Peter T. Daniels wrote on Sunday, April 17, 2005 5:47 PM

> Richard Wordingham wrote:
>> Peter T. Daniels wrote on Sunday, April 17, 2005 2:30 PM

>Maybe if you make it clear to your Thai audience that language and script
>are two very different things, you'll get your points across more easily.

> Thai is a good candidate for such a discussion because of all the
> historical letters that are hanging around not doing much of anything.

As I get to know the language better, I'm beginning to think they have their
uses, just as gender does in Latin. Tagging Pali/Sanskrit words as such
does help word division.

(This post is best viewed as UTF-8.)

>You could only be bringing in Hebrew and Aramaic if you're not
>distinguishing script and language. There's no reason to suppose /'/ and
>/`/ were lost in Hebrew and Aramaic until (in some varieties) recently, and
>several distinctions that couldn't be recorded with the Phoenician
>consonant inventory were preserved at least until the time of the
>Septuagint.

There's no reason to believe ayin had been lost before the text was pointed.
Many of the minor peculiarites of the lamedh guttural verbs may however be
late, as the taw of the endings is dagesh (plosive) rather than raphe
(fricative) despite the pathah between the guttural and the taw.
Unfortunately for me, I don;t think there's any reason to believe it was a
glottal stop.

The syllable final aleph had been lost by the time of the pointing, for it
does not receive shewa. Moreover, there was some confusion between lamedh
he (originally stems ending in /y/ and /w/, which has mostly been absorbed
by the vowels) and lamedh aleph verbs. The alpeh had been quiescent long
enough for the taw of verb endings to be raphe rather than dagesh. The same
goes for the raphe in _ma:la:kh_ 'messanger, angel'. Words like _ro:'sh_
'head' suggest that the quiescence had been going on for a long time (though
perhaps the original vowels isn't as in Arabic _ra's_), and the vocalism of
_ri:'sho:wn_ 'first' does not look recent. Does the Septaguint show any
evidence that final aleph as still pronounced? I know it shows that resh
could still be gemimate (e.g. _Sarra_ for 'Sarah'), but does it show any
sign of geminate gutturals?

>> >> vowels (but not between vowels) had been absorbed into such vowels,
>> >> lengthening them. That's how Latin 'a' derives from the letter for a
>> >> glottal stop, aleph.

>> > No, Greek alpha is /a/ instead of */'/ because Greek doesn't have /'/
>> (or the other consonant sounds represented by the Phoenician letters
>> that were turned into vowel letters).

>> How about digamma for /w/? Surely you haven't forgotten that 'f', 'u',
>> 'v'
>> and 'y' all derive from waw?

> Nothing to do with the transfer from Phoenician to Greek.

So where does 'f' come from? Where does digamma come from? Where does
upsilon come from?

>> What about the use of he for /e/ and heth for
>> /h/? Was that because Greek /h/ was [x] at the time and therefore closer
>> to
>> the sound of he than of heth? I admit I can't think of any argument that
>> /h/ wasn't [x] at the time. Just how arbitrary was the use of aleph, he
>> and
>> ayin for /a/, /e/ and /o/ rather than some other pemutation?

> There was no /h/ in the variety of Greek for which the alphabet was
> derived. That's why the two breathings were invented, centuries later.

> He > e, heth > long e.

Ultimately, yes. As my source are somewhat tertiary, I'll just quote from
Liddell & Scott:

'The old alphabet had only one sign (ε [epsilon]) for the sounds of _e_ both
long and short, till the long vowel η [eta], with ω [omega], was introduced
from the Samian into the Athenian alphabet in the archonship of Euclides
(B.C. 403) together with ω [omega], ξ [xi], ψ [psi]. The sign Η [eta],
before it represented long ε [epsilon], was used for the rough breathing, as
ΗΟΣ [eta, omicron, signa] for _hós_ [I gave up the struggle for polytonic
Greek]. When the same Η [eta] became a vowel, it was divided, so that [left
half of capital eta] represented the rough [right half of eta] the smooth
breathing, whence came the present signs for the breathings.'

Is this account untrue? If it's true, why was eta used for the rough
breathing? If it's untrue, where does 'h' come from?

> Meaning what? Pali and Sanskrit are two different languages that can be
> written with any of the scripts that developed from Brahmi. What do you
> want to compare with Thai (writing)?

This discussion was essentially about Pali and Sanskrit written in the Thai
script (traditional way, as an abugida, not the alphabetic way). Alone's
argument that nikkhahit might be considered a consonant is that
Pali/Sanskrit nikkhahit can be replaced by ngo ngu, at least before ho hip
(to which I can add the sibilants). Nikkhahit is relevant because it is
most conveniently lumped in with the tone marks as a vowel-like feature.
(Irritatingly, the Windows IME for Thai treats nikkhahit as a vowel, and
won't allow Sanskrit <si.mha> 'lion' to be typed in naturally - the Thai
Royal Institute Dictionary online was forced to enter it as <so suea, sare
ue, ho hip>. I'm pretty sure this is in accord with a Thai standard, so
they can't blame Unicode.) The pronunciation of a final nikkhahit does
depend on whether one is reading Pali or Sanskrit.

I think I've come across an interesting consequence of this pre-computer
convention that sara ue is sara i + nikkhahit. In Devanagari at least, the
Sanskrit stem _linga_ of 'lingam' can be written <liGga> or <li.mga>. In
Thai, the latter may be written <lo ling, sara ue, kho khwai>, while the
former is <lo ling, sara i, ngo ngu, kho khwai>. The commoner Thai
derivative is a hybrid of the naive spelling pronunciations - <lo ling, sara
ue, ngo ngu, kho khwai, thanthakhat>! Of course, the obvious derivative <lo
ling, sara i, nogo ngu, kho khwai, thanthakhat> (which is the rarer
derivative) is a homophone of the word for 'monkey'!

One reason for Pali and Sanskrit to be spoken of in the same breath is that
the spelling of Pali loan words in Thai is often Sanskritised - the 3-say
distinction ç v. .s v. s is often restored, and geminates are often
'restored' to double ro ruea plus consonant. Thus Pali _dhamma_ becomes
Thai <dhrrm> /tham/; Sanskrit _dharma_ would become Thai *<dhrrm\> /than/
(with a silenced final consonant). Hypersankritisation is not unknown, and
hybrid Pali-Sanskrit compounds compounded in P/S fashion exist.

Richard.