Four

41. A rule from the West Bengal Bangla Academy states: "In the present habitual tense, for regular customary actions, verbs take the 'a'-ending: Which paper do you read? We hear, think, cut in this manner." However, following this convention, I'm finding it quite jarring to write 'bihar' in 'bishwasathe joge jetha y biharo'.
That's not the point. The question is: can we consider that rule's statement as a comprehensive declaration? If I want to drop the 'o'-inflection in favor of the 'a'-inflection for the second person in the present habitual tense, shouldn't this apply universally? Because I notice that when the root verb ends in a vowel, the 'o'-inflection becomes necessary, as in "I know, you give alms." Again, even when the root is a derived verb, it takes the 'o'-inflection, as in "the students you teach," etc. Then why should we drop the 'o'-inflection only for second-person consonant-ending roots? Where's the difficulty in this neutralization of inflections?
Where there's potential for confusion, the 'o'-vowel can certainly be added. In the Bangla Academy's spelling convention, the 'a'-inflection was adopted in writing because others didn't want to call it the 'o'-inflection based on pronunciation. Whatever the case, when there's potential for confusion, writing according to pronunciation is surely appropriate.
Therefore, unless it means 'tui', instead of keeping the 'a'-ending, we'll add the 'o'-vowel. "Tui etai kor. Tumi etai koro." We'll do this even in the present habitual tense.

42. Remember the difference between 'emonki' and 'emon ki'. 'Emonki', with this spelling, connects to yes-no questions. "Did Messi score?" "Yes, even Messi scored." But 'emon ki' and 'ki emon' are the same thing—"What's so great about who played well?" Even someone as glum as Moheshbabu burst out laughing. What happened that prevented you from coming to the meeting that day? The difference is clear in these two examples.
The same principle applies to 'boi-ki'. It won't be 'boiki'. 'Boi ki' can exist without the hyphen. But its usage is different. "I've disobeyed my teacher; what else but failure could be my due!" Whether it's a particle or a pronoun, if there's doubt about that too, if the yes/no question can't be asked either, as in this emonki/emon-ki matter, then as an alternative, see which syllable gets stressed in usage. Where 'ki' gets the stress, unhesitatingly add the long vowel. In the example "emonki Moheshbabur," the stress in 'emonki' falls on 'e' (pronounced 'æ'), so 'k' takes the short 'i'. In "emon ki ghotol," the stress is on 'ki' itself, so 'k' takes the long vowel. In "se jabe boi-ki" too, the stress in 'boi-ki' (pronounced 'boiki') falls on 'bo', so 'ki' takes the short 'i'. In "byorthota boi ki ar amar prapyo," the stress in 'boi ki' falls on 'ki', hence the long vowel.
I favor writing 'emon-ki' and 'boi-ki' with hyphens (though the first is standardly written as 'emonki' without a hyphen) because the particle 'ki' joining them has altered the conventional meaning of the words 'emon' and 'boi'. I'll write 'emon ki' and 'boi ki' separately because here they're separate words, so the current meanings of 'emon' and 'boi' remain intact. The interrogative pronoun 'ki' then becomes the subject of the verb. Though that's correct, in modern spelling, under the pretext of avoiding redundancy and economy, attacks on punctuation are increasingly common.
From hoyil to hol>holo, apostrophes are no longer needed. Do we see much use of hyphens nowadays? But what's the problem with writing emonki, boiki? Not everyone may favor hyphens as I do. Hyphens to avoid morphological confusion, as you write 'na-i' or 'ta-i'. In this convention, the refrain of the song "Dukher timire" would have to be written as 'tobe ta-i hok'.

43. Since Bengali has no capital letters, hyphens are needed to separate proper nouns from inflections, as in "Rabindranath Choturongo-e dekhiyechen." The purpose here is to indicate the uninflected form of the name. With foreign names, this becomes essential, as in "Kant-er darsho n." Some compound words also take hyphens. There's no end to the utility of hyphens (or hyphen-er). Suniti Kumar was so fond of hyphens that he used them even in words like 'bhasha-prakash'.

44. I was surprised to see "Ghora douritechhe" in Ishwarchandra Vidyasagar's Bornopotichoy. We don't say 'douritechhe', we say 'douraitechhe'.
And that's precisely why the surprise was even greater. I do kori, pori, pori, lori, gori, mori, sori (move aside), chori, nori, dhori, bhori, hori (snatch); but not douri—dourаi.
Let's compare with some other verbs. I do makhi and I do makhai—the two have different meanings. Similarly pori and porai, sori and sorai, chori and chorai. Here the first applies to actions on oneself, the second to others. I read myself, I teach others; I climb myself, I make others climb. Yet when running, I myself dourаi, I make others run.
Why? Is 'dourano' then a denominative verb? A verb 'dourano' from the noun 'dour'?
It seems so. Yet running is such an essential action—can we accept that it wouldn't have its own Bengali root form? In Sanskrit, alongside koroti, khadoti, gocchoti, hasoti, there's dhavoti equally present. In Hindi, as they say main korunga, marunga, khaunga, similarly bhagunga. So in Bengali, ami douri, tumi douro, tui dour, etc., should be correct.
And that's why "Ghora douritechhe" should be right. (I move. He's 'choli'techhe. I run. The horse is 'douri'techhe.)
Vidyasagar wrote 'douritechhe', yet generally 'douraitechhe' is the established convention. Why? Jogeshchandra indicated a simple method for extracting the root verb from verb forms. Remove the final 'i' from the first person present habitual form, and what remains is the root verb. For instance, ami pori. (Remove 'i' from pori and we get por, i.e., √por); ami porai (remove 'i' from porai and we get pora, i.e., √pora). Now the question is: what is the root verb in question? √dour or √doura?
I checked Haricharan. This is an intransitive Bengali root. √dour as it happens, alternatively √doura is also possible. Both roots denote the same meaning: to run, to rush. Samsad also says the same thing. There, both "Ghora douritechhe" and "Ghora douraitechhe"—precisely these two sentences are given as examples. In both forms, the verb here is intransitive.
In Bengali, causative formation doesn't happen with nijanta, we form causative roots by adding the 'a' suffix to the main root. For instance—√kor>√kora, √por>√pora, √has>√hasa, √saj>√saja. Koritechhe, koraitechhe; sajitechhi, sajaitechhi. Rajshekhar calls our root verb 'doura'. According to him, this is a root of his designated bouva-class. But when showing verb forms, he writes—bouva-class √doura alternatively becomes bouv-class; that is, √dour. Then this form is anijanta and only archaic. From 'douritechhe' the colloquial form should be 'dourchhe'. Like koritechhe>korchhe, dekhitechhe>dekhchhe, shunitechhe>shunchhe.
But 'dourchhe' is not an established word in colloquial Bengali. On the other hand, both archaic and colloquial forms of √doura exist—se douraitechhe>dourchhe. Sometimes nijanta usage also occurs. Se ghorakе douraitechhe>dourchhe. However, in modern Bengali, instead of using the √doura root to convey nijanta meaning, we use conjunctive compound verbs. For instance: What's the point of making me run?

45. I've noticed something about 'koro/kooro'. Koro, bolo, lekho, otho, dao, khao... these are forms of the present imperative. The future imperative forms would be kooro, boloo, likho, utho, diyo, kheyo. But with negative imperatives, Bengali doesn't use the present imperative, only the future imperative. For "Don't do this," we say: eta kooro na (we don't say 'koro na'). For "Don't listen to him," we say: or kotha shuno na (we don't say 'shono na'). For "Don't write my name on stone," we say: pashaner buke likho na amar nam (we don't say 'lekho na'). For "Don't eat fruits at night," we say: rate phol kheyo na (we don't say 'khao na').
When 'na' is added to the present imperative, that 'na' is not negative but like an ornamental particle. Its function is to make the positive imperative more emphatic. For instance—do my work a bit quickly na, listen to me once na, tell him once na to help, write a couple of lines recommending the boy na, etc. In none of these examples is the action forbidden; rather, the request becomes stronger with the addition of 'na'.

46. In Bengali, negative imperatives always take the form of future imperatives; adding 'না' to present imperatives doesn't create negation but rather emphasizes the positive command. Now the question arises:
1. Does the absence of form mean the absence of meaning too? Will negative imperatives in Bengali always denote future tense? If I tell someone 'কোরো না', 'বোলো না', or 'শুনো না', will they assume I'm forbidding them from doing, saying, or hearing something in the future, and that I have no objection if they do, say, or hear it at this very moment?
2. If the answer to the above question is 'no'—that is, if we accept that the negative form of present imperatives can indeed be expressed in Bengali—then the second question becomes: How? Will the same verb form denote both present and future imperatives? In that case, wouldn't the temporal meaning of the imperative depend on context?
It's also difficult to accept that positive imperatives have a present form while negative ones don't. The example 'ওকে ছুঁয়ো না' is used to explain that although the form of the negative imperative is future tense, the meaning here indicates present tense. It seems there's truly no other way to convey or understand present imperatives without context.
However, browsing through Gitabitan, I see that in negative imperatives, the temporal boundary between present and future is often unclear, shrouded in mist. If I say:
এতদিন যা সঙ্গোপনে ছিল তোমার মনে মনে
আজকে আমার তারে তারে শুনাও সে বারতা॥
আর বিলম্ব কোরো না গো, ওই-যে নেবে বাতি।
...then it's clear that the prohibition against delay refers to the present moment. 'বিলম্ব কোরো না' must be read in conjunction with 'আজকে' from the previous line and the present imperative 'শুনাও'. Though the form is future, the meaning is present imperative. But if I say:
এখন আমার বেলা নাহি আর, বহিব একাকী বিরহের ভার–
বাঁধিনু যে রাখী পরানে তোমার সে রাখী খুলো না, খুলো না॥
...then despite the presence of 'এখন', the meaning of future imperative becomes dominant. The rakhi is already bound to the heart, and there's no immediate anxiety about it being untied right now. The main tenor of the entire lyric poem is the apprehension of possible future forgetfulness. The temporal consistency is maintained with the future imperative meaning, aligning with the future tense verb 'বহিব' in the previous line. Again, if I say:
দূরে যাব যবে সরে তখন চিনিবে মোরে--
আজ অবহেলা ছলনা দিয়ে ঢেকো না॥
...then although the previous line contains two future tense verb forms, the use of 'আজ' makes the imperative seem to express present meaning. Secondly, there's no doubt about recognition in the future; what seems unbearable is the present neglect veiled by deception. Still, a doubt about temporal meaning persists.

47. Words like 'দারিদ্র্য', 'বৈদগ্ধ্য' must certainly have y-phola. There's a difference between the y-phola in সূর্য্য and the y-phola in বৈদগ্ধ্য. The first contains double ya—that is, gemination of ya; the second has y-phola due to the ষ্ণ্য suffix. The first is √সৃ+য (active voice), the second is বিদগ্ধ+ষ্ণ্য (expressing state or quality). The committee formed by Calcutta University in 1936 for Bengali spelling reform, at Rabindranath's initiative, presented their first recommendations in May of that year. One rule stated: gemination of consonants after repha will be dropped. For example, instead of writing সর্ব্ব, কর্ম্ম, নির্দ্দয়, সূর্য্য, one should write সর্ব, কর্ম, নির্দয়, সূর্য.
But if dropping the gemination creates difficulties in understanding etymology, the gemination should remain as it is. 'কার্ত্তিক' (কৃত্তিকা+ষ্ণ) was cited as an example. In the second edition of the booklet published in October of that year, even this exception was removed. It was declared that no consonant after repha would have gemination. Even 'কার্ত্তিক' would have to be written as 'কার্তিক'. This is now the accepted practice.
Dropping gemination after repha and dropping y-phola from suffixes are different matters. Besides indicating patronymic, both ষ্ণ and ষ্ণ্য suffixes are used to express abstract quality. ষ্ণ has only 'অ', while ষ্ণ্য has 'য'. লঘু+ষ্ণ = লাঘব, গুরু+ষ্ণ = গৌরব. However, the ষ্ণ suffix is more commonly used to form adjectives from nouns: পৃথিবী>পার্থিব, হেম>হৈম, বিধি>বৈধ, চক্ষুঃ>চাক্ষুষ, সূর্য>সৌর. On the other hand, the ষ্ণ্য suffix is more frequently used to form abstract nouns from adjectives. অনুকূল>আনুকূল্য, ললিত>লালিত্য, বিপরীত>বৈপরীত্য, পণ্ডিত>পাণ্ডিত্য, সুকুমার>সৌকুমার্য, উজ্জ্বল>ঔজ্জ্বল্য, প্রচুর>প্রাচুর্য.
Some institutional newspapers nowadays no longer follow this rule. When the original adjective ends with a conjunct consonant, they casually drop the y-phola after adding the suffix. Removing the y-phola means the newly formed noun can no longer be called ষ্ণ্য-suffixed; it appears like a ষ্ণ-suffixed word. For example—দারিদ্র, বৈচিত্র, বৈশিষ্ট, যাথার্থ, দৌরাত্ম, স্বাতন্ত্র. I've seen the first two spellings in Chalantika and/or Samsad dictionary, but not the others. 'We'll write 'লালিত্য' but not 'বৈশিষ্ট্য', we'll write 'জাড্য' but not 'দার্ঢ্য'—the logic behind this thinking is probably that when the final letter is a conjunct consonant, adding or not adding y-phola amounts to the same thing; it makes no significant difference in Bengali pronunciation. So why complicate the spelling unnecessarily!
Unless a rule like 'when forming abstract nouns from adjectives, if the final letter of the word is a conjunct consonant, then ষ্ণ will be used instead of (or alternatively to) ষ্ণ্য' gains universal acceptance, how can we write 'বৈদগ্ধ' by removing the y-phola?
But the source of these two suffix names, ষ্ণ and ষ্ণ্য, for me (and perhaps for many others) is Baman Dev Chakrabarty's grammar book. But what was Baman Dev's source? Neither Vyakarana Kaumudi (Vidyasagar), Paniniyam (Lahiri-Shastri), nor Bhasha-Prakash Bangala Vyakarana (Suniti Kumar) mentions ষ্ণ or ষ্ণ্য. Instead, they primarily have অণ্‌ and ষ্যঞ্‌. Baman Dev shows words like 'কৌতূহল', 'লাঘব' formed with the ষ্ণ suffix, while other grammarians place these same words under অণ্‌, citing Paninian sutras "হায়নান্ত-যুবাদিভ্যো'ণ্‌" (for 'কৌতূহল') and 'ইগন্তাচ্চ লঘুপূর্বাৎ' (for 'লাঘব'). Again, 'বৈদগ্ধ্য' is under ষ্ণ্য suffix in Baman Dev, while other grammarians place it under ষ্যঞ্‌ suffix, citing the Paninian sutra 'বর্ণদৃঢ়াদিভ্য ষ্যঞ্‌ চ'.
The other grammatical traditions we hear about besides Panini are now extinct or obsolete. Many post-Paninian grammarians composed their works based on the Ashtadhyayi. Hemchandra Bhattacharya informs us that Bhattoji Dikshita wrote 'Siddhantakaumudi' by systematizing Panini's sutras, which is now followed almost everywhere. In Hemchandra's 'Samagra Vyakarana Kaumudi' too, ষ্ণ is called ষণ্ (অণ্), ষ্ণ্য is called ষ্যণ্ (ষঞ্), ষ্ণি is called ষিণ্ (ইঞ্), ষ্ণায়ন is called ষায়ণন্ (ফক্), ষ্ণেয় is called ষেয়ণ্ (ঢক্), ষ্ণিক is called ঠক্/ঠঞ্, ষ্ণীয় is called ছ, and so on.
All this is quite confusing. Simple things become complex. Better than this is how Samsad dictionary shows etymology, dropping the it, calling ষ্ণ as অ, ষ্ণ্য as য, ষ্ণিক as ইক, ষ্ণেয় as এয়, etc. This is much simpler and safer. But even there, there are problems. অ sometimes represents taddhita ষ্ণ (মাধব), sometimes kridanta ঘঞ্ (ভাব), অপ্ (ভব), অচ্ (ক্ষয়), খল্ (সুলভ). If we keep saying অ-অ-অ-অ for everything, can that really be called knowing the suffix!

48. Dipankar-babu says he writes his own name incorrectly. It should be দীপঙ্কর. I don't know what Acharya Suniti Kumar said about this. By the rules of consonant sandhi, দীপংকর is not incorrect spelling. When any of the twenty sparsh varnas from cha-varga to pa-varga follows, the ম্ at the end of the first word is replaced by the fifth letter of that varga. For example—সম্+চয় = সঞ্চয়, মৃত্যুম্+জয় = মৃত্যুঞ্জয়, বসুম্+ধরা = বসুন্ধরা, সম+নিহিত = সন্নিহিত, কিম্+পুরুষ = কিম্পুরুষ.
But in the case of ka-varga, there's an optional provision for anusvara. That is, just as ঙ্ can occur, so can anusvara. অহঙ্কার or অহংকার, সঙ্গীত or সংগীত, সঙ্ঘাত or সংঘাত—both forms are acceptable. In modern Bengali, writing anusvara instead of ঙ্ has become the prevalent practice. It must be remembered that in the case of original words, and even in derived words where the final letter of the first word is not ম্, anusvara will not replace ঙ্. Words like শঙ্কা, কঙ্কাল, শঙ্খ, শৃঙ্খল, শৃঙ্গার (শৃঙ্গ+√ঋ+অ), সঙ্গ (√সঞ্জ্+অ), রঙ্গ, ব্যঙ্গ, ভঙ্গুর (√ভঞ্জ্+উর) will not have anusvara. Therefore, anusvara can comfortably be used in দীপংকর or শুভংকর.
In the spelling rules proposed by Calcutta University (1936), under the section titled 'Anusvara in place of ঙ্‌ in sandhi' it stated—"If ক খ গ ঘ follow, then anusvara or optionally ঙ্‌ should be used in place of ম্‌ at the end of the word—'অহংকার, ভয়ংকর, শুভংকর, সংখ্যা, সংগম, হৃদয়ংগম, সংঘটন' or 'অহঙ্কার, ভয়ঙ্কর' etc." It also stated—"According to Sanskrit grammatical rules, when a vargiya letter follows, anusvara or the fifth letter of the subsequent varga occurs in place of ম্‌ at the end of the word, such as—'সংজাত, স্বয়ংভূ' or 'সঞ্জাত, স্বয়ম্ভূ'. Following this rule everywhere in Bengali might hinder pronunciation, but using anusvara before ka-varga will not hinder it; rather, it will simplify spelling." The latter part explains why we don't write spellings like 'সংজাত'. Suniti-babu was also on this committee.
Would it be appropriate to write সংগ, বংগ, ভংগ? সঙ্গ is formed from √সঞ্জ্ (to unite) + ঘঞ্; ভঙ্গ is formed from √ভঞ্জ্ (to break) + ঘঞ্. Originally, there's no ম্, not even ঙ্. বঙ্গ—I don't know how this word was formed. It seems to be an original proper noun. There's no question of anusvara.

49. Suniti Kumar was somewhat particular about anusvara in Bengali. For instance, his view was that if we write 'বাং