The power of instruction is seldom of much efficacy except in those happy dispositions where it is almost superfluous. -Edward Gibbon
When I first moved into my apartment, I couldn’t figure out how to open the front door of the building. Half the time, my key turned perfectly in the lock—half the time, it jammed, and I had to wriggle it around randomly until I got it to click into position. Twice, the key got completely stuck, putting me in the humiliating position of having to ask the property manager to help me extract it.
Over time, I internalized the correct depth and angle I needed to insert the key at—to the point where, but for my immortalizing the memory in this blog post, I surely wouldn’t remember ever having struggled. Yet I can’t help but wonder: How many other “doors” are there in my life that, having long since learned how to open, I’ve forgotten used to stump me?
The truth is, to get through our days, we deploy all kinds of skills mastered through unconscious feedback loops: subtle cues we get from the environment, or the sensations of our body, or the expressions on other people’s faces. Through these processes, we learn to walk, drive, and dance; to pause and laugh at the appropriate times during stories exchanged at a cocktail party. For most of us, these behaviors eventually feel natural—but observe a child or adolescent who has not yet acquired them, and you’ll realize they’re anything but.
And while a guide can often point you in the right direction (“Here, try holding the key like this”), there are inevitably nuances that resist verbal description: implicit understanding which can be achieved only through a gradual process of trial, error, and correction.
This is the issue that’s been gnawing at me ever since I figured out how to gain entry to my own home—an issue which troubles me greatly as an educator: Most complex skills can’t actually be taught.
Barrier #1: The Curse of Knowledge
The first reason complex skills can’t be taught is because the very people most qualified to do so are often the least capable of communicating their wisdom to beginners.
You’ve probably experienced this at some point when talking to someone with deep expertise is a highly specialized domain. Known as the curse of knowledge, the basic issue arises when people get so good at something that they forget what it’s like not to be good at it. (Or, to be a little more generous to experts: They have internalized the patterns of their discipline so deeply that the world now appears to naturally fall into categories which are in fact constructed; more on that in Part 2.)
With practice, an expert can learn to anticipate the most common gaps in student understanding—but this is rarely, I think, a case of truly seeing the material through untrained eyes; the teacher just learns through experience where comprehension tends to break down.
For example, you can surely appreciate why a beginning reader might struggle to differentiate between the symbols ‘b’ and ‘d’—both of which look like a stick with a half-circle on the bottom. But being literate yourself, you can’t actually stop yourself from seeing these symbols as letters, and associating them with their respective sounds; your own education has imbued these symbols with unalienable meaning.
So the well-intentioned expert tries to make things simple: They come up with clever mnemonics, like how the shape of the word ‘bed’ looks like a bed (for differentiating ‘b’ and ‘d.’). They invent acronyms like BEEF and STEAL. They use metaphors and exercises to catalyze understanding. (Think of the meditation teacher’s metaphors, or a tutorial for seeing magic eye images.) But these techniques ultimately are just a catalyst—not the thing itself. True understanding is internal and undistillable. As psychologist and Substack-bad-boy
describes:Computer people have a good word for this kind of thing: lossy compression. You simply can’t fit a thought into a sound wave. Something’s gotta go, and what goes is its ineffable essence, its deep meaning. You have to hope that the other person can reconstruct that essence with whatever they have lying around in their head. Often, they can’t.
The curse of knowledge, in other words, is the cognitive limitation that leads us to give instructions like “Insert the key and turn clockwise” when the true recipe for success is more like “Insert the key at an angle of five degrees right of vertical until you feel resistance, then withdraw it two millimeters while applying a slight clockwise pressure until the key turns, then hold it there while opening the door”—though even this second version surely leaves out some even subtler aspects of the process.
And in truth, even if a teacher were able to break a complex skill down into that ridiculous level of specificity, it would only help so much. This is because true mastery entails not only knowing what discrete sub-skills are required; it entails seamlessly integrating and applying those skills.
Barrier #2: Expertise is a Menu, Not a Map
Let’s expand on our running metaphor. It’s true that, like learning to open a finicky door, learning a complex skill requires certain implicit or embodied understanding. But opening one door is a simple skill. Complex skills, by contrast, are less like internalizing the use of one key to one door and more like internalizing the use of a set of keys to many doors—each with its own idiosyncratic entry requirements and reasons to pass through.
Alternatively, you could say that expertise is more like a special vocabulary or toolkit or menu of options than it is an instruction manual or a map from A to B. As acclaimed author
reflects:The difference between me, now, and the writer I was when I first started is that now I can see more potential choices in a swath of text. I have a stronger, more reliable, system of preferences. And more trust in my gut response—more confidence that everything goes back to that.
Artist and educator Betty Edwards puts it this way:
As each new skill is learned, you will merge it with those previously learned until, one day, you are simply drawing—just as, one day, you found yourself simply driving without thinking about how to do it.
So yes, complex skills are made up of discrete sub-skills—however difficult to articulate. But those sub-skills are so diverse, so subtly different in application, so deeply integrated in the unconscious, that they can only be acquired through years of deliberate practice; lift the curse of knowledge, and you’re still stuck with a static list for what, in reality, is a dynamic interaction. (M.C. Richards: “Let no one be deluded that a knowledge of the path can substitute for putting one foot in front of the other.”)
Indeed, experts in domains as wide-ranging as chess and medicine even show different eye-tracking patterns compared to beginners—per the authors: “[E]xperts, when compared with non-experts, had shorter fixation durations, more fixations on task-relevant areas, and fewer fixations on task-redundant areas; experts also had longer saccades and shorter times to first fixate relevant information.” Mastery, in other words, leads people to look more efficiently, more holistically, and at more relevant information in the context of their given domain.
So the issue isn’t just experts forgetting what it’s like to be a beginner—it’s that the synthesis required for complex skills is so deep that it alters the very nature of perception. This synthesis is built from the inside—not taught from the outside. That’s a feat no curriculum can achieve on its own.
But Wait—There’s More!
In this post, I’ve argued for the essential impossibility of teaching complex skills. While expert guidance can help point learners in the right direction, it is ultimately practice which reveals the relevant sub-skills, and which helps integrate those sub-skills into a coherent “menu.”
In spite of (in light of?) this difficulty, it’s obvious that some curricula are better than others. In Part 2 of this series, I’ll dive into curriculum from the student perspective, and in Part 3, I’ll offer a speculative vision of what the ideal curriculum might look like.