From: tgpedersen
Message: 33208
Date: 2004-06-11
> On Thu, 10 Jun 2004 09:27:32 +0000, tgpedersenthe
> <tgpedersen@...> wrote:
>
> >A long time ago I vented an idea here that perhaps the vowels of
> >Greek alphabet were not reinterpretations of the signs forPhoenician
> >laryngeals, as usually assumed, but that Greek at the time stillhad
> >the PIE laryngeals, and that the signs for laryngeals were onlythat
> >reinterpreted as vowels later, as Greek lost the laryngeals.
> >
> >Jens recently, and (I think) Miguel then, sensibly pointed out
> >the Greek letters for vowels were also used for the e/o reflexesof
> >the ablaut vowel.omikron/omega)
> >
> >But then I got to think of the fact that Greek has two vowels for
> >each of /e/ and /o/. Yes, they are used to distinguish length, but
> >were they always? The letters themselves (epsilon/eta,
> >don't seem to be related.Yes, that's very detailed, but a detailed conjecture is still a
>
> You're acting as if the origin of the Greek letters were
> something ill-understood, when actually the facts are known
> very precisely.
>
> Greek alpha /a/ comes from Phoenician *?alf (*?alpH). Since
> Greek did not have a glottal stop, the letter name (meaning
> "ox") was taken as acrophonic for Greek /a/ (and /a:/).
>
> Greek (at least most dialects) *did* have a phoneme /h/, and
> to write it, the Greeks could choose between Phoenician *he:
> (laryngeal /h/) and *h.E:t (pharyngeal /h./). They opted
> for the latter, which left */he:/ "hey!" free to stand for
> Greek /e/, /E:/ or /e:/.
>
> For the vowels /i(:)/ and /u(:)/, Phoenician offered the
> semivowel letters *yo:d "hand" /j/ and *waw "hook" /w/.
> Since Greek had no phoneme /j/, the leter iO:ta was
> naturally adapted as /i/. Greek (at the time) *did* have a
> phoneme /w/, so the Phoenician letter *waw was split
> graphically into two variants, Y /u/ and >|, |< /w/ (the
> latter becoming F under the influence of preceding E). The
> new letter /u/ was added at the end of the alphabet, after
> T. Note that when the Greek alphabet was adopted by the
> Phrygians (who *did* have a phoneme /j/), the Phrygians used
> the same trick, adopting straight iO:ta I for /i/, while
> using the other Greek variant ("crooked" iO:ta, S) for the
> phoneme /j/.
>
> That only left /o/ (/O:/, /o:/) unaccounted for, and the
> Greeks rather arbitrarily adopted the remaining Phoenician
> letter that to their ears started with a vowel (*¿E:n "eye")
> to represent it (it may have helped that "eye" is Greek
> ophthalmos). Note that Iberian (which didn't have /h/)
> adopted Phoen. */h.E:t/ (H) for /o/.
>
> One variety of Greek that did not have the phoneme /h/ was
> Ionian. There, the Greek letter /hE:ta/ H was pronounced
> /E:ta/, and, again by the acrophonic principle, this
> resulted in H being used to write the long vowel /E:/ (which
> in Ionian continues both pre-Greek */e:/ and */a:/). This
> prompted the subsequent invention of a corresponding vowel
> sign for /O:/, which is merely a graphical variation on O
> (perhaps originally two circles on top of each other). The
> letter /O:/ was added at the end of the alphabet (after the
> other letters invented by the Greeks, those being (in Ionia)
> u: phe: khe: pse:).
>
> In the other (non-Ionian) epichoric alphabets, H remained in
> use as /h/, and was so transmitted to the Etruscans (and
> then the Romans), and the /O:/ sign was unknown, until the
> official adoption of the Ionian alphabet by most Greek city
> states in the 4th. century BC.
>