The Language of Three (inter alia)

[Davush]

This thread is for my random experimental/philosophical/artistic ideas which aren't exactly naturalistic, but might be fun to try out nonetheless.

—

A recent idea I've had is one where the language is characterised by its features (mostly) occuring in sets of three – in the phonology, grammar, lexicon – hence its name, The Language of Three. It may be related to, or descend from, the Language of Five.


Phonology

Consonants
/p t k/
/pr tr kr/
/ps ts ks/
/m n (ɲ)/
/s h (ç)/
/r l (ʎ/)
(/w j/)

The plosive + /r s/ series can be treated as single phonemes for the purposes of analysis rather than clusters, as they are the only allowed onset clusters and function as single consonants for stress and other purposes. The plosive + /s/ series historically derive from *pj *tj *kj. (This also allows a nice 3x3 system).

Also, *mj *nj>, /ɲ/ *rj *lj > /ʎ/, *sj *hj > /ç/.

The maximum syllable is CVC
Only three consonants appear in coda:
/n s r/

/w j/ are better analysed as non-syllabic /i u/

Vowels
A simple three vowel system:
/a i u/

With the diphthongs:
/ai̯ au̯ i̯a i̯u u̯a u̯i/

/ai au/ are generally realized /e o/ in all but the most formal/archaising speech.

Plosive + /ia iu/ > /Psa Psu/

The Vowel Cycle

Vowels undergo shifts according to the following cycle. The cycle begins from the inherent vowel of the root/stem.

a > i > u > (> a)

(I'm not sure what exactly the cycle will be used for yet, but let's say Irrealis - Imperfective - Perfective for now).

Additionally, the inherent vowel is usually predictable from the consonants. Roots containing a palatal (which includes /ps ts ks/ have i, those with a labial have u, all else have a.

So:
*san- : san-, sin-, sun-
*kits- : kits-, kuts-, kats-
*tum- : tum-, tam-, tim-

On top of that, the thematic suffix vowel also undergoes the cycle, with the starting vowel determined from the stem that has already gone through the cycle:

san-i-, sin-u, sun-a
kits-u-, kuts-i-, kats-u
etc.

The Language of Five

This will have a 5x5 system in the plosives at least:

/p pʰ p' b mb/
/t tʰ t' d nd/
/c cʰ c' ɟ nɟ/
/k kʰ k' g ng/
/kʷ kʷʰ kʷ' gʷ ngʷ/

…

It's kind of fun to do something that's not aiming at/emulating strict naturalism for once!

[Davush]

I'll call this language Áiuaiu /ˈeβeu/, which I think is befitting the 'theme of three'.

Verbs
As mentioned, verbs are characterised by their stem vowel changing. I actually quite like the tripartite Irrealis - Imperfective - Perfective thing, so I'll keep that for now. However, I don't like the inherent root/stem vowel, so verbs are now basically bi/triconsonantal roots.

Irrealis: -a-
Imperfective: -i-
Perfective: -u-

Before the person endings are added, a thematic vowel is inserted. This is determined according to the 'vowel cycle' (a > i > u), so:

Irrealis/Subjunctive: -a-...-i- (tami-)
Imperfective: -i-...-u- (timu-)
Perfective: -u-...-a- (tuma-)

Plural subjects will probably have a different thematic vowel or something else going on (maybe plural subjects are prefixes...?).

The person markers are:

1 -n
2 -s
3 -0 (or -r)

Additionally, augments can be added to the imperfective/perfective forms for further nuance. The irrealis does not accept augments. These are the same vowels corresponding the Irrealis - Imperfective - Perfective stem vowels, and largely carry the same/similar function.

á- hypothetical, conditional
ú- past perfective, anterior
í- present progressive, continuative

Examples using T-M 'to eat'.

tímun - I eat, I am eating
átimun - I might be eating
útimun - I was eating, I have been eating
ítimun - I am (still) eating

túman - I ate
átuman - if I had eaten
útuman - I had eaten
ítuman - I had been eating

The verbs to be and to have, which can underlyingly be noted as *-N and *-H (i.e., they are mono consonantal):

Irrealis: ánin, ánis, áni(r) // áhin, áhis, áhi(r)
Imperfective: ínun, ínus, ínu(r) // íhun, íhus, íhu(r)
Perfective: únan, únas, úna(r) // úhan, úhas, úha(r)

There will probably be more verb shenanigans (I'm thinking a three-leve honorific system, maybe).

Anyway...

A hakáina ta ínu (áni?) un Áiuaiu.
/a haˈcena ta ˈinu (ˈani) un ˈeβeu/
This language is (might/will be?) Áiuaiu.

[Davush]

As usual, I was adding things and taking things out, not liking any of the results. The word /vka/ [fka] popped into my mind, which I really liked, so i thought perhaps this comes from earlier *uká, with unstressed initial *u > /v/. Then I thought why not have unstressed initial *i > /s/, in parallel with the *pj *tj *kj > /ps ts ks/ series.

So I thought instead of having the 'contemporary' language have a semi-phonemic orthography reflecting the surface pronunciation, why not have the orthography actually be very archaic! This could be fun, as it could function as a kind of pan-dialectal form.

Additionally, the contemporary pronunciations seem quite different from what the orthography would indicate, due to a few simple but pervasive sound changes.

The 'archaic' language had:

/p t k/ <p t k>
/ᵐp ⁿt ᵑk/ <np nt nk>
/pɾ tɾ kɾ/ <pr tr kr>
/s x/ <s x>
/m n/ <m n>
/ɾ l/ <r l>
(/w j/) <i u>

/a i u/ <a i u>
/ai au/ <ai au>

Syllable structure is CGV, where G is non-syllabic /i u/. Unsure about coda consonants.

The following sound changes occurred:

I. Voicing of prenasalised consonants:

/ᵐp ⁿt ᵑk/ > /b d g/

II. Palatalization before /i j/, probably via /ç ʝ/. Unstressed initial /i/ > /s~z/.

/p t k/ > /pˢ tˢ kˢ/
/ᵐp ⁿt ᵑk/ > /bᶻ dᶻ gᶻ/
/s x/ > /ç/
/m n/ > /ɲ/
/r l/ > /z ʎ/

III. Labialization before /u w/, leading to velarization. Unstressed initial /u/ > /f~v/.

/p b m/ > /pˠ bˠ mˠ/ (unstressed <pu npu> > /f v/)
/t d n/ > /tˠ dˠ nˠ/
/k g/ > /kf gv/
/s x/ > /sˠ f/ (unstressed /su/ > /hu/ > /f/)
/r l/ > /ɣ ɫ/

IV. Unstressed /a/ > /ə/. Unstressed /i u/ tend to elide. Subsequent epenthesis with /ɨ/. Word-final/coda /mˠ lˠ ɲ ʎ/ > /u i/.

Some examples:

<kiáipra> /ˈkˢepɾə/
<iláuni> /zloɲ ~ zloi/
<kítanki> /ˈkˢitəgᶻ/
<málusi> /ˈmałɨç/
<nkautrúmia> /goˈtruɲə/
<npíkatu> /ˈbᶻikətˠ/
<trátinu> /ˈtɾatˢɨnˠ/
<ukuáni> /fkᶠaɲ ~ fkᶠai/
<untiá> /vdᶻa/
<ruáku> /ɣakᶠ/
<npuáliani> /ˈbˠaʎaɲ ~ ˈbˠaʎai/
<áiluri> /ˈełɨz/
<ikiáitaru> /ˈskˢetəɣ/
<tárimu> /ˈtazɨmˠ ~ ˈtazu/

Phew! Now the 'contemporary' phonology could be arranged like this:

/p t k b d g/
/pˢ tˢ kˢ bᶻ dᶻ gᶻ/
/pˠ tˠ kᶠ bˠ dˠ gᶹ/
/m mˠ n nˠ ɲ/
/f v s z ç h/
/r ɣ l lˠ ʎ/

/a e i o u/ + (ə ɨ)

[Arayaz]

I really like this.

[Davush]

I really like this.

Thanks!

Some grammatical notes
There won't be extensive morphology - mostly just agglutinative. The language will be quite head-initial.

Pronouns
Pronouns come in three forms: independent, prefixed, and suffixed.
The prefixed forms are used as subject markers on the verb; the suffixed forms as objects, possessive pronouns, and they also attach to prepositions and particles.

1sg: paxau / xa- / -p
2sg: sanai / na- / -t
3sg: itan / i- / -n

Verbs take (mostly) pre-verbal TAM clitics. I have one so far: naia- used for the future.

xa-naia-tami-n
1sg-FUT-eat-3sg
/xᵊne:taɲɨn/
I will eat it

na-naia-sani-p
2sg-FUT-see-1sg
/nᵊne:sanɨp/
You will see me

Another idea I've had that's been sticking around for a while is a generalized "adverbial" marker, that has many uses. On nouns, it can be used to indicate locatives or instrumentals, on adjectives it just forms plain adverbs, and on verbs it forms adverbial participles. Let's say it is an infixed /ʔ/ for now.

kuti - a stick
ku'ti - with a stick

xanaiatapakit ku'ti
xa-naia-tapak-it ku'ti
1sg-FUT-hit-2sg stick<ADV>
/xᵊne:tapəkst kuɂts/
I will hit you with a stick

xanaiasa'nit...
xa-naia-sá<'>n-it…
1sg-FUT-see<ADV>-2sg
/xᵊne:saɂɲt/
Since I'm going to see you…

I quite like the Vʔ sequences, so maybe vowels will come in three (of course) types:
Plain, Glottalised, Aspirated where glottalised and aspirated can be realised as /VʔV VhV/ sequences or as creaky/breathy voice.

[Creyeditor]

I am pretty sure that I had a similar idea once. Along with a binary language and a square/quaternary, etc.