Project Limestone - Esseintial Scratchings

A forum for all topics related to constructed languages
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Sun 15 Mar 2015, 17:47

0. Project Limestone

Over Christmas I conceived of a simple phonology for a Híí language and posted it. Then I thought no more of it and it lay sleeping at the back of my head for about two months.

In the last few days, the phonology has caught my fancy again, and I've done some work on it. Hence this scratchpad.

This thread may or may not die pretty soon. One reason why I'm putting this here is so that I have a record in case my phone decides to die on me.

Hey Des, what's the language called?

At the time I toyed with several similar-sounding names, namely *so witty* Olıommtoohwa, Olıosstoohwa and Olıommstoohwa. However, I'm not quite satisfied with any of those, and the phonology has taken a somewhat different direction, so I'm going to give it one of those cool project names for now and refer to it as Project Limestone.

Current ideas for a name include Ḿḿkonísstowo Óyoḿḿnokísstowo Óyawaḿḿnokísstowo and variations thereof. Long names are cool.

In-World

Project Limestone is descended from the same protolanguage as Hííenununóóoþa, Proto-Híí. It therefore belongs to the Híí family of languages.

Project Limestone and Hííenununóóoþa are probably quite distantly related, but they share some areal features, such as an abundance of long resonants and some pretty serious consonant clusters.

I envisage Project Limestone to be spoken by the main enemies of the speakers of Hííenununóóoþa and their allies the Ałýýla. Otherwise, I know little about them.

Phonology coming up!
Last edited by DesEsseintes on Wed 18 Mar 2015, 18:13, edited 1 time in total.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Sun 15 Mar 2015, 17:53

1. Phonology

This is still pretty rudimentary, but maybe I'll edit stuff in if/as the lang grows.

The Inventory

This has recently been posted elsewhere, but here it is:

/m n/ m n
/p t k/ p t k
/s/ s
/l j w/ l y w
/a i o/ a ı o

The inventory is quite small, with 9 consonant phonemes, and 3 vowel phonemes.

Some notes:
- stops are typically neither aspirated nor voiced
- all consonants save /j w/ can occur geminate - geminate consonants are written double
- /l/ only occurs geminate intervocalically (historically intervocalic single /l/ lenited to /j w/)
- geminate /m n s l/ are syllabic next to another consonant or when accented
- /j w/ cannot cluster with other consonants, but may be permissible next to syllabic resonants
- short /m n/ assimilate to PoA of following stop but geminate (and syllabic) /mm nn/ do not, so sequences such as kmmt nnp are possible

The h-Question

In my original conception, /h/ was a part of the inventory but only occurred in coda position. I have removed it for now, but if I choose to reinstate it, it will devoice any preceding vowel or syllabic /m n l/.

The e-Question

I like the three-vowel /a i o/ system, but I have considered two possibilities:

a) introducing phonemic /e/ - this would allow for cool stuff like words ending in -ewe

b) have front vowel allophones of /a/ as follows:
a aa aı → [æ ɛː eɪ̯] e ee eı / _{y}ı


At the moment I'm leaning towards adopting neither, or perhaps b).

Accent

I'm thinking pitch-accent with primary accent - marked by high pitch - placed normally towards the end of the word, and every second or third syllable before that receiving secondary accent.

It's worth bearing in mind here that syllabic resonants can carry accent.

High pitch is marked with an acute accent. Long vowels, diphthongs and syllabic resonants have the accent marked on all graphemes involved - áá óí ńń śś, etc.

As ĺĺ is quite ugly, I'm considering using łł for the accented syllabic lateral approximant.

Sample Lexicon

Here are some semantically void word forms to give a feel for what I'm going for:

kıksííwıpńńa
kśśımıp
kśśımıopo
weyíllpotstoyííwa
akńńwanáppıyıp
oyolpeyıpśśkawa
ıpsısskóówona
yíkspıstsı → yıkspıstśśı
ńńtsawa
śśtnomo
ḿḿseyıwı
ḿḿkonísstowo
Ḿḿkonísstowo
Óyoḿḿnokísstowo
Óyawaḿḿnokísstowo
wopíppıyık
nıkśśwa
áátnn
sıwıssóókawa
okkıyílltsıwık


So What's Next?

I don't know. Hopefully something.

Tomorrow I'm gonna choose a colour and make the headings pretty...
Last edited by DesEsseintes on Wed 18 Mar 2015, 18:16, edited 1 time in total.
User avatar
kanejam
greek
greek
Posts: 871
Joined: Fri 07 Jun 2013, 06:50
Location: NZ

Re: Project Limestone - Esseintial Scratchings

Post by kanejam » Tue 17 Mar 2015, 08:47

Awesome! More! I always enjoy your threads [:)]

How is high pitch on a syllabic /s/ realised? Or is it voiced in the nucleus?

I find myself not quite satisfied with the pitch accent, maybe I just need to see more of the language. Also I like the language without a phonemic /e/, but maybe could rarely appear allophonically? Maybe lowering near /w/ so you still get ewe (which I agree is cool).
Spoiler: show
Found another diachronic source of syllabic resonants; apparently some Japanese dialects realise umai and uma as mmai and mma (with syllabic m).
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Tue 17 Mar 2015, 17:20

kanejam wrote:Awesome! More! I always enjoy your threads [:)]
[<3]
How is high pitch on a syllabic /s/ realised? Or is it voiced in the nucleus?
Syllabic /s/ is not voiced. I guess I'm just hissing at a slightly higher pitch and/or with greater intensity.
I find myself not quite satisfied with the pitch accent, maybe I just need to see more of the language.
Which aspect in particular fails to please? We would hate to disappoint! [:)]
Also I like the language without a phonemic /e/, but maybe could rarely appear allophonically? Maybe lowering near /w/ so you still get ewe (which I agree is cool).
I'm seriously considering a compromise whereby /e/ is marginally phonemic in that it contrasts with /a/ only word-finally, and furthermore word-final e will condition harmony across any vowel or resonant as follows:

a → e / _Ce#

I think that's delightfully idiosyncratic.

If I choose to use this, I will also have e occur allophonically word-internally as outlined above.
Found another diachronic source of syllabic resonants; apparently some Japanese dialects realise umai and uma as mmai and mma (with syllabic m).
Thanks! [;)]
wintiver
sinic
sinic
Posts: 220
Joined: Tue 09 Oct 2012, 02:37

Re: Project Limestone - Esseintial Scratchings

Post by wintiver » Wed 18 Mar 2015, 15:11

First and foremost, I love the initial geminates. I read a bit about Pohnpeian and how they have some initial geminates (my favorite of which is the word for "to vomit" which is <mmwus>.

Anyway, my only question is, if you only have /i/ why do you represent it as <ı>? Is this purely aesthetic? Looking over some of the word forms you presented I find myself looking for that overdot on at least a few forms. But alas, none found. I feel like <ı> would be good for /ɨ/ or /ɯ/, but seeing as though you have a very stripped down /i a o/ system I find it odd I guess.

And just a few random quick questions:
  • Are the stops never aspirated? Are they never voiced? I feel like initials or finals would be aspirated perhaps and medials would more likely be voiced, or maybe lenited?
  • Since your phonemic inventory is so small there must be a fair amount of allophony right? You did mention /m/ and /n/ and how they shift in POA unless they are geminate and are marked with pitch accent.
  • Does /o/ resist realization as /u/ all the time? If it's geminate and high pitch for instance? As in your word form "sıwıssóókawa". I'm just curious because that's pretty neat if it's strongly anti-/u/ or perhaps a better way to put it would be /u/-realization-resistant?
Anyway, as always, interesting stuff.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Wed 18 Mar 2015, 16:26

Thanks for the feedback, wintiver. [:)]
Anyway, my only question is, if you only have /i/ why do you represent it as <ı>? Is this purely aesthetic?
Yes, I avoid i in my languages purely for cosmetic reasons. As an added excuse, it makes for a more marked difference with í.
Are the stops never aspirated? Are they never voiced? I feel like initials or finals would be aspirated perhaps and medials would more likely be voiced, or maybe lenited?
For the moment at least, I'm resisting both aspiration and voicing, although I might have aspirated realisations of /p t k l/ as [pʰ tʰ kʰ ɬ] word-initially.

If I choose to go for /h/, that phoneme would cause devoicing of vowels and syllabic resonants, and aspirate stops before devoiced short vowels.
Does /o/ resist realization as /u/ all the time?
Yes. The prime candidate for [ɪ~ʏ~ɯ̯̽~ʊ] realisations would actually rather be /i/ if any.

However, I don't think I'll be spending too much time working out detailed allophony for this lang for now. First, I want to try and get the morphology off the ground and see how I can extrapolate elements from it for Hííenununóóoþa.
wintiver
sinic
sinic
Posts: 220
Joined: Tue 09 Oct 2012, 02:37

Re: Project Limestone - Esseintial Scratchings

Post by wintiver » Wed 18 Mar 2015, 17:04

I look forward to seeing what you have in store morphologically.

I hope it's going to be agglutinating. If the word forms you presented hint at anything I figure it'll be agglutination.
wintiver
sinic
sinic
Posts: 220
Joined: Tue 09 Oct 2012, 02:37

Re: Project Limestone - Esseintial Scratchings

Post by wintiver » Wed 18 Mar 2015, 17:07

DesEsseintes wrote:The prime candidate for [ɪ~ʏ~ɯ̯̽~ʊ] realisations would actually rather be /i/ if any.
Also really cool. I won't push you to come up with a reason or justification but just to throw it out there - I feel like the phoneme would be /ɨ/, but that's assuming that there's a more or less even distribution from [i~ɪ to u~ʊ], which come to think of it, is probably not the case.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Wed 18 Mar 2015, 18:18

wintiver wrote:I hope it's going to be agglutinating. If the word forms you presented hint at anything I figure it'll be agglutination.
Yep, agglutinating it is.
...pssht, it's also going to be polysynthetic, but don't tell anyone!
wintiver
sinic
sinic
Posts: 220
Joined: Tue 09 Oct 2012, 02:37

Re: Project Limestone - Esseintial Scratchings

Post by wintiver » Wed 18 Mar 2015, 18:21

Psh, I'd never leak a polysynthetic secret.

Or in your language... uh... ḿḿsawatosıpśśkaıyílltsıwa? Heh.
User avatar
thaen
roman
roman
Posts: 1170
Joined: Sat 04 Jun 2011, 21:01
Location: Plano

Re: Project Limestone - Esseintial Scratchings

Post by thaen » Wed 18 Mar 2015, 22:33

DesEsseintes wrote:
wintiver wrote:I hope it's going to be agglutinating. If the word forms you presented hint at anything I figure it'll be agglutination.
Yep, agglutinating it is.
...pssht, it's also going to be polysynthetic, but don't tell anyone!
I knew you couldn't resist the call of Algonquian polysynthesis! [:P]
:con: Nillahimma
:con: Øð!
:con: Coneylang

I am the Great Rabbit. Fear me, O Crabs!
Spoiler: show
ı θ ð ʃ ɲ ŋ ʔ ɛ ə ø ʑ ɕ ʷ ʲ ⁿ
User avatar
Lao Kou
korean
korean
Posts: 5471
Joined: Sun 25 Nov 2012, 10:39
Location: 蘇州/苏州

Re: Project Limestone - Esseintial Scratchings

Post by Lao Kou » Thu 19 Mar 2015, 02:28

DesEsseintes wrote:...pssht, it's also going to be polysynthetic, but don't tell anyone!
Knock me over with a feather. [;)]

My lips are sealed. [:x]
道可道,非常道
名可名,非常名
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Thu 19 Mar 2015, 18:18

Some Ideas

łłtse is bird. It ends in -e which is probably underlyingly -ay, although it may be -ə(y)*.
I'm considering a phoneme /ə/ which can: a) cause syllabic resonants; b) become -e word-finally; c) disappear without a trace.

I want the plural of this word to be łłtseyewe or lltséyewe (it depends on what kinds of accent patterns I go for).

So now I'm thinking the plurals of this class of nouns will be formed with -ewe if the last vowel in the root is e o, and -awa if it is a ı or a syllabic resonant.

śśtnomo young man → pl. śśtnomóewe
ókokss friend; ally → pl. ókokssawa or ókoksawa
ospsíwı (← *ospsííl) falcon → pl. ospsíílawa

I like what's going on in that last example.
User avatar
kanejam
greek
greek
Posts: 871
Joined: Fri 07 Jun 2013, 06:50
Location: NZ

Re: Project Limestone - Esseintial Scratchings

Post by kanejam » Fri 20 Mar 2015, 05:03

Firstly, I'm disappointed that Limestone doesn't have it's own title colour yet!
DesEsseintes wrote:łłtse is bird. It ends in -e which is probably underlyingly -ay, although it may be -ə(y)*.
Oo! I really like the e# coming from /aj#/. You could possibly have /aj/ [ej] non-word-finally as well. I have to say I quite like the /a i o/ inventory. If you were going to expand it, though, I think it would be neat to have a few of the phonemes in very near complementary distribution, but not quite. As an example, there are a few Sepik languages (I think) with an epenthetic [ɨ] but they have a handful of words with an [ɨ] in between two consonants which wouldn't make an illegal cluster and so can't be epenthetic. Another example would be semi-phonemic /m/ in Wichita which occurs only in the two words kammac to grind corn and camma:ci to hoe, to cultivate (according to wiki).

I'm not quite sure where I'm going with this, but basically a barely phonemic /e/ or an allophonic [e] would both be equally cool. Up to you, of course, and if you like the -awa/-ewe alternation then add it.

I'll wait until you settle the accent patterns until I comment. And I'm still not sure how viable tone on voiceless syllabic fricatives will be. It might devolve into a bit of a stress accent.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Fri 20 Mar 2015, 09:07

Firstly, I'm disappointed that Limestone doesn't have it's own title colour yet!

[:O] But it does!
And I'm still not sure how viable tone on voiceless syllabic fricatives will be. It might devolve into a bit of a stress accent.

Perhaps you are right, but I don't think having one sound behave oddly would necessarily bring the whole pitch-accent system crashing down.

Tentative way out: In the case of a word like śśtnomo śś-tno-mo, I guess an important indicator that the pitch accent is indeed on the first syllable would be low intonation on the last two syllables.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Sat 21 Mar 2015, 13:17

2. Some Additions to the Phonology

Let's look at the inventory again:

/m n/ m n
/p t k/ p t k
/s/ s
/l j w/ l y w
/a e i o/ a e ı o

Astute readers will have noticed that I have added /e/. However, it is only phonemic word-finally. All word-internal es are allophones of /a/. See the section on morphophonological processes below for more details.

Some Phonotactics

Although Project Limestone shows a preference for open syllables, it also permits some pretty spectacular consonant clusters.

The maximal syllable is sth along the lines of (C)(S)V(S)(C)(S) where S is any of /m n s l/ and C must be a stop if adjacent to an S. V can be a long or short vowel, or a syllabic resonant. The second C can be geminate if there is no final S.

As an example of an extreme cluster that can result from this syllable structure, there is psálkstsawa psálks-tsa-wa. However, such clusters are thankfully quite rare.

Geminate consonants frequently occur at morpheme boundaries.

Some Morphophonological Processes

Here are a few principles that affect how words end up looking. This is very much a WIP, and I hope I'll be able to add quite a lot more morfofo as I go along.

1. ıı oo → ıyı owo / _(C)#, _CC$

The long vowels ıı oo break up: a) word finally; b) before a word-final consonant; c) when they precede a complex syllable coda.

e.g. ópıı + Ø → ópıyı (cf. ópıı + pa → ópııpa)

2. ı o → y w / V_V

This is pretty straightforward.

e.g. ńńıo + ıp → ńńıwıp

3. a → e / _Ce#

Remember that phonemic /e/ only occurs word-finally. This rule states that if there is an /a/ in the penultimate syllable, it will shift to e.

sínna + me → sínneme

The common plural suffix -ewe is underlyingly /awe/.

4. a aa → e ee / _{ye yı}
aı → eı

(constraint: but sometimes not after /p k w/)

These rules show the conditions for allophonic e. These rules may not always apply though, especially when the /a/ is stressed and/or follows /k p w/. I'm not entirely sure how it's going to work.

/ajípptii/ → eyípptıyı - bee

Note also that rule 3 can feed rule 4, as follows:

łłtsay + ewe → łłtseyewe

5. l → w / {a o}_V, ı_ı

Non-geminate /l/ is not very popular inter-vocalically in Project Limestone and shifts to /w/ in most situations. It survives only after a high vowel and before a non-high vowel.

íl + ıssa → íwıssa (cf. íl + aksa → ílaksa)

I'm going to post this now. Then I'm going to try and assign meanings to some of the examples I've used.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Sun 29 Mar 2015, 14:29

More Plurals

While work on my verb post slowly progresses, I've slightly refined my plural formation patterns for Class I nouns. I've now decided that the last vowel of the stem decides whether the ending is -awa or -ewe.

Word-final vowels -a -ı -o do not affect harmony, and -a elides before the plural ending.

This results in quite a lot of variation in forming plurals, so I'm pretty pleased.

1. Nouns with last stem vowel /a i/ take the plural ending -awa:

nntsáí - porcupine
nntsáíkıawa

kńńkısa - trout
kńńkısawa

wíso
wísoawa

Nouns in /aji/ surface as -eyı but ı → Ø / y_V so we get:

kattsḿḿeyı - moose
kattsḿḿayawa

Nouns ending in /ii/ have this surface as -ıyı in the singular but this reverts to ıı in the plural:

níyı - son; (one's) child
nííawa

2. Nouns with last stem vowel /o/ take the plural ending -ewe:

óksı - boy
óksıewe

nííwo - hunter
nííwokıewe

kósko - dog
kóskoewe

Nouns ending in /oo/ have this surface as -owo in the singular but this reverts to -oo in the plural:

wawátstowo - spirit rock
wawatstóóewe

3. Words ending in -e form a class of their own, and form plurals by adding -yewe:

wáwıstse - woman
wáwıstseyewe

Further Complications?

I'm considering having nouns ending in -sı -tsı that lose this ending in the plural but have compensatory lengthening/gemination of a preceding stop, resonant or vowel. Tentative examples:

sómtsı - brother
sómmewe

íyıntsı
íyınnawa or íínnawa

tíksı
tíkkawa

I can't decide if I like these.
User avatar
DesEsseintes
cleardarkness
cleardarkness
Posts: 4370
Joined: Sun 31 Mar 2013, 12:16

Re: Project Limestone - Esseintial Scratchings

Post by DesEsseintes » Mon 30 Mar 2015, 17:07

Unresolved Conundra of Limestone Phonology

In this post I'm going to list some of the things I can't decide on regarding the phonology, in hope of clearing my thoughts and perhaps getting some feedback.

What's the Aesthetic?

As I explained in the OP, Project Limestone was almost entirely inspired by Blackfoot. However, as things have progressed, I feel there is a slightly African - by which I mean Bantu, I guess - feel to the language. This is something I've come to like.

Word-Final Stops?

I originally planned to allow words ending in stops -p -t -k but as of yet, I haven't come up with any forms using final stops.

I like the look of words such as kśśımıp and endings such as -yıp -wık, but having all words end in vowels is also appealing.

Stop clusters

Geminate stops are not only allowed but common. Limestone phonotactics also leads to stops coming into contact across syllable boundaries. I'm undecided whether to allow stops to cluster without restrictions or not.

The first solution would be simply to allow all combinations: pt pk tp tk kp kt.

Another solution would be to allow pt pk kp kt but have t assimilate to a following stop: tp tk → pp kk. The reason would be that tp tk are simply the most unpleasant clusters in my opinion. A variant here would be to have t → ts before another stop. (That's what happens in Hííenununóóoþa.)

The third option would be to disallow heterorganic clusters, with all stops assimilating to a following stop.

More on Allophonic 'e'

I have the following rules:
a aa → e ee / _{ye yı}
aı → eı

constraint: but sometimes not after /p k w/

I find myself breaking these rules already, especially when it comes to the diphthong. I think áí looks particularly delightful in this language, and so I think these rules might only apply to unstressed syllables, which would help to restrict the distribution of e even further, which I like.

Another thing I'm doubting is whether to allow aa → ee. ee feels out of place and I'm pretty sure I'll drop it.

I'm not even sure about allowing long /a/, as long /i/ and /o/ seem to be underlyingly /ij/ /ow/ respectively, leaving no justification or need for long /a/.

I hope this post isn't too boring, but I just wanted to put this here anyway. This is just a scratchpad after all.

Opinions and suggestions would be welcome and appreciated.
shimobaatar
darkness
darkness
Posts: 8684
Joined: Fri 12 Jul 2013, 22:09
Location: PA → IN

Re: Project Limestone - Esseintial Scratchings

Post by shimobaatar » Mon 30 Mar 2015, 19:01

DesEsseintes wrote:As I explained in the OP, Project Limestone was almost entirely inspired by Blackfoot. However, as things have progressed, I feel there is a slightly African - by which I mean Bantu, I guess - feel to the language. This is something I've come to like.
I know what you mean, but I can't exactly put my finger on what it is that makes the language look/feel Bantu-ish.
DesEsseintes wrote:I originally planned to allow words ending in stops -p -t -k but as of yet, I haven't come up with any forms using final stops.

I like the look of words such as kśśımıp and endings such as -yıp -wık, but having all words end in vowels is also appealing.
I'd say allow stops word-finally, but make them a lot less common than word-final vowels.
DesEsseintes wrote:The third option would be to disallow heterorganic clusters, with all stops assimilating to a following stop.
This option appeals to me the most.
DesEsseintes wrote:I find myself breaking these rules already, especially when it comes to the diphthong. I think áí looks particularly delightful in this language, and so I think these rules might only apply to unstressed syllables, which would help to restrict the distribution of e even further, which I like.

Another thing I'm doubting is whether to allow aa → ee. ee feels out of place and I'm pretty sure I'll drop it.

I'm not even sure about allowing long /a/, as long /i/ and /o/ seem to be underlyingly /ij/ /ow/ respectively, leaving no justification or need for long /a/.
I would restrict the rules to unstressed syllables and disallow ee. I would either recommend disallowing /aː/ aa or having it be realized as [e] e or something.
DesEsseintes wrote:Opinions and suggestions would be welcome and appreciated.
Hopefully some of what I've said helps.

Looking back, I don't think I've commented on this thread before. As always, it looks great, and it's always nice when we get to see more. I've just never had anything really constructive to say before now.
User avatar
cedh
metal
metal
Posts: 311
Joined: Wed 07 Sep 2011, 21:25
Location: Tübingen, Germany
Contact:

Re: Project Limestone - Esseintial Scratchings

Post by cedh » Mon 30 Mar 2015, 23:40

DesEsseintes wrote:I like the look of words such as kśśımıp and endings such as -yıp -wık, but having all words end in vowels is also appealing.
I would agree with shimobaatar: Allow final stops, but make them rare.
Limestone phonotactics also leads to stops coming into contact across syllable boundaries. I'm undecided whether to allow stops to cluster without restrictions or not.

The first solution would be simply to allow all combinations: pt pk tp tk kp kt.

Another solution would be to allow pt pk kp kt but have t assimilate to a following stop: tp tk → pp kk. The reason would be that tp tk are simply the most unpleasant clusters in my opinion. A variant here would be to have t → ts before another stop. (That's what happens in Hííenununóóoþa.)

The third option would be to disallow heterorganic clusters, with all stops assimilating to a following stop.
Of these, I also like the third option best. It would be a bit reminiscent of Inuktitut.
I have the following rules:
a aa → e ee / _{ye yı}
aı → eı

constraint: but sometimes not after /p k w/

I find myself breaking these rules already, especially when it comes to the diphthong. I think áí looks particularly delightful in this language, and so I think these rules might only apply to unstressed syllables, which would help to restrict the distribution of e even further, which I like.

Another thing I'm doubting is whether to allow aa → ee. ee feels out of place and I'm pretty sure I'll drop it.
What about aa → aı instead?
Post Reply