Devanagari/Sanskrit: an unknown ligature

marrish

Senior Member
اُردو Urdu
Hello,

I thought I could read this script well and I didn't have any problems till date. Here in this picture there is a ligature I can't read.
Screen Shot 2015-02-13 at 00.23.04.png
I mean that after pitr.mi... Is it supposed to be tra?

Thank you.
 
  • marrish

    Senior Member
    اُردو Urdu
    Sure! कथसरित्सागर आदितस्तङ्गः ६/३०

    अन्योन्यं निजवाणिज्यकलाकौशलवादिनाम्। कचिञ्चं वणिजां मध्ये वणिगेकोऽब्रवीदिदम्।।
    अर्थैः संयमावानर्थान्प्राप्नोति कियदद्भुतम्। मया पुनर्विनैवार्थं लक्ष्मीरासादिता पुरा।।
    गर्भस्थस्य च मे पूर्वं पिता पञ्चत्वमागतः। मन्मातुश्च तदा पापैर्गोत्रजैः सकलं हृतम्।।
    ततः सा तद्भयाद्गत्वा रक्षन्ती गर्भमात्मनः। तस्थौ कुमारदत्तस्य पितृमि...स्य वेश्मनि।।
     

    Gope

    Senior Member
    Tamil
    Yes, it's a tr (त्र ).

    I found this text in google books by doing a word search :)

    https://books.google.com/books?id=l...तस्य&pg=PT25#v=onepage&q=कुमारदत्तस्य&f=false

    Personally I have never seen this ligature before. Looks like some archaic symbol from the past. It is not used in modern Devanagri any more.

    tarshkya jii has clearedthe mistery. More likely it is a misprint, one of the reasons why traditional scholars always believed in oral transmission.
     

    marrish

    Senior Member
    اُردو Urdu
    My version of the text is by Pt. Durgaprasad and Kashinath Pandurang Parab, Bombay 1889. Might be "archaic" :) On the margin, by 11th century AD the oral transmission would not be relevant I think.

    Thank you both for informing me, it is enough to move forward with these parables. Probable typos are on my part as I have re-typed this.
     

    Gope

    Senior Member
    Tamil
    My version of the text is by Pt. Durgaprasad and Kashinath Pandurang Parab, Bombay 1889. Might be "archaic" :) On the margin, by 11th century AD the oral transmission would not be relevant I think.

    Thank you both for informing me, it is enough to move forward with these parables. Probable typos are on my part as I have re-typed this.
    The same book can be read here: https://archive.org/stream/KathaSaritSagaraOriginalText/katha_sarit_sagara#page/n25/mode/1up
    on p.16 you can see it is printed correctly as पितृमित्रस्स . The letter in your original post can only be read as मिन्त्रस्य (mintrasya), very probably a typo and not an archaic version of त्र. Sorry for prolonging the discussion.:)
     

    Dib

    Senior Member
    Bengali (India)
    I am reasonably sure that it is supposed to be "ttr", i.e. mittrasya, and not a misprint. Whether the character is archaic, I don't know. You'd probably know that (certain?) consonants (I am sure about the stops - not sure about the others - have to check the grammar books later) can de optionally geminated/degeminated in Sanskrit right before "r" and after a short vowel. So, mittra = mitra, pattra = patra, etc. It does not matter whether it was etymologically a "tr" or a "ttr", both forms are acceptable. In modern Hindi, Bengali, and (some?) other modern Indian languages the form without geminate is preferred in spelling, but the form with geminate is preferred in pronunciation. Bengali, however, used both forms in spelling till probably 50-100 years ago (I am sorry, I am not sure about exactly when the forms with geminates were abolished).

    This same optional geminate/degeminate is available right after r, and there y and v also take part in this game. Thus sUryya = sUrya, sarvva = sarva, etc.

    Just to be extra sure, keep an eye for whether your edition uses more of these optional geminates.
     
    Last edited:

    Gope

    Senior Member
    Tamil
    Dib dada is right about त being geminated before र optionally. Perhaps if marrish Saahib came across other instances of त्र written in the way he cited in his original post that would end any speculation about printing mistakes.:)
     

    Dib

    Senior Member
    Bengali (India)
    I am reasonably sure that it is supposed to be "ttr", i.e. mittrasya, and not a misprint. Whether the character is archaic, I don't know. You'd probably know that (certain?) consonants (I am sure about the stops - not sure about the others - have to check the grammar books later) can de optionally geminated/degeminated in Sanskrit right before "r" and after a short vowel. So, mittra = mitra, pattra = patra, etc. It does not matter whether it was etymologically a "tr" or a "ttr", both forms are acceptable.

    If you have access to Whitney's "Sanskrit Grammar", please refer to sections 229 and 232. It says, Indian grammarians had a lot of disagreement about this sort of gemination - whether and where they are allowed or required. Pāṇini describes it as optional for the first consonant in any consonant cluster after any vowel (even if the vowel is in the previous word). Prātiśākhya's require it mandatorily. But both of them acknowledge authorities who deny its existence altogether. Given such disagreements, Pāṇini's position of optionality is, in my opinion, the most accurate.
     

    tarkshya

    Senior Member
    Marwari
    If you have access to Whitney's "Sanskrit Grammar", please refer to sections 229 and 232. It says, Indian grammarians had a lot of disagreement about this sort of gemination - whether and where they are allowed or required. Pāṇini describes it as optional for the first consonant in any consonant cluster after any vowel (even if the vowel is in the previous word). Prātiśākhya's require it mandatorily. But both of them acknowledge authorities who deny its existence altogether. Given such disagreements, Pāṇini's position of optionality is, in my opinion, the most accurate.

    This is valuable information. So if this three pronged symbol is indicative of germination, does it mean that plain old त्र should be spoken with a non-germinated sound?
     

    Dib

    Senior Member
    Bengali (India)
    This is valuable information. So if this three pronged symbol is indicative of germination, does it mean that plain old त्र should be spoken with a non-germinated sound?

    The short answer is "probably yes" for Sanskrit. Other languages may have their own standards, e.g. in Bengali it is always geminated whether written as such or not.

    The long answer is more nuanced, and ambivalent even for Sanskrit. What is clear from all the disagreement among the grammarians, is that there was no phonological distinction between geminates and nongeminates in this position in Sanskrit, as there is none in its modern descendents*. It is perfectly possible that the actual phonetic realization was just a geminate, like we have it now. However, because there was no phonological contrast with the non-geminate, the contemporary linguists/grammarians really had no way to prove it - their own perceptions could deceive them, and they unfortunately didn't have the instrumentation that modern linguists/phoneticians enjoy. Hence, some would analyze it confidently as a geminate, some as non-geminate equally confidently, and others would take the middle ground by being ambivalent to the analysis. The spelling variations could then simply attest to this confused state of the analysis, and not necessarily representative of variable phonetic realities. This is, of course, just an alternative scenario that sounds perfectly plausible to me, and consistent with the speech habit of modern IA speakers. The other alternative is also possible, that the spelling differences represent actual phonetic difference, but then you have to discredit some ancient grammarian or other.

    ---

    *Well, one of my Bengali teachers in school did try to teach us an artificial distinction between "Cri" and "CCri" sequences claiming the first is valid when "ri" represents Sanskrit vocalic R and the second when there was a consonantal "r" in Sanskrit. This may be accurate for Sanskrit (except the question of pronunciation of vocalic R of course), but Bengali has no vocalic R. So, this is basically puritanical hogwash. Just to be clear: I am not saying "Cri" and "CCri" may not contrast in any language. I am just saying, they do not in Bengali and Sanskrit (unless ri is conflated with vocalic R), and to the best of my knowledge - Hindi.
     
    Last edited:

    tonyspeed

    Senior Member
    English & Creole - Jamaica
    tarshkya jii has clearedthe mistery. More likely it is a misprint, one of the reasons why traditional scholars always believed in oral transmission.

    Changes in oral transmission leave no traces and go completely unnoticed...It makes the transmitters seem infallible without checks and balances.
     

    tarkshya

    Senior Member
    Marwari
    .. but Bengali has no vocalic R. ..

    Thanks Dib. I concur that as long as there is no phonological distinction between germinate and non-germinate pronunciation, it really is just an academic question.

    Now let me take the discussion off topic just a tiny bit (sorry about that, but don't want to open a new thread for every single question that comes up in the middle of discussions). You mentioned that Bengali has no vocalic R. By vocalic-R I am assuming you are mentioning the R in Sanskrit words like मातृ , पितृ (maatri, pitri) etc. But like every IA language, Bengali must have a ton of Sanskrit words in unaltered forms, some of them containing vocalic-Rs. So do you mean to say that Bengali pronunciation of these words consciously differs from Sanskrit?
     

    Dib

    Senior Member
    Bengali (India)
    Thanks Dib. I concur that as long as there is no phonological distinction between germinate and non-germinate pronunciation, it really is just an academic question.

    As far as native speakers are concerned - yes. But then, this is applicable to descriptive grammar in general. On the other hand, for a language learner, these details are important to fully master the foreign "accent".

    You mentioned that Bengali has no vocalic R. By vocalic-R I am assuming you are mentioning the R in Sanskrit words like मातृ , पितृ (maatri, pitri) etc. But like every IA language, Bengali must have a ton of Sanskrit words in unaltered forms, some of them containing vocalic-Rs. So do you mean to say that Bengali pronunciation of these words consciously differs from Sanskrit?

    In Bengali, Sanskrit words are borrowed in unaltered form only in spelling*. The pronunciations are most often quite different. Bengali vowel phonology is significantly different from Sanskrit, and the consonant phonology also has some noticeable differences. Some of these are well-known among the neighboring populations through stereotypes - like Bengali is full of o, and b for v, and j for y, and sh for s, etc. Like all stereotypes, they are not completely accurate, but they do represent the general picture. On top of that, Bengali pronunciation of Sanskrit consonant clusters is also sometimes quite unintuitive for the non-native. There are some interesting situations too. Sanskrit "svaamii" is spelt as such, but pronounced "shami" (a is short but with the same quality as Hindi "aa"). In some dialects/sociolects, there is a version "shoami" or "soami", which is closer to the Sanskrit in pronunciation, but is considered nonstandard (even rustic), and is spelt out phonetically as "soyami" if ever written. Same goes with soad ~ shad (< svaada), duar ~ dar (< dvaara), etc. duar is, however, considered poetic, rather than rustic. So, what is really the "unaltered" form, is often tricky to determine in Bengali.

    In the particular case of vocalic r, it is not a conscious difference. Overwhelming majority of Bengalis would use "ri" also in Sanskrit texts, even many Sanskrit scholars (if not most). Also in other cases, I can bet that most Bengalis do not know (and do not care) about what the Sanskrit pronunciation was. The only significant Sanskritism that Bengali has borrowed in its prescriptive standard pronunciation is the barrage of initial consonant clusters (like sk, pr, kl, etc.), or rather maybe only half a barrage, because many clusters have a distinctive Bengali pronunciation like I mentioned before. If someone uses more Sanskritism than that, they will likely be "corrected" by the polite, and made fun of by the not-so-polite. So, you may like to call that "conscious" at some level, I guess, though these actions mostly do not depend on the person's acquaintance with Sanskrit.

    ----

    *I would argue, this is the case also in other modern IA languages. They all adapt the pronunciation in some way or other. Hindi, for example, by applying its schwa deletion rules. But, most of them are not as radical as Bengali (and other Eastern languages), I believe.
     
    Last edited:
    Top