Re: {भारत�यविद�वत�परिष त�} Re: Information Theoretic analysis of Sanskrit langu age

[Dr. P. Ramanujan]

I think you may consult a similar study in the 1990s by Lars Martin Fosse
from University of Oslo.

Ramanujan
On Mon, Nov 12, 2012, Aniket Deshmukh <danik...@gmail.com> said:

> Respected Scholars and members of भारतीयविद्वत्परिषत्,
>
> I am a final year undergraduate student from the Department of Electrical
> Engineering, Indian Institute of Technology (IIT) Hyderabad, INDIA. I'm
> currently working on project- "Information Theoretic Analysis of evolution
> of Sanskrit Language". I have a doubt pertaining to my project regarding
> literary works in Sanskrit. I humbly request you to clear the doubt, which
> I have described in the upcoming paragraphs.
>
> In my project,I evaluated various Information Theory parameters and tested
> them on Sanskrit text files of different sizes. I collected a text database
> of Sanskrit literature - from Rig Veda to Gita Govinda and Padyavali and
> based on results from the Information Theory parameters, studied the
> evolution of Sanskrit language.The database(text files - (Unicode) I have
> collected is from sanskritdocuments.org.
>
> As I'm commenting on the evolution of the language, I need to confirm,
> whether particular literature belong to a particular period. I understand
> that most of these works were composed over a span of time - from a few
> decades to even a few centuries.Below in the table are the eras when I
> considered these works were composed for my study. I request you to tell me
> if any of these works have faced a substantial change over the ages after
> their composition or do they still stay almost true to the original? If
> there have been any major changes I request you to provide me with a some
> further information about it so that I can adjust my study accordingly.
>
>
> Literature Time
> 1. Rig Veda 2000 BC-500 BC
> 2. Ramayana 200 BC – 200 AD
> 3. Mahabharata 200 BC – 200 AD
> 4. Raghuvamsham 300 AD – 700 AD
> 5. Meghadutam 300 AD – 700 AD
> 6. Kumarasambhavam 300 AD – 700 AD
> 7. Gita Govinda 1200 AD – 1300 AD
> 8. Padyavali 1500 AD – 1600 AD
>
>
> Thank you for taking your valuable time for helping me out.
>
> Hope to hear from you soon.
>
> Yours sincerely,
>
> --
> Deshmukh Aniket
> Final Year Undergraduate Student
> Department of Electrical Engineering
> IIT Hyderabad
> Web: https://sites.google.com/a/iith.ac.in/aniket/
> Phone no: +91-9502580449
>
> --
> निराशीर्निर्ममो भूत्वा युध्यस्व विगतज्वरः।। (भ.गी.)
> to subscribe go to the link below and put a request
> https://groups.google.com/group/bvparishat/subscribe
> To unsubscribe from this group, send email to
> bvparishat+...@googlegroups.com
>
>
>
> Email secured by Check Point
>
>
>
>

--
Dr. P. Ramanujan
AD-IHLC,
C-DAC, Bangalore (KP)
ra...@cdac.in
080-25246350, 66116444 (Off.-Dir.)



-------------------------------------------------------------------------------------------------------------------------------

This e-mail is for the sole use of the intended recipient(s) and may
contain confidential and privileged information. If you are not the
intended recipient, please contact the sender by reply e-mail and destroy
all copies and the original message. Any unauthorized review, use,
disclosure, dissemination, forwarding, printing or copying of this email
is strictly prohibited and appropriate legal action will be taken.
-------------------------------------------------------------------------------------------------------------------------------

[Ajit Gargeshwari]

As I'm commenting on the evolution of the language, I need to confirm, whether particular literature belong to a particular period. I understand that most of these works were composed over a span of time - from a few decades to even a few centuries.Below in the table are the eras when I considered these works were composed for my study. I request you to tell me if any of these works have faced a substantial change over the ages after their composition or do they still stay almost true to the original? If there have been any major changes I request you to provide me with a some further information about it so that I can adjust my study accordingly.

Namaste,

All dates on ancient texts  such as vedas are relative I don't think any scholar can give an absolute date. He or she can only give you valid and good reasons why they assign a particular. You can read several of the books and articles published and arrive at you conclusion.

Except for the Vedas, several of Kalidasas works puranas etc have come down in various version followed different Manuscript traditions. Critical edition on a few Puranas and Kalidasas's work have been published you can read them

Are the works original or have interpretations crept a Critical edition of a given work would be the best guide.

The dates you have assigned can be found in any standard history book but a Scholar might and another might not.

The areas pertaining to your question wide vast and there have been several arguments and debates among scholars with no unanimity on dates and variation of texts.

Collections of documents found in Sanskrit.org are not critical editions or don't have all the printed editions and Manuscripts. You have to get a hold of the original books either printed or in manuscripts or both on specific authors or vedic literature and then apply "Information Theory parameters "  One cannot apply information theory and make any statistical analysis on a single text data base.

As to Theoretical Analysis of evolution of Sanskrit Language there a wide range of books published which one choose and read.

My answers are just pointers and I hope it has partially answered your question.
Regards
Ajit Gargeshwari

[Sivasenani Nori]

On Mon, Nov 12, 2012, Aniket Deshmukh <danik...@gmail.com> said:

> Respected Scholars and members of भारतीयविद्वत्परिषत्,
>

> I am a final year undergraduate student ... from (IIT) Hyderabad ... I'm

> currently working on project- "Information Theoretic Analysis of evolution
> of Sanskrit Language". I have a doubt pertaining to my project regarding
> literary works in Sanskrit. I humbly request you to clear the doubt,

 

Mr. Deshmukh

 

Good luck to you in your endeavour.

 

I gather that you do not want comments on the dates considered by you. Even so, you must know about the war between Pulakesin II and Harshavardhana and the edict that was issued after the war, where Kalidasa - along with Bharavi - is mentioned by name. So going upto A.D. 700 for Kalidasa is quite a stretch.

 

Regarding the other question: Except Ramayana and Padyavali (about which I do not know), the received wisdom is that the other texts have not changed. Why? In case of the Rigveda there are various techniques like anukramanis, embedding of string length within the metatext (learnt along with the text), vikriti paaThas like krama and ghana have all kept the text intact. A commentary is considered to be a kavacham which ensures the integrity of the original text. This was supplied in the 14th century through Sayana's Bhashya (though there were other earlier commentaries, we do not have them in full).

 

Kalidasa's works have the commentary of Mallinatha Suri and a number of others, ensuring their integrity. The various differences in readings are also well documented. Usually electronic corpora do not incorporate the paaThaantaras.

 

Jayadeva's work also has commentaries, besides being quite late (at best requiring two transcriptions to survive till modern times).

 

Ramayana due to its length, has been commentated upon only quite late - about 17th century. By that time, quite a few differences have emerged in the various paaThas. There are two very interesting linguisitic analyses of Ramayana, apart from the papers of Dr. Nil Madhav Sen - which to my knowledge are the first detailed analyses in English of the language of Ramayana. The first, a classical linguistic study is by Dr. Satyavrata Sastri, which was carried out even before the Baroda critical edition of Ramayana was published. The second, and far more interesting work is by Dr. L. A. VanDaalen called Valmiki's Sanskrit. There he seemingly starts to do what Dr. Satya Vrat Sastri did, that is evaluate the language of Valimiki in light of Panini, but quickly takes a different turn. He starts looking closely at the sections considered to be interpolations (entire Uttara Kanda, huge swathes of Bala kanda, and many sargas in between according to some) and compares the results on un-Paninian usages there with other considered-to-be-genuine sections. It turns out that in the considered-to-be-interpolated sections there is a huge increase in un-Paninian usages. He investigates this further by comparing the story. For instance in the incident about Jatayu, there are two sargas which go through the same narrative material but in one, Jatayu reveals that it is Ravana who has taken Sita in the southern direction. Now, Dr. Van Daalen argues, if it is already known that it is Ravana who has taken Sita in a southern direction, the elabroate search party being constituted etc. does not hold the attention of the reader to justify such a lengthy explanation. So, he puts the sarga which reveals that it was Ravana who abducted Sita as an interpolation. Then he compares the use of language in that sarga with other considered-to-be-interpolations and his proposition holds.

 

Dr. Van Daalen going against the grain of received wisdom in historiography (which holds that the more corrupt the language, the older the language) claims that Valmiki's Sanskrit has very few un-Paninian usages, but it was the later interpolators with their weak language skills and regional variations - and bad narrative - who made interpolations and made Ramayana what it is today. I believe further analyses of Ramayana, especially to find out the effect of time on language, necessarily needs to be informed by Dr. Van Daalen's work.

Finally, I would urge one thing, though unsolicited. There is a danger that in these kind of daichronic analyses, the researcher sits in judgment - one who kind of decides Valmiki / Kalidas / X knew Sansrkit well, or did not read Asthadhyayi on the conjugation of the distant past tense or reduplication of roots or . . . . That is getting the causation backwards, just like Dr. Van Daalen showed in terms of Paninian usage and possible-interpolations. Under the sutra, prishodaraadi etc. Patanjali explains clearly that SishTas are the pramaaNa for the form of a word that is saadhu and that Vyakarana only helps identify such SishTas. In other words, good Sanskrit is what is used by great people and what Maharshi Panini taught us is the essence of such good usage. Now, if occasionally there are deviations, the right attitude is neither to blame Panini (that he did not describe properly) nor the user (that he did not know Panini). What is the right attitude? Our ancestors call such usages aarshaprayogas - the usage of Rishis. They are what they are - language has a force of its own and sometimes some forms manifest themselves in the words of great people, in the works of great poets (for instance the word asaampratam in Kalidasa's Kumarasambhavam when Brahma talks about his inability to kill Tarakasura whom he himself strengthened: vishavriksho'pi samvardhya svayamacchettumasaampratam).

 

This does not mean that I suggest that there be no analysis. No. Those Rishis themselves would be happy if their children show diligence. The presentation of the analysis ought to be informed of the directionality of cause and effect and at all times, respectful for those who have made the way for us to walk. Modern education, excellent in other spheres, seems to be sorely lacking in inculcating respect and gratitude for the Rishis - or, it might be that respect is reserved only for work in the more mundane sciences - like focussing on Newton's achievements rather than failure to anticipate Einstein, or indeed Einstein's theory of relativity rather than failure to give a complete Grand Unified Field theory.

 

Maybe an analogy from material science would clarify the view that the science of grammar as exounded by Panini is one of the most comprehensive ever - not merely in terms of any language, but compared to other hard, modern sciences as well. For instance, the yield strength of steel is usually about a tenth of what it ought to be if we were to compute it as per the forces operating at the atomic level (which are the real reason for the strength of steel). This is usually attributed to impurities and granularity. However the exact effect of specific impurities on reducing the strength is not quite known, though it would be hugely useful to do so (then Eiffel tower could have one-tenth the material used). We do know the yield strength of steel, but do not know why it is one-tenth of what it ought to be. Similarly flying in aerolplanes is so routine that one would be justified in expecting that the 'science' underlying flying is fully understood. Not quite. We have very little understanding of the turbulent layer of air flow near the skin of the aircraft. We simply know how to predict the behaviour as long as the aricraft doesn't approach the speed of sound too closely (that is why commercial aircraft don't usually go beyond 0.9 mach - around the speed of sound, we know the behaviour of air flow changes). Our lack of fundamental understanding is hidden behind various numerical constants. Like with strength of steel, we can predict the effect but not explain it fully in terms of fundamental science. Just like we do not focus on our lack of complete understanding of material science (in solids like steel, or gases like air), but rather celebrate the great advancements possible due to our understanding, with Panini, our ancestors celebrated what was understood ignoring the miniscule fraction of good words that were not 'understood'.

 

Regards

N. Siva Senani

[Vidyasankar Sundaresan]

It is very welcome news to me that such projects are being done in the
IIT campuses. More power to you and your advisor. The more such
projects are done in India, by Indian students, the better.

A number of pertinent points have been made by others on this thread.
Let me add the caveat that proof reading of the texts available on
sanskritdocuments.org has not always been done well. It is crowd
sourced and therefore there is quite a lot of variability in the
contents. A lot depends on the person who encoded the text originally.
People have not always been able to go back and edit their
contributions to fix mistakes, so please be careful when using them.
(This is just a statement of fact, not a criticism of those who have
contributed e-texts to the archive, for I happen to be one of those
who has been lax about proof-reading the few documents that I have
contributed to it.)

Re: dates - gItagovinda should be at least about half a century
earlier than the time period you have mentioned. A little bit of
research will tell you that there are literary and inscriptional
pieces of evidence about the gItagovinda right around 1200 AD. If the
Padyavali you mention is by Rupa Goswamin, then you possibly again
have to move the period earlier by about 40-50 years at both termini.
However, please note that the Padyavali is an anthology of devotional
verse put together by the Goswamin, so the dates of the actual verses
in it probably span a good chunk of time, from the 11th-12th century
to the 15th century.

I notice that you have only picked poetic works and nothing in prose.
I would hypothesize that prose will show other kinds of trends in the
evolution of the language that verse does not reveal, because it does
not have to conform to meter. There are many prose works available,
starting from the brAhmaNa portions of the veda through the various
philosophical texts, as also texts like daSakumAracarita, kAdambarI,
vAsavadatta etc. Of course, I understand yours is a bachelor's degree
project, not a PhD thesis, so you have to define the scope of your
research more narrowly. This may be something to at least explore in a
preliminary fashion for future extension of your work.

With best wishes for your project,
Vidyasankar

On Nov 12, 10:42 am, Aniket Deshmukh <daniket1...@gmail.com> wrote:
> Respected Scholars and members of भारतीयविद्वत्परिषत्,
>

> I am a final year undergraduate student from the Department of Electrical

> Engineering, Indian Institute of Technology (IIT) Hyderabad, INDIA. I'm

> currently working on project- "Information Theoretic Analysis of evolution
> of Sanskrit Language". I have a doubt pertaining to my project regarding

[Vijnasu]

Dear Aniket,

Its good to take epics to study information theory, but don't come to concluding the dates of those texts.

Bhagawan Shri Ramachandra appeared in 24th Treta Yuga, which would mean approx 14 million years ago. Apparently mentioned in Vayu Purana and possibly Bhagawatam. Date of Vedas cannot be fixed as it was there from the time immortal.