Bengali: Difference between জ/য, ত/ৎ, ঞ/ণ/ন, র/ড়/ঢ় and শ/ষ/স and usage rules

[rehanelbengali]

Could anyone please enlighten me on the difference between জ/য, ত/ৎ, ঞ/ণ/ন, র/ড়/ঢ় and শ/ষ/স and which one should be used when?

 

[Dib]

ৎ = ত্ just a different shape of the glyph. By convention, some consonant clusters are written by the first (ৎস in উৎস ‘source’), some by the second (ত্ব in গুরুত্ব ‘importance’). At the end of a word in -t in Sanskrit, the spelling is invariably ৎ (অর্থাৎ, যাবৎ, etc.).

ঞ has the theoretical value of a palatal nasal, like Spanish ñ, which appears in Bengali (and Sanskrit) before the palatal affricates অঞ্চল ‘region’, বাঞ্ছা ‘desire’, কুঞ্জ ‘grove’, ঝঞ্ঝা ‘storm’. In all these cases, it is just an allophone of our normal n. So, just pronounce it as n, and you are good. Just remember the spelling rule - in conjuncts before চ, ছ, জ, ঝ.

Same happens theoretically also after চ্ in the (in Bengali) rare word যাচ্ঞা ‘request’. In Sanskrit, you have it also after জ্ (জ্ঞ) but in Bengali that cluster is pronounced as গ্যঁ.

The only word where you may see a stand-alone ঞ in Bengali is in some spelling variants of ভুঁইয়া, e.g. ভুঞা (now a surname, but originally ‘landowner’, ‘lord’), a hold-over from Middle Bengali spelling presumably.

Similarly, ণ is theoretically the retroflex n and appears before retroflex stops in clusters. Though in Bengali pronunciation, it does not occur independently of a following retroflex stop, in Sanskrit it did. So, in Sanskrit words it occurs more freely and you just have to know which word is spelt with ণ and which with ন। Sometimes this has been artificially extended to non-Sanskrit words too to distinguish otherwise homographs, e.g. the old unit of weight is often spelt মণ ‘~38 kg’ to distinguish it from মন ‘mind’.

ন is the default n-letter in Bengali, and should be used in all new (non-Sanskrit) words.

The situation of র/ড় is somewhat similar to ন/ণ but not a clash between Bengali and Sanskrit but between different dialects of Bengali. Traditionally, say before 1950’s, র and ড় were pronounced consistently distinctly in whole of South-West Bengal, including Calcutta - the first primarily as an alveolar trill, the second as a retroflex/post-alveolar tap. So, the spellings show that. In Calcutta itself this distinction is losing ground since after independence and the big influx of Eastern Bengalis who lacked the distinction. We who have roots in the countryside, however, still maintain the distinction natively.TV and radio presenters also largely maintain it (at least in WB) - I am guessing by training, if they didn’t have it natively, because that is the prescriptive standard.

As for ঢ় that is a rather marginal phoneme, limited to a few Sanskrit loans like গাঢ় ‘thick’ (liquid) or ‘dark’ (colour). Those of us who have a distinct ড়, we may pronounce it as an aspirated ড়। But if there is no vowel pronounced after it, it is pronounced as a simple ড় as in আষাঢ় ‘the third month of Bengali calendar’.

শ/ষ/স is also primarily a Sanskrit distinction not maintained in Bengali pronunciation. In clusters, just like ঞ/ণ/ন the distribution is predictable. But elsewhere, you just have to know the spelling. However, it is a bit worse than ণ/ন actually because the s-letter is often ‘restored’ even in the spelling of pure Bengali words based on their Sanskrit etymon. So, সাত ‘seven’ (<সপ্ত), ষাট ‘sixty’ (<ষষ্টি), শ’ ‘hundred’ (<শত), unlike কান ‘ear’ (<কর্ণ) where the ণ is not ‘restored’ in the spelling. In fact, শোনে/শুনে ‘he/she hears’ (<শৃণোতি) shows both - the restoration of শ and non-restoration of ণ in the spelling. New (non-Sanskrit) words tend to use স and avoid ষ, but শ also shows up often enough, especially if the original had an ‘sh’ sound (স্টেশান < station). It’s a mess. I am sorry.

য/জ - I believe, you get the drift by now: Sanskrit distinction, not maintained in Bengali pronunciation, you just have to know the spelling, sometimes etymologically restored in spelling even in proper Bengali words. Additionally, in BD Bengali I see a trend of using য for (Islamic religious?) words that would be pronounced with a z in Urdu/Farsi, e.g. যাকাত or যাকাৎ. I am more used to spelling such words with a জ, like জাকাত।

 

[rehanelbengali]

ৎ = ত্ just a different shape of the glyph. By convention, some consonant clusters are written by the first (ৎস in উৎস ‘source’), some by the second (ত্ব in গুরুত্ব ‘importance’). At the end of a word in -t in Sanskrit, the spelling is invariably ৎ (অর্থাৎ, যাবৎ, etc.).

ঞ has the theoretical value of a palatal nasal, like Spanish ñ, which appears in Bengali (and Sanskrit) before the palatal affricates অঞ্চল ‘region’, বাঞ্ছা ‘desire’, কুঞ্জ ‘grove’, ঝঞ্ঝা ‘storm’. In all these cases, it is just an allophone of our normal n. So, just pronounce it as n, and you are good. Just remember the spelling rule - in conjuncts before চ, ছ, জ, ঝ.

Same happens theoretically also after চ্ in the (in Bengali) rare word যাচ্ঞা ‘request’. In Sanskrit, you have it also after জ্ (জ্ঞ) but in Bengali that cluster is pronounced as গ্যঁ.

The only word where you may see a stand-alone ঞ in Bengali is in some spelling variants of ভুঁইয়া, e.g. ভুঞা (now a surname, but originally ‘landowner’, ‘lord’), a hold-over from Middle Bengali spelling presumably.

Similarly, ণ is theoretically the retroflex n and appears before retroflex stops in clusters. Though in Bengali pronunciation, it does not occur independently of a following retroflex stop, in Sanskrit it did. So, in Sanskrit words it occurs more freely and you just have to know which word is spelt with ণ and which with ন। Sometimes this has been artificially extended to non-Sanskrit words too to distinguish otherwise homographs, e.g. the old unit of weight is often spelt মণ ‘~38 kg’ to distinguish it from মন ‘mind’.

ন is the default n-letter in Bengali, and should be used in all new (non-Sanskrit) words.

The situation of র/ড় is somewhat similar to ন/ণ but not a clash between Bengali and Sanskrit but between different dialects of Bengali. Traditionally, say before 1950’s, র and ড় were pronounced consistently distinctly in whole of South-West Bengal, including Calcutta - the first primarily as an alveolar trill, the second as a retroflex/post-alveolar tap. So, the spellings show that. In Calcutta itself this distinction is losing ground since after independence and the big influx of Eastern Bengalis who lacked the distinction. We who have roots in the countryside, however, still maintain the distinction natively.TV and radio presenters also largely maintain it (at least in WB) - I am guessing by training, if they didn’t have it natively, because that is the prescriptive standard.

As for ঢ় that is a rather marginal phoneme, limited to a few Sanskrit loans like গাঢ় ‘thick’ (liquid) or ‘dark’ (colour). Those of us who have a distinct ড়, we may pronounce it as an aspirated ড়। But if there is no vowel pronounced after it, it is pronounced as a simple ড় as in আষাঢ় ‘the third month of Bengali calendar’.

শ/ষ/স is also primarily a Sanskrit distinction not maintained in Bengali pronunciation. In clusters, just like ঞ/ণ/ন the distribution is predictable. But elsewhere, you just have to know the spelling. However, it is a bit worse than ণ/ন actually because the s-letter is often ‘restored’ even in the spelling of pure Bengali words based on their Sanskrit etymon. So, সাত ‘seven’ (<সপ্ত), ষাট ‘sixty’ (<ষষ্টি), শ’ ‘hundred’ (<শত), unlike কান ‘ear’ (<কর্ণ) where the ণ is not ‘restored’ in the spelling. In fact, শোনে/শুনে ‘he/she hears’ (<শৃণোতি) shows both - the restoration of শ and non-restoration of ণ in the spelling. New (non-Sanskrit) words tend to use স and avoid ষ, but শ also shows up often enough, especially if the original had an ‘sh’ sound (স্টেশান < station). It’s a mess. I am sorry.

Thank you very much. I really appreciate it.

Additionally, in BD Bengali I see a trend of using য for (Islamic religious?) words that would be pronounced with a z in Urdu/Farsi, e.g. যাকাত or যাকাৎ. I am more used to spelling such words with a জ, like জাকাত।

I think this comes from the respect we, Muslims, give to words from Arabic origin. Spelling it with য is a way to say that one should ideally pronounce it with Z (and many of us who do make a distinction between J and Z in loanwords do pronounce it as such), in contrast to জ used for J.

 

[Dib]

I think this comes from the respect we, Muslims, give to words from Arabic origin. Spelling it with য is a way to say that one should ideally pronounce it with Z (and many of us who do make a distinction between J and Z in loanwords do pronounce it as such), in contrast to জ used for J.

Yes, I guess so. It is paralleled by the distinction of s/sh indicated in words of similar background by using ছ, like writing ছালাম instead of সালাম, which is the only spelling I would write as a 'Westerner'.

 

[rehanelbengali]

Yes, I guess so. It is paralleled by the distinction of s/sh indicated in words of similar background by using ছ, like writing ছালাম instead of সালাম, which is the only spelling I would write as a 'Westerner'.

I think using চ/ছ for the S-sound might have more to do with influence from local dialects which pronounce those 2 letters as “S”. I have also seen the S-sound being spelt as চ and ছ with English loanwords like three-piece, which I have seen written variously as থ্রি-পিচ, থ্রী-পীচ, থ্রি-পিছ, থ্রী-পীছ, etc.

 

[Dib]

The only word where you may see a stand-alone ঞ in Bengali is in some spelling variants of ভুঁইয়া, e.g. ভুঞা (now a surname, but originally ‘landowner’, ‘lord’), a hold-over from Middle Bengali spelling presumably.

‍Just remembered - there is another word with a stand-alone ঞ: নঞর্থক, which is a part of grammar jargon (don't use in any other context!) and means 'negative' as in a negative sentence (নঞর্থক বাক্য)। Don't ask me how to pronounce it. Nobody knows. haha 🙃

 

[rehanelbengali]

‍নঞর্থক, which is a part of grammar jargon (don't use in any other context!) and means 'negative' as in a negative sentence (নঞর্থক বাক্য)। Don't ask me how to pronounce it. Nobody knows. haha 🙃

According to Google, it’s pronounced নোই্অঁর্‌থক.

 

[Dib]

Fair enough! :-D