Apr 072014


Although originally penned in Gurmukhi script, Guru Granth showcases many Indic and Middle Eastern languages extant when it was composed 300 to 500 years ago.  The beginnings of Sikhi emerge from Punjab but Sikhs are now found all over the globe. Many newer generations of Sikhs are more at home in the language and cultural context of their world wide settlements.

How then to uncover the universality and the timelessness of the message of Guru Granth.  Clearly, it would be by forging a connection with the Guru Granth.  This requires that the message be available to all irrespective of their ethnicity or primary language.

A two-step approach is the key:  First, a comprehensive attempt to transliterate the Guru Granth initially at least in Roman script so that a non-Punjabi knowing reader can read correctly; and secondly, a thoughtful translation into English, but ultimately into many of the languages of the world. Neither task is simple.  Each mandates a multi-author, multi-year effort.  Examples from religious and secular literature are explored to highlight the rewards and pleasures as well the pitfalls and problems in translation and transliteration of sacred literature.

Video of Presentation

Body of Paper


It’s an honor to be starting the day-long conversation on how Guru Granth speaks to us today.  I need not tell you that for it to be universal and timeless, the Guru Granth must speak to us today in America as it did to countless others centuries ago in Punjab. Otherwise it loses all relevance.

Other speakers will parse the message; I will talk about how we come to terms with the message.

Today, there is perhaps no continent or country where Sikhs are not.  Wherever we have ventured, we have taken our lifestyle, family values, cuisine, song and dance, and our enterprising spirit; along with Sikhi – a unique, universal and timeless heritage that makes us what we are.

We now enjoy the globally connected existence of a nation without walls.  And now more than ever, Sikhs are growing up outside the linguistic and cultural cocoon of Punjab.

The mythological antecedents of India shaped us, not because they were essential to Sikhi, but because mythology was the overarching cultural context of India. This is now alien to a new generation of Sikhs.

We, in the diaspora, dearly value our mother tongue, Punjabi, but within our lifespan it has diminished to a transactional presence, limited to social banter, music and humor.  We are not comfortable enough to pick up a book of poetry, history or philosophy in it, so we usually don’t.  In English, too, our command of the language is largely transactional.  So, the education of the mind is often effectively stalled in both languages, even though I recognize the growing number of exceptions to this in the new generation of Sikhs.

The repository of our spiritual heritage,  Guru Granth, traditionally penned in the Gurmukhi script, contains little of present-day modern colloquial Punjabi.  With copious references to mythology, Gurbani showcases many Indic and Middle Eastern languages extant when it was composed 300 to 500 years ago.

Guru Granth does not endorse mythology but is written in the vernacular and frames the message to resonate with the average Indian of that time. Why?  Clearly, no matter the topic, teaching is best couched in the culture, context and language of the student or else the lesson is lost.

Two imperatives emerge when we engage with the Guru Granth Sahib.

Translations connect us to Gurbani without disconnecting us from the modern world in which we live.   But there are many languages in Guru Granth, including Arabic and Persian with Semitic roots, and also Sanskrit, and Braj from the tree of Indo-European languages.

Remember that India is a country of nation states that were for much of history semi- autonomous, each with its own narrative.  Add to this mélange the fact that Indian languages, like others, show many regional and dialectical variations.  English, French and Spanish are classic examples that are chockfull of such structural intricacies.

Moreover, Sikhs are and will remain a minority no matter where they live, even in India.  This reduces Punjabi in Gurmukhi script effectively to a minority language that is not commonly studied.  For many Punjabis, including Sikhs, Punjabi is mostly a spoken language today.

That’s why there is a generation of Sikhs, who do not comfortably read Gurmukhi.  True that a minimal commitment of a day or so would enable anyone to read it competently but human inertia being what it is, many Sikhs remain relatively clueless in reading the Guru Granth.  They need translations to comprehend Guru Granth and transliterations into Arabic or Roman script to read it. So, translation has an equally troublesome twin beside it and that is transliteration.

Also keep in mind that Guru Granth is mystical revealed poetry, teeming with allegories, metaphors and analogies, as all good poetry is.  In this potent mix the enormity of the task of translating Guru Granth into modern English or any other language becomes obvious. 

We need different sets of professionals to do things right:  For translation we need linguists and scholars of Sikhi who know the two languages intimately– their lexicon, grammar and historical-cultural context– and who can seamlessly travel between the two.  For transliteration we need masters of phonetics in the two spoken languages and the cultures involved. These must be mavens of the phonemes involved in the exchange.

Transliteration frustrates us but should be easier to tame. Let’s first dive into the issue of translations of Gurbani.

Some Translation Issues:

Think a moment: any conversation, no matter how simple, no matter if it is with an arch enemy or a soul mate.  Isn’t it accompanied by some thought about what the other person really means or understands? Understanding the other demands tuning into (translating) the other’s moods, gestures, body language and words, and mining them for meaning.

Translation remains the only effective insight into another mind.  War and peace stem from translating or mis-translating each other.

The literary output of past civilizations comes to us via translations.  That’s how we know of Homer, the greatest poet of ancient Greece, and of Virgil and Ovid, of similar standing in Rome, or Kalidasa, the preeminent playwright and poet of ancient India.

How accurately does a translation capture the mind and insights of a poet and the beauty of his meter and language?  Such questions are rarely laid to rest, but they give birth to new scholars of the original language and also the one in which a translation is done. Countless new PhDs result.

Times change as do cultures and languages; the vernacular becomes opaque, literary language even more so. For instance today, just a few hundred years later, Chaucer’s Canterbury Tales defy comprehension without translation into modern English.  Similar hurdles abound in engaging with classics, such as the writings of Plato, or German and Latin Masters. I offer you a line from gurbani: “Bako Shubh Rasna.”  Shubh rasna, of course, speaks of a kind generous tongue but bakko as speech and as directing one to speak has now colloquially morphed into bakwas that is best translated as cheap or trash talk in American English.

How good is any translation?  This admits no easy answer but it deserves an exploration.  As examples, let’s revisit two classics and then we will segue into translations of Guru Granth.

Omar Khayyam’s Poetry & New Testament Bible:

A Persian poet and astronomer, Omar Khayyam, lived around 1050. Some of his quatrains (Rubaiyat) have seen at least 15 translations into English and into many other languages. Why so many English versions?  Obviously, scholars saw a lack of fidelity.  Some derisively labeled the popular version by Edward FitzGerald as the “The Rubaiyat of FitzOmar.”  FitzGerald himself published five editions in 30 years with significant variations among them.

My second example, even more instructive, comes from Christianity.

Many versions of the New Testament Bible exist. Followers of John Wycliffe gave us the first English translation of the Christian Bible but it was banned in 1409. King Henry VIII authorized an English translation; later another version (The Bishop’s Bible) came in 1568.  The puritans who were part of the Church of England did not approve these versions.

In 1604, King James I convened the Hampton Court Conference for a new English translation.  This — the Authorized Version of the Bible — was prepared between 1604 and 1611by 47 scholars, all from the Church of England.  Keep in mind: seven years and 47 scholars!

Opposition to this Bible surfaced quickly.   Hugh Broughton, a Hebraic scholar, condemned it in 1611, saying that “I would rather be torn in pieces by wild horses than that this abominable translation should ever be foisted upon the English people.”

But a hundred years later, it became the Bible in all Anglican and Protestant denominations.  It remains unchallenged today, except by The Roman Catholic Church.

My purpose is not to judge any scripture but to explore problems inherent in translation and transmission of a heritage.  Many Sikh sites on the Internet are abuzz these days with translation projects.  I welcome them and I also wonder.

Translations of Gurbani:

Our sacred writings are cast in inspired poetry that, to us, is divine.  And I don’t need to tell you the difficulty in deciphering the mind of a poet when he plays with words, language and meter. For example, take the root word chinta, meaning worry.  It occurs in Guru Granth as chint, chinta, chinti, even as Chind;  I suppose the variations reflect the demands of versification.  Or look at the verse from Aasa ki Vaar: Gyan ka badha munn rahay; Gur bin gyan na hoye. Is the word gur here poetic shorthand for Guru?  Should “gur” be translated as Guru since Guru is the source of gyan or should the word be literally rendered as technique or technology?  The Guru gives us the Gur that is method or technique.  Similar concerns abide in another line: Ete channan hondya(n) Gur bin ghor andhaar. In understanding the verse either word, Guru or Gur, would suffice. The translational difficulty here is likely harmless, but not always.

Yet, critical are good faith efforts to translate the poetry of Guru Granth to capture its message; how else would we understand or adopt it as a blueprint for our life.

When my interest in the Guru Granth awakened, my intimacy with its language and grammar was minimal.  My stumbling eased when I discovered the 1966 UNESCO publication, an English translation of selections from Guru Granth and related writings by five iconic masters of the grammar and lexicon of Sikh scriptures: Trilochan Singh, Jodh Singh, Kapur Singh, Bawa Harkishen Singh and Khushwant Singh, and edited by an English poet, George Fraser.  I find this by far the best translation, way better than any that I have seen.  It captures the magic, even though now the language seems a little archaic, and the book remains incomplete.

The early 1970’s saw complete translations of Guru Granth in English. (Ernst Trump’s translation was way earlier, but it was incomplete.)  Manmohan Singh’s phraseology was often awkward, and the meaning not always clear.  As translations by Gopal Singh, Trilochan Singh, Sant Singh Khalsa, Pritam Singh Chahil, and Kartar Singh Duggal appeared I eagerly pounced on them, but was left at sea by the language, style or clarity.

In time, I graduated to exegesis in Punjabi by Bhai Vir Singh and Professor Sahib Singh.  At times they, too, appeared to mix mythological lore with the pristine purity of the Guru’s message.

All existing translations bar two are solo efforts – one person’s endeavor.  Exceptions are the UNESCO publication and the four-volume Shabdarth in Punjabi which is not a translation, but a guide to difficult words and concepts in the Guru Granth; it is published by the SGPC and no single author is identified.

Of many that are possible, I offer you brief examples where the traditional translations often leave me baffled.

Should one literally interpret Farid’s recommendation to kiss the feet of the enemy? Or, for that matter, what to make of the traditional take on the cycle of birth and death; or that even our smallest action is controlled and prewritten by God, which would then leave us no free will and no option to act otherwise.  I don’t quite see that a Creator — that gurbani assures us repeatedly cannot be measured, has no form, shape, color, caste or gender — sits out there somewhere micromanaging my puny existence, keeping track of all my sins committed or contemplated, and yet all of my actions are in accord with God’s prewritten dossier on me.  It strikes as a wholly anthropomorphic model of God as the supreme puppeteer.

Such matters often leave one wondering what exactly the Guru meant. Literal rendition leaves us entirely lost while an interpretation free of the cultural context seems to go beyond the mandate of a translator. Perhaps we need a literal rendering accompanied by a footnote with interpretation of the poetry in modern English.

As I see it, living a life in Hukum, like walking in the shadow of the Creator, transcends Gurbani’s literal rendering.  To me it asks us to live wholly in the present – in the moment – to have the courage to change the things we can change, to accept with serenity (as Hukum) what we cannot change, and the wisdom to know the difference.

Translating Gurbani is a Never Ending Process:

Gurbani is mystical poetry, full of allegories, analogies and metaphors, seldom to be literally translated.

A translator has to know two cultures intimately: their languages, idioms and traditions, the land and the people, the history and mythology that have shaped them.  And then the translator has to navigate between the two realities seamlessly.   In the process an early loss is the inability to capture the rhythmic flow and cadence of inspired poetry that transcends the literal rendition.

Given the richness of the original language, grammar and mythology, any translation project promises to be a life-long unfinished quest.    A translator needs to merge the cold-blooded mind of an analyst and grammarian with the warm joyous heart of a poet in an existence of faith.  A daunting task but surely, many dedicated translators will come out of it steeped in Sikhi.

I offer you a brief detour:  Even when the language is not so alien or abstruse, differences in interpretation between equally brilliant minds are not uncommon.  Look at the laws of any country.  Without plausible and differing interpretations of the same law a society would not need thousands of lawyers, so many different layers of judiciary, and the courts would never be so busy striving mightily to ferret out the truth.

For example: What exactly did the framers of our Constitution really mean – Is ours (USA) a Christian nation?  How is the line between Church and State to be interpreted?  Do differences in interpretation of civil rights exist or don’t they? And have some such understandings changed with time?

This says to me then that I, or any Sikh, will always have to struggle to make sense of what the Guru likely meant from an inadequate translation, no matter how good it appears to be.  And that becomes the lifelong path of a Sikh.

But when I get lost I am reassured by Gurbani that my smallest, hesitant step towards the Guru would be reciprocated by the Guru covering miles towards me.  In other words, grace would pervade and prevail.  And that with reading and cogitation a sense of the poetry would emerge.

When I realized this, I knew that I was on my way home.

That’s how I grew to like “less than perfecttranslations — that don’t seem so easy or adequate.  They place the onus on me.  I then stop and wonder if the Guru could have meant what the translator implies.  If the translations had been excellent, I might never have made the struggle my own.

Guru Granth tells us (p. 594) “Dithay mukt na hoveyee jichhar sabd na karay vichhaar,” it is not the sight of the Guru Granth but thoughtful engagement with the Word that will liberate us.

No interpretation may be guaranteed to be totally true forever. The best scholar or translator, like an honest lawyer, can only guarantee sincerity of effort, not purity of result.

Translation initiatives?  Yes, embrace them but beware of the rocky road ahead.   What we translate today is NOT for ever; it would need retranslating and tweaking by every new generation. Explore the translations, and keep at hand the original text of Gurbani.  Will it be easy?  Never!  Is it necessary? Like breath to life!

Transliteration Issues:

This now brings us to the second, but I think the more manageable riddle of Tranliteration.

Just look at many imaginative ways that the opening alphanumeric of Guru Granth Sahib – Ik Oankaar – is rendered in Roman script – Oankaar, Omkar, Onkar, Ongkar ….  more such variations abound.

A phonetics expert could teach us the correct standard enunciation devoid of the baggage of regional variations and could help us record it precisely in Roman or any other script so that a non-native speaker of Punjabi could sound it out accurately, precisely and reliably.

A phoneme is defined as the smallest contrasting unit in the sound system of a language that is capable of conveying a distinct meaning.  The American language system recognizes a set of 20 to 60 distinctive phonemes or sound units – a different number for each spoken language that can be captured in Roman script.

Rules of pronunciation, in English, often seem to be arbitrarily derived from geography, social class, and, from the worldwide British colonial experience over two centuries.  Nevertheless, there are standards of phonetics that are both trustworthy and replicable.

Phonetics is not always a perfect science or art.  For example some tribal languages have guttural sounds and clicks as distinct parts of their lexicon and there is no way to render them adequately in the Roman alphabet as we know it.

I don’t know if anyone has systematically identified how many and which specific phonemes capture the Punjabi language, and if any gaps remain. Clearly, some sounds in the Gurmukhi alphabet and in Gurbani are not easily rendered into Roman script.    But in transliterations of Gurbani today there are as many systems as there are people doing them.  In today’s expanding global reality perhaps Roman script, the most commonly used alphabet system, could stretch its dimensions beyond its 60 phonemes.

But in this process we do start with a supreme advantage.  Punjabi, like most Indic languages, is a precisely designed phonetic language and to determine the variety of phonemes that capture should not be that complex if we can find dedicated experts in phonetics to take on the task.

All languages have blind spots.  A simple example: the Punjabi ear does not distinguish between the sounds of “v” and “w” because the language does not make the distinction, whereas English has no phonemes for the distinctions between the hard “d”, the soft “d” and for the combined “dh.” A phonetics expert can capture these distinctions but likely cannot show them in written form in the Roman script as it exists today.  Exactly how and by what standards a transliteration is done into the Roman script may spell the difference between war or peace, success or failure.

Both the writer and the reader need to be on the same page.  An arbitrary decision on pronunciation serves little purpose except to sow discord.  Please note that what follows next is not a judgment of what is right or wrong.  It is merely an example where an amateur like me can easily go astray.

Look at the expression “Guru Fateh” – universally used by Sikhs.  I have seen “Guru” spelled as “Goroo” and if I was learning to read by sounding it out, as in grade school, I would be lost.

I have seen “Fateh” transcribed as “Fatih, Fatah, Phatih or Phatah.” Now look at how a reasonably sane English educated non-Sikh would sound it out.   Certainly “Guru Fatah” reminds us of the Palestinian organization “Al Fatah,” while the last choice here “Guru Phatah” pushes us towards a totally unacceptable rendering in Punjabi where “Phatah” means “torn” like a piece of cloth, and to say the Guru is torn is not so good a greeting; it sounds almost blasphemous.


In summary, the issue is how to standardize the way to capture sounds from the languages of gurbani that don’t normally exist in English into Roman script.  Remember that “sounding it out” is how all of us mastered the fundamentals of ABC and the joy of reading.

The sole purpose of translation and transliteration is to enhance communication of the message, particularly with those who are on unfamiliar territory. We are talking here of the reasonably educated common man or woman, not one at home with the intricacies of linguistics and phonetics, or the time and energy to pursue them.

There is no question that we need both a standardized translation and transliteration of Gurbani and these are not what a single scholar, no matter how good, can or should handle alone.  We need linguists steeped in Sikhi as well as mavens of phonetics to work in tandem and produce a standardized body of knowledge that remains a work in progress and puts us on the path to progress.  This is eminently doable.

Translation and transliteration are very different species of animal but each demands our full, enduring and clearheaded engagement.  Nothing less will do.


About the Author


I.J. SINGH came to the United States in 1960 on a Murry & Leonie Guggenheim Foundation fellowship. He received a PhD in anatomical sciences from the University of Oregon Medical School (now Oregon Health Sciences University), and a DDS from Columbia University. He is a professor emeritus of anatomical sciences at New York University.


He serves on the Editorial Advisory Boards of the Sikh Review (Calcutta) as well as Nishaan (New Delhi), and writes a regular internet column on Sikhi


 Posted by at 12:00 pm

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>