||"dawnbreak in the west"|
Friday, March 31, 2017
Whilst we’re fooling around on the southern and eastern Black Sea coast, I thought I’d check out “ergative-absolutive” languages. Split ergative languages are, or were, all over those shores: Kartveli "Georgian", the Indo-Aryan languages, and Knesian “Hittite”. Some scholars say the same of proto-Indo-European.
We who study Greek and Latin (and Arabic) are accustomed to accusative languages; we rarely see much else. Our languages have a nominative, for the subject (“He rests”). If the verb is transitive then we have an accusative, for the object (“something affects Him”). But that's not the only way to go about it. For a start, one could just do without case entirely and rely on word-order, like western Romance and her bastard child English. Or, one could rethink the whole case system...
Ergative languages seem more picky about that sentence’s subject. They see a restricted transitive-subject, which takes the ergative, implying that the subject does stuff; and the intransitive-subject plus the object, the absolutive, which means the guy is just there, either getting pushed by events or merely existing.
But pure ergativity is rare, as far as I know found only in the Pyrenees and even then the Basques might just be trolling Spaniards. On the other side of the Mittelmeer, the Caucasus mixes it up.
According to B George Hewitt, “Georgian – Ergative, Active, or What?” (heh) ed. N Bennett et al., Subject, Voice and Ergativity (SOAS, 1995), 197f: Georgian often expands the “absolutive”'s scope to the subject of transitive verbs, thus making a familiar nominative of it. In such sentences, Georgian for its object uses the dative. The Georgian language does have a special case for the subject of a transitive aorist (its object is then absolutive, like Basque), but this is so restricted that Georgians themselves don’t call its subject “ergative”. They call it the “narrative” case.
JC Catford, “the Classification of Caucasian Languages” ed. Sydney Lamb et al., Sprung from Some Common Source (CA: Stanford U Press, 1991), 232f. writes that North(west) Caucasian languages, like Kabardian, are optionally ergative. Such languages offer the ergative way to say “the boy is reading the book”, which would imply he was actively into it. Alternatively Kabardian offers the “intransitive” way, which to me sounds like the book was more interested in being read than the boy was in reading it. We’ve all been there.
Dennis Campbell, Zeitschrift fur Assyriologie 98 (de Gruyter, 2008), 262-94 sees “split ergativity in Hurrian” as well.
In and around Pontus, we have a wonderful Bronze Age window in the Hittite tablets. These are generally Late Bronze Age affairs. But several are ritual in nature, and these translate older texts. Sometimes they translate Middle Bronze Hittite. Recently Petra Goedegebuure from Chicago has been going through this evidence.
She notes that the Hittites had ways to “individualise” / “individuate” a noun, to mark it as atomic. As a root word nepis, “sky”, can be divided. But a poet (or, perhaps, a better-informed Hittite) might want to hint otherwise: that nobody can split the sky. The composer of the Ullikummi myth went with nepisa; older Hittite used nepisants. She then notes the resemblance between the earlier individuating suffix -ants and the newer suffix for an ergative subject.
Further illustrating how the Hittites changed their minds, Goedegebuure brings from a Netherworld ritual, the sentence “let this town and house become a ram”. Here we are fortunate to have its original and then its Neo-Hittite translation. Either version is cuneiform, so littered with those annoying Sumerograms, but not in what matters: the nominative house, normal parnaš, old atomic parnanzaš.
Kāššaza URU-az parnanzašš-a UDU.A.LUM DÙ-ru
As seen here, the Hittites held on to individuation, but did it another way – by “thematizing the neuter stem”. Thus the Neo-Hittites freed up the -ant/-anz suffix for use elsewhere. They went for marking the transitive subject, at least on neuters. That is: for split ergativity.
I don’t know whether Proto-Indo-European was ergative. I don’t even dare discuss the Indo-Hittite protolanguage. But Hittite, as of 1500 BC, was accusative. I suspect Hurrian as the force which pulled Hittite toward split ergativity.
On this site
Property of author; All Rights Reserved