Re: Stress placement systems

H. S. Teoh wrote:
> On Wed, Sep 20, 2006 at 07:55:04PM +0100, R A Brown wrote:
[snip]
>>It appears in fact to have been similar to the pitch system of Vedic
>>Sanskrit. But alas the Greeks were less clear in their descriptions
>>than the ancient Indian phoneticians were.
>
>
> That's very interesting. Perhaps they are relics of a pitch accent
> system in PIE?
That would seem to suggest it. But, alas, I have not kept up with the
latest thinking on PIE. I note that 'Stress System Database' gives:
{quote}
Indo-European (protolanguage) 	12..89/1L 		Halle & Vergnaud 1987:72
Syllable "weight" determined by lexical accent
{unquote}

I note "weight" is put between quotes which suggests to me that it has
nothing to do with syllabic quantity. But there is no reference to a
pitch accent.

Maybe others on the list are more informed regarding PIE word accent.

[snip]
>>
>>I don't know of the top of my head, but they were originally
>>introduced by the Alexandrian grammarians of the 2nd cent BCE as aids
>>to pronounce Homer correctly (Epic dialect);
>
> How much does Homer differ from Attic in this respect (accents)?
Much the same - which, I guess, is why conventionally texts in other
dialects, e.g. Doric, are normally printed with the same accentuation.
It is assumed that anciently all dialects had much the same pitch accent
  system.

[snip]
>
>>As for Aeolic, we are told by IIRC more than one authority that they
>>treated all words according to the 'recessive accent' system of Greek
>>verbs.
>
> Cool. I wonder what caused the differentiation.
Simplification, I suspect.

Besides verbs, there were also some sets of nouns & adjectives that used
the recessive accent, and many vocatives did also. I guess the Aeolians
simply extended this to all words, thus dispensing with the apparently
arbitrary lexical positionings of the pitch.

We should perhaps explain for those who know little or no ancient Greek
that the term 'recessive accent' referred to a system whereby the pitch
accent a word was placed as far back from the end as the final _vowel_
(*not* syllable) allowed, i.e. if final vowel was short, the accent was
third from the end, and if it was long the accent was on the preceding
vowel. Diphthongs were normally treated as long, but _final_ -ai and -oi
were treated as short.

The recessive accent was, therefore, phonologically determined. Of
course when the system of long and short vowels broke down during the
Hellenistic period, the verb accent of Koine remained in the same place,
but it was no longer phonologically conditioned and we have a
morphologically conditioned accent.

That stress accent of modern Greek is lexically & morphologically
determined (and not determined by syllabic weight).

>
>>>Also, how confident are we that Koine continues to use pitch
>>>accents?
>>
>>I am sure the Greeks themselves were still using pitch accent and I
>>suspect the more discerning L2 speakers tried to do things properly -
>>but the change took place during this period.
>
> Ahhh, I see. So the foreigners substituted stress for pitch, and
> eventually the natives also started adopting it. Hmm, this gives me
> con-world ideas. :-)
Sort of - but I suspect there was a more complicated inter-reaction
between L1 and L2 speakers over the whole period. But I feel sure the
internationalization of Greek was a main factor in bringing about this
change.

[snip]
>>Graffiti at Pompeii where some Greek words are written in Roman script
>>show quite clearly that fricativization had already occurred in
>>colloquial speech by the middle of the 1st cent CE.
>
> Wow, this is interesting. I hadn't expected it to have happened this
> early. I remember reading a hypothesis that perhaps it had begun even in
> Classical times: is this plausible, or do we know for sure that it came
> later?
There are hints that it may have begun in some dialects as early as
Classical times. For example:
- the Athenians mocked the Spartans because they said things like 'sios'
instead of 'theos' "god". That is, the Lakonian dialect had _s_ where
Attic and others has _th_. I has been speculated that this Lakonian _s_
was in fact [T].
- there is some evidence that in Boeotian, Elean & Pamphylian the voiced
plosives had developed a fricative pronunciation by the 4th cent BCE.
(e.g. using vau ('digamma') where we would expect beta).

[snip]

> Now, something I've always wondered about is why [p^h] fricativizes to
> [f] rather than [P]. Or perhaps [P] was an intermediate form that
> eventually became [f]?
It may well have been, we have no way of telling.

>
>
>>As for stress accent - we do not have clear evidence that stress
>>accent was the norm until the 4th cent CE; but there are indications
>>as early as the 2nd cent CE that transition from pitch to stress had
>>begun.
>
>
> I see. So basically it started around Hellenistic times, or going into
> the Byzantine period.
Yep - by the Byzantine period the stress system was the norm as far as
we can tell.

[snip]
>>
>>But the Erasmian pronunciation, though considerably closer, to the
>>ancient, is only a makeshift. Whether one uses Erasmian or modern
>>pronunciation, it is normal to use the modern stress system.
>
>
> That's true. How confident are we about the actual values of vowels in
> Classical Greek, though? (I understand the reconstructed values are not
> quite the same as the Erasmic?)
Correct - as I understand it, Erasmus's reconstruction was closer to
Koine than Classical Attic. We cannot, of course, be 100% confident
about any ancient reconstruction but the evidence for the approximate
vowel & diphthong values of Attic is fairly strong. I would suggest
getting hold of Sidney Allen's "Vox Graeca" (ISBN 0521040213 [hardback],
0521335558 [paperback]).

> Now, during my Koine course, I was told that there is evidence even in
> the New Testament itself that itacization was starting to happen (e.g.
> with some words misspelling iota for eta).
Well, |ei| (epsilon-iota) had become the same as 'long iota' [i:] by the
3rd century BCE. The process had clearly begun earlier as we see
confusions in spelling between |ei| ~ |i| in the late 4th cent BCE.
AFAIK confusion in spelling between eta & iota don't occur until about
150 CE,

> I'm curious about the extent
> of this, and whether the loss of the optative has something to do with
> the collapse of /oi/ to [i].
/oi/ became /y/ first. Confusion between |oi| and |y| (upsilon) is
certainly found in the 3rd cent CE; there is evidence that the change of
/oi/ to /y/ had already occurred in Boeotian as early as the 3rd cent BCE.

But AFAIK there is no evidence of confusion of /oi/ and /i:/ in Koine
until the unroundng of /y/ in the Byzantine period, probably around the
end of the millennium CE.

I suspect restricted use of the optative in Koine (it is used in the New
Testament) was part of process of simplification. in fact in Classical
language:
- it could be used instead of indicatives in subordinate clauses if the
main verb was past. writers like Xenophon did this, but others like
Plato & Thucydides did not and simply used the indicative.
- it could be used instead of the subjunctive in subordinate clauses if
the main verb was past. Writers like Thucydides ignored this.
- it was used in remote or unlikely future conditions (If I were to
become president of the USA, I would ... etc  ).
- to express wishes relating to the future (Oh that he would come!)

Quite simply the optional uses of optative in subordinate clauses after
main verbs in past tense was simply dropped. Why keep a complicated
option when a simpler choice is available?

The use of the optative in conditions is rare in Koine - the use of
_ean_ plus subjunctive for the protasis & simple future for apodosis
suffices for any future condition. It still holds on to express wishes.
But such a limited use was almost certain to lead to the dropping of the
mood and other ways of expressing wishes being developed.

> Also, I understand that the iota subscript wasn't actually used until
> Byzantine times, correct?
Correct - it was a Byzantine practice.

> In uncial writing it was written as long vowel
> + iota --- my question is, was it actually pronounced in Koine, or had
> it become irrelevant already? What about in Classical times?
Certainly pronounced in Classical times. But confusion in spelling
between eta-iota and epsilon-iota occurs as early as the 4th cent BCE;
it seems that with long-alpha+iota and omega-iota the iota had become
silent by the 2nd cent BCE.

[snip]

> How well-described is the Vedic system?
I'm not certain, but Panini IIRC does give rather clear descriptions
that is ever found by the Alexandrian grammarians.

> Recently, through learning
> Russian, I'm beginning to appreciate just how much nuance there is in
> actual speech, that is never recorded in the textbooks. If one went
> merely by textbook descriptions and second-hand information, one would
> pronounce Russian very strangely indeed! And were it not for actual,
> living, native speakers, one could be easily deluded to think that one's
> pronunciation is quite accurate.
:-)

> So now I'm wondering about just how
> much information has been preserved about these ancient tongues, or how
> wildly inaccurate our attempted reconstructions are. :-)
Exactly!

Alas the neither the Greeks nor the Romans were the world's best
phoneticians. In the case of Latin we do have great deal more
information (partly because of the different way that the Romancelangs
have treated certain sounds), but with Greek it would be IMO a very
foolish person who would claim complete accuracy. I am quite sure that
our attempted reconstructions would indeed sound very strange to Greeks,
whether in the Classical period or the Koine (even tho, of course, in
the Koine there was more regional variation   :)

--
Ray
==================================
ray@carolandray.plus.com
http://www.carolandray.plus.com
==================================
Nid rhy hen neb i ddysgu.
There's none too old to learn.
[WELSH PROVERB}