Representations

“I” never leaves language: pointing at an object and naming it doesn’t name the object, but it catalogs the representation of that object and stores it as “name” — or a label if you will.

Considering that language itself is computational, the question becomes: when to apply a certain computation? Let’s say you are a researcher and you have made an observation. That observation gets stored as a representation with a label in your brain. Now, you can talk and/or write about the representation of that observation when retrieving its stored label. That’s the why: the stories that are told after the computation. So when do you choose to tell them?

Let’s pause for a second and reflect on that:

I never leaves language: pointing at an object and naming it doesn’t name the object, but it catalogs the representation of that object and stores it as “name” — or a label if you will.

That quote is simply a representation. Not of an object or an experience that we can point at with our fingers, but a representation of another representation; an abstraction — or an emerging representation as a result of multiple representations dynamically evolving in my brain and statically stored as labels or “names” for easy retrieval. For instance, to communicate or to linguistically process a thought.

All you can do with the emerging representation quoted above (this entire piece, actually) is to store it by labeling it — can be a single word like “abstraction”, “representation”, “philosophy”, “nonsense” or something more descriptive. My guess is that a human tends to apply the least computationally intensive method for storing representations. That’s why there are eleven instances of the pronoun we in this text: two combined letters compressing and thus erasing billions of humans into an emerging representation so I can make a point in a few words.

That said, “least computationally intensive” looks different for each one of us because not every human has the same processing power and computes the same names or labels in exactly the same way. Similar, perhaps, but not exactly the same. Anyway, after storing and labeling, you move to the story you tell. A representation of the emerging representation quoted above. And so on.

To most people, what I am describing here does not matter in the slightest. Because they don’t see this insight as a tool that can actually shape their reality, but as mere words that don’t pay the bills, feed or clothe them. Food and clothing can be pointed at: the labels “food” and “clothing” store their representation. “Money”, on the other hand, is a label for an emerging representation which we then map onto objects made from metals and paper that can be pointed at with our fingers.

Here’s the difference in a nutshell:

• Our senses experience stuff outside of our own bodies and store representations of those experiences as names or labels.
• In our brain, storing representations of those experiences can lead to emerging representations: representations that have formed through computations with different representations. “Money” is the result such a computation.

If a label like “money” simply stores an emerging representation — the result of a computation that does not come from pointing at something outside of the human body — how can it then travel between brains?

But here is what’s fascinating: it does travel between brains because otherwise I wouldn’t have been able to use it as an example. My encounter was this: somebody pointed at pieces of metal and paper and said “money”. Then, they produced a new representation where “money”, “food”, and “clothing” are intertwined. That is an emerging representation that combines both:

1. a representation of stuff that can be pointed at (primary)
2. an emerging representation (secondary)

This new, more complex emerging representation (tertiary) gets catalogued again, perhaps labeled as “reality” or some other kind of name that fits the computation at a given time.

Basically, that’s how we move from simplicity (a representation of something that can be pointed at) to complexity (a representation of something that can be pointed at interacting with another representation leading to emerging representations; emerging representations then interacting with other emerging ones or with representations that can be pointed at, leading to more emerging representations etc.).

And if that isn’t enough already, an emerging representation is a little different than a standard representation of things that can be pointed at. For starters, it can shapeshift. Food is something that ends up in our belly (ideal) or on the floor (shit happens). Clothes are something we wrap our bodies in. But money? Can be metal and paper; can be plastic; or it can even be a string of ones and zeros — leading us to the question of labels.

Representations of stuff that can be pointed at are in harmony with the static labels we use to store their representation in. Even when we switch the labels. We don’t have to call food “food”, but the label we use is still storing a representation of the stuff that ends up in our belly — or on the floor. By contrast, emerging representations are always the result of a computation of computations — they are thus dynamic and fundamentally incompatible with the static labels we use to store them in. Instead, we have to compute them in order for them to emerge. In short, they have to be modeled.

So you might want to ask: who gets to model them? And that’s a miscalibrated question because you are modeling them whether you want or not. And accepting that fact or not is what decides on how complex, emerging representations with labels like “reality” look to you. Your ability to compute and store representations is either going to trap you inside statically labeled dynamic representations — or enable you to model your own.