Thinking through writing ru en

Thinking through writing is the practice of translating one's own thoughts into written form to uncover gaps in reasoning and construct a clearer picture of the world.

It’s what I mostly do here on this blog — and, to the best of my ability, at work.

I’ve finally found time to elaborate on this.

It may sound like a truism, but every word in the definition is essential:

Practice implies consistency.

A significant part of the benefit of thinking through writing comes from doing it on a day-to-day basis — thanks to emergent effects of accumulating written artifacts.

Still, even a one-time act of formalizing thoughts on paper will be beneficial — give it a try :-)

Translation implies a change in the form and the organization of information [ru]. You'll gain nothing if you start "literally" writing down images from your head. Translation — and, by extension, transformation — is a key element of thinking through writing — we'll talk about it later.

Own thoughts is an obvious component: if you write not thoughts or thoughts but not your own, then this is a completely different story, although also useful.

Written form implies that the work done should remain in the material world. For the purposes of this essay, let's interpret the writing form in a broad sense — as any form of recording information on material carriers, from text to graphs, diagrams, and even paintings.

As the other common practices, like "doing sports" or "learning languages", thinking through writing does not deliver one specific kind of benefit, especially instantly. Instead, it acts gradually in many directions. Because of this, it is difficult to explain the coolness of this practice in a short and concise way.

I won't try to outsmart the universe and take the long way by talking about each direction separately.

Let's start with the heuristic.

Every activity should produce an artifact

I can’t say that this consideration — and this consideration alone — pushed me toward thinking through writing. But, if memory serves me well, it helped me see the problem clearly: I was (and still am) engaged in a great deal of intellectual work, but often, there was nothing left after this work — my thoughts, decisions, and conclusions would blur over time, dissolved in the stream of consciousness.

One day, I thought something like "Why don't I sometimes see the results of my colleagues' work, despite the fact that I knew — the work was done? Why can’t I often recall the results of my own work? Why do I have to re-think the same things over and over again?". And then it snapped into place: because there were no artifacts, no "material" traces of the work done.

From a materialistic point of view, producing artifacts is a very sound and simple idea.

If the world is entirely material and governed by cause and effect, then the essence of meaningful activity within it is the creation of artifacts — whether it’s a house built, a book written, or even a scratch on a table. In other words, it is the act of producing changes in the material world that continue to exert influence through causal chains.

If, however, you thought about something but did not reflect the result of your thoughts in reality, then consider those mental efforts non-existent, lost.

Formally speaking, when you think, you’re altering the properties of neurons in your brain — a change of sorts. But it’s like building a sandcastle before the tide: the wave will come, and nothing will remain. So it is with unwritten thoughts — give it a day or two, and all that’s left are tattered fragments.

That’s why, for someone with a materialistic worldview, it should be logical to convert all meaningful thoughts into a lasting form. And historically, the most convenient and enduring form for thoughts has been text.

As I’ve already said, it’s useful to put your thoughts into any medium — be it TikTok videos, musical notation, or manga. Working with such formats may be more challenging, but it’s still worthwhile: each of us can recall something non-textual that had a deep impact. Therefore, all the reflections in the following text can safely apply to any way of presenting information.

That's why we may consider the alternative title of this essay as Thinking with artifacts.

By the way, personally, I would prefer to blog in the form of manga, but that’s beyond my capabilities :-D

Now, let's talk about the areas where the practice of creating written artifacts proves beneficial.

Verification of mental models

Every act of complex thinking involves models [ru]. You either modify existing ones, create new ones, or use them to alter or build other models. To exaggerate a bit, every thought is, in a way, a model of something.

In our minds, models are stored as (mental) images/representations/patterns/gestalts/embeddings — a fuzzy, inexact format unique to the brain. And we operate mainly with these images.

One can force oneself to think in a formal language (like logical propositions), but it’s very energy-consuming, slow, and you can’t fully switch your thinking to such rails. We live in a vague world with incomplete information about it, which naturally requires us to use fuzzy tools.

When you move your thoughts to the paper, you change the form of their representation, as well as the rules for describing the model you work with. For example, written text requires you to use a predefined vocabulary (which has accumulated the experience of all language speakers), arrange statements in a convenient order, control their consistency, etc.

In essence, you are formalizing your knowledge — translating it from a less formal form (mental images) to a more formal one (text).

In the process, you do two important things.

Firstly, you consciously go through all aspects of your mental model. To write down a thought, you have to assess every aspect of it, all implicit assumptions and undefined concepts. This helps to uncover many errors simply by checking the target mental model with other models in your brain.

That's why we can say: "what is not written down is not fully understood".

This is hard work — the required skills should be developed through practice. And this is probably the main reason why most people struggle with written communication: no practice — no skill.

Secondly, you build an alternative model to your mental one. If we have multiple models, we can compare them and identify differences and common parts. In other words, we can refine both of our models by comparing them with each other.

I already wrote in detail about model verification. You can find links to those posts in the essay Life and work with models [ru].

Refinements will start as soon as you begin writing the first sentence. And it will continue throughout the whole process of translating thoughts into text. Even when the text seems finished, it makes sense to reread it as a whole, compare it with the content of your mind, and correct the text, the thoughts, or both.

With time, the results of such formalization will become your personal external knowledge base — a record of your worldview — part of your exocortex [ru].

In my case, the knowledge base is this blog.

Knowledge base

People forget everything. It’s still unclear whether this is a bug or a feature, but that’s the way it is.

Some things are not scary to forget and even useful. Some are inconvenient but not critical — they can always be restored. And some are better not forgotten.

Forgetfulness leads to troubles:

• It slows down the decision-making process — you need to spend time recalling or searching for forgotten information in external sources.
• It damages our mental world models — we always forget important nuances. And inaccurate data fed into an inaccurate model leads to poor predictions.

My «favorite» example of forgetfulness is the persistent inability of people to remember the reasons and consequences of decisions made six months or more ago. Between us, I think this is utter nonsense, but I’ve seen it everywhere. As a result of such forgetfulness, people start running in circles, discussing the same issues over and over again.

If you have ever faced this and are looking for a way to solve the problem at its root, check out my post on the practice of writing Requests For Comments.

Forgetfulness weakens our ability to adapt and, thus, our competitiveness. And nowadays, the future of every person depends precisely on the quality and speed of decision-making.

The knowledge base that emerges as a result of the practice of thinking through writing saves your mental models-thoughts and thus increases the quality and speed of decision-making.

In case your brain is not running idle (I hope it is), ideas are constantly buzzing in your head, distracting you from current tasks and irritating you with constant reminders of unfinished or underdeveloped thoughts.

An external knowledge base unloads the brain. It no longer has to constantly replay the same thoughts. As soon as you start transferring them from your head to paper, you will feel lighter and freer in your thinking.

Cluttering with ideas was one of the reasons I took a sabbatical 2019-2021 [ru]. The effect of clearing my head through the blog was amazing.

Of course, one can’t offload all models from the mind. However, it is possible to keep only their simplified versions optimized for routine tasks. If a complex task arises, the full model can always be retrieved from the records.

So, even by its existence, the knowledge base brings a lot of value. But its efficiency can be multiplied if you make it public.

Public knowledge base

We put in more effort when our work goes public. This is not true for everyone, but most likely, it is true for you :-)

And the well-done formalization of thoughts leads to better models.

People will compare your models with theirs and provide feedback by telling you about their mental models. The more models available for verification, the more discrepancies with reality you will find, and thus, your models will become even better. And not only yours but also those of your readers!

This is a real way to make the world a little better, one small step at a time.

As a bonus, the public knowledge base simplifies discussions. It’s faster and easier to share a link to ready-made reasoning than to write them out every time from scratch. Your arguments will be fuller, clearer, more convincing, and, importantly, more stable in comparison with extracting knowledge from memory.

"Your personal brand," if I may, also benefits quite well from such activities.

And personally, I'm waiting for the moment when my children will read these posts. I hope it will be amusing.

Training communication skills

It's not the most obvious aspect of thinking through writing, but it's still important.

By translating thoughts into text, you practice the skill of expressing ideas to other people. It is true even if you write for yourself (i.e., for your future self).

In our days, it is nearly impossible to achieve anything significant without the ability to communicate your ideas.

Complications

Of course, there are some.

Writing is hard. We all know that from school. Spelling, punctuation, grammar, and the rest don't come naturally. Fortunately, we live in the 21st century, and automated editors are at our fingertips.

It is hard to formalize thoughts, especially at the start. The key is not to give up. Maybe read a couple of books on the topic. I can't recommend anything specific, but I know there are good ones.

Last but not least, writing takes a lot of time. With practice, the process speeds up, but it never becomes instantaneous. An education in journalism should help with this, but I don't have one.

At the beginning of following this practice, I could spend a week on large posts (okay, they were really large). Now, writing an essay can take from 4 hours to a couple of days, depending on complexity and inspiration. There can be breaks of a week or more. So I try to write about what itches the most — then it goes faster. For example, the original version of this post in Russian took about 6-8 hours, split over 2 days. The translation to English took about 4-6 hours, but mostly because of my level of English :-)

I don't consider time spent a problem. Rather, it’s an inevitable evil. Remember: every activity should leave an artifact. It’s better to spend time writing down a few thoughts than to ponder many if nothing remains of them in a month.

Besides, one can cheat a little.

When you work on a new topic, you can think and write at the same time. This way, you not only save time but also close the fast feedback loop of validating your written and mental models.

I would even say that this practice is a "textbook example" of thinking through writing.