A Good Translator is Never Satisfied: What it’s Like to Work in Literary Translation

We’re now wrapping up Series 1 of our Think in Translation podcast and blog. Thank you to all who’ve listened and participated in our debate on literary translation! At the start of this project we sent out a two-question survey in the hopes of getting a better sense of what it is about translation and translated books that most interests our readers and listeners. The following blog post is the first of two parts (we’ll do one post per question) in which Allan Cameron, Vagabond’s founder as well as a writer and translator himself, responds to the questions we received.

Without further ado...

Thanks to those of you who replied to our questionnaire on what you would like to know about translation. Your questions were either technical or professional, and you all went for the key issues – unsurprisingly. I will start with question one and the responses in order of the number of times the subject came up.

The Question: What would you like to know about the process of translating novels?

Literal meaning vs style

I found that half of the respondents were interested in the age-old technical question of how translators resolve the incompatibility between literal meaning and style. The answer is probably that they can’t (except perhaps when translating between very close languages such as Spanish and Portuguese, or Scottish and Irish Gaelic). But they can get close, and they can resolve problems creatively. Languages do things very differently, and they are not consistent in their use of nouns and adjectives in particular. Modern national languages, which are in a sense the inventions of lexicographers and grammarians, are relatively more consistent.

Gaelic, for instance, uses different words to denote colours according to the noun they’re qualifying. Since the beginning of the nineteenth century Scottish Gaelic has been widely taught in schools (with a lengthy hiatus between the 1872 Education Act and fairly recent times), so the colours have become more standardised. This is not necessarily a gain, as the old system would have worked very well for speakers, but not perhaps for translators. A layer of complexity has been lost, and the existing history of languages goes from complex to simple, rather than the more intuitive opposite. But even national languages are infinitely more complex than invented languages, such as Esperanto.


In the questions asked of us, it was often implied that the conflict is between “literal translation” and style. A literal translation, if this means a word-for-word translation perhaps using a dictionary, would never produce anything of worth (as Google Translate must surely have discovered by now). The conflict is between the two languages, which to a greater or lesser extent are inherently different. Words have differently shaped semantic fields, so rarely does one word from one language fit exactly with one word from the other language. There may be more than one word in a language for a particular meaning (or in the jargon, more than one sememe for a given semanteme). In Javanese, there were five different “languages”. This meant that instead of denoting a difference of social status by address (“tu” or “vous”, for instance), you would choose the appropriate noun to describe an object according to the status of your interlocutor. To some small extent, this is true of English, so some would call a spade a spade, while others may call it a gardening tool or a bloody shovel. Wikipedia has just informed me that this reference goes all the way back to Plutarch and Ancient Greek. It then worked its way through several languages, proving that ideas really do travel well through translation.

Walter Kaufmann, an academic, Nietzsche expert, and translator from German to English, claimed that the mistranslation of Übermensch as “superman” profoundly misled readers of Nietzsche’s works, not only in English. He insisted on translating it as “overman”, meaning a person who can push beyond their own limitations. I don’t know German, but as I understand it, Mensch does not mean “man”, but that can mean “man” or “woman”, further complicating what may have seemed a relatively small problem to the original English translator. But that is not all; grammar and syntax also change and categorise more complex concepts differently. They affect the way we see things and how we build arguments, and these differences can be more complex and more difficult to identify. Tense, for instance, can vary greatly from one language to another (practically disappearing in East Asian languages and becoming incredibly specific in some African ones).


What are we trying to do with literary translation?

So let’s approach the argument from a different direction. What are we trying to do with literary translation? We’re attempting to reproduce the sensation of a certain work of art, using different material. Just as you would go about copying a statue in marble in a different way than you would if you were copying it in bronze, you have to translate Tolstoy into English in a different way than you would do into French. But in both cases you want to get as close as you can to the spirit of the original work. This is a translation, not a rewrite. I often resolve this problem by saying that a translation has to get as close as possible to the syntax and grammar of the target language, and remain as close as possible to the culture of the source language. I’m not a fan of resetting books in different places and times, or adopting British dialects, as the power of translation is that it does come from a different culture and it should retain that unfamiliarity. I should undermine my own dictum on syntax, grammar and culture, by saying that no pronouncement on translation should be applied unthinkingly. There are always exceptions.

My awareness of this problem has increased through my more recent work as a publisher of translations, which has highlighted how syntax varies from one language to another. Translations from Neo-Latin languages tend towards excessive hypotaxis (widespread use of subordinate clauses, hypothetical clauses and clauses in apposition), whereas translations from Latvian and Lithuanian tend towards excessive parataxis. This of course concerns translation into English, because these stylistic choices are not odd in their original languages, and were they odd in the original, then that oddness should be reproduced in the English translation. But what is plain or odd in original is not necessarily plain or odd in English. Every translation involves a series of ad hoc decisions, and the more literary the translation the more difficult the translation is likely to be, but that is the reason why this work is so fascinating and also occasionally frustrating. Good translators should never be entirely happy with their work.


I have spent a great deal of time on this matter, as it was the question most respondents were interested in, but also because it goes to the very core of what translation is and much more could be said about it. This could be the subject of another lengthier blog post.


How to become a translator

Roughly one-third of our survey respondents asked us questions about the profession of translation and how to get into it. This is also a very important question, but of a very different nature. Again my answer will not be the final word, but this time I fear that I’ll be the bearer of bad news. Firstly translation of literary fiction is a tiny part of the publishing output in English-speaking countries. We read less literary fiction than many countries (most European countries, I would say) and we read a great deal less in translation. This means less work for translators, though the global English-language output is much higher, slightly redressing the balance.

Some wanted to know how to get into this field, and I would encourage them to persevere, with all the provisos of course, because like writing translation is a very isolating experience and not suited to everyone. I’m not talking about the intellectual challenge, because clearly no one would be interested if they didn’t find that attractive. What may be less clear is whether they would want to work in a field where there is no need to ever see another human face. In the many years I have worked as a translator, I’ve met the authors on some occasions but rarely the publishers (it is a very international business, because each language pair is a closed world, so the publisher is usually in another country, often the USA). Sometimes, the relationship with the author turned into a friendship, but the author is unlikely to live round the corner, unless the translator goes to live in the country of the source language (which is a possibility and quite a few do).


The upside for translators is that they can easily carry their business around on their backs and choose any location or follow the vagaries of their partners’ careers, but it’s an upside with a significant downside: they may find themselves working on their holidays. It’s too easy to do. Just get this one chapter out of the way before you go home.

And how do you start? Like so many other professions, it sometimes chooses you. I was offered a book by Polity Press, and I’m not exactly sure how they got my name. The fact that I was asked means that their usual Italian translator must have been busy, so happenchance played its part. I found book translation an interesting challenge, and the book I started with was by an eccentric academic who specialised in historical analysis mainly drawn from literature. The book contained both prose and poetry, and it was a very good place to start. In at the deep end. After that, work just continued to flow. I never woke up one morning and said that I wanted to be a book translator. I’m very glad that it happened, but it probably wasn’t what I wanted to do at the time.

There are courses in translation now, and these may help. They will get you into a network too, and universities will want to show that their courses lead to work in the field, but for many there will be neither the time nor the money to follow that route. Publishers go by results, and if you can get that first break and they like your work, then you will progress. It is the first break that is difficult, and how difficult will depend on supply. If you know a lesser-known language, you may have a better chance, particularly now publishers are casting their nets wider. There is more work from French, German, Italian and Spanish, but there are also many more translators to compete with. Rates for lesser-known languages may be higher, and to be honest, publishers may be more forgiving of shoddy works as they probably spent a lot of time seeking out their translator.



Moving from commercial to literary translation

One of our respondents asked the question, can a translator switch from commercial translation work, such as engineering and marketing, to translating novels. I can answer this one positively: yes, they can because that was what happened to me. I was working on a PhD and teaching part-time at a university, but also very short of money. I started commercial translation with car parts (a step up from running the dishwasher at a local restaurant) and ended up mainly with legal and financial translations which paid better. Commercial translation* is initially interesting, because there is some specialist vocabulary to learn, but after a bit it can become dull. When you start book translation, you’ll take a cut in income (compared with commercial work), but as your confidence and speed build up, book translation is not as badly paid as some people claim (though I’ve rarely met a translator of any kind who was happy with his or her financial lot).

Another respondent asked how long it takes to translate a novel. This varies greatly according to the complexity of the text. As I’ve already suggested, experience really matters. The linguistic problems of translating from Italian into English are going to be different from those of German into English. The bad fit is another kind of bad fit, but the number of problems is perhaps about the same – very high but probably finite. The translator will eventually develop an instinct for turning a sentence around. Every writer writes in a different way, but each is circumscribed by the device being used – their language. When I started, I thought that there would no pay rise in the business, but effectively there is. Familiarity and experience increase speed and quality.


The process: From selection to publication to marketing

About one-quarter of our respondents wanted to know how we choose our translated novels.** This is a very interesting question, and given the number of languages and the number of novels each language has, the process cannot be exhaustive. It is at best a compromise between the random and intuition, but in this sense it is only a more extreme version of publishing novels written in English. Many good novels are not published, and many bad novels are. I have never published a translated novel which has really disappointed me, but I have very rarely published a novel originally written in English which has. I have also published translated novels which, in my opinion, probably don’t measure up to the original, possibly after having put in a great deal of editorial time. Editors of a translated novel are cut off from the original, unless they know the original language. All they can do is sense whether something is missing, and this is an unpleasant experience I have gone through, though almost always it has been counterbalanced by a sense that something has also been achieved and a door opened on a unique reading experience.


One of the reasons we (at Vagabond Voices) concentrate on “lesser-used languages” is that we can pick up untranslated classics, such Antanas Škėma’s White Shroud and A.H. Tammsaare’s I Loved a German, but we have also translated from Italian, German and Russian.

The actual process of selection overlaps with the question asked by a small number of respondents who wanted to know about the process as whole. So I will run quickly through the entire process from selection to publication and marketing. We first go through a number of readers’ reports or synopses, possibly with a short sample translation. If we’re interested, we can often ask for a larger sample translation, though this will vary from country to country. If we’re lucky, someone at Vagabond Voices will know another language which the book has already been translated into. This is quite common, as translations tend to come out in Italian, French and German long before they do in English. This is ideal and means we can get a very good idea of the book. With the synopses and sample translation, we are taking what might seem to be a very risky step into the unknown. In practice, these can provide us with quite a bit of reliable information.

Once the decision has been made to publish the book, we need to secure the rights from the foreign publisher or, in some cases, the author to whom the translation rights have reverted. Then we have to find a translator. This may be a translator known to us and who we enjoy working with, or may be a translator recommended by another publisher, which is usually a reliable source. A rate has to be agreed and a contract signed. Then we, the publisher, just have to wait. We can’t do anything until the translation has been completed, sometimes with a delay. I have never been able to complain too much about this, as in the decades I worked as a translator I was forever going through deadlines.


Once the translation is complete, we commence the editorial process. At Vagabond Voices, we always have two editors working on our novels, and this is even more important with translated novels. Some translators submit perfect or near perfect texts, some deliver serious problems that take a great deal of our time to put right. And many of course are somewhere in the middle. But the editorial process can never wholly compensate for a less than perfect translation, unless the editor knows the language the novel was translated from. It can, however, ensure that the original author’s voice does find a coherent replica in the English language.

The translation then continues, like any other book, through the cover design, the layout, the blurbs, the publicity material, the sales campaign, the marketing and publication. The translated novel’s production process is considerably longer for obvious reasons, and there are more things that can go wrong, but where there is high demand, as you would expect in small European countries, but also for some reason in more widely spoken languages as German, French, Italian and Spanish, publishers quite happy to brave the added risks. In English-speaking countries, translated novels have always been the reserve of a few dedicated publishers, but during the crisis that hit publishing in the early 2000s (in the wake of the demise of the Net Book Agreement in 1997) these either went out of business or were bought up by larger companies. One exception was Pushkin Press which coincidentally was founded in the same year that the Net Book Agreement was abolished. Around 2010 a number of new publishers became active in the field, and appear to be surviving and possibly thriving. Lower printing costs and the possibility of better stock management provided by digital printing have made a contribution to viability.

I hope that these comments will have gone some way to answering the respondents’ questions, which are probably representative of the queries many other people would have liked to raise. They have not exhausted the subject, and I would refer you to all our podcasts, but particularly Adam Freudenheim’s as it covers many of the issues raised here and also examines the translation of children’s books, an important area of publishing we’re not involved in.

In the second part of this series (to be released on 26 July) I’ll be addressing the second question in our survey: What type of translated novels do you most like? It’s clear that a number of the respondents are translators, and already know quite a bit about the trade, so I’ll mainly be trying to collate the information and see if there’s anything I can distil from the responses. I think that there is.


* To hear about Katy Derbyshire’s transition from commercial translation to literary translation, listen to episode six of our Think in Translation podcast.

** To learn about how Pushkin Press discovers books to translate, listen to episode seven of our Think in Translation podcast.

About the author: Allan Cameron is a writer and translator as well as the founder of Vagabond Voices. Read his full bio.  

Our Think in Translation project has been made possible thanks to The Space and Creative Scotland. To learn more about the project, visit the Think in Translation home page.