Jengief

This forum is for constructed languages, both those invented by UniLang members and those already existing.

Moderators: Ashucky, Dormouse559

User avatar
Kshaard
Posts: 209
Joined: 2012-06-08, 17:19
Real Name: Nott Tellun
Gender: male
Location: Xan
Country: GB United Kingdom (United Kingdom)

Jengief

Postby Kshaard » 2013-05-25, 13:26

As it's the language I've been recently working on the most, I figured I should probably make Jengief a thread.

Jengief is an Ergative-Absolutive language with basic word order ABS ERG verb. It lacks all pre/postpositions, instead using 8 cases plus nouns to describe location.

BilabialLabiodentalDentalAlveolarPalato-alveolarAlveolo-palatalPalatalVelar
Nasalm--n---(ŋ)
Plosivep b--t d---k g
Affricate---ts dztʃ dʒtɕ dʑ--
Fricative-f vθ ðs z-ɕ ʑç jx ɣ
Trill---r r̝̊ r̝----


FrontNear-frontCentralBack
Closei y--u
Near-close-ɪ--
Mide ø-əo
Open-mid---ʌ
Near-openæ---
Open---ɑ

Plus 6 diphthongs - [ai̯ au̯ ei̯ eu̯̯ oi̯̯ ou̯] and two long vowels [iː uː]

Syllable structure is (C)V(C), and only single letters can end the syllable (so not cj, gj, kj, ng, nk, rs, rz, sj, xj or zj)

Orthography:
A [ɑ]; Ä [æ]; B [v] initial or medial and single; [b] final or double; C [ts]; Cj [tɕ]; D [ð] initial or medial and single; [d] final or double; E [e]; Ę [ə]; F [f]; G [ɣ] initial or medial and single; [g] final or double; Gj [j]; I [ɪ]; Ia/Ie/Io/Iu [iː]; J [dʒ]; K [x] initial or medial and single; [k] final or double; Kj [ç]; M [m]; N [n]; Ng [ŋg]; Nk [ŋk]; O [o]; Ö [ø]; Ǫ [ʌ]; P [f] initial or medial and single; [p] final or double; Q [tʃ]; R [r]; Rs [r̝̊]; Rz [r̝]; S [s]; Sj [ɕ]; T [θ] initial or medial and single; [t] final or double; U [u]; Ua/Ui/Uo [uː]; Ü [y]; V [v]; X [dz]; Xj [dʑ]; Y [i]; Z [z]; Zj [ʑ]

Next time I will explain verbs.
Project Zelia is not dead - it is only sleeping. Long live project permanently unnamed!

Xupaca - Gakxiirou - Tawʒessò - Ōrgadūvk - Maxyrteser

dan3697
Posts: 52
Joined: 2012-05-10, 0:05
Real Name: Dan
Gender: male
Location: Geneva
Country: US United States (United States)

Re: Jengief

Postby dan3697 » 2013-05-26, 15:41

Kshaard wrote:"It lacks all pre/postpositions, instead using 8 cases plus nouns to describe location."


Hm, this idea pleases Jabba. But on a more serious note, I am intrigued that someone has found great use for all of those obscure (and frankly, unloved) cases. I will attempt to observe your success from the shadows. :)
If Hangul is based on shapes of mouth positions, then the Koreans must be Martians. Do I like, get a show on the History channel now?
.اگاگاھاۋاگىيۇد تاچامىيۇداب

My Conlangs: Vérédæntsk, Tagawaha, Слувидроватскиј, Dareqar, Matoran (reconstruction of the Bionicle language), Krndmaz, Þęndeś, Sutspiki, Gwornligh

I've formulated a conjecture as to the nature of language change. Every language change began with one guy saying, "I'm gonna say this a certain way, let's see if it catches on..."

User avatar
Kshaard
Posts: 209
Joined: 2012-06-08, 17:19
Real Name: Nott Tellun
Gender: male
Location: Xan
Country: GB United Kingdom (United Kingdom)

Re: Jengief

Postby Kshaard » 2013-05-26, 19:29

I wouldn't call them obscure - for example to say something like 'on', you would have to use 'the top-LOC (thing)-GEN' - the eight cases are absolutive, ergative, genitive, dative, ablative, instrumental, vocative and locative.

So anyway, verbs.

For verbs with stem ending in -C
InfinitiveImperativePastPresentFuture
Positive-a-e-erzi-erzę-arza
Negative-as-es-ersi-ersę-arsa


For verbs with stem ending -V
InfinitiveImperativePastPresentFuture
Positive-a-e-vvi-vvę-avva
Negative-as-es-ffi-ffę-affa

NB - Some vowel combinations eg. <ue> are impossible and must change to a different vowel in order to be accepted eg. <ü>. Vowel mutation will be covered in greater detail soon.

For verbs with stem ending in -CC
InfinitiveImperativePastPresentFuture
Positive-a-e-exxji-exxję-axxja
Negative-as-es-eccji-eccję-accja

<Xxj> and <Ccj> are pronounced just the same as <Xj> and <Cj> respectively, but were geminated in the proto-language.

There is no passive voice - to make something passive you would use the ergative alone in a sentence, or if the sentence was transitive, you would have to use an active construction.

To make a verb into a noun (ie. a sort of gerund) you add -v to the infinitive, or -av to a negative infinitive and decline like a third declension noun (You can tell this language was inspired by Latin). Speaking of nouns, those will be coming next time.
Project Zelia is not dead - it is only sleeping. Long live project permanently unnamed!

Xupaca - Gakxiirou - Tawʒessò - Ōrgadūvk - Maxyrteser

User avatar
Kshaard
Posts: 209
Joined: 2012-06-08, 17:19
Real Name: Nott Tellun
Gender: male
Location: Xan
Country: GB United Kingdom (United Kingdom)

Re: Jengief

Postby Kshaard » 2013-05-30, 9:42

(Referring to the consonant table in 'phonology')
OK, [j] isn't a fricative, but I didn't want to go to the trouble of making another row "Approximants" just for that, so this is how the table shall stay.

1st declension of nouns:
AbsolutiveErgativeGenitiveDativeAblativeInstrumentalVocativeLocative
Singular-ab-is-ia-ov-a-y
Plural-ǫb-ois-oue-oia-ov-öa-oi


So, in the first declension the plural is formed by putting an <o> before the ending. Vowel mutations!

2nd declension (DLCAA indicates 'double last consonant and add'):
AbsolutiveErgativeGenitiveDativeAblativeInstrumentalVocativeLocative
Singular-ob-es-u-ieDLCAA -eDLCAA -öt-y
Plural-aub-äs-au-aieDLCAA -ä-auaDLCAA -auet-ai


In the second declension, plural is formed with an <a>.

3rd declension:
AbsolutiveErgativeGenitiveDativeAblativeInstrumentalVocativeLocative
Singular-av-an-ad-avi-avi-a-agy
Plural-ava-ana-adda-avia-avia-äa-a-agia


In the third declension, plural is formed by adding an <a> to the end of the word.

And, since they are practically the same thing as nouns, I figured I would also show the pronouns. They lack the last three cases but do have the others.
First person:
AbsolutiveErgativeGenitiveDativeAblative
Singularabqisüiaov
Pluralagabqasiueoiaava


Second person:
AbsolutiveErgativeGenitiveDativeAblative
Singularavvaanadakaviokav
Pluralavvaiainadaikavioikava


Third person:
AbsolutiveErgativeGenitiveDativeAblative
Singularacabaxanaxaacieaccav
Pluralacaibaxainaxaiacieiaccava

As before, doubled consonants are pronounced the same, but in the proto-language they were geminated.
Project Zelia is not dead - it is only sleeping. Long live project permanently unnamed!

Xupaca - Gakxiirou - Tawʒessò - Ōrgadūvk - Maxyrteser

User avatar
Kshaard
Posts: 209
Joined: 2012-06-08, 17:19
Real Name: Nott Tellun
Gender: male
Location: Xan
Country: GB United Kingdom (United Kingdom)

Re: Jengief

Postby Kshaard » 2013-06-08, 10:57

At long last, here is how the vowel mutations work:

+aeiouy
aaäaiauauai
eęeeięeuei
iiaieiioiuy
oǫöoioouoi
uuaüuiuouui
yiaieyioiuy


The first two vowels in a cluster of vowels take mutation priority, eg.
The 2nd declension vocative ending -öt can be expressed as 'oet'. To make it plural, you add an <a> to the beginning to make 'aoet'. But, <ao> is not a proper vowel combination in Jengief, so you must change it to <au>, thus the plural of -öt is -auet.

If this isn't clear enough, I will try to explain in greater detail.
Project Zelia is not dead - it is only sleeping. Long live project permanently unnamed!

Xupaca - Gakxiirou - Tawʒessò - Ōrgadūvk - Maxyrteser

User avatar
Kshaard
Posts: 209
Joined: 2012-06-08, 17:19
Real Name: Nott Tellun
Gender: male
Location: Xan
Country: GB United Kingdom (United Kingdom)

Re: Jengief

Postby Kshaard » 2018-12-03, 0:25

So I decided to revisit/completely overhaul this language a few months ago, combining it with some of the features of my very first conlang, and it is now occupying a decent chunk of my attention.

Also, this forum still exists apparently! Reddit and discord haven't quite eaten it up yet, I see.

Phonology is basically as it was in 2013; a couple of minor changes were made in the consonants - /t d/ are now dental rather than alveolar, /j/ is now /ʝ/, /x ɣ/ tend to be realised /h ʁ/ in quick speech, the /ŋk ŋg/ allophony is not really a thing any more. There now exists another long vowel /y:/ and a slew of new diphthongs - three beginning with /ø/ and four ending with /y/. The "vowel mutation" system has been totally revamped to actually make sense now, and the rules governing phonotactics have been changed up a bit.

Ergative-absolutive is still a thing, but the word order somehow got flipped around to something more like SOV. No doubt because I wasn't checking my notes carefully enough. Still got a whole pile of cases with varying usefulness, but I decided to make them a little bit less latinate: we now have absolutive, ergative, comitative, instrumental, locative, possessive, possessed, and comparative (coincidentally still eight of them. Maybe I just like that number?) Also there are now prepositions (because I'm lazy apparently) which tend to take either the locative or the comitative respectively, depending on whether the relationship they represent is spacio-temporal or not.

Rules governing grammatical number have shifted around quite a bit. The unmarked noun is by default paucal, and this can either be changed morphologically to singular and/or plural, or those senses can be inferred from the determiner which is used.

Pronouns are now regular because I'm lazy or something.

Every single word class aside from the conjunction particles (ötę, siz and co) can be made negative by undergoing a regular suffixation, and these negative senses "stack" instead of cancelling each other, e.g.,

Panna zjoir qy c uettö.
he-ABS-SG loudly less - sing-PST
Panna zjoiros qy c uettö.
he-ABS-SG loudly-NEG less - sing-PST

Both of these sentences mean effectively "He sang less loudly than before", even though the second one contains a negative in the form of zjoiros "quietly". One would use this construction if he were now singing what would be considered "quietly" rather than simply "less loudly than before".

Verbs now have extra tenses (distant past, distant future and indefinite), no infinitive form, and a two-level aspect system borrowed from my first conlang (inchoative vs. perfect vs. neither, then long vs. short). Also their morphology looks totally different now.

Determiners work quite differently now. The two most common types are, as before, indefinite (ra/rat/-/r) and definite (ęra/ęrat/ęr/ęr), but rather than an using an English-style system to work out which to use, this is based on whether or not the item has been brought up in conversation yet. Indefinite is used whenever a new item is introduced, and definite is used to reintroduce an old item. Neither is used if the item is currently being discussed. These two can also stack up with other determiners before the noun, e.g.,

Qüs e gjoti r obpec . . .
1SG-PSV this all INDEF cart
"All these carts of mine which we haven't mentioned yet . . ."

Plus there are now one or two homophones to spice things up a bit. Not quite sure why I was so afraid of them back when I was first into conlanging. Ah well. se vs. -se "four" vs. "exactly". it vs. it "through" vs. "as much as possible". ü vs. ü "on" vs. "that".

Plus there's that thing with word classes in Jengief being rigidly tied to semantics and the interesting cultural outcomes of that rigidity. But it's twenty past midnight and I've no idea why I just typed all this stuff.

Enjoy, ask questions if you like, and so on.
Project Zelia is not dead - it is only sleeping. Long live project permanently unnamed!

Xupaca - Gakxiirou - Tawʒessò - Ōrgadūvk - Maxyrteser


Return to “Conlangs”

Who is online

Users browsing this forum: No registered users and 1 guest