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

> Piotr can give you better references. I've read "A Vedic Grammar
> for Students" by Arthur A. Macdonell and "The Sanskrit Language"
> by T. Burrow.

I own these two. The problem is that they are based on historical
linguistics not the authentic vedic texts so they can't really help
me establish the actual pronunciation since it's historical
linguistics that I'm questioning. That would be circular. I guess
I'll end up reading the pratishakhyas to see if it is possible to
really know how they were pronounced.

> > > David: Sanskrit has only long 'e' and 'o', and they're clearly
> > > phonetic realizations of the diphthongs 'ai' and 'au' (or 'ay'
> > > or 'av') in a closed syllable, replaced by 'ay' and 'av' in

open

> > > syllables.

The modern Indian pronunciation of 'e' is not as long as the
'ai' in 'bait' although most Westerners pronounce it that way.
Similarly,
'o' is not as long as in 'boat'. Both of these are somewhat in between
'bot' and 'boat' and 'bet' and 'bait'. In the same way, the
Sanskrit 'ai'
sounds to most Westerners like 'e' as in "Jai Ram > Je Ram" although
they are slight dipthongs.

> Greek has three phonemically independent short vowels 'a', 'e'
> and 'o', and three corresponding long vowels.

Are the corresponding long vowels also written like 'a', 'e' and 'o'
but pronounced longer?

So my point is: if Greek had a short 'e' and a long 'e' and
Sanskrit's were in between, then can you really say with certainty
that they weren't more like the short form?

> If by "(mis)pronounce 'e' and 'o' as long syllables" you mean
> that they pronounce them as diphthongs, with the 'e' like the
> diphthong in English 'they' [ey], and the 'o' like the one in
> English 'know' [ow], then you're right. They should be
> pronounced as pure vowels, [e] and [o],

OK, great. I think we're on the same page here, at least.

> but that is not the
> same as length and 'e' and 'o' were definitely long vowels in
> Vedic times. We can be sure of this because the Vedas
> were written in poetic meters, each based on a specific
> arrangement of long and short syllables.

The point about the meters is very telling. The Greeks had short and
long versions of 'e' and 'o' right? How were they pronounced? Do we
have an examples of them being used metrically like the Sanskrit?

Is the reconstruction of PIE e,a,o necessarily connected to the
meter? These could be two separate issues: length vis a vis PIE
reconstruction vs. length as used in chanting....? I'll have to think
about this.

> An open syllable ends in a vowel and a closed one in a
> consonant.

OK, that's what I thought you meant. The Sanskrit terms are 'laghu'
light and 'guru' heavy. Unless they also differentiate them in other
ways. I've studied Sanskrit but not the writings of the original
grammarians.

> To further clarify
> on length, closed syllables are always long, and open
> syllables are also long if they contain a long vowel.

> na-ya-ti (CV-CV-CV) 'he leads, guides'
> *nay-tram (CVC-CCVC) > netram 'eye'
> *nay-tr- (CVC-CV) > netr 'leader'

Hmm, this is interesting. I never thought about it that way: that
guru syllables are heavy (closed) because their vowels deconstruct
into a consonant conjunct. I wonder if the Sanskrit grammarians saw
it that way.

> If you have a Sanskrit primer it will be covered under "Sandhi".

Oh, I see what you mean now. I thought you meant something else.
You're talking about this:

> na-ya-ti (CV-CV-CV) 'he leads, guides'
> *nay-tram (CVC-CCVC) > netram 'eye'

My problem with this, is that I think it's backwards. The root 'ni'
which means 'to lead' is lengthened to become 'netram' or 'eye'. *nay-
tram is an analysis after the fact. Assuming that the whole idea of
it being guru because it analyzes into a consonant conjunct is even
valid to begin with. Cool though.

> That's a really neat chart! I'm glad you pointed it out.

Yeah, its very neat. I just wish they had more intermediate positions.

> I've been having more fun with it than a child with one of those
> "the cow goes mooo" toys.

LOL!

Um, why is this distinction between long and short vowels relevant to
the discussion about how PIE merged e, a, o into Sanskrit 'a' ?
I was going to post more about this but I think I'll wait until I
understand better why it matters. :-)

I gather that maybe the long versions of e,a,o didn't collapse into
short 'a' in Sanskrit but behaved differently somehow?

> > MD: This is my main issue: I can accept this kind of "only one
> vowel"
> > thinking in a theoretical sense but to apply it to a language

that

> > clearly demonstrates a, e, o in massive amounts of historical
> > documents in order to try to reconstruct something that is

supposed

> > to have REALLY HAPPENED in a real time and place, strikes me as
> > absurd. And to make it one of the pillars of the entire PIE
> > reconstruction seems very shaky to me. Theory and practice need

to

> be kept distinct.
>
> Well, actually we are talking about what really happened.
> 'ay' and 'e' really are part of the same phonemic cluster.

I can understand that.

> The problem I think is that the concept of a phoneme must
> really be well understood to fully understand why Sk. 'e'
> and 'o' aren't parallel to Gk. 'e' and 'o', and unfortunately
> I've always had a hard time explaining it well, especially if
> I can't use my hands. (-:

Yes, this is the crux of the matter. Could you upload to the library
a video of yourself explaining it with all the hand-waving?

> Anyway I've listed some Greek and
> Sanskrit correspondences below. Maybe they will help.

Yes, these are excellent!

> Gk. Sk.
>
> a = a
> e a
> o a
> a: a:
> e: a:
> o: a:

OK, so the long versions of the Greek collapse to the long
Sanskrit 'a:'

> ai e:/ay
> ei e:/ay

And their dipthongs collapse to corresponding Sanskrit dipthongs
(ai, au) or long vowels (e, o); is this correct?

I guess the central point of this whole argument is when Hock
says: "the correspondence of Sanskrit a- vowels to Latin and Greek e-
, a-, and o- vowels cannot be explained by means of a regular sound
change that "split" the 'a' of Sanskrit into the e, a, o of the other
languages."

My question is: Why not? Apart from the fact that it's really messy.
3 > 1 is certainly preferable in a theoretical sense than 1 > 3.
Are there any other reasons? Are there any environment where this
does not happen or where it is explainable in other ways?

Do you know of any really long lists of where this happens plus any
exceptions?

> But what relevance does the modern pronunciation have? Vowel length
> is not distinguished in the modern pronunciation of Latin either.

In most languages this would not be relevant. But Sanskrit
pronunciation is supposedly very conservative and most of the pandits
I've heard pronounce 'e' as a pure vowel and 'ai' as a very slight
dipthong. The distinction between short/long may only refer to
prosody and have no relevance to the vowel changes.

> In
> Vedic and Classical Sanskrit, <e> and <o> always represented a long
> vowel (IPA /e:/ and /o:/). In Prakrits such as Pali, <e> and <o>

were

> long in open syllables, short in closed syllables. In modern Hindi,
> <e> and <o> usually represent short closed [close-mid] /e/ and /o/,
> respectively.

Thanks, this is valuable information. Do you have any documentation
about the pronunciation of the Prakrits? Or is this just a
reconstruction?

> You're confused: <ai> and <au> are different sounds altogether.

Me: No, I'm not confused. You're confused.
You: No, you're confused!
Me: Look, you're really confused!!

Actually, I wasn't confused: my point about the .wav sound
for 'maila' was how short it was -- and I didn't have many examples
to choose from so 'maila' was all I had. If a supposed dipthong is
pronounced practically as a pure vowel, then Sanskrit 'e' and 'o' may
have lain between Greek long and short. Actually, now that I think
about it, David's posting of the entire range of sound change may
have clarified this whole long/short thing for me. Hock's article
only posted the short e, a, o.

OK, so now the question has changed: what about Sankrit vowels: i and
u? What happens with them in other IE languages like Greek?

> After */ai/ and */au/ had gone to /e:/
> and /o:/ (written <e> and <o>)

This brings up a related question on symbology I've had reading some
posts here: what do the / / and the < > signs represent? As in: /e/
and <e>.

> The analysis of the ancient Sanskrit grammarians differs from the
> modern analysis in several respects. In traditional grammars
> (ultimately based on Pa:n.ini), the citation form of roots (the
> "normal" form, one might say), normally corresponds to what in

modern

> usage would be the reduced form (zero grade, Nullstufe). The normal
> grade of modern usage is what the ancient Indian grammarians called
> "gun.a". The "vr.ddhi" grade of traditional Indian granmmar usually
> corresponds with modern "lengthened grade" (Dehnstufe, vrddhi), but
> sometimes (due to Brugmann's Law) with modern "o-grade".

Thanks for clarifying that. I figured it was something like that.
See, my problem is I'm questioning the very fundamentals of
historical linguistics and to do so I'm trying to get an idea of what
the original states and thinking about the languages were. But I find
that all discussions about this are heavily filtered through 19th
century linguistic concepts. They *might* be totally valid. Since I
doubt I'm another Saussure they probably are. :-) But I want to prove
it myself because these preconceptions might be hiding something
that happened 4000 years ago. So when I hear things like "traditional
view[s] ...failure to recognise gun.a as the normal grade", I have to
wonder: is this a failure or is it recognition of some pattern that
linguists have missed in the intervening 2000 years?

Normal > guna > vriddhi makes sense to me. What's wrong with it?
Isn't it relative? And perhaps it even varies with language?

--

--- In cybalist@..., "P&G" <petegray@...> wrote:
> You are confusing modern Indian pronunciation and ancient Sanskrit,
Michael.
> There have been changes in pronunciation over time, as with any
language.

Actually, I'm trying to use modern Indian pronunciation to
illustrate my points since I don't have any links to recordings of
ancient Sanskrit. I'm not sure how it has changed but I'm aware of
the issue.

> In pre-Sanskrit the increasing length gave the patterns:
> i ai a:i
> u au a:u
> This became in Sanskrit:
> i e ai
> u o au

Yes, but what is the evidence for this? The oldest attested languages
are Sanskrit and Greek. (We'll leave out Hittite for now since it's
controversial. Although it would be interesting to see what its
vowels do. What does Iranian do?)

Why do we say the Greek form is older?

> By this time we are way outside anything that can be
called "Sanskrit", but
> this new pronunciation affects the way Indians read Sanskrit.

It might. It probably does. But what is the evidence for it?

> Colin Masica discusses modern vowel systems on pages106 - 121 of
his book
> "The Indo-Aryan Languages". I quote from page 110-111:

Aarrgghh! That's two references to Masica. I thought I could get by
with only Beekes, Hock and Anttila. Looks like I'll have to get
Masica too.

> (2) the only source of the Sanksrit "e" is the combination a + i,
and the
> only source of "o" is a + u.

But its not the only source! Guna is a more common occurance than
sandhi, I think. Sandhi is irrelevant in this situation because it
would not be reflected in the word changes between Greek and Sanskrit
but a word created by Guna/Vriddhi acting on a root is directly
relevant.

--
--- In cybalist@..., "Piotr Gasiorowski" <gpiotr@...> wrote:
> Two languages only??? Who says the IEists base their
reconstructions on Greek and Sanskrit, and nothing else?

Well, that was rhetorical since Greek was the only example I had. But
I'd be interested in seeing evidence from other languages as well.

>It so happens that the abstract analysis of Sanskrit morphemes (or
internal reconstruction within Sanskrit, which amounts to the same
thing) agrees wery well with the comparative data. For example,<

Thanks for the examples. But what is to say it didn't go the other
direction?

>Diphthong-smoothing of this kind is common cross-linguistically.<

Are there examples of it going the other way?

For example, Old French /ai/, /au/ became /E, e/, /O, o/ in
historical times. The English vowel of <law, cause, caught> also
derives from the ME diphthong /au/.


> *keu, *geu, *gHeu > *cau, *jau, *jHau > co, jo, ho

Here's one of my fundamental problems: where are the attested
languages on the left side of the equation? Shouldn't we start from
what we know or at least start from what we know, work backwards then
forwards again to what we know?

> There's much more other evidence, but I have to refer you to any
handbook of IE or any good historical grammar of Sanskrit.

Please do so. If I don't own them yet, I'll buy them. I've mentioned
five in this thread alone. As you know, the best way to learn is
usually reading combined with class room lectures and questions. This
is my classroom time! To show that I'm doing my homework also -- the
inspiration for this thread was my reading of Hock's article refuting
Misra -- which I finally got. (Why are linguistics articles so @#$%
ing hard to find?)

Ah, I got your list in the other post. I own Whitney. I guess I'll
have to order Masica.

> Stress is stress, and quantity is quantity. In Vedic poetic metres
there are both stressed and unstressed <e>'s and <o>'s (stress was
free and _independent_ of vowel length at that time), and they
_always_ count as long.

Yes, forgive my imprecise language.

> As regards the comparative aspect, Skt. <e> and <o> correspond to
diphthongal *ai and *au in Iranian and even in Mitanni Indo-Aryan
(Mit.IA aika- = Skt. eka-), and to the diphthongs *ei/*oi/*ai and
*eu/*ou/*au in the rest of Indo-European.

This is very central, I think. Although I suppose it could just as
conceivably have gone from e > ai in Iranian/Mitanni. I think the
most crucial point I've seen so far is the Greek evidence where you
have many sounds collapse into one. It could possibly go the other
way, but you'd need to see some kind of reason for it to do so.

> Hope you find it sufficient now.

Hehe, I don't expect to get closure on this issue for a long time.
But you have given me some things to think very deeply about.

The value of these discussions is two-fold: it inspires me to wade
thru the texts and it highlights some of the main issues so that they
jump out at me when I'm reading. You have clarified many things for
me and highlighted many of the central points that this issue hangs
on.

My deepest thanks.

<wanders away with hand to furrowed brow>