If we consider speech with only one listener and look firstly at the micro level, we find it is fear of wrong word choice, wrong gesture, expression, demeanor, or tone of voice that limits our speech because a misstep with any one of these may transgress interpersonal limits.
At the meso level, it is either fear of offending or embarrassing (our understanding of) the “personality” of our listener or the fear of an actual flareup from our listener.
At the macro level, it is the fear of introducing a largish idea with sociological or career implications that might disturb, embarrass, or anger our one listener.
With more than one listener, the analysis is much the same though the numbers of people make it more complex, until we get to so many people we are speaking to an audience. Then it becomes simpler in some ways because the micro and meso levels will be less prominent due to distance between speaker and audience and there being no clear single target of our tone of voice or phraseology.
On the other hand, an audience’s response can be more complex and problematic because more than one person can become angry at us.
Human beings thus are stuck in a game that is controlled by how most of us listen most of the time.
Stated differently, human beings have magnificent speech and communicative capabilities, but rarely get to use them to their full, best effect because one or more of the many speech limits outlined above will cause us either to hold our tongues or else risk creating a disruption in the mind(s) of our listeners.
This seems like a Big Problem to me. I do not want to spend my life constrained by those rules. FIML can help us overcome this problem but even FIML cannot do it all.
We must also recognize that our very comprehension of meaning itself is grounded in fear.