Re: How does Armenian fit into the Indo-European family?


"Darkstar" <darkstar100@xxxxxxxx> wrote...
John Atkinson wrote:
"Richard Wordingham" <jrw0602@xxxxxxxxxxx> wrote...
> On Mar 10, 7:37 pm, "Darkstar" <darkstar...@xxxxxxxx> wrote:

>> Peter T. Daniels wrote:
>> > On Mar 10, 7:46 am, "Darkstar" <darkstar...@xxxxxxxx> wrote:
>> > > Hi, everyone! Long time no see.
>>
>> > > How in the world is Armenian supposed to fit into the IE >> > > family?
>> > > Cf. /mek/ -*/oynos/, /yerku/ - */duwo/, /yerek/ -*/treyes/.
>> > > What's that stuff about comparing Armenian to the Hellenic >> > > group?
>> > > Are
>> > > there any other theories concerning its origin?
>>
>> > The change from *dva to _erk_ is perfectly regular.

erkow being the Classical Armenian equivalent of modern yerku.

>> In what way? I see no regularity here.
>
> I think you'll find it's *dw > er and *s > ku (or is it just /k/?),
> with the protoform being more like *dwo:s. A better fit is > obtained
> if you assumed that PIE has *sW, a labialised sibilant, a > hypothesis
> associated for purely historical reasons with the glottalic>
As I understand it, it's *d >(e) r and *w > k

> hypothesis.

It's certainly regular. Compare Classical Armenian erkar "long" <
dweh2ro-, erknc^'im "I fear" < *erki-nc^'inm <*dwi-n-sk-, erkun
"birth-pangs < *h1dwon-.

On its own, initial *w- becomes g-. The voiced stops, *b *d *g always
become voiceless, p t k. Presumably, -w- assimilated and became
voiceless too, hence > -k- instead of -g-. (cf skownd < *kywon-t-,
dog -- palatized stop ky > s because Armenian's a satem language.) The
rhotacisation of the initial stop before -k- is less explicable, but
hardly unknown in other languages. Initial e- is a prosthetic vowel.

John.

Hi, John.
Thanks. That's what I expected to get -- a more specific answer.

Anyway, it's not that I say they are irregular, I just mean the
reconstruction is too long and unnatural, therefore subject to
Ockham's razor. Generally, using prosthetic phonemes is a bad habit --
they're quite rare and are often the first sign of ad hoc
conjecture...

Tell that to the Basques!

There are lots of other languages that don't permit word-initial /r/

Intrinsically, it could reconstructed in the following way:
/erkou/ <-- */eskou/ (if rhotacism) <-- */THeskou/ (if there was a
lost light dental as in spoken Eng. /get'm/ or Spanish /ciu'a/ <-- /
ciudad/) <-- */THesHou/ (if /k/ or /kh/ is from aspiration). That
would be more like it.
Or, if no rhotacism, simply:
/erkou/ <-- */THerkou/ <-- */THerHou/

Consequently,
/erekh/ <-- */THerekh/ <-- */THereH/

To me, it's enough evidence to wonder about the origins of Armenian.
It might have branched off earlier from other IE languages, more or
less like Etruscan. But maybe I'm wrong, and there are better
reconstructions -- I don't know yet. Just thinkin'.

The following are the main universally-accepted shifts from PIE to Armenian:

(1) *s > 0 initially and between vowels. (Final -s > k', but only in plural suffixes, so this's probably a replacement, not a sound shift).

(2) Satem: PIE labiovelars > velars, palatals > fricatives or affricates

(3) *p > h or 0

(4) Quasi-Grimm's law: plain voiceless stops > voiceless aspirates, voiced stops > voiceless stops, voiced aspirates > voiced stops.

(5) Glides: *j > 0, *w > g (like in Celtic)

If you're going to play reconstruction, I reckon you should at least take these into account.

J.

.

Relevant Pages

  • Re: How does Armenian fit into the Indo-European family?
    ... What's that stuff about comparing Armenian to the Hellenic ... There are lots of other languages that don't permit word-initial /r/ ... plain voiceless stops> voiceless aspirates, ...
    (sci.lang)
  • Re: How does Armenian fit into the Indo-European family?
    ... What's that stuff about comparing Armenian to the Hellenic ... There are lots of other languages that don't permit word-initial /r/ ... plain voiceless stops> voiceless aspirates, ...
    (sci.lang)
  • Re: How does Armenian fit into the Indo-European family?
    ... Romance languages which have borrowed words starting with /w/ from ... another, since gw, if it existed, became g well before Armenian was ... It has nothing to do with phonemes. ... into a lot of other things in the mainstream IELs (except in Celtic, ...
    (sci.lang)
  • Re: Khoisan is very old
    ... neighbouring Bantu languages. ... adjacent to the click languages. ... Probably the social norms of personal interaction are as ... for instance, textbooks of Armenian. ...
    (sci.lang)
  • Re: How does Armenian fit into the Indo-European family?
    ... >>> Isn't it strange it's so different in Armenian? ... contact with German "confederates". ... > from the mainstream IE languages, or whether the latter share any ... It has nothing to do with phonemes. ...
    (sci.lang)
Quantcast