(Conlangs) Q&A Thread - Quick questions go here

A forum for all topics related to constructed languages
Salmoneus
MVP
MVP
Posts: 1634
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 02 Oct 2018 12:38

Ahzoh wrote:
02 Oct 2018 04:27
1-NOM 3-ABL run = I run away from him
1-NOM 3-ACC run-APL = I make him run
That's not an applicative, that's a causative.

An applicative is a voice that moves a syntactic oblique into object position, without changing its semantic role (indeed, it often marks the role on the verb). What you have here, where the subject is demoted to the object and the oblique is promoted to subject*, isn't blurring the lines, it's just plain not an applicative.


*the structure of the change is: I-subj run because-of-him-OBL > he-subj makes-run me-ACC. The fact that you can have a 'from him' oblique for the intransitive is irrelevent, because it's a different argument from the one you're promoting in the transitive.


Remember, a voice doesn't change the core semantics of the verb, it just rearranges the syntax. "He eats me" and "I am eaten by him" have the same basic semantics, just different role assignments; likewise, "He eats because of me" and "I make him eat", or "I fly over it" and "I overfly it". If you have some process that turns "I run away from him" into "I force him to run", you're completely changing the semantics, and that's not a new voice, that's just a totally unrelated sentence.

Of course, you could use applicatives to form causatives. I-nom run-ACT because of him > I-nom run-CAU him-ACC. That is, "I run because of him" is turned by an applicative into "I for-run him" (which doesn't quite work in English because we don't have morphological causative applicatives; we maybe have some suppletive ones, like "I act because of her" > "I obey her" (though that doesn't quite work either, because it's a specific form of causation only)).

User avatar
Ahzoh
korean
korean
Posts: 5764
Joined: 20 Oct 2013 02:57
Location: Toma-ʾEzra lit Vṛḵaža

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh » 02 Oct 2018 14:17

Salmoneus wrote:
02 Oct 2018 12:38
Ahzoh wrote:
02 Oct 2018 04:27
1-NOM 3-ABL run = I run away from him
1-NOM 3-ACC run-APL = I make him run
That's not an applicative, that's a causative.
Perhaps I was making too great of a leap. I looked at applicative solely as a valency increasing process.
Of course, you could use applicatives to form causatives.
Although I am not looking to derive a pure morphological causative but have a single applicative that can convey, comitative (together with), adjutative (on behalf of), causative (because of), and superlative meanings. But that seems to require the promoted object to be in cases other than accusative, which defeats the point.
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Šat Wərxažu (Vrkhazhian) [ WIKI | CWS ]

User avatar
Reyzadren
greek
greek
Posts: 458
Joined: 14 May 2017 10:39
Contact:

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Reyzadren » 02 Oct 2018 15:53

Ahzoh wrote:
02 Oct 2018 14:17
Although I am not looking to derive a pure morphological causative but have a single applicative that can convey, comitative (together with), adjutative (on behalf of), causative (because of), and superlative meanings. But that seems to require the promoted object to be in cases other than accusative, which defeats the point.
Well, you don't need to adhere to pure linguistics for your conlang with regards to causatives or applicatives etc, you can just claim your conlang has something that "has approximately or implies these meanings", as natlangs also often don't neatly fit within linguistics terminology sometimes. For example, in that other natlang that I natively speak, there is an affix that has a similar effect that you propose - let's just call it the K-suffix here in the gloss.

* Comitative (~instrumental)
1SG ACT-key-K door = I locked the door with a key
* On behalf of (~pseudo-trigger mechanics)
1SG ACT-read-K 3SG book = I read a book for you
* Causative (~transitivity elevator)
1SG ACT-jump-K ball = I made the ball jump
* Superlative (~attribute-giver)
1SG ACT-long-K 3SG = I lengthen it

The left shows what you wanted, and the right (bracket) shows what that other natlang that I speak does, which seems rather close. If it pleases you that such feature or idea is attested in natlangs, then...

You got ANADEW-ed. [:P]
Image Soundcloud Profile | Image griuskant conlang

User avatar
Ahzoh
korean
korean
Posts: 5764
Joined: 20 Oct 2013 02:57
Location: Toma-ʾEzra lit Vṛḵaža

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh » 02 Oct 2018 16:23

Well, I've realized that I couldn't really derive a causative from an applicative voice because an applicative promotes an oblique object to direct object.
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Šat Wərxažu (Vrkhazhian) [ WIKI | CWS ]

User avatar
Creyeditor
mongolian
mongolian
Posts: 4489
Joined: 14 Aug 2012 19:32

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Creyeditor » 02 Oct 2018 17:25

Ahzoh wrote:
02 Oct 2018 16:23
Well, I've realized that I couldn't really derive a causative from an applicative voice because an applicative promotes an oblique object to direct object.
Indonesian -kan is either an applicative or a causative, depending on the verb it attaches to. Also some (Amazonian IIRC) languages have a transitivizer, which is basically an applicative or a causative.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

User avatar
jimydog000
cuneiform
cuneiform
Posts: 104
Joined: 19 Mar 2016 04:14
Location: Australia

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by jimydog000 » 02 Oct 2018 18:33

I have ku'et [kuʔet]: collection of something or mass uncountable noun modifier
këng [kɜŋ]: seed

Together they make këngku'et: grain.
Or ku'etkëng or two words... maybe?
“Comparison is the death of joy.” -Mark Twain

User avatar
Dormouse559
moderator
moderator
Posts: 2789
Joined: 10 Nov 2012 20:52
Location: California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Dormouse559 » 02 Oct 2018 19:36

jimydog000 wrote:
02 Oct 2018 18:33
I have ku'et [kuʔet]: collection of something or mass uncountable noun modifier
këng [kɜŋ]: seed

Together they make këngku'et: grain.
Or ku'etkëng or two words... maybe?
What is your question exactly? If it's about what order to put the morphemes in and whether to treat them as one word or two, those are language-specific issues. There are some tendencies regarding where modifiers go in relationship to a modified term, or "head". If the head tends to go after the modifiers, that is called "head-final" (e.g. këngku'et); if the head goes before the modifiers, that is called "head-initial" (e.g. ku'etkëng). Natlangs often prefer one "head directionality" but not to the complete exclusion of the other. For example, English is largely head-initial, but noun phrases are head-final when it comes to adjectives.

Treating the compound as one word or two probably depends on your conlang's phonology. If individual words have a distinct stress pattern or some other distinguishing feature, and the compound has the stress pattern or feature of an individual word, it likely makes sense to analyze it as a single word. If the term has a different pattern, maybe the individual-word pattern twice over, it may be more logical to analyze it as two words. Depending on the particulars of your conlang, the difference between a single word and two words could be largely arbitrary. See French, where there are few, if any, phonological markers of word boundaries.

Salmoneus
MVP
MVP
Posts: 1634
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 03 Oct 2018 13:31

Ahzoh wrote:
02 Oct 2018 14:17
Perhaps I was making too great of a leap. I looked at applicative solely as a valency increasing process.
So far as I'm aware, it's only used for oblique>object processes. Though you raise a good point: I don't know if there's a general term for a valency-increasing process, other than 'valency-increasing process'.
Of course, you could use applicatives to form causatives.
Although I am not looking to derive a pure morphological causative but have a single applicative that can convey, comitative (together with), adjutative (on behalf of), causative (because of), and superlative meanings. But that seems to require the promoted object to be in cases other than accusative, which defeats the point.
[/quote]

Huh? Why? The point of the applicative is that the semantic role of the object is marked on the verb, so that the verb CAN be in the plain accusative.
And of course it's OK if the marking of role is vague enough that those meanings - which are all, after all, rather close together semantically - marked the same way.

Just imagine if English had a 'for' applicative prefix:
"I for-ate him" > I ate for him (on his behalf, to his benefit), I ate because of him, etc.

----

You say you've 'realised' that you can't make an applicative into a causative. But that's not what I said. I pointed out that applicatives often ARE used as causatives, and showed you how: I-nom him-ACC ate-APP = I ate because of him (i.e. he made me eat).

If you really need a causative that matches English exactly, with the cause as the subject, you can then passivise that applicative: he-NOM ate-APP-PASS me-OBL.

----

Or, as others have said, you could have a suffix that's a generic valency-increaser, and have the new role assignment dictated lexically - say, creating causatives specifically out of involuntary verbs.

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 03 Oct 2018 16:13

Whereas Ddoean's adjectives used to be formed with suffixes similar to <-xi>, they're now formed with suffixes based on the formative noun's declension. For example, roua, meaning water, becomes rouayak (/ˈru.ɑ.ʔɑk/), meaning aquatic. The vowel featured in the suffix is the verbal conjugation's characteristic vowel as well. How unusual is this relationship?

How odd would system consisting of all near-front and near-back vowels be? For example I'm toying with the idea of switching Ddoean's /i u e ø ɑ/ system to something like /ɪ ʊ e ø ɑ/. The mid /e/ and /ø/ are directly underneath the near-front /ɪ/ and /ʏ/ on the chart. Are they front or near-front?
Last edited by yangfiretiger121 on 03 Oct 2018 20:07, edited 1 time in total.
Alien conlangs (Font may be needed for Vai symbols)

User avatar
jimydog000
cuneiform
cuneiform
Posts: 104
Joined: 19 Mar 2016 04:14
Location: Australia

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by jimydog000 » 03 Oct 2018 17:59

Dormouse559 wrote:
02 Oct 2018 19:36
What is your question exactly? If it's about what order to put the morphemes in and whether to treat them as one word or two, those are language-specific issues. There are some tendencies regarding where modifiers go in relationship to a modified term, or "head". If the head tends to go after the modifiers, that is called "head-final" (e.g. këngku'et); if the head goes before the modifiers, that is called "head-initial" (e.g. ku'etkëng). Natlangs often prefer one "head directionality" but not to the complete exclusion of the other. For example, English is largely head-initial, but noun phrases are head-final when it comes to adjectives.

Treating the compound as one word or two probably depends on your conlang's phonology. If individual words have a distinct stress pattern or some other distinguishing feature, and the compound has the stress pattern or feature of an individual word, it likely makes sense to analyze it as a single word. If the term has a different pattern, maybe the individual-word pattern twice over, it may be more logical to analyze it as two words. Depending on the particulars of your conlang, the difference between a single word and two words could be largely arbitrary. See French, where there are few, if any, phonological markers of word boundaries.
Thanks, that is what I was asking.
“Comparison is the death of joy.” -Mark Twain

User avatar
Shemtov
runic
runic
Posts: 3057
Joined: 29 Apr 2013 04:06

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Shemtov » 03 Oct 2018 23:07

Can a language with primary and secondary aspects have a null secondary aspect marker?
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

User avatar
Ahzoh
korean
korean
Posts: 5764
Joined: 20 Oct 2013 02:57
Location: Toma-ʾEzra lit Vṛḵaža

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh » 03 Oct 2018 23:16

Salmoneus wrote:
03 Oct 2018 13:31
Huh? Why? The point of the applicative is that the semantic role of the object is marked on the verb, so that the verb CAN be in the plain accusative.
That's why I can't have distinctive applicatives without some additional affixes markers. With just an applicative there are many situations where context wouldn't be able to distinguish between, say, "outdo", "activate", "with-do", and "for-do".
You say you've 'realised' that you can't make an applicative into a causative. But that's not what I said. I pointed out that applicatives often ARE used as causatives, and showed you how: I-nom him-ACC ate-APP = I ate because of him (i.e. he made me eat).
I thought in a causative the cause has to be in the subject role.
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Šat Wərxažu (Vrkhazhian) [ WIKI | CWS ]

User avatar
Dormouse559
moderator
moderator
Posts: 2789
Joined: 10 Nov 2012 20:52
Location: California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Dormouse559 » 04 Oct 2018 00:11

Ahzoh wrote:
03 Oct 2018 23:16
You say you've 'realised' that you can't make an applicative into a causative. But that's not what I said. I pointed out that applicatives often ARE used as causatives, and showed you how: I-nom him-ACC ate-APP = I ate because of him (i.e. he made me eat).
I thought in a causative the cause has to be in the subject role.
You could pile voices on top of each other. Passivize that applicative: he-NOM me-OBL ate-APP-PASS = He was for-eaten by me. / He made me eat.

User avatar
Ahzoh
korean
korean
Posts: 5764
Joined: 20 Oct 2013 02:57
Location: Toma-ʾEzra lit Vṛḵaža

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Ahzoh » 04 Oct 2018 00:35

I'm not sure what case to put the actual recipient when an applicative turns a ditransitive like give into tritransitive because my language is secundative.

I gave him a book because of John (made me do it)= 1-NOM John-ACC give-APL book-INSTR 3ms-???
Image Ӯсцьӣ (Onschen) [ CWS ]
Image Šat Wərxažu (Vrkhazhian) [ WIKI | CWS ]

Salmoneus
MVP
MVP
Posts: 1634
Joined: 19 Sep 2011 19:37

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Salmoneus » 04 Oct 2018 02:18

Ahzoh wrote:
03 Oct 2018 23:16
Salmoneus wrote:
03 Oct 2018 13:31
Huh? Why? The point of the applicative is that the semantic role of the object is marked on the verb, so that the verb CAN be in the plain accusative.
That's why I can't have distinctive applicatives without some additional affixes markers. With just an applicative there are many situations where context wouldn't be able to distinguish between, say, "outdo", "activate", "with-do", and "for-do".
Why is that a problem? Besides, you already said you were thinking of just one applicative.

Consider, incidentally, the Germanic all-purpose causative/applicative "be-" affix...
You say you've 'realised' that you can't make an applicative into a causative. But that's not what I said. I pointed out that applicatives often ARE used as causatives, and showed you how: I-nom him-ACC ate-APP = I ate because of him (i.e. he made me eat).
I thought in a causative the cause has to be in the subject role.
OK, yes, using 'causative' as the name of a voice, you're right. I meant it more in the general sense of a construction to show the cause. Looking it up, apparently have a cause as the applied object isn't actually that common, though I'm sure it must happen (given that benefactives and malefactives and comitatives and substitutives are all found and they're semantically very close to causes (eg. "I did it on behalf of John" is very close to "John had me do it")).

And apparently it's VERY common for the same affix to be both an applicative and a causative, depending either on the verb or the on the nouns (in Javanese apparently sometimes it's based on animacy).

Regarding your question on cases: whatever case you want. Or none - why does your language even need to have quadrivalents?

User avatar
Shemtov
runic
runic
Posts: 3057
Joined: 29 Apr 2013 04:06

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Shemtov » 04 Oct 2018 03:10

Shemtov wrote:
03 Oct 2018 23:07
Can a language with primary and secondary aspects have a null secondary aspect marker?
Just making sure people see the question.
Many children make up, or begin to make up, imaginary languages. I have been at it since I could write.
-JRR Tolkien

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 04 Oct 2018 09:31

Repost cause skipped:

Whereas Ddoean's adjectives used to be formed with suffixes similar to <-xi>, they're now formed with suffixes based on the formative noun's declension. For example, roua, meaning water, becomes rouayak (/ˈru.ɑ.ʔɑk/), meaning aquatic. The vowel featured in the suffix is the verbal conjugation's characteristic vowel as well. How unusual is this relationship?

How odd would system consisting of all near-front and near-back vowels be? For example I'm toying with the idea of switching Ddoean's /i u e ø ɑ/ system to something like /ɪ ʊ e ø ɑ/. The mid /e/ and /ø/ are directly underneath the near-front /ɪ/ and /ʏ/ on the chart. Are they front or near-front?
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Dormouse559
moderator
moderator
Posts: 2789
Joined: 10 Nov 2012 20:52
Location: California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Dormouse559 » 04 Oct 2018 15:42

Shemtov wrote:
03 Oct 2018 23:07
Can a language with primary and secondary aspects have a null secondary aspect marker?
I see no reason why not. Null markers are possible for any grammatical category.

yangfiretiger121 wrote:
04 Oct 2018 09:31
Whereas Ddoean's adjectives used to be formed with suffixes similar to <-xi>, they're now formed with suffixes based on the formative noun's declension. For example, roua, meaning water, becomes rouayak (/ˈru.ɑ.ʔɑk/), meaning aquatic. The vowel featured in the suffix is the verbal conjugation's characteristic vowel as well. How unusual is this relationship?
I don't understand the description. What do verbal conjugations have to do with the noun and adjective you cite?

yangfiretiger121 wrote:How odd would system consisting of all near-front and near-back vowels be? For example I'm toying with the idea of switching Ddoean's /i u e ø ɑ/ system to something like /ɪ ʊ e ø ɑ/. The mid /e/ and /ø/ are directly underneath the near-front /ɪ/ and /ʏ/ on the chart. Are they front or near-front?
Well, you've given a phoneme inventory, and the symbols representing phonemes only give a partial picture of reality. That wouldn't always be important for questions about vowel distribution, but there isn't a terribly big difference between /i u/ and /ɪ ʊ/. That said, your vowels look fine.

/e ø/ are front vowels. The vowel chart's shape is an abstraction of the places where the cardinal vowels are articulated in the mouth, with /e ø/ being shown as the most fronted vowels possible at their height. And just a note: /ɑ/ is back, not near-back.

yangfiretiger121
sinic
sinic
Posts: 233
Joined: 17 Jun 2018 03:04

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by yangfiretiger121 » 04 Oct 2018 16:50

Dormouse559 wrote:
04 Oct 2018 15:42
yangfiretiger121 wrote:
04 Oct 2018 09:31
Whereas Ddoean's adjectives used to be formed with suffixes similar to <-xi>, they're now formed with suffixes based on the formative noun's declension. For example, roua, meaning water, becomes rouayak (/ˈru.ɑ.ʔɑk/), meaning aquatic. The vowel featured in the suffix is the verbal conjugation's characteristic vowel as well. How unusual is this relationship?
I don't understand the description. What do verbal conjugations have to do with the noun and adjective you cite?
Essentially, adjective declensions now have the same height-based characteristic vowels as the verb conjugations (/i/ for first declension/conjugation, /ɑ/ for second declension/conjugation (roua (noun)/rouayak (derivative adjective)), and /ø̞/ for third declension/conjugation). How odd is this arrangement?
Dormouse559 wrote:
yangfiretiger121 wrote:How odd would system consisting of all near-front and near-back vowels be? For example I'm toying with the idea of switching Ddoean's /i u e ø ɑ/ system to something like /ɪ ʊ e ø ɑ/. The mid /e/ and /ø/ are directly underneath the near-front /ɪ/ and /ʏ/ on the chart. Are they front or near-front?
Well, you've given a phoneme inventory, and the symbols representing phonemes only give a partial picture of reality. That wouldn't always be important for questions about vowel distribution, but there isn't a terribly big difference between /i u/ and /ɪ ʊ/. That said, your vowels look fine.

/e ø/ are front vowels. The vowel chart's shape is an abstraction of the places where the cardinal vowels are articulated in the mouth, with /e ø/ being shown as the most fronted vowels possible at their height. And just a note: /ɑ/ is back, not near-back.
I used local broad transcription rather than /e̞ ø̞/ for some reason. Just to confirm, are /e̞ ø̞/, in fact, front vowels despite being directly underneath /ɪ ʏ/ and Wikipedia's note that "rounded front vowels are often centralized, which means that often they are in fact near-front"? For the record, there's a similar note about unrounded back vowels being centralized as well.
Alien conlangs (Font may be needed for Vai symbols)

User avatar
Dormouse559
moderator
moderator
Posts: 2789
Joined: 10 Nov 2012 20:52
Location: California

Re: (Conlangs) Q&A Thread - Quick questions go here

Post by Dormouse559 » 04 Oct 2018 17:13

yangfiretiger121 wrote:
04 Oct 2018 16:50
Essentially, adjective declensions now have the same height-based characteristic vowels as the verb conjugations (/i/ for first declension/conjugation, /ɑ/ for second declension/conjugation (roua (noun)/rouayak (derivative adjective)), and /ø̞/ for third declension/conjugation). How odd is this arrangement?
And the adjective declensions are based on the noun declensions. Well, I can't tell you how odd that is. Like many things, it's plausible. A language only has so many vowels, and if it has an emphasis on thematic vowels, it could easily end up with heavy overlap.

yangfiretiger121 wrote:I used local broad transcription rather than /e̞ ø̞/ for some reason. Just to confirm, are /e̞ ø̞/, in fact, front vowels despite being directly underneath /ɪ ʏ/ and Wikipedia's note that "rounded front vowels are often centralized, which means that often they are in fact near-front"? For the record, there's a similar note about unrounded back vowels being centralized as well.
The visual relationship of the vowels is less important for naming their features than their position on the chart. Imagine the chart as a rectangle. The left vertical line of the chart symbolizes the frontmost part of the vowel space in the mouth, and therefore the vowels on that line are "front". /e̞ ø̞/ are front because /e ø/ are on that left vertical line, and all a lowering diacritic does is move them down that line. The operative word in Wikipedia's note is "often". There is a tendency to centralize front rounded vowels and back unrounded vowels, but that need not always be the case.

Post Reply