Re: [tied] Re: Sanskrit and e, a, o

----- Original Message -----
From: michael_donne

To: cybalist@yahoogroups.com

Sent: Tuesday, March 26, 2002 3:46 AM

Subject: [tied] Re: Sanskrit and e, a, o

> Is there any reason it can't have gone from e > oi, ai, etc. Apart from the 1 > many direction -- that's a different issue.

It isn't a different issue but part of the reason why. You'd have to assume the following splits:

*e > ei, oi, ai

*o > eu, ou, au

*a > e, o, a

*a: > e:, o:, a:

-- all of them unconditioned (an unprecedented example of irregular split on such a scale) but with all sorts of mysterious coincidences, the same vowels appearing in different branches in the same lexico-morphological environments.

>>The vowel was monophthongised in some languages
>>(e.g. > Slavic *e^, Arm. e:
>>[both different from inherited *e] ...

>How are they different from *e?

>> but in such cases there is sufficient evidence of its derivation
from an older diphthong ...

>What is this evidence?

In Slavic, for example, old *oi and *ai became <e^>, a tense front vowel, originally with an open articulation (I suppose one could compare it to the American English vowel of "past"), while the reflex of *e was a lax mid front vowel (as in "pest"). Some of the instances of <e^> derive from pre-Slavic *e: (let's call them "monophthongal"), but the evidence that in the remaining cases <e^> stands for an original diphthong includes the following facts:

(1) It alternates with prevocalic -oj-, e.g. pojo~ : pe^ti, derived from the same root in different environments (*poj-V- : *poi-C-).

(2) "Diphthongal" <e^> (< *oi, ai) does not cause the so-called "first palatalisation" of a preceding *k or *g (into <c^, z^>), while a "monophthongal" <e^> (< *e:) does. This is because the first palatalisation took place before front vowels, and at the time when it operated the first element of the diphthongs *oi, *ai was not yet front. The "second palatalisation" of *k and *g before _new_ front <e^> took place much later, giving different results (namely, <c> [ts] and <3> [dz]).

(3) The RUKI rule in Slavic operates after "diphthongal" <e^>, but not after the "monophthongal" <e^>. This is because at the time when it operated the former was a diphthong with a palatal offglide (<..i>), thus satisfying the condition of the rule.

(4) Old loanwords with foreign *ai have Slavic <e^>, e.g. ce^sarI < Gmc. *kaisar- 'emperor' (Caesar).

(5) "Monophthongal" <e^> corresponds regularly to a front monophthong in most other branches of IE (and to an <a:> that palatalises a preceding velar in Sanskrit), whereas "diphthongal" <e^> corresponds to dipthongs (and to an <e> that does _not_ palatalise a preceding velar in Sanskrit, see (2) above).

> What alphabet are the Mitanni texts in? Cuneiform? Did it distinguish between e and ai?

It's a cuneiform syllabic script (not an alphabet), and yes, it did make the distinction. The spelling used for *aika- is <a-i-ka->.

> What would a back 'a' sound like as opposed to a front 'a'?

Check the IPA vowel diagram. The difference in quality is like that between standard British English "a" in <last> (or American English "o" in <lot>) and the "a" in "cat".

> Misra passed away some time ago.

Did he? I didn't know. When was that?

Piotr