If you’re currently plateaued at something, chances are you’re hitting some constraint. There is something that is stopping you from getting better at the thing, and if you don’t identify that constraint and try to fix it you’ll get nowhere.
—David R. MacIver
My brother-in-law recently showed me a clip of a Swedish TV show that visited the small town where my wife, Johanna, grew up. The show hosts got out of their car, walked up to an old drunk outside a supermarket, and asked, “Is it possible to find love in this town?” “Yes,” said the drunk, “but you need to have a . . . peculiar taste.”
It isn’t exactly a place that encourages the life of the mind. When I met Johanna, far from there, in a university town, she made a deep and perplexing impression on me: the precision of her attention and the intensity of her curiosity were unlike anything I had seen, and yet, despite this, she was almost completely ignorant. If I mentioned the First World War, she would say, “When was that?” But if I explained an area I was researching, it would take her five minutes to cut through to the deep, underlying question that had eluded me for months.
One day we were out for a walk in the pocket park a block from my apartment. She was complaining about how little use she had of her classes in teacher ed. The faculty didn’t seem to have a grasp of learning psychology, nor of the practical skills necessary to do a proper job as a teacher. Besides, there had to be better ways of going about education than what she had been exposed to—this was a topic she often returned to.
I said, “You know you can just teach yourself, right? You don’t need a course.”
She stopped mid-step on the gravel path.
Later that day, Johanna biked up to the university library and packed a bag full of books picked semi-randomly from the shelves of education history. Then she sat down to read. She read every spare minute for a year, two years, three . . .
I find these transitions lovely. Somebody notices a limitation in themselves, something they lack or have misunderstood, something they need to learn or unlearn—and then they learn it, and discover, on the other side, a richer world. For Johanna, her blindspot was a lack of understanding of herself and her agency. But these hidden limitations can take many forms. It can be a skill you lack; a lacuna of knowledge; an attitude that is coming between you and the things that you value; a needless constraint; a fear. Often they have the character of doors hiding in plain sight: obvious once you see them, but quite possible to walk straight past for years (or a lifetime).
Once I catch sight of a limitation and set to work overcoming it, I invariably say, “I can’t believe I didn’t do this before. It feels so alive.” That’s how I felt when I noticed I wasn’t a great listener, and decided to practice. And it is what I felt when at 30 I admitted that I was, in many ways, a weak writer, and chose to stare at that flaw.
I assume there are many more opportunities to grow like this if I can only get better at seeing my blindspots. How do I do that?
Looking at things that break
But what we need is an approach, a language which is adequate for the subject... —Oliver Sacks
These days, Johanna and I collaborate when writing. On Escaping Flatland, we do all ambitious pieces together (the small weird ones I do on my own). A consequence of this is that we’ve had to put words to many of the things that I had previously done intuitively. We walk in circles on the meadow below the house, hashing out what it means for an idea to be mature, what the difference is between a “draft” and an “approach,” and so on.
In itself, making the process explicit in this way has helped us spot limitations—like how I would start an essay before I had gathered my thoughts, and then get attached to what I had written in a way that made it hard to refine my thinking.
But what is particularly useful is this habit Johanna has. She writes a note every time she feels that something is off, however subtle that feeling is. It might be a confusion in the writing, or a misunderstanding between us, or anything else that goes wrong—a slight frown will play across her face, then she’ll reach for her workbook and make a note. She is on the hunt for things that don’t seem right. All the big, frustrating failures, but also the hairline fractures that might hint at a deeper problem.
Darwin, who kept a notebook where he wrote down facts that contradicted him, observed that frustrating, cognitively dissonant things were the first to slip his memory. This is Johanna’s impression, too. If she doesn’t note those reactions immediately, they slip her by. It is like a rabbit that shoots up from the grass when you walk in the meadow. “You have to grab hold of her when she leaps,” Johanna writes in the margin of a draft. “And then you skin her quick.”
By skinning, she means to articulate why something feels off, to flesh it out, so you can see if it contains information that will help further the work. Let me give you an example. In the first version of this essay, we talked about “finding bottlenecks.” It was about what in management is known as the theory of constraint. But something about this felt off to Johanna. Why was that? A bottleneck refers to the slowest station on a production line, which, because production is linear, sets the speed of the entire line. To speed up production, you have to find the bottleneck and ease it. But is that a good metaphor for the types of things we are talking about? Does essay writing have bottlenecks like that? Does being a good partner? Bottlenecks form when there are clearly separated tasks that follow each other in a set order. Most things are not like this. In writing, it is quite possible to start typing before you’ve done the tasks that should come before it, and, besides, I’m never sure how to tell if the preparation is finished or not. In other words, the reason talking about bottlenecks felt off was that it mapped badly to the problems we cared about. So we went back to the drawing board.
Often the questions that arise as Johanna skins a rabbit are not relevant to the project at hand, but still seem promising. These she saves in her workbook and returns to later.
Johanna genuinely likes finding problems. I have more complicated feelings about it. If I discover, say, that the narrative that holds together a draft bores me, my first reaction is, “But it has to work. It took me a week.” These days, though, I don’t act on that thought. Instead, I tell Johanna about the problem. Then I put on my jacket and go see if the cranes are passing by going north (one of Johanna’s flaws is that she forgets that people get upset by these kinds of things, so I need some time alone before I can talk about it). Why did I fail? The narrative in the last essay felt so alive—why can’t I achieve that now? I hear the cranes call through the fog. When I get back, Johanna has made coffee and we talk it over. The pain point I wanted to ignore reveals itself as a source of insight.
What are the characteristics of a good narrative frame? How could we have evaluated that the one we settled on was boring sooner? In the light of that, what should we do with this essay?
For some reason, meditating on failure brings out our process, and our assumptions, in a way that success does not. (It is also useful to contrast success and failure, ideally several examples of each.) This reminds me of Paul Broca, the 19th-century French anatomist. He mapped the brain by sawing open the heads of people with aphasia and comparing their brains to normal brains so he could see which part was damaged in those who had lost their speech. Malfunctions mapped the function of the brain.
Essays that spring fully formed into my head and magically just work (not common!) do not tell us much about the mysterious process that goes on inside the head as an idea matures, etc. But studying the frustrating cases, the failures, the dead ends, gives an abundant stream of insights. Every frustration and failure tells something about the process and its limitations. It is not always obvious exactly what it tells us. Was it a fluke? Or is there a pattern? But by mapping the breakdowns month after month, we have been able to fill in our blindspot about how good essays evolve (or fail to evolve) from insights to finished pieces. Our process has, as a consequence, been overhauled.
Many high performers go a step further and actively push things until they break. They want to increase the number of failures they can learn from. Musicians speed up their playing to figure out which finger positions make them stumble, so they can practice that. Factories speed up production lines to see which station can’t keep up, so they know what the engineers need to work on. Startup founders set (seemingly) impossible goals to pinpoint the assumptions that makes it seem impossible, so the assumptions can be tested or circumvented. They want to accelerate the rate at which they find needless limitations.
Writing about this makes me curious about doing this in more domains (for example in parenting), to increase my sensitivity to problems and hard-to-spot limitations.
Studying others to see what they do that is beyond you
One day in 2022, as I was doing the dishes, Johanna called to me, turning her laptop to show a video. Fog. A dog walking on a path through a garden—the greenery was so abundant it was otherworldly. It was like one of Anselm Kiefer’s mad sprawling art installations, where he turns an entire army base into a surreal wonderland, except it was all plants.
“What is that?”
“It is from this guy called Dan Pearson,” said Johanna. “It is his farm.”
This was about a year after we had bought our house. Johanna was thinking about what to do with the six acres we now had to tend. She knew nothing about gardening, but from suddenly having land she was attracted by the challenge of using it well. She had begun teaching herself garden design a few months before by skimming textbooks, handbooks, overviews, and so on—working through 10 to 20 books, in a month or two. She was mapping the field. Between her reading sessions, she would walk our grounds. She observed the forest of maple, pine, and oak; the meadow; and the spaces around the houses, to get a sense of what areas of gardening and design were relevant in our context.
This taught her plenty. But learning from books soon hit diminishing returns. What they taught was not high resolution enough to help her explain why some gardens are much more appealing than others, nor how to create a garden that felt right for our specific plot of land. Using what she had learned, she was unable to solve the design problems we faced to her satisfaction. There was something she lacked. Reading the textbooks more closely didn’t tell her what it was.
Noticing this, Johanna shifted to massive input learning. People who perform at the highest levels have deep stores of tacit knowledge that textbooks miss. By looking at thousands of pictures of great gardens, Johanna hoped to see patterns that were too subtle or complex to be reduced to a concept and communicated in a book.1
She did this mainly by curating an Instagram feed where she could see the season unfold in gardens that resonated with her. It was like an AI train run; I could see the capabilities emerge. At first, she couldn’t even tell where one plant ended and another began when they bled into each other in an arrangement, but after seeing the same plants in multiple constellations, they became individuals to her. She learned their names and their growing patterns. Our land was suddenly awash with names and we saw it anew.
She mainly looked at gardens that struck her as exceptional (Great Dixter, Rousham Garden, Sissinghurst, Beth Chatto’s gardens). But she also, because of the messiness of the process, ended up seeing gardens that felt off to her, which was useful, too, since articulating why they felt off helped make certain design patterns and concepts salient.
This kind of massive input learning was what she was doing the morning when she called out to me, asking me to look at the video of a dog walking through Dan Pearson’s Hillside farm. The garden was unlike anything she had seen so far. To talk with Christopher Alexander, Hillside had an unusual degree of garden-landscape fit. It was well-integrated on all levels Johanna could perceive—the colors, the arrangement of the plants, the sections, and the echoes between them—and it all sat perfectly in the landscape, the larger wholeness.
Johanna wanted to know what Pearson knew that she did not and which allowed him to do this. As far as this was possible for her to learn. Lacking direct access to him, she poured through everything he had written over the last 40 years (hundreds of diary-like pieces detailing his process and work), as well as every interview and lecture available, and, first and foremost, pictures and films of the gardens. Each source in itself was of limited use. You can’t learn gardening from an interview, or from looking at pictures. But the sources were limited in different ways, so bouncing between them, they illuminated each other. A lecture could highlight patterns in a garden that Johanna had failed to see; while carefully studying hundreds of pictures from the same garden could bring out things that Pearson’s words only gestured at. Johanna would imitate what Pearson described doing which was also illuminating—going out into our fields looking, making trial beds, and sketching in the way he does—it helped her test her understanding and opened questions that enriched her reading.
By getting a deep sense of what it means to do exceptional work in gardening, Johanna’s limitations came into view. She now had a sense of what Pearson could do that she could not; what he saw that she did not; whom he had studied that she had not. This pointed out things she needed to learn to approach the level of skill, awareness, and taste she was after. It was a recursive process. The more she learned, the better she could see how little she knew.
Dan Pearson would note, for instance, that when he was making the final plans for his big flower garden (after five years of preparations!) he walked the land, imagining, feeling inwards, where he would want height and air, respectively. These kinds of throwaway remarks were useful insights into the process; they made things visceral and real. But they also raised new questions. What precisely was Pearson considering when he decided that a particular spot needed high plants, and another needed air?
Johanna had entered a loop where her constraints came into view—and dissolved as she approached them, revealing further, subtler, or deeper constraints.
When you find people who perform at a high level, and whose work you love, there are layers upon layers of insight to be gleaned from their work and the process behind it. If they are at the top of their field, it is sometimes the case that no one else has the mental models they have, it can’t be learned any other way; it can’t be found in a textbook.
Learning to see these layers takes time. They come into view one after the other as you attend to the work. Exactly how long this can be worth doing before you hit diminishing returns is itself something you may learn by studying others; it is not intuitive. Most people stop too soon. When it comes to the highest achievements in a domain, you don’t know how much there is to see there until you learn to see it. Our six-year-old, Maud, sighs at Johanna, saying, “Oh, but mom, you are always looking at the same pictures.” Yes. It’s true. She has returned to some of the pictures of Hillside for more than a year now.
But although they might look like a tangle of flowers to a child’s eye, they are dripping with patterns to be found.
Feedback from others
We are about 3000 words in now, so it might be time to remind ourselves of what we are doing. We are talking about how to find limitations—those often hard-to-spot doors that can open a new world to us. Many times, we sort of know where they are, but fail to see them fully until we take a step back and spell out what we’re doing and where that fails. Other times, we don’t know what we don’t know. Then studying others can help us see the landscape and get a sense of what is possible.
You can also do the inverse of the last strategy and ask other people to look at you. Getting people to honestly share their feedback with you can be a way to extend your perception so that you see more problems and possibilities. They might see things about you that are in your blindspot.
The point isn’t to do what other people tell you. Just as they see things that you don’t, you see things they don’t. Often, when people tell you what to do they don’t understand what you are going for and project their interests onto you. This will kill your best ideas if you let it. But used right, feedback from (competent) others is a way to get more data so you can see further.
If you read about people who have done exceptional work, they have usually had friends and mentors who felt comfortable giving them direct and deep feedback.
PHILIP GLASS: People talk about their lessons with Boulanger as being like having a brain transplant. It was like that. [. . .] It was very intense. I usually started around seven in the morning and finished at seven at night. [. . .]
CLAIRE CHASE: If you were to locate the most essential lesson that she gave you, what would it be?
GLASS: [laughing] Makes me laugh when you say that. She was full of advice. The thing that she said like a mantra was, “Il faut faire un effort.” You must make an effort. She said that constantly. She made me feel lazy. [. . .] I thought I knew everything. Of course, I knew nothing. I went to her. I took about what I thought were my best pieces. I was a Juilliard graduate and had a master’s degree, it didn’t matter what it was. I put my music on the rack of the piano. And she could speed read. [Glass indicates that it took her a second to read a page]. She was reading like that through three or four pieces. Finally, she stopped and she said, “That measure there was written by a real composer.”
Getting that kind of exact, deep, and demanding feedback is invaluable if you want to figure out what your limitations are, what you need to work on, and what your potential is. But it can be tough. It is tempting to shy away from it.
In 2014/15, I wrote a novel. Reading an early draft Johanna noticed that the basic premise didn’t hold together. Addressing this problem, however, would have meant throwing out four or five months of my work, which she thought I couldn’t handle. So instead she gave me superficial feedback (talking about descriptions and sentences) and looked for ways to save the project. This was useful and helped me find my voice, but it didn’t, ultimately, save the project. It kept me from learning what I needed to learn.
When Johanna learned to trust me enough to give more sincere feedback, I did, indeed, find it stressful. If she pointed out that I had failed to do my research, I would say things like, “But this essay has to ship next week. What you are saying means I have to throw out everything—” And she would say, “Except one paragraph.” I would come up with reasons why doing what I had done was good, why I didn’t have time to make such a drastic rewrite—claiming we had to ship next week was just something I made up as an excuse not to look at my mistakes—and so on. My stress made me unable to think about the substance of what she was saying. When you act like I did, you encourage people to lie to you.
A lot of people send me drafts of blog posts asking for feedback, and it is interesting to notice when I feel compelled to give my deepest insights and when I feel like honest feedback isn’t what is asked for. When I get some version of, “What do you think?” I tend to read that as, “Am I ok?” Yes, you are, and I don’t mind telling you. But I get much more excited about talking craft when people write stuff like: “I really want to get better at writing. That would be so much fun. Could you look at this thing I wrote and see what I need to work on?” If you want to grow, it is important to think about what kind of energy you project when asking for feedback, so that people feel comfortable giving care in a demanding way, and so they know at what level you want to be seen. Most people will default to giving praise and pointing out surface-level details. It is up to you to prompt them to reveal deeper problems.
It took Johanna and me a few years to figure out how to be sincere about the flaws we saw in each other. I’m not sure what changed to make it less uncomfortable. Partly it was exposure training, I guess. If you get enough feedback, you figure out that having your failures pointed out doesn’t hurt you. It is just data.
But there was also a deeper mental shift for me. I transferred my loyalty away from the thing in front of me and toward what I could achieve. Much as working with a piano teacher is not, fundamentally, about learning songs, but about using songs to push yourself; I now think of our projects, not as ends in themselves, but as means to help us improve the underlying process and ourselves. This helps put me in the right frame of mind. I want this essay to turn out well, of course. But the goal of being honest with each other when we examine the drafts isn’t to figure out how to make a particular draft work, but to see what deeper possibilities it points out and then pursue those, even if it means trashing the draft.
When I look at the kids Johanna and I was when we met, when I look at our confusion and our fumbling attempts at being sincere with each other, I feel a deep gratitude toward them. When they found their limitations, they could transcend them, and this brought us into being. We have the same responsibility to the people we will be a year from now. Looking for limitations is about extending care to your future self.
As we walk the grounds around our house, our daughters in tow, we talk about what we need to do next to keep growing. What are we not seeing? We spell out what we are trying to achieve and how we go about it; we study others; we give and take feedback. Then we write it all down in this essay, to see if you can help us see more clearly.
The paintings in the essay are both from Monet’s water lily series.
This type of massive input learning works particularly well in domains like gardening, where there are no towers of abstractions (as in mathematics), but just tons of names and messy patterns. It works for language learning. José Rincón says it works well in biology.
An extreme example of this was Casey Handmer’s contribution to the Vesuvius Challenge, where they sought to read the content of petrified papyrus rolls that had been scanned using a particle accelerator:
The early months were a slog of splotchy images. Then Casey Handmer, an Australian mathematician, physicist and polymath, scored a point for humankind by beating the computers to the first major breakthrough. Handmer took a few stabs at writing scroll-reading code, but he soon concluded he might have better luck if he just stared at the images for a really long time. Eventually he began to notice what he and the other contestants have come to call “crackle,” a faint pattern of cracks and lines on the page that resembles what you might see in the mud of a dried-out lakebed. To Handmer’s eyes, the crackle seemed to have the shape of Greek letters and the blobs and strokes that accompany handwritten ink.
Man, sometimes when I read your writing I get the sense that more is possible in my own life. I'm in the thick of that process of figuring out how to truly live in partnership with others. Reading this felt like lighting the way a little more.
One thing I can contribute to this is that this applies to physical skills as well!
I spent years doing yoga and pilates. A trainer taught me barbells, and my strength sky-rocketed. A yoga teacher taught me training regiments for backbends and splits, and I made more progress in 6 weeks than I had in 7 years. I think having gone through that process prepared me for reading this post. It really is possible.