I agree with that. But the conclusion is very interesting. If something cannot be expessed within a language, it also cannot be described used any formal system, or mathematics in general. Which should be very distrurbing to modern naturalists, to whom (to the first approximation) everything is just clouds of particles (whose behavior can be described using mathematics).
The problem is that we are convinced that any experience or object can be defined within language.
The truth is that somethings are undefinable. And as soon as you attempt to define it, label it, you have already lost it.