The author is on to something. I would put a different emphasis. I think higher level expertise tends to not map well to words at all. There are some people who develop an incredible pedagogy, but it is a tiny minority of experts.
Frankly, the convo with Herbie isn’t super informative. What they are talking about is necessarily vague.
“In order to know how good you are at something requires exactly the same skills as it does to be good at that thing in the first place,” Cleese elaborates, “which means — and this is terribly funny — that if you are absolutely no good at something at all, then you lack exactly the skills you need to know that you are absolutely no good at it.”
Of course, that isn't quite true. If the result or outcome of an actvity is publicly observable or measurable, then a secondhand sense of judgement can be developed.
There are multiple definitions of "good at {something}".
For example, the n men in the last men's 100m Olympics sprint were not the n men with the best sprint technique. (Some were on both lists, but some weren't.)
I've come across this before. I call it ineffable knowledge.
The expert may know that they know something, yet isn't always able to
articulate it. In creative areas like music, cinema and sound design
people are able to make extraordinary connections and skilfully master
tools (virtuosity) without necessarily being able to explain. With
music the skill resides in motor circuits and the nervous system
itself which may not be amenable to introspection except in dreams. [1]
I first encountered this via analytical knowledge elicitation for
making "blackboard" expert systems - the most "talented" experts (who
we were trying to learn from) would always say "let me just show you".
That's when I realised that the oft repeated phrase "Those who can do,
those who can't teach" is possibly the most ill-informed sentiment
ever expressed. To teach one needs to be an expert, know that one is
an expert, and communicate the nuances to others. It specifically
requires that you understand the ways in which your students are not
expert in N different ways, for M individual students.
As the author describes, there's a point where vocabulary and 'vista'
means you see and think about things in a qualitatively different
way. Someone in the top 1000 is playing a different game than someone
in the top 10,000. And each of these is unique because expertise
individuates past a certain point. In some sense PhD's are all unique
experts in a field of one.
This is frustrating if encountering people, say in interviews or
debates, and you realise that they are not actually able to grasp what
it is you know within some "test/frame" they are trying to apply to
you. Whose 'fault' is that? Surely ones own. So the next milestone is
learning to see that gap and be a good communicator,
A take-away I get from the article is that recruiters are poor at
their job because they are not expert enough. Cookie-cutter HR is a
bane for business.
[1] Descartes' Error: Emotion, Reason, and the Human Brain. Damasio
I'd have to look up the book again, which is sitting in a shelf somewhere at home, about knowledge management. In there, if memory serves well, someone says that people who achieved true mastery of a something look like they have no clue whatsoever about said topic. I always understood that in the sense that those people are beyond talking about formalisms (usually the hallmark of "expert", just throw abbreviations and lingo around) and know to a high degree of certainty what they don't know.
That being said, I did train /teach supply chain management before in my life. And that showed me that being good in a subject has nothing to do with teaching that subject.
EDIT: After having read the essay beginning to end, I'd like to add one more observation: If you are new to a situation and discuss it with the resident expert, you can measure your level of mastery if that discussion of some serious potential challenges end with you, the new guy, saying "Fuck". And the resident expert responds with a slight smile and a nod. Usually, following this revelation that you now have a peer realizing the same issues, the really productive work can start. I love those moments, not the least because they are so damn rare.
Author here. Quite the contrary, actually — I’ve read all the above sources. On top of that, with the exception of Chambliss (which is a landmark work, to be clear), and Dreyfus (which I prefer to look at adaptations of; Accelerated Expertise has several updated or derivative scales that expertise researchers in the military use), I’ve talked to collaborators of these researchers, and in the case of Klein, am actively in contact with and am following the rest of the work of his community (https://naturalisticdecisionmaking.org/).
I’ve updated the bottom of the post to link to the many other articles I’ve written on expertise. These are either summaries of papers, or summaries of branches of the research, or pointing out obvious holes in the popsci representations of the research, or synthesise various (mostly military-funded) approaches to accelerating expertise.
This particular essay is a note from personal application.
This is related to what I call the Google approach, but that's probably a dated term as it relates to the original Google search.
Basically, if you try to sell people on a really cool algorithm or other geeky stuff, they simply won't care because they don't understand.
Put all that nerdery behind a simple interface that lets people type stuff in and get a good answer and suddenly you've got a business.
The trick then is, figure out a business model where you won't be held back by the lack of expertise of your customers. Only talk with them about their desires, not what you're doing to achieve them.
Following this thread could lead to all sorts of weird places like a group of programmers getting jobs in different places and then automating what they are asked to do. But instead of giving that automation to people who won't appreciate it, just use it to complete your day's work early and spend the hour further automating. Again the difficulty shifts from convincing your customer/boss to understand your ideas to somehow hiding the cleverness behind some kind of wall so that they feel like they're getting value for money and have no need to micromanage you into mediocrity.
And then you realise this is basically what software companies do.
Kind of like Coase's theory of the firm but with expertise mismatch as a friction cost.
The most effective way to get non technical people excited about a technology is first, show them a working version of it, then show them what they can do with it. Most people don’t care how innovative something is unless it benefits their life in some appreciable way
One of the biggest mistakes I see DIY youtubers and bloggers make is jumping right into the process of building something without first showing the end result they're pushing towards. A few do this under the mistaken impression that it builds suspense, but that only works for those who already have solid followings. For most, it just means they get skipped over, with the audience moving on to something else where the payoff to them for watching is clearer.
re: Federer's backhand. A one-handed backhand has some flaws that are exposed by Nadal's crazy topspin. The sweet spot is to hit it waist high, like a baseball batter. If the ball jumps very high, then you (1) stand way behind the baseline and wait for the ball to drop to waist level, or (2) hit a weak slice from shoulder level. On (1) you're so far back that Nadal can run you around the court. On (2) Nadal will destroy your weak return. Hitting the ball immediately after it bounces is risky. You have less time to get in position, the ball can bounce erratically, and you generally take a shorter swing. Often people hit a slice on the rise. Hitting w/ topspin on the rise requires impeccable timing and finesse. When it works it's beautiful.
The author's core point is that climbing higher in any skill tree will often equip practitioners with a deeper index of understanding across that discipline. Reaching the "vocab point" as described in the article means one can participate in extensive discussion around the subject and skill; even when lacking the nuanced vocabulary to articulate certain observations or feelings, there is an intuitive and implicit recognition of the skill's shared language, themes and motifs.
I think this is an excellent mental model to evaluate our capabilities. I find the ability to analogize, to tell stories about vastly complex things in a way that is simple and digestible to a novice, is another hallmark of expertise. Richard Feynman used to say something like, "If I can't explain a subject simply, I don't understand it well enough" (paraphrased).
I'm the least technical team member in a scientific simulation company, yet I never feel out of place. In part, that's because I've invested effort into my ability to learn, but it's much more due to everyone else's ability and motivation to help me understand what's going on, to walk me through some of the layers of physical and digital abstractions, to call my attention to the history behind a dependency.
Analogies are the original abstraction, and they allow us to quickly and intuitively create a shared, contextual vocabulary for learning transfer. I wonder how much further there is to climb up this metaphorical tree!
18 comments
[ 105 ms ] story [ 97.9 ms ] threadFrankly, the convo with Herbie isn’t super informative. What they are talking about is necessarily vague.
“In order to know how good you are at something requires exactly the same skills as it does to be good at that thing in the first place,” Cleese elaborates, “which means — and this is terribly funny — that if you are absolutely no good at something at all, then you lack exactly the skills you need to know that you are absolutely no good at it.”
For example, the n men in the last men's 100m Olympics sprint were not the n men with the best sprint technique. (Some were on both lists, but some weren't.)
The expert may know that they know something, yet isn't always able to articulate it. In creative areas like music, cinema and sound design people are able to make extraordinary connections and skilfully master tools (virtuosity) without necessarily being able to explain. With music the skill resides in motor circuits and the nervous system itself which may not be amenable to introspection except in dreams. [1]
I first encountered this via analytical knowledge elicitation for making "blackboard" expert systems - the most "talented" experts (who we were trying to learn from) would always say "let me just show you".
That's when I realised that the oft repeated phrase "Those who can do, those who can't teach" is possibly the most ill-informed sentiment ever expressed. To teach one needs to be an expert, know that one is an expert, and communicate the nuances to others. It specifically requires that you understand the ways in which your students are not expert in N different ways, for M individual students.
As the author describes, there's a point where vocabulary and 'vista' means you see and think about things in a qualitatively different way. Someone in the top 1000 is playing a different game than someone in the top 10,000. And each of these is unique because expertise individuates past a certain point. In some sense PhD's are all unique experts in a field of one.
This is frustrating if encountering people, say in interviews or debates, and you realise that they are not actually able to grasp what it is you know within some "test/frame" they are trying to apply to you. Whose 'fault' is that? Surely ones own. So the next milestone is learning to see that gap and be a good communicator,
A take-away I get from the article is that recruiters are poor at their job because they are not expert enough. Cookie-cutter HR is a bane for business.
[1] Descartes' Error: Emotion, Reason, and the Human Brain. Damasio
That being said, I did train /teach supply chain management before in my life. And that showed me that being good in a subject has nothing to do with teaching that subject.
EDIT: After having read the essay beginning to end, I'd like to add one more observation: If you are new to a situation and discuss it with the resident expert, you can measure your level of mastery if that discussion of some serious potential challenges end with you, the new guy, saying "Fuck". And the resident expert responds with a slight smile and a nod. Usually, following this revelation that you now have a peer realizing the same issues, the really productive work can start. I love those moments, not the least because they are so damn rare.
The mundanity of excellence by Chambliss
Dreyfus model of skill acquisiton
http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA084551&Location=U...
Sources of Power by Gary Klein
Peak by Ericsson
Surprised not to see any nods to the giants whose shoulders this is standing on.
I guess it's disappointing is all.
I’ve updated the bottom of the post to link to the many other articles I’ve written on expertise. These are either summaries of papers, or summaries of branches of the research, or pointing out obvious holes in the popsci representations of the research, or synthesise various (mostly military-funded) approaches to accelerating expertise.
This particular essay is a note from personal application.
There is no requirement for a blog post to list every article the author read!
Accusing someone of plagarizing is a very serious accusation, one that should be made with more care than you are making.
Basically, if you try to sell people on a really cool algorithm or other geeky stuff, they simply won't care because they don't understand.
Put all that nerdery behind a simple interface that lets people type stuff in and get a good answer and suddenly you've got a business.
The trick then is, figure out a business model where you won't be held back by the lack of expertise of your customers. Only talk with them about their desires, not what you're doing to achieve them.
Following this thread could lead to all sorts of weird places like a group of programmers getting jobs in different places and then automating what they are asked to do. But instead of giving that automation to people who won't appreciate it, just use it to complete your day's work early and spend the hour further automating. Again the difficulty shifts from convincing your customer/boss to understand your ideas to somehow hiding the cleverness behind some kind of wall so that they feel like they're getting value for money and have no need to micromanage you into mediocrity.
And then you realise this is basically what software companies do.
Kind of like Coase's theory of the firm but with expertise mismatch as a friction cost.
I think this is an excellent mental model to evaluate our capabilities. I find the ability to analogize, to tell stories about vastly complex things in a way that is simple and digestible to a novice, is another hallmark of expertise. Richard Feynman used to say something like, "If I can't explain a subject simply, I don't understand it well enough" (paraphrased).
I'm the least technical team member in a scientific simulation company, yet I never feel out of place. In part, that's because I've invested effort into my ability to learn, but it's much more due to everyone else's ability and motivation to help me understand what's going on, to walk me through some of the layers of physical and digital abstractions, to call my attention to the history behind a dependency.
Analogies are the original abstraction, and they allow us to quickly and intuitively create a shared, contextual vocabulary for learning transfer. I wonder how much further there is to climb up this metaphorical tree!