|Abstract:||This paper reanalyzes the so-called reduplication with fixed segmentism cases in Korean and shows, inter alia, that the fixed segmentism occurring in examples such as, talkak/talkatak ''rattling'' is not of the phonological type as previously claimed but of the morphological type. The arguments for this emendation regarding the type of fixed segment are based on three findings: 1) the suffixes that occur with the fixed segment are affixational in origin; 2) the proposed TETU ranking based on Place Markedness Hierarchy does not really work, due to its inability to choose /t/ over other coronals, while its revision with addition of the Sonority Cline Constraints of Lombardi (2003) results in a ranking paradox; 3) The complementary distribution, cited also as evidence for the phonological nature of the fixed segment is not phonologically conditioned. In addition, I examine the relevant issues in the diagnostics for the reduplicative fixed segmentism given by Alderete et al. (1999) with a detailed comparison of the fixed segmentism types in Nancowry and Korean. Some suggestions for origins of the reduplicative suffixation are given, comparing their explanatory adequacy with that of the Optimality Theory.