Distributed computation, metaphor, enculturation, and consciousness

How enculturation bootstraps consciousness through metaphor, resulting in society as a distributed computer

Through group interactions, we're incentivized to behave a certain way. At a very elementary level, this includes things like status-based interactions. There are also in-group out-group dynamics. Through these dynamics, we end up with shared state. That shared state can be mediated by any number of communication mechanisms. Facial expressions, grunts, etc. These communications give us personal responses that are trapped in our heads. We can do our best to communicate those responses to others. Ultimately through this communication we end up with shared state as well as mechanisms for changing that shared state. Voila, computations.

The distinction between individual vs shared mental activity is sort of the difference between what could exist without communcation, vs what can't. The concept of enculturation plays a big role here. Apes and even dogs can understand human language when trained enough. Does this change the way they think? Many would say yes. There's a fundamental difference between the kinds of thoughts you only have in a vacuum versus the ones that you have access to after enculturation. Look at what Hellen Keller had to say about the matter... access to the enculturated world was not just cool and neat for her. It wasn't just a slight enhancement on what she already had.. From the way she described it, it appears to have been the difference between essentially a void and having a rich internal life. How? I would propose it gave her the ability to do internal symbolic manipulation. In other words, computations which allowed her to take part in the distributed system that is society.. giving her access to download a huge amount of distributed data which she could then run operations on.

I used to be extremely skeptical of the claim that there was some sort of lacking consciousness without language, but the key is that it's not just vocabulary that's missing when you don't have language.. it's all the information that language give you access to. It's being part of a huge distributed computer. Then the question is: how does it give you access to this? Well if you read Julian Jaynes' bicameral mind theory he provides a mechanism - which is essentially through metaphor. By developing a metaphor which allows us to model our own internal states, we gain the ability to manipulate those internal states. While there are things about Jaynes' theory that are certainly iffy like the timeline, etc, this basic idea that consciousness could arise from metaphor arrives very close to where cognitive linguistics has landed on the matter, who have recognized the fundamental role that metaphor plays in consciousness.

So then the questions are: how could these metaphors behave in a computational manner? how do they get communicated between individuals to provide this new computational way of processing the world, etc.? How do minds which possess no defined concepts originally end up driven into a state where there are defined concepts which can also be manipulated from the outside via stimulus. Ultimately the process of enculturation then would be layering metaphor on top of metaphor on top of metaphor. But what's the basis for the first metaphor? How does enculturation get a foothold with which to bootstrap this new form of processing?

Let’s hypothesize that as non-enculturated consciousnesses, we interact with the world, but we don’t process it via defined concepts, only qualia. Perhaps there are some genetically pre-defined concepts, but perhaps their aren’t so let’s assume the latter for now and see if we can construct a model for how metaphor and consciousness could be bootstrapped without pre-defined concepts. Some might argue that there can be no concept without interacting with the outside world first, which would make the idea of a genetically-defined “concept” impossible.

If we accept this view, then we’d have to suppose that from the outside, genetically defined responses might cause behavior which makes the organism appear to have a concept, where in reality the *genes* have a concept, which would mean the locus of consciousness is not through the individual but for the species as von Uexkull describes. But each of these computational realms is sealed off from each other. An individual cannot interact directly with genetic concepts. Rather than genes imparting concepts to individuals, they must impart behavior which encourages the formation of a concept. The genes can’t code for the conscious concept of “up/down”. So if we took these as given, we would say that genes only code for sensory neurons and other anatomy, as well as brain anatomy which, through interaction with the environment, drives the brain into a pattern of activity which *can* represent the conscious concept of “up/down”. This is the basic concept of embodiment.

So with those assumptions, we have to determine the way that the most fundamental concepts could grow from only sensory stimulus as well as “deliberate” arrangement of stimulus from the outside, which could drive various internal states in such a way as to create a representation… essentially bootstrapping the entire world of meaning and semantics through metaphor.

From here we can explore many directions. Can we model this with the Free Energy Principle model and show outside agents arranging stimulus of another agent in combination with a shared environment so as to induce such a binding? Can we use a description of the two brains as a dynamical system to model how such bindings could have originally occurred and self-propagated? Perhaps such a description could involve wave dynamics where signals reverberate among agents and the environment. Can we then describe how a full-fledged self-propagating system of arranging and replicating internal states of other agents could arise, getting bootstrapped off of such bindings? Perhaps that would arise from a category-theory description of relationships among previously mentioned reverberations, as well as a model sort of like a pi-calculus which could be derived from such a description.