Demystification versus Understanding

I am getting really interested in distinctions between types of knowledge these days. I think these distinctions are very important to the invisible structure of mental models.

One distinction whose importance I have come to increasingly appreciate, is the one between demystification and understanding. Both are types of appreciative knowledge. I define them as follows:

• To demystify something is to understand it to a level where you no longer feel anxious about your ignorance.
• To understand something is to have the same priorities as experts regarding that something.

The latter is in fact an implicit chicken-egg definition of expertise.

There is a shallow sense in which I can come across as very “knowledgeable.” Very few important things utterly mystify me that do not also mystify everybody else. When I encounter a new idea, I usually have some way to parse it. I am rarely at a loss over what to make of it. But this knowledge is only slightly deeper than the knowledge of a librarian who knows how to classify a book on any subject in a catalog.

So what is real understanding? Why is having the same priorities as experts a good test?

An example of a subject that mystifies me today is holography. It is a very important idea, and not just because we might get true 3D TV out of it some day. It is important in an information-theoretic sense, for understanding esoteric things like black holes and the difference between two and three dimensions. It’s been on my to-do list for quite a while to demystify it for myself. I expect it will take a day of reading.

But this level of understanding is far short of anything you could call actual knowledge. To demystify airplanes, you just need some very simple explanations based on high school physics, and a picture of an aerofoil section with some 101 fluid mechanics equations next to it. But to actually know how planes work, you need to understand many more things. These are crucially important but not mystifying details. For example, did you know that for really high-performance jet engines, turbine blades are effectively a single crystal? Or that at high velocities a phenomenon called aeroelastic coupling can make the metal wings of airplanes flutter wildly?

Almost certainly, you will not be able to guess what questions seasoned airplane designers ask when designing a new plane, based on a high-school understanding of how planes work.

You generally learn about these important-but-not-mystifying things when you go deep enough into a subject that you have to get on the same page with experts about what is important. At that point (assuming you are not just agreeing blindly), you are an expert yourself.

Expertise is, in a sense, a consensus of priorities.

This is why experts get exasperated with novices who stick their noses into expert domains. The novice is generally on a confidence high — Dunning-Kruger variety — that is the result of having achieved a level of demystification. But this does not mean he knows enough to contribute usefully.

One way to understand this is that knowledge is cumulative and expertise is the ability to operate at the edge of the known. Newton’s laws are certainly the most important element in designing spacecraft orbits, but they are also the oldest piece of the puzzle. They are so well understood by now that the coarse, first order paper-napkin calculations for a Mars mission can be done by any aerospace engineering graduate student. Getting the basic mission profile computed, along with fuel and weight calculations, would take an experienced aerospace engineer perhaps a couple of weeks.

But to land a heavy rover on Mars, something at the edge of our capabilities, all the relevant knowledge is at the edge.  Which means the experts involved are probably arguing about priorities and trade-offs that non-experts are not even competent to recognize or frame, let alone resolve correctly.

So in general, Russell was correct: when the experts disagree, the lay person had best reserve judgment.

But there is an exception to the rule. Expertise also comes with taking many basic things for granted. So when radical changes happen, sometimes it is the naive novice, wrestling with the basics, who ends up innocently asking the right questions. You can only re-examine foundational assumptions if they are not ingrained second nature for you.

Thinking like a novice: the Zen idea of “beginner’s mind” is really hard for an expert. Which is one reason disruptive changes are often triggered by relative outsiders and smart novices. But not so often as romantics like to think. I suspect “experts thinking like novices” happens more often than novices serendipitously asking the brilliant right questions.