Intuition Machine
Published in

Intuition Machine

On Abstraction, Creation and Subtraction

Photo by Dragos Gontariu on Unsplash

The curse of knowledge is that experts become unconscious of their current knowledge and forget how they arrived at that knowledge.

The ‘Curse of Knowledge’ & foundational knowledge in teaching

One of the greatest flaws of current education is the assumption that comprehension leads to competence. It’s a completely upside-down model of learning!

Explaining complex narratives is also very difficult to get across when you begin with the conclusion and work yourself backward. This is how proofs are presented in mathematics. There’s a statement and you find a way to discover why it is true.

It’s a very different activity from model making. In model making, you start with parts and you build a whole with the parts. To understand how to create, you begin with mastery of the parts and how they compose with each other to get to the whole.

The first kind of thinking of decomposing a whole into parts is what you do when you attempt to explain something. But explanations fall on deaf ears if the listener does not understand the vocabulary.

To understand vocabulary is to know the meaning of individual words and know how words interrelate and compose together to make sentences. To be fluent in a language is to know how to create sentences in that language.

We don’t learn language by just understanding grammatical structure. We don’t learn musing by understanding music theory. We learn by doing. We learn by creating.

Unfortunately, we educate students only to explain. To recite back the information they previously received. We don’t teach students to create, rather we only train them to imitate and mimic.

Thus we only have students with conscious incompetence. They know what to mimic, but they lack the competence to create. Never do they reach unconscious competence.

Humans also have an inclination to find additive solutions over subtractive solutions.

It’s easier to spot the difference on the left than on the right.

But I must ask, how do you arrive at emergent subtractive solutions via a collective consensus mechanism?

Adding is favored over subtracting in problem-solving

The reason bureaucracies become more complex is that it’s easier to get consensus on adding something new than removing something old. Even in the realm of biological evolution, additive solutions are the more pervasive solution. Only through creative destruction do we actually do see subtractive solutions.

However, organism development and growth is not only additive processes but also a subtractive one. A pluripotent stem cell just happens to be expressible in so many surprising ways.

Are collective subtractive processes only possible through a kind of pluripotency?

Classical reality is constructed via a subtractive pluripotent quantum reality. It’s just counter-intuitive, but that’s the strange nature of reality.

What about consciousness and attention? Are they not also subtractive processes? How does our mind gravitate to a single sequence of thoughts even when there is a boundless multitude of concerns that are present in everyday perception?

Subtractive methods are in fact a superset to the notion of forgetting. How is it that humans leverage forgetting to create more powerful and more useful abstractions?

The weakness of Artificial Neural Networks is their inability to systematically forget. This is also related to the inability of these networks to also systematically create new abstractions. An abstraction process is also a subtractive process. Neural networks do not create simple models of this world because they are essentially a collective additive process. Neural networks are incapable of abstraction via subtraction.

Christopher Alexander in the formulation of design noticed the importance of not only additive processes but subtractive processes as well.

But what is it about subtractive processes that are so difficult? Well, have you ever wondered why every time you move to a new home that you accumulated even more junk? Why don’t you just get rid of the stuff you don’t use? Well, it’s because you just don’t know if you’ll ever use again the stuff you rarely use. I guess it’s time for me to consult Marie Kondo to see if any of her principles might be enlightening.

Managing technical debt is a subtractive process. Garbage collecting memory that has no references is a subtractive process. Death itself is a subtractive process. To effectively perform a subtractive process requires that decisions are made across many different scopes. There is a considerable amount of mental simulation required to understand the causal consequence of removing something. That is why best practices for refactoring code require a comprehensive regression test scaffolding. This is quite unlike a bottom-up consensus mechanism.

A consensus must be made across multiple scales in space and scales in time. Furthermore, it is a destructive process where progress erases the past. Rollbacks are many times, not a feasible option. In short, it is damn hard and unfortunately, biological brains tend towards what takes the least effort. Which is indeed the paradox. What usually takes the least effort requires less and not more.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store