Random ideas: Morphosyntax

[TBPO]

Spoiler:

I fixed the case system:
Ergative - Initiator
Nominative - What undergoes changes
Dative - Goal (physical/abstract) or delimiter
Oblique - Background or context
Instrumental - Method (physical/abstract)

man-ERG run hour-PL-DAT two-ADJ
Man runs (with own will) for 2 hours.

man-NOM run
Man runs (for unknown reason)

wolf-ERG chase rabbit-DAT
The wolf chase the rabbit.

man-ERG go shop-DAT run-INS
Man goes to shop while running.

3SG-ERG walk near front-OBL shop-ADJ
He/she/it walks in the front of the shop.

knife-NOM fall table-OBL floor-DAT
Knife fell of the table onto the floor.

[...]

1) This case system is inspired by Okuna and nominative is often used as subject, so it's still nominative.

The absolutive, too, is often used as the subject.

Your definition of the nominative as "what undergoes changes" aligns almost perfectly with the absolutive.

The way I would analyze your system would be an ergative-absolutive, probably split-S, system that allows the usage of the dative in place of the accusative in a transitive clause, depending on semantic nuances of the verb.

Thanks, I fixed it.

[Creyeditor]

It's an incipient dative-genitive merger.

Well, could it be stable?

[Knox Adjacent]

Sure, in the meantime it just groans aloud that 1st step down a sudsy slope.

[Omzinesý]

Years ago when I knew basically nothing about linguistics, I had an idea of coding a feature, say aspect, with turning diphthongs around.

kaum 'to kill' vs. kuam 'to be killing'
tien 'to hit' vs. tein 'to be hitting'

I have no idea how such a system could develop.

[VaptuantaDoi]

Years ago when I knew basically nothing about linguistics, I had an idea of coding a feature, say aspect, with turning diphthongs around.

kaum 'to kill' vs. kuam 'to be killing'
tien 'to hit' vs. tein 'to be hitting'

I have no idea how such a system could develop.

Here's my proposal

*kumú ~ kúmu
*kəmú ~ kúmə
*kamú ~ kúma
*kámu ~ kúma
*káu̯mu ~ kúa̯ma
kaum ~ kuam

[Omzinesý]

Years ago when I knew basically nothing about linguistics, I had an idea of coding a feature, say aspect, with turning diphthongs around.

kaum 'to kill' vs. kuam 'to be killing'
tien 'to hit' vs. tein 'to be hitting'

I have no idea how such a system could develop.

Here's my proposal

*kumú ~ kúmu
*kəmú ~ kúmə
*kamú ~ kúma
*kámu ~ kúma
*káu̯mu ~ kúa̯ma
kaum ~ kuam

One pair of sounds can accindentally be like that but as with all diphthongs. It would need much analogy I think.

[VaptuantaDoi]

One pair of sounds can accindentally be like that but as with all diphthongs. It would need much analogy I think.

Well, you can get as many pairs that way as vowels you started with. Maybe */i u e o æ ɑ/ → */ie~ei uo~ou ia~ai ua~au ue~eu io~oi/.

[Salmoneus]

The only difficult bit here is explaining why every verb in the entire language has a diphthong in it!

But something close could certainly be very plausible.

--------------

To think about the entire language, let's imagine four verb classes:

Class 1: simple-onset monosyllables: sat, kip
Class 2: glide-onset monosyllables: pyat, nwik
Class 3: bisyllables with sonorant medials: kawam, tayan, boron
Class 4: bisyllables with obstruant medials: makat, fupur

Sound changes (lenition) have made class 3 very common. The two vowels in a root must match due to harmony.

One verb form is just the root; another takes the suffix -a.

Hence:
sat / sata
kawam / kawama
etc
------------------

First, with stress on the penult, unstressed non-final vowels are dropped when this does not produce illegal clusters. Hence:
kawm / kwama
makat / makata [mk and kt are illegal]

Second, vowels lengthen in monosyllables, but not in existing diphthongs or in bisyllables (all for moraic reasons):
Hence:
sa:t / sata
pya:t /pyata
kawm / kwama
makat / makata

Third, long vowels break into diphthongs:
sayt / sata
pyayt / pyata
kiwp / kipa
kawm / kwama
makat / makata

Fourth, triphthongs simplify to diphthongs by losing initial glides:
payt / pyata

Fifth, stressed open vowels lengthen and diphthongise:
mawkat /makawta

Sixth, stress moves to the initial syllable, and complex codas aren't allowed in unstressed syllables, leading to metathesis:
mawkat / makwata

Eighth, final unstressed vowels are dropped.

This gives us three patterns. Classes 2 and 3 show metathesis:

payt, pyat
kawm, kwam

Class 4 shows metathesis in the other direction but maintaining the relationship between diphthong shape and grammatical function:

mawkat, makwat

Only Class 1 shows monophthongs in the secondary form:
sayt/ sat

Simple analogy then extends the metathesis to the secondary form of class 1 verbs, and glides develop into full vowels:
sait : siat
pait : piat
kaum : kuam
maukat : makuat

[Creyeditor]

Years ago when I knew basically nothing about linguistics, I had an idea of coding a feature, say aspect, with turning diphthongs around.

kaum 'to kill' vs. kuam 'to be killing'
tien 'to hit' vs. tein 'to be hitting'

I have no idea how such a system could develop.

The closest I can think of is Rotuman, which has two word forms CVCV and CVVC depending on the (at least not purely phonological) context. But then there's stuff like vowel coalescence, so I don't know what actually happens.

[Visions1]

You know, Kebreni uses this a whole lot.

[Arayaz]

You know, Kebreni uses this a whole lot.

Kebreni is possibly the least natural-seeming language I've seen published in writing. Mark Rosenfelder says that all of its wild stuff was derived diachronically, but I haven't been able to find said diachronics...

[Visions1]

In some posts they are there, just hiding somewhere on the end of the page (e.g. Lé), while Verdurian and its proto-Eastern relatives get a lot of discussion.
And to be honest, I'd give that title to Boghetan. Maybe.
Or some IRL lang.

[Omzinesý]

The only difficult bit here is explaining why every verb in the entire language has a diphthong in it!

Maybe it's enough that everyone verb that has a diphthong has such a process. In English the number of words with diphthongs is big.

[Visions1]

Wait till you see Lithuanian. I refuse to believe the language was not influenced by their blood alcohol content.

[Omzinesý]

Wait till you see Lithuanian. I refuse to believe the language was not influenced by their blood alcohol content.

What's so strange in it?

[Visions1]

A lot of the diphthongs are formed with a sort of slurred sound to them, based on where they start and end. There are lots of diphthongs in speech. Many singleton vowels in the script just get pronounced as diphthongs.

Of course, I could be wrong. I love how it sounds in any case.

[TBPO]

A conlang in which are not subject and object, but short cause, long cause, short effect and long effect. Speakers are speaking 2 or more simultaneouns expressions, each about other person.

For example:
1SG-2SG anger-personal.heresy hate-small.size satisfaction-wound prison-depression.
I hit him.

[Creyeditor]

I don't understand your example. Coild you explain it?

[TBPO]

I don't understand your example. Coild you explain it?

First word is the participant declaration, and it declarates which participant correstond to which morpheme in next 4 words. The second word is the short cause, with the first morpheme being the short cause of first participant, and second morpheme being of the second. The third word is the long casue, the fourth is the short effect and fifth is the long effect.

[Khemehekis]

This causation string reminds me of the conlang Malat (see http://www2.cmp.uea.ac.uk/~jrk/conlang.html ).