I think what you're witnessing here is not the "truth" but John La Valle trying to bypass a limit by redefining or reframing a problem.
That is, what he does is teach language patterns and language interventions, so having a person consider their problems in terms of what he teaches is of benefit to him as a teacher (source of knowledge, source of solution) and to the client hearing his words as the client (if he buys it) will feel it is possible to bypass his or her limiting beliefs.
It's a reframe, a shell game, a trick to get someone to think a certain way and do something. It isn't "the truth."
Beliefs are more than simply language. Can a non-verbal person have a belief? You bet. A belief is simply a fixed representation of reality that makes us expect a certain outcome in terms of cause and effect.
It can involve distortions created by way of language, but I do not believe that it has to. Language can certainly affect beliefs, but I think that relying solely on language is an unnecessary limitation.
My distortion of John Grinder's prescription for belief change is as follows:
1) Elicit from the client the counter example to a limiting belief that had the client experienced it, it would completely negate and transform the limiting belief.
2) Arrange an experience for the client that provides the counter example in such a way as you are not perceived as being involved.
3) Get out of the way.
Beliefs are "frozen experiences." I do not know if the only way they are frozen is by linguisitc transformations. However, they do exist because of our ability to create a model of reality that is not reality and respond to it as if it were reality. Is that model primarily or totally linguistic in nature? No. Because we build our models of reality before we learn language, however, we do have our own natural coding system.
