How to remember books and build useful knowledge

[Julianek]

@1000000CHF

How do you internalize books? You make notes? Deliberate practice? I’m asking seriously, as I tend to read a lot, but I’m unable to memorize anything that I read for longer than a week.

This is a very important question! My answer will be quite long, sorry about that.

I read a lot of books (way too much, according to @ma0), both because I enjoy it, and because I am curious about how things work. But as you rightly said, there comes a point where you figure out that you forgot more than 80% of what you have read (and I may be underestimating it). This is quite a sour taste: what’s the point of reading non-fiction, if you won’t remember any of it? All this time spent reading has been wasted (except for the possible enjoyment you had while reading).

Note: I am only talking about non-fiction books here. Fictions are a strongly different category, where the focus is enjoyment, and also possibly to get a glimpse on human nature).

So I decided I needed a way to remember what I read. Here is what I came up with:

1. The Feynman Technique

First, before remembering something/internalizing it, you should be able to understand it. In many cases, it is way harder than it seems. How many times did I try to talk to a friend about something I was reading, only to realize that I was not able to explain it in simple terms (i.e, I don’t really understand it)? Needless to say that if you don’t understand something, you won’t be able to internalize it in a useful way (or even worse, you may base future knowledge on it, without realizing that your assumption is shaky).

The most efficient way to check if I understand something is to use the following technique devised by Richard Feynman:

1. Choose a concept you want to learn about
2. Pretend you are teaching it to a student in grade 6 (i.e, a young student, reasonably intelligent, but who does not know advanced concepts).
3. Identify gaps in your explanation; Go back to the source material, to better understand it.
4. Rinse and Repeat.

2. Semantic Tree / Latticework of Mental Models

The second fundamental principle is that you will never be able to remember isolated facts. Do you remember when you were a student and you learned a lot of stuffs just before an exam, only to forget it once the exam was over? You don’t want to do that again. This works to get credentials, but it does not build useful knowledge.

Enter Charlie Munger and Elon Musk:

it is important to view knowledge as sort of a semantic tree — make sure you understand the fundamental principles, ie the trunk and big branches, before you get into the leaves/details or there is nothing for them to hang on to. - Elon Musk

You’ve got to have models in your head. And you’ve got to array your experience ‑ both vicarious and direct ‑ on this latticework of models. You may have noticed students who just try to remember and pound back what is remembered. Well, they fail in school and in life. You’ve got to hang experience on a latticework of models in your head - Charlie Munger

You get the idea: before reading about anything, it is important to ask if you already know the big ideas about the topic the book is about. Ideally what you want to do is to pick the big disciplines, and to learn and master the fundamental principles about each discipline.

If you want to read about biology, read Darwin first. Economics? Start with Adam Smith, and go from there. Those will be the “trunk” of your semantic tree - the basics on which you can attach everything else. I noticed that I remember a lot of things if I do it in this way.

But it is not enough to know principles, you need also to see how they interact between them, and how to make a synthesis. I’ll give examples below.

In practice

In real life, I take a lot of notes. Plain, simple, markdown text files which I organize using Obsidian (there are other cool similar softwares, like Roam for instance. The important feature is that Obsidian allows you to make links between notes, and also to see all the notes linking to the current link.

This way, you quickly realize that the notes that are linked a lot form the principal branches of your knowledge tree. Almost all your existing knowledge is based on them, so you’d better get them right.

Here is a snapshot of the graph of my notes:

The big nodes with a lot of links are currently about Investing (especially the study of competitive advantages), Biology and Psychology.

Here is how I proceed:
For instance, let’s say I want to read about biology. I start with the basics: one of the fundamental principles of biology is evolution and natural selection, so I will write the following note:

Natural selection is the following process:

• Some traits in a species present a lot of variance
• These traits are transmissible by heredity
• Because the population of a species increase geometrically, individuals are bound to struggle to survive because of a scarcity of resources
• Given a specific environment, some variations of a given trait provides a [[Competitive Advantage]] to the individuals possessing these variations
• This advantage will help their bearer to reproduce more easily, or at least to survive to be old enough to be able to reproduce. Disadvantaged individuals in this environment will be outnumbered after a few generations and go extinct.
• Over many generations, the accumulation of trait divergence and incremental advantages lead to the creation of new species

Then a bit later, I will learn about the Red Queen Effect, which derives from evolution, so I may write the following note, linked to the first one:

Because of evolution and species’ adaptation to their environment, species expend a lot of resources and efforts to just stay at the same place. For instance, consider lions and antilopes.

• Imagine one lion get a genetic mutation that allows him to run faster to catch antilopes. This brings him such a advantage over other lions that his gene is very likely to spread over the future population of lions offsprings.
• On the other hand, in the antilope population, one antilope will get a genetic mutation where it can for instance run faster to escape lions. This improves its odds of surival and reproduction, and soon enough the gene spreads to the entire antilope population
• Net effect: after many generations, both lions and antilopes are way faster than before, but none has a considerable advantage over the other. There has been a lot of expenditures just to stay at the same competitive place. This is the Red Queen Effect.

Then later, when I read about investing, I realize that I can find similar effects not only in biology , but also in business. For instance, I wrote here about how Swiss providers of 3a third pillars are in effect commodity businesses. Each provider will try to improve its offer. They will reinvest all of their profits in a new platform, or in digital offering, or whatever, but at the end the important factor is the price, which means that on a net business basis all their efforts will neutralize each other : they will have spent a lot of capital just to stay at the same competitive place: the profits of the new offerings do not accrue to their bottom line, but to the customer advantage.

So i write a third note, with two links; one toward the Red Queen effect, and another one toward commodity businesses (in the investing section). This is an example of how you can do a ssynthesis between disciplines. Every time you do that, your latticework of mental models get bigger and stronger, and you will remember stuffs better.

Misc

This also implies that you should choose your books carefuly: start with the fundamentals, master them. When you get a new book, figure out quickly if is going to build upon you knowledge. Sometimes, you will find texts that will shake your assumptions: this is quite important because you need to figure out if the author is right: if he is, you need to adjust your tree. If he is wrong, just ditch the book, it is not worth it.

Another thing: you don’t want to read all the books on earth. You want to read and re-read the best ones. So figure out what the best ones are