« A bit of Tangut | Main | Wolves, wolps, and lupes »

01/09/2008

Comments

Glen Gordon

[this is good]

Genitives are an interesting world to delve into. So much to say, but I also need sleep. So I'll be as brief as possible.


You divide an ending *-esio up into thematic *-e-, **-si- and **-o, but may I suggest instead: genitive *-(e/o)s plus relative pronominal stem *yo-. I reckon that this ending, added in pre-IE, would have helped to disambiguate the nominative from the genitive in thematic stems nicely.


Then on to the next topic: "morphological redetermination". Etruscologists already stumbled over this a while ago. Example: Uni, the name of a goddess equivalent to Greek Hera or Roman Juno, receives the genitive -al to become Unial "of Uni". However, Unial in turn can mean "(temple) of Uni" and hence with the addition of the inessive postclitic -thi, we get Unial-thi meaning "in (the temple) of Uni". Strange as this all seems, this is amply attested as a quirky feature of Etruscan morphology. Chalk it up to yet another Aegeo-Anatolian areal feature, I say.


As for genitives in *-si, I don't remember seeing that without the terminating vowel. Either I'm delirious from lack of sleep, or possibly have you confused this ending with the issue of those other case endings formed with locative particle *bhi (*-bhi/*-bhyoh1)? This is what Plank lists for genitive variants in Indo-European (Plank, Double Case: Agreement by Suffixaufnahme (1995), p.244): click here. He attributes Greek -ou to *-osyo not *-eso or *-oso (note the Mycenaean example), however Gothic -is is assigned to *-eso.

PhoeniX

I was doing my Iliad resit today (which without a doubt, I failed) but what I did realise was the Greek -ou, or rather the lack there of. Throughout the Iliad you almost exclusively see -oio, which indeed safely goes back to *osio.

It's interesting that Attic lost this intervocalic iota though, that's not common practice.

as for genitives in *-si, I don't remember seeing that without the terminating vowel.

I'm not really sure what you mean. Anatolian clearly has this -ssi- suffix, which is generally only followed by consonants.

Or are you referring to:
I am quite sure there's other Indo-European languages that have
evidence for a -si- genitive like Sanskrit does, maybe Slavic
languages? I'm not sure, but if you know, please leave a message.


Because then, I meant -si- with  a terminating vowel. Since Sanskrit has that too .

Splitting up the genitive as *os-io is probably more plausible. Interesting to once again see Greek and Indo-Aryan in the same boat. This further strengthens my 'Graeco-Aryan' hypothesis.

It's just the sudden realisation of a possible Indo-European correspondence to the Anatolian genitival suffix -ssi-  which got me really excited. If the *o-si-o is indeed wrong, it makes me wonder where the genitival Anatolian suffix -ssi- did come from.

Glen Gordon

Phoenix: "Throughout the Iliad you almost exclusively see -oio, which indeed safely goes back to *osio."

Hooray! I'm a geek :P

 

Phoenix: "I'm not really sure what you mean. Anatolian clearly has this -ssi- suffix, which is generally only followed by consonants."

I'm not talking about the quality of the vowel itself. The Anatolian languages are reflecting *-i- in their genitives and it's common to attribute this to an adjectival origin instead of directly from PIE *-ós

I'm however refering to your etymological claim mentioned above: "An *e thematic vowel, *-si- suffix, and an *-o ending." As far as I know, there is no such morpheme **-si- used as a genitive in Common IE, nor is there any good motivation for the **-o ending. The prototypical genitive in PIE remains *-ós. It is secondarily reduced to *-s in some athematic paradigms or elaborated upon with *-yo, which I deduce is an archaic unmarked oblique case form of the relative pronoun stem *yo- introduced after Syncope in Late IE (although I'm pretty sure most linguists don't analyse it quite this way... yet :P).

Hence I think we should divide the genitive as *-(o)s-yo instead. The form *-(o)so is just an eroded form of earlier *-(o)syo and unless I'm missing something, I would hesitate to attribute *-(o)so to PIE itself. So while your etymology assumes unattested morphemes **-si- and **-o, I can explain it with attested morphemes *-(o)s and *yo-. Semantically, *X-s-yo would in effect mean "of/with that of X". I believe this was a necessary circumlocution to avoid nominative-genitive case merger in thematic paradigms if this stem weren't agglutinated. (Right before the fragmentation of PIE, it seems to me that PIE quickly became an increasingly synthetic language. So IE speakers seem to have began to shy away from simple endingless oblique forms like Late IE *yo, in favour of more complex ones, using tactics such as agglutinating a declinable stem to receive case endings like intervening *-sm̥- < *sem- "one". But I digress.)

 

Phoenix: "This further strengthens my 'Graeco-Aryan' hypothesis."

Actually, I hate to burst your bubble, but it's standard theory that Hellenic and Indo-Iranian dialects were indeed side-by-side in the Proto-IE linguistic area. The "centum-satem" isogloss line divides them.

 

Phoenix: "If the *o-si-o is indeed wrong, it makes me wonder where the genitival Anatolian suffix -ssi- did come from."

Have you read this pdf by Melchert entitled Genitive case and Possessive Adjective in Anatolian?

PhoeniX

Actually, I hate to burst your bubble, but it's standard theory that Hellenic and Indo-Iranian dialects were indeed side-by-side in the Proto-IE linguistic area. The "centum-satem" isogloss line divides them.

There's not much bursting of bubbles going on there. I often assume that anything I assume has already been assumed by someone in the past. :D

But often centum-satem is seen as a very strict line dividing the two dialectal regions to almost have no influence any more. And I think that, looking at Greek and Indo-Aryan it is clear that that is absolutely not the case. The verbal system is just too similar.

I'm currently thinking that a great part of our reconstructions  of the current Indo-European verbal system is far more exclusive to this graeco-aryan dialect than 'real'  PIE. The verbal systems displayed by both Gothic and Tocharian are so different from Greek and Indo-Aryan, it's difficult to assume, and no reason to do so that Gothic and Tocharian just threw the rather complex stuff right out of the window (And for Tocharian, assuming a new system infinitely more insane than any of the other systems that developed). I touched upon this in my post about reduplication, and if I find more I'll discuss it too ;-)

Have you read this pdf by Melchert entitled Genitive case and Possessive Adjective in Anatolian?

I hadn't! Thanks.
Glen Gordon

Phoenix: "But often centum-satem is seen as a very strict line dividing the two dialectal regions to almost have no influence any more."

True. I blame lingering PIE tree diagrams for that, the ones published from yesteryear that are used and reused in new books. They corrupt us modern neophytes into resurrecting the "centum-satem split" myth until we're finally talked down from the clouds with a little wave theory. But there are also lots of discussions on "centumized" words wandering into satem dialects like Balto-Slavic and vice versa. (Read Fortson, Indo-European Language and Culture: An Introduction (2002), p.365.) We should expect wanderwords happening between both sides of the isogloss line and it would be rather odd if this didn't happen. The big problem that IEists have is actually proving that specific words really are examples of loans between neighbouring dialects of PIE rather than products of internal processes. So breathe easy. What you're saying isn't risqué at all. Be strong, be proud, say it loud! :)

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)