Greatest Literature banner

translation graphic

Does it matter which translation you read?

Almost half the works listed in The Greatest Literature of All Time were composed in languages other than English. Much of what we read in great books—old or contemporary—are the words of relatively unknown translators, rather than the words written directly by the great authors whose names we revere.

Is this really a problem? A translation is just the original author's words put into a language we understand, isn't it? What's the difference?

A lot.

If we consider all the problems a translator faces, it's a wonder we can ever understand a work rendered into a second language. This has led some to argue we can never really appreciate a work unless we read it in the original language. When we enjoy a translated work we are responding to a different work, that of the translator, while the original necessarily remains alien to us. An extreme view is that all works are untranslatable in principle.

There is something to be said for this extreme view but it is debatable—and will be debated in literary criticism and philosophy for longer than any of us will live and read. In the meantime a more pragmatic approach may be to assume that when we read a translation we are reading a joint work of the author and the translator. How much weight we want to give each will vary, depending on the translation, the period and our personal knowledge.

We can also take comfort in realizing translation does not produce the only problems we face in understanding literature. Perhaps it does not pose even special problems. Even a work read in the original language is understood only through the several filters we bring to it: the many biases due to our culture, our class, our times, and so on. Yet somehow great works manage to speak to us through all that. The layer of interpretation that a translator adds is just one more factor to consider.

Also, it seems in practice we seldom find that different translations of any work of literature give us radically different ideas about it. More often we discover only slightly different highlights and shadows.

It is okay not to worry about any of this. A personal appreciation of a novel, poem or play based entirely on one translation, with no idea of how accurately it renders the author's intentions, is quite legitimate. You are engaged with a particular piece of art or entertainment, Y's translation of X, and that can be a worthwhile experience.

But it is equally legitimate, and quite rewarding in a different way, to study various translations of X, or to seek out the translation that makes the most of X—or possibly even makes more of it than existed in the original.

In case you are interested in this latter approach and wish to get a better idea of how to judge translations, let's look briefly at some of the problems translators face, which we'll divide into problems of meaning and problems of style (although in practice it is not so easy to separate the two).

Problems of meaning

Literary writing is full of connotation. In addition to denoting action or presenting objective description, most creative writing also implies, suggests, understates, ironically overstates, conceals, reveals, satirizes, glorifies, mythologizes, lulls or alerts. Take a few simple sentences in English:

All eyes on him, the Great Man slouched into the suddenly cooler room. "My hero," she said and turned her back. Everyone became peculiarly engrossed in their drinks.

Now imagine this passage being translated into another language literally, missing the connotations, as perhaps carried out by poorly programmed robot. It might end up as the other language's equivalent of:

With poor posture, the worthy gentleman entered the room whose heating had unaccountably failed and he was covered in eyeballs. A woman praised his heroism and rotated her body in the opposite direction. The beverages were found noteworthy by all in an odd fashion.

Not quite the same thing?

Even if the robot translator understood the idioms and their connotations in English, it would still face the daunting task of finding idioms to express those senses in the second language. Quite often an expression such as "all eyes were on him" does not have an exact equivalent in another language and so the translator has to find a colloquial alternative that means "everyone was looking at him". So why not just say "everyone was looking at him"? Because that too does not exactly express what the writer was trying to say. The writer wanted to give readers the visual image of dozens of pairs of eyes turned somewhat expectantly toward one person. But "everyone was looking at him" is too neutral, carries less impact and could give an altogether different impression. A translator who wants to put across in the second language what the writer got across in his own language might have to choose an expression very different from the literal meaning but expressing the same sense—in this case, the sense of a person entering upon an unnerving social situation.

The same thing happens with translations into English, of course. It may in fact be a worse problem, because English has a larger vocabulary with more shades of meaning than any other language. You could write an entire book on whether to call something "big", "large" or "huge", let alone "massive", "gargantuan", "humongous" or any of the other hundred-plus related terms found in a thesaurus, each of which has a slightly different connotation, each of which combines with other words to produce further connotations.

The translator strives to understand every nuance in the original text and then figure out how to communicate compactly each of those nuances in the translation. But the hard truth is that he or she can never be entirely successful. The translator has to make delicate choices as to which nuances to keep and which to drop in order to keep the overall meaning intact.

It's a daunting task, even an impossible one in many cases. Some things just don't translate well. One prominent translator estimated that only eighty percent of any literary work is translatable into another language. He considered his work a success if he managed to get fifty or sixty percent of it into a given translation.

Problems of style

Literary works have their own rhythms. Writers use certain words whose sounds or lengths or beats go together to create certain moods. They choose long sentences, short sentences or mixtures of sentence lengths. They use elaborate language, colloquial language or workmanlike language. They speed up or slow down the reader as fits their purpose. Their phrases are terse or elaborate, or some effective combination of both. They use words that sound fussy in their own tongue when it suits them and they use blunt street language when it makes their stories move another way.

By style I don't necesarily mean something artsy or overly decorative. Consider the difference between this:

Then I saw he had a gun.
Pointing at me.
And I saw my life passing before me.
Ending in my death. Damn.


It was at that precise moment I noticed he was pointing a revolver at me.
I would like to say my whole life flashed before me and I suppose it might have. But if it did flash before me, I wasn't watching, for I couldn't take my eyes off the end of the gun.
Was this all there was then?


I had always wondered if when I faced death my life would pass before me as in a film. I found out when I realized his hand was holding a gun, pointed at me.
Apparently my life was to be a short film with the most memorable part being "The End".
What, no final credits?

Each of these is written to depict the same scene in a modern, colloquial fashion with an ironic tone and very little description. Yet even within this similarity of genre, the styles vary greatly, creating different impressions of the narrator, different moods, and different levels of tension.

Much of these differences are not directly transferable to another language. The translator has to find an approximation. Either that or give up trying and stick to getting across the main points without reproducing the writer's style. For a writer who expresses himself through the subtle use of style, the latter choice by a translator can eviscerate the entire work. Yet all translators have to do it to some extent.

In the translation of poetry and drama, which often have stricter rules of form, these problems are multiplied. Poetry often ends up being translated into prose instead, or into poetry with far different rhyming schemes, metre, alliteration, line lengths and aural techniques—in effect, becoming newly composed poetry in the second language.

With drama, the translator has to consider whether to be concerned not only with how the words are read but with how they are spoken and how they work in a performance, which requires a knowledge of stagecraft. Often translators are forced to choose readability over performability, or vice versa, in any given dramatic passage or for the entire play.

This brings up the very important and slippery matter of feel for language. In our daily conversation we follow all sorts of rules of expression that are not taught in school and of which we are barely conscious. These are not rules of right or wrong grammar, but rather of right-sounding and wrong-sounding language. For example, when native English-speakers string together a series of items, we generally put the item with the most syllables at the end, as "the good, the bad and the ugly" or "Tom, Dick and Harry". Yet in some other languages, the longest item comes first. If we heard someone refer to "the ugly, the good and the bad", we'd have a feeling—without quite knowing why—that the speaker's first language was not English.

There are hundreds, maybe thousands, of other such unwritten rules of expression in English and other languages. The translator has to have a native feel for both languages he or she is dealing with in order to reproduce the text in the second language so that it sounds natural—again a task that can be only partially fulfilled at best.

Selecting a translation

Knowing all this and wanting to read the great works in translations that are closest to the intentions of the great authors, how do we select the best translations?

The short answer is that there is no "best" translation.

No translation can express everything exactly as the original work did. Every translator makes numerous choices—sacrificing style for meaning, connotation for economy, humour for literal sense, literal sense for humour, and so on—on practically every page.

Judging translations is like judging works in their original language. (Or choosing the "greatest literature of all time", for that matter.) There is no one standard of excellence to which all works can be compared. It becomes a very subjective process. Some of us get certain rewards from some works that we don't get from others, and so we "like" those works. Over time certain works are accepted as being valued by readers more than other works. These become the "great" works. Similarly with translations, particular efforts become known as classics over time. And, as with original works, translations can go in and out of fashion over the years.

Interestingly, the popularity of particular translations is not always dependent on how well they reflect the original work. No one thinks the King James version of the Bible, for instance, in its majestic English style of the early seventeenth century reflects the speech of the ancients who wrote the books in Hebrew and Greek. There's little doubt that what the original authors thought—their meanings—was also subtly changed in that translation. Yet the King James Bible has been an acknowledged classic in the English-speaking world for four centuries.

Alexander Pope's creative renditions of Homer's tales were standards for a couple of centuries and are still quoted as great poetry, but they have been replaced by translations more faithful to the Greek for serious students. Yet this does not mean that Pope's Iliad is not as good as, say, Richmond Lattimore's more faithful rendition. Just that it's different. Time will tell which one is still being read a hundred years from now—though this may not be particularly important to us now.

So, in selecting a translation to read here and now, we should not necessarily be swayed by what's popular or critically acclaimed.

What we do have to consider is what we want from a translation. Given that trade-offs and compromises are implicit in every translation, we want to find the set of trade-offs and compromises that best match our own needs.

If we just want a good, enjoyable read—as the author's original readers may have appreciated—we don't need to worry about how literally accurate a translation is. Often I find translators who are novelists, poets or playwrights themselves produce livelier versions to meet these requirements than do academics.

However, if we want to know the original author's thoughts and feelings as deeply as possible, we may have to be more concerned about scholarly accuracy and less about verve and stylishness.

Or if we want to experience the time and place out of which a work emerged, we may opt for a translation that offers complete adherence to the original form, however difficult and non-modern it may seem.

In short, we have to recognize how much translations can vary, find out what each offers, and pick one that seems to come closest to what we are seeking. Read the reviews. Browse the books in stores and libraries. On literary sites like this one and on book-selling sites, read available sample pages.

And, of course, read the articles on this site that compare translations. They provide a starting point for a few great works. Prominent translations of the works are surveyed and brief passages from several are presented for comparison. You'll probably also  be able to discern which ones I prefer, but that's just me. You may have better tastes.

— Eric