Split ergativity and case question

If you're new to these arts, this is the place to ask "stupid" questions and get directions!
Post Reply
Nachtuil
greek
greek
Posts: 595
Joined: 21 Jul 2016 00:16

Split ergativity and case question

Post by Nachtuil »

Hello everyone,
Recently I have become interested in creating a split ergative language based on Sumerian where I'll have two animate genders and an inanimate gender and the animate genders use a nominative accusative structure but inanimate subjects use an ergative structure.

When I look at Sumerian, it has only absolutive and ergative case endings, the absolutive ending being a null suffix. There apparently is no nominative or accusative case suffox. Is it normal for split negative language to just reuse the the absolutive and ergative case endings to show nominative accusative constructions? I would expect to see nominative, accusative, ergative and absolutive case endings but realtiy defies these expectations.

I have been using the Wikipedia page on Sumerian grammar which is good and a Google books page on sumerian which explains it but not with the clear and explicit glosses that I so desperately crave.

I know the answer is out there but I can't grasp it regrettably so I am hoping to rely on the wisdom of the board here for a direct answer. Thank you for your patience. :)
Last edited by Nachtuil on 08 Nov 2016 09:03, edited 2 times in total.
Salmoneus
MVP
MVP
Posts: 3030
Joined: 19 Sep 2011 19:37

Re: Split ergativity and case question

Post by Salmoneus »

I don't have statistics to support this, but I think it would be not uncommon to have only two core cases.

You have the agent of a transitive and the patient of a transitive - that's two cases. In nom-acc languages, the subject of an intransitive take the case of the transitive agent; in erg-abs language, that of the transitive patient. Still two cases. It's just that sometimes the subject takes one and sometimes it takes the other, but it's still choosing from two options.

That's not necessarily the case, because of how ergativity can arise. If it arises from nominalisation, the ergative may be, or come from, a genitive, giving you three cases. If it arises from instrumental structures, the ergative may be, or come from, an instrumental, again giving you three cases. So a three-case solution should be possible, certainly.

But the plain two-case solution, which is all that is really needed and doesn't involve inventing a separate case for some agents of transitive verbs that isn't used elsewhere, is always going to be a logical alternative.

Four cases probably occurs somewhere, but seems way too many. Why would absolutive and nominative be distinguished?
User avatar
Frislander
mayan
mayan
Posts: 2088
Joined: 14 May 2016 18:47
Location: The North

Re: Split ergativity and case question

Post by Frislander »

Basically what Sal said, with a few further points of my own.

If the split occurs at the animate-inanimate boundary, then I find it very unlikely that's you'll even be using the ergative very much - inanimate transitive agents are extremely uncommon, both cross-linguistically and within languages, and some languages forbid them totally (e.g. Algonquian languages). If you do want them to be present, then I would follow Sal and have it be an instrumental case, e.g. in "the knife cut the bread" (not a terribly natural sentence), someone is definitely using the knife to do the cutting.

Alternatively, perhaps the system originated as a kind of fluid-S marking with a marked "agentive case" to denote a subject which fulfills an agent-like role. Then perhaps the system later breaks down, giving a marked nominative system in animates and an ergative system in inanimates.

Another further word on Sumerian: firstly what we know about Sumerian is severely restricted by the nature of the writing system, so everything you read should be taken with a pinch of salt. Secondly, don't just take the one-word case-terminology used in the article as-is: it will not be telling the full story. Finally, I don't see any evidence that Sumerian nouns aren't completely ergative: that is, that the split is only found in relation to pronouns/verb-marking vs. full nouns, in which case a separate set of accusative markers would not be necessary.
User avatar
KaiTheHomoSapien
greek
greek
Posts: 641
Joined: 15 Feb 2016 06:10
Location: Northern California

Re: Split ergativity and case question

Post by KaiTheHomoSapien »

Hittite has an ergative case only in the neuter (as opposed to the "common" gender). So Hittite has this set up:

COMMON

nom: -as
acc: -an
(no erg)

NEUTER

nom: -an
acc: -an
erg: -anza

Because like most older IE languages, the nominative and accusative have the same form in the neuter, so this is essentially an "absolutive" (just IE grammar tradition calls them nom/acc) and an additional ergative case was developed for transitive subjects.
Image
Nachtuil
greek
greek
Posts: 595
Joined: 21 Jul 2016 00:16

Re: Split ergativity and case question

Post by Nachtuil »

Thank you everyone for the responses. I am still contemplating and processing. Perhaps I will set it so one animate gender uses ergative as well as the inanimate gender. I would have imagined inanimate using ergative in sentences like "the snow covered their tracks" which I suppose could also covered with a passive "their tracks where covered in snow". Using the instrumental case for inanimate agents would definitely work. I will look into how Hittite handles things further.
Nachtuil
greek
greek
Posts: 595
Joined: 21 Jul 2016 00:16

Re: Split ergativity and case question

Post by Nachtuil »

I am reading about ergativity in some linguistic PDFs but I am on the verge of just using a full on ergative-absolutive system with restrictions on the inanimate agents or a tripartite system again with restrictions on inanimate agents. When I have more time I will work it out thoroughly.
Nachtuil
greek
greek
Posts: 595
Joined: 21 Jul 2016 00:16

Re: Split ergativity and case question

Post by Nachtuil »

I think I will go with a split afterall.
First and second person lead phrases will get nominative accusative marking and third person negative absolutive. Inanimates, in theory, will be be restricted from being agents but I need to research still thougy I understand I may need to use a passives and or antipassives.
Curlyjimsam
sinic
sinic
Posts: 221
Joined: 01 Sep 2010 15:31
Location: UK
Contact:

Re: Split ergativity and case question

Post by Curlyjimsam »

Nominative and absolutive will have the same endings significantly more often than not, I think. (This marking will often be zero.) In fact, no instances come to mind where they are different, but that doesn't mean there aren't any.

Having the same endings for ergative and accusative is rare but not unattested.

You can imagine a language where nominative=ergative and absolutive=accusative in terms of morphological marking, but I'm not sure if this ever actually happens.
The Man in the Blackened House, a conworld-based serialised web-novel
Post Reply