Light Verbs, Serial Verb Constructions, & other verb odditie

A forum for all topics related to constructed languages
Post Reply
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Light Verbs, Serial Verb Constructions, & other verb odditie

Post by eldin raigmore »

Thread title is supposed to be:
Light Verbs, Serial Verb Constructions, & other verb "oddities", in Conlangs and Natlangs.

Solarius wrote:Kobon is really cool also- verbs are a small closed class, so you have to do a bunch of serial verb constructions.

Micamo wrote:
Solarius wrote:Kobon is really cool also- verbs are a small closed class, so you have to do a bunch of serial verb constructions.
Here's an example from Kalam, which is closely related to Kobon:

yad am mon pk d ap ay-p-yn (Pawley 1980)
1s go wood hit hold come put-PERF-1SG
I fetched firewood.

Unfortunately, good luck finding any first-hand material for these languages online! I've only been able to find second-hand fragments like the above.

EDIT: Nevermind!

And I'd been searching in vain for "Kalam" all this time. Screw you, google search algorithms.

gach wrote:
Micamo wrote:EDIT: Nevermind!
Wonderful, thanks for that. Kalam's verbal semantics is magnificent. I've too had my eyes on them for a while but unfortunately the stuff Foley cites is in these unpublished manuscripts that might randomly pop up or then might not.

XXXVII wrote:
Micamo wrote:Here's an example from Kalam, which is closely related to Kobon:

yad am mon pk d ap ay-p-yn (Pawley 1980)
1s go wood hit hold come put-PERF-1SG
I fetched firewood.
Okay... that is simply wonderful!

In terms of conlanging... how would you even determine what types of things would need to be included in a "closed class" for verbs?

Micamo wrote:
XXXVII wrote:Okay... that is simply wonderful!

In terms of conlanging... how would you even determine what types of things would need to be included in a "closed class" for verbs?
Foley claims that only about 25 of Kalam's verbs are commonly used: These verbs are "generic verbs" with very broad meaning that depends on context. Serialized generic verbs often have a somewhat idiomatic meaning:
Spoiler:
Image
It's also possible to have an adjunct (that is, non-argumental) nominal further narrow down the verb's meaning:
Spoiler:
Image

eldin raigmore wrote:
XXXVII wrote:In terms of conlanging... how would you even determine what types of things would need to be included in a "closed class" for verbs?
(My answer is not as fascinating, IMO, as Micamo's and your discussion about Kalam; but here it is.)

You might take, for instance, the 300 most common light verbs of Korean; or the twelve, or three, or whatever number, most common light verbs of any language the majority of whose "verbs" (do I mean "lexical verbs"?) are two-word lightverb+contentword phrases rather than one-word verbs.

A closed class is just a word-class to which new members are not synchronically being added, by coinage or by borrowing or by derivation or whatever processes make open classes grow. Small classes tend to be closed and closed classes tend to be small, but the exact meaning of "small" is relative. For instance, consider numeral classifiers in those languages that have them. The number of genders (concordial noun-classes) in which the gender is lexically inherent in the noun seems to be limited to about 40 among natlangs that have them but don't have numeral classifiers; but the number of numeral classifiers seems to be limited to a much larger number, about 400, among natlangs that have them rather than some other kind of noun-class system.

Just as numeral classifiers are the only nounlike word-class that can inflect for grammatical number (e.g. singular or plural) in natlangs that have numeral classifiers, so also light verbs are the only verblike word-class that can inflect for most verbal accidents (e.g. agreement (with person or nominal accidents of participants), aspect, evidentiality, mirativity, modality/mode/mood, pluractionality, polarity, tense, valency, validationality, voice) among natlangs most of whose "semantic verbs" are lightverb+contentword phrases.

But light verbs are semantically "light" rather than semantically "empty". To qualify as a "light verb", a word needs to be grammatically usable as the only verb in a clause.
For most "semantic verbs" in such natlangs, most of the weight of the semantic content is carried by the content-word: which needn't be, and probably usually isn't, grammatically (i.e. morphologically and syntactically) a verb; but also might be pretty much any "part-of-speech" or word class.

[hr][/hr]

Some languages have serial verb constructions.
Apparently the source Micamo and you have been discussing says most (or is it just "many"?) of Kalam's semantically content-heavy "semantic verbs" are light-verb series -- serial verbs.
I am not clear whether they sometimes, or even often, include a content-word that isn't grammaticaly a verb; but I take it that the comment Micamo quoted about the "adjunct nominal" means they do sometimes occur (the "adjunct nominal" being such a content-word).
I am not clear whether there is never more than one such "adjunct nominal", but my impression from the quotes cited is that there either never is, or seldom is, more than one non-verb content-word such as an "adjunct nominal".

[hr][/hr]

Does any of that help anyone, or inspire anyone? (Even someone besides XXXVII or Micamo?)

[hr][/hr]

BTW should this sub-thread be moved to some other thread than "What did you accomplish today?"?
Like the Conlangs Q&A Thread?
Or perhaps split to its own thread?

gach wrote:
eldin raigmore wrote:Some languages have serial verb constructions.
Apparently the source Micamo and you have been discussing says most (or is it just "many"?) of Kalam's semantically content-heavy "semantic verbs" are light-verb series -- serial verbs.
There are apparently vast differences in the frequency of the verbs in Kalam. The exact quote from Foley about the statistics is: "Kalam has under 100 verb stems and, of these, only about twenty-five are commonly used. Almost every action, process or state is categorized to one of these twenty-five verbs, which Pawley (1980) calls 'generic verbs'." So even though there are only few verbs to begin with, some of them must still be much rarer than others or in other words have more specific functions.
I am not clear whether they sometimes, or even often, include a content-word that isn't grammaticaly a verb; but I take it that the comment Micamo quoted about the "adjunct nominal" means they do sometimes occur (the "adjunct nominal" being such a content-word).
I am not clear whether there is never more than one such "adjunct nominal", but my impression from the quotes cited is that there either never is, or seldom is, more than one non-verb content-word such as an "adjunct nominal".
The whole list of examples in Micamo's second scan gives such noun-verb compounds. It's not entirely clear from the glosses, but I get the idea that only the last word in each of the examples is a verb while the preceding ones are nominal complements to it. Most of the given examples are of the form noun verb but there are also many three word ones of the form noun noun verb, a very nice example being jwn bobam ay- (head dandruff put) for "have dandruff".

And by the way, the source here is Foley's The Papuan languages of New Guinea which is a fascinating book and definitely worth reading.
BTW should this sub-thread be moved to some other thread than "What did you accomplish today?"?
Like the Conlangs Q&A Thread?
Or perhaps split to its own thread?
It would certainly be worth archiving in a more accessible place than here. Do we have a thread on serialisation?

eldin raigmore wrote:
gach wrote: .... (other stuff I also want to thank you for, but in particular: .... ) The whole list of examples in Micamo's second scan gives such noun-verb compounds. It's not entirely clear from the glosses, but I get the idea that only the last word in each of the examples is a verb while the preceding ones are nominal complements to it. Most of the given examples are of the form noun verb but there are also many three word ones of the form noun noun verb, a very nice example being jwn bobam ay- (head dandruff put) for "have dandruff". ....
Thank you, @gach. It looks like I had misunderstood earlier.


[hr][/hr]

Edit:
gach wrote:
me wrote:BTW should this sub-thread be moved to some other thread than "What did you accomplish today?"? Like the Conlangs Q&A Thread? Or perhaps split to its own thread?
It would certainly be worth archiving in a more accessible place than here. Do we have a thread on serialisation?
Apparently not.

User avatar
Yačay256
greek
greek
Posts: 648
Joined: 12 Aug 2010 01:57
Location: Sacramento, California, USA

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Yačay256 »

What are you asking about? If you are interested in complex predicates, check out this link about Wagiman and its coverbs.
¡Mñíĝínxàʋày!
¡[ˈmí.ɲ̟ōj.ˌɣín.ʃà.βä́j]!
2-POSS.EXCL.ALIEN-COMP-friend.comrade
Hello, colleagues!
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by eldin raigmore »

Yačay256 wrote:What are you asking about? If you are interested in complex predicates, check out this link about Wagiman and its coverbs.
Thanks!

I started this thread in response to a sub-thread of the "What did you accomplish today?" thread. My first post here just quotes every post made to that subthread, before I started this thread.

Yes, I think complex clausal nuclei (not necessarily complex predicates -- their nuclei might be very simple), especially those that are phrases, are of central interest to this thread.
Coverbs in general might or might not be. Wagiman's probably are.
Unfortunately there doesn't appear to be any way I can read the full text of the publication you linked to.
User avatar
Yačay256
greek
greek
Posts: 648
Joined: 12 Aug 2010 01:57
Location: Sacramento, California, USA

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Yačay256 »

eldin raigmore wrote:Unfortunately there doesn't appear to be any way I can read the full text of the publication you linked to.
Sorry about that; hopefully this link, this one on Yoruba or this one on the Dravidian Language Malto will be of more use to you.
¡Mñíĝínxàʋày!
¡[ˈmí.ɲ̟ōj.ˌɣín.ʃà.βä́j]!
2-POSS.EXCL.ALIEN-COMP-friend.comrade
Hello, colleagues!
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Micamo »

This might also help.

So here's a question for the thread: Why do some languages (Yoruba, Edo) require their verbs to be uninflected to participate in Serial Verb Constructions, while others (Lango, Yimas, Kalam) apparently don't have this restriction?

Are there any natlangs that require uninflection for some, but not all, of their SVCs?
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by eldin raigmore »

Yačay256 wrote:Sorry about that; hopefully this link,
Thanks, but I still couldn't follow through on any link that had any interesting details about the coverbs.

Yačay256 wrote:this one on Yoruba
Turning clauses (rather than smaller phrases) into compound words is interesting, but I'm not sure I would have put it in this thread. But I think it possible I could be wrong. And, of course, Thanks! anyway.

Yačay256 wrote:or this one on the Dravidian Language Malto will be of more use to you.
Thanks! [:)]

Having lots of "verb-forms" in a clause, but all but the last of them non-finite, is pretty interesting.

Before looking into the paper, I wouldn't have thought it's what I had in mind when I started this thread. (But I see that the stems of non-finite verbforms can be combined with each other, or with the roots of finite verbs, to form compound verbs! That's exactly up our alley.)

But I would have thought it's worth discussing.
Especially the fact that there's derivational morphology that can turn a verb-root into:
Something that functions the way the nuclear verbs of adjunct clauses (clauses used as adverbs) function (in languages that have such clauses);
Supines (another kind of verbal adverb, for instance those that function as purpose-clauses would in languages that have purpose-clauses);
"TAM-relative" verbs, that function as the nuclear verbs of "TAM-relative clauses" would function, that is, as sentential adverbs that are temporal auxiliaries or aspectual auxiliaries or modal auxiliaries;
Relativization, marking the verb to perform a function like that performed by the nuclear verb in a relative clause, which is a clause-used-as-an-adjective;
Participalization, turning the verb into a participle, which is a kind of verbal adjective;
Or complementization, making the verb function as the nuclear verb of a complement clause (a clause used as if it were a noun) would function.

That's a lot of derivational morphology, and yes, I "like" it all.
I can't tell yet (because I haven't yet finished carefully reading the whole document) whether Malto has biclausal and multiclausal sentence constructions, but I think it probably does; allowing only one finite verb per sentence does not prohibit allowing several subordinate clauses in the same sentence, provided the subordinate clauses all have non-finite verbs for their nuclei. Requiring all such subordinate clauses to precede the finite verb which is the nucleus of the main clause, is interesting, though. Also, I think maybe Malto would have problems allowing a subordinate clause to have a further-subordinated clause subordinate to it.

I will have to finish the paper later, but it looks fascinating.


[hr][/hr]

Micamo wrote:This might also help.
No; I couldn't link to it.
Thanks just the same.

Micamo wrote:So here's a question for the thread: Why do some languages (Yoruba, Edo) require their verbs to be uninflected to participate in Serial Verb Constructions, while others (Lango, Yimas, Kalam) apparently don't have this restriction?
I can state wholeheartedly and without reservation that I do not know.

Micamo wrote:Are there any natlangs that require uninflection for some, but not all, of their SVCs?
Maybe Malto, depending on the exact meaning of your question?
User avatar
Thrice Xandvii
runic
runic
Posts: 2698
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Thrice Xandvii »

I just posted this in the source thread before I realized that it had moved here:
gach wrote:It would certainly be worth archiving in a more accessible place than here. Do we have a thread on serialisation?
If we don't yet... we should. I am completely fascinated by this stuff!

I think I am actually going to have it so that Sijaam Tû is a descendant of a language that had a closed-class of verbs. That would explain some of the serial verb constructions that have survived into the language, as well as the reasoning for many of the simpler and older verbs in the language being represented by logographs that have survived past the invention of a new script that better represents the sound values of the language (as an alphabet). I would actually give Sijaam a closed verb class, but I feel like I have grown too attached to a number of the verb stems in use, and some of the constructions for a few verbs that simply wouldn't have been part of such a system since their meanings are a bit too focused and unique.
Image
User avatar
gach
MVP
MVP
Posts: 513
Joined: 07 Aug 2013 01:26
Location: displaced from Helsinki

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by gach »

Micamo wrote:Why do some languages (Yoruba, Edo) require their verbs to be uninflected to participate in Serial Verb Constructions, while others (Lango, Yimas, Kalam) apparently don't have this restriction?
Without delving deeper into the matter I'd say that what we call SVCs don't form such a homogeneous class as we might first think. The amount of inflections that can or must appear on the individual serialised verbs could just be a free parameter that varies from language to language. The inclusion of inflections into individual components of SVCs might indicate a border line between something like inner and outer inflections, but whether such a division is meaningful would need much more data to be properly addressed.

Skou provides an example of a language where all verbs in a SVC must be inflected for the subject using bound agreement. The forms of bound agreement vary greatly from one verb to another and many verbs don't have it at all, but nevertheless if it can appear it must. Verbs are also marked for the person of the subject by pronominal proclitics, but as clitics these are outer inflections and only have to apply once for the entire SVC. You can see this in the following example where all the three verbs are inflected for the subject person and number either by bound prefixes or stem alternations while the subject clitic ke= only appears once in the beginning of the verb complex.

Ke ke=k-atà k-o ti báng.
he SG3.NF=SG3.NF-run SG3.NF-go.seawards SG3.NF:go beach
"He's running to the beach."

This isn't the whole story, however, since also the pronominal clitics can sometimes appear on multiple verbs of what seem to be SVCs. See the following example which is given two possible translations

Ne líhi náti ne=ne ne=pang-pang ka.
we garden new PL1=PL1:go PL1=PL:chop-RED NEG
"We didn't go and clear it away to make a new garden." ~
"We went but didn't clear it away to make a new garden."

The second translation indicates that the presence of the second pronominal clitic can in fact mark the breaking up of the SVC into two separate predicates restricting the scope of the negation only to the latter of them. However, the first translation where the scope of the negation spans over both of the verbs still remains valid. This could indicate that the two verbs still belong into a SVC forming one single predicate and that the pronominal clitics are also tolerated within such close knit verb chains.

I have the simplified version of this without the optional agreement clitics within a SVC in Nooníí kisnk. The basic agreement pattern in the language is that objects are marked by bound prefixes which must repeat on each verb in a SVC while the subjects don't really have agreement at all but are marked by proclitic pronouns attached to the first verb in the chain. A typical sentence with a transitive SVC will thus look like

O=k-stkaa k-on n=tók.
SG1.S=SG2.O-see SG2.O-go/be.up REF=tree
"I saw you up in the tree."

Returning to the example languages you listed, what SVC internal morphology did you have in mind for listing Yimas in there. Quickly glancing over the grammar, the only morphology in the language that can find intruding a chain of verb stems creating a SVC are the simultaneity connector -ra and the sequential connector -mpi. The rest of the verb morphology applies to the full serialised complex treating it as an extended verb stem. Examples with and without these morphemes are

Nawn ya-ŋa-awa-ta-n?
who PL.CL5-SG1.D-excrete-put-PRES
"Who's urinating on me?"

Na-n-munta-ra-wapal-k.
SG3.O-SG3.A-call-SIM-ascend-IRR
"She called out to him while going up."

Awt ŋa-kra-awl-mpi-waraca-ŋa-n.
fire IMP-PL1.D-get-SEQ-return-BEN-IMP
"Bring back fire for us!"

These are explained as suffixes on the preceding verb stem and the sequential marker -mpi can certainly exist as a suffix on isolated verbs. Still, within the SVCs these morphemes could also be analysed as mere connectors required for the construction of certain kinds of verb chains and not any more tightly connected to the preceding verb than to the following one.
ImageKištaLkal sikSeic
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Micamo »

An example of what I'm talking about with Edo:

*Èvbàré òré Òzó lé-rè khién(-rèn)
food FOC Ozo cook-PST.PERF sell(-PST.PERF)
"It's the food which Ozo has cooked and sold."

If you remove both past perfective suffixes (or either verb), the sentence becomes grammatical. The question, I suppose, is why this restriction applies in Edo but not in other langs with SVCs.
gach wrote:Without delving deeper into the matter I'd say that what we call SVCs don't form such a homogeneous class as we might first think. The amount of inflections that can or must appear on the individual serialised verbs could just be a free parameter that varies from language to language. The inclusion of inflections into individual components of SVCs might indicate a border line between something like inner and outer inflections, but whether such a division is meaningful would need much more data to be properly addressed.

Skou provides an example of a language where all verbs in a SVC must be inflected for the subject using bound agreement. The forms of bound agreement vary greatly from one verb to another and many verbs don't have it at all, but nevertheless if it can appear it must. Verbs are also marked for the person of the subject by pronominal proclitics, but as clitics these are outer inflections and only have to apply once for the entire SVC. You can see this in the following example where all the three verbs are inflected for the subject person and number either by bound prefixes or stem alternations while the subject clitic ke= only appears once in the beginning of the verb complex.

Ke ke=k-atà k-o ti báng.
he SG3.NF=SG3.NF-run SG3.NF-go.seawards SG3.NF:go beach
"He's running to the beach."

This isn't the whole story, however, since also the pronominal clitics can sometimes appear on multiple verbs of what seem to be SVCs. See the following example which is given two possible translations

Ne líhi náti ne=ne ne=pang-pang ka.
we garden new PL1=PL1:go PL1=PL:chop-RED NEG
"We didn't go and clear it away to make a new garden." ~
"We went but didn't clear it away to make a new garden."

The second translation indicates that the presence of the second pronominal clitic can in fact mark the breaking up of the SVC into two separate predicates restricting the scope of the negation only to the latter of them. However, the first translation where the scope of the negation spans over both of the verbs still remains valid. This could indicate that the two verbs still belong into a SVC forming one single predicate and that the pronominal clitics are also tolerated within such close knit verb chains.

I have the simplified version of this without the optional agreement clitics within a SVC in Nooníí kisnk. The basic agreement pattern in the language is that objects are marked by bound prefixes which must repeat on each verb in a SVC while the subjects don't really have agreement at all but are marked by proclitic pronouns attached to the first verb in the chain. A typical sentence with a transitive SVC will thus look like

O=k-stkaa k-on n=tók.
SG1.S=SG2.O-see SG2.O-go/be.up REF=tree
"I saw you up in the tree."
Interesting examples from Skou: Thanks!
Returning to the example languages you listed, what SVC internal morphology did you have in mind for listing Yimas in there. Quickly glancing over the grammar, the only morphology in the language that can find intruding a chain of verb stems creating a SVC are the simultaneity connector -ra and the sequential connector -mpi. The rest of the verb morphology applies to the full serialised complex treating it as an extended verb stem. Examples with and without these morphemes are

Nawn ya-ŋa-awa-ta-n?
who PL.CL5-SG1.D-excrete-put-PRES
"Who's urinating on me?"

Na-n-munta-ra-wapal-k.
SG3.O-SG3.A-call-SIM-ascend-IRR
"She called out to him while going up."

Awt ŋa-kra-awl-mpi-waraca-ŋa-n.
fire IMP-PL1.D-get-SEQ-return-BEN-IMP
"Bring back fire for us!"

These are explained as suffixes on the preceding verb stem and the sequential marker -mpi can certainly exist as a suffix on isolated verbs. Still, within the SVCs these morphemes could also be analysed as mere connectors required for the construction of certain kinds of verb chains and not any more tightly connected to the preceding verb than to the following one.
I was mostly referring to the normal inflectional morphology that appears on the verbs: But the -ra and -mpi suffixes are a good point. Perhaps word-internal root serialization should be considered a different beast from whole verb serialization?
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
Salmoneus
MVP
MVP
Posts: 3033
Joined: 19 Sep 2011 19:37

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Salmoneus »

My layman's interpretation: 'serial verb construction' is used in cases where two or more verbally things are used in a construction in a way that is idiomatically and morphologically somewhere between a compound verb and two fully independent verbs.
User avatar
gach
MVP
MVP
Posts: 513
Joined: 07 Aug 2013 01:26
Location: displaced from Helsinki

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by gach »

Micamo wrote:Perhaps word-internal root serialization should be considered a different beast from whole verb serialization?
Yeah, that sounds like a good convention. There might not be any true boundaries between the two phenomena but at least provisionally it's good to acknowledge a possible distinction between the two.
Salmoneus wrote:My layman's interpretation: 'serial verb construction' is used in cases where two or more verbally things are used in a construction in a way that is idiomatically and morphologically somewhere between a compound verb and two fully independent verbs.
That sounds like it catches the basic motivation to use the term.
ImageKištaLkal sikSeic
thetha
mayan
mayan
Posts: 1545
Joined: 29 Apr 2011 01:43

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by thetha »

Micamo wrote: So here's a question for the thread: Why do some languages (Yoruba, Edo) require their verbs to be uninflected to participate in Serial Verb Constructions, while others (Lango, Yimas, Kalam) apparently don't have this restriction?
Yoruba doesn't inflect verbs at all, what do you mean?
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Micamo »

Teddy wrote:Yoruba doesn't inflect verbs at all, what do you mean?
I could have sworn Yoruba had a tense-aspect suffix like Edo did: My bad. Just Edo then, I guess.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
k1234567890y
mayan
mayan
Posts: 2400
Joined: 04 Jan 2014 04:47
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by k1234567890y »

an interruption:

I have created the following structure for Lonmai Luna:
"V1 nel V2" means "doing V1 and V2 simultaneously"

"V1 sum V2" means "doing V1 then V2"

for example:

dewa awat kor oli bola on imon sel sum takal disi on imai sel wampla.
rain "shower" on to body the of.INALIENABLE 1.SG and.then make(causative) cloth the of.ALIENABLE 1.SG be.wet.
The rain showered on my body and (then) made my clothes wet.

alen Kirino lusi nel leksi de fodo alen Kuroneko
PND Kirino discuss and.simultaneously argue INDEF accompany.with PND Kuroneko
Kirino discusses and argues something with Kuroneko
Can this be seen as an instance of SVC?
I prefer to not be referred to with masculine pronouns and nouns such as “he/him/his”.
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by Micamo »

Only if the english translations also count as SVCs, and I'd say they don't.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by eldin raigmore »

k1234567890y wrote:an interruption:
I have created the following structure for Lonmai Luna:
"V1 nel V2" means "doing V1 and V2 simultaneously"
"V1 sum V2" means "doing V1 then V2"

for example:
dewa awat kor oli bola on imon sel sum takal disi on imai sel wampla.
rain "shower" on to body the of.INALIENABLE 1.SG and.then make(causative) cloth the of.ALIENABLE 1.SG be.wet.
The rain showered on my body and (then) made my clothes wet.

alen Kirino lusi nel leksi de fodo alen Kuroneko
PND Kirino discuss and.simultaneously argue INDEF accompany.with PND Kuroneko
Kirino discusses and argues something with Kuroneko
Can this be seen as an instance of SVC?
I wouldn't see it that way, but perhaps it could be seen that way.
For SVCs there is usually not any conjunction (unless I am mistaken).
"Nel" and "sum" are (verb-to-verb) coordinating conjunctions in Lonmai Luna, as you've described and exemplified, I think.

The verb series in an SVC can be thought of as describing a single action, with just the one agent (grammatically one, that is; the agent-phrase may actually refer to a group of people or things, but if so that group is the agent of every component verb of the SVC), and if they are (or it is) transitive, usually just the one patient IIANM.

If the constituent "sub-verbs" aren't simultaneous they are usually listed in the order in which they occur (or occurred or will occur).
But the addressee is expected to be able to tell which pairs of them are simultaneous with each other and which, instead, are consecutive one after the other. The speaker's SVC supplies no help on that score; if the addressee needs the help, s/he has to ask for clarification.

I could be wrong about any particular exceptional SVC language. Or, I could be wrong about SVC languages in general. But this is what I presently understand SVC to mean.

[hr][/hr]

All of that being said, I think something like your "nel/sum" alternation does happen in some natlang(s) or other; and I think it's a cool idea, even if it's one you just made up out of whole cloth.

Could you have
V1 nel V2 nel V3 sum V4 nel V5 nel V6 sum V7 nel V8 nel V9
?
What would it mean?
Could it ever be at all ambiguous or misleading?
Where would the agent(s?) and patient(s?) go?
User avatar
gach
MVP
MVP
Posts: 513
Joined: 07 Aug 2013 01:26
Location: displaced from Helsinki

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by gach »

I'd also rather be talking about coordination of separate independent predicates.

The next bit goes mostly under the title of other oddities. Here's an example from Choctaw that parallels the construction of complex verb stems in Yimas using the morphemes -ra and -mpi except here the all the verbs remain separate words

Sa-baa-washooha-t hilha-t taloowa-h!
SG1.II-COM-play-SS dance-SS sing-NPST
"Sing, dance, and play with me!"

This is interesting since morphologically only the last of the three verbs is finite as evidenced by the fact that only it carries a tense suffix. The two preceding verbs are treated as dependent forms that share all the inflectional categories indicated for the finite verb and are both marked by the same subject suffix -t. However, all the three verbs are treated syntactically as components of a single complex predicate and both of the inflectional prefixes are applied to this string as if it were a single verb stem.
ImageKištaLkal sikSeic
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by eldin raigmore »

gach wrote:This is interesting since morphologically only the last of the three verbs is finite as evidenced by the fact that only it carries a tense suffix. The two preceding verbs are treated as dependent forms that share all the inflectional categories indicated for the finite verb and are both marked by the same subject suffix -t. However, all the three verbs are treated syntactically as components of a single complex predicate and both of the inflectional prefixes are applied to this string as if it were a single verb stem.
Typically in SVCs only one of the verbs is finite. Or, at least, so I (mis?)understand.
User avatar
k1234567890y
mayan
mayan
Posts: 2400
Joined: 04 Jan 2014 04:47
Contact:

Re: Light Verbs, Serial Verb Constructions, & other verb odd

Post by k1234567890y »

eldin raigmore wrote:
k1234567890y wrote:an interruption:
I have created the following structure for Lonmai Luna:
"V1 nel V2" means "doing V1 and V2 simultaneously"
"V1 sum V2" means "doing V1 then V2"

for example:
dewa awat kor oli bola on imon sel sum takal disi on imai sel wampla.
rain "shower" on to body the of.INALIENABLE 1.SG and.then make(causative) cloth the of.ALIENABLE 1.SG be.wet.
The rain showered on my body and (then) made my clothes wet.

alen Kirino lusi nel leksi de fodo alen Kuroneko
PND Kirino discuss and.simultaneously argue INDEF accompany.with PND Kuroneko
Kirino discusses and argues something with Kuroneko
Can this be seen as an instance of SVC?
I wouldn't see it that way, but perhaps it could be seen that way.
For SVCs there is usually not any conjunction (unless I am mistaken).
"Nel" and "sum" are (verb-to-verb) coordinating conjunctions in Lonmai Luna, as you've described and exemplified, I think.

The verb series in an SVC can be thought of as describing a single action, with just the one agent (grammatically one, that is; the agent-phrase may actually refer to a group of people or things, but if so that group is the agent of every component verb of the SVC), and if they are (or it is) transitive, usually just the one patient IIANM.

If the constituent "sub-verbs" aren't simultaneous they are usually listed in the order in which they occur (or occurred or will occur).
But the addressee is expected to be able to tell which pairs of them are simultaneous with each other and which, instead, are consecutive one after the other. The speaker's SVC supplies no help on that score; if the addressee needs the help, s/he has to ask for clarification.

I could be wrong about any particular exceptional SVC language. Or, I could be wrong about SVC languages in general. But this is what I presently understand SVC to mean.

[hr][/hr]

All of that being said, I think something like your "nel/sum" alternation does happen in some natlang(s) or other; and I think it's a cool idea, even if it's one you just made up out of whole cloth.

Could you have
V1 nel V2 nel V3 sum V4 nel V5 nel V6 sum V7 nel V8 nel V9
?
What would it mean?
Could it ever be at all ambiguous or misleading?
Where would the agent(s?) and patient(s?) go?
for "S ((V1 nel V2 nel V3) sum (V4 nel V5 nel V6) sum (V7 nel V8 nel V9)) O", it is interpreted as "S does/did "V1+V2+V3" on O, then S does/did "V4+V5+V6" on O, then S does/did "V7+V8+V9" on O"

here I assume that all Vs take the same object.

"sum" and "nel" should always be used with verbs or verb phrases, and they must always follow a verb/verb phrase, moreover, phrases start with "nel" and "sum" don't have their own subjects, so phrases like "*alen Nepgear lewep bahi nel alen Nepu-nepu seplar" are ungrammatical, and phrases like "alen Nepgear lewep bahi nel seplar" are grammatical.
I prefer to not be referred to with masculine pronouns and nouns such as “he/him/his”.
Post Reply