So no one can go beyond the boundaries of theory? This relies on the premise of music being formed by the use of a certain theory and nothing more. As I’ve already mentioned, there’s a wealth of theory, applicable to different tone systems, etc.
In a programming language, what boundaries are there? Ultimately, it depends on the language, doesn’t it? A person, no matter how gifted or imaginitive, can’t program something that is not inherently possible because of the language itself.
Music (I’m using it as a broad term, so it becomes synonymous with “sound”) never had to nurture a dependency on a definite language (that we, or atleast I know of), because they were subsequently developed around it, to control it’s use and to explain it (easily resolved by switching, intermerging, creating your own or forgetting about tone relationships and rhytmic spacing). It wasn’t born from a man-made theory. It gave birth to all of them.
The only boundaries (from man’s point of view) are sonic, and even they are susceptible to perception. So there is no definite theory, or way of grammar that let’s you “speak” music, that we/I know of. It’s like when you tame something from the wild, you only tame certain aspects of it, while being totally unbeknowing to the rest.
If you draw the lines, music is then, by itself or by part, a programming language. One that we create a perceptual understanding of, as we differ on amongst ourselves, because we’ve had no way of learning the one true syntax (if such a thing exists), and because we have no definite knowledge through perceiving. Above music, there’s another, greater part (reality or whatever), which has another dependency and so on.
I hope you get my point. Musical theories only serve as a guideline.
ROCK ON 
