Conlang Random Thread

[Xwtek]

Also, is there any improvement using pants instead of (short-flapped) loincloth when riding a horse (Or in this conworld horse-dog, i.e. pony-sized dog. In Sakha, this creature is called xwtek)?

If your thighs are naked, wouldn't they chafe against the animal?

Okay. Then people are wearing pants when riding. Loincloths were worn when farming. Xwtek is hairier than horse and this may cause chafing problem even worse. (The climate is below 10 degrees Celsius in average, and may drop below zero during winter).

Also, is there any adaptation for cold with only pants? My sentient creature is human in name only (they use magic, and in general looks more like a brown-skinned elf).

[Qwynegold]

I can't remember exactly, but I think it was that if the S argument is typically an agent, but has absolutive case, then it's morphologically ergative but semantically nominative-accusative.

Are you sure you didn’t mean the A argument? Because the S argument is the argument of an intransitive verb, and so you don’t have any choice in how it’s case-marked: it will always be marked with the absolutive in an ergative language, whether it’s ‘semantically nominative-accusative’ or ‘semantically ergative-absolutive’.

No, I did mean the S argument. I don't know, it seems like this would imply that sentences like "1SG-ABS run" are not normally used in ergative languages.

So, using your terminology, Chukchi is ‘semantically nominative-accusative’.)

It's not my terminology, it's something I was told on the forum. But it was a long time ago, so I might be misremembering something. Or the person who said that could've been mistaken.

This can be summarized in a chart like this:

Code: Select all

           Intr-A Intr-P Tr-A Tr-P
Monovalent 1,X    2,X    3,X  4,PASS
Divalent   *6,Y   7,CAUS 5,Y  5,Y

Intr-A means an intransitive verb which takes an agent-like S. Intr-P means an intransitive which takes a patient-like S. Tr-A means a transitive verb with at least an agent argument. Tr-P means a transitive verb with at least a patient argument.

The Y thingy can only be used on verbs that are intransitive by default. The X thingy is normally used on intransitives, but it can also be used on a transitive verb to remove the object.

Does all of this make sense? What should I call X and Y?

For the most part, this makes sense, after a couple of read-throughs. The only things I’m confused about are:

• You say that Y can only be applied to intransitive verbs. However, your glosses imply that sing is a transitive verb here, yet you apply Y to it. Was this meant to read ‘Y … can only be used on verbs that are transitive by default’?

Oh no. *headdesk* It was supposed to be Y can only be applied to transitive verbs. I'll change the original post...

• What exactly does Y do? You say that X simply marks an intransitive verb, and removes the object of a transitive verb, but you never say what Y does.

Y just marks that a verb is indeed divalent. But it's zero-marked.

• How do you translate a simple sentence like ‘I see you’, or ‘I break the window’? For these neither X nor Y seem to apply.

1SG-ERG 2SG-ABS see.TR-Y
I see you.

1SG-ERG window-ABS break.TR-Y
I break the window.

• What exactly makes sentence 6 (*1SG-ERG marathon-ABS run.INTR-Y ‘I run a marathon’) illegal? How can that sentence be translated legally?

Because I decided to have such a constraint that you can't use Y on intransitives. That sentence you'd have to paraphrase or say something like 1SG-ABS marathon-PROL run.INTR-Y (prolative) or 1SG-ABS marathon-ADE run.INTR-Y (adessive).

As for naming: I’m not sure about Y until the above issues are clarified, but X applied to transitive verbs is pretty clearly an antipassive , which is very typical of ergative languages. X applied to intransitive verbs seems could be called an ‘intransitivity marker’, although that would be pretty weird — I believe that transitivity markers are more usual.

Heh, I was thinking like "have I reinvented an antipassive or mediopassive voice or something?" So there are two uses for X. Should I treat it like two different verb categories that are marked with homonymous morphemes, or should I just call X the antipassive and make a note that "it's only really an antipassive when used like this and not when used like that"?

Thanks for taking your time answering my questions!

[Qwynegold]

Xwtek is hairier than horse and this may cause chafing problem even worse.

Hmm, I think hair might actually decrease chafing. This is why people have pubic hairs. Or so I've heard.

Also, is there any adaptation for cold with only pants?

Sorry, I don't quite understand what you're asking here.

[bradrn]

I can't remember exactly, but I think it was that if the S argument is typically an agent, but has absolutive case, then it's morphologically ergative but semantically nominative-accusative.

Are you sure you didn’t mean the A argument? Because the S argument is the argument of an intransitive verb, and so you don’t have any choice in how it’s case-marked: it will always be marked with the absolutive in an ergative language, whether it’s ‘semantically nominative-accusative’ or ‘semantically ergative-absolutive’.

No, I did mean the S argument. I don't know, it seems like this would imply that sentences like "1SG-ABS run" are not normally used in ergative languages.

Ergative languages are distinguished by using the absolutive case in intransitive sentences. If it doesn’t do this, then it’s not an ergative language. So: Yes, sentences like “1SG-ABS run” are normally used in ergative languages.

So, using your terminology, Chukchi is ‘semantically nominative-accusative’.)

It's not my terminology, it's something I was told on the forum. But it was a long time ago, so I might be misremembering something. Or the person who said that could've been mistaken.

Sorry!

This can be summarized in a chart like this:

Code: Select all

           Intr-A Intr-P Tr-A Tr-P
Monovalent 1,X    2,X    3,X  4,PASS
Divalent   *6,Y   7,CAUS 5,Y  5,Y

Intr-A means an intransitive verb which takes an agent-like S. Intr-P means an intransitive which takes a patient-like S. Tr-A means a transitive verb with at least an agent argument. Tr-P means a transitive verb with at least a patient argument.

The Y thingy can only be used on verbs that are intransitive by default. The X thingy is normally used on intransitives, but it can also be used on a transitive verb to remove the object.

Does all of this make sense? What should I call X and Y?

For the most part, this makes sense, after a couple of read-throughs. The only things I’m confused about are:

• You say that Y can only be applied to intransitive verbs. However, your glosses imply that sing is a transitive verb here, yet you apply Y to it. Was this meant to read ‘Y … can only be used on verbs that are transitive by default’?

Oh no. *headdesk* It was supposed to be Y can only be applied to transitive verbs. I'll change the original post...

• What exactly does Y do? You say that X simply marks an intransitive verb, and removes the object of a transitive verb, but you never say what Y does.

Y just marks that a verb is indeed divalent. But it's zero-marked.

• How do you translate a simple sentence like ‘I see you’, or ‘I break the window’? For these neither X nor Y seem to apply.

1SG-ERG 2SG-ABS see.TR-Y
I see you.

1SG-ERG window-ABS break.TR-Y
I break the window.

• What exactly makes sentence 6 (*1SG-ERG marathon-ABS run.INTR-Y ‘I run a marathon’) illegal? How can that sentence be translated legally?

Because I decided to have such a constraint that you can't use Y on intransitives. That sentence you'd have to paraphrase or say something like 1SG-ABS marathon-PROL run.INTR-Y (prolative) or 1SG-ABS marathon-ADE run.INTR-Y (adessive).

That makes a lot more sense now! So, to summarise:

• Y is used to mark ordinary transitive verbs.
• X is used to mark ordinary intransitive verbs. If X is used with a transitive verb, it removes the object (absolutive) and promotes the subject (ergative) to an object i.e. an antipassive.

As for naming: I’m not sure about Y until the above issues are clarified, but X applied to transitive verbs is pretty clearly an antipassive , which is very typical of ergative languages. X applied to intransitive verbs seems could be called an ‘intransitivity marker’, although that would be pretty weird — I believe that transitivity markers are more usual.

Heh, I was thinking like "have I reinvented an antipassive or mediopassive voice or something?" So there are two uses for X. Should I treat it like two different verb categories that are marked with homonymous morphemes, or should I just call X the antipassive and make a note that "it's only really an antipassive when used like this and not when used like that"?

I don’t really know how to answer this; I would say this is more of a philosophical question. I would personally call X and Y an ‘intransitivity marker’ and ‘transitivity marker’ respectively, and say that when the intransitivity marker is used on a transitive verb, it gives an antipassive meaning.

An alternative system I was thinking about is that Y could be marked, while X could be unmarked. In that case, there would be no need to analyse a special ‘intransitivity marker’: you could just say that the presence of a transitivity marker shows that the verb is transitive, and when it is removed the verb takes on an antipassive meaning.

Thanks for taking your time answering my questions!

You’re welcome! I do find this topic particularly interesting, so I’m always happy to talk about it.

[zompist]

Okay. Then people are wearing pants when riding. Loincloths were worn when farming. Xwtek is hairier than horse and this may cause chafing problem even worse. (The climate is below 10 degrees Celsius in average, and may drop below zero during winter).

Also, is there any adaptation for cold with only pants? My sentient creature is human in name only (they use magic, and in general looks more like a brown-skinned elf).

If you've got a separate species, there's no reason you can't have them adapted to the cold. Humans prefer warm temperatures because we evolved in a warm savannah. If your species evolved in a cold climate, they will be adapted to it and shouldn't need extra clothing. On the other hand, they may be ill adapted to anything warmer!

Also, humans can definitely be adapted to cold— Inuit, for instance, are most comfortable at temperatures not far above freezing. They complain about the heat and take off their shirts when it's 13° C out.

You might consider, however, why or whether your species would lose their fur, assuming their ancestors were mammals. Fur is pretty good in a 10°C climate, naked skin less so. If they have no fur, however, I'd expect an insulating layer of fat, like walruses. (Or maybe a high metabolism that generates a lot of heat, but at the cost of very high food requirements.)

[bradrn]

I know of at least five strategies to mark complement clauses:

• Use of a complementiser particle (used in e.g. English)
• Use of an interrogative word (e.g. English)
• Addition of an affix or clitic to the predicate (e.g. Tariana)
• Other morphological change (e.g. Jarawara modifies the final vowel of the predicate)
• Other strategies which link two clauses without actually making a complement clause (Dixon calls these ‘complementation strategies’; e.g. serial verb constructions, nominalization, purposive clause linking, apposition etc.)

Aside from these, are there any other strategies used?

[Xwtek]

Xwtek is hairier than horse and this may cause chafing problem even worse.

Hmm, I think hair might actually decrease chafing. This is why people have pubic hairs. Or so I've heard.

Thanks. That factor and the fact that Native American are sometimes seen wearing loincloth while riding horse, I decided to revert back to loincloth wearing people.

[Xwtek]

You might consider, however, why or whether your species would lose their fur, assuming their ancestors were mammals. Fur is pretty good in a 10°C climate, naked skin less so. If they have no fur, however, I'd expect an insulating layer of fat, like walruses. (Or maybe a high metabolism that generates a lot of heat, but at the cost of very high food requirements.)

Humans are initially adapted to a climate that is either tropical or having a big swing of temperature. (i.e. 30°C at day and -10°C at night. The reason of the big swing is because a single day can correspond to months in earth time). Initially, humans will hibernate at night, and will develop a warm den to fight the winter/night instead of wearing fur. This changes when humans begin to use fire and living in world with smaller temperature swing. Human is also not a single species, but is actually a genus. And despite both having language, Sakhan and other people in neighboring plane isn't all in the same species. People sometimes arrange a marriage between two worlds, even if technically they're of the different species. The offspring is usually highly regarded for strength and intelligence, but is usually sterile. My OC, Xwtek, also used as my name, and is named for titular horse-dog xwtek, is a hybrid. (My profile picture is Kirishima Eijirou, though. I can't draw)

[Ahzoh]

Anyone good at making syntax trees?

Not sure if I treed my sentence correctly:
[S [NP [N Māte] [NEG tē]] [VP [V zagā] [^CP [C bi] [VP [N ʾamak] [V ṣessarnākku] [PP [P ʾina] [N lāmīnu]]]]]]

Sentence in question is:
Māte tē zagā bi ʾamak ṣessarnākku ʾina lāmīnu.
"nothing could convince me otherwise."

Gloss:
what-MASC NEG be.able\ACT.PRES COMP 1s.OBL <REL>poke\ACT.PRES-SUBJ.DAT across finger\FEM.OBL

[cedh]

Anyone good at making syntax trees?

Not sure if I treed my sentence correctly:
[S [NP [N Māte] [NEG tē]] [VP [V zagā] [^CP [C bi] [VP [N ʾamak] [V ṣessarnākku] [PP [P ʾina] [N ]]]]]]

Sentence in question is:
Māte tē zagā bi ʾamak ṣessarnākku ʾina lāmīnu.
"nothing could convince me otherwise."

Gloss:
what-MASC NEG be.able\ACT.PRES COMP 1s.OBL <REL>poke\ACT.PRES-SUBJ.DAT across finger\FEM.OBL

That looks fairly OK at first glance, although some people might insist that ʾamak and lāmīnu should appear within extra NP layers. Also, I'm not sure about tē 'NEG' - is it really used adnominally as a quantifier "no, none", or does it rather apply to the verb after it, i.e. "anything would not be able..."?

Could you post a literal (and possibly over-detailed) English translation of your sentence, so that it's easier to understand how the language actually does this?

[Ahzoh]

That looks fairly OK at first glance, although some people might insist that ʾamak and lāmīnu should appear within extra NP layers. Also, I'm not sure about tē 'NEG' - is it really used adnominally as a quantifier "no, none", or does it rather apply to the verb after it, i.e. "anything would not be able..."?

Seems kind of stupid to consider one noun as an entire phrase... and yes, the negative particle is being use adnominally as the quantifier "no, none", negative particle would otherwise go after the verb.

Could you post a literal (and possibly over-detailed) English translation of your sentence, so that it's easier to understand how the language actually does this?

"what no is-able that me [they-]might-poke-to across finger"
It is SOV and the infix <ess> indicates an anaphoric reference (and also relative clauses), so in this sentence, it is indicating that the subject in the complimentizer phrase is the same as the subject of the main phrase.

[zompist]

Anyone good at making syntax trees?

Not sure if I treed my sentence correctly:
[S [NP [N Māte] [NEG tē]] [VP [V zagā] [^CP [C bi] [VP [N ʾamak] [V ṣessarnākku] [PP [P ʾina] [N lāmīnu]]]]]]

Sentence in question is:
Māte tē zagā bi ʾamak ṣessarnākku ʾina lāmīnu.
what-MASC NEG be.able\ACT.PRES COMP 1s.OBL <REL>poke\ACT.PRES-SUBJ.DAT across finger\FEM.OBL
"nothing could convince me otherwise."

Correctly by X-Bar or Minimalism, no, but why would you want that?

It's fine as a representation of surface structure. For more you need some syntactic theory. I'd just suggest one principle to you: don't add a node unless it's useful and justified in terms of your language. In this case: does your language even have VPs? That is, are there syntactic facts that require grouping the verb with the object, rather than just making subject, object, verb children of the S?

It's maybe odd to have an S at one level and CP at another. What I'd expect is that the lower CP consists of a COMP plus an S. Or, if you're trying to be simple, I'd expect just an S down there.

Finally you might consider marking that anaphoric reference with subscripts. This is one area where syntactic trees can clarify things. In this case it might require adding a dummy node-- something like

nothing_i able [that (Ø)_i me might poke]

where the subscript i indicates identity of reference, and ø indicates a node that performs a syntactic function (it's the subject of 'poke') but doesn't get vocalized.

[cedh]

Taking both your answer and Zompist's remarks into account, how about something like this:

[S [NP_i [N [Māte what_MASC]] [Q [tē no]]] [V [zagā be.able_ACT.PRES]] [CP [C [bi COMP]] [S [NP_i^ [(Ø) *t*_i]] [NP [ʾamak 1SG_OBL]] [V [ṣessarnākku <REL>poke_ACT.PRES]] [PP [P [ʾina across]] [NP [lāmīnu finger_FEM.OBL]]]]]]

[Ahzoh]

Correctly by X-Bar or Minimalism, no, but why would you want that?

It's fine as a representation of surface structure. For more you need some syntactic theory. I'd just suggest one principle to you: don't add a node unless it's useful and justified in terms of your language. In this case: does your language even have VPs? That is, are there syntactic facts that require grouping the verb with the object, rather than just making subject, object, verb children of the S?

Aw hell, this is why I hate syntax trees. I don't know what would make it a syntactic fact that justifies having the object always grouped with the verb except Object Bonding, which maybe Vrkhazhian has. My categorization of things into NP, PP, AP, VP, etc. is based off of the wiki on head-directionality parameters where it talks about types of phrases.

Is it correct to consider it a verb phrase if it consists of a verb modified by an adverb or preposition phrase behaving like an adverb?

It's maybe odd to have an S at one level and CP at another. What I'd expect is that the lower CP consists of a COMP plus an S. Or, if you're trying to be simple, I'd expect just an S down there.

I thought S stood for "sentence", as in the whole thing, not individual clauses.

[Ahzoh]

Taking both your answer and Zompist's remarks into account, how about something like this:

[S [NP_i [N [Māte what_MASC]] [Q [tē no]]] [V [zagā be.able_ACT.PRES]] [CP [C [bi COMP]] [S [NP_i^ [(Ø) *t*_i]] [NP [ʾamak 1SG_OBL]] [V [ṣessarnākku <REL>poke_ACT.PRES]] [PP [P [ʾina across]] [NP [lāmīnu finger_FEM.OBL]]]]]]

It looks nice, and intimidates me as I also made a tree for a much more complicated sentence which is evidently completely wrong.

ʾĀwe bēdun bessād ʾutu kamā ʾutu śū ʾāwi zessagā ʾutu bi bēdun bessādāh.
man-MASC.NOM book-FEM.OBL-PL <REL>read\ACT.PRES NEG stand\ACT.PRES NEG on man-MASC.OBL <REL>be.able\ACT.PRES NEG COMP book-FEM.OBL-PL <REL>read\ACT.PRES-SUBJ
"A man who does not read books does not stand on (have advantage over) a man who is not able to read books."
man_i [(Ø)_i books read not] stand on man_j [(Ø)_j able not that {(Ø)_j books might-read}]

Although frankly I can't decide if the preposition phrase is modifying the verb or should be filling its own object slot (which would result in the whole thing coming before the verb).

[cedh]

That tree looks good to me.

Although frankly I can't decide if the preposition phrase is modifying the verb or should be filling its own object slot (which would result in the whole thing coming before the verb).

To me (obviously as an outsider) it feels as if both variants might be equally possible. It probably depends on whether you see "stand on" as a phrasal verb or as a verb plus preposition. In any case, if the PP does come after the verb, your tree looks suitable to me.

[Ahzoh]

That tree looks good to me.

Although frankly I can't decide if the preposition phrase is modifying the verb or should be filling its own object slot (which would result in the whole thing coming before the verb).

To me (obviously as an outsider) it feels as if both variants might be equally possible. It probably depends on whether you see "stand on" as a phrasal verb or as a verb plus preposition. In any case, if the PP does come after the verb, your tree looks suitable to me.

There aren't really phrasal verbs like that, the closest to that are the applicative voice markers. I generally treat the PP as an adverb rather than an object. Given that PPs can also modify nouns like adjectives, this is consistent with adverbs and adjectives behaving as one word class.

[zompist]

Correctly by X-Bar or Minimalism, no, but why would you want that?

It's fine as a representation of surface structure. For more you need some syntactic theory. I'd just suggest one principle to you: don't add a node unless it's useful and justified in terms of your language. In this case: does your language even have VPs? That is, are there syntactic facts that require grouping the verb with the object, rather than just making subject, object, verb children of the S?

Aw hell, this is why I hate syntax trees. I don't know what would make it a syntactic fact that justifies having the object always grouped with the verb except Object Bonding, which maybe Vrkhazhian has. My categorization of things into NP, PP, AP, VP, etc. is based off of the wiki on head-directionality parameters where it talks about types of phrases.

I've written a whole book on this, which I mention only to say that it takes a book to explain— it's hard to explain things in one post!

There's nothing wrong with including the VP. But I'd omit it unless it actually makes a unit in your language. Are there things that you can do just to the VP?

There are in English, e.g. VP Deletion, which deletes a VP

Bert will make a salad, and Ernie will make a salad too.
:> Bert will make a salad, and Ernie will too.

A VP can also be a sentence fragment that answers a question:

"What did you do today?"
"Made a conlang."

By contrast you cant do things in English with the subject + V. If you have "Bert made a salad", there's very little you can do with "Bert made" as a unit.

I thought S stood for "sentence", as in the whole thing, not individual clauses.

The thing is, you can have the same structure at different levels:

Bert will eat the salad.
Ernie is certain that it's desirable that Bert will eat the salad.

The easiest way to express this is to allow an S at different levels. You could equally well just have a CP at both levels.

[Ahzoh]

A VP can also be a sentence fragment that answers a question:
"What did you do today?"
"Made a conlang."

Does the behaviour of my relative/anaphoric infix <ess> indicate the presence of a VP? I can use it to omit the subject in following sentences leaving just the syntactic object and its verb (plus any modifiers).

John_i has a boat. G<ess>ot_i into the boat. A crocodile_j bit the boat. S<ess>wam_j away.

John_i saw what today? S<ess>aw_i a crow.

It is a very multi-purpose infix.

[Xwtek]

A VP can also be a sentence fragment that answers a question:
"What did you do today?"
"Made a conlang."

Does the behaviour of my relative/anaphoric infix <ess> indicate the presence of a VP? I can use it to omit the subject in following sentences leaving just the syntactic object and its verb (plus any modifiers).

Probably not. If I read the example correctly, it's just a Same Subject marker. It just means that your language have a switch reference marking.