Ad blocker interference detected!
Wikia is a free-to-use site that makes money from advertising. We have a modified experience for viewers using ad blockers
Wikia is not accessible if you’ve made further modifications. Remove the custom ad blocker rule(s) and the page will load as expected.
Sangi Consonant MutationsEdit
Sangi has two kinds of consonant mutation. The first is "phonological mutation", a process which occurs at morpheme boundaries, affecting only the phonology of a word and not its meaning. The second is "morphological mutation", which changes the meaning of the word as well as its phonology.
Initial Plosive MutationsEdit
Sangi, like most of the Celtic languages, has a series of initial consonant mutations which change the sound of the word and can change the meaning as well. There are four types of mutation, all of which only affect non-nasal plosives; nasal, aspirant, spirant and soft.
Like Welsh, rather than Irish, initial consonant mutation is marked by the new pronunciation, not the base consonant plus another to represent it.
- Negative adjective forms, e.g. pic>m̌ic
- Nouns following the number 1 - Although not indicated in writing, the final "n" of "an" (one) assimilates becoming the same as the following nasal, and becomes "m" when followed by "pm-" mutation. For example "an toc" > "an ňoc"
- Locative noun stems, e.g. toc>þoccanda (towards a dog)
- Male nouns after the numbers 3 and 9, e.g. ți mendo > ți ŵendo (three men)
- Feminine nouns after the word "an" (and), e.g. an mana > an ŵana (and a woman)
- Positive superlative adjective forms, e.g. pic>wicissi (biggest)
- Nouns follwoing the number 2, e.g. su tuś > su suś (two dogs)
- Comparative forms following a feminine noun, e.g. mana picer > mana bicer (bigger woman)
- Equative forms following a masculine noun, e.g. mano piþisse > mano biþisse (man of the same bigness)
- Adjectives used as adverbs following verb-nouns, e.g. maneŵar picil > maneŵar bicil (he is a big man, literally "he is man bigly)
Mutation only affects nouns and adjectives so when these become verbs, the mutation cannot occur if the verb form is given a prefix.
I-affection occurs when the last consonant of the stem is immediately followed by the vowels i and e (long, short and diphthong forms) and also the consonant j. This does not occur however when the -i, -e or -j is part of a morphological suffix unless, although there are some exceptions to this rule.
I-Affection does not affect geminate l, though.
The "pf" form occurs with a singular "p" mutating, "ff" with "pp".
Like I-affection, A-affection occurs when the last consonant of the stem is immediately followed by a vowel, in this case a. Also, like I-affection, this process does not occur with the majority of morphological suffixes.
A slight irregularity in this system occurs as “w”, “h” and the velar stops are affected in the above way by a following “o”. A-Affection doesn’t affect consonant clusters, unlike I-Affection, except when the final consonant is an h, in which case the h, unexpectedly, becomes an the a-mutated form of the preceding consonant, or the final consonant of this if it is a cluster itself, e.g. nh>nd, rh>rź, etc.
With geminate finals, only the second consonants will undergo a-affection.
This mutation occurs when forming the plural of the four noun stems (see below).
When the plural stem is weakened or strengthened then the retroflex form of the plural is used as the stem consonant. The plural forms of retroflex consonants are the retroflex equivalents of the plural forms of the alveolar counterparts with 2 exceptions. Retroflex plural forms of alveolar stems are preceded with a retroflex "r" when the stem is retroflex and the plural of retroflex "-tt" is retroflex "d".
Stem gradation is a weakening of the final consonant of the stem to form different noun and verb stems which are used to built up the meaning of the noun or verb as well as to derive words from different classes of words.
pp > p > b > v > f > w > h > hi > he
tt > t > d > j > i > e*
cc > c > g > j > i > e
m > mb > mm > mp > mi > me
m̌ > mp > m̌m̌ > mf > m̌i > m̌e
n > nd > nn > ns > nt > ni > ne*
ň > ŋ > ng > nc > nct > ŋi > ŋe
ŋ > ng > ŋŋ > ŋc > ŋi > ŋe
ŋ̌ > nc > ŋ̌ŋ̌ > ŋ̌i > ŋ̌e
l > ld > ll > ls > lt > li > le*
ł > łd > łł > łs > łt > łi > łe
r > rd > rr > rs > rt > ri > re*
ř > řd > řř > řs > řt > ři > ře*
s > t > d > j > i > e
sp > sw > sl > śi > śe
st > ss > ś > si > se
sc > ś > si > se
h > p > b > w > hi > he
Ø > c > g > j > i > e (if the Ø was originally a c)
Ø > t > d > j > i > e (if the Ø was originally a t)*
- These gradation series can also be applied to retroflex consonants. So if the sequence was originally VVt# then the t is reinstated as retroflex, removing the ambiguity caused by the process VVt#>V#. The “j” however is pronounced as “hj” and written “ĵ”, while the “i” and the “e” are pronounced as normal.
It should be noted that “s”, “tt”, “cc” and “Ø” all eventually weaken to the same pattern, causing a certain amount of ambiguity if the gradation is carried through enough times. Also, grades of the pattern “Ci” and “Ce” can never be the general stem, being analysed as “C-i/e” where C is graded, but “i” and “e” can be strengthened.