"To have" in Wena/Hibuese/Ngehu

[Imralu]

Languages employ a lot of different strategies to express "having" something and I just wanted to talk about how it's expressed in my conlang that I've variously called Wena, Hibuese, Ngehu etc. etc. (I also keep moving the setting. Terrible!)

Wena essentially doesn't have a lexical category of verbs other than, arguably, the non-inflecting copular / predicate marking particle i (e after another i, which is used to link noun phrases. You can think of a sentence such as "I write books" as essentially being structured like "I am a writer of books": Na i ngegyo ya ndyolya (1S COP writer GEN book). Obviously you can syntactically find a predicate and you could argue that i ngegyo "is a writer" is simply a verb meaning "writes", but there is no distinction in lexical categories and could then end up saying i mba is a verb meaning "is a house". There is also no grammatical distinction between a possessor and an object, as the equivalent of objects are essentially present as genitive modifiers. The genitive marking can even be dropped when context allows, forming a loose compound such as ngegyo ndyolya ("book writer") (as opposed to a "tight" compound which is written as one word and distinguished by stress) and with certain words, the genitive marking is more often dropped than not (that generally depends more on the head than the modifier).

There are three words that basically express different shades of "having".

• nyu - owner (expresses legal ownership)
• za - holder, wearer, carrier, one who has ... on/with oneself (expresses physical location or at least temporary control over the object)
• ne - one who has ... without volition; one who is endowed with ...; undergoer, victim (expresses non-volitional possession, ornative case, passive voice)

I'm curious about gaps I may not have thought about in the semantic space of "having" and it would be cool if someone could give some examples of "having" that might not fit neatly into the nyu, za, ne categories so I can think about how to express them.

Here are some examples to show them a bit better:

Na i nyu (ya) mba.
1S COP owner (GEN) house
I have/own a house.
I'm a home owner.

Funny coincidence: nyumba means "house" in Swahili.

Na i za (ya) nggu wo.
1S COP holder (GEN) money GEN.2S
I've got your money.

Na i ne (ya) hugo.
1S COP PASS (GEN) brother
I have a brother

Generally the ya would be dropped in these sentences.

If you're confused about there is overlap between passive marking and possession, essentially the idea here is that, "to have a murderer" means "to be murdered", "to have a writer" means "to be written" etc. There's also a little bit of a parallel in the use of a kind of a past participle formed from a noun to form ornative adjectives in English and some romance languages, like "dark-haired".

ne (ya) hwi (ye) dwo
PASS (GEN) hair (ATTR) black.thing
"black-haired (person, animal)"

Compare the transformation of active to passive with both verby and nouny nouns.

Hi e vwe nga.
DEM COP seer GEN.1S
This (person) sees me.
*This is my seer / a seer of mine.

Na i ne vwe.
1S COP PASS seer
I am seen.
*I have a seer.

Hi e hwi dwo nga.
DEM COP hair black.thing GEN.1S
This is my black hair / a black hair of mine.

Na i ne hwi dwo.
1S COP PASS hair black.thing
I have black hair.
I am black-haired

[bradrn]

Firstly, I love the idea of having a purely noun-based language! As it happens, I’ve been thinking about this same idea for the past day or two, and independently ended up at a very similar design to what you have (although I attempted to base it around a monovalent existential rather than a divalent copula).

There are three words that basically express different shades of "having".

• nyu - owner (expresses legal ownership)
• za - holder, wearer, carrier, one who has ... on/with oneself (expresses physical location or at least temporary control over the object)
• ne - one who has ... without volition; one who is endowed with ...; undergoer, victim (expresses non-volitional possession, ornative case, passive voice)

I'm curious about gaps I may not have thought about in the semantic space of "having" and it would be cool if someone could give some examples of "having" that might not fit neatly into the nyu, za, ne categories so I can think about how to express them.

Dixon and Aikhenvald’s Possession and Ownership lists six different meanings possessive constructions may have: the ‘core’ meanings are (a) ownership (in terms of control of property), (b) whole-part relations and (c) kinship relations, while less core meanings include (d) association in general (Paul’s dentist), (e) locations/orientations (top of the mountain), and (f) attribution and properties (the man’s temper). In English, ‘have’ may be used for any of these meanings (a)–(f). It looks like your nyu covers (a), and ne covers (c) and possibly some cases of (b),(d),(f); I’m even less about za, but it may be (a),(e). So, if I’m interpreting your three words correctly, you seem to have covered the semantic space fairly well here.

One other thing I note about this is that your words appear to have some degree of overlap. I see this as possibly having a semantic impact: you can use different possessive words to emphasise particular aspects of your ownership. For instance, you might use nyu to emphasise that this is yours and no-one else’s, whereas you might use ne to emphasise that you did not actively choose to own it. (This reminds me particularly of noun classifiers: in Classifiers, Aikhenvald gives the example from Burmese that you might use myiʔ tə yaʔ ‘river one place’ for a picnic destination, whereas myiʔ tə tan ‘river one line’ gets used for a river on a map.)

[Pabappa]

There is also no grammatical distinction between a possessor and an object, as the equivalent of objects are essentially present as genitive modifiers.

I like this, but Im not sure Im reading it right .... are you saying the possessor aligns syntactically with the patient of a verb? so if I own a car, Im the "object", not the car?

I like the three types of genitives too .... my conlang basically does that, but I strongly lean in favor of the unmarked possessive, using the others only for emphasis or for a few nouns that just by tradition canonically take one of what i call the augmented possessives. I like the term "ornative case" .... never heard of it before ... I might use that term somewhere too.

The augmented possessives in Poswa are a theoretically open class, being created formulaically from free morphemes pushed through certain sound rules, so while I can give you suggestions, some of them are so off-the-map that I suspect you wont have a use for them.

So here we go:
-mb- borrowed, temporarily owned
-amb- beloved, dearly held
-epp- supported, favored by (e.g. "my team" ... i suspect literally no natlangs have a distinct morpheme for this, but for some reason Ive always wanted it)
-mp- created (e.g. i own it because i made it; used for one's children as well)
-p- desired, wanted (i dont own it yet, but i've made up my mind which one i want)

I dont know if youll want any of these. Poswa's grammar is such that I felt they would be a good fit, but I wouldnt try something like this even in my other conlangs, because it would be far more convenient to just use ordinary verbs. I have plenty more, but their meanings are even further off from the standard genitive sense, e.g. /-b-/ means "exposed, shown" and is used in a manner similar to English sentences like "that, you know, bracelet ..."

Also .... can a single noun take more than one possessor argument? e.g. if i leave a huge mess on the carpet, but its your job to clean it up, is it my mess (nyu or za) or yours (ne)? Or both?

[Imralu]

Firstly, I love the idea of having a purely noun-based language! As it happens, I’ve been thinking about this same idea for the past day or two, and independently ended up at a very similar design to what you have (although I attempted to base it around a monovalent existential rather than a divalent copula).

Ooh, I lack the imagination to see how that would work. How do you say "X is Y"?

Dixon and Aikhenvald’s Possession and Ownership lists six different meanings possessive constructions may have: the ‘core’ meanings are (a) ownership (in terms of control of property), (b) whole-part relations and (c) kinship relations, while less core meanings include (d) association in general (Paul’s dentist), (e) locations/orientations (top of the mountain), and (f) attribution and properties (the man’s temper). In English, ‘have’ may be used for any of these meanings (a)–(f). It looks like your nyu covers (a), and ne covers (c) and possibly some cases of (b),(d),(f); I’m even less about za, but it may be (a),(e). So, if I’m interpreting your three words correctly, you seem to have covered the semantic space fairly well here.

Ah thanks, that's useful. I generally function better with examples, so I'll make some and someone can correct me if it's not exactly what the category meant.

(A) I have a house. Na i nyu mba.
(B) I have a head. Na i ne zyo. The house has a roof. Mba i ne vuzwi.
(C) I have a sister. Na i ne lago.
(D) I have a dentist. Na i ne zyehondi gembwa.
(E) The house has a roof. Mba i ne vuzwi. The mountain has a top. Mwoda i ne vuzwi.
(F) I think this would generally just be expressed more directly as "man is temperamental.person" rather than deriving an abstract noun phrase from it and then saying that he has it. That sounds like two unnecessary steps to get back to the first position. I may find a way to do this that seems useful though and provides some different nuance.

Za doesn't really seem to be covered unless that is what is meant by (E). I see the example "top of the mountain" as being an example of (b), no? I mean, it could be both; parts of something are generally located in the same place as them unless we're talking about something collective and dispersed like an organisation. In any case, za is like when someone talks about having something on them or somehow in their possession even if not theirs. If you can talk about having someone else's something, then it's necessarily the za kind of possession. Turkish shows this difference with the genitive case/possessive suffixes for most kinds of possession, but the locative case for locative possession.

(Benim) para-m var.
(my) money-my existent
I have money.
Na i ne nggu.

Bende para var.
at-me money existent
I have money (on me)
Na i za nggu.

(Senin) para-n bende
(your) money-your at-me
I have your money. / Your money is with/on me.
Na i za nggu wo

One other thing I note about this is that your words appear to have some degree of overlap. I see this as possibly having a semantic impact: you can use different possessive words to emphasise particular aspects of your ownership. For instance, you might use nyu to emphasise that this is yours and no-one else’s, whereas you might use ne to emphasise that you did not actively choose to own it. (This reminds me particularly of noun classifiers: in Classifiers, Aikhenvald gives the example from Burmese that you might use myiʔ tə yaʔ ‘river one place’ for a picnic destination, whereas myiʔ tə tan ‘river one line’ gets used for a river on a map.)

Hmm, yeah, but can't really think of any examples that would illustrate that well. Na i ne mba means "I have a house" in the sense that I am not homeless, but if you want to indicate that you own it (not renting for example), you'd say Na i nyu mba. If you're squatting in or occupying a house that is not really "yours", you'd say na i za mba, but you can use ne for this too. I think basically ne is the most versatile of the three. I guess it doesn't explicitly mark non-volition, but the other two generally do imply volition as you can sell your property and release things that you have in your possession.

All of these kinds of possession are equivalent to the same kind of genitive relationship, if we take it the other way. There are three different ways to say "I have a house", but only one way to say "my house" (mba nga) unless you really want to get much deeper into it and say mba ye nenyu nga (house which is my legal possession), mba ye neza nga (house which is in my possession). You might notice that nenyu and neza are basically compounds, passivised forms of nyu and za; nyu = owner, nenyu = "ownee" (i.e. legal possession, "thing that is owned", "thing that has an owner"). za "holder", neza "holdee" ("thing that is held", "thing that has a holder").

There is also no grammatical distinction between a possessor and an object, as the equivalent of objects are essentially present as genitive modifiers.

I like this, but Im not sure Im reading it right .... are you saying the possessor aligns syntactically with the patient of a verb? so if I own a car, Im the "object", not the car?

Hmm, not exactly. I don't mean a semantic possessor is equivalent to an object, I mean a syntactic possessor. It depends on what you're saying syntactically.

In "I own a car", I'm the subject. Genitives (or non-explicitly marked modifiers that can optionally be marked as genitives) are red here. Notice that "car" is syntactically the possessor of the word "owner", which is its possessum. If we're talking semantically, the possessor is "I" and "owner", as "I" and "owner" are one and the same entity in this sentence and obviously the car is the semantic possessum.

Na i nyu (ya) dyazo.
1S COP owner (GEN) car
I own a car.
I'm the owner of a car.
I'm a car owner.

What I meant was that transitivity is syntactically identical to (genitive) possession.

Hi e dyazo nga.
DEM COP car GEN.1S
This is my car.

Hi e ngu nga.
DEM COP killer GEN.1S
This kills me.
This is my killer.

You could of course analyse i ngu (e ngu after a word ending in i as above) as a transitive verb meaning "kill" and the word nga as the accusative form of na. It also seems logical to analyse i nyu as a verb meaning "own" and the "ya" in the above sentence as an optional accusative marker (and when it's not there, maybe that's noun-incorporation into the verb!?), but if you do that, for consistency's sake you should then also analyse i dyazo (or e dyazo as a transitive verb meaning "be the car of", which takes the accusative object nga "me" in the above sentence, and that would be, quite frankly, a bit silly. You'd also have to say that removing the i or e is the completely regular way to derive an agent noun (kill → killer, be a car → car "be-er", i.e. car) or as an unexpressed head of a relative clause "That which ...". It's really much more simple to just regard i ngu and i dyazo as COPULA + NOUN and nga and all other apparent objects as genitives modifiers rather than accusative objects, especially as this "object" structure is not limited to the predicate and can occur within the subject too:

Dyazo nga i da.
car GEN.1S COP big.entity
My car is big.

Ngu nga i da.
killer GEN.1S COP big.entity
The one who kills me is big.
My killer is big.

Nyu (ya) dyazo i na.
owner (GEN) car COP 1S
The owner of the car is me.
The car owner is me.
The one who owns the car is me.

I like the three types of genitives too ....

They're not really genitives. They're the equivalent of "have" except as agent nouns. Three different kinds of "havers". There's only one kind of genitive (marked by ya or a few irregular changes to things such as na → nga).

So here we go:
-mb- borrowed, temporarily owned
-amb- beloved, dearly held
-epp- supported, favored by (e.g. "my team" ... i suspect literally no natlangs have a distinct morpheme for this, but for some reason Ive always wanted it)
-mp- created (e.g. i own it because i made it; used for one's children as well)
-p- desired, wanted (i dont own it yet, but i've made up my mind which one i want)

Wow, that's seriously cool!

Also .... can a single noun take more than one possessor argument? e.g. if i leave a huge mess on the carpet, but its your job to clean it up, is it my mess (nyu or za) or yours (ne)? Or both?

I wouldn't use any of nyu, za or ne for that unless I'm saying "I have a mess" or "you have a job to clean it up" and I just can't imagine that being the normal way to express it. As I said, they're not genitives. "My" and "your" would be expressed the same way, with the genitive ya or its irregular "contractions" (ya na → nga; ya wa → wo).

[bradrn]

Firstly, I love the idea of having a purely noun-based language! As it happens, I’ve been thinking about this same idea for the past day or two, and independently ended up at a very similar design to what you have (although I attempted to base it around a monovalent existential rather than a divalent copula).

Ooh, I lack the imagination to see how that would work. How do you say "X is Y"?

I did say ‘attempted’, and you’ve just found the point where I gave up because I couldn’t get it to work. Stuff like ‘I am happy’ can easily enough be said as ‘There exists a happy me’, but you can’t translate equalities such as ‘I am bradrn’ into that form.

On the other hand, it goes the other way around as well. If you only have a copula, how would you translate existentials? I suppose you could use a dummy pronoun as in English, or simply omit the subject, but to me that seems awfully like having a dedicated existential construction rather than representing it using an ordinary copula construction.

And actually, I’ve just thought of a way that you can have no verbs at all! The only verb in your examples is the copula, and it only appears after the first NP, so you can just get rid of it and express copula clauses with pure apposition. That would also give you a simple way to get existentials — just make a clause with only one NP rather than multiple.

Dixon and Aikhenvald’s Possession and Ownership lists six different meanings possessive constructions may have: the ‘core’ meanings are (a) ownership (in terms of control of property), (b) whole-part relations and (c) kinship relations, while less core meanings include (d) association in general (Paul’s dentist), (e) locations/orientations (top of the mountain), and (f) attribution and properties (the man’s temper). In English, ‘have’ may be used for any of these meanings (a)–(f). It looks like your nyu covers (a), and ne covers (c) and possibly some cases of (b),(d),(f); I’m even less about za, but it may be (a),(e). So, if I’m interpreting your three words correctly, you seem to have covered the semantic space fairly well here.

Ah thanks, that's useful. I generally function better with examples, so I'll make some and someone can correct me if it's not exactly what the category meant.

(A) I have a house. Na i nyu mba.
(B) I have a head. Na i ne zyo. The house has a roof. Mba i ne vuzwi.
(C) I have a sister. Na i ne lago.
(D) I have a dentist. Na i ne zyehondi gembwa.
(E) The house has a roof. Mba i ne vuzwi. The mountain has a top. Mwoda i ne vuzwi.
(F) I think this would generally just be expressed more directly as "man is temperamental.person" rather than deriving an abstract noun phrase from it and then saying that he has it. That sounds like two unnecessary steps to get back to the first position. I may find a way to do this that seems useful though and provides some different nuance.

That sounds about right. Note that this is a categorisation of possession rather than having, which is why (F) is best expressed without have.

I see the example "top of the mountain" as being an example of (b), no? I mean, it could be both; parts of something are generally located in the same place as them unless we're talking about something collective and dispersed like an organisation.

My best guess at this point that (B) refers to actual objects, whereas (E) refers to locations: Dixon says that ‘The door of the cabin’ is (B), but ‘The front of the van’ is (E). And I suppose ‘the top of the mountain’ could be either. But I’m not entirely sure for this one.

Za doesn't really seem to be covered unless that is what is meant by (E). … In any case, za is like when someone talks about having something on them or somehow in their possession even if not theirs. If you can talk about having someone else's something, then it's necessarily the za kind of possession. Turkish shows this difference with the genitive case/possessive suffixes for most kinds of possession, but the locative case for locative possession.

(Benim) para-m var.
(my) money-my existent
I have money.
Na i ne nggu.

Bende para var.
at-me money existent
I have money (on me)
Na i za nggu.

(Senin) para-n bende
(your) money-your at-me
I have your money. / Your money is with/on me.
Na i za nggu wo

Good point — now that you explain it, it does look like za isn’t covered!

One other thing I note about this is that your words appear to have some degree of overlap. I see this as possibly having a semantic impact: you can use different possessive words to emphasise particular aspects of your ownership. For instance, you might use nyu to emphasise that this is yours and no-one else’s, whereas you might use ne to emphasise that you did not actively choose to own it. (This reminds me particularly of noun classifiers: in Classifiers, Aikhenvald gives the example from Burmese that you might use myiʔ tə yaʔ ‘river one place’ for a picnic destination, whereas myiʔ tə tan ‘river one line’ gets used for a river on a map.)

Hmm, yeah, but can't really think of any examples that would illustrate that well. Na i ne mba means "I have a house" in the sense that I am not homeless, but if you want to indicate that you own it (not renting for example), you'd say Na i nyu mba. If you're squatting in or occupying a house that is not really "yours", you'd say na i za mba, but you can use ne for this too. I think basically ne is the most versatile of the three. I guess it doesn't explicitly mark non-volition, but the other two generally do imply volition as you can sell your property and release things that you have in your possession.

That’s exactly the sort of thing I was talking about — and I think that’s a pretty good example, actually.

[Imralu]

I did say ‘attempted’, and you’ve just found the point where I gave up because I couldn’t get it to work. Stuff like ‘I am happy’ can easily enough be said as ‘There exists a happy me’, but you can’t translate equalities such as ‘I am bradrn’ into that form.

There exists a bradrny me. ?

On the other hand, it goes the other way around as well. If you only have a copula, how would you translate existentials? I suppose you could use a dummy pronoun as in English, or simply omit the subject, but to me that seems awfully like having a dedicated existential construction rather than representing it using an ordinary copula construction.

Yeah, there is this structure that I use sometimes:

Nwa i mba.
one COP house
There is a house.

We i mba.
PL COP house
There are houses.

Mbwo i zwa.
substance COP water
There is water.

Basically the words nwa, we and mbwo are the least likely to be interpreted as definite in the subject (i.e. not "The one is a house.") so that's why they're used over other words.

But usually I just omit the subject.

I mba. = There's a house. Or: There are houses.

That leaves another possibility for expressing 'have' which is basically exactly how Turkish expresses "have" with possessives:

I mba nga.
COP house GEN.1S
I have a house.
Something is my house.
There is my house.

In a weird way, that's basically the "active" equivalent of the following "passive" sentence with ne.

Na i ne mba.
1S COP PASS house
I have a house.

Compare with these sentences where a passive meaning is formed either via subject omission or by using ne and promoting the genitive modifier to the subject:

I le nga.
COP love^A GEN.1S
I am loved.
Someone loves me.
I have a "lover".

Na i ne le.
1S COP PASS love^A
I am loved.
I have a "lover"

By "lover" I simply mean "someone who loves", not somone who I'm having a (-n extramarital) sexual relationship with. This is the issue with glossing everything as faithfully as possible using agent nouns as agent nouns are incredibly irregular in English and may have unexpected meanings, forms or may simply be absent. A "waiter" is not really just someone who or something that waits (mbo). A shoulder is not someone who or something that should (he). A beer is not simply someone who or that which is. Describing someone who wants as a "wanter" or someone who understands as an "understander" just sounds stupid. So I've started glossing things like this as le "love^A", mbo "wait^A", nwimu "understand^A", he "should^A" with A standing for "agent" as well as da "big^E" with "e" standing for "entity" or "big^O" with "o" standing for "one". It's much more compact and less distracting than other strategies. Writing "love.AG" in the gloss would most likely be interpreted as meaning that le is the agentive form of another word glossed as "love", which it is not - it is an underived, uninflected root. The other alternative would be simply to drop any indications like this, as many people have advised me to in the past, but then the glosses simply don't make sense (mostly thanks to English's huge amount of zero derivation I guess) as people would assume in a sentence like le nga i da "love GEN.1S COP big" that le means the abstract noun "love", making the sentence mean "my love is big", when it actually means "the one who loves me is big", so: love^A GEN.1S COP big^E is the best solution I've come up with that satisfies me, and if someone doesn't like the superscripts and thinks they're unnecessary, they can just feel free to ignore them.

And actually, I’ve just thought of a way that you can have no verbs at all! The only verb in your examples is the copula, and it only appears after the first NP, so you can just get rid of it and express copula clauses with pure apposition. That would also give you a simple way to get existentials — just make a clause with only one NP rather than multiple.

Yeah, I've thought of strategies like that before. I don't like that that would force me to mark at least one of the bounds of a noun phrase, maybe with an article, and I kind of don't like relying on something like "second position". I've thought about marking predicates with something supersegmantal like a tone contour or whatever, or just include a predicative case that functions in exactly in the same way as other nominal cases, but I find the easiest for my purposes with this lang is to just have a tiny copula that usually doesn't even add an extra syllable to the sentence. My goal has never been to have a conlang where I can say "Hey everyone, look, no verbs!" but just not to have to make decisions like "Should I have a basic verb meaning 'annoy' and derive the noun 'nuisance' from it or have a basic noun "nuisance" and derive a verb from it, or a basic adjective 'annoying' and derive both a noun and a verb from it?" I just find these kind of decisions so tedious and completely unnecessary. I know this can bring huge amounts of nuance into different structures and most natlangs take advantage of this, but it's really not necessary and most of the difference between sentences like "He hunts pigs" and "He's a pig hunter" is simply aspectual and there are other ways to express that than dividing the whole lexicon into nouns and verbs just for an aspectual distinction. Having a copula doesn't bother me. In my other main lang, the content words can be most easily thought of as verbs and they can be used in noun phrases by placing them after an article.

That sounds about right. Note that this is a categorisation of possession rather than having, which is why (F) is best expressed without have.

Ah, yes, and now it makes perfect sense why the za kind is missing. Za is about having as in, we use the word "have" for that in (a lot of) languages that have a have-verb, but that kind of possession is not usually expressed in terms of a genitive/possessive construction. If I have a house or a sister, it follows that they're my house and my sister, but if I have your money, it's not my money. If I have a knife on me, it doesn't mean it's necessarily my knife. It could be, but that's not what we're even talking about. It's not saying whose knife it is, only basically where it is or under whose control it is.

My best guess at this point that (B) refers to actual objects, whereas (E) refers to locations: Dixon says that ‘The door of the cabin’ is (B), but ‘The front of the van’ is (E). And I suppose ‘the top of the mountain’ could be either. But I’m not entirely sure for this one.

I guess you could make a philosophical distinction here and say, for example, that "the roof of the house" and "the door of the cabin" are (B) and "top of the mountain" and "front of the car" are (E) because a "roof" and "door" are distinct things, but the "top" of a mountain is not a discrete, identifiable part of a mountain but simply defined by its spatial location in regards to the rest of the mountain. The front of a car is generally identifiable as the "front" even if it's removed or placed somewhere else, but I guess just the fact that the word "front" is spatially defined is enough to make it (E). If I felt inclined to make this kind of distinction in my conlang, I'd think carefully about all cases and decide which structure to use, but I don't and I just can't be bothered thinking about this distinction again.

That’s exactly the sort of thing I was talking about — and I think that’s a pretty good example, actually.

Ah, OK. We can do this kind of thing in English too though, at least the "have" vs. "own" distinction is very easy, and the distinction betwen "I have a knife" and "I carry a knife" is also an option.

[Raphael]

I did say ‘attempted’, and you’ve just found the point where I gave up because I couldn’t get it to work. Stuff like ‘I am happy’ can easily enough be said as ‘There exists a happy me’, but you can’t translate equalities such as ‘I am bradrn’ into that form.

There exists a bradrny me. ?

[...]

[...]

[...]

That leaves another possibility for expressing 'have' which is basically exactly how Turkish expresses "have" with possessives:

I mba nga.
COP house GEN.1S
I have a house.
Something is my house.
There is my house.

Would "bradrn is my name" meet your criteria?

[bradrn]

I did say ‘attempted’, and you’ve just found the point where I gave up because I couldn’t get it to work. Stuff like ‘I am happy’ can easily enough be said as ‘There exists a happy me’, but you can’t translate equalities such as ‘I am bradrn’ into that form.

There exists a bradrny me. ?

Nope, that doesn’t sound quite right to me.

Would "bradrn is my name" meet your criteria?

I suppose that would work well — for that example. But what about things like ‘This man is a doctor’?

On the other hand, it goes the other way around as well. If you only have a copula, how would you translate existentials? I suppose you could use a dummy pronoun as in English, or simply omit the subject, but to me that seems awfully like having a dedicated existential construction rather than representing it using an ordinary copula construction.

Yeah, there is this structure that I use sometimes:

Nwa i mba.
one COP house
There is a house.

We i mba.
PL COP house
There are houses.

Mbwo i zwa.
substance COP water
There is water.

Basically the words nwa, we and mbwo are the least likely to be interpreted as definite in the subject (i.e. not "The one is a house.") so that's why they're used over other words.

But usually I just omit the subject.

I mba. = There's a house. Or: There are houses.

Thanks! This makes a lot of sense — it’s just a dummy NP, so similar to ‘there’ in English ‘there is’.

That leaves another possibility for expressing 'have' which is basically exactly how Turkish expresses "have" with possessives:

I mba nga.
COP house GEN.1S
I have a house.
Something is my house.
There is my house.

I’ve also used this construction in at least one of my conlangs. And actually, now that I check, it looks like I’ve used this for locatives (like your za) as well — that is, stuff like ‘There is food at home’ for ‘The food is at home’.

By "lover" I simply mean "someone who loves", not somone who I'm having a (-n extramarital) sexual relationship with. This is the issue with glossing everything as faithfully as possible using agent nouns as agent nouns are incredibly irregular in English and may have unexpected meanings, forms or may simply be absent. A "waiter" is not really just someone who or something that waits (mbo). A shoulder is not someone who or something that should (he). A beer is not simply someone who or that which is. Describing someone who wants as a "wanter" or someone who understands as an "understander" just sounds stupid. So I've started glossing things like this as le "love^A", mbo "wait^A", nwimu "understand^A", he "should^A" with A standing for "agent" as well as da "big^E" with "e" standing for "entity" or "big^O" with "o" standing for "one". It's much more compact and less distracting than other strategies. Writing "love.AG" in the gloss would most likely be interpreted as meaning that le is the agentive form of another word glossed as "love", which it is not - it is an underived, uninflected root. The other alternative would be simply to drop any indications like this, as many people have advised me to in the past, but then the glosses simply don't make sense (mostly thanks to English's huge amount of zero derivation I guess) as people would assume in a sentence like le nga i da "love GEN.1S COP big" that le means the abstract noun "love", making the sentence mean "my love is big", when it actually means "the one who loves me is big", so: love^A GEN.1S COP big^E is the best solution I've come up with that satisfies me, and if someone doesn't like the superscripts and thinks they're unnecessary, they can just feel free to ignore them.

I don’t know — something like ‘lover’, ‘wanter’, ‘shoulder’ seems fine to me, given that the context is known.

And actually, I’ve just thought of a way that you can have no verbs at all! The only verb in your examples is the copula, and it only appears after the first NP, so you can just get rid of it and express copula clauses with pure apposition. That would also give you a simple way to get existentials — just make a clause with only one NP rather than multiple.

Yeah, I've thought of strategies like that before. I don't like that that would force me to mark at least one of the bounds of a noun phrase, maybe with an article, and I kind of don't like relying on something like "second position".

How would it force you to mark any boundaries? It’s just apposition! And, if you have fairly strict word order within an NP (with a language like this, I assume you do), there’s no ambiguity either.

My best guess at this point that (B) refers to actual objects, whereas (E) refers to locations: Dixon says that ‘The door of the cabin’ is (B), but ‘The front of the van’ is (E). And I suppose ‘the top of the mountain’ could be either. But I’m not entirely sure for this one.

I guess you could make a philosophical distinction here and say, for example, that "the roof of the house" and "the door of the cabin" are (B) and "top of the mountain" and "front of the car" are (E) because a "roof" and "door" are distinct things, but the "top" of a mountain is not a discrete, identifiable part of a mountain but simply defined by its spatial location in regards to the rest of the mountain. The front of a car is generally identifiable as the "front" even if it's removed or placed somewhere else, but I guess just the fact that the word "front" is spatially defined is enough to make it (E). If I felt inclined to make this kind of distinction in my conlang, I'd think carefully about all cases and decide which structure to use, but I don't and I just can't be bothered thinking about this distinction again.

Dixon seems to love categorising these sorts of fine, semi-philosophical semantic distinctions. It certainly makes sense to me to merge (B) and (E) — the closest analogy I can find is instrumental vs comitative, which plenty of languages merge as well.

That’s exactly the sort of thing I was talking about — and I think that’s a pretty good example, actually.

Ah, OK. We can do this kind of thing in English too though, at least the "have" vs. "own" distinction is very easy, and the distinction betwen "I have a knife" and "I carry a knife" is also an option.

I didn’t realise we could do that in English, but now that I think about it, you’re completely right with this example!

[Imralu]

Would "bradrn is my name" meet your criteria?

I suppose that would work well — for that example. But what about things like ‘This man is a doctor’?

How could "bradrn is my name" work with a monovalent existential verb? That's a bivalent copula linking "bradrn" with "my name"!? I don't see how it's any different from "this man is a doctor" except that the former identifies one thing as one other thing, whereas the latter identifies one thing as belonging to a category.

Thanks! This makes a lot of sense — it’s just a dummy NP, so similar to ‘there’ in English ‘there is’.

Yeah, but I generally don't use it though.

I’ve also used this construction in at least one of my conlangs. And actually, now that I check, it looks like I’ve used this for locatives (like your za) as well — that is, stuff like ‘There is food at home’ for ‘The food is at home’.

I wouldn't use it for that though.

Zyamo i lu mba.
food COP LOC house
The food is at home.

I zyamo ye lu mba
COP food ATTR LOC house
There's food at home.

Na i za zyamo ye lu mba.
1S COP hold^A food ATTR LOC house
I've got food at home.

I don’t know — something like ‘lover’, ‘wanter’, ‘shoulder’ seems fine to me, given that the context is known.

Haha, what does this mean then?
He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

What about this?
He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

How would it force you to mark any boundaries? It’s just apposition! And, if you have fairly strict word order within an NP (with a language like this, I assume you do), there’s no ambiguity either.

Because noun phrases in Wena are full of apposition. Without something to mark where they begin or end, you wouldn't know where the subject noun phrase ends and where the predicate noun phrase begins.

Is mba da (house big^E) "big house" (one noun phrase with compounding apposition) or "house is big" (subject noun phrase + predicate noun phrase). Let's try it with four nouns: Would hi le mbingge da (DEM love^A dog big^E) be "this is a big-dog lover" (i.e. "this person loves big dogs" hi e le mbingge da), "this lover is a big dog" (hi le i mbingge da or "this dog lover is big" (hi le mbingge i da)? There's already a lot of lexical vagueness in this language. It really wouldn't be able to handle syntactical vagueness and it also wouldn't be an option to sacrifice these kind of loose compounds, so a marker to separate the subject from the predicate is pretty necessary. Making it an article would mean that both the subject and the predicate noun phrase would have to be marked, but a copula is very economical and can simply sit in between them.

Dixon seems to love categorising these sorts of fine, semi-philosophical semantic distinctions. It certainly makes sense to me to merge (B) and (E) — the closest analogy I can find is instrumental vs comitative, which plenty of languages merge as well.

Instrumental and comitative is a much less fine distinction though.

[KathTheDragon]

How could "bradrn is my name" work with a monovalent existential verb? That's a bivalent copula linking "bradrn" with "my name"!? I don't see how it's any different from "this man is a doctor" except that the former identifies one thing as one other thing, whereas the latter identifies one thing as belonging to a category.

"There_is bradrn, my name"

[bradrn]

Would "bradrn is my name" meet your criteria?

I suppose that would work well — for that example. But what about things like ‘This man is a doctor’?

How could "bradrn is my name" work with a monovalent existential verb? That's a bivalent copula linking "bradrn" with "my name"!? I don't see how it's any different from "this man is a doctor" except that the former identifies one thing as one other thing, whereas the latter identifies one thing as belonging to a category.

Oops, you’re totally right. I shouldn’t make these posts when I’m not concentrating well.

I don’t know — something like ‘lover’, ‘wanter’, ‘shoulder’ seems fine to me, given that the context is known.

Haha, what does this mean then?
He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

No idea, but only because I’m not at all familiar with the syntax of your language. Besides, there’s plenty of natlangs where it’s hard to deduce meaning from a gloss e.g. here’s a Burmese sentence:

be

which

ha

thing

hma

even

ma

NEG

cai.

like

hpu꞉

NEG

What about this?
He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

Are you sure the gloss is correct for this one?

How would it force you to mark any boundaries? It’s just apposition! And, if you have fairly strict word order within an NP (with a language like this, I assume you do), there’s no ambiguity either.

Because noun phrases in Wena are full of apposition. Without something to mark where they begin or end, you wouldn't know where the subject noun phrase ends and where the predicate noun phrase begins.

Is mba da (house big^E) "big house" (one noun phrase with compounding apposition) or "house is big" (subject noun phrase + predicate noun phrase). Let's try it with four nouns: Would hi le mbingge da (DEM love^A dog big^E) be "this is a big-dog lover" (i.e. "this person loves big dogs" hi e le mbingge da), "this lover is a big dog" (hi le i mbingge da or "this dog lover is big" (hi le mbingge i da)? There's already a lot of lexical vagueness in this language. It really wouldn't be able to handle syntactical vagueness and it also wouldn't be an option to sacrifice these kind of loose compounds, so a marker to separate the subject from the predicate is pretty necessary. Making it an article would mean that both the subject and the predicate noun phrase would have to be marked, but a copula is very economical and can simply sit in between them.

That makes sense — if you have freer word order you do get a lot of ambiguity.

Dixon seems to love categorising these sorts of fine, semi-philosophical semantic distinctions. It certainly makes sense to me to merge (B) and (E) — the closest analogy I can find is instrumental vs comitative, which plenty of languages merge as well.

Instrumental and comitative is a much less fine distinction though.

That’s why I said it’s the ‘closest’ analogy I can find rather than an ‘exact’ analogy.

[Imralu]

How could "bradrn is my name" work with a monovalent existential verb? That's a bivalent copula linking "bradrn" with "my name"!? I don't see how it's any different from "this man is a doctor" except that the former identifies one thing as one other thing, whereas the latter identifies one thing as belonging to a category.

"There_is bradrn, my name"

Hmm, yeah, well then, you could also do "There's me, a doctor." and if it can be used like "There's X, a Y", then you may as well just analyse it as a copula meaning "X is a Y".

No idea, but only because I’m not at all familiar with the syntax of your language. Besides, there’s plenty of natlangs where it’s hard to deduce meaning from a gloss e.g. here’s a Burmese sentence:

be

which

ha

thing

hma

even

ma

NEG

cai.

like

hpu꞉

NEG

Yeah, good point, but in that sentence, my main sticking point is that I don't know if cai means "like" as in "be fond of" or "similar to" and I prefer glosses that would make that distinction. Of course, when you have the translation with it as well, it's generally clear. What does that mean, by the way? What don't you like? Which of these things is not like the others?

Haha, what does this mean then?
He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

What about this?
He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

Are you sure the gloss is correct for this one?

Yeah. I just deliberately made it ambiguous. One different word in the sentence but the same gloss. The first one means "The one who should love me is waiting for the waiter to bring pork shoulder." The second one means "The one who should love me is waiting for the waiter to bring what should be pig." I'd prefer to gloss these like this:

He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
should^A love^A GEN.1S COP wait^A GEN.C waiter COP bring^A GEN shoulder pig
The one who should love me is waiting for the waiter to bring pork shoulder.

and

He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
should^A love^A GEN.1S COP wait^A GEN.C waiter COP bring^A GEN should^A pig
The one who should love me is waiting for the waiter to bring what should be pig.

That makes sense — if you have freer word order you do get a lot of ambiguity.

It's not even that free in the way people usually mean with "free word order". Reversing things changes the meaning and it's actually pretty strict to produce a given meaning, but the relatively free compounding (or juxtaposition) means that the reverse word order is usually also grammatically possible, even though it may make no semantic sense. It often does make sense though, just a different sense. Like zi mba (past^E house) means "former house", "that which was a house" etc., but mba zi (house past^E) would mean more like a "house of former things" (whence the "tight" compound mbazi "museum").

[bradrn]

How could "bradrn is my name" work with a monovalent existential verb? That's a bivalent copula linking "bradrn" with "my name"!? I don't see how it's any different from "this man is a doctor" except that the former identifies one thing as one other thing, whereas the latter identifies one thing as belonging to a category.

"There_is bradrn, my name"

Hmm, yeah, well then, you could also do "There's me, a doctor." and if it can be used like "There's X, a Y", then you may as well just analyse it as a copula meaning "X is a Y".

That’s why I dislike using these special constructions in languages like this — there’s very little difference between ‘language with exactly one verb plus a special copulative construction’ and ‘language with two verbs’.

No idea, but only because I’m not at all familiar with the syntax of your language. Besides, there’s plenty of natlangs where it’s hard to deduce meaning from a gloss e.g. here’s a Burmese sentence:

be

which

ha

thing

hma

even

ma

NEG

cai.

like

hpu꞉

NEG

Yeah, good point, but in that sentence, my main sticking point is that I don't know if cai means "like" as in "be fond of" or "similar to" and I prefer glosses that would make that distinction. Of course, when you have the translation with it as well, it's generally clear. What does that mean, by the way? What don't you like? Which of these things is not like the others?

Translation: (I) don’t like anything. Burmese glosses (or at least the ones in the grammar I’m using) seem to get really ambiguous, especially when they use zero-anaphora. Some other highlights:

thu.

3p.G

pai'hsan

money

pyan

again

pei꞉

give

lai'

away

pa

POL

More: show

Please give his money back (to him).

hna.

I

hsi.

place

ḵou

ALL

hsan

rice

lei꞉

four

loun꞉

CLF

win

enter

ṯe.

REL

poun꞉

container

yu

bring

hke.

distal

More: show

Bring me the can that holds four tins of rice.

Haha, what does this mean then?
He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

What about this?
He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
shoulder lover GEN.1S COP waiter GEN.C waiter COP bringer GEN shoulder pig

Are you sure the gloss is correct for this one?

Yeah. I just deliberately made it ambiguous. One different word in the sentence but the same gloss. The first one means "The one who should love me is waiting for the waiter to bring pork shoulder." The second one means "The one who should love me is waiting for the waiter to bring what should be pig." I'd prefer to gloss these like this:

He le nga i mbo zyu zazozyamo i zazo ya ndingye mbimbi.
should^A love^A GEN.1S COP wait^A GEN.C waiter COP bring^A GEN shoulder pig
The one who should love me is waiting for the waiter to bring pork shoulder.

and

He le nga i mbo zyu zazozyamo i zazo ya he mbimbi.
should^A love^A GEN.1S COP wait^A GEN.C waiter COP bring^A GEN should^A pig
The one who should love me is waiting for the waiter to bring what should be pig.

Now that was just mean. But I still think that ‘shoulder’ etc. in glosses are fine — they get disambiguated by the translations, just like my Burmese examples.

That makes sense — if you have freer word order you do get a lot of ambiguity.

It's not even that free in the way people usually mean with "free word order". Reversing things changes the meaning and it's actually pretty strict to produce a given meaning, but the relatively free compounding (or juxtaposition) means that the reverse word order is usually also grammatically possible, even though it may make no semantic sense. It often does make sense though, just a different sense. Like zi mba (past^E house) means "former house", "that which was a house" etc., but mba zi (house past^E) would mean more like a "house of former things" (whence the "tight" compound mbazi "museum").

Ah, that makes sense then.

[Imralu]

Translation: (I) don’t like anything. Burmese glosses (or at least the ones in the grammar I’m using) seem to get really ambiguous, especially when they use zero-anaphora. Some other highlights:

thu.

3p.G

pai'hsan

money

pyan

again

pei꞉

give

lai'

away

pa

POL

More: show

Please give his money back (to him).

hna.

I

hsi.

place

ḵou

ALL

hsan

rice

lei꞉

four

loun꞉

CLF

win

enter

ṯe.

REL

poun꞉

container

yu

bring

hke.

distal

More: show

Bring me the can that holds four tins of rice.

Ah, yeah. What is the G ub 3p.G?

Now that was just mean. But I still think that ‘shoulder’ etc. in glosses are fine — they get disambiguated by the translations, just like my Burmese examples.

I can just imagine if I posted this anywhere ...

He mbo!
shoulder waiter
Wait!

... that people would be confused and one member here may even react angrily.

A noun phrase on its own with no copula is an "appellative" clause, essentially with an implied "You are ..." in the sentence, just like when someone says "Idiot!" The only examples I can think of for this in English are insults. Anyway, this makes the difference between wa i he mbo (2S COP shoulder waiter) "You should wait" and he mbo (shoulder waiter) "Wait!" and a very forceful imperative can even be made as more or less just an indicative with just (ga) mbo (future^E waiter) "You (will) wait!"

BTW, if I'm going to gloss by just indiscriminately adding -er to verbs, do you have a suggestion for how to gloss "entity" into words like da (big^E), lu (located^E), nggwo (southern^E), ga (future^E) etc.? If I can't find a compact solution, I think I'm just going to stick with the superscript E and then I may as well continue using a superscript A on words glossed as English verbs.

[bradrn]

Translation: (I) don’t like anything. Burmese glosses (or at least the ones in the grammar I’m using) seem to get really ambiguous, especially when they use zero-anaphora. Some other highlights:

thu.

3p.G

pai'hsan

money

pyan

again

pei꞉

give

lai'

away

pa

POL

More: show

Please give his money back (to him).

hna.

I

hsi.

place

ḵou

ALL

hsan

rice

lei꞉

four

loun꞉

CLF

win

enter

ṯe.

REL

poun꞉

container

yu

bring

hke.

distal

More: show

Bring me the can that holds four tins of rice.

Ah, yeah. What is the G ub 3p.G?

G is genitive, as marked by an induced creaky tone (transcribed with a full stop).

(Also, I assume that ‘ub’ is a mistyped ‘’ here? It took me a while to figure out what you meant by a ‘G ub 3p.G’…)

Now that was just mean. But I still think that ‘shoulder’ etc. in glosses are fine — they get disambiguated by the translations, just like my Burmese examples.

I can just imagine if I posted this anywhere ...

He mbo!
shoulder waiter
Wait!

... that people would be confused

Yes, that’s definitely a confusing sentence.

and one member here may even react angrily.

Are you serious? I appreciate that I’m a relative newcomer here, but I find it hard to believe that anyone here who would actually get angry at a sentence with an odd gloss.

(Anyway, for me, ‘other people might get angry’ isn’t really sufficient motivation to change something like this.)

BTW, if I'm going to gloss by just indiscriminately adding -er to verbs, do you have a suggestion for how to gloss "entity" into words like da (big^E), lu (located^E), nggwo (southern^E), ga (future^E) etc.? If I can't find a compact solution, I think I'm just going to stick with the superscript E and then I may as well continue using a superscript A on words glossed as English verbs.

da big.thing
nggwo southerner
ga future (which is a noun in English too)

Not sure what lu would mean in English, so I can’t give a gloss for that one yet.

[Imralu]

G is genitive, as marked by an induced creaky tone (transcribed with a full stop).

(Also, I assume that ‘ub’ is a mistyped ‘’ here? It took me a while to figure out what you meant by a ‘G ub 3p.G’…)

Ah, I would have understood "GEN", but if you're using it a lot and you never need to gloss just "gender" (e.g. G4 = fourth gender), works well.

Yeah, ub was meant to be in, but I guess my right hand was misplaced.

He mbo!
shoulder waiter
Wait!

... that people would be confused

Yes, that’s definitely a confusing sentence.

Do you think should^A wait^A would be a less confusing gloss?

and one member here may even react angrily.

Are you serious? I appreciate that I’m a relative newcomer here, but I find it hard to believe that anyone here who would actually get angry at a sentence with an odd gloss.

Yeah, well, he's not around much these days and I don't think he's necessarily angry, but the words he uses seem very angry if you do things he doesn't believe are necessary. It's just tiring. For a while he was railing against people writing ä in narrow transcription for the low central vowel, even though that's exactly what is prescribed by the IPA and insisted that "a" (the symbol for the low front vowel) was close enough (for phonemic transcription, sure, but not for narrow). He told me "Stop it!" when I was using quotation marks around the translation under my glosses. (I assume that's what he wanted me to stop.) Anyway, any time I gloss with more high fidelity, it tends to bring out the people who think it's impossible not to have a lexical distinction between nouns and verbs. Usually they misquote a bit of Zompist's CCK at me as if I've never read that or thought about it. The member in question once went on a bit of a rant because I was using "be" in my glosses of my conlang where the content words are more verby. For example, I'd gloss things like oko be.dog, tta be.large etc., and I was told that these cannot truly be verbs and that what is happening here is that there is a zero-copula. I'd love for them to go through my lexicon and identify all of the content words which include a zero copula and those which don't - I have a funny feeling that they'd decide that there is a zero copula with any word that is translated with an English noun or adjective, and that the answers they reach would depend on my translation (e.g. "nuisance" or "annoy", "teach" or "teacher"). I started simply dropping "be" in these glosses most of the time just to avoid everything getting derailed with tedious conversations. It is more compact, but it's less information.

(Anyway, for me, ‘other people might get angry’ isn’t really sufficient motivation to change something like this.)

It means I don't enjoy being here or contributing when that happens.

da big.thing
nggwo southerner
ga future (which is a noun in English too)

Not sure what lu would mean in English, so I can’t give a gloss for that one yet.

Yeah, that's much less compact. How do you find these?

Mba wo i nggwo zyi nga.
house GEN.2S COP southerner GEN.DEF GEN.1S
Your house is to the south of mine.

Na i zongwo nggwo.
1S COP walker southerner
I'm walking southwards.

Nyoga i ga.
future COP future
The future will come

Ga i ga.
future COP future
What will be will be.
What will come will come.
Que será será.

Nyoga i nyoga.
future COP future
The future is the future.

If I'm going to gloss ga as "future", I may as well just gloss it as "FUT" - compact and it will distinguish it from nyoga "future". If I do that, I may as well just gloss nggwo as "south" and lu as "LOC"

And here is lu with my superscript notation and then my "fuck it, more compact, less information" gloss.

Na i ga lu.
1S COP future^E located^E
1S COP FUT LOC
I will be there / present / in attendance

Lu mba i ba da u gi e lu zwazo.
located^E house COP extreme^E large^E ADV mild^E COP located^E river
LOC house very big ADV mild COP LOC river
The one in/at the house is bigger than the one in/at the river.

Na i nggwa lwe u lu mba zyi lazye wo.
1S COP habitual^E sleep^A ADV located^E house GEN.DEF mother GEN.2S
1S COP HAB sleep ADV LOC house GEN.DEF mother GEN.2S
I sleep at your mother's house.

Na i ze mwe lwe wi lu.
1S COP no^E able^E/can^A sleep^A ADV-COP located^E
1S COP NEG can sleep ADV-COP LOC
I can't sleep when there's someone there.

De i lu ma?
DEF COP located^E what/who
DEF COP LOC what/who
Where is he/she/it?

De i nggwo.
DEF COP southern^E
DEF COP south
He/she/it is in the south.

With the "fuck it" kind of glossing though, it's not clear for example that lu nggwo would be redundant (essentially meaning "entity which is located at an entity located in the south"). I've been using "fuck it" glossing most of the time on the board, but for teaching materials or whatever (which I'm just writing for myself and never really intend to share with anyone), my glosses are as clear as possible.

[bradrn]

He mbo!
shoulder waiter
Wait!

... that people would be confused

Yes, that’s definitely a confusing sentence.

Do you think should^A wait^A would be a less confusing gloss?

Possibly, but then I’d just wonder what ^A means. (I’m still not entirely clear on that, actually.)

and one member here may even react angrily.

Are you serious? I appreciate that I’m a relative newcomer here, but I find it hard to believe that anyone here who would actually get angry at a sentence with an odd gloss.

Yeah, well, he's not around much these days and I don't think he's necessarily angry, but the words he uses seem very angry if you do things he doesn't believe are necessary. It's just tiring. For a while he was railing against people writing ä in narrow transcription for the low central vowel, even though that's exactly what is prescribed by the IPA and insisted that "a" (the symbol for the low front vowel) was close enough (for phonemic transcription, sure, but not for narrow). He told me "Stop it!" when I was using quotation marks around the translation under my glosses. (I assume that's what he wanted me to stop.) Anyway, any time I gloss with more high fidelity, it tends to bring out the people who think it's impossible not to have a lexical distinction between nouns and verbs. Usually they misquote a bit of Zompist's CCK at me as if I've never read that or thought about it. The member in question once went on a bit of a rant because I was using "be" in my glosses of my conlang where the content words are more verby. For example, I'd gloss things like oko be.dog, tta be.large etc., and I was told that these cannot truly be verbs and that what is happening here is that there is a zero-copula. I'd love for them to go through my lexicon and identify all of the content words which include a zero copula and those which don't - I have a funny feeling that they'd decide that there is a zero copula with any word that is translated with an English noun or adjective, and that the answers they reach would depend on my translation (e.g. "nuisance" or "annoy", "teach" or "teacher"). I started simply dropping "be" in these glosses most of the time just to avoid everything getting derailed with tedious conversations. It is more compact, but it's less information.

(Anyway, for me, ‘other people might get angry’ isn’t really sufficient motivation to change something like this.)

It means I don't enjoy being here or contributing when that happens.

Wow, that sounds incredibly annoying. I can definitely understand now why you would want to avoid a repeat of that.

da big.thing
nggwo southerner
ga future (which is a noun in English too)

Not sure what lu would mean in English, so I can’t give a gloss for that one yet.

Yeah, that's much less compact. How do you find these?

…

If I'm going to gloss ga as "future", I may as well just gloss it as "FUT" - compact and it will distinguish it from nyoga "future". If I do that, I may as well just gloss nggwo as "south" and lu as "LOC"

And here is lu with my superscript notation and then my "fuck it, more compact, less information" gloss.

…

With the "fuck it" kind of glossing though, it's not clear for example that lu nggwo would be redundant (essentially meaning "entity which is located at an entity located in the south"). I've been using "fuck it" glossing most of the time on the board, but for teaching materials or whatever (which I'm just writing for myself and never really intend to share with anyone), my glosses are as clear as possible.

I think that in general I prefer to be as explicit as possible with my glosses, even at the expense of compactness. After thinking about it a bit, here’s how I think I would gloss some of your examples:


Mba wo i nggwo zyi nga.
house GEN.2S COP southern GEN.DEF GEN.1S
Your house is to the south of mine.

Na i zongwo nggwo.
1S COP walker southern
I'm walking southwards.

Nyoga i ga.
future COP future
The future will come

Lu mba i ba da u gi e lu zwazo.
located.thing house COP extreme large.thing ADV mild.thing COP located.thing river
The one in/at the house is bigger than the one in/at the river.

De i nggwo.
DEF COP southern
He/she/it is in the south.

?lu nggwo
located.thing southern

True, all the .things do get repetitive, but they also give a much better idea of what the language is really like.

[quinterbeck]

This is great!

Personally, I think your superscript notation is perfectly transparent, and helps me to parse the sentences correctly.

Zyamo i lu mba.
food COP LOC house
The food is at home.

I zyamo ye lu mba
COP food ATTR LOC house
There's food at home.

Na i za zyamo ye lu mba.
1S COP hold^A food ATTR LOC house
I've got food at home.

What function is ATTR serving in these? Is it grammatical to say

I zyamo lu mba
COP food LOC house

and what would it mean if so?

And here is lu with my superscript notation and then my "fuck it, more compact, less information" gloss.

Na i ga lu.
1S COP future^E located^E
1S COP FUT LOC
I will be there / present / in attendance

Lu mba i ba da u gi e lu zwazo.
located^E house COP extreme^E large^E ADV mild^E COP located^E river
LOC house very big ADV mild COP LOC river
The one in/at the house is bigger than the one in/at the river.

Na i nggwa lwe u lu mba zyi lazye wo.
1S COP habitual^E sleep^A ADV located^E house GEN.DEF mother GEN.2S
1S COP HAB sleep ADV LOC house GEN.DEF mother GEN.2S
I sleep at your mother's house.

Na i ze mwe lwe wi lu.
1S COP no^E able^E/can^A sleep^A ADV-COP located^E
1S COP NEG can sleep ADV-COP LOC
I can't sleep when there's someone there.

De i lu ma?
DEF COP located^E what/who
DEF COP LOC what/who
Where is he/she/it?

De i nggwo.
DEF COP southern^E
DEF COP south
He/she/it is in the south.

What is your ADV particle doing exactly? It seems to me that it;s either heading predicates, or heading a noun which is then modified with a copula+argument phrase. I'm noticing that in Lu mba i ba da u gi e lu zwazo the second copula is e rather than i, what's going on there?

What if I substituted ADV for ATTR and got

Na i nggwa lwe ye lu mba zyi lazye wo.
1S COP habitual^E sleep^A ATTR located^E house GEN.DEF mother GEN.2S

Is that grammatical? If so what would it mean?

I'm very curious to know more about the syntax involved!

[Imralu]

Yes, that’s definitely a confusing sentence.

Do you think should^A wait^A would be a less confusing gloss?

Possibly, but then I’d just wonder what ^A means. (I’m still not entirely clear on that, actually.)

Yeah, I'll explain that in anything I write. I guess I'll change my signature slightly to reflect it. It's my way of indicating the agent nount of the glossed word, so I can gloss mwe as "can^A" rather than "entity.that.can" or making awkward things like "canner" and "shoulder" for "entity.that.can" or "entity.that.should". I think .AG is the normal way to gloss this, except I think that glosing he as should.AG may imply that he is an agent noun derived from some other word in Wena, when it is in fact an underived root, and what I want to do is merely indicate that the translation in the gloss should be understood as whatever agent noun we can imagine for the word "should". A superscript ^A needs an explanation anyway as it's not part of the Leipzig rules, but I like that it's much more compact and makes for much more readable glosses, I think.

I think that in general I prefer to be as explicit as possible with my glosses, even at the expense of compactness.

Interesting that you say that but then you still glossed both ga and nyoga as "future". Ga is basically "entity which will be", whereas nyoga is just the word for the future. I could also gloss ga as "will^A".

Anyway, I forgot to mention it, but I prefer "entity" over "thing" because the latter implies inanimacy and I suppose the only implication of the word "entity" that is unwanted is one of countability, but it doesn't feel like a very strong implication to me.

True, all the .things do get repetitive, but they also give a much better idea of what the language is really like.

Yeah, but that's why I'd rather use ^E. It requires an explanation of course, but once it's there, the glosses are, I think, easier to read.

This is great!

Personally, I think your superscript notation is perfectly transparent, and helps me to parse the sentences correctly.

Thanks!

Zyamo i lu mba.
food COP LOC house
The food is at home.

I zyamo ye lu mba
COP food ATTR LOC house
There's food at home.

Na i za zyamo ye lu mba.
1S COP hold^A food ATTR LOC house
I've got food at home.

What function is ATTR serving in these? Is it grammatical to say

I zyamo lu mba
COP food LOC house

and what would it mean if so?

Yeah, the ye is not strictly necessary there. Without it, we basically have a loose compound made up of several nouns. In some cases, the meaning could be quite different with or without ye, as ye specifies "which is", which is just one of the multiple possibilities of freely compounded nouns. For example, mo ye da (consume^A ATTR big^E) means an eater/drinker who is big, whereas a mo da (consume^A big^E) could also be an eater of big things or, by extention, someone who eats a lot.

In this case, zyamo lu mba could potentially be interpreted to mean more or less a different kind of food, something called "at-home food" or something, whereas zyamo ye lu mba is very clearly nothing other than food which is at home. We could go even further and have za zyamo lu mba, a compound noun phrase of four nouns basically meaning an "at-home food haver". You can often put in ye and ya in these kind of compounds to disambiguate how the elements should relate to each other: za (ya) zyamo (ye) lu (ya?) mba "haver of food which is a located thing (of?) a house". Lu ya mba is possible, but pretty weird. In compound noun phrases, the main stress always falls on the last noun, so dividing things up with particles allows other elements to be stressed. As phrases become more lexicalised, these particles get dropped and when fully lexicalised, "tight" compounds are formed, e.g. ndwé ya díngo "protector of child(ren)" → ndwe díngo → ndwédingo "babysitter", "nanny". I've marked the stresses with acute accents.

I zyamo u lu mba would probably be the most usual way to say "there is food at home", but I deliberately avoided writing it with u simply so I didn't confuse anyone by changing that u to ye in the third sentence, as having Na i za would instantly make an u phrase have a different interpretation, although again, the difference is fairly subtle.

Na i za zyamo u lu mba.
1S COP hold^A food ADV located^A house
I have food (when I'm) at home.

In this sentence, "at home" tells the location of my having, not simply the location of the food. It seems like a very subtle difference, but it would kind of imply that I have food when I'm at home. Since I'm presumably telling you "I have food at home" while I'm not at home, it doesn't mean that I necessarily have food now as "at home" is the location of my having food. In the existential sentence i zyamo u lu mba it's perfectly unambiguous that "at home" is the location of the food (to be incredibly anal: the location of the foods existence), but to if I use na i za and want to make it clear that I'm only talking about the location of the food itself, I would either use ye or this slightly more complex sentence with u:

Na i za zyamo u de i lu mba.
1S COP hold^A food ADV DEF COP located^A house
I have food at home.

This sentence essentially means "I have food, with it being at home." Because I didn't really want to get into these little details yesterday, I just decided to use ye in all three sentences. If any of that is confusing, I'm about to explain u.

Lu mba i ba da u gi e lu zwazo.
located^E house COP extreme^E large^E ADV mild^E COP located^E river
LOC house very big ADV mild COP LOC river
The one in/at the house is bigger than the one in/at the river.

Na i nggwa lwe u lu mba zyi lazye wo.
1S COP habitual^E sleep^A ADV located^E house GEN.DEF mother GEN.2S
1S COP HAB sleep ADV LOC house GEN.DEF mother GEN.2S
I sleep at your mother's house.

Na i ze mwe lwe wi lu.
1S COP no^E able^E/can^A sleep^A ADV-COP located^E
1S COP NEG can sleep ADV-COP LOC
I can't sleep when there's someone there.

What is your ADV particle doing exactly? It seems to me that it;s either heading predicates, or heading a noun which is then modified with a copula+argument phrase. I'm noticing that in Lu mba i ba da u gi e lu zwazo the second copula is e rather than i, what's going on there?

E is simply the form of the copula after a word ending in i. After a word ending in u, the adverbial particle also dissimilates to become o. When u i (ADV COP) would occur in sequence, we get wi.

There are essentially three possible structures with the adverbial particle and it's easiest to explain by showing a whole sentence structure. U can either introduce a noun phrase understood as an adverb (sentence type (1)), or a clause understood as an adverb (sentence types (2) and (3)).

(1)
A i B u C
A is B while also being C. / A is B, with A being C. (Since A and B are equivalent, it's more or less saying that A, B and C are all descriptions of the same thing, but B is generally the focus description and C is more background information that modifies how the "A is B" relationship is understood).

(2)
A i B u C i D
A is B, with C being D.

(2)
A i B wi C
A is B, with there being C.

It's not quite that simple as the adverbial clause is also explicitly relevant to the relationship between A and B. It's generally not just an addition of information. For example, the type one sentence ...

Na i ngebwi u nwiho.
1S COP sing^A ADV happy^E
I sing happily.

... does not simply mean "I sing AND I am also happy" but that the happiness is relevant to the singing.

The type two sentence ...

Na i ngebwi u byo i wa.
1S COP sing^A ADV topic COP 2S
I sing about you.

... makes it explicitly clear that my singing is about you, not that you are just some other contextually understood topic, but that you are the topic relevant to my being a singer.

The type three sentence ...

Na i vwe zyi wi gyazi.
1S COP see^A GEN.DEF ADV-COP previous.day
I saw him/her/it yesterday.

... does not simply assert the existence of yesterday independent of me seeing him/her/it (like "I see it, oh and also yesterday existed), but it provides the time frame for my being a seer of him/her/it.

The comparative form in that sentence is possibly what's confusing you. Basically, there are no dedicated comparative words. To compare degrees of something, you have to correlate ba (extreme^E) with gi (mild^E), with one being in an adverbial clause.

Na i ba da u gi e wa.
1S COP extreme^E large^E ADV mild^E COP 2S
I'm bigger than you.

Translated more literally, it's basically like saying "I'm very large, with what is slightly (large) being you." The adverbial clause could also be u gi da i wa, but because da appeared after ba, that is what gi is assumed to be modifiying, so it can be dropped from that clause. Again, the adverbial phrase modifies the main clause, meaning that it doesn't indicate that I am "very large" or that you are "slightly large" in any more objective sense. Maybe neither of us are large at all. All I'm saying is that I'm larger than you and "-er than ..." is basically expressed as ba ... u gi e .... "Less ... than ..." is expressed as gi ... u ba i ....

Na i ba le wo u gi (le) zyi.
1S COP extreme^E love^A GEN.2S ADV mild^E (love^A) GEN.DEF
I love you more than I love him.
("I intensely love you while mildly (loving) him.")

Na i ba le wo u gi (le wo) e(/i) de.
1S COP extreme^E love^A GEN.2S ADV mild^E (love^A GEN.2S) COP GEN.DEF
I love you more than he (does/loves you).
("I intensely love you with the one who mildly (loves you) being him.")

What if I substituted ADV for ATTR and got

Na i nggwa lwe ye lu mba zyi lazye wo.
1S COP habitual^E sleep^A ATTR located^E house GEN.DEF mother GEN.2S

Is that grammatical? If so what would it mean?

Yeah, that's also grammatical and the meaning is more or less the same in this case because the word lwe before ye is still just another description of the subject na. Here's an example where it would be very different.

Na i zi vwe ya ngo ye lu mi ya mba.
1S COP past^E see^A GEN person ATTR located^E interior GEN house
I saw someone (who was) inside the house.

Na i zi vwe ya ngo u lu mi ya mba.
1S COP past^E see^A GEN person ADV located^E interior GEN house
I saw someone (when I was) inside the house.

Na i gumbe ya hu ye ba nda.
1S COP punch^A GEN man ATTR extreme^E strong^E
I punched a very strong man.

Na i gumbe ya hu o ba nda.
1S COP punch^A GEN man ADV extreme^E strong^E
I punched a man very strongly.

[quinterbeck]

Aha, very interesting! You're right that the comparative sentence was confusing me.

What distinguishes ya and ye? Are there other particles that describe noun relationships?