of Milgram’s Parisians were off in much the same way as Eratosthenes. Eratosthenes, so to speak, mapped the one way streets around Alexandria, Athens, and the Mediterranean, where he lived most of his life. But he failed to get any details right once he set his sights on distal terrain, like India or southern Africa. And just as with Eratosthenes, while Parisians didn’t think their map was perfect, they thought it was pretty much correct. The difference here is that, unlike Eratosthenes, they had seen the real thing. They should have known how far they were off. But they still confidently asserted that the reality in their head corresponded with what was out there. It didn’t.

Share Meaning Lab

So what exactly had Eratosthenes seen that informed his map? That answer begins, as it so often does for these sort of things, with Aristotle. In his Meteorologika, Aristotle documented an interest in the earth’s surface. He was the first one to provide an estimate of the circumference of the earth (it was way off). But most of his work ultimately didn’t go much further than speculation. After all, how could he possibly know what the rest of the world was like when he’d never seen it?

So when Aristotle began tutoring a young pupil named Alexander around 340 bce, there were many questions about what the earth looked like but not many answers. Alexander, for his part, took a certain pragmatically-inclined interest in questions about the surface of the earth. This was probably connected to the fact that he would, over the next couple decades, go on to conquer most of it.

Alexander’s campaigns went throughout most of Asia minor—northeastern Africa, Macedonia and the Hellenic peninsula, the eastern Mediterranean, Arabia, and into India, just north of Bombay. These campaigns started in 336 BCE and continued until 323 BCE, the year of Alexander’s death. Eratosthenes was born about four decades later in 276 BCE and became head librarian in Alexandria at the age of forty. 

By the time Eratosthenes took his station at Alexandria, all of Alexander’s military personnel had returned from their travels, compared notes with one another, and had written up reports about what they had seen. And where do you think they kept those reports once they’d been filed away? The Library of Alexandria.

Eratosthenes had what Aristotle did not. He had data. 

Which is what puts Eratosthenes on the same playing field as the rest of us. We can only use the data available to us to paint a picture of what’s out there. Like Milgram’s Parisians, and Tolman’s rats, we each construct a map of reality as we go about our business in the world. And also like them, we assume that what our version looks like is the real thing. The risk we run is to mistake the representation for the reality. We might think that we are exempt from the kind of mistakes Eratosthenes made, because we simply have access to so much reliable information. But that isn’t a guarantee. In the Geographika, Eratosthenes wrote of why he was unable to consider the possibility of another unknown expanse of land, which might conceivably exist outside his own realm of knowledge. It was, in a literal sense, a foreign world. “For myself, however, I must speak of what is in our own.”

Leave a comment

If you enjoyed this post, please consider buying a paid subscription. I just finished my PhD and started doing this Substack—and I’d love for it to provide an income stream to support my goals as writer. Right now all my posts are free, in an effort to grow my subscriber count. But in the coming months, I’ll start switching over to publishing most posts as premium content (while my podcasts will remain free). If you purchase a subscription now, I’ll give you 50% off for the next year (regardless of whether you buy a monthly or yearly subscription). That price will go up once I start to paywall the content.

Either way, I appreciate you taking the time to read my work!

Get 50% off for 1 year

Over the course of this week, I’ve been sending thank you notes to those who have been a part of my life over the past few years. I say “few” because there were a lot of people who helped me throughout grad school, and I never really got around to thanking most of them. I’m reaching out to friends who I’ve been meaning to text but haven’t. I’m sending brief notes to the people who came on my podcast and said something that really stuck with me. I won’t get to everyone. But I find that in thinking about my experiences with others and what I really value about them, I gain more clarity on my past year than I would from a more self-centered reflection.

Plus—what better way to start the New Year than with an inbox full of replies from the people you care about?

I don’t think we spend nearly enough time telling the people we care about how much they mean to us. I think we wait until it’s too late. And we regret it.

Share

As Kierkegaard said, life is understood backwards but lived forward. Standing as a kind of purgatorial middle-ground between those two orientations is the week between Christmas and New Year. It’s a time for reflection on who we’ve been, as well as an opportunity to think about who we want to be. In thinking about who I want to be in 2023, one thought comes to mind: a better friend.

The more I think about it, the more I’m convinced that the hardest problem of adult life is making and maintaining friendships.

It’s not that friendship is, in general, the hardest part of adult life. Developing one’s career, selecting a mate, spending time with family, engaging in meaningful experiences, staying fit—these are all difficult. The difference is that we expend a lot of energy trying to be better at them. We care a great deal about how to be better in our professional and personal lives. But our social lives are a different story. As a society, we devote very little intellectual capital to how to maintain a robust and thriving social circle.

The further I get into adulthood, the more it feels like one big psychology experiment to see how willing we all are to hold onto friendships as an increasingly large number of variables are stacked against us.

Everyone, myself included, is so damn consumed with doing their own thing, it becomes harder and harder to relate to what other people are up to. We become more physically distant—whether we find ourselves moving somewhere new or someone we’re close to moving away. We also become more interpersonally distant. It can be tough to relate to someone where you both have well-defined expectations about who the other person is which are based on a version of yourself you no longer relate to. Time does that to a relationship.

And it’s not like social media has really solved the problem, has it?

This year, I want to find more and better ways of working against that schema. I want to make time for the people I really care about. I want to actually take the time to catch up with someone when I say “Hey! Let’s catch up soon!” I want to lift my head up from my own desk long enough to see what other people have going on in their own lives. Career? Family? Physical health? Absolutely, there’s things on those fronts which I’m thinking about in the coming year. But that’s exactly the point. For me, they’re not in danger of falling by the wayside. Friendship feels a lot more like it is.

And I’ve got a couple concrete ideas about what to do about that.

The first is that I’ve resolved to make birthdays a time where I go out of my way to reconnect with friends. I’m not super big on birthdays. But I’ve gotten pretty excited about the idea of using them as a yearly opportunity to get together with someone who it isn’t especially easy for me to get together with. I’ve got my thirtieth birthday coming up this year. My vision for it is to bring together a small group of people featuring friends from every era of my life—Seattle, LA, Boston, Oxford, etc—and celebrate this decennial milestone with the people who have meant to most to me throughout the years. And for as many different versions of “me” as I can make happen.

The second is that I’ve recently been testing out using a Calendly page to schedule calls with friends I want to catch up with. Everyone makes fun of me when I send them the link. (As they should.) But the truth is that so far it’s made the conversion from “Hey! Let’s catch up soon!” to actually catching up soon a lot higher. I’m still working on the details.

Other things I’m thinking about include: finding occasions for social connection which don’t revolve around alcohol; using social media in a more intentionally engaged way to connect with friends/people I care about (which I think it’s safe to say is no longer the default mode on most of these platforms); and how to encourage conversations toward more meaningful, rather than superficial, topics.

Anyway, these are just a couple examples of things I’m thinking about. I’d love any ideas from what other people find most effective in their own social lives.

Subscribe now

Thanks for reading and subscribing to Meaning Lab. It means a lot to have you here. I just started this Substack in earnest at the end of November, and I’m really excited about the coming year. Friendship is definitely one of the subjects I’ll be covering—along with much, much else. And if you want to work on Strategies for Making and Maintaining Friendships™ together, then hit me up! In 2023, I’ll be in Vietnam from January through April; in Europe (with trips to the US and UK) from late April through August; and in London from September onward.

Share Meaning Lab

There are books that were more impactful for me personally, ones that I’ve recommended to more people, and ones that I simply liked more. But Diaz’s work stands out as—I’m gonna say it, but please know that I’m not saying it lightly—a work of true genius.

First of all: it’s genuinely fun and engaging. It has all the tension of a murder mystery. The novel is a kind of puzzle. You are given four separate texts. Essentially, four different version of the same story—or maybe the same story, maybe a slightly different one... it can be hard to know whose version of the facts to trust. But what you think is happening in the beginning of the novel is not what ends up happening later on. It’s up to you, as the reader, the put the pieces together.

The main thing, though, is that the novel covers SO MUCH. It’s layered: there’s theme upon theme upon theme, and all of them are substantive and all of them feel urgently fascinating. For example, the book is about the way we tell the story of success in America. But it’s also about the construction of texts! How do we translate events from life itself into our narrative rendering of them? Throughout the four parts of the novel, you see how the contents of one text comes to influence the next one. But there’s so much more as well. For example, it’s also about the role women play in this story, and how they’re often written out of it.

Trust is one of those books that you read and think, “Wow, that was incredible.” Then, after a little while, you start to think about it again and realize it was even more incredible than you first thought. Every time you revisit it, you realize that your previous assessment was a tragic undervaluation and that there’s actually even more to the book when you really think about it. It’s one of those books where you can’t help but ask: How on earth did someone actually create this?!

Another notable thing about the book (and something that Diaz often comments on when talking about the work) is that it deals with the theme of money. Specifically, money. There are shelves and shelves worth of Great American Novels about things like status, wealth, important people, success, and power. But none of them actually talk about money itself. It’s just there in the background, a supporting structure for the larger story. In Trust, the characters actually talk about money: how they made it, what they do with it, that sort of thing. It’s one of those instances where as you’re reading you feel like—okay, how have I never seen this before? It’s at once the most intimate concept in the world, something we all interact with many, many times a day, and yet you’ve never seen it treated in this way. Incredible!

This, in my reading, is the meaning of the title: it is the double entendre of something that refers to both a specific financial instrument and the foundation of all human relationships. Generational wealth, the social fabric of society, the reliability of a story’s narrator, who gets to be the narrator in the first place, money, power, and the emotional bonds between human beings—it’s all connected.

Subscribe now

In college, I read a book called “How to be Rich” by J Paul Getty. Back then, I tried to read as many different kinds of books as possible, and this definitely counted as something different. It was a work by a rich guy (the American oil tycoon who endowed the Getty museum in LA, just down the road from where I went to school) not about how to become rich... But, like, once you were rich, how should you act to play it off like it’s no big deal? Lord knows why I felt compelled to read this. But I was extremely gratified to find myself instantly transported back to this Getty text by one of Diaz’s narrators. Diaz absolutely NAILED the voice of a plutocrat trying to make sense of their own success. (Don’t worry, though; the book isn’t solely based on the perspective of rich white men.) And, importantly, it’s not a caricature of that voice, either. It’s a rendering by someone who truly, profoundly understands the perspective and is able to represent it in the strongest, most sympathetic way possible. But unlike Getty’s book itself—and this is where it really gets deconstructed—we get to see this voice juxtaposed to... I won’t give anything away, but let me just call it different sides of the same story.

Diaz’s first novel was In the Distance, a reimagining of the American Western. (If you know me, you know that I’ll read any book with a cowboy in it.) Basically, it was the most significant revitalization of this literary genre since Cormac McCarthy. And I WOULD compare Diaz to McCarthy. The only problem is that Diaz has way more range. It’s not that my man Cormac has no range: All the Pretty Horses is a very different western from Blood Meridian, and The Road is not a western at all (at least not superficially; I have a theory about westerns which would support the claim that it is… but that’s a theory for another day). Though Diaz’s two books both seem to penetrate to the beating heart of the American story, they are worlds apart. Most of McCarthy’s work is set in New Mexico. Diaz is from... Sweden, Argentina, and the US, depending on how you slice it, and his work (especially In The Distance) reflects that. If McCarthy’s native milieu had been Stockholm and Buenos Aires then he could’ve written a very similar book to In the Distance. But he never in a million years could ever have written something like Trust!

Another thing I love about Diaz is that he’s a PhD. Diaz did his degree in literary criticism at NYU. He studied Borges. His first book (non-fiction) was his own highly technical interpretation of this great South American writer. I tried to read it (just out of curiosity). I didn’t get very far. It’s not really meant for someone who isn’t professionally trained as a literary critic. But, in a way, this makes his current work even more special.

As someone who just finished my doctorate, I look to Diaz as a symbol of what’s possible in life post-PhD. His work is as cerebral as you’d expect from an academic in the best case scenario. It is, above all, interesting. But it also has soul. Though it’s not overtly sentimental or emotional, it is real. It seeks to capture something that matters in a way that accounts not just for facts but for the broader human experience.

Ultimately, I’m not 100% confident that I’d recommend this book to everybody. I’m a big believer in the power of the right book for the right person at the right time. If I were going to recommend a book to every human being on the planet, it would be Project Hail Mary by Andy Weir. That’s the most fun you can have while reading (or in my case listening to; I highly recommend the audiobook!) words in the English language. I may not have told every person on earth about it, but I’ve definitely mentioned it to a majority of the ones I’ve personally come in contact with. Anyway, what I am confident in is that if you read the above and thought “that sounds right up my alley!” then you will absolutely, unequivocally be glad you read this book. There’s nothing like it… Trust me. (Sorry.) Honestly, though, now’s the time to get into his work. Hernan Diaz is a name your grandchildren will learn in school.

Share

I hope you find something good for your next read. If you happen to find it through the above links, I get a referral fee. Thanks!!

Also—if you enjoyed this post, please consider buying a paid subscription. I just finished my PhD and started doing this Substack—and I’d love for it to provide an income stream to support my goals as writer. Right now all my posts are free, in an effort to grow my subscriber count. But in the coming months, I’ll start switching over to publishing most posts as premium content (while my podcasts will remain free). If you purchase a subscription now, I’ll give you 50% off for the next year (regardless of whether you buy a monthly or yearly subscription). That price will go up once I start to paywall the content. Either way, I appreciate you taking the time to read my work!

Get 50% off for 1 year

“I only write when inspiration strikes. Fortunately it strikes at nine every morning.”

—William Faulkner, allegedly

I call this the “9am theory.” The 9am-ers believe that productivity is a function of consistency. Wake up at the same time every morning, get to work, and you won’t fail to create something worthwhile. They believe that waiting for inspiration is the strategy of amateurs. As Steven Pressfield says in The War of Art (great book, by the way): “The most important thing about art is to work. Nothing else matters except sitting down every day and trying.”

But not everyone endorses that theory. I call its competitor the “Lion theory,” based on the following tweet:

These two camps subscribe to different theories about the distribution of productive output. The lion theory says that productivity comes in clumps. Asking yourself to be productive every day isn’t going to lead to meaningful productivity. It’s going to lead to monotony. You’ll feel like you’re getting plenty of stuff done. But most of that stuff wasn’t worth doing in the first place.

While the 9am theory says something slightly different. Meaningful productive may appear to come in clumps. But that’s only because you don’t know what’s going to hit and what isn’t. Your responsibility as the creator is, as Pressfield and Faulkner suggest, to show up every day, to be consistent. Whether or not what you make resonates with others is not in your control. Consistency is.

This reminds me of a more general point made by Lebanese-American Twitter user and former author Nassim Nicholas Taleb. We expect the world to work according to a bell curve. We wake up every day for our nine-to-five. Most days our work-level is moderate. Some days it’s a lot, some days it’s less. The 9am theory says that the more consistent we are, putting in that moderate work, the more productive we’ll be.

But the lion theory suggests that our work-level should not be normally distributed. It should, instead, be distributed bimodally. Spend a lot of time doing almost no work. Watch Netflix. Call a friend. And then, once the timing is right, transcend your lethargic state and go all-out. Taleb calls this a “barbell strategy” because the graph looks like a weight, and he likes weights.

Personally, I can relate to both of those theories. In my PhD, I spent my entire first year (half of which was waylaid by the overture of the pandemic) doing nothing. Quantifiably nada. Then I spent my entire second year working on my research. After twelve months of that, I took another six off. Then I finished writing my dissertation in the next six. That’s pretty bimodal: periods of nothing, followed by periods of intense work. But on the other hand, I did those long stretches of productivity by working at my projects at the same time every day: 9am to noon.

So it seems reasonable to assume there’s merit to both of these approaches. As Paul Bloom recently told me, the number one dictum of productivity hacking is: know thyself. Figure out what works for you; don’t worry about what anyone else does. When someone tells you about an “optimal” routine, remember that optimality is contextual, and depends on the specifics of the person and what they’re trying to do.

That being said, there ought to be a generalizable principle here. There ought to be circumstances in which the lion theory makes more sense, and circumstances in which the 9am theory does. What are they?

Subscribe now

A key assumption in the lion theory is the existence of hot streaks. If productivity is bimodal—with distinct periods of sprints and rests—then the main time to sprint would be when you’re on a hot streak. When the iron is hot, strike.

Whether hot streaks are actually thing has been a long running debate in the behavioral sciences. For a long time, it enjoyed a prominent seat in the theatre of conventional wisdom. Then, kinda famously, it was “debunked” by a paper from the 1980s, referring to this received wisdom as the “hot hand fallacy.”

The authors of this paper were particularly interested in the case of basketball. If a player has made their last five shots, are they more likely to make the sixth? They looked at data from the Philadelphia 76ers and found no evidence of a correlation between the number of previous shots a player made and the probability they’d make the next one. In other words, there’s no such thing as a “hot hand”—we only think there is because true randomness results in clusters, and because we humans are so adept at recognizing patterns we try to spot them in random sequences even when they’re not actually there.

While it’s true that people do tend to find illusory patterns while reading the tea leaves of randomness, that’s far from the final verdict on hot streaks. The best current evidence comes from a group of data scientists doing a large-scale analysis across many different creators across many years in several domains.

The authors of this paper, which was published in Nature, looked at the individual career histories of artists, scientists, and film directors. For artists, they used a database of auction sales for 3,480 artists, which tracked both how often they created new works and how much those new works sold for. For scientists, they looked at publication records for 20,040 scientists, measuring the number of citations their papers earned 10+ years after publication. For cinematographers, they looked at IMDB ratings of films by 6,233 directors. They wanted to know: do their successes come randomly? Or do they tend to cluster—above and beyond what you’d expect by clustering via simple randomness?

The first thing they found was that hot streaks were actually quite common. In fact, almost everyone they looked at had at least one hot streak in their career! They compared their statistical hot-streak model with a “random impact” model—where there was no discernible logic to periods of exceptional success—to see which one better explained the patterns of data. It was clear that creator’s successes tended to cluster more than what would be expected by randomness. According to their model, 91% of artists, 90% of scientists, and 82% of directors experienced at least one hot streak in their career.

Okay, so when did these hot streaks occur?

Did they tend to come earlier in a career or later on? Now, the authors discovered that this appears to be random. They couldn’t find a pattern.

In other words: hot streaks are themselves non-random. But when that hot streak hits is... totally random! They also found that these hot streaks were pretty consistent in their length: 5.7 years for artists, 3.7 years for scientists, and 5.2 years for directors. Again, the length of the streak was independent of when it happened in their career.

Finally, they looked at the rate of productivity during these hot streaks. Were creators producing a larger number of things during their hot streak? Perhaps the increase in quality could be explained via increased quantity. But no, they didn’t find any such pattern. There was no difference in the number of works they created during their hot streak. The works were just simply better—or at least better received.

But wait… is there really no logic to when to hot streaks occurred? That’s what these data scientists thought at first. Even with their fancy model and all their data, they couldn’t figure it out. As it turns out, they had to build an entirely new, even fancier model to find the trend. It wasn’t until they published a follow-up, three years later, that they provided an answer.

Share

So what predicts when a creator will enter a hot streak?

The authors started with the same databases they used in the initial experiment: auctions, citation counts, and IMDB. But this time they combined that information with a new model.

For the artists, they trained a neural network to look through the artist’s entire collection of work. The details of this are a bit complicated, but trust me when I say the entire process was really clever. The neural network took into account low-level features, like brush strokes, as well as higher-level conceptual features—and how these changed across as artist’s works.

The neural network for the directors and scientists worked similarly. For films, it took into account a wide range of data such as plot information, cast members, and genre. For scientists, the neural network took into account features such as research topic and co-authorship. Basically, what they had was an automated assessment of stylistic change. For each domain, they could say how closely an artist, scientist, or director’s latest work resembled their old work.

Then, with the information about stylistic change in hand, they could compare that with the onset of a hot streak. (A slightly more technical definition of stylistic change can be found at this footnote.)1

Here’s what they found:

During a hot streak, creators underwent very little stylistic change. Their best cluster of works all tended to look alike. They were more predictable, more consistent.

But prior to a hot streak, they were all over the place. Each creator seemed to be trying out a bunch of different stuff. They were experimenting, unsure of what would work and what wouldn’t. It didn’t matter whether they looked at artists, scientists, or directors, the pattern was the same—a period of adventurous experimentation followed by a period of focused successes.

Now, if you’ve studied machine learning, you’ll have seen this kind of pattern before. It’s called the explore-exploit trade-off. Finding the actions of highest value requires you first to figure out what actions are possible (explore) and then to dial in on the ones that you think are most lucrative (exploit). Hot streaks, in this framing, are a period of focused exploitation brought on by a period of open-minded exploration.

And both of these stages are necessary. The authors also looked at whether exploitation alone could predict a hot streak. It couldn’t. If the creator just focused in on one thing, unless they had spent their time exploring the space of possibilities, they wouldn’t be exploiting the highest value option.

It’s also worth noting that this preceding phase seems to require true exploration. In other words, it’s not that the creator is getting closer and closer to their big success with each successive attempt. They’re not following a linear path, starting off by doing something that looks nothing like what they’ll eventually do and slowly refining it over time. To test this, the data scientists looked at whether the similarity metric during the exploratory lead-up phase came to look more like the work produced during the creator’s hot streak. What they found that was the successful style of work was less likely to resemble the most recent or the most popular in the exploratory phase.

In short, they were just testing things out; eventually, one of them hit it big. The message is clear: let your explorations be truly unencumbered. Then, when you’ve found something that works, make the most of it.

Share Meaning Lab

Let’s bring this back to productivity. We’ve got the lion theory and the 9am theory. Which one is better supported by these data?

One big thing we’ve learned is that productivity is bimodally distributed. At least productivity weighted by how successful the product in question turns out to be. Sometimes you’re on fire (i.e., exploiting), while sometimes you’re still figuring things out (i.e., exploring). There’s a lot of wisdom in knowing which phase you’re in at a particular moment.

If you’re in an explore phase, it makes more sense to allow yourself to be erratic, to do things a bit differently, to not go too hard on yourself when something doesn’t land the way you want. But then when you’re in an exploit phase, when you’ve found something that works—you gotta make moves.

That’s in favor of the lion theory. But only slightly.

The other thing we saw is that overall levels of productivity didn’t change based on whether or not you’re in a hot streak. When you set aside the quality/success of the output, you’re still generating the same amount of stuff. That goes more toward the 9am theory, which would say if you keep your effort consistent, eventually you’ll find something worth doing.

Ultimately, what the hot streak evidence suggests is that meaningful productivity depends on a lot more on what you do than when you do it. And what you really need to get right is whether you’re in an exploratory or exploitative phase.

Both theories, I think, can accommodate both phases.

The lion theory naturally lends itself to separate between explore and exploit phases. When you’re resting, you could be thinking, taking walks, talking to friends and colleagues, watching YouTube videos, reading Substack articles—all of which could stir the creative pot of your next move. Then when you sprint, you exploit a single option which seems like the best at the time. Once you enter rest mode again, you can evaluate how it went and start the processes anew.

For the 9am theory, exploration looks like working on a different kind of project than you normally do. Exploitation looks like working on whichever project is most important at the time.

I think the answer of which is superior ultimately depends on which of these suits your preferred style of exploration versus exploitation. Speaking for myself: if I’m in an exploratory period, I like to work like a lion. A lot of doing nothing. Intermittent bursts of fascination with something new. Lazy afternoon browsing unfamiliar library shelves. Keeping my interests as wide as possible. Then if I’m in an exploitative phase: it’s time to sit down and get to work. I have to say no to the allure of novelty and focus on what’s already in front of me. I try to do that everyday at 9am whether I especially want to or not.

There are many ways to be productive, and as Paul Bloom says you have to find the one that works for you and your present needs. Based on this evidence, the only cardinal sin is not to respect the freedom needed during your periods of exploring or the focus needed during your periods of exploiting.

So I think it’s fair to say that productivity can be turned on like a faucet. But while meaningful productivity—the kind that produces the work you’ll be know for in the long-run—isn’t exactly random, you can’t just turn it on at will, either. You just have to explore until you find the right path to exploit.

Leave a comment

P.S., I just found the polling option on Substack. Let’s give it a try:

If you voted, thanks. Looks like the results will be publicly available 3 days after the post goes live. Check back then I guess?

If you enjoyed this post, please consider buying a paid subscription. I just finished my PhD and started doing this Substack—and I’d love for it to provide an income stream to support my goals as writer. Right now all my posts are free, in an effort to grow my subscriber count. But in the coming months, I’ll start switching over to publishing most posts as premium content (while my podcasts will remain free). If you purchase a subscription now, I’ll give you 50% off for the next year (regardless of whether you buy a monthly or yearly subscription). That price will go up once I start to paywall the content. Either way, thanks for reading!

Get 50% off for 1 year

But that didn’t seem like the right explanation. Their actions certainly reflected a concern for my happiness. And it’s not like they were forcing me to study something just because it would lead to a job that was prestigious or met some conventional definition of success. In fact, my personal happiness seemed like a huge priority for them.

The more I thought about this, the more I realized that it wasn’t out of a disregard for my happiness—but because they held a different theory about how to achieve it.

Our society has two fundamental beliefs about happiness: (1) that we can become more happy, and (2) that more happy is a desirable thing to be. And yet—I think most people would agree, when you frame it this way, that one of the most efficient ways to become less happy is to spend a great deal of time worried about your own happiness.

This presents a bit of riddle. We all want to be happy. But the key to pursuing it is... not pursuing it. How do you get more of something without trying to get more of it?

Subscribe now

One of the most successful frameworks used in modern artificial intelligence is called reinforcement learning. The basic idea is simple. If an action leads to a good outcome, do it again; if it leads to a bad outcome, don’t. It is a definition of intelligence that places reward maximization at its heart. Simply put, there is something that you want, and intelligent behavior consists in getting as much of it as possible.

At the core of reinforcement learning is what’s known as a “policy.” For example, if your agent is a robot that plays basketball, then its reward comes in the form of points. The more baskets the robot makes, the more points it gets. The more likely it is to win the game, the more intelligently it behaved. The policy is the robot’s playbook.

A policy says, in mathematical abstraction, “This is where I am right now. This is what I have to do next to maximize my points.” In basketball, a good policy might be to get the ball, dribble it toward the basket, and toss in a lay-up. Each time the robot does this, it looks at how effective it was in getting points, and adjusts its behavior to do better next time. The robot might start off bad, but using reinforcement learning it could become better over time.

The hard part is that there’s no direct way of knowing the best policy. You have to try out different policies, and figure out which one is most effective. Is the best policy to drive toward the basket? Or should you sit back and shoot jumpers? How do you know which is going to work out better next time around? Will the same policy work against a different opponent?

In general, there are two strategies for how to learn a policy.

The first is called on-policy. It’s the more straightforward of the two strategies. On-policy means that the robot uses the same information to make decisions and evaluate whether or not they were the right ones. Basically, it’s going to make decisions based on what it thinks will most quickly increase its number on the scoreboard. If its current policy says to drive toward the basket and that results in a lot of points right away, then it’ll will be more likely to keep going with that same plan in the future.

The second strategy is called off-policy. This means that the robot is using different information to make decisions than it is to evaluate them. The robot could make decisions based on, for instance, its time of possession on the ball. Or how many passes it completes before taking a shot. Or how little time it lets the opponent spend in its own half. At the end of the game, it could then look back at its play based on the different policies and see if focusing on something else actually made it more likely to win.

At first, it might seem like the better strategy is always going to be on-policy. How could you achieve your goal more effectively by focusing on something totally irrelevant? But that’s not necessarily true. The empirical fact in artificial intelligence research is that some problems are better solved by off-policy methods.

And framed in another way, this makes total sense. It’s actually a pretty neat solution to Goodhart’s Law, which says that “once a measure becomes a target, it ceases to be a good measure.” How do you prevent a good measure from becoming a target? Simple. Pick something else as the target. Sometimes the best way to attain a goal is indirectly.

Leave a comment

So, is happiness better pursued with an on-policy strategy, or an off-policy one?

It depends on what you mean by “happiness.”

Psychologists often distinguish between two kinds of happiness. The first is “subjective” well-being. This is a straightforward definition of hedonism: in any given moment, do you find yourself pleased with what you have? The second kind is “eudaimonic” well-being. This is an idea that goes back to Aristotle, that well-being is about a cultivation of virtue and skill in service of contributing to the larger, more worthy goals of humanity. The first kind of well-being is a vertical cross-section of happiness in the moment; the second takes a broader view of overall life-satisfaction. In a Christian context, this is often thought of as the difference between happiness and joy.

A recent study led by happiness research kingpin Kennon Sheldon put these two kinds of well-being to the on-policy/off-policy test. Do people who pursue subjective well-being feel more subjective well-being? What about for eudaimonic well-being?

Sheldon and his co-authors gave people questionnaires to measure both their subjective and eudaimonic well-being—how happy they were feeling in the moment, according to both definitions. They also gave participants a questionnaire to measure how motivated they were to change these respective aspects of their well-being. If an on-policy strategy works well, then you’d expect that the more motivated they are in their pursuit, the higher they’ll score on current well-being.

And for the pursuit of eudaimonic well-being, that’s exactly what they found. The more people prioritized eudaimonic well-being the more of it they felt. But the pursuit of subjective well-being? It was also correlated with subjective well-being—but in the opposite direction! The more they claimed to pursue it, the less of it they had over all.

But here’s the true test of the off-policy strategy: Did the students who were more motivated to pursue eudaimonic well-being also achieve more subjective well-being as a by-product? Of course they did! The correlation was even stronger than with achieving eudaimonic well-being itself. There was no such by-product effect for the pursuit of subjective well-being.

Now, it should be said that this is a correlational study. The authors ran it on mostly white, college-age students enrolled in a psychology course. So this study isn’t the final word in the discussion. But it’s nonetheless a clear test of on-policy versus off-policy approaches. If the on-policy strategy worked for subjective well-being, then pursuing more of it should lead to getting more of it.

As Sheldon and his co-authors put it in another paper: “the pursuit of happiness involves trying out different kinds of goals, values, behaviors, and activities, to determine which ones bring one satisfaction and happiness. Ironically (and reassuringly), the best happiness boosting behaviors tend be the ones that focus on long-term self-improvement and on deepening connections with others, just as most lay and eudaimonic theories of ‘a life well-lived’ have long proposed.”

Naturally, as Sheldon alludes to, these researchers weren’t the first to make this observation. My favorite characterization of the off-policy theory of happiness comes from John Stuart Mill, in a passage from his Autobiography:

Ask yourself whether you are happy, and you cease to be so. The only chance is for you to have as your purpose in life not happiness but something external to it. Let your self-consciousness, your scrutiny, your self-interrogation, exhaust themselves on that; and if you are otherwise fortunately circumstanced you will inhale happiness with the air you breathe, without dwelling on it or thinking about it, forestalling it in imagination, or putting it to flight by fatal questioning.

The reason, then, that my parents never told me to pursue happiness directly was that they, like Mill, believe in an off-policy approach to happiness. When someone tells you that you should “do what makes you happy,” they’re advocating for an on-policy approach—making decisions and evaluating them by the same metric. That’s exactly what my parents didn’t want me to do. And while my parents didn’t learn this from reading Mill, the surprising thing about this position on happiness that it is shared—in some version or another—by practically every other philosopher who has weighed in on the matter.

Another of my favorites comes from Bertrand Russell. He says more or less the same thing as Mill, but with a certain flair of nonchalance in contrast to Mill’s solemn weightiness. Russell writes in The Conquest of Happiness: “Fundamental happiness depends more than anything else upon what may be called a friendly interest in persons and things.” He continues, “let your interests be as wide as possible, and let your reactions to the things and persons that interest you be as far as possible friendly rather than hostile.”

Share

Oh, and by the way—happiness isn’t the only thing the off-policy strategy applies to!

Earlier this week, I published a podcast episode on the perils of gamification, featuring game designer Adrian Hon. “Gamification” is the term for when we assign points to our actions in real life in an effort to incentive doing the right actions by maximizing our points. For example, learning a language via Duolingo. Learning a new language is hard, because it usually involves a lot of boring vocab learning. Duolingo uses game-inspired features like experience points, leaderboards, and streak counts to make the drudgery of vocabulary learning more fun. And the trend in modern life is toward increased gamification: from our fitness goals, to our diets, to the number of books we read.

In his latest book, Adrian argues that the problem with this is when we start to care more about the fictional game elements (for instance, keeping our streak alive) than we do about our original goal (learning the language). The reason Adrian thinks we should be concerned about this is because, psychologically-speaking, this problematic shift can happen pretty quickly.

Gamification starts off as an off-policy approach to learning. We make our immediate decisions according to what’s going to get us the most points, or hit our goal of 10,000 steps in a day, or gets us the most likes on a Tweet. Then in the long-run this will translate into having learned a language, or maintaining our overall health and fitness, or the feeling that people are listening to us and care about what we’re doing. But at some point, we shift into on-policy learning. We forget why we started in the first place. We’re just after the points.

Neither on- nor off-policy learning is a one-size-fits-all approach. For example, what if our basketball player decided to try an off-policy strategy based on maximizing possession? Maybe it would just takes the ball to the corner and hang onto it for dear life. That’s not going to help you win the game any more than naively dribbling the ball straight toward the basket. The off-policy approach may help you avoid Goodhart’s Law. But it’s possible for it to lead you too far astray, as well.

That’s why I like framing personal goals in terms of meaningfulness. It’s a lot easier not to get lost in your own pursuit of personal happiness. As Mill says, focus on something larger than yourself. One day you’ll wake up to realize that you inhale happiness with the air you breathe.

Get 50% off for 1 year

But nonetheless here we are. Last month, I defended my dissertation.

And so now I’m in the somewhat startling position of having done what I set out to do. I find myself faced with a familiar question, but one whose answer feels a lot less straightforward than it used to be.

Now what?

The month before I began my PhD, in October 2019, I sat down with an idea. The concept was to reach out to people I admired—mostly academics and authors—and ask them about the decisions they made when they were in my position. What did they do when they were grad students that set them up for success later on? Sure, I wanted to know about their success, in some sort of career-prestige sense. But I also wanted to understand how they thought about what it means to make a substantive contribution to their field, whatever that may have looked like to them. I envisioned it as a podcast, which I called Cognitive Revolution.

People, I was surprised to learn, were incredibly generous with their time. The project didn’t always go as well as I hoped. There’s a lot that I could’ve done better, and the pandemic actually stifled my show when it seemed to bolster this kind of project for so many others. But I got to talk to many of my heroes, a lot of whom were the ones who inspired me to pursue cognitive science and social psychology in the first place.

I started the project with the vague idea that it would be a useful exercise in “audience building.” It seemed like the kind of thing that was done by other authors who had taken a path like the one I envisioned for myself. It was clear to me since I was an undergrad that I cared at least as much about telling stories about research findings than actually doing the research itself. And I’ve always known that I wanted to write non-fiction pop-psych books as a part of my career. But I also knew that going directly into writing wasn’t the right move, either. I wanted to have something to say. And I felt that developing actual expertise in a field I cared about would give me that.

The Cognitive Revolution podcast allowed me the opportunity to explore the different versions of what that can look like, and how different people have constructed something resembling a coherent career from the disparate pieces of whatever they’ve found, in retrospect, that they’d managed to accomplish. What I thought was going to a means of building an audience was more like adding a second major to my degree. I got a lot out of it. But it was only incidental whether anyone else did as well.

Somewhere along the line, though, I began to feel I was reaching a point of diminishing returns on that project. It’s not that there was nothing left for me to learn. But it seemed like I had gotten all the information that I was going to get out of asking people how they went about doing whatever it was they did. I still am drawn to people’s personal stories, absolutely. But the original concept of Cognitive Revolution no longer represents the dimension of growth that I see myself moving along. It’s time to do something else.

And so I’m starting a new project. It’s a podcast; it’s a blog. It’s the Substack you’re reading now. I call it Meaning Lab.

In Meaning Lab, I’ll take a cognitive science perspective on the pursuit of meaning in work, life, and relationships. Each week, I’ll publish a podcast interview with an author, scientist, or academic about how their work has uncovered some interesting or unexpected aspect of meaning—where it comes from, how it works, what exactly it means to find more of it in one’s daily activities. I’ll also publish a weekly piece from my own perspective delving into what psychological research tells us about the mechanisms underlying how we make sense of the world and our place in it.

There are, above all, two reasons I want to talk about meaning.

First, I just think it’s the coolest concept in all of cognitive science. The enterprise of meaning-making is the single most interesting thing that minds can do. To take one example, we humans can take arbitrary sequences of squiggles and lines and dots and use those to represent our entire experience of the world. Human language is amazing. It’s something I’ve been interested in for a long time (for instance, my undergrad thesis was on “Computational models of jazz improvisation inspired by language”). But another example of meaning is how we reflect on our own experiences to create stories about what we’ve done, who we’ve done it with, and why it was worth doing. And meaning isn’t just important for esoteric things, like the study of linguistic semantics, or more practical things, like what research says about how to get more fulfillment out of your work—but the full range of human experience, from music, to art, to ideas, to the basic infrastructure of cognition, to what brings us all together in organized society. In a very real sense, our minds are designed for meaning-making.

The second reason is that I think the idea of meaning is able to give us a more nuanced vocabulary for talking about our experience of the world. This, in my estimation, is something we really need. I’m skeptical of the way we normally talk about some of the routine psychological concepts of work and life.

For instance, happiness. The concept just seems very flimsy to me. As if the best of all possible lives is one in which you attain a permanent state of placid appeasement. Ice cream for every meal. It’s a one-dimensional definition of what it means to be human. Feelings like heartache and profound sadness may not be especially gratifying in the moment. But they’re at least as important in giving texture to the experience of a human life. The concept that reflects that much more directly, in my opinion, is not “happiness,” but meaning.

Which leads me to another of the usual constructs that I think we’ve misunderstood: habit. So much of our discourse about work, and how to be better at it, has to do with developing an optimal habitual routine. The reason for this is that the promise of good habits is frictionless productivity. In the best case scenario, we’d be able to do the right thing without ever having to think about what exactly it is. The problem is that reliance on habit puts us on autopilot. That might be fine when you’re flying a simple route. But when life requires flexibility, contemplation, or creativity, our habits—good or bad—work against us rather than for us.

These are kinds of arguments and ideas I want to explore on this Substack. Eventually, I’ll really be trying to do this blog/podcast as a premium product. Looking forward, I think at some point I’ll do most of my posts paywalled, with the podcasts (or at least, like, the first 60 minutes of them) free. So, in the future I will be asking you to shell out some dough to support my work. For now, I want to focus on making sure the work is as high-quality as possible, as well as growing my free subscriber base before dialing in the paid content. That said, if you do want to support up front, I’d really appreciate it! This is the move I’m trying to make post-PhD, so your contribution will help me be able to solidify doing this kind of thing full-time. Even signing up for a month of paid, then cancelling makes a big difference! There’s a button below, which I believe says “subscribe now” for non-subscribers and “upgrade to paid” for free subscribers—so please do feel free to use such a button however you see fit, including leaving it completely untouched.

Subscribe now

At any rate, I’m glad to have you here. I think it’s gonna be a lot of fun. I’m excited. New episodes of the Meaning Lab podcast will begin next week.

It was at this point I discovered that I had not in fact satisfied the university's standards for a “general education.” I was missing a class, and now in the final term of undergraduate, I was too late to enroll in one that would rectify this oversight.

The fix was to take an online class during the summer after my senior year. I was able to walk at graduation. I was able to start the job I had lined up. So this seemed like a good solution. And anyway, I would use the method which had got me through English classes in high school. I would forgo reading the actual book, opting instead to browse through a summary online.

This was the context for my first encounter with Joseph Conrad’s Heart of Darkness.

It’s the only book I remember being assigned during the course. There must have been others. But I think, even in the moment, I was struck by just how little I cared about it. I knew this was a much vaunted work in the canon of English language literature. But I couldn’t drum up a single shit to give about it.

I was working at a venture-backed start-up in Silicon Valley. I had no use for Conrad. I had no use for the Congo. And I had even less use for some esoteric work where the entire second section is some guy who isn’t even the narrator going on about some story which had, from what I could tell, only a tenuous connection to the actual “plot”. What I did have use for—even took pride in—was my ability to perpetrate bullshit. This was the muscle I decided I would exercise during this course. Not the literary, actual reading-of-books one.

In putting together the piece you’re reading now, I went back through my college docs and found my original essay. I can see why I chose not to read the book. The relevant assignment was only a single page. Hardly seems worth it to scrutinize an entire novel just to put together a few hundred words on it. I wrote about Conrad’s conception of “human nature.”1 Instead of going in-depth on the text, I connected its major (read: obvious) themes with general insights from cognitive science. This sort of move is often regarded by the grading class (i.e., grad students) as evidence of “critical thinking.”

The instructor singled out my contributions on the class discussion board as being especially incisive. I got an A in the course, and officially received my diploma. But I never made it into the Heart of Darkness. I didn’t even make it to the delta of the Congo River. I never even ordered the book.

“Everyone gets everything he wants. I wanted to see the Congo. And for my sins, they let me.”

This is the line Anthony Bourdain uses to open his exposition in his episode on the Congo for the television show Parts Unknown. I’m not sure what it means. But it’s a badass thing to say.

The episode is structured, self-consciously, around Conrad’s Heart of Darkness. Bourdain quotes it throughout in the voiceover. He makes reference to the characters in the footage from the trip, for example pointing out which buildings in the modern Congo were the inspiration for the book’s “inner station.” Bourdain makes it clear that Heart of Darkness looms large in his imagination.

I’ve watched every episode of Bourdain’s Parts Unknown at least twice. And it’s common for Bourdain to riff on the themes and imagery of a book or movie to structure an episode. For example, this shot from Bourdain’s Tokyo episode clearly alludes to the shots of the hotel bar at which Bill Murray and Scarlett Johansson find themselves in Lost in Translation (one of my own favorite movies).

But the Congo episode is one of the ones I’ve returned to most often. It is the episode that perhaps capture’s Bourdain worldview most directly, in its most naked form. The cinematic climate of the episode is when the crew attempts to make dinner while drifting down the Congo River. This is their descent into madness, the closest they get to penetrating into their own heart of darkness.

In the show, we see Tony attempt to prepare coq au vin for dinner while on board. His logic is that this dish will be an effective use of the scraggly chickens they brought on the boat. But, it turns out, the knife is dull. It’s an ineffective tool for butchering the chickens, let alone slaughtering them. Then the generator goes down. No generator, no fire. And without heat there’s no coq au vin, just chickens in wine. The scenes are tense. Tony curses on camera. They don’t eat the meal until, like, two in the morning.

That Tony obviously was so personally moved by Heart of Darkness was the first clue that I ought to revisit the work. But I didn’t appreciate just how deep this connection went until I read the excellent In The Weeds by Tom Vitale, Tony’s long-time producer.

He writes:

To call doing a show in the Congo Tony’s ‘obsessive dream project’ was a gross understatement. He always said you tend to see your life as a book or a movie, and for Tony that story—if there had to be just one—was Joseph Conrad’s 1902 novella Heart of Darkness. The book, set in the Belgian Congo, and its cinematic reinterpretation Apocalypse Now, about the Vietnam War, had been recurring motifs throughout Tony’s work right from the beginning. Pretty much every river trip we ever filmed contained a Kurtz reference or some kind of homage to the theme of descending into madness deep in the jungle.

In the episode, we see their crew encounter logistical difficulties. They enter the Congo from the border with Rwanda. Their first experience in country is the war zone around Goma. Then they fly from there to Kisangani, where they’ll be able to access the river. As Vitale writes, “the Congo was in ruins: much of the country was a patchwork of lawless rebel controlled territories.” He continues:

Thinking it through logically it really demanded the question, Why the fuck were we going to the Congo? Was it just about satisfying one of Tony’s biggest obsessions? Or was it a way for him to use the show to shine a light on something bigger? When I asked Tony why he wanted to go to the Congo, he said, ‘Don't worry about why, did he tell you why in Apocalypse Now? No. He wanted a mission. He got one.’

He’s referencing the opening quote he uses in the episode. It’s a riff on the opening line from Apocalypse Now. They arrive at Kisangani, known to Conrad as the “Inner Station.” Tony points this out for the camera.

In fact, probably the very spot we were standing at that moment was the setting for his [Marlow's] encounter with the infamous Kurtz, an ivory trader gone mad, worshiped by the local population as some kind of god. Much like Marlow, Tony saw Kurtz as an alluring and enigmatic figure. Was he monster, demagogue, or prophet? Kurtz had gone to the Congo with “good intentions,” but ended up decorating his house with severed human heads. Kurtz, what the book represented, as well as Tony’s infatuation with it, were all somewhat... ambiguous and open to interpretation. Conrad quit after his first trip as steamboat captain on the Congo River, the horrors of what he witnessed having left him in a deep depression. Now 123 years later, it was our turn to embark on a river journey toward our own indefinite narrative conclusion.

Vitale summarizes the progression of their trip: “Over the last couple of days, Tony’s references had increasingly switched from Heart of Darkness to Apocalypse Now.”

Then there’s the dinner scene. In the book, Vitale gives us a behind-the-scenes look. Tony had instituted a rule that everyone had to kill their own chicken to contribute to dinner. According to Vitale’s account:

‘Tom, [Tony said] you might want to observe how it’s done, so you don’t cause the poultry any unnecessary suffering. Jerry, let’s go.’ Jerry put down his camera and—being a farm boy from Iowa—nonchalantly popped off the chicken’s head like it was a bottle cap.

‘Clean kill, Jerry, clean kill,’ Dan said, having adopted the role of execution commentator.

Holding the bloody severed chicken head, its beak still slowly opening and closing, Jerry asked, ‘What do we do with these?’

Tony looked proud, and I decided my best course of action was to slip overboard unnoticed, or at least disappear for a bit. I turned around to sneak away and slammed my head into the low beam so hard I fell back flat against the floor. My first thought was, maybe if I’m paralyzed, I won’t have to kill a chicken! Regrettably, I was uninjured.

‘It’s batter up, Tom,’ Tony said.

‘Please, I really don’t want to do this,’ I said.

‘If we don’t kill those chickens, we’ll starve to death,’ Tony said, thrusting the bloody knife in my direction. ‘You want to eat a raw onion for dinner?’

As much as I didn’t want to kill a chicken, I also didn’t want to look like a chicken in front of Tony. I know none of this makes any sense, but such refined notions as logic had been left a long ways upriver. Besides, both cameras were pointed in my direction—which in case you didn’t know has a way of making people do stupid things—so I took a deep breath and resigned myself to participating in Tony’s bizarre blood rite. I reached down and grabbed the chicken, who stopped struggling and stared up at me. As Tony had prophesied, I could tell the chicken was thinking, ‘Why me, Tom?’ Hovering, Tony said, ‘Don’t be gentle, you’re not going out on a date, C’mon, just do it!’ I looked away and plunged the knife into the chicken’s neck and sawed and sawed as hard as I could. Over the ringing in my ears and the horrified screams of the other chickens, I could hear voices shouting, “Hold it higher so I can get a better shot!” and “Kill! Kill! Kill!" The Congolese boat guys were all watching the spectacle and must have been wondering what the hell was going on. When I looked down, it seemed like the knife was barely penetrating, let alone going all the way through, and the chicken was still looking up into my eyes, and there was blood.

‘Saw harder!’ Dan laughed maniacally.

I was really trying, back and forth with the knife, and the chicken was still looking at the me. ‘Do it faster!’ Tony commanded. ‘You're making him suffer!’ And that's when the panic set in. ‘Stop it, stop!’ I yelled through tears. I felt myself let go of the knife, and I ran to the other side of the boat. Had it been an axe or even something sharp, it would have been bad enough. But that fucking knife was so dull, and I just couldn’t stand the thought of the poor chicken not seeing another sunrise.

When the red mist cleared, I felt pretty pathetic for freaking out over a chicken, considering I was in a country where horrific acts of cruelty and violence were facts of life. Worst of all, I’d failed the test. I was too chicken to kill the chicken, to chicken to say no, then too chicken to finish the job. The sad truth is I was the chicken all the along. I’d managed to hold it together very well on the Congo shoot thus far—and for the last ten years of working on the show—but Tony had finally broken me. He had a way of always getting what he wanted. Perhaps a consolation prize Tony put a smudge of blood on my forehead and said, ‘Now you can join our treehouse.’

With the chickens slaughtered, it was now time to prepare dinner. This is the footage that makes it into the show.

Trying to focus over the din of two clattering outboard engines and now a clanking, sputtering generator, Tony placed the first chicken on the cutting board, raised his knife, and the lights cut out, plunging us into complete darkness. ‘Jeee-zuuus,’ Tony said, drawing out the syllables for dramatic emphasis. ‘I need some fucking light so I can see what the fuck I'm doing!’ I felt around along the floor to my backpack and found a headlamp. Moose did the same, and soon a few beams of illumination cut through the inky void. Dez went to investigate why the generator was still whirring away but not providing power. Within a minute or so, the bulbs flickered on to reveal a pissed-off looking Tony. ‘This knife is as dull as a soup spoon,’ he said, going to work butchering the chickens. ‘Get me the machete, and I need another pot.’ Before either could be found, the lights wavered, dimmed, and failed again. ‘Fuuuuck,’ he moaned. As power returned, I was faced with an even more pissed-off Tony, arms folded. Wiping some blood off his Rolex to check the time, he said, ‘Tom if we’re going to eat at all, you need to unfuck this situation!’

Like it was my fault the generator wasn’t cooperating with Tony’s ‘Full Metal Julia Child’ fantasy of effortlessly preparing a jungle-style coq au vin.

At length, dinner was made.

While attempting to choke down a zinc-flavored onion, my mind drifted toward Kurtz. Maybe there’d been a method to the madness all along? Was Tony's point that—as in Heart of Darkness and Apocalypse Now—deep in the jungle everyone goes a little bit crazy? That we all have a Kurtz somewhere within us? We’d arrived in the Congo with good intentions, and ended up doing something as vulgar as playing with our food in a country of starving people... on TV.

The first thing that stood out to me in returning to the text is Marlow, in the beginning of the book, begins by noting what the Thames river must’ve been like a thousand years ago—in a sense, the original Heart of Darkness.

In her book The Dawn Watch, an exploration of the parallels between Conrad’s work and the currents of globalization, Maya Jasanoff writes:

When the Thames reminded Marlow of the Congo, he wasn’t simply saying: Look, Africa is more primitive than England. He was saying that history is like a river. You can go up or you can go down. You can ride the current by nesting Marlow’s experience in Africa inside the telling of his story in England, Conrad warned his readers against any complacent notion that savagery was as far from civilization as there was from here. What happened there and what happened here were fundamentally connected. Anyone could be savage. Everywhere could go dark.

My own interpretation differs slightly. One way to look at it, as Jasanoff does, would be in terms of the discontinuities of history. Like a river, one can move with or against it. In constant fluctuation, but with parts that are clearly up-stream or down-stream in accord with the river’s flow. But another way would be to see Marlow’s is, instead of the discontinuity of history, the continuity of human nature. A kind of There but for the grace of God go I. How you behave depends a lot more on when and where you were born than who you are as a person.

Think London is so civilized? You should’ve seen it back in the day.

Think Congo is so brutal? You should see your own heart, under different conditions.

As a fresh college grad, these questions felt distant—or at the very least abstract. Now, more than half a decade later, they feel more pressing. I feel much better acquainted with my own propensity for madness. I feel a much greater need to symbolize darkness, to have a concrete way of reckoning with the inherent complicatedness of human life.

There’s a line from Heart of Darkness, which has come to be a kind of refrain for me on this front: “It was written I should be loyal to the nightmare of my choice.” Where was this written? Well, as far as I can tell, it’s canonical rendering is in Conrad’s novel itself. But there’s something about this notion that I can’t extricate from my mind.

Loyal to the nightmare of one’s choosing…

I think this line speaks to me because it directly contradicts the fundamental dictum of millennial life: you can always do better. In our place as the generation of infinite choice, we’ve been brought up with the idea that, since all possible outcomes are available to us, we should be able to find the one to which we are best suited. In short, not only to follow our dreams, but to reasonably expect to attain them.

This is the promise of optimality. Take mating, for example. Because, in theory, we have access, via dating apps, to the set of all possible mates, we expect that the mate with which we end up ought to be the very best possible option. The presence of suboptimality signals that we’ve not yet found that option and ought instead to keep looking. So we keep looking, until we find the ‘perfect’ choice—the choice entirely devoid of nightmare.

The same is true of careers. Millennials are more happy to switch companies, even industries or job descriptions, than any generation before them. The same is true of cities. The ‘global citizen’ seeks constantly to find the right product-market fit between the themselves and the milieu in which they live. This is the incessant quest for optimality.

But that’s not how life actually works, is it? There’s no such thing as the perfect choice, the option devoid of imperfection and suboptimality. If one only develops loyalty to the outcome of one’s dreams, then guess what. You’re never going to find anything to be loyal to—or, even at a less morally-upstanding level, to be content with. No, see, the trick is to realize is that the choice isn’t between nightmares and dreams, but only different kinds of nightmares. It is perhaps a dire a way to put it. But it paints the disparity in stark enough contrast to make it comprehensible.

The presence of suboptimality does not mean that you are on the wrong path. It is in the nature of paths to be suboptimal. Therefore, if something feels “wrong” about it—well, to some degree, that’s just par for the course.

The real key, then, is to accept the tradeoffs. If there’s an aspect of your current situation you don’t like, the appropriate question is not: “Would I trade this for something better?” It is: “Would I trade this for a different set of downsides?”

Loyal to the nightmare of your choosing.

To me, this isn’t a dire sentiment at all. It’s liberating. We don’t have to shy away from the presence of bad things, from negative emotions, from situations that feel like they ought to be improvable. Life doesn’t have to be a constant quest for optimality. This quest itself is only one of the many choices of possible nightmares. I don’t know about you. But that kinda makes me feel better.

Maybe that’s what Tony thinks it means for everyone to “get what they want.” We all chase after something. And when we find it, we discover it’s still, in its way, a nightmare. As far as I can tell, fame is like this. Success is like this. Wealth is like this. Having a family and settling down in the suburbs definitely seems like this. Doing a PhD is absolutely like this. But that discovery is only surprising—only detrimental to one’s evaluation—if one works from the premise that optimality is attainable. The sooner we let go of that notion, the sooner we become loyal to the nightmare of our choosing.