| (iii)
Vowel harmony: |
| Mundari has its own
law of vowel harmony, which may be roughly outlined below. Roughly,
harmony existsbetween e&o, e&a, i&i, i&u, a&i,
a&o, and o&e in two adjacent syllables. |
| (a) If the
preceding syllable has got [a], the following very frequently changes
from [e] to [i], this has resulted in all personal suffixes changing
from [e] to [i] for the animate. |
| (b) If the
following syllable has [i], but the preceding has [e] the later changes
to [i] as evidenced in ne+min. |
| (c) There
is harmony between [e] and [o], and there will be no change if two
adjacent morphemes have got the two in that order, and so is between
[e] and [a]a. However, sometimes both [e] and [o] may change, e.g.,
enko>inku. |
| (d) Elimination
of the final consonant: |
| In such cases as
opposed to elision, the consonants leave their traces on adjacent
sounds. |
| The final [N]
disappears leaving the trace of nasalization on the preceding vowel.
When any morpheme comes after it beginning with a vowel, that vowel
also is ifself nasalized. |
| e.g., |
|
|
| Occasionally the
consonant [g] may be introduced between [N] and the following
vowels. |
| e.g., |
|
|
| The
tense markers ending with /d/ are most exposed to assimilation, when
the following suffix may hav either a vowel or a consonant. Some typical
and common examples are illustrated. Here this is to be remebered
that the final /d/ is phonetically a checked sound [?d] so that this
checking plays important role. |
| (a) When the
sound [i] as insertion of direct object follows ked, akad, led, tad,
jad etc., the forms become like the following, employing two sub-rues: |
| (i) deletion
of /d/, and (ii) in case of /e/ preceding /d/, deletion of /e/ also: |
|
ked + i
|
=
|
ki? or
ki
|
|
|
akad +
i
|
=
|
aka?i,
simplified as akai
|
They all
lose
|
|
led + i
|
=
|
li? ro
li
|
the final
|
|
tad + i
|
=
|
ta?i or
tai
|
consonant
|
|
jad + i
|
=
|
ja?i or
jai
|
|
|
|
| Since /d/ is preglottalized
in all above cases, the glottal check remains even after insertion
of /i/. |
| (b) When the
sound [a] follows: |
|
ked+ate
|
=
|
ke?ate
|
|
ked+ad
|
=
|
ke?ad,simplified
to /kead/
|
|
led+a
|
=
|
la?
|
|
jad+am
|
=
|
ja?m
|
|
|
| (c) When the
sound [e] follows: |
|
|
| (d) The sound
[d] is lost even when a consonant follows: |
|
ked+ci
|
=
|
ki?ci
|
|
jad+ko
|
=
|
ja?ko or
jako
|
|
ked+ke
|
=
|
kere
|
|
ked+ko
|
=
|
ketko
|
|
|
+a?
ć?
