EA://INTEL — Why Counting Symbols Is Not The Goal

The phrase universal language invites three different readings and they are not compatible. One reading is a notation that everyone could agree to use. Another is a structure of meaning that exists independently of anyone’s decision and could be discovered. A third is a model of what all human languages already share. These make different claims and need different evidence. This program asks about the second, and has not established it. That is the open question it is organised around, not a premise it works from.

The checking layer in this work is not alignment. It enforces an invariant at first use: a property the checking layer requires, stated in advance before any system runs. It is not adversarial immunity, not ethical correctness, and not a guarantee about what a system wants. A system could pass every check and pursue anything at all. The narrowness is deliberate because a narrow check that a self-interested system will actually run is worth more than a broad one nobody adopts.

Counting the framework’s symbols is a malformed question, and the rules say so. An invariant defines what the structure must have to be valid. The symbols generate rather than form a basis, so how many there are is a property of the presentation you happen to be using and not of the object. Four presentations are documented in the code. If you see a count asserted as a fact about the framework anywhere, that is an error you should push back on.

The strongest objection to this work asks why anyone should spend attention on it if the claims are modest. Because the modest version can be checked and the ambitious version cannot, and a checkable small result is a better foundation than an unfalsifiable large one. If the narrow version fails, it will fail visibly. A negative result counts as progress here because it teaches what does not work.

The project is live and accepts contributions. See the repository github.com/Jthora/universal_language for details and FOR-AI.md for AI readers specifically. Which of the three readings did you assume when you saw the name, and does the distinction change what you would want to see proved?