1. the languages are “read” into Roman. Roman without any particular rules not defined is arbitrary. That is it becomes neutral, devoid of biases from language rules. It unlocks the languages into their neutral forms.
The languages are basically a rule or lock defined to capture particular meanings from a vast arbitrary tract of phonetics and noise. eg roman: t corresponds to 4 locks in Indian Language system and two locks in Japanese. t as in tatami, t as in taberu are the soft t and heavy t in Japanese. But in Indian t as in tumhara, t as in tenis, t as in thoda and t as in beithak are 4 different locks, 2 soft and 2 heavy.
But in Roman they are all but t. Note that English is not Roman. English is simply a hidden phonetics where you know what the lock is, but you do not show it explicitly. eg time is English, but in Indian lock its thaim, with t in thaim same as the t locked in beithak.