The word you hear is, on this account, the fourth thing to have happened, not the first. Indian philosophy of language, developed most fully by grammarians like Bhartrihari, describes speech as descending through four progressively grosser levels before it becomes an audible sound. Para is speech in its most subtle form, undifferentiated and identical with pure awareness, prior to any distinction between a thing to be said and a way of saying it. Pashyanti is the stage at which an intention to communicate first arises, seen rather than said, a kind of pre-verbal seeing of what will be spoken. Madhyama is the mental formulation, the sentence as it takes shape internally before any sound is produced. Only at vaikhari, the final stage, does the word leave the body as air is shaped by the vocal apparatus into a sound another person can hear.
The scheme is doing more than describing anatomy. It is making a claim about where meaning actually lives. If audible speech, vaikhari, were the whole of language, then translation, mishearing, and the gap between what someone means and what they manage to say would be unexplained oddities. The four-level model treats them as expected: the sound you hear is already several removes from the subtlest level at which the meaning existed, and something is inevitably lost or reshaped at each descent.
This is also why mantra practice treats the sound of a syllable as carrying force independent of whether the meaning is consciously grasped. If the deepest level of speech, para, is continuous with pure consciousness itself, then a sound correctly formed at vaikhari retains a connection to that root regardless of whether the listener’s intellect processes a translation. The chain runs the whole way down; understanding at the surface level was never the only channel the tradition claimed was operating.
The model gives Sanskrit grammar a philosophical stake that most linguistics does not carry: getting the sound of a mantra precisely right is not merely aesthetic correctness, it is fidelity to a structure that is claimed to run from ordinary speech all the way back to awareness itself. That claim is the reason, taken up next in this series, that Sanskrit pronunciation was preserved with such extraordinary precision for so long.
Part 22 of 100 — Series I: From the Beginning
Series I: From the BeginningThe Five KoshasIndexFAQThe Timeline