Nine Phoneme Conlang (with oodles of allophones)

[ZedSed]

Hey guys. I haven't posted in an embarrassingly long time. Sorry for that. Life has been like the mean cowboy making me dance as it shoots at my feet. It hasn't given me much time to get into my favorite pastime. So, carrying on...

I was looking at Linguifex's eight phoneme conlang (awesome btw!) and it reminded me of an idea I had ages ago. One of my old conlangs that was inspired by a good idea, but at that time I was too naïve about how to design a conlang, so basically it came out shite. Nonetheless, the original idea was still pretty good, I think.

So, here I am years later with something of a clue about how to go about this whole conlanging shenannigannery, and I think I'm going to resurrect this dusty old idea of mine and put it to some use!

Nine phonemes; five phonemic consonants and four vowels, resulting in roughly an additional thirty consonant allophones, the result of a surprisingly simple set of sound-change rules.

I've got the system written up and ready to go. Would anybody like to hear more?

[k1234567890y]

sounds nice (: I have ever had a conlang with only nine phonemes too, however, I have not worked on it for many years...

[ZedSed]

Awesome! Awesomely awesome! I was hoping someone would be curious.

To kick off, I will begin with the phonemic inventory:

h k l t ʔ i u a ã

Not much to look at yet, but this set has a lot of potential. You'll see what I mean soon.
The five consonants interact with surrounding vowels and may undergo one or more of three processes;
(1) Palatalization
(2) Labialization
(3) Nasalization/Resonation

This is true for the vowels too.
They tend to fuse, forming the allophones /e/ and /o/ (or /ẽ/ and /õ/ if they belong to a nasalized vowel cluster).

I have only just begun to work on the grammar and lexicon of this conlang. So far, I feel like it should be some kind of polysynthetic or agglutinative language, maybe something like Inuktitut. Not sure yet.

[ZedSed]

The full list of consonant allophones (not including syllable-codas) are as follows:

ʔiV → jV / where V ≠ i
ʔuV → wV / where V ≠ u
jṼ → nṼ
wṼ → mṼ
hiV → çV / where V ≠ i
huV → ɸV / where V ≠ u

(it's gonna get really long-winded if I keep repeating “where V ≠ i or u”. I'll abbreviate this with ***

hṼ → ɣṼ
çṼ → ʝṼ
ɸṼ → βṼ
kiV → cV / ***
kuV → kwV / ***
kṼ → gṼ
cṼ → ɟṼ
kwṼ → gwṼ
liV → ʎV / ***
luV → ʟwV / ***
lṼ → ŋṼ
ʎṼ → ɲṼ
ʟwṼ → ŋwṼ
tiV → tsV / ***
tuV → pV / ***
tṼ → dṼ
tsṼ → dzṼ
pṼ → bṼ

Edit: To explain the 3-star rule above, there is a rule I should have explained earlier called the “Vowel Solidity” rule. I will add a post later to explain it in-depth, but for now I think it's sufficient to say that long vowels (i.e. / ii, uu, aa, ãã /) "protect themselves" and resist being a part of these processes. They remain “solid”.

[k1234567890y]

looks nice (: maybe its descendants will have much more phonemes compared to this language (:

also, I assume that the voicing process caused by the nasal vowel was actually P > NP > B / _Ṽ in the past, where P is a voiceless obstruent, NP is a cluster of nasal+obstruent or a nasalized obstruent, and B is a voiced obstruent.

But these are my speculations, you can think of anything else.

[ZedSed]

looks nice (: maybe its descendants will have much more phonemes compared to this language (:

also, I assume that the voicing process caused by the nasal vowel was actually P > NP > B / _Ṽ in the past, where P is a voiceless obstruent, NP is a cluster of nasal+obstruent or a nasalized obstruent, and B is a voiced obstruent.

But these are my speculations, you can think of anything else.

I was thinking along the same lines. This makes for a fairly interesting foundation for a proto-lang, and I feel it could branch out in a lot of different directions.

There are some more allophones for consonants that are in syllable-coda position. The way I've defined a syllable coda is dead simple. “If it isn't immediately followed by a vowel, then it's a coda.”

Coda sound-change rules are a little different from the previous rules because they are progressive, not regressive. Codas change form to agree with a preceding vowel by being palatalized or labialized (not nasalized). Also, in some cases their coda-position allophone is slightly different when compared with their syllable-initial allophones. Lastly, unlike the regressive rules listed above, the progressive coda processes don't cause fusion with, and in fusing with, annihilation of the preceding vowel. The vowel remains unaffected.

Here they are:

ih → iç / iʝ
uh → uɸ / uβ
ik → ic / iɟ
uk → uk͡p / ug͡b
il → iʎ / iɲ
ul → uʟ / ŋ͡m
it → is / iz
ut → up / ub

[ZedSed]

The Vowel Solidity Rules

The rules are not so complex but they enforce themselves in a few different contexts.

To summarize the way they work, they are the small set of rules that prevents an / i / or / u / vowel from being subject to the regressive palatalization and labialization processes as they relate to syllable-initial consonants and vowels.

Edit: Note: They do not affect the palatalization or labialization of coda consonants. Coda consonants will always change to agree with the vowel that precedes them.

So, the rules are:

(1) If there is only one short / i / or / u / vowel comprising the entire given vowel cluster, the vowel will “remain solid”.
(2) If the / i / or / u / vowel is long (doubled), it will “remain solid”.
(3) If the vowels / a / and / ã / are together in one cluster, the former will change to agree with the latter. Or, to put it more simply, the right-most
/ a / or / ã / vowel dominates.
(4) The vowels / i / and / u / will agree with the rest of the vowel cluster in being either oral or nasal. These vowels are pronounced orally by default when neither / a / nor / ã / is present in the vowel cluster.

[ZedSed]

Orthography

I don't want to keep writing with IPA characters, because it's annoying jumping between keyboards, so I'm going to create a simpler orthography:

/ ʔ / [ ' ] / h / [ h ] / k / [ k ] / l / [ l ] / t / [ t ]
/ ɣ / [ gh ] / g / [ g ] / ŋ / [ ng ] / d / [ d ]
/ j / [ y ] / ç / [ ch ] / c / [ c ] / ʎ / [ lh ] / ts / [ ts ]
/ n / [ n ] / ʝ / [ jh ] / ɟ / [ j ] / ɲ / [ nh ] / dz / [ dz ]
/ w / [ w ] / ɸ / [ f ] / kw / [ kw ] / ʟw / [ lw ] / p / [ p ]
/ m / [ m ] / β / [ v ] / gw / [ gw ] / ŋw / [ ngw ] / b / [ b ]

/ kp / [ x ] / gb / [ q ] / ʟ / [ ll ] / ŋm / [ ñ ] / s / [ s ] / z / [ z ]

Not beautiful, but not too bad. It'll do for now atleast.

[Man in Space]

I was looking at Linguifex's eight phoneme conlang (awesome btw!) and it reminded me of an idea I had ages ago.

I am quite honored.

I'm extremely impressed by your allophonic rules.

[ZedSed]

I'm extremely impressed by your allophonic rules.

Thanks! I was inspired at first by the way English words are forced to fit into the limited phoneme set of Hawaiian by its speakers. Put together with a few other languages (Caddo, Blackfoot, Vietnamese, etc..) and their sound-change processes, I arrived at this weird mix of rules.

It's not natural having all of these rules together like this, of course, but I think almost every rule I've given has some natural counterpart somewhere in the world.