[dawg-]

>If you went to a physical therapist and he told you to move your tongue/lips/lungs in certain specific motions, which happened to produce noises, you wouldn't judge the exercise by the semantic content of those noises.

So the brain actually does have a special motor planning area for speech movements that is intermeshed with all the other parts of our language system, including semantics. As an example, when people have Apraxia (loss of coordination) they can have two different kinds, depending on where the lesion is: oral apraxia (loss of ability to move the jaw/tongue/cheeks/lips in NON speech tasks) or verbal apraxia (loss of coordination specifically for articulation movements). You are on to something interesting with your comment about "low-level tongue movements" vs. movements associated with communication.

I wonder what an fMRI would look like on someone reciting these Vedic hymns (in a language they do not fully possess) vs. someone reciting hymns in their native language? And what if you had an English-speaking Christian recite hymns in another language they don't know, like German or Latin or Greek?