Re: Vardelm's Scratchpad (NP: Devani OVS order)

[Vardelm]

I have no shame in telling you that some of this will end up in Kala's grammar, which I'm in the process of rewriting.

Specifically the active/stative bits.

I feel seen.

By all means, take bits & pieces that work for you. "Fair use" and "creative borrowing", right?

[Vardelm]

Jin: Arguments

Also known as "Pronouns, Determiners, & Content Words (Verbs)".


Pronouns & Determiners

	Active Pronouns	Stative Pronouns	Active Determiners	Stative Determiners
1st - singular - plural	t’inaya t’ana	t’anji t’anjin	t’aya t’a	t’i t’an
2nd	saqûmbu	sundiwa	sambu	siwa
3rd	qasim	jida	qam	ja
Proximal	mbundi	mbiwa	mbi	mba
Distal	!ntandi	!ntiwa	!nti	!nta

Notes:

• More pronouns/determiners coming in the future
• I prefer to think of the "determiners" as "determinative pronouns", which means that are a pronoun that introduces a subordinate clause. Think of them link a pronoun + a relative pronoun wrapped into 1.
• This structure is nearly identical to the Salishan & Wakashan languages. I will be glossing the "determiners" in the same way as Wikipedia does right now, which is as though they were introducing a subordinate clause, IE "that which...", "you who..." etc.

!ñcadaja qasim !ñcada-ja bite-INTRAN qasim 3P.ACT It/they bite.

!ñcadaja qam q’alab !ñcada-ja bite-INTRAN qam that.which.is.3P.ACT.DET q’alab dog That which is a dog bites. Dogs bite.

!ñcadaja qam q’alab !t’azud !ñcada-ja bite-INTRAN !nti these.which.are.PROX.ACT.DET q’alab dog !t’azud black These that are dogs and are black bite. These black dogs bite.

!ñcada qam q’alab !t’azud ja ndalu !ñcada bite !nti these.which.are.PROX.ACT.DET q’alab dog !t’azud black ja that.which.is.3P.STA.DET ndalu man These that are dogs and are black bite that which are men. These black dogs bite men.

Notes:

• Pronouns used as stand-alone arguments
• Determiner always introduce a content word (a "verb")
• Verbs ("content words") used as arguments must be preceded by a determiner (a "determinative pronoun")
• All verbs following a determiner refer back to that last determiner

[Vardelm]

Jin: Singulative & Plurative Markers

mishati ja ndalu mishati thirsty ja that.which.is.3P.STA.DET ndalu man Men are thirsty. / Men thirst.

mishati ja ndaluwi mishati thirsty ja that.which.is.3P.STA.DET ndalu-wi man-the.SNG The man is thirsty. A (particular) man is thirsty.

mishati ja ndaluma mishati thirsty ja that.which.is.3P.STA.DET ndalu-ma man-the.PLR The men are thirsty. Some (particular) men are thirsty.

mishati yod ndalu mishati thirsty yod one.who.is.DET ndalu man A (non-specific) man is thirsty.

Notes:

• Verbs that describe a pronoun/determiner (AKA "nouns") are collective by default
• The singulative suffix based on final phone:
  • -C = -i
  • -i = -wi
  • -u = -wi
  • -a = -hi
• The plurative suffix based on final phone:
  • -VC = -Vm (final vowel reduplicated)
  • -V = -ma
• Tone register will eventually play a role in the suffix vowels
• The articles mark for number and are definite, specific, or both.
• The pronoun/determiner yod "one" may be used for an indefinite, non-specific argument

mishati mba ndalu mishati thirsty mba this.who.is.PROX.DET ndalu man These men are thirsty.

mishati mba ndaluwi mishati thirsty mba this.who.is.PROX.DET ndalu-wi man-the.SNG This man thirsty.

Notes:

• The demonstrative determiners add specificity/definition to an argument
• As a result, the plurative marker is not needed to indicate specific/definite groups
• I might have the plurative still used in a formal register
• The singulative is still added to indicate the reference of an individual

[Vardelm]

Jin: Singulative & Plurative Markers II

Singulative & plurative markers ain't just be used on verb arguments!


Predicates without a singulative/plurative suffix


Active Verbs

iquma qam ñgondahi iquma limp qam 3P.ACT.DET ñgonda-hi warrior-the.SNG The warrior limps.

gunshisa qam umbabiwi ja chapitu gunshisa sell qam 3P.ACT.DET umbabi-wi merchant-the.SNG ja 3P.STA.DET chapitu fruit The merchant sells fruit.

Stative Verbs

mbulah t'anji mbulah happy t'anji 1P.STA I am happy.

halamb t'anji qam muñka halamb smell t'anji 1P.STA qam 3P.STA.DET muñka smoke I smell smoke.

dambik ja untaqi saqumbu dambik behind ja 3P.STA untaq-i camel-the.SNG saqumbu 2P.ACT The camel is behind you.

Notes:

• Active verbs without a singulative or plurative suffix have habitual or gnomic meaning.
• Stative verbs have their intrinsic, stative meaning.
• Stative verbs that would be prepositions in other langs may seem odd since they are "backwards". What would usually be the preposition's object is instead cast in the active (more "agentive") role since it is "causing" the stative argument to be behind.

Predicates with a singulative/plurative suffix

Active verbs become episodic

!nlajak qam untaqi !nlajak bray qam 3P.ACT.DET untaq-i camel-the.SNG The camel brays (habitually; he's a grump camel).

!nlajaki qam untaqi !nlajak-i bray-SNG qam 3P.ACT.DET untaq-i camel-the.SNG The camel is braying (now).

Notes:

• Active verbs WITH a singulative or plurative suffix take on an episodic meaning; they are now a specific instance of an event.

Singulative vs plurative subjects

buch'adi qam sambiwi buch'ad-i run-SNG qam 3P.ACT.DET sambi-wi boy-the.SNG The boy is running.

buch'adam qam sambiwi buch'ad-am run-PLR qam 3P.ACT.DET sambi-wi boy-the.SNG The boy is running and running.

buch'adam qam untaqam buch'ad-am run-PLR qam 3P.ACT.DET untaq-am camel-the.PLR The camels are running (separately).

buch'adi qam untaqam buch'ad-i run-SNG qam 3P.ACT.DET untaq-am camel-the.PLR The camels are running (as one; as a pack).

Notes:

• A plurative verb suffix for a singulative subject can be used to indicate continuativity (an action of extended time) or spatial distributivity ("running all over").
• There may be an iterative aspect prefix at some point, and this suffix would be used with that as well.
• A plurative suffix on a verb for a plurative subject generally means the subjects are doing the activity, but independently of each other.
• A singulative suffix with a plurative subject means the subjects are acting as one or at least doing the action at the same time and in the same manner.

Singulative objects & distributivity

shipalahi qam t'umbadi ja untaqi shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-i servant-the.SNG ja 3P.STA.DET untaq-i camel-the.SNG The servant is feeding the camel.

shipalahama qam t'umbadi ja untaqi shipala-ma feed-PLR qam 3P.ACT.DET t'umbad-i servant-the.SNG ja 3P.STA.DET untaq-i camel-the.SNG The servant is feeding & feeding the camel.

shipalahi qam t'umbadam ja untaqi shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-am servant-the.PLR ja 3P.STA.DET untaq-i camel-the.SNG The servants are feeding the camel.

shipalahama qam t'umbadam ja untaqi shipala-ma feed-PLR qam 3P.ACT.DET t'umbad-am servant-the.PLR ja 3P.STA.DET untaq-i camel-the.SNG The servants are (each) feeding the camel.

Notes:

• Transitive, active predicates with a singulative object (stative argument) behave almost identically to intransitives
  • Plurative predicate + singulative subject = continuative or spatial distributive
  • Singulative predicate + plurative subject = unified, coordinated action
• Example 3 probably means just 1 servant is feeding the camel, but the servants as a whole are taking care of that & other tasks
• For this predicate, example 4 is a bit ridiculous, unless they are all feeding the camel a small amount for some reason (a ceremonial reward for the camel???)
• Stative predicates function the same, except the "object" is the agentive argument

Plurative objects & distributivity

shipalahi qam t'umbadi ja untaqam shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-i servant-the.SNG ja 3P.STA.DET untaq-am camel-the.PLR The servant is feeding the camels.

shipalahama qam t'umbadi ja untaqam shipala-ma feed-PLR qam 3P.ACT.DET t'umbad-i servant-the.SNG ja 3P.STA.DET untaq-am camel-the.PLR The servant is feeding each of the camels.

shipalahi qam t'umbadam ja untaqam shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-am servant-the.PLR ja 3P.STA.DET untaq-am camel-the.PLR The servants are feeding the camels.

shipalahama qam t'umbadam ja untaqam shipala-ma feed-PLR qam 3P.ACT.DET t'umbad-am servant-the.PLR ja 3P.STA.DET untaq-am camel-the.PLR The servants are (each?) feeding (each of?) the camels.

Notes:

• Example 1: the servant is feeding the camels as whole, perhaps by just putting out a pile of .... whatever camels eat.
• Example 2: the servant is feeding each individual camel separately.
• Example 3: same as example 1 but with plural subjects. Perhaps only 1 or a few are actually performing the task.
• Example 4: all of the servants are probably involved since the action is plurative. They are feeding the camels, but it is ambiguous whether they are feeding all of the camels at once or individually.
• Stative predicates function the same, except the "object" is the agentive argument

DONE! (I think...)
Figuring out how to organize those interacting ideas & how to present them was a brain teaser!

[Vardelm]

Jin:Voice & Valency I

Voice	Active	Stative
Intransitive	V+ja C+aj	V+ya C+iya
Transitive	V+nza C+anza	V+mba C+imba
Mediopassive	V+sh C+ish	V+la C+ala
Passive	V+qi C+iq	V+ta C+at
Reflexive	V+jun C+ajun	V+bu C+abu
Reciprocal	V+nzim C+unzim	V+bim C+ubim
Impersonal	V+mas C+umas	V+lah C+alah

Voice	Active	Stative
Intransitive	VS !ñcada-ja bite-INT qam 3P.DET qalab-i dog-the.SNG The dog bites.	VO wani-ya see-INT ja 3P.STA ndalu-wi man-the.SNG The man sees.
Transitive	VSO !ñcada-nza bite-TRN qam 3P.DET qalab-i dog-the.SNG ja 3P.DET ndalu-wi man-the.SNG The dog bites the man.	VOS wani-mba see-TRN ja 3P.STA ndalu-wi man-the.SNG qam 3P.DET qalab-i dog-the.SNG The man sees the dog.
Mediopassive	VO !ñcada-sh bite-MDP ja 3P.DET ndalu-wi man-the.SNG The man is bitten.	VS wani-la see-MDP qam 3P.DET qalab-i dog-the.SNG The dog is seen.
Passive	VOS !ñcada-qi bite-PAS ja 3P.DET ndalu-wi man-the.SNG qam 3P.DET qalab-i dog-the.SNG The man is bitten by the dog.	VSO wani-ta see-PAS qam 3P.DET qalab-i dog-the.SNG ja 3P.STA ndalu-wi man-the.SNG The dog is seen by the man.
Reflexive	VS !ñcada-jun bite-REF qam 3P.DET qalab-i dog-the.SNG The dog bites itself.	VO wani-bu see-REF ja 3P.STA ndalu-wi man-the.SNG The man sees himself.
Reciprocal	VS !ñcada-nzim bite-REC qam 3P.DET qalab-am dog-the.PLR The dogs bite each other.	VO wani-bim see-REC ja 3P.STA ndalu-ma man-the.PLR The men see each other.
Impersonal	V !ñcada-mas bite-IMP There is biting.	V wani-lah see-IMP There is seeing.

Notes:

• Voice/valency is marked by suffixes
• Singulative/plurative is suffixed after voice/valency (root + voice suffix + number suffix)
• Content words (verbs) will have a default voice that is not marked; other voices are marked by adding the appropriate suffix
• Each voice defines if an active and/or stative argument is required (which is why I mention "valence")
  • The table shows word orders: VS, VO, VSO, etc. These indicate the required arguments & their order (S = active, O = stative)
• Voices:
  • Intransitive = basic intransitive voice, roughly unergative for active verbs and unaccusative for stative verbs
  • Transitive = basic transitive voice typical for active & stative; has the basic word orders listed in the 1st post on word order/syntax
  • Mediopassive = can be used as a prototypical mediopassive where the single argument is affected; middle voice where no agent is implied or passive where an agent is implied
  • Passive = not your daddy's passive; this is a transitive voice which has the same roles as the transitive voice, but the order & topicality is reversed
  • Reflexive = prototypical; argument is both agent & patient
  • Reciprocal = prototypical; arguments affect each other
  • Impersonal = no mention of participants
• As mentioned, the mediopassive & passive are maybe a bit different than usual, mostly in that the passive is not a valency-reducing operation
• The mediopassive & passive basically "flip" the required arguments & their positions from the intransitive & transitive voices
  • mediopassive corresponds to intransitive
  • passive corresponds to transitive
• This post is about predicates; the next post will detail the voices within an argument phrase (under a determiner)
• I plan to tweak the current suffixes a bit from what appears now

[Vardelm]

Jin:Voice & Valency II

Verbs / content words use all of the same voices listed in the post above when they appear in an argument phrase, subordinate to a determiner.

qalab jida qalab dog.STA.INT jida 3P.STA It's a dog.

ja qalab ja 3P.STA.DET qalab dog.STA.INT That which is a dog...

Notes:

• When a verb/content word is used with a determiner, the 1st argument in its valency requirements is taken by the determiner itself, and so doesn't appear after the verb
• qalab "to be a dog" is a stative,class membership verb with VO order
• In example 2 (just an argument phrase) the stative argument O is filled by ja "that.which.3P.STA.DET"

ja qalab ja 3P.STA.DET qalab dog.STA.INT !t'azud black.STA.INT That which is a dog and is black... The black dog...

Notes:

• The same rule applies for multiple verbs under a single determiner
• !t'azud "to be black" is also a stative verb with VO order, so it's O argument is also filled by the determiner

ja qalab !ñcada ja ndalu ja 3P.STA.DET qalab dog.STA.INT !ñcada bite.ACT.TRN ja 3P.STA.DET ndalu man That which is a dog and bites that which is a man... The dog that bites men...

Notes:

• Same rule. Yep, even for transitives!
• !ñkada "to bite" is an active verb that defaults to transitive voice (thus no -nza suffix) with VSO order
• The S argument is taken by the preceding determiner ja
• That leaves the O argument, filled by ja ndalu "that which is a man"

[Vardelm]

So there's that moment when you have worked on a conlang for quite a while and have a bunch of ideas that seem like they should work, but are just convoluted enough that it's just not working. And then, you make one little change while revamping the language that allows the rest of it to be recast in a way that keeps those original ideas but helps them to actually work. That's kinda cool.


Jin: The Linker

Jin has (so far) only 1 preposition, which I call "the linker". It originates from a grammar of ǂʼAmkoe (ǂHȍã). I should go back & look through it again since I think the usage here is different. I seem to remember a few other languages that have a single adposition referred to as a "linker".

Notes:

• The linker is al- and is cliticized to the word/phrase that it modifies. The "-" indicates that it's a clitic.
• The linker essentially creates genitive or oblique phrases and serializes verbs
• The linker can attached to a pronoun, determiner, or verb
• If attached to a pronoun or determiner, the phrase is linked to the last determiner in the sentence
• If attached to a verb, that verb (and associated arguments) are linked to and modify or append a previous predicate verb. Which one is a matter of context & pragmatics, but I think mostly it will be the sentence's main predicate verb.

Genitives & Possession/Ownership

buch'adi t’izh untaqi buch'ad-i run-SNG t’izha mine.that.is.1P.ACT.POS.DET untaq-i camel-the.SNG Mine which is a camel is running. My camel is running.

buch'adi qam untaqi al-t'anji buch'ad-i run-SNG qam 3P.ACT.DET untaq-i camel-the.SNG al-t'anji of.LNK-1P.STA The camel that is of me is running.

Notes:

• When used to indicate possession with a linked pronoun, a stative pronoun is used (for now...)
• This form isn't used as often as example sentence 1. Sentence 1 could be translated as "my camel" while sentence 2 as "the camel that is mine". Sentence 2 might be used for emphasis.

shipalahi qam t'umbadi ja untaqi al-ja ndaluwi shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-i servant-SNG ja 3P.STA.DET untaq-i camel-SNG al-ja of.LNK-3P.STA.DET ndalu-wi man-SNG The servant is feeding the man’s camel.

Notes:

• This example shows the linker attached to a determiner (al-ja) which then has the verb ndalu-wi "the man"
• This argument phrase is linked back to the previous determiner, which is "ja" for ja untaqi "that which is a camel"
• The link then indicates a genitive relation: in this case, possession/ownership

shipalahi qam t'umbadi ja untaqi al-dambik qam nimbani shipala-hi feed-SNG qam 3P.ACT.DET t'umbad-i servant-SNG ja 3P.STA.DET untaq-i camel-SNG al-dambik of.LNK-behind qam 3P.ACT.DET nimban-i house-SNG The servant is feeding the camel behind the house.

Notes:

• Here the linker is attached to the verb dambik "to be behind"
• Because the linker is attached to a verb, it modifies the last predicate, which is shipalahi "to feed"
• This is how oblique/adjunct phrases are added to a sentence
• Phrases linked in this manner need to come at the end of the sentence, after the active & stative arguments

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

WHEW!!!!

This might be my final post for a short while. These ideas have been floating in my head for several years, but I haven't found a way to make them really work and describe them until now. Most of what I'll be doing for this language from here out will be completely new. I had a verb template sketched out on the last board which will serve as a springboard for the rest of the verbal inflections, but even that will need to be heavily modified.

[Ares Land]

Kudos! That's very interesting, and I do like the inflected pronouns.

I'd be very interested in seeing how those voices work in detail, I confess I have some trouble wrapping them around my head. But all in good time of course.

You probably talked about this on the old board or the new, but I think I missed it: are the latter two languages related in any way? And who speaks them?

[Vardelm]

Kudos! That's very interesting, and I do like the inflected pronouns.

Thank you. Funny, I actually hadn't thought of them as "inflected", but yeah, obviously you're correct. Might be just because I was comparing to Imralu's Iliaqu (Ngolu) language, which IIRC has a bunch of locational type noun cases on the pronouns as well as the morphosyntactic type cases.

I'd be very interested in seeing how those voices work in detail, I confess I have some trouble wrapping them around my head. But all in good time of course.

That's not a bad idea for a next post. It wouldn't involve any new design work & could help define that a little better.

You probably talked about this on the old board or the new, but I think I missed it: are the latter two languages related in any way? And who speaks them?

Actually, all 3 languages are related, but only in that they are set in the same world. The plan is to have 4 demi-human races plus humans. I want to do something similar to how Tolkien took the idea of elves, dwarves, etc. and made a history as though those had been actual cultures rather than just treating them as children's tales. The idea is to also take inspiration from various spirits/mythical beings from around the world, instead of focusing just on European myths. Each race will come from 1 of 4 contients. The 4 demi-human folk are:

• dwarves (European-ish; northern continent)
• deva (Indic to SE Asia influence, maybe some Central to South American; eastern continent)
• jin (from djinn, of course, inspired by African and Middle-eastern cultures; southern continent)
• yokai (also from yokshini & yakshas, representing a sort of pan-East Asia and some North American; western continent)

The yokai language is the least developed so far and I haven't posted anything about it, but I have some general ideas. I would say that the races might eventually not be recognizable to their real-world inspiration, but those being are a jumping-off point, and we'll see where it goes. I have a few general ideas for the cultures and the world history, but the world right now is mostly about the languages. I have a sort of "stage 1" plan where I make the languages, do ethnographic write-ups of the 4 cultures, develop the magic system, and create a world map. All of this will represent the ancient history of the world, rather than any sort of "modern day" (which would be a Middle-ages type era) in which stories might be set.

[Vardelm]

So...... bugger. A post I wrote either didn't get posted or got eaten. It was supposed to go right before the post on the linker, which would show why the linker isn't used for phrases like "my camel", etc. I even refered to the possessive pronouns in the linker post. Bleh. Here's another go at that.


Jin: Possessive Pronouns & Determiners

	Active Pronouns	Stative Pronouns	Active Determiners	Stative Determiners
1st Possessive - singular - plural	t’inazh t’anzha	t’izhi t’anizh	t’izha t’azha	t’izh t’azh
2nd Possessive	suqanzha	sunizh	suzha	suzh
3rd Possessive	qisazh	jadizh	qizh	jazh

dambik ja untaq qam nimbani dambik behind t’azh our.3P.STA.POS.DET untaq camel qam 3P.ACT.DET nimban-i house-SNG Our camels are behind behind the house.

buch'adi t’inazh buch'ad-i run-SNG suqanzha yours.2P.ACT.POS Yours is running.

Notes:

• The possessive pronouns & determiners work just like the personal & demonstratives, except they denote possession/ownership
• I don't love the forms of them right now, but meh: 1st draft

I think this is a bit shorter than the original post, but not much since it's fairly straightforward.

[Vardelm]

I'd be very interested in seeing how those voices work in detail, I confess I have some trouble wrapping them around my head. But all in good time of course.

This comment had me thinking (in a good, creative problem solving way). I did some work to revise the voice system from it's original form on the old board. However, it needs more detail, explanation, & examples. While thinking about examples, I decided the stative vs dynamic distinction needed to be more defined, and the same for causativity or other (de-)transitivitizing effects. I believe I have something that works better now, but I'd love feedback on it.


Jin: More Voice & Valency


Default Voices

	Active	Stative
Intransitive	VS ikinsu "mourn" nzabulu "relax"	VO marah "red" mimut "dead"
Transitive	VSO k'alis "cut" nd'api "write"	VOS !nlabun "know" wind'a "inside"

Notes:

• Verbs all have a default voice that is either active or stative and either intransitive or transitive
• Verbs in their default voice do not have a voice suffix
• The voice suffixes straddle the line of inflection & derivation

Example: mimut "dead"

	Active	Stative
Intransitive	VS mimutaji qam untaqi mimut-aj-i dead-ACT.INT-SNG qam 3P.ACT.DET untaq-i camel-SNG "The camel is dying."	VO mimut ja untaqi mimut dead ja 3P.STA.DET untaq-i camel-SNG "The camel is dead."
Transitive	VSO mimutanzahi qam ndaluwi ja untaqi mimut-anza-hi die-ACT.TRN-SNG qam 3P.ACT.DET ndalu-wi man-SNG ja 3P.STA.DET untaq-i camel-SNG "The man killed the camel." (not worry about tense right now)

Notes:

• In the intransitive stative, mimut "dead" appears with no voice suffix because this is its default voice
• For the intransitive active, the suffix -aj is added. Adding active voice suffixes to statives generally (always?) makes them dynamic
  • I actually don't know that I love this. I would almost rather add the stative intransitive suffix -iya and use the ja stative determiner to indicate the change from stative to dynamic. Not sure what the change to active would be, though.
• The transitive voice adds -anza, and of course adds a second argument. Verbs that go from default intransitive to taking a transitive voice suffix will often add causativity ("to die" -> "to kill"), at least on the active side.

	Active	Stative
Intransitive	VS mimutaji qam untaqi mimut-aj-i dead-ACT.INT-SNG qam 3P.ACT.DET untaq-i camel-SNG "The camel is dying."	VO mimut ja untaqi mimut dead ja 3P.STA.DET untaq-i camel-SNG "The camel is dead."
Transitive	VSO mimutanzahi qam ndaluwi ja untaqi mimut-anza-hi die-ACT.TRN-SNG qam 3P.ACT.DET ndalu-wi man-SNG ja 3P.STA.DET untaq-i camel-SNG "The man killed the camel."
Antipassive	VS mimutiñguwi qam ndaluwi mimut-iñgu-hi die-ACT.ANP-SNG qam 3P.ACT.DET ndalu-wi man-SNG "The man killed (something)."
Mediopassive	VO mimutishi ja untaqi mimut-ish-i die-ACT.MDP-SNG ja 3P.STA.DET untaq-i camel-SNG "The camel was killed (by something)."
Passive	VOS mimutishi ja untaqi qam ndaluwi mimut-iq-i die-ACT.PAS-SNG ja 3P.STA.DET untaq-i camel-SNG qam 3P.ACT.DET ndalu-wi man-SNG "The camel was killed by the man."
Reflexive	VS mimutajuni qam ndaluwi mimut-ajun-i die-ACT.REF-SNG qam 3P.ACT.DET ndalu-wi man-SNG "The man killed himself."
Reciprocal	VS mimutunzimi qam ndaluma mimut-unzim-im die-ACT.REC-PLR qam 3P.ACT.DET ndalu-ma man-PLR "The men killed each other."

Notes:

• One major-ish change I made from posts above was to add the antipassive voice. This was so that there could be a clear distinction between the "intransitive" voices, which don't imply any sort of 2nd actor, and all of the transitive & *passive voices, all of which do imply a 2nd actor. Adding the antipassive allows the transitive passive voices to have VSO, VS, VO, VOS variants.
• All of the voices in the transitive range (transitive, antipassive, mediopassive, passive, reflexive, & reciprocal) will have the same semantics in terms of causativity or whatever makes them transitive. The stative side won't ever have causativity.
• Not every verb will use every voice. Here, the stative transitive voices aren't used at all.

-----

I'm going to cut this post short here & provide more examples in a another post.

[bradrn]

I’m not too sure I understand how this system works. Are you saying that there are four affixes to derive active/stative transitive/intransitive verbs — and if so, then where do the various intransitive/transitive/antipassive/mediopassive/etc. affixes fit in?

(And also, wouldn’t an active vs stative contrast be a distinction of aspect rather than voice?)

[Vardelm]

I’m not too sure I understand how this system works. Are you saying that there are four affixes to derive active/stative transitive/intransitive verbs — and if so, then where do the various intransitive/transitive/antipassive/mediopassive/etc. affixes fit in?

There are a total of 16 "voice" affixes:

• active intransitive
• active transitive
• active antipassive
• active mediopassive
• active passive
• active reflexive
• active reciprocal
• active impersonal
• ------------------------
• stative intransitive
• stative transitive
• stative antipassive
• stative mediopassive
• stative passive
• stative reflexive
• stative reciprocal
• stative impersonal

The "voice" affixes actually do the derivation; there aren't any other affixes to "derive active/stative transitive/intransitive verbs". These "voice" affixes ride the line between derivation & inflection. They change the verb's default active/stative status as well as it's voice & valency, and in doing so can change the semantics a bit in terms of causativity, forming resultatives, etc.

(And also, wouldn’t an active vs stative contrast be a distinction of aspect rather than voice?)

Probably?

The thing here is that there's not 1 "active voice" and 1 "stative voice". Instead, there are 8 voices that happen to work on active verbs or derive an active verb from a stative, and there are 8 voices that happen to work on stative verbs or derive an stative verb from an active.

Does this make more sense?

[bradrn]

That makes a lot more sense now, thank you!

And I think I found a typo: in a couple of places a sentence has ja but it’s glossed as qam.

[Vardelm]

That makes a lot more sense now, thank you!

Excellent! As I develop this more, stating at the outset that there are 16 voices might be worthwhile.

And I think I found a typo: in a couple of places a sentence has ja but it’s glossed as qam.

Commencing repairs.

[Ares Land]

Well, that's really a great system!

I just have a few remarks/quibbles:

What is the impersonal voice? I suppose it's used for forms akin to 'it rains'?

and there are 8 voices that happen to work on stative verbs or derive an stative verb from an active.

I'm not sure I understand the stative voices. I thought maybe a stative transitive changed a stative into an active transitive, but that doesn't seem to be the case?

As far as I can see from your example, the mediopassive is really a passive. Mediopassive is a difficult label, but the medio- part implies that the agent is benefitted, or somewhat affected by the action. (Some ideas: for 'I kill for my benefit' ie. 'I hunt', 'I kill myself for my benefit', ie. 'I work hard').

Oh, and in such a system with lots of productive morphological categories, there tends to be a lot of 'defective' verbs. So for instance, a given verb may only use 4 out of the 8 voices given, for obvious or less obvious reasons. So don't be unduly bothered if you can't figure what the mediopassive or the antipassive should be for a given verb, or if you find that a verb is going to be always mediopassive.
You're problably already aware of this already (and in this case you can just dismiss that comment) but I had trouble figuring this out myself so maybe it'll help.

[Vardelm]

Well, that's really a great system!

Thanks, glad you like it, and I'm happy the added information was useful for understanding it.

It came about by wanting a system of verbal endings like person agreement, but wanting to stay away from any sort of person agreement: nothing for person, gender, or number.

What is the impersonal voice? I suppose it's used for forms akin to 'it rains'?

Yep, exactly. I think it may be used to form the equivalent of gerunds & infinitives as well, possibly through a new pronoun/determiner instead of qam/ja. I think one used for subordinate clauses would work. (I'm blanking on the term for such clauses right now, meaning "a full clause that could stand on it's own as a complete sentence but is used as a subject or object. Maybe "subordinate" is correct?)

I'm not sure I understand the stative voices. I thought maybe a stative transitive changed a stative into an active transitive, but that doesn't seem to be the case?

Nope. I'll work on getting a couple more examples posted that use the stative transitives, so maybe that will help. Stative transitives would be examples like "I see her" or "I know English". They don't have a dynamic component. Adpositions from other languages would be stative transitives in Jin, as another quick example.

As far as I can see from your example, the mediopassive is really a passive. Mediopassive is a difficult label, but the medio- part implies that the agent is benefitted, or somewhat affected by the action. (Some ideas: for 'I kill for my benefit' ie. 'I hunt', 'I kill myself for my benefit', ie. 'I work hard').

Basically, yes. I just needed some kind of label, and I'm already using "passive" voice for the 2nd fully transitive voice (VOS for active and VSO for stative). "Mediopassive" seemed like an acceptable fit, but admittedly it might be be exactly what people think of as "mediopassive". I would note that mediopassives tend to vary from language to language, correct? I figured since mediopassives are sometimes used as a passive, and the "medio" part of it points to there not being a separate, specificed agent lent itself well here. Perhaps something like "semi-passive" might work? Basically, it's a passive without an agent mentioned.

Oh, and in such a system with lots of productive morphological categories, there tends to be a lot of 'defective' verbs. So for instance, a given verb may only use 4 out of the 8 voices given, for obvious or less obvious reasons. So don't be unduly bothered if you can't figure what the mediopassive or the antipassive should be for a given verb, or if you find that a verb is going to be always mediopassive.
You're problably already aware of this already (and in this case you can just dismiss that comment) but I had trouble figuring this out myself so maybe it'll help.

Yep, I'm not concerned with many verbs not using all of the voices, but it's good to see this affirmed. I'm perhaps more concerned about each verb using what voices it does in a more-or-less correct manner and eventually trying to make sure I'm not just duplicating English verbs (although that will largely be the case at the moment since I'm making up words on the fly).

[cedh]

As far as I can see from your example, the mediopassive is really a passive. Mediopassive is a difficult label, but the medio- part implies that the agent is benefitted, or somewhat affected by the action. (Some ideas: for 'I kill for my benefit' ie. 'I hunt', 'I kill myself for my benefit', ie. 'I work hard').

Basically, yes. I just needed some kind of label, and I'm already using "passive" voice for the 2nd fully transitive voice (VOS for active and VSO for stative). "Mediopassive" seemed like an acceptable fit, but admittedly it might be be exactly what people think of as "mediopassive". I would note that mediopassives tend to vary from language to language, correct? I figured since mediopassives are sometimes used as a passive, and the "medio" part of it points to there not being a separate, specificed agent lent itself well here. Perhaps something like "semi-passive" might work? Basically, it's a passive without an agent mentioned.

A prototypical "passive voice" is intransitive, i.e. it usually doesn't have an agent as a syntactic core argument, and if there is an overt agent, it is typically marked as oblique in some way. And a symmetrical transitive voice, where the agent swaps roles with the patient but still retains its core status, is generally not called "passive voice" but rather "inverse voice" or similar. Therefore, maybe you could re-label your "passive" as "inverse", and your "mediopassive" as "passive". Although I suppose the relative frequency of usage might also play a role in finding the best terms...

[Vardelm]

And a symmetrical transitive voice, where the agent swaps roles with the patient but still retains its core status, is generally not called "passive voice" but rather "inverse voice" or similar. Therefore, maybe you could re-label your "passive" as "inverse", and your "mediopassive" as "passive".

I rather like this suggestion.

Wouldn't systems like that usually call the usual transitive voice "direct"? Or, are there systems where "inverse" is used, but not "direct"? This is definitely not the typical "direct-inverse" system that I'm familiar with, which is based on hierarchy, IE Ojibwe, Cree, etc.

EDIT: Using the label "symmetric voice" for the current "passive voice" might be an option as well.

[cedh]

And a symmetrical transitive voice, where the agent swaps roles with the patient but still retains its core status, is generally not called "passive voice" but rather "inverse voice" or similar. Therefore, maybe you could re-label your "passive" as "inverse", and your "mediopassive" as "passive".

I rather like this suggestion.

Wouldn't systems like that usually call the usual transitive voice "direct"? Or, are there systems where "inverse" is used, but not "direct"? This is definitely not the typical "direct-inverse" system that I'm familiar with, which is based on hierarchy, IE Ojibwe, Cree, etc.

EDIT: Using the label "symmetric voice" for the current "passive voice" might be an option as well.

You're right that this is not quite a "direct-inverse" system. But natlangs with symmetrical "passives" can have a variety of very different voice systems. For instance, one of the defining characteristics of Austronesian alignment is that these languages have several symmetrical transitive voices. There, they are often called "agentive voice" and "patientive voice", and I think these terms should be quite suitable for your conlang too. (I'm also using them in my current project Lerudųrunį.)