Jaynesian consciousness and Turing completeness

Discussion of Julian Jaynes's first hypothesis - that consciousness (as he carefully defines it) is based on language, and related topics.
Post Reply
minnespectrum
Posts: 19
Joined: Tue Jul 11, 2023 3:12 pm

Jaynesian consciousness and Turing completeness

Post by minnespectrum »

There is an important concept in computing theory called Turing completeness, which says that any system that is Turing complete can simulate any other system. If you have a computer language (or even other types of systems, like Conway’s Game of Life or Minecraft), and it has access to an infinite amount of memory (such as an infinite tape with symbols, in Turing’s original model), then you can compute literally anything. Presumably, you could even run a simulation of our universe, if (as some but not all physicists believe) the laws of physics are computable.

Of course, this is an idealization (since nothing you could practically build would have infinite data storage). Nonetheless, it’s rather remarkable how many things are Turing-complete when you wouldn’t expect them to be.

This does not mean that all programming languages are equally powerful in practice, though, only in theory. “Turing tar pits” (https://en.wikipedia.org/wiki/Turing_tarpit) are languages where you could program anything, but would have an extraordinarily difficult time doing so, since you would need a lot of code to program common tasks that would be trivial in other languages.

Obviously, computer language is not the same thing as natural human language, but there does seem to be a parallel to the idea of Turing completeness. With English, and probably most other modern languages, it does seem theoretically possible to express any thought one could possibly wish to communicate.

Yes, it’s true that there is often a lot that is lost in translation between languages, but that’s due to practical concerns. If you wanted to ensure all of the nuance present in the original (let’s say it was a French poem) made it into the English translation, you probably actually could. It’s just that the translated text would have to be several times longer in order to break down and explain the “French-isms” to an English audience, and most readers would find this inconvenient and annoying. The longer you allow your translation to be, the less would get lost. (As an extreme example, given enough time, you could just teach your English audience to read French, and then understanding this particular poem would become trivial for them)

Since the same is true for English-to-French translations, we could say the two languages are theoretically equivalent in terms of what they can express. It’s just that they are practically different because some commonly used French words may not have a simple way of being expressed in English, or vice versa. Since people don’t generally like overly complicated explanations (see: Occam’s razor) this means that English and French speakers may have some degree of difference in how they think, despite both being conscious.

It seems likely that “linguistic completeness” (or whatever you would call the analogue of Turing completeness for natural languages) is a necessary condition for Jaynesian consciousness, but not a sufficient one. Going by the Turing tarpit example above, one could imagine a language where it is possible to express concepts associated with consciousness, but very difficult since these require extremely long words or phrases; meanwhile, bicameral-style thinking in this language would be expressed with short, simple words.

A person speaking such a language could develop Jaynesian consciousness, but it would seem to be far less likely. Even “Occam’s Razor”—to a person speaking this language instead of ours—would favor the bicameral mentality. So, languages that are theoretically equal in expressive power could still influence people in different directions.
Moderator
Site Admin
Posts: 408
Joined: Sat Feb 26, 2005 1:03 pm
Contact:

Re: Jaynesian consciousness and Turing completeness

Post by Moderator »

It' s great to see you applying Jaynes's theory in all of these different areas.
Post Reply

Return to “1.0. Hypothesis One: Consciousness Based On Language”