A Mīmāṃsaka has a powerful device against the Vaiśeṣika claim that śabda (linguistic sound) is momentary. In fact, we do communicate and communication presupposes that we know what is said and this would not be the case unless we already knew the śabda our discussants utter. Hence, śabda must be fix.
What has, then, the burden of explaining changes in accent, lenght, tone, etc.? An easy way-out would be to distinguish between word (śabda) and sound (nāda) and to say that only the latter changes accordingly. But this duality may be too metaphysical for a down-to-earth pragmatist. Hence, Rāmānujācārya (I am not sure about how frequent this idea was among Mīmāṃsakas) suggests instead that it is only the manifestatory (phonetic) effort which varies, thus making the permanent śabda appear slightly different.
Heuer from Leeds to UCL
46 minutes ago