Random phonology/phonemic inventory thread [2011–2018]

A forum for all topics related to constructed languages
cromulant
cuneiform
cuneiform
Posts: 143
Joined: 13 Jan 2012 00:04

Re: Random phonology/phonemic inventory thread

Post by cromulant »

Porphyrogenitos wrote:Sometimes I think about using a "staggered" stop inventory like this:

Code: Select all

    kʼ
  t k
b d
ɓ
It's a little obnoxious but it's typologically justified.
Yes, due to more sonorous stops being easier at the front of the mouth and less sonorous stops being easier at the back. /b t d k/ is a fairly common thing and this just takes it a step further. I have had musings in similar directions.

Here is a particularly obnoxious one I came up with recently. It's a phonology based on the letters of my name, nothing too serious.

Code: Select all

    k
  d g
m n
Here's another from some time back:

Code: Select all

  tʼ kʼ
p t 
  d  g
ɓ ɗ
It might seem like a violation that it has /p g/ but no /b k/. Think of it as /p t k b d g/, but with /b k/ "exaggerated" in terms of their sonority or lack of. /tʼ ɗ/ are then thrown in to balance it out.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

In a follow-up to my previous phonology.

/t̪~θ (t)s (t)ɬ (t)ʃ k~x/
/t̪’ t͡s’ t͡ɬ’ t͡ʃ’ k’ ʔ/
/n r j/

/i a o/ in mid, high or low tone.

Syllable structure is CV(C), where any consonant may appear in the onset or coda, except that /j/ may not appear after /i/.

The plain plosives/affricates are realised as fricatives escept when following another member of the same class, when they take their stopped pronunciation. Additionally between vowels the same consonants are realised as voiced (fricatives).

Vowels are normally hort but are allophonically lengthened in the following environments (which are definitely not just a clone copy of Scottish English [;)] ):
  • Before a word-boundary
  • Before a morpheme boundary
  • Before a voiced fricative
  • Before /r/
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Another phonology inspired by (but not adhering to) my 26-syllable challenge from a while back:

/m n/
/t k/
/ɸ s h/
/l/

/i a ə u/

(C)V syllables

Code: Select all

 i   a   ə(ː)  u
(si) ta  tɨ(ː) tu
 ki  ka  ka(ː) ku
 hi  ha  hə(ː) ɸu
 mi  ma  mə(ː) mu
 ni  na  nɨ(ː) nu
 si  ha  sɨ(ː) su
 li  la  lɨ(ː) lu
 hi  ha  hə(ː) hu

All consonants except /h/ are automatically geminated after stressed /ə/. (This is a real feature found in some Austronesian languages.)

/ə/ is [ɨ] after alveolars; it is [a] after /k/, but remains underlyingly distinct from /a/ due to gemination effects.

Historical /ti/ merged with /si/.

Historical /p/ has debuccalized to [h] except before /u/, where it became [ɸ]; however, debuccalized /p/ remains underlyingly distinct from historical /h/ because it surfaces as [ɸː] after stressed /ə/.
wintiver
sinic
sinic
Posts: 214
Joined: 09 Oct 2012 03:37

Re: Random phonology/phonemic inventory thread

Post by wintiver »

/m n ɲ ŋ/ m n ň ŋ
/p t t͡s t͡ɬ t͡ʃ k ʔ/ p t c tl č k '/ʔ
/b d/ b d
/s ɬ x h/ s ł x h
/l w j/ l w y
/r/ r

/i iː/ i í
/e eː o oː/ e é o ó
/a aː/ a á

/ḭ ḭː/ į į́
/ḛ ḛː o̰ o̰ː/ ę ę́ ǫ ǫ́
/a̰ a̰ː/ ą ą́

(C)V(C) is he basic syllable structure. There is a strong aversion to put labial consonants in coda position. Anytime a syllable ends with a labial consonant it is the byproduct of assimilation to place of articulation.

Creaky voice is phonemic and indicated via an ogonek.

Glottal stops are only phonemic intervocalically and in final position.

All consonants save for the stops and /h/ can form geminates.

Syllabic nasals, laterals and trills exist in simple (C)V or V(C) syllable structures.

Some words:
/ra̰ː/ rą́: knife
/kaːdi/ kádi: spider
/n̩tir/ ntir: house
/xeʔ/ xeʔ: this (pronoun)
/ŋe/ ŋe: that (pronoun)
/haɬḛː/ hałę́: to bleed
/poʔat/ po'at: sharp, acrid smell
/kewta/ kewta: thought; world-view
User avatar
Osia
cuneiform
cuneiform
Posts: 83
Joined: 15 Sep 2016 01:16
Location: The Observable Universe

Re: Random phonology/phonemic inventory thread

Post by Osia »

Frislander wrote:In a follow-up to my previous phonology.

/t̪~θ (t)s (t)ɬ (t)ʃ k~x/
/t̪’ t͡s’ t͡ɬ’ t͡ʃ’ k’ ʔ/
/n r j/

/i a o/ in mid, high or low tone.

Syllable structure is CV(C), where any consonant may appear in the onset or coda, except that /j/ may not appear after /i/.

The plain plosives/affricates are realised as fricatives escept when following another member of the same class, when they take their stopped pronunciation. Additionally between vowels the same consonants are realised as voiced (fricatives).

Vowels are normally short but are allophonically lengthened in the following environments (which are definitely not just a clone copy of Scottish English [;)] ):
  • Before a word-boundary
  • Before a morpheme boundary
  • Before a voiced fricative
  • Before /r/
Ooh! I love the feel of this. The allophony between the stops and fricatives is lovely, along with the lateral affricates without an /l/. I also like the vowel lengthening rules.

A conlang I've been thinking of developing, based on Dravidian and Australian languages, but with heavy influences from Dyirbal and Arrente in particular. A tentative name is Mbirundhing, or something similar.

/m ŋ n̪ n ɳ ɲ/ <m ŋ nh n rn ñ>
/p k t̪ t ʈ c ʔ/ <b g dh d rd j '>
/ⁿb ⁿg ⁿd̪ ⁿd ⁿɖ ⁿɟ/ <mb ng ndh nd nrd nj>
/h/ <h>
/w l̪ l ɭ j/ <w lh l rl j>
/ɾ ɽ/ <ṙ rṙ>
/ð̞ z̞~ɹ ɻ/ <ð z r>

There is partially productive retroflex harmony, where dentals are neutral.

/a i u ɚ/ <a i u er>

With retroflex harmony, /a/ alternates with /ɚ/.

Allophony:

Vowels are nasalized before prenasalized, and adjacent to nasals and /h/.
The glottal stop is likely non-phonemic, as it only serves to seperate vowels across syllable and word boundaries, and before/after vowels phrase initially and finally.
Unvoiced stops are only devoiced at the beginning of an utterance, elsewhere they are voiced but unapproximated.
The distinction between /z̞~ɹ/ and /ɻ/ could be said to be allophonic, as they are only conditioned by harmony if there are alveolars or retroflexes elsewhere in the word, and if there are none, the two are in free variation.
The /a/ is realized as /ɔ/ adjacent to peripherals and /ɛ/ adjacent to palatals.
The /t/ may be optionally realized as /s/ before /i/.
The /h/ is realized as [ç] before /i/, [xʷ] before /u/, and [ʂ] before /ɚ/.

Syllable structure is something like CV(N,D), where N is a homoorganic nasal, and D is any stop. The coda nasal only occurs word finally, and varies with vowel nasalization. This may prenasalize initial consonants in the following word. Stress is word intial.

Grammatically, if I develop this further, it will be ergative and agglutinating, with noun like adjectives and SOV syntax.

Some sample words
rdirdi onomotapoeia for rainfall
-ŋu- to go
-mbaha- to walk
-zaziṙu- to hunt, to fish
she/her/hers
ConWorkShop
:eng: : [tick] :esp: : [:S] ASL: [>_<]
[<3] Navajo, Tlingit, Polysynthesis, Semantics
Auvon
cuneiform
cuneiform
Posts: 104
Joined: 27 Aug 2016 08:48

Re: Random phonology/phonemic inventory thread

Post by Auvon »

tjetjerû

/p t c k
ts tʃ tɕ
s ʃ ɕ x
m n ɲ
ɾ
l ʎ
ɹ
i ɯ u
e ɤ o
ɛ ɑ ɔ/

p t tj k
c č ć
s š ś x
m n nj
r
l lj

i û u
e ô o
è a ò


Syllable structure is (C)(C)V(C) where the only permitted codas are an obstruent /p t c k ts tʃ tɕ s ʃ ɕ x/ followed by an approximant /l ʎ ɹ/. Intervocalically, obstruents are voiced. /tʃ tɕ c/ are respectively: apicopostalveolar domed with delayed release (sometimes labialized to a degree), laminopostalveolar palatalized with delayed release, and palatal (with slightly delayed release).
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

/p t k ʔ/
/b~β~m d~ɾ~n/
/s x h/
/z~ɹ/

/i ĩ ɯ ɯ̃ u ũ/
/e ɤ o/
/a ã/

/˥ ˩ ˥˩/

Voiced stopped are realised as nasals before nasal vowels, while non-nasalised voiced consonants are realised as sonorants intervocalically.

Syllable structure is CV.
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

I came up with an idea for a certain kind of contrast:

Dental/alveolar vs. labialized palatoalveolar plus, potentially, vs. palatal. The labialized palatoalveolar would be slightly retroflex but the main contrast would be the labialization - specifically, roundedness. I'm imagining a consonant inventory kind of like this:

/m n nʷ ɲ ŋ ŋʷ/
/p t tʃʷ cç k kʷ/
/f s ʃʷ ç x xʷ h/
/v z ʒʷ ʝ ɣ ɣʷ/
/l ɹʷ j w/

Or, well, in reality of course, the various contrasts probably wouldn't be that thoroughgoing, so it would probably be more like:

/m n nʷ ɲ ŋ ŋʷ/
/p t tʃʷ k kʷ/
/v s ʃʷ ç x xʷ h/
/l~ɾ ɹʷ j w/
User avatar
sangi39
moderator
moderator
Posts: 3024
Joined: 12 Aug 2010 01:53
Location: North Yorkshire, UK

Re: Random phonology/phonemic inventory thread

Post by sangi39 »

Porphyrogenitos wrote:I came up with an idea for a certain kind of contrast:

Dental/alveolar vs. labialized palatoalveolar plus, potentially, vs. palatal. The labialized palatoalveolar would be slightly retroflex but the main contrast would be the labialization - specifically, roundedness. I'm imagining a consonant inventory kind of like this:

/m n nʷ ɲ ŋ ŋʷ/
/p t tʃʷ cç k kʷ/
/f s ʃʷ ç x xʷ h/
/v z ʒʷ ʝ ɣ ɣʷ/
/l ɹʷ j w/

Or, well, in reality of course, the various contrasts probably wouldn't be that thoroughgoing, so it would probably be more like:

/m n nʷ ɲ ŋ ŋʷ/
/p t tʃʷ k kʷ/
/v s ʃʷ ç x xʷ h/
/l~ɾ ɹʷ j w/
Seems reasonable enough, even with the full contrast. In English, for example, at least /ʃ/ and /r/ are usually slightly rounded in some if not most dialects and contrastive palatals doesn't seem unreasonable either, see Hungarian.
You can tell the same lie a thousand times,
But it never gets any more true,
So close your eyes once more and once more believe
That they all still believe in you.
Just one time.
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

Here's the phonology of my newest project.

/p t~s k ʔ/ <p t~s k ‘>
/b~β~m d~ɾ~n/ <b~m d~n>
/s~x h/ <s~x h>

/iə ɯə uɵ/ <ie ue uo>
/e ɤ o/ <i e u>
/a ɒ/ <a o>

Additionally all vowels nay be nasalised: this also causes nasalisation of adjacent voived stops.

Both /t/ and /x/ are realised and written as <s> before /iə e/: additionally /x/ is also realised as <s> when following /e/. Additionally, when not nasalised, the voiced stops /b d/ are realised as [β ɾ] in intervocalic position.

Syllable structure is CV(C)(C), where final CC clusters are restricted to /ʔ/ + /p t k b d x/ and /h/ + /p t k b d/. Roots are rather constrained in shape, most taking the form CV(C)C(V), where the CC cluster is again restricted to the set described above. Additionally the diphthongs can never occur word-finally, however they can occur underlyingly. When suffixes are added some instances of root-final /e o/ break to /iə uɵ/ while all instances of final /ɤ/ break to /ɯə/, while all consonant-final roots insert /ɤ/.

Additionally there is a tone system that operates at the word-level. There are three basic contours: level, rising and falling. With monosyllabic words the rising and falling tones are realised as contours on a single vowel, while in polysyllabic words the difference in tone is shown by the relative pitches of different syllables, as can be seen when suffixes are added, e.g. [bâʔ] <bà‘> with a falling tone becomes [báʔɤ̀ʔ] <bá‘e‘> when the 3rd person possessive suffix is added, while [bǎʔ] <bá‘> with a rising tone becomes [bàʔɤ́ʔ] <ba‘é‘> in the same environent.
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Thinking about Iroquoian stuff today

/t k kʷ ʔ/ <t k q x>
/s h/ <s h>
/n j ɹ ɰ/ <n y r w>
/n̥ j̊ ɹ̥ ɰ̊/ <ṅ ẏ ṙ ẇ>

/a e i o ʌ̃/ <a e i o u>
/eː iː oː ʌ̃ː/ <ee ii oo uu>
User avatar
DesEsseintes
mongolian
mongolian
Posts: 4331
Joined: 31 Mar 2013 13:16

Re: Random phonology/phonemic inventory thread

Post by DesEsseintes »

Porphyrogenitos wrote:Thinking about Iroquoian stuff today

/t k kʷ ʔ/ <t k q x>
/s h/ <s h>
/n j ɹ ɰ/ <n y r w>
/n̥ j̊ ɹ̥ ɰ̊/ <ṅ ẏ ṙ ẇ>

/a e i o ʌ̃/ <a e i o u>
/eː iː oː ʌ̃ː/ <ee ii oo uu>
I like how you’ve reduced the vowel inventory. And having ɰ rather than w is a nice twist.

If (and this applies only if) you’re going for Mohawk/Oneida style phonotactics, what is the reason you choose to define n̥ j̊ ɹ̥ ɰ̊ as phonemes? In those languages these would be analysed as hn hj etc. Perhaps you’re planning very different phonotactics?

Hope to see this taken further. [:)]
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

DesEsseintes wrote: I like how you’ve reduced the vowel inventory. And having ɰ rather than w is a nice twist.

If (and this applies only if) you’re going for Mohawk/Oneida style phonotactics, what is the reason you choose to define n̥ j̊ ɹ̥ ɰ̊ as phonemes? In those languages these would be analysed as hn hj etc. Perhaps you’re planning very different phonotactics?

Hope to see this taken further. [:)]
The /ɰ/ instead of /w/ is inspired by Passamaquoddy-Maliseet, which has /kʷ/ and /w/, but its /w/ is a lot less rounded than English /w/, so it might as well be /ɰ/ (and this is probably the case for a lot of other Algonquian langs, I'm guessing) - which I think makes a nicer contrast of /k kɰ kʷ kʷɰ/, especially after you factor in whatever voicing rules might be operative.

I threw in /n̥ j̊ ɹ̥ ɰ̊/ pretty much cause I wondered what it would be like to cross-pollinate Iroquoian and Muskogean phonetics. I'll probably look to some Muskogean languages as well for inspiration if I take this any further, but I probably won't stray too far from Iroquoian phonotactics, either. The voiceless resonants might only end up being rather marginally phonemic, idk.
Solarius
roman
roman
Posts: 1173
Joined: 30 Aug 2010 01:23

Re: Random phonology/phonemic inventory thread

Post by Solarius »

/p t k/<p t k>
/s/<s>
/m n ŋ/<m n ng>
/r ɾ/<rr d>
/ʋ ɹ j/<v r j>

/i y e ø æ ɑ ɵ ɨ o u/<i ü e ö ä a õ y o u>
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

With that many vowels and a scant list of consonants, I'm curious if you had any allophony in mind?
Image
Solarius
roman
roman
Posts: 1173
Joined: 30 Aug 2010 01:23

Re: Random phonology/phonemic inventory thread

Post by Solarius »

Thrice Xandvii wrote:With that many vowels and a scant list of consonants, I'm curious if you had any allophony in mind?
Well, there's a system of roundedness harmony but obviously that's not what y're asking.

I'm not super sure about what to do wrt: allophony there. I was thinking maybe some light frication on the rhotic approximant in stressed syllables and intervocalic voicing though.
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Random phonology/phonemic inventory thread

Post by Thrice Xandvii »

Porphyrogenitos wrote:Another phonology inspired by (but not adhering to) my 26-syllable challenge from a while back:

/m n/
/t k/
/ɸ s h/
/l/

/i a ə u/

(C)V syllables

Code: Select all

 i   a   ə(ː)  u
(si) ta  tɨ(ː) tu
 ki  ka  ka(ː) ku
 hi  ha  hə(ː) ɸu
 mi  ma  mə(ː) mu
 ni  na  nɨ(ː) nu
 si  ha  sɨ(ː) su
 li  la  lɨ(ː) lu
 hi  ha  hə(ː) hu

All consonants except /h/ are automatically geminated after stressed /ə/. (This is a real feature found in some Austronesian languages.)

/ə/ is [ɨ] after alveolars; it is [a] after /k/, but remains underlyingly distinct from /a/ due to gemination effects.

Historical /ti/ merged with /si/.

Historical /p/ has debuccalized to [h] except before /u/, where it became [ɸ]; however, debuccalized /p/ remains underlyingly distinct from historical /h/ because it surfaces as [ɸː] after stressed /ə/.
I didn't say anything before, but I think this setup is pretty nifty. The fact that some syllables are essentially the same except for the gemination effects (as well as them be related to schwa), is pretty cool. I'm a fan of the idea!

I still plan to remake/revive one of my languages again sometime, and an idea like this (or something related to consonant lengthening in stressed syllables as dormouse mentioned in another thread ages ago) might be where I go with it to help keep it distinct from my other languages and their phonology while still having some other phonations and a minimal inventory.
Image
Porphyrogenitos
sinic
sinic
Posts: 401
Joined: 21 Jul 2012 08:01
Location: Buffalo, NY

Re: Random phonology/phonemic inventory thread

Post by Porphyrogenitos »

Thrice Xandvii wrote:I didn't say anything before, but I think this setup is pretty nifty. The fact that some syllables are essentially the same except for the gemination effects (as well as them be related to schwa), is pretty cool. I'm a fan of the idea!

I still plan to remake/revive one of my languages again sometime, and an idea like this (or something related to consonant lengthening in stressed syllables as dormouse mentioned in another thread ages ago) might be where I go with it to help keep it distinct from my other languages and their phonology while still having some other phonations and a minimal inventory.
Thanks! One other thing I might add to it that I didn't specify before: I think that in this system, unstressed /a ə/ merge into [ə] (with no gemination effects, as it's an unstressed syllable), and unstressed /i u/ merge into [ɨ]. If I were to make a syllabic writing system for this phonology, I would probably use the a-series and the i-series for the respective unstressed vowels - or maybe I'd use the "underlying" values, if there was alternation due to stress shifts. I'm undecided on what the stress system would specifically look like, though.

In any case, feel free to use whatever you want of the idea!
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Random phonology/phonemic inventory thread

Post by Frislander »

/p t t͡ɕ k/
/t’ t͡ɕ’ k’ ʔ/
/ⁿb~m ⁿd~n/
/ɓ ɗ/
/s h/
/s’/
/ɾ/

/z̩ v̩/
/ɪ ʊ/
/a/
/aɪ̯ aʊ̯/

Each vowel phoneme may also appear nasalised. The prenasalised stops are realised as pure nasals before the nasal vowels.

Syllable structure is CV, that is, both the onset and coda are obligatory and the are no coda consonants nor consonant clusters.

There is a process of fortition/glottalisation which affect the initial consonant of all suffixes which begin with a certain subset of consonants.

Code: Select all

Base Mutated
p  | ʔ
t  | t’
t͡ɕ|  t͡ɕ’
k  | k’
ⁿb | ɓ
ⁿd | ɗ
s  | s’
This process is lexically conditioned, that is, there is not way to predict which particular lexemes with cause the mutation based on their phonetic forms alone.

There is a pitch-accent system, whereby an underlying downstep may be present at any syllable boundary. If the downstep is word-initial, then the whole word takes a low-falling tone, otherwise there is a high/rising tone which drops down to a low level immediately following the downstep until the end of the word.

----

/p t t͡s k ʔ/
/s h/
/m n~ŋ/
/w j/

The alveolar nasal is realsed as velar when it occurs word finally and before /k w/, and some speakers also show that realisation intervocalically as well.

/i ɨ u/
/ɛ a ɔ/

Each vowel may appear in oral, nasal, breathy or creaky. The oral-nasal contrast is levelled before nasal coda consonants.

Syllable structure is C(G)V(C), where G is an optional glide which may follow another non-glide consonant and a coda C may be any non-glottal consonant. In addition the sequences *ji and *ij are reduced to /i/ and the sequences *wu and *uw reduced to /u/.

Primary stress is on the first syllable of the root, with secondary stresses being applied to even numbered syllables counting outwards from it in both directions. There is a limited height-harmony system whereby unstressed syllables assimilate to the height of an adjacent stressed syllable: This is the preceding syllable if the vowel in question is after the primary stress and the following one if before.
quixoticxenoglot
hieroglyphic
hieroglyphic
Posts: 49
Joined: 29 Nov 2014 05:51
Location: Milwaukee, USA

Re: Random phonology/phonemic inventory thread

Post by quixoticxenoglot »

It's been a really long time! But hi, I'm back, if anyone remembers me (and I don't blame you if you don't).

Here's a wee little unnamed something that's mostly serving as an exercise in morphology right now:

/m n/ m n
/p t ts tʃ k/ p t c ç k
/f s ʃ/ f s ş
/w l j/ w~u l y~i

/i ə a u/ i e~o a u

Some allophony:
V / ʔV / #_
n / ŋ / _{w k}
n / m / {m p}
ə / e / {j i [e]}C(l)_
ə / o / {w u [o]}C(l)_
ə / e / j_
ə / o / w_
{aw əw} / aʊ̯ / _
{aj əj} / ɑɪ̯ / _
a / ä / _

Stress is nonphonemic, but always falls word-initially, and causes a bit of further vowel allophony, where /i ə u/ are [i ɐ u] when stressed and [ɪ ə ʊ] otherwise; /a/ [e o] are unaffected.

Syllables are (C)(C)V(C), where:
- vowel hiatus is prohibited;
- allowed initial clusters are /ml mj nw nj pl pj tw tj tsw tʃw kw kl kj fl fj sw sl ʃw ʃl/;
- clusters of */mw nl pw tl tsl tsj tʃl tʃj fw sj ʃj/, including across syllable boundaries, are strictly prohibited;
- */wVw lVl jVj/ are prohibited, as well as */ji ij wu uw/;
- stops and affricates may not occur in the coda;
- consonant gemination is prohibited.

Morphophonological processes include:
- alternation of /u a i/ with /w l j/, and of /w l j/ with /f s ʃ/;
- metathesis of */fn sn ʃn fm sm ʃm/ to /nf ns nʃ mf ms mʃ/;
- reduction of geminated consonants to single ones;
- assimilation of */sʃ stʃ ʃs ʃts/ to /s ʃ ʃ s/.

And finally, here's a sample:

And on the pedestal, these words appear: My name is Ozymandias, king of kings; look upon my works, ye mighty, and despair!

Lu tausaye faukluşo mun çeşe: şelef çel Wosimantyas mla mlaşe; wo pyen nacu tauspyen kauloti çel lu cunfaipyen!
/lu tawsajə fawkluʃə mun tʃəʃə ʃələf tʃəl wəsimantjas mla mlaʃə wə pjən natsu tawspjən kawləti tʃəl lu tsunfajpjən/
[lʊ taʊ̯sɑje faʊ̯klʊʃo mun tʃɐʃə ʃɐləf tʃəl wosɪmantjas mla mlaʃə wo pjen natsʊ taʊ̯spjen kaʊ̯lotɪ tʃəl lʊ tsuɱfɑɪ̯pjen]

Code: Select all

lu  taus-ai-e   fauklu-şe    mun  çe-şe   şelef çe-l    Wosimantyas mla  mla-şe   wo  pyen nacu   taus-pyen kaulo-ti    çe-l    lu  cuf-na-ai-e-pyen
and see-TRA-PAS pedestal-LOC word 1SG-LOC name  1SG-POS Ozymandias  king king-LOC VOC 2PL  strong see-2PL   product-LAT 1SG-POS and be_sad-AUG-TRA-2PL
"And words by me on the pedestal become seen: my name is Ozymandias, king over kings; o you strong ones, see my products and become depressed!"
Locked