The term is usually used by scholars of the Germanic languages: it is particularly important in the history of the Germanic languages because inflectional suffixes with an /i/ or /j/ led to many vowel alternations that are still important in the morphology of the languages.
I-mutation took place separately in the various Germanic languages from around 450 or 500 CE in the North Sea area and affected all the early languages,[1] except for Gothic.[2] It seems to have taken effect earliest and most completely in Old English and Old Norse. It took place later in Old High German; by 900, its effects are consistently visible only in the spelling of Germanic */a/.
In Middle Korean, I-backward-sequenced vowels (ㅐ, ㅔ, ㅒ, ㅖ, ㅚ, ㅟ, ㅢ) were diphthongs, i.e. ㅐ[ɐj], ㅔ[əj], ㅒ[jɐj], ㅖ[jəj], ㅚ[oj], ㅟ[uj], ㅢ[ɯj~ɰi]. However, in early modern Korean, they are monophthongized by umlaut, i.e. ㅐ[ɛ], ㅔ[e], ㅒ[jɛ], ㅖ[je], ㅚ[ø], ㅟ[y] with only one exception: ㅢ.[8] However, in late modern Korean, ㅟ is diphthongized to [ɥi].[9] Also, ㅚ is unstable and standard Korean allows to pronounce both [ø] and [we].[10][11]
In modern Korean language, there are two types of I-mutation, or I-assimilation: I-forward-assimilation (ㅣ 순행 동화) and I-backward-assimilation (ㅣ 역행 동화). Assimilation occurs when ㅣ is in front of (forward) or behind (backward) the syllable. In standard Korean, only a few words are allowed to assimilate, however, exceptions are often observed in some dialects and casual usage.[12] I-forward-assimilation adds [j] sound, but I-backward-assimilation causes vowel to umlaut.
^Starke, Frank (1990). Untersuchung zur Stammbildung des keilschrift-luwischen Nomens. Studien zu den Boǧazköy-Texten. Vol. 31. Wiesbaden: Harrassowitz. ISBN3-447-02879-3.