Newsgroups: sci.lang,sci.lang.japan
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!udel!gatech!howland.reston.ans.net!ix.netcom.com!netcom.com!alderson
From: alderson@netcom.com (Richard M. Alderson III)
Subject: Re: Pitch Accent vs. Tone (was How did Korean lose the tones?)
In-Reply-To: koontz@cam.nist.gov's message of 12 Jan 1995 20:24:21 GMT
Message-ID: <aldersonD2DCBs.B48@netcom.com>
Reply-To: alderson@netcom.com
Fcc: /u52/alderson/postings
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <1995Jan6.215248.9102@galileo.physics.arizona.edu>
	<3es05f$hbf@pheidippides.axion.bt.co.uk>
	<1995Jan10.000227.7127@midway.uchicago.edu>
	<3f0m3a$m6d@pheidippides.axion.bt.co.uk> <3f435l$33j@ss1.cam.nist.gov>
Date: Sat, 14 Jan 1995 00:07:04 GMT
Lines: 106
Xref: glinda.oz.cs.cmu.edu sci.lang:34299 sci.lang.japan:21597

>In article <3f0m3a$m6d@pheidippides.axion.bt.co.uk>, donald@srd.bt.co.uk
>(Donald Fisk) writes:

>|> : Nope.  The *intonation* of Greek has certainly changed over time,
>|> : but Ancient Greek was never a "tone language" in the way that term
>|> : is currently used in linguistics.
>|> 
>|> What were the accents for then if not tones?

And in article <3f435l$33j@ss1.cam.nist.gov> koontz@cam.nist.gov 
(John E Koontz) writes:

[A long and learned discourse on pitch accent systems.  I excerpt for comment
two paragraphs.  --rma]

>In still more interesting systems you might have more than one contour, e.g.,
>L, H and HL words.  I believe that Ancient Greek falls into a class something
>like this, though I'm not sure of the details.

>So to return to the original question, what are the accent marks in Ancient
>Greek for if not to mark tone?  There are three:  grave (low stressed), acute
>(high stressed), and circumflex (falling, on long vowels only).  The location
>of the accent mark is restricted to the last two (?) syllables.  In essence,
>as far as I've ever been able to learn, the accents serve to code word pitch
>contours of L (grave) and HL (possibly also H) (acute and circumflex).  The
>circumflex is essentially a mark of a "hl" sequence on a long vowel, i.e., a
>long vowel in which the first mora, as opposed to the last, is accented.

In the following, I will use the simple noun "Greek" to signify the dialect of
Athens in the 6th and 5th centuries BCE.  This is the language taught at most
universities as "Ancient" or "Classical" Greek.

The Greek accent is fully lexical in nouns, pronouns, and adjectives; certain
verb forms require a particular placement of the accent, but for the most part
the accent in verbs is "recessive"--that is, it is placed as far from the end
of the word as possible given certain phonological constraints.

Some definitions are in order, to simplify the discussion.

	ultima		the final syllable of the word
	penult		the next to final syllable (penultimate "almost last")
	antepenult	the next to penultimate

Using these allows us not to worry about whether the antepenult is "second from
last" or "third from last" (like "this Thursday" vs. "next Thursday").

Only one syllable within the word bears the accent.

The accent may never fall further from the end of the word than the antepenult.
If it falls on the antepenult, it is *always* acute.

If the ultima contains a long vowel or a diphthong other than -oi or -ai, the
accent may fall only on the penult or ultima.  In this case, the accent on the
penult must be an acute; the ultima may bear either an acute or a circumflex.

If the ultima contains a short vowel or one of the diphthongs -oi or -ai, and
the penult is long and accented, the accent will be a circumflex.

An additional kicker is Vendrye`s' Law:  In a word with a final sequence of
short-long-short the accent will recede from the penult to the antepenult.  For
example, the word for "human being":

	nom. sg. a'nthrO:pos
	gen. sg. anthrO:'po:

This can be analyzed, like John Koontz's Siouan data, as a mora accent.  The
possible patterns are (using his notation, with additional x for don't care):

	Monosyllables:  h lh hl
	Disyllables:	x.h x.lh x.hl h.l h.ll hl.l
	Trisyllables:	x.x.h x.h.l h.l.l x.lh.ll h.ll.l

The last pattern in the trisyllables is that of a'nthrO:pos above, and is the
only one in which the accent is more than three morae from the end of the word.

Those long syllables in which the first mora is accented are the circumflexes.
This was described by the ancient Greek grammarians as "perispomenon" =
"bending around" with a rise and a fall in the same syllable.

A special note must be made of the monosyllables:  Unaccented monosyllables do
exist, but formed a phonological word with another constituent.  The masculine
and feminine nominative (singular and plural) of the definite article is
proclitic, that is, acts as part of the following word for accentual purposes;
most others are enclitic, forming a phonological word with the preceding word.

There are a few disyllabic enclitics, most notably the third single present of
"to be."

This leaves only the grave accent to be discussed.  This appears only on the
ultima when that is acute and another accented word follows.  The ancient
grammarians are not entirely in agreement on it, some describing it as a loss
of the acute, others as a lowering of the pitch of the acute.  I interpret this
as a difference between descriptions of the phonological and phonetic roles of
the grave:  It is as if no accent appears (phonologically) but a lowered place-
holder pitch is actually produced.

(I should also note that the ancient grammarians were unanimous in describing
the accent in the same terms with which music tones were described.)

Thus Greek clearly has a simple pitch accent, similar to Siouan (or, as I noted
in a previous post, Japanese).
-- 
Rich Alderson   You know the sort of thing that you can find in any dictionary
                of a strange language, and which so excites the amateur philo-
                logists, itching to derive one tongue from another that they
                know better: a word that is nearly the same in form and meaning
                as the corresponding word in English, or Latin, or Hebrew, or
                what not.
                                                --J. R. R. Tolkien,
alderson@netcom.com                               _The Notion Club Papers_
