drdhaval2785 / SanskritSorting

Codes written by Dr. Dhaval Patel for Sanskrit Natural Language Programming
2 stars 1 forks source link

Gasuns' Reverse Sanskrit Dictionary Issues #36

Closed gasyoun closed 9 years ago

gasyoun commented 9 years ago

Reverse Sanskrit Dictionary, the book I'm working on in InD. Please see pages 223-224 in the book. To some maybe @funderburkjim has an answer, to some only @drdhaval2785

1) #púṣpa--kárṇa# kárṇa has an accent, while none of the other similar words has:

#lomaśá--karṇa#
#śaśá--karṇa#
#bhāsá--karṇa#

Is karṇá (in #sthūṇā--karṇá# and #api-karṇá#) a different word than karṇa? Can we add accents to similar unaccented words?

2) #kaumbhakarṇa#, #khārjurakarṇa# why not #kaumbha-karṇa#. And if we have #śakaṭī--karṇa#, can't we make #śākaṭīkarṇa# as #śākaṭī-karṇa# for sure?

3) If #śyāparṇa# is unsplit because of sandhi, than good, because we will be able to show the four type of "roofs", if MW markup has it

4) #vi-ṣama---karṇa#, #vi-stīrṇa---karṇa# why not #vi-ṣama--karṇa# with two hypens only?

5) what is the difference between #śvā--karṇa# and #śvā-karṇa#?

6) #sa--karṇa# no accents at all

7) #adhirūḍhā-kárṇa# why adhi not split off as vi in other cases?

8) interesting to see #kucika--rṇa# between #plīhā-kárṇa# and #śú ci--karṇa#

For comparison: Из Штиля (509 стр.) From Stiehl (page 509)

madhya-vartin#
svaira-vartin#
antaḥ-pura-vartin#

From Gasuns Из Гасунса

#mádhya--vartin#
#samīpa--tara---vartin#
#antáḥ--pura-vartin#
#dūrá--vartin#
#nāti---dūra-vartin#
#svaira--vartin#
#cakrá--vartin#
drdhaval2785 commented 9 years ago

Q1 - I would not advise adding accent to anything. Sanskrit has very large volume of rules on how to decide accent. Those rules are called 'phiT sUtrANi'. There are changes in accent based on which kind of samAsa it is. So, accent is quite contextual.

Q2, 4, 5, 6, 7, 8 - not applicable to me

Q3 - This is not a samAsa.

gasyoun commented 9 years ago

A1 - accent is already there. I can't remove it. But I could consider adding to similar backgrounds. How well are you versed in the phiT sUtrANi? Is it a machine doable task anytime in the future?

drdhaval2785 commented 9 years ago

Not at all machine doable in near future. This branch is not at all taught recently in Sanskrit Grammar. I will have to look at the grammar texts. One starting point can be https://www.dropbox.com/s/immot20ahxdqwh7/Phit-Sutra-s%20from%20Siddhanta%20Kaumudi%20with%20Skt%20Commentary%20Subodhini%20-%20%20Pansikar%201908.pdf?dl=0

gasyoun commented 9 years ago

Never heard of even an attempt, maybe Amba has heard, wonder I. So not in next 3-4 years, got it. It's the deep ocean, and not the shallow waters we are in now, got it. If only I would have some brain, too bad I have none.

Shalu411 commented 9 years ago

Namaste What do the authors of those dictionaries say about accents? What base is theirs for taking accents on to these words? We have rules for everything in traditional grammar - but it is important to know if those authors followed them or is it some other "vyavasthA" or system which they followed. First of all I need the accents list. and their representatioon scheme before I can approach for help.

funderburkjim commented 9 years ago

Regarding hyphens in mw.xml: The number of hyphens has no consistent significance.

funderburkjim commented 9 years ago

Regarding accents in MW. I prepared a summary of all accent occurrences in mw.xml at https://www.dropbox.com/s/ylyimo0loyxwms0/accentcount.zip?dl=0

Only udAtta (slp1 = vowel + /) and svarita (slp1 = vowel + ^) occur.

Comparing to the scans of MW, the accents occur only in IAST representations, not in Devanagari.

The udAtta in scans appears as an 'acute' accent over the vowel; for example: image

The svarita in scans appears as a 'grave' accent over the vowel; for example: image

Incidentally, in the IAST in the scans of MW, there occur many instances of a vowel with a circumflex diacritical mark. For example, image

This has nothing to do with accents, but represents the fact that, in this example, the a with circumflex is really a long 'a' that has resulted from a simple vowel replacement sandhi.

Incidentally, the reason that the acute and grave accents are interpreted in SLP1 as udAtta and svarita is that at some early time (ca. 2006) in the process of developing monier.xml, Peter Scharf determined that this was the proper interpretation. It is likely that Peter has further information regarding the 'meaning' of accents in this dictionary. Dhaval or Shalu might want to discuss this with Peter.

funderburkjim commented 9 years ago

regarding 'What do the authors of those dictionaries say about accents?'

One potential source would be the preface material for the dictionary. This preface material is in the documentation for each dictionary

For MW the url for the preface material is http://www.sanskrit-lexicon.uni-koeln.de/scans/csldoc/dictionaries/prefaces/mwpref.html. I don't know whether in fact Monier-Williams discusses the principles of the dictionary's accent usage in the preface.

Monier-Williams also wrote a Sanskrit Grammar (https://archive.org/details/APracticalGrammarOfSanskrit). Based on using a search for 'accent' within the online display's "Search within this text" function, there are several places where accents are discussed.

drdhaval2785 commented 9 years ago

Closing the issue. https://github.com/sanskrit-lexicon/CORRECTIONS/issues/31 is where we can continue discussion further. Not a sorting issue any more