zompist bboard

a congress of convoluted conworldery
It is currently Wed Jul 26, 2017 9:29 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 25 posts ] 
Author Message
PostPosted: Fri Mar 31, 2017 11:44 am 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
Ive been naively using the accusative case for the agents of passive verbs for basically my whole conlanging experience. I seem to have conceived of passive voice as basically just an "opposite" of the active voice, so it seemed logical to me that there would be no difference in how the agent and patient were marked; only the meaning would be changed.

But looking at other languages I get the impression that this is a bad idea and doesnt line up with how natlangs generally do things, or even conlangs for that matter. It seems that there are a lot of possible choices. Ancient Greek used the genitive, Latin used the ablative, and it seems that in Late Latin they even used the accusative ... but this last one might not mean much, since the accusative also took over for the nominative in most nouns in Vulgar Latin. Meanwhile, Finnish seems to lack the use of agents in passive sentences altogether.

I expect that the answer to which case to use is basically up to a matter of taste, as there doesnt seem to be a wrong answer if even the accusative is attested. I'd be interested in advice, though, and examples of what other people have chosen (including from natlangs that I havent been able to find information on). I kind of like the accusative solution, to be honest, but I would be happier if there were at least one example of in a natural language other than the marginal case of Vulgar Latin which was losing its entire case system anyway.

Another idea, at least for some of my conlangs, would be to use the locative. This may seem an odd choice, but the locative in Poswa and Pabappa has a very wide range of meanings, as it resulted from the coalescence of several previously existing noun cases and retained most of their meanings. One of those meanings was what I call "becausative", which essentially means "because of ..." ... technically, this is not the locative but a secondarily inflected, "chiral" form of it, but I dont want to get bogged down in details. The idea of a becausative case seems natural to me, and I've been using it for about 20 years now, though it seems curiously difficult to find in natlangs. Anyway I think that would be potentially a good option as well, and would rarely cause confusion even in those words where it is identical to the locative.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Fri Mar 31, 2017 12:13 pm 
Smeric
Smeric
User avatar

Joined: Thu Oct 29, 2015 6:44 am
Posts: 2060
Location: suburbs of Mrin
I would think that the reason for why a prepositional case might be used is that the action is being done by the agent, so the agent needs a case mean by (basically instrumental). The canonical accusative has something done to it in contrast. I have no idea what is going with that Greek genitive, though. Maybe its related to this:
Wikipedia wrote:
Depending on the language, specific varieties of genitive-noun–main-noun relationships may include...
  • participation in an action:
  • as an agent ("She benefited from her father's love") – this is called the subjective genitive (Compare "Her father loved her", where Her father is the subject.)
  • as a patient ("the love of music") – this is called the objective genitive (Compare "She loves music", where music is the object.)

_________________
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť


Top
 Profile  
 
PostPosted: Fri Mar 31, 2017 12:16 pm 
Avisaru
Avisaru
User avatar

Joined: Mon Feb 29, 2016 6:34 am
Posts: 885
Location: The North
If you're using the "accusative" for "passive" agents, then that's not so much an active-passive distinction as a case of Austronesian alignment.

If you're going to permit agents in passive constructions (and that's not a given, as you say above with Finnish), then it's always in some kind of oblique case/preposition, which as you say can be pretty much anything: genitive, locative, instrumental etc. Your "becausative" case (often referred to simply as a "causative" case, which I think I have seen in a natlang or two in lieu of or in addition to a morphological causative on the verb, sounds like a good candidate too. The point of this is that a true passive demotes an agent from the status of a core argument and is a valency-decreasing voice: if you're using an accusative for the agent, then it's not a true passive since it's not valency decreasing.

_________________
https://frislander.tumblr.com/

First known on here as Karero


Top
 Profile  
 
PostPosted: Fri Mar 31, 2017 12:23 pm 
Smeric
Smeric
User avatar

Joined: Thu Oct 29, 2015 6:44 am
Posts: 2060
Location: suburbs of Mrin
Some deponent and semi-deponent verbs in Latin take the accusative or dative instead of the ablative. I think that the case used is mainly a case (no apologies for the wordplay) of semantics rather than taste.

_________________
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť


Top
 Profile  
 
PostPosted: Fri Mar 31, 2017 1:57 pm 
Sumerul
Sumerul
User avatar

Joined: Wed Dec 29, 2004 11:57 am
Posts: 3338
Location: Elsewhere
What cases do you have already? If we could see lists then we might have more idea of what might be appropriate - though I think Frislander's right to say there's quite a lot of freedom here.

_________________
@jsbaker750
seven-fifty.net


Top
 Profile  
 
PostPosted: Sat Apr 01, 2017 12:47 am 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
Frislander wrote:
If you're using the "accusative" for "passive" agents, then that's not so much an active-passive distinction as a case of Austronesian alignment.
OK thank you. I actually read the Wikipedia article for Austronesian alignment earlier, but misread a crucial part of it and didn't even pick up on the similarity. Still, although I love that system, it seems that it would be highly inconvenient to use in a verb-final, suffixing language like literally all of my conlangs are, since, while Im sure agent-trigger sentences are the predominant type, the listener wouldnt know for sure which argument was the agent and which was the patient until the last syllable of the last word in the sentence. I do still have a morpheme, -m, which acts like English "be-" (think of befriend*) and could thus fulfill a similar role of getting the accusative to take on the syntactical agent role while remaining grammatically the patient


Frislander wrote:
Your "becausative" case (often referred to simply as a "causative" case, which I think I have seen in a natlang or two in lieu of or in addition to a morphological causative on the verb, sounds like a good candidate too.
Well, I only used the name "becausative" since I had already used "causative" for its opposite, i.e. the "causing a(n)..." case. That case may well be entirely nonexistent in natlangs, since I imagine it is usually expressed by a verb, but I dont see it as being particularly unwieldy in the example sentences I've written, such as

Plebum lappabel, puppiep.
The rain fell down, causing a flood.

Poswa has verbal causatives as well, and in fact, all they really are is a verbal affix on top of a noun in the causative case.

Using the "becausative" (I have also called it "circumstantial" just to give it a more formal sounding name) along with the -ž- passive verb ending would produce Poswa sentences like

Fappivem, pypappebeži.
Because of the snowball, I was hit in the face by it.
Which could be more concisely translated as
The snowball hit me in the face.

Probably the pause, marked by a comma, would be omitted in most instances of this type of speech.

Since inanimate objects can only be the syntactical agent in a passive verb, essentially they are case-marked when they are the agent and not when they are the patient. Does this make Poswa a split-ergative language, even though the case marking that goes on the promoted inanimate noun is not a separate "ergative" case but rather a special use of a different case?

I suppose that this special use will in fact be so common that it could well outnumber all other uses of the becausative/circumstantial case, or whichever case I end up going with, and perhaps I could simply rename the case to "ergative" and explain that Poswa's ergative case, like its other noun cases, has a wide assortment of extended uses in addition to its basic and most common function.


Frislander wrote:
The point of this is that a true passive demotes an agent from the status of a core argument and is a valency-decreasing voice: if you're using an accusative for the agent, then it's not a true passive since it's not valency decreasing.
This is probably the part Im having the most trouble accepting, since I have some other weird things going on in at least Poswa, and probably soon other languages, where inanimate objects can never be the agents of a transitive verb, but can serve as the syntactical agent of any passive verb, such as "I was hit in the face by the falling lampshade". To me, those sentences seem like they should be just as simple to express, from the speaker's point of view, as the normal type such as "i hit the lampshade". But perhaps what I'm really thinking is that I don't really like the animacy/inanimacy distinction I've come up with, and want a setup more like English where any noun can be the agent of any verb. I'll have to do some more thinking on that, since I really like animacy.


meþru wrote:
Some deponent and semi-deponent verbs in Latin take the accusative or dative instead of the ablative. I think that the case used is mainly a case (no apologies for the wordplay) of semantics rather than taste.
Thanks, that's good to know, too, and it reminds me that I've already done something very similar with ordinary verbs, where there are some verbs that govern the genitive or the locative case instead of the accusative. So if ordinary active verbs can govern more than one case, surely passive verbs could too. (Verbs governing the locative are generally metaphorical analogs of literal locatives, e.g. "to need" is LOC + an otherwise unanalyzable verb stem, which strongly resembles the English phrase "I'm in need of...."

Curlyjimsam wrote:
What cases do you have already? If we could see lists then we might have more idea of what might be appropriate - though I think Frislander's right to say there's quite a lot of freedom here.

In Poswa (my main conlang), there are nominative, accusative, locative, possessive, essive, instrumental, all of which have quite a variety of meanings since the parent language two steps back had had perhaps as many as 21 cases, of which only six were locatives and all the rest described various syntactical relationships. The essive could probably be better described as a partitive case, except that it also means "made of, composed of".






*I get that this could be analyzed another way, since friendship is a reciprocal process.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Sat Apr 01, 2017 9:39 am 
Sanno
Sanno
User avatar

Joined: Thu Jan 15, 2004 5:00 pm
Posts: 16045
Location: One of the dark places of the world
"Symmetrical passives" like this are found in Indonesia, in a greatly reduced form of "austronesian alignment". However, these languages also generally have an asymmetric passive like that of English.

Soap wrote:
Frislander wrote:
If you're using the "accusative" for "passive" agents, then that's not so much an active-passive distinction as a case of Austronesian alignment.
OK thank you. I actually read the Wikipedia article for Austronesian alignment earlier, but misread a crucial part of it and didn't even pick up on the similarity. Still, although I love that system, it seems that it would be highly inconvenient to use in a verb-final, suffixing language like literally all of my conlangs are, since, while Im sure agent-trigger sentences are the predominant type, the listener wouldnt know for sure which argument was the agent and which was the patient until the last syllable of the last word in the sentence.

Why is that a problem? As the old joke goes, you sometimes have to wait a very long time in a language like German before you finally, at the last moment, discover what the verb is...


Anyway, the most common cases for semantic agents of passive verbs are, so far as I'm aware, and logically enough, the genitive and the instrumental. It is presumably not a coincidence that this aligns with the case syncretisms of ergative languages.

_________________
Blog: http://vacuouswastrel.wordpress.com/

But the river tripped on her by and by, lapping
as though her heart was brook: Why, why, why! Weh, O weh
I'se so silly to be flowing but I no canna stay!


Top
 Profile  
 
PostPosted: Sat Apr 01, 2017 10:40 am 
Smeric
Smeric
User avatar

Joined: Thu Oct 29, 2015 6:44 am
Posts: 2060
Location: suburbs of Mrin
What is the difference between your causative and nominative? BTW, causative or casual case are the mre usual names for your "becausative".

_________________
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť


Top
 Profile  
 
PostPosted: Wed Apr 05, 2017 7:52 pm 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
I hadnt noticed this thread until now, but it seems to be helping me (lol), especially SHG's post at the bottom. Even by coincidence, the Yukulta language being described there happens to have a similar case syncretism to what I was considering using in Poswa.


Salmoneus wrote:
"Symmetrical passives" like this are found in Indonesia, in a greatly reduced form of "austronesian alignment". However, these languages also generally have an asymmetric passive like that of English.
Yeah, I will probably keep in at least one morpheme that simply flips the Subject and Object and remains an active-voice verb. That feature has been in my conlangs for over 20 years.

Salmoneus wrote:
Why is that a problem? As the old joke goes, you sometimes have to wait a very long time in a language like German before you finally, at the last moment, discover what the verb is...
Yes and Im not afraid to make a language that is clumsy and inefficient on purpose. In fact I get a lot of pleasure out of that. For example, most of Poswa's verb inflections are two syllables long, and some are considerably longer, even though they evolved from monosyllabic ones and the extra syllables carry almost no additional information. But, I only do that with features I like. I think mixing SOV word order with a grammatical quirk that marks the identities of the S and O on the verb would be highly annoying even if the vast majority of sentences were of one type. In fact I created a rule awhile back that if the first noun in a sentence is not the subject, it must be marked for the person of the subject, so that e.g. things like "banana-2ps eat-2ps" for "You ate the banana" show up.

mèþru wrote:
What is the difference between your causative and nominative? BTW, causative or casual case are the mre usual names for your "becausative".
Nominative is always bare stem, (be)causative adds an infix -i- and a suffix -m, though that may not be apparent since there are so many sound changes built into the morphology, such that the -i- is really better described as a consonant mutator.

I used the name "becausative" because I had already used "causative" for something else. Either way, though, I will try to come up with a new name for both of them. I may end up calling the becausative the "ergative" at least in some of my languages.

Part of the reason why I have so many noun cases and have a hard time figuring out how to use them is that Im determined to make a language that is entirely without pronouns and has very few particles ... right now I have four conjunctions, basically corresponding to "if", "and", "or", and "but", and every other word in the entire language is either a noun or a verb (or both). Even "hello" is a verb. There is no word for "because" or "therefore", so I have to use inflections instead, or in some cases, fully inflected verbs.

Thank you all again for your help. I still havent decided what to do but I have a much clearer picture than I did a week ago.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Wed Apr 05, 2017 11:35 pm 
Sumerul
Sumerul
User avatar

Joined: Sat Jul 19, 2008 1:55 am
Posts: 2791
Location: British Columbia / Colombie Britannique, Canada
Soap wrote:
Ancient Greek used the genitive, Latin used the ablative,
Actually, for animate agents, which are the prototypical agents when we talk of the agents of passive verbs, Latin generally used the preposition ab... Though its true that for inanimate agents it sometimes used the ablative alone.

Quote:
and it seems that in Late Latin they even used the accusative ... but this last one might not mean much, since the accusative also took over for the nominative in most nouns in Vulgar Latin.

[...] I kind of like the accusative solution, to be honest, but I would be happier if there were at least one example of in a natural language other than the marginal case of Vulgar Latin which was losing its entire case system anyway.
Do you happen to have a source or examples for the use of the accusative for agents of passive verbs in Late Latin? If you mean using the preposition ab with the accusative, that would not be very surprising (early medieval Latin often replaces the ablative with the accusative like this), but if you mean the accusative alone, then...

_________________
ZBB Conlang Index


Top
 Profile  
 
PostPosted: Thu Apr 06, 2017 5:00 am 
Smeric
Smeric
User avatar

Joined: Sun Nov 04, 2012 3:40 pm
Posts: 1679
Location: Anarcho-Centrist Island
Isn't this basically just syncretism between two different cases? Seems okay to me (although maybe don't call the form "accusative").

_________________
陳第 wrote:
蓋時有古今,地有南北;字有更革,音有轉移,亦勢所必至。

R.Rusanov wrote:
seks istiyorum
sex want-PRS-1sg

Read all about my excellent conlangs
Basic Conlanging Advice


Top
 Profile  
 
PostPosted: Thu Apr 06, 2017 7:05 am 
Smeric
Smeric
User avatar

Joined: Thu Oct 29, 2015 6:44 am
Posts: 2060
Location: suburbs of Mrin
I meant what are the semantic differences between nominative and what you call the causative (not what you called the becausitive).

_________________
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť


Top
 Profile  
 
PostPosted: Wed Apr 12, 2017 10:03 am 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
Well Ive kind of hit a wall, I think, since it turns out there is no single correct path forward.

The only thing Ive changed my mind on is that I think now that it does make sense to consider passive verbs a valency-decreasing operation, since an inanimate noun cannot choose to act on an animate. So I may use the intransitive endings rather than the transitive endings, at least when the agent is an inanimate noun. This would give the language the odd quirk of having passive sentences sometimes be a syllable shorter than active ones.

e.g.

Pepopep vebi.
I saw the car.

But

Pepoppem vi.
I was seen by the car.

That seems a little too weird for me right now, but I'll sleep on it.

Quote:
Do you happen to have a source or examples for the use of the accusative for agents of passive verbs in Late Latin? If you mean using the preposition ab with the accusative, that would not be very surprising (early medieval Latin often replaces the ablative with the accusative like this), but if you mean the accusative alone, then...
I believe it was with a preposition, yes, but as I said I wasnt really leaning heavily on Latin here because it was losing its case system anyway.

Quote:
Isn't this basically just syncretism between two different cases? Seems okay to me (although maybe don't call the form "accusative").]
I can see that, yes, but technically (not that it matters) this was present from the formation of the case system, and therefore not syncretism, but an original trait.

mèþru wrote:
I meant what are the semantic differences between nominative and what you call the causative (not what you called the becausitive).
The causative would just mean "causing X", more or less the inverse of the becausative. I'd think it'd be at least attested in natlangs, but it seems to be either rare or completely nonexistent. Im not going to let it worry me though, the causative is just the right-chiral form of the accusative anyway, meaning it evolved from a special use of the accusative.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Thu Apr 20, 2017 1:54 am 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
OK, I think what I'm going with is basically this:

1) All passive verbs imply 3rd person agents. This is a side effect of the lack of pronouns in the language: since there are no words for "I" or "you", there cannot be inflected forms of them either, and the marking of 1st and 2nd person must be done entirely on the verb. Since it would be to silly to have two verbal affixes, where one means "I verb you" and another means "you are verbed by me", I will simply say that all passive verbs imply a 3rd person agent even if none is given in the sentence.

2) The agent of a passive verb can be omitted if the context is clear. Therefore, if the agent of a given passive verb was named in a previous sentence, it can be in *any* noun case, because the simple fact that it was mentioned is enough to establish that it will automatically be the agent of any passive verb in an upcoming sentence, up until another 3rd person noun is introduced.

3) My first instinct is that this new idea would mean that the most sensible case for the agent of a passive verb, when not introduced in a previous clause, is actually the nominative, since it is the likeliest form to appear in such a previous clause. But since 3rd person entities can also be the *patient* of a passive verb, I need to figure out which noun case to use for them as well.

I would probably be best off to sleep on it a little bit more and pick out the most logical option among the various oblique cases, and among those, I'm still leaning towards the dative case, which is what I have been calling becausative up above, and seems to have the meaning that most closely resembles an ergative case in a traditional split-ergative system.

I will try to use established names for my grammatical peculiarities whenever possible, but since I'm so fond of doing nearly everything by way of noun and verb inflection, some things will probably always remain unnamable. e.g. one of my favorite noun inflections translates roughly as "Get me a ...", e.g. from wawampaelum "balloon" is formed wawampaelioppub "Get me a balloon!" I could call it the impulsive imperative possessive, but sometimes I think it's just a lot better to describe it by using it in a sentence.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Sun May 14, 2017 6:20 pm 
Visanom
Visanom
User avatar

Joined: Thu May 24, 2007 11:34 pm
Posts: 6477
Location: Stockholm
Sorry for such a late reply, but the agent can be expressed with the genitive case in Finnish. Although this happens very seldomly.

Mies tuli koiran puremaksi.
mies-0 tul-i-0 koira-n pur-ema-ksi
man-NOM come-PST-3SG dog-GEN bite-INF3-TRANSL
The man was bitten by the dog.

INF3 = third infinitive
TRANSL = translative case

_________________
Image
Most recent quiz:
Eurovision Song Contest 2017


Top
 Profile  
 
PostPosted: Sat Jun 10, 2017 9:32 pm 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
Qwynegold wrote:
Sorry for such a late reply, but the agent can be expressed with the genitive case in Finnish. Although this happens very seldomly.

Mies tuli koiran puremaksi.
mies-0 tul-i-0 koira-n pur-ema-ksi
man-NOM come-PST-3SG dog-GEN bite-INF3-TRANSL
The man was bitten by the dog.

INF3 = third infinitive
TRANSL = translative case
Sorry, I read your reply the day you posted it, but just now came and noticed that even the verb apparently has a case marker on? I'm guessing "bite" in this sentence isn't really a verb at all but a sort of verbal noun?

That's an interesting idea.

But I think I am going to stick with what I decided about halfway through the thread and use the -m noun case on the "object" and the nominative case on the "subject" of the verb. This lines up with another idea I had to use the locative case (also marked with -m) for verbs in which the subject of the verb is affected but the object is not. (e.g. "I punched the brick wall" and "I punched the piece of paper" would have the same verb, but different cases for the object).

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Sun Jun 11, 2017 1:28 am 
Sumerul
Sumerul
User avatar

Joined: Sat Jul 19, 2008 1:55 am
Posts: 2791
Location: British Columbia / Colombie Britannique, Canada
Soap wrote:
Qwynegold wrote:
Sorry for such a late reply, but the agent can be expressed with the genitive case in Finnish. Although this happens very seldomly.

Mies tuli koiran puremaksi.
mies-0 tul-i-0 koira-n pur-ema-ksi
man-NOM come-PST-3SG dog-GEN bite-INF3-TRANSL
The man was bitten by the dog.

INF3 = third infinitive
TRANSL = translative case
Sorry, I read your reply the day you posted it, but just now came and noticed that even the verb apparently has a case marker on? I'm guessing "bite" in this sentence isn't really a verb at all but a sort of verbal noun?

Hmm? The only genitive case marker I see is attached to "dog", not the verb "bite".

_________________
ZBB Conlang Index


Top
 Profile  
 
PostPosted: Sun Jun 11, 2017 1:28 am 
Sumerul
Sumerul
User avatar

Joined: Sat Jul 19, 2008 1:55 am
Posts: 2791
Location: British Columbia / Colombie Britannique, Canada
Soap wrote:
Qwynegold wrote:
Sorry for such a late reply, but the agent can be expressed with the genitive case in Finnish. Although this happens very seldomly.

Mies tuli koiran puremaksi.
mies-0 tul-i-0 koira-n pur-ema-ksi
man-NOM come-PST-3SG dog-GEN bite-INF3-TRANSL
The man was bitten by the dog.

INF3 = third infinitive
TRANSL = translative case
Sorry, I read your reply the day you posted it, but just now came and noticed that even the verb apparently has a case marker on? I'm guessing "bite" in this sentence isn't really a verb at all but a sort of verbal noun?

Hmm? The only genitive case marker I see is attached to "dog", not the verb "bite".

_________________
ZBB Conlang Index


Top
 Profile  
 
PostPosted: Sun Jun 11, 2017 10:10 am 
Smeric
Smeric
User avatar

Joined: Mon Feb 17, 2003 11:03 am
Posts: 2097
Location: displaced from Helsinki
It's the translative case ending -ksi on the nominalised verb puremaksi. It's used on nominalised verbs together with the finite auxiliary tulla ("come/become") to form a few of these not as commonly used passive like inchoatives,

Tähtee-t tuli-vat syö-dy-(i)-ksi.
leftover-PL come-PL3 eat-PST.PASS.PART-(PL)-TRANSL
"The leftovers got eaten."

Tähtee-t tuli-vat koir-i-en syö-m-i-ksi.
leftover-PL come-PL3 dog-PL-GEN eat-MA-PL-TRANSL
"The leftovers got eaten by the dogs."

I'm labelling the -m(a)- affix simply as MA in the glosses since it's a bit more complex than just being the core of the adverbial infinitives traditionally called the third infinitives. It's also used as the so called agent participle which is the counterpart of the past passive participle when the agent of the verb is specified (using the genitive case and/or possessive suffixes),

avat-tu ovi
open-PST.PASS.PART door
"opened door"

vartija-n avaa-ma ovi
guard-GEN open-MA door
"door opened by a guard"

(sinu-n) avaa-ma-si ovi
(SG2-GEN) open-MA-SG2.POSS door
"door opened by you"

In my view the -m(a)- affix in the passive inchoative constructions is specifically used as a participle since the distribution of it and the past passive participle is the same there as it's with the attributively used participles in the second group of examples.

There's also a GEN + toimesta construction for passive agents in Finnish (lit. "by someone's action", GEN + action-ELA). However, this is a calque from the passive agent constructions used e.g. in English or Swedish and is generally bad style. With this construction you can say things like

Juhla-t avat-tiin pormestari-n toime-sta.
festival-PL open-PASS mayor-GEN action-ELA
"The festival was opened by the mayor."

But really, if you need to state the agent, it's more idiomatic to just use an active transitive sentence and alter the word order accordingly if the patient is the sentential topic,

Juhla-t avas-i pormestari.
festival-PL open-SG3.PST mayor
"The festival was opened by the mayor."


Top
 Profile  
 
PostPosted: Fri Jul 07, 2017 9:32 am 
Sanci
Sanci

Joined: Wed Jul 05, 2017 7:00 am
Posts: 21
Frislander wrote:
If you're using the "accusative" for "passive" agents, then that's not so much an active-passive distinction as a case of Austronesian alignment.

If you're going to permit agents in passive constructions (and that's not a given, as you say above with Finnish), then it's always in some kind of oblique case/preposition, which as you say can be pretty much anything: genitive, locative, instrumental etc. Your "becausative" case (often referred to simply as a "causative" case, which I think I have seen in a natlang or two in lieu of or in addition to a morphological causative on the verb, sounds like a good candidate too. The point of this is that a true passive demotes an agent from the status of a core argument and is a valency-decreasing voice: if you're using an accusative for the agent, then it's not a true passive since it's not valency decreasing.


What about Indonesian passive:

Aku membeli buku : I bought a book
Buku(nya) dibeli Dio : The book was bought by Dio
Dio membelikan-ku buku : Dio bought me a book
Aku dibelikan Dio buku : ?? (same meaning as above but with passive voice.)


Top
 Profile  
 
PostPosted: Fri Jul 07, 2017 9:35 am 
Sanci
Sanci

Joined: Wed Jul 05, 2017 7:00 am
Posts: 21
Frislander wrote:
If you're using the "accusative" for "passive" agents, then that's not so much an active-passive distinction as a case of Austronesian alignment.

If you're going to permit agents in passive constructions (and that's not a given, as you say above with Finnish), then it's always in some kind of oblique case/preposition, which as you say can be pretty much anything: genitive, locative, instrumental etc. Your "becausative" case (often referred to simply as a "causative" case, which I think I have seen in a natlang or two in lieu of or in addition to a morphological causative on the verb, sounds like a good candidate too. The point of this is that a true passive demotes an agent from the status of a core argument and is a valency-decreasing voice: if you're using an accusative for the agent, then it's not a true passive since it's not valency decreasing.


What about Indonesian passive:

Aku membeli buku : I bought a book
Buku(nya) dibeli Dio : The book was bought by Dio
Dio membelikan-ku buku : Dio bought me a book
Aku dibelikan Dio buku : ?? (same meaning as above but with passive voice.)

Although Indonesian is Austronesian, I don't think Indonesian also have Austronesian alignment as Indonesian is a non Philipines Nuclear Malayo-Polynesian language. But the agent of the passive verb is positioned as if it is an object.


Top
 Profile  
 
PostPosted: Sun Jul 09, 2017 6:41 pm 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
gach wrote:
It's the translative case ending -ksi on the nominalised verb puremaksi. It's used on nominalised verbs together with the finite auxiliary tulla ("come/become") to form a few of these not as commonly used passive like inchoatives,

Tähtee-t tuli-vat syö-dy-(i)-ksi.
leftover-PL come-PL3 eat-PST.PASS.PART-(PL)-TRANSL
"The leftovers got eaten."

Tähtee-t tuli-vat koir-i-en syö-m-i-ksi.
leftover-PL come-PL3 dog-PL-GEN eat-MA-PL-TRANSL
"The leftovers got eaten by the dogs."
I like this idea too, and I might want to use it in a different conlang.

Akangka wrote:
What about Indonesian passive:

Aku membeli buku : I bought a book
Buku(nya) dibeli Dio : The book was bought by Dio
Dio membelikan-ku buku : Dio bought me a book
Aku dibelikan Dio buku : ?? (same meaning as above but with passive voice.)

Although Indonesian is Austronesian, I don't think Indonesian also have Austronesian alignment as Indonesian is a non Philipines Nuclear Malayo-Polynesian language. But the agent of the passive verb is positioned as if it is an object.
I like that. I dont think I've ever seen a language where the passive verb form is actually shorter (at least in this one verb) than the active verb form. The setup I'm looking at for my conlangs is similar in that the passive and active constructions are also about the same length, but different because it has partly free word order and obligatory case marking of nouns (though the nominative case marker is zero).

As a side effect of my intentional insistence that Poswa be entirely without pronouns, the only verb endings available for sentences with 3rd person agents and 1st or 2nd person patients are the passive verb endings. Furthermore, another side effect of the lack of pronouns is that, just by hearing the first word of the sentence, the listener knows whether the agent of the sentence is a 3rd person entity or not. This may seem bizarre and unworkable, but I think I can follow the model of any language in which inanimate nouns are not allowed to be the subject of a transitive verb. The difference is merely that in the Poswa worldview, the entire world apart from the speaker and listener (even including deities) consists of inanimate objects.

This means that it would be highly inconvenient to use a formula for the passive verbs that makes passive sentences signficantly longer than active ones. I admire the symmetry of the Indonesian setup, and I think that I will go with a symmetric system but have it require a different noun case ending for the 3rd person agent and the 3rd person patient, so that the listener knows which way the sentence is going just from hearing the first word. I think that natlangs generally would not need this distinction because I dont think there exists a natlang anywhere on Earth in which all 3rd person entities are lumped together with inanimates and therefore require passive verbs.

By assigning a different case to the 3rd person agent in a passive sentence, the speaker also has the ability to appreviate the sentence by omitting the verb, as is done in English. ("I just got bitten by a flea!" / "By a what?" / "By a flea!") This makes sense because it mirrors the active-voice construction of the same verb, which corresponds to English constructions like ("I just picked some apples." / "What kind?" / "Green apples!") ... in Poswa, the sentence "green apples" would require an accusative case marker on the word for apple, just as it would in the first sentence.

Below are the Poswa translations of the sentences above; note that -o marks the 1st person and that a 3rd person noun, whether acting as agent or patient, takes a suffix that agrees with the verb and therefore also ends in -o. -m(b)- is the case marker I'm using for the agent of a passive verb,* and -p- is the case marker for the accusative case.
:
Tipambo tšuppažo!
I just got bitten by a flea!

Pampis?
By (a) what?

Tipam!
By a flea!

Tatepo weliuppabo.
I just picked some apples.

Buweppis?
What kind?

Tatep lippa!
Green apples!

There is no purple in the English glosses on the second set of sentences because English uses a zero morph for this function, considering it the basic setup, in use for almost all verbs, whereas in English passive verbs are rare and are marked by an additional morpheme. By contrast, in Poswa active and passive sentences are on equal ground and both have the same number of morphemes.

The symmetry of the system is also reflected in that, in both sets of sentences, the only verbal person markers that ever appear are the 1st person markers, because even when the person is being bitten by a flea, in Poswa the flea is lower on the animacy hierarchy and therefore this is considered a 1st person passive activity rather than a 3rd person active one. It is no different in the Poswa worldview than picking apples, except that, since the word order is the same on both, the first word of the sentence (which is also the only noun in both cases) is marked with different noun cases to distinguish the two paths of action. The verb marker is also different, which means that the sentence could also be abbreviated to a verb-only form, as in

Bupis?
(The flea) did what (to you)?

Tšuppažo!
It bit me!

(identical to the locative, so I would call it the locative, but others might analyze it as case syncretism as discussed earlier in the thread, particularly since there will likely be more use of this case as agent-of-passive-verb than as a proper locative)

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
PostPosted: Fri Jul 14, 2017 7:30 am 
Sanci
Sanci

Joined: Wed Jul 05, 2017 7:00 am
Posts: 21
Actually, in Indonesian, there is symmetry because we conflate the doer and the direct object, thus looking like it is just swapped. "By" is translated "oleh," however, it is optional and the passive can be done just by placing the original subject on direct object slot. In my opinion, it is only Austronesian alignment if the there is trigger like locative trigger, benefactive trigger, etc.

Soap wrote:
the only verb endings available for sentences with 3rd person agents and 1st or 2nd person patients are the passive verb endings.


Why? You can also make your verbs agree to the direct object. By the way I feel like this is a Direct-Inverse system, but I might be wrong. I think that Indonesian active-passive system, Austronesian focus system, and Direct-Inverse systems are so similiar.


Top
 Profile  
 
PostPosted: Fri Jul 14, 2017 8:58 am 
Avisaru
Avisaru
User avatar

Joined: Mon Feb 29, 2016 6:34 am
Posts: 885
Location: The North
Akangka wrote:
Soap wrote:
the only verb endings available for sentences with 3rd person agents and 1st or 2nd person patients are the passive verb endings.


Why? You can also make your verbs agree to the direct object. By the way I feel like this is a Direct-Inverse system, but I might be wrong. I think that Indonesian active-passive system, Austronesian focus system, and Direct-Inverse systems are so similiar.


This is a kind of direct-inverse system yes. Some languages do handle things this way because they only ever mark the person of one argument and they want to keep the most "important" argument (i.e. either the speaker or listener) prominent whenever it is present. I also think you are right to note the similarity with the Indonesian passive.

_________________
https://frislander.tumblr.com/

First known on here as Karero


Top
 Profile  
 
PostPosted: Sat Jul 15, 2017 9:49 pm 
Osän
Osän
User avatar

Joined: Sun Feb 16, 2003 2:57 pm
Posts: 11057
Location: šinambiem bambam
Akangka wrote:
Soap wrote:
the only verb endings available for sentences with 3rd person agents and 1st or 2nd person patients are the passive verb endings.

Why? You can also make your verbs agree to the direct object.
In another conlang, I'll probably try that, and Im pretty sure I know how to evolve double person marking diachronically, but I really like the "inefficient on purpose" setup I have in Poswa right now. Generally I design languages to be concise and powerful but having a few weak spots can enhance the language's character.

I think Poswa could be analyzed as lacking passive verbs altogether, but that would require explaining why the verb only ever marks the agent or the patient, but not both. I'll stick with just calling these verbs passive for the time being.

Frislander wrote:
Akangka wrote:
By the way I feel like this is a Direct-Inverse system, but I might be wrong. I think that Indonesian active-passive system, Austronesian focus system, and Direct-Inverse systems are so similiar.


This is a kind of direct-inverse system yes. Some languages do handle things this way because they only ever mark the person of one argument and they want to keep the most "important" argument (i.e. either the speaker or listener) prominent whenever it is present. I also think you are right to note the similarity with the Indonesian passive.
I like this. I wasnt aware that a setup resembling mine existed anywhere in the world, or that it had a name. I will try to use this system in my other languages whenever possible.

I want to make a language with asymmetrical gender relations, meaning that men and women are placed in different animacy levels and therefore cause verbs to conjugate differently depending on the gender of the agent (and possibly patient). I've been trying to do this for quite a while now but I have a difficult time figuring out where to start.

_________________
<newsboy> Tiger Woods charged with DUI in Florida 18:48:40
<newsboy> Tiger Kills Employee at U.K. Zoo 18:56:46


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 25 posts ] 

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group