P. Lal the historian had used the term "transcreation" in 1964 referring to ancient Sanskrit texts requiring contemporising to be valid and relatable. Umberto Eco's "Translation as Negotiation", states that all translation is a process of negotiation between the "letter" and the "spirit" which means more or less the same.

Both say, translation is a creative act not a linguistic test of loyalty and authenticity to the original. And both focus on cultural context to focus on the key factor that influences translation.

In our region, the general trend is towards translating Bangla to English and vice versa. While translating from other languages to Bangla is important for accessing globally relevant texts, literary or otherwise, translating to English makes less sense to me because the two languages are parked in different worlds.

My experience is neither language travels comfortable to each other's zone. However, many authors consider translation to English as a key to global status. "I am translated to English therefore I exist".

My own experience and learning

My first literary language is probably English which I learnt at around 3 years and Bengali around 8/9 years. My linguistic structure is English I was told by my linguist colleague but I know enough Bengali to be a fiction writer too. I learnt Urdu in 1964 which I didn't forget and in today's world thanks to you tube, Hindustani has made both Hindi and Urdu the dominant language of billions. We are influenced by two languages we don't call our own but who are always with us in this digital world.

This constant presence of language produces a cultural flow we don't recognise. I write Bangla using English syntax my publisher's editors, I myself recognise that. But I know both the source and the transferring language well enough to practise literature in both. That has helped me in recognising the limitations of what can be translated successfully with nuances intact and what can be called more of a textual transfer than translation.

Bengali and English may both be Indo-Aryan rooted but the divergences are linguistically, culturally and socially very wide. The idiomatic difference reflects two different world which is why poetry which I what I translate most, travels not so well either way. "Prufrock" doesn't travel to Bangla at all because the tonal part is missing just as Jibananda doesn't travel to English for the same reason.

I myself have translated poet Syed Shamshul Huq's "Boisakhe rochito Pangtimala" into English calling it" Stanzas of Summer and spring" a non- literal translation. When the poet was asked if he agreed, he said, "It conveys what I meant to say".

When another poet Al-Mahmud asked me to translate his poems to English I refused. I explained that his tone, language and rhythm were very rural and Bengal village culture driven which missed English cultural markers to hang the musical pegs of his poems. Unlike SSH's poetry, which was highly urbanised, slangy and structured with a syntax closer to English including their very urban imageries, Al-Mahmud is entirely rural. It doesn't travel.

Can everything be translated?

I love Kamal Kumar Majumder but won't try translating him because he is very Bengali and the tonal notes would almost all be withered in translation. I refused Akhtauzuzaman Ilyas's request to translate "Khowabnama" because it again is not an inter-linguistic literary product.

This issue of inter linguistic limitation is a reality I myself have encountered. It's a cultural not skills gap. Translators should accept the limits of translation which are not just linguistically structural but as others have noted cultural and socio-psychological.

English translations from fellow European languages are successful because both linguistic and cultural similarity plays a major role. Russian, French and German etc -basically European languages travel best because it's the same lingo-cultural zone. Can't say that about translations to and from Bangla.

My own personal experience is, of all the books that I have read, the one I have read the most -is a novel by Krishan Chander titled "dadr pool ke bacche" which was translated into Bangla by Mustafa Harun. It's also interesting that in translation. It becomes, "Bhogobaner sathe kicchukkhon", far from the original title but much closer to the spirit and story of the book. It worked because it's from the same language family, had same cultural markers and shared the same mental framework with the readers.

So, translation has two broad categories. a. Necessary translations. -Works that need to be translated because of their significance. They succeed as transfers to another language. A. Milton's "Paradise Lost" should be translated to Bangla no matter how it reads in another language. B. Artistic translations. Works that work in both languages. In the case of English to Bangla and vice versa, it doesn't do too well for reasons explained above. It's up to the author, translator and the readers to decide.

Intra language translations

As a social science and history researcher, most of my work is in this sector. It involves a great deal of field research where over the last half century of work, I find in issue of translation outside language comes up.

Much of our work is in the informal sector but literature is formal, polite and dominantly middle or upper class- the traditional text book class. The uphold the formal language ideology and insist on translating voices from the field on their own terms. In this process of translation, the socio-cultural context is ignored as the translator dominates the translation through socio-political power dynamics not language or social culture.

I will focus on a single word to explain. The word "gondogol" has suffered considerable controversy while travelling from the rural society to the urban academic and political zone facing high hostility. It relates to 1971 history and its how many rural folks describe that phase - "gondogoler bochor". But formal spaces call 1971 "muktijuddho", swadhinota juddho etc. Gondogol is considered demeaning to 1971 history.

This literally means that the formal world can't translate the meaning of the word effectively when it travels from the informal to their world. However, the same language in such cases uses translation, not linguistic but social. Thus, translation is also influenced by class, community, space and cultural identity within a single language.

Take another example, my mother's family is from Hooghly so they use "noon" to mean salt. My father is from Noahkhali which calls salt "lobon" but here is the interesting part. Both actively use the term to mean "nimakharam" meaning a traitor. "Lobonharam" and "Noonharam" sounds so weak. This word doesn't challenge power relations or the construction of history and culture hence is not challenged.

My final words are that even within a single language, the issue of translation exists and we need to explore this matter too as we assume universality of our so-called mother language when it carries so much socio-cultural history in the process of its construction.

This article draws from the presentation made by the author at an international conference on Translating organised by Behala College. Kolkata on 23-25- June 2026.