I-affection or Umlaut is a phonological phenomenon which is an integral part of the Sindarin language.
The phenomenon refers to the vowels of a word to be 'slid' higher and fronter to another sound, usually e, drawn because of a following i (usually near the end of the word) in a sort of vowel harmony. This started to occur after the stage called "Old Sindarin".
Umlaut has a grammatical significance in Sindarin since it is how plural of nouns is formed.
The following verbs show how the vowels of the word-stems slid into e because of the affection of i.
- gonod "count" + -ia > genedia "to reckon"
- aglar "glory" + -ia > egleria "to praise"
- said "private" + -ia > seidia- "to set aside"
The same effect is seen with other endings such as -ien or -ian(d) as seen in
Another ending which causes umlaut is -il
It also can be triggered by compounded words
- See also: Sindarin#Pluralization
The Primitive Quendian plural ending -ī was retained in Quenya and Old Sindarin. In the later stages the *-i affected the previous vowels, especially the preceding one
Then, all final vowels were apocoped in Sindarin, including the plural markers. All traced of plural were simply the affected vowels.
To summarize this, the envisioned history of word adar pl. edair would be like this:
- OS *atari "fathers" > *edeiri > *edeir > edair
In a different form it is seen in Ilkorin. For example adar pl. edrin, aman pl. emnin, boron pl. burnin, gangel pl. genglin and thorn > thurin.
The phenomenon is common in Germanic and Welsh, on which Sindarin is based on. For example the plural of W. bard (< ancient *bardos) is beird (< *bardoi) The same process is responsible for some plurals in English, such as man > men or foot > feet.
- ↑ Vinyar Tengwar, Number 42, July 2001 p.20
- ↑ J.R.R. Tolkien, "Addenda and Corrigenda to the Etymologies — Part One" (edited by Carl F. Hostetter and Patrick H. Wynne), in Vinyar Tengwar, Number 45, November 2003 p.5
- ↑ 3.0 3.1 Bill Welden, Jim Allan, On the formation of plurals in Sindarin, published in An Introduction to Elvish pp. 62-67