Re: Stress placement systems

Rob Haden wrote:
> On Tue, 19 Sep 2006 19:34:43 +0100, R A Brown <ray@...>
> wrote:
>
>
>>On Classical Greek we read the primary *stress* is: "12/2R". I
>>understand this to mean "on last syllable if heavy, else on next to last
>>if heavy, else next to last".
>>
>>To put it politely, this is rubbish - because:
>>1. Ancient Greek did not, as far as we know, have word stress; there
>>possibly was phrasal stress, but we can merely guess how that might have
>>worked.
>
>
> Quite right.
Thanks    :)

> In fact, I would argue that Ancient Greek lacked even
> lexical pitch -- i.e. the pitch was phrasal in nature.  My main piece of
> evidence for this is the use of the grave accent.  It indicates that,
> where a high pitch would be pronounced in isolation, it is not pronounced
> in the given phrase.
The recessive accent on verbs, neuter nouns, exocentric compounds 7 one
or two other categories can be defined phonologically. But on other
groups, where the accent can be so defined, whether the word is
proparoxytone, paroxytone, properisomenon or perispomenon must surely be
a matter of lexis. It seems odd to me if _in these groups_ oxytones are
then not lexical but phrasal; these words do have final stress in modern
Greek.

> As a result, it's no surprise that the grave is
> typically used for prepositions, pronouns, adjectives, and genitive
> nouns.  Basically, these kinds of words tended to be treated as clitics,
> at least on the prosodic level.  I think this echoes the situation of
> latest PIE.
Do we, in fact, know what a final grave means? I have heard/read it
suggested that it indicated that the vowel was not raised as much as one
would expect.

I think the case for prepositions is certainly strong (the enclitic
pronouns are in any case enclitic - accents occur only on the
non-enclitic forms). Presumably the graves on the definite article would
be regarded in the same way - it makes sense, though the final
circumflexes, where they occur on pronouns & the definite article cannot
be considered this way.

Certainly the situation is phrasal in that enclitics affect the pitch
accentuation of the whole phrase - but should enclitics be considered as
separate words?

The final grave business is certainly an interesting one. If only we had
time travel ........   ;)


BTW as far as I can understand it, Hayes (1995) got his info on ancient
Greek stress from Sauzet (1989) & Golston (1990). It seems that in their
approach:
The accent consists of a HL _pair_, the H being the high pitch (marked
by the familiar accents of Greek texts) and L being the following low
tone; and it is, according to them, the L part of the pair that attracts
stress.
As the the recessive accent is concerned, according to their approach:
- Final consonants [not consonant clusters, but just the final consonant
itself] are extrametrical;
- A moraic trochee is constructed at the final three morae of a word;
- thus if the final syllable is light, the rise in pitch is the
antepenultimate syllable & the stress on the penultimate; if, on the
other hand, the final syllable is heavy, then the stress is on the last
syllable & the rise tone on the penultimate.

Even if this analysis were true (and I do not think it is), the 'Stress
System Database' would still be faulty in that it still does not account
for the *very* large number of words that do not have recessive accent.

However, I do not know why Sauzet & Golston think that stressed syllable
is the one _following_ the syllable with the (written) pitch accent. I
guess it is something to do with a supposed analysis of Greek meter. I
find it, however, very difficult to accept their analysis. If there was
this word stress in ancient Greek, why the heck did that not prevail as
pitch accent died out in the late Hellenistic period?

The modern Greek stress system is clearly derived from the ancient pitch
accent system!!!!!

I do not understand Sauzet & Golston's approach (assuming that I've
basically got it right above), not why the 'Stress System Database' puts
such heavy reliance on Hayes after the warning Hayes himself gave about
checking the sources.

--
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}