View source for Sangi/Phonological changes

=Phonological changes=

The sound changes shown below are used to create base (non-compound) words in Sangi from a Germanic English root.

==Consonant Changes==

===Initial consonant changes===

{| class="wikitable"
|-
! English
! Before C
! Before l
! Before r
! Before V
! Before a
! Before u
! Before ŭ
! Before ə:
! Before i/e/ə
|-
| p
| h
| Ø
| Ø
| Ø
| Ø
| Ø
| Ø
| Ø
| Ø
|-
| b
| p
| p
| p
| p
| p
| p
| p
| p
| p
|-
| m
| -----
| -----
| -----
| m
| m
| m
| m
| m
| m
|-
| f
| b
| w
| b
| w
| w
| w
| w
| w
| w
|-
| v
| f
| f
| f
| f
| f
| f
| f
| f
| b
|-
| w
| -----
| -----
| -----
| v
| Ø
| Ø
| Ø
| Ø
| v
|-
| þ
| t
| t
| t
| ti
| ti
| ś
| ti
| ti
| t
|-
| ð
| d
| d
| d
| di
| di
| di
| di
| di
| d
|-
| t
| s
| s
| ---
| s
| s
| s
| s
| s
| ć
|-
| d
| t
| t
| ---
| t
| t
| t
| t
| t
| ǵ
|-
| n
| -----
| -----
| -----
| n
| n
| n
| n
| n
| n
|-
| r
| -----
| -----
| -----
| r
| r
| r
| r
| r
| r
|-
| s
| s
| s
| ş
| s
| h
| s
| f
| h
| d
|-
| z
| -----
| -----
| -----
| z
| z
| z
| z
| z
| ð
|-
| l
| -----
| -----
| -----
| l
| l
| l
| l
| l
| l
|-
| ś
| ś
| ś
| ś
| ś
| ś
| tj
| ś
| ś
| sc
|-
| ź
| -----
| -----
| -----
| zi
| zi
| zi
| zi
| zi
| zj
|-
| j
| -----
| -----
| -----
| ć
| ć
| ć
| ć
| ć
| Ø
|-
| c
| c
| h
| h
| h
| s
| h
| s
| h
| s
|-
| g
| c
| c
| c
| c
| c
| c
| c
| c
| z
|-
| ŋ
| -----
| -----
| -----
| -----
| -----
| -----
| -----
| -----
| -----
|-
| h
| -----
| -----
| -----
| Ø
| Ø
| Ø
| Ø
| Ø
| Ø
|-
| ć
| t
| t
| ț
| t
| t
| t
| t
| t
| c
|-
| ǵ
| g
| g
| ḑ
| g
| g
| g
| g
| g
| g
|}

'w' also becomes 'Ø' before o unless the w becomes v in which case the w disappears and the o becomes an a.

===Medial Consonant Changes===

{| class="wikitable"
|-
! English
! Before C
! Before l
! Before r
! Between V
! (After C) Before a
! (After C) Before u
! (After C) Before ŭ
! (After C) Before ə:
! (After C) Before i/e/ə
|-
| p
| h
| h
| h
| v
| v
| v
| v
| v
| v
|-
| b
| p
| p
| p
| p
| p
| p
| p
| p
| p
|-
| m
| m
| m
| m
| m
| m
| m
| m
| m
| m
|-
| f
| p/w (after V)
| p
| p/w (after V)
| pi
| pi
| s
| s
| pi
| p
|-
| v
| f
| b
| f
| bi
| bi
| bi
| bi
| bi
| b
|-
| w
| -----
| -----
| -----
| v
| Ø
| Ø
| Ø
| Ø
| v
|-
| þ
| t
| t
| t
| si
| si
| si
| si
| si
| t
|-
| ð
| d
| d
| d
| zi
| zi
| zi
| zi
| zi
| d
|-
| t
| t
| t
| ț
| d
| t
| t
| t
| t
| ć
|-
| d
| d
| d
| ḑ
| t
| d
| d
| d
| d
| ǵ
|-
| n
| n
| n
| ņ
| n
| n
| n
| n
| n
| n
|-
| r
| -----
| -----
| ŗ
| r
| r
| r
| r
| r
| r
|-
| s
| s
| s
| ș
| z
| s
| f
| f
| s
| d
|-
| z
| þ
| þ
| ẓ
| s
| z
| z
| z
| z
| ð
|-
| l
| l
| l
| ļ
| l
| l(/j/i)*
| l(/j/i)*
| l(/j/i)*
| l(/j/i)*
| l(/j/i)*
|-
| ś
| ś
| ś
| ś
| ś
| ś
| ś
| ś
| ś
| sc
|-
| ź
| z
| z
| z
| zi
| zi
| zi
| zi
| zi
| z
|-
| j
| -----
| -----
| -----
| j(/Ø)*
| j(/Ø)*
| j(/Ø)*
| j(/Ø)*
| j(/Ø)*
| Ø
|-
| c
| c
| h
| c
| c
| h
| h
| h
| h
| s
|-
| g
| gC>Cj
| c
| c
| þ
| c
| c
| c
| c
| z
|-
| ŋ
| g
| ŋ
| g
| ŋ
| ŋ
| ŋ
| ŋ
| ŋ
| ŋ
|-
| h
| -----
| -----
| -----
| Ø
| Ø
| Ø
| Ø
| Ø
| Ø
|-
| ć
| t
| c
| ț
| ci
| ti
| ti
| ti
| ti
| c
|-
| ǵ
| d
| g
| ḑ
| gi
| di
| di
| di
| di
| g
|}

'w' also becomes 'Ø' before o unless the w becomes v in which case the w disappears and the o becomes an a.
* See the table of Other Consonant Changes

===Final Consonant Changes===

{| class="wikitable"
|-
! English
! After C
! After V
! After VV
|-
| p
| h (sounded as "x")
| h (sounded as "x")
| h (sounded as "x")
|-
| b
| b
| b
| b
|-
| m
| m
| m
| m
|-
| f
| p
| p
| p
|-
| v
| w
| w
| w
|-
| w
| -----
| -----
| -----
|-
| þ
| t
| t
| t
|-
| ð
| d
| d
| d
|-
| t
| Ø
| Ø
| Ø
|-
| d
| s*
| s*
| s*
|-
| n
| n*
| n*
| n*
|-
| r
| -----
| -----
| -----
|-
| s
| r*
| r*
| r*
|-
| z
| g
| g
| g
|-
| l
| l*
| l*
| l*
|-
| ś
| sc
| sc
| sc
|-
| ź
| z*
| z*
| z*
|-
| j
| -----
| -----
| -----
|-
| c
| Ø
| Vc>VV
| VVc>Vj, V¹V²(V³)c>VV(V²)j¹
|-
| g
| c
| c
| c
|-
| ŋ
| -----
| ŋ
| ŋ
|-
| h
| -----
| -----
| -----
|-
| ć
| t*
| t*
| t*
|-
| ǵ
| d*
| d*
| d*
|}

¹ The forms of this new vowel complex can be found in th table for Vowel Changes.
* Retroflex after VVC#>VrC#

===Consonant Cluster Changes===

{| class="wikitable"
|-
! 
! Initial
! 
! Medial
!
! Final
|-
! English
! Before V
! Before i/e/ə
! General
! Before i/e/ə
!
|-
| sl
| sj
| ś
| sj
| ś
| ---
|-
| sp
| sp
| sp
| ff*
| ff*
| sp
|-
| st
| st
| st
| ss*
| ss*
| ss
|-
| sc
| sc
| ś
| sc*
| ś*
| ś
|-
| spr
| spr
| spr
| pr*
| pr*
| -----
|-
| str
| şţ
| şţ
| şş*
| şş*
| -----
|-
| scr
| scr
| scr
| cr*
| cr*
| -----
|-
| spj
| sp
| sp
| sp
| sp
| -----
|-
| stj
| st
| st
| st
| st
| -----
|-
| scj
| sc
| sc
| sc
| sc
| -----
|-
| scw
| scv
| scv
| śv*
| śv*
| -----
|-
| spl
| spi/spj
| spi/spj
| spi/spj
| spi/spj
| -----
|-
| sm
| śim
| śim
| sm*
| sm*
| -----
|-
| sn
| śin
| śin
| sn*¹
| sn*¹
| -----
|-
| cw
| p
| p
| p
| p
| -----
|-
| hw
| cw
| cw
| cw
| cw
| -----
|-
| VVm
| -----
| -----
| Vmm
| Vmm
| Vmm
|-
| mp
| -----
| -----
| m̌
| m̌
| m̌
|-
| mp (before a schwa, which will disappear after this cluster)
| -----
| -----
| mb
| mb
| mb
|-
| mb
| -----
| -----
| mp
| mp
| mp
|-
| nt
| -----
| -----
| nn¹
| nn¹
| nn¹
|-
| nd
| -----
| -----
| nt¹
| nt¹
| nt¹
|-
| ns
| -----
| -----
| nd¹
| nd¹
| nd¹
|-
| nz
| -----
| -----
| ng
| ng
| ng
|-
| nć
| -----
| -----
| ň¹
| ň¹
| ň¹
|-
| nǵ
| -----
| -----
| ns¹
| ns¹
| ns¹
|-
| nc
| -----
| -----
| ŋ̌
| ŋ̌
| ŋ̌
|-
| ng
| -----
| -----
| ŋ
| ŋj
| nc
|-
| ps
| -----
| -----
| ps
| ps
| pr
|-
| ts
| -----
| -----
| tt¹
| tt¹
| ss¹
|-
| cs
| -----
| -----
| cs¹
| cs¹
| cr
|-
| lp
| -----
| -----
| ł
| ł
| ł
|-
| rp
| -----
| -----
| ř
| ř
| ř
|}

* These also occur in the spoken and written language at the beginning of words when the final sound of the previous word was a vowel, e.g. “an e ssil – and yet still”. It is a kind a sandhi. This process also happens after a prefix which ends in a vowel.

¹ These sounds, or there alveolar parts, become retroflex under the normal conditions, i.e. preceded and/or followed by an "r"

===Other Consonant Changes===

{| class="wikitable"
|-
! English
! 
|-
| Cj
| C, VCCV¹, VCV²
|-
| Cw
| Cv
|-
| Cl
| Ci, Cj, Velar-l³
|-
| Cr
| Cr
|-
| Vct
| Vit, Vp#
|-
|}

¹ This applies only to plosives, sibilants and nasals in there basic form.

² The j affects the fricative consonants (from English) in the following way: f > p; v > b; th > t; dh > d; ch > c; gh > g; l > jj; r > s.

³ This change means that the only Cl combinations are cl, gl and hl.  After all other consonants l becomes i but becomes j before i to avoid the creation of a long vowel and causes palatalisation of the preceding consonant. If l is preceded by an original initial “p”, now absent, it is still either “i” or “j”.


====Palatalisation====

Palatalisation is a change which occurs after the new word has been made from the original English. It occurs when a word ends in an -i but is then followed by a vowel ending, e.g. “-a” for female nouns in the nominative case. It also occurs when and l becomes a j. It only affects a small number of consonants and should not be confused with I-affection, which is a morpho-phonological process. 

t > ć

d > ǵ

s > ś

z > ź

When the stem undergoes any form of consonant mutation, the original combination of “Cj” or “Ci” re-emerges. It is this combination which goes through the mutation, not the palatalised form. The “j” is, however, ignored under these processes if the new form can be palatalised, if it can not then the “j” is taken into account and the stem will not mutate, for example “-tj-” becomes “-ć-” under palatalisation. When subjected to plural mutation “t” becomes “ð”. “ð” cannot be palatalised so the plural form of “-tj-” remains “-tj-” and therefore becomes “-ć-”, even though “-ć-” can undergo plural mutation. The “t” of the same “-tj-” combination, when subjected to consonant gradation, becomes “d”, which can be palatalised. The new sequence will be “-dj-” which becomes “-ǵ-”. This process does not occur for -Ci- sequences. Here the -i- re-emerges and the consonant is changed regardless. The -i- only disappears again when the new consonant can be palatalised otherwise it remains intact. So the sequence -ti- will change to “-ć-” when palatalised and the “-ti-” sequence will become “-ði-” in the plural and remain as such. If the original "-ti-" sequence were weakened, however, it would become “-di-” which would palatalise to “-ǵ-”. If the sequence were “-Cji-” then it is seen as “-Cj-” followed by a “-i-”. This means the "C" will palatalise but will not undergo any further form of mutation as the "-i-" is seen as the stem ending.

==Vowel Changes==

{| class="wikitable"
|-
! English
! General
! Before a single stop, r, l, s
! Before l (only l)
! Before m (only m)
! Before final t - Weak I - Weak II
! Before final c - Weak I - Weak II
! Weak I when final
! Weak II when final
|-
| a
| a
| a
| a
| a
| a – at – ad
| â – ac – ag
| a
| a
|-
| e
| e
| e
| e
| e
| e – et – ed
| ê – ec – eg
| e
| e
|-
| i
| i
| i
| i
| i
| i – it – id
| î – ic – ig
| i
| i
|-
| o
| o
| o
| o
| o
| o – ot – od
| ô – oc – og
| o
| o
|-
| u
| u
| u
| u
| u
| u – ut – ud
| û – uc – ug
| u
| u
|-
| a:
| a
| aC
| al
| a
| a – att – add
| aj – ai – ae
| a
| a
|-
| e:
| ö
| öC
| öl
| ö
| ö – ött – ödd
| öj – öi – ö
| ö
| ö
|-
| i:
| i
| iC
| il
| i
| i – itt – idd
| ij – î – ie
| i
| i
|-
| o:
| o
| oC
| ol
| o
| o – ott – odd
| oj – oi – oe
| o
| o
|-
| u:
| u
| uC
| ul
| u
| u – utt – udd
| uj – ui – ue
| u
| u
|-
| ŭ
| a
| a (forms following retroflex, not word finally)
| a
| a
| a – at. – ad.
| aj – ai – ae
| a
| a
|-
| ə
| e
| e (forms following retroflex, not word finally)
| e
| e
| e – et. – ed.
| ej – ei – ê
| ê
| ê
|-
| ö
| e
| eC
| el
| e
| e – ett – etd
| ej – ei – e
| er
| erd
|-
| ai
| â
| â
| â
| ar
| â – ât – âd
| âj – âi – âe
| â
| ai
|-
| ei
| ê
| ê
| ê
| er
| ê – êt – êd
| êj – êi – ê
| ê
| ei
|-
| oi
| î
| î
| î
| ir
| î – ît – îd
| îj – jî – îe
| î
| oi
|-
| au
| û, labial-â
| û, labial-â
| or
| or
| û/â – ut./ât – ud./âd
| ûj/âj – ûi/âi – ûe/âe
| û/â
| uh/ah
|-
| əu
| labial-û, ŷ
| labial-û, ŷ
| yr
| yr
| û/ŷ – ut./yt. – ud./yd.
| ûj/ŷj – ûi/ŷi – ûe/ŷe
| û/ŷ
| uh/yh
|-
| iə
| y
| y
| y
| yr
| y – yt – yd
| ŷj – ŷ – ŷe
| yr
| yrd
|-
| uə
| ä
| ä
| ä
| ä
| ä – ät – äd
| öj – öi – öe
| är
| ärd
|-
| aiə
| ai
| ai
| â
| ai
| ai – ait – aid
| âij – âî - âie
| ar
| ae
|-
| eiə
| ei
| ei
| ê
| ei
| ei – eit – eid
| êij – êî - êie
| er
| ê
|-
| oiə
| oi
| oi
| ô
| oi
| oi – oit – oid
| ôij – ôî – ôie
| or
| ô
|-
| auə
| au
| au
| û
| au
| au – aut – aud
| âuj – âui – âue
| au
| ahi
|-
| əuə
| ou
| ou
| û
| ou
| ou – out – oud
| ôuj – ôui – ôue
| ou
| ohi
|-
| (ə)m
| ü
| -----
| -----
| -----
| -----
| -----
| em
| emm
|-
| (ə)n*1
| en
| -----
| -----
| -----
| -----
| -----
| end
| enn
|-
| (ə)l*
| ô
| -----
| -----
| -----
| -----
| -----
| ô
| ôr
|-
| ə(r)* (final form of the schwa)
| î
| -----
| -----
| -----
| -----
| -----
| î
| îr
|}

* Syllabic “m”, “n”, “l”, and “r”. The inserted “e” is lost when the consonant is followed by a vowel though. It acts more as an epenthetic vowel that still causes change in consonants when forming new words from English roots. When a compound is formed and the first word ends in “î” (original “r”) and the next begins with a vowel, the original “r” will be restored and the “î” lost. A special circumstance for this change is when forming surnames in which the second begins when “l”. The “r” returns causing the “l” to become retroflex and the “r” to disappear again.

1. Word finally after a dental consonant, the syllabic "n" disappears creating a retroflex version of the preceding consonant.

The changes of /ə/ are quite complex. It always becomes /e/ before a dental, which when non-final become a retroflex. However, when two /ə/s occur in two syllables, one after the other, then the first one will be lost unless it is before a dental in which case the second one is lost, but if both appear before a dental then both are retained, e.g. /eləfənt/>/elpen/, etc.

[[Category:Sangi]]