Inicio Traductores Intérpretes Ventajas Servicios Eventos Asociación Requisitos
Artículos y humor Cómo contactarnos
www.appliedlanguage.com/articles/training_translators.shtml
Training Translators
Do translators need to be educated in the art and science of translation? Is
it possible to train a translator in an academic setting? Is doing so
worthwhile, from the translator's perspective or from that of a potential
employer? The current state of affairs in the translation industry in the
Academic Training in Principle
The basic approach to academic training for translators seems to remain
unchanged from the time of the
Academic training rarely if ever includes any theoretical work, though this could well be a result of the relative lack of material to present to students; a cogent theory of translation remains to be developed. Academic training also tends to limit or exclude entirely matters related to terminology research and development, glossary or term database design and maintenance, or various computer technologies, running the gamut from basic word processing to localization of software source code. Academic training also rarely if ever includes content courses on, for instance, law, medicine, computer science, or other subjects that translators inevitably translate material on but almost have extensive, in-depth, formal academic training in. Last, academic training seems to stay away from the development or writing skills or cultural sensitivity, two areas which translators have to deal with every day of their working lives.
In essence, translation training then consists of giving students plenty of time to do very short translation assignments, then analyzing those assignments in every last detail, and ultimately discussing the nature of their work, often with an extreme focus on student errors, in great depth. In other words, students get to make mistakes that have no ramifications in their professional careers, and presumably they learn from their mistakes, as well as acquiring familiarity with terminology and various kinds of source material.
This system of education for translators places an extreme emphasis on the teaching faculty, making them responsible for selecting texts that reflect what the translation industry is currently asking translators to work on, evaluating student translations in a manner consistent with what the industry expects a translator to be able to produce, and discussing all student errors and other textual problems so as to guide the student toward developing translation skills that will meet with the industry's demands and expectations.
Faculty must therefore be very proficient translators themselves, as well as capable educators. Needless to say, not everyone can teach, regardless of how well they know their subject matter. Merely possessing a mastery of a subject does not qualify someone to teach it. Teaching is a very difficult task. Imparting knowledge and experience to another person requires not only mastery of the subject matter, but also mastery of the communication of knowledge.
Teaching translation is a very labor-intensive process, since each student's weekly assignments have to be evaluated in great detail and with extreme precision. In addition, course materials should not be recycled; the translation profession change too quickly to allow last year's translation texts to be used more than perhaps a couple of years in a row. Having students work on patents from five years ago may mean they will not learn the currently accepted format for a patent; having students work on a hardware or software manual from even three years ago will deny them much needed current terminology and subject knowledge. The same obviously holds for all subject areas. And by extension, the same will hold for any tests the students take.
Translation faculty must as a result have considerable time and enthusiasm for their subject, as well as intimate familiarity with the current state of affairs in the industry. This does not mean, however, that faculty should be working professionals who just happen to teach a few classes here and there. Students require and deserve full-time attention, meaning that faculty at best should be doing part-time work as freelance translators. As an aside, I do teach one translation course each spring (entitled "The Business of Translation"); one of the reasons I do not teach more than that is my freelance business doesn't allow me to. Covering the demands of my clients and the needs of the students in that one class pretty much absorbs all my working time. To teach more than that would, for me at least, be irresponsible.
Finally, we have the issue of the administration of a translation program. Again, the people involved in running the program should have experience in the translation profession, as well as considerable ability to manage and oversee an academic program. This may seem obvious, but it bears mentioning because the supply of such people is sufficiently small that some programs do not have such people in their administrative ranks. While having any administrator is arguably better than canceling a program, the limitations of someone who does not know the translation profession in a position of authority in a translation program can lead to misunderstandings about the time, money, and academic resources (e.g.: dictionaries, glossaries, software, etc.) that such a program requires.
Is This The Best We Can Do?
The above approach has been in use for thousands of years, give or take a bit of technology. The Thebes School of Scribes did not have the Internet, word processors, or MAT to contend with, but their methodology was roughly the same. Though I agree that practice is an essential element of training translators, I am convinced that other forms of learning should play an integral role.
First, the volume of translation. Students who spend one or two years translating a thousand words per week are wholly unprepared for the daily demands of professional translation. The average translator produces in the neighborhood of two- to three-thousand words per day, with many doing considerably more. While students should certainly on occasion study one short text very intently as an exercise in learning how to analyze and parse a text, then create the best translation possible, they should also work on translating a normal day's work for a practicing translator.
Students need to acquire the speed and accuracy of a professional. This includes learning to type quickly, knowing how to work efficiently in current software applications and on the Web, and understanding what to do when a text does not yield to translation, whether because the printing is illegible or the writing is unintelligible. Translators who lack these skills will find the market unwelcoming and uncomfortable.
Translation students also need to become very comfortable in the subject field or fields they will work in. Because most translators lack a thorough education in anything other than their languages, part of their education should include subject knowledge. This cannot come directly from the translation texts themselves, as the students simply do not generally translate texts of sufficient length or depth. Developing a thorough understanding of computer science by reading 1,000 words per week on the subject is just not possible. So content coursework, presented in both of the student's languages should be integrated into any translation program.
What's more, professional translators specialize. No translator works in every subject area; it is simply not practical to try to develop that much expertise and linguistic knowledge in that many subjects. Translators generally focus on a few related subject areas, depending on their backgrounds and interests, then cultivate their specialized knowledge and language skills so as to tailor them for translation in those fields. The same should occur in a translation program. After an introductory period of basic translation exercises and development of secondary skills like word processing and terminology management, students should pick a subject area and focus on that. A translation program should therefore offer academic tracks—such as in medicine, law, social science, natural science, finance, computers/localization, the specific breakdown does not matter—then work through both a concerted study of general material on those subjects, in both of the student translator's languages, of course, as well as translation exercises based on what the market is having translators do in those areas. This will not only develop their subject knowledge to a point at which they can confidently approach a text, but will also provide them with the terminology and knowledge of writing style necessary to create an accurate, readable translation.
Subject fields should be selected based on what the market is demanding from
translators. Certain fields, like computers/localization, are strong enough for
all major languages that they should exist for all translation students. It may
not be practical, however, to have certain translation languages working on
certain subjects. For instance, how much medical research or fundamental
science is done in
Therefore, having all students in all language combinations work through the same set of exercises in broad categories of experience is not practical or fair to the students. A student with no interest in financial translation who is studying a language combination with minimal demands for such work will not benefit from such efforts as much as she might from doing more in an area that interests her and is in demand. Similarly, the market rarely if ever asks for translators who simply have a smattering of ability and familiarity with a wide range of texts, so graduating students who fit that description is less efficient than graduating students who can handle one subject very well.
Next, technology. Different programs integrate various computer technologies and translation tools into the curriculum, with each language combination often functioning as a separate entity, and therefore each student getting a different level of training with these technologies. All students need to know how to do high-level word processing, basic DTP and HTML work, as well as deal with terminology and glossary databases, and MAT/MT tools. These technologies should be introduced through special classes, of course, but then should be a part of the daily translation curriculum. Students should be expected to create complex word-processing documents for their translations, to do database and terminology projects, to deal with HTML files, and even to open and translate text strings within software code. This will prepare them for the demands of the translation industry. Anything less will simply give them more to learn after they graduate, defeating the purpose of the training they have paid for and worked through.
Finally, theory. I realize there is as yet no well-developed theory of translation, but there is certainly enough theory within linguistics and psychology on the subjects of language, terminology, and such that translation students would benefit from an introduction to this kind of material as a way to ground them in what it is they are doing and give them some broad, general ideas of how to approach a text and translate it well. Too much theory is unnecessary, and each program, depending on its duration, will have to decide how much is sufficient. But no theory at all will leave translation students without a model to use to unify their knowledge and develop a deeper understanding and appreciation for their work.
JOHN FELSTINER
Translating Neruda: The Way To
http://www.sup.org/book.cgi?book_id=%201327
In 1980, after more than 20 years working on a translation of Pablo Neruda’s
dense The Heights of Macchu Picchu, John Felstiner, a professor of English at Stanford,
knew that he had broken new ground when he published his account of his slow
familiarization with that grand, rhetorical poem. Re-reading his lucid
Translating Neruda: The Way to Macchu Picchu is to enter the translator’s
workshop, and follow Felstiner’s to-ing and fro-ing between his sense of
English and Neruda’s Chilean Spanish.
His challenge was to learn Spanish, put himself as much as possible into Neruda’s shoes, and read all the poetry up to 1945 when Neruda wrote this poem about his rebirth as a Latin American and as a Communist. As an academic, Felstiner also read all the criticism, even uncollected pieces of Neruda’s prose. Reading Neruda is already a Herculean task (the current complete works will reach 5,000 pages), but translating him, despite such intimate knowledge, is even harder.
Felstiner brought with him a love of close reading, a capacity for admiration of the great poets, and refreshingly, a kind of honesty often lacking in academia. As an American, Felstiner caught what Neruda meant in the 1950s and 1960s, an antidote to the ‘sterile and effete symbolism of T. S. Eliot and the Wastelanders’ (Felstiner citing J. L. Grucci in 1942). He locates Neruda’s visionary poetics as somewhere between Robinson Jeffers and Walt Whitman (the latter translated by Neruda), and often contrasts him with T. S. Eliot, whose Waste Land Neruda had read and digested while posted to Ceylon in Chile’s diplomatic corps. Incidentally, that reading of Eliot resulted in Neruda sounding so different to Spanish ears in his staggering Residence on Earth poems of the 1920s and 1930s.
Two related questions occurred to me while re-reading: has Felstiner’s study stood the test of time and is his choice of The Heights of Macchu Picchu the best Neruda? To start with, I now found Felstiner’s biographical chapters too Nerudan, following the mytho-poetic line that Neruda offered his readers in his memoirs. Felstiner did not write a critical biography (we must await Adam Feinstein’s for that). Second, although Felstiner is acute on echoes of Neruda’s earlier work in The Heights of Macchu Picchu, he gives little sense of how Neruda became Neruda, what he read and his implicit criticism of Latin American and French poets, as if Neruda came out of nothing, was unique; so there is little here about that bizarre mix in Neruda of the traditional and modern. If the early chapters have dated, however, the central chapter, finding and justifying an English voice for Neruda, is still as alive and suggestive as it was in 1980. But why did Felstiner pick on Las Alturas de Macchu Picchu?
It seems that he came to this Neruda poem through Nathaniel Tarn’s
translation of 1966, and that like Neruda himself, Felstiner was charmed by
that potent talismanic site, Macchu Picchu. Furthermore, he liked the idea that
this was a watershed poem in Neruda’s life, a self-critique of his earlier
introspective verse and a rebirth into a new committed self and poet.
Let’s begin with the Inca site, one of the tourist wonders of the world. Neruda
climbed up to it from the station below in 1943, the war raging in Europe, on
his way home to Chile after three hectic years in Mexico (he had arrived there
four days after Trotsky’s assassination, befriended Diego Rivera, got Siqueiros
off the hook and jail by sending him to paint a mural in Chile, and almost had
a fist fight with Octavio Paz). Neruda found the jungle ruins so impressive
that all that he had seen in
Like any tourist today, Neruda was dazzled, and has said so in prose.
However, his poem was written two years after his ascent, in 1945, just after
officially joining the Chilean Communist party, soon after the atom bombs had
fallen, and a new war, the cold one, was beginning. Despite the magic title
then, Neruda’s poem tells readers to ignore the ruins and concentrate on the
wretched anonymous and exploited builders who built the Andean mountain-top
sanctuary.
Neruda’s political vision had crystallised during those years between the
ascent in 1943 and 1945, the writing of the poem. ‘At Macchu Picchu’, he said,
‘I felt Chilean, Peruvian and American.’ What Felstiner does not delve deep
into is Neruda’s emotional Communist Party rebirth that is at the heart of the
poem. The poem’s title, I am suggesting, makes the poem seem more exotically
Latin American than it actually is.
A further reason that seems to have stirred Felstiner is that many other translators have felt the same attraction to this epic-ishy poem. Felstiner had read them all — H. R. Hays, 1948, ‘Waldeen’, 1949, Angel Flores, 1950, George Kubler 1960, Ben Bellit, 1961, Nathaniel Tarn, 1966, and Tom Raworth, 1971, focusing particularly on how bad Bellit’s version was, and adopting and criticising Tarn’s. But this reading of earlier renderings brought him up against Neruda’s sensuousness, his untranslatable use of vowel sounds.
Having read so many versions into English, Felstiner had come up against the old trap, Frost’s dictum that poetry is what gets lost in translation. In an implicit way, Felstiner defends bilingual texts so that, like Raworth’s prose version, the translation functions as a way into the Spanish. He doesn’t actually say that — his book, naturally, ends with his translation of the long poem, proof of his apprenticeship. But he is explicit about working against earlier translations, a mark of his honesty.
Another way of working with translation that Felstiner emphasizes is voice, Neruda’s incantatory reading voice. There are three recordings of Neruda reading Alturas de Macchu Picchu; once you have heard Neruda read, as I have, you read him hearing his voice in your head, and this poem is a litany. Felstiner solves punctuation problems in the poem by listening to Neruda reading aloud, his pauses. But there is a further problem concerning Neruda’s voice, especially in this poem, that Felstiner only indirectly faces up to and that is that the poem was written to be read aloud; it is declamatory, loud and public.
A word for word rendering brings you up against this language problem; Neruda’s words enter through the ear, not the critical eye; he is close to Dylan Thomas in this aspect. Felstiner, acutely honest, admits that at moments the poem seems ‘bloated, its metaphors merely jostling each other’, that they surprise him but are not ‘true’ so that Neruda ‘virtually parodies himself’ and Felstiner as a reader, any reader, becomes ‘careless of what they signify’. Here is the crux of the translator’s dilemma, for they ‘sound’ good when read or heard aloud, but mean little when read more critically on the page, especially Canto IX, without verbs, where the verse mimics the construction of a wordy Macchu Picchu: ‘Sidereal eagle, vineyard of mist. / Bulwark lost, blind scimitar. / Starred belt, sacred bread...’ and so on for 41 lines. The point being that the difference between struggling to understand what Neruda means and scribbling down a literal version doesn’t exist, for Neruda is building a city of sounds to be heard by listeners (already the words of the title Macchu Picchu are gorgeous sounds).
So a reader of The Heights of Macchu Picchu does not get any insights into
the magical place itself, but into Neruda’s visionary self-understanding about
new forms of love (moving away from eroticism towards a patriotism directed at
his patria
Felstiner’s 1980 book, then, has not dated when it comes to letting us into the translator’s workshop, a place of sweat and labour that is usually out of bounds. Yet, there is still Felstiner’s own version into English to come. Here we brush another puzzle, for it is also based on choices and whims (failures and mistakes), like all the previous translations, and the two published since (Jack Smitt, 1991 and Stephen Kessler, 2001 — the last not yet seen; so far ten versions into English of this poem).
Here are two examples of Felstiner pushing Neruda beyond Neruda: one is at the start of the poem where Neruda has ‘Hundí la mano turbulenta y dulce / en lo más genital de lo terrestre’, translated as ‘I plunged my turbulent and gentle hand / into the genital quick of the earth’ where the unique Spanish turning of an adjective ‘genital’ into a noun by adding the neuter ‘lo’ is interpreted as ‘quick’. Felstiner knew this was a risk, admits it ‘tunes up Neruda’s Spanish slightly’, but why tune it up, what dissatisfied Felstiner with Neruda’s Spanish? Neruda uses that ‘lo’ construction twice in one line, he didn’t write ‘de la tierra’ but ‘lo terrestre’, which the English is forced to ignore completely.
My earlier point suggests that Neruda did it for ‘sound’ (the vowels ‘o’ and ‘e’) not for sense, while Felstiner worked with sense, adding ‘quick’. The later version by Jack Smitt, who presumably read Felstiner’s, corrects ‘quick’ to ‘the genital matrix’, slightly closer to Neruda, but still ignoring the point that for Neruda the matrix was sound itself.
One more example of Felstiner’s tuning up comes from the last canto, number
XII. From the metaphoric heights of Macchu Picchu, Neruda tells the silenced
builders that he wants their passion, their iron, their volcanoes, that
Stalinist steel. He says ‘file the knives you kept by you’, a call to
revolution and revenge from the pre-Colombian origins of
Here then is the paradox at the heart of Felstiner’s book; no matter how
much hard work you put in, starting from scratch on the spot in Chile in 1967
to publication in
I suggest this is mainly because Neruda turned towards, was reborn into, a bardic mode where voice creates the poetry, not words on pages. He wasn’t always like that, though it’s an element in all his earlier poetry. By the end of Felstiner’s book, I wondered if, like me, he didn’t prefer the earlier poems of Residence on Earth, for one of his chapters is an acute reading of ‘Dead Gallop’, a poem from that Neruda collection. Felstiner did say as an aside that Neruda’s ‘Widower’s Tango’ (my own favourite) is ‘splendid’; perhaps Felstiner was challenged by the long Macchu Picchu poem, but found the shorter, more surrealist ones finer poems?
At the end of his book, Felstiner admits to two thoughts. First, that he felt ‘a strange sense of having authored’ Neruda’s poem himself, that is, through translation he became a poet, and second, that his version is sometimes better than Neruda’s, and here I agree, where the risky translator is an ideal editor (at one point he regretted that Neruda didn’t have a sensitive editor). This brings us back to his ‘genital quick’ and the feeling that Neruda’s Spanish now sounds like an ‘uncannily good translation’ of Felstiner’s own translation.
Felstiner’s pioneering study on Neruda argued that a translator works at more than a poem — translating is a discipline that ‘requires every resource: history, biography, tradition, theory, philology, prosody’, as he wrote in 1998, but by then he had moved on to Paul Celan’s biography, and later to Celan’s Selected Poems and Prose, 2000. Re-reading his Nerudan phase brought me deep into the process of translating, often more exciting than the end product.
Jason Wilson translated Octavio Paz’s Itinerary
for Menard Press, 1999 and compiled Buenos Aires: A Cultural and Literary
Companion for Signal Books, 1999. He is professor of Latin American
literature at University College London.
Untranslatable Word In
Zoellick Challenges
By NEIL KING JR. in Washington and JASON DEAN in
Staff Reporters of THE
December 7, 2005; Page A1
In late September, Deputy
Secretary of State Robert Zoellick spoke to a packed house of the National
Committee on U.S.-China Relations in
The words were in italics in
the written text of the speech, and Mr. Zoellick underscored them upon
delivery. He uttered the word "stakeholder" seven times in all. Chinese officials and academics, who had felt
baffled for months about the Bush administration's true view of their country,
seized on the term.
There was only one problem:
What does it mean? The Chinese language has no corollary for
"stakeholder."
Thus began the great
translation scramble. Emails zipped across the Pacific within hours of the
speech's delivery. The State Department kicked things off with its own
translation, posted on a Chinese-language
State-run academies sent
scholarly delegations to
The bigger issue is what
Mr. Zoellick's speech gave a
long list of the Bush administration's concerns with
"
In publications and on
Internet chat sites, others offered a rosier interpretation, suggesting
translations with meanings such as "joint operator" and
"partner." If Mr. Zoellick was thinking of
"This word
'stakeholder' is relatively easy to understand in English," Wang Jisi,
director of the Institute of International Strategic Studies at the Communist
Party's Central Party School, explained last month in a leading Chinese
magazine. "It means shareholder. As a shareholder, you have to carry a
certain risk. If you and I cooperate, the share price might rise, and everyone
will benefit; if you don't work hard, and you're not responsible, everyone will
lose."
Mr. Zoellick declined
interview requests to comment on the matter. A State Department official said
the
The word "stake"
in a betting context first popped up in 16th-century
The word became trendy in
the late 1990s as politicians and others began touting "stakeholder
capitalism" and a "stakeholder society" in which companies would
work for the benefit not only of shareholders but also of workers and
communities.
The dustup in
The phrase
"win-win" was an enigma to many Chinese officials before negotiations
in 1999 over the country's accession to the World Trade Organization. Now the
phrase, whose Chinese translation is closer to "twin win," is
ubiquitous in official Chinese speeches.
Then came talk in
In
A Chinese Ministry of
Foreign Affairs official says there is no official Chinese translation yet of
"stakeholder."