Send sanskrit mailing list submissions to
[email protected]
To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.cs.utah.edu/mailman/listinfo/sanskrit
or, via email, send a message with subject or body 'help' to
[EMAIL PROTECTED]
You can reach the person managing the list at
[EMAIL PROTECTED]
When replying, please edit your Subject line so it is more specific
than "Re: Contents of sanskrit digest..."
Today's Topics:
1. questions by Srinivas (Jay Vaidya)
2. schutva sandhi (Jay Vaidya)
3. Translation (Jay Vaidya)
4. algorithm - philosophical issue (Jay Vaidya)
----------------------------------------------------------------------
Message: 1
Date: Wed, 9 Feb 2005 15:03:48 -0800 (PST)
From: Jay Vaidya <[EMAIL PROTECTED]>
Subject: [Sanskrit] questions by Srinivas
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
Questions and answers:
(1)
>> kaH + Aste -> ka Aste (the a+A -> A rule,
>> a prior rule, cannot be triggered)
>> OPTIONALLY kaH + Aste -> kaYAste (who sits?)
>> --'Y' stands for the laghu-prayatna-tara 'y'
>> in which a very weak palate-contact is maintained.
> Question: a+A -> A rule, why can it not be
> triggered? Is it because we already applied
> a "point of no return" type of sandhi here?
Correct. The chain that terminates with "ka(Y)Aste" is
a "point of no return" rule. Only the approximately
166 rules stated after this can now be applied. On
checking them, only one is relevant -- the one about
changing the pronunciation of 'a' to close-mouthed.
And remember, the "point of no return" is for all
rules, not only sandhi.
> I thought we have the option of "kAsthe", like
> "saishha" had.
saH + eshhaH -> sa eshhaH (only if necessary for
versification) is a pre-point-of-no-return rule (from
within the first 3700). So the conversion of sa eshhaH
-> saishhaH is OBLIGATORY. The example "saishha
dAsharathI rAmaH..." is a verse line that would have
been too long without the sandhi.
(2)
> 'Y' stands for the laghu-prayatna-tara 'y:
> Is this 'Y' an obsolete kind of alphabet? Does
> it belong to some specific style of Sanskrit
> such as Vedic or Pali or something?
By the way, 'Y' is not a part of the Itrans internet
standard, but something I just put there to represent
the laghu-prayatna-tara. It is not obsolete. It is one
of those few orphan letters that did not get their own
spot in the recited alphabet, or a respectable written
symbol. Other examples are the jihvA-mUlIya k- and
kh-, or the upa-dhmAnIya p- and ph-, as well as the
laghu-prayatna-tara 'v'. All of these were
non-obsolete sounds in spoken sa.nskR^ita. 'L^i' just
barely escaped this orphan treatment!
(3)
> Having found that the sandhi's can be applied
> recursively like this, where it it customary
> to stop the process, say, at level 2, level 3 ,
> etc.?
Don't stop! keep recursing till nothin is applicable.
While the algorithm (at least the first 3700 lines of
it) is recursively applied, the programmer (pANini)
had to take care that there are no infinite loops. At
any time, most likely only one rule is applicable.
For example:
You are the poet thinking of the verse-line (in
sa.nskR^ita, of course)
he this dasharatha-son king (is)
(form of 'tad') (form of 'etad') (form of
'dasharatha') (form of 'rAjan')
Processes internal to words happen before processes
external to the word.
Lets look at the first word:
(form of 'tad')
In the first (or first 4) 3700-rule-go-through you
find that 'tad' is a "sarvanAma" and a "prAtipadika",
and that based on your meaning something called
"prathamA" and "ekavachana" is expected. There is
nothing else to do.
Based on this understanding, in the next
3700-rule-go-through you find out that you need to put
the termination
tad + su
There is nothing else to do
In the next three go-throughs you find out that the
'u' of su is a marker to be remembered and deleted,
and 'su' is a "vibhakti".
There is nothing else you can do till that is done.
In the next two go-arounds, armed with this previous
knowledge you make the tad 'a'-ending and change the
first letter to s-
sa+s
There was nothing else to do in all of the 3700 rules.
Given that most processes internal to words to not
interact with the processes external to the word, you
could go through separate circular rule-searches for
each word, but for ease, I will write them parallel
processed:
sa-s eshha-s dAsharathi-s rAj..
(The endings of dAsharathI and rAjA come from the
last-283 processes. Since they do not interact with
the first two words, I will not discuss them for
simplicity.)
the -s of sa-s and eshha-s are dropped in the next two
3700-go-arounds (since the two are not interacting, I
will write them parallel-processed.
sa eshha dAsharathi.. rAj..
There was nothing else to do before this is done.
In the next 3700-go-around you find another rule
triggered, but no other than that:
saishha dAsharathi.. rAj...
In the next 3700-go-around you find that there is
nothing to do. You may now enter the No-looking-back
final 283 rules. You get the endings for rAj... and
dAsharathi... (in that order! :-) -- not really
important)
saishha dAshrathI rAjA
Then you keep going until you come to the final rule
until you come to the last, "make all 'a'
close-throated"
saishh'a' dAsh'a'r'a'thI rAjA
(You find three 'a' make them close-throated)
There is no looking back, and no more rules.
You say
"saishha dAsharathI rAjA"
Given the algorithm's operator precedence, there is
never any UNSTATED optionality during any
3700-go-through. There is always only one thing you
can do in any go-around. In the absence of STATED
options there is only one valid final form. Could
there have been infinite loops? The idea that you go
recursively till there is NOTHING to do means that a
bad programmer could have introduced possible infinite
loops. It was pANini's business to take care that
there were none. And there are none. Only, he did
leave 9 programming errors in the no-look-back section
(you have to look back, or you get stuck with
non-final-sounding forms); but no one has had the
courage to rewrite the whole delicately balanced
algorithm. We just flag the offending rules as "you
may peek behind!".
However, sometimes there are STATED options. These do
lead to more than one valid potential final forms.
> We seem
> to have several chains of recursive aplications,
> within each chain (starting from some node till the
> "most-derived" node) several valid potential final
> forms.
Recursive applications of the algorithm only lead to a
single chain of events, unless an option is STATED.
The exact operator-precedence is too complicated for
the purpose of this message.
(4)
> Where can I buy/copy/download panini's works, etc.
> and "kAshikA" that you referred to in your message?
I was going to direct you to the Rashtriya sanskrit
vidyapeeth website (sansknet.org) where the kAshikA
(and many other documents) were posted. But the site
is dead. The kAshikA is a commentary on the
ashhTAdhyAyI, so all of pANini's rules are found
within the body. I can send you a zipped archive of
the kAshikA that I downloaded from sansknet if you
want.
pAnini's ashhTAdhyAyI (without commentary) is
available at the sanskrit.gde.to site.
I bought Hindi translations of these works from
Chaukhamba Prakashan, Varanasi-1. I understand that an
English translation of pANini by sarojA kAtre is also
available (where?).
Dhananjay
__________________________________
Do you Yahoo!?
Yahoo! Mail - Helps protect you from nasty viruses.
http://promotions.yahoo.com/new_mail
------------------------------
Message: 2
Date: Wed, 9 Feb 2005 15:26:26 -0800 (PST)
From: Jay Vaidya <[EMAIL PROTECTED]>
Subject: [Sanskrit] schutva sandhi
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
avaxId-dIpakaH
> the combination that u r talking about is
> saH+chAdakaH after sandhi it should become
> sashchAdakaH,
> this is know as either a visarga sandhi or a
> Shchuthva sandhi.
> shubhamasthu
---
tanna bhavati | etad-tadoH vishhesham idam | paraM tu
sakash-chhAdakaH, asash-chhAdakaH ity-etayoH vishhaye
sushhThu vadati bhavAn yad visarjanIya-shchutve
bhavata eva|
---
That (visarga sandhi of that type and shchutva) does
not happen. There are special rules for forms of etad
and tad. However, you are correct when the special
rules do not apply -- visarga sandhi and shchutva do
happen. e.g., "That dear/little concealer" ... sakaH +
chhAdakaH =
sakashh-chhAdakaH;
"not-he the concealer"... asaH + chhAdakaH =
asash-chhAdakaH
just as you say.
(Raman has discussed these exceptions briefly.)
Dhananjay
__________________________________
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail
------------------------------
Message: 3
Date: Thu, 10 Feb 2005 09:07:11 -0800 (PST)
From: Jay Vaidya <[EMAIL PROTECTED]>
Subject: [Sanskrit] Translation
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
shloka presented by JKM Rao, slight modifications to
interpretations by A Raman:
QUOTED:
snigdhashshAmAM gurubhisupalaiH parvataM
mAlyava.ntam.h
pashyesuttaMbhitamiva purashkeshapAshaM pR^ithivyAH
tatrAdyApi pratijharajalaiH jarjarAprasthabhAgAH
sItAbharturpR^ithutarashucassUcayaMtyashrupAtAn.h
END QUOTE
Preferable:
snigdhash"y"Am"a"M gurubhi"r"upalaiH...
... parvataM mAlyavantam.h
pashye"r"uttambhitamiva ...
... pura"shh"keshapAshaM pR^ithivyAH |
tatrAdyApi pratijharajalaiH ...
... jarjarApra's'thabhAgaH
sItAbharturpR^ithuthutarashuchas-
... sUcayantyashrupAtAn ||
snigdha- = wet/cool
shAmAM = ??? = shyAm'a'm = dusky
gurubhiH = (because of) big
upalaiH = rocks
parvataM = mountain
mAlyava.ntam.h = Malyavanta
pashyeH= you may please see(d^Rish vidhyartha 2nd
sing)
uttaMbhitam= ud+staMbhitam (special sandhi rule) =
pulled up
iva=like
purashhkeshapAshaM= fore-lock of hair
pR^ithivyAH = earth's
tatra=there
adya api = even now
pratijharajalaiH = by each of the (multitude of)
cascades
jarjarA-prasthabhAgAH = eroded (made old) foreparts
sItAbhartuH = Sita's husband = Rama
pR^ithutara = very opulent
shucaH= grieving
sUcayaMti= indicate (plural)
ashrupAtAn.h = flow of tears
The meaning is mostly as JKMR, AR stated:
Please look at the Malyavanta mountain, which, cool
and dark due to big rocks, is like a pulled-up
forelock of the earth's hair. There, even today, the
foreparts, which are eroded by the many cascades,
indicate falling tears of opulent grief for Sita's
husband.
---
This verse looks like it were from the meghadUta, but
my quick search of the version on the sanskrit.gde.to
site does not show it. But the original does have "you
do this, do that" kind of "polite order"
constructions: "nartayethAH, bhajethAH," etc.
Where is the verse from?
dhana.njayaH
__________________________________
Do you Yahoo!?
All your favorites on one personal page � Try My Yahoo!
http://my.yahoo.com
------------------------------
Message: 4
Date: Thu, 10 Feb 2005 09:26:39 -0800 (PST)
From: Jay Vaidya <[EMAIL PROTECTED]>
Subject: [Sanskrit] algorithm - philosophical issue
To: [email protected]
Message-ID: <[EMAIL PROTECTED]>
Content-Type: text/plain; charset=us-ascii
Re pAnini's algorithm:
Philosophically, the algorithm is only to RECOGNIZE
correct forms, not to form them by native speakers.
Native speakers learn -- who knows by what exact
psychological trick -- the correct forms without
explicit algorithms.
pANini knows and stresses that speech by native
speakers is more fundamental than his algorithm.
Dhananjay
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
------------------------------
_______________________________________________
sanskrit mailing list
[email protected]
http://mailman.cs.utah.edu/mailman/listinfo/sanskrit
End of sanskrit Digest, Vol 23, Issue 9
***************************************
