Indo-Persian translations


back to issue

THE rendering of Sanskrit texts into Persian constitutes one of the largest translation movements in world history. Sanskrit and Persian coexisted as languages and cultural systems on the subcontinent for hundreds of years, chiefly between the 14th and 18th centuries CE. During this period, intellectuals and poets performed hundreds of translations and adaptions of Sanskrit stories, knowledge systems, and philosophies into the Persian language. This sustained movement of Sanskrit based ideas, narratives, and even words into Persian resulted in a distinctive realm of Persianate culture on the subcontinent that is often characterized by the modern descriptor Indo-Persian.

Today, however, Persian translations of Sanskrit materials are largely forgotten. Few Indians know that some of the most beautiful versions of the Ramayana are written in Persian or that Ayurvedic medicine was once a topic of sustained interest among Persian-speaking intellectuals. Most Indo-Persian translations are severely understudied; many moulder away in manuscript libraries, unpublished and in want of sustained philological attention.

I analyze this fascinating and relatively untouched archive of Indo-Persian translations. I read both Sanskrit and Persian, and this rare language combination affords me the ability to examine translated Persian works alongside their Sanskrit originals. My current scholarship focuses on Persian translations of the two Indian epics, the Mahabharata (including the Bhagavad Gita) and the Ramayana. In addition, my research includes Persian adaptations of other Sanskrit works typically thought of today as ‘Hindu scriptures’, including many Upanishads and Puranas (e.g., the Bhagavata Purana).

Few scholars have been able to read Sanskrit and Persian since the early Orientalists. In addition to the difficulty of learning both tongues, modern nationalist assumptions have dissuaded students from acquiring both languages. Persian is generally considered a Middle Eastern, Islamic tongue, whereas Sanskrit is the classical language of ancient India. Universities typically teach the two in separate departments (‘Near/Middle East’ versus ‘South Asia’). Perhaps more importantly, few people today conceptualize Sanskrit and Persian in tandem, either as languages or as larger cultural traditions. We moderns far too often assume that our state of affairs is the way it has always been, especially when the evidence to the contrary is primarily found in manuscripts that few can access.

Thus far, examining Sanskrit and Persian texts side-by-side has proven rewarding. I have highlighted an entirely overlooked aspect of the Mughal Empire, namely the sustained imperial engagement with Sanskrit intellectuals and texts. I have also drawn out an under-appreciated facet of India’s literary past by showing how many early modern Sanskrit thinkers accepted patronage from Islamic rulers and engaged in cross-cultural encounters, including religious debates. For me, the core value of my work is reconstructing India’s early modern past and thereby recovering the multi-layered identities and complex composite networks realized during that historical period.


Studying the past always has implications in the present, and some of the aspects of India’s history that I have recovered clash with contemporary popular visions of precolonial India. Moderns in all nations are invested in how they imagine their history, but the subject seems constituted in particularly bizarre ways on the subcontinent. For instance, Indian politicians regularly boast about alleged western-style scientific achievements on the ancient subcontinent, including plastic surgery and genetic science. But rarely does one hear political leaders taking pride in the real splendours of premodern India, such as its almost unfathomable linguistic and cultural diversity and the vast depth of literary production in many languages. On the contrary, as Sheldon Pollock has discussed eloquently in many publications, India’s premodern learned traditions are in danger of being forgotten on the subcontinent altogether. Many Indians, it seems, are keen to disregard their rich premodern literary and cultural heritage in favour of a past fabricated for the sole purpose of claiming itself as the precursor of western scientific advances.

Even among rejected pasts there is a hierarchy, and Indo-Persian is at the bottom. In large part, this low status is because most Indians now associate Persian narrowly with Islam, and many wish to sideline India’s Islamic cultural and imperial history. This link between language and religion (Persian and Islam) is not fully inaccurate, but it is highly limiting. The Indo-Persian world encompassed people of diverse religious backgrounds, including many Hindus who wrote in Persian, worked in Persian-language courts, and read texts – including their own religious works such as the Bhagavad Gita – in Persian. The evidence for Hindu participation in the Indo-Persian realm is overwhelming, but it is either dismissed or unknown by many contemporary people who think of language and religion as inextricably linked.

More broadly, predominant views of premodern India are overly religious and communal. This is a problem that originated (and, in some circles, is still perpetuated) by westerners who posited a timeless, spiritual East upset by endemic Hindu-Muslim conflict. Emphasizing the divisive religiosity of the subcontinent frequently served western, specifically British, imperial and cultural interests. For example, alleging that the period of so-called Hindu rule was upset by the invasion of foreign Islamic peoples justified British colonialism as rescuing India from her Muslim oppressors. Although colonialism ended decades ago, a similar vision of a magnificent but interrupted Hindu past that must be reinstated galvanizes some in contemporary India. In particular, the BJP has led efforts to brand the Bhagavad Gita as a national book and include it within school curriculum. Such agendas are undergirded by the collapse of Indian and Hindu into a single identity, a process that sidelines Muslims, Christians, and others in India.


Indo-Persian translations offer up a disruptive, unsettling past in today’s political climate in two respects. The texts I study point to the absurdity in modern claims about India’s unchanging Hindu identity and long-standing communal divisions. At the same time, however, my work does not suggest a simple, catchy alternative. The sound bites of modern politics are ill-suited to accurately describe India’s diverse literary history, and indeed realizing this disconnect is part of the value of studying this past. In what follows, I point up some key aspects of the lost world of Indo-Persian translations that belie modern expectations and ought to give us pause in our assumptions about India’s past and present.


Indo-Persian thinkers often categorized Sanskrit texts differently than we do today. For example, acting under royal orders, translators in Akbar’s court rendered all eighteen books of the epic Mahabharata into Persian in the mid-1580s. This extensive translation took place under the supervision of Naqib Khan, a historian, and involved several other Mughal figures, including Mulla Shiri, Sultan Thanisari, and Badauni (a poet, fiscal administrator, and secretary, respectively). The Mughal translators were assisted by Brahmans who could read Sanskrit (the two groups communicated with one another orally in Hindi). A colophon to the translations lists the names of some of these Brahman assistants, including Deva Mishra, Shatavadhana, Madhusudana Mishra, Caturbhuja, and Shaykh Bhavan.

In a modern edition, the Persian Mahabharata constitutes over 2,000 pages of printed text (not including the Harivamsa, which was also translated into Persian but remains unpublished). But, for Akbar’s translators, the Mahabharata was not chiefly a Hindu work, as it is commonly thought of in modern India, but rather was described as a purported history (tarikh in Persian) of pre-Islamic India. Akbar’s court also understood the epic as a work of kingly advice focused on bazm o razm, the feast and the fight that epitomized Persianate rulership (the work was renamed Razmnama, Book of War, in Persian). Through the Mahabharata, Emperor Akbar wanted to learn about ancient Indian history and rulership.

The Ramayana too was not primarily considered a Hindu religious text in Persian. Rather, after its initial rendering into Persian on Akbar’s orders in the late 1580s, Rama’s tale quickly became rich fodder for poets looking to craft either martial narratives or love sagas. During the first quarter of the 17th century, Giridhar Das crafted his version of the tale, called simply Ramayan, to celebrate the war hero Rama. Around the same time, Saad Allah Masih Panipati retold the epic as a romance, appropriately named Dastan-i Ram u Sita (The Story of Rama and Sita). There are more than two dozen distinct Persian Ramayanas known today, and many of them are captivating masnavi poems.


The Bhagavad Gita was a difficult text for Indo-Persian thinkers to place, although few if any saw it as the central, unique text of Hinduism. Akbar’s translators who rendered the Sanskrit Mahabharata into Persian cut the majority of the Gita from the work. Eliding the Gita was not without precedent in the premodern world; the old Javanese version of the epic similarly shortened the Bhagavad Gita. Akbar’s translators did not explain their reason for abbreviating this section, although the decision seemed to bother nobody at the time. There are hundreds of Razmnama manuscripts extant in India today, and, with only a single exception I have identified to date, all of them maintain the elision of the Gita.1 Translators in Akbar’s court were little interested in theological texts in general. But when they desired to know more about Hindu religious thought, they endeavoured to translate not the Gita but the Atharva Veda (after attempts by three different intellectuals, this translation failed).

Several early modern writers rendered the Bhagavad Gita into Persian as a stand-alone text, much as the work is typically encountered today. We are unclear about the authorship of many Persian Bhagavad Gitas, although Muslim names commonly feature in attributions, including Abul Fazl and Fayzi, two literary stars of Akbar’s court, and Abdur Rahman Chishti (his Gita is titled Mirat al-Haqaiq, Mirror of Truths). Many of the independent Persian Bhagavad Gitas identify the text as a Hindu work but not necessarily a unique one. The Mughal prince Dara Shikuh, for example, translated (or at least oversaw the translation of) the Gita into Persian, but he was arguably more interested in the Yoga Vasishtha and the Upanishads (his translation is titled Sirr-i Akbar). Additionally, for Dara, even his engagement with the Gita must be understood within his larger vision of religious unity. Dara Shikuh saw little distinctive about Hindu thought and instead strove to show that, at their cores, Hindu and Islamic ideas were comparable.


From the beginning, Persian translations of the Sanskrit epics and the Bhagavad Gita enjoyed a mixed Muslim and Hindu audience. Persian was the language of administration in Mughal India, and many Hindus learned Persian for professional reasons from the late 16th century onward. These Hindus constituted part of the wider Persianate reading public, and some also wrote in Persian alongside their Muslim counterparts. For instance, the first quarter of the 17th century witnessed two poetic retellings of the Ramayana in Persian: one by the Hindu Giridhar Das and the other by the Muslim Saad Allah Masih Panipati.

Sometime in the 17th or 18th centuries, the number of Hindus who could read Persian reached an all-time high. At this point, more Hindus could probably read Persian than had access to Sanskrit. By the 18th century, many Hindus were more likely to encounter their religious stories and theological texts in Persian translations rather than in the Sanskrit originals. The majority of surviving manuscripts of Indo-Persian translations date to the 17th, 18th, and 19th centuries, and they offer substantial evidence that these works were repeatedly encountered by Hindus. Most obviously, many 18th and 19th century Indo-Persian translations were copied by people with Hindu names. Additionally, Persian versions of the epics and the Gita were updated according to the interests, religious and otherwise, of their new readers.


Some Hindu scribes altered aspects of the Persian translations of the epics to add a stronger Hindu context. For example, some copyists replaced Perso-Arabic words with Sanskrit-derived equivalents, such as renaming khuda and allah (God, a ubiquitous character in the Persian versions of the epics) as bhagavan or shri bhagavan (God). Some Hindus added honorifics (ji and jiu) to Krishna’s name, and some also provided basic salutations to Krishna, Ganesha, or other Hindu deities at the opening of Persian copies of the epics. These invocations are in Sanskrit, e.g., om ganeshaya namah (Homage to Ganesha), although frequently written in Perso-Arabic script. One Razmnama manuscript now housed in the Oriental Research Library in Srinagar (Acc. No. 2246, 19th century?) contains small images of individual Hindu gods preceding each book of the epic.

Yet an equal number of scribes and readers emended Persian translations of the Sanskrit epics to introduce a mix of Hindu and Islamic ideas. For example, many copies of the Persian Mahabharata and other translated Sanskrit texts open with praises to both Hindu and Muslim deities. It is not uncommon to see om ganeshaya namah followed by bismillah al-rahman al-rahim (In the name of God, the Merciful and the Compassionate). Sometimes the Hindu salutation is in Devanagari and the Islamic invocation in Perso-Arabic script, but just as often they are both expressed in Perso-Arabic letters.

Many later changes to the Persian epics have nothing to do with religion but rather concern cultural preferences. For example, numerous later copies of the Persian Mahabharata omit some of the Persian poetry that the Mughals had introduced into the translation. Certain passages on royal advice in the Razmnama are often abridged in 18th and 19th century copies, which is unsurprising given that this advice was directed specifically at Akbar. Occasionally scribes wrote Sanskrit words and even entire verses in Devanagari on the margins of translations, which signals that some later readers had access to both Persian and Sanskrit renditions of the epics. Nawal Kishore lithographed select chapters of the Persian Mahabharata in the late 1800s, and these published manuscripts show such marginal notations.


Given this brief survey of Indo-Persian translations of the Indian epics and the Bhagavad Gita, there are a few points worth emphasizing that intervene in current debates about Indian historiography and disrupt our modern assumptions. Religion featured less prominently and less divisively in how early moderns thought about the Ramayana, the Mahabharata, and the Bhagavad Gita, as compared to many in India today. Indo-Persian intellectuals largely viewed the Indian epics as historical and literary materials rather than as religious texts. When questions of religion arose in Indo-Persian translations, they were not explosive. For instance, Akbar’s translators judged the Gita inappropriate in the context of the Persian Mahabharata and so quietly abbreviated the work. However, several later Persianate intellectuals translated the Gita as a stand-alone Hindu text, even though, for some of them, the content of the work was hardly uniquely Hindu.


Religious identity was also not determinative of textual interests. It did not strike anybody in the 17th century as odd that both Hindus and Muslims read and retold the Ramayana repeatedly in Persian, resulting in more versions of the Ramayana in Persian than in possibly any other vernacular language in premodernity. Most Persian translations of the Bhagavad Gita were attributed to Muslim authors. The Razmnama was a joint translation project that involved Brahmans who could read Sanskrit and Muslim Mughals who knew Persian (the two groups both spoke Hindi).

Especially in the 18th and 19th centuries, Hindus voraciously consumed Persian translations of Sanskrit texts. This was not an attempt to recapture or reclaim their heritage or religion but rather a practical decision. Persian was a more accessible language than Sanskrit for many Hindus of the day. Hindu scribes enacted light changes to some of the texts they copied, which shows that these works were still in active circulation and being carefully and dynamically read. But I have not found any sense of resentment or even consideration that Persian Bhagavad Gitas and Ramayanas existed largely because non-Hindus had produced them. Anxiety over perceived outsiders representing (and misrepresenting) Hinduism is largely a modern concern.


Last, Persian translations of Hindu texts and their early modern receptions lack any shame about their status as cross-cultural projects. Each translation strikes a different balance between fidelity to its Sanskrit source(s) and adopting the structures and cultural expectations of Persian. Many works end up with a highly uneven landscape in this regard, especially involving religion. For example, in the Razmnama’s abridged Bhagavad Gita, Krishna is portrayed as a prophet of Allah. But elsewhere in the same translation, Krishna is treated as an Indian deva and once is even equated with khuda. Such inconsistency was not a problem for early moderns who seemed to think far less in exclusive theological binaries than we tend to today. Some slightly later intellectuals preferred to use different terminology for God in the Persian Mahabharata or decided to add appropriate honorifics. But even 18th and 19th century Hindus actively engaged with these translations rather than denying, condemning, or ignoring them.

Indo-Persian translations are not an easy history for contemporary India. This diverse body of materials cannot be smoothly incorporated into pre-existing narratives about Hindu identity or communalism, nor does it suggest an alterative coherent vision of the past. And yet, it is precisely that disruptive quality, which contradicts assumed timeless ideas – such as the relationship of religion to language and the religious nature of specific texts – without offering a simple substitution, that is of crucial value to India today.



1. A single Razmnama manuscript, copied for one Lala Hukumat Rai Jeeva in the early 1720s, includes a full translation of the Bhagavad Gita inserted into book 6 of the epic. The copy is now in the Maulana Azad Library at Aligarh Muslim University, No. Persian/Ikhbar 159(be).