The additive [n] has appeared in the formation of compound nouns, derived words and phrases in the contemporary Korean. In these cases, the surfaced [n] neither originally exists nor is inserted during their derivative process. I suggest that this phonological phenomenon is caused by the sonority meditation by changing [j] to [?] within the prosodic phrase. Because the initial [j] of a following second syllable is not a vowel, [j] is changed to [?] by decreasing its sonority in order to satisfy the syllable contact constraint. Another motivation of the surfaced [n] is found at the boundary of the following word with initial /i/. The phonetic motivation of the inserted [n] is explicitly not clear and [n] also does not appear in front of the nominative case ``i`` and designated case ``i-``. We notice that /i/-initial Sino-Korean did not represent [n] insertion. Therefore /j/ and /i/ undergo different phonological processes. Historically, as the domain of initial sound law was extended to the non-initial position within a prosodic phrase, this restructuring process was gradually expanded through many kinds of words and so both conservative forms and innovative forms coexist.