by David Bellos
It would therefore be quite wrong to see the progressive interpenetration of English, French, German, and Italian together with terms and phrases from the ancient source tongues, Latin and Greek, and (in the writings of Volodine) Russian and Chinese, too, as the sole product of what is now called globalization. In any case, globalization does not spread only English into other languages and cultures: it could just as well be exemplified by the spread of pizza language and the vocabulary of pasta into corner stores and fast-food joints the world over. It is also the result of long efforts by translators to raise their national languages to international status. They did not necessarily seek to make their translations sound authentically foreign. Indeed, if that is what they were really trying to do, their success has made mincemeat of the ambition, because the words they imported or mimicked have now become part of the receiving language to such an extent that they are no longer foreign at all.
No less than 40 percent of all the headwords in any large English dictionary are imports from other languages. A foreignism—be it a word, a turn of phrase, or a grammatical structure that is brought into our marvelously and infuriatingly malleable tongue by a translator seeking to retain the authentic sound of the original—has its path already mapped out. Either it will be disregarded as a clumsy, awkward, or incomplete act of translation, or it will be absorbed, reused, integrated, and become not foreign at all.
However, contemporary efforts to produce translations into English that keep something authentically foreign about them are not strictly comparable to the kind of translators’ campaigns in centuries past that made German more like English, French more like Italian, Syriac more like Greek, and so forth. The for-eignizers of today are not struggling to make English an international language, because English is the international language of the present. To some degree, they are seeking to enrich English with linguistic resources afforded by languages that are distant from it. “One subliminal idea I started out with as a translator was to help energize English itself,” Richard Pevear stated in an interview in The New Yorker.10 That creative, writerly project rests on a wish to share with readers some of the feelings that Pevear has when reading a Russian novel. He has also often said that he is not a fluent speaker of the language and relies on his partner to provide a basic crib that he then works into a literary version.11 Something similar may be true of other proponents of awkward and foreign-sounding translation styles. The project of writing translations that do the least “ethnocentric violence” to the original thus runs the risk of dissolving into something different—a representation of the funny ways foreigners speak.
The natural way to represent the foreignness of foreign utterances is to leave them in the original, in whole or in part. This resource is available in all languages and has always been used to some degree in every one of them.12
It is not easy to represent the foreignness of foreign languages in complete seriousness. It takes the wit of Chaplin or Celentano to do so for comic effect without causing offense.
What translation does in the first place is to represent the meaning of a foreign text. As we shall see, that’s quite hard enough.
SIX
Native Command: Is Your Language Really Yours?
Translators traditionally and now almost by iron rule translate from a foreign language into what is called their mother tongue. In translation-studies jargon this is called L1 translation, as opposed to L2 translation, which is translation out toward a learned or other tongue. But what exactly is a mother tongue?
We all start with a mother and it seems obvious that we first learn language in her arms. The language that your mother speaks to you is therefore what you are “born into,” which is all that can be meant when instead of “mother tongue” we call it a native language.
It is an axiom of language study that to be a native speaker is to have complete possession of a language; reciprocally, complete possession of a language is usually glossed as precisely that knowledge of a language that a native speaker has. In spite of the obvious fact that speakers of the same language use it in infinitely varied ways and often have quite different vocabularies and language habits at the levels of register, style, diction, and so forth, we proceed on the assumption that only native speakers of (let us say) English know English completely and that only native speakers of English are in a position to judge whether any other speaker is using the language “natively.”
We also know, from observation and self-observation, too, that native speakers make grammatical and lexical mistakes and find themselves lost for words from time to time. In what is now a conventional view of language use, the slips and stumbles in the speech of a native speaker are themselves part of what it means to possess the language natively. Teachers of foreign languages are expert in distinguishing between mistakes that language learners make and those that are characteristic of native speech; and for a native speaker of any language, there are some kinds of errors made by others that sound not just wrong but not native. But let us put these practical and effective uses of the distinction between “native” and “nonnative” aside. Other, much more difficult issues are involved in using terms such as mother and native to name the way we are more or less at home in the language we call our own.
We do not have to learn our mother tongue from a mother. It can be acquired just as effectively from siblings, from an au pair, or from the kids next door. What matters for normal human development is that there be a language available in our immediate environment in infancy, for no child invents a language by itself, without input from outside. We acquire our first language from whatever sources are available in our infant environment. Some children do it faster than others, some acquire wider vocabularies than others, but all children normally achieve communicative competence within a relatively narrow time band, between the ages of one and three. But the language that is acquired in those early stages of development may or may not turn out to be the one in which as adults we feel most at home. Great numbers of people the world over are not particularly skilled users of the language taught to them by their infant environment. In many circumstances, formal education replaces the infant language with one that goes on to be used in adult life as the operative means of communication.
From the disappearance of Latin as a spoken language in around the sixth and seventh centuries C.E. until the age of Descartes, Newton, and Leibniz, no mother ever spoke Latin to her child, and no child was ever born into a Latin-speaking home. However, Latin was learned by young males of the higher social classes throughout Christianized Europe for well over a thousand years. Throughout that long period, Latin was the language in which all educated Europeans operated in thought, formal speech, and writing, for purposes as varied as diplomacy, philosophy, mathematics, science, and religion. The language was taught by means of writing, and it was also spoken—in schools, monasteries, churches, chancelleries, and law courts—as the verbalization of a written idiom. All speakers of Latin in the period of its use as the primary form of communication had at least one other mother tongue, but these vernaculars were not used as tools for elaborated thinking or expression. But if a clear distinction can be made between the language learned from your mother and the language in which you operate most effectively for highborn males in Western Europe between 700 and 1700 C.E., the very concepts of “mother tongue” and “native speaker” need to be looked at again.
Examples of the difference between “first learned language” and “operative language” can be found almost anywhere. I can find several in my own family. My father learned to speak in Yiddish, the language of his mother and of his environment in London’s East End some ninety years ago. Once he started going to school, he acquired English. There is no question that he could soon do far more with it than he ever could with his mother tongue. Similarly, the mother of my children spoke Hungarian as an infant but acquired French when she moved to France at the age of five. Neither of these cases involved the loss of the mother tongue.
Descartes, Newton, and Leibniz also remained everyday speakers of their “native languages,” respectively, French, English, and German.
In many modern cases, the mother tongue that is supplanted by a learned language for higher-level activities remains only “mother’s tongue,” used exclusively for interaction with the older generation. Yiddish and Hungarian remained for my two relatives the way they spoke with their mothers and served almost no other purpose in adult life. That is fairly typical of first-generation immigrants in countries such as France, Britain, and the United States, many of whom possess a mother tongue that is stuck at the state of sophistication achieved around the age of five. But that was certainly not true of Descartes or Newton, who also wrote in French and English, respectively; it may well not be true of many millions of other bilingual speakers in the world today.
Throughout our lives we retain more or less strong emotions about the language in which we first learned songs, nursery rhymes, games, and playgroup or family rituals. These are foundational experiences, and the language in which they were experienced must surely be forever lit by the warm glow of our earliest reminiscences. But it does not automatically follow that the language of our earliest memories has any special importance as a language for what we may go on to become, or for what we take to be our personal identity.
When the first-learned language is overlaid by a language of education, it ceases to be at the forefront of an individual’s development. What is learned in the second but increasingly firstlike language are the foundational techniques of writing and counting, as well as all-important systems like the rules of baseball, alongside song lyrics and the bruising business of social interaction outside the family circle. All this sudden learning can of course be translated back into a first language, especially if the family environment supports parallel development and parents or siblings take the time to teach the child how to express all these new things in the family idiom; but without such support, few children would bother to do something so manifestly pointless (pointless, because unrelated to the social and personal uses of the newly acquired skills).
One problem with using the expression “mother tongue” to name the language in which an adult operates most comfortably is that it confuses the history of an individual’s acquisition of linguistic skills with the mystery of what we mean by the “possession” of a language. But it also does something more insidious: it acts as a suggestion that our preferred language is not just the language spoken to us by a mother but is, in some almost mystical sense, the mother of our selfhood—the tongue that made us what we are. It is not a neutral term: it is burdened with a complex set of ideas about the relationship between language and selfhood, and it unloads that burden on us as long as we take the term to be a natural, unproblematic way of naming our linguistic home.
We may all be born with the potential to acquire a language and a need to do so—with what some linguists have called a “language-acquisition device”—hardwired in our brains. But in practice, we are not born into any particular language at all: all babies are languageless at the start of life. Yet we use the term native speaker as if the contrary were true—as if the form of language acquired by natural but fairly strenuous effort from our infant environment were a birthright, an inheritance, and the definitive, unalterable location of our linguistic identity. But knowing French or English or Tagalog is not a right of birth, even less an inheritance: it is a personal acquisition. To speak of “native” command of a language is to be just as approximate, and, to a degree, just as misleading as to speak of having a “mother tongue.”
The curious ideology of these language terms is brought into clearer focus by British and American universities, which, when seeking to appoint someone as a professor of languages, conventionally state that “native or quasi-native competence” is required in the language to be taught. What can “quasi-native” possibly mean? In practical terms it means “very, very good.” Implicitly, it means that you can be very good at French or Russian or Arabic even if it is not your birthright. But the most obvious implications of the formula are, first, that a distinction can be made between those who were “born into” the given language and those who were not; and, second, that for the purposes of high-level instruction in the language this distinction is of no consequence. But that creates a curious problem. If the latter holds, how can the former be true?
Language scholars distinguish between sentences that are grammatically and lexically “acceptable” and “unacceptable” by appealing to the intuitive judgments of “native speakers.” “Native-speaker competence” is the criterion most commonly invoked for determining what it is that the grammar of a language has to explain. Now it may seem obvious that “Jill loves Jack” is a sentence of English and that “Jill Jack loves” is not, and that a grammar of English should explain why the first is acceptable and the second is not. But to ground the boundaries of what is and is not English on the judgments of native speakers alone creates a somewhat mind-bending circularity to the whole project of writing a grammar. How do we judge in the first place whether the English spoken by some individual is “native” or not? Only by appealing to the grammar, itself established by reference to the judgments of “native speakers” themselves. Yet there is no regular way for distinguishing unambiguously between native and nonnative speakers of any tongue. Most often we don’t even use any formal tests, we just take people’s word for it. And, as a result, we often make mistakes.
That is to say, speakers of English cannot reliably ascertain whether another person speaking the language acquired it in the cradle, or at school, or by some other means. And we are even less able to separate the “natives” from the “others” when it comes to written expression. I am sometimes mistaken for French when speaking the language. But I am not a “native speaker” in the commonly accepted meaning of the word: I learned French at school, from a mild-mannered teacher called Mr. Smith. When French people exclaim with surprise, “But I thought you were French,” I still blush with pride, like the good schoolboy I was. But what such flatterers really mean is not that I speak “native French” but that they took my speech to indicate a particular nationality. Nationality is of course one of the few things that most people acquire by birth—either because of the nationality of their parents (“by right of blood,” jus sanguinis) or because of where they were born (“by right of soil,” jus soli).1 The relatively short history of the European nation-state founded on linguistic uniformity has resulted in a fairly profound confusion of language with nationality, and of “native-speaker competence” with country of origin.
The passport you hold doesn’t have anything to do with your competence as a translator, nor does the language that you learned in your infant environment. What matters is whether you are or feel you are at home in the language into which you are translating. It doesn’t really help to call it “native,” and it helps even less to insist that you can translate only into a “mother” tongue. The paths by which speakers come to feel at home in a language are far too varied for the range of their abilities to be forced into merely two slots (“native” and “nonnative”), however broad or flexible the definitions of those slots may be.
Knowing two languages extremely well is generally thought to be the prerequisite for being able to translate, but in numerous domains that is not actually the case. In the translation of poetry, drama, and film subtitles, for example, collaborative translation is the norm. One partner is native in the “source-text language,” or L1, the other is native in the “target language,” or L2; both need competence in a shared language, usually but not necessarily L2. Also, the target-language translator needs to be or to believe he is in expert command of the language of the genre—as a playwright, as a poet, as a skilled compressor of meanings into the very restricted format of subtitles, and so on. Even in the translation of prose fiction, there are celebrated translation teams—Richard Pevear and Larissa Volokhonsky, for example, who together have produced new E
nglish versions of many classic works of Russian literature. A different form of collaborative transmission is involved in my own work with the novels of Ismail Kadare, who writes in Albanian, a language I do not possess beyond phrase-book level. I work from the French translations done by the violinist Tedi Papavrami and then raise my own queries with both Papavrami and Kadare, through the medium of French, which Kadare speaks well enough to discuss allusions, references, questions of style, and so on.
In cultures other than those of Western Europe, the prejudice against translating into a language that is not native is less profound, and in some places it is rejected outright. For many decades, Soviet Russia insisted that the speeches of its UN delegates be interpreted not by native speakers of the other official languages but by Russian speakers who were expert interpreters and translators into Spanish, French, English, Arabic, and Chinese. The translators’ school in Moscow developed a theory—or a cover story—to justify this politically motivated practice, according to which the essential skill of an interpreter is her complete comprehension of the original.2 Most professionals disagree, regarding unreflecting fluency in the target language as the real key to getting away with the almost unimaginably brain-taxing act of simultaneous interpretation—but for more than forty years the Russian booths at the UN were indeed staffed with what are called “L2 interpreters,” who coped with the job very well.3
L2 translation—writing in a language that is not “native”—is also quite widespread for languages that do not belong to the small group of Western tongues with established traditions of teaching one another’s languages in schools and long-standing two-way translation relations. Few “native” writers of English, French, Spanish, or German are fluent readers of Tamil, Tagalog, Farsi, or Wolof, and among them fewer still wish to devote their time to translation. For writers in these and most of the other languages of the world, the only way to get an international hearing is to put the work into a world language learned at school or else through emigration or travel. The effort often backfires. L2 translations from contemporary China and Albania are notoriously dreadful. Many of the sillier examples of translation mistakes in commercial material and tourist signage are visibly produced by L2 translation. But it would be futile to insist that the iron rule of L1 translation be imposed on all intercultural relations in the world without also insisting on its inescapable corollary: that every educational system in the world’s eighty vehicular languages devote significant resources to producing seventy-nine groups of competent L1 translators in each cohort of graduating students. The only alternative to that still-utopian solution would be for speakers of the target languages to become more tolerant and more welcoming of the variants introduced into English, French, German, and so forth by L2 translators working very hard indeed to make themselves understood.