Rkou Scratchpad

[Xwtek]

I was thinking of a language for what is basically Orcish race in my conworld.

Here is the phoneme inventory

Consonant Inventory

Code: Select all

 m n
 p t k ʔ
   r w h

Vowel Inventory

Short Vowel (All of them have breathy voiced and creaky variant)

Code: Select all

ɪ ʏ   ʊ
ɛ œ ə ɔ
æ     ɑ

Long Vowel (Only modal voiced vowel)

Code: Select all

iː yː ɨː uː
eː øː əː oː
ɛː       ɔː
    a:

Nasalized short vowel (Has creaky variant)

Code: Select all

ɪ̃ ʊ̃
ɛ̃ ɔ̃

Nasalized long vowel (Only modal voiced vowel)

Code: Select all

ĩː ũː
ẽː õː
ɛ̃ː ɔ̃ː

Rhotacized vowel (Only modal voiced vowel)

Code: Select all

    u˞ː
ɜ˞ː ɔ˞ː

In unstressed syllable, only vowel allowed is /i u e o a ĩ ũ ã/ with modal phonation unless overridden.

Note that adjacent to /h/ or /ʔ/ or cluster containing it, vowel phonation is not phonemic and it's always breathy and creaky, in order, even if vowel normally cannot carry it.

The phonotactics is C(C)V where all allowed consonant cluster is (* means it is not allowed word initially).

Cr pr tr kr mr nr *wr
rC rp rt rk *rm *rn rw
Cw tw kw nw
Ch ph th kh mh nh rh
Cʔ pʔ tʔ kʔ nʔ mʔ nʔ rʔ

Vowel hiatus is allowed unless the vowel is identical. But word cannot start with vowel.
Only nasal vowels occur adjacent to nasal consonant or cluster containing it.

Rkou is not tonal. However, pitch varies depending on vowel and its surroundinɡ.
1. Breathy, or glottalized vowels are pronounced with low pitch
2. Nasal modally voiced short vowels are pronounced with medium pitch
3. Oral modally voiced vowels are pronounced with high pitch
4. Rhotacized vowel is pronounced with low falling pitch if the consonant begins with /p t k ʔ h/, otherwise, it is pronounced with low pitch
(The next rules uses T for /p t k ʔ h/ and D for /m n r w/)
5. Nasalized long vowel is pronounced with low falling pitch at environment T_D, low raising tone at environment D_T, medium pitch at environment T_T, and otherwise uses low pitch
6. Oral long vowel is pronounced with high falling pitch at environment T_D, high raising tone at environment D_T, high pitch at environment T_T, and otherwise uses medium pitch

[mèþru]

It seems like the language is transitioning either from or to a tonal stage though. More likely the latter.

[Vilike]

Only nasal vowels occur adjacent to nasal consonant or cluster containing it.

So, are they phonemic? A common question.

Are there voiced allophones of the stops?

[Xwtek]

Only nasal vowels occur adjacent to nasal consonant or cluster containing it.

So, are they phonemic? A common question.

Yes, in a environment that there is no nasal consonant adjacent to it, it is contrastive.

[Xwtek]

I decided to revise the phonology.

1. Rkou will get voiced plosives but will lack nasal consonant. So the consonant inventory becomes

Code: Select all

p t k ʔ
b d ɡ
  r w h

There is no longer nasality constraints. However, voiced plosives becomes nasal consonant the consonant directly adjacent to the consonant is nasal.

2. Rkou will get coda. The syllable coda is simpler than the initial though, although still CC. Also I disallow hiatus

So the Phonotactics of Rkou becomes:

C(C)V((C)(C)CV)*(C)(C).

Allowed clusters:

Code: Select all

Initial
     p   t   k   ʔ   b   d   ɡ   r   w   h 
rC   rp  rt  rk  rʔ  rb  rd  rɡ      rw  rh
Cr   pr  tr  kr      br  dr  ɡr 
Cw       tw  kw          dw  ɡw          hw
Cʔ   pʔ  tʔ  kʔ      bʔ  dʔ  ɡʔ      wʔ
Ch   ph  th  kh      bh  dh  gh 

Medial
     p   t   k   ʔ   b   d   ɡ   r   w   h 
rC   rp  rt  rk  rʔ  rb  rd  rɡ      rw  rh
Cr   pr  tr  kr      br  dr  ɡr      wr
Cw       tw  kw          dw  ɡw          hw
Cʔ   pʔ  tʔ  kʔ      bʔ  dʔ  ɡʔ      wʔ
Ch   ph  th  kh      bh  dh  gh 
pC   pp  pt  pk      bb  bd  bg
tC   tp  tt  tk      db  dd  dg
kC   kp  kt  kk      gb  gd  gg
rpC     rpt rpk                 rpr
rtC rtp     rtk                 rtr rtw
rkC rkp rkt                     rkr rkw
rbC                     rbd rbg rbr
rdC                 rdb     rdg rdr rdw
rgC                 rgb rgd     rgr rgw
wtC wtp     wtk                 wtr wtw
wkC wkp wkt                     wkr wkw
wdC                 wdb     wdg wdr wdw
wgC                 wgb wgd     wgr wgw

Final
rC   rp  rt  rk      rb  rd  rg
wC       wt  wk          wd  wg  wr
CC   pp  tt  kk      bb  dd  gg  rr

3. Rkou no longer has stress reduction.

[Xwtek]

Phonology Inventory

Consonant:

	Labial	Alveolar	Velar	Glottal
Voiceless	p	t	k	ʔ
Voiced	b	d	ɡ
Liquid		r	w	h

Vowel:
Modally voiced vowel [/cell]

ɪ

ɪ̃

iː

ĩː

ʏ

y:

ɨː

ʊ

ʊ̃

u:

ũː

ʉ̞˞ː

eː

ẽː

øː

ə

əː

o:

õː

ɛ

ɛ̃

ɛː

ɛ̃ː

œ

ɜ˞ː

ɔ

ɔ̃

ɔː

ɔ̃ː

æ

aː

ɑ

ɑ˞ː

ɑɪ

Adjacent to cluster /ʔ/, modally voiced vowel becomes creaky. Similarly to /h/ except to breathy voiced vowel. For short vowel (oral only for /h/), it caused merger. However, for long vowels, no merger occurs. If the vowel is adjacent to both /ʔ/ and /h/, the consonant in the same syllables as the vowel decides.

Creaky voiced

ɪ	ɪ̃	ʏ		ʊ	ʊ̃
ɛ	ɛ̃	œ	ə	ɔ	ɔ̃
æ				ɑ

Breathy voiced

ɪ	ʏ		ʊ
ɛ	œ	ə	ɔ
æ			ɑ

By the way, the table syntax is too clunky and too nonstandard to be generated via 3rd party tools

[Xwtek]

Phonotactics
Initial
Rkou allows all consonant to begin the syllable. Plus these cluster:
{p, t, k, b, d, g, r}{r, h, w, ʔ}
The constraint is that identical consonant cluster is disallowed, and /w/ cannot be in the same cluster with /p, b/.
Final
Rkou allows all nonglottal cluster to end the syllable. Additionally, /r/, /w/ and identical consonant can follow a stop or /r/. /w/ cannot occurs in the same cluster with labial consonant
Medial
Rkou allows consonant of form {r, w}SS{r, w, h, ʔ}. The constraint is that the stop may not be identical if all four position occurs, (There is no problem with that if there is only 3 consonants filled, though) no /w/ in cluster if /p/ or /b/ occurs in the same cluster, and the stops must agree in voicing.
Assimilation.
If two morpheme is combined, it is assimilated so the cluster is permitted
1. If there is one voiced stop, the other stop in the same cluster becomes voiced, too.
2. If there is labial stop somewhere in the cluster, drop all instances of /w/.
3. Check the length of the cluster
◦ 1-2 : finished
◦ 3-4 : check pattern (C=/ptkbdgrwhʔ/, S=/ptkbdg/, L=/rw/, G=/ʔh/) (Repeat this until nothing matches):
▪ S:S : Shorten the first stop => ggb → gb
▪ SrS : Move r to after second S => krt → ktr
▪ SLG: Drop G
▪ SLL: If there is at least 1 /w/, drop one of them. Else drop one of the /r/.
▪ LLS: If there is at least 1 /w/, drop one of them. Else drop one of the /r/.
▪ rrL: Turn into rL
▪ C:CC: Shorten the first consonant
▪ LC:C: Shorten the second consonant
▪ LSwS: Drop the w
▪ LSrS: Turn the cluster to LSSr
▪ LSLL: If there is at least 1 /w/, drop one of them. Else drop one of the /r/.
▪ LSLG: Drop the glottal phoneme.
▪ Oherwise: Pass to the next

[Xwtek]

Allophones
Consonants

1. Directly adjacent to a nasal vowel, or in the same syllable as one, /b d g/ becomes [m n ŋ]
2. Directly adjacent to a rhotacized vowel, or in the same syllable as one, /t d n r/ becomes [ʈ ɖ ɳ ɽ]
3. Cluster /pʔ tʔ ʈʔ kʔ/ becomes ejectives, even if it is actually on different syllable. Example: öht-’-a /œ̤t-ʔ-ɑ/ (check-PL.SBJ-3SG.OBJ) “he checks it”, becomes [œ̤tʼɑ̰].
4. /k g ŋ/ is palatized into /c ɟ ɲ/ adjacent to front vowel.
5. [ɹ] is inserted after rhotacized vowel if the cluster is not already full

Vowels

1. Modally voiced nasal vowel and breathy nasal vowel is on free distribution. Generally speaking, modally voiced one is preferred.
2. In closed syllable, /ɪ/, /ʏ/ and /ʊ/ is often realized as /ë/, /ø̈/ and /ö/ (i.e. slightly centralized high-mid vowel). In the same environment, /iː/, /yː/, /uː/, and /øː/ is often realized as /ɪː/, /ʏː/, /ʊː/, and /œː/. Diphthonɡ /ɑɪ/ is often raised to /ʌɪ/. /ɑ˞/ becomes /ɔ˞/
3. Adjacent to /k/, /g/ and /w/, the central vowel becomes back unrounded /ɨː/ > /ɯː/, /ə/ > /ɤ/, and /əː/ > /ɤː/
4. In the same environment, /ɜ˞/ and /ɑ˞/ merɡes into /ɔ˞/
5. Note to English speaker, V + r contrasts with rhotacized vowel here. For example: ôk /ʔɑ˞ːk/ [ʔɔ̰˞ːɹk] means “orc”, but oork /ʔɔːrk/ [ʔɔ̰ːrk] means “I agreed.” and ôrk /ʔɑ˞ːrk/ [ʔɔ̰˞ːɽk] means, “firewood”

Tones
Rkou language is not tonal, however, many syllables is pronounced with certain tone. Rkou has 6 phonetic (I lack a better word for this, this is not exactly phonemic, but it is also rarely surface as is) tones. H M L HL MH LH. Here’s how tone is assigned to each syllable. (Note that Rkou also has tone sandhi that further affect the pronunciation)

1. Rkou short vowel that is not followed or preceded by /r/ or /w/ can only be assigned three tones H M L. It depends the voicing of the vowel and the adjacent consonant, and the nasality of the vowel. (in this table +V means presence of /bdg/ or absence of /ptk/ in cluster near the vowel, -V is anything not +V ), Note that we use surface voicing realization here.

   	Oral			Nasal
   	Breathy	Modal	Creaky	Breathy/Modal	Creaky
   +V	L	M	L	L	L
   -V	L	H	M	M	L

2. Rkou long vowel, or short vowel followed or preceded by /r/ or /w/, on other hand can be assigned all six tones. It is considerably more complex too. Here we use notation Z for presence for /bdg/ in cluster, S for presence of /ptk/ or the consonant is simply /ʔ/ and /h/, and L for cluster that don’t follow that specification

   	Oral			Nasal
   	Breathy	Modal	Creaky	Breathy/Modal	Creaky
   S_S	M	H	L	L	L
   S_Z	L	HL	M	M	L
   S_L	M	H	M	M	L
   Z_S	LH	MH	L	L	L
   Z_Z	L	M	L	L	L
   Z_L	L	M	L	L	L
   L_S	M	H	LH	LH	M
   L_Z	L	M	L	L	L
   L_L	M	M	L	L	L

3. Rhotacized is always low, except at adjacent to /ptk/ (It becomes mid instead)

[Xwtek]

Tone sandhi
After each syllabe is given a tone, tone in second syllable and beyond is mutated based the previous tone. Here is the rule.

	Current	H	M	L	HL	LH	MH
Previous
H		H	M	M	HL	M	H
M		H	M	L	HL	MH	MH
L		M	M	L	M	LH	LH
HL		M	L	L	M*	LH	LH
LH		H	H	L	HL	M*	H
MH		H	M*	M	HL	M*	H

After the tonal sandhi is done, the tone M* make the previous syllable have tone M. Example:
diːtɔ̰k → diː(MH)tɔ̰k(M) →diː(MH)to̰k(M*)→ diː(M)to̰k(M)