This study shows that the function of kakarimusubi (focus concord constructions) with zo changed during the first half of the Late Middle Japanese era.

It also identifies the following three changes and characteristics based on a comparison between Early Middle Japanese and Late Middle Japanese:

(i) Zo is less likely to attach to the case elements of time, place, and subject.

(ii) In both periods, zo tended to occur in the preverbal position.

(iii) In Early Middle Japanese, kakarimusubi with zo tends to take a different word order from the basic one. Meanwhile, in Late Middle Japanese, kakarimusubi with zo shows little such tendency.

These changes and characteristics resulted from the modification of the focalization function of zo. In Late Middle Japanese, kakarimusubi with zo was mainly used in two ways:

(i) Zo marks elements that arise in preverbal position in basic word order, expressing focus without recalling an alternative set.

(ii) Zo marks any element in a sentence expressing the focus-recalling alternative set.

This paper systematically classifies the combinations of root sounds in two-syllable mimetics (C₁V₁C₂V₂, V₁C₂V₂, C₁V₁V₂) and their derivatives in the Akita dialect by examining the correspondence between C₁ and C₂, V₁ and V₂, C₁ and V₁, and C₂ and V₂. The analysis aims to identify the distinctive features of each of these sound combinations. In conclusion, the following characteristics can be said to play a role in differentiating dialectal mimetics from those of Standard Japanese:

1) The incorporation of sounds unique to the Akita dialect, such as [tso], [çe̝], [ɸe̝], [ʃe̝], and [gwa].

2) The use of phonetic features that are unique to the Akita dialect, such as the centralization/merging of /i/ and /u/ and the merging of /i/ and /e/ after certain consonants, which are also present in common vocabulary.

3) The frequent use of voiced and sonorant consonants.

4) A tendency toward using the vowel /a/.

5) Among the sounds unique to the Akita dialect, there exist sounds that are infrequently used as well as those that, although present in the dialect, are not utilized in the roots of mimetics.

These findings underscore the rich phonetic diversity inherent in the mimetic expressions of the Akita dialect.

By focusing on the CVCV structure within these sound combinations, this study proposes a novel framework for extracting and analyzing the unique features of mimetic roots in the Akita dialect. This framework not only advances our understanding of mimetic expressions of the Akita dialect but also highlights the broader potential for analyzing phonetic elements in other dialects.