Destruction & Creation
or why should you keep an open mind.
I was invited to give a speech to a group of people about a text of my choosing. I picked “Destruction and Creation” by John Boyd. This is the adapted version of that same speech.
For today's conversation, I would like to talk about a text by a man named John Boyd.
Boyd was a fire pilot and wrote a wide variety of books and essays on diverse topics usually read by those who are interested in both military and business strategy. There's a particular piece I want to bring to your attention, an essay called Destruction and Creation.
Boyd begins this paper by pointing out that the world, and you may have noticed this, is a very complex place. In order to deal with such ever-changing and complex reality, we create mental models, interpretations of reality, that facilitate decision-making.
One must wonder why we do this. Why do we come up with these interpretations of the world? Well, the answer is, in a Darwinian sense, survival. However, Boyd has a specific definition of survival, which is paramount to understanding his ideas in the text.
Surviving on our own terms means “increasing the capacity for independent action”. We do what we want to do, as much as we want to do it.
We are, however, as history shows us, willing to constrain that individual action in order to gather the skills and resources (present in other people, with the desire for independent action as well) that allow for the surpassing of a bigger obstacle that may be in the way of our own independent action.
This is the key to cooperation, the belief in a higher capacity for independent action in the future, by willingly restraining your current capacity.
So, with this aim in mind, we start taking action, doing stuff, and our internal model shifts in order to encapsulate that new view of reality into an upgraded mental model.
The process we go about in order to create our mental models is the result of 2 different, smaller processes: starting from a whole and breaking it down into the different components (also called Deduction) or doing the opposite, starting with individual components and assembling a concept with them (also called Induction).
If we take a previously existing concept in our minds and separate it, we’re undergoing a process of Destructive Deduction. We break the layer of meaning that once connected all of those individual pieces, being left with a series of “particulars”, floating around in a “sea of Anarchy”, inviting self-doubt and confusion. However, if we link those same "particulars” in a new way, creating a new concept that fully replaces the previous one, we’re indulging in what Boyd calls Constructive Induction.
Imagine a Lego set with a given number of pieces, forming a car. If you pick those pieces apart and then assemble them into a flower, you have completed this process of destructive deduction and constructive induction. However, it is important to notice that, this was only possible by destroying the original concept (the car) without changing the individual components (the pieces). You cannot have the pieces assembling both the car and the flower at the same time. In order for something to be created, something else has to be destroyed.
This of course applies to much more than Lego blocks. It’s the process we use to create mental models that allow for our survival, as we’ve seen.
Yet, even though it provides extreme value and should be something that happens on a steady basis, the truth of the matter is that once we’re “happy” with our new concept, we stop looking for alternative models outside of us, turning our direction inward, focusing on improving the model we already have.
We start to add more pieces to our Lego model, instead of picking it apart and building something different.
This looks like a great idea, right? Doing the “inner work”?
Truthfully, it blinds us from the process of full destruction of an idea and makes us slaves of our current view. Tweaks in a certain mental model don’t allow for the full power that comes with Destructive Deduction and Constructive Induction.
And so, even though we turn within, we do it while carrying a suspicion that we may be proven wrong, that our model may not match reality. So, instead of being open to that same reality, free from the internal expectations of “being right”, we test our own model against it, hoping to arrive at a perfectly crafted model, that comes from us and not from reality.
Whenever there’s a mismatch, we blame a small incongruency in our model instead of adopting a new one. This is a mistake and Boyd turns to Physics and Mathematics to explain to us why.
In 1931, a young mathematician called Kurt Gödel shocked the whole field by proving it was impossible to embrace mathematics within a single logic system.
In order for us to understand how he did this and why this is relevant to the topic we’re discussing today, there are a few notions we need to understand.
The first thing we need to grasp is what a “logic system” means. Whenever you have a set of assumptions about the world that does not lead to a logical contradiction, you’re in the presence of what we call a “system of axioms”.
If you want to fully explore that system (to, say, prove that it does not lead to a logical contradiction) you may list all of its theorems (the list of conclusions you can have based on the system’s components).
We have these systems for almost all areas of life, from the way we relate with others to how we deal with numbers and Mathematics.
Güdel pointed out that no consistent system of axioms whose theorems can be listed is capable of proving all truths about the arithmetic of natural numbers. Not only that but also that even if a system is consistent, the consistency itself cannot be proved by it.
This applies to our own worldview as well.
No matter the system of convictions you pick, or the mental model you build, it will always lead to ideas that cannot be proven by the system itself. You’ll always need to uncover a system beyond the current one to do that.
“Over and over this cycle must be repeated to determine the consistency of more and more elaborate systems”, Boyd says, pointing our attention toward the infinite loop of destruction and creation. Instead of being content with refining whatever system you already have, dive into all of the things wrong with it.
Destroy, Create, and repeat.
He even tries to explain to us what’s the problem with focusing merely on our current model, based on Heisenberg’s discoveries.
In 1927, he noted that we could not determine both the position and the speed of a particle. The closer we get to its position, the less we know about its speed and vice-versa. As mass gets smaller, uncertainty gets bigger. As we narrow our focus to our mental model, uncertainty in the outside world gets bigger.
“Both the observer and what’s being observed influence each other” is a principle in the world of atoms and in the world of mental models. Keeping an old model will influence reality and vice-versa.
The problem is that once our model becomes an isolated system (with no interaction with reality) it becomes subdued to the 2nd Law of Thermodynamics.
In case you’ve skipped chemistry lessons in high school, let me remind you that this law claims that the state of entropy of the entire universe, as an isolated system, will always increase over time. Entropy is “the unavailability of a system’s thermal energy for conversation with mechanical work”. In layman’s terms: the degree of disorder and confusion in a given system.
That’s why Boyd points out that “entropy must increase in any system that cannot communicate in an ordered fashion with other systems or environments external to itself”. If our mental model (our system) closes itself and abstains from communicating with an external environment (reality), entropy and confusion will both arise.
So, what's Boyd's point?
By mixing Gödel, Heisenberg and the Law of Thermodynamics we understand that any continuous, exclusively inward-oriented effort to improve that match between our fixed mental model with observed reality will only increase the degree of mismatch.
In order to diminish this, we need to create a higher, broader concept for representing reality. We need to be focused on reality as well, not just our own perception.
It’s the acceptance and embrace of an eternal loop of destruction and construction of our model, based on our contact with reality, that will guide us into a better worldview.
Destructive deduction will unfold, shifting towards creative induction to stop chaos and satisfy order, until the process begins again.
It’s our purpose to develop such mental flexibility. To develop a genuine desire to indulge in such a loop, opening our minds to what reality offers.
That’s my desire for all of you.
Thank you.
Love it! Reminds me of our chat in which we also talked about legos. Great how you tied in physics with philosophy here.
A thought i had: for some reason it also feels the creativity point is relevant to the current LLM ai models, as in, that there can grow a bigger and bigger mismatch—or less creative output—the longer an ai model reiterates over itself without new/external input.