Except, rereading, it seems to make more sense to think of metalanguage in computing terms. We can think of a statement as a series of instructions for running a program. Rather than a noun, the metalanguage would be an action: a reasoning process by which we somehow evaluate statements. Except that that still doesn't explain what we do when we run them. Or how we somehow seem able to overcome the limitations of computers.
1) I really don't understand the "let's not shoot the messenger comment." I'm guessing it's non-essential, though I don't know who I shot lol.
2) I agree that we can make sensical and non-sensical statements with plain English, and that the non-sensical statements do not render English inoperable or useless.
The syntax and rules of operation for English determine what is sensical and what isn't. Statements are relayed back to the syntax and processed according thereto to determine if it is meaningful in a way consistent with it. Thus, at the syntactic level there is indeed a "reasoning" process by which statements are evaluated, but the syntax itself is structural, i.e. it imposes constraints upon what can be considered meaningful.
How would you determine correctness in the first place? Maybe the syntax -- or language rules -- that you speak of are created experimentally?
One model of reality that I'm thinking about goes like this:
There's a Turing machine and a Programmer.
The Turing machine doesn't know very much about rules or syntax. It just gets instructions from somewhere, which it runs automatically. There's no syntax-checking or filtering at that level. The machine occasionally gets stuck because of the Halting Problem, so this requires intervention from the Programmer to reset it. The Programmer might also have additional powers, such as being able to replicate itself, perhaps conjuring a
higher self into existence as a workaround if it gets stuck resetting the machine in an infinite loop. Alternatively, it creates and delegates a
lower self, but I guess that would be pretty similar.
The 'instructions' could be message data that we get from our senses in serialised form, presumably coming from another programmer entity, whom we don't have direct access to, but
only via the message tape.
Rapid multiplication of the programmer selves could then pave the way for creating complex mental structures, out of something that had absolutely minimalistic rules. Far from being a nuisance, the undecidable parts of the software are what allow both sides (message and the messenger) to exist.
4) If you run software with code that does not conform to the syntax of its programming language, it will be evaluated as an invalid input. If valid, how those statements are expressed is a product of both their relation to their governing syntax, and also in relation to other object-level statements governed by the same syntax that may affect them (e.g. if-then or "conditional" statements). I'm not sure if I fully responded to what you were saying, here. I'm at lunch on an iPhone.
Edit: Linking this to subjectivity and objectivity, consider a governing syntax of Reality in total as it relates to its internal components. As we perceive real content and subsequently process and model that content, we can either model that content in a way that is consistent with the syntax of Reality in total, or in a way that is inconsistent. Because the structural syntax of Reality in total necessarily distributes to all of its content, if our model is consistent, then it is objectively valid, else we have an inconsistent, invalid model that provides us with no objective value. In this way, we can consider this process in terms of a fundamental utility function, where utility is defined upon consistency and congruency with Universal syntax.
When a computer checks a piece of software for syntax errors before compiling or running it, the processor is running software the entire time. What we call syntax would therefore be some complex pattern of learned behaviour.
When you suggest that perhaps "the syntax -- or language rules -- that I speak of are created experimentally," you have to remember that, given this possibility, there must still be an unconditional and unchanging structure at play, i.e.
what defines a rule.
In his theory, Langan describes a "one-to-many" mapping of real/Universal syntax, which would allow for the simultaneous possibility of various conditional syntactic systems at the "many" level while maintaining an unchanging syntax archetype at the "one" level. The general structure of syntax or 'rule' still applies, but how this is expressed differs within the mapping.
When you talk about the Programmer creating a 'higher self,' basically you're talking about omnipotence. To create a 'higher self' would imply the creation of a self which is totally unbound by the syntax of the 'lower self,' but this is paradoxical to the fact that the 'lower self' must be unbound by the syntax of the 'higher self' in order to create it. If the Programmer can actually do this, then he was omnipotent all along, and any 'higher self' is simply one of a many diversified essence of the 'omnipotent self' [archetype].
