The Tacit Manifesto
I take all the help I can get from the parts of my brain that cannot speak.
Let me back-up this claim by telling three stories.
Story One: The Moniac

In 1949, William Phillips (later known for the Phillips Curve) was a freshly graduated economist. During his studies, he was taken by the ideas of Keynes and the Keynesian model of economies. However, he had a hard time calculating details using the model; the way the various parameters interacted over time was complicated to analyze by hand.
After he graduated, he decided to build a machine that would simulate a nation’s economy using Keynesian economics. He spent the summer in his garage working with tubing, pumps, valves, and tanks, and emerged with a hydraulic analog computer that he called the Moniac.
The Moniac solved nine simultaneous differential equations and, amazingly, is accurate to 2%. But, frustratingly for the people who used it, the Moniac could never reveal why it produced the graphs that it did. For the most part, the numbers jibed with reality, but there was no way to determine the path that lead to a particular result. And, every now and then, it produced wildly different answers when fed the same parameters twice.
Hence the frustration: when it produced an unexpected answer, was it wrong, or was it doing something not understood by its operators?
In many ways, the Moniac behaved like a human expert.
Story Two: The Dreyfus Brothers
Stuart Dreyfus was an engineer and early programmer. Hubert Dreyfus was a philosopher, best known for his work in the 1960s and 70s on the limits of artificial intelligence. You therefore have to wonder just how the pair landed a contract with the US Air Force to study why their pilot training wasn’t as effective as the Air Force wanted.
The brothers wrote an 18 page report, A Five-Stage Model of the Mental Activities Involved in Directed Skill Acquisition.
The paper argues that the acquisition of skill is a process of gradually internalizing experience, building first explicit understanding and later a more tacit, intuitive model.
Imagine someone learning tennis. They have an instructor, who shows them how to hold the racket, where to put their feet, and so on. For the first few weeks, these instructions are in the front of their mind; there’s not much capacity left to play a decent game. But, gradually, all this explicit learning becomes part of them; they stand a particular way without consciously deciding to, and so on. This frees them up to receive more knowledge–how to serve, perhaps. As time goes on, the mechanical side of the game becomes almost a reflex, so the instructor starts working on tactics and other advanced topics. An expert tennis player is not consciously thinking about their feet or the trajectory of the ball; that’s all being handled by the hidden, intuitive part of their brain. The knowledge has become tacit.
Bringing it closer to home: when someone learns to drive, every motion is a conscious effort. They are remembering what the pedals do, and when to signal, and LOOK IN YOUR MIRROR! The rest of us learn never to try to make small talk with a novice driver. But if you’re an experienced driver, you can probably drive from home to your grocery store without actually being that conscious of the mechanics of driving. Sure, you’re looking out for hazards and reading street signs, but did you signal that last turn? No idea; that would have been done automatically, reflexively, below the conscious level.
The Dreyfus paper is not to be taken as a literal description of learning; no one graduates from stage two to stage three with a certificate and a bunch of flowers.
Instead, it’s a compelling justification of what we all know: as you become more proficient at something, you handle more of it automatically. Tacitly.
Story Three: “We Know More Than We Can Tell”
The title of this section is a quote from the 1966 book The Tacit Dimension by Michael Polyani,
Polyani was an expert in many fields, from chemistry to philosophy to economics. In the latter two he favored explanations that did not rely on absolutes: central sources of power or information.
In the late 1940s he formulated the ideas of post-critical thinking, which argued that a purely objective understanding of how things worked was incomplete; there was always a personal slant injected by a person’s tacit beliefs.
In The Tacit Dimension, Polyani talks about this in more direct terms. When we ride a bicycle, or read a web page, or catch a ball, we can’t describe exactly how we are doing it. In normal bicycling, the rider makes about two to five subtle steering corrections each second. But if you asked them, they’d tell you they were steering straight ahead.
We only survive because we can do all of these sophisticated things tacitly–without thinking about them.
The Manifesto
So this is somewhat tongue in cheek; as one of seventeen writers of the Manifesto for Agile Software Development, I’ve always fancied going solo.
But, at the same time, I’m also serious. When I look at my life, I lean into situations where I can experience things and get feedback. And, as I get older, I’m learning to accept that my intuition should play a major role in my decision making.
In practice, this means that I constantly use little tricks to try to accelerate the creation of tacit knowledge.
I try to learn new things, because the payback in terms of ideas per day is so much higher.
When I come across something I want to incorporate into my mental model, I try to use as many senses as I can. If I’m reading, I’ll stop and reread the passage aloud, so I’m exercising both speech and hearing areas in my brain. I’ll scribble notes, and draw diagrams or sketches, engaging motor, spatial, and visual centers. I keep a daybook, where I write down things that happen, sketch the structures I’m working on, and doodle. Using pen and paper somehow seems to generate sensations that embed themselves more deeply.
I have a rule that I don’t start something until I know how to tell I’m done; I seek feedback, because feedback (positive and negative) reinforces the tacit knowledge.
I also have a rule that when I start something, I try to take the smallest steps I can, each tested with feedback. That way, I know when I stray off the path (which I do more often than not) sooner rather than later.
I also try hard to treat mistakes as a positive thing. At the very least, they’re a negative reinforcement of some tacitly-help belief. But, every now and then, they turn out to be the head of a previously unnoticed trail.
And, when I’m working in areas where I have some experience, I try to listen to my intuition. Your subconscious has no voice; it can’t shout “that code is wrong". Instead, it affects you at a more primitive level: you might feel uneasy, or you might want to stand up and pace, or you may feel a little queasy.
In The Gift of Fear, Gavin Becker says when he talks with people who were mugged or otherwise attacked, they described a sense of something being wrong prior to the assault. Being sophisticated, modern people, they shrugged that feeling away rather than crossing the street or turning around. He helps them to listen to that quiet feeling.
In a far less dramatic setting, that’s what I try to do, too. I’ve come to accept that my subconscious brain is faster and better at pattern matching than the part of me that does explicit thinking.
This may sound like “look at me–I do it good.” It really isn’t. Instead, I’m trying to acknowledge that what works for me may not work for you. Everyone has to find their own path, because we’re all different (if for no other reason than we each have a unique set of experiences, and hence we each have different tacit knowledge). So when I talk about what I do, the intent is to give you some food for thought; things you might be able to adapt.
My gut tells me that might be a good idea..,
In other news, my new book, simplicity, comes out March 5th. But, if you want to get it before then, it’s available now in the Pragmatic Bookstore. It will be in beta for a month or two before hitting all the regular channels later in the year.
Thanks for sharing, Dave! Just got the book. Your post reminded me of the 4 stages of competence, and I think it fits the current scenario where LLM's are becoming both a boost to perceived productivity and a barrier to acquired knowledge.
I'll share my comments once I read it! :)
I really like this. What is meant by "formal tuition"?