The Miniature Conlangs Thread

Hallow XIII · Post by **Hallow XIII** » Fri Dec 20, 2013 8:59 am

After recent discussions on the state of conlanging and especially the deluge of phonologies, plus a good deal of inspiration from miekko, I have decided this thread is needed. Not only is everyone advised to start with the phonology, but there are also several places you can show it off in. The thing is, though, phonologies are only worth so much, and if you are going to get a cool idea from one it is usually a natlang one because conphonologies are more or less the same over and over.

Grammar, on the other hand, whether it be syntax or morphology, is far more interesting and the conlangs that I have had opportunity to study generally all include at least one interesting idea that I wouldn't have thought of. Sadly, those are rarely shown off because there is a certain pressure to have complete languages. So with that in mind, this thread is for showing off interesting grammar tidbits you have come up with, whether it be syntax or morpho(phono)logy or anything in between those two. All I ask is that you keep the phoneme lists elsewhere.

Clarifying Edit: your posts here need not belong to a complete conlang; it is rather intended as a place to show off interesting grammatical ideas you would otherwise not have presented to the public. No matter how small.

Hallow XIII · Post by **Hallow XIII** » Fri Dec 20, 2013 9:00 am

So, to kick this thread off, here is something I have been toying with recently for my con-Old Chinese (a mostly monosyllabic clicklang). Basic grammatical relations are expressed by a series of polyexponential formatives that code the following, respectively:

Participants/Diathesis
Tense/Aspect/Mood/Evidential
Polarity/Politeness

With varying exponentiality for different paradigms. For instance, the last formative may co-occur with a preceding politeness marker (and possibly another one for referent rather than addressee politeness, I don't know). Similarly, the TAME formative does not necessarily code for all these categories simultaneously, making the various combinations a defective paradigm.

Bonus Feature: The language also features a pluractionality marker rather than overt marking of either nominal plurality or iterative aspect. Which one is meant must be inferred from context or can be clarified with adverbials or, in some cases, by using voice operations.

Example:

3NF>3NF.PASS hit PLURACT IND.PFV.PST.SENSORY AFFIRMATIVE.POLITE
"He was hit multiple times (I see)."

Yaali Annar · Post by **Yaali Annar** » Fri Dec 20, 2013 9:48 am

Fusing verbs and adjectives together is a common Thing in my language. Usually, what I do is making adjectives as stative verbs. In its bare form, an adjective behaves as a predicate instead of an attribute.

So for example:

puppy roll > puppy rolls
puppy fluffy > puppy is fluffy

To make the adjective into an attribute, the participle marker is added:

puppy roll-PARTICIPLE > puppy who rolls > rolling puppy
puppy fluffy-PARTICIPLE > puppy who is fluffy > fluffy puppy

You can also employ syntax based marking instead. For example verb/adjective before noun becomes attributive/participle:

roll puppy > rolling puppy
fluffy puppy > fluffy puppy
eat puppy > puppy who eats

---

The other approach in combining verb with adjective is making the verb behave adjectiv-ey instead. In this paradigm, the bare form of the verb becomes participle:

puppy fluffy > fluffy puppy
puppy eat > puppy who eats

Thus, to make the verb and adjective into predicates a marker is required.

puppy fluffy-FINITE > puppy is fluffy
puppy eat-FINITE > puppy eats

Matrix · Post by **Matrix** » Fri Dec 20, 2013 3:01 pm

In Ishdes, adjectives are prefixed onto nouns and verbs. Those can also take the suffix -so to themselves become adjectives. For example, you could have ishiru, which could mean "beautiful woman" (or, indeed, an Ishirai woman, the Ishirai being the speakers of Ishdes) or "the woman is beautiful". Ish- means "beautiful", ir means person, and -u is the feminine singular. Thus, adjectives and the adjectivizing suffix can act copulaic.

Jess · Post by **Jess** » Fri Dec 20, 2013 4:26 pm

This is a great idea for a thread. I like making phonologies as much as the next conlanger but all my long-term projects are sustained by grammar that interests me.

With Classical Prakyũ, I am exploring Suffixaufnahme, except I'm trying to extend it into a productive system of affix absorption and deletion, only a few usages of which would fit the traditional definition of Suffixaufnahme. And then I had to complicate it by giving CP a Possessed case instead of a normal Genitive. The most basic use of suffixaufnahme in CP is to mark definiteness; an indefinite noun phrase generally does not have suffixaufnahme:

son.POSS.ABL earth
"from a son of the earth"

son.POSS.ABL earth.ABL
"from the son of the earth"

son.POSS earth.ABL
"from the son of the earth"

The last example shows that the main case of the entire noun phrase may only be marked on the dependent element of the phrase (though it probably makes more sense to interpret this as marking "son.POSS earth" as a unit). ABL could be almost any case (CP has a lot).

This happens with nouns modified by adjectives too, not just nouns modified by other nouns:

cat.ACC ugly.ACC
an ugly cat

cat ugly.ACC
the ugly cat

Note that *cat.ACC ugly would be ungrammatical.

This is just scratching the surface of Weird Noun Phrase Shit in CP. Overall I am trying to make it like an "inverse Navajo" with no verbal morphology but a ton of nominal morphology. To this end I've been reading up on case-stacking in natlangs, such as the batshit way that Kayardild uses cases. It is a long, slow work in progress because it is probably by far the weirdest grammar I've ever tried to construct, maybe even unrealistically weird.

gach · Post by **gach** » Fri Dec 20, 2013 5:59 pm

This is a nice start of a thread. I'm constantly facing the problem of getting ideas for grammar but lacking actual forms to give the needed segments because I can't make the final decisions on the sound history or the morphophonemics.

Kišta will have a category of distancing in its finite verbs. It contrasts unmarked plain forms with the so called distanced forms marked with one single verb final morpheme. It's hard to pinpoint its core meaning too accurately since it appears in a variety of different uses:

- If the speaker wants to background the information conveyed by the sentence or just feels detached from it, the distancing marker can be used to indicate pragmatic distance.

- Temporal distancing is often achieved through the use the distancing marker either on its own for distant past,

he once young-COP.SG3-DIST
"He was young once."

or with irrealis for distant future,

still come.back-IRR-SG1-DIST
"I'll still come back"

- Inferential statements always receive marking for distancing,

here-INSTR disease pass-SG3-DIST
"A disease has passed here (judging from what's evident)."

- On questions marking for distancing signals polite requests or cautious enquiries,

yet part-ACC-SG2.POSS pay-Q.SG2-DIST
"Did you already pay your part?"

I guess dialectically distancing can also be used on imperatives for the effect of polite requests but I wouldn't call this standard for the majority of the speakers.

Morphologically the only change that the use of distancing imposes on the verb is that the perfective and imperfective aspects merge, or in other words the perfective suffix can never appear together with the distancing suffix.

Yaali Annar · Post by **Yaali Annar** » Sat Dec 21, 2013 9:21 pm

This feature is totally unnatural. But anyway... syntax based aspect (or tense) encoding.

So, this language has fixed word order except that the verbs can be positioned anywhere in the sentence depending on the aspect. So for example you can have:

drink puppy.NOM milk.ACC > puppy haven't drunk milk
puppy.NOM drink milk.ACC > puppy is drinking milk
puppy.NOM milk.ACC drink > puppy have drunk milk

Tiamat · Post by **Tiamat** » Sun Dec 22, 2013 12:09 pm

I was thinking about making one of my conlangs get rid of prepositional phrases completely. What I was thinking of having take over the function of preposition is a series of locational verbs and relative clauses. Some examples to illustrate my idea.

man.nom house-obl be.at-3
The man is in/at a house

house-obl be.at-rel man.nom dog-obl see-pst-3sub(-3obj)
the man in/at the house saw the dog

man.nom house-obl be.at-rel dog-obl see-past-3sub(-3obj)
the man say the dog in the house (the dog is in the house)

man.nom house-obl be.at-3-comp dog.obl see-past-3sub(-3obj) (comp means some sort compartmentalizer that gives the CP a temporal adverbal meaning)
the man saw the dog in the house (he saw the dog while he was in the house)

Thoughts? Questions?

cromulant · Post by **cromulant** » Sun Dec 22, 2013 12:29 pm

Rather than having grammatical role slots on the verb that inflect for person, this language has person slots that inflect for grammatical role.

qiihoskeh · Post by **qiihoskeh** » Sun Dec 22, 2013 4:10 pm

Vortex wrote:I was thinking about making one of my conlangs get rid of prepositional phrases completely. What I was thinking of having take over the function of preposition is a series of locational verbs and relative clauses.

Thoughts? Questions?

I've used variations of that idea for several conlangs. What I always have to think about is how to handle things like "into" and "away from" and "down from on top of".

Tiamat · Post by **Tiamat** » Sun Dec 22, 2013 4:20 pm

qiihoskeh wrote:
Vortex wrote:I was thinking about making one of my conlangs get rid of prepositional phrases completely. What I was thinking of having take over the function of preposition is a series of locational verbs and relative clauses.

Thoughts? Questions?
I've used variations of that idea for several conlangs. What I always have to think about is how to handle things like "into" and "away from" and "down from on top of".

I was thinking about having direction and location marking on the verb. So in my third example instead of having a separate clause have it be kind of another argument.

Man.nom house-obl dog-obl see-past-loc-3sub(-3obj)
the man saw the dog in the house (he saw the dog while he was in the house)

Then have different suffixes that express the specifics of the location or direction.

gach · Post by **gach** » Sun Dec 22, 2013 4:35 pm

Directional verbs work and more complex local notions can always be expressed using local nouns as in "from the top of ..."

Jess · Post by **Jess** » Sun Dec 22, 2013 4:58 pm

Yaali Annar wrote:This feature is totally unnatural. But anyway... syntax based aspect (or tense) encoding.

So, this language has fixed word order except that the verbs can be positioned anywhere in the sentence depending on the aspect. So for example you can have:

drink puppy.NOM milk.ACC > puppy haven't drunk milk
puppy.NOM drink milk.ACC > puppy is drinking milk
puppy.NOM milk.ACC drink > puppy have drunk milk

Coincidentally enough, this is the other thing I was thinking about posting, although Classical Prakyũ only distinguishes perfective (OV) vs. imperfective (VO) this way.

Some other word-order changes come into play for CP's equivalent of passive voice:

guard.NOM kick prisoner.ACC - "The guard was kicking the prisoner"
prisoner.ACC kick guard.NOM - "The prisoner was getting kicked by the guard"
prisoner.ACC kick - "The prisoner was kicked/being kicked" (the syntactic aspect distinction only occurs in transitive sentences)

Probably not a "true" passive voice since the patient's accusative marking never changes, but it serves the same function of topicalizing the patient and optionally not specifying the agent. I don't think it counts as ergativity either because prisoner.NOM kick "The prisoner was kicking" is also a perfectly normal sentence.

KathTheDragon · Post by **KathTheDragon** » Sun Dec 22, 2013 6:34 pm

One of my earlier conlangs also did that 'passive'-word-order trick.

Nortaneous · Post by **Nortaneous** » Sun Dec 22, 2013 10:58 pm

qiihoskeh wrote:
Vortex wrote:I was thinking about making one of my conlangs get rid of prepositional phrases completely. What I was thinking of having take over the function of preposition is a series of locational verbs and relative clauses.

Thoughts? Questions?
I've used variations of that idea for several conlangs. What I always have to think about is how to handle things like "into" and "away from" and "down from on top of".

Easy: derivational affixes attached to nouns. "I am in the house" would be 1S house-in. Then locative clauses are formed like relative clauses -- "the man in the house is drinking coffee" would be man REL house-in coffee drink -- and, if you don't use locative cases for these, verbs of motion as statements of intent -- "I am walking into the house" would be 1S in.order.to house-in walk, and "walk" would always be monotransitive.

I have a conlang that does this -- it doesn't have any locative adpositions or cases -- but it does have postpositions for things like the benefactive.

Matrix · Post by **Matrix** » Sun Dec 22, 2013 11:17 pm

While I was on my way home from Pathfinder earlier today, I came up with this: [tθ’~ǀ]. The two sounds sound rather similar to me, so I think they'd make a good couple of sounds to be in free variation.

faiuwle · Post by **faiuwle** » Sun Dec 22, 2013 11:48 pm

I actually posted about this once before (long ago when C&C was still pruned, so it isn't around anymore), but I'm trying to revive an old conlang where I marked tense on the topic noun instead of the verb, and othe nouns can take tense if they differ from the topic:

2S-PRES-TOP 1s.NOM NEG-like
"I don't like you."

2S-PAST-TOP 1S.NOM NEG-like
"I didn't use to like you."

2S-PAST-TOP 1S.NOM-PRES NEG-like
"I don't like the way you used to be."

2S-PRES-TOP 1S.NOM-PAST NEG-like
"I wouldn't have liked the way you are now, in the past."

2S-PRES-TOP 1S.NOM-FUT DUB-like
"I probably won't like the way you are now in the future."

I'll need to get back to working on it after I fix my conlang dictionary program up to do what I want it to.

Yaali Annar · Post by **Yaali Annar** » Mon Dec 23, 2013 12:12 am

qiihoskeh wrote:
Vortex wrote:I was thinking about making one of my conlangs get rid of prepositional phrases completely. What I was thinking of having take over the function of preposition is a series of locational verbs and relative clauses.

Thoughts? Questions?
I've used variations of that idea for several conlangs. What I always have to think about is how to handle things like "into" and "away from" and "down from on top of".

I recall, Chinese has locative verbs. Like for example:

你在哪里
you be.at where

But then again it might be interpreted as zero copula.

vec · Post by **vec** » Mon Dec 23, 2013 6:31 am

Yaali Annar wrote:This feature is totally unnatural. But anyway... syntax based aspect (or tense) encoding.

So, this language has fixed word order except that the verbs can be positioned anywhere in the sentence depending on the aspect. So for example you can have:

drink puppy.NOM milk.ACC > puppy haven't drunk milk
puppy.NOM drink milk.ACC > puppy is drinking milk
puppy.NOM milk.ACC drink > puppy have drunk milk

My conlang Lomanin does this, exactly that way:

loma y ve Lomanin "I speak Lomanin"
y loma ve Lomanin "I spoke Lomanin"
y ve Lomanin loma "I will speak Lomanin"

Imralu · Post by **Imralu** » Mon Dec 23, 2013 1:48 pm

Parts of Speech
Ngolu essentially has four parts of speech:

interjections - closed-ish
particles - closed
nominals - closed
verbals - open

Nominals are a closed class that encode ...

case (nominative, accusative, dative, locative, ablative, genitive, possessive, vocative, causative, benefactive, instrumental, comitative, topical)
person (1st, 2nd, 3rd)
number (singular, plural)
gender (mostly semantic: inanimate, animate, masculine - the latter restricted only to initiated men)
definiteness/specificity (definite, specific, non-specific)

... with not all combinations being possible. First and 2nd persons are always definite and lack the inanimate gender. Animate and inanimate merge in the non-specific. Genitive is only distinguished from possessive in the masculine gender and in the non-specific, non-masculine.

Verbals take the role of nouns, verbs, adjectives and even adverbs and to some extent prepositions of other languages, with meanings such as "to be a house" etc.

Syntax
Ngolu clauses generally take the form:

PREDICATE (ARGUMENT) (ARGUMENT) (ARGUMENT) (ARGUMENT) (ARGUMENT) (...)

Predicates are made up of a verbal phrase. A verbal phrase may be one verbal, or a series of verbals strung together. Each verbal has a set argument structure which defines the meaning of the cases of arguments when used with that verbal as the predicate. A compound verbal phrase's argument structure is determined by more complex rules that I won't put here and haven't fully worked out yet.

Arguments consist of a nominal, optionally followed by a verbal phrase. For example, the xu (NOM.3s.DEF.INAN(.REL)) essentially means "it" or "the thing" when used alone and "the thing (that)" when followed by a verbal phrase. When followed by a verb such as "fall" or "be a house", the resulting phrase means "the thing that", "the (thing that is a) house".

Any or all arguments may be omitted. A predicate-only sentence indicates the existence of the action or entity described by the verbal phrase without defining any of the participants.

deliberately.kill NOM.3s.DEF.MASC ACC.3s.DEF.INAN
He kills it.

deliberately.kill ACC.3s.DEF.INAN
It is killed (by someone).

deliberately.kill
Something or someone is killed.

be.house NOM.3s.DEF.INAN.REL be.this1 POS.1s.MASC
This is my house.

be.house POS.1s.MASC
I have a house.

be.house
There's a house. / It's a house.

In informal Ngolu, predicativeless sentences may occur.

NOM.1s.MASC LOC.3s.DEF.INAN.REL forest
I'm in the forest.

In formal Ngolu, this can be avoided by filling the predicate with a fairly semantically empty verb such as 'be an entity', or a verb indicating aspect.

be.entity NOM.1s.MASC LOC.3s.DEF.REL be.forest
I am in the forest.

be.current NOM.1s.MASC LOC.3s.DEF.REL be.forest
I am in the forest.

An alternative is to use create a verbal phrase from a nominal using the morpheme VRB.

VRB-LOC.3s.DEF.REL be.forest NOM.1s.MASC
I am in the forest.

VRB is frequently used to place a nominal phrase inside a verbal phrase as a modifier.

light.fire NOM.3s.DEF.MASC ACC.3s.DEF.INAN.REL be.house VRB-POS.1s.MASC
He set my house on fire.

be.house (VRB-NOM.3s.SPEC.REL) be.large NOM-3s.DEF.INAN be.that3
That's a big house.

move.IPFV be.that2 NOM.3s.DEF.MASC.REL deliberately.kill (VRB-ACC.3p.NSPC.INAN) be.fish
The fisherman is coming here.

prettydragoon · Post by **prettydragoon** » Mon Dec 23, 2013 5:10 pm

This language no verbs.

Well, almost. I've been thinking about a conlang with no lexical verbs. It would use light verbs with an appropriate noun or adjective instead.
Such as

do word ::= speak
do food ::= eat
make food ::= cook
do foot ::= walk

KathTheDragon · Post by **KathTheDragon** » Mon Dec 23, 2013 6:21 pm

In other words, there is a very small closed class of verbs?

Melteor · Post by **Melteor** » Mon Dec 23, 2013 9:23 pm

One thing that popped into my mind considering my present lang Tailevu - I began thinking about describing dedicated positions in sentences for information structural purposes. For instance, having a totally different place for pronouns (PRO) which are already given information, from repeated or new NPs. Like to the left.

TOP PRO verb_phrase NP NP...

Edit: The rest of this is a different thing but just shows how it could work.

There would be three further privileged positions - immediately adjacent (to the right) of the verb, where any constituents are said to be more closely "affiliated" with the verb, and also at the extreme right edge of the sentence, which would be the "focus" (FOC). These two features can fall on one argument (VO VS), different arguments (VOS VSO), or none (OSV SOV) - because the verb itself cannot be focused in the focal position. There would be an optional "stressed" topic position in front of PRO for maximal disjuncture from the verb and to avoid focusing.

The use of the cases would allow the word order to be relatively free (mostly VO, except with topics - and intransitives), above-mentioned "postions" notwithstanding. VSO VOS SVO* are the canonical orders with full NPs. *SVO being a topical disjuncture.

One quirk is that NPs are followed by a 3rd person PRO with case marking, and the PRO constituents would only jump to the front when given. Which means that often PROs would be located inside the right edge noun phrases, and a few would be on the left in the PRO domain.

Another interesting feature would be that since the verb is in initial and the arguments are in final position - it would make some sense to restructure the alignment to ergative-absolutive, to emphasize the similarity of the intransitive subject and transitive object appearing in the same final "focused" position.

prettydragoon · Post by **prettydragoon** » Tue Dec 24, 2013 2:09 pm

KathAveara wrote:In other words, there is a very small closed class of verbs?

Yes, exactly!

roninbodhisattva · Post by **roninbodhisattva** » Wed Dec 25, 2013 12:42 am

Playing around with a little idea from something similar in Dinka. Basically, we have something like V2 but instead of the verb being in second position there's a separate auxiliary/particle that occupies that position. Something like:

X Aux S V O

And then the Aux actually agrees with the thing to its left, no matter the function. I'd be analyzing this position as a C head, and then the position the thing goes to as its specifier. With that in mind, complementizers would supres V2, say like they do in German, but you'd get and order like this:

C+Aux S V O

There might be another layer of agreement, as well, where you get subject verb agreement, but then also subject - C agreement, so in a clause where the subject moves to Spec-CP, you get something that looks like this:

S Aux-Agr.S V-Agr.S O

But when the object moves to Spec-CP you get the following:

O Aux-Agr.O S V-Agr.S

zompist bboard

The Miniature Conlangs Thread

The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread

Re: The Miniature Conlangs Thread