Sanskrit - Printable Version

Sanskrit - Printable Version

+- Forums (http://india-forum.com)
+-- Forum: Archives (http://india-forum.com/forumdisplay.php?fid=7)
+--- Forum: Library & Bookmarks (http://india-forum.com/forumdisplay.php?fid=21)
+--- Thread: Sanskrit (/showthread.php?tid=861)

Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Sanskrit - Guest - 01-07-2004

Sanatana means the eternal (For all time).
Dharma is derived from the root Dhr which means to uphold.
Dharma is the eternal laws of the universe that are upheld by Isvara (God).

Sanatana Dharma refers to the eternal laws of the universe both spiritual & scientific.

Our religion only qualifies for this title Sanatana Dharma. Since it originates from Isvara (God) himself.

All other religions are man made and emphasize one aspect or another of Sanatana Dharma.

Sanskrit - Guest - 01-28-2004

Sanskrit Optical Character Recognition Research at University of Buffalo , NY

> UB researchers, versed in Sanskrit and computer science, develop tools to
> bridge the digital divide
>
> Software will boost Web access to Indian-language documents
>
> So, you think searching for things in English on the Internet is
> frustrating?
>
> Well, try searching for documents written in ancient Sanskrit, modern
Hindi
> and any of dozens of Indian and South Asian languages that are based on
the
> beautiful, intricate symbols of the Devanagari script.
>
> The ability to put this valuable content online from printed sources in
> Devanagari, requires optical character recognition (OCR), the tool
necessary
> to turn any text document into a digital one.
>
> "The lack of a good OCR for Devanagari has made it very difficult to make
> available on the Web the vast majority of Devanagari documents," said Venu
> Govindaraju, Ph.D., associate director of the University at Buffalo's
Center
> of Excellence in Document Analysis and Recognition (CEDAR) and UB
professor
> of computer science and engineering.
>
> Now, with funding from the National Science Foundation, Govindaraju and
his
> UB colleagues are taking a major step toward boosting online access to
these
> documents.
>
> The UB researchers happen to share not only expertise in machine-print and
> handwriting recognition, but also a rare passion for -- and fluency in --
> Sanskrit and other Indian languages.
>
> Their project, funded under a $487,000 grant from the National Science
> Foundation's International Digital Libraries initiative, endeavors to make
> Devanagari documents, ranging from ancient Sanskrit masterpieces, such as
> the Bhagavadgita and the Vedas, to contemporary documents in Hindu,
Marathi
> and other Indian languages, easily accessible on the Web.
>
> The researchers, based at CEDAR, have created a software tool that is the
> first step in developing OCR for Devanagari, ultimately allowing documents
> in these scripts to be widely searchable on the Web.
>
> It will be presented by Govindaraju, who is the principal investigator, on
> March 11 at the 13th International Workshop on Research Issues on Data
> Engineering in Hyderabad, India.
>
> The UB researchers expect to make it available for free on the Web by the
> end of March.
>
> "We are developing machine technologies to read Devanagari documents,
> whether they are contemporary documents written in Hindi or ancient
> documents that were handwritten on palm leaves," said Govindaraju.
>
> The project, which involves collaboration with the Indian Statistical
> Institute in Kolkata, one of India's premier research institutions, takes
an
> important step toward bridging the digital divide between the developed
> world and some developing nations, according to the UB researchers.
>
> "The half-billion people around the world whose main language is Hindi, or
> based on Devanagari, are totally missing out on the 'information
> revolution,'" said Govindaraju. "In IT, the native languages all have
taken
> a back seat."
>
> While Sanskrit has been considered a "dead" language, he noted that in his
> native India a movement to revive it both in written and spoken forms has
> been gaining ground and in certain regions, schools are including Sanskrit
> in their curricula.
>
> He and his UB colleagues on the project are among those in the U.S. who
have
> rediscovered the language; they teach Sanskrit to their own children and
> hold classes in it at the Hindu Cultural Society of Western New York.
>
> "The Indian civilization is 5,000 years old," said Govindaraju. "So there
> are many, many documents written in Devanagari script, but if we want to
> include them in a digital library in order to preserve access to them, we
> need to develop software that recognizes the script."
>
> OCR, the UB researchers explain, essentially "trains" the computer to
> correctly interpret the images of a particular alphabet based on "truthed"
> data, that is, numerous scanned images of characters or words and their
> interpretation recorded by humans who have visually examined the original
> images.
>
> About 15 years ago, UB's CEDAR, the largest research center in the world
> devoted to developing new technologies that can recognize and read
> handwriting, developed the first comprehensive OCR for handwritten
documents
> in English.
>
> That turned out to be a milestone, spurring numerous new research projects
> into handwriting recognition that led to some of the applications now
taken
> for granted, such as personal digital assistants.
>
> "Similarly, we are expecting that the development of benchmarked OCR for
> Devanagari will trigger a groundswell of research in machine-reading
> technologies for these Indian languages," said Govindaraju.
>
> To develop benchmarked OCRs, the UB researchers have constructed a dataset
> of 400 pages of Hindi and Sanskrit documents from books and periodicals,
> both ancient and contemporary, that is representative of the huge variety
of
> documents available in these languages.
>
> The researchers have used the tool they developed to record information
> about these documents that indicate how OCR for Devanagari should
interpret
> each word. The researchers also plan to develop character databases and
> on-line dictionaries, text corpora and other tools for linguistic analysis
> that will be invaluable to the OCR community.
>
> "The availability of our truthing and evaluation tool together with the
> availability of new truthed Devanagari data, will spur greater research in
> the development of Devanagari OCR," said Srirangaraj Setlur, Ph.D., senior
> research scientist at UB's CEDAR and co-investigator.
>
> Vemulapati Ramanaprasad, Ph.D., senior research scientist at UB's CEDAR,
> also is co-investigator.
>
> In the future, the UB researchers plan to extend the scope of this tool to
> include OCR evaluation for other Indian languages such as Kannada,
> Malayalam, Tamil and Telugu, that do not use the Devanagari script, as
well
> for as Arabic and Urdu.
>
> Contact: Ellen Goldbaum, goldbaum@b...
> Phone: 716-645-5000 ext 1415
> Fax: 716-645-3765

Sanskrit - Guest - 02-19-2004

Folks, for sometime now I have been pondering whether we should change our national language to Samskrit? What do members think? What are the pros and cons?

Sanskrit - Guest - 02-19-2004

Sanskrit can be one of the national language, but very limited people can speak or understand Sanskrit. There are couple of villages and families in India, Thailand and Malayasia who only communicate in Sanskrit. Schools can add Sanskrit as another subject.

Sanskrit - acharya - 03-13-2004

> We know what Sanskrit is, and what it isn't.

I do not understand the impetuousness in saying the above. However, I
do understand what you mean as a linguist.

But perhaps what you do not know is how the indigens see it....or is
it that one takes no cognizance of the fact that the indigen might a
opinion of their own on matters of such importance?

> It's just a question of terminology.
> I mean no disrespect, but a "dead" language is not one
> that's completely disused and forgotten; it's one that isn't the
first
> native language of anyone, which undergoes no or little change, and
> whose use is restricted to particular purposes.

Please do remember, over centuries, sanskrit has been used on a daily
basis by millions of Indians as part of their daily ritual. Every
significant event in many a Indian's life is touched and sanctified
by it and is perhaps incomplete without it. The relationship has
grown deeper than with "mere" language of communication - it has
grown into a cultural-bond. It is not a "professional competence" to
develop - much the attitude a modern western linguist (MWL) takes.

IMHO, neglecting cultural significance of a language is not the best
way of studying it. This is what the MWLs do. What excuse do they
present? -"Hey, I classify it as a dead language, so what's your
problem if I call it so?" IMHO, that attitude has got arrogance
written all over it and the backlash of that is what a MWL feels...as
one "wondering" professional Sanskritist found out sometime ago.

Allow me to emphasize. We are not discussing a particle travelling at
the speed of light. That can be done quite un-emotionally, and since
the particle has little say on the issue, what we conclude might be
taken as correct. However, we are disscusing about a language. One
significant aspect of a language is expressing emotions. Further we
are not discussing some yesterday-born-today-dead language like
Esperanto. We are discussing a 5000+ year old language (disregarding
the modern-linguistic-classroom-classification of "Vedic Sanskrit and
classical-sanskrit", it is _Sanskrit_ afterall isn't it?), one which
the natives have taken _extra-care_ to preserve it unchanged, and
have lavished on it highest praise - clearly indicating that they
fully well understand and appreciate the value and significance of
the language AND of what they are doing to it.

IMHO, it is naivety to expect no reaction when one tramples over
centuries old carefully preserved tradition (often on the face of
onslaught) with loosely formed speculative theories of "linguistics"
with utter disregard to what the cultural significance of the
language and what the native users/protectors of the language have to
say about it.

I understand, and, I am not being emotional here. I fully appreciate
the "Einsteinian" attempts by linguists to find the elusive unifying
base - the *PIE. But, I am not convinced that the PIE (without the *)
ever existed. OTOH, I am also equally convinced that it _DOES_ exist,
but only in the minds of the protagonists who have bought the theory.

Please note: Nothing personal here. I just expressed what in my
opinion is the error the MWL commits when he/she calls Sanskrit a
dead langauge. Unlike other languages, the "deadness" of it was not
out of disuse, but out of intention - to preserve and pass on the
cultural ethos via an unambiguous and "timeless" medium in an attempt
to stem the ravages that time causes to culture. In its so-
called "deadness" originates the sublime continuity of a culture and
a tradition - as one found nowhere else on earth.

> As Old and Middle Indic vernaculars gradually
> changed into the Modern Indo-Aryan languages, Sanskrit was
preserved
> unchanged by artificial means.

Clearly, it wasn't a "cultural-error" that the language got
preserved, it was a "cultural-intention".

So, please help me understand why, in your opinion did Sanskrit
undergo little change? What "artificial means" were used to preserve
it?

Sanskrit - Guest - 03-13-2004

Can we all, Samskruta fans do a big favor to ourselves by calling samskrutha, samskrutha and NOT saNskrit?

In conversations with desis I always use "samskrutha" when I refer to it. I am sure this makes sense to all desis. Then, why dont we just be ourselves, throw the gora's yoke...............

TO BEGIN WITH PLEASE SPELL SAMSKRUTHA PROPERLY IN THE TITLE OF THIS THREAD. MODERATOR, I URGE YOU TO DO SO ASAP

Discalimer: I have no formal training in Samskrtha but I am making an effort to learn at home.

Sanskrit - Sunder - 03-13-2004

QUOTE(manju @ Mar 13 2004, 10:56 PM) Can we all, Samskruta fans do a big favor to ourselves by calling samskrutha, samskrutha and NOT saNskrit?

TO BEGIN WITH PLEASE SPELL SAMSKRUTHA PROPERLY IN THE TITLE OF THIS THREAD. 
Bhavathu Bhavathu.. Idham shabdhasya angila akshara-progyogah matram saukaryartham asthi smah <!--emo& Wink

--><img src='style_emoticons/<#EMO_DIR#>/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif' /> Asya shabdasya akshara-prayogah "Sanskritham", Samskrutham adhi varthethe.

Ko'apyapacharam na bhavathi. Thasmaath, shantham bhava bho.

Sanskrit - Guest - 03-14-2004

QUOTE(Sunder @ Mar 13 2004, 11:08 PM) <div class='quotetop'>QUOTE(manju @ Mar 13 2004, 10:56 PM) Can we all, Samskruta fans do a big favor to ourselves by calling samskrutha, samskrutha and NOT saNskrit?

TO BEGIN WITH PLEASE SPELL SAMSKRUTHA PROPERLY IN THE TITLE OF THIS THREAD. 
Bhavathu Bhavathu.. Idham shabdhasya angila akshara-progyogah matram saukaryartham asthi smah <!--emo& Wink

--><img src='style_emoticons/<#EMO_DIR#>/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif' /> <img src='style_emoticons/<#EMO_DIR#>/laugh.gif' border='0' style='vertical-align:middle' alt='laugh.gif' />

Sanskrit - Guest - 03-14-2004

Samskruta, Samskrutha, Sanskrita etc...

It depends upon which transliteration scheme you use. There is nothing correct about any of the above transliterations.

Rishi and Rushi both have the vowel Ri(or Ru) which are rendered incorrectly. It is neither Ri nor Ru but something in between. Remember Maha+Rishi becomes Maharshi after the Sandhi, it doesn't become Maharishi or Maharushi.

About Sam vs San in Sanskrita or Samskrita, the Anuswar or the dot is kept for nasals appearing before the consonants that don't fall in the five major consonant groups. If any consonant falls in the five major groups (ka-vrga, cha-varga ... pa-varga) then any nasal coming before such a consonant takes the form of nasal of that group.

Consonant "sa" is not in the five major groups, it is an "Antahstha". That is why a dot or an "Anuswar" is used in writing Sanskrita. But the nasal is "definitely not" "m" as in Samskrita!! That would work if "s" were labial or "Pa-varg". But "s" is actually a dental consonant also called "dantya sa". The nasal of the dantya group is close to "n" and not "m" which is of labial group. So of all possible spellings Sanskrita is more correct than samskrita.

About "tha" vs "ta" again it depends upon your transliteration scheme. In regular transliteration "t" stands for dental ta-varga, while a t with a dot under it stands for retroflex Ta-vrga. In Itrans scheme the dental "t" is transliterated as "t" while the retroflex t is written as capital "T". In north Indian casual transliteration both dental and retroflex "t" are written simply as "t". Since english doesn't distinguish between these two classes of consonants. In north "tha" is kept for aspirated version of the consonant such as in swastha (healthy). In south "th" is often used for dental "ta" and "t" for retroflex "ta". There is nothing right or wrong about any of these choices. These are the wages of trying to use english to write Sanskrit which is far superior in linguistic phonetic aspects.

Sanskrit - Guest - 03-14-2004

Also ditto what Sunder said:

"Ko'apyapacharam na bhavathi."

Sanskrit - Sunder - 03-14-2004

QUOTE(rhytha @ Mar 14 2004, 09:47 PM)<div class='quotetop'>QUOTE(Sunder @ Mar 13 2004, 11:08 PM) Bhavathu Bhavathu.. Idham shabdhasya angila akshara-progyogah matram saukaryartham asthi smah <!--emo& Wink

--><img src='style_emoticons/<#EMO_DIR#>/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif' />Â Asya shabdasya akshara-prayogah "Sanskritham", SamskruthamÂ adhi varthethe.

Ko'apyapacharam na bhavathi. Thasmaath, shantham bhava bho.

can any one translate it plz? Huh

</div>
So be it.. The english spelling of this word is only for conventional convenience. <!--emo& Wink

--><img src='style_emoticons/<#EMO_DIR#>/wink.gif' border='0' style='vertical-align:middle' alt='wink.gif' /> The word can be transliterated as "Sanskritham", Samskrutham etc.

There is no harm in it whatsoever. So,be cool.

PS: If ever you wish to learn samskrit, here is an excellent site.
http://vedamu.org/home.asp

http://vedamu.org/Sankrit/sankritmain.asp

Sanskrit - Guest - 03-15-2004

Four letters from Sanskrit alphabet have proven to be especially troublesome as far as correct pronunciation goes. These are:

Troublesome vowels: Ri(or Ru), Li (similar to Ri but hardly ever used), ai, Visarga
Troublesome consonat: Murdhanya Sha

My thoughts on these are given below. But before that lets clarify what is a vowel and a consonant:

Vowel: Open your mouth and put it in a certain configuration. A configuration involves amount of opening of the mouth, shape of the lips (normal, rounded or stretched sideways), position of tounge inside the mouth etc. Make sure the configuration is static, that is no parts of your mouth are moving. Then release a sound through your vocal cords. The sustained sound that you hear is a vowel. Vowels like a,aa,i.ii, e, o etc are simple.

Consonants: Consonants involve touching part of your palate, toungue, lips and teeth together and then releasing a sound. The sound becomes a vowel when it is sustained. But just the initial sound which is dominated by the initial location of the blockage is the consonant. A consonant can be pronounced only with a vowel. Depending upon where the toungue touches the mouth (or where the intial blockage is created) we get the five major consonant classes:
ka-varga: toungue touching the root of the mouth
cha-varga: tounge touching lower palate
ta(murdhanya) varga: toungue touching the upper palate
ta(dantya) varga: toungue touching the base of the teeth
Pa-varga: lips touching each other

Each varga has 5 consonants. In each varga the first consonant is unvoiced-unaspirated, second is unvoiced-aspirated, third is voiced-unaspirated, fourth is voiced-aspirated and fifth is nasalized.

ka is unvoiced while ga is its voiced counterpart. The difference between voiced and unvoiced is that vocal cords are vibrating from the very beginning for a voiced consonant. Try to speak ka and ga. You will notice that toungue touches the same location but in the case of ga there is extra "resonance" felt at the throat due to vocal cords vibrating from the very beginning.

An aspirated consonant is one which has an associated "ha" sound attached to it like kha and gha. Kha is unvoiced and gha is its voiced counterpart.

Other consonants like ya, ra , la, va are intermediate consonants along with sha (talavya), sha(murdhanya) , sa (dantya) and ha. Ha is the last true consonant. Ksha, tra, jna are composite consonants.

Vowel Ri or Ru 

The way it is written in english it seems to be a consonat "r" combined with a vowel "i" or "u". But in Sanskrit it is supposed to be a vowe!! Lets keep this in mind. IMHO the correct way to pronounce this vowel is to keep your tounge in the same position as when pronouncing "r" but not touching the palate and then make a guttural sound. Even better, start with a sustained sound of "a" and then slowly curl up your toungue up while trying to keep your vocal cords vibrating the same way. When the toungue is in an upward position but not touching the palate, then sustained sound that you hear is the vowel "Ri" or "Ru". To justify this remember that Maha+Rishi=Maharshi after sandhi. The "r" in Maharshi comes close to the sound of the vowel "Ri" or "Ru".

Vowel Li 

Just as in the case of Ri but don't curl up the toungue as much as in pronouncing "r" but little less as in pronouncing "l". This vowel is hardly ever used.

Vowel ai 

This vowel is often butchered by many people. It is a single vowel, but is often pronounced as two vowels "a" followed by "i". This not what is meant by a vowel! While pronouncing a vowel, the configuration of the mouth should not change. But for most people intially the mouth is in the configuration for pronouncing "a" and then changes to "i". Why have a vowel for "ai" then if two separate vowels can do the task? Try to speak english word Hat. Letter "a" as pronounce in the english word "hat" comes close to the vowel "ai". You will noticed that mouth can be put in a steady configuration while pronouncing this vowel. And it is clearly different from vowel "e" (similar to letter a in english word ape)

Visarga 

This is another big casualty. Most people emit a heavy "ha" sound when pronouncing a visarga. You will mostly hear "Ramaha gacchati" etc. If the consonant "ha" could do the task why invent a vowel for it? The reason is lost on most people. It is so bad that most school teachers insist on a heavy "ha" or "aha" pronunciation for the visarga which IMHO is completely wrong.

The literal meaning of visarga is an ending. The dissolution of universe at the end of time can also be called a visarga. The word "Visarjana" in the sense of discarding or ending comes from the same root.

Word Rama with the vowel "a" at the end has a smooth traling off sound as the vowel ends. The smooth trailing nature of the vowel is not mandated for the vowel. The vowel can possibly end in a different way. Suppose that insted of ending the vowel "a" smoothly, you end it abruptly, sharply cutting off the sound. In my opinion Visarga is just that. Sharply cutting off the sound of the vowel rather than letting it smoothly trail off to zero sound level.

If you try it yourself, this sharp ending will sound almost like a "h" sound. But it is not a full "ha" sound. And if you are careful you can end the vowel sharply without actually giving a hint of the letter "h" in it. I have heard some british clipped accent where many words are ended sharply like that. It may have been a common Indo-European trait in languages.

Murdhanya Sha 

Talavya "sha" as in english word sharp and sanskrit word shankar and dantya sa as in satya are easy to pronounce. It is the third "sh" also called murdhanya sh that thows most people off balance. It occurs in words like "Dhanusha" or "Purusha" etc. But this one is simpler. The sequence talavya, murdhanya and dantya is a give away to the correct pronunciation. In Talavya sha the toungue almost touches the lower palate, while in the dntya sa it comes close to the teeth. Murdhanya sha must be in between. Murdhnya ta (or retroflex ta) is pronounced with the toungue curling upwards. If you do the same with the toungue curled upwards and almost touching the upper palate and release a sha kind of sound that will be murdhanya sha.

Sanskrit - Guest - 03-15-2004

Adding to the Murdhanya sha topic above:

There were two more kinds of "sha" in Vedic litearture (as if three were not enough! <!--emo& Smile

--><img src='style_emoticons/<#EMO_DIR#>/smile.gif' border='0' style='vertical-align:middle' alt='smile.gif' /> )

They were called :

Jihvaamuliya or Kanthya (throat based)
and
Upupadhmaniya (labial).

This completes the story of the "sha" by putting them in all five consonant classes and not just three as it is now.

Jihvamuliya in my opinion would be pronounced almost like an urdu kha with a dot under it as in the urdu word "khat" meaning letter. Upupadhmaniya would be pronounced almost like the english "f" but using only the lips, like the sound you make while blowing out a candle.

Sanskrit - Guest - 03-15-2004

On Visarga and Anuswara 

Although both these are put togther with the vowels they are actually "vowel modifiers".

When a vowel has a nasal sound then Anuswara comes into play.

When a vowel ends sharply/abruptly then it has a visarga.

Sanskrit - Guest - 03-21-2004

Hi Ashok Kumar,

Excellent points! ...always great to read your posts on samskrtam.

I have a few points/questions:

1. Would it be better to transliterate the vowels Ri/Li as R and L in in English?
For example : Krshna & Klptam
Since these vowels represent the "pure" R and L sounds?

Also other indo-european languages like Serbian & Croatian have the vowel R and they transliterate them as R in english. Example:

Srbija (serbia)
Brcko (town in former yugoslavia)
Hrvatska (official name of croatia)

Dr. Walter H. Maurer a linguist has stated that the vowel L was present in proto-indo-european and is infact present in the unaccented pronunciation of L in English words like bottle and myrtle.

2. Also since the anusvara is written "AM". Shouldn't it be samskrtam instead of sanskrtam?

Sanskrit - Guest - 03-21-2004

Ashok Kumar,

Thank you very much! I wish they had explained these things in school when they taught sanskrit. Sanskrit in school was a nightmare - mug up subhashitani and the 3x8 tables (declinations?)

Can you elaborate on the 3x8 tables too..

Thanks

Sanskrit - G.Subramaniam - 03-21-2004

http://www.samskrita-bharati.org

offers low cost long distance learning of sanskrit

Sanskrit - G.Subramaniam - 03-22-2004

Samskrita - Bharati offers 4 training segments for correspondence course
each costs $100 and runs 6 months

G.S

Sanskrit - Guest - 03-28-2004

Mitradena,

Thanks. Sorry for a late reply. I didn't know anyone was reading these threads. <!--emo& Smile

--><img src='style_emoticons/<#EMO_DIR#>/smile.gif' border='0' style='vertical-align:middle' alt='smile.gif' />

1. Good points about Ri and Li. Parallels with other Indo-European languages do make good sense although I am personally not familiar with them.

2. As far as transliteration in english goes, we just have to stick to a convention. There are many transliteration schemes used in scholarly circles.
Of major schemes in use at present, the oldest is CSX (complex sanskrit encoding). CSX uses a charcter set larger than pure ASCII with dots, bars over/under the letters etc. CSX is not easy to type on a normal keyboard. For this reason ITRANS transliteration scheme was developed. ITRANS uses only ASCII characters and is phonetically more appealing to Indians who are used to Sanskrit like pronunciation. A similar ASCII based transliteration was developed by Kyoto Univ and is now called Harvard-Kyoto transliteration scheme.

If I were to recommend a scheme I would recommend ITRANS. An associated Windows software called "Itranslator" is a great tool to write in Sanskrit with Devanagari as well as CSX ouput. CSX is still the transliteration scheme of choice for scholarly publications.

Website for Windows software for Sanskrit writing:
http://www.omkarananda-ashram.org/Sanskrit/Itranslt.html

Website for a Windows software for Monier Williams Dictionary:
http://members.ams.chello.nl/l.bontes/

3. Regarding Anuswara , without any Sandhi it is usually written with trailing "Am" with labial (pa-varga) ma at the end. But this nasal changes form when Sandhi is employed. The general rule is that the anuswar takes the form of the nasal of the group to which the following consonant belongs.

For exmple:
sam+gama=Sa(n)gama. Here after the sandhi (n) is now ka-varga nasal, same as in Ganga.
sam+chaya=sa(n)chaya, here the nasal (n) is now cha-varga nasal as in pancha.
sam+tosha=sa(n)tosha, here (n) is dental (ta-varga)
sam+patti =sampatti, here the nasal is same as m (pa-vatga), so it doesn't change form.

The second s in sanskrita is a dental consonant although not in the five major classes of consonants. Therefore nasal of sam should sound closer to the dental (n) than m. But since s is not a pure consonant in the 5 major classes, it is written as a dot.