On agency

Birch in a forest, Gustav Klimt, 1903

---

Last May, when our oldest daughter Maud turned seven, I wrote:

I wish I had a book that I could put in her hands, and it helps her learn what many never learn, or learn too late, namely, that the possibilities are much bigger than you think, that you can live more deeply, and truly, and that you can solve almost any problem if you put your mind to it. A book about how to handle being sentenced to freedom, and to handle it effectively, and authentically, and responsibly.

It is late May again, Maud is eight now, and I’ve decided to write down a glimpse of the book I imagined.

Last year, when I talked about learning “how to handle being sentenced to freedom,” a phrase I borrowed from Sartre, I meant roughly what people these days call “cultivating high agency.” But I need to define my words, since some ways the phrase high agency is used feel foreign to me, and depressing.

Agency, as I see it, is an amalgamation of two skills, or mental dispositions: autonomy and efficacy.

1. Agency requires the capacity to formulate autonomous goals in life—the capacity to dig inside and figure out what wants to happen through you, no matter how strange or wrong it seems to others. In other words, it requires autonomy (which was what I was getting at when I said “authentically, and responsibly”).

2. Agency also requires the ability and willingness to pursue those goals. It requires the “will to know,” the drive to see reality as it is, so you can manipulate it deftly and solve the problems you want to solve, instead of fooling yourself that certain problems are “unsolvable.” In other words, efficacy (“handle it effectively”).

Or phrased negatively, the opposite of agency can mean one of two things. Either (1) doing what you are “supposed to do,” playing social games that do not align with what, on reflection, seems valuable to you and/or (2) being passive or ineffective in the face of problems (assuming your problems can’t be solved, that someone else should solve them, or working on things that do not in a meaningful way address the problem.)

I wrote about ways of figuring out what you genuinely care about in this post:

Rationality is an underrated way to be authentic

Agency is often framed as a hard-edged, type-A, aggressive approach. But over the last year, as I’ve been thinking about writing this essay, I’ve talked to a lot of highly agentic people, and I’ve read biographies about and interviews with people whose agency I admire and . . . hard-edged does not fit what I’ve seen. Often, agency is almost gentle—an attunement to the world and the self, a feeling out the details of reality, and a finding of the path of least resistance. There is sometimes considerable force involved, hard work, but it is like the force of a river being pulled toward the sea.

For the sake of brevity, I’ll mainly use examples from the life of the German filmmaker Werner Herzog, but the patterns I choose to focus on recur in most of the lives I looked at.

Problems are solvable

Werner Herzog started writing screenplays and submitting proposals to producers and television stations when he was “about 14.” I also wrote screenplays at that age, and dreamed of making films, but it didn’t even strike me that I could actually attempt what Herzog did.

I would read about him in the encyclopedia at the local library, and I would read about Tarkovsky and Kubrick and Cassavetes, but I thought of them as near-mythical figures; it didn’t strike me that I could look at what they did and do it myself. I thought what I wanted to do wasn’t doable, so I didn’t even try.

Now, 22 years later, I’ve lived long enough to have learned that most things are actually doable if you care enough. I rarely feel blocked in the way I did when I was younger. But it is interesting that it was something I had to learn: that problems are solveable; that if I direct my attention to the problem and learn to understand it, and act on what I learn, the problem will, sooner or later, cave in.

I think the main reason I didn’t realize my problems were solvable was a lack of imagination. I’d never seen anyone solve problems in an agentic way, and I failed to imagine that things I hadn’t seen done could be done by me.

I’m not sure why Herzog, at 14, growing up among the ruins of a bombed out Munich, could imagine himself directing a film, but he could. He was agentic enough to realize that the fact that kids until then had never been allowed to make films was just a problem, and thus could be presumed solvable. Herzog’s puberty was late—he looked like a young child until he was sixteen—so he assumed the producers would never take a bet on him if they saw him; he would have to make and negotiate deals over letter and telephone.

After three years of trying this again and again, he managed to get two film producers interested. They liked one of his proposals and were willing to accept Herzog as a first-time director. A meeting was booked.

Herzog recounts the meeting in Paul Cronin’s A Guide to the Perplexed:

When I walked into their office, I saw the two of them sitting behind a huge oak desk. I remember it second by second. I stood there as they looked beyond me, waiting, as if the father had come into town with his child. The first one shouted something so abusive I immediately wiped it from my memory, while the other slapped his thigh and laughed, shouting, “Aha! The kindergarten is trying to make film nowadays!” The entire encounter lasted fifteen seconds, after which I turned and left the office, knowing full well I would have to become my own producer.

Just imagining these sorts of reactions is painful and a big reason people hesitate to try things. But for Herzog, it was information that helped him get clarity on which strategy to try next.

Werner Herzog:

Two days later I filled out the necessary paperwork, paid a few dollars to register the company, and founded the Werner Herzog Filmproduktion.

Compare this reaction with how I behaved when, at 22, I realized that I wanted to write essays. I stumbled into essay writing after a friend asked me to write a piece for a poetry journal he edited, and I fell in love with it right away; it took me only a few hours to realize that thinking on the page was much more exhilarating than film or music or poetry or programming or anything else I had tried up until then.

This was in 2012/13, around the time I started dating Johanna who is now my wife. And she told me that, given how alive I seemed to feel when I wrote essays and given how alive the result was, I should consider writing more of them. But I said there was no point doing that. I was involved in the literary world in Stockholm at this point so I knew how it worked—there are no magazines that pay for essays in Sweden. “No one has ever made a living as an essayist,” I said.

Johanna likes to tease me about this, now that we pay our bills by writing essays. And I wish I could say that I listened to her back then, but I didn’t. I was sure I knew which problems were solvable and which weren’t—it is almost arrogant if you think about it. I gave up writing essays and instead wrote novels, because that was what publishers seemed to want. But I never cared for novels enough to get good at it. My writing fizzled out.

What made my decision especially unagentic is the fact that what I wanted to do wasn’t even hard: I could have found a decent-paying day job and used that to fund my essay writing. It was as simple as that. And then I could have iterated on this first solution until I found a better one (perhaps if no one reads essays in Sweden, I shouldn’t write in Swedish?). The American composer Philip Glass funded his music by starting various companies where he could work intensely for two weeks, then take two weeks off to write—a moving company, a plumbing business. After his international breakthrough, the opera Einstein on the Beach (1976), the 39-year-old Glass spent two years working as a cab driver in New York to pay off the debts from the tour. That is as good a way as any to do it.

Herzog, after founding his company at 17, spent the final two years of high school working the night shift at a steel factory to save up money for productions:

I did Punktschweissen, the kind of electrical welding that doesn’t require the precise skills of a welder — which is much trickier and takes years to master. [...] I can scarcely remember my last two years at school; I was so tired, working every night until six in the morning, saving every penny. They threatened to throw me out because occasionally I would sleep through class. “It would be justified if you kick me out because I can’t translate a phrase from Latin,” I told my teachers, “but it would be a scandal if you did so because I’m working harder than anyone else.”

The best advice I can offer to those heading into the world of film is not to wait for the system to finance your projects and force it to decide your fate. If you can’t afford to make a million-dollar film, raise $10,000 and produce it yourself. That’s all you need to make a feature film these days. Beware of useless, bottom-rung secretarial jobs in film-production companies. Instead, so long as you are able-bodied, head out to where the real world is. Roll up your sleeves and work as a bouncer in a sex club or a warden in a lunatic asylum or a machine operator in a slaughterhouse. Drive a taxi for six months and you’ll have enough money to make a film.

Why didn’t I see that it would be easy to fund essay writing in this way? It was, as I’ve said, partly a lack of imagination, partly a fear of looking stupid. But it was also that my thinking was bundled. I had conflated “being a writer” with “having a publisher” and “getting a salary from my writing.” These are not the same thing.

If someone had pointed this out, it would have been clear to me that the part I cared about was thinking on the page—I would have never picked having a publisher or getting paid to write over getting to write interesting things. But I didn’t realize I needed to get clarity around what precisely I was trying to achieve; hence, I had a fuzzy bundled picture of my goal which made me miss the obvious solution. Without a clear idea of what it would mean to solve the problem, it was hard to evaluate options and make consistent progress.¹

Overcoming this sort of confusion is one reason why I’ve come to value thinking-through-writing, which I write about here:

How to think in writing

Most problems are solvable. But that doesn’t mean the solution will look like you hope it will; there will be trade offs. While Herzog made Signs of Life (1968), Even Dwarfs Started Small (1970), Fata Morgana (1971), Land of Silence and Darkness (1971), and Aguirre, the Wrath of God (1972), he had to live at home with his mom to afford making the films. He was 31 years old when he moved out. Most people would never consider a solution like that, but sometimes, that is what high agency looks like. You have to keep your eyes locked on your values so that you don’t sacrifice the wrong thing.

Looking at the problem

At the heart of agency lies a willingness to question defaults. To be agentic, you have to treat “how things are supposed to be done” as just one option among many.

Or, no, that formulation isn’t deep enough. When I think about friends of mine who struggle to be agentic, the problem isn’t precisely that they do the default thing; it’s that they fail to understand their problems and the solution space. They act in incoherent or ineffective ways because their mental model of the situation is too limited to show them a way out. They are not attuned enough to figure out what they want and how the world works. To be agentic, you have to really look at the problem and at the solution space and accept the responsibility of learning what is necessary to make the problem go away.

If you forget about how your problems are “supposed to be solved” and just look at the goal—what is the shortest path from here to there? What is the fastest way to get the information you need to find that path?

If you have a clear understanding of the goal, there are often paths that lead there that are much shorter than the default path. A good question to ask is: what is the simplest solution that could possibly work?

If you read interviews and biographies of people with high agency, you can get plenty of examples of them asking the equivalent of this question and finding simple, creative solutions to complicated problems. Since it is hard to say anything general about what it means to find a simple solution, I recommend looking at plenty of examples to get a feel for what it means to act in this way.

For a sample of that, I recommend (apart from the Herzog books I’ve quoted from) this piece (~28,000 words) about how Patrick McKenzie (patio11) and a group of people on a Discord server in 2021 became “the clearinghouse for vaccine location information for the United States of America.” That piece is filled with hard problems solved with simple solutions.

The reason the Discord server was set up was that it had been revealed, in January 2021, that the bureaucracies involved with the vaccination program in California had lost track of the vaccine doses that had been sent to the state: only 27 percent of the doses had been administered—73 percent were in a freezer, somewhere, and people often had to make “40, 50, 60 phone calls” to find a pharmacy that could give them a shot. (It is interesting to consider that tracking the vaccine stock was such a complex problem that the state couldn’t handle it, yet a group on Discord was able to solve it by being agentic.)

During the first day of the project, the team needed to figure out how to get information about which pharmacies in California had the vaccine in stock. The definition of success, in other words, was clear: they wanted to track down where the vaccine doses that had been sent to California were now.

(If you want to practice, stop here and think about how you’d go about setting up a system to track the vaccine inventory in California.)

Patrick McKenzie:

I resolved this ambiguity in a very startup-y way: I googled for the phone number of the Walgreens at Fourth and Townsend in San Francisco. I have been to San Francisco on business before; that Walgreens is right next to the city’s main Caltrain station. (Like everyone living in Japan, I assume that a city is centered not at its geographical midpoint but at the train station. [...]) I called at 9:30am and asked to speak to the pharmacist. Without identifying myself or giving any preamble I asked, ‘Could a 65 year old get the Covid vaccine, and, if so, how?’ The pharmacist told me that they didn’t have it but to check back in two weeks.

I reported to the team that the US healthcare system would happily give vaccine inventory information to literally anyone calling. We immediately started calling the healthcare providers in the spreadsheets we had compiled overnight.

There are several good things about looking for the simplest solution that could possibly work. The first is that the types of problems that require agency are often hard, so if you make the problem any harder than necessary (as we can assume the state was doing, by insisting that they had to follow “process”), you might not make it. Or you make it, but you spend so much resources that you have to sacrifice something else (like, in this case, human lives).

If there is a “default solution” to a problem, that solution is often one of the costly solutions that needs to be avoided.

This is because the “default solution” usually has all sorts of bells and whistles that are not necessary for the concrete outcome you are looking for. When I was thinking about making films, for example, I thought that to do so I had to go to film school—this is the default solution. But film school tries to be a general preparation for all sorts of filmmakers, meaning you’ll have to spend time learning things that aren’t necessary for your specific projects—and so it takes years. That seemed too costly to me, so I gave up filmmaking.

Herzog, on the other hand, realized that film school was just one option among many. And the better solution, given his goals, was to take the money he had saved up by working as a welder and fund his own films, then send the films to festivals and use the prize money to fund ever more ambitious projects. He did the actual work and got rapid feedback from reality, learning precisely what he needed to solve his current problems. This was faster.

Herzog:

If I want to explore something, I never think about attending a class; I do the reading on my own or seek out experts for conversations.

Put another way: Herzog mapped the actual landscape and found the shortest path to his specific goals. (And his goals were very specific. He never wanted to be “a filmmaker,” he wanted to make this film and then that film.) Working in a goal-directed way, he gathered the information he needed; he found people who could help him; and he learned the necessary skills,² duct-taping one solution after the other until he had the film he wanted. Then he went on to the next project.

People with high agency tend to be obsessed with finding simpler solutions, to the point where normal people think they are idiots. As a case in point, consider the impression Herzog made on the team of Hollywood professionals who were sent to assist him during the production of Rescue Dawn (2006):

At every turn, crew members let [Herzog] know that they considered his directing habits strange, impulsive, even amateurish. They couldn’t comprehend why Herzog insisted on grabbing the machete himself when the sound crew wanted to capture the sound of slashed reeds. [...] They were irritated when Herzog declared that someone’s unfinished makeup looked “good enough,” and that he couldn’t wait for it to be perfect, because he liked the way the tropical light was filtering through the treetops.

Herzog had his eyes clearly locked on the goal, capturing the images he knew the film needed, and he didn’t care if the way he did that was “unprofessional.” Doing it “the proper way,” as the team insisted, would have been too expensive and time consuming—he would never have been able to amass a body of work so rich and beautiful as he has if he had followed the Hollywood process. The professionals, having too many preconceived ideas about how to go about things, wasted resources and missed the light in the trees.

Herzog:

The Americans were always nervous, telling me I wasn’t shooting sufficient coverage. I took my assistant aside and asked, “What do they mean by coverage? I have insurance coverage for my car, but coverage when making a film?” They want me to get a range of intermediate shots, close-ups and reverse angles, all for safety’s sake. But I have always filmed only what I need for the screen, and nothing else. When you do open heart surgery, you don’t go for the appendix or toenails, you go straight for the beating heart.

Epilogue

Earlier I painted myself as an unagentic person, but that doesn’t feel quite true anymore. Let me end by saying a few words about what changed. There have been a series of experiences that have helped me realize more of my agency, but I think the most important one was becoming a father (which is ironically circular since I started this essay by thinking about what I wanted to tell our daughter Maud about agency—and I end up realizing she was the one who made me agentic).

The reason having Maud in my life made me more agentic was that it was the first time I experienced what it means to surrender to my values. I had a lot of idiosyncratic opinions and values when I was younger, too, but I held onto them in a rather flimsy way. Whenever things got too hard or people disapproved of what I was doing, I tended to give up and do the normal thing instead.

When Maud was two, Johanna and I decided that we wanted to homeschool her. This is not only illegal in Sweden, but also seen as something akin to child abuse. So when we told my parents we were going to leave the country to homeschool their grandchild, they were shocked, afraid, heartbroken, embarrassed, and did what they could to talk us out of it. I remember lying on the sofa, reading their emails, and thinking there was no way to ever repair this. If I had experienced it before Maud, I would have caved in after 30 seconds. But in this case, caving in was unforgivable; I must never fail Maud. So I had to sit through the experience of having nearly everyone I knew either quietly disapprove of what we were doing or actively try to talk us out of it, crying. This was very empowering. Because I got to experience what it was like to stand my ground and do what I knew needed to be done, and see how good the outcome was.

And not only that. The decision to leave Sweden to homeschool also made me agentic in another way: there were so many problems that we just had to solve. We had to find a new country to live in and learn the language. And since the move pushed us down into relative poverty for four years, we needed to become self-reliant and learn how to renovate old windows and start companies and wrestle with bureaucracies that threatened to force us to sell the farm and so on and on and on—endure an endless barrage of challenges that I did not think I was capable of enduring. And though there were moments where I feared that the stress would break me, there was something very profound about experiencing, again and again, that problems that seemed unsolvable and overwhelming to me were solvable, always solvable. It changed my priors.

I remember when we discovered that the brick wall in the kitchen was in bad shape and needed to be redone, and I was like, “I can’t do this. I can’t hammer a hole from the kitchen straight out to the garden. We have to sell the house.” But then I just did it and it completely changed my view of what a wall is—it is something understandable, something I can take apart and put back together and manipulate to better fit our needs.

And everything is like that.

Everything can be learned, everything can be understood and reshaped to fit our values.

Looking up, five years later, I can’t believe how much my life has changed and how much more open the world feels, how much more filled with possibility and beauty. I see Maud and her little sister Rebecka (who was born after we moved to the island) playing on the trampoline on the edge of the meadow. And I think about the fact that Johanna and I are now actually, somehow, feeding our family by writing essays, and that brings a wetness to my eyes. I look at the beech trees and I think about how much beauty this world is capable of and how much possibility there is out there and how much more I can still do to nurture that beauty and that possibility.

It was there all along, that possibility; I could have reached for it all these years. The thing that kept my hand from rising was in my head. What I needed was to let my care grab hold of me and pull me away like a river—or like a drop of water running down the trunk of a beech tree, feeling out the details of this world, the bark, the lichens, looking for the path of least resistance, running down, down, into the dark wet earth we call home.

For more on this topic, see “6 lessons about agency I learned working at an art gallery” and “Thoughts on agency.”

Acknowledgements

This piece was written in collaboration with Johanna Karlsson. Esha Rana did the copy edits. The essay also benefited from conversations and comments from Celine Nguyen, Sasha Chapin, Ha Tran Nguyen Phuong, Alexander Obenauer, Gena Gorlin, James Giammona, Alexis Gauba, Adhitya K R, Catherine Brewer, Yashvardhan Jain, Amit Parekh, Philip Chin, Tim Rutherford, Sindi Stefanova, Kiri McCrory, Tam Minh, Matt Joass, Julius Henning, Tom Critchlow, and Alexandra Heller.

1. These days, when I want to do something hard, I like to start by describing the problem in as much detail as possible. What would it mean for this problem to be solved? What does the world look like when I’ve made the problem go away? Not what the solution should be (not “I want to sign with a major Swedish publisher”), but the effect I want the solution to achieve (“I will consider the solution successful if and only if I have food on the table, good friends, and 30 hours a week to write… but, as long as I get that, I’m agnostic about what the specific solution is”).

Having clear goals makes it much easier to find creative solutions to the problems.

What I’m saying here is in tension with an old post of mine called “Everything that turned out well in my life followed the same design process,” where I say (if I am to caricature myself a bit) that all the good stuff in my life have come through a process of unfolding where I paid attention to interesting opportunities and iterated, without worrying where the process would take me. So, which is it? Should you follow your curiosity and trust the process? Or should you define a clear goal and pursue it agentically?

Walking up and down the road by our farm, I conclude that it is a false dichotomy. I can have goals and unfold (it might even be that I need goals to unfold), but it has to be a specific kind of goal.

What tends to go wrong is, as I said, when I have goals that define what the solution will look like. But the other type of goal is compatible with unfolding—the goal that defines what I want to achieve, what functions I want your solution to have. To give myself a chance, I need to be free to iterate and learn from the context and try different approaches toward the goal. But I don’t need to be free of goals. I think I was a bit confused about this in the essay from last year.

2. As is common among agentic people, Herzog often chose to do a lot of things himself—writing, producing, directing, holding the boom mic, recording music, sometimes filming, sometimes editing—since this type of vertical integration allowed him to move faster and save money. When working with Hollywood teams, this insistence of doing things himself was seen as amateurish and strange. But he could never have done what he did if he worked in the default Hollywood way, relying on large teams and specialized professionals.