The volitional verb suffix was originally む(mu, /mu/). This regularly shifted to a nasalized /ũ/, which then lost its nasalization and was reinterpreted as regular う(u, /u/) by roughly the Kamakura period (1185–1333).[1][2][3]
This suffix grammatically attached to the 未然形(mizenkei, “irrealis or incomplete form”) of the verb. Regular sound changes then gave rise to the modern よう(yō, /joː/) ending, initially via fusion of the vowel sounds, and then by a reanalysis of the resulting fused -yō as the suffix to be added to the mizenkei stem. This reformulation as verb stem + suffix -yō started in the late Muromachi period (1336–1573) and continued through the Edo period in 1603–1868).[1][2][3]
上一段活用(kami ichidan, “upper monograde”) verbs with mizenkei stems ending in -i:
The irregular verb する(suru) underwent further changes. The classical mizenkei stem ended in -e, and せ(se) in classical Japanese was pronounced more as she. Around the same time that the fused shō was separating again into a verb stem + suffix, the mizenkei stem shifted to end in i:
The irregular verb 来る(kuru) has a mizenkei stem ending in -o. This stem includes no front-vowel sound like /e/ or /i/ that might produce the palatal glide in /joː/, and here, the final -yō arose as a parallelconstruction with the above verb paradigms:
Meanwhile, for regular classical 四段活用(yodan katsuyō, “quadrigrade conjugation”) verbs, the mizenkei stem ends in -a. This ultimately fused with the -u suffix form to produce -ō, resulting in the modern regular 五段活用(godan katsuyō, “quintigrade conjugation”).
The volitional form of type 1 (godan) verbs are formed with う(-u) instead of よう(-yō), with further sound changes.
In traditional Japanese grammar, this is a 助動詞(jodōshi, “auxiliary verb”), with the sole form よう(yō) as the 終止形(shūshikei, “terminal form”) and the 連体形(rentaikei, “adnominal form”). Morphologically, this is an uninflecting inflectionalsuffix attaching to the stem of vowel-stem verbs.