Unpredictable Patterns #98: Illusions, concepts and thinking
Drawing, moving around in concepts, spaces, memory palaces and illusions as illustrations
Dear reader,
This week’s note is about images and spaces and how all of our thinking starts and ends in three dimensions. When we forget that vision is our first sense, and a key sense for survival, and that it has massive impact on the evolution of thinking, we also become vulnerable to illusions of different kinds. The best way to avoid this may actually be very simple: to draw.
Thinking in images
As has been established over and over again, we think in images. This is not strange, given that vision is so important for us and such a central component of survival. Evolution is also a sparse designer, in the sense that it tries to do more with what little it has - so using vision as the basis of organising concepts is quite natural.
This has a number of different consequences, that we often ignore - because we tend to think that abstract thought is best expressed in words. Drawing a problem seems childish, but writing a 2000 word essay on it is somehow much more insightful (and yes, I do realise the irony of writing that essay about exactly why words and images are valued wrong).
In reality, drawing a problem can be a very efficient way of understanding it more in-depth. One key technique is sketching - creating a draft image of the problem you are working on. A good sketch is unfinished, open and invites new angles, new ideas. Drawing two different sketches of the same problem forces you to apply different mental models, and creates an interesting dynamic where the problem is captured in the sketch, to the best of our ability.
And it is not important to be able to draw. The thing that holds us back most is probably the belief that we need to sketch like Leonardo da Vinci (if you can - kudos!) - when in reality you only need to be able to doodle. Developing your sketching capacity is interesting also in that you can create your own sketching language with specific uses of basic forms like circles, squares, arrows etc — how do you want to use them? What do they mean to you?
A personal confession here is that I am very bad at drawing - I am a linguistic thinker (bet you did not see that one coming, huh? The 3000 word essays did not give it away?), but that only makes it much more important for me to sketch, since there is an interesting phenomenon at play here: images capture us, they - to quote Wittgenstein - hold us hostages. We get stuck in an unarticulated picture, and unless we draw it and draw alternatives to it we will not be able to escape the maze of the image.
Images are mazes, rather than labyrinths. The distinction is a recent one, but a useful one: mazes have no one path out, you can get caught in a maze for ever. A labyrinth, on the contrary, has a single path to the center if you can find it. Many conceptual images have no center, no single path to the kernel of what the concept really means. You can get lost in them forever and lose track of what you were trying to do in the first place.
So, we draw and we end up with a clearer set of ideas about what the landscape of the concept looks like. That is the better way to think our way through problems - since the way we think about problems ultimately is - at some level - spatial.
An image toolbox
When thinking with images, it is helpful to have a series of different stock images to think in - like a basic vocabulary in a language. This allows us to start somewhere when we sketch, and yet ask interesting questions. There are numerous well-known visual thinking systems, like mindmaps, and these are great to explore and use - but the image elements are even simpler. Some examples might be helpful.
The cartesian plane and the simple two-dimensional chart
Te cartesian plane is a mental model that is well-known to everyone who has had the benefit of some maths. It is a coordinate system that specifies points in a plane by two different points, and allows for a lot of cool mathematical stuff. It is also the grandfather of the simple two dimensional functional chart is one of the simplest ways of thinking about two different variables and how they relate to each-other.
The two dimensional diagram is a powerful image, and holds hosts of economists hostage - and has for a long time - but it is easy to see that it may not be the best of images for understanding more complex problems. Even so - it is a great way to start. It allows you to try to reduce a problem to the two key dimensions you are thinking about, and then to really explore the different possibilities in that two-dimensional space.
Let’s say that we want to understand how technology develops over time, and that we want to remain ambiguous - so we just speak of the power of technology in general and want to understand how it changes. We can then sketch out a two-dimensional graph and start playing around with ideas. We might end up with something like the below, to chart different hypothesis.
In this one image, then, we have four different theories of the development of technological power, and we can certainly come up with more. The red curve suggests an exponential take-off, something like the singularity. The green curve suggests that technology grows fast in power and then levels off. The black dotted line is a linear increase and the blue line is a punctuated equilibrium with plateaus. We can easily come up with more possible charters, and so also discover new possible contours of a theory. Adding, say, an inverted U-curve gives us a theory that says that technology rises quickly in power, but technological power then declines. Just drawing that curve forces us to think through why that could happen - and to fit the image to a narrative. Maybe technology grows in complexity, becomes brittle and so ultimately power decreases and disappears?
The image invites a new narrative and a new mental model, and helps us think through the problem in new ways.
The trap here, of course, is that most problems are not two-dimensional at all, they are n-dimensional and the reduction to two dimensions may mean that we miss something fundamental about the problem overall.
Other helpful models in the same genealogical image tree include the trusted consultant two-by-two, orienting different items in a field defined by supposed opposites. Often drawing a simple two-by-two challenges the way you think about different concepts too. Let’s play around with a simple model for political stances and rhetoric that looks at orientation in time and image of the state:
This simple model can be used to classify different political parties on the basis of if they talk about society as a system that should work for everyone or a project we all participate in, and if the majority of their rhetoric is oriented towards the future or an image of the past. I will not fill it out now, but you can do that for yourself as an exercise with your favorite politicians.
The two-by-two needs to start from two fairly independent variables to be interesting, but if it does, it can often reveal interesting things about how a field of actors or concepts is clustered.
Trees
Trees are also essential images to work with. There are many different examples of good trees that can help us think about different problems. The perhaps most basic is the decision tree, and these come in many shapes and forms. What they do, essentially, is to break down a big decision into smaller ones that can be addressed in sequence, and then you can run through the tree to see where you end up. Figure out an opponents decision tree in a negotiation or a game is essential to winning the game.
Other kind of trees are capability trees - that break down a capability you want to acquire into its component parts. An example could be a capability tree for a public policy function in a company or one of the skill trees in popular strategy games.
Trees have been the subject of an earlier note, so we will not dig more into them, but just not that there are casual trees as well — breaking down an event into the root causes that brought it about and of course evolutionary trees. Networks are other good examples of a tree-like form that can highlight social connections or subject connections in different situations.
These are just two examples - there are many other basic image classes that you play around with. Building your on little library or tool box of images you play with is a great way to kickstart analysis and thinking about almost anything.
Arguments and illusions
If our thinking is visual, it is probably also the case that we can learn something from visual illusions. Specifically, we can learn something about how conceptual confusion may have its roots in visual illusions. There are plenty of examples of visual illusions, and they often amaze us - but we do not immediately apply the insights that they can give us also to our thinking.
The strong argument we could make here is that all visual illusions represent thinking traps of different kinds. I am not sure that argument holds up, but it is instructive to look at a couple of different visual illusions to explore it further.
We will start with our old friend the duck-rabbit.
The duck-rabbit is an example of a picture that can be seen in two different ways - and both are equally correct. And there are concepts just like this as well. Light has both wave-nature and particle-nature, it makes no sense to argue that it is one or the other, just like it makes no sense to argue that the duck-rabbit is a rabbit or a duck. Yet, this is likely one of the most common thinking traps we fall into. We believe that something has to be one or the other, that there is no middle. We even have that idea framed as a principle of logic: tertium non datur. There is no third: it is a duck or a rabbit, it cannot be anything else.
This is, as the illusions shows us, clearly wrong. Our concepts are visual, not logical, and so applying binary values will lead us to end up with useless conclusions. When we leave out the way our concepts evolved over time and try to put them into propositional logic we end up making up arbitrary limits for thinking, and that gets us stuck. Even worse: we can start to disagree about the “true” nature of the duck-rabbit. As cartoonist Paul Noth has it:
These conflicts all originate from a lack of understanding that our concepts have a visual evolutionary origin.
Other visual illusions are more pernicious. The lines on the wall here are actually parallel but we see them as angled:
Imagine trying to figure out what is and what is not equal overall - we obviously get stuck arguing about this because our concepts of equality are related to our concepts of the parallel.
At heart our concepts are spatial, visual and so we get stuck in ways that can be described by looking at how we get stuck in images.
Or we can see things that do not exist:
Just like we can find patterns and causes in noise and build conspiracy theories of different kinds on the loosest possible associations.
We can detect motion where there is none:
Just as we can detect change where there has been no real change in a situation, or believe that things are moving when they are not in analysing a situation.
In other examples we can miss the message or pattern in a situation because we are focusing too hard on the wrong things. The hidden message here could be an insight in an financial analysis or a legal memo:
The way we pay attention really matters, because we pay attention in a way that is informed by our visual skills and those skills underpin so much of our intellectual understanding of the world.
We are horrible at understanding relative sizes as well, and are likely to see the same thing as larger or smaller depending on the context:
The application of this in thinking is self-evident. We relate size to what we compare to, not to any sense of absolute size and so often make mistakes.
These are just a few examples, and the point is not to suggest that they are exhaustive - quite the opposite - rather that it is always interesting to ask of any particular intellectual problem how we can think about it visually, or spatially.
Images and memory
A last example of how images and spaces are connected to our thinking is memory. As documented thoroughly by Frances Yates in her fantastic book The Art of Memory, memory experts often rely on and design so-called memory palaces - three dimensional rooms that they can move through as they remember something. Here is Joshua Foer - who wrote a book about memory competitions - explaining what happens in the brains of memory experts:
“Surprisingly, when the mental athletes were learning new information, they were engaging regions of the brain known to be involved in two specific tasks: visual memory and spatial navigation.”
Our memories are not flat lists, they are rooms, spaces, that we navigate through in different ways, and they are distorted, changed and we can get lost in them if we do not pay attention to their visual and spatial nature.
We are stuck in a very peculiar trap here - the abstraction trap - where our ability to abstract out key features of a concept has led us to believe that concepts are defined, almost like geometry. The abstraction trap lies at the heart of philosophy and science, and suggests that there is something pure about the non-visual and flat, the simple function or formula - but the opposite is true. There is something dead about concepts that are not illustrated, concepts you cannot move through and visualise in different ways. Thought is three-dimensional or at least visual, and if we forget that we run the risk of losing ourselves in false simplicity dressed up as pure thought.
So what?
The take away from this meditation is not that difficult to grasp, but it is difficult to put into practice. Draw more. Ask about the spatial configuration of a concept or a problem — admit that many problems have, as Wittgenstein noted, the form “I cannot find my way about”. If you do, there is a chance that you can break free from the implicit images and spaces that have kept you hostage.
But it takes a lot to get there. You need to conquer your fear of seeming childish when you draw a problem, and you need to rediscover the thrill of putting pen to paper to try to sketch a situation out. And you probably already know that this is the right thing to do — it is the sublimated insight that we think visually and spatially that has led us down the path to Powerpoint, a poor tool for any visual thinking, but the only one that we feel has enough respectability to allow us to think visually.
Imagine how much more interesting a presentation would be if the presenter had drawn a number of images rather than flash ready-made Powerpoints without nuance or direct engagement in them. And if you could respond in drawings as well!
Slides often end up being a worse alternative that a memo or a note, because we just put words on the slides, and treat the words as if they had the explanatory power of an image. The result becomes a vague and often hard-to-understand mixture of writing and drawing that collapses into the illusion of consensus.
Now, slides can be done well — but it is rare, and very, very hard. But how much better it would be if we just dared draw a little. And that is a lesson I will need to practice myself.
Thanks for reading,
Nicklas