Methodology
The translator’s goal is to produce very precisely the English text that the author requires. That may no longer be quite the same as the French book; revisions may already have been contemplated.
Surprise! I don’t like to read in advance the text that I am translating, especially if it is long. I just begin with the first sentence – I want to find out what the book is by translating it.
The first stage of translation is one that I slightly dislike; I dictate my first draft by oral recognition software (having had tennis elbow from too much typing), and I usually have the sense that the first draft is hopelessly clumsy and irremediably inferior to the original.
Oral recognition software is remarkable but quite frustrating, as English is rich in homophones (they’re / their / there), so my first pass in revision is to sort out the – sometimes witty – confusions introduced by the software. The second pass tends to be reassuring; my first draft was not, after all, so bad and will form a workable basis for further drafts.
Next comes the part that I enjoy most: turning my second draft into good English. On the whole, I’m confident that it is accurate, at least to the extent that I’ll be able to tell from the English text where I have gone astray. At this point, as I revise, again starting from the first sentence, the ideas of the book begin to become clear, and I have a sense of discovery that tends to promote creative solutions.
Over the course of this process, I am storing up questions for the author: Why was this said here? What nuance was intended there? This passage is ambiguous, but perhaps no ambiguity was intended? I save these questions up (some answer themselves as I progress), and ask them chapter by chapter, often writing to French authors in French, asking that they put their answers next to the question in responding by email.
Once they have realised how serious I am, they usually oblige, though the problem may not be solved in one exchange. This dialogue with expert authors can be one of the privileges of translation, and very educational.
By now, I have not only a fairly accurate and idiomatic text but a clear grasp of the arguments of the book and how they fit together. More questions tend to follow at this stage, when solutions are being refined and arguments honed in English.
The answers I receive to my questions vary a great deal. In some cases, I ask about a cryptic phrase, receive a long paragraph in answer and, if the author is agreeable, substitute the long paragraph for the cryptic remark in the translation. In this way, the book can be improved by the close scrutiny inevitable in the process of translation (see ‘Tributes’). Such scrutiny can sometimes bring to light errors in the original.
Wherever possible, I look at the artworks described, since a description can be misleading without visual confirmation. And, with some thirty years of art-history translation behind me, I am sometimes able to offer an idea or illustration that helps the argument. The definitive translation is printed out on clean paper, after successive print-outs on the back of paper already printed on one side (previous final drafts). This ‘pristine’ print-out helps me to read my translation as if I were reading a book and makes errors or infelicities stand out. So I may have to print out again…
One of the odder features of translation is the need to use published English translations when the author cites a non-English original.
This can be very demanding: for example, a single sentence from Hegel, for which my sole guide is a page number in the German collected works, and which contains words that might easily be found anywhere in Hegel’s corpus.
Sometimes the author can help; sometimes there is an electronic corpus. I have, on occasion, found myself scanning through a fat volume, watching the argument progress, and thinking ‘getting warmer’, ‘getting colder’, etc., till the sentence is found.
Sometimes, too, the quotation in its published English form doesn’t fit the vocabulary used by the author in the rest of the book. It can be a relief to find that there is no existing translation and that I can therefore make my own.
I don’t like to submit a part of even a very long work until I have completed the translation of the whole.
The translation process is a learning one: a decision about how to translate a particular word, made in chapter one, has to be revised in the light of developments in chapter five; vocabulary has to be made consistent throughout the translation; quotations by disparate translators of the same author have to be modified and notes specifying this have to be added to those of the author.
This is not the kind of difficulty that could be solved by reading through the whole work first; it is very much a matter of practical detail.
Translation theory is helpful when teaching students translation, but most translation solutions are ad hoc: this works here. I have sometimes had to translate a particular quotation one way in one work, and another way in a different work to fit a different context.
The finalised translation should read as if it had been written in English. That can be a tall order.
It should also be just as the author wants – within limits. All translators will tell you stories of authors who think they know English better than they do. I was once asked, by an author whom I admire, to correct my translation of ‘les morts’ from ‘the dead’ to ‘the deads’.
With any luck, the definitive translation then passes through the hands of a copy-editor. It is a specialisation that I admire and my translation is almost invariably improved by receiving its own close scrutiny. A good copy-editor is a splendid ally.
Translators have to be fairly humble. I have rarely seen a translation that could not be improved, and have often benefited from suggestions made by my family, my colleagues, my authors or my copy-editors. But when translators do a good job, they can also be very proud: it’s a demanding art.