The Miniature Conlangs Thread

[qiihoskeh]

Here's yet another idea. This is probably not very clear ....

For each argument of the verb, the verb's argument structure class specifies the possible genders as well as the semantic role. There are 4 genders: animate, inanimate, location, and cause. Each noun has an implicit gender (excluding cause); if it doesn't match the required gender for the argument, it takes a conversion clitic (=A, =I, =L, and =C).

The verb appears last in the clause and takes a suffix specifying the syntactical function of the clause (such as -H main, -D attributive, etc.). Arguments can mostly appear in any order. Relative clauses are internally headed with the relativized noun taking the R- determiner prefix and the verb taking the -R function suffix. In a participial clause, the relativized noun appears first; it doesn't take the R- prefix and the verb takes the -D suffix. Any conversion clitic on the noun is relevent to the clause immediately containing the noun. The clitic appears on the participial or relative clause verb when it's relevent to the containing clause.

(1) cat small-H. "The cat is small."
(2) cat small-D hungry-H. "The small cat is hungry."
(3) cat man=L with-H. "The cat is with the man."
(4) girl man=L cat with-D=I see-H. "The girl sees the man the cat is with."

If the number of arguments is less than (or greater than) the normal valence of the verb, the verb takes a prefix specifying the number (T- trivalent, B- bivalent, U- univalent, and M- impersonal).

(5) boy=C soup B-spicy-H. "The boy has made the soup spicy."
(6) book U-give-D old-H. "The books being given away are old."

[zyxw59]

In my conlang, the singular and plural merged in nouns, leaving the dual to take over as the plural. But in verbs and pronouns, the singular and plural remained distinct, and the dual instead took on a meaning of conflict or contrast.
e.g.

Code: Select all

ətə̄       rað       iš          ətə́ji
ətə̄       ra-ð      is-j        ətə́j-i
PN.1p.DIR go-PRT.1p house-DIR.p PN.1p.GEN-DIR.p
We went to our (shared) house

vs.

Code: Select all

ī     ráði          iš          ýji
ī     ra-ði         is-j        ýj-i
PN.1d.DIR go-PRT.1d house-DIR.p PN.1d.GEN-DIR.p
We each went to our own house

(DIR is direct case- the result of a merger of nominative and accusative)

Another thing that happened was that all stative verbs in the proto-lang (a group that included adjective-type words) function as adjectives in the descendant lang, regardless of whether they are actually adjective-y.
e.g.

Code: Select all

un        ətɒllá     æňs
un-0      ətɒllá-0   æňs-0
PN.1s-DIR know-DIR.s live-DIR.s
I know I'm alive

One very common stative verb was rah (now the 'adjective' ra), meaning 'to go' as a dynamic verb, but used to form both perfect and future tenses as a stative verb. When used with dynamic verbs, it uses a relative clause (although the verb is usually in the infinitive if the subject doesn't change) introduced by either zɛ (that.GEN.s) for the perfect, or zɔr (that.LOC.s) for the future. But when used with stative verbs/adjectives, the adjective is just put in the genitive or locative case, respectively:

Code: Select all

un        ra       zɔr          ɛðmá
un-0      ra-0     zɔr          ɛðmá
PN.1s-DIR go-DIR.s that.I.LOC.s eat
I will eat

un        ra       zɛ           ɛðmá
un-0      ra-0     zɛ           ɛðmá
PN.1s-DIR go-DIR.s that.I.GEN.s eat
I have eaten

un        ra       rɛ́zur
un-0      ra-0     rɛz-ur
PN.1s-DIR go-DIR.s heal-LOC.s
I will be healthy

un        ra       rɛž
un-0      ra-0     rɛz-j
PN.1s-DIR go-DIR.s heal-GEN.s
I was healthy

One more thing that I like is that the way of negating a verb comes from the construction 'not … of all', where 'of all' is 'all' in the genitive. The 'not' part has been dropped, so 'ðoľ', the genitive of 'ðol', 'all', now means 'not'.

[gach]

Here's a simple case system I'm working for a language with polypersonal agreement. The system consists of three cases: the basic unmarked form and two suffixially marked forms, the oblique and the local. The unmarked case is the most common form and the core arguments of verbs for example are left in this form as their roles are already coded on the verb. The oblique case functions as a general adverbial marker with uses ranging from instrumental and local adverbials to deriving more lexically set manner and time adverbs. The local case is a more specific marker for local adverbials but gets used for all modes of location and movement (at/to/from/via).

There are two overlaps in the system both involving the oblique case. Firstly both the locative and lative meanings ("at" and "to") can be coded both by the oblique case as well as the local case and the oblique coding is especially common for lative adverbials. The choice between the two possibilities is lexically determined by the associated verb so you get contrasting coding strategies like

come-PST-SG3 village-OBL
"He came to the village."

enter-PST-SG3 house-LOC
"He entered the house."

while the opposite choices for case marking are never allowed.

Secondly the oblique case can be used to mark underspecified or unexpected objects in cases where the agreement patterns on the verb aren't enough to fully specify the roles of the arguments. A sentence like "The child saw the stranger." could be constructed as

child stranger see-PST-SG3.S-SG3.O

This is a correctly formed sentence but it has actually two possible and equally correct readings: "The child saw the stranger." and "The stranger was the child." To make clear which of the participants is to be understood as the object it can be marked with the oblique case as

child stranger-OBL see-PST-SG3.S-SG3.O

If there is difference in the definiteness of the subject and object, this will differentiate between the subject and the object. Subjects are much more commonly definite than objects so between a definite and an indefinite argument the indefinite one is interpreted as the object by default. If we've already been talking about the child but are introducing the stranger as new information, the sentence

child stranger see-PST-SG3.S-SG3.O

which lacks any case marking can only be interpreted as "The child saw a stranger." On the other hand it's possible that we've been talking about the stranger and are introducing the child into the discourse as the subject of the sentence. Now we have a case where the definiteness hierarchy is contrary to the intended interpretation of the argument roles. To get the reading "A child saw the stranger." we have to mark the unexpected definite object with the oblique case as

child stranger-OBL see-PST-SG3.S-SG3.O

Using animacy hierarchies to differentiate between subjects and objects left underspecified by the agreement on the verb should also work reducing the need to employ oblique marking on objects.

[Click]

I am working on a descendant of one conlang of mine, and several interesting ideas popped into my head while I was being outside today.

Firstly, the language loses the ancestral polypersonal agreement system and replaces it with a new one which marks subject person and transitivity, so there are two sets of personal endings for each tense now – one used with transitive verbs and another one used with intransitive ones.
The transitive set originates from ancestral person affixes marking the subject acting on a third-person object¹ and the intransitive set likewise originates from ancestor’s person affixes marking only the subject that narrowed to marking intransitive verbs.

The intransitive set later extends to nouns where it marks possession, replacing the ancestral system of distinct possessive prefixes.

Plausible?

• ¹ The language doesn’t distinguish between singular and plural with third person arguments.

[GreenBowTie]

quick conjuring conlang i threw together for a story where some characters summon demons:

phonology
p b t d k g <p b t d k g>
ɸ β θ s z x ɣ <f v th s z kh gh>
m n ŋ <m n ng>
l r j w <l r j w>
a e i o u y <a e i o u y>

syllable structure: (C)(C)V(L)(C) [L = l, r, j, w, m, n, ng]

accent on penultimate syllable

nouns, pronouns, and adjectives
inflectional suffixes (sg, pl)
nominative: -el, -ul
accusative: -a, -o
dative: -er, -ur
genitive: -it, -yt
vocative: -e, -u

these inflections are found on all nouns and pronouns, and on predicative adjectives (attributive adjectives do not decline)

pronoun roots:
1st person: kor-*
2nd person: vus-
3rd person animate: tir-*
3rd person inanimate: dur-*
interrogative animate: thir-*
interrogative inanimate: zyr-*
relative animate: sal-
relative inanimate: zol-
* the r becomes l in the dative

nouns can also have suffixes for articles and determiners; these follow the case endings (the first form in each set is used after a vowel; the second is used after a consonant)
indefinite article: -wa, -a
definite article: -k, -ak
this (animate): -f, -af
this (inanimate): -v, -av
that (animate): -m, -em
that (inanimate): -n, -en

animate nouns are living beings (people, animals, gods, spirits), and certain natural forces that move on their own (water and other liquids, fire, wind, etc.). the heart is also listed as animate (because it beats on its own; other organs are inanimate), as are corpses (because they were once alive). plants are inanimate, even if they move.

verbs
verbs conjugate only for subject agreement, not for tense. verb stems usually end in vowels
(sg, pl)
1st person: -ra, -ro
2nd person: -se, -su
3rd animate: -ri, -ry
3rd inanimate: -re, -ru

the language is pro-drop. if the subject is present rather than deleted, the verb is not conjugated and has no suffix.

imperative forms:
1st person (“let’s”): -laro
2nd singular: -si
2nd plural: -sy
imperative suffixes are never deleted, even when the subject is present

tense may optionally be indicated by a particle (basically a special form of adverb):

past: fer
present: uth
future: vyk

syntax
the language’s basic structure is SOV. datives generally precede accusatives. adjectives and adverbs precede their referents, and the verb tense particles, if any, come at the beginning of the verb phrase

relative pronouns appear at the beginning of the relative clause and are declined. the rest of the clause continues. following the verb, the relative pronoun appears again, this time not declined.

the language is prepositional, not postpositional. objects of most prepositions are accusative. the genitive can function as an adjective (though it does not take any additional case endings).

the past participle is formed with the suffix -myn, turning a verb into an adjective (which then may take case endings if appropriate).

vocabulary
and (conj): me
be (v): sa
bind (v): tse
creature, living being (n anim): krin
deep (adj): ghur
freeze (v): stu
give (v): fo
heart (n anim): wong [ŋ, not ŋg]
life (n anim): kert
live (v): koru
offer (v): jur
offering (n, inan): jyrt
spirit (n, anim): klathan
take (v): vri
to, attached to, at (prep): glo
together (adv): sry
want (v): baw
will (n inan): bon

Klathanu ghuritak, jyrtav vrisy me koryt wongo sry tsesy!
Spirits of the deep, take this offering and bind our hearts together!

Klathanu ghuritak, stumyn krinuraf kertak fosy!
Spirits of the deep, give these frozen creatures life!

Klathanu ghuritak, krinytaf wongo glo koryt bona tsesy!
Spirits of the deep, bind these creatures' hearts to our will!

turns out i made a couple mistakes when i put the sentences together for the story lol. they're corrected here though

[Chagen]

Oh shit I'm making a poly-lang and I didn't even intend to.

Only one sentence:

ūqtaŋt kçiððil qagat
bear javelin-DIMUN hit-3S.AGT-3SP-ANIMAL.CLASS
The bullet hit the bear

Here both the 3S markers for object and subject are null, -at being a suffix indicating that an animal and thus the bear is the focus of the utterance.

Of note is that the word for "bullet" is the word for "javelin", kçid, with the dimunitive suffix -[W]zVL, applied. The [W] part means to lenite the /d/ into /ð/, the /z/ assimilates to create a geminate, and then the root word's vowel is inserted in the V slot. The word for bear is literally "stealing-honey" (<waq> being the root for "steal", which is then weakened into <ūq>_, and the incorporated word for "honey", <taŋ> is suffixed. The final /t/ indicates that it's an animal (and yes it is related to the animal class verbal inflection -at). So it's literally "the animal that steals honey" Came up with that for...no reason whatsoever, I just made it in literally 10 seconds.

The language has a kind of wierd agreement where verbs agree with whatever the speaker and the verb deem to be the most "important" argument. Here the verb is marked for the bear as focus because what you hit is more important than what you hit it with. If you for whatever reason wanted to stress that the bullet as opposed to anything else hit the bear, you would mark it with the suffix -uŋx indicating that the focus is a weapon. So it's basically focus-marking on the verb.

If you wanted to mark the fact that the bullet hit the bear as opposed to missing it, then you would use the suffix -irq, which is to mark the actual action as the focus.

I have zero idea on what I could do to expand this though.