Some years ago Edith Grossman translated Don Quixote, and she has an interesting essay about the process in the latest issue of Guernica. I particularly liked this passage on dealing with difficult words:
What was I to do about the inevitable lexical difficulties and obscure passages? These occur in prodigious numbers in contemporary works and were bound to reach astronomical proportions in a work that is four hundred years old. As I’ve said, normally when I translate I dig through countless dictionaries and other kinds of references—most recently Google—for the meaning of words I don’t know, and then my usual practice is to talk with those kind, patient, and generous friends who are from the same country as the author, and preferably from the same region within the country. As a last step in my lexical searches, I generally consult with the original writer, not for the translation of a word or phrase but for clarification of his or her intention and meaning. But Don Quixote clearly was a different matter: none of my friends came from the Spain of the early seventeenth century, and short of channeling, I had no way to consult with Cervantes. I was, I told myself in a tremulous voice, fervently wishing it were otherwise, completely on my own.
Two things came to my immediate rescue: the first was Martín de Riquer’s informative notes in the Spanish edition of the book I used for the translation (I told García Márquez, whose Living to Tell the Tale I worked on immediately after Don Quixote, that Cervantes was easier to translate than he was because at least in a text by Cervantes there were notes at the bottom of the page). Riquer’s editorial comments shed light on countless historical, geographical, literary, and mythical references, which I think tend to be more obscure for a modern reader than individual lexical items. Throughout his edition, Riquer takes on particularly problematic words by comparing their renderings in the earliest translations of Don Quixote into English, French, and Italian, and I have always found this—one language helping to explicate another—especially illuminating. The second piece of invaluable assistance came from an old friend, the Mexican writer Homero Aridjis, who sent me a photocopy of a dictionary he had found in Holland when he was a diplomat there: a seventeenth-century Spanish-English dictionary first published by a certain gentleman named Percivale, then enlarged by a professor of languages named Minsheu, and printed in London in 1623. The dictionary was immensely helpful at those dreadful times when a word was not to be found in María Moliner, or in the dictionary of the Real Academia, or in Simon and Schuster, Larousse, Collins, or Williams. I do not mean to suggest that there were no excruciatingly obscure or archaic phrases in Don Quixote—it has a lifetime supply of those—but despite all the difficulties I was fascinated to realize how constant and steady Spanish has remained over the centuries (as compared with English, for example), which meant that I could often use contemporary wordbooks to help shed light on a seventeenth-century text.
Thanks for the link, Denny!